In e na ional Jou nal o
Molecula Sciences
A icle
Toxici y o Ca bon Nanoma e ials—Towa ds Reliable Viabili y
Assessmen ia New App oach in Flow Cy ome y
Tomáš Malina 1,2,* , Ka eˇ ina Poláko á1,* , Co dula Hi sch 3, Ladisla S oboda 4,5 and Radek Zboˇ il 1,4


Ci a ion: Malina, T.; Poláko á, K.;
Hi sch, C.; S oboda, L.; Zboˇ il, R.
Toxici y o Ca bon Nanoma e ials—
Towa ds Reliable Viabili y
Assessmen ia New App oach in
Flow Cy ome y. In . J. Mol. Sci. 2021,
22, 7750. h ps://doi.o g/10.3390/
ijms22147750
Academic Edi o : Vladimi N.
U e sky
Recei ed: 29 Ap il 2021
Accep ed: 14 July 2021
Published: 20 July 2021
Publishe ’s No e: MDPI s ays neu al
wi h ega d o ju isdic ional claims in
published maps and ins i u ional a il-
ia ions.
Copy igh : © 2021 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license (h ps://
c ea i ecommons.o g/licenses/by/
4.0/).
1Regional Cen e o Ad anced Technologies and Ma e ials, Czech Ad anced Technology and Resea ch
Ins i u e (CATRIN), PalackýUni e si y Olomouc, Šlech i el˚u 27, 779 00 Olomouc, Czech Republic;
[email p o ec ed]
2Depa men o Physical Chemis y, Facul y o Science, PalackýUni e si y Olomouc,
17. Lis opadu 12/1192, 771 00 Olomouc, Czech Republic
3Labo a o y o Pa icles-Biology In e ac ions, Empa, Swiss Fede al Labo a o ies o Ma e ials Science
and Technology, Le chen elds asse 5, 9014 S . Gallen, Swi ze land; [email p o ec ed]
4Nano echnology Cen e, Cen e o Ene gy and En i onmen al Technologies, VŠB−Technical Uni e si y
o Os a a, 17. Lis opadu 15/2172, 708 00 Os a a-Po uba, Czech Republic; ladisla [email p o ec ed]
5IT4Inno a ions Na ional Supe compu ing Cen e , VŠB–Technical Uni e si y o Os a a,
17. Lis opadu 15/2172, 708 00 Os a a, Czech Republic
*Co espondence: [email p o ec ed] (T.M.); [email p o ec ed] (K.P.)
Abs ac :
The scope o applica ion o ca bon nanoma e ials in biomedical, en i onmen al and
indus ial ields is ecen ly subs an ially inc easing. Since
in i o
oxici y es ing is he i s essen ial
s ep o any comme cial usage, i is c ucial o ha e a eliable me hod o analyze he po en ially
ha m ul e ec s o ca bon nanoma e ials. E en hough esea che s al eady epo ed he in e e ence
o ca bon nanoma e ials wi h common oxici y assays, he e is s ill, un o una ely, a la ge numbe
o s udies ha neglec his ac . In his s udy, we in es iga ed in e e ence o ou bio-p omising
ca bon nanoma e ials (g aphene acid (GA), cyanog aphene (GCN), g aphi ic ca bon ni ide (g-C
3
N
4
)
and ca bon do s (QCDs)) in commonly used LIVE/DEAD assay. When a s anda d p ocedu e was
applied, ma e ials caused a ious ypes o in e e ence. While posi i ely cha ged g-C
3
N
4
and QCDs
induced alse esul s h ough he c ea ion o ee agglome a es and in insic luo escence p ope ies,
nega i ely cha ged GA and GCN led o alse signals due o he complex quenching e ec o he
luo escen dye o a LIVE/DEAD ki . Thus, we de eloped a new app oach using a speci ic ga ing
s a egy based on addi ional con ols ha success ully o e came all ypes o in e e ence and lead o
eliable esul s in LIVE/DEAD assay. We sugges ha he newly de eloped p ocedu e should be a
manda o y ool o all in i o low cy ome y assays o any class o ca bon nanoma e ials.
Keywo ds: ca bon nanoma e ials; low cy ome y; cell iabili y; cy o oxici y; in e e ence
1. In oduc ion
Ca bon nanos uc u es ank among he mos p omising ma e ials in he ield o nan-
o echnology. They include he well-known ulle enes and ca bon nano ubes; howe e ,
he a en ion o esea che s has ecen ly shi ed mo e owa ds 2D ca bon nanoma e ials
(CNMs) and ca bon do s [
1
,
2
]. As o 2D ma e ials, he disco e y o g aphene in 2004 [
3
]
was he key poin , which led o he syn heses o a la ge numbe o new eme ging de i a-
i es [
4
]. These de i a i es possess se e al ex ao dina y p ope ies ha a e p omising o
a b oad spec um o applica ions [
4
,
5
]. This applies mos ly o hose de i a i es ha o e -
come g aphene’s hyd ophobici y h ough selec i e unc ionaliza ion. Highly hyd ophilic
de i a i es a e especially a ac i e o a ious ields o biomedical esea ch [
4
–
7
]. Fu he -
mo e, g aphi ic ca bon ni ides (g-C
3
N
4
) a e a new class o g aphene-like ma e ials ha
o e desi able op ical p ope ies simila o ano he ype o eme ging ca bon nanoma e ial
o he las decade—ca bon do s (CDs) [
8
,
9
]. Bo h o hese ma e ials a e highly p omising
In . J. Mol. Sci. 2021,22, 7750. h ps://doi.o g/10.3390/ijms22147750 h ps://www.mdpi.com/jou nal/ijms
In . J. Mol. Sci. 2021,22, 7750 2 o 15
o applica ions in biosensing and bioimaging because, compa ed o he luo escen ino -
ganic semiconduc o s, hey consis mainly o en i onmen ally iendly elemen s such as
ca bon (C), hyd ogen (H) and ni ogen (N), gi ing hem a emendous ad an age in hei
biocompa ibili y [8,10–12].
The po en ial ad e se e ec s o nanoma e ials (NMs) on human heal h ha e been o
gene al conce n in ecen yea s. An unde s anding o he cellula consequences o NMs
a e di ec con ac
in i o
is he i s impo an s ep and a c ucial p emise o hei sa e
and success ul use in biomedical applica ions [
13
]. Fu he mo e, he pa adigm o he
oxicology o he 21s cen u y is o eliably es on he basis o high h oughpu
in i o
cell
cul u e-based models o minimize animal use [
14
,
15
]. One o he mos impo an endpoin s
o
in i o
NMs es ing is acu e cy o oxici y. Based on his esul , i is possible o de ine he
dose- esponse cha ac e is ics o nanoma e ials and u he ocus on he unde s anding o
he NMs’ cellula in e ac ions wi h addi ional assays [
16
]. The e o e, i is c ucial o ha e a
wo king me hodology o his endpoin o p oduce eliable and jus i ied esul s.
I is widely known ha NMs cause in e e ence wi h s anda d iabili y assays [
17
–
20
],
which u he highligh s he need o a eliable and alid p ocedu e o NMs
in i o
es ing. Wha is ala ming is ha s udies e e ing o nanoma e ial oxici y do no gene ally
ake his in o ma ion in o accoun (o e 85% o pape s) [
18
,
20
,
21
]. This ac can cause a
huge p oblem o u u e sa e and sus ainable applica ions o speci ic nanoma e ials and
he e o e needs o be add essed as soon as possible because di e en NMs can cause
a ious ypes o in e e ence [
20
]. Ca bon nanoma e ials a e no excep ion, and in e e ence
o CNMs wi h common oxici y assays such as Alama blue, neu al ed, MTT and WST-1
assay has al eady been epo ed [
22
–
27
]. The e o e, o CNMs, using assays based on
spec oscopic de ec ion is ecommended only wi h g ea cau ion. Flow cy ome y is a
as and highly accu a e echnique p o iding in o ma ion abou indi idual cells in he
whole popula ion [
28
,
29
]. Thus, i ep esen s a sui able al e na i e o
in i o
iabili y
es ing. The LIVE/DEAD assay uses wo luo escence p obes (P opidium Iodide (PI) and
Calcein-AM) o dis inguish be ween he popula ion o dead and ali e cells. PI is able o
in e cala e in o he DNA o dead cells wi h a up u ed memb ane, while ac i e es e ases in
ali e cells ans o m he non- luo escen calcein-AM o highly luo escen calcein. On he
o he hand, he e is al eady a epo showing in e e ence o nanoma e ials e en wi h
low cy ome y [
30
]. The e o e, a ques ion should be aised whe he CNMs can also cause
in e e ence wi h low cy ome y and, mo e impo an ly, i he e is a way o o e come i .
In his s udy, we in es iga ed po en ial in e e ence eac ions o new p omising CNMs
such as 2D g aphene acid (GA), cyanog aphene (GCN), g aphi ic ca bon ni ide (g-C3N4)
and 0D ca bon do s (QCDs) in a basic low cy ome y assay (LIVE/DEAD). The i s in e -
e ence eac ion was igge ed by he in e ac ion o he ma e ial wi h he used luo escence
p obes. The second and mo e complica ed was he in e e ence o nanoma e ials inside o
he cells ha caused changes in he luo escence p ope ies o he p obes. In he case s udy
wi h human skin ib oblas cells (BJ), we o e came bo h o hese in e e ences and ob ained
eliable esul s o he LIVE/DEAD assay o hese CNMs wi h a newly de eloped p o ocol
using addi ional con ols.
2. Resul s and Discussion
2.1. P ope ies o Ma e ials
The ou ep esen a i es o CNMs we e selec ed o demons a e how di e en p ope -
ies o ma e ials could in luence hei in e e ence wi h
in i o
es ing. The size, su ace
cha ge and shape o ma e ials a e known o gene ally in luence no only
in i o
assays.
The cha ac e iza ion o ma e ials is summa ized in Table 1. I should be no ed ha Table 1
only p esen s in o ma ion gi en in ou p e ious pape s. GA and GCN had simila p ope -
ies in he su ace cha ge (ze a po en ial o
−
32 and
−
30 mV o GA and GCN, espec i ely),
size (<500 nm acco ding o DLS) and shape (bo h GA and GCN we e mono/ ew laye
shee s) (Table 1). Fu he mo e, nei he o hem exhibi ed luo escence p ope ies. On he
o he hand, as men ioned be o e, bo h QCDs and g-C
3
N
4
a e known o hei luo escence
In . J. Mol. Sci. 2021,22, 7750 3 o 15
quali ies and especially QCDs as ep esen a i es o ca bon do s ha e wide emission spec a
wi h s ong luo escence signals [
9
]. Bo h QCDs and g-C
3
N
4
exhibi ed posi i e su ace
cha ge (+40 mV o QCDs and +24 mV g-C
3
N
4
), bu hey had a comple ely di e en size
and shape. While QCDs we e e y small sphe e nanodo s (5 nm) [
12
], he g-C
3
N
4
we e
loose agglome a es wi h i egula shapes wi h he Z-a e age o 880 nm (acco ding o
DLS) [
31
,
32
]. To ha e a g ea e awa eness o he shape o NMs, we included ou own TEM
images, which a e displayed in Figu e 1.
Table 1. Cha ac e iza ion o ca bon nanoma e ials g-C3N4, QCDs, GA and GCN.
g-C3N4[31,32] QCDs [12] GA [7] GCN [7]
Size (DLS, nm) 880 5 200 300
Ze a po en ial (mV) +24 +40 −32 −30
Shape loose agglome a es wi h
i egula shape sphe e mono/ ew laye shee s
Figu e 1.
T ansmission elec on mic oscopy images o (
a
) QCDs (scale 20 nm); (
b
) g-C
3
N
4
(scale
500 nm); (c) GA and (d) GCN (bo h scales 100 nm).
2.2. Op ical Mic oscopy Imaging and MTT In e e ence
The i s impo an ac o in luencing possible in e e ence o nanoma e ials is hei
beha io in cell cul u e media. Al hough we had in o ma ion abou physico-chemical
p ope ies o s udied nanoma e ials (Table 1), hese can quickly change due o he in e -
ac ion wi h p o eins p esen in he cul u e medium (p ecisely in FBS) [
33
]. Simple op ical
In . J. Mol. Sci. 2021,22, 7750 4 o 15
mic oscopy images p esen ed in Figu es S1 and S2 showed ha he colloidal beha io o
NMs in cul u e media di e ed g ea ly, depending mainly on he su ace cha ge o ca bon
nanoma e ials. Fo he posi i ely cha ged g-C
3
N
4
and QCDs, ee agglome a es o NMs
we e obse ed o samples ea ed wi h 300
µ
g/mL (Figu e S1b,c). Fo he g-C
3
N
4
sample,
agglome a ion also occu ed a a concen a ion o 50
µ
g/mL (Figu e S1c). I should be no ed
ha p io o op ical mic oscopy imaging, he samples we e washed and obse ed in PBS,
meaning ha a huge amoun o ee agglome a es was al eady washed away. When we
obse ed he BJ cells ea ed wi h QCDs and g-C
3
N
4
in mo e de ail, we ound ha bo h
ma e ials a bo h concen a ions we e ei he in e nalized o a ached o he memb ane o
he cells (Figu e S2b,c). On he o he hand, in he samples ea ed wi h nega i ely cha ged
GA and GCN, he e we e ewe agglome a es seen ou side o he cells (mos ly only in
samples ea ed wi h 300
µ
g/mL), bu mo e impo an ly, he size o he agglome a es
was much smalle han in he case o g-C
3
N
4
and QCDs (Figu e S1d,e). Howe e , again,
bo h ma e ials we e seen up aken o a ached o he memb ane o he cells e en o a
concen a ion o 50
µ
g/mL and cells we e comple ely co e ed wi h GA and GCN o he
concen a ion o 300 µg/mL (Figu e S2d,e).
A he beginning o ou s udy, we wan ed o show he inapp op ia eness o using
he common oxici y assay o CNMs. In Figu e S3, i is clea ly shown ha especially
in he samples ea ed wi h GA and GCN, he nanoma e ials could massi ely in e e e
wi h any ype o spec opho ome ic o spec o luo ome ic e alua ion as he supe na an s
a e comple ely da k when compa ed o con ol samples. E en o he g-C
3
N
4
and QCDs
samples, he colo in supe na an s is di e en om he con ol samples. To assess he
po en ial in e e ence, we pe o med an MTT assay as an example. We included Blank
con ols (emp y wells wi h he same ea men as o wells wi h cells) o y o a oid
in e e ence, as i is a s anda d p o ocol o his ype o assay. F om he esul p esen ed
in Figu e S4, i is clea ha GA, GCN and g-C
3
N
4
samples caused in e e ence, as hei
MTT iabili y alues we e all below 90% o con ol (Figu e S5). The mos p onounced
d op was obse ed o he g-C
3
N
4
sample, as he MTT iabili ies we e unde 50% o
bo h concen a ions, which no mally indica es a huge cy o oxic e ec . Howe e , i was
de ini ely due o he s ong in e e ence. The whi e colo o he s ock solu ion in e e ed
wi h he blue/pu ple colo o he dissol ed o mazan c ys als because he mic oscopy
imaging showed almos no dead o damaged cells (Figu es S1 and S2). Addi ionally,
Figu e S4 shows ha cells ea ed wi h he g-C
3
N
4
sample we e much b igh e han he
un ea ed con ol. Fo he GA and GCN samples, we again obse ed almos no dead cells
in mic oscopy images (Figu es S1 and S2). On he o he hand, we saw many NMs ei he
co e ing he cells’ su ace o being inside o he cells, which could po en ially cause a
dec ease in he MTT signal h ough in e ac ion wi h o mazan c ys als, as was al eady
epo ed o o he CNMs [
25
,
27
]. We could no de e mine any in e e ence o he QCDs
sample, as all he alues we e simila o he un ea ed con ol and he e was no change
obse ed in he colo (Figu es S4 and S5).
2.3. In e e ence o CNMs in Fo wa d and Side Sca e P o iles in Flow Cy ome y
Di e en colloidal beha io o CNMs in cul u e media was con i med by low cy om-
e y in he o wa d and he side sca e p o iles as well (Figu e 2). The ex a popula ion
o e en s, besides popula ions o cells and deb is, occu ed only in hose samples ea ed
wi h g-C
3
N
4
and QCDs, which can be solely assigned o agglome a ed NMs
(Figu e 2b,c).
We p o ed his hypo hesis by using spike-in con ols, he i s c ucial con ols ha ha e o
be de ini ely used, whe e a simila popula ion o e en s was obse ed especially o he
g-C
3
N
4
and sligh ly also o he QCDs sample (Figu e 2d). Spike-in con ols a e mean o
mimic he highes possible concen a ion o ee NMs in samples by adding he app op ia e
olume o NMs s ock solu ions in wa e o he nega i e con ol sample igh be o e he
measu emen . The e was also a clea shi in he cells’ popula ion sca e p o ile compa ed
o he un ea ed sample, which indica es ei he in e naliza ion o memb ane a achmen o
bo h g-C
3
N
4
and QCDs samples. E en hough we did no see o he samples ea ed wi h
In . J. Mol. Sci. 2021,22, 7750 5 o 15
GA and GCN any e en s ep esen ing ee agglome a es in sca e p o iles (including spike-
in con ols), he cells’ popula ion shi ed again g ea ly, compa ed o he un ea ed sample
(Figu e 2). These esul s ag eed well wi h obse a ions om op ical mic oscopy images.
Figu e 2.
Fo wa d and side sca e p o iles o samples: (
a
) nega i e con ol o BJ cells (4
×
same sam-
ple wi h speci ic ga es o each NMs); BJ cells ea ed o 24 h wi h (
b
) 50
µ
g/mL and (
c
) 300
µ
g/mL
o ca bon nanoma e ials and (
d
) spike-in con ols ( om le o igh : g-C
3
N
4
, QCDs, GA and GCN).
The popula ion o cells is highligh ed in ed, he popula ion o deb is is ma ked blue and he
emaining black popula ion in samples g-C3N4and QCDs is conside ed NMs agglome a es.
As i is known om he li e a u e [
34
], posi i ely cha ged NMs show signi ican ly
highe in e ac ion wi h p o eins (especially albumin and globulins ha a e p esen in
FBS) han nega i ely o neu al cha ged NMs. The eason behind his phenomenon is
mos p obably he ac ha he mos abundan p o eins in FBS (albumin and globulins)
ca y a nega i e ne cha ge a physiological pH [
35
]. The e o e, a nanoma e ial wi h
posi i e ze a po en ial would p e e en ially in e ac wi h such p o eins. I was al eady
epo ed ha he ze a po en ial o NMs wi h he p o ein co ona is nega i e in mos cases,
which u he suppo s he hypo hesis o highe adso p ion o p o eins on NMs wi h
a posi i ely cha ged su ace [
35
,
36
]. Rega ding g-C
3
N
4
, i was unde s andable ha he
in e ac ion o la ge posi i ely cha ged NMs wi h p o eins esul ed in he o mula ion o
agglome a es o a size signi ican enough o be seen in mic oscopy images and sca e
p o iles
(Figu es 2, S1 and S2).
Su p isingly, e en 5 nm QCDs p oduced agglome a es wi h
p o eins big enough o be de ec ed by hose echniques. As men ioned be o e, one o he
easons is de ini ely he s ong posi i e su ace cha ge o QCDs (+40 mV) esul ing in a
massi e in e ac ion wi h p o eins. Howe e , as Glancy e al. ecen ly epo ed, he p o ein
co ona o sub-10 nm nanopa icles is mo e complex and nanopa icles can se e mo e as
ca go on a p o ein a he han as a ca ie o he p o ein, as is usually he case o la ge
NMs [
37
]. Hence, he agglome a es o QCDs in he cul u e medium could be a mix o
se e al nanopa icles in combina ion wi h a ious p o eins, which could explain he size o
hose agglome a es.

In . J. Mol. Sci. 2021,22, 7750 6 o 15
2.4. In e e ence o CNMs in Spike-in Con ols
Ha ing in mind a o ma ion o agglome a es o g-C
3
N
4
and QCDs in cul u e media
hese CNMs a e amous o hei ex ao dina y luo escence p ope ies [
12
,
32
]. Logically,
i was c ucial o check i hose agglome a es could in e e e wi h he luo escence p obes (PI,
Calcein) used in he LIVE/DEAD assay. This was ca ied ou using spike-in con ols again.
Fi s , we measu ed he spike-in con ols o all NMs and analyzed hem in a do plo
o ed channel (FL 2: ex. 488/em. 700 nm) o PI-posi i e cells agains g een channel
(FL 3: ex. 488/em. 527 nm) o calcein posi i e cells. To see i he ee NMs can in e e e
wi h he luo escence p obes, we ga ed ou he popula ion o deb is shown in
Figu e 2
(blue popula ion o e en s). The do plo s o he LIVE/DEAD assay a e shown in Figu e 3.
The i s impo an hing o obse e is ha e en a e we disca ded he deb is popula ion o
e en s, he e we e s ill some e en s o uns ained cells (uns ained ga e) wi h he luo escence
in ensi y in bo h channels lowe han 10
3
in he log scale in nega i e and posi i e con ol
samples (Figu e 3a). This was no included in he e alua ion o iabili y. Then, we c ea ed
ga es acco ding o he posi i e and nega i e con ol samples o dead ( ed, PI-posi i e)
and ali e (g een, calcein posi i e) cells (Figu e 3a) and used hem o e alua ion o he
iabili y o spike-in con ols o NMs (Figu e 3d). As expec ed o he nega i ely cha ged
GA and GCN, no popula ion o ee agglome a es o NMs was p esen o in e e e wi h
he esul s (Figu es 2d, 3b and S7b), as he iabili y o spike-in con ols o bo h o hese
ma e ials emained o e 90%, simila o he nega i e con ol
(Figu e 3d).
Howe e , o pos-
i i ely cha ged g-C
3
N
4
and QCDs, an addi ional popula ion o e en s appea ed in he
sca e p o iles (Figu e 2d) as well as in he do plo s
(Figu es 3c
and S7c). This popula ion
ep esen s he agglome a es o NMs because he only di e ence be ween hose samples
and he nega i e con ol (o posi i e con ol in he case o spike-in PC samples) was he
addi ion o NMs igh be o e he measu emen . The spike-in con ol o g-C
3
N
4
gene a ed
a huge numbe o agglome a es, compa able o he amoun we saw in he sample a e
24 h o ea men (Figu e 2c,d). Fu he mo e, he agglome a es in e e ed g ea ly wi h
he e alua ion as he iabili y o he g-C
3
N
4
spike-in sample d opped o
42% (Figu e 3d).
On he o he hand, in spike-in con ol o QCDs, he amoun o agglome a es was signi i-
can ly lowe han a e 24 h (Figu e 2c,d). This is mos p obably due o he di e ence in he
size o g-C
3
N
4
and QCDs because he o mula ion o agglome a es akes mo e ime wi h
ul a-small nanopa icles and he spike-in con ols a e measu ed immedia ely a e adding
he NMs. Howe e , e en a lowe amoun o agglome a es was esponsible o he dec ease
in iabili y o he QCDs spike-in con ol sample o 74% (Figu e 3d).
Spike-in con ols a e a necessa y ool in low cy ome y and should be used in e e y
measu emen , whe e any in e e ence o NMs is expec ed. Only when spike-in con ols
a e used, ga ing— he mos impo an s ep in low cy ome y—is pe o med p ope ly. I is
c ucial ha we ha e in o ma ion only abou he popula ion o cells in he sample. Ne e he-
less, spike-in con ols canno gi e in o ma ion abou cells’ sca e p o iles, cha ac e izing
he cell size and g anula i y (complexi y) [
38
]. O cou se, he size and g anula i y o cells
a e in luenced a e 24 h o ea men wi h NMs. Howe e , his is no he case wi h spike-in
con ols as he NMs a e added jus be o e he measu emen .
E en hough spike-in con ols helped massi ely wi h choosing he speci ic ga es
co ec ly (Figu e 2), hey did no gi e us any in o ma ion abou he beha io o NMs
inside he cells o on hei memb ane. Gi en ha all NMs somehow in luenced he cells’
sca e p o iles (Figu e 2), we needed o in oduce ano he impo an ype o con ol named
nanoma e ial posi i e con ol (NM PCs). In hese con ols, we mimicked he si ua ion
whe e all he cells ea ed wi h NMs o 24 h would be dead o ge in o ma ion abou he
p o ile o dead cells wi h NMs on hei memb ane o inside. We hea -killed he cells ea ed
wi h NMs o 24 h be o e he measu emen . We did no need any con ol o ali e cells
ea ed wi h NMs o 24 h as, acco ding o he op ical mic oscopy, he majo i y o he cells
we e ali e in all samples (Figu e S2).
In he nex chap e s, we will gi e de ailed in o ma ion on how i is possible o se-
lec co ec ga ing, a oiding di e en ypes o in e e ence, and ge eliable esul s in
In . J. Mol. Sci. 2021,22, 7750 7 o 15
LIVE/DEAD assay o all ou ypes o ou chosen CNMs ollowing ou newly de eloped
app oach using a combina ion o bo h addi ional con ols. Fo a s anda d p ocedu e, we ap-
plied he well-known me hod o ga ing using nega i e (o e 90% iabili y) and posi i e
con ol (unde 10% iabili y). In ou new app oach, ga es we e se as ollows: iabili y o
nega i e and spike-in con ols o e 90%, he iabili y o posi i e and nanoma e ial posi i e
con ols (NM PCs) unde 10%.
Figu e 3.
Do plo showing LIVE/DEAD assay o : (
a
) nega i e and posi i e con ol o BJ cells; (
b
) BJ
cells spike-in con ols o GA (
le
) and GCN (
igh
) and (
c
) BJ cells spike-in con ols o g-C
3
N
4
(
le
)
and QCDs (
igh
). Ga es we e selec ed acco ding o he NC and PC samples. (
d
) The e alua ion o
he iabili y o spike-in con ols’ samples (n= 3). The e en s in he ali e ga e a e shown in g een,
e en s in he dead ga e in ed and e en s in he uns ained ga e a e highligh ed in blue.
In . J. Mol. Sci. 2021,22, 7750 8 o 15
2.5. In e e ence o QCDs in LIVE/DEAD Assay
As i is seen in Figu e 2, o samples ea ed wi h QCDs, he e was a clea change in he
sca e p o iles and ano he popula ion o agglome a es eme ged o he samples ea ed
wi h 50
µ
g/mL and especially 300
µ
g/mL. Tha is why i was qui e challenging o ga e only
he popula ion o cells as agglome a es migh also in e mingle o some ex en . Using he
spike-in con ol, we could obse e he p o ile o ee agglome a es in he do plo o ed
agains g een channels (Figu es 3c and S7c). The e o e, when he e we e almos no e en s
in he do plo o he sample a e 24 h, which co ela ed wi h he ee agglome a es in he
do plo o spike-in con ol, we knew we had he co ec ga e, as i displayed he popula ion
o cells wi h as ew agglome a es as possible (Figu e 2). Wi h his ga ing, we pe o med
LIVE/DEAD assay e alua ion using do plo s o wo luo escen channels.
Fi s , we analyzed he samples wi h s anda d ga ing using only nega i e and posi i e
con ol. F om Figu e 4a–e, i is clea ha he e was he in e e ence o op ical p ope ies o
QCDs [
12
,
31
,
32
] e en hough he iabili y o NM PCs was unde 10%, which was co ec
o hea -killed cells (Figu e 4e). In e ac ion o QCDs wi h cells esul ed in he shi in
he luo escence in ensi y in he ed de ec o (FL 2: ex. 488/em. 700 nm), which was
s ong enough o mo e some o he cells om he ali e ga e o he dead ga e (Figu e 4b).
Addi ionally, al hough we ca e ully ied o ga e only he popula ion o cells (Figu e 2),
he e we e s ill some e en s ep esen ing ee agglome a es le in he samples, as he ia-
bili y o spike-in con ol o QCDs dec eased o 86% (Figu e 4d,e). The e o e, he e was an
in e e ence wi h he e alua ion as he luo escence shi and he p esence o agglome a es
in ga ing acco ding o he con ol samples esul ed in dec easing he iabili y o 70 and
32% o samples ea ed wi h 50 and 300 µg/mL o QCDs, espec i ely (Figu e 4e).
Thus, he e we e wo challenges o he LIVE/DEAD assay o BJ cells ea ed wi h
QCDs. Fi s was he in luence o he op ical p ope ies o QCDs, and, second was he
p esence o ee agglome a es in cul u e media ha we e no ga ed ou o e alua ion.
To show how o o e come hese challenges, we analyzed he same da ase wi h a new
ga ing p o ocol acco ding o he addi ional con ols, which is displayed in Figu e 4 –j.
Using hese ga es, he whole popula ion o ali e cells emained in he ali e ga e despi e he
luo escence shi (Figu e 4g) and he e en s ep esen ing ee agglome a es we e now in
he uns ained ga e, and we e no included in he e alua ion (Figu e 4i). This se up esul ed
in 98 and 80% o he iabili y o BJ cells a e 24 h o incuba ion wi h 50 and 300
µ
g/mL o
QCD, espec i ely (Figu e 4j).
The e o e, due o he in e e ence in he analysis acco ding o s anda d ga ing, QCDs
could be alsely conside ed oxic o BJ cells, al hough he eliable iabili y o BJ cells did
no d op unde 80% e en o concen a ion 300 µg/mL o QCDs.
2.6. In e e ence o g-C3N4in LIVE/DEAD Assay
Fo he g-C
3
N
4
sample, e en be o e he LIVE/DEAD assay i sel , we obse ed ano he
challenge. The in e ac ion o cells wi h 50
µ
g/mL and especially 300
µ
g/mL o g-C
3
N
4
caused he side sca e alues o hese samples o shi ou o he de ined scale (Figu e 2).
We ha e no obse ed his phenomenon wi h any o he ma e ial be o e. As o he igh
e alua ion, we needed only he popula ion o cells (pa ame e s o measu emen we e se
acco ding o he con ols and hose canno be changed du ing he measu emen ). Tha is
why we used a do plo p o ile o o wa d sca e alues agains alues o he luo escen
channel (FL–1: ex. 405/em. 528 nm) o ga ing he popula ion o cells whe e we did no
expec any inc ease in he luo escence in ensi y o calcein and PI (Figu e S6). E en wi h
he use o ano he de ec o , he e was an o e lap be ween some e en s o he popula ions o
ee NMs and e en s om he popula ion o cells (Figu e S6b). We used he same p ocedu e
as we did o he QCDs samples o y o ga e only he popula ion o cells wi hou any ee
agglome a es o g-C
3
N
4
(Figu e S6a). Howe e , a a concen a ion o 300
µ
g/mL, some o
g-C
3
N
4
ee agglome a es had o be ga ed as well as he e was no clea line ha would
sepa a e hem om he cells (Figu e S6b).
In . J. Mol. Sci. 2021,22, 7750 9 o 15
In he s anda d ga ing, e en hough he luo escence in ensi y o g-C
3
N
4
was no
as dis inc i e as o QCDs, he e was a shi in some cells om he ali e o he dead ga e
in he sample ea ed wi h 300
µ
g/mL (Figu e 5b). A mo e se ious p oblem was ee
agglome a es ha we e ga ed in he popula ion o cells, as was men ioned abo e, and we e
now p esen ed in he dead ga e. This was p onounced in he spike-in con ol g-C
3
N
4
sample, which had 88% iabili y (Figu e 5d,e). Al hough he dec ease in iabili y was
sligh e han o QCDs, i was ano he indica ion ha he e was in e e ence in he sample.
Fu he mo e, he e en s p esen ed in Figu e 5d showed a do plo p o ile compa able o he
one in he spike-in con ol o g-C
3
N
4
displayed in Figu e 3c. Ne e heless, because o he
men ioned o e lap o agglome a es wi h cells in he samples a e 24 h (Figu e S6), we had
o include hem, as we could no a o d o lose in o ma ion abou hose cells. Thus, due o
he s anda d ga ing acco ding o he posi i e and nega i e con ol, e en s ep esen ing
ee agglome a es in he dead ga e we e esponsible o he d op in he iabili y o samples
ea ed wi h 50 and 300 µg/mL o g-C3N4 o 86 and 63%, espec i ely (Figu e 5e).
Figu e 4.
Do plo showing LIVE/DEAD assay o : (
a
,
) nega i e and posi i e con ol o BJ cells
and (
b
–
d
,
g
–
i
) BJ cells ea ed wi h QCDs samples wi h (
a
–
d
) s anda d ga ing and (
–
i
) new ga ing
app oach. (
b
,
g
) Samples ea ed wi h 50
µ
g/mL (
le
) and 300
µ
g/mL (
igh
) o QCDs o 24 h;
(
c
,
h
) Samples ea ed wi h NM PC 50
µ
g/mL (
le
) and NM PC 300
µ
g/mL (
igh
); (
d
,
i
) Spike-in
con ol o QCDs. (
e
,
j
). The e alua ion o he iabili y o BJ cells ea ed wi h QCDs and addi ional
con ol samples (n= 3). The e en s in he ali e ga e a e shown in g een, e en s in he dead ga e in ed
and e en s in he uns ained ga e a e highligh ed in blue.
By applying new ga ing acco ding o addi ional con ols p esen ed in Figu e 5 –j,
we managed o a oid his in e e ence. Using speci ic ga es in he LIVE/DEAD do plo ,
we e alua ed only he popula ion o cells. E en s ep esen ing ee agglome a es o g-C
3
N
4