Lab on a Chip
PAPER
Ci e his: Lab Chip,2016,16,586
Recei ed 18 h No embe 2015,
Accep ed 16 h Decembe 2015
DOI: 10.1039/c5lc01415h
www. sc.o g/loc
Handheld eal- ime PCR de ice†
Ch is ian D. Ah be g,
a
Bojan Robe Ilic,
b
And eas Manz
a
and Pa el Neužil*
acd
He e we epo one o he smalles eal- ime polyme ase chain eac ion (PCR) sys ems o da e wi h an
app oxima e size o 100 mm ×60 mm ×33 mm. The sys em is an au onomous uni equi ing an ex e nal
12 V powe supply. Fou simul aneous eac ions a e pe o med in he o m o i ual eac ion chambe s
(VRCs) whe e a ≈200 nL sample is co e ed wi h mine al oil and placed on a glass co e slip. Fas , 40 cycle
ampli ica ion o an amplicon om he H7N9 gene was used o demons a e he PCR pe o mance. The
s anda d cu e slope was −3.02 ±0.16 cycles a h eshold pe decade (mean ±s anda d de ia ion) co e-
sponding o an ampli ica ion e iciency o 0.91 ±0.05 pe cycle (mean ±s anda d de ia ion). The PCR de-
ice was capable o de ec ing a single deoxy ibonucleic acid (DNA) copy. These esul s u he sugges ha
ou handheld PCR de ice may ha e b oad, echnologically- ele an applica ions ex ending o apid de ec-
ion o in ec ious diseases in small clinics.
The in en ion o polyme ase chain eac ion (PCR) 32 yea s
ago is conside ed o be one o he g ea es in en ions o he
las cen u y.
1
O e he yea s, many a ian s o he o iginal
sys em ha e been de eloped. One o he mos impo an
ad ancemen s is he eal- ime PCR analysis sys em.
2
The ap-
p oach enables eal- ime PCR ampli ica ion, moni o ing, and
quan i ica ion o he numbe o deoxy ibonucleic acid (DNA)
copies in he sample unde conside a ion. This me hod is
commonly e e ed o as quan i a i e PCR (qPCR).
2
The main
ad an age o eal- ime PCR is he elimina ion o any pos -p o-
cessing, such as elec opho esis o hyb idiza ion o de ec he
PCR p oduc .
The PCR eac ion is pe o med by he mal cycling in he
p esence o speci ic oligonucleo ides, he enzyme polyme ase,
ee nucleic acids and bi alen sal s such as MgSO
4
o MgCl
2
.
This cock ail is commonly e e ed o as he PCR mas e mix.
The de ec ion o PCR p oduc ampli ica ion is conduc ed by
moni o ing he luo escence ampli ude du ing he PCR. In
he p esence o an in e cala ing dye, such as SYBR G een I,
he luo escence ampli ude is p opo ional o he concen a-
ion o he DNA amplicon, he p oduc o he PCR. In o de
o e i y ampli ica ion speci ici y, upon PCR comple ion, em-
ploymen o an in e cala ing dye enables pe o ming mel ing
cu e analysis (MCA). Ano he imp o emen o he PCR is he
addi ion o he e e se ansc ip ase enzyme o he PCR cock-
ail, o ming e e se ansc ip ion PCR (RT-PCR).
PCR has become he me hod o choice o he de ec ion o
DNA and RT-PCR o de ec RNA. These wo eac ions ha e
e olu ionized gene ics. Fu he mo e, PCR has many di e se
applica ions in in ec ious disease diagnos ics o de ec ion o
i uses o bac e ia,
3
in o ensic science,
4,5
pa e ni y es s,
6
secu i y applica ions
7
and my iad o he comme cial applica-
ions.
8
Comme cial sys ems a e ypically a he la ge able-
op ools used o high h oughpu mass sc eenings and a e
imp ac ical o use in poin -o -ca e applica ions (POC), whe e
he mos impo an sys em pa ame e s a e po abili y and
powe consump ion. The ques o a minia u ized PCR e -
sion sui able o POC diagnos ics was ini ia ed a Law ence
Li e mo e Labo a o ies
9,10
mo e han wo decades ago.
Ag awal e al. ha e de eloped a pocke -sized con en ional
PCR sys em
11
ha equi es ex ensi e sample pos -p ocessing
o iden i y he p esence o an amplicon. In con as , eal- ime
PCR elimina es he need o sample p ocessing once PCR is
comple ed.
Real- ime PCR sys ems consis o a hea e , empe a u e
senso , and luo escence exci a ion and de ec ion uni . Tem-
pe a u e cycling is pe o med by hea ing and cooling o sam-
ples. Wi hin he PCR p ocess, he cooling a e is one o he
p ima y limi ing ac o s. Bulky comme cial sys ems ha e la ge
hea capaci ies, hence hea emo al is challenging, and is ypi-
cally accomplished by using a he moelec ic coole (TEC),
commonly known as he Pel ie elemen . Since hese bulky
sys ems consume a conside able amoun o powe , hey a e
gene ally unsui able o ield es ing POC applica ions.
586 |Lab Chip,2016,16,586–592 This jou nal is © The Royal Socie y o Chemis y 2016
a
KIST-Eu ope, Mic o luidics G oup, Campus E7.1, 66111 Saa b ücken, Ge many.
E-mail: [email p o ec ed]
b
Na ional Ins i u e o S anda d and Technology (NIST), Cen e o Nanoscale
Science and Technology, 100 Bu eau D i e, MS 6201, Gai he sbu g, MD 20899-
6201, USA
c
B no Uni e si y o Technology (BUT), Cen al Eu opean Ins i u e o Technology
(CEITEC), Technická 3058/10, CZ-616 00 B no, Czech Republic
d
No hwes e n Poly echnical Uni e si y (NPU), School o Mechanical Enginee ing,
Depa men o Mic osys em Enginee ing, 127 Wes Youyi Road, Xi'an Shaanxi,
710072, PR China
†Elec onic supplemen a y in o ma ion (ESI) a ailable. See DOI: 10.1039/
c5lc01415h
Open Access A icle. Published on 16 Decembe 2015. Downloaded on 23/06/2016 07:07:56.
This a icle is licensed unde a
C ea i e Commons A ibu ion 3.0 Unpo ed Licence.
View A icle Online
View Jou nal
| View Issue
Lab Chip, 2016, 16,586–592 | 587This jou nal is © The Royal Socie y o Chemis y 2016
Small PCR ins umen s a e o en based on mic o luidic
de ices, so-called “lab-on-a-chip”de ices.
12
These sys ems
comp ise wo majo g oups, spa ial-domain and ime-domain
PCRs. On he one hand, ime-domain PCRs ha e a single
hea e wi h samples placed in di ec con ac . He e, empe a-
u e cycling is ca ied ou by changing he hea e elemen
empe a u e. On he opposi e end o he spec um, he
spa ial-domain PCR has se e al hea e s, each held a a di e -
en empe a u e. In his scena io, empe a u e cycling is
accomplished by mo ing samples be ween hea e s.
A ypical ep esen a i e o a spa ial-domain sys em is he
con inuous PCR-on-a-chip.
13
Wi hin his sys em, he sample
low in he mic o luidic chip is posi ioned o e he hea e s,
each kep a a di e en empe a u e. The sample lows
h ough ubes, he eby achie ing he mal cycling. He e, PCR
du a ion is only limi ed by he low a e and he hea ans e
be ween he sample and he side walls, o bo h hea ing and
cooling. The wo majo d awbacks o a low- h ough PCR a e
sys em complexi y and a high likelihood o sample- o-sample
c oss-con amina ion.
An al e na i e e sion was in oduced a ew yea s ago
whe e he sample was in he o m o a i ual eac ion cham-
be (VRC).
14–16
The VRC sel -assembly sys em consis s o a
wa e d ople co e ed wi h mine al oil, p e en ing wa e e ap-
o a ion om he sample. In his scena io, he wa e d ople
con ained he PCR mas e mix wi h a p e-de e mined num-
be o DNA copies. He e, he VRC wi h DNA was sepa a ed
om he mic omachined silicon hea e s by a disposable,
hyd ophobically-coa ed mic oscope co e slip. To elimina e
sample- o-sample con amina ion, he glass co e slip was a
disposable pa o he sys em, and he e o e each co e slip
was a single use componen . The sample con ained magne ic
pa icles which acili a ed sample mo ion be ween hea e s.
17
A pocke -size eal- ime PCR sys em capable o p ocessing a
single sample was in oduced a ew yea s la e .
18
The sys em
had an in eg a ed minia u ized op ical de ec ion uni , LCD
display and con ol elec onics. One o he key ea u es was
he implemen a ion o lock-in ampli ica ion o op ical signal
p ocessing.
17
The lock-in ampli ica ion ea u e allowed o
ambien sys em ope a ion wi hou ligh p o ec ion, he eby
ende ing he sys em obus and use - iendly. One o he sys-
em d awbacks was p ocessing a single sample a a ime and
u he mo e, he de ice was bulky.
A p ac ical sys em o conduc PCR o POC applica ions e-
qui es simul aneous p ocessing o 4 o mo e samples. Diag-
noses o clinical samples should be concu en ly conduc ed
wi h posi i e and nega i e con ol samples, he eby elimina -
ing alse nega i e o posi i e e en s.
In ou wo k, we in oduce a new po able PCR sys em
(Fig. 1) capable o concu en ly analyzing ou ≈200 nL ol-
ume samples. The sys em speed is de e mined by he hea ing
and cooling a es. The hea ing a e collec i ely a ises om
he VRC he mal capaci ance (H) and he dissipa ed Joule
hea . The a e o passi e cooling applied wi hin ou sys em is
gi en by he he mal ime cons an (τ) o he sys em, which is
gi en by H/G, whe e Gis he he mal conduc ance. Since he
speci ic hea o wa e is excep ionally la ge, he he mal p op-
e ies o ou sys em a e s ongly dependen on he sample
wa e olume. Consequen ly, a smalle sample size esul s in
a as e sys em.
19
Sys em samples consis o a nega i e con ol, also called
no empla e con ol (NTC), a posi i e con ol, and wo sam-
ples o in e es . The ou sample sys em a chi ec u e ep e-
sen s he minimal numbe o samples equi ed o p ac ical
applica ions. Ou sys em conduc s 40 PCR cycles in less han
≈35 min, while simul aneously p ocessing he esul s. Fu -
he mo e, ou po able eal- ime PCR is capable o de ec ing
a single DNA copy.
The PCR pe o mance was e alua ed by de ec ing a com-
plemen a y DNA om he a ian in luenza i us (H7N9) as
well as wo human ansc ip s, hypoxan hine
phospho ibosyl ans e ase (HPRT) and glyce aldehyde-3-
phospha e dehyd ogenase (GAPDH). To he bes o ou
knowledge, ou sys em pla o m ep esen s he smalles eal-
ime PCR sys em.
Ou PCR ins umen has wo key ea u es:
1. The ou samples a e in he VRC o m and a e placed
on a disposable glass co e slip o e mic omachined silicon
hea e s. Upon PCR comple ion, he single-use, disposable
glass elemen is emo ed and a new glass co e slip is placed
on op o he silicon hea e .
2. The luo escence exci a ion/de ec ion sys em is based
on a lock-in ampli ie , he eby ende ing he sys em immune
o ambien ligh . The PCR ins umen is equipped wi h a
g aphical 84 ×48 pixel liquid c ys al display (LCD) wi h a di-
agonal size o ≈38.1 mm o show he eac ion p og ess and
Fig. 1 (A) CAD design d awing o he handheld PCR. The illus a ion
shows a display wi h a compa men accommoda ing 4 samples in he
VRC o m. (B) Fab ica ed and assembled comple e eal- ime PCR de-
ice packaged wi hin a 3D p in ed casing.
Lab on a Chip Pape
Open Access A icle. Published on 16 Decembe 2015. Downloaded on 23/06/2016 07:07:56.
This a icle is licensed unde a
C ea i e Commons A ibu ion 3.0 Unpo ed Licence.
View A icle Online
588 |Lab Chip,2016,16,586–592 This jou nal is © The Royal Socie y o Chemis y 2016
inal esul s. The cap u ed da a is s o ed in an in e nal mem-
o y and can be uploaded o ex e nal p ocessing ia a uni e -
sal se ial bus (USB). The sys em is powe ed by an ex e nal 12
V ba e y.
Sys em se up
Ou cu en sys em has wo new key ea u es: an in eg a ed
op ical head and simpli ied con ol elec onics.
1. In eg a ed op ical head
A luo escence de ec ion sys em o a single spo equi es a
ligh sou ce, h ee il e s (exci a ion, dich oic mi o and
emission) and a de ec o . We edesigned he o iginal
head
20,21
wi h 5 il e s o he ou uni s (see Fig. 2). Each
measu emen spo is illumina ed wi h a ligh emi ing diode
(LED) wi h a p incipal emission wa eleng h o 470 nm and a
luminous in ensi y in he ange o 7.2 cd o 12 cd. Ligh
passes h ough an exci a ion band pass il e wi h a cen e
wa eleng h o ≈470 nm and a band pass o ≈40 nm,
blocking ligh om he LED wi h wa eleng hs longe han
≈490 nm. Ligh is hen e lec ed o o a long pass dich oic
mi o wi h a cu -o wa eleng h o ≈495 nm, and ocused by
a lens wi h a ocal leng h o ≈3.1 mm, a nume ical ape u e
o ≈0.68 and an an i e lec i e coa ing in he ange o ≈350
nm o ≈700 nm. The emi ed luo escence (F) is collima ed
by he same lens, passing h ough he dich oic mi o . The
esidual blue ligh is supp essed by a long pass emission il-
e wi h a cu -o wa eleng h o ≈510 nm, and luo escence is
cap u ed by he con en ional silicon pho odiode wi h a adi-
an sensi i e a ea o ≈7.5 mm.
2
The c oss sec ion schema ic
o he handheld PCR sys em illus a ing he op ical pa h is
shown in Fig. S1 (ESI†). The esul ing pho ocu en is
con e ed in o ol age using an ul a-low bias cu en
ope a ional ampli ie wi h dielec ically-isola ed ield e ec
ansis o inpu s (diFET) as a ansconduc ance ampli ie . In
his con igu a ion, he ou sample sys ems sha e op ical il-
e s. Fou LEDs a e moun ed in wo pai s. Wi hin each pai ,
he LEDs we e pa allel, hus equi ing only wo exci a ion
and wo dich oic il e s o all 4 LEDs. Finally, he e is only a
single emission il e o all pho odiodes. A po en ial expan-
sion o eigh sys ems would equi e addi ional LEDs and
pho odiodes wi h ampli ie s.
2. Con ol elec onics
A p e iously published sys em
18
had one lock-in ampli ie o
a single luo escence de ec ion sys em and a second one o
empe a u e measu emen . This scena io was e y ine icien
since luo escence was moni o ed o ≈2 s du ing each PCR
cycle. The second lock-in ampli ie o empe a u e measu e-
men was used du ing he en i e PCR ope a ion. Ou cu en
sys em employs a single lock-in ampli ie o moni o he sam-
ple empe a u e and cap u e luo escence om all ou spo s.
The sys em hea e s we e connec ed in a se ial–pa allel
combina ion, whe ein he sys em con olled he a e age em-
pe a u e o all ou hea e s. We used a simila AC biased
Whea s one b idge o con e he esis ance o he esis ance
empe a u e de ec o (RTD) in o a DC ol age as p e iously
desc ibed.
18
The con ol elec onics o he op ical sys em was a simpli-
ied e sion o ou p e ious wo k.
18
He e, each PCR sys em
(spo ) had i s own LED, collima ing lens and a pho odiode
wi h a espec i e ansconduc ance ampli ie while he op i-
cal il e s a e sha ed. We ac i a ed one LED a a ime, he eby
eeding he signal o a single, co esponding ans-
conduc ance ampli ie . Ou pu s o he ou ampli ie s we e
connec ed oge he and p ocessed as one signal. The com-
ple e schema ic o he PCR sys em is shown in Fig. S2 (ESI†).
The inciden pho ocu en s om he pho odiodes we e
con e ed in o ol age using ou dedica ed ope a ional am-
pli ie s. The ampli ie ou pu s we e connec ed oge he and a
single composi e signal was u he p ocessed. The c oss alk
be ween ou measu ed spo s was minimized since one LED
was ac i a ed a a ime; he e o e he esul ing o al ampli-
ude o he p ocessed pho ocu en o igina ed om a single
dedica ed PCR eac ion. All impo an de ices a e lis ed in
Table S1 (ESI†).
Expe imen al
A ypical eal- ime PCR p o ocol wi h an in e cala ing dye
such as SYBR-G een I ini ia es wi h a ho s a o ac i a e he
polyme ase. PCR cycles consis o dena u a ion, annealing
and ex ension s eps. The luo escence ampli ude is measu ed
a he end o he ex ension s ep o a pe iod o ≈2 s. We con-
olled he hea e empe a u e using he p opo ional
in eg a i e-de i a i e (PID) closed eedback loop me hod. The
amoun o hea deli e ed h ough dissipa ed Joule hea ing
was con olled using a pulse-wid h modula ion (PWM) ech-
nique. The las ≈2 s o he ampli ica ion cycle we e used o
Fig. 2 Schema ic illus a ion o he in eg a ed op ical head. Blue
a ows show he op ical pa h om one LED wi h il e s o he VRC.
The g een a ow shows he op ical pa h o exci ed luo escence o he
pho odiode. Ligh emi ed om a blue LED passes h ough he blue
il e in o de o emo e he g een po ion o emi ed ligh , hen
e lec s o o he dich oic mi o h ough a donu -shaped hea e and
is ocused on he sample by an asphe ical lens. Exci ed luo escence is
collima ed by he same lens, passing h ough he dich oic mi o wi h
he blue po ion il e ed ou by a g een il e . Residual g een ligh in-
e ac s wi h a pho odiode and induces a pho ocu en , which is u he
p ocessed.
Lab on a ChipPape
Open Access A icle. Published on 16 Decembe 2015. Downloaded on 23/06/2016 07:07:56.
This a icle is licensed unde a
C ea i e Commons A ibu ion 3.0 Unpo ed Licence.
View A icle Online
Lab Chip,2016,16,586–592 | 589This jou nal is © The Royal Socie y o Chemis y 2016
luo escence moni o ing (see Fig. 3). In his s ep, ollowing
empe a u e s abiliza ion, we moni o ed he du y cycle o he
PWM and calcula ed i s a e age. Du ing he las ≈2 s, he
eedback loop was disconnec ed, he empe a u e was no
moni o ed and he a e age alue o he PWM was employed.
Du ing he sys em de elopmen phase, we moni o ed he
hea e empe a u e and ound ha he me hod desc ibed
abo e gi es us a empe a u e a ia ion o less han ±0.5 °C
o e a ≈2 s ime in e al. The sample empe a u e ollows
he hea e empe a u e wi h a ≈1.5 s delay,
22
he e o e a ±0.5
°C empe a u e a ia ion a he hea e does no a ec he
PCR pe o mance. The measu ed empe a u e p o ile om
≈6 PCR s eps is shown wi hin he sys em liquid c ys al dis-
play (LCD) in Fig. 4A.
Du ing he las wo seconds o he ex ension s ep, he
luo escence measu emen sys em was ac i a ed. Sequen-
ially, each LED was indi idually powe ed o ≈0.5 s and he
emi ed luo escence was cap u ed by he espec i e pho odi-
ode and lock-in ampli ie . A he comple ion o an ampli ica-
ion cycle, he sys em was swi ched back in o he empe a u e
measu emen mode, ini ia ing he s a o a new cycle. The
PCR ampli ica ion cu es we e plo ed o each spo and he
cap u ed ampli ude o luo escence was displayed on he
LCD display. The ypical PCR ampli ica ion cu e is shown in
Fig. 4B. Di ec ly ollowing he PCR p ocess, mel ing cu e
analysis (MCA) was pe o med. Since empe a u e measu e-
men du ing he MCA is ime consuming, we pe o med his
s ep wi hou a eedback loop. In he cou se o he PCR, we
moni o ed and eco ded he PWM du y cycle o h ee ixed
empe a u e poin s: dena u a ion, annealing and ex ension
empe a u es. These h ee poin s we e used o calcula e he
equi ed du y cycle o empe a u e scans anging om
≈68 °C o≈94 °C wi hou a closed eedback loop. The sys em
was s abilized a ≈68 °C. The empe a u e o each hea e was
hen g adually inc eased o ≈95 °C while he luo escence in
each o he ou spo s was sequen ially measu ed. The mea-
su emen o each spo equi ed a du a ion o ≈0.5 s, hence
≈2 s was equi ed o measu e all 4 VRCs. This measu emen
se up allowed o luo escence measu emen om each spo
wi h an o se o ≈0.25 °C be ween adjacen VRCs. Once he
MCA was comple ed, we s abilized he sample a he as-
sumed empe a u e o ≈95 °C and hen measu ed he ac ual
empe a u e. Di ec ly ollowing his, he sys em's cen al p o-
cessing uni (CPU) pe o med wo co ec ions. Fi s , he MCA
was ecalcula ed based on he ac ual inal empe a u e. Sec-
ond, he co ec ion accoun s o he empe a u e o se be-
ween indi idual VRCs. Consequen ly, he MCAs o each
spo we e ecalcula ed acco dingly. Finally, a nega i e de i a-
i e alue o luo escence wi h espec o empe a u e (−dF/
dT) was calcula ed.
Fig. 3 PCR he mal p o ile o a single ampli ica ion s ep. The hea e
empe a u e ( ed pa ) is linea ly p opo ional o he buil -in lock-in
ampli ica ion ou pu , which was cap u ed wi h an oscilloscope. While
pe o ming dena u a ion a ≈93 °C, annealing a ≈56 °C and mos o
he ex ension s ep a ≈72 °C, he buil -in lock-in ampli ie is u ilized o
measu e he a e age empe a u e o all ou hea e s. In he las wo
seconds o he ex ension s ep, he lock-in ampli ie is used o sequen-
ially p ocess he luo escence signal (g een pa ) om he 4 measu e-
men spo s (ci cled a ea). These da a a e s o ed in hei o iginal o ma
in analog- o-digi al con e e uni s (ADC uni s) wi hin he memo y o a
mic ocon olle . Once luo escence is measu ed, he hea e is
powe ed by he a e age alue o pulse wid h modula ion ob ained
du ing he ex ension phase.
Fig. 4 (A) An assembled PCR sys em wi h ou VRCs, each consis ing
o a ≈0.5 μL sized sample co e ed wi h ≈1.5 μL o M5904 mine al oil,
showing he PCR empe a u e p o ile (p o ocol) wi hin he LCD
display. The de ice size is 82 mm ×45 mm ×20 mm (leng h, wid h and
heigh ). Scale ba is 40 mm. The p o ocol s a ed by a “ho s a ” o
ac i a e he polyme ase enzyme o ≈10 min a ≈95 °C, ollowed by
40 cycles o PCR ampli ica ion. Each ampli ica ion cycle consis ed o 3
s eps. Fi s , dena u a ion o ≈10 s a ≈95 °C, hen annealing o ≈10 s
a ≈50 °C and he las s ep was ex ension o ≈15 s a ≈68 °C (ins ead
o ypical ≈72 °C), du ing which luo escence was measu ed. The LCD
display shows 6 cycles. A he comple ion o each PCR ampli ica ion
cycle, luo escence ampli ude a each spo was calcula ed and
ampli ica ion cu es we e plo ed in (B). We placed NTC a posi ion 1, a
low concen a ion o complemen a y DNA (cDNA) om H7N9 HA
gene a posi ion 3, a medium concen a ion a posi ion 2 and he
highes concen a ion (posi i e con ol) a posi ion 4. The esul s show
ha he PCR eac ion was success ully accomplished wi hou PCR
ampli ica ion o he NTC sample. Fu he mo e, he esul s p o e ha
he samples we e no c oss con amina ed, he eby elimina ing alse
posi i e ou comes. Posi i e con ol esul s a posi ion 4 indica e he
absence o alse nega i e esul s, he eby showing a success ul PCR
ampli ica ion p ocess.
Lab on a Chip Pape
Open Access A icle. Published on 16 Decembe 2015. Downloaded on 23/06/2016 07:07:56.
This a icle is licensed unde a
C ea i e Commons A ibu ion 3.0 Unpo ed Licence.
View A icle Online
590 |Lab Chip,2016,16,586–592 This jou nal is © The Royal Socie y o Chemis y 2016
The de ice pe o mance was e alua ed using syn he ic
complemen a y DNA (cDNA) o he hemagglu inin o he
H7N9 a ian in luenza i us. Fo wa d and e e se p ime s
we e chosen as sugges ed ea lie :
23
o wa d p ime :
TACAGGGAAGAGGCAATGCA, e e se p ime :
AACATGATGCCCCGAAGCTA, gi ing a o al amplicon leng h
o 104 base pai s wi h a mel ing empe a u e o ≈81.1 °C, as
measu ed using a comme cial eal ime PCR sys em.
The PCR mas e mix was p epa ed by mixing ≈2μLo
Fas S a DNA Mas e SYBR G een I, ≈2μL o MgCl
2
solu ion,
≈2μL o sample HA (5 ×10
−5
ng μL
−1
), ≈0.3 μLo ≈400 mg
mL
−1
BSA solu ion and p ime s in a inal concen a ion o
≈1.8 ×10
−6
mol L
−1
. We added deionized (DI) wa e wi h a e-
sis i i y highe han 18 MΩcm a 25 °C o c ea e he inal
olume o ≈20 μL. The NTC sample had ≈2μL HA gene ol-
ume eplaced wi h DI wa e .
Fi s , we pe o med basic eal- ime PCR wi h di e en
con en s o cDNA pe μL as shown in Table 1 wi h NTC a po-
si ion 1 and di e en con en s a posi ions 2 o 4. The ampli-
ica ion cu es a e shown in Fig. 5A. Once he PCR p ocess
was comple ed, we also conduc ed a MCA (no shown he e).
The MCA shape is no sui able o pe o ming high-
esolu ion analysis;
24
ne e heless, i does show ha a spe-
ci ic DNA was ampli ied wi h he mel ing empe a u e o
83.36 ± 0.63 °C (mean ± s anda d de ia ion), in close p oxim-
i y o he measu ed alue o T
M
≈81.1 °C. The ma ginal di -
e ence in he T
M
alues is due o he unce ain y o calib a-
ion p ecision o he comme cial PCR sys em used as a
benchma k ool. The T
M
esolu ion is su icien o e i y spe-
ci ic DNA ampli ica ion; howe e , i may no be sui able o
pe o ming high- esolu ion mel ing cu e analysis.
25
We hen pe o med a se ies o ou iden ical measu e-
men s using an HA con en o ≈5×10
−5
ng in a μL o cDNA
(see da a in Table 2 and a g aphical ep esen a ion in
Fig. 5B). Di e en PCR loca ions esul ed in pe o mance a -
ia ion, consequen ly p oducing mean C
T
alues in he ange
o ≈8 o≈9.6 wi h a s anda d de ia ion anging be ween
≈0.8 and ≈1.5. The di e ence in C
T
alues a posi ions 2 o 4
migh be caused by impe ec ions due o manual VRC place-
men . The VRC a posi ion 1 appea s o ha e a lowe hea
ans e a e han VRCs a o he posi ions. We a ibu e he
a ia ion o a slowe ansi ion om ampli ica ion o sa u a-
ion o he DNA ampli ica ion cu e (Fig. 5B). We u he p e-
sume ha hea e de ec s a posi ion 1 could gi e ise o a
empe a u e ha is di e en in compa ison wi h he o he
h ee posi ions, consequen ly leading o a di e ing PCR
e iciency. Addi ionally, his disc epancy could be a ibu ed
o he s ess induced du ing chip- o-PCB solde ing, gi ing
ise o bending o he chip. The silicon chip de o ma ions
could cause bo h a ia ions in he in e media e oil laye
hickness and a di e ing hea a e.
Finally, we ob ained s anda d PCR cu es. The samples
we e p epa ed by he ollowing p ocedu e. We mixed a sam-
ple wi h cDNA co esponding o 12 500 copies pe 200 nL ol-
ume. This sample was dilu ed 10×, yielding 1250 copies pe
200 nL olume; he nex dilu ion yielded 125 copies pe 200
nL olume. The las wo dilu ions had a s a is ical numbe o
12.5 copies pe 200 nL and 1.25 copy pe 200 nL olume, e-
spec i ely. The PCR esul s as well as he no malized da a a e
shown in Fig. 5C and D, demons a ing ha ou po able
eal- ime PCR is able o de ec a single DNA copy wi h an ex-
cellen e iciency o 0.91 ± 0.05 pe cycle (mean ± s anda d
de ia ion), which is well wi hin he equi ed ange o PCR e -
iciency be ween 0.8 and 1.0.
Discussion
The PCR p o ocol consis ed o a ≈10 min ho s a a ≈95
°C, ollowed by 40 cycles o ≈10 s a ≈95 °C, ≈10 s a ≈50 °C
and ≈15 s a ≈68 °C. Once he PCR ampli ica ion was com-
ple ed, we conduc ed an MCA wi h a scan a e o ≈0.2 °Cs
−1
.
This p o ocol equi ed a o al ime o ampli ica ion o less
han 35 minu es wi h an addi ional ≈150 s o he MCA. The
ul ima e speed o he PCR was no he p ima y a ge o his
wo k. Ne e heless, he ime equi ed o DNA ampli ica ion
can be sho ened by using di e en ypes o ho s a s,
Taqman chemis y (o bo h)
19
o e en no using he ho s a
a all.
26
He e we used he same hea e as in ou p e ious wo k. The
dissipa ed Joule hea Pdepends on he squa e o ol age V:
P=V
2
/R,whe eRis he hea e esis ance. The Joule hea dissi-
pa ion and he consequen hea ing a e we e enhanced wi h
ei he an inc eased ol age bias o by lowe ing he hea e esis-
ance. In a p e ious wo k,
22
we inc eased he bias ol age up
o 20 V using an ex e nal powe supply. He e, he hea e is
powe ed using a 12 V ex e nal powe supply in ei he an
AC/DC con e e o a ba e y ype con igu a ion. A change in
his ol age would equi e addi ional space o a s ep up ol -
age con e e . The addi ional ea u e would u he equi e
ei he a edesign o he hea e on a mic omachine, equi ing a
new mask o he me al li hog aphy le el, o he use o a
hicke me al laye wi h a lowe shee esis ance. Bo h cases
would equi e he ab ica ion o new PCR chip a chi ec u es.
14
The cu en PCR chip layou is shown in Fig. S3 (ESI†) and he
chip ab ica ion p ocess in sec ion 5 (ESI†).
In p inciple, he undamen al limi a ion in he speed o
he de ice is de e mined by he hea ans e be ween he
hea e and he sample, which is ≈1.5 s o each empe a u e
s ep. Ne e heless, he de ice can s ill un as as as i s p ede-
cesso achie ing ≈9.5 s pe PCR cycle, s ill being conside ed
as one o he as es eal- ime PCRs demons a ed a ha
poin in ime.
19
Table 1 Typical esul s wi h NTC (posi ion one) se ing as nega i e con-
ol and h ee di e en sample concen a ions a posi ions 2 o 4. The
sample a posi ion 4 se es as posi i e con ol
Posi ion Mean C
T
S anda d de ia ion Concen a ion HA (ng μL
−1
)
1—— —
2 28.7 1.5 ≈5.0 ×10
−8
3 27.0 1.0 ≈7.5 ×10
−8
4 20.0 1.0 ≈5.0 ×10
−6
Lab on a ChipPape
Open Access A icle. Published on 16 Decembe 2015. Downloaded on 23/06/2016 07:07:56.
This a icle is licensed unde a
C ea i e Commons A ibu ion 3.0 Unpo ed Licence.
View A icle Online

Lab Chip,2016,16,586–592 | 591This jou nal is © The Royal Socie y o Chemis y 2016
Fu he mo e, we enhance he sys em obus ness by no in-
co po a ing mo ing pa s. The sys em ligh sou ce consis s o
4 LEDs wi h an es ima ed li e ime o mo e han 50 000
hou s. A single 40 cycle PCR un equi es each LED o
ope a e o less han 1 min. The mos ulne able pa is he
mic omachined silicon chip moun ed di ec ly on he
main PCB. We en ision a new e sion o ou sys em, cu -
en ly unde de elopmen , wi h he b i le silicon chip
moun ed on a dedica ed PCB. The e o e, i he agile pa is
damaged, a eplacemen silicon chip can be easily ex-
changed. In o de o limi in e e ence be ween PWM pulses
and empe a u e sensing signals, he layou o he mic o-
machined silicon chip will inco po a e elec ical shielding
be ween in eg a ed hea e s and senso s. In his scena io, he
chip size will inc ease o ≈18 mm ×18 mm. Also, his
con igu a ion p o ides addi ional space in o de o u he
modi y he op ical housing and acili a e he emo al o
he PCB om he op ical pa h. The cu en e sion o he
sys em exhibi ed a la ge sel -induced luo escence due o he
PCB being illumina ed by he blue LEDs. Finally, we plan o
educe he complexi y o he op ical housing by educing
he numbe o pa s. This would also allow us o eplace
he silicon chip once a ia ions in PCR e iciency a e
disco e ed.
Fig. 5 (A) Single PCR un as i appea s in he PCR display. These PCR ampli ica ion cu es show esul s om he 4 posi ions o he PCR de ice.
We used complemen a y DNA (cDNA) om he H7N9 HA gene o es ing pu poses. AU s ands o a bi a y uni s. Once he PCR was comple ed,
he MCAs we e pe o med wi h a mel ing empe a u e alue o 83.36 ±0.63 °C (mean ±s anda d de ia ion). Measu emen unce ain ies emana e
om he manual placemen o he d ople . Sligh d ople misalignmen s a he hea e cause empe a u e a ia ions be ween a ious expe imen al
uns, he eby a ec ing he o e all PCR e iciency. Also, due o hese empe a u e a ia ions, we eadily obse e a sligh shi in he measu ed
mel ing empe a u e. (B) Measu emen s pe o med a he 4 posi ions wi h iden ical concen a ion o all samples pe o med h ee imes o
supp ess andom e o . The co esponding ex ac ed c i ical h esholds (C
T
) a e shown in Table 2. A g ea e alue o C
T
a posi ion 4 sugges s a
lowe ampli ica ion e iciency a ha posi ion. This may be caused by a empe a u e a ia ion due o a non-op imized bonding p ocess o he sili-
con chip o he PCB. (C) PCR esul s om posi ion 1 wi h he calcula ed numbe o cDNA copies in he sample om ≈12 500 down o ≈1.25. These
numbe s a e calcula ed by a 10×dilu ion s a ing om ≈12500. Since only whole numbe s o cDNA copies pe sample exis , ac ional alues imply
he s a is ically mos p obable alue. Each expe imen was pe o med h ee imes. (D) Ex ac ed s anda d eal- ime PCR cu e om esul s in
Fig. 5C showing he C
T
alue as a unc ion o LOG (cDNA concen a ion). The slope o −3.02 ±0.16 cycles a h eshold pe decade (mean ±s an-
da d de ia ion) co esponds o he PCR e iciency o 0.91 ±0.05 pe cycle (mean ±s anda d de ia ion).
Table 2 Resul s o c i ical h eshold om 4 measu emen s a each PCR
loca ion wi h iden ical concen a ion o he HA gene. The g aphical ou -
pu is shown in Fig. 5B. The disc epancy be ween esul s om indi idual
samples was p obably caused by sample misalignmen wi h espec o
he hea e as hey we e placed manually
Posi ion Mean C
T
S anda d de ia ion Concen a ion HA (ng μL
−1
)
1 9.6 1.5 ≈5.0 ×10
−5
2 9.0 1.2 ≈5.0 ×10
−5
3 9.5 0.6 ≈5.0 ×10
−5
4 8.0 0.8 ≈5.0 ×10
−5
Lab on a Chip Pape
Open Access A icle. Published on 16 Decembe 2015. Downloaded on 23/06/2016 07:07:56.
This a icle is licensed unde a
C ea i e Commons A ibu ion 3.0 Unpo ed Licence.
View A icle Online
592 |Lab Chip,2016,16,586–592 This jou nal is © The Royal Socie y o Chemis y 2016
Fu he mo e, in u u e expe imen s, we plan o use ei he
mic oscope glass co e slips wi h up on lyophilized PCR o
a single s ep RT-PCR mas e mix. In his con igu a ion, he
pipe ed ≈200 nL olume sample will be en i ely composed
o DNA (RNA).
Conclusions
We designed and es ed one o he smalles eal- ime PCR de-
ices. I has a leng h o ≈100 mm, a wid h o ≈60 mm and a
heigh o ≈33 mm and weighs only ≈90 g. The de ice mea-
su ed 4 PCRs simul aneously in less han ≈35 min, including
MCA. The sample was p ocessed in he o m o a i ual eac-
ion chambe (VRC) whe e 200 nL o a sample was placed on
a disposable glass co e slip co e ed wi h mine al oil o p e-
en wa e e apo a ion om he sample. The sample only in-
e ac s wi h he glass co e slip o elimina e possibili ies o
sample- o-sample c oss-con amina ion; he glass elemen was
a single-use, disposable sys em componen . Ou nega i e con-
ol es s u he demons a e he lack o c oss con amina ion
be ween samples. The sys em depic ed in Fig. 4 was success-
ully u ilized o a leas 100 dis inc PCR uns. We ha e dem-
ons a ed i s pe o mance by ampli ying he cDNA o an HA
gene o he H7N9 a ian in luenza i us and displayed he e-
sul s on an in eg a ed LCD display. We demons a ed he ca-
pabili y o simul aneously unning 4 samples a a ime wi h
good ep oducibili y. The PCR e iciency was demons a ed
by ob aining a PCR s anda d cu e in he ange o 12500 o
1.25 copies wi h an achie ed slope o −3.02 ± 0.16 cycles a
h eshold pe decade (mean ± s anda d de ia ion). The alue
co esponds o a PCR e iciency o 0.91 ± 0.05 pe cycle (mean
± s anda d de ia ion). The sys em was also capable o
de ec ing a single DNA copy wi hin he sample.
The cap u ed da a was subsequen ly ans e ed o a pe -
sonal compu e (PC) ia a USB in e ace o u he p ocess-
ing. This iny eal- ime PCR de ice is a p omising diagnos ic
sys em o emo e clinics as well as a ool o educa ional in-
s i u ions demons a ing he powe o a eal- ime PCR as
“seeing is belie ing”. The sys em h oughpu can be doubled
using a single channel mul iplexing me hod as demons a ed
ea lie .
27
Acknowledgemen s
P. Neužil acknowledges pa ial inancial suppo om he
Cen al Eu opean Ins i u e o Technology (CEITEC), g an
numbe CZ.1.05/1.1.00/02.0068. The au ho s g a e ully ac-
knowledge he NIST CNST NanoFab s a o help ul discus-
sions and assis ance wi h de ice ab ica ion. This a icle iden-
i ies ce ain comme cial equipmen , ins umen s, and
ma e ials o speci y he expe imen al p ocedu e. Such iden i i-
ca ion does no imply ecommenda ion o endo semen by
he Na ional Ins i u e o S anda ds and Technology, no does
i imply ha he equipmen , ins umen s, and ma e ials iden-
i ied a e necessa ily he bes a ailable o he pu pose.
Re e ences
1 K.Mullis,F.Faloona,S.Scha ,R.Saiki,G.Ho nandH.E lich,
Cold Sp ing Ha bo Symp. Quan . Biol., 1986, 51(P 1), 263–273.
2 R.K.Saiki,D.H.Gel and,S.S o el,S.J.Scha ,R.Higuchi,
G.T.Ho n,K.B.MullisandH.A.E lich,Science, 1988, 239,
487–491.
3S.YangandR.E.Ro hman,Lance In ec . Dis., 2004, 4, 337–348.
4 R.Deco eandJ.J.Cassiman,J. Med. Gene ., 1993, 30, 625–633.
5 M.A.JoblingandP.Gill,Na . Re . Gene ., 2004, 5, 739–751.
6M.A.Jobling,A.PandyaandC.Tyle Smi h,Z. Rech smed.,
1997, 110, 118–124.
7 S.S.Iqbal,M.W.Mayo,J.G.B uno,B.V.B onk,C.A.Ba and
J. P. Chambe s, Biosens. Bioelec on., 2000, 15, 549–578.
8 M. A. Valasek and J. J. Repa, Ad . Physiol. Educ., 2005, 29,
151–159.
9 P.Belg ade ,J.K.Smi h,V.W.WeednandM.A.No h up,
J. Fo ensic Sci., 1998, 43, 315–319.
10 P. Belg ade , S. Young, B. Yuan, M. P imeau, L. A. Ch is el, F.
Pou ahmadi and M. A. No h up, Anal. Chem., 2001, 73,391–391.
11 N. Ag awal, Y. A. Hassan and V. M. Ugaz, Angew. Chem., In . Ed.,
2007, 46, 4316–4319.
12 P.Neuzil,S.Giselb ech ,K.Laenge,T.J.HuangandA.Manz,
Na . Re . D ug Disco e y, 2012, 11, 620–632.
13 M. U. Kopp, A. J. Mello and A. Manz, Science, 1998, 280,
1046–1048.
14 P. Neuzil, J. Pippe and T. M. Hsieh, Mol. BioSys ., 2006, 2,
292–298.
15 P.Neuzil,T.M.HsiehandJ.Pippe ,US Pa ., 8216855, 2012.
16 J. Pippe , T. M. Hsieh and P. Neuzil, US Pa ., 8124033, 2012.
17 J. Pippe , M. Inoue, L. F. Ng, P. Neuzil, Y. Zhang and L. No ak,
Na . Med., 2007, 13, 1259–1263.
18 P.Neuzil,L.No ak,J.Pippe ,S.Lee,L.F.NgandC.Zhang,Lab
Chip, 2010, 10, 2632–2634.
19 P.Neuzil,C.Zhang,J.Pippe ,S.OhandL.Zhuo,Nucleic Acids
Res., 2006, 34, e77.
20 L. No ak, P. Neuzil, J. Pippe , Y. Zhang and S. Lee, Lab Chip,
2007, 7,27–29.
21 P.Neuzil,L.No akandJ.Pippe ,Pa . Appl., WO2008024080A,
2008.
22 P. Neuzil, W. Sun, T. Ka asek and A. Manz, Appl.Phys.Le .,
2015, 106.
23 V. M. Co man, M. Eickmann, O. Land , T. Bleicke , S. B ünink,
M. Eschbach-Bludau, M. Ma oso ich, S. Becke and C. D os en,
Eu o Su eill., 2013, 18(16), 20461. A ailable online: h p://www.
eu osu eillance.o g/ViewA icle.aspx?A icleId=20461.
24 G. H. Reed and C. T. Wi we , Clin. Chem., 2004, 50,1748–1754.
25 L.Zhou,J.Vande s een,L.Wang,T.Fulle ,M.Taylo ,B.Palais
and C. T. Wi we , Tissue An igens, 2004, 64, 156–164.
26 J. S. Fa a and C. T. Wi we , Clin. Chem., 2015, 61, 145–153.
27 C. D. Ah be g and P. Neuzil, Sci. Rep., 2015, 5, 12595.
Lab on a ChipPape
Open Access A icle. Published on 16 Decembe 2015. Downloaded on 23/06/2016 07:07:56.
This a icle is licensed unde a
C ea i e Commons A ibu ion 3.0 Unpo ed Licence.
View A icle Online