UNIVERSIT Y O F COPENHAGEN
F A C U L T Y O F H E A L T H A N D M E D I C A L S C I E N C E S
Dissec ing Glycosyl ans e ase
Ac i i y in Human Mesenchymal S em
Cells o En o ce Cellula Mig a ion
Mas e ’s Thesis, Medicine (12,5 ECTS)
Amalie Dahl Haue
Supe iso : Hans H. Wandall
Submission Da e: Sep embe 12 h 2016
Name o depa men : Depa men o Cellula and Molecula Medicine
P ima y si e o ac i i y: Copenhagen Cen e o Glycomics
Au ho : Amalie Dahl Haue, BSc, BA
KU Use name: hq750
Ti le: Dissec ing Glycosyl ans e ase Ac i i y in Human
Mesenchymal S em Cells o En o ce Cellula Mig a ion
Supe iso : P o esso Hans H. Wandall, MD, PhD
Submission da e: Sep embe 12 h 2016
Wo d coun : 5.825
The p o ided da a concludes my ecen esea ch aining as a summe s uden .
Resea ch aining pe iod: June 8 h 2016 – Augus 19 h 2016
Resea ch aining place: Labo a o y o D . Robe Sacks ein
77 A enue Louis Pas eu , Bos on, MA 02115, USA
___________________________________________________________________
Ti le page illus a ion has been c opped om EMBL Symposium Pos e . Symposium held 5 Decembe 2011 in Heidelbe g, Ge many.
i
P eamble
The s udy o glycans da es back o Emil Fische ’s wo k sys ema izing he desc ip ion
o hexose suga s by making ac ual 3D-s uc u es accessible in a 2D- o ma . Today,
oughly a cen u y la e , he ield o glycobiology is p omising as a diagnos ic and
he apeu ic ool. Yea s o in ensi e and sys ema ic s udy ha e e en gi en ise o he
e m glycomedicine, demons a ing he g ea p og ess and p omising u u e o he
ield.
Thanks o combined insigh , e o and hospi ali y o D s. Robe Sacks ein, Di ec o
o P og am o Excellence in Glycosciences and Hans H. Wandall, P o esso o
Glycomedicine, I was e y p i ileged o spend The Summe o 2016 as a esea ch
ainee in The Labo a o y o D . Robe Sacks ein. The pu pose o his epo is o
p esen and e alua e my esea ch aining expe ience.
Fo mally, he epo will cons i u e The Mas e ’s Thesis o my Medical Deg ee.
The expe imen s ha I conduc ed du ing my ecen esea ch aining we e in
acco dance wi h D . Sacks ein’s he apeu ical mission, including weekly
consul a ions and w i en consen ag eeing o his in ellec ual p ope y. Pos Doc
Nandini Mondal pe o med ou inely supe ision and labo a o y manage Kyle C.
Ma in p o ided echnical as well as heo e ical assis ance.
S ud.med. Amalie Dahl Haue
Panumins i u e , Københa n
Den 11. sep embe 2016
ii
Resumé
Fo måle med næ æ ende opga e e a p æsen e e min e a ing som medicin-
s ude ende og somme gæs i e o skningslabo a o ium il enh e med medicinske
in e esse , in en ione og isione i e o ma de op ylde “Recommenda ions o he
Conduc , Repo ing, Edi ing and Publica ion o Schola ly Wo k in Medical
Jou nals”.
Læ ingsmæssig ha de opholde o komponen e , h o a den ene a a unde søge en
speci ik kulhyd a s uk u og den anden a a ud ide mi molekylæ biologiske
me odekendskab.
Den speci ikke kulhyd a s uk u kaldes he sialyal ed lewis x binding de e minan , og
be egne en s uk u de indgå i den klassiske besk i else a cellulæ mig a ion unde
så el no mal ysiologiske o hold som sygdomsmæssige p ocesse .
His ologisk e den ud yk i adskillige humane æ . Dog e den ikke ud yk i humane
mesenkymale s amcelle . Ren opologisk e he sialyal ed lewis x binding
de e minan ud yk på mange p o eine he unde CD44, h ilke e e p o ein de
g unde de s gennemg ibende kompleksi e e in e essan a s ude e i biologiske og
ysiologiske sys eme . Ud idelse a mi molekylæ biologiske me odekendskab
in ol e ede indsig i o in i o me ode . Disse me ode e henholds is p ak isk
hånd e ing a humane mesenkymale s amcelle sam enzyma isk behandling a in ak e
celle , he unde også humane mesenkymale s amcelle . Sids næ n e o egik ed a
induce e glykosyle ingen a memb anbundne p o eine ed an endelse en me ode, de
be egnes exo ucosyla ion.
Opholde s æsen ligs e ekspe imen elle und a a ed o søg på a slukke o i e
indi iduelle glycosyl ans e ase i humane mesenkymale s amcelle a i s and il a
de ek e e disk e e æno ype .
Opga en e inddel i em sek ione . Den ø s e del indeholde en gene el in oduk ion
il glykobiologi. Den anden sek ion o binde den gene elle glykobiologi med de
ekspe imen elle a bejde, som jeg ud ø e unde mi ophold. Den edje sek ion
p æsen e e og disku e e de da asæ , jeg indsamlede unde opholde . I den je de
sek ion ems illes en syn ese a ho edkonklusione ne. Den em e og sids e sek ion
p æsen e e en de alje ede besk i else a henholds is ma e iale og me ode , som jeg
kons ue ede og a bejdede med unde opholde .
iii
Abs ac
The aim o he p esen hesis is o communica e my summe esea ch aining
expe ience as a Medical S uden o anyone wi h an in en ion o ea and a ision o
p e en in a o ma he s i es o ul ill he “Recommenda ions o he Conduc ,
Repo ing, Edi ing and Publica ion o Schola ly Wo k in Medical Jou nals”.
The wo main componen s o he esea ch aining expe ience we e (i) o s udy one
pa icula classic glycan modi ica ion and (ii) o expand my echnical expe imen al
epe oi e.
The glycan s uc u e in ques ions is he sialyla ed lewis x binding de e minan , which
is a s uc u e known o acili a e cellula mig a ion in heal h and disease.
His ologically, i is exp essed in many issues o man, bu i is cha ac e is ically
absen om human mesenchymal s em cells. Topologically, he sialyla ed lewis x
binding de e minan is exp essed on many p o eins, including CD44, a p o ein o
g ea biological in e es owing o i s complexi y.
Expansion o my echnical expe imen al epe oi e coun ed wo, namely wo king wi h
human mesenchymal s em cells and in i o enzyma ic p epa a ion o in ac cells, by
en o ced cell su ace glycosyla ion bene i ing om me hod known as
exo ucosyla ion.
Scien i ically, I was able o de ec only modes pheno ypes by knockou o ou
indi idual glycosyl ans e ase genes exp essed in human mesenchymal s em cells.
The hesis has i e sec ions. The i s sec ion is a gene al in oduc ion o glycobiology
wi h bo h a s uc u al and a unc ional ocus. The second sec ion links gene al
glycobiology wi h he expe imen al wo k I conduc ed as a esea ch ainee. The hi d
sec ion p esen s and e alua es he da a ha I collec ed du ing my aining. A syn hesis
o he main conclusions will be p esen ed in sec ion ou . The i h and inal sec ion
p esen s a de ailed desc ip ion o he ma e ials and me hods ha I p epa ed and
wo ked wi h du ing my s ay.
1
CONTENTS
THEORETICAL INTRODUCTION 2
GLYCOBIOLOGY AT A GLANCE 2
GLYCANS AT WORK: THE MULTI-STEP PARADIGM OF CELLULAR MIGRATION 3
BRIDGE 5
EXPERIMENTAL MOTIVATION 5
THE SIALOFUCOSYLATED LACTOSAMINE DETERMINANT 5
EXPERIMENTAL OBJECT 7
RESULTS 8
DISCUSSION 11
CONCLUSION 13
MATERIALS AND METHODS 15
HUMAN MESENCHYMAL STEM CELLS (MSCS) 15
CHARACTERIZATION OF GLYCOSYLTRANSFERASES EXPRESSED IN HUMAN MSCS 15
CONSTRUCTION OF VECTORS FOR TRANSFECTION 16
TRANSFECTION OF MSCS 16
ENZYMATIC PREPARATION OF HUMAN MSCS 17
CHARACTERIZATION OF TRANSFECTED MSCS BY PCR ON GENOMIC DNA 17
CHARACTERIZATION OF TRANSFECTED MSCS BY FLOW CYTOMETRY 17
LITERATURE 19
2
Theo e ical In oduc ion
Glycobiology a a Glance
Glycosyla ion is a pos - ansla ional modi ica ion p esen on memb ane bound
p o eins and lipids o bo h p o- and euka yo es and a e ca ego ized depending on hei
ela ion o he p o ein backbone o lipid s uc u e. The glycop o ein amily coun s N-
glycosyla ion, O-glycosyla ion and glycosaminoglycans cha ac e is ically displaying
epea ing disaccha ide uni s. Glycosaminoglycans a e also quali a i ely di e en om
he o me wo, as hey exis bo h indi idually and p o ein bound. Also, glycosamino-
glycans a e linea and syn hesized by a dis inc class o enzymes. As wi h glyco-
p o eins, glycolipids can be modi ied by he e minal glycan sialic acid. I a glycolipid
is sialyla ed i is called a ganglioside (Oh subo & Ma h, 2006). The enzyma ic
machine y esponsible o he c ea ion o N- and O-glycosyla ion esides in he Golgi-
appa a us. Phylogene ically, hey all belong o he glycosyl ans e ase amily which
occupies app oxima ely 1% o he human genome (Rini J, 2009).
The glycosyl ans e ases unc ion in a s ep-wise ashion, whe e indi idual suga
moie ies o polysaccha ides a e added o he non- educing end o he de eloping
glycop o ein o -lipid. A a molecula le el glycosyla ion is unique in ha he
s uc u es a e no di ec ly ansla ed om he genome. In o he wo ds, i is nei he
possible o p edic a glycan s uc u e om an amino acid sequence, no possible o
de i e he enzyma ic cascade ha p ecedes a speci ic glycan s uc u e. This
in e dependency be ween he molecules p esen in he mic o milieu and he e ol ing
glycan s uc u e is e med mic ohe e ogenei y. Func ionally, pos - ansla ional glycan
modi ica ions ha e a subs an ial impac as ac o s in biological sys ems, as hey may
be media o s o nea ly all biological p ocesses including cell adhesion, molecula
a icking and clea ance, ecep o ac i a ion, signal ansduc ion and endocy osis
(Oh subo & Ma h, 2006).
Figu e 1. Common classes o glycans: Modi ied om Va ki A. 1997. FASEB J.
Downloaded om:
h p://www.ncbi.nlm.nih.go /books/NBK1931/ igu e/ch1. 6/? epo =objec only
3
Classical examples o glycop o eins ha media e cell adhesion a e he in eg ins
de ined by hei RGD-binding mo i (A naou , 1990). Func ionally, i has been
s ipula ed and e iden ially suppo ed ha glycans a e implica ed in molecula
a icking and clea ance by a a ian o endocy osis, ha is cha ac e is ically
cla h in-independen (S anley, 2014). A classic example o glycans in ol ed in
ecep o ac i a ion s ems om s udies o he Ashwell-Mo ell ecep o , which belongs
o he amily o asialoglycop o ein ecep o s (G ewal e al., 2008). O he examples
ha mani es he biological signi icance o glycosyla ion in signal ansduc ion has
e ol ed om s udies o he No ch ecep o , which is a ansmemb ane p o ein wi h
inhe en ca aly ic ac i i y acili a ing homophilic ecep o in e ac ions (Takeuchi &
Hal iwange , 2014).
Glycans a Wo k: The Mul i-S ep Pa adigm o Cellula Mig a ion
In a clinical-physiological con ex , he signi icance o glycans da es all he way back
o Ka l Lands eine s disco e y o ed blood cells abili y o agglu ina e and
subsequen ly mapping o he AB0-blood g oup sys em (Bayne-Jones, 1931). To da e,
he ield o glycobiology is apidly e ol ing as an empi ical pla o m wi h diagnos ic
and he apeu ical pe spec i es bene i ing om sophis ica ed and con inuously
e ol ing gene ic enginee ing echnologies (Galluzzi e al., 2014; Hudak & Be ozzi,
2014).
The mul i-s ep pa adigm o cellula mig a ion is an example o a physiological model
ha is s ic ly dependen on ecep o signaling in ol ing glycan s uc u es.
Speci ically, i in ol es he in e ac ion o selec ins exp essed on he endo helium and
he sialo ucosyla ed lac osamine binding de e minan (in oduced in g ea e de ail
below) exp essed on ci cula ing cells, mainly leukocy es. Indeed, he in e ac ion o
ci cula ing cells wi h he endo helial bed is a ubiqui ously physiological phenomenon.
Examples include cellula homing, a concep o iginally in oduced o explain he
obse ed lux o lymphocy es be ween blood and lymph (Sacks ein, 2012) and
con e sely ec ui men o neu ophils om he ma ow o si es o in ec ion in
sys emic ci cula ion (Bo egaa d, 2010).
O iginally, he mul i-s ep model o cellula mig a ion was de eloped o desc ibe he
ex a asa ion o leukocy es o he ascula u e as a physiological componen o s ess
(Sp inge , 1994). Du ing emb yogenesis, he ecep o -ligand in e ac ions o selec ins
and he sialo ucosyla ed lac osamine exp essed on ci cula ing cells ope a e o o m
cell-cell adhesion esponsible o he coloniza ion o lymphoid o gans by
lymphocy es. Addi ionally, he sialo ucosyla ed lac osamine is an impo an media o
o he in lamma o y esponse as well as a s uc u al ac o o a malignan hallma k,
namely me as asis. In ac , he selec ins ha e been asc ibed he p ima y unc ion o
glycan ecogni ion in mammalian immune unc ion (Ma h & G ewal, 2008).
4
O iginally, h ee consecu i e s eps de ined he mul i-s ep pa adigm o cellula
mig a ion. I is in he i s s ep ha he binding o he sialo ucosyal ed lac osamine
and he selec ins is p ominen . O he impo an media o s o cellula mig a ion he
chemokines (e.g. TNF- and IL-1) and in eg ins (e.g. LFA-1 and VLA-4), ope a ing
chemoa ac i ely h ough G-p o ein-coupled ecep o s and binding be ween opposing
ansmemb ane ype I p o eins, espec i ely (Sp inge , 1994).
Figu e 2. The mul i-s ep model o cellula mig a ion: Sacks ein R, Immun. Re iew, 2009.
11
MFI was educed o 38%, 26% and 48% o baseline le el in knockou o ST3GalTIII,
-IV and –VI, espec i ely. Thus, he da a sugges s ha ST3GalTIV is he majo
sialyl ans e ase in gene a ion o he sLex de e minan in human MSCs ( igu es 1.3
and 1.4). Explici ly knockou o ST3GalTIV yields he la ges HECA-452 nega i e
popula ion ( igu e 1.4, le panel).
Figu e 1.4. De ec ion o cell su ace sialyla ed and unsialyla ed lac osamines in ST3GalT knockou s.
Le panel: Exo ucosyla ion wi h FT6, hen s aining wi h HECA-452 mAb, speci ic o sLex
Righ panel: Exo ucosyla ion wi h FT6, hen s aining wi h an i-CD15 mAb, speci ic o Lex
MFI: Mean luo escence in ensi y
Discussion
The esul s p o ide e idence ha mul iple enzymes a e in ol ed in he syn hesis o
sLex binding de e minan in human MSCs. The da a sugges s, ha he e is no a one o
one ela ion be ween a pa icula s uc u e (i.e. he disaccha ide bond) and a pa icula
enzyme (i.e. a glycosyl- o sialyl ans e ase). The e o e, a numbe o biological
egula o y mechanisms may explain he disc e e pheno ypes. In addi ion, he ac ha
a numbe o conce ning echnical challenges emain unse led, complica e a igo ous
analysis. The biological as well as echnical conside a ions will be co e ed in g ea e
de ail below.
The echnical challenges a e pa ly due o he igo ous dependency on app op ia e
con ols. We did no , o ins ance ha e an app op ia e con ol in de ec ion o he ype
2 lac osamines by ECL ( igh panel, igu e 1.2). Consequen ly, he da a does no
assess i s aining wi h ECL o human MSCs co ela es wi h he expec ed ECL
speci ici y. Howe e , he da a sugges s ha na i e human MSCs p esen wi h ype 2
lac osamine binding epi opes, as hey a e ECL posi i e ( igh panel, igu e 1.2).
Co espondingly, na i e MSCs a e CD15 nega i e ( igh panel, igu e 1.3). In e ec ,
he majo i y o ype 2 lac osamine epi opes on human MSCs a e un ucosyla ed. Thus,
12
his piece o da a co ela es wi h he ac ha na i e human MSCs exp ess no (1,3)
ucosyl ans e ases (addendum om D . Sacks ein). Mo eo e , as an i-CD15 mAb
ecognizes he unsialyla ed ype 2 lac osamine (Lex), he da a unde sco es ha na i e
human MSCs a e highly sialyla ed (discussed in g ea e de ail below).
Ano he echnical challenge is ha i is no o iously di icul o assess speci ically how
enzyma ic p epa a ion o cells a ec s he cellula pheno ype including iabili y.
Explici ly, p io o de ec ion o he sLex binding de e minan , cells we e
exo ucosyla ed and p io o de ec ion o he Lex binding de e minan , cells we e
neu aminidase ea ed. The e o e, he h ee epo e s (i.e. HECA-452, ECL and an i-
CD15 mAb) a e no eadily compa able.
The cu en s a egy in dissec ion o he glycsosyl ans e ase edundancy is s ic ly
dependen on con inuous in i o a ailabili y o human MSCs. In e ec o being a
p ima y human cell line, he MSCs p esen a e a majo expe imen al oppo uni y, as
hey closely esemble he heo e ical basis o he enzyma ic dissec ion. Howe e ,
p ac ically i is no he mos e icien way o assay cell su ace glycosyla ion as
human MSCs a e a ely passed o mo e han a mon h in i o (ex i o). Thus mos
likely, es ablishmen o a s able knockou cell line is es ic ed by he limi ed
eplica ion po en ial o MSCs in cul u e.
In o de o ex end he in i o eplica ion po en ial o human MSCs, we a emp ed o
op imize he g ow h condi ions by subs i u ing he media componen o e al bo ine
se um (FBS) o pla ele lysa e. In casual e alua ion by di ec ligh mic oscopy o e a
pe iod be ween 24 and 48 hou s, we obse ed al e ed mo phology mani es ed as
smalle cells and less spindle o ma ion accompanied by accele a ed g ow h.
Besides, a conside able limi o he quali y o he da a is he quan i y o cells ha we e
included o analysis. All indi idual his og ams ep esen coun s o less han 5.000
cells. Fo compa ison, i is epo ed ha a cell popula ion o 1 x 106 indi idual cells is
analyzed o publica ion (Dyks a e al., 2016).
The modes pheno ype in he 4GalT1 knockou popula ion is a concei able
consequence o he echnical challenges ou lined abo e o a mani es a ion o
glycosyl ans e ase edundancy. I may e en be a combina ion o bo h.
Taken oge he , he da a sugges s ha enzyma ic edundancy accoun o lack
pheno ype in he 4GalT1 knockou s as an isoenzyme, namely 4GalT2 is exp essed
in human MSCs ( igu es 1.1 and 1.2). A suppo ing obse a ion is ha he p esen
da a se iden i ies ha he exp ession le el o 4GalT2 is simila o he exp ession
le els o ST3GalTIII and –IV ( igu e 1.2) one o which, p esumably, accoun s o
c ea ion o mos sialyla ed ype 2 lac osamines in human MSCs ( igu e 1.4).
In sum, enzyma ic dissec ion o sialyla ed cell su ace glycans is in e nally cohe en
as MFI in non- ans ec ed human MSCs is equal in he wo un ela ed measu emen s
(le panels o igu es 1.1 and 1.2, espec i ely). Fu he , he da a sugges s ha human
MSCs a e highly sialyla ed, as s aining o na i e MSCs wi h HECA-452 is posi i e
(le panel, igu e 1.4). The cu en model sys em does no disc imina e he speci ic
13
backbone o he adjacen glycan s uc u es. The e o e, he sialic acid may be
p edominan ly exp essed on ei he N-linked, O-linked glycans o glycolipids. Fo
e e ence (da a no included), he abundance o sialic acid on human MSCs was
assessed speci ically by s aining wi h indi ec ly FITC conjuga ed Maackia Amu ensis
Lec in II (MAL II), a lec in ha speci ically ecognizes (2,3) sialic acids.
Consis en ly, epo ing wi h MAL II sugges s ha human MSCs a e highly sialyla ed.
Knockou o he indi idual sialyl ans e ase genes educes he MFI o 85%, 43% and
69% (ST3GalTIII, -IV and -VI, espec i ely) o baseline le el co esponding o MAL
II s aining o non- ans ec ed human MSCs.
Based on he wid h o he his og ams, he da a p esen s an unexpec ed a ia ion in
assessmen o he sLex and Lex binding de e minan s, espec i ely. Possibly, he
a ia ion e lec s cellula senescence, as he wo se s o expe imen s we e conduc ed
wo weeks apa ( igu e 1.3). Fu he mo e, li e a u e epo s ha MSCs exp ess a β-
galac osidase ac i i y ela ed o senescence (see Ma e ials & Me hods and igu e 0.3
o de ails). Thus, he p esence o lac osamine de e minan s dec eases as cell age and
a such may accoun o an al e ed pheno ype un ela ed o gene ic enginee ing.
The key s ep in es ablishmen o a sensi i e e alua ion o glycosyl ans e ase ac i i y
in human MSCs consis s in op imiza ion o he ans ec ion p ocedu e. Fo
en ichmen s a egies, in oduc ion o a g een luo escence p o ein (GFP) ag on he
cas9 endonuclease enables en ichmen based on luo escen ac i a ed cells so ing
(FACS) me hodologies. Al hough he echnique has been success ully epo ed (Yang
e al., 2015), i does no enable sus ained moni o ing o a ge gene exp ession le el as
he cas9 endonuclease is only ansien ly exp essed. In con as , con inuous
moni o ing o he ans ec ion e iciency can be ob ained by in oduc ion o a epo e
gene a he clea age si e o he cas9 endonuclease. Po en ially an e en mo e
sophis ica ed s a egy is o gene a e a human mesenchymal s em cell line de icien o
sialyla ed cell su ace epi opes. The pe spec i e can ei he be ob ained by co-
ans ec ion o mul iple plasmids each a ge ing he indi idual genes, cons uc one
plasmid ha ha bo s mo e han one gRNA o knockou o selec ed enzymes ups eam
o he enzyma ic pa hways in ol ing ca alysis o glycosyl- and sialyl ans e ases,
espec i ely (S een o e al., 2014).
Conclusion
In conclusion, he epo p esen s e idence ha syn hesis o he ype 2 lac osamine
de e minan in human MSCs is a complex and highly egula ed p ocess; consis en
wi h con en ional e alua ion o glycosyla ans e ase ac i i y and speci ici y (Wandall
e al., 1997). Addi ionally, he epo emphasizes some echnical challenges o he
expe imen s (see Discussion o de ails).
14
Al hough only a mino e ec was obse ed in knockou o β4GalT1, he ans ec ion
esul ed in an expec ed dec eased p esence o ype 2 lac osamines epi opes on he cell
su ace o human MSCs. In addi ion, i canno be s essed enough ha lack o posi i e
con ols o assess β4GalT1 knockou pheno ype muddles he conclusion e en u he .
Rega dless, i appea s ha β4GalT1 is dispensable o c ea ion o ype 2 lac osamines
in human MSCs. A such a “ escue pheno ype” domina ed by exp ession o β4GalT2
may explain he mino e ec o knocking ou β4GalT1.
Knockou o he indi idual sialyl ans e ases exp essed in human MSCs, esul ed
expec edly in educed p esence o sialyla ed ype 2 lac osamines epi opes on he cell
su ace. Fu he , he da a sugges s ha al hough edundan ly exp essed, he
sialyl ans e ases do no ha e equal enzyma ic capaci y. Tha is, he sialyl ans e ase
knockou da a sugges s ha sialyla ion o ype 2 lac osamine in human MSCs is
domina ed by ac i i y o ST3GalIV. In compa ison, knockou o he sialyl ans e ases
in human MSCs has a g ea e pheno ypic impac han knockou o β4GalT1.
In sum, we iden i ied a leas ou enzymes in ol ed in syn hesis o he ype 2
sialyla ed lac osamine in human MSCs.
In addi ion o he igo ous dependency on p ope con ols, he immedia e echnical
challenges include wo. These a e op imizing managemen o human MSCs in i o
and op imizing he ans ec ion p ocedu e pe se. The o me can be ob ained by
adjus ing he speci ic media o mula ion. Al hough no e iden ially suppo ed, a
conside able e ec was obse ed by subs i u ing FBS o pla ele lysa e. Pe spec i es
on op imiza ion o he ans ec ion p ocedu e include applica ion o FACS p eceded
by in oduc ion o a GFP- ag and inducing he exp ession o a epo e gene.
Acco ding o he biological pe spec i e (see Expe imen al Mo i a ion o de ails), he
es ablishmen o a obus assay based on he cu en s a egy, will gene a e a
desc ip i e da ase . Mo eo e , he he apeu ical pe spec i es demands gene a ion o
an ope a ional pla o m, whe e a pa icula physiological e ec (i.e. en o ced cellula
mig a ion) is ob ained by gene ic enginee ing o human MSCs. Explici ly, he la e
eadou is unc ional. The o me is s uc u al. Despi e he quali a i e di e ence
be ween he wo eadou s (i.e. unc ional and s uc u al), hey a e dependen on each
o he as gene ic dissec ion o enzyma ic egula ion is necessa y in o de o bene i
he apeu ically om physiological models assessed biochemically. Thus, gene ic
dissec ion o he glycosyl ans e ase ac i i y in human MSCs demands ca e ul and
pe sis en a en ion in o de o ob ain en o ced cellula mig a ion and, subsequen ly
di e en ia ion o issues we e hey a e needed o immuni y and issue epai .
15
Ma e ials and Me hods
Human Mesenchymal S em Cells (MSCs)
All epo ed cellula assays h oughou we e conduc ed in human mesenchymal s em
cells (MSCs). Consequen ly, hei cellula cha ac e is ics will be in oduced b ie ly.
Expe imen al wo k wi h human MSCs has been epo ed o hal a cen u y (Cos e o,
Che ez, Ba oso-Moguel, & Pome a , 1955). They a e de ined by hei abili y o gi e
ise o a b oad se o di e en ia ed mesenchymal cell ypes including os eoblas s,
adipocy es, myocy es and chond ocy es. A such, hey ha e a obus and cha ac e is ic
di e en ia ion po en ial. By de ini ion, MSCs a e ca ego ized as non-hema opoie ic
s em cells. Pheno ypically, hey a e de ined by he exp ession o CD105, CD73 and
CD90 and absence o CD11b, CD14, CD45 and CD34. Addi ionally, he e is
consensus ha human MSCs exp ess CD44 (Uccelli, Mo e a, & Pis oia, 2008).
MSCs a e pa icula ly in e es ing as a esea ch pla o m in he a emp o ad ance
he apeu ical op ions o egene a i e medicine, as MSCs consequen ly o hei
in ini e cell cycle ha e po en ial o os e issue epai . In connec ion o hei
mic oen i onmen , MSCs also ha e po en an i-p oli e a i e and -in lamma o y
e ec s (Uccelli e al., 2008).
As s em cells age, hey al e pheno ypically a a molecula le el as well as
me abolically. MSCs a e no excep ion. Fo ins ance exp ession o he umo
supp esso p53 is pa o he mesenchymal s em cell senescence esponse. Also,
MSCs unde go i e e sible changes and ypically exhibi –galac osidase ac i i y
ela ed o senescence. Physiologically, MSCs ha e a ema kable capaci y o inhibi ing
he immune esponse, making hem – seemingly – e en mo e p omising in acili a ing
issue egene a ion in man (Dominici e al., 2006).
In cul u e, human MSCs a e adhe en , which is an addi ional de ining cha ac e is ic o
MSCs (Dominici e al., 2006). Howe e , in compa ison wi h o he adhe en cell lines
(e.g. HaCaT cell line (Boukamp e al., 1988)) hey a e li ed easily. Mo phologically,
hey a e cha ac e is ically s a -shaped and exhibi some simila i ies wi h ib oblas .
Cells om he a e age dono may su i e ex i o o abou 8 passages,
co esponding o he o de o 40 days (pe sonal obse a ions).
Cha ac e iza ion o Glycosyl ans e ases Exp essed in Human MSCs
Exp ession le els o he enzymes in ques ion we e assessed by quan i a i e
polyme ase chain ea ion (qPCR) and measu ed ela i e o he exp ession o
glyce aldehyde 3-phospha e dehyd ogenase (GAPDH) (Schmi gen & Li ak, 2008).
Ampli ica ion o he genes in ques ion was se up o PCR in 12 L using he SYBR®
Selec Mas e Mix (appliedbiosys ems, by Li e Technologies). P ime s we e p e-
pai ed a a s ock concen a ion o 10 M (5 M o o wa d and e e se p ime ). Fo
eac ions, p ime s we e dilu ed o a inal concen a ion o 1.25 M. The PCR
empla es consis ed in cDNA ansc ibed in i o om mRNA o wo indi idual
dono s. The ela i e exp ession le el o each gene was based on he geome ic mean
de i ed om eplica es o h ee and was exp essed ela i e o GAPDH, se ing he
16
unc ion o a housekeeping gene. The ampli ica ion was pe o med wi h The
S epOnePlus™ Real Time Sys em (The moFishe Scien i ic) and analyzed using
G aphPad P ism.
Cons uc ion o Vec o s o T ans ec ion
Plasmid ec o s o ans ec ion o MSCs we e p epa ed in acco dance wi h he
CRISPR/cas9- echnology (Cong e al., 2013). P ime s a ge ed a he genes o in e es
we e designed using he UCSC Genome B owse . Sea ch was limi ed o exon one and
p ime s we e o de ed om Applied Biosys ems™. By e alua ing each se o p ime s
a he da abase a ailable a c isp .mi .edu, op imal on a ge and minimal o a ge
e ec s we e ensu ed. The plasmid cons uc s we e analyzed by es ic ion diges ion
and sequencing p io o ans ec ion. The plasmid o choice was pX330 (Addgene),
which has been designed by he manu ac u e such ha i ha bo s sequences encoding
nuclea ion loca ion signals and a p omo o ups eams o he modi ied CRISPR gene
locus. Fo selec ion pu poses, he manu ac u e has also designed he plasmid o
exp ess an ampicillin esis ance gene. The DNA inse s we e designed wi h o e hangs
compa ible wi h clea age mo i e o he es ic ion enzyme minimizing isk o sel -
liga ion. The es ic ion enzyme was BbsI (New England BioLabs®). Res ic ion
clea age was pe o med in a bu e composed o 50 mM NaCl, 10 mM T is-HCl, 10
mM MgCl2 and 100 g/mL BSA. Reac ion was pe o med a pH 7.9 a 37°C o 1
hou .
Fo liga ion, 50 ng o linea ized plasmid was incuba ed wi h 37.5 ng DNA inse
co esponding o a s oichiome ic a io 3:1 be ween inse and linea ized ec o
backbone. As a means o in oduce Gibbs ene gy in o he sys em and ini ial
challenges p oducing he plasmid a ge ing he gene ic locus o in e es , DNA inse s
we e phospho yla ed p io o annealing. Running bu e was 50 mM T is-HCl, 10
mM MgCl2, 1 mM ATP and 10 mM DTT. Reac ion was pe o med a pH 7.5.
Fo ampli ica ion o he liga ed plasmid, i was ansduced in o One Sho ® S bl3™
Chemically Compe en E. coli (in i ogen™, by Li e Technologies) and cul u ed on
aga pla es wi h ampicillin a 100 g/mL. DNA was isola ed upon binding o a silica
memb ane and elu ed by an inc easing sal concen a ion g adien in acco dance wi h
he manu ac u es p o ocol (©Qiagen). In o de o e i y inse ion o ele an guide
RNA (gRNA) in o he ec o , samples we e sen o sequencing p io o ans ec ion
(Biopolyme s Facili y @ Ha a d Medical School, h ps://genome.med.ha a d.edu/).
T ans ec ion o MSCs
The MSCs we e ha es ed om human dono s. Dono s we e included acco ding o
he Human Expe imen a ion and E hics Commi ees o Pa ne s Cance Ca e
Ins i u ions (Dyks a e al., 2016). The mononuclea ac ion was isola ed by Ficoll-
Hypaque densi y sepa a ion, and pla ed a 2-5 106 cells/mL in T-175 cul u e lasks.
Cells we e cul u ed in Dulbecco’s Modi ied Eagle Medium (DMEM), 10% pla ele
lysa e and 1 % An ibio ic-An imyco ic (Gibco®, by Li e Technologies) dilu ed as
p esc ibed o a inal concen a ion o 100 U/mL o penicillin and s ep omycin,
espec i ely. Cells we e incuba ed a 37°C o 18 hou s and washed in PBS, whe eby
17
isola ion o he adhe en cell popula ion, ha is he MSCs, was ob ained.
Cells we e g own con luen in T-175 lasks, co esponding o 500.000 cells pe lask.
P io o ans ec ion, cells we e li ed wi h 0.05% ypsin and 0.5 mM EDTA dilu ed
in phospha e-bu e ed saline (GIBCO® PBS). T ans ec ion was pe o med using
Lipo ec amine® 3000 T ans ec ion Reagen (in i ogen™, by Li e Technologies)
using 3 g o p e-pai ed plasmid cons uc (as desc ibed abo e).
Enzyma ic P epa a ion o Human MSCs
Fo exo ucosyla ion, cells we e ea ed wi h ecombinan ucosyl ans e ase 6
exp essed in Pichia pas o is exp ession sys em (in e nal de elopmen ). Reac ion
olume was 30 L. Reac ion bu e was dilu ed in Hanks Balanced Sal Solu ion
(HBSS) om 25X s ock (in e nal o mula ion). Addi ionally, GDP- ucose was added
o a wo king concen a ion o 1 mM GDP- ucose. Cells we e ea ed o 1 hou a
37°C and eac ions we e s opped by dilu ion in HBSS. Fo neu aminidase ea men ,
cells we e ea ed wi h Neu aminidase (Sialidase) om Clos idium pe ingens
(Roche). Enzyme was dilu ed 1:10 co esponding o a wo king ac i i y o 10 U/mg.
Reac ion was pe o med in PBS o 1 hou a 37°C and s opped by dilu ion wi h PBS.
Cha ac e iza ion o T ans ec ed MSCs by PCR on Genomic DNA
Media was emo ed om ans ec ed MSCs and cells we e washed. Subsequen ly,
cells we e li ed wi h ypsin as desc ibed abo e. Then, genomic DNA was isola ed
wi h Pu eLink® Gennomic DNA Ki (In i ogen, by li e echnologies) acco ding o
manu ac u es p o ocol and sample ype ca ego ized as mammalian cell lysa e. The
isola ed DNA se ed as a empla e in he PCR. Reac ion was p epa ed wi h olume
picomoles o o wa d and e e se p ime , a o al o 10 mM dNTPs and 1 L o DNA
polyme ase. Reac ion olume was 50 L and PCR was pe o med in MJ Resea ch
PTC-200 Pel ie The mal Cycle DNA Engine Dual Alpha Blocks. P og am was
ini ial dena u a ion, one cycle a 94 o 2 minu es, hen 30 cycles o ampli ica ion
composed o 94°C o 30 seconds, 56°C o 30 seconds and 72°C o 30 seconds.
Final ex ension was ensu ed by one cycle a 72°C o 5 minu es.
Ampli ied PCR p oduc s we e analyzed on a 2% aga ose gel p epa ed wi h 3%
e hidium b omide. Gel an a 100 V o 30 minu es. Running bu e was T is-ace a e-
EDTA.
Cha ac e iza ion o T ans ec ed MSCs by Flow Cy ome y
Upon ans ec ion, cells we e e-g own in T-175 lask ill a con luency o >80%,
co ela ing wi h an incuba ion ime o app oxima ely 4 weeks. O no e is ha , cellula
doubling ime was slowed signi ican ly a e ans ec ion. Cellula de elopmen was
moni o ed on a egula basis by di ec ligh mic oscopy. When cells we e con luen ,
hey we e washed and li ed as desc ibed abo e. Cells we e di ided acco ding o he
mode o assessmen . Fo ECL-s aining, cells we e incuba ed wi h Fluo escein labeled
E y hina C is agalli Lec in (ECL, ECA) (Vec o Labo a o ies) dilu ed 1:4000 in
PBS, incuba ed a 4°C o 30 min. Fo s aining agains he HECA-452 epi ope, cells
we e incuba ed wi h FITC an i-human/mouse Cu aneous Lymphocy e An igen
An ibody (Biolegend®) dilu ed 1:10 and incuba ed a 4°C o 30 min. Fo assessmen
18
o CD15 epi opes, cells we e incuba ed wi h FITC an i-human CD15 (SSEA-1)
An ibody (Biolegend®). As a s aining con ol, FITC Mouse IgM, Iso ype C l
An ibody (Biolegend®) was used upon dilu ion 1:50 in PBS. Da a was acqui ed using
FC 500 Se ies Flow Cy ome e (Beckman Coul e ) and analyzed by FlowJo®.
19
Li e a u e
Albe s B, J. A., Lewix J, Ra M, Robe s K and Wal e P. (2007). In acellula
Vesicula T a ic Molecula Biology o The Cell (pp. 776-777): Ga land
Science.
A naou , M. A. (1990). S uc u e and unc ion o he leukocy e adhesion molecules
CD11/CD18. Blood, 75(5), 1037-1050.
Bayne-Jones, S. (1931). DR. KARL LANDSTEINER NOBEL PRIZE LAUREATE
IN MEDICINE, 1930. Science, 73(1901), 599-604.
doi:10.1126/science.73.1901.599
Bo egaa d, N. (2010). Neu ophils, om ma ow o mic obes. Immuni y, 33(5), 657-
670. doi:10.1016/j.immuni.2010.11.011
Boukamp, P., Pe usse ska, R. T., B ei k eu z, D., Ho nung, J., Ma kham, A., &
Fusenig, N. E. (1988). No mal ke a iniza ion in a spon aneously immo alized
aneuploid human ke a inocy e cell line. J Cell Biol, 106(3), 761-771.
Cong, L., Ran, F. A., Cox, D., Lin, S., Ba e o, R., Habib, N., . . . Zhang, F. (2013).
Mul iplex genome enginee ing using CRISPR/Cas sys ems. Science,
339(6121), 819-823. doi:10.1126/science.1231143
Cos e o, I., Che ez, A., Ba oso-Moguel, R., & Pome a , C. M. (1955). [Fib oblas ic
sys em. VI. Mo phology o mesenchymal cells in i o]. A ch Ins Ca diol
Mex, 25(2), 125-170.
Cummings RD, E. J. (2009). P inciples o Glycan Recogni ion. In C. R. Va ki A,
Esko JD, e al. (Ed.), Essen ials o Glycobiology. Cold Sp ing Ha bo (NY):
Cold Sp ing Ha bo Labo a o y P ess.
Dominici, M., Le Blanc, K., Muelle , I., Slape -Co enbach, I., Ma ini, F., K ause, D.,
. . . Ho wi z, E. (2006). Minimal c i e ia o de ining mul ipo en
mesenchymal s omal cells. The In e na ional Socie y o Cellula The apy
posi ion s a emen . Cy o he apy, 8(4), 315-317.
doi:10.1080/14653240600855905
Dyks a, B., Lee, J., Mo ensen, L. J., Yu, H., Wu, Z. L., Lin, C. P., . . . Sacks ein, R.
(2016). Glycoenginee ing o E-Selec in Ligands by In acellula e sus
Ex acellula Fucosyla ion Di e en ially A ec s Os eo opism o Human
Mesenchymal S em Cells. S em Cells. doi:10.1002/s em.2435
Galluzzi, L., Vacchelli, E., B a o-San Ped o, J. M., Buque, A., Seno illa, L., Ba acco,
E. E., . . . K oeme , G. (2014). Classi ica ion o cu en an icance
immuno he apies. Onco a ge , 5(24), 12472-12508.
doi:10.18632/onco a ge .2998
G ewal, P. K., Uchiyama, S., Di o, D., Va ki, N., Le, D. T., Nize , V., & Ma h, J. D.
(2008). The Ashwell ecep o mi iga es he le hal coagulopa hy o sepsis. Na
Med, 14(6), 648-655. doi:10.1038/nm1760
Hudak, J. E., & Be ozzi, C. R. (2014). Glyco he apy: New Ad ances Inspi e a
Reeme gence o Glycans in Medicine. Chem Biol, 21(1), 16-37.
doi:10.1016/j.chembiol.2013.09.010
Lo ling, J., & Holge sson, J. (2009). Co e saccha ide dependence o sialyl Lewis X
biosyn hesis. Glycoconj J, 26(1), 33-40. doi:10.1007/s10719-008-9159-z
Manzella, S. M., Hoope , L. V., & Baenzige , J. U. (1996). Oligosaccha ides
con aining be a 1,4-linked N-ace ylgalac osamine, a pa adigm o p o ein-
speci ic glycosyla ion. J Biol Chem, 271(21), 12117-12120.
20
Ma h, J. D., & G ewal, P. K. (2008). Mammalian glycosyla ion in immuni y. Na Re
Immunol, 8(11), 874-887. doi:10.1038/n i2417
Mou an , A. E. (1946). A new human blood g oup an igen o equen occu ence.
Na u e, 158, 237.
Oh subo, K., & Ma h, J. D. (2006). Glycosyla ion in cellula mechanisms o heal h
and disease. Cell, 126(5), 855-867. doi:10.1016/j.cell.2006.08.019
Rini J, E. J., Va ki A. (2009). Glycosyl ans e ases and Glycan-p ocessing Enzymes.
In C. R. Va ki A, Esko JD, e al. (Ed.), Essen ials o Glycobiology: Cold
Sp ing Ha bo Labo a o y.
Sacks ein, R. (2009). Glycosyl ans e ase-p og ammed s e eosubs i u ion (GPS) o
c ea e HCELL: enginee ing a oadmap o cell mig a ion. Immunol Re ,
230(1), 51-74. doi:10.1111/j.1600-065X.2009.00792.x
Sacks ein, R. (2012). Glycoenginee ing o HCELL, he human bone ma ow homing
ecep o : swee ly p og amming cell mig a ion. Ann Biomed Eng, 40(4), 766-
776. doi:10.1007/s10439-011-0461-8
Schmi gen, T. D., & Li ak, K. J. (2008). Analyzing eal- ime PCR da a by he
compa a i e C(T) me hod. Na P o oc, 3(6), 1101-1108.
Sp inge , T. A. (1994). T a ic signals o lymphocy e eci cula ion and leukocy e
emig a ion: he mul is ep pa adigm. Cell, 76(2), 301-314.
S anley, P. (2014). Galec ins CLIC ca go inside. Na Cell Biol, 16(6), 506-507.
doi:10.1038/ncb2983
S een o , C., Benne , E. P., Schjoldage , K. T., Vakh ushe , S. Y., Wandall, H. H.,
& Clausen, H. (2014). P ecision genome edi ing: a small e olu ion o
glycobiology. Glycobiology, 24(8), 663-680. doi:10.1093/glycob/cwu046
Takeuchi, H., & Hal iwange , R. S. (2014). Signi icance o glycosyla ion in No ch
signaling. Biochem Biophys Res Commun, 453(2), 235-242.
doi:10.1016/j.bb c.2014.05.115
Uccelli, A., Mo e a, L., & Pis oia, V. (2008). Mesenchymal s em cells in heal h and
disease. Na Re Immunol, 8(9), 726-736. doi:10.1038/n i2395
Va ki A, K. R., Toole BP. (2009). Glycosyla ion Changes in Cance . In C. R. Va ki
A, Esko JD, e al. (Ed.), Essen ials o Glycobiology. Cold Sp ing Ha bo
(NY): Cold Sp ing Ha bo Labo a o y P ess.
Wandall, H. H., Hassan, H., Mi go odskaya, E., K is ensen, A. K., Roeps o , P.,
Benne , E. P., . . . Clausen, H. (1997). Subs a e speci ici ies o h ee
membe s o he human UDP-N-ace yl-alpha-D-galac osamine:Polypep ide N-
ace ylgalac osaminyl ans e ase amily, GalNAc-T1, -T2, and -T3. J Biol
Chem, 272(38), 23503-23514.
Yang, Z., Wang, S., Halim, A., Schulz, M. A., F odin, M., Rahman, S. H., . . .
Clausen, H. (2015). Enginee ed CHO cells o p oduc ion o di e se,
homogeneous glycop o eins. Na Bio echnol, 33(8), 842-844.
doi:10.1038/nb .3280
