Cha ac e iza ion o Oligonucleo ide
Mic oa ay Hyb idiza ion
Mic oa ay Fab ica ion by Ligh -Di ec ed in si u Syn hesis –
De elopmen o an Au oma ed DNA Mic oa ay Syn hesize ,
Cha ac e iza ion o Single Base Misma ch Disc imina ion and he
Posi ion-Dependen In luence o Poin De ec s on Oligonucleo ide
Duplex Binding A ini ies
Von de Uni e si ¨a Bay eu h
zu E langung des G ades eines
Dok o s de Na u wissenscha en (D . e . na .)
genehmig e Abhandlung
on
Thomas Naise
gebo en in Bay eu h
1. Gu ach e : P o . D . Alb ech O
2. Gu ach e : P o . D . Jose K¨as
3. Gu ach e : P o . D . Thomas Fische
Tag de Ein eichung: 14.12.2007
Tag des Kolloquiums: 04.07.2008
Abs ac
The p esen hesis ocuses on nucleic acid hyb idiza ion be ween ee- loa ing a ge se-
quences and complemen a y end- e he ed oligonucleo ide p obes on he su ace o DNA
mic oa ays.
Hyb idiza ion expe imen s we e pe o med on oligonucleo ide mic oa ays (DNA Chips)
which we e ab ica ed wi h an au oma ed syn hesis appa a us (de eloped in he amewo k
o he p esen hesis). The wo king p inciple o he mic oa ay syn hesize is based on a
pho ochemically con olled in si u syn hesis p ocess [Fod91]. By means o he combina o-
ial app oach up o 25000 di e en (a bi a y) p obe sequences can be ab ica ed in pa allel
– s a ing om nucleo ide building blocks (NPPOC-phospho amidi es [Has97]) – di ec ly
on he su ace o he mic oa ay. G ea lexibili y wi h ega d o he choice o p obe se-
quences is achie ed by use o ’ i ual pho omasks’ [SG99] on he basis o a spa ial ligh
modula o (Digi al Mic omi o De ice, DMDTM, Texas Ins umen s Inc.). A mic oscope
p ojec ion pho oli hog aphysys em is employed o p ojec he ’ i ual masks’ (i.e. he pho-
omask images shown on he DMDTM) on o he su ace o he mic oa ay subs a e. Spa-
ially con olled pho odep o ec ion o pho olabile NPPOC p o ec i e g oups ( ollowed by
coupling o a u he nucleo ide building block) enables massi ely pa allel syn hesis o
DNA p obe sequences. In he au oma ed syn hesis p ocess mic oa ays a e ou inely ab i-
ca ed o e nigh . Compa able in si u syn hesis sys ems a e cu en ly ope a ed only a e y
ew ins i u ions a ound he wo ld.
We i s epo he applica ion o phospho us dend ime subs a es [LB03] in he in si u
syn hesis o DNA mic oa ays. Wi h he phospho us dend ime unc ionaliza ion we ob-
ained supe io esul s in ega d o sensi i i y, su ace homogenei y, signal/backg ound-
a io and eusabili y o he mic oa ays.
We pe o med mic oa ay hyb idiza ion expe imen s o in es iga e he impac o single
base de ec s (delibe a ely in oduced single base misma ches and single base bulges)on he
binding a ini y o oligonucleo ide duplexes. This is pa icula ly in e es ing wi h ega d o
geno yping mic oa ays which a e inc easingly employed as a molecula diagnos ics ool
o he de ec ion o single nucleo ide polymo phisms (SNPs).
In a numbe o expe imen s we in es iga ed he la ge in luence o he single-de ec posi ion
[Wic06; Poz06; Nai06b] on duplex binding a ini y. The o igin o his posi ional depen-
dence – which is appa en ly no in ag eemen wi h he ( wo-s a e) nea es -neighbo model
– had no been iden i ied so a . We disco e ed ha he in luence o he de ec posi ion is
no es ic ed o single base misma ches bu can also be obse ed o single base bulge de-
i
ec s. On he basis o he double-ended zippe model [Gib59; Ki 69] (assuming luc ua ing
end-domain-opening o he oligonucleo ide duplex) we could ep oduce he expe imen-
ally obse ed posi ional in luence. Mo eo e , ou heo e ical in es iga ions on he zippe
model indica e a signi ican posi ional in luence in ega d o he con ibu ions o he in-
di idual Wa son-C ick nea es -neighbo pai s o he Gibbs ee ene gy o oligonucleo ide
duplex o ma ion. The p esen wo k p o ides o he i s ime a heo e ical app oach o
he posi ional-dependen nea es -neighbo model (PDNN) o Zhang e al. [Zha03].
In he in si u syn hesis p ocess o DNA mic oa ays andom poin -mu a ionsa e in oduced
in o he mic oa ay p obe sequences. We ha e shown – expe imen ally and by means o a
nume ical model – ha syn hesis- ela ed de ec s signi ican ly a ec mic oa ay hyb idiza-
ion cha ac e is ics.
Wi h ega d o single base misma ch disc imina ion, we disco e ed signi ican di e ences
be ween DNA/DNA- and RNA/DNA hyb idiza ion: expe imen al esul s indica e an im-
p o ed disc imina ion o pu ine-pu ine misma ch base pai s in RNA/DNA-duplexes.
Fo he expe imen ally obse ed, unexpec edly high s abili y o G oup II single bulges
[Zhu99] we p o ide an explana o y app oach on he basis o he zippe model.
The selec ion o app op ia e (speci ic and sensi i e) p obe sequences is o c ucial impo -
ance o success ul applica ion o DNA mic oa ay echnology. Ou expe imen al esul s
con i m p e ious esul s [Lue03] which show ha only a small ac ion (in piecewise sec-
ions abou 20-30%) o a long cRNA a ge sequence is a ailable o hyb idiza ion wi h
he complemen a y mic oa ay p obes. Reduced binding a ini ies a e assumed o o igi-
na e om he in luence o a ge seconda y s uc u e. Using so wa e ools o an isense
oligonucleo idedesign (accoun ing o a ge accessibili y) we we e able o p edic e icien
mic oa ay p obes. We disco e ed e idence ha mechanically s able seconda y s uc u es
(e.g. double-helical sec ions) in e e e wi h he mic oa ay su ace (s e ical hind ance) and
hus esul in educed mic oa ay binding a ini ies.
ii
Ku zzusammen assung
In de o liegenden A bei wu de die Hyb idisie ung einzels ¨angige RNA- und DNA-
Ta ge -Sequenzen mi den ¨u die einzelnenSequenzen spezi ischen Oligonukleo id-P obe-
Sequenzen au de Obe l¨ache on DNA-Mic oa ays un e such .
Die hie bei e wende en Oligonukleo id-Mic oa ays wu den mi els eines im Rahmen
diese A bei en wickel en Mic oa ay-Syn hese-Sys ems au de Basis eines au oma i-
sie en, pho oli hog aphischkon ollie en Syn hesep ozesses [Fod91] he ges ell : Mi Hil-
e eines kombina o ischen Ve ah ens wu den – ausgehend on chemisch modi izie en
NPPOC-Phospho amidi Basenbaus einen [Has97] – in pa allele Weise bis zu 25000 un-
e schiedliche ( ei w¨ahlba e) P obe-Sequenzen in si u au dem Mic oa aysubs a syn-
he isie . Eine hohe Flexibili ¨a hinsich lich de Auswahl de P obe-Sequenzen wi d du ch
die Ve wendung i uelle ”Pho omasken” [SG99] – au de Basis eines Mik ospiege-
la ays (DMDTM Digi al Mic omi o De ice, Texas In umen s Inc.) – e eich . Mi els
eine Mik oskop-P ojek ions-Pho oli hog aphie-Kon igu a ion wi d das Bild des Spa ial
Ligh Modula o s au die Subs a obe l¨ache abgebilde , um die En sch¨u zung pho olabile
NPPOC-Schu zg uppen – und dami die nach olgende Ankopplung wei e e Basenbaus ei-
ne – ¨aumlich kon ollie zu s eue n.
Mi den in unse en Expe imen en e s mals bei eine in si u Syn hese e wende en Phos-
pho us-Dend ime -Subs a en [LB03] konn en im Ve gleich mi ande en Linke /Space -
Molek¨ulen die bes en Resul a e in Hinsich au Sensi i i ¨a , Homogeni ¨a , Signal/Un e -
g und-Ve h¨al nis und Wiede e wendba kei , e ziel we den. Mi dem Mic oa ay-Syn he-
size k¨onnen in einem au oma isie en P ozess DNA Mic oa ays mi Tausenden on be-
liebig w¨ahlba en P obe-Sequenzen p ak isch ¨ube Nach he ges ell we den. Ve gleichba e
Sys eme s ehen bislang nu wenigen Fo schungsein ich ungen zu Ve ¨ugung.
Anhand on Hyb idisie ungsexpe imen en wu de un e such , wie sich (geziel eingebau e)
Einzelbasen-De ek e au die Bindungsa ini ¨a on Oligonukleo id-Duplexen auswi ken.
Dies is in Hinsich au die Anwendung on SNP-Mic oa ays in e essan , die zu De ek-
ion on Single Nucleo ide Polymo phismen – gene isch beding en Va ia ionen einzelne
Basenpaa e – in zunehmenden Maße in de molekula en Diagnos ik eingese z we den.
In eine Reihe on Expe imen en lag das Augenme k au dem s a ken Ein luss de De-
ek posi ion [Wic06; Poz06; Nai06b] au die Bindungsa ini ¨a . Die U sache diese o en-
sich lich im Wide sp uch zum wo-s a e nea es -neighbo -Modell s ehenden Posi ionsab-
h¨angigkei konn e bislang nich e kl¨a we den. Unse e Expe imen e zeigen e s mals, dass
die Posi ionsabh¨angigkei nich nu bei Misma ch-De ek en [Wic06; Poz06; Nai06b], son-
de n in e gleichba e S ¨a ke auch bei single bulge De ek en au i . Au de Basis eines
iii
Zippe -Models des Oligonukleo id-Duplexes, bei dem eine luk uie ende pa ielle Dena-
u ie ung de Duplexenden angenommen wi d (die auch zu olls ¨andigen Dissozia ion
¨uh en kann), konn e de expe imen ell beobach e e Posi ionsein luss ep oduzie we den.
Da ¨ube hinaus zeigen unse e heo e ischen Un e suchungen (au de G undlage des Zip-
pe Modells) einen signi ikan en Posi ionsein luss hinsich lich de Gewich ung de ein-
zelnen nea es -neighbo -Bei ¨age zu Duplexs abili ¨a au . Die o liegende A bei lie e
dami e s mals einen heo e ischen Ansa z ¨u das posi ional-dependen nea es -neighbo
Modell (PDNN) on Zhang e al. [Zha03].
Ve u sach du ch S eulich und ande e Ein l¨usse we den im Ve lau de in si u Syn hese
zu ¨allige Punk mu a ionen in den Mic oa ay-P obe-Sequenzen gene ie . Expe imen ell
und in nume ischen Modellen konn e gezeig we den, dass diese Syn hesede ek e maß-
geblich die Hyb idisie ungseigenscha en en sp echende Mic oa ays beein lussen.
Eine de aillie e Analyse des Ein lusses de einzelnen Misma ch-Basenpaa e au die Bin-
dungsa ini ¨a zeig hinsich lich de Misma ch-Disk iminie ung signi ikan e Un e schiede
zwischen DNA/DNA- und RNA/DNA-Hyb idisie ung au , die wah scheinlich au un e -
schiedliche Duplexs uk u en zu ¨uckzu ¨uh en sind.
F¨u die expe imen ell beobach e e, e gleichsweise hohe S abili ¨a on G oup II single
bulge [Zhu99] De ek en konn e ein E kl¨a ungsansa z au de Basis des Zippe -Modells
ge unden we den.
F¨u die Du ch ¨uh ung on Mic oa ayexpe imen en is die Auswahl geeigne e P obe-
Sequenzen mi eine hohen Bindungsa ini ¨a hinsich lich de dazu komplemen ¨a en Ta -
ge -Sequenzen on en scheidende Bedeu ung. Wi konn en ¨uhe e Resul a e [Lue03] be-
s ¨a igen, wonach – e mu lich du ch den Ein luss de Ta ge sekund¨a s uk u – nu ein
ela i kleine Teil (abschni sweise e wa 20 bis 30%) eine meh e e hunde Nukleo i-
de langen cRNA Ta ge -Sequenz ¨u die Hyb idisie ung mi den Mic oa ay-P obes zu
Ve ¨ugung s eh . Au de G undlage eines So wa e Tools ¨u das Design on An isense-
Oligonukleo iden (Be ¨ucksich igung de Ta ge sekund¨a s uk u ) konn en die expe imen-
ell bes imm en Hyb idisie ungse izienzen de Mic oa ay-P obe-Sequenzen ep oduzie
we den. Da ¨ube hinaus en deck en wi Hinweise da ¨u , dass mechanisch s abile Sekun-
d¨a s uk u en (z.B. doppelhelikale Abschni e) du ch Wechselwi kung mi de Mic oa ay-
Obe l¨ache – au g und on s e ische Hinde ung de Duplexbildung– die Bindungsa ini ¨a
he abse zen.
i
Con en s
1 In oduc ion 1
2 Fundamen als 7
2.1 Nucleic Acids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.1 The Double-Helix S uc u e . . . . . . . . . . . . . . . . . . . . 8
2.1.2 S abilizing In e ac ions . . . . . . . . . . . . . . . . . . . . . . . 9
2.1.3 Di e ences be ween DNA and RNA . . . . . . . . . . . . . . . . 14
2.2 Biological Func ions o Nucleic Acids . . . . . . . . . . . . . . . . . . . 15
2.2.1 The Cen al Dogma o Molecula Biology . . . . . . . . . . . . . 15
2.2.2 Genomic DNA . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.2.3 Genes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2.4 Gene Exp ession . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2.5 Exp ession Regula ion . . . . . . . . . . . . . . . . . . . . . . . 19
2.2.6 Biological Func ions o RNA . . . . . . . . . . . . . . . . . . . 22
2.3 Nucleic Acid Hyb idiza ion . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.3.1 Kine ics o Nucleic Acid Hyb idiza ion . . . . . . . . . . . . . . 24
2.3.2 The Nea es -Neighbo Model . . . . . . . . . . . . . . . . . . . 27
2.3.3 Zippe -Model o he Oligonucleo ide Duplex . . . . . . . . . . . 30
2.3.4 Fu he Models o he DNA Mel ing T ansi ion . . . . . . . . . . 34
2.4 Des abiliza ion o Oligonucleo ide Duplexes by Poin De ec s . . . . . . 34
2.4.1 Single Base Misma ches . . . . . . . . . . . . . . . . . . . . . . 35
2.4.2 Single Base Bulges . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.4.3 In luence o he De ec Posi ion . . . . . . . . . . . . . . . . . . 39
2.5 Solid-Phase Syn hesis o Nucleic Acids . . . . . . . . . . . . . . . . . . 41
2.5.1 P inciples o Solid-Phase Chemical Syn hesis . . . . . . . . . . . 41
2.5.2 Nucleic Acid Syn hesis by he Phospho amidi e Me hod . . . . . 42
2.6 DNA Mic oa ays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
2.6.1 Mic oa ay Applica ions . . . . . . . . . . . . . . . . . . . . . . 47
CONTENTS
2.6.2 The De elopmen o DNA Mic oa ay Technologies . . . . . . . 49
2.6.3 Cha ac e is ics o Mic oa ay Hyb idiza ion . . . . . . . . . . . . 51
2.6.4 Fu he Reading . . . . . . . . . . . . . . . . . . . . . . . . . . 53
2.7 DNA Chip Fab ica ion by Ligh -Di ec ed In Si u Syn hesis . . . . . . . . 54
2.7.1 Pho oli hog aphic Con ol o he Combina o ial Syn hesis P ocess 54
2.7.2 ”Maskless” Pho oli hog aphy and Combina o ial Chemis y . . . 57
3 De elopmen o he DNA Mic oa ay Syn hesize 61
3.1 Mo i a ion and O e iew . . . . . . . . . . . . . . . . . . . . . . . . . . 61
3.2 The Maskless Mic op ojec ion Pho oli hog aphy Sys em (MPLS) . . . . . 63
3.2.1 The UV Ligh Sou ce . . . . . . . . . . . . . . . . . . . . . . . . 63
3.2.2 Digi al Mask P ojec ion Using a Digi al Mic omi o De ice . . . 65
3.2.3 The Image P ojec ion Op ics . . . . . . . . . . . . . . . . . . . . 66
3.2.4 UV-Sensi i e Pho och omic Films . . . . . . . . . . . . . . . . . 69
3.2.5 Ch oma ic Co ec ion o he P ojec ion Op ical Sys em . . . . . . 70
3.2.6 UV Ligh In ensi y and Uni o mi y o Illumina ion . . . . . . . . 71
3.2.7 Op ical Sys em Pe o mance Tes ing . . . . . . . . . . . . . . . . 73
3.2.8 Ou look - Fu he Possible Applica ions . . . . . . . . . . . . . . 78
3.3 The Fluidics Sys em . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
3.3.1 The Syn hesis Cell . . . . . . . . . . . . . . . . . . . . . . . . . 80
3.3.2 A gon Bubble T apping . . . . . . . . . . . . . . . . . . . . . . 83
3.4 Au oma ed Mic oa ay Syn hesis . . . . . . . . . . . . . . . . . . . . . . 84
3.5 Pe o mance o he Mic oa ay Syn hesize . . . . . . . . . . . . . . . . 84
4 Ligh -di ec ed in si u Syn hesis o DNA Mic oa ays 87
4.1 Ligh -Di ec ed in si u Syn hesis o DNA Mic oa ays . . . . . . . . . . . 87
4.2 P epa a ion o Phospho us Dend ime Subs a es . . . . . . . . . . . . . 91
4.3 No ewo hy Cha ac e is ics o he Mic oa ays . . . . . . . . . . . . . . . 94
4.3.1 Au o luo escence o he Chip Su ace . . . . . . . . . . . . . . . 94
4.3.2 Hyd ophilici y o DNA Mic oa ay Fea u es . . . . . . . . . . . . 95
4.3.3 Hyb idiza ion wi hou De e gen - Unspeci ic Adso p ion . . . . . 96
4.3.4 I e e sible Ta ge Adso p ion . . . . . . . . . . . . . . . . . . . 96
4.3.5 Robus ness o he Phospho us Dend ime Su ace Coa ing . . . . 97
5 DNA Mic oa ay Analysis 99
5.1 Hyb idiza ion Signal Acquisi ion - Expe imen al Se up . . . . . . . . . . 99
5.1.1 The Hyb idiza ion Chambe . . . . . . . . . . . . . . . . . . . . 100
5.1.2 Epi luo escence Mic oscope . . . . . . . . . . . . . . . . . . . . 102
i
CONTENTS
5.1.3 Image Acquisi ion wi h an EM-CCD Came a . . . . . . . . . . . 103
5.2 Quan i a i e Analysis o Mic oa ay Hyb idiza ion Signals . . . . . . . . 103
5.3 Real- ime Moni o ing o Mic oa ay Hyb idiza ion . . . . . . . . . . . . 105
5.3.1 Hyb idiza ion Bu e . . . . . . . . . . . . . . . . . . . . . . . . 106
5.3.2 Mic oa ay Washing P ocedu es . . . . . . . . . . . . . . . . . . 106
6 In luence o Poin -De ec s on Oligonucleo ide Duplex Binding A ini ies 109
6.1 Mo i a ion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
6.2 Concep ion o he Mic oa ay Hyb idiza ion Expe imen s . . . . . . . . . 111
6.3 DNA Mic oa ay Design . . . . . . . . . . . . . . . . . . . . . . . . . . 111
6.3.1 Chip Design - Quan i a i e Analysis o Hyb idiza ion Signals . . 113
6.3.2 Single Base De ec Expe imen s . . . . . . . . . . . . . . . . . . 114
6.4 Hyb idiza ion Assays and Image Analysis . . . . . . . . . . . . . . . . . 115
6.4.1 Oligonucleo ide Ta ge s . . . . . . . . . . . . . . . . . . . . . . 115
6.5 Dominan In luence o he De ec Posi ion . . . . . . . . . . . . . . . . . 115
6.6 Misma ch Disc imina ion in DNA/DNA Duplexes . . . . . . . . . . . . . 121
6.6.1 Expe imen al Resul s . . . . . . . . . . . . . . . . . . . . . . . . 121
6.6.2 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
6.7 In luence o Flanking Base Pai s . . . . . . . . . . . . . . . . . . . . . . 128
6.8 Misma ch Disc imina ion in DNA/DNA and RNA/DNA Duplexes . . . . 132
6.8.1 Ou line o he Expe imen . . . . . . . . . . . . . . . . . . . . . 132
6.8.2 Resul s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
6.8.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
6.9 Single Base Bulge De ec s . . . . . . . . . . . . . . . . . . . . . . . . . 137
6.9.1 S a is ical Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 137
6.9.2 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
6.10 Compa ison o Single Base Misma ches and Single Base Bulges . . . . . 143
6.11 Binding A ini ies o Duplexes Con aining Mul iple De ec s . . . . . . . 146
6.11.1 Resul s and Discussion . . . . . . . . . . . . . . . . . . . . . . . 147
7 Modeling he In luence o Poin De ec s on Duplex S abili y 151
7.1 The Double-Ended Zippe Model . . . . . . . . . . . . . . . . . . . . . . 151
7.2 S ochas ic Simula ion o Oligonucleo ide Duplex S abili y . . . . . . . . 153
7.2.1 S ochas ic Simula ion wi h he Gillespie Algo i hm . . . . . . . . 153
7.2.2 Simula ion Resul s . . . . . . . . . . . . . . . . . . . . . . . . . 155
7.3 Pa i ion Func ion App oach o he Double-Ended Zippe Model . . . . . 159
7.3.1 Implemen a ion o he Pa i ion Func ion App oach (PFA) . . . . 160
ii
Glossa y
oligonucleo ide
sho nucleic acid s and
pe ec ma ch
duplex consis ing o wo comple ely complemen a y s ands; de ec - ee duplex
p obe
a mic oa ay p obe is used o de ec /iden i y one speci ic nucleic acid a ge se-
quence; p obes a e ypically oligonucleo ide p obes (leng h<100 n ) o se e al hun-
d ed n long cDNA sequences; p obes a e e he ed in a egula a angemen (a ay)
– wi hin he mic oa ay ea u es – on he solid suppo
p obe sequence mo i
in he p esen wo k his exp ession is used o he pe ec ma ching p obe sequence
ha is complemen a y o he oligonucleo ide a ge sequence employed in a single
base de ec hyb idiza ion expe imen . Single base de ec p obes a e de i ed om
he ’p obe sequence mo i ’ by subs i u ion, inse ion o dele ion o a single base.
The p obe sequence mo i may be sho e han he a ge oligonucleo ide used in he
expe imen . Hyb idiza ion signals om he comple e se o single base de ec p obes
co espond o he ’de ec p o ile’.
single base bulge
de ec in a nucleic acid duplex which o igina es om a su plus unpai ed base in one
o he wo s ands; he su plus base can adop a s acked-in con o ma ion o a looped-
ou con o ma ion and can esul in signi ican educ ion o he binding a ini y
single base misma ch
de ec in a nucleic acid duplex which o igina es om a non-Wa son-C ick base pai ;
he educed binding a ini ies is employed o de ec ion o SNPs and poin -mu a ions
a ge
ee nucleic acid sequence whose iden i y and abundance a e o be de ec ed in he mi-
c oa ay assay; o de ec ion a ge sequences a e commonly labeled wi h luo escen
dyes o wi h bio in
xi
Chap e 1
In oduc ion
Almos all cells o he human body, ega dless o he cell ype, con ain he same gene ic
ma e ial. Howe e , owing o epigene ic ac o s (e.g. CpG me hyla ion) he cell ypes di e
in hei gene exp ession – o example, genes which a e s ongly exp essed1in one cell
ype, may no be exp essed in o he s. Knowledge on gene exp ession is he key o un-
de s anding he indi idual gene unc ions and he complex in e ac ions be ween he abou
20,000 o 25,000 genes o he human genome.
DNA mic oa ays a e a key echnology o massi ely pa allel analysis o gene exp ession.
The wo king p inciple o DNA mic oa ays is based on nucleic acid hyb idiza ion: se-
quen ial Wa son-C ick base pai ing be ween he bases o wo complemen a y nucleic acid
s ands esul s in he o ma ion o a ela i ely s able double-helical duplex. Nucleic acid
hyb idiza ion is highly speci ic – al eady a single misma ched (non-Wa son-C ick) base
pai can signi ican ly educe he binding a ini y [Nel81; Pa 82].
The sequence-speci ic hyb idiza ion be ween complemen a y s ands is employed o he
pu pose o molecula ecogni ion (Fig. 1.1): su ace- e he ed single-s anded p obes (o
known sequences) a e employed as sequence-speci ic sca enge s o complemen a y a ge
sequences in solu ion. Hyb idized a ge molecules (bound o he su ace) can be de ec ed
by means o adioac i e o luo escen dye labels.
On DNA mic oa ays he same de ec ion p inciple is applied in pa allel ashion (Fig. 1.2).
Owing o he high speci ici y o nucleic acid hyb idiza ion housands o e en millions o
di e en a ge sequences can be de ec ed simul aneously. DNA mic oa ays comp ise a
egula a ay o mic oa ay ea u es, small a eas, each o which is co e ed wi h su ace-
e he ed single-s anded DNA p obes o a well-known sequence. Indi idual mic oa ay
ea u es (and hus he co esponding p obe and a ge sequences) can be iden i ied by hei
posi ion on he mic oa ay.
1Gene exp ession – he con e sion o gene ic in o ma ion in o gene p oduc s – can be unde s ood as
’gene ac i i y’.
1
In oduc ion
Figu e 1.1: Nucleic acid hyb idiza ion be ween su ace- e he ed p obe s ands and
complemen a y a ge s ands in solu ion. Nucleic acid hyb idiza ion is based on se-
quen ial Wa son-C ick base pai ing be ween complemen a y sequences o nucleo ides
and esul s in he o ma ion o a ela i ely s able double-helical nucleic acid du-
plex. Nucleic acid hyb idiza ion is e e sible (dissocia ion is a o ed by inc eased
empe a u es) because he indi idual binding in e ac ions (hyd ogen bonding and
base s acking in e ac ions – no co alen bonds in ol ed) be ween he base pai s a e
ela i ely weak. Ta ge s s ands a e labeled by co alen linkage o a luo escen dye,
o al e na i ely, by bio inyla ion.
In a gene exp ession p o iling expe imen he messenge RNA (mRNA) sequences (indi-
ca o s o he indi idual genes ansc ip ional ac i i ies) a e isola ed om he biological
sample, ampli ied (i necessa y, e.g. by in i o ansc ip ion), and labeled o de ec ion.
Subsequen ly he complex mix u e o a ge sequences o be analyzed is applied (in hy-
b idiza ion bu e solu ion) on o he su ace o he mic oa ay. The a ge s ands can eely
di use a ound and in e ac wi h he su ace- e he ed mic oa ay p obes, un il hey a e cap-
u ed by a complemen a y p obe and o m a s able duplex.
A e emo al (washing-o ) o unhyb idized a ge s, he hyb idiza ion signal, which p o-
ides in o ma ion on he quan i y o he indi idual a ge sequences, is commonly de ec ed
by means o luo escen ma ke s. Compa ison o he hyb idiza ion signals wi h he co -
esponding hyb idiza ion signals om a e e ence sample (by dual-colo analysis on he
same chip, o by means o wo single-channel mic oa ays) enables iden i ica ion o genes
ha ha e been up- o down egula ed. Some comme cial pla o ms enable gene exp ession
p o iling on a genome-wide scale.
Geno yping analysis is a u he impo an mic oa ay applica ion: Single nucleo ide poly-
mo phisms (SNPs) – sequence a ia ions in which single nucleo ides di e be ween he
membe s o a species2(o e en be ween he wo alleles in diploid cells) – ha e a s ong
in luence on he pheno ype. SNPs a e esponsible o he majo i y o gene ic a ia ions
2The human genome con ains abou 3 million SNPs. Thus, abou one in a housand base pai s is
subjec o his ype o inhe i able gene ic a ia ion.
2
In oduc ion
AB C
Figu e 1.2: Nucleic acid hyb idiza ion on he mic oa ay. Th ee di e en su ace-
e he ed p obe species a,band ca e loca ed sepa a e om each o he wi hin he
co esponding mic oa ay ea u es A, B and C. A complex mix u e o di e en a ge
sequences is applied o he mic oa ay su ace. D i en by di usion (o ac i e mix-
ing) a ge s mo e a ound and in e ac wi h he di e en p obes. I a a ge mee s
a complemen a y p obe, a s able duplex can a ise. Thus, he a ge ge s cap u ed
by he complemen a y p obe. A e he hyb idiza ion he unbound a ge s can be
emo ed by washing-o . The emaining hyb idiza ion signal ( luo escen signal) o
he hyb idized p obes p o ides in o ma ion on he quan i ies o he indi idual a ge
sequences. In his example we obse e hyb idiza ion signals only a ea u es A and
B. We conclude ha he sample mix u e con ains a ge s sequences ha a e comple-
men a y o he p obes aand b. The sample does no con ain a ge s complemen a y
o p obe c.
wi hin a single species. They a e also associa ed wi h a p edisposi ion o a a ie y o dis-
eases. Mo eo e , SNPs a e associa ed o indi iduals’ esponse o pa hogens, chemicals,
d ugs, accines, and o he agen s. SNP mic oa ays make use o he speci ici y o el-
a i ely sho 12 o 30me oligonucleo ide p obes o de ec single misma ched base pai s
o igina ing om SNPs [Con83]. Geno yping a ays a e a aluable ool in genomics e-
sea ch, pha maceu ical esea ch (wi h a ocus on he indi idual esponse o pha maceu ical
agen s) and now inc easingly in medical diagnos ics.
Fu he applica ions o DNA mic oa ays include esequencing assays3and he iden i ica-
ion o pa hogens.
Lab-scale ab ica ion o DNA mic oa ays on he basis o s anda d echniques equi es
conside able echnical and inancial e o s.4To p o ide a lexible and a o dable basis o
DNA mic oa ay hyb idiza ion expe imen s we de eloped a DNA mic oa ay syn hesize
3Resequencing a ays a e used o he sea ch o mu a ions wi h espec o a well-known e e ence
sequence. An impo an applica ion is he iden i ica ion o (possibly new) i us s ains.
4These include, o example, he acquisi ion o a mic oa ay spo ing obo ( o be ope a ed in a clean
oom en i onmen ) and conside able unning expenses o p esyn hesized mic oa ay p obes.
3
In oduc ion
based on he wo k o Singh-Gasson e al. [SG99]. Based on a pho ochemically con olled
in si u syn hesis p ocess, he DNA p obe sequences a e syn hesized om nucleo ide build-
ing blocks, di ec ly on he su ace o he mic oa ay. The use o expensi e ch omium
pho omasks (and associa ed mask alignmen ) is ci cum en ed by means o a spa ial ligh
modula o (DMDTM) ob ained om a comme cial ideo p ojec o . Compa able in si u syn-
hesis sys ems a e cu en ly ope a ed only a a ew ins i u ions a ound he wo ld.
E en hough DNA mic oa ays ha e become a well-es ablished echnology, he unde lying
physicochemical p inciples o DNA mic oa ay hyb idiza ion a e no ye ully unde s ood
[Le 05; Poz06]. Fo example, an un esol ed p oblem in he applica ion o DNA mic oa -
ays is he lack o p edic abili y o he hyb idiza ion e iciency o DNA mic oa ay p obes.
The mal s abili y o oligonucleo ide duplexes (in solu ion-phase) is well desc ibed by he
nea es -neighbo model [C o64; Tin73; B e86; F e86], which is accoun ing o hyd ogen
bonding and also o base-s acking in e ac ions be ween adjacen base pai s. The mody-
namic pa ame e s o nea es -neighbo double s o base pai s we e de i ed om solu ion-
phase hyb idiza ionexpe imen s [San98]. The nea es -neighbo model is widely employed
o he p edic ion o duplex mel ing cha ac e is ics (mel ing empe a u es, Gibbs ee ene -
gies o duplex o ma ion) – o example, o he design o PCR p ime s and o he design
o DNA mic oa ay p obe sequences. The la e applica ion, howe e , is ques ionable: on
DNA mic oa ays, due o a ious su ace-e ec s and ab ica ion- ela ed e ec s, he e a e
signi ican di e ences wi h espec o solu ion-phase hyb idiza ion [Hel03; Le 05; Bin06;
Poz06]. Mo eo e , he seconda y s uc u e o long a ge sequences esul s in a es ic ed
a ge accessibili y. Thus, he binding a ini y o indi idual mic oa ay p obes is also go -
e ned by he complex a ge seconda y s uc u e [Lue03; Ra 05].
In con as o solu ion-phasehyb idiza ions udies, ecen mic oa ays udies[Wic06; Poz06;
Nai06b] epo a la ge in luence o he posi ion o single base misma ch de ec s on he hy-
b idiza ion signal. A posi ion dependen in luence o single base de ec s is no conside ed
by he ( wo-s a e) nea es -neighbo model5and hasn’ been explained so a . Acco ding o
Pozhi ko e al. [Poz06] he e is li le e idence ha mic oa ay hyb idiza ion e iciencies
can be accu a ely p edic ed wi h so wa e ools on he basis o nea es -neighbo he mo-
dynamic pa ame e s de i ed om solu ion-phase expe imen s.
In he expe imen al pa o he p esen hesis pa icula in e es is on he in luence o poin
de ec s (single base misma ches and single base bulges) on mic oa ay binding a ini ies.
We sys ema ically in es iga e he in luence o de ec ype and de ec posi ion on p obe-
a ge binding a ini ies. In he same con ex we in es iga edi e ences be ween RNA/DNA
5The nea es -neighbo model, on he basis o misma ch base pai nea es -neighbo pa ame e s [All97],
is also employed o misma ched duplexes [All97; San04]. The model does no conside a posi ion
dependen in luence, excep o he ou e mos base posi ions.
4
In oduc ion
and DNA/DNA hyb idiza ion. Ou heo e ical in es iga ion on he in luence o poin de-
ec s on duplex binding a ini ies is based on a zippe model [Gib59; Ki 69] o he oligonu-
cleo ide duplex.
Fu he expe imen s add ess a a ie y o poo ly unde s ood in luences on DNA mic oa ay
hyb idiza ion. These include:
• andom de ec s in he mic oa ay p obes (gene a ed by he in si u syn hesis p ocess)
a ec he hyb idiza ion cha ac e is ics [Job02; Ga 02; Hel03]
• he complex seconda y s uc u e o long a ge sequences (widely belie ed o be a main
ac o in luencing he e iciency o hyb idiza ion [Lue03])
•nonspeci ic c oss-hyb idiza ion
•di usion limi a ion – local deple ion o he hyb idizing a ge molecules can esul in
inhomogeneous hyb idiza ion signal in ensi ies and slowed-down hyb idiza ion kine -
ics [Pap06; Dan07]
5
In oduc ion
6
Chap e 2
Fundamen als
2.1 Nucleic Acids
Fo i s ou s anding ole in molecula biology DNA (deoxy ibonucleic acid) is o en e-
e ed o as he ”molecule o li e”. Like a bluep in genomic DNA con ains he he edi a y
in o ma ion, ins uc ions o g ow and sus ain all o ms o li e.
In he highe euca yo ic o ganisms he genomic DNA is densely packed on ch omosomes
inside he cell nucleus. Each ch omosomecomp ises a single double-helical DNA molecule.
The leng h o he human ch omosomes is a ying be ween 50×106and 250 ×106base
pai s (co esponding o leng hs be ween 1.7 and 8.5 cm). O e all, s e ched end- o-end,
he DNA helix con ained in a single human diploid cell is abou 2 me e s long. The in o -
ma ion densi y in he densely packaged nucleus is abou 1021 bi /cm3 1 ( o compa ison:
he in o ma ion s o age densi y on a DVD disc is abou 109bi /cm2).
The biological unc ion o DNA is he sa e s o age o gene ic in o ma ion. Genomic DNA
is basically a ead-only memo y and in his way compa able o he CD-ROM d i e o a
compu e . Pa s o he genome ( he genes) a e ead in he ansc ip ion p ocess o p oduce
RNA ansc ip s o he DNA sequence. RNA is a a he ola ile in o ma ion ca ie in he
ongoing p ocessing o gene ic sequence in o ma ion. In he abo e analogy i is he e o e
compa able wi h he wo king memo y (RAM) o a compu e . Howe e , RNA is mo e e -
sa ile: i s no jus an in o ma ion ca ie bu a he (in o m o unc ional RNA) a c ucial
pa o he ansla ional machine y and in ol ed in egula o y p ocesses.
1The es ima e is based on a cell olume o 8 µm3and a genome size o 3·109base pai s - which is
abou he size o he human genome.
7
Fundamen als
2.1.1 The Double-Helix S uc u e o Nucleic Acids
The s uc u e o deoxy ibonucleic acid (DNA) was disco e ed by James Wa son and F an-
cis C ick in 1953. A ew mon h ea lie , Linus Pauling epo ed a iple-helix model o
he DNA-s uc u e, which assumed ha he phospha e g oups a e a anged in he in e io
o he helix. The model was based on high esolu ion elec on mic og aphs showing he
DNA as cylind ic ib ils wi h a diame e o 1.5 nm. The w ong assump ion o a iple helix
o igina ed om he inco ec measu emen o a oo high packaging densi y.
Wa son and C ick showed ha unde physiological condi ions DNA has indeed a double-
helical s uc u e (Fig. 2.1). The hyd ophobic bases a e loca ed in he cen e , whe eas he
hyd ophilic phospha e g oups a e loca ed a he ou side o he helix. The disco e y o Wa -
son and C ick elies on he wo k o Rosalind F anklin, whose X-Ray s uc u al analysis o
DNA ib es p o ed ha DNA has indeed a helical s uc u e.
Figu e 2.1: Wa son-C ick model o he DNA double-helix. Canonical (Wa son-
C ick) base pai s comp ise ei he adenine ( ed) and hymine (blue), o guanine (g een)
and cy osine (yellow) bases. The suga -phospha e-backbones o he wo s ands
(shown in g een and cyan) o m a igh -handed double helix. The ideal B-DNA
s uc u e was gene a ed wi h he make na web-se e which is based on he NAB
(Nucleic Acid Builde ) by Tom Macke [Mac98]. Image isualiza ion was pe o med
wi h he UCSF Chime a molecula modeling sys em [Pe 04]. A h ee-dimensional
s e eo iew o he DNA s uc u e is shown in he appendix, in Fig. B.19.
8
Nucleic Acids
Ano he impo an hin was p o ided by E win Cha ga . Acco ding o Cha ga ’s ule
he nucleobases A and T, jus as he nucleobases C and G always occu wi h he a io o
abou 1:1, independen o he biological o igin o he DNA.
2.1.2 Nucleic Acid Duplex S uc u e - S abilizing In e ac-
ions
The DNA double-helix shown in Fig. 2.1 is composed o wo complemen a y single-
s anded DNA molecules. I ’s well-known ha he duplex s abili y o igina es om in e -
s and hyd ogen bonding be ween complemen a y base pai s A·T and C·G (see Fig. 2.2).
Howe e , i is less well-known ha a simila deg ee o s abiliza ion o igina es om π-π
in e ac ion be ween closely-s acked a oma ic bases (π-s acking) [Koo01].
N
N
N
O
H
H
N
N
N
N
N
O
H
H
HN
N
N
NN H
H
N
N
O
O
CH3
H
Adenine Thymine
Guanine Cy osine
Figu e 2.2: Canonical (Wa son-C ick) base pai s A·T and C·G, comp ise a bi-
cyclic pu ine base (adenine o guanine) and a monocyclic py imidine base ( hymine
o cy osine). A·T is s abilized by wo, C·G by h ee hyd ogen bonds.
A DNA molecule is basically a lexible polyme -chain2made up o nucleo ide monome s
(as shown in Fig. 2.3).
A nucleo ide consis s o a he e ocyclic base (i.e. adenine, cy osine, guanine o hymine
- in RNA he hymine is eplaced by u acil) and a pen ose suga - ing (2-deoxy ibose in
DNA and ibose in RNA), which in conjunc ion wi h a phospha e g oup cons i u es he
suga -phospha e-backbone o he DNA molecule. Apa om he nucleobases lis ed abo e,
u he nucleobases occu na u ally in RNA (e.g. pseudou idine in ans e -RNA).
Fig. 2.3 shows ha subsequen nucleo ides a e linked ia a phosphodies e bond (i.e. o e
he phospha e g oup) be ween he 3’- and 5’-ca bons o he deoxy ibose suga s. Because
2He e one needs o dis inguish be ween he highly lexible single s anded molecule (pe sis ence leng h
alues p o ided in he li e a u e ange om lp≃0.5 nm o 1.3 nm [Koh06]) and he signi ican ly
mo e igid double-s anded DNA duplex (lp≃45-50 nm [Hag88]).
9
Fundamen als
RNA
DNA
PROTEIN
ansla ion
ansc ip ion
DNA eplica ion
e e se
ansc ip ion
RNA eplica ion
di ec ansla ion
o DNA
Figu e 2.7: F ancis C icks Cen al Dogma o Molecula Biology. The e a e gene al
ans e s o biological sequence in o ma ion (solid a ows) and specialized ans e s
(dashed a ows). A gene al ans e o sequence in o ma ion is om DNA ia he
ansc ip ion p ocess o messenge RNA. mRNA is ansla ed in o a polypep ide chain
which olds in o a p o ein. Ano he gene al ans e is he eplica ion o DNA du ing
cell di ision. Specialized ans e s a e ela ed o i us ep oduc ion (e.g. e e se
ansc ip ion) o ha e be pe o med in i o (e.g. di ec ansla ion o DNA sequences
in o p o eins).
2.2.2 Genomic DNA
The genomic DNA con ains he he edi a y in o ma ion o an o ganism. La ge pa s o he
genome a e a anged as genes, o ganiza ional uni s ha a e ansc ibed in o one o se e al
gene p oduc s. The unc ion o o he noncoding pa s o he genome, p e iously e med
”junk DNA” is less well unde s ood.
In he simple p oca yo ic o ganisms (e.g. bac e ia) he DNA is packaged in ing-shaped
ch omosomes and plasmids, which a e esiding in he cy oplasm. In he mo e complex
euca yo ic o ganisms he DNA is con ained in he nucleus, well-sepa a ed om he cy o-
plasm (see Fig. 2.8). Ch omosomes con ain he DNA in a highly compac , hough o de ed
and accessible o m. The double-helical DNA ilamen con ained on a single ch omosome
can be se e al cen ime e s long. Enla ged o a diame e o 2 mm he DNA ilamen would
ex end o e a leng h o abou 30 km.
In conjunc ion wi h a complex o his one p o eins, ac ing as spool a ound which he DNA
double-s and is wound up ( oughly wo supe helical u ns o abou 80 base pai s a ound
he cylind ical his one oc ame ), he DNA o ms a nucleosome. Coun less nucleosomes
condense in o an o de ed supe s uc u e, o ming a ch oma in ib e wi h a diame e o
abou 30 nm. The ch oma in ib e (which ia ce ain domains is connec ed o he nuclea
ma ix p o eins) o ms innume able loops which compose he s uc u e o he ch omosome.
16
Biological Func ions o Nucleic Acids
The deg ee o ch oma in condensa ion is la gely de e mined by he cell-cycle. The ch o-
ma in s uc u e, since i de e mines he accessibili y and eadabili y o genes, has a s ong
in luence on gene ac i i y. T ansc ip ional ac i e egions co ela e wi h an open ch oma in
s uc u e (euch oma in).
2.2.3 Genes
A gene can be unde s ood as a unc ional uni o he gene ic ma e ial, which con ains he
blue p in o a gene p oduc . A gene p oduc can be one o se e al p o eins (o subuni s
o p o eins) o a unc ional RNA, e.g. mic oRNA (miRNA), ibosomal RNA ( RNA), o
ans e RNA ( RNA).4
2.2.4 Gene Exp ession
Gene exp ession can be unde s ood as gene ac i i y. I desc ibes how much o a gene p od-
uc is p oduced om each pa icula gene. The gene ac i i y can be egula ed a di e en
s ages, e.g. a he ansc ip ion ini ia ion, o pos - ansc ip ional, a he mRNA le el.
The a ious cell ypes o a highe o ganism all con ain he same gene ic in o ma ion. How-
e e , he gene ac i i ies a e di e en , depending on he equi emen s o he pa icula cell
unc ion.
T ansc ip ion
T ansc ip ion equi es opening o double helix s uc u e i s . I is assumed ha he e-
duced duplex s abili y wi hin he P ibnow-box (comp ising he sequence mo i TATAAT)
suppo s he opening o he ansc ip ion bubble. The ansc ip ion bubble ex ends o e
abou 18 base pai s.
T ansc ip ion ini ia ionis ollowedby he elonga ion p ocess, in which an RNA-copy o he
sense-s and (only he sense-s and encodes he sequence in o ma ion o he gene p oduc )
is ansc ibed un il a e mina o sequence a he end o he gene is eached.
Du ing elonga ion, he holoenzyme slides along he ope on om 5’ o 3’ di ec ion (wi h
espec o sense s and - see Fig. 2.8). The co ec nucleo ides o he assembly o he
mRNA s and a e ecognized by complemen a y base pai ing wi h he coding s and. RNA
polyme ase joins hese nucleo ides wi h he g owing RNA s and. A p oo eading mecha-
nism eplaces inco ec ly added nucleo ides.
4These don’ se e as empla es o he syn hesis o polypep ide s ands bu a he cons i u e a
c ucial pa o he cells molecula machine y o , like miRNA, a e in ol ed in he egula ion o he
exp ession o o he genes.
17
Fundamen als
Figu e 2.8: T ansc ip ion (A) and ansla ion (B). RNA polyme ase opens a an-
sc ip ion bubble and p oduces a RNA copy o he sense s and while sliding in 5’ o 3’
di ec ion (wi h espec o he sense s and) un il ansc ip ion e mina ion is encoun-
e ed. A e poly-adenyla ion (no shown) he mRNA is eleased om he nucleus
h ough he nuclea po es. T ansla ion o he mRNA sequence in o a polypep ide
sequence (B) is pe o med in he cy oplasm. Ribosomes mo e along he mRNA in
5’ o 3’ di ec ion, he eby ansla ing he gene ic code in o a polypep ide sequence.
P o eins eme ge om olding o he polypep ide chains.
The ansc ip ion ends when a e mina o sequence is encoun e ed. Then he ansc ip-
ion complex comes apa , he ansc ip ion bubble collapses and he mRNA s and is e-
leased. S ill in he nucleus he (euca yo ic) mRNA unde goes poly-adenyla ion (addi ion
o a poly-A- ail a he 3’-end). By binding he poly(A)-binding p o ein (PABP) he poly-A-
ail p o ec s he mRNA om deg ada ion and inc eases he ansla ion o he mRNA. The
poly-A sequence is echnically employed o he speci ic ex ac ion o mRNA sequences
wi h poly-T unc ionalized magne ic beads.
T ansla ion
Messenge RNA (mRNA) is used as a empla e o he syn hesis o p o eins. Single
s anded RNAs simila like polypep ide chains can old and ha e he capabili y o o m
complex e ia y s uc u es, simila as p o eins. Ribosomal RNA ( RNA), he mos abun-
dan RNA in cells, is no a simple in o ma ion ca ie like DNA, bu a he olds i sel in o
a complex ”nanomachine” which is c ucial o he syn hesis o polypep ide chains.
Like iny obo s ibosomes slide along he mRNA s ands (downs eam om he 5’- o
18
Biological Func ions o Nucleic Acids
he 3’-end) and ansla e he nucleic acid sequences ia he gene ic code in o polypep ide
sequences (Fig. 2.8). The molecula ecogni ion o he codons (base iple s encoding o
amino acids) is pe o med wi h ans e RNA ( RNA), ano he unc ional RNA s uc u e
(Fig. 2.10). The an icodon, an exposed base iple a he end o he an icodon a m o
he RNA, can speci ically bind ia base pai ing5 o a complemen a y codon sequence on
he mRNA s and. Upon binding he co esponding amino acid which was ca ied by he
RNA o he si e o polypep ide syn hesis is a ached o he g owing polypep ide chain.
Subsequen ly he ibosome mo es on o he nex codon and simul aneously eleases he
discha ged RNA.
T ansla iono an mRNA s and is pe o med by many ibosomes simul aneously. While he
ansla ion p ocess is going on he mRNA s and is deg aded by nucleases in he 5’→3’ di-
ec ion.
2.2.5 Exp ession Regula ion
The unc ions o a cell (e.g. exp ession o s uc u al and egula o y p o eins, di e en ia-
ion, con ol o he li e cycle, adap ion o en i onmen al in luences) a e la gely con olled
by gene egula o y ne wo ks. T ansc ip ion ac o p o eins ( ia speci ic p o ein-DNA bind-
ing) can ac i a e, ampli y o inhibi he ansla ion o he a ge ed gene(s) and hus con ol
he co esponding gene ac i i y.
Pos - ansc ip ional egula o y mechanisms include al e na i e splicing, RNA silencing,
an isense supp ession, and he egula ion o mRNA s abili y.
DNA mic oa ays enable simul aneous in es iga ion o he ac i i y o many genes, on a
genome-wide scale. Gene exp ession p o iles (which a e encoding he complex in e ac-
ions be ween genes) a e an impo an ool o he in es iga ion gene egula o y pa hways
(→ unc ional genomics). Exp ession p o iling has also eme ged as a p omising diagnos ic
ool o iden i ica ion o cance ypes o sub ypes, hus enabling a well-di ec ed he apeu ic
esponse.
Exp ession egula ion a he ansc ip ion le el
The mos p ominen egula ion mechanism is ansc ip ion ini ia ion. In p oca yo es es-
sen ially only he holoenzyme RNA-polyme ase (composed o se e al subuni s) is di ec ly
in ol ed in he ansc ip ion p ocess. In euca yo es a la ge machine y o p o eins (includ-
ing se e al holoenzymes) needs o o m an ini ia ion complex be o e he ansc ip ion can
commence.
5F equen ly an icodons con ain he ela i ely unspeci ically binding nucleo ides inosine o pseudou i-
dine. Unspeci ic binding accoun s o he degene acy o he gene ic code.
19
Fundamen als
In he simple p oca yo ic o ganisms (e.g. bac e ia) he ini ia ion o ansc ip ion is egu-
la ed by ac i a o s and ep esso s. This shall be explained in he ollowing on he example
o he egula ion o he lac ose genes o he bac e ium E. Coli, which has been in es iga ed
by Jacob and Monod [Jac61].
E. Coli can diges bo h ood sou ces - glucose and lac ose. To conse e esou ces he lac-
ose me abolism is only ac i a ed i only lac ose and no glucose is a ailable. In case bo h
suga s a e a ailable E. Coli gi es p e e ence o glucose since i is he mo e e icien sou ce
o ene gy. Only i he glucose is deple ed and lac ose is s ill p esen in he medium, E. Coli
begins o exp ess he gene o he p o ein β-galac osidase, an enzyme which is equi ed o
he diges ion o lac ose.
The gene o β-galac osidase lacZ is combined wi h wo u he genes, lacY and lacA
(auxilia y genes, also equi ed o lac ose diges ion) in a unc ional uni called ope on
(Fig. 2.9A). The ope on is ypical o p oca yo ic o ganisms. Apa om he coding se-
quences o he p o ein(s) he ope on con ains he p omo e sequence. This is ecognized
by RNA-polyme ase and enables binding o he RNA-polyme ase o he double-s anded
DNA. The p omo e con ains he P ibnow-box wi h he sequence mo i TATAAT ( ypical
o p oca yo es), and he so-called ope a o . In he lac ope on he ope a o is a binding si e
o a ep esso p o ein. The ep esso p o ein, when bound o he ope a o si e, p e en s
RNA polyme ase om binding o he p omo e si e (Fig. 2.9D).
Ano he sequence mo i , adjacen o he p omo o , se es as speci ic binding si e o he
ac i a o p o ein CAP, which suppo s he binding o RNA-polyme ase o he p omo e si e
(Fig. 2.9C).
The unc ion o he egula o y p o eins (ac i a o and ep esso ) is con olled by he abun-
dance o glucose and lac ose, espec i ely. The ac i a o CAP (a ecep o o cyclic AMP)
can only bind o CAP binding si e (p o ein-DNA in e ac ion) upon binding o cyclic AMP,
which is abundan in he absence o glucose. The lac ep esso p o ein can only bind o he
ope a o si e i lac ose is no a ailable, since he binding a ini y o he ep esso p o ein o
DNA is signi ican ly dec eased by a con o ma ional change induced om he p esence o
allolac ose.
•Glucose and lac ose a ailable: In he p esence glucose he ac i a o canno bind o
he CAP si e. Since he ep esso can nei he bind, he exp ession can occu a a low
basal le el (Fig. 2.9B).
•Lac ose a ailable/glucose una ailable: The ac i a o (CAP) can only bind nea he
p omo e si e i glucose is no a ailable (Fig. 2.9C). In he p esence o lac ose only,
he ac i a o inc eases he lacZ exp ession by a ac o o abou 40 compa ed o he
basal le el [P a02].
20
Biological Func ions o Nucleic Acids
lacY lacA
T ansc ip ion
p omo e
CAP
si e ope a o
lacZ
p omo e
CAP
si e ope a o
lacZ
ep
p omo e
CAP
si e ope a o
lacZ
CAP RNA
Polyme ase
A
D
C
B
p omo e
CAP
si e ope a o
lacZ
RNA
Polyme ase
glucose
&
lac ose
a ailable
glucose
una ailable
lac ose
a ailable
lac ose
una ailable
basal exp ession
exp ession ac i a ed
exp ession inhibi ed
Figu e 2.9: The lac ope on (A) and he lac exp ession (B-D). The lac ope on
comp ises he CAP ac i a o si e, he p omo e and he genes lacZ,lacY and lacA.
The la e a e ansc ibed as a single mRNA. The exp ession le el o he lac genes
is con olled by he abundance o glucose and lac ose, espec i ely. Ac i a o and
ep esso p o eins which can bind o speci ic binding si es (p o ein-DNA in e ac ion),
con ol he binding RNA polyme ase. See ex o de ails. Figu es we e adap ed om
[P a02].
21
Fundamen als
•Lac ose una ailable: The lac ep esso can only bind o he ope a o si e i lac ose is
no a ailable. In his case he ep esso is bound o he ope a o si e p e en ing he
binding o he RNA-polyme ase, no ma e i he ac i a o is bound o he CAP si e
(Fig. 2.9D). The exp ession o he lac genes is inhibi ed.
2.2.6 Biological Func ions o RNA
Figu e 2.10: S uc u e o phenylalanine ans e RNA ( isualiza ion o 4TNA.pdb
[Hin78] wi h UCSF Chime a). T ans e -RNA is employed in he ansla ion p ocess
as a sequence speci ic ehicle o amino-acids. The an icodon-a m (nea he lowe
edge o he image) con ains a uni o 3 nucleo ides co esponding o a codon on he
mRNA s and. The amino-acid (no shown he e) is a ached o he accep o s em
(uppe igh end) wi h he cha ac e is ic CCA 3’- e minal g oup.
The biological unc ion o RNA is mo e e sa ile han ha o DNA:
•In he p ocess o gene exp ession messenge RNA (mRNA) is employed as a empla e
o polypep ide syn hesis. RNA, unlike DNA, is a ola ile in o ma ion ca ie wi h a
a he limi ed li e ime.
•Mic o-RNAs (miRNA) ha e egula o y unc ions. Via he RNA in e e ence (RNAi)
mechanism hey can speci ically inhibi he exp ession o he co esponding a ge
genes.
•An isense-RNAs (aRNA) ha e egula o y unc ions. An aRNA sequence is p oduced i
he noncoding (an isense)-s and o a gene sequence is also being ansc ibed. Thus he
22
Nucleic Acid Hyb idiza ion
aRNA is complemen a y o he mRNA o he pa icula gene. By base-pai ing be ween
he complemen a y RNA s ands he ansla ion o he co esponding polypep ide-
sequence is inhibi ed. In he ansgenic Fla Sa T M oma o an isense RNA is em-
ployed o supp ess he exp ession o an enzyme in ol ed in e hylene p oduc ion. The
signi ican educ ion o e hylene delays he ipening and spoiling o he oma o.
•RNA sequences, simila as polypep ide chains, can old in o complex seconda y and
e ia y s uc u es. Ribosomal RNA and ans e RNAs ( RNA) a e essen ial pa s o
he ansla ion machine y (see sec ion 2.2.4).
2.3 Nucleic Acid Hyb idiza ion
Two complemen a y (o pa ially complemen a y) nucleic acid s ands S1and S2can bind
ia base pai ing and o m a s able nucleic acid duplex D. The double-helical duplex s uc-
u e is s abilized by hyd ogen bonding and base s acking in e ac ions.
S1+ S2
hyb idiza ion
−−−−−−−*
)−−−−−−−
dissocia ion D (2.1)
The o ma ion o nucleic acid duplexes is commonly called hyb idiza ion since usually nu-
cleic acid s ands om di e en sou ces (e.g. DNA p obes and RNA a ge s) a e in ol ed.
Owing o he non-co alen cha ac e o he s abilizing in e ac ions nucleic acid hyb idiza-
ion is e e sible: In he modynamic equilib ium he duplex o ma ion is balanced by du-
plex dissocia ion (also called duplex dena u a ion o mel ing). Lowe empe a u es and
inc eased ionic s eng hs (up o 1 M [Na+]) a o duplex o ma ion. Wi h inc easing em-
pe a u e o educed ionic s eng h o he hyb idiza ion bu e solu ion he duplexes a e
inc easingly des abilized. Depending on he pa icula duplex sequence, nucleic acid du-
plexes can ha e a e y dis inc mel ing ansi ion. Owing o he coope a i e cha ac e o
he duplex binding, he ac ion o mel ed duplexes can change om close o 0% o 100%
wi hin a empe a u e ange o a ew Kel ins.6
Only a small ac ion o he duplexes is in a pa ially dena u ed in e media e s a e. The e-
o e, he hyb idiza ion/mel ing ansi ion is equen ly desc ibed as a wo-s a e ansi ion.
An impo an cha ac e is ic o he nucleic acid hyb idiza ion is i s ou s anding sequence
speci ici y. Al eady a single misma ched base wi hin an oligonucleo ide duplex can esul
in a signi ican ly educed bindinga ini y. Molecula ecogni ion by nucleic acid hyb idiza-
ion is employed by na u e (e.g. in RNA in e e ence) and by a ious molecula biology
applica ions:
6Fo oligonucleo ide duplexes he wid h o he mel ing ansi ion is dec easing wi h inc easing Gibbs
ee ene gy o he duplex, hus wi h inc easing duplex leng h.
23
Fundamen als
•DNA mic oa ays
•Fluo escen in si u hyb idiza ion(FISH): sequencespeci iclabelingo mRNA sequences
wi hin cells.
•P ime sequences a e used as s a ing poin s o nucleic acid eplica ion (e.g. in PCR
o dideoxy sequencing). Fo his pu pose he p ime s a e hyb idized o he empla e
s ands.
•Molecula beacon p obes: his ype o hai pin-shaped nucleic acid p obe con aining a
luo opho e-quenche -pai becomes luo escen upon hyb idiza ion wi h a complemen-
a y a ge sequence.
•An isense RNA sequences (sequence-speci ic silencing o mRNA ansc ip s)
•RNA in e e ence (sequence-speci ic silencing o mRNA ansc ip s)
2.3.1 Kine ics o Nucleic Acid Hyb idiza ion
The widely used wo-s a e model o nucleic acid hyb idiza ion assumes ha he single
s anded species S1and S2a e in equilib ium wi h he duplexes D.
S1+ S2
k+
−*
)−
k−
D (2.2)
Equa ion 2.2 doesn’ desc ibe elemen a y base pai ing p ocesses and is he e o e alid only
i he e a e no signi ican ly popula ed in e media e s a es. The wo-s a e model is a eason-
able app oxima ion, o example, o sho linea duplexes. The zippe model o DNA
duplex mel ing ansi ion, which conside s indi idual base pai ing and base pai dissocia-
ion e en s, is desc ibed in sec ion 2.3.3.
In he ollowing, o simplici y’s sake, we assume ha a e duplexes a e no sel -com-
plemen a y and ha olding o single s anded species (in as and base pai ing) can be
neglec ed.
Duplex o ma ion is a second o de eac ion, whe eas he dena u a ion is a i s o de eac-
ion. d[D]
d =−k−[D] + k+[S1][S2] (2.3)
In equilib ium (wi h d[D]/d =0) we ob ain he equilib ium cons an K(as desc ibed by
he law o mass ac ion).
K=k+
k−
=[D]
[S1][S2]=[D]
([S1]0−[D]) ·([S2]0−[D]) (2.4)
24
Nucleic Acid Hyb idiza ion
The Gibbs ee ene gy o duplex o ma ion ∆G◦
D(◦is e e ing o s anda d condi ions) is
ela ed o he equilib ium cons an Kby
∆G◦
D=−R·T·ln K.(2.5)
I he comple e empe a u e dependence o he binding a ini y - e.g. om expe imen ally
de e mined plo s o 1/Tm e sus ln (CT/4) - is known, he Gibbs ee ene gy ∆G◦
Dcan be
de e mined ia he an’ Ho equa ion:
1
Tm
=R
∆H◦
D
·lnCT
4+∆S◦
D
∆H◦
D
(2.6)
Tmis he mel ing empe a u e o he duplex - he empe a u e a which pe de ini ion (in
he modynamic equilib ium) 50% o he duplexes a e dissocia ed. CTis he o al concen-
a ion o nucleic acid s ands.
F om he o al en halpy ∆H◦
Dand en opy changes ∆S◦
D he Gibbs ee ene gy change
∆G◦
Do he mel ing ansi ion can be ob ained wi h
∆G◦
D= ∆H◦
D−T·∆S◦
D.(2.7)
Al e na i ely ∆H◦
Dand ∆S◦
Dcan be p edic ed om sequence-dependen nea es -neighbo
he modynamic pa ame e s (see sec ion 2.3.2).
F ac ion o hyb idized duplexes
The ac ion o hyb idized oligonucleo ides Fb [Koe05] ( ac ion bound) is a quan i y
which is di ec ly accessible om expe imen s (e.g. ia he hyb idiza ion signal in ensi y in
mic oa ay assays o ia he hypoch omici y in UV-abso p ion-based measu emen s). Fb
can be de i ed om he modynamic quan i ies (e.g. ia he equilib ium cons an K).
Fb =[D]
min([S1]0,[S2]0)(2.8)
[S1]0and [S2]0a e he ini ial concen a ions o single-s anded species S1and S2. How
Tm,∆G◦
Dand Fb a e ela ed and in luenced by expe imen al pa ame e s (duplex leng h,
sequence composi ion, de ec s, sal concen a ion, nucleic acid concen a ion and empe -
a u e) is well discussed in [Koe05].
I he ac ion bound F b is compa ed o mic oa ay hyb idiza ion signals one needs o
conside ha mic oa ay hyb idiza ion is a ec ed by many pa ame e s, which a e no ac-
coun ed o in he simple model desc ibed abo e.
25
Fundamen als
con o ma ions an open link can adop [Gib59].
In C. Ki els double-ended zippe model [Ki 69] he zippe is consis ing o N bonds (co -
esponding o he base pai s) ha can only be opened om he ends. The pa i ion unc ion
is de e mined by summa ion o e he s a is ical weigh s o all pa ially unzipped duplex
s a es. Wi h he pa i ion unc ion he s a is ical mechanics o he duplex, e.g. he a e -
age numbe o open bonds (co esponding o he deg ee o pa ial duplex dena u a ion) is
accessible. Ki el showed ha he assumed degene acy o pa ially unzipped duplex s a es
(a ising om o a ional eedom o unpai ed nucleo ides) - in DNA his degene acy may
be on he o de o 104- gi es ise o a mel ing ansi ion in he quasi-one-dimensional
sys em.7No phase ansi ion can occu in he non-degene a e case (when he numbe o
o a ional deg ees o eedom equals 1).
Zocchi e al. [Zoc03] epo ed ha a zippe -model based on end-domain opening desc ibes
well he empe a u e dependence o he a e age numbe o unzipped base pai s de e mined
in UV abso p ion expe imen s. Howe e , hey also epo ha hei analysis o ansi ion
pa ame e s indica es ha , apa om end-domain opening, bubble o ma ion is also impo -
an o he dena u a ion p ocess.
Deu sch e al. [Deu04] employed he double-ended zippe model o a s a is ical mechan-
ics based desc ip ion o mic oa ay hyb idiza ion signals.
End-unzipping o he duplexhas also been assumed by Ambj¨o nssonand Me zle [Amb05]
o a model o in es iga e he blinking dynamics o molecula beacons ( luo opho e-quen-
che pai included in a aying duplex sec ion).
Base pai s a he duplex ends a e s abilized by s acking in e ac ion wi h only one neigh-
bo ing base pai , whe eas base pai s in he in e io o he duplex a e s acked be ween wo
neighbo ing base pai s. The s abilizing s acking in e ac ions om bo h sides p e en in-
e nal dena u a ion. The e o e unzipping is (la gely) es ic ed o he duplex ends (end
aying) as shown in Fig. 2.14A. S uc u al cons ain s a ising om he double helix s uc-
u e may impose u he es ic ions o in e nal bubble o ma ion. The in luence o he
helical s uc u e on duplex s abili y is, howe e , no well unde s ood.
Dena u a ion bubbles
The abo e s a emen s, howe e , do no apply o he dena u a ion o long duplexes. These
dena u e ia he o ma ion o dena u a ionbubblesin he in e io o he duplex(see Fig. 2.14B
and C). This is due o se e al easons:
•due o an exponen ial dec ease o he base pai dissocia ion p obabili y owa ds he
7Cues a and Sanchez [Cue04] discuss why Van Ho e’s heo em (simply in e p e ed: ”No phase ansi-
ions occu in 1D pa icle sys ems wi h sho - ange pai in e ac ions”) doesn’ apply o he mel ing
ansi ion o nucleic acid duplexes.
32
Nucleic Acid Hyb idiza ion
cen e o he duplex, end-domain opening is es ic ed o duplex-ends ⇒ hus, long
duplexes can only dena u e ia he o ma ion o dena u a ion bubbles.
•occu ence o ela i ely weakly bound (AT- ich) subsequences in a long duplex
•inc eased mel ing empe a u es o long duplexes ⇒ he inc eased en opy con ibu ion
(−T∆S) esul s in des abiliza ion o he nea es neighbo in e ac ions ∆GNN
A
C
B
Figu e 2.14: Dena u a ion o sho duplexes (A) occu s mainly ia end-domain
opening. In long duplexes (B) end-domain opening does’ ex end in o he middle o
he duplex. Ra he , dena u a ion bubbles, o ming a weakly bound sec ions in he
in e io o he duplexes, p opaga e and (C) me ge wi h he open end- egions. A
inc eased empe a u es dena u a ion bubble o ma ion leads o dissocia ion o long
duplexes.
The ele ance o in e nal dena u a ion bubble o ma ion depends on duplex leng h and,
in pa icula , on he indi idual sequences (i.e. on he dis ibu ion o mo e/less s able
NN pai s). To p o ide a coa se es ima e: o duplexes wi h l < 15 base pai s end- aying
is expec ed o be he p e ailing mode o nucleic acid dena u a ion, ice e sa, o long
and in e media e size duplexes wi h l > 100 base pai s bubble o ma ion is expec ed o
be ele an o mo e impo an han end-domain opening [Blo03].8Howe e , Zocchi e al.
[Zoc03] epo ed ha dena u a ion bubbles may be ele an also in he dena u a ion p ocess
o sho duplexes.
8Blossey e al. [Blo03]: ”On a he sho DNA sequences (∼100 bp’s) he loop en opy con ibu ion
is no e y impo an as loops a e a e and sho and he DNA dena u es mainly h ough unbind-
ing om he edges. A desc ip ion based on he 1D Ising model wi h app op ia e expe imen ally
de e mined ene gy pa ame e s is he e o e su icien [...].”
33
Fundamen als
2.3.4 Fu he Models o he DNA Mel ing T ansi ion
Fu he well-es ablished models o he DNA mel ing ansi ion a e he Poland-Sche aga
(PS) model [Pol66] and he Pey a d-Bishop-Dauxois (PBD) model [Dau93].
The Poland-Sche aga model desc ibes he helix-coil ansi ion in long polynucleo ide du-
plexes. The duplex comp ises al e na ing double-helical segmen s and dena u a ion bub-
bles. The PS model is essen ially a one-dimensional Ising-model. Conside a ion o he
a ious bubble con igu a ions gi es ise o an en opic e m. This esul s in an e ec i e
long ange in e ac ion, so ha in he PS model a phase ansi ion may occu [Blo03].
The PBD model ep esen s a Hamil onian app oach. In he PBD model coope a i i y e -
ec s - a ising om anha monic nea es neighbo s acking in e ac ions - esul in a dis inc
mel ing ansi ion.
An o e iew on heo e ical models o he nucleic acid mel ing ansi ion is p o ided wi h
e e ence [Zho06].
Fu he eading on he DNA mel ing ansi ion:
• he mal dena u a ion o DNA [Wa 85] and DNA oligome s [Zoc03]
•DNA b ea hing dynamics [Amb06]
•zippe models [Ki 69; I a04]
•end-dena u a ion [Amb05]
•misma ches and bubbles [Zen06]
• he modynamic p ope ies o DNA sequences [Koe05]
• u he ela ed publica ions [ E06; E e07]
2.4 Des abiliza ion o Oligonucleo ide Duplexes by
Poin De ec s
A high disc imina ion capabili y be ween simila sequences is impo an in geno yping ap-
plica ions, whe e single nucleo ide polymo phisms(SNPs), a ia ions o single bases, a e
he subjec o in e es . SNPs la gely de e mine gene ic indi iduali y, bu also disposi ion o
gene ically caused diseases o esponse o medicamen s, and a e he e o e o g ea in e es
no only o gene ic esea ch bu also o medical diagnos ics and he apy. SNPs can be
de ec ed (using DNA mic oa ays) by hyb idiza ion wi h sho oligonucleo ide p obes. Al-
eady a single misma ching (MM) base pai (owing o he SNP) can esul in a signi ican
dec ease o duplex s abili y [Nel81; Pa 82; Con83]. The impac o a MM base pai on
34
Des abiliza ion o Oligonucleo ide Duplexes by Poin De ec s
duplex binding a ini y is is de e mined by he leng h o he duplex [Koe05], he ype o
misma ch base pai [All97], he in luence o neighbo ing bases [All97] and by he posi ion
o he de ec (wi h espec o he duplex ends) [Wic06; Poz06; Nai06b].
In his s udy we also in es iga e single base bulges, ano he ype o poin de ec , o igina -
ing om single base inse ions and dele ions. The inse ion o a su plus (unpai ed) base
in o one o he duplex s ands esul s in a small bulge in he egula duplex s uc u e. Sim-
ila ly a single base dele ion c ea es a bulged base in he opposi e s and. Like single base
misma ches base bulges can signi ican ly educe duplex binding a ini y.
2.4.1 Single Base Misma ches
Figu e 2.15: S uc u e o T·G misma ches in a B-DNA duplex (X- ay di ac ion
da a 113D.pdb [Hun87]). G een a ows indica e he T·G misma ches.
S uc u al in es iga ions (NMR and X- ay s udies) ha e shown ha single misma ch base
pai s (see Fig. 2.15) misma ches in oduce li le o e all s uc u al dis o ion on he double
helical duplex s uc u e [Hol91; Cog91; Ske93].
Conside a ion o single base misma ches in he nea es neighbo model
The nea es neighbo model has been ex ended beyond Wa son-C ick base pai s o include
single base misma ch (MM) de ec s [All97; San04]. F om UV mel ing expe imen s Allawi
e al. [All97] ha e es ablished a comple e da abase o MM single base MM he mody-
namic pa ame e s o DNA/DNA duplexes. The (mos ly) des abilizing MM p opaga ion
35
Fundamen als
pa ame e s (a comple e able is p o ided in [San04]) a e used o duplex ee ene gy calcu-
la ions jus like he Wa son-C ick p opaga ion pa ame e s. Mos des abilizing MM nea es -
neighbo pai s a e AC/TC (∆G◦
37=1.33 kcal/mol), TC/AA (∆G◦
37=1.33 kcal/mol), TC/AC
(∆G◦
37=1.05 kcal/mol) and GT/CC (∆G◦
37=0.98 kcal/mol). Leas des abilizing a e GG/CG
(∆G◦
37=-1.11 kcal/mol) and GT/CG (∆G◦
37=-0.59 kcal/mol). An o de o DNA/DNA base
pai s abili ies (based on [All97]) is p o ided in [San04]:
G·C>A·T>G·G>G·T≥G·A>T·T≥A·A>T·C≥A·C≥C·C
The s udy o Allawi e al. [All97] also e eals a s ong impac o closing base pai s ( he
base pai s enclosing he MM base pai ) - closing C·G base pai s a e mo e s abilizing han
A·T base pai s.
The wo-s a e nea es neighbo model doesn’ accoun o he MM posi ion wi hin he
duplex sequence. Acco ding o San aLucia [San04] ”[...] wi h he excep ion o he e minal
and penul ima e posi ions, he he modynamics o a gi en misma ch in a gi en con ex is
independen o i s posi ion in a duplex, con a y o common opinion”. This, howe e , is no
in ag eemen wi h ecen obse a ions o a s ong in luence o de ec posi ion on duplex
binding a ini y [Kie99; Do 03; Wic06; Poz06; Nai06b].
2.4.2 Single Base Bulges
De ec s o igina ing om inse ion o dele ion o a base esul in bulged duplexes as shown
in Fig. 2.16. Base bulges a e a equen s uc u al mo i in RNA s uc u es e.g. in RNA and
RNA. I is assumed ha bulges may play a ole in nucleic acid-p o ein binding [Wu87].
Single bulged bases can adop looped ou (Fig. 2.16) o in ahelically s acked con o ma-
ions [Yoo01; Ba 06]. Acco ding o Woodson and C o he s [Woo88] ”[...] e idence om
se e al labo a o ies sugges s ha ex ahelical pu ines a e gene ally s acked in o he helix,
while ex ahelical py imidines a e in equilib ium be ween s acked and uns acked s a es
[...]” (in his con ex ”ex ahelical base” has he meaning ”bulged base”).
The he modynamics o bulged duplex was i s in es iga ed by Fink and C o he s [Fin72].
They epo ed a des abilizing ee ene gy (25◦C) o 2.8 kcal/mol o a single base bulge.
Wa ell and cowo ke s [Ke93; Ke95; Zhu99] in es iga ed he he modynamics o single
base bulges on a la ge numbe o DNA and RNA sequence mo i s. The ela i e s abil-
i y o bulged RNA duplexes was in es iga ed in empe a u e g adien gel elec opho esis
(TGGE) expe imen s. Fo RNA bulges hey epo an un a o able ee ene gy (wi h espec
o he bulge- ee e e ence duplex) δ∆G◦
37 be ween 2.85 and 4.8 kcal/mol.
Wa ell and cowo ke s obse ed, ha he s abili y o bulged duplexes is inc eased i he
36
Des abiliza ion o Oligonucleo ide Duplexes by Poin De ec s
Figu e 2.16: Single base bulge (looped ou cy osine base, shown in yellow) in a
RNA helix s uc u e (X- ay di ac ion da a 1DQF.pdb [Sun00]).
bulged base has a leas one iden ical neighbo ing base. They ca ego ized bulged duplexes-
depending on he iden i y o he bulged base and he duplex sequence - in wo g oups:
•G oup I: he bulged base has no iden ical neighbo ing bases
•G oup II: he bulged base has a leas one iden ical neighbo ing base
Acco ding o [Zhu99] he local a e age ee ene gy con ibu ion o a DNA base bulge can
be exp essed as:
∆G◦
37,(XNZ)·(X0−Z0)= 2.72 kcal/mol + 0.48∆G◦
37,(XZ)·(X0Z0)+δg (2.18)
Fo he ee ene gy o RNA single bulges a simila ela ion was de i ed [Zhu99]:
∆G◦
37,(XNZ)·(X0−Z0)= 3.11 kcal/mol + 0.40∆G◦
37,(XZ)·(X0Z0)+δg (2.19)
No a ion: The unpai ed base N is enclosed by he base pai s X·X’ and Z·Z’.
∆G◦
37,(XZ)·(X0Z0)is he s acking ene gy o he base pai double (XZ)·(X0Z0).
The s abilizing con ibu ion o degene a e G oup II bulges δg is -0.4 kcal/mol
o DNA and -0.3 kcal/mol o RNA (in bo h cases δg=0 kcal/mol o G oup I
bulges).
37
Fundamen als
AGGCGTACGTA GTTTCCAGAG
TCCGCATGCAT CAAAGGTCTC
C
AGGCGTACGTA GTTTCCAGAG
TCCGCATGCAT CAAAGGTCTC
AGGCGTACGT AGTTTCCAGAG
TCCGCATGCA TCAAAGGTCTC
A
A
A
B
G oup I base bulge
G oup II base bulge
degene a e con o ma ion
Figu e 2.17: Posi ional degene acy o base bulges. (A) G oup I bulge. The non-
degene a e bulged base (C, shown in g ey) has no iden ical neighbo bases. (B)
G oup II bulge. The bulged base A has an iden ical neighbo , gi ing ise o posi ional
degene acy [Ke95] o he bulge con o ma ion. The inc eased numbe o possible bulge
con o ma ions (he e wo a he han only one in A) ep esen s an inc ease in en opy,
esul ing in a s abiliza ion o he degene a e Goup II bulge wi h espec o he non-
degene a e G oup I bulge.
The expe imen allyobse ed ee ene gy di e encebe ween G oup Iand degene a eG oup
II bulges o -0.4 and -0.3 kcal/mol ( o DNA and RNA, espec i ely) is in good ag eemen
wi h he simpli ied en opic es ima e o a wo-posi ion degene acy o -RT·ln(2)=-0.43
kcal/mol (a 37◦C) [Zhu99].
Znosko e al. [Zno02] epo an inc eased s abili y o py imidine single bulges wi h espec
o pu ine single bulges (0.4 kcal/mol on a e age). This s udy, based op ical mel ing expe -
imen s (UV abso p ion) on RNA duplexes, p o ided di e en equa ions (w i en he e in
he no a ion o [Zhu99]) o he bulge ee ene gies o py imidines (eqn. 2.20) and pu ines
(eqn. 2.21).
∆G◦
37,(XNZ)·(X0−Z0)= 3.9 kcal/mol + 0.10∆G◦
37,(XZ)·(X0Z0)+δg.(2.20)
∆G◦
37,(XNZ)·(X0−Z0)= 3.3 kcal/mol −0.3∆G◦
37,(XZ)·(X0Z0)+δg (2.21)
He e, δg is 0 and -0.8 kcal/mol o G oup I and G oup II bulges, espec i ely. The epo ed
s abiliza ion o G oup II bulges δg=-0.8 kcal/mol is signi ican ly la ge han he p e iously
epo ed s abiliza ion om [Zhu99] (δg=-0.3 o -0.4 kcal/mol), hus aises ques ions abou
he mechanisms unde lying G oup II bulge s abiliza ion.
Tu ne [Tu 92] sugges ed ha he s abili y o a bulged duplex could depend on he p ox-
imi y o he bulge wi h espec o he helix end. Znosko e al. [Zno02] didn’ ind e idence
o an in luence o bulge posi ion on duplex s abili y.
38
Des abiliza ion o Oligonucleo ide Duplexes by Poin De ec s
2.4.3 In luence o he De ec Posi ion
Kie zek e al. [Kie99] in es iga ed he e ec o he posi ion o a single misma ch wi hin
sho RNA duplexes (op ical mel ing expe imen s). They obse ed ha ”[...] mo ing he
posi ion o he misma ch owa d he end o he helix enhances he s abili y o U·U and
A·A misma ches by ∼0.5 kcal/mol pe each posi ion close o he helix end [...]”. Fo A·A
misma ches he obse ed end is less ob ious han o U·U misma ches, G·G misma ches
we e ound o be insensi i e o he posi ion wi hin he helix. Since he s udy was pe o med
wi h hep ame duplexes (enabling he compa ison o h ee MM posi ions) he da a base o
he obse ed MM posi ional in luence is a he limi ed.
Do is e al. [Do 03] obse ed a simila posi ional in luence o 2-base and 3-base mis-
ma ch p obes (wi h espec o cRNA a ge s) on CodeLinkTM 3D gel a ays. They also e-
po a s ong co ela ion (including he posi ional in luence) be ween solu ion-phase mel -
ing empe a u es and mic oa ay hyb idiza ion signals o he MM duplexes.
Recen mic oa ay s udies [Wic06; Poz06; Nai06b; Nai06a] using ex ensi e se s o p obe
sequences ha e shown a e y dis inc in luence o misma ch posi ion and bulge posi ion
[Nai06a], espec i ely. The disc imina ion be ween MM and PM is signi ican ly mo e dis-
inc o de ec s nea he cen e o he duplex han o de ec s nea he duplex ends.
In e es ingly, om solu ion phase hyb idiza ion s udies (apa om [Kie99] and [Do 03])
an in luence o de ec posi ion is no been epo ed. In he nea es neighbo model only
e minal and penul ima e MM posi ions a e conside ed o be less des abilizing han MMs
in he in e io o he duplex [Pey99; San04]. I is no clea whe he he posi ional in luence
has been o e looked in p e ious solu ion-based s udies o , i di e en expe imen al condi-
ions a e he eason, why a dis inc posi ional in luence has only been desc ibed ecen ly,
ypically o mic oa ay-based expe imen s.
Typical cha ac e is ics o s udies no epo ing an in luence o de ec po-
si ion [Ke95; All97; Pey99; Sug00]:
•mos ly solu ion-phase hyb idiza ion
•p esyn hesized oligonucleo ide p obes ( hus con aining a negligible ac ion o syn he-
sis de ec s)
•small p obe se s (<100 p obes) in es iga ed
• he de ec is ypically es ic ed o one o ew posi ions (commonly in he cen e o he
duplex), no sys ema ical a ia ion o he de ec posi ion
•in mos s udies a he sho duplexes ≤10 bp (li le ma gin o a ia ion o de ec
posi ion) we e employed
39
Fundamen als
•expe imen al me hod: measu emen o he mel ing cu es by UV abso bance spec-
oscopy ( o an assumed wo-s a e mel ing ansi ion he measu ed ac ion o dissoci-
a ed base pai s is equal o he ac ion o dissocia ed duplexes)
•duplex ee ene gies a e de i ed om mel ing cu e analysis
Acco ding o [Pey99] binding a ini y con ibu ions o misma ches mo e han h ee base
pai s om he end a e independen o he posi ion.9[Pey99]: ”Consequen ly, i can be
concluded ha he nea es -neighbo model is a good app oxima ion o bo h Wa son-C ick
pai s and all single misma ches.”
Typical cha ac e is ics o s udies epo ing an in luence o de ec posi ion
[U a02; Do 03; Wic06; Poz06; Nai06b]:
•mos ly mic oa ays s udies
•mic oa ays in [Wic06; Poz06; Nai06b] a e ab ica ed by in si u syn hesis - p obes can
he e o e con ain a conside able amoun o syn hesis de ec s
•duplex leng h be ween 16 and 25 bp
•expe imen al me hod ( ypically): measu emen o mic oa ay hyb idiza ion signals
(mos ly luo escence in ensi y)
•measu emen o binding a ini y a ia ions depending on de ec ype, de ec posi ion
and closing base pai s. The PM/MM hyb idiza ion signal a io is a di ec measu e o
he MM disc imina ion.
•mic oa ay s udies a e a o able o la ge scale sys ema ic in es iga ions o MM dis-
c imina ion (imp o ed s a is ics - many di e en sequences, ”di ec compa ison” o
binding a ini ies ob ained in he same expe imen )
The posi ional in luence appea s o be mos p onounced in [Wic06; Poz06; Nai06b]. How-
e e , his may be owing o he ac ha he expe imen al design o hese pa icula s udies
enables a mo e sys ema ic and ex ensi e in es iga ion o he posi ion dependence han he
o he s udies.
Expe imen al esul s in [Kie99] and [Do 03] indica e ha a posi ional in luence is no lim-
i ed o mic oa ay s udies bu can be obse ed in solu ion-phase hyb idiza ion s udies as
well.
9This pa icula s udy was pe o med wi h a ela i e small se o 51 ela i ely sho 9-12me duplexes.
40
Solid-Phase Syn hesis o Nucleic Acids
Modeling o he posi ional in luence
Pozhi ko [Poz02] conside ed misma ch posi ional in luence empi ically in an algo i hm
o inding speci ic oligonucleo ide p obes o species iden i ica ion.
Binde [Bin06] ies o explain he posi ional in luence wi h a zippe model in which he
misma ch a ec s he base pai ing o Wa son-C ick base pai s in he duplex sec ion be ween
he MM and he duplex end. The e o e, he impac o a misma ch on duplex s abili y is
ge ing smalle as i s posi ion is close o he duplex end.
Howe e , he assumed base pai opening p obabili y (as shown in Fig. 10 in [Bin06])
doesn’ accoun o he ac ha end aying unde hyb idiza ion condi ions is la gely con-
ined o he wo [And06] o h ee [Lei92] ou e mos base pai s.
Like Binde we use a zippe based model in ou analysis, howe e , we accoun o he ac
ha he end aying is la gely es ic ed o he ou e mos base pai s and ha he base pai
opening p obabili y is exponen ially dec easing owa ds he cen e o he duplex. Pa ial
dena u a ion o inne base pai s is conside ed as a a e s ochas ic e en .
2.5 Solid-Phase Syn hesis o Nucleic Acids
In molecula biosciences syn he ic nucleic acid sequences a e employed in many o appli-
ca ions. Fo example, as p ime s o he ampli ica ion o DNA sequences by polyme ase
chain eac ion (PCR), as a ge -speci ic p obe molecules on DNA mic oa ays (o in luo-
escen in si u hyb idiza ion wi hin biological specimens), o as double-s anded RNAs o
gene silencing in RNA in e e ence applica ions.
Syn he ic nucleic acid sequences a e commonly p oduced in a solid-phase syn hesis ap-
p oach.
2.5.1 P inciples o Solid-Phase Chemical Syn hesis
Solid-phase syn hesishas i s been employed o he ab ica ion o polypep idesequences10
[Me 63]. In solid-phase syn hesis he polyme -chains o be syn hesized a e end- e he ed
o a solid subs a e. This enables e icien sepa a ion o uncoupled building blocks (in
solu ion) om he su ace- e he ed syn hesis p oduc s, a e a syn hesis s ep has been com-
ple ed.
Coupling o monome building blocks (see Fig. 2.18) is pe o med ia eac i e e minal
g oups. A emo able chemical p o ec ion g oup p e en s uncon olled polyme iza ion o
10 Fo he de elopmen o he solid-phase polypep ide syn hesis R.B. Me i ield ecei ed he Nobel
P ize in chemis y in 1984.
41
Fundamen als
mRNA
ed luo escen
cDNA a ge s
mRNA
mRNA
isola ion
mic oa ay
hyb idiza ion
e e se
ansc ip ion
labelling
cance cells no malcells
combine
a ge s
g een luo escen
cDNA a ge s
Figu e 2.23: Dual colo mic oa ay expe imen . In his example he exp ession
p o ile o cance cells is compa ed o a e e ence sample o no mal cells. Complex
mix u es o messenge RNAs (mRNAs) a e isola ed om each sample and luo es-
cen ly labeled ia e e se ansc ip ion labeling. The a ge s om he cance cell
a e labeled wi h a g een luo escen dye, whe eas he a ge s om he e e ence
sample a e labeled wi h a ed luo escen dye. The a ge s a e combined and hy-
b idized on he same mic oa ay. Analysis and compa ison o he wo colo -channels
enables iden i ica ion o up- and down- egula ed genes. (Adap ed om Wikipedia:
h p://en.wikipedia.o g/wiki/DNA mic oa ay)
Figu e 2.24: Single Nucleo ide Polymo phisms (SNPs) a e gene ic a ia ions
o single base pai s be ween membe s o he same species, o e en be ween
he wo copies o a ch omosome pai . The DNA s and in 1 di e s om
he DNA s and in 2 by a single base pai . Geno yping assays enable high-
h oughpu sc eening o single nucleo ide polymo phisms. (Sou ce: Wikipedia,
h p://en.wikipedia.o g/wiki/Single nucleo ide polymo phism)
48
DNA Mic oa ays
2.6.2 The De elopmen o DNA Mic oa ay Technologies
An ea ly me hod (1975) o he analysis o complex nucleic acid mix u es is he Sou he n
blo [Sou75]. The eby he mix u e o uniden i ied DNA agmen s ( a ge s) is sepa a ed
by gel elec opho esis, ans e ed and immobilized on a lexible nylon memb ane. Fo
iden i ica ion o he a ge s adioac i ely o chemically labeled p obes (wi h well-known
sequences) a e incuba ed wi h he memb ane, hus enabling hyb idiza ion wi h he com-
plemen a y a ge sequences.
The so-called do blo is a simila echnique, in which he (unsepa a ed) a ge sample is
di ec ly applied on o he memb ane as ”do s”. A e ixa ion he iden i ica ion o he a ge
sequences is pe o med by hyb idiza ion wi h a labeled p obe sequence (o a mix u e o
labeled p obes).
Minia u iza ion and pa alleliza ion ha e e ol ed he do blo in o he high h oughpu
mac oa ay echnique. Wi h he help o au oma ed me hods se e al housand millime e -
sized nucleic acid spo s can be immobilized on a nylon memb ane ( ypically 10 o 20 cm
in size). He e, di e en om he blo ing echniques desc ibed abo e, he known p obe se-
quences (e.g. cDNA o syn he ic oligonucleo ide p obes) a e immobilized on he solid sub-
s a e, whe eas he a ge s a e applied in hyb idiza ion solu ion. Au o adiog aphic analysis
and he la ge quan i y o p obe ma e ial p o ide a high sensi i i y. Howe e , adioac i e
labeling wi h 32Po 33P( equi ing p ecau ious handling) and he need o la ge quan i ies
o p obe and a ge ma e ial a e se ious disad an ages o he mac oa ay echnique.
By using igid subs a es a he han lexible nylon memb anes, a signi ican minia u iza-
ion was achie ed, gi ing ise o DNA mic oa ay echnology. Mic oa ays a e commonly
p oduced on chemically unc ionalized glass subs a es - equen ly a mic oscope slide o -
ma is employed. The use o glass subs a es, which, unlike nylon-memb anes, ha e low
au o- luo escence, enables highly sensi i e de ec ion o luo escen ly labeled a ge s.
Di e en ypes o DNA mic oa ays ha e been de eloped in se e alindependen app oaches:
•In 1995 Schena e al. [Sch95] epo ed he i s gene exp ession assay on a p in ed
mic oa ay. They employed a con ac p in ing echnique o deposi ion o iny spo s
(abou 0.1-0.2 mm in diam.) o nucleic acids p obes (cDNA p obes) on a chemi-
cally unc ionalized glass subs a e. This now widely-used echnique is also known
as spo ing. The spo ing solu ion wi h he p e ab ica ed nucleic acid p obes is de-
posi ed on he su ace by a pin. A capilla y gap a he ip o he pin eleases a small
(and ep oducible) amoun o he spo ing solu ion when he pin is ouching he sub-
s a e su ace. Chemical unc ionaliza ion o he subs a e (e.g. wi h amino-, epoxy-
o aldehyde-g oups) and he p obe molecules (e.g. by a achmen o an amino g oup)
49
Fundamen als
enable ixa ion (immobiliza ion) o he p obes. cDNA mic oa ays a e mainly used in
gene exp ession assays. Apa om cDNA and PCR p oduc s, p esyn hesized oligonu-
cleo ide p obes can be immobilized on mic oa ays (→oligonucleo ide mic oa ay).
Mic oa ay obo s (a aye s) a e commonly employed o a ully au oma ed ab ica ion
p ocess.
•Al eady se e al yea s ea lie Fodo e al. [Fod91] de eloped a pho oli hog aphically
con olledcombina o ialchemis yapp oach o he ab ica ion o high-densi yoligonu-
cleo ide mic oa ays.11. Owing o he simila i y o he pho oli hog aphic ab ica ion
p ocess wi h semiconduc o ab ica ion echniques, hese mic oa ays a e commonly
called DNA chips. Unlike he spo ing app oach he ligh -di ec ed in si u syn hesis ap-
p oach doesn’ equi e p e ab ica ed p obes o deposi ion. The p obe molecules (DNA
oligonucleo ides) a e ab ica ed in si u, i.e. nucleo ide by nucleo ide, on he mic oa -
ay subs a e. The massi ely pa allel syn hesis o up o a million di e en sequences on
he same chip is di ec ed by UV ligh exposu e. In he combina o ial syn hesis p ocess
ch ome masks p o ide a sequence speci ic exposu e scheme (spa ially es ic ed o he
pa icula mic oa ay ea u es) o con ol he sequence o nucleo ide couplings o each
p obe sequence indi idually.
•Ink-je echniques (based on piezoelec ic deposi ion) a e used o in si u syn hesis o
mic oa ays [Bla96] (by deposi ion o phospho amidi es) and also o spo ing o p e-
syn hesized DNA [Sch98].
•A a he no el echnique is he elec ochemical in si u syn hesis o DNA mic oa ays
[Mau06]. The eby nucleic acid coupling is con olled by acid gene a ion on a CMOS
add essable elec ode a ay.
Depending on he ype o p obes employed DNA mic oa ays (no o be con used wi h
o he ypes o mic oa ays, e.g. p o ein mic oa ays) can be ca ego ized in o wo g oups:
cDNA mic oa ays
This ype o mic oa ay comp ises immobilized cDNA p obes o PCR p oduc s. Ow-
ing o he a ailabili y o cDNA and PCR p oduc s om biological sou ces, cDNA
a ays (mic oa ays and mac oa ays) a e equen ly p epa ed by biological labs.
Since he long p obe sequences ( ypically one hund ed o se e al hund ed n long)
a e no sui able o disc imina ion be ween simila sequences (e.g. o he iden i ica-
ion o single base MMs) he applica ion o cDNA mic oa ays is es ic ed o gene
exp ession p o iling.
11 ”High densi y” e e s o a high densi y o mic oa ay ea u es (up o one million pe cm−2)
50
DNA Mic oa ays
Oligonucleo ide mic oa ays
Oligonucleo idemic oa ays comp ise syn he ically ab ica ed p obesequences which
a e ypically be ween 15 and 100 bases long. Unlike cDNA mic oa ays, oligonu-
cleo ide mic oa ays enable disc imina ion o e y simila genes belonging o he
same gene amily. Sho oligonucleo ides (<30 n ) owing o hei high disc imina-
ion capabili y a e used o geno yping and esequencing applica ions. Long oligonu-
cleo ide p obes (∼60 n ) ha e he ad an age o p o iding a high sensi i i y o he
de ec ion o low abundance ansc ip s. Oligonucleo ide mic oa ays a e ab ica ed
by immobiliza ion (spo ing) o p esyn hesized oligonucleo ides, o by in si u syn-
hesis.
A de ailed o e iew o mic oa ay ypes and ab ica ion me hods is p o ided in [Gao04].
2.6.3 Cha ac e is ics o Mic oa ay Hyb idiza ion
Li e a u e epo s a la ge disc epancy be ween hyb idiza ion cha ac e is ics in bulk solu-
ion and on he mic oa ay su ace [Hel03; Bin06; Poz06]. Since NN he modynamic
pa ame e s we e de e mined in solu ion-phase expe imen s - unde ”ideal hyb idiza ion
condi ions”, he nea es -neighbo model doesn’ necessa ily pe o m sa is ac o y o he
p edic ion mic oa ay binding a ini ies.
Acco ding o Bhano e al. [Bha03], he loss o ansla ional ene gy and en opy o mic o-
a ay-bound p obes (wi h espec o hyb idiza ion o ee s ands in bulk solu ion), and he
cons ain ha a ge s can app oach he p obes only om one hal -space, is independen
o he sequence. Thus, wi h espec o bulk-solu ion, hyb idiza ion equilib ium cons an s,
equilib ium cons an s o mic oa ay hyb idiza ion a e mul iplied by he same sequence-
independen ac o . The di e ence be ween solu ion-phase and su ace-phase hyb idiza-
ion is o li le consequence o speci ici y and sensi i i y when equilib ium is achie ed.
Howe e , hyb idiza ion kine ics (which is di e en o su ace- and solu ion-phase hy-
b idiza ion) has a p onounced e ec on speci ici y and sensi i i y [Bha03].
Le icky and Ho gan [Le 05] e iewed physicochemical aspec s o DNA mic oa ay hy-
b idiza ion. In pa icula hey discussed di e ences wi h espec o solu ion-phase hy-
b idiza ion.
On DNA mic oa ays (wi h espec o solu ion hyb idiza ion)mel ing empe a u es [Hel03]
a e signi ican ly educed. Addi ionally, signi ican ly b oadened hyb idiza ion iso he ms
(de ia ing om Langmui - ype cha ac e is ics) [Bin06] a e obse ed. Mo eo e , on DNA
mic oa ays a s ong in luence o he posi ion o single base MMs on duplex binding a ini-
ies [Wic06; Poz06; Nai06b] is obse ed.
51
Fundamen als
An ”ideal mic oa ay” (in e ms o speci ici y and a ge quan iza ion)
would equi e he ollowing cha ac e is ics:
• a ge -speci ic hyb idiza ion: i.e. p obes hyb idize only wi h he complemen a y a ge
species
• he e is no in a-s and base pai ing leading o o ma ion o p obe o a ge seconda y
s uc u es.
•all p obe- a ge pai s ha e app oxima ely he same binding a ini y
• he e is a simple (e.g. linea ) ela ion be ween he hyb idiza ion signal and he concen-
a ion o he co esponding a ge sequence
Real mic oa ays de ia e om he ”ideal mic oa ay” (abo e) in se e al
aspec s:
•complex a ge mix u es gi e ise o compe i i e hyb idiza ion p ocesses [Poz06] (un-
speci ic a ge / a ge and p obe/ a ge c oss hyb idiza ion)
• he use o long ela i elylong a ge sequences ( ypically be ween 100 and se e al hun-
d ed n long) esul s in a ge seconda y s uc u e o ma ion and inc eased po en ial o
c oss hyb idiza ion. Bo h p ocesses compe e wi h he speci ic p obe- a ge hyb idiza-
ion. Ta ge seconda y s uc u e can p e en p obe- a ge hyb idiza ion, hus leading o
alse nega i es. Unspeci ic c oss-hyb idiza ion can esul in alse posi i es.
•su ace e ec s (e.g. elec os a ic e ec s and s e ical hind ance) can (in conjunc ion
wi h he a ying leng h/seconda y s uc u e o indi idual a ge s) a ec he quan i a-
i eness o he measu emen (binding a ini y is a unc ion o he amoun o hyb idized
a ge s, a ge leng h and a ge s uc u e)
•p obes a e con ined o a small a ea on he mic oa ay (⇒di usion-limi a ion e ec s)
•syn hesis de ec s (o igina ing om in si u syn hesis) a ec binding a ini ies
•labeling o he a ge sequences (e.g. wi h la ge luo escen dye molecules like Cy3
a ached a andom posi ions) may a ec binding a ini ies
Mic oa ay hyb idiza ion - a di usion d i en p ocess
Mic oa ays a e o en ab ica ed on mic oscope slides wi h dimensions o abou 75 mm×
25 mm. The hyb idiza ion solu ion ( en o se e al hund ed µl) is inse ed in o he gap
be ween he mic oa ay and a co e glass, hus o ming a hin ilm wi h a hickness o 20-
100 µm. This is be e illus a ed by he ollowing compa ison in which he mic oa ay is
assumed o be enla ged o he size o a oo ball ield. On his scale, he liquid ilm co e-
sponds o a puddle be ween 2 and 10 cm deep. The size o a mic oa ay ea u es may be
isualized by a socce ball.
Hyb idiza ion in such a con igu a ion is a slow p ocess since di usion is he domina ing
anspo mechanism o he a ge s. The hyb idiza ion o a a ge wi h he co espond-
ing p obe is usually limi ed by he slow di usion p ocess [Pap06]. Wi h he Eins ein-
52
DNA Mic oa ays
Schmoluchowki ela ion we ind ha he a e age dis ance a a ge molecule (wi h a molec-
ula di usion coe icien o 10−11 m2/s [Pap06]) is a eling in an o e nigh hyb idiza ion
is app oxima ely 1 mm. In mic oa ay assays, by di usi e anspo alone, he equilib ium
can’ be eached on a easonable ime scale.
No el chao ic mic omixing echniques, e.g. based on su ace acous ic wa es (SAW) [Toe03],
can e y e icien ly gene a e mic oagi a ion in he capilla y gap und hus o e come he di -
usion limi a ion.
By scaling down he dimensions o he mic oa ay (⇒inc eased a io be ween he di u-
sion coe icien and he mic oa ay su ace) he hyb idiza ion equilib ium can be eached
on a ealis ic ime scale [Dan07].
2.6.4 Fu he Reading on he Technical and Physical P in-
ciples o DNA Mic oa ays
•Sensi i i y, speci ici y, c oss hyb idiza ion (→de ec ion o alse posi i es) [Bha03;
Bin06]
•Poin de ec s (misma ches [Dod77; Wal79; Nel81; All97], base bulges [Ke95; Zhu99;
Zno02]), in luence o de ec posi ion [Do 03; Wic06; Poz06], disc imina ion capabili y
[U a03; Lee04]
•Seconda y s uc u e o p obes and a ge s (→de ec ion o alse nega i es) [Lue03;
San04]
•Mic oa ay ab ica ion (immobiliza ion, in si u syn hesis) [Sch99; Sch02; Gao04]
•Quali yo he p obesequences - syn hesisde ec s [Ga 02; Job02; Ric04; Bin06] →he -
e ogenei y o binding a ini ies
•Ta ge p epa a ion [Sch02] ( a ge leng h, luo escen labeling, composi ion o he a -
ge mix u e, ype o nucleic acid a ge - DNA o RNA)
•Compe i i e e ec s [Bin06]
•Su ace densi y o he p obes [Pe 02; Wa 00; Le 05] (s e ic hind ance [Hal05], elec-
os a ic epulsion [Vai02; Bin06])
•A achmen o he p obes [Sch02], linke /space [Bei99], linea and dend ime ic linke s
[Cam06]
• a ious hyb idiza ion pa ame e s [Sch02] (e.g. ionic s eng h, empe a u e, pH, block-
ing eagen s) [Koe05]
•Washing cha ac e is ics [Poz07]
•Mic oa ay size [Dan07], di usion-limi ed a ge anspo [Pap06], mixing [Gu 05;
Toe03]
53
Fundamen als
2.7 DNA Chip Fab ica ion by Ligh -Di ec ed In
Si u Syn hesis
Ligh -di ec ed in si u syn hesis o DNA mic oa ays was de eloped a ound 1990 by Fodo
and cowo ke s [Fod91]. Sho ( ypically ≤25me ) oligonucleo ide p obe sequences a e
syn hesized nucleo ide by nucleo ide on he su ace o he mic oa ay. Spa ially add ess-
able pho o-dep o ec ion enables a massi e pa allel syn hesis o a bi a y DNA p obe se-
quences on a single mic oa ay.
Ligh -di ec ed in si u syn hesis is basically a solid-phase syn hesis p ocess (see sec ion
2.5.1), equi ing phospho amidi e eagen s wi h pho o-labile p o ec ion g oups. Spa ially
con olled pho o-dep o ec ion is achie ed wi h a pho oli hog aphic p ocess and he use o
phospho amidi e eagen s wi h pho olabile p o ec ion g oups. P obe sequence in o ma ion
and mic oa ay geome y is encoded in he pho omasks.
Today, comme cial high densi y oligonucleo ide mic oa ays ( ab ica ed wi h high esolu-
ionpho omasks)ha eup o6.5millionp obes (wi h a ea u e size o 5µm). Ligh -di ec ed
in si u syn hesis can also be employed o he syn hesis o polypep ide sequences [Fod91]
(→p o ein mic oa ays) o o he combina o ial chemis ies.
2.7.1 Pho oli hog aphic Con ol o he Combina o ial Syn-
hesis P ocess
Fo pa allel syn hesis o di e en p obe sequences spa ial con ol o he phospho amidi e
coupling eac ion is equi ed. This is achie ed by a spa ially con olled pho o-dep o ec ion
o he pho olabile 5’-p o ec ion g oup (chemical s uc u e shown in Fig. 2.19). The pho o-
clea age gene a es a hyd oxy-g oup a he 5’-ends o he exposed sequences and hus de-
e mines whe e on he mic oa ay (i.e. a which mic oa ay ea u es/p obe sequences - see
Fig. 2.25 h and k) he nex phospho amidi e building block (p o ided in he subsequen
coupling s ep) will elonga e he sequence.
The ab ica ion o a mic oa ay comp ising a bi a y N-me sequences equi es 4×N de-
p o ec ion/coupling s eps. I is necessa y o p o ide all coupling al e na i es (X=A, C,
G and T) in each ”nucleo ide laye ”. Thus, he ligh -di ec ed combina o ial syn hesis
comp ises a se ies o 4×N pho o-dep o ec ion DXiand associa ed nucleo ide coupling
54
DNA Chip Fab ica ion by Ligh -Di ec ed In Si u Syn hesis
s eps CXi:
1.DA1/CA1→DC1/CC1→DG1/CG1→DT1/CT1
2.DA2/CA2→DC2/CC2→DG2/CG2→DT2/CT2
⇒...
N. DAN/CAN→DCN/CCN→DGN/CGN→DTN/CTN
Spa ially con olled pho o-dep o ec ion is shown in mo e de ail in Fig. 2.25 whe e each
p obe s and symbolizes an indi idually add essable mic oa ay ea u e (whe eas in eali y
each ea u e comp ises millions o iden ical p obes). Assuming a s epwise coupling e i-
ciency c he yield Y= cNo p obes which a e ee o syn hesis de ec s is dec easing wi h
he powe o he p obe leng h N. P obes con aining de ec s canno be epai ed o emo ed
as in common solid-phase syn hesis (capping, unca ion, HPLC sepa a ion). Syn hesis de-
ec s (i.e. single base misma ches, inse ions anddele ions) will he e o e a ec mic oa ay
hyb idiza ion [Job02].
The leng h o he mic oa ay-p obes is de e mined by he applica ion. Sho e 15-25me
p obes p o ide a high disc imina ion capabili y be ween PM and MM and a e he e o e
sui able o SNP de ec ion and esequencing assays. Longe p obes a e less disc imina i e
bu a he mo e sensi i e(inc eased binding a ini y), and a e he e o e a o able o de ec-
ion o low abundance mRNAs in exp ession p o iling applica ions. Typically p obes on
high densi y oligonucleo ide mic oa ays ha e a leng h o ≤25 n , howe e , he ab ica-
ion/applica ion o a ays wi h longe 40-60 n p obes has also been epo ed.
The syn hesis cycle
Fo ligh -di ec edinsi usyn hesispho olabilephospho amidi e eagen s, α-me hyl-6-ni o-
pipe onyloxyca bonyl (MeNPOC) [Pea94; McG97] o [2-(2-ni ophenyl)-p opyloxyca -
bonyl]-2’-deoxynucleoside (NPPOC) phospho amidi es [Has97] a e used.
The MeNPOC-chemis y (employed in he ab ica ion o A yme ix GeneChips R
) has a
s epwise yield o 92 o 94% [McG97]. Signi ican ly be e coupling yields ha e been e-
po ed o NPPOC phospho amidi es [Bei99]. Nuwaysi e al. [Nuw02] epo ed s epwise
chemical yields be ween 96 and 98%.12
Use o NPPOC phospho amidi e eagen s (chemical s uc u e shown in Fig. 4.1) has been
epo ed in [Has97; Bei99; Nuw02; Bau03; Wol04; Woe06]. MeNPOC phospho amidi es
eagen s we e used in [Pea94; McG97; SG99; Lue02].
12 S epwise syn hesis yields o NPPOC phospho amidi es acco ding o [Nuw02]: NPPOC-A( ac) 96%,
NPPOC-C(ibu) 99%, NPPOC-G(ipac) 97%, NPPOC-T 98%
55
Fundamen als
Figu e 2.25: Ligh -di ec ed in si u syn hesis o DNA mic oa ays [Fod91]. The
spa ially con olled combina o ial chemis y app oach enables pa allel syn hesis o
a bi a y p obe sequences. In he (non-op imized) coupling scheme shown he e, he
p obe sequences a e syn hesized ”laye by laye ”: o co e all coupling-al e na i es,
in each ”nucleo ide laye ” phospho amidi e-couplings a e pe o med in he o de A,
C, G, T. In he abo e se ies o images each p obe s and symbolizes an indi idually
add essable mic oa ay ea u e (whe eas in eali y each ea u e con ains millions o
p obes). Syn hesis o he i s nucleo ide laye (a- ). (a) The subs a e is ini ially
unc ionalized wi h pho o-labile p o ec ion g oups (depic ed as blue balls). Spa ially
con olled UV exposu e (use o pho omasks) is es ic ed o hose ea u e a eas whe e
phospho amidi e building blocks a e o be a ached in he subsequen coupling s ep.
(b) Pho o-clea age o he p o ec ion g oups c ea ed hyd oxyl-moie ies, which a e he
binding si es o he subsequen adenosine-phospho amidi e coupling s ep (c). In he
coupling s ep only one building block can a ach o each dep o ec ed s and. Fu -
he couplings a e p e en ed by new p o ec ion g oups (impo ed wi h he building
blocks). (d) Pho o-dep o ec ion o hose p obes which equi e cy osine a he i s
base posi ion. (e) Coupling o cy osine-phospho amidi e. The i s nucleo ide laye
is comple ed a e dep o ec ion and coupling o G nucleo ides (no shown) and T
nucleo ides ( ). The second laye is syn hesized upon he i s laye (g-l). The de-
p o ec ion/coupling scheme is con inued un il he inal leng h o he oligonucleo ide
p obes is eached (m).
56
DNA Chip Fab ica ion by Ligh -Di ec ed In Si u Syn hesis
Owing o he he 5’-a achmen o he NPPOC p o ec ion g oups he in si u syn hesis is
pe o med in 3’→5’ di ec ion. The e o e he p obes a e ypically 3’- e he ed a a he mi-
c oa ay su ace. Howe e , 5’- e he ed p obes can be syn hesized (in 5’→3’ di ec ion)
wi h modi ied phospho amidi e eagen s (ca ying 3’-NPPOC p o ec ion g oups) [Alb03].
5’- e he ed mic oa ay p obes a e, unlike 3’- e he ed p obes, a ailable o enzyma ic mod-
i ica ion.
In he ollowing we e e o he 5’-NPPOC phospho amidi e chemis y (Fig. 4.1) which
has been employed in his wo k.
The syn hesis cycle in he ligh -di ec ed syn hesis p ocess (Fig. 4.2) is e y simila o he
scheme employed o oligonucleo ide syn hesis on CPG-suppo s (shown in Fig. 2.20).
Exposu e wi h UV ligh (λ=350-380 nm) induces pho o-dep o ec ion and enables cou-
pling o he nex phospho amidi e building block. A capping s ep (as shown in Fig. 2.20),
esul ing in unca ed s ands a he han in s ands con aining single base MMs, is o a
a he limi ed alue in mic oa ay syn hesis ( unca ed s ands canno be emo ed) and is
he e o e omi ed. Coupling and oxida ion s eps a e pe o med in he same way as in he
oligonucleo ide syn hesis on CPG-suppo s.
Pho o-dep o ec ion o NPPOC esul s in sho -li ed in e media e s a es. Acco ding o
[Wal01] an aci-ni o in e media e is in acid-base equilib ium wi h i s anion. The uns able
anion can agmen , hus esul ing in he desi ed dep o ec ion eac ion (comple e emo al
o he NPPOC g oup). Howe e , ia a compe ing eac ion pa hway, he aci-ni o in e -
media e can also o m a ni oso p oduc , which is no emo ed om he phospho amidi e
esidue, hus p e en ing pho o-dep o ec ion.
To p omo e he desi ed eac ion pa hway he pho o-dep o ec ion needs o be pe o med in a
sol en p o iding su icien p o on accep o s. The e o e he basici y o sol en ace oni ile
is inc eased by addi ion o a mild base (e.g. pipe idine [Bei99]).
2.7.2 Combina ion o ”Maskless” Digi al Pho oli hog aphy
and Combina o ial Chemis y
Ligh -di ec ed in si u syn hesis o DNA mic oa ays wi h high esolu ion pho omasks has
been de eloped and is employed on an indus ial scale by A yme ix Inc.. High cos s o
ch omium masks, conside able echnical e o o he mask alignmen , and he lack o
lexibili y (a new se o pho omasks is equi ed o each new mic oa ay design) ha e so
a p e en ed lab-scale applica ion o he pho omask-based ab ica ion echnique.
The use o compu e -con olled spa ial ligh modula o s as ” i ual pho omasks” can ci -
cum en he limi a ions ela ed o he use ch omium pho omasks and hus p o ide g ea
lexibili y o cus om mic oa ay ab ica ion.
57
The Mic oa ay Syn hesize
Figu e 3.2: (a) Pho og aph o he maskless mic oscope p ojec ion pho oli hog aphy
sys em ( op iew). Along he op ical pa h (do ed whi e line): UHP lamp housing, UV
cold mi o s, shu e , band pass il e s (g een and UV), DMD and d i e elec onics,
ube lens, mic oscope (Zeiss Axio e 135) and he eac ion cell, which is moun ed
on o he sample holde .
(b) D awing o he li hog aphy sys em: Ul a High P essu e lamp (UHP) powe ed by
ideo p ojec o (VP2), plano-conca e silica lens (L1), plano-con ex lens (L2), UV cold
mi o F1, ligh ap (LT), plano-con ex lens (L3), UV cold mi o (F2), shu e (S),
bandpass il e s o UV (F3) and g een (F4) illumina ion, plano-con ex lens (L4), old
mi o s (M1 and M2), DMD and d i e elec onics o he As oBeam p ojec o (VP1),
ube lens (L5), in ini y co ec ed mic oscope (ICM), mi o /beamspli e -assembly
(M3), 5×(0.25 NA) Flua mic oscope objec i e (FO), subs a e o be pa e ned (PS).
Technical de ails a e p o ided in Appendix B.3.
64
The Maskless Mic op ojec ion Pho oli hog aphy Sys em (MPLS)
250 W UHP lamp o ano he ideo p ojec o (Op oma EP 758). Since UHP lamps equi e
specialized powe supplies (in eg a ed in he ideo p ojec o ) he Op oma p ojec o is now
employed as a lamp powe supply.
Due o he equi emen o high UV ansmission we couldn’ use he highly op imized op-
ics2o he ideo p ojec o . Fo he pho oli hog aphy sys em a new UV illumina ion op ics
had o be designed: he lamp module o he Op oma EP758 p ojec o was buil in o an
ai cooled housing and connec ed ia an ex ension cable o he lamp d i e o he Op oma
p ojec o (VP2). The a c o he 250 W UHP lamp is loca ed a he inne ocal poin o
he ellip ical lamp e lec o . To e icien ly collima e he s ongly di e gen beam, a plano-
conca e di ac ion lens (L1) ( =50 mm, 25.4 mm diam., used silica) is placed be ween
he lamp window and he ou e ocal poin o he e lec o .
E icien il e ing o he nea UV wa eleng h band equi ed o he pho o-dep o ec ion
eac ion p o ed o be di icul owing o he high he mal load. Fil e ing is he e o e pe -
o med in se e al s eps: A dich oic il e om he Op oma lamp module (F1) (o iginally
designed as a UV p o ec ion il e ) is employed as a UV cold mi o o cu down he is-
ible ligh in ensi y o abou 10 pe cen . UV ligh below 400 nm is e icien ly e lec ed.3
In a ed adia ion is il e ed using ano he UV cold mi o (O iel) (F2). Finally a band
pass in e e ence il e (F3) (bk-370-35-B, In e e enzop ik Elek onik GmbH) is used o
selec ing he wa eleng h band in he me cu y i-line egion (λ= 365 nm) equi ed o he
pho o-dep o ec ion eac ion. Taking in o accoun ha he me cu y i-line is conside ably
b oadened due he high ope a ion p essu e o he lamp, we had o use a ela i ely wide
band pass il e (peak ansmission Tmax = 60% a 370 nm, FWHM: 33 nm) o achie e
a su icien ly high UV ansmission. Use o a b oadband il e (colo glass UG-5, Scho ,
ansmission be ween 230 and 430 nm and abo e 650 nm, Tmax = 90% a 350 nm ) would
esul in se e e ch oma ic abe a ion.
3.2.2 Digi al Mask P ojec ion Using a Digi al Mic omi o
De ice
The DMD is a spa ial ligh modula o commonly used o image gene a ion in DLP ideo
p ojec ion sys ems ( o echnical de ails on DMD echnology see sec ion B.1). In ou se up
we use a DMD wi h XGA esolu ion con aining 1024×768=786432 squa e mi o s (16 µm
in size wi h a pi ch o 17 µm) ha can be il ed by an angle o +10◦o -10◦ ela i e o he
2Op imized o high ligh h oughpu and uni o mi y o illumina ion.
3The ansmission spec um o he dich oic il e shows a dis inc cu o a 415 nm - om 420 o
700 nm he ansmission is ≥90%. The e lec i i y in he i-line ange couldn’ be measu ed wi h
he spec opho ome e a ailable. Howe e , a simple expe imen wi h a 100 mW UV-LED (Nichia
NCCU033) shows ha he UV e lec i i y is (coa sely es ima ed) be ween 60 and 80%.
65
The Mic oa ay Syn hesize
no mal axis o he chip. The wo posi ions a e e e ed o as on- and o -s a e: Mi o s in
he on-s a e e lec he inciden ligh pe pendicula o he DMD su ace in o he p ojec ion
op ical sys em, whe eas mi o s in he o -s a e e lec ligh a an angle o 40◦ ela i e o
he DMD no mal axis in o a ligh ap (Figu e 3.3).
+10° -10°
40°
20°
20°
ON OFF
I
P
T
I
Figu e 3.3: Spa ial ligh modula ion wi h a Digi al Mic omi o De ice. Mi o s
in he on-s a e (blue) e lec he inciden ligh (I) in a di ec ion no mal o he DMD
su ace in o he p ojec ion op ics (P). Mi o s in he o -s a e (g een) e lec he ligh
unde an angle o 40◦wi h espec o he no mal axis in o a ligh ap (T).
The DMD is o ien ed pe pendicula o he op ical axis o he p ojec ion sys em. The mi-
c omi o s il a ound hei diagonal axis. We ha e o a ed he DMD by 45◦a ound he
op ical axis, so ha he inciden beam and he e lec ed beam lie bo h in he ho izon al
plane o he se up (Figu e 3.4).
Technical de ails on he modi ica ion o he DLP ideo p ojec o a e p o ided in Appendix
B.2. Fo be e accessibili y o he mic omi o a ay he DMD boa d had o be emo ed
om he p ojec o chassis and econnec ed o he d i e boa d ia a 148 pin ex ension ca-
ble. Because he d i e elec onics o he p ojec o emains unchanged, all so s o ideo
signals can be used o con ol he image display. Connec ion o a PC wi h a dual-head
g aphics ca d p o ed o be use ul, as one sc een can be used o con ol pu poses (e.g. o
unning he DNA syn hesis con ol p og am which au oma es and coo dina es pho oli ho-
g aphic pa e n display and he luidics sys em) while he o he one is ese ed o pa e n
display.
3.2.3 The Image P ojec ion Op ics
To educe he mic oa ay size o a ew mm2we op ed o a mic oscope p ojec ion app oach.
Reduced dimensions o he mic oa ay a e bene icial o mic oa ay hyb idiza ion due o
educed di usion imes [Dan07] and educed ma e ial equi emen s (syn hesis eagen s
66
The Maskless Mic op ojec ion Pho oli hog aphy Sys em (MPLS)
P
T
I
a)b)c)
Figu e 3.4: Ro a ed DMD a angemen in he maskless mic oscope p ojec ion
li hog aphy se up. The DMD is o a ed by 45◦a ound he op ical axis, so ha he
il ing axis o he mi o s is e ical. The inciden beam and he e lec ed beam lie
bo h in he ho izon al plane o he se up.(Le image) F om he mi o s in on-posi ion
(a anged as a ”X”) he incoming ligh (I) is e lec ed owa ds he p ojec ion op ics
(P). Mi o s in he o -posi ion e lec he ligh in o a ligh ap (T). (Cen e image)
View o he DMD om he p ojec ion op ics. (Righ image) View o he DMD om
he ligh ap.
and nucleic acid sample size). By educing he image a ea, he illumina ion in ensi y is
inc eased by a simila ac o : a 250 W UHP lamp does su ice in o de o keep he ime e-
qui ed o op ical dep o ec ion in a easonable ela ionship o he o al u no e ime o he
chip syn hesis. Use o he mic oscope also p o ides supe io con ol o he image ocusing
and mechanical s abili y. Image d i occu ing om he mal expansion o he op ical pa s
has p e iously been desc ibed as se ious p oblem in he ligh -di ec ed syn hesis p ocess,
equi ing ac i e con ol o ocusing, e.g. by means o an image-locking echnique [Ric04].
An impo an aspec in he design o he li hog aphy sys em is image con as . In ligh -
di ec ed mic oa ay syn hesis s ay ligh is much mo e c i ical han o example wi h pho-
o esis . Pho o esis , ha ing a s ong nonlinea exposu e cha ac e is ics, doesn’ espond
o small s ay ligh in ensi ies below a h eshold alue. In mic oa ay syn hesis he e is no
h eshold and s ay ligh induced e o s can accumula e o e many exposu e s eps. Wi hin
he o al exposu e ime o abou wo hou s, s ay ligh causes base inse ion e o s, a ec -
ing mos o he syn hesized DNA s ands.
The whole syn hesis p ocess in ol es abou 80 exposu es wi h di e en mask pa e ns, i
ex ends o e abou 6.5 hou s. Mask alignmen equi es he mal and mechanical s abil-
i y. To make use o he maximum pixel esolu ion o he se up (which is 3.5 µm wi h a
5×mic oscope objec i e) no mo emen s caused by ib a ions, ension elease, o he mal
expansion la ge han abou 1 µm (in he on ocal plane o he objec i e)can be ole a ed.
67
The Mic oa ay Syn hesize
Figu e 3.5: The p ojec ion op ics sys em. Inciden ligh (I) ( il e ed - ei he nea
UV o he exposu e o g een o ocusing); ligh ap (T); ube lens (TL); beam
spli e (BL); mic oscope objec i e (MO); syn hesis cell (S).
The mic omi o a ay (DMD) is placed in he image plane (loca ed ou side he mic oscope
ame) o he in e ed mic oscope. Wi h in ini y co ec ed mic oscope objec i es, a ube
lens (TL) is necessa y o p ojec he image o he DMD o in ini y. The adjus men o
he dis ance be ween DMD and ube lens, which does no exac ly equal he nominal ocal
leng h o 164.5 mm (as speci ied by he manu ac u e ), is c ucial o he calib a ion o he
se up, as explained la e (in Sec. 3.2.5).
A mo able hal mi o /hal beamspli e op ical elemen (BS), loca ed a he posi ion o he
mic oscope’s luo escence il e block, is used o e lec he ligh in o he objec i e back
ape u e. Using he beamspli e pa , he ligh e lec ed om he su ace o he mic oa ay
subs a e can be coupled in o he mic oscope. This is employed o exac ocusing and
di ec obse a ion o he p ojec ed image h ough he eyepiece. Fo pho opa e ning, he
mi o pa is used (exchange is achie ed by sliding he pla e by hand). In p inciple o
his pu pose a dich oic beamspli e ( e lec ion o UV ligh and educed e lec ion o isi-
ble ligh ) could be used. Howe e , he use o a beamspli e pla e ( o pho o-dep o ec ion)
u ned ou o be p oblema ic since e en a small amoun o e lec ion a he backside o he
pla e can p oduce ghos images, and hus signi ican ly a ec he image con as .
Among se e al objec i es(MO) es ed, we ound he Zeiss Flua 5×(0.25 NA) as mos sui -
able o DNA chip ab ica ion, pa icula ly o i s supe io UV ansmi ance and i s la ge
back ape u e allowing o e icien ligh collec ion. O e a wo king dis ance o 12.5 mm
he image o he DMD is p ojec ed on o he DNA syn hesis subs a e - a chemically unc-
ionalized glass su ace - inside he syn hesis cell (S).
A 10×(0.30 NA) Plan Neo lua and a 20×(0.5 NA) Plan Neo lua objec i e (Zeiss) we e
success ully used o u he educe he image size. Diminished con as makes hese ob-
68
The Maskless Mic op ojec ion Pho oli hog aphy Sys em (MPLS)
jec i es less sui able o ligh di ec ed mic oa ay ab ica ion. Howe e , pa e ning o pho-
o esis - ha ing lowe equi emen s on con as - should be simple wi h hese highe mag-
ni ica ion objec i es.
A a wa eleng h o 365 nm he di ac ion limi o he 5×(0.25 NA) Flua objec i e is
R=λ/(2 ·NA) = 0.73 µm. Howe e , a signi ican ly la ge dis ance be ween adjacen
ea u es is necessa y o achie e a su icien local con as o he ligh -di ec ed ab ica ion
p ocess.
Re lec i e objec i es ha e he ad an age o a high UV ansmission and a e no subjec
o ch oma ic abe a ions. We he e o e es ed image p ojec ion wi h a 15×(0.28 NA)
Schwa zschild ype e lec i e objec i e (Ealing). Howe e , a sa is ac o y image con as
o e he whole ield couldn’ be achie ed. Also, in he gi en op ical sys em, owing o a
na ow back ape u e, he ligh h oughpu h ough he e lec i e objec i e is e y limi ed.
3.2.4 Fab ica ion and Applica ion o UV-Sensi i e Pho o-
ch omic Films
Fo e alua ion o he imaging quali y a as and simple me hod o gene a ing pa e ns
upon UV exposu e is equi ed. Pho og aphic ilms and pho o esis u ned ou o be no
e y use ul due o di icul handling and p ocessing e o s. The e o e we ha e de eloped
a UV-sensi i e ilm based on he pho och omic dye spi opy an. Spi opy an unde goes a
s uc u al change when exposed o UV-ligh . This esul s in a s ongly inc eased ligh
abso p ion in he isible ange.
P epa a ion o pho och omic ilms:
We dissol ed 10 mg o spi opy an dye (1’,3’-dihyd o-1’,3’,3’- ime hyl-6-ni ospi o
[2H-1-benzopy an-2,2’-(2H)-indole], Ald ich, Ca .: 27,361-9) in 1 ml o PMMA
pho o esis (E-beam esis PMMA 200 k; AR-P 641.04, All esis GmbH, S aus-
be g, Ge many) and spincoa ed a hin ilm ( hickness abou 1 µm) on o a mic oscope
slide. O he esis s - we also ied wi h Mic oChem PMMA and Mic oChem SU-8
50 - wo k equally well. The pho o esis is used as a ca ie ma e ial only. A e
spincoa ing, and b ie hea ing on a ho pla e (1 minu e a 100◦C) he slides a e eady
o use.
We ound hese pho och omic ilms o be a well-sui ed imaging ma e ial. Unlike wi h
pho o esis o pho og aphic ma e ial no de eloping o o he p ocessing is equi ed. Unde
UV exposu e he ilm changes om anspa en o an almos opaque pu ple. Wi h he
in ensi ies we usually apply (50-100 mW/cm2) his happens wi hin seconds. The p ocess
can be e e sed by hea ing o by illumina ion wi h b igh ligh (a isible wa eleng hs).
69
The Mic oa ay Syn hesize
Unless he spi opy anhas been bleached wi h high i adia ion doses, he ilms can be eused
se e al imes.
Fo a small exposu e dose he op ical densi y inc eases almos linea ly wi h he dose o UV
ligh . Fo la ge doses D he op ical densi y OD app oaches sa u a ion.
OD=ODsa (1−exp (−cons .·D)) (3.1)
Upon e y high exposu e, pho odeg ada ion o he pho och omic dye (bleaching) esul s in
educed OD alues. Since he linea exposu e cha ac e is ics o he spi opy an dye a e e y
simila o ha o NPPOC phospho amidi e eagen s, spi opy an ilms a e a e y use ul ool
o es ing and e alua ion o he UV op ical sys em.
3.2.5 Ch oma ic Co ec ion o he P ojec ion Op ical Sys-
em
Since he dep h o ocus DOF=λ/NA2is only abou 6 µm o he 5×(0.25 NA) Flua objec-
i e (a λ=365 nm), i is necessa y o pe o m p ope ocusing each ime a new pa e ning
subs a e is moun ed on he sample holde . The ocus ange p o iding op imum con as is
e en smalle han he dep h o ocus, hus pe ec ocusing o he pa e n on o he su ace
is c ucial. I can be achie ed by obse ing he back e lec ion o he p ojec ed image ( om
he pa e ning su ace) h ough he mic oscope eyepiece. This is easy o pe o m wi h is-
ible ligh , bu a he di icul wi h UV ligh .
I he back- e lec ed image o he pa e n is pe ec ly ocussed in isible (g een) ligh , his
usually is no ue o UV a he same ime. This is owing o ch oma ic abe a ion. Lon-
gi udinal ch oma ic abe a ion causes an axial ocus shi usually esul ing in a comple ely
blu ed image in UV. In he ollowing we desc ibe a me hod o he co ec ion o his lon-
gi udinal ch oma ic abe a ion, so ha ocusing o he nea UV image can be pe o med by
obse a ion ( h ough he eyepiece) and ocus adjus men unde g een ligh illumina ion.
Using pho och omic ilms as a con ol o he quali y o he p ojec ed UV pa e n, we ound
ha he ch oma ic abe a ions can be compensa ed by ine-adjus men o he dis ance dbe-
ween he DMD and he ube lens (see Fig. 3.2). The dis ance dis oughly he nominal
ocal leng h o he ube lens o 164.5 mm. A e ocusing wi h g een ligh , he ilm is ex-
posed wi h a con ol pa e n in UV and subsequen ly inspec ed on a ligh mic oscope. The
dis ance dnow can be adjus ed i e a i ely un il he pa e ns imaged on he spi opy an slide
indica e pe ec ocusing. Jus a small de ia ion o a ew millime e s om he nominal o-
cal leng h o he ube lens is necessa y o ch oma ic co ec ion. The ole ance o d, wi hin
which a good co ec ion is achie ed, is only a ew en hs o a millime e wide. Once he
70
The Maskless Mic op ojec ion Pho oli hog aphy Sys em (MPLS)
ch oma ic co ec ion p ocedu e has been accomplished, ocusing can always be pe o med
unde illumina ion wi h g een ligh .
Cau ion! The abo e op ical adjus men depends on he eye ocal leng h o he expe imen e
who pe o med he adjus men . In daily use o he mic oa ay syn hesize , when ocusing
on he mic oa ay subs a e is pe o med, de ia ing eye ocal leng hs (nea / a sigh edness)
o o he pe sonnel using he equipmen do ma e and need o be accoun ed o .
3.2.6 UV Ligh In ensi y and Uni o mi y o Illumina ion
Fo measu ing he in ensi y a he image plane we used a lase powe senso (PS10Q,
Cohe en Inc.). The he mopile senso was placed in he ocal plane o he mic oscope
objec i e. To measu e he mean in ensi y, a comple ely whi e image was displayed on he
DMD. Wi h he measu ed o al powe o 7.8 mW we de e mined he in ensi y in he image
plane as 87 mW/cm2.
To s udy he uni o mi y o he illumina ion we p ojec ed he image on o a sc een. The
in ensi y was measu ed a di e en egions o he p ojec ed image. An asymme ic la ge
scale de ia ion wi h a peak in ensi y o abou 140% o he mean in ensi y is obse ed. This
is due o he con igu a ion o he illumina ion sys em: The UHP lamp’s a c gap is o ien ed
pa allel o he op ical axis, p o iding a e y inhomogeneous illumina ion p o ile. Fo his
eason in a ideo p ojec ion sys em an in eg a o elemen , e.g. an in eg a o od (which is a
ligh guide wi h a ec angula c oss sec ion) o a ly-eye lens a ay is employed o gene a e
a e y uni o m illumina ion. Using he in eg a o od o he As oBeam p ojec o u ned
ou o be no easible as he glass od abso bs mos o he UV ligh .
We decided o la en he illumina ion p o ile by using only a small homogeneous sec ion
o he ligh cone o illumina ing he DMD. This way we sac i ice abou 80% o he ligh .
Ne e heless, he emaining 20% o ligh allow pho o-dep o ec ion o be pe o med in a
easonable ime. Al e na i ely, i such pa s we e a ailable, a qua z in eg a o od o an
in eg a o pla e ( ly-eye lens a ay [Sun05]) could be used o achie e signi ican ly highe
ligh in ensi ies.
To a ain a mo e uni o m illumina ion we employ he DMD o in ensi y le eling, simila
as desc ibed by Huebschman e al. [Hue04]. Fo hispu pose we ha e c ea ed an ”in ensi y
le eling mask”. The black and whi e images ( o be used as a pho omasks) can easily be
le eled o educe in ensi y a ia ions o abou ±10% by pixelwise mul iplica ion wi h his
mask. To gene a e he in ensi y le eling mask, a ully illumina ed image (all mi o s in he
on-s a e) is p ojec ed on he sc een (as desc ibed abo e, s ill wi hou using he mic oscope
objec i e) and pho og aphed wi h a Nikon Coolpix 4500 digi al came a. Deskewing he
71
The Mic oa ay Syn hesize
aw image using s anda d image p ocessing so wa e esul s in a 1024×768 pixel image,
which inally has o be in e ed and adjus ed in b igh ness and con as . The le eling mask
is hen p ojec ed on o he sc een and a pho ome e is used o measu e uni o mi y o illumi-
na ion. In an i e a i e way image b igh ness and con as a e adjus ed o achie e a uni o m
in ensi y wi hin mos o he image a ea. Con ou plo s o he ligh in ensi y be o e and
a e in ensi y le eling a e shown in Fig. 3.6. Only in he ou e mos co ne s o he image
(comp ising abou 10% o he o al image a ea) he in ensi y is educed o abou 50% o
he mean in ensi y. This is due o igne ing: Ligh e lec ed om he co ne s o he DMD,
which a e loca ed close o he edge o he en ance pupil, is pa ially blocked by he ape -
u es o he ube lens espec i ely he mic oscope objec i e. Applying in ensi y le eling
we achie ed a mean ligh in ensi y o 76 mW/cm2.
125
120.6429
116.2857
111.9286
107.5714
103.2143
98.8571
90.1429
81.4286
68.3571
20 40 60 80 100 120 140 160 180 200
20
40
60
80
100
120
140
104
101.1429
98.2857
95.4286
92.5714
89.7143
86.8571
81.1429
66.8571
66.8571
20 40 60 80 100 120 140 160 180 200
20
40
60
80
100
120
140
a) b)
Figu e 3.6: Uni o mi y o illumina ion. (a) In ensi y con ou map be o e in ensi y
le eling. (b) A e in ensi y le eling. Using he ube lens, he image o he DMD was
p ojec ed on o a sc een, wi hou he mic oscope objec i e in place, and pho og aphed
wi h a digi al came a. Vigne ing om he mic oscope objec i e is neglec ed he e bu
his e ec is small compa ed o igne ing o he ube lens.
The in ensi y alues men ioned abo e we e achie ed using an in e e ence il e wi h a
FWHM o 33 nm and a maximum ansmission o 60% a a cen e wa e leng h o 370 nm.
Using a na ow i-line il e (FWHM 12 nm a a cen e wa eleng h o 365 nm; 35% maxi-
mum ansmission) p o ided signi ican ly lowe in ensi ies (abou one nin h o he in ensi y
achie ed wi h he b oad il e ). The demand o a wide il e can be explained by he s ong
line b oadening due o he high ope a ion p essu e o he UHP me cu y a c lamp.
72
The Maskless Mic op ojec ion Pho oli hog aphy Sys em (MPLS)
3.2.7 Op ical Sys em Pe o mance Tes ing wi h UV-Sensi i e
Pho och omic Films
Ligh -di ec ed syn hesis o DNA mic oa ays equi es ha he image is p ojec ed on o a
subs a e inside an ine eac ion chambe , so ha eac ions can ake place unde a mois-
u e ee a gon a mosphe e. The syn hesis subs a e, a 0.17 mm hickness mic oscope co e
glass, is o ming he window o he eac ion cell. Hence he image has o be p ojec ed on o
he inne ace o he window. Fo image ocusing (see Sec. 3.2.5) we use he small ac-
ion o g een ligh which is e lec ed back om he imaging su ace in o he mic oscope.
Applying a simila app oach o con as measu emen is no p ac icable because he ou e
ace o he co e glass con ibu es o back- e lec ion as well. Mul iple e lec ions in he
mic oscope sys em (e.g. om a beamspli e ) may deg ade he image con as u he .
Con as a ios o 1:3000 (as can be ound in p oduc speci ica ions o ideo p ojec ion
sys ems) usually e e o he ull-on/ ull-o con as ob ained by compa ing he in ensi-
ies o comple ely black espec i ely whi e images. On ou se up (placing a pho ome e
in o he ocal plane o he mic oscope objec i e) we measu ed a ull-on/ ull-o con as o
abou 3400:1. This means ha he DMD chip wi h he mi o s in he o -posi ion e lec s
only abou 0.03% o he exposu e in ensi y on o he imaging subs a e. This means ha
he amoun o ligh sca e ed by he DMD housing and by he mi o s in he o -posi ion is
negligible.
Much mo e ele an o DNA mic oa ay syn hesis is he local con as [Kim04] be ween
neighbo ing ea u es. The local con as is diminished by ligh -sca e ing and di ac ion
om mi o s in he on-s a e, bu also by op ical abe a ions, which cause dis o ions o
he poin sp ead unc ion. I also depends on he ea u e geome y (i.e. ea u e size and
ea u e spacing). Re lec ions wi hin he imaging op ics cause la e. This could possibly be
imp o ed by using UV an i- e lec ion coa ed op ical su aces (DMD window, ube lens).
The pa e ns used o mic oa ay syn hesis ypically ha e an a ay s uc u e wi h a pi ch o
17 µm o less. To ob ain an es ima e o he s ay ligh induced e o a e we ha e measu ed
he image con as a high spa ial equencies.
We ound ha he UV-sensi i e ilms we al eady used o adjus men o he UV op ics (see
sec ion 3.2.5) a e e y well sui ed o es ing he pe o mance o he pho oli hog aphy sys-
em. Fo isual inspec ion o he pa e ns we used an op ical mic oscope (Olympus IX81)
equipped wi h an au oma ed X-Y ansla ional s age and wi h a high esolu ion CCD cam-
e a (C9100 EM-CCD, Hamama su Pho onics).
Pa e ns o egula ly spaced line pai s (a pai comp ises a black and a whi e ba o equal
wid h), we e imaged on o pho och omic ilm (Fig. 3.7). The spa ial equency o he pa -
e n was a ied be ween 14 and 70 line pai s pe millime e (lp/mm). Using an exposu e
73
The Mic oa ay Syn hesize
3.3 The Fluidics Sys em
The modi ied al e block o a comme cial DNA syn hesize (Applied Biosys ems ABI
381A)cons i u es hemaincomponen o he luidicssys em(Fig. 3.12). Amic ocon olle -
ope a ed solenoid al e d i e ( echnical de ails desc ibed in Appendix B.8) enables con-
ol o he luidics sys em ia he RS232-in e ace o he con ol PC.
To ensu e wa e - and oxygen- ee condi ions he mic oa ay syn hesis is pe o med in an
ai igh low cell (Fig. 3.13) unde an ine a gon a mosphe e. A gon gas p essu e is em-
ployed o d i e he eagen anspo . A de ailed schema ic o he luidics sys em is shown
in Fig. 3.12.
3.3.1 The Syn hesis Cell
Technical equi emen s
• esis ance o he agg essi e sol en s MeCN, THF and py idine
•use o chemically ine ma e ials (no a ec ing DNA syn hesis)
• igh sealing (no seeping o eagen s below he gaske ) is necessa y o enable com-
ple e exchange o eagen s (e.g. o a oid con amina ion wi h wa e le o e om he
p e ious eac ion s ep)
•negligible dead olume ( equi ed o as and comple e exchange o eagen s)
• he 0.17 mm mic oa ay subs a e mus cons i u e a window o he cell
•p e en ion o gas bubble s icking a he edges o he cell olume
•ligh e lec ion and sca e ing mus be a oided
Implemen a ion
The cell olume is o med by a s eamlined cu ou (shown o example in Fig. 3.15) in an
app ox. 1 mm hick shee o polydime hylsiloxane(PDMS) silicone ubbe . PDMS is used
o i s chemical ine ness and sealing capabili y.7The agile mic oa ay subs a e (diam.
22 mm ound co e glass) can be eliably sealed wi h li le o ce, hus wi hou he isk
o ac u e. The DNA chip subs a e is employed as an op ical window (Fig. 3.14). Pho-
omasks a e p ojec ed wi h a mic oscope objec i e on o he inne su ace o he subs a e,
whe e he DNA p obes a e syn hesized.
7E en hough PDMS appea s o be chemically ine , we obse ed ( e e sible) sol en swelling o he
PDMS gaske upon exposu e o e ahyd o u ane and py idine. To p e en excessi e de o ma ion
o he syn hesis olume, oxida ion and capping s eps should no be longe han necessa y. An
al e na i e THF- (and wa e -) ee oxidize solu ion (enabling phospho amidi e syn hesis on PDMS
su aces) has been desc ibed in [Moo05].
80
The Fluidics Sys em
A gon
A gon
A gon
Ac i a o Cap A Cap B Dep. Soln.
A gon
Oxidize
A gon
MeCN
24
23
20 19 18
11
109
8
76
AX
T
C
G
A gon
Ven
21
22
154
3
2
A gon
MeCN
Oxidize
Dep . Soln.
Cap A
Cap B
Ac i a o
A
T
Was e
X
G
C
8
7
6
5
1
4
3
2
14
13
12
11
10
9
Was e
A gon
17
15
Syn hesis Cell
Was e
16
T-Piece
A
B
Figu e 3.12: Schema ic o he luidics sys em. The al e block has been adop ed
om a comme cial oligonucleo ide syn hesize . The syn hesis cell has eplaced he
syn hesis column employed in s anda d oligonucleo ide syn hesis. Val e numbe s co -
espond o hose used in he syn hesis con ol so wa e. (A) Val e block. (B) Reagen
s o age bo les. A gon gas p essu e ( ia al es 1, 15, 18-21, 23-24) is employed o
d i e he eagen anspo .
81
The Mic oa ay Syn hesize
Figu e 3.13: Schema ic o he syn hesis cell. Sy inge needles o m he in- and
ou le o he low cell. A PDMS gaske wi h a s eamlined cu ou o ms he syn hesis
cell olume which is sandwiched be ween he chip subs a e and a glass pla e. The
assembly is placed on an in e ed mic oscope. UV ligh om he objec i e is en e ing
he cell h ough he chip subs a e. Mask pa e ns a e p ojec ed on o he inne ace
o he subs a e, whe e he in si u syn hesis akes place.
Figu e 3.14: The syn hesis cell on he mic oscope. The assembly is moun ed on
a p ecision-adjus able aluminium suppo . The mic oa ay subs a e ( ound co e
glass) is loca ed abo e he mic oscope objec i e. Use o anspa en ma e ials (poly-
ca bona e, PDMS and glass) simpli ies handling and enables isual con ol.
82
The Fluidics Sys em
To p e en he a achmen o gas bubbles (a gon gas employed o d i e he luidics sys em
ends o o m bubbles upon p essu e elie ) a he edges o he PDMS cell, we pu e o in
making a cell wi h e y smoo h edges. This is achie ed using a sha p-edged punching ool
(see appendix B.4) ab ica ed (elec ical discha ge machining) by he mechanics wo kshop
o he uni e si y. Smoo h su aces also imp o e he eagen exchange be ween consecu i e
syn hesis s eps. Fo he pu pose o chemical ine ness he uppe side o he syn hesis cell
consis s o a glass mic oscopy slide which is glued on o a 10 mm hick block o UV ab-
so bing Mak olon R
plas ics. Back- e lec ion (and back-sca e ing) o UV ligh om he
in e aces is educed (by index ma ching) wi h a hin laye o PDMS employed as glue.
In- and ou le a e o med by sy inge needles which a e connec ed o he al eblock ia
PTFE ubing (Fig. 3.14). Mo e de ailed in o ma ion on he cons uc ion o he syn hesis
cell is p o ided in appendix B.4.
The design o he cell is op imized o DNA in si u syn hesis wi h ligh -di ec ed pho o-
dep o ec ion. I enables a e y small eagen consump ion o ca. 40 mg o each NPPOC-
phospho amidi e o a 25me syn hesis [Nai06b].
3.3.2 A gon Bubble T apping
The occasional o ma ion o a gon bubbles, owing o p essu e elie du ing he eagen
anspo owa ds he syn hesis cell ( he sol en MeCN is sa u a ed wi h a gon) ep esen ed
a se ious p oblem o he mic oa ay syn hesis. Bubbles which ha e become apped in he
syn hesis olume (Fig. 3.15A) do locally inc ease he s ay ligh in ensi y du ing he UV
exposu e o a ec syn hesis eac ions (coupling e c.) since he subs a e su ace benea h
he bubble is no co e ed by he eagen s.
The ”a gon bubble p oblem” has been esol ed wi h a cle e ly de ised echnique:
•La ge bubbles a e cap u ed by a T-piece bubble ap (Fig. 3.15C) which is in eg a ed
in he inle line.
•Small bubbles (<2 mm diam.), owing o he inc eased channel wid h in he syn hesis
a ea, ha e he endency o ge s uck in he syn hesis olume (Fig. 3.15A). By employing
a sho suc ion pulse he small bubbles a e pushed in o he inle egion o he syn hesis
cell (Fig. 3.15B), whe e hey ge eliably apped.
This me hod o bubble ca ching is highly eliable. In he c i ical s eps o he syn hesis
p ocess he occu ence o bubbles wi hin he syn hesis a ea is p e en ed almos comple ely.
83
The Mic oa ay Syn hesize
A
C
B
Figu e 3.15: Bubble apping echniques. Top iews o he syn hesis cell olume
(A) and (B) – in- and ou le holes a e shown a he igh and he le end o he
s eamlined cell olume. Du ing eagen supply small a gon bubbles can ge in o he
syn hesis cell (A). They can be emo ed om he syn hesis a ea (dashed box) by a
applying a sho suc ion pulse. Bubbles a e mo ed in o he na ow inle egion o he
chambe (B), whe e hey a e apped due o a mo e a o able su ace ene gy. (C)
La ge bubbles ( oo la ge o ge apped in he inle egion) a e cap u ed in a ”T-piece
bubble ap” be o e hey each he syn hesis cell. Ven ing o he accumula ed gas is
achie ed by occasionally opening he al e o he en ing line.
3.4 Au oma ed Mic oa ay Syn hesis - Con olle -
Ha d- and So wa e
The o iginal ABI 381A DNA syn hesize con ol ha dwa e has been subs i u ed by a pe -
sonal compu e based con olle . Fully au oma ed ligh -di ec ed in si u syn hesis is pe -
o med wi h he syn hesize con ol so wa e DNASyn, which is desc ibed in de ail in
appendix B.7. DNASyn in eg a es luidics con ol ( ia an ex e nal mic ocon olle -based
solenoid al e d i e - echnical de ails a e p o ided in appendix B.8) wi h he ” i ual
pho oli hog aphy mask” p ojec ion.
3.5 Pe o mance o he Mic oa ay Syn hesize
An a o dable mic oa ay syn hesize sys em o lab-scale ab ica ion o DNA mic oa ays
has been de eloped om he ollowing widely a ailable componen s:
•Oligonucleo ide syn hesize (second-hand)
84
Pe o mance o he Mic oa ay Syn hesize
•DLP ideo p ojec o (second-hand)
•In e ed esea ch mic oscope (second-hand)
•Pe sonal compu e
•Mic ocon olle -based solenoid al e d i e (home-buil )
•Op ical componen s: Op ical able, il e s, lenses, mi o s
The highly lexible mic oa ay syn hesis sys em enables massi ely pa allel in si u syn he-
sis o almos a bi a y p obe sequences. New mic oa ay designs can be de eloped wi hin
hou s and au oma ically syn hesized o e nigh . The syn hesis o a 25me mic oa ay e-
qui es abou 6.5 hou s (plus 1.5 hou s o he inal dep o ec ion s ep ou side he syn hesis
appa a us). Wi h ou mic oscope-p ojec ion-li hog aphy se up he size o he mic oa ays
has been educed o a o al a ea o <10 mm2. Owing o he minia u iza ion, he cos s
o syn hesis eagen s (NPPOC-phospho amidi es, MeCN, ac i a o , oxidize , e hanol) a e
abou 50 Eu os pe mic oa ay syn hesis. Mo eo e , he small a ea o he mic oa ay en-
ables hyb idiza ion wi h a e y small amoun o a ge solu ion (in p inciple less han 10 µl
a e equi ed). The high s abili y o ou mic oscope-p ojec ion-li hog aphy se up (wi h e-
spec o image d i ing o igina ing om he mal expansion e c.) is bene icial o he quali y
o he syn hesized DNA p obes.
In p inciple each mic omi o -pixel (in o al 1024×768) could be used o syn hesize a mi-
c oa ay ea u e. Howe e , he need o a high local con as and expec ed di icul ies wi h
he image analysis o he small densely-packed ea u es (image dis o ions e c.) equi e
he use o composi e ea u es consis ing o 5×5 DMD pixels (4×4 pixel ea u e a ea plus
1 pixel sepa a ion gap). In he co ne s o he syn hesis a ea (i.e. he imaging ield de ined
by he DMD chip) DNA p obe quali y is su e ing om igne ing (→ educed exposu e
in ensi y) and unco ec ed cu a u e o ield (→ educed local con as ). Fo quan i a i e
in es iga ions o p obe- a ge binding a ini ies, a maximum numbe o abou 25000 mi-
c oa ay ea u es is cu en ly achie able.
85
The Mic oa ay Syn hesize
86
Chap e 4
Ligh -di ec ed in si u Syn hesis o
DNA Mic oa ays
4.1 Ligh -Di ec ed in si u Syn hesis o DNA Mi-
c oa ays
Reagen s
RayDi eTM 3’-phospho amidi es NPPOC-dA( ac), NPPOC-dC(ib), NPPOC-dG(ipac) and
NPPOC-dT (see Fig. 4.1) ca ying pho olabile 5’-ni ophenylp opyloxyca bonyl p o ec-
i e g oups we e pu chased om Sigma-P oligo (Hambu g, Ge many).
Ace oni ile (ROTISOLV R
o DNA syn hesis, wa e <10 ppm, Ca l Ro h GmbH, Ge -
many); Ac i a o 42TM, 0.25 M (P oligo R
); iodine based oxidize (pa no. 401732, Ap-
plied Biosys ems); T ap-PakTM molecula sie e bags (Applied Biosys ems); wa e - ee a -
gon (≤0.5 ppm H2O)
Pho o-dep o ec ion is ca ied ou in a mildly basic (dep o ec ion) solu ion o 25 mM pipe i-
dine (99%, Ald ich) in wa e ee ace oni ile. Al e na i ely, he use o dime hylsul oxide
(DMSO) has been epo ed [Woe06].
Final base dep o ec ion is pe o med (a oom empe a u e o abou 90 minu es) in a 1:1
mix u e o e ylenediamine (analy ical g ade, Fluka) and e hanol (analy ical g ade, VWR,
Ge many).
UV glue (No land op ical adhesi e 60, Edmund op ics) is used o ix he chip on o a s ain-
less s eel suppo .
87
Mic oa ay Syn hesis
NO2
O
CH3
O
O
NCCH3
CH3
CH3CH3
N
P
O
B
O
O
N
N
N
NH( ac)
N
NH
O
O
CH3
N
NH
N
N
NH(ipac)
O
N
N
NH(ib)
O
B=
Adenine( ac) Thymine
Guanine(ipac) Cy osine(ib)
ib=isobu y yl
ac= e -bu ylphenoxyace yl
ipac=isop opylphenoxyace yl
Figu e 4.1: 5’-[2-(2-Ni ophenyl)-p opyloxyca bonyl]-2’-deoxynucleoside phospho-
amidi es. Simila as nucleosides, nucleoside phospho amidi es comp ise nucleobases
and deoxy ibose suga . Addi ionally, phospho amidi es con ain a phospho us g oup,
which, when chemically ac i a ed, can eac wi h he hyd oxy g oup o a g owing (de-
p o ec ed) oligonucleo ide s and, This coupling eac ion c ea es he phospha e g oup
in he suga -phospha e backbone. Va ious p o ec ion g oups enable a con olled syn-
hesis o oligonucleo ide chains wi hou he isk o unwan ed side eac ions. The
pho olabile NPPOC g oup (blue) subs i u es he 5’-hyd oxyl o he pen ose ing. I s
emo al (dep o ec ion) enables coupling o ano he building block. The phospho us
g oup is p o ec ed by a diisop opylamino g oup ( ed) (→phospho amidi e) and a
2-cyanoe hyl p o ec ion g oup (g een). Fu he p o ec ion g oups (ib, ac, ipac) a e
necessa y o p e en side eac ions o he exocyclic amine g oups o he nucleobases
du ing he in si u syn hesis p ocess. All p o ec ion g oups a e emo ed a he end o
he syn hesis.
P epa a ion o he Mic oa ay Syn hesis
Ligh -di ec ed in si u syn hesis was pe o med wi h NPPOC-phospho amidi es [Has97;
Bei99; Nuw02; Nai06b] which di e om he commonly used acid-labile DMT-p o ec ed
phospho amidi esby he pho o-clea able5’-ni ophenylp opyloxyca bonylp o ec iong oup
(NPPOC).
Phospho amidi e eagen s a e highly sensi i e o wa e . To minimize con amina ion wi h
wa e , NPPOC-phospho amidi e solu ions - 40 mM in wa e - ee MeCN - a e p epa ed
only immedia ely be o e he s a o he syn hesis. Dep o ec ion solu ion, oxidize 1and
ac i a o a e mo e s able and can emain on he syn hesize o p olonged imes. Con ami-
na ion wi h wa e is pa icula ly c i ical o he phospho amidi e/MeCN solu ion con ained
in he s o age bo les. Once deg ada ion due o a small amoun o wa e has s a ed, he
phospho amidi es unde go au oca aly ic deg ada ion [K o04]. To minimize wa e con ami-
na ion in c i ical eac ion s eps, molecula sie e bags (T ap-PakTM) a e added o he MeCN
s o age bo le and o he ac i a o s o age bo le. Fu he hin s on phospho amidi e han-
1The oxidize solu ion i sel con ains a conside able amoun o wa e (se e al pe cen )
88
Ligh -Di ec ed in si u Syn hesis o DNA Mic oa ays
dling p ocedu es a e p o ided in appendix B.5.1.
The p epa a ion o he au oma ed syn hesis should be pe o med wi h he syn hesis sc ip
P epSyn.p g, which is execu ed by he con olle so wa e DNASyn. The P epSyn-sc ip
includes he p epa a ion o he phospho amidi e solu ions, p iming o he eagen supply
lines and checklis unc ionali y (ins alla ion o he syn hesis cell, op ics, a gon p essu e,
al e unc ion, eagen a ailabili y).
The au oma ed syn hesis cycle
An ini ial pho o eac i e monolaye is c ea ed by coupling o NPPOC-dT-phospho amidi e
o hehyd oxyl-g oupso he dend ime unc ionalizedsubs a e. The syn hesiscycle, o be
epea ed 4×25=100 imes o he syn hesis o a mic oa ay wi h 25me p obes2, comp ises
phospho amidi ecoupling,phosphi e ies e bond oxida ion, and pho o-dep o ec ion.
•Phospho amidi e coupling is ca ied ou o one minu e wi h a 1:1 mix u e o 40 mM
NPPOC-amidi e solu ion in wa e - ee MeCN and ac i a o solu ion (Ac i a o 42TM,
0.25M)
•A iodine based oxidize solu ion (ABI) is employed o abou 40 s (a e e e y i h
coupling s ep) o oxidize uns able phosphi e ies e bonds, hus o o m s able phos-
pho ies e linkages
•The pho o-dep o ec ion s ep (exposu e dose 7 J/cm2a λ=370 nm) is pe o med in
a 25 mM solu ion o pipe idine (Sigma-Ald ich) in MeCN. Pipe idine [Bei99] p o-
ides he mildly basic condi ions necessa y o pho oclea age o he NPPOC p o ec ion
g oup [Wol04].
Be ween he indi idual eac ion s eps ex ensi e washing o he al e block and o he syn-
hesis cell/supply line is pe o med. I is, o example, absolu ely necessa y o emo e
ace amoun o wa e ( om p e ious oxida ion s eps) om he luidics sys em p io o he
nex coupling eac ion. Al e na ing insing wi h pu e MeCN and lushing wi h a gon gas
is e y e icien o emo e emaining eagen s om he p e ious eac ion s ep. Howe e , i
is impo an ha solid esidues a e no allowed o d y on he subs a e su ace.
The inal coupling s ep is ollowed by comple e pho o-dep o ec ion o he whole mic oa -
ay, o emo e all emaining NPPOC p o ec ion g oups, and by a inal oxida ion s ep.
Capping o un eac ed binding si es by ace yla ion is commonly employed in oligonu-
cleo ide syn hesis o p e en he syn hesis o s ands con aining poin de ec s. Because
o he a he limi ed bene i s o a capping in ligh -di ec ed mic oa ay ab ica ion (see sec-
ion 2.7.1) we do no apply capping in ou DNA Chip syn hesis scheme.
2In p ac ice, owing o mask op imiza ion, only abou 80 cycles a e equi ed.
89
Mic oa ay Syn hesis
Figu e 4.6: We ing cha ac e is ics o he mic oa ay su ace (mic oscope image).
Hyd ophilic ea u es (size abou 20 µm) a e co e ed by a closed hin ilm o wa e .
Regions be ween ea u e blocks a e co e ed wi h iny d ople s.
4.3.3 Hyb idiza ion wi hou De e gen - Unspeci ic Adso p-
ion
Omi ance o he su ac an (Tween-20TM o SDS) esul ed in e y s ong su ace abso p-
ion on he en i e mic oa ay su ace - also in he egions whe e no p obes ha e been
syn hesized. Subsequen addi ion o 0.01% Tween-20 on he same mic oa ay esul ed in
p obe-speci ic hyb idiza ion. Hyb idiza ion in his pa icula expe imen was pe o med
wi h MES hyb idiza ion bu e a oom empe a u e.
4.3.4 I e e sible Ta ge Adso p ion
In MES hyb idiza ion bu e a empe a u es >55◦C a ge s end o bind i e e sibly o he
subs a e su ace, making a euse o he mic oa ays impossible. The luo escence in ensi y
is pa icula y high be ween he ea u es (see Fig. 4.7). This sugges s ha a ge s which ha e
dissocia ed om he p obes a e cap u ed by eac i eg oups a he subs a esu ace adjacen
o he ea u es. The p oblem seems o be ela ed o he use o he MES hyb idiza ion bu e
a high empe a u es (>55◦C) . Using 5×SSPE bu e ins ead, we do no obse e his
cha ac e is ics. Howe e , we ound ha o en (e en a high empe a u es o 70◦C) he
hyb idiza ion signals can no be comple ely emo ed. This p oblem, which has also been
epo ed by Hu e al. [Hu05], could be owing o s able duplexes which do no comple ely
dissocia e a he empe a u es applied. I is also possible ha hyb idized a ge s ha e an
inc eased p obabili y o bonding o unblocked eac i e si es a he mic oa ay su ace. In
96
No ewo hy Cha ac e is ics o he Mic oa ays
Figu e 4.7: Fluo escence mic og aph o i e e sible adso p ion. The ea u e blocks
in he cen e o he image ha e unde gone dissocia ion in pu e MES hyb idiza ion
bu e . A empe a u es o abou 60◦C a he han o de ach om he su ace he
luo escen ly labeled a ge s ha e i e e sibly bound o he mic oa ay su ace. The
b igh es signal is isible in he gaps be ween he ea u es. A he le edge o he
image ano he ea u e block (wi h ano he sequence mo i ) is shown, which has been
hyb idized a e dissocia ion condi ions ha e been applied ( hus demons a ing ha
he o he p obes on he mic oa ay main ained hei hyb idiza ion capabili y).
ei he case he a ge s can be emo ed comple ely i RNA a ge s a e used a he han DNA
a ge s. An alkaline s ipping p ocedu e (sodium hyd oxide) will selec i ely deg ade RNA
a ge s (in o nucleo ides), whe eas DNA p obes emain una ec ed [Hu05].
4.3.5 Robus ness o he Phospho us Dend ime Su ace Coa -
ing
Fig. 4.8 demons a es ha he phospho us dend ime unc ionaliza ion (sec ion 4.2) o ms
a s able ne wo k on he glass su ace. Pa s o he dend ime coa ing (au o luo escence
unde blue exci a ion) ha e come o he su ace a e ha sh ea men wi h an unsui able
s ipping bu e . The obus closed- ilm s uc u e shown in Fig. 4.8 is a he unexpec ed
since he chemis y o he su ace- unc ionaliza ion would a he sugges a monomolecula
laye o unconnec ed dend ime molecules. Howe e , dend ime s bound o he aminosilane
laye possibly o m a densely in e wo en ne wo k. I is u he possible, ha he unc ion-
aliza ion wi h he aminosilane APTES esul s in he o ma ion o a s able mul i-laye ilm.
97
Mic oa ay Syn hesis
Figu e 4.8: Fluo escence mic og aph o he phospho us dend ime subs a e. Use o
an unsui able s ipping bu e (10 minu es in boiling 0.1 M Na2CO3solu ion) e ealed
he s able ne wo k s uc u e o he su ace coa ing. I appea s ha he phospho us
dend ime ne wo k emained in ac , e en hough he coa ing is comple ely de ached
om he glass su ace.
98
Chap e 5
DNA Mic oa ay Analysis
5.1 Hyb idiza ion Signal Acquisi ion - Expe imen-
al Se up
Mic oa ay hyb idiza ion assays we e pe o med in a empe a u e-con olled hyb idiza ion
chambe . The design o he low- h ough ype chambe is simila o ha o he syn hesis
cell (see sec ion 3.3.1). Ins alla ion on an epi luo escence mic oscope se up enables eal
ime moni o ing o he hyb idiza ion signal. A sensi i e elec on mul iplying CCD-came a
(EMCCD) is used o image acquisi ion.
B
A
Figu e 5.1: Mic oa ay analysis se up. (A) Mo o ized luo escence mic oscope wi h
EMCCD-came a (bo om le ). (B) Hyb idiza ion chambe on he XY-s age o he
mic oscope.
99
Mic oa ay Analysis
5.1.1 The Hyb idiza ion Chambe
Design conside a ions:
• eal ime moni o ing ( equi es a window in o he sealed chambe and low backg ound
luo escence)
• eagen exchange (e.g. o eplace he hyb idiza ion bu e by a washing solu ion)
•high mechanical s abili y o minimize de ocusing and xy-d i ing o he image upon
he mal expansion
• empe a u e con ol
AB
Figu e 5.2: Hyb idiza ion chambe assembly (A). Inle /ou le ubes en e om he
op. The mic oa ay is loca ed a he bo om. Pa (B) shows he mic oa ay on i s
s ainless s eel suppo (lying in he on ) and he s ainless s eel op pla e (leaning
agains he b ace) wi h he PDMS gaske . The mic oa ay slide is p essed agains
he aluminium b ace wi h wo as ening sc ews.
The hyb idiza ion chambe is made om a 1.5 mm hick PDMS gaske . The chambe ol-
ume (abou 120 µl) is o med by a 10 mm diam. hole (cu om a shee o PDMS wi h a
punching ool). Ci cula cu -o s a he inle and ou le openings (see Fig. 5.2B) p e en
s icking o ai bubbles inside he chambe olume.
The mic oa ay wi h i s s ainless s eel suppo cons i u es he bo om side o he hyb idiza-
ion olume. This con igu a ion, using he chip subs a e as window, enables obse a ion
o he hyb idiza ion signal wi h an in e ed mic oscope. A s ainless s eel pla e o ms he
uppe side o he hyb idiza ion olume. S ainless s eel is used because i is esis an o
he hyb idiza ion bu e (no sal co osion). Also impo an , since he s eel pla e is in he
backg ound o he mic oscope ield o iew: he s eel pla e isn’ luo escen and doesn’
adso b nucleic acid a ge s.
A lexible The mo oilTM hea e (Minco) (wi h a 15 mm diam. opening in he cen e - o
100
Hyb idiza ion Signal Acquisi ion - Expe imen al Se up
inle /ou le ubes) is glued on o he uppe side o he s eel pla e. Tempe a u e is measu ed
wi h a pla inum esis o (P -100) which is ixed wi h he mal adhesi e a he edge o he
s eel pla e. In- and ou le ubes (loca ed a opposing ends o he chambe olume) pene a e
he s eel pla e om he op side (see Fig. 5.2A). To a oid dead olume and co osion and o
achie e eliable sealing, he PFA ubes a e di ec ly connec ed o he pla e by p ess- i ing.1
Tempe a u e Con ol
PT100
PC
P o iLabExpe 3.0
PID-Tempe a u e-Con olle
Meilhaus
RedLab
USB Measu emen Module
Toellne
TOE 8951
Powe Supply
USB 2.0
A/D in
D/A ou
Minco
The mo oil
hea e
P 100
Tempe a u e
Senso
Resis ance ->Vol age
Con e e
analog
Remo e
Con ol
Hyb idiza ion Chambe
Figu e 5.3: Con ol o he hyb idiza ion empe a u e. Hea ing o he hyb idiza ion
chambe is pe o med wi h a Minco The mo oilTM hea e which is in he mal con ac
wi h he hyb idiza ion solu ion ia a co osion esis an s ainless s eel pla e. The
empe a u e is measu ed wi h a P -100 senso (in he mal con ac wi h he s ainless
s eel pla e). The esis ance is con e ed in o a ol age signal ha is p opo ional o
he empe a u e. The ol age is ead by an A/D inpu channel o he RedLab USB
measu emen module. A so wa e based PID- empe a u e con olle ( un as a PC
applica ion gene a ed wi h P o iLabExpe 3.0 - see Fig. B.16) by compa ing he
ac ual empe a u e and he se empe a e, de e mines he con ol ol age (ou pu ia
he RedLab D/A ou pu ) ha is used o ope a e he emo e con olled hea e powe
supply.
Tempe a u e ismeasu edwi haP -100 esis o and con e edin oa empe a u e-p opo ional
ol age signal. A USB measu emen module (ME-Redlab, Meilhaus) is employed o sig-
nal acquisi ion wi h a pe sonal compu e . A so wa e-based PID-con olle (see appendix
Fig. B.16) designed wi h P o iLab-Expe 3.0 (ABACOM Elec onics-So wa e) enables
use -de ined empe a u e p o iles and empe a u e- eco ding. The hea ing powe o he
oil hea e is p o ided by a emo e-con olled powe supply (TOE 8951, Toellne Elec-
onic Ins umen e GmbH) which is con olled ia he D/A-ou pu o he USB-module.
A es wi h a calib a ed P -100 esis o - b ough in he mal con ac wi h he ou side o he
mic oa ay subs a e - showed ha he empe a u e a he mic oa ay su ace is con olled
1The ubes - ou e diam. 1.2 mm in . 0.8 mm we e d awn h ough he 1 mm diam. inle /ou le
moun ing holes in he s eel pla e (→s able p ess- i -connec ion).
101
Mic oa ay Analysis
wi h an accu acy o app oxima ely ±1◦C. The empe a u e can be held cons an wi hin a
a ia ion o <0.2◦C.
5.1.2 Epi luo escence Mic oscope
The mic oa ay hyb idiza ion signal is acqui ed by epi luo escence mic oscopy (p inciple
shown in Fig. 5.4). Real ime moni o ing o he hyb idiza ion signal is pe o med wi h an
Figu e 5.4: Epi luo escence mic oscopy. The ligh o a b igh me cu y a c lamp (A)
is il e ed by he exci a ion il e (E) and e lec ed by he dich oic mi o (D) h ough
he mic oscope objec i e (M) on o he luo escen ly labeled sample (F). Fluo escen
dye molecules abso b he exci a ion ligh , and en e an exci ed elec onic s a e. Due
o he S okes shi he emi ed ligh has a longe wa eleng h han he exci a ion
ligh . The Cyanine 3 (Cy3) dye used h oughou his s udy has a peak abso p ion
a 550 nm (g een) and shows yellow o o ange luo escence emission (wi h a peak a
570 nm). A ac ion o he luo escence signal (emi ed in all di ec ions) is collec ed
by he mic oscope objec i e M and ansmi ed h ough he dich oic mi o (D). The
ba ie il e (B) passes only he luo escence ligh o he came a (C).
Olympus IX81 in e ed esea ch mic oscope (Fig.5.1). The luo escence o Cy3 labeled
a ge s is imaged using an UPlanApo 10×0.40 NA mic oscope objec i e (Olympus) and
he U-MWG 2 il e se (Olympus).
102
Quan i a i e Analysis o Mic oa ay Hyb idiza ion Signals
5.1.3 Image Acquisi ion wi h an EM-CCD Came a
High esolu ion image acquisi ion was pe o med wi h a sensi i e Hamama su EM-CCD
C9100-02 elec on mul iplying came a.
Came a speci ica ions:
•Pel ie cooling: -50◦C
•Gain ac o : 800
•Read-ou noise: <1 elec on .m.s. a high gain mode
•Dynamic ange: 14 bi
•Full esolu ion: 1000×1000 pixels
Came a and mic oscope (shu e , il e , exposu e, ocus, XY-s age e c. ) we e con olled
by he SimplePCI (Compix Inc.) image acquisi ion so wa e.
Shading co ec ion
Une en luo escence exci a ion and luo escence collec ion, owing o igne ing (la ge
blockage o o -axis ligh ays) yield luo escence mic og aphs ha a e b igh e a he cen-
e and da ke a he edges. In ensi y g adien s due o shading can be a signi ican sou ce
o e o o quan i a i e analysis o hyb idiza ion signals.
Shading co ec ion (using he SimplePCI se ing Ra io shade co ec ion) is he e o e pe -
o med by di iding he specimen image (mic oa ay) h ough a luo escence e e ence im-
age, which is acqui ed by imaging a uni o mly luo escen su ace. As desc ibed by Model
e al. [Mod01] spa ially uni o m luo escence is ob ained om a hin laye o luo escen
dye (e.g. 20 µl hyb idiza ion solu ion wi h 100 nM o Cy3 labeled a ge s) sandwiched
be ween a mic oscopy slide and a co e glass.
5.2 Quan i a i e Analysis o Mic oa ay Hyb idiza-
ion Signals
Fluo escence mic og aphs o he hyb idiza ion signal a e sa ed as 16-bi g ayscale TIFF
images. Shading co ec ion is pe o med du ing image acquisi ion.
Quan iza ion o ea u e in ensi ies is ca ied ou wi h he Ja a p og am ScanRA ( echni-
cal de ails in appendix B.10). The so wa e (which was de eloped as pa o his hesis)
enables au oma ic analysis o mic oa ay ea u e in ensi ies. To de ine ea u e posi ions a
103
Mic oa ay Analysis
Figu e 5.5: Raw hyb idiza ion signals as imaged wi h he Hamama su EM-CCD
came a (o iginal esolu ion o he image 1000×1000 pixel - size educed o 500×500).
Fo he image acquisi ion he 16me -mic oa ay emained in he hyb idiza ion solu ion
(1 nM Cy3-end-labeled RNA oligonucleo ide a ge ). The hyb idiza ion empe a u e
was 30◦C.
eadou g id (Fig. 5.6C) is placed on he mic oa ay image. Then he p og am in eg a es
pixel in ensi y alues o e he in eg a ion boxes loca ed a he g id poin s in cen e o he
ea u es. The size o he in eg a ion boxes should be chosen o p e en in eg a ion o e
ea u e bounda ies (Fig. 5.6D). The exac placemen o he eadou g id equi es o a ion
o he image, so ha he mic oa ay g id is app ox. aligned wi h he sc een axis. Consid-
e ing small image dis o ions an o hogonal g id o e enly spaced poin s is no sui able o
de e mine ea u es posi ions (Fig. 5.6A). Ra he , a quad ila e al g id (de ined by he ou
co ne poin s) is sui able o accoun o i s o de dis o ions o he mic oa ay image.
Mic oa ay hyb idiza ion signals (16-bi in ensi y alues) a e a e aged o e he in eg a-
ion boxes o p o ide a 16-bi mean in ensi y alue. The s anda d de ia ion o he pixel
in ensi y alues p o ides in o ma ion abou he homogenei y o he indi idual mic oa ay
ea u e in ensi ies. La ge s anda d de ia ions can indica e de ec s (e.g. luo escen pa i-
cles o sc a ches on he mic oa ay su ace) o bad alignmen o he eadou g id. A e age
b igh ness, s anda d de ia ion o he ea u e b igh ness and he posi ion o he indi idual
ea u es a e sa ed in comma-sepa a ed alue (CSV) o ma .
104
Real- ime Moni o ing o Mic oa ay Hyb idiza ion
A B
CD
Figu e 5.6: Mic oa ay analysis wi h ScanRA. Readou o he hyb idiza ion signal
in ensi ies o a ea u e block. (A) Ro a ion o he image. (B) An o hogonal eadou
g id doesn’ ma ch all ea u e posi ions exac ly i he a ay is sligh ly dis o ed. (C)
A quad ila e al eadou g id de ined by he ou co ne poin s is a good i s o de
app oxima ion o small image dis o ions. (D) In eg a ion boxes (blue) a e loca ed
a he g id poin s in he cen e o he ea u es.
Time se ies o luo escence mic og aphs can be analyzed in ba ch mode – he eadou g id
needs o be de ined only once. To accoun o d i ing o he image owing o he mal ex-
pansion o he hyb idiza ion chambe (i he empe a u e has been a ied signi ican ly), he
posi ion o he i s (uppe le ) co ne poin has o be p o ided manually o abou i e
images. The d i o se s o he o he images a e de e mined by linea in e pola ion.
5.3 Real- ime Moni o ing o Mic oa ay Hyb idi-
za ion
Mic oa ay hyb idiza ion is usually ollowed by one o se e al washing s eps o emo e
unhyb idized a ge s (see Fig. 2.22). Washing is necessa y o he de ec ion o small hy-
b idiza ion signals since hese a e o he wise no isible wi hin he luo escen backg ound
o he hyb idiza ion solu ion. This is ypically he case o exp ession p o iling expe imen s
whe e housands o di e en nucleic acid a ge s comp ise he hyb idiza ion solu ion.
Howe e , in mos o he expe imen s pe o med his s udy only a single a ge species is
con ained in he hyb idiza ion solu ion. A a a ge concen a ion o 1 nM he concen a ed
luo escence o he hyb idized a ge s (su ace-bound in he mic oscope ocal plane) can
be well-dis inguished om he backg ound luo escence o he hyb idiza ion solu ion.
105
Single Base De ec s - Mic oa ay Expe imen s
0 2 4 6 8 10 12 14 16
0
0.05
0.1
0.15
0.2
T T G A C T T T C G T T T C T G
De ec posi ion
Hyb idiza ion
signal (a.u.)
5’-AACTCGCTATAATGACCTGGACTG-Cy3-3’
3'-TATTACTGGACCTGAC-5’
Ta ge oligonucleo ide
P obe sequence mo i
(complemen a y o a
sec ion o he a ge )
Se o poin -mu a ed p obe sequences,
de i ed om common p obe sequence mo i
3'-T TTACTGGACCTGAC-5’
3'-T TTACTGGACCTGAC-5’
3'-T TTACTGGACCTGAC-5’
3'-T TTACTGGACCTGAC-5’
3'-TA TTACTGGACCTGAC-5’
3'-TA TTACTGGACCTGAC-5’
3'-TA TTACTGGACCTGAC-5’
3'-TA TTACTGGACCTGAC-5’
3'-T TTACTGGACCTGAC-5’
A
C
G
T
-
A
C
G
T
De ec posi ion 2
3'- ATTACTGGACCTGAC-5’
3'- ATTACTGGACCTGAC-5’
3'- ATTACTGGACCTGAC-5’
3'- ATTACTGGACCTGAC-5’
3'-T ATTACTGGACCTGAC-5’
3'-T ATTACTGGACCTGAC-5’
3'-T ATTACTGGACCTGAC-5’
3'-T ATTACTGGACCTGAC-5’
3'- ATTACTGGACCTGAC-5’
A
C
G
-
TA
C
G
T
Pe ec ma ch (PM) p obe
Single base misma ch (MM) p obes
Single base dele ion p obe
Single base inse ion p obes
De ec posi ion 1
De ec posi ion 1
De ec posi ion 2
Hyb idiza ion signals
De ec p o ile
Hyb idiza ion wi h he
Ta ge sequence
Fea u e a angemen on
he mic oa ay
}
}
Figu e 6.1: Design o he expe imen : a comp ehensi e se o poin -mu a ed p obes
is de i ed om a common p obe sequence mo i which is complemen a y o he a ge
sequence (p obe sequences a e shown o he i s wo de ec posi ions only). Fo each
de ec posi ion hese include 3 single base misma ches (MMs - shown in ed), 4 single
base inse ions (g een), one single base dele ion ( ed) and one pe ec ly ma ching
(PM) con ol p obe (blue). To enhance quan i a i e analysis, p obe se s a e a anged
on he mic oa ay as a compac ea u e block. Hyb idiza ion signal in ensi ies om
hyb idiza ion wi h he a ge sequence a e plo ed e sus de ec posi ion. The de ec
p o ile shows ela i e binding a ini ies (i.e. he disc imina ion be ween he de ec
hyb idiza ion signal and he co esponding PM hyb idiza ion signal) as a unc ion o
de ec ype and de ec posi ion.
112
DNA Mic oa ay Design
•ex ac ion o he de ec posi ional dependence,
•compa ison o he binding a ini ies o di e en de ec ypes, and on he
•iden i ica ion o u he in luen ial pa ame e s.
The indi idual chip designs employed di e in selec ion and spa ial a angemen o he
p obe sequences.
6.3.1 Mic oa ay Design Conside a ions o Quan i a i e Ana-
lysis o Hyb idiza ion A ini ies
Se e al ac o s a ec quan i a i e analysis o mic oa ay hyb idiza ion signals: Spa ial
a ia ions o he pho o-dep o ec ion in ensi y and op ical abe a ions a ec ing he imag-
ing con as can esul in g adien s o he p obe DNA quali y (as indica ed in Fig. 6.2B).
Depending on hei posi ion on he mic oa ay, p obes con ain a a ying deg ee o andom
syn hesis e o s. The co ne s o he ec angula syn hesis a ea a e mos a ec ed, since he e
he UV exposu e dose, due o igne ing, is signi ican ly smalle han in he cen e o he
syn hesis a ea.
G adien s on he luo escence in ensi y also a ise om op ical igne ing in he luo es-
cence mic oscope. This is la gely compensa ed by shading co ec ion (see sec ion 5.1.3).
To minimize impai men s by g adien s, p obes o which hyb idiza ion signals a e o be
compa ed di ec ly we e a anged in closely spaced ea u e blocks (as shown in Figs. 6.1
and 6.2).
Local a ge deple ion du ing hyb idiza ion (see sec ion 8.5) can likewise esul in posi ion-
dependen g adien s o he hyb idiza ion signal in ensi y. In ea u e blocks wi h iden ical
(o e y simila ) p obe sequences, owing o he compe i ion o he p obes o he same pool
o a ge s, ea u es in he cen e o he block (su ounded by 8 compe ing ea u es) - unde
un a o able hyb idiza ion condi ions [Pap06] - can ha e smalle hyb idiza ion signals han
equi alen ea u es a he edges o he ea u e block.
Con ol ea u es (comp ising pe ec ma ching p obes) which a e e enly dis ibu ed o e
he ea u e block, a e employed o indica e hyb idiza ion signal g adien s: he a ia ion o
he PM signals (e.g. in Fig. 6.4A) shows he magni ude o ea u e-posi ion dependen bias.
Usually he impai men o he hyb idiza ion signal by such g adien s is ela i ely small,
esul ing in a ia ion o he con ol-p obe in ensi ies which is ypically smalle han 5-10%
o he PM hyb idiza ion signal in ensi y. Howe e , i he hyb idiza ion kine ics is e y as
- hus incoming a ge s a e p e e en ially cap u ed by he p obes a he edge o he ea u e
block - spa ial a ia ions o he hyb idiza ion signal o up o 50% o he PM in ensi y can
occu [Pap06]. Un a o able condi ions a ec ing quan i a i e measu emen a e a oided by
113
Single Base De ec s - Mic oa ay Expe imen s
using ela i ely sho p obes ( a he 16me s han 25me s), using sequences wi h mode a e
binding a ini ies, and by applica ion o su icien ly s ingen hyb idiza ion condi ions.
6.3.2 Single Base De ec Expe imen s
A
C
G
T
A
C
G
T
A
C
G
T
A
C
G
T
A
C
G
T
A
C
G
T
A
C
G
T
A
C
G
T
A
C
G
T
A
C
G
T
A
C
G
T
A
C
G
T
A
C
G
T
A
C
G
T
A
C
G
T
A
C
G
T
246 8 10 12 14 16
911 13 15
135 7
AB1
16
Figu e 6.2: Mic oa ay ea u e a angemen (A) o he single base misma ch
expe imen (compa e wi h Fig. 6.3) and (B) o he di ec compa ison o a ious
de ec ypes. In A he ea u e block comp ises 16 MM posi ions. The subs i u ion
base is ei he A, C, G o T. Depending on he p obe sequence mo i he subs i u ions
will esul in one PM and h ee MM p obes. The design in Bincludes one PM, h ee
MM, ou single base inse ion and one single base dele ion p obe ou each o he
16 de ec posi ions. The 9 p obes belonging o each posi ion a e andomly a anged
in a 3×3 ma ix (depic ed by dashed boxes o de ec posi ions 1 and 16). In his
a angemen , as shown in (B), he g adien - ela ed a ia ion wi hin he closely spaced
3×3 ea u e g oup (belonging o a pa icula de ec posi ion) is signi ican ly smalle
han he a ia ion be ween ea u es (belonging o di e en de ec posi ions) which
a e loca ed u he apa .
Single base misma ches
To in es iga e he posi ional dependence o single base misma ches and he impac o he
misma ch ype, we designed mic oa ays con aining comp ehensi e se s o MM p obes
de i ed om a se ies o wen y- i e 16me p obe sequence mo i s. As desc ibed abo e,
posi ion and ype o he misma ch base pai we e sys ema ically a ied, allowing us la e
o dis inguish be ween he domina ing posi ional dependence and o he in luen ial ac o s.
The ea u es a e a anged in g oups o ou , co esponding o he ou possible subs i uen
bases (A, C, G and T) a a pa icula base posi ion. A g oup comp ises h ee misma ch
p obes plus one pe ec ma ch p obe used o con ol. Six een o hese ea u e g oups
(one o each base posi ion) a e a anged in a squa e ea u e block comp ising in o al 64
ea u es (Figs. 6.3 and 6.2A).
114
Hyb idiza ion Assays and Image Analysis
Single base bulges
P obes con aining single base inse ions and dele ions, owing o an unpai ed unpai ed nu-
cleo ide o m bulged duplexes (see Fig. 2.16) wi h educed s abili y. A comp ehensi e
s udy on he impac o single base inse ions was pe o med. The expe imen comp ised
abou 1000 single base inse ion p obes (inse ion base ype and posi ion sys ema ically
a ied) de i ed om wel e 20 o 25me p obe sequence mo i s. The ea u e a angemen
is simila o ha in Fig. 6.2A.
Di ec compa ison o single base MMs and single base bulges
P obe se s we e de i ed om 16me p obe sequence mo i s, complemen a y o he a ge s
in Table 6.1. Fo each o he 16 possible de ec posi ions a subse o 9 p obes (comp ising
ou single base inse ions, one base dele ion, h ee MMs and one PM p obe) has been
c ea ed. To p e en ha egula a angemen o he de ec ypes can c ea e a sys ema ic
bias on measu emen (e.g. due o inc eased a ge deple ion nea he PM p obes), he
subse s o 9 p obes we e andomly a anged in 3×3 ma ices as shown in Fig. 6.2B.
6.4 Hyb idiza ion Assays and Image Analysis
6.4.1 Oligonucleo ide Ta ge s
DNA and RNA a ge oligonucleo ides (Tab. 6.1) we e syn hesized by MWG Bio ech AG
(Ebe sbe g, Ge many) and by IBA Nucleic Acids Syn hesis (G¨o ingen, Ge many). 5’-Cy3
ma ke s we e a ached in he inal coupling s ep o he oligonucleo ide syn hesis ia cou-
pling o Cy3-phospho amidi e. The 3’-Cy3 modi ica ions we e p oduced pos syn he ically
by linkage o amino- eac i e NHS-es e s.
Gibbs ee ene gies ∆G◦
37 andmel ing empe a u esTmo he PM-duplexes(p edic ed wi h
he DINAMel se e - wo-s a e hyb idiza ion) a e p o ided in Tab. 6.2. Ta ge seconda y
s uc u e could no be a oided comple ely - in pa icula o he longe sequences and o
he mo e s able RNA sequences. Possible a ge oligonucleo ide seconda y s uc u e (loop
and hai pin o ma ion) was in es iga ed wi h he DINAMel Se e [Ma 05] (see Tab. 6.2).
6.5 Dominan In luence o he De ec Posi ion
The ”de ec p o ile” plo s (plo s o he no malized hyb idiza ion signal s. de ec posi ion
- e.g. in Figs. 6.4 and 6.15) show ha he dominan pa ame e de e mining oligonucleo ide
p obe- a ge -a ini y - on he mic oa ay su ace - is he posi ion o he de ec .
115
Single Base De ec s - Mic oa ay Expe imen s
Table 6.1: Fluo escen ly labeled DNA and RNA a ge oligonucleo ides
Name Ta ge sequence (5’→3’) Label Leng h (n )
URA DNA ACTACAAACTTAGAGTGCAG... 5’-Cy3 38
...CAGAGGGGAGTGGAATTC
NIE DNA ACTCGCAAGCACCACCCTATCA 3’-Cy3 22
LBE DNA GTGATGCTTGTATGGAGGAA... 3’-Cy3 30
...TACTGCGATT
PET DNA ACATCAGTGCCTGTGTACTAGGAC 3’-Cy3 24
BEI DNA ACGGAACTGAAAGCAAAGAC 3’-Cy3 20
COM DNA AACTCGCTATAATGACCTGGACTG 5’-Cy3 24
NCO DNA TAGTGGGAGTTGTTAGTGATGTGA 3’-Cy3 24
PET RNA ACAUCAGUGCCUGUGUACUAGGACA 5’-Cy3 25
LBE RNA GUGAUGCUUGUAUGGAGGAA 5’-Cy3 34
...UACUGCGAUUCGAU
COM RNA AACUCGCUAUAAUGACCUGGACUG 5’-Cy3 24
Table 6.2: Gibbs ee ene gies and mel ing empe a u es o PM duplexes and a ge
seconda y s uc u es (DINAMel se e [Ma 05]), T=37◦C, [Na+]= M, s and con-
cen a ion 1 nM. The a ge s COM (DNA) and NCO don’ o m ele an seconda y
s uc u es. Fo RNA/DNA duplexes no da a on duplex s abili y is a ailable (NDA).
PM duplex Ta ge seconda y s uc u e
Ta ge Duplex ∆G◦
37 in Tm∆G◦
37 in Tm
name ype kcal/mol in ◦C kcal/mol in ◦C
URA DNA/DNA -48.1 77.5 -0.1 40.1
NIE DNA/DNA -29.2 67.1 0.5 27.6
LBE DNA/DNA -36.6 70.7 -1.16 45.3
LBE RNA/DNA NDA NDA -7.1 63.0
PET DNA/DNA -29.6 66.1 -1.23 54.5
PET RNA/DNA NDA NDA -1.23 54.5
BEI DNA/DNA -24.2 59.6 0.08 35.1
COM DNA/DNA -28.7 64.5 - -
COM RNA/DNA NDA NDA -0.1 37.4
NCO DNA/DNA -28.7 65.1 - -
116
Dominan In luence o he De ec Posi ion
A
T
G
C
A
T
G
C
1
2
Figu e 6.3: Fluo escence mic og aph o wo neighbo ing ea u e blocks in he 16me
misma ch expe imen . The shading-co ec ed image shows wo ea u e blocks co e-
sponding o wo di e en 16me p obe sequence mo i s (3’-TTGAGCGATATTACTG-
5’ o he le , and 3’-TATTACTGGACCTGAC-5’ o he igh ) bo h hyb idiz-
ing wi h he luo escen ly labeled a ge sequence COM (5’-Cy3-AACTCGCTATA-
ATGACCTGGACTG-3’). The di e en hyb idiza ion signal in ensi ies o he wo
ea u e blocks a e owing o di e en binding a ini ies o he wo p obe sequence
mo i s. The ea u e size is 21 µm. Each ea u e block comp ises all single base mis-
ma ches ha can occu in he co esponding p obe sequence mo i . G oups o ou
ea u es (as indica ed by he ma ked g oups 1 and 2) co espond o each one o he
16 possible misma ch base posi ions. As indica ed by he le e s be ween he ea u e
blocks he uppe mos ow o ea u es in each g oup co esponds o an A base a he
co esponding base posi ion, ollowed by p obes wi h C, G and T (see also Fig. 6.2).
The b igh es ea u e wi hin each g oup co esponds o he pe ec ma ching p obe.
Nonhyb idized a ge s in he hyb idiza ion solu ion con ibu e o he backg ound in-
ensi y be ween he ea u es. The ”misma ch de ec p o ile” o he p obe sequence
mo i 3’-TATTACTGGACCTGAC-5’ is shown in Fig. 6.4.
De ec s nea he duplexends a e dis inc ly less des abilizing han de ec s in he cen e o he
duplex. As shown in Fig. 6.4 he hyb idiza ion signals o he indi idual misma ch p obes
a e lined-up along he ough-like ”mean p o ile” cu e (solid black line). A pa abolic i
can p o idea easonable app oxima ion o he a e age posi ion dependence ob ained om
a la ge numbe o di e en sequence mo i s (as shown in [Wic06; Poz06]). The disc imi-
na ion be ween PM and MM hyb idiza ion signals is la ges i he de ec is loca ed in he
middle o he duplex. Fo 16me duplexes (as shown in Fig. 6.4) a single base misma ch
(MM) in he cen e ypically yields 0-40% o he pe ec ma ch (PM) hyb idiza ion signal,
whe eas a he duplex ends de ec s ha e signi ican ly less impac on he hyb idiza ion sig-
nal.
The disc imina ion be ween PM and poin -mu a ed p obes depends on he s abili y o he
pa icula p obe sequence mo i : The mo e s able 25me p obes (shown in Fig. 6.15) a e
less disc imina i e han he sho e 16me p obes (Figs. 6.4A and 6.19). Reduced dis-
c imina ion is also obse ed (see Fig. 6.5) o sequences which a e s abilized by a high
CG-con en .
117
Single Base De ec s - Mic oa ay Expe imen s
0 5 10 15
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
Misma ch base posi ion
Hyb idiza ion
signal (a.u.)
0 5 10 15
-0.05
0
0.05
Misma ch base posi ion
De ia ion om
he mean (a.u.)
0 5 10 15
T A T T A C T G G A C C T G A C
Misma ch base posi ion
A
B
Chighe
lowe
Figu e 6.4: The misma ch de ec p o ile (A) (hyb idiza ion signal e sus de ec
base posi ion) was ob ained om he analysis o he hyb idiza ion signals o he
ea u e block shown in he igh pa o Figu e 6.3. The p obe sequence mo i 3’-
TATTACTGGACCTGAC-5’ is complemen a y o he a ge oligonucleo ide COM.
The di e en ypes o base subs i u ions a e highligh ed by di e en ma ke s (A ed
c osses; C g een ci cles; G blue s a s; T cyan iangles). The black line indica es
he mean p o ile (mo ing a e age o all misma ch hyb idiza ion signals o e posi ions
p−2 o p+ 2). PM p obes (g ey symbols) a e used as a con ol o de ec sys ema ic
bias (g adien e ec s) on he hyb idiza ion signal. The a ia ion o he PM p obe
in ensi ies also p o ides an es ima e o he e o o he measu emen . Bias- ela ed
de ia ions be ween dis an ea u es, owing o g adien e ec s, a e expec ed o be
la ge han he e o s be ween he compac ly a anged ea u es co esponding o
he same de ec posi ion. (B) De ia ion p o ile. The s ong posi ion dependen
componen o he hyb idiza ion signal was elimina ed by sub ac ion o he mean
p o ile. In he ollowing he hyb idiza ion signal de ia ion om he mean p o ile
is e e ed o as δImp.(C) Compa ison o mean misma ch hyb idiza ion signals
(a e age o he h ee misma ch hyb idiza ion signals a a pa icula de ec posi ion)
a he si es o C·G base pai s o mean MM hyb idiza ion signals a he si e o adjacen
A·T base pai s. A ma ke ( ed s a : A·T; blue ci cle C·G) is se in he uppe ow i
he hyb idiza ion signals o he misma ches a he co esponding si e is highe han
ha a he adjacen si e; o he wise a ma ke is se in he lowe ow. We no iced ha
misma ched base pai s subs i u ing a C·G base pai usually ha e sys ema ically lowe
hyb idiza ion signals han misma ches subs i u ing a neighbo ing A·T base pai .
118
Dominan In luence o he De ec Posi ion
The posi ional in luence obse ed in he mean p o iles is la gely de e mined by he de ec -
o-end dis ance, bu is supe imposed by a sequence dependen con ibu ion. The a ia ion
o he shapes o he mean inse ion p o iles in Fig. 6.5 indica es ha he impac o a de ec
is a ec ed by he s abili y o he local sequence en i onmen (i.e. no only by he nex
nea es neighbo base pai s). We disco e ed ha single base bulge de ec s, o igina ing
om single base inse ions (Fig. 6.15) and dele ions (Fig. 6.19) - wi hin he indi idual
de ec p o iles - display he same posi ional dependence as single base misma ch de ec s.
An a emp o explain he o igin o de ec posi ional in luence is made in sec ion 7.
To in es iga e o he ac o s in luencing oligonucleo ide duplex binding a ini y (e.g. de ec
ype and de ec neighbo hood) he domina ing posi ional in luence needs o be elimina ed.
Design (selec ion and a angemen o p obes) and analysis o ou expe imen s enable sep-
a a ion o he di e en in luen ial ac o s.
0 5 10 15 20 25 30 35 40
0
0.2
0.4
0.6
0.8
1
T G A T G T T T G A A T C T C A C G T C G T C T C C C C T C A C C T T A A G
Inse ion base posi ion
Hyb idiza ion signal (a.u.)
2
134
Figu e 6.5: The impac o de ec s is a ec ed by he local sequence en i onmen .
No malized single base inse ion p o iles (hyb idiza ion signal plo ed e sus he in-
se ion base posi ion) o ou 25me p obe sequence mo i s complemen a y o he
same a ge sequence (URA - shown below). The p obe mo i s 1 o 4 hyb idize o di -
e en sec ions o he a ge oligonucleo ide. Mean p o iles (bold lines) we e ob ained
om he mo ing a e age o he pa icula inse ion p o iles (indi idual hyb idiza ion
signals a e shown as ain g ey symbols - p o ile 4 is shown in de ail in Figu e 6.15A).
The mean p o iles 1 o 3 ha e a dis inc minimum be ween base posi ions 15 o 20.
The s abilizing CG- ich egion be ween base posi ions 15 and 33 is he eason o he
educed MM disc imina ion in p o ile 4.
Discussion
We obse e a domina ing in luence o he de ec posi ion on duplex binding a ini y. De-
ec s loca ed in he cen e o he oligonucleo ide duplexes a e signi ican ly mo e des abi-
lizing han de ec s a he ends.
119
Single Base De ec s - Mic oa ay Expe imen s
S ong in luence o MM posi ion has been epo ed p e iously mainly by o he mic oa ay
based s udies, bu also om hyb idiza ion expe imen s in solu ion.
•F om op ical mel ing s udies (on 7me RNA/RNA duplexes in solu ion) Kie zek e
al. [Kie99] epo a 0.5 kcal/mol s abiliza ion inc emen pe each base posi ion ha
he de ec is close o he helix end. A posi ional in luence was obse ed o U·U and
A·A, whe eas he G·G misma ch s abili y was la gely una ec ed by he posi ion.
•Do is e al. [Do 03] ound a simila posi ional in luence o 2-base MM and 3-base
MM p obes on CodeLink 3D gel a ays. They also epo a s ong co ela ion (in-
cluding he posi ional in luence) be ween solu ion-phase mel ing empe a u es and
mic oa ay hyb idiza ion signals o he MM duplexes.
•Mo e ecen ly Wick e al. [Wic06] and Pozhi ko e al. [Poz06] epo ed a s ong
in luence o he de ec posi ion on he binding a ini y o single base MM duplexes
on DNA mic oa ays.
In acco dance wi h [Poz06] we ha e iden i ied MM posi ion ( ela i e o he duplex ends)
as he s onges in luen ial ac o on he hyb idiza ion signal, when compa ed o MM- ype
(de e mined by he misma ch base pai X·Y) and nea es neighbo s.5
To ou knowledge only wo s udies ([Kie99] and [Do 03]) epo a de ec posi ional in lu-
ence o hyb idiza ion in solu ion. This may pa ly be due o he una ailabili y o a la ge
numbe o o app op ia e p obes o a sys ema ic s udy. So a he s ong posi ional in lu-
ence, mos ly obse ed in mic oa ay expe imen s, is unexplained.
The obse a ion o a s ong posi ion dependence is in con lic wi h he wo-s a e nea es -
neighbo model o DNA duplex he mal s abili y, whe e he he modynamics o in e nal
misma ches is ea ed as independen o he MM posi ion [San04]. Also, oligonucleo ide
duplex s abili y p edic ion so wa e (based on a mul i-s a e model) unde es ima es he MM
posi ional in luence when compa ed o mic oa ay hyb idiza ion assays [Wic06].
Fo single base bulge de ec s we obse ed a e y simila posi ion dependence as o single
base misma ches. Also, he magni udes o he impac s o he MMs and base bulges on
he hyb idiza ion signal a e e y simila (apa om he ela i e high binding a ini y o
G oup II bulges). This consis ency sugges s a common o igin o he posi ional in luence,
expec ed o be independen o he de ec ype.
S e ical c owding a he su ace as discussed by Pe e son e al. [Pe 02] could possibly
in oduce a posi ional dependence on he hyb idiza ion signals o de ec p obes. Reduced
accessibili y o he p obes su ace-bound 3’-ends can in p inciple dec ease he impac o
5Acco ding o he nea es -neighbo model he lanking base pai s owa ds bo h sides o he misma ched
base pai X·Y– jus like he misma ched base pai i sel – de e mine he base s acking in e ac ions.
120
Misma ch Disc imina ion in DNA/DNA Duplexes
de ec s loca ed nea he 3’-end, and hus esul in inc eased hyb idiza ion signals o he
co esponding p obes. This, howe e , uns con a y o he la gely symme ical in ensi y
p o iles obse ed (Fig. 6.4) and he e o e does no p o ide a sa is ac o y explana ion o
he in luence o de ec posi ion.
Focusing on indi idual p obe sequence mo i s, we obse e, ha he posi ional in luence
is no simply a unc ion o he de ec - o-end dis ance: i a he has a sequence-dependen
con ibu ion. This indica es ha he misma ch disc imina ion could be a ec ed by he s a-
bili y o he nea es neighbo pai s be ween he de ec and he p oxima e duplex end.
The obse ed in luence o he duplex sequence and he symme y o he de ec posi ional
in luence wi h espec o bo h duplex ends sugges ha end-domain opening (i.e. sequen ial
unzipping o he double-helix om he duplex ends) is he key mechanism o unde s and-
ing he in luence o de ec posi ion on duplex s abili y.
6.6 Misma ch Disc imina ion in DNA/DNA Du-
plexes
6.6.1 Expe imen al Resul s
Fo s a is ical analysis o MM ype and nea es -neighbo in luences he supe imposed po-
si ional in luence needs o be elimina ed. This is achie ed by sub ac ion o he (mo ing
a e age) mean p o ile. In he ollowing he hyb idiza ion signal de ia ion om he mean
p o ile is e e ed o as δImp. The esul ing posi ion-independen de ec p o ile ( o sim-
plici y we keep using he exp ession ”de ec p o ile”) comp ising de ec - ype and lanking
base pai in luences only, is shown in Fig. 6.4B.
In he ollowing we use he no a ion o he misma ch base pai X·Yconsis ing o he mis-
ma ched base Xin he p obe sequence and he base Yin he a ge sequence. In ou
expe imen s he sys ema ic a ia ion was es ic ed o he bases Xin he mic oa ay p obe
sequences. Since we had only a limi ed se o luo escen ly labeled a ge oligonucleo ides
a ailable (see Tab. 6.1) - he a ge sequences wi h he bases Y emained unchanged.
To in es iga e how he pa icula MM- ypes X·Ya ec duplex s abili y we measu ed p obe-
a ge -a ini ies o 25 di e en p obe sequence mo i s (dis ibu ed o e h ee di e en mi-
c oa ays). The PM hyb idiza ion signals o he 16me p obe sequence mo i s display a
s ong a ia ion (up o a ac o o 20). The absolu e hyb idiza ion signals om di e en
p obe se s a e he e o e no di ec ly compa able. Howe e , since he ela i e in ensi ies
(o he a ious MM p obes) wi hin he p obe se s a e la gely una ec ed by his a ia ion,
we can no malize he ”posi ion-independen de ec p o iles” by di ision by hei s anda d
121
Single Base De ec s - Mic oa ay Expe imen s
mu a ions he sequence o he guide s and was p ese ed) pu ine-pu ine MMs esul ed
in he leas silencing o gene ac i i y, whe eas U·G, C·U and U·U misma ches esul ed in
a e y e icien gene silencing (see Fig. 6.8c).8”A a o ed model is ha pu ine-pu ine
misma ches dis up RISC ac i i y by p e en ing he o ma ion o a con en ional A- o m
helix be ween he guide s and and he a ge mRNA, a s uc u al equi emen o RISC-
media ed clea age” [RL06]. In e es ingly, he epo ed educed s abili y o pu ine-pu ine
misma ches is in good ag eemen wi h he indings o Pozhi ko e al.. Howe e , he in-
e ed RNA/RNA MM s abili y o de in Fig. 6.8c, like ha in in [Sug00], is no no malized
wi h espec o he co esponding PM s abili ies, bu a he e lec s he absolu e impac o
he MM base pai s in a gi en duplex sequence.
Di e ences be ween MM disc imina ion in DNA/DNA hyb idiza ion and RNA/DNA hy-
b idiza ion a e no su p ising since DNA/DNA duplexes (unde he expe imen al condi-
ions employed) occu as B- o m helices, whe eas RNA/DNA and RNA/RNA duplexes
commonly occu as A- o m helices (see Fig. 2.5).
The appa en disc epancy be ween he s abili y o de s in he s udies discussed abo e (see
Fig. 6.8)mo i a eda sys ema iccompa isono singlebaseMMdisc imina ioninDNA/DNA
and RNA/DNA duplexes (see sec ion 6.8).
6.7 In luence o Flanking Base Pai s on Single Base
Misma ch Binding A ini ies in DNA/DNA Mi-
c oa ay Hyb idiza ion
Due o s acking in e ac ions he des abilizing impac o a misma chde ec no only depends
on he MM base pai X·Y, bu also on he lanking Wa son-C ick base pai s A·Aand B·B
on bo h sides o he de ec . [Alk82; Sug86].
50−A Y B −30
30−A XB −50
Fo a sys ema ic s udy o he nex -nea es -neighbo in luence he misma ch hyb idiza ion
signal da a was ca ego ized no only acco ding o he he misma ch ype (as discussed in
sec ion 6.6), bu also acco ding o he lanking base pai s a bo h sides o he misma ched
base pai .
8[Sch06]: ”Misma ches o be well accommoda ed in an A- o m RNA/RNA helix (py imi-
dine:py imidine, py imidine:pu ine, o pu ine:py imidine) displayed in e media e le els o disc imi-
na ion, whe eas pu ine:pu ine misma ches, expec ed ei he o des abilize he helix o o p omo e a
s able, bu nonhelical, con o ma ion, silenced he epo e leas .”
128
In luence o Flanking Base Pai s
The e a e 16 neighbo hood classes (combina ions o A·Aand B·B) o each o he 12
misma ch ypes X·Y.
-1
-0.5
0
0.5
1
1.5
Misma ch base pai X .Y
De ia ion om he mo ing a e age p o ile (a.u.)
5'-TYT-3'
3'-AXA-5'
5'-TYG-3'
3'-AXC-5'
5'-TYC-3'
3'-AXG-5'
5'-TYA-3'
3'-AXT-5'
5'-GYT-3'
3'-CXA-5'
5'-GYG-3'
3'-CXC-5'
5'-GYC-3'
3'-CXG-5'
5'-GYA-3'
3'-CXT-5'
5'-CYT-3'
3'-GXA-5'
5'-CYG-3'
3'-GXC-5'
5'-CYC-3'
3'-GXG-5'
5'-CYA-3'
3'-GXT-5'
5'-AYT-3'
3'-TXA-5'
5'-AYG-3'
3'-TXC-5'
5'-AYC-3'
3'-TXG-5'
5'-AYA-3'
3'-TXT-5'
AA CA GA TA AC CC GC TC AG CGGGTG AT CT GT TT
Figu e 6.9: Dis ibu ion o he median hyb idiza ion signal alues (de ia ion om
he mo ing a e age p o iles) o he a ious MM neighbo hoods classes (see legend) as
shown in Figs. A.12 - A.22. Red symbols deno e C·G neighbo s only, blue symbols
deno e A·T neighbo s only. G een symbols co espond o mixed neighbo s. The max-
imum alue o abou 1.3 a.u. o A·G MMs (g een up-poin ing iangle) is p obably
an ou lie (only a single measu emen was a ailable o ha pa icula MM class),
whe eas he alue o 0.74 ( ed s a ) is based on 10 measu emen s. The signi icance
o indi idual da a poin s (which can be a ec ed by lack o expe imen al da a) can be
e alua ed om he co esponding his og ams in Figs. A.12 - A.22.
Spli ing o he expe imen al da a in o 192 subse s (see Figs. A.12 - A.22) esul s in a
ela i ely small s a is ical base o he indi idual MM classes (→la ge s a is ical e o s and
sequence dependen bias). The signi icance o indi idualda a poin s can be e alua ed om
he co esponding his og ams in Figs. A.12 - A.22.
The median alues o he neighbo hood-dependen MM hyb idiza ion signal9dis ibu ions
a e shown in Fig. 6.9. To in es iga e i he expe imen ally obse ed in luence o lank-
9no malized hyb idiza ion signals, posi ional in luence elimina ed
129
Single Base De ec s - Mic oa ay Expe imen s
AA CA GA TA AC CC GC TC AG CG GG TG AT CT GT TT
-5
-4
-3
-2
-1
0
1
2
3
AA CA GA TA AC CC GC TC AG CG GG TG AT CT GT TT
0
1
2
3
4
5
6
δ ∆G (kcal/mol)
Base pai X.Y
Base pai X.Y
∆G (kcal/mol)
5'-TYT-3'
3'-AXA-5'
5'-TYG-3'
3'-AXC-5'
5'-TYC-3'
3'-AXG-5'
5'-TYA-3'
3'-AXT-5'
5'-GYT-3'
3'-CXA-5'
5'-GYG-3'
3'-CXC-5'
5'-GYC-3'
3'-CXG-5'
5'-GYA-3'
3'-CXT-5'
5'-CYT-3'
3'-GXA-5'
5'-CYG-3'
3'-GXC-5'
5'-CYC-3'
3'-GXG-5'
5'-CYA-3'
3'-GXT-5'
5'-AYT-3'
3'-TXA-5'
5'-AYG-3'
3'-TXC-5'
5'-AYC-3'
3'-TXG-5'
5'-AYA-3'
3'-TXT-5'
A
B
37 37
Figu e 6.10: In luence o lanking base pai s on MM duplex s abili y. (A) Gibbs ee
ene gies ∆G◦
37 o misma ched and pe ec -ma ching DNA/DNA inucleo ide duplexes
we e calcula ed om MM nea es -neighbo pa ame e s [All97]. C·G lanking base
pai s ( ed ma ke s) a e consis en ly s abilizing, whe eas A·T lanking base pai s (blue
ma ke s) ha e a des abilizing in luence. (B) Gibbs ee ene gy inc emen s δ∆G◦
37
be ween MM and co esponding PM duplexes. In he wo-s a e nea es -neighbo
model he disc imina ion be ween single base MM and PM duplexes only depends on
he iden i y o he a ec ed inucleo ide sequence (MM base pai and lanking base
pai s). δ∆G◦
37 does no depend on he es o he duplex sequence o on he posi ion
o he de ec (unless he de ec is loca ed a a e minal posi ion).
130
In luence o Flanking Base Pai s
0123456
-1
-0.5
0
0.5
1
1.5
Hyb idiza ion signal (a.u.)
δ ∆G37 (kcal/mol)
0123456
-1
-0.5
0
0.5
1
1.5
Hyb idiza ion signal (a.u.)
δ ∆G37 (kcal/mol)
5'-TYT-3'
3'-AXA-5'
5'-TYG-3'
3'-AXC-5'
5'-TYC-3'
3'-AXG-5'
5'-TYA-3'
3'-AXT-5'
5'-GYT-3'
3'-CXA-5'
5'-GYG-3'
3'-CXC-5'
5'-GYC-3'
3'-CXG-5'
5'-GYA-3'
3'-CXT-5'
5'-CYT-3'
3'-GXA-5'
5'-CYG-3'
3'-GXC-5'
5'-CYC-3'
3'-GXG-5'
5'-CYA-3'
3'-GXT-5'
5'-AYT-3'
3'-TXA-5'
5'-AYG-3'
3'-TXC-5'
5'-AYC-3'
3'-TXG-5'
5'-AYA-3'
3'-TXT-5'
AA
CA
GA
AC
CC
TC
AG
GG
TG
CT
GT
TT
Misma ch
Base Pai
Flanking
Base Pai s
A
B
Figu e 6.11: Compa ison o MM hyb idiza ion signals (no malized wi h espec o PM hy-
b idiza ion signals - hus ep esen ing a measu e o MM disc imina ion) wi h p edic ed Gibbs
ee ene gy inc emen s δ∆G◦
37. Hyb idiza ion signals (as shown in Fig. 6.9) a e ca ego ized
acco ding o MM base pai ype and acco ding o lanking base pai s. Each da a poin ep e-
sen s he median alue o a dis ibu ion o hyb idiza ion signals (in de ail shown in Figs. A.12
o A.22). We obse e a signi ican co ela ion be ween he MM hyb idiza ion signal and he
p edic ed Gibbs ee ene gy inc emen δ∆G◦
37. Pa (A) highligh s he in luence o lanking
base pai s on MM disc imina ion. Flanking A·T base pai s on bo h sides o he de ec (blue
symbols) esul (on a e age) in smalle hyb idiza ion signals han C·G-only ( ed symbols) o
mixed lanking base pai s (g een symbols). Howe e , he in luence o lanking base pai s is
li le consis en compa ed wi h he in luence o he MM base pai ype, which is highligh ed
in (B): The disc imina ion o G·G, A·A and T·G misma ches is la ge han p edic ed by MM
nea es -neighbo pa ame e s om [All97] and la ge han in a simila expe imen in [Wic06].
131
Single Base De ec s - Mic oa ay Expe imen s
ing base pai s on binding a ini ies is in ag eemen wi h MM nea es -neighbo pa ame e s
[All97], we compa ed ou expe imen al da a (Fig. 6.9) o p edic ed ee ene gy inc emen s
be ween MM and PM duplexes (Fig. 6.10): The MM nea es -neighbo pa ame e s om
[All97] p edic a s abilizing in luence o C·G lanking base pai s. Fig. 6.10A shows a
consis en ly inc eased s abili y o hose duplexes wi h C·G nex nea es neighbo s only,
whe eas a sys ema ically dec eased s abili y is seen o duplexes wi h A·T nea es neigh-
bo s only. Fo he p edic ed di e ence δ∆G◦
37 be ween PM and MM ee ene gies - which
is expec ed o be e lec ed in he expe imen ally de e mined MM disc imina ion - his con-
sis ency is somewha educed (see Fig. 6.10B). The compa ison o δ∆G◦
37 wi h expe i-
men ally de e mined hyb idiza ion signals in Fig. 6.11A con i ms a signi ican in luence
o lanking base pai s. On a e age, lanking A·T base pai s esul in smalle hyb idiza ion
signals han C·G o mixed lanking base pai s. Howe e , he in luence o he MM-base
pai s X·Yon he MM binding a ini y (see Fig. 6.11B) is dis inc ly mo e consis en han
he in luence o lanking base pai ypes. A la ge scale in es iga ion o lanking base pai
in luence (based on a much la ge se o oligonucleo ide a ge sequences/p obe sequence
mo i s) would be necessa y o inc ease he s a is ical signi icance o he abo e esul s.
6.8 Misma ch Disc imina ion in DNA/DNA and
RNA/DNA Duplexes - a Di ec Compa ison
To in es iga e i he abo e esul s om DNA/DNA hyb idiza ion also apply o hyb idiza-
ion o RNA/DNA duplexes we pe o med a di ec compa ison be ween DNA/DNA hy-
b idiza ion and RNA/DNA hyb idiza ion (employing DNA a ge s and equi alen RNA
a ge sequences - see Tab. 6.1 ) on he same mic oa ay.
6.8.1 Ou line o he Expe imen
The expe imen is basically iden ical wi h he expe imen s desc ibed in sec ion 6.6. Hy-
b idiza ion assays a e conduc ed wi h luo escen ly labeled DNA a ge s and co esponding
RNA a ge sequences (Table 6.1). To a oid ab ica ion- ela ed a ia ion o he hyb idiza-
ion signals he DNA and RNA hyb idiza ion assays we e pe o med on he same chip, i s
wi h RNA a ge oligonucleo ides and - a e egene a ion o he mic oa ay wi h NaOH
(selec i e deg ada ion o RNA a ge s) - wi h he co esponding DNA a ge s.
Th ee di e en mic oa ays we e ab ica ed, each one ocussing on one pa icula a ge
sequence (COM,PET and LBE). The indi idual mic oa ays comp ise single base MM
and inse ion p obes (→single base bulges) o 6 di e en p obe sequence mo i s (p obing
132
Misma ch Disc imina ion in DNA/DNA and RNA/DNA Duplexes
di e en 16 o 20me subsequences o he a ge sequence).
Two eplica es o each ea u e block p o ide a es o he ep oducibili y o he mea-
su emen . The subse s o da a ob ained om he indi idual mic oa ays we e analyzed
independen ly o check he consis ency o he obse ed esul s: apa om small sequence-
ela ed biases he h ee mic oa ays p o ided basically he same esul s. Hyb idiza ion was
pe o med wi h 1 nM a ge solu ions in 5×SSPE (0.01% Tween-20TM). Hyb idiza ion
empe a u es we e 30◦C o PET and LBE and 40◦C o COM ( o he a ge sequence
COM he empe a u e had o be inc eased o 40◦Csince local deple ion led o inhomoge-
neous hyb idiza ion - see sec ion 8.5).
6.8.2 Resul s
The in luence o he de ec posi ion is e y simila o he DNA/DNA and he RNA/DNA
binding a ini ies (see Fig. A.1). Howe e , he e a e small, hough ep oducible di e -
ences, as he compa ison be ween eplica e ea u e blocks (see Figs. A.2 - A.7) shows. Fo
single base bulges no de ec ype speci ic di e ences be ween RNA/DNA and DNA/DNA
hyb idiza ion we e ound.
We obse ed ha unde equi alen hyb idiza ion condi ions he hyb idiza ion signal om
RNA a ge s is on a e age abou 1.3 imes b igh e han ha o he co esponding DNA
a ge s. This is an icipa ed: RNA a ge s ha e a sligh ly la ge binding a ini y han DNA
a ge s since s acking in e ac ions a e s onge in A- o m RNA/RNA and RNA/DNA du-
plexes han in B-DNA duplexes.10
Di e ences be ween MM s abili ies in DNA/DNA and RNA/DNA du-
plexes
The MM disc imina ion in RNA/DNA duplexes (Fig. 6.12B) is e y simila o ha in
DNA/DNA duplexes (Fig. 6.12A). Howe e , a close look e eals sys ema ic di e ences
be ween DNA/DNA and RNA/DNA hyb idiza ion. A s a is ical analysis (Figs. 6.12 and
6.14) e ealed ha pu ine-pu ine MMs a e less s able in RNA/DNA duplexes (Fig. 6.14c)
han in DNA/DNA duplexes (Fig. 6.14b). Th ee independen expe imen s (pe o med on
di e en mic oa ays and wi h di e en p obe/ a ge sequences) p o ided he same ends.
The dec ease o pu ine-pu ine MM s abili ies becomes ob ious in he anking o de o
di e ences be ween RNA/DNA and DNA/DNA MM s abili ies (Fig. 6.14d). The la ges
di e ences be ween RNA/DNA and DNA/DNA MMs a e obse ed o he MM- ypes G·A
and A·G (which a e mo e s able in DNA/DNA duplexes) and, wi h e e sed sign, o he
MM- ype T·G, which is signi ican ly mo e s able in RNA/DNA duplexes.
10 Binding a ini ies: RNA/RNA >RNA/DNA >DNA/DNA
133
Single Base De ec s - Mic oa ay Expe imen s
AA
AC
AG
CA
CC
CT
GA
GG
GT
TC
TG
TT
DNA/DNA misma ch hyb idiza ion signal
-0.5 0 0.5 1 1.5
AA
AC
AG
CA
CC
CU
GA
GG
GU
TC
TG
TU
Hyb idiza ion signal (a.u.)
RNA/DNA misma ch hyb idiza ion signal
A
B-0.5 0 0.5 1 1.5
Hyb idiza ion signal (a.u.)
Figu e 6.12: Compa ison o DNA/DNA and RNA/DNA misma ch hyb idiza ion
signals - s a is ical analysis. (A) MM- ype ela ed in luence in DNA/DNA oligonu-
cleo ide duplexes. The posi ional in luence was elimina ed by sub ac ion o he
mo ing a e age MM p o ile. Subsequen no maliza ion was pe o med by di ision
h ough he mean hyb idiza ion signal o he pa icula MM p o ile. (B) MM- ype
ela ed in luence in RNA/DNA oligonucleo ide duplexes.
134
Misma ch Disc imina ion in DNA/DNA and RNA/DNA Duplexes
-1 -0.5 0 0.5 1
AA
AC
AG
CA
CC
CT
GA
GG
GT
TC
TG
TT
Hyb idiza ion signal (a.u.)
Di e ences be ween MM hyb idza ion signals
o RNA/DNA and DNA/DNA duplexes
Figu e 6.13: Di e ences be ween RNA/DNA and DNA/DNA MM binding a ini ies.
La ges di e ences be ween RNA/DNA and DNA/DNA ha e been ound o he MM-
ypes T·G, G·A and A·G.
6.8.3 Discussion
Ou in es iga ion on he impac o MM- ypes in DNA/DNA oligonucleo ide duplexes e-
ealed ha single base misma ches subs i u ing C·G base pai s a e mo e des abilizing han
misma ches subs i u ing A·T base pai s.
Howe e , his seemingly plausible esul (shown in Fig. 6.6) is no in gene al ag eemen
wi h p e ious wo k [Sug00; Wic06; Poz06; Sch06] on he in luence o he MM ype on
binding a ini ies.
Ou di ec compa ison (”di ec ” in he sense o using he same p obe sequences on he
same mic oa ay) be ween DNA/DNA and RNA/DNA hyb idiza ion on mic oa ays e-
eals - o RNA/DNA duplexes- an inc eased des abiliza ion o pu ine-pu ine misma ches,
wi h espec o o he MM ypes. Howe e , we did no obse e such a dis inc impac o
pu ine-pu ine MMs as epo ed in [Poz06] and [Sch06]. Ra he he MM s abili y o de was
e y simila o ha o DNA/DNA hyb idiza ion.
F om MM s abili yo de s in Figs. 6.14c and 6.14b (and Fig. 6.8e) we in e ha he s abili y
o MMs in RNA/DNA duplexes is de e mined by wo ac o s:
•In RNA/DNA duplexes pu ine-pu ine MMs end o be mo e des abilizing (wi h espec
135
Single Base De ec s - Mic oa ay Expe imen s
T U T U U T⋅ ≥ > ⋅ ≈ ⋅ ≈ ⋅ ≈ ⋅ ≈ ⋅ ≈ > ⋅ ≥ ≥ ⋅ >C C C C C CG A G G A A A A A G G G⋅ ⋅ ⋅ ⋅
T C C C T T C T T C T C⋅ > ⋅ ≥ ⋅ ≥ ⋅ > ⋅ ≈ ⋅ > ⋅ ≈ ⋅ ≥ ≈ > >G A G A A A G G A G G A⋅ ⋅ ⋅ ⋅
G A G A G A A A G A G G⋅ ⋅ ⋅ ⋅> ⋅ ≥ ⋅ > ≥ ⋅ ≈ ⋅ > ≈ ⋅ ≈ ⋅ ≈ ⋅ ≥ ⋅ ≥T T T C T C T C C T C C
G A A G G A A A A G G G⋅ ⋅ ⋅ ⋅> ⋅ > ≈ ⋅ ≈ ⋅ ≈ > ⋅ > ⋅ ≥ ⋅ ≥ ⋅ ≈ ⋅ >T T C T T C C T T C C C
a) DNA/DNA hyb idiza ion (la ge da a se )
b) DNA/DNA hyb idiza ion (small da a se o di ec
compa ison wi h RNA/DNAhyb idiza ion)
c) RNA/DNA hyb idiza ion (small da a se - equi alen
o he DNA/DNA da ase in b)
d) Di e ence be ween RNA/DNA and DNA/RNA
hyb idiza ion signals. U acil is ea ed as hymine.
(TG o GT posi i e; AC o GA nega i e)
Figu e 6.14: Ranking o de s o DNA/DNA MM s abili ies in compa ison wi h ha
o RNA/DNA MMs. (a) Fo compa ison he DNA/DNA MM s abili y o de om
an independen expe imen (Fig. 6.8) is shown he e again. (b) As an icipa ed he
anking o de o DNA/DNA MMs ob ained om he smalle da a se which is used
o he di ec compa ison be ween DNA/DNA and RNA/DNA hyb idiza ion (Fig.
6.12A) is e y simila . The anking o de o RNA/DNA misma ch s abili ies (c)
(ex ac ed om Fig. 6.12B) e eals signi ican di e ences wi h espec o (b). In pa
(d) MM- ypes a e o de ed acco ding o he hyb idiza ion signal di e ences be ween
RNA/DNA and DNA/DNA MMs (ex ac ed om Fig. 6.12 A and B). Pu ine bases
a e highligh ed in blue.
o o he MM- ypes) han pu ine-pu ine MMs in DNA/DNA duplexes.
•The in luence o he ”a ec ed base pai ” - he base pai which has been subs i u ed by
he MM base pai - is he o he ac o ha de e mines he impac o he MM ype. In he
expe imen s he PM hyb idiza ion signal is used as a e e ence alue o he educ ion
o he hyb idiza ionsignal due he MM de ec . In ag eemen wi h [Wic06] we obse ed
ha MMs a ec ing C·G base pai s a e mo e disc imina ing han MMs a ec ing A·T
base pai s.
In he o de o RNA/DNA misma ch s abili ies (Fig. 6.14c) he la e e ec is supe im-
posed by he des abilizing e ec o pu ine-pu ine MMs, whe eas in DNA/DNA duplexes
(Fig. 6.14b - ou esul s - in ag eemen wi h [Wic06] - see Fig. 6.8d) an inc eased des abi-
liza ion o pu ine-pu ine MMs is no obse ed.
An explana ion o he obse ed di e ences be ween DNA/DNA and RNA/DNA binding
a ini ies is, ha pu ine-pu ine MMs cause la ge s e ic hind ance in he A- o m hyb id
duplexes han in he B- o m DNA/DNA duplexes.
In his s udy, like in [Poz02], a des abilizing impac o pu ine-pu ine MMs was obse ed
in RNA/DNA hyb idiza ion. Howe e , we ound only a sligh ly inc eased des abiliza ion
wi h espec o he co esponding pu ine-pu ine MMs in DNA/DNA duplexes, whe eas
[Poz02] and [Sch06] epo ed ha pu ine-pu ine MMs - in absolu e e ms - a e he mos
136
Single Base Bulge De ec s
disc imina ing MMs wi h espec o o he MM- ypes.11 Fu he s udies will be necessa y
o esol e he emaining disc epancy.
A mo e de ailed u u e in es iga iono MM s abili ies should also ocus on he in luence o
he lanking base pai s. This, howe e , will equi e a signi ican ly la ge da abase o MM
hyb idiza ion signals.
6.9 Single Base Bulge De ec s
Single base inse ions and dele ions, owing o a su plus unpai ed base in one o he wo
s ands, esul in bulged duplexes, which like MM duplexesha e a educed binding a ini y.
In duplexes wi h single base inse ion p obes he bulged base is loca ed on he su ace-
bound p obe s and, whe eas in duplexes wi h single base dele ion p obes he bulged base
is loca ed on he a ge s and.
The posi ional dependence o he inse ion in ensi y p o iles (Figu e 6.15A) is e y simila
o he misma ch in ensi y p o ile in Figu e 6.4, hough he indi idual inse ion p o iles ( o
example he p o ile o C-inse ions - g een ci cles in Figu e 6.15) show la ge de ia ions
om he (mo ing a e age) mean p o ile.
Hyb idiza ion signals can be signi ican ly inc eased o e wo o mo e consecu i e de ec
posi ions. In pa icula , base inse ions nex o iden ical bases (G oup II bulges [Zhu99])
esul in sys ema ically inc eased binding a ini ies - in compa ison o inse ions o non-
iden ical bases (G oup I bulges). In he no a ion o Zhu e al. [Zhu99] bulged bases wi hou
an iden ical neighbo ing base (Fig. 2.17A) a e de ined as G oup I bulges, whe eas bulges
wi h a leas one iden ical neighbo ing base (Fig. 2.17B) a e e e ed o as G oup II bulges.
Inc eased s abili y o duplexes wi h G oup II bulges in solu ion-phase expe imen s has
been desc ibed by Ke e al. [Ke95]. Fig. 6.15C demons a es he sys ema ically inc eased
binding a ini y o G oup II bulges in DNA mic oa ay hyb idiza ion.
6.9.1 S a is ical Analysis
The obse ed s abiliza ion o G oup II bulges (in compa ison o G oup I bulges) in ou mi-
c oa ay expe imen s is su p isingly la ge (see discussion below): G oup II bulges loca ed
nea he cen e o 16me p obes o en show hyb idiza ion signals wi h a simila in ensi y
as he co esponding PM p obe, whe eas G oup I bulges a he same de ec posi ion ha e a
signi ican ly smalle binding a ini y, wi h a simila le el as single base MMs a he co e-
11 These s udies, howe e , in es iga ed only DNA/RNA hyb idiza ion and RNA/RNA hyb ids (RNAi:
A- o m helix be ween he guide s and and he a ge mRNA), espec i ely. No compa ison wi h
DNA/DNA hyb idiza ion was made.
137
Summa y/Zusammen assung
allein au g und de ge ing ¨ugigen S abilisie ung in olge diese En opiezunahme zu e -
kl¨a en. Unse E kl¨a ungsansa z be uh au eine du ch den bulge-De ek e u sach en Blo-
ckade des Zippe -Mechanismus: Die du ch den bulge-De ek he o ge u ene Ve schiebung
zwischen den Einzels ang-Sequenzen ( ameshi ) e hinde ein schnelles Schließen (zip-
ping up) des Duplex. Diese Ba ie e kann beim Vo liegen eines G oup II bulges – au g und
de Posi ionsen a ung – schnelle ¨ube sp ungen we den18 als bei G oup I bulge-De ek en
(bei welchen keine Posi ionsen a ung o lieg ).
Die Bindungsa ini ¨a zwischen P obe- und Ta ge -Sequenzen wi d seh s a k on de Se-
kund¨a s uk u de Ta ge -Sequenzen beein luss [Lue03]. F¨u ein Expe imen zu Un e -
suchung des Ein lusses solche Sekund¨a s uk u en (Abschni 8.6), wu den luo eszenz-
ma kie e cRNA-Ta ge s mi eine L¨ange on 300 bzw. 800 Nukleo iden he ges ell . Bei
diesen L¨angen sind s abile in amolekula e Sekund¨a s uk u en zu e wa en, die in den
dazugeh¨o igen Sequenzabschni en eine Hyb idisie ung mi komplemen ¨a en Mic oa ay-
P obes e hinde n. Ta s¨achlich konn e in dem iling-a ay-Expe imen 19 nu au e wa 20
bis 30% de L¨ange diese Ta ge -Sequenzen eine signi ikan e Hyb idisie ung e ziel we -
den.
Mi Hil e on S old [Din04], einem So wa e-Tool welches u. a. zum Au inden e ek i e
An isense Oligonukleo idedien , wu de un e such , wiesich die in olge de Sekund¨a s uk-
u e minde e Zug¨anglichkei on g oßen Teilen de Ta ge sequenz au die Bindungsa i-
ni ¨a de einzelnen P obesequenzen auswi k . Unse e E gebnisse zeigen, dass die mi Hil e
on S old au heo e ische G undlage (un e Be ¨ucksich ung des Bol zmann-Ensembles
on Ta ge -Sekund¨a s uk u en) e mi el en Bindungsa ini ¨a en mi unse en expe imen-
ell bes imm en Hyb idisie ungssignalen ko elie sind. Unse e E gebnisse legen nahe das
S old auch zum Au inden e izien e Mic oa ay-P obe-Sequenzen geeigne is . Wei e e
Mic oa ay-Hyb idisie ungsexpe imen e mi ande en Ta ge -Sequenzen sind e o de lich
um die im Rahmen de o liegenden A bei gewonnenen E gebnisse zu un e maue n.
Au de Basis des double-ended Zippe -Modells [Gib59; Ki 69] wu de ein he modyna-
misches Modell des Oligonukleo id-Duplexes en wickel (Kapi el 7), um die expe imen-
ellen E gebnisse, inbesonde e den s a ken Ein luss de De ek posi ion, genaue zu un e -
suchen. Im Gegensa z zum in de P axis am h¨au igs en e wende en wo-s a e nea es -
18 Die S abilisie ung on G oup II bulges be uh de e h¨
oh en Wah scheinlichkei , dass eine de iden-
ischen Basen eine g¨
uns ige Kon o ma ion einnimm , bei de ein asches Fo sch ei en des Zipping-
P ozesses m¨
oglich is .
19 Das iling-a ay-Expe imen beinhal e einen Sa z on 25me P obe-Sequenzen die en lang de seh
iel l¨
ange en Ta ge -Sequenz ela i zueinande e se z angeo dne sind. Diese A on Expe imen
e olg den Zweck, die Bindungsa ini ¨
a de einzelnen Ta ge -Be eiche zu sondie en.
240
Zusammen assung
neighbo Modell we den beim Zippe Modell auch die an den Enden pa iell dena u ie -
en Duplexkon o ma ionen be ¨ucksich ig . Ausgehend on den nea es -neighbo Wechsel-
wi kungen benachba e Basenpaa e we den ¨u die einzelnen Duplexkon o ma ionen die
s a is ischen Gewich e und da aus schließlich die Zus andssumme be echne . Die heo e i-
schen Be ach ungen zeigen, dass die Zus andssumme beim Vo liegen on Einzelde ek en
umso g ¨oße is , je n¨ahe de De ek bei den Duplexenden lieg . Dies bes ¨a igen die expe-
imen ellen E gebnisse: Oligonukleo id-Duplexe mi endnahen De ek en sind s abile als
en sp echende Duplexe mi in de Mi e liegenden De ek en. Eine nume ische Analyse des
De ek -Posi ionsein lussesau die Bindungsa ini ¨a zeig , dass die Oligonukleo idsequenz,
in diesem Fall als Ab olge un e schiedliche s a ke nea es -neighbo -Wechselwi kungen
be ach e , wie bei auch expe imen ell beobach e , einen signi ikan en Ein luss au die Po-
si ionsabh¨angigkei de Bindungsa ini ¨a haben kann. Dies wi d o allem o ensich lich,
wenn inne halb de Duplex-Sequenz s ¨a ke e und schw¨ache e NN-Paa e ungleichm¨aßig
e eil sind.
Um die expe imen ell bes imm en Hyb idisie ungssignale mi den au heo e ische Ba-
sis e mi el en Duplexs abili ¨a en e gleichen zu k¨onnen wu de in einem Mic oa ay-Hy-
b idisie ungsexpe imen (Abschni 7.4) die L¨ange de P obes – und somi die Gibbs-
Ene gie ∆Gde DNA-Duplexe – sch i weise a iie . Wi beobach en einen sigmoidalen
Zusammenhang θ(∆G)zwischen dem An eil hyb idisie e P obes und de eien En -
halpie de Duplexe ∆G.¨
Ube einen ela i wei en ¨
Ube gangsbe eich nimm das Hyb i-
disie ungssignal n¨ahe ungsweise linea mi de eien En halpie de Duplexe zu. Dami
weich das expe imen elle E gebnis deu lich on einem heo e ischen Ve lau ab, de du ch
die Langmui -Adso p ionsgleichungbesch ieben wi d - diese weis einen e gleichsweise
schmalen ¨
Ube gangsbe eich au . Die Disk epanz konn e anhand eine nume ischen Simu-
la ion mi dem Ein luss on Syn hesede ek en e kl¨a we den: Die in den Expe imen en
o liegende b ei e Ve eilung on Bindungsa ini ¨a en, die du ch eine a iable Anzahl on
De ek en in de P obe-Sequenz he o ge u en wi d (die sich zudem an un e schiedlichen
Posi ionen be inden), esul ie in einem s a k e b ei e en ¨
Ube gangsbe eich in θ(∆G).
Die un e such e Posi ionsabh¨angigkei on De ek en kann auch au die meh ode we-
nige s a ken NN-Wechselwi kungen on Wa son-C ick-Basenpaa en ¨ube agen we den.
Unse e Un e suchungen in Abschni 7.5 zeigen: Duplexe, die aus iden ischen NN-Paa en
zusammengese z , und somi au de G undlage des wo-s a e nea es -neighbo Modell
he modynamisch ¨aqui alen sind, weisen im Zippe -Modell die g ¨oß e S abili ¨a dann au ,
wenn die s abils en NN-Paa e in de Mi e des Duplex und die schw¨achs en NN-Paa e en -
sp echend an den Enden des Duplexes angeo dne sind. Bei Raum empe a u sind die E -
gebnisse des Zippe -Modells mi denen des wo-s a e nea es -neighbo Modells p ak isch
241
Summa y/Zusammen assung
iden isch. E s mi zunehmende Tempe a u is in olge de e s ¨a k en Dena u ie ung an
den Duplexenden die besch iebene Posi ionsabh¨angigkei zu beobach en. Dieses E gebnis
lie e e s mals eine heo e ische G undlage ¨u das bislang nu au empi ische Basis be-
sch iebene posi ionsabh¨angige nea es -neighbo Modell (PDNN).
Im Rahmen de o liegenden A bei wu de au de Basis on handels¨ublichen Kompo-
nen en ein lexibles Sys em zu in si u-Syn hese on DNA-Mic oa ays en wickel . Au -
g und seine echnischen M¨oglichkei en(bei e gleichsweise nied igen In es i ionen),abe
auch weil es im Gegensa z zu komme ziellen Mic oa ay-Pla o men keine Black-Box-
Technologie da s ell , d¨u e das hie im De ail besch iebene Sys em eine in e essan e
Ausgangsbasis ¨u die En wicklung on Mic oa ay-Syn hesize n sein. Eine (e l. au ei-
ne ”Open Sou ce”-Basis be iebene) Wei e en wicklung des Mic oa ay-Syn hesesys ems
w¨a e w¨unschenswe , dami diese iel e sp echende und ielsei ig einse zba e Zukun s-
echnologie bald b ei e Anwendung inden kann.
In Hinblick au die zunehmende Bedeu ung de DNA-Mic oa ay Technologie is ein un-
die es Ve s ¨andnis de zug unde liegenden physikalisch-chemischen Zusammenh¨ange e -
o de lich. Vo allem in Hinblick au die Un e suchungen zu De ek ion on Punk mu a-
ionen wu de in de o liegenden A bei dazu beige agen.
242
Summa y/Zusammen assung
244
Bibliog aphy
[AB03] G. Al an-Bonne , A. Libchabe , and O. K iche sky. Bubble dynamics in double-
s anded DNA. Physical Re iew Le e s, 90(13):138101, Ap il 2003.
[Alb03] T. J. Albe , J. No on, M. O , T. Richmond, K. Nuwaysi , E. F. Nuwaysi ,
K. P. S engele, and R. D. G een. Ligh -di ec ed 5 ’- 3 ’ syn hesis o complex
oligonucleo ide mic oa ays. Nucleic Acids Resea ch, 31(7):e35, Ap il 2003.
[Alk82] D. Alkema, P. A. Hade , R. A. Bell, and T. Neilson. E ec s o lanking GC base-
pai s on in e nal wa son-c ick, GU, and nonbonded base pai s wi hin a sho
ibonucleic-acid duplex. Biochemis y, 21(9):2109–2117, 1982.
[All97] H. T. Allawi and J. San aLucia. The modynamics and NMR o in e nal GT
misma ches in DNA. Biochemis y, 36(34):10581–10594, Augus 1997.
[Amb05] T. Ambjo nsson and R. Me zle . Blinking s a is ics o a molecula beacon ig-
ge ed by end-dena u a ion o DNA. Jou nal o Physics-Condensed Ma e ,
17(49):S4305–S4316, 2005.
[Amb06] T. Ambjo nsson, S. K. Banik, O. K iche sky, and R. Me zle . Sequence sensi-
i i y o b ea hing dynamics in he e opolyme DNA. Physical Re iew Le e s,
97(12):128105, Sep embe 2006.
[And06] D. And ea a, S. Sen, J. L. P. Lus es, S. A. Ko alenko, N. P. E ns ing, C. J.
Mu phy, R. S. Coleman, and M. A. Be g. Ul a as dynamics in DNA: ” aying”
a he end o he helix. Jou nalo he Ame ican Chemical Socie y, 128(21):6885–
6892, May 2006.
[App65] J. Applequis and V. Damle. The modynamics o helix-coil equilib ium in
oligoadenylicacid om hypoch omici ys udies. Jou nal o he Ame ican Chem-
ical Socie y, 87(7):1450–&, 1965.
[Ba 06] A. Ba hel and M. Zacha ias. Con o ma ional ansi ions in na single u idine
and adenosine bulge s uc u es: A molecula dynamics ee ene gy simula ion
s udy. Biophysical Jou nal, 90(7):2450–2462, Ap il 2006.
[Bau03] M. Baum, S. Bielau, N. Ri ne , K. Schmid, K. Eggelbusch, M. Dahms,
A. Schlaue sbach, H. Tahedl, M. Beie , R. Guimil, M. Sche le , C. He mann,
J. M. Funk, A. Wixme en, H. Rebsche , M. Honig, C. And eae, D. Buchne ,
245
BIBLIOGRAPHY
E. Moschel, A. Gla he, E. Jage , M. Thom, A. G eil, F. Bes a e , F. Obe -
meie , J. Bu gmaie , K. Thome, S. Weiche , S. Hein, T. Binnewies, V. Foi zik,
M. Mulle , C. F. S ahle , and P. F. S ahle . Valida ion o a no el, ully in eg a ed
and lexible mic oa ay bench op acili y o gene exp ession p o iling. Nucleic
Acids Resea ch, 31(23):e151, 2003.
[Bea81] S. L. Beaucage and M. H. Ca u he s. Deoxynucleoside phospho amidi es a new
class o key in e media es o deoxypolynucleo ide syn hesis. Te ahed on Le -
e s, 22(20):1859–1862, 1981.
[Bei99] M. Beie and J. D. Hoheisel. Ve sa ile de i a isa ion o solid suppo media o
co alen bonding on DNA-mic ochips. Nucleic Acids Resea ch, 27:1970–1977,
1999.
[Ben02] R. Ben e s, C. M. Niemeye , D. D u schmann, D. Blohm, and D. Woh le. DNA
mic oa ays wi h PAMAM dend i ic linke sys ems. Nucleic Acids Resea ch,
30(2):e10, Janua y 2002.
[Bha03] G. Bhano , Y. Louzoun, J. H. Zhu, and C. DeLisi. The impo ance o he mo-
dynamic equilib ium o high h oughpu gene exp ession a ays. Biophysical
Jou nal, 84(1):124–135, Janua y 2003.
[Bin04] H. Binde , T. Ki s en, M. Loe le , and P. F. S adlle . Sensi i i y o mic oa ay
oligonucleo ide p obes: Va iabili y and e ec o base composi ion. Jou nal o
Physical Chemis y B, 108(46):18003–18014, 2004.
[Bin06] H. Binde . The modynamics o compe i i esu ace adso p ion on DNA mic oa -
ays. Jou nal o Physics-Condensed Ma e , 18(18):S491–S523, 2006.
[Bla96] A. P. Blancha d, R. J. Kaise , and L. E. Hood. High-densi y oligonucleo ide
a ays. Biosenso s & Bioelec onics, 11(6-7):687–690, 1996.
[Blo03] R. Blossey and E. Ca lon. Repa ame izing he loop en opy weigh s: E ec on
DNA mel ing cu es. Physical Re iew E, 68(6):061911, Decembe 2003.
[B e86] K. J. B eslaue , R. F ank, H. Blocke , and L. A. Ma ky. P edic ing DNA du-
plex s abili y om he base sequence. P oceedings o he Na ional Academy o
Sciences o he Uni ed S a es o Ame ica, 83(11):3746–3750, June 1986.
[Cam06] A. M. Caminade, C. Padie, R. Lau en , A. Ma a al, and J. P. Majo al. Uses o
dend ime s o DNA mic oa ays. Senso s, 6(8):901–914, Augus 2006.
[Ca 06] E. Ca lon and T. Heim. The modynamics o RNA/DNA hyb idiza ion in high-
densi y oligonucleo ide mic oa ays. Physica A-S a is ical Mechanics and i s
Applica ions, 362(2):433–449, Ap il 2006.
[Cha05] C. Y. Chan, C. E. Law ence, and Y. Ding. S uc u e clus e ing ea u es on he
s old web se e . Bioin o ma ics, 21(20):3926–3928, Oc obe 2005.
246
BIBLIOGRAPHY
[Che07] W. W. Chen, S. Ki iha a, and Y. Miyamo o. Fab ica ion o h ee-dimensional
mic o pho onic c ys als o esin-inco po a ing TiO2 pa icles and hei e ahe z
wa e p ope ies. Jou nal o he Ame ican Ce amic Socie y, 90(1):92–96, Jan-
ua y 2007.
[Chi05] P. Y. Chiou, A. T. Oh a, and M. C. Wu. Massi ely pa allel manipula ion o
single cells and mic opa icles using op ical images. Na u e, 436(7049):370–
372, 2005.
[Cog91] J. A. H. Cogne , J. Gaba oa pa, M. Leb e , G. A. Vande ma el, J. H. Vanboom,
and G. V. Fazake ley. Solu ion con o ma ion o an oligonucleo ide con aining
a GG misma ch de e mined by nuclea -magne ic- esonance and molecula me-
chanics. Nucleic Acids Resea ch, 19(24):6771–6779, Decembe 1991.
[Con83] B. J. Conne , A. A. Reyes, C. Mo in, K. I aku a, R. L. Tepli z, and R. B. Wal-
lace. De ec ion o sickle-cell be a-s-globin allele by hyb idiza ion wi h syn he ic
oligonucleo ides. P oceedings o he Na ional Academy o Sciences o he Uni ed
S a es o Ame ica, 80(1):278–282, 1983.
[C a71] M. E. C aig, D. M. C o he s, and P. Do y. Relaxa ion kine ics o dime o -
ma ion by sel complemen a y oligonucleo ides. Jou nal o Molecula Biology,
62(2):383–&, 1971.
[C i70] F. C ick. Cen al dogma o molecula biology. Na u e, 227(5258):561–&, 1970.
[C o64] D. M. C o he s and B. H. Zimm. Theo y o mel ing ansi ion o syn he ic
polynucleo ides: E alua ion o s acking ee ene gy. Jou nal o Molecula Biol-
ogy, 9(1):1–&, 1964.
[Cue04] J. A. Cues a and A. Sanchez. Gene al non-exis ence heo em o phase an-
si ions in one-dimensional sys ems wi h sho ange in e ac ions, and physical
examples o such ansi ions. Jou nal o S a is ical Physics, 115(3-4):869–893,
May 2004.
[Dan07] D. S. Dandy, P. Wu, and D. W. G ainge . A ay ea u e size in luences nucleic
acid su ace cap u e in DNA mic oa ays. P oceedings o he Na ional Academy
o Sciences o he Uni ed S a es o Ame ica, 104(4):8223–8228, Feb ua y 2007.
[Dau93] T. Dauxois, M. Pey a d, and A. R. Bishop. Dynamics and he modynamics o
a nonlinea model o DNA dena u a ion. Physical Re iew E, 47(1):684–695,
Janua y 1993.
[De 05] G. De a, H. Moench, E. Fische , H. Giese, U. Hech ische , G. Hensle , A. Ko-
e be , U. Niemann, F. C. Noe emann, P. Peka ski, J. Pollmann-Re sch, A. Ri z,
and U. Weichmann. Uhp lamp sys ems o p ojec ion applica ions. Jou nal o
Physics D-Applied Physics, 38(17):2995–3010, Sep embe 2005.
[Deu04] J. M. Deu sch, S. Liang, and O. Na ayan. Modelling o mic oa ay da a wi h
zippe ing. P ep in q-bio.BM/0406039 1, 2004. a Xi :cond-ma /0304567.
247
BIBLIOGRAPHY
[Din01] Y. Ding and C. E. Law ence. S a is ical p edic ion o single-s anded egions in
RNA seconda y s uc u e and applica ion o p edic ing e ec i e an isense a ge
si es and beyond. Nucleic Acids Resea ch, 29(5):1034–1046, Ma ch 2001.
[Din03] Y. Ding and C. E. Law ence. A s a is ical sampling algo i hm o RNA sec-
onda y s uc u e p edic ion. Nucleic Acids Resea ch, 31(24):7280–7301, De-
cembe 2003.
[Din04] Y. Ding, C. Y. Chan, and C. E. Law ence. S old web se e o s a is ical olding
and a ional design o nucleic acids. Nucleic Acids Resea ch, 32:W135–W141,
July 2004.
[Din05] Y. Ding, C. Y. Chan, and C. E. Law ence. RNA seconda y s uc u e p edic ion
by cen oids in a bol zmann weigh ed ensemble. RNA-A Publica ion o he RNA
Socie y, 11(8):1157–1166, Augus 2005.
[Dod77] J. B. Dodgson and R. D. Wells. Syn hesis and he mal mel ing beha iou o
oligome -polyme complexes con aining de ined leng hs o misma ched da.dg
nucleo ides. Biochemis y, 16(11):2367–2374, 1977.
[Do 03] D. R. Do is, A. Nguyen, L. Giese , R. Lockne , A. Lublinsky, M. Pa e son,
E. Touma, T. J. Sende a, R. Elghanian, and A. Mazumde . Oligodeoxy ibonu-
cleo ide p obe accessibili y on a h ee-dimensional DNA mic oa ay su ace and
he e ec o hyb idiza ion ime on he accu acy o exp ession a ios. BMC
Bio echnology, 3:6, 2003.
[E e07] R. E e ae s, S. Kuma , and C. Simm. Uni ied desc ip ion o poly- and oligonu-
cleo ide DNA mel ing: Nea es -neighbo , poland-she aga, and la ice models.
Physical Re iew E, 75:041918, 2007.
[Fin72] T. R. Fink and D. M. C o he s. F ee-ene gy o impe ec nucleic-acid helices 1.
bulge de ec . Jou nal o Molecula Biology, 66(1):1–&, 1972.
[Fod91] S. P. A. Fodo , J. L. Read, M. C. Pi ung, A. T. S ye , L.and Lu, and D. So-
las. Ligh -di ec ed, spa ially add essable pa allel chemical syn hesis. Science,
251(4995):767–773, 1991.
[F e86] S. M. F eie , R. Kie zek, J. A. Jaege , N. Sugimo o, M. H. Ca u he s, T. Neilson,
and D. H. Tu ne . Imp o ed ee-ene gy pa ame e s o p edic ions o RNA
duplex s abili y. P oceedings o he Na ional Academy o Sciences o he Uni ed
S a es o Ame ica, 83(24):9373–9377, Decembe 1986.
[Gao01] X. L. Gao, E. LeP ous , H. Zhang, O. S i anna i , E. Gula i, P. L. Yu,
C. Nishiguchi, Q. Xiang, and X. C. Zhou. A lexible ligh -di ec ed DNA chip
syn hesis ga ed by dep o ec ion using solu ion pho ogene a ed acids. Nucleic
Acids Resea ch, 29(22):4744–4750, 2001.
[Gao04] X. L. Gao, E. Gula i, and X. C. Zhou. In si u syn hesis o oligonucleo ide mi-
c oa ays. Biopolyme s, 73(5):579–596, Ap il 2004.
248
BIBLIOGRAPHY
[Ga 02] P. B. Ga land and P. J. Se a inowski. E ec s o s ay ligh on he ideli y o pho-
odi ec ed oligonucleo ide a ay syn hesis. Nucleic Acids Resea ch, 30(19):e99,
Oc obe 2002.
[Gib59] J. H. Gibbs and E. A. Dima zio. S a is ical mechanics o helix-coil ansi ions in
biological mac omolecules. Jou nal o Chemical Physics, 30(1):271–282, 1959.
[Gil77] D. T. Gillespie. Exac s ochas ic simula ion o coupled chemical- eac ions. Jou -
nal O Physical Chemis y, 81(25):2340–2361, 1977.
[Gla06] M. Glaze , J. A. Fidanza, G. H. McGall, M. O. T ulson, J. E. Fo man, A. Suseno,
and C. W. F ank. Kine ics o oligonucleo ide hyb idiza ion o pho oli hog aphi-
cally pa e ned DNA a ays. Analy ical Biochemis y, 358(2):225–238, No em-
be 2006.
[Go 81] O. Go oh and Y. Tagashi a. S abili ies o nea es -neighbo double s in double-
helical DNA de e mined by i ing calcula ed mel ing p o iles o obse ed p o-
iles. Biopolyme s, 20(5):1033–1042, 1981.
[Gue87] M. Gue on, M. Kochoyan, and J. L. Le oy. A single-mode o DNA base-pai
opening d i es imino p o on-exchange. Na u e, 328(6125):89–92, July 1987.
[Gu 05] Z. Gu enbe g, H. Mulle , H. Habe mulle , A. Geisbaue , J. Pippe , J. Felbel,
M. Kielpinski, J. Sc iba, and A. Wix o h. Plana chip de ice o pc and hy-
b idiza ion wi h su ace acous ic wa e pump. Lab On A Chip, 5(3):308–317,
2005.
[Hag88] P. J. Hage man. Flexibili y o DNA. Annual Re iew o Biophysics and Biophys-
ical Chemis y, 17:265–286, 1988.
[Hal04] A. Halpe in, A. Buho , and E. B. Zhulina. Sensi i i y, speci ici y, and he
hyb idiza ion iso he ms o DNA chips. Biophysical Jou nal, 86(2):718–730,
Feb ua y 2004.
[Hal05] A. Halpe in, A. Buho , and E. B. Zhulina. B ush e ec s on DNA chips: he mo-
dynamics, kine ics, and design guidelines. Biophysical Jou nal, 89(2):796–811,
Augus 2005.
[Has97] A. Hasan, K. P. S engele, H. Gieg ich, P. Co nwell, K. R. Isham, R. A. Sach-
leben, W. P leide e , and R. S. Foo e. Pho olabile p o ec ing g oups o nucleo-
sides: Syn hesis and pho odep o ec ion a es. Te ahed on, 53(12):4247–4264,
1997.
[Hel03] G. A. Held, G. G ins ein, and Y. Tu. Modeling o DNA mic oa ay da a by using
physical p ope ies o hyb idiza ion. P oceedings o he Na ional Academy o
Sciences o he Uni ed S a es o Ame ica, 100(13):7575–7580, June 2003.
[Hel06] G. A. Held, G. G ins ein, and Y. Tu. Rela ionship be ween gene exp ession
and obse ed in ensi ies in DNA mic oa aysa modeling s udy. Nucleic Acids
Resea ch, 34(9):e70, 2006.
249
BIBLIOGRAPHY
[Sin84] N. D. Sinha, J. Bie na , J. Mcmanus, and H. Kos e . Polyme suppo oligonu-
cleo ide syn hesis .18. use o be a-cyanoe hyl-n,n-dialkylamino-/n-mo pholino
phospho amidi e o deoxynucleosides o he syn hesis o DNA agmen s sim-
pli ying dep o ec ion and isola ion o he inal p oduc . Nucleic Acids Resea ch,
12(11):4539–4557, 1984.
[Ske93] J. V. Skelly, K. J. Edwa ds, T. C. Jenkins, and S. Neidle. C ys al-s uc u e o
an oligonucleo ide duplex con aining G.G base-pai s in luence o mispai ing on
DNA backbone con o ma ion. P oceedings o he Na ional Academy o Sciences
o he Uni ed S a es o Ame ica, 90(3):804–808, Feb ua y 1993.
[Sou75] E.M. Sou he n. De ec ion o speci ic sequences among DNA agmen s sepa-
a ed by gel elec opho esis. J Mol Biol., 98:503–517, 1975.
[Sug86] N. Sugimo o, R. Kie zek, S. M. F eie , and D. H. Tu ne . Ene ge ics o in e -
nal GU misma ches in ibooligonucleo ide helixes. Biochemis y, 25(19):5755–
5759, Sep embe 1986.
[Sug95] N. Sugimo o, S. Nakano, M. Ka oh, A. Ma sumu a, H. Nakamu a, T. Ohmichi,
M. Yoneyama, and M. Sasaki. The modynamic pa ame e s o p edic s abili y
o RNA/DNA hyb id duplexes. Biochemis y, 34(35):11211–11216, Sep embe
1995.
[Sug00] N. Sugimo o, M. Nakano, and S. Nakano. The modynamics-s uc u e ela ion-
ship o single misma ches in RNA/DNA duplexes. Biochemis y, 39(37):11270–
11281, 2000.
[Sun00] M. Sunda alingam and Y. Xiong. C ys al s uc u e o domain II o x-lae is
soma ic 5s RNA in wo con o ma ions. Biophysical Jou nal, 78(1):311A–311A,
Janua y 2000.
[Sun05] C. Sun, N. Fang, D. M. Wu, and X. Zhang. P ojec ion mic o-s e eoli hog aphy
using digi al mic o-mi o dynamic mask. Senso s and Ac ua o s A-Physical,
121(1):113–120, 2005.
[Tin73] I. Tinoco, P. N. Bo e , B. Dengle , M. D. Le ine, O. C. Uhlenbeck, D. M.
C o he s, and J. G alla. Imp o ed es ima ion o seconda y s uc u e in
ibonucleic-acids. Na u e-New Biology, 246(150):40–41, 1973.
[Toe03] A. Toegl, R. Ki chne , C. Gaue , and A. Wix o h. Enhancing esul s o mic oa -
ay hyb idiza ion ough mic oagi a ion. Jou nal o Biomolecula Techniques,
14:197–204, 2003.
[Tu 92] D. H. Tu ne . Bulges in nucleic acids. Cu en Opinion in S uc u al Biology,
2:334–337, 1992.
[U a02] H. U akawa, P. A. Noble, S. El Fan oussi, J. J. Kelly, and D. A. S ahl. Single-
base-pai disc imina ion o e minal misma ches by using oligonucleo ide mi-
c oa ays and neu al ne wo k analyses. Applied and En i onmen al Mic obiol-
ogy, 68(1):235–244, Janua y 2002.
256
BIBLIOGRAPHY
[U a03] H. U akawa, S. El Fan oussi, H. Smid , J. C. Smoo , E. H. T ibou, J. J. Kelly,
P. A. Noble, and D. A. S ahl. Op imiza ion o single-base-pai misma ch dis-
c imina ion in oligonucleo ide mic oa ays. Applied and en i onmen al mic o-
biology, 69(5):2848–2856, 2003.
[Vai02] A. Vain ub and B. M. Pe i . Coulomb blockage o hyb idiza ion in wo-
dimensional DNA a ays. Physical Re iew E, 66(4):041905, Oc obe 2002.
[ E06] T. S. an E p, S. Cues a-Lopez, and M. Pey a d. Bubbles and dena u a ion in
DNA. Eu opean Physical Jou nal E, 20(4):421–434, Augus 2006.
[Vic00] T. A. Vicke s, J. R. Wya , and S. M. F eie . E ec s o na seconda y s uc u e
on cellula an isense ac i i y. Nucleic Acids Resea ch, 28(6):1340–1347, Ma ch
2000.
[Vij01] R. A. Vijayend an and D. E. Leckband. A quan i a i e assessmen o he e o-
genei y o su ace-immobilized p o eins. Analy ical Chemis y, 73(3):471–480,
Feb ua y 2001.
[Wal79] R. B. Wallace, J. Sha e , R. F. Mu phy, J. Bonne , T. Hi ose, and K. I aku a.
Hyb idiza ion o syn he ic oligodeoxy ibonucleo ides o phi-chi-174 DNA e ec
o single base pai misma ch. Nucleic Acids Resea ch, 6(11):3543–3557, 1979.
[Wal01] S. Walbe , W. P leide e , and U. E. S eine . Pho olabile p o ec ing g oups o
nucleosides: Mechanis ic s udies o he 2-(2-ni ophenyl)e hyl g oup. Hel e ica
Chimica Ac a, 84(6):1601–1611, 2001.
[Wa 85] R. M. Wa ell and A. S. Benigh . The mal-dena u a ion o DNA-molecules:
a compa ison o heo y wi h expe imen . Physics Repo s-Re iew Sec ion o
Physics Le e s, 126(2):67–107, 1985.
[Wa 00] J. H. Wa e son, P. A. E. Piunno, C. C. Wus , and U. J. K ull. E ec s o oligonu-
cleo ide immobiliza ion densi y on selec i i y o quan i a i e ansduc ion o hy-
b idiza ion o immobilized DNA. Langmui , 16(11):4984–4992, May 2000.
[Wes07] E. M. Wes e hou and B. Be khou . A sys ema ic analysis o he e ec o a ge
na s uc u e an na in e e ence. Nucleic Acids Resea ch, 35(13):4322–4330,
2007.
[We 68] J. G. We mu and N. Da idson. Kine ics o ena u a ion o DNA. Jou nal o
Molecula Biology, 31(3):349–&, 1968.
[We 91] J. G. We mu . DNA p obes: Applica ions o he p inciples o nucleic-acid hy-
b idiza ion. C i ical Re iews in Biochemis y and Molecula Biology, 26(3-
4):227–259, 1991.
[Wic06] L. M. Wick, J. M. Rouilla d, T. S. Whi am, E. Gula i, J. M. Tiedje, and S. A.
Hashsham. On-chip non-equilib ium dissocia ion cu es and dissocia ion a e
cons an s as me hods o assess speci ici y o oligonucleo ide p obes. Nucleic
Acids Resea ch, 34(3):e26, 2006.
257
BIBLIOGRAPHY
[Woe06] D. F. Woell. Neue pho olabile Schu zg uppen mi in amolekula e Sensibil-
isie ung - Syn hese, pho okine ische Cha ak e isie ung und Anwendung ¨
u die
DNA-Chip-Syn hese. PhD hesis, Uni e si ae Kons anz, 2006.
[Wol04] D. Woll, S. Walbe , K. P. S engele, T. J. Albe , T. Richmond, J. No on,
M. Singe , R. D. G een, W. P leide e , and U. E. S eine . T iple -sensi ized
pho odep o ec ion o oligonucleo ides in solu ion and on mic oa ay chips. Hel-
e ica Chimica Ac a, 87(1):28–45, 2004.
[Won04] C. W. Wong, T. J. Albe , V. B. Vega, J. E. No on, D. J. Cu le , T. A. Richmond,
L. W. S an on, E. T. Liu, and L. D. Mille . T acking he e olu ion o he sa s
co ona i us using high- h oughpu , high-densi y esequencing a ays. Genome
Resea ch, 14(3):398–405, Ma ch 2004.
[Woo88] S. A. Woodson and D. M. C o he s. S uc u al model o an oligonucleo ide
con aining a bulged guanosine by NMR and ene gy minimiza ion. Biochemis y,
27(9):3130–3141, May 1988.
[Wu87] H. N. Wu and O. C. Uhlenbeck. Role o a bulged-a esidue in a speci ic RNA
p o ein-in e ac ion. Biochemis y, 26(25):8221–8227, Decembe 1987.
[Yil04] L. S. Yilmaz and D. R. Nogue a. Mechanis ic app oach o he p oblem o hy-
b idiza ion e iciency in luo escen in si u hyb idiza ion. Applied And En i on-
men al Mic obiology, 70(12):7126–7139, Decembe 2004.
[Yoo01] J. S. Yoo, H. K. Cheong, B. J. Lee, Y. B. Kim, and C. Cheong. Solu ion s uc-
u e o he SL1 RNA o he m1 double-s anded RNA i us o saccha omyces
ce e isiae. Biophysical Jou nal, 80(4):1957–1966, Ap il 2001.
[Zen06] Y. Zeng and G. Zocchi. Misma ches and bubbles in DNA. Biophysical Jou nal,
90(12):4522–4529, 2006.
[Zha03] L. Zhang, M. F. Miles, and K. D. Aldape. A model o molecula in e ac ions on
sho oligonucleo ide mic oa ays. Na u e Bio echnology, 21(7):818–821, July
2003.
[Zha07] L.Zhang, C. L.Wu, R. Ca a, and H.T.Zhao. F ee ene gy o DNA duplex o ma-
ion on sho oligonucleo ide mic oa ays. Nucleic Acids Resea ch, 35(3):e18,
Feb ua y 2007.
[Zho06] H. Zhou, Y. Zhang, and Z. Ou-Yang. Handbook o Theo e ical and Compu-
a ional Nano echnology, chap e Chap e 9: Theo e ical and Compu a ional
T ea men s o DNA and RNA Molecules, pages 419–487. Ame ican Scien i ic
Publishe s, 2006.
[Zhu99] J. Zhu and R. M. Wa ell. The e ec o base sequence on he s abili y o RNA
and DNA single base bulges. Biochemis y, 38(48):15986–15993, 1999.
[Zim60] B. H. Zimm. Theo y o mel ing o he helical o m in double chains o he DNA
ype. Jou nal o Chemical Physics, 33(5):1349–1356, 1960.
258
BIBLIOGRAPHY
[Zno02] B. M. Znosko, S. B. Sil es i, H. Volkman, B. Boswell, and M. J. Se a. The mo-
dynamic pa ame e s o an expanded nea es -neighbo model o he o ma ion
o RNA duplexes wi h single nucleo ide bulges. Biochemis y, 41(33):10406–
10417, 2002.
[Zoc03] G. Zocchi, A. Ome zu, T. Ku iabo a, J. Rudnick, and G. G une . Duplex-
single s and dena u a ion ansi ion in DNA oligome s. 2003. a Xi :cond-
ma /0304567.
259
BIBLIOGRAPHY
260
Appendix A
Expe imen al Da a
261
Expe imen al Da a
A.1 Expe imen al Da a
A.1.1 Compa ison Be ween MMs in RNA/DNA and DNA/DNA
Duplexes
0 2 4 6 8 10 12 14 16
0
0.2
0.4
0.6
0.8
T A T T A C T G G A C C T G A C
0 2 4 6 8 10 12 14 16
0
0.1
0.2
0.3
0.4
T A T T A C T G G A C C T G A C
0 2 4 6 8 10 12 14 16
0
0.5
1
1.5
T T G A G C G A T A T T A C T G
0 2 4 6 8 10 12 14 16
0
0.5
1
T T G A G C G A T A T T A C T G
0 2 4 6 8 10 12 14 16
0
0.2
0.4
0.6
A G T C A C G G A C A C A T G A
0 2 4 6 8 10 12 14 16
0
0.05
0.1
0.15
0.2
0.25
A G T C A C G G A C A C A T G A
0 2 4 6 8 10 12 14 16
0
0.5
1
1.5
C G A A C A T A C C T C C T T A
0 2 4 6 8 10 12 14 16
0
0.1
0.2
0.3
C G A A C A T A C C T C C T T A
D
C
AB
Figu e A.1: Di ec compa ison o DNA/DNA and RNA/DNA misma ch hyb idiza-
ion signals (see sec ion 6.8). Pa s A-D compa e de ec p o iles o di e en sequence
mo i s (sequences shown a he bo om o he plo s). Hyb idiza ions o RNA a -
ge s ( op image) and equi alen DNA a ge s (bo om image) we e pe o med subse-
quen ly on he same mic oa ays. The de ec posi ional in luence is e y simila o
DNA/DNA and RNA/DNA hyb idiza ion. Howe e , he e a e sys ema ic di e ences
be ween he binding a ini ies o he a ious MM ypes in DNA/DNA and RNA/DNA
duplexes. The hyb idiza ion signal (in a.u.) is plo ed e sus de ec posi ion. Subs i-
u ion bases A ( ed c oss), C(g een ci cle), G (blue s a ) and T (cyan iangle) ei he
esul in 3 MM duplexes and one PM duplex a e e y de ec posi ion; Hyb idiza-
ion signals o duplexes wi h single base dele ions (yellow line); mo ing a e age MM
hyb idiza ion signal (black line).
262
Expe imen al Da a
0 2 4 6 8 10 12 14 16
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
G A T A T T A C T G G A C C T G
0 2 4 6 8 10 12 14 16
0
0.2
0.4
0.6
0.8
G A T A T T A C T G G A C C T G
0 2 4 6 8 10 12 14 16
0
0.2
0.4
0.6
0.8
1
1.2
G A T A T T A C T G G A C C T G
0 2 4 6 8 10 12 14 16
0
0.1
0.2
0.3
0.4
0.5
G A T A T T A C T G G A C C T G
0 2 4 6 8 10 12 14 16
0
0.5
1
1.5
A G C G A T A T T A C T G G A C
0 2 4 6 8 10 12 14 16
0
0.2
0.4
0.6
0.8
1
A G C G A T A T T A C T G G A C
0 2 4 6 8 10 12 14 16
0
0.5
1
1.5
A G C G A T A T T A C T G G A C
0 2 4 6 8 10 12 14 16
0
0.2
0.4
0.6
0.8
1
1.2
A G C G A T A T T A C T G G A C
0 2 4 6 8 10 12 14 16
0
0.5
1
T T G A G C G A T A T T A C T G
0 2 4 6 8 10 12 14 16
0
0.2
0.4
0.6
0.8
1
T T G A G C G A T A T T A C T G
0 2 4 6 8 10 12 14 16
0
0.5
1
1.5
T T G A G C G A T A T T A C T G
0 2 4 6 8 10 12 14 16
0
0.5
1
T T G A G C G A T A T T A C T G
Figu e A.2: Fo de ails see Fig. A.1.
263
Expe imen al Da a
0 5 10 15 20
0
0.5
1
1.5
2
2.5
G C G A T A T T A C T G G A C C T G A C
0 5 10 15 20
0
0.5
1
1.5
G C G A T A T T A C T G G A C C T G A C
0 5 10 15 20
0
0.5
1
1.5
2
2.5
G C G A T A T T A C T G G A C C T G A C
0 5 10 15 20
0
0.5
1
1.5
2
G C G A T A T T A C T G G A C C T G A C
0 5 10 15 20
0
0.5
1
1.5
2
2.5
3
T T G A G C G A T A T T A C T G G A C C
0 5 10 15 20
0
0.5
1
1.5
2
2.5
T T G A G C G A T A T T A C T G G A C C
0 5 10 15 20
0
0.5
1
1.5
2
2.5
T T G A G C G A T A T T A C T G G A C C
0 5 10 15 20
0
0.5
1
1.5
T T G A G C G A T A T T A C T G G A C C
0 2 4 6 8 10 12 14 16
0
0.2
0.4
0.6
0.8
1
T A T T A C T G G A C C T G A C
0 2 4 6 8 10 12 14 16
0
0.1
0.2
0.3
0.4
0.5
0.6
T A T T A C T G G A C C T G A C
0 2 4 6 8 10 12 14 16
0
0.2
0.4
0.6
0.8
T A T T A C T G G A C C T G A C
0 2 4 6 8 10 12 14 16
0
0.1
0.2
0.3
0.4
T A T T A C T G G A C C T G A C
Figu e A.3: Fo de ails see Fig. A.1.
264
Expe imen al Da a
0 2 4 6 8 10 12 14 16
0
0.1
0.2
0.3
0.4
C A C G G A C A C A T G A T C C
0 2 4 6 8 10 12 14 16
0
0.05
0.1
0.15
C A C G G A C A C A T G A T C C
0 2 4 6 8 10 12 14 16
0
0.1
0.2
0.3
0.4
0.5
C A C G G A C A C A T G A T C C
0 2 4 6 8 10 12 14 16
0
0.05
0.1
0.15
0.2
C A C G G A C A C A T G A T C C
0 2 4 6 8 10 12 14 16
0
0.2
0.4
0.6
A G T C A C G G A C A C A T G A
0 2 4 6 8 10 12 14 16
0
0.05
0.1
0.15
0.2
0.25
A G T C A C G G A C A C A T G A
0 2 4 6 8 10 12 14 16
0
0.2
0.4
0.6
A G T C A C G G A C A C A T G A
0 2 4 6 8 10 12 14 16
0
0.05
0.1
0.15
0.2
0.25
A G T C A C G G A C A C A T G A
0 2 4 6 8 10 12 14 16
0
0.2
0.4
0.6
T G T A G T C A C G G A C A C A
0 2 4 6 8 10 12 14 16
0
0.1
0.2
0.3
T G T A G T C A C G G A C A C A
0 2 4 6 8 10 12 14 16
0
0.1
0.2
0.3
0.4
0.5
0.6
T G T A G T C A C G G A C A C A
0 2 4 6 8 10 12 14 16
0
0.05
0.1
0.15
0.2
0.25
T G T A G T C A C G G A C A C A
Figu e A.4: Fo de ails see Fig. A.1.
265
Expe imen al Da a
A.1.3 Single Base Misma ches in DNA/DNA Duplexes - S a-
is ical Analysis o In es iga e he In luence o he
Flanking Base Pai s
−1.5 −1 −0.5 0 0.5 1 1.5
0
5
10
15
20 MM base pai : AA
µ= −0.22
−1.5 −1 −0.5 0 0.5 1 1.5
0
5
10
15
20 MM base pai : CA
µ= −0.11
−1.5 −1 −0.5 0 0.5 1 1.5
0
5
10
15
20 MM base pai : GA
µ= 0.27
−1.5 −1 −0.5 0 0.5 1 1.5
0
5
10
15
20 MM base pai : AC
µ= −0.29
−1.5 −1 −0.5 0 0.5 1 1.5
0
5
10
15
20 MM base pai : CC
µ= −0.24
−1.5 −1 −0.5 0 0.5 1 1.5
0
5
10
15
20 MM base pai : TC
µ= −0.26
−1.5 −1 −0.5 0 0.5 1 1.5
0
5
10
15
20 MM base pai : AG
µ= −0.081
−1.5 −1 −0.5 0 0.5 1 1.5
0
5
10
15
20 MM base pai : GG
µ= −0.33
−1.5 −1 −0.5 0 0.5 1 1.5
0
5
10
15
20 MM base pai : TG
µ= −0.23
−1.5 −1 −0.5 0 0.5 1 1.5
0
5
10
15
20 MM base pai : CT
µ= −0.13
−1.5 −1 −0.5 0 0.5 1 1.5
0
5
10
15
20 MM base pai : GT
µ= 0.059
−1.5 −1 −0.5 0 0.5 1 1.5
0
5
10
15
20 MM base pai : TT
µ= 0.069
Figu e A.11: All misma ch base pai ypes X·Y. Measu ed hyb idiza ion signal
dis ibu ions (occu ence e sus de ia ion o he pa icula hyb idiza ion signal om
he mean p o ile) as a unc ion o he MM base pai alone, i.e. independen o he
lanking base pai s. µdeno es he median alue o he dis ibu ions. A box-whiske
plo o he dis ibu ions is shown in Fig. 6.6.
On he ollowing pages (Figs. A.12 - A.22) his da a is ca ego ized acco ding o he ype
o lanking base pai s. Owing o he es ic ed se o a ge sequences a ailable o his
s udy he sizes o he da a se s measu ed o he indi idual de ec con igu a ions a e e y
di e en . µdeno es he median alues o he dis ibu ions.
The median alues o he nea es neighbo pai dependen subse s a e compa ed in Fig. 6.9.
272
Expe imen al Da a
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAT−3´
3´−AAA−5´
µ= −0.39
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAG−3´
3´−AAC−5´
µ= −0.39
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAC−3´
3´−AAG−5´
µ= 0.0021
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAT−3´
3´−CAA−5´
µ= −0.0072
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAG−3´
3´−CAC−5´
µ= −0.24
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAA−3´
3´−CAT−5´
µ= −0.1
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAT−3´
3´−GAA−5´
µ= −0.57
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAC−3´
3´−GAG−5´
µ= −0.087
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAA−3´
3´−GAT−5´
µ= −0.27
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAG−3´
3´−TAC−5´
µ= −0.15
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAC−3´
3´−TAG−5´
µ= −0.22
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAA−3´
3´−TAT−5´
µ= −0.12
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAA−3´
3´−AAT−5´
µ= −0.055
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAC−3´
3´−CAG−5´
µ= −0.33
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAG−3´
3´−GAC−5´
µ= −0.26
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAT−3´
3´−TAA−5´
µ= −0.036
Figu e A.12: A·A misma ches. Measu ed hyb idiza ion signal dis ibu ions ca ego-
ized acco ding o he lanking base pai s.
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAT−3´
3´−ACA−5´
µ= −0.24
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAG−3´
3´−ACC−5´
µ= 0.4
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAC−3´
3´−ACG−5´
µ= 0.21
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAT−3´
3´−CCA−5´
µ= 0.064
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAG−3´
3´−CCC−5´
µ= 0.18
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAA−3´
3´−CCT−5´
µ= −0.074
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAT−3´
3´−GCA−5´
µ= −0.47
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAC−3´
3´−GCG−5´
µ= −0.24
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAA−3´
3´−GCT−5´
µ= −0.29
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAG−3´
3´−TCC−5´
µ= −0.09
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAC−3´
3´−TCG−5´
µ= −0.22
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAA−3´
3´−TCT−5´
µ= −0.15
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAA−3´
3´−ACT−5´
µ= −0.29
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAC−3´
3´−CCG−5´
µ= −0.35
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAG−3´
3´−GCC−5´
µ= −0.035
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAT−3´
3´−TCA−5´
µ= −0.26
Figu e A.13: C·A misma ches. Measu ed hyb idiza ion signal dis ibu ions ca ego-
ized acco ding o he lanking base pai s.
273
Expe imen al Da a
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAT−3´
3´−AGA−5´
µ= 0.68
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAG−3´
3´−AGC−5´
µ= 0.33
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAC−3´
3´−AGG−5´
µ= 0.017
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAT−3´
3´−CGA−5´
µ= 0.51
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAG−3´
3´−CGC−5´
µ= 0.27
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAA−3´
3´−CGT−5´
µ= −0.28
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAT−3´
3´−GGA−5´
µ= −0.33
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAC−3´
3´−GGG−5´
µ= 0.28
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAA−3´
3´−GGT−5´
µ= 0.18
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAG−3´
3´−TGC−5´
µ= 0.56
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAC−3´
3´−TGG−5´
µ= 0.16
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAA−3´
3´−TGT−5´
µ= −0.00077
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAA−3´
3´−AGT−5´
µ= −0.17
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAC−3´
3´−CGG−5´
µ= 0.68
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAG−3´
3´−GGC−5´
µ= 0.4
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAT−3´
3´−TGA−5´
µ= 0.35
Figu e A.14: G·A misma ches. Measu ed hyb idiza ion signal dis ibu ions ca ego-
ized acco ding o he lanking base pai s.
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TCT−3´
3´−AAA−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TCG−3´
3´−AAC−5´
µ= −0.26
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TCC−3´
3´−AAG−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GCT−3´
3´−CAA−5´
µ= −0.47
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GCG−3´
3´−CAC−5´
µ= 0.087
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GCA−3´
3´−CAT−5´
µ= −0.32
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CCT−3´
3´−GAA−5´
µ= −0.2
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CCC−3´
3´−GAG−5´
µ= −0.25
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CCA−3´
3´−GAT−5´
µ= −0.18
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ACG−3´
3´−TAC−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ACC−3´
3´−TAG−5´
µ= −0.16
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ACA−3´
3´−TAT−5´
µ= −0.48
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TCA−3´
3´−AAT−5´
µ= −0.28
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GCC−3´
3´−CAG−5´
µ= −0.11
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CCG−3´
3´−GAC−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ACT−3´
3´−TAA−5´
µ= −0.58
Figu e A.15: A·C misma ches. Measu ed hyb idiza ion signal dis ibu ions ca ego-
ized acco ding o he lanking base pai s.
274
Expe imen al Da a
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TCT−3´
3´−ATA−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TCG−3´
3´−ATC−5´
µ= −0.087
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TCC−3´
3´−ATG−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GCT−3´
3´−CTA−5´
µ= −0.32
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GCG−3´
3´−CTC−5´
µ= −0.63
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GCA−3´
3´−CTT−5´
µ= −0.31
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CCT−3´
3´−GTA−5´
µ= −0.094
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CCC−3´
3´−GTG−5´
µ= −0.042
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CCA−3´
3´−GTT−5´
µ= 0.094
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ACG−3´
3´−TTC−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ACC−3´
3´−TTG−5´
µ= −0.086
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ACA−3´
3´−TTT−5´
µ= −0.22
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TCA−3´
3´−ATT−5´
µ= −0.1
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GCC−3´
3´−CTG−5´
µ= −0.15
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CCG−3´
3´−GTC−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ACT−3´
3´−TTA−5´
µ= −0.37
Figu e A.16: T·C misma ches. Measu ed hyb idiza ion signal dis ibu ions ca ego-
ized acco ding o he lanking base pai s.
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TGT−3´
3´−AAA−5´
µ= −0.1
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TGG−3´
3´−AAC−5´
µ= −0.11
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TGC−3´
3´−AAG−5´
µ= −0.26
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GGT−3´
3´−CAA−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GGG−3´
3´−CAC−5´
µ= 0.74
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GGA−3´
3´−CAT−5´
µ= 0.1
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CGT−3´
3´−GAA−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CGC−3´
3´−GAG−5´
µ= −0.028
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CGA−3´
3´−GAT−5´
µ= 1.3
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AGG−3´
3´−TAC−5´
µ= −0.11
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AGC−3´
3´−TAG−5´
µ= 0.046
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AGA−3´
3´−TAT−5´
µ= −0.23
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TGA−3´
3´−AAT−5´
µ= −0.11
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GGC−3´
3´−CAG−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CGG−3´
3´−GAC−5´
µ= −0.55
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AGT−3´
3´−TAA−5´
µ= −0.25
Figu e A.17: A·G misma ches. Measu ed hyb idiza ion signal dis ibu ions ca ego-
ized acco ding o he lanking base pai s.
275
Expe imen al Da a
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TGT−3´
3´−AGA−5´
µ= −0.36
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TGG−3´
3´−AGC−5´
µ= −0.22
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TGC−3´
3´−AGG−5´
µ= −0.24
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GGT−3´
3´−CGA−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GGG−3´
3´−CGC−5´
µ= −0.17
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GGA−3´
3´−CGT−5´
µ= −0.15
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CGT−3´
3´−GGA−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CGC−3´
3´−GGG−5´
µ= −0.2
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CGA−3´
3´−GGT−5´
µ= −0.33
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AGG−3´
3´−TGC−5´
µ= −0.49
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AGC−3´
3´−TGG−5´
µ= −0.37
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AGA−3´
3´−TGT−5´
µ= −0.77
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TGA−3´
3´−AGT−5´
µ= −0.33
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GGC−3´
3´−CGG−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CGG−3´
3´−GGC−5´
µ= −0.26
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AGT−3´
3´−TGA−5´
µ= −0.51
Figu e A.18: G·G misma ches. Measu ed hyb idiza ion signal dis ibu ions ca ego-
ized acco ding o he lanking base pai s.
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TGT−3´
3´−ATA−5´
µ= −0.35
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TGG−3´
3´−ATC−5´
µ= −0.41
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TGC−3´
3´−ATG−5´
µ= −0.22
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GGT−3´
3´−CTA−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GGG−3´
3´−CTC−5´
µ= 0.056
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GGA−3´
3´−CTT−5´
µ= −0.1
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CGT−3´
3´−GTA−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CGC−3´
3´−GTG−5´
µ= −0.11
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CGA−3´
3´−GTT−5´
µ= −0.15
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AGG−3´
3´−TTC−5´
µ= −0.31
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AGC−3´
3´−TTG−5´
µ= −0.17
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AGA−3´
3´−TTT−5´
µ= −0.23
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TGA−3´
3´−ATT−5´
µ= −0.27
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GGC−3´
3´−CTG−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CGG−3´
3´−GTC−5´
µ= −0.52
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AGT−3´
3´−TTA−5´
µ= −0.45
Figu e A.19: T·G misma ches. Measu ed hyb idiza ion signal dis ibu ions ca ego-
ized acco ding o he lanking base pai s.
276
Expe imen al Da a
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TTT−3´
3´−ACA−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TTG−3´
3´−ACC−5´
µ= 0.053
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TTC−3´
3´−ACG−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GTT−3´
3´−CCA−5´
µ= 0.12
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GTG−3´
3´−CCC−5´
µ= −0.052
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GTA−3´
3´−CCT−5´
µ= 0.099
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CTT−3´
3´−GCA−5´
µ= −0.13
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CTC−3´
3´−GCG−5´
µ= −0.4
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CTA−3´
3´−GCT−5´
µ= −0.048
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ATG−3´
3´−TCC−5´
µ= −0.34
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ATC−3´
3´−TCG−5´
µ= −0.42
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ATA−3´
3´−TCT−5´
µ= −0.34
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TTA−3´
3´−ACT−5´
µ= 0.38
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GTC−3´
3´−CCG−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CTG−3´
3´−GCC−5´
µ= −0.071
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ATT−3´
3´−TCA−5´
µ= 0.091
Figu e A.20: C·T misma ches. Measu ed hyb idiza ion signal dis ibu ions ca ego-
ized acco ding o he lanking base pai s.
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TTT−3´
3´−AGA−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TTG−3´
3´−AGC−5´
µ= 0.24
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TTC−3´
3´−AGG−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GTT−3´
3´−CGA−5´
µ= 0.2
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GTG−3´
3´−CGC−5´
µ= −0.018
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GTA−3´
3´−CGT−5´
µ= 0.17
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CTT−3´
3´−GGA−5´
µ= −0.1
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CTC−3´
3´−GGG−5´
µ= 0.00061
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CTA−3´
3´−GGT−5´
µ= 0.48
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ATG−3´
3´−TGC−5´
µ= −0.11
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ATC−3´
3´−TGG−5´
µ= −0.15
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ATA−3´
3´−TGT−5´
µ= 0.075
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TTA−3´
3´−AGT−5´
µ= 0.28
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GTC−3´
3´−CGG−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CTG−3´
3´−GGC−5´
µ= 0.14
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ATT−3´
3´−TGA−5´
µ= 0.27
Figu e A.21: G·T misma ches. Measu ed hyb idiza ion signal dis ibu ions ca ego-
ized acco ding o he lanking base pai s.
277
Expe imen al Da a
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TTT−3´
3´−ATA−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TTG−3´
3´−ATC−5´
µ= −0.12
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TTC−3´
3´−ATG−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GTT−3´
3´−CTA−5´
µ= −0.083
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GTG−3´
3´−CTC−5´
µ= −0.036
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GTA−3´
3´−CTT−5´
µ= 0.25
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CTT−3´
3´−GTA−5´
µ= 0.99
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CTC−3´
3´−GTG−5´
µ= 0.48
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CTA−3´
3´−GTT−5´
µ= 0.69
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ATG−3´
3´−TTC−5´
µ= −0.082
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ATC−3´
3´−TTG−5´
µ= −0.2
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ATA−3´
3´−TTT−5´
µ= −0.0012
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TTA−3´
3´−ATT−5´
µ= −0.017
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GTC−3´
3´−CTG−5´
no da a a ailable
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CTG−3´
3´−GTC−5´
µ= 0.34
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−ATT−3´
3´−TTA−5´
µ= 0.52
Figu e A.22: T·T misma ches. Measu ed hyb idiza ion signal dis ibu ions ca ego-
ized acco ding o he lanking base pai s.
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAT−3´
3´−AGA−5´
µ= 0.68
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAG−3´
3´−AGC−5´
µ= 0.33
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAC−3´
3´−AGG−5´
µ= 0.017
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAT−3´
3´−CGA−5´
µ= 0.51
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAG−3´
3´−CGC−5´
µ= 0.27
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAA−3´
3´−CGT−5´
µ= −0.28
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAT−3´
3´−GGA−5´
µ= −0.33
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAC−3´
3´−GGG−5´
µ= 0.28
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAA−3´
3´−GGT−5´
µ= 0.18
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAG−3´
3´−TGC−5´
µ= 0.56
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAC−3´
3´−TGG−5´
µ= 0.16
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAA−3´
3´−TGT−5´
µ= −0.00077
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−TAA−3´
3´−AGT−5´
µ= −0.17
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−GAC−3´
3´−CGG−5´
µ= 0.68
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−CAG−3´
3´−GGC−5´
µ= 0.4
−1.5 −1 −0.5 0 0.5 1 1.5
0
2
4
6
5´−AAT−3´
3´−TGA−5´
µ= 0.35
Figu e A.23: G·A misma ches. Measu ed hyb idiza ion signal dis ibu ions ca ego-
ized acco ding o he lanking base pai s.
278
Expe imen al Da a
A.1.4 Single Base Inse ions - S a is ical Analysis
−0.4 −0.2 0 0.2
0
5
10
5´−T T−3´
3´−AAA−5´
µ= 0.023
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−T G−3´
3´−AAC−5´
µ= 0.048
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−T C−3´
3´−AAG−5´
µ= 0.0049
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−T A−3´
3´−AAT−5´
µ= −0.009
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−G T−3´
3´−CAA−5´
µ= 0.031
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−G G−3´
3´−CAC−5´
µ= 0
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−G C−3´
3´−CAG−5´
µ= −0.12
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−G A−3´
3´−CAT−5´
µ= −0.064
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−C T−3´
3´−GAA−5´
µ= 0.0063
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−C G−3´
3´−GAC−5´
µ= 0.015
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−C C−3´
3´−GAG−5´
µ= −0.012
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−C A−3´
3´−GAT−5´
µ= −0.098
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−A T−3´
3´−TAA−5´
µ= 0.017
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−A G−3´
3´−TAC−5´
µ= 0
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−A C−3´
3´−TAG−5´
µ= −0.041
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−A A−3´
3´−TAT−5´
µ= −0.075
G oup: I
Figu e A.24: Inse ions o adenine bases - in luence o he neighbo ing base pai s.
Dis ibu ion o hyb idiza ion signal in ensi ies (de ia ion om he mean p o ile in
a.u.). µdeno es he median alue o he dis ibu ion. G oup II inse ions ha e a
leas one iden ical neighbo base, whe eas G oup I inse ions don’ ha e an iden-
ical neighbo . G oup II inse ion ha e consis en ly inc eased hyb idiza ion signals
compa ed o G oup I inse ions.
279
Expe imen al Da a
−0.4 −0.2 0 0.2
0
5
10
5´−T T−3´
3´−ACA−5´
µ= 0
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−T G−3´
3´−ACC−5´
µ= 0.045
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−T C−3´
3´−ACG−5´
µ= −0.043
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−T A−3´
3´−ACT−5´
µ= 0.0075
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−G T−3´
3´−CCA−5´
µ= 0.012
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−G G−3´
3´−CCC−5´
µ= 0.12
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−G C−3´
3´−CCG−5´
µ= 0.012
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−G A−3´
3´−CCT−5´
µ= 0.032
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−C T−3´
3´−GCA−5´
µ= −0.054
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−C G−3´
3´−GCC−5´
µ= −0.0087
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−C C−3´
3´−GCG−5´
µ= −0.039
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−C A−3´
3´−GCT−5´
µ= −0.057
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−A T−3´
3´−TCA−5´
µ= −0.069
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−A G−3´
3´−TCC−5´
µ= 0.025
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−A C−3´
3´−TCG−5´
µ= −0.051
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−A A−3´
3´−TCT−5´
µ= −0.066
G oup: I
Figu e A.25: Inse ions o cy osine bases - in luence o he neighbo ing base pai s.
Dis ibu ion o hyb idiza ion signal in ensi ies (de ia ion om he mean p o ile in
a bi a y uni s).
−0.4 −0.2 0 0.2
0
5
10
5´−T T−3´
3´−AGA−5´
µ= 0
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−T G−3´
3´−AGC−5´
µ= −0.088
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−T C−3´
3´−AGG−5´
µ= 0.0059
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−T A−3´
3´−AGT−5´
µ= 0.014
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−G T−3´
3´−CGA−5´
µ= −0.013
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−G G−3´
3´−CGC−5´
µ= −0.093
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−G C−3´
3´−CGG−5´
µ= 0.048
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−G A−3´
3´−CGT−5´
µ= −0.021
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−C T−3´
3´−GGA−5´
µ= 0.085
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−C G−3´
3´−GGC−5´
µ= 0.034
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−C C−3´
3´−GGG−5´
µ= 0.12
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−C A−3´
3´−GGT−5´
µ= 0.09
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−A T−3´
3´−TGA−5´
µ= −0.033
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−A G−3´
3´−TGC−5´
µ= −0.063
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−A C−3´
3´−TGG−5´
µ= 0.14
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−A A−3´
3´−TGT−5´
µ= 0
G oup: I
Figu e A.26: Inse ions o guanine bases - in luence o he neighbo ing base pai s.
Dis ibu ion o hyb idiza ion signal in ensi ies (de ia ion om he mean p o ile in
a bi a y uni s).
280
Expe imen al Da a
−0.4 −0.2 0 0.2
0
5
10
5´−T T−3´
3´−ATA−5´
µ= 0.032
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−T G−3´
3´−ATC−5´
µ= −0.081
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−T C−3´
3´−ATG−5´
µ= −0.035
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−T A−3´
3´−ATT−5´
µ= 0.0054
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−G T−3´
3´−CTA−5´
µ= 0.0062
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−G G−3´
3´−CTC−5´
µ= −0.08
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−G C−3´
3´−CTG−5´
µ= −0.11
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−G A−3´
3´−CTT−5´
µ= −0.0068
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−C T−3´
3´−GTA−5´
µ= 0.028
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−C G−3´
3´−GTC−5´
µ= −0.029
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−C C−3´
3´−GTG−5´
µ= 0
G oup: I
−0.4 −0.2 0 0.2
0
5
10
5´−C A−3´
3´−GTT−5´
µ= 0
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−A T−3´
3´−TTA−5´
µ= 0.01
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−A G−3´
3´−TTC−5´
µ= −0.027
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−A C−3´
3´−TTG−5´
µ= −0.0097
G oup: II
−0.4 −0.2 0 0.2
0
5
10
5´−A A−3´
3´−TTT−5´
µ= 0.043
G oup: II
Figu e A.27: Inse ions o hymine bases - in luence o he neighbo ing base pai s.
Dis ibu ion o hyb idiza ion signal in ensi ies (de ia ion om he mean p o ile in
a bi a y uni s).
281
Suppo ing In o ma ion
0 0.2 0.4 0.6 0.8 1
0
500
1000
1500
Iin
Iou (a.u.)
Measu ed in ensi y
Fi : Iou =Iin2.2
Figu e B.5: Gamma unc ion o he As oBeam p ojec o . The in ensi y esponse
Iou on he image b igh ness Iin (no malized on a maximum alue o 1) ollows a
powe law wi h an exponen o 2.2. Fo an image b igh ness la ge han abou 80% o
he maximum alue a cu -o is obse ed. The posi ion o he cu -o depends on he
con as and b igh ness alues chosen in he As oBeams ”Display Se ings Menu”.
B.3 Op ics o he Mic oscope P ojec ion Pho o-
li hog aphy Sys em
•UHP: Philips UHP-lamp 250W 1.35 TOP 222 H4 ellip ical e lec o ellip ical e lec o geome y: majo
axis ∼80 mm, mino axis ∼50 mm)
•L1: plano-conca e lens: =50 mm, diam. 25 mm (silica), placemen be ween UHP lamp window and
he ou e ocal poin o he ellip ical e lec o
•L1-L2: 145 mm
•L2: plano-con exlens: =50 mm, diam. 50 mm
•L2-F1: 120 mm
•F1: UV cold mi o (UV ba ie il e om he Op oma p ojec o lamp module)
•F1-L3: 165 mm
•L3: plano-con exlens (BK7): =100 mm, diam. 50 mm
•F1-F2: 215 mm
•F2: UV cold mi o (O iel)
•F2-F3: 165 mm
•F3: UV band pass (bk-370-35-B, In e e enzop ik Elek onik GmbH), diam. 25.4 mm
•F2-L4: 250 mm
•L4: plano-con exlens (BK7): =125 mm, diam. 50 mm
•L4-M1: 170 mm
288
Op ics o he Mic oscope P ojec ion Pho oli hog aphy Sys em
UHP
L1
L2
F1
LT
L3
F2
LT
S
F3
F4 L4
M1
M2
L5
DMD M3
FO
PS
VP1
VP2
PC
ICM
Figu e B.6: Schema ic o he mic oscope p ojec ion pho oli hog aphy sys em.
•M1: mi o
•M1-M2: 380 mm
•M2: mi o
•M2-DMD: 60 mm
•DMD-L5: ca. 164.5 mm, o be ine-adjus ed
•L5: ube lens, Ca l Zeiss, =164.5 mm
•M3: mi o /beam spli e
289
Suppo ing In o ma ion
B.4 Fab ica ion o he Syn hesis Cell
Figu e B.7: Punching ool ( op) o he ab ica ion o he PDMS gaske (cen e ).
The ool, p oducing a diamond-shaped cu ou ( he cell olume) wi h clean edges,
is essen ial o smoo h ope a ion o syn hesis appa a us. Wi e-cu EDM (elec ical
discha ge machining) has been employed o p oducing he sha p-edged s uc u e in
ha dened s eel. Dimensions o diamond-shaped cell olume: leng h 16 mm; wid h
5 mm. The ou e edge o he gaske was cu wi h ano he (smalle ) e sion o he
punching ool.
Pa names a e e e ing o Fig. 3.13.
•The op-pla e is made om a 10 mm hick pla e o anspa en Mak olon R
plas ics
(polyca bona e). P oduce ou apped holes o as ening sc ews (no oo a away
om he cen e o he pla e, o enable p ope sealing ac ion). Fu he , wo holes o
as ening he cell-assembly on he p ojec ion li hog aphy se up a e equi ed.
•Inle and ou le ubes a e made om sy inge needles (0.9×40 mm). By using a d illing
machine as a ”la he” he plas ic adap e o he sy inge needle is educed o a cylind ic
bi as shown in Fig. 3.13.
•P oduce holes o inle /ou le needles. (diam. 1 mm on he uppe side o he op pla e).
A he bo om side o he op-pla e he needle (blun end nea he coupling) should
p o ude 1 mm. The needles a e as ened wi h epoxy glue.
•To ob ain a anspa en and chemically ine (sol en esis an ) su ace, a glass mi-
c oscopy slide is glued on o he lowe side o he op-pla e. Be o e gluing (wi h ans-
pa en PDMS silicone ubbe ), he slide needs o be cu in 3 pieces o p oduce gaps o
he as ening sc ews. Mo eo e , wo 1 mm diam. holes o he inle /ou le ubes ha e
o be d illed in o he glass slide by using a diamond ool. By gluing he glass slide on o
290
Fab ica ion o he Syn hesis Cell
he op-pla e he gaps be ween he needles and he glass a e sealed wi h PDMS (a oid
ge ing PDMS in o he needles!). PDMS (Dow Co ning Sylga d R
184) was pu chased
om Wo ld P ecision Ins umen s.
•The bo om-pla e is made om 5 mm aluminum. The exposu e window should no
be oo la ge (ideally implemen ed as a long hole) o achie e p ope sealing ac ion by
p essing he Chip-subs a e/PDMS-gaske agains he op-pla e.
•Fab ica ion o he PDMS-gaske : PDMS Sylga d 184 (Dow Co ning) is mixed ho -
oughly ( a io be ween elas ome base and cu ing agen : 10:1), degassed and pou ed
in o a glass pe i dish. Cu ing o 20 minu es a 80◦C. A cus om-made punching ool
(Fig. B.7) is used o p oduce he s eamlined cu ou o ming he syn hesis olume.
•Connec o s: PFA (PTFE) ubes (in e nal diam. 0.8 mm) i igh ly on he 0.9 mm
diam. sy inge needles. PTFE ube end i ings (UNF 1/4” 28 G) p o ide a emo able
connec ion wi h he luidics sys em.
291
Suppo ing In o ma ion
B.5 Technical No es on Ligh -di ec ed DNA Chip
Syn hesis
B.5.1 Handling o Phospho amidi e Reagen s
The coupling e iciency o phospho amidi e eagen s is e y sensi i e o con amina ion
wi h (e en ace amoun s o ) wa e . To main ain low mois u e condi ions he ollowing
p ecau ions should be conside ed:
•S o age unde mois u e ee condi ions a -20◦C. Use d y a gon a mosphe e and desic-
can .
•Open s o age bo les only in glo e box unde d y a gon a mosphe e. Use silica gel
beads o main ain a low mois u e con en in he glo e box.
•Use o en-d ied glass wa e o minimize su ace-adso bed wa e .
•Dissol e phospho amidi es only immedia ely be o e syn hesis.
•Use d y MeCN wi h <10 ppm o wa e .
•Use molecula sie e bags (in he MeCN s o age bo le and in he ac i a o solu ion) o
adso b wa e om he sol en .
•Phospho amidi e solu ions should be used he same day as p epa ed.
Solu ion s abili y and deg ada ion pa hways o deoxy ibonucleoside phospho amidi es in
MeCN a e discussed in [K o04].
B.5.2 Addi ional No es on he Syn hesis
P io o he i s phospho amidi e coupling he subs a e is soaked in MeCN o abou
2 minu es. The ini ial coupling is pe o med o 1 minu e and hen epea ed once. Acco d-
ing o Richmond e al. [Ric04] an inc ease o he coupling ime (o he i s base only)
om 20 s o 6 h esul ed in an 80% inc ease in he amoun o ull-leng h p obes.
Coupling and exposu e ime, washing s eps and image quali y a e he key pa ame e s o
high quali y syn hesis. Acco ding o [Ric04] he numbe o e o - ee p obe sequences
could be inc eased 100- old by making se e al echnical imp o emen s on hei syn hesis
appa a us. Imp o emen s include he ex ension o he coupling ime om 20 o 60 s and
o he exposu e ime om 50 o 150 s, addi ional a gon d ying s eps and modi ica ions on
he p ojec ion op ical sys em (image-locking).
Upon p olonged exposu e he sol en s e ahyd o u ane (THF) and py idine cause signi i-
can swelling o he PDMS gaske . Exposu e o hese sol en s (con ained in oxidize and
capping eagen s) should he e o e be minimized.
292
Technical No es on Mic oa ay Dend ime Subs a e P epa a ion
B.6 Technical No es on Mic oa ay Dend ime Sub-
s a e P epa a ion
Figu e B.8: (A) Te lon slide holde o up o 12 ound co e glasses. The s ainless
s eel pin secu es he glasses. Fo use wi h dichlo oe hane he nylon sc ews should
be eplaced by s ainless s eel sc ews. (B) Subs a e unc ionaliza ion in a 500 ml
g adua ed cylinde equi es abou 250 ml eagen solu ion.
•Fo dend ime unc ionaliza ion o he mic oa ay subs a es a compac slide holde
o handling o up o 12 co e glasses was de eloped. Pa s o he e lon (PTFE)
slide holde a e assembled wi h s ainless s eel sc ews and can hus wi hs and a ba h
in dichlo oe hane solu ion. The holde enables as and ho ough washing and d y-
ing o he slides. Use o he holde s esul ed in signi ican ly inc eased quali y o he
subs a es and enabled educ ion o he eagen consump ion.
•To minimize eagen consump ion (e hanol analy ical g ade, dend ime s in dichlo o-
e hane) he subs a e unc ionaliza ion is pe o med in a 500 ml g adua ed cylinde .
Th ee slide holde s (wi h 36 slides in o al) a e imme sed in abou 250 ml o solu ion.
•D ying o he slides unde a ni ogen s eam should be pe o med in such a way ha
he liquid is blown away om he cen e o he slides. D ying o d ople s on he su ace
has o be a oided because his can p oduce i emo able s ains.
293
Suppo ing In o ma ion
B.7 Technical No es on he Syn hesize Con ol
So wa e DNASyn
The ligh -di ec ed ab ica ion o a DNA mic oa ay has been ully au oma ed. The syn he-
size con ol so wa e DNASyn in eg a es con ol o he luidics sys em wi h he maskless
mic opho oli hog aphy sys em (including image display, shu e and il e con ol).
Figu e B.9: The g aphical use in e ace o DNASyn. The bu ons in he le panel
enable manual access o use -de ined mac o unc ions. The ex box a he igh shows
he code o he syn hesis sc ip loaded.
DNASyn was implemen ed in Ja aTM. I is unning wi h Windows XP P o essional (and
is also expec ed o wo k wi h Window98). The use wi h Windows XP Home o Windows
Vis a is no ecommended since hese ope a ing sys ems won’ allow di ec access o he
ha dwa e po s ia he ke nel mode d i e Use Po .
294
Technical No es on he Syn hesize Con ol So wa e DNASyn
B.7.1 Basic Fea u es
Manual
ope a ion
( ia GUI) au oma ic
mode
Syn hesis
sc ip
(syn hesis p ocedu e o
he pa icula DNAchip)
Mask display
(1024x768 XGA)
S anda d
mac os
( luidics pa ame e s e c.)
Fil e and Shu e
con ol
( ia pa allel po )
Mask iles
(jpeg images) G aphical Use In e ace
Syn hesis sc ip in e p e e
Fluids con ol
( ia se ial po )
DNASyn
Figu e B.10: Concep o he DNASyn mic oa ay syn hesis con ol so wa e.
DNASyn includes a lexible mac o p og amming language o he au oma ed con ol o
he syn hesis p ocess, and a g aphical use in e ace (GUI) o manual con ol o a ious
syn hesize unc ions (see Fig. B.9). The mac o language comp ises only a small numbe
o basic commands.
Keywo ds
START Begin o he main p og am
END End o he main p og am
MACRO mac oname {...}Mac o heade
PRINT n no e DNASyn shows ex no e in ou pu line n
// commen Commen in he sou ce code
WAIT n Wai o n seconds
VX Y Val e ope a ion X: al e numbe ; Y: 0=close 1=open
DISPLAY imagename.jpg Vi ual mask display
DISPLAY AGAIN Display he p e ious image again
SHUTTER ON/OFF Shu e con ol
FILTER GREEN/UV Fil e change con ol
•Swi ching o solenoid al es ( luidics ope a ions) is pe o med wi h he V X Y com-
mand.
•The DISPLAY imagename.jpg command loads he JPG image om he syn hesis di-
ec o y and shows i on he DMD. The keywo d AGAIN is used o eload he p e ious
image.
•The WAIT ncommand (ndu a ion in seconds) is used o ime con ol o he syn hesis
p ocesses.
295
Suppo ing In o ma ion
•Commen s begin wi h // ollowed by a space cha ac e .
Mac os
Typical ou ines (e.g. amidi e coupling o pho o-dep o ec ion) can be combined o mac o
commands, as shown in he ollowing example.
MACRO inse 20
{
//Rinse syn hesis cell wi h MeCN o 20 s - his is a commen
V 18 1
V 2 1
V13 1
V17 1
WAIT 20
V 2 0
V13 0
V17 0
V18 0
}
Mac o commands can be called om he main p og am and om wi hin o he mac os.
Manual con ol ( ia bu on-click in he con ol panel) is also based on mac o commands.
Mos con ol panel bu ons a e assigned a mac o unc ion. Mac o codes o hese unc ions
a e lis ed (and can be modi ied i necessa y) in he ile unc ions.p g.
A syn hesis p og am comp ises a lis o mac os (a lib a yo s anda dmac osandaddi ional
use -de ined mac os) and he main p og am. S anda d mac os desc ibe ou ine syn hesis
p ocesses. Basically hey a e no di e en om use -de ined mac os, bu since heyinclude
c i ical ime pa ame e s (du a ion o luidics p ocesses, exposu e imes e c.) and since hey
may be called om o he mac os, modi ica ions in s anda d mac os should be conside ed
cau iously. Upon loading a syn hesis p og am ( ile ex ension .p g) he pa se o DNASyn
ini ially eads he main p og am (be ween he commands START and END). In he nex
s ep mac o calls a e subs i u ed by he co esponding mac o codes. To conside nes ed
mac os his is epea ed un il all mac os a e esol ed. A comple ely esol ed syn hesis
p og am o a 25me a ay syn hesis ypically comp ises abou 40000 commands.
F equen ly used mac o unc ions
lush lush syn hesis cell wi h a gon
low X low eagen X h ough he syn hesis cell
inse X ill MeCN in o he s o age bo le o eagen X
296
Technical No es on he Syn hesize Con ol So wa e DNASyn
inse block inse al e block wi h MeCN
lush block lush al e block wi h a gon
p ime X ill he ube be ween he s o age bo le X
and he al e block wi h eagen X
e e se lush as lush o he syn hesis cell
wi h a gon in e e se di ec ion
dep o ec pho odep o ec ion
couple X coupling o he phosho amidi e X
oxidize oxidiza ion o phosphi e bonds
Numbe -ex ensions o he unc ions name (e.g. lush10) speci y he du a ion o he ope a-
ion (in seconds).
B.7.2 Communica ions be ween he Con ol PC and he Syn-
hesize Ha dwa e
Fo se ial communica ion wi h he solenoid al e con olle he Ja a Communica ions API
(Sun Mic osys ems) is employed. The communica ions pa ame e s ha e been se o he
equi emen s o he al e con olle (see below).
The con ol o he shu e and il e -change ia he pa allel po has been implemen ed wi h
a Ja a na i ecode. Di ec con ol o he pa allel po equi es he ja a packagepa po . The
lib a y pa po .dll needs o be ins alled in he di ec o y Sys ems32/d i e s. Wi h pa po
he channels o he pa allel po can be se and ead in a s aigh o wa d way. Fo di ec
access on he I/O po s (use mode) he d i e Use Po (w i en by Tomas F anzon) needs
o be ins alled ( o his pu pose Use po .sys needs o be copied o Sys em32/d i e s).
Possibly he Windows98 compa ibili y mode needs o be enabled. Wi h he execu able
Use po .exe he access o he pa allel po (base add ess $387) is se enabled.
B.7.3 Dual Sc een Suppo
DNASyn p o ides dual sc een suppo o display he con ol panel and he pho oli hog a-
phy mask pa e ns on di e en de ices - TFT moni o and ideo p ojec o (DMD), espec-
i ely. This equi es he use o a dual iew g aphics ca d and ex ension o he Windows
desk op on o he second display. The con ol panel is displayed on he p ima y sc een
(TFT-moni o wi h 1280×1024 pixels). Display o he pho oli hog aphy masks on he
seconda y display ( ideo p ojec o ) is achie ed by opening a window a he co espond-
ing desk op coo dina es - no u he p og amming icks a e necessa y. The Class Dis-
playF ame, an ex ension o he Ja a Class JWindow enables display o he masks wi hou
297
Suppo ing In o ma ion
Figu e B.15: The eadou g id is exac ly posi ioned on he mic oa ay ea u es.
A e aging o e he eadou boxes yields he hyb idiza ion signals o he indi idual
mic oa ay ea u es.
304
Tempe a u e Con ol o he Hyb idiza ion Chambe
B.11 Tempe a u e Con ol o he Hyb idiza ion
Chambe
DA0
DA1
DO0
DO1
DO2
DO3
DR0
DR1
DR2
DR3
RES
CNT
C7
C6
C5
C4
C3
C2
C1
C0
B7
B6
B5
B4
B3
B2
B1
B0
A7
A6
A5
A4
A3
A2
A1
A0
DI3
DI2
DI1
DI0
CH7
CH6
CH5
CH4
CH3
CH2
CH1
CH0
PMD1
PMD-1008
RUN
STP
E1
E2
E3
E4
YT1
Y( )
ND1
1.23
E0
E1 A
SUB1
-
Di e ence
be ween
se empe a u e
and ac ual
empe a u e
Manual
empe a u e
se ing
Ac ual empe a u e
display
E0
E1 A
MUL1
*
E0
E1 A
MUL2
*
E0
E1 A
MUL3
*
E
RST A
INT1
d(+ )
E
RST A
DIF1
d(-)
E0
E1
E2
A
ADD1
+
PID-Con olle
KP
KI
KD
RUN
STP
E1
E2
E3
E4
YT2
Y( )
ND2
1.23
ND3
1.23
SR1
ND4
1.23
Con ol ol age
E0
E1
A
MUL4
*
E0
E1 A
MUL5
*
FW1
W
FW2
W
Ain
Bin A< B
A= B
A> B
AVG1
Ve gl.
E0
E1
A
MUL6
*
Maximum empe a u e se ing
SR2 ND5
1.23
EN
IN
LiH
LiLClpL
ClpH
A
LIM1
Limi e
FW3
W
FW4
W
0...5V
SR3 ND6
1.23
LE1
Ain
Bin A< B
A= B
A> B
AVG2
Ve gl.
E0
E1 A
MUL7
*
ND7
1.23
ND8
1.23
LED1
T1
G1 E A
KT1
KT
UP
DN
RST
Z
CO1
0 0 0 1
T2
ND9
1.23
Tempe a u e p og am con ol
S1
AND1
&
E0
E1
SEL
A
REL1
S2
manual/au oma ic
selec o
E0A
FRM1
F
FW5
W
O se in V
PON1
R
un p og am
FW6
W
FW7
W
FW8
W
FW9
W
E1
E2
Add
RST
REC1
MWR
G2
AND2
&T3
T4 EXOR1
1
=
S3
Da a Reco de
T5
E0 A
FRM2
F
E0 A
FRM3
F
E A
KT2
KT
LED2
E A
MW1
MW
Signal smoo hing
Neues File
Ou pu : se - and ac ual empe a u e
In eg a o - ese
( o p e en s ong o e shoo )
E0 A
FRM4
F
E0 A
FRM5
F
Hea ing cu en limi e
Plo ing on/o
CK
RST
U/D
ENT
ENP RCO
Q3
Q2
Q1
Q0
ZBIN1
Zähle (4)
T6
a
b
c
d
e
g
a
b
c
d
e
g
EN
IN
S0
S1
S2
A7
A6
A5
A4
A3
A2
A1
A0
ADMX1
ADMX
S0
S1
S2
S3
g
e
d
c
b
a
Seg7-Dekode
FW10
W
P og am selec ion display
E A
KT3
KT
E A
KT4
KT
E A
KT5
KT
E A
KT6
KT
E A
KT7
KT
E A
KT8
KT
Meilhaus
Redlab
USB
Measu emen
Module
D/A ou
A/D in
p og am selec o
Mul iplexe
p og am ime
Tempe a u e p og am ables
ime s empe a u e se ing
To modi y empe a u e p og ams
make co ec ions he e
PID pa ame e
ou pu
Figu e B.16: So wa e-based PID- empe a u e con olle . Implemen a ion wi h P o iLab
Expe 3.0 (ABACOM GbR). The RedLab measu emen module (Meilhaus) is employed o
inpu /ou pu o analog signals. Tempe a u e can be se manual o in a p og am mode. P o-
g ams a e en e ed as ables (P o iLab-Func ion ”Ko ek u abelle”) o ime e sus empe a u e
( ecompila ion necessa y). Be ween wo successi e empe a u e se -poin s he empe a u e is
a ied linea ly. The empe a u e con olle applica ion is un on he ”mic oscope con ol PC”
in pa allel wi h he image acquisi ion-so wa e SimplePCI (Compix Inc.).
305
Suppo ing In o ma ion
B.12 cRNA Seconda y S uc u es
∆G37
o= -301.8 kcal/mol
Minimum ee ene gy
seconda y s uc u e
5'
3'
U
A
U
A
A
G
C
A
G
A
GC
U
G
G
U
U
U
A
G
U
G
A
A
C
C
G
UCA
G
A
U
C
C
G
C
U
A
G
C
G
C
U
A
C
CG
G
U
C
G
C
C
A
C
C
A
U
GG
U
G
A
G
C
A
A
G
G
G
C
G
A
G
G
A
G
C
U
G
U
U
C
A
C
C
G
G
G
G
U
G
G
U
G
C
C
C
A
U
C
C
U
G
G
U
C
G
A
G
C
U
G
G
A
C
G
G
C
G
AC
G
U
A
A
ACGGCC
A
CAA
GUUCAGC
GUG
UCCGG
CGAGGGC
G
A
GG
G
C
G
A
U
G
C
C
A
C
C
U
ACGGCAA
G
C
U
GAC
CC
U
G
A
A
G
U
U
C
A
U
C
U
G
C
A
C
C
AC
C
G
G
C
A
A
GC
U
G
C
C
C
G
U
G
C
C
C
U
G
G
C
C
C
A
C
C
C
U
C
G
U
G
A
C
C
AC
C
C
U
G
A
C
C
U
A
C
G
G
C
G
U
G
C
A
G
U
G
C
U
U
C
A
G
C
C
G
C
U
A
C
C
CCGACC
A
C
A
UGAAGCAG
C
ACG
A
C
U
U
C
U
UC
AAGUCC
G
C
C
A
U
G
C
C
C
GAA
G
G
CUAC
G
U
C
C
A
G
G
A
G
C
G
C
ACC
A
U
C
U
U
C
U
UC
A
A
G
G
A
C
G
A
C
G
G
C
A
ACUAC
A
A
G
A
CC
C
G
C
G
C
C
G
A
G
G
UGAA
G
U
U
C
G
A
G
GG
C
G
A
C
A
C
C
CU
G
G
U
GA
A
C
C
G
C
A
U
C
G
A
G
C
U
G
A
A
G
G
G
CAUC
G
A
C
U
U
C
AAG
G
A
G
GA
C
G
G
C
AACA
U
C
C
U
G
G
G
G
C
A
CAAG
C
U
G
G
AG
U
A
CAA
C
U
A
C
A
A
C
A
G
CC
AC
A
AC
G
UCU
AU
AU
C
A
U
G
G
C
C
GA
C
A
A
G
C
A
GAA
G
A
A
C
G
GCA
U
C
A
A
G
G
U
G
A
A
C
U
U
C
AAG
A
U
C
C
G
C
C
A
CAA
C
A
UCGAGGACGGCAGCGUG
CAGCUCGCCGA
C
C
ACUACCA
G
C
A
G
A
A
C
A
C
C
C
C
C
A
U
C
G
G
C
G
A
C
G
G
C
C
C
C
G
U
G
C
U
G
C
U
G
C
C
C
G
A
C
A
A
C
C
A
C
UA
C
C
U
G
A
G
C
ACCC
A
G
U
C
C
G
C
C
C
UG
A
G
CAA
A
G
A
C
C C C
A
A
C
G
A
G
A
AGCGC
GAUCA
C
A
U
G
G
U
C
C
U
G
C
U
G
G
A
G
U
U
C
GU
G
A
C
C
G
C
C
G
C
C
G
G
G
AU
C
A
C
U
C
U
C
G
G
C
A
U
G
G
A
C
G
A
G
C
U
G
U
A
CAAG
U
C
C
G
G
A
C
U
C
A
GAU
C
U
C
G
A
G
U
G
C
G
U
G
A
G
U
G
C
A
U
C
UC
C
A
U
C
C
A
CG
U
U
G
G
C
C
A
G
25
50
75
100
125
150
175
200
225
250
275
300
325
350
375
400
425
450
475
500
525
550
575
600
625
650
675
700
725
750
775
800
825
Figu e B.17: The minimum ee ene gy (MFE) seconda y s uc u e o he eGFP
cRNA a ge sequence T2 – see sec ion 8.6.2 – was calcula ed on he S old web se e
[Din04]. Owing o in as and base pai ing la ge pa s o he sequence a e una ailable
o hyb idiza ion o DNA mic oa ay p obes. The base numbe ing 1 o 825 co e-
sponds o bases 556 o 1380 o he eGFP-Tub plasmid sequence (see sec ion 8.6.2).
Compa e wi h he cen oid s uc u e in Fig. B.18. G een do s ep esen base pai s
common in he MFE and cen oid s uc u es. Blue do s ep esen base pai s p esen
only in he MFE s uc u e.
306
cRNA Seconda y S uc u es
∆G37
o= -200.66 kcal/mol
Ensemble Cen oid
5'
3'
U
A
U
A
A
G
C
A
G
A
G
C
U
G
G
U
U
U
A
G
U
G
A
A
C
C
G
U
C
A
G
A
U
C
C
G
C
U
A
G
C
G
C
U
A
C
C
G
G
U
C
G
C
C
A
C
C
A
UG
G
U
G
A
GC
A
A
G
G
G
C
G
AGGA
G
C
U
G
U
U
C
A
C
C
G
G
G
G
U
G
G
U
G
C
C
C
A
U
C
C
U
G
G
U
C
G
A
G
C
U
G
G
A
C
G
G
C
G
A
C
G
U
A
A
ACGGCC
A
CAA
GUUCAGC
G
U
G
UCC
G
G
C
G
A
G
G
G
C
G
A
G
G
G
C
G
A
U
G
C
C
A
C
C
U
ACGGCA
A
G
C
U
G
A
C
C
C
U
G
A
A
G
U
U
C
A
U
C
U
G
CACCACCGGC
A
A
GCU
GCCCGUGCCCUGGCC
C
A
C
CCUC
G
U
GACC
A
C
C
CU
G
A
C
C
U
A
C
G
G
C
G
U
G
C
A
G
U
G
C
U
U
C
A
G
C
C
G
C
U
A
C
C
CCGACC
A
C
A
UGAAGCAG
C
ACG
A
C
U
U
C
U
UC
A
AGUCC
G
C
C
A
U
G
C
C
C
GAA
G
G
CUAC
G
U
C
C
A
G
G
A
G
C
G
C
ACC
A
U
C
U
U
C
U
UC
A
A
G
G
A
C
G
A
C
G
G
C
A
ACUAC
A
A
G
A
CC
C
G
C
G
C
C
G
A
G
G
UGAAGUUC
G
A
GGGCGA
CACCCUGGUGAACC
GC
A
U
C
G
A
G
C
U
G
A
A
G
G
G
CAUCGACU
U
C
A
A
G
G
A
G
G
A
CG
GCA
A
C
A
U
C
CU
G
G
G
GC
A
C
A
AG
C
U
G
G
AG
U
A
CAA
C
U
A
C
A
A
C
A
G
C
CACAACGUCUAUAUCAUGGCCGACAAGCAGAAGAA
C
G
G
C
A
U
C
A
A
G
G
U
G
A
A
C
U
U
C
A
A
G
A
U
C
C
G
C
C
A
C
A
A
C
AUCGAGGACGGCAGC
GUG
CAGC
UCGCCG
A
C
CACUACCA
G
C
A
G
A
A
C
A
C
C
C
C
C
A
U
C
G
G
C
G
A
C
G
G
C
C
C
C
G
U
G
C
U
G
C
U
G
C
C
C
G
A
CAACCACUACCUGAGCACCCAGUCCGCCCUGAGCAAAGACCCC
A
A
C
G
A
G
A
A
G
C
G
C
G
A
U
C
A
C
A
U
G
G
U
C
C
U
G
C
U
G
G
A
G
U
U
C
G
U
G
A
C
C
G
C
C
G
C
C
G
G
G
AU
C
A
C
U
C
U
C
G
G
C
A
U
G
G
A
C
G
A
G
C
U
G
U
A
C
A
A
G
U
C
C
G
G
A
C
U
C
A
G
A
U
C
U
C
G
A
G
U
G
C
G
U
G
A
G
U
G
C
A
U
C
U
C
C
A
U
C
C
A
C
G
U
U
G
G
C
C
A
G
25
50
75
100
125
150
175
200
225
250
275
300
325
350
375
400
425
450
475
500
525
550
575
600
625
650
675
700
725
750
775
800
825
Figu e B.18: Cen oid seconda y s uc u e [Din05] o he eGFP cRNA a -
ge sequence T2. The cen oid s uc u e was calcula ed on he S old web se e
[Din04; Cha05] om a Bol zmann-weigh ed s uc u e ensemble. ”The cen oid s uc-
u e can be conside ed as he single s uc u e ha bes ep esen s he cen al endency
o he se ” [Cha05]. Compa e wi h he minimum ee ene gy seconda y s uc u e in
Fig. B.17. G een do s ep esen base pai s common in he MFE and cen oid s uc-
u es. Red do s ep esen base pai s p esen in he cen oid s uc u e, no howe e
in he MFE s uc u e.
307
Suppo ing In o ma ion
B.13 3-D Visualiza ion o Nucleic Acid S uc u es
Figu e B.19: B-DNA s uc u e - s e eo iew (use c oss-eye- echnique o 3D e ec ).
S e eo images o he ideal B-DNA s uc u e we e c ea ed wi h UCSF Chime a.
Figu e B.20: A-RNA s uc u e - s e eo iew (use c oss-eye- echnique o 3D e ec ).
S e eo images o he ideal A-RNA s uc u e we e c ea ed wi h UCSF Chime a.
308
3-D Visualiza ion o Nucleic Acid S uc u es
Figu e B.21: Top iews o he helix s uc u es - B-DNA (le ) and A-RNA ( igh )
- demons a e signi ican di e ences in base s acking
309
Suppo ing In o ma ion
310
Lis o Publica ions
Lis o Publica ions
T. Naise , T. Mai, W. Michel, A. O . A e sa ile maskless mic oscope p ojec ion pho-
oli hog aphy sys em and i s applica ion in ligh -di ec ed ab ica ion o DNA mic oa ays.
Re iew o Scien i ic Ins umen s, 77(6): 063711, 2006.
W. Michel, T. Mai, T. Naise , A. O . Op ical s udy o DNA su ace hyb idiza ion e eals
DNA su ace densi y as a key pa ame e o mic oa ay hyb idiza ion kine ics.
Biophysical Jou nal, 92(3):999-1004, 2007.
T. Naise , O. Ehle , J. Kayse , T. Mai, W. Michel, A. O . Impac o poin -mu a ions on
he hyb idiza ion a ini y o su ace-bound DNA/DNA and RNA/DNA oligonucleo ide-
duplexes: compa ison o single base misma ches and base bulges.
BMC Bio echnology 2008, 8:48
Submi ed manusc ip s
T. Naise , J. Kayse , T. Mai, W. Michel, A. O . DNA hyb idiza ion o su ace bound
p obes: poin de ec s in expe imen and model.
Submi ed o Phys. Re . Le .
T. Naise , J. Kayse , T. Mai, W. Michel A. O . Poin de ec s and he s abili y o su ace
bound oligonucleo ide duplexes - expe imen s and model.
Submi ed o Biophysical Jou nal
311
Lis o Publica ions
312
Danksagung
Danksagung
Als e s es m¨och e ich mich bei meinem Dok o a e , He n P o . D . Alb ech O , ¨u die
g oßa ige Be euung meine Dok o a bei bedanken. Vielen Dank Alb ech , ¨u die ausge-
sp ochen eundscha liche Zusammena bei mi Di . Danke ¨u die g oßen F ei ¨aume die
Du mi bei de Ausges al ung de o liegenden A bei gew¨ah has , und auch da ¨u , dass
es p ak isch jede zei m¨oglich wa Dich um Ra zu agen und mi Di wissenscha liche
P obleme zu e ¨o e n.
Bei meinen Mi -Dok o anden Timo Mai, Pablo Fe nandez, Ha ish Bokkasam, J´e ome
Goidin, und ganz besonde s bei Wol gang Michel, mi dem ich ¨ube die e gangen Jah e
das B¨u o ge eil habe, m¨och e ich mich eben alls seh he zlich ¨u die eundscha liche
Zusammena bei und die angenehme und an egende A bei sa mosph¨a e in unse e A bei s-
g uppe bedanken. Diese Dank gil ebenso P amod Pulla ka und Jo di So iano-F ade a,
die mi beide als Pos docs w¨ah end de e gangen Jah e s e s mi iel Ra und Ta zu Sei e
s anden, und ein wich ige Quell de Mo i a ion ¨u mich wa en – und siche imme bleiben
we den. Wa ’ne olle Zei mi Euch! Zum gu en A bei sklima in de A bei sg uppe haben
auch die Diplomanden Oli e Ehle , Jona Kayse , Benjamin T ¨ankle und Philipp Baaske
ih en Bei ag geleis e . Die gesellige Zei bei unse en EP1-Ka eepausen – zusammen
mi Michael K¨uken, E nes o Nicola, S en R¨udige , Snigdha Thaku , Cy il Colombo und
Robe o Be nal – we d’ ich nie e gessen F eunde!
Du ch ih e kompe en e echnische Un e s ¨u zung bei molekula biologischen, so wa e ech-
nischen und elek onischen P oblemen haben auch Tobias Mumme , And ea Hanold, Ral
Pihan und Paul Hu ych ganz wesen lich zum Gelingen diese A bei beige agen. Hab ie-
len Dank! Auch J¨u gen Gmeine om Leh s uhl EP II m¨och e ich ¨u seine Un e s ¨u zung
in F agen de o ganischen Chemie an diese S elle meinen Dank aussp echen. F¨u eine
he aus agende echnische Un e s ¨u zung m¨och e ich mich auch bei He n K ej schi und
seinem Mechanikwe ks a -Team bedanken.
Besonde s he zliche Dank gil de gu en Seele des Leh s uhls Ma go Lenich ¨u eine olle
Un e s ¨u zung in adminis a i e Hinsich , abe auch ¨u iele ne e Un e hal ungen zwis-
chendu ch. Danke Ma go !
F¨u ih en Bei ag zu gu en A bei sa mosph¨a e am Leh s uhl EP I m¨och e ich mich auch
bei He n D . Uwe Schmelze , He n P o . D . Pasche und seinen Mi a bei e n - meinen
F eunden und Kollegen - Jens F¨u s , Wol gang Kellne , Ral Lang und And eas Win e
seh he zlich bedanken.
Bei P o . D . F ank J¨uliche m¨och e ich mich da ¨u bedanken, dass e mi einen meh mona i-
gen Gas au en hal in seine A bei sg uppe am MPIPKS in D esden, und somi in e essan e
313
