a dna-based nano-immunoassay for the label-free detection of glial fibrillary acidic protein in...
TRANSCRIPT
1
2
3Q3Q1 Q4
4
5
6Q57
8Q69
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Nanomedicine: Nanotechnology, Biology, and Medicinexx (2014) xxx–xxx
nanomedjournal.com
NANO-00933; No of Pages 8
RO
OF
A DNA-based nano-immunoassay for the label-free detection of glialfibrillary acidic protein in multicell lysates
M. Ganau, MD, PhDa,b,1, A. Bosco, PhDb,1, A. Palma, PhDc,d, S. Corvagliaa,b,P. Parisse, PhDb,e, L. Fruk, PhDf, A.P. Beltrami, MDc, D. Cesselli, MDc,d,
L. Casalis, PhDb,⁎, G. Scoles, PhDb,c,d
aUniversity of Trieste, Trieste, ItalybNanoInnovation Lab–Elettra Sincrotrone Trieste S.C.p.A, Trieste, Italy
cDepartment of Medical and Biological Sciences, University of Udine, Udine, ItalydInstitute of Pathology, University Udine Hospital, Udine, Italy
eINSTM, Trieste ST Unit, Trieste ItalyfKarlsruhe Institute of Technology DFG, Centre for Functional Nanostructures, Karlsruhe, Germany
Received 21 January 2014; accepted 15 April 2014
RECTED P
Abstract
We have developed a quantitative approach to eventually enable precise and multiplexing protein analysis of very small systems, down toa single or a few cells. Through DNA-directed immobilization of DNA–protein conjugates we immobilized antibodies specific for a certainprotein of interest, on a complementary DNA nanoarray fabricated by means of nanografting, a nanolithography technique based on atomicforce microscopy (AFM). The proof of concept was realized for glial fibrillary acidic protein (GFAP), a biomarker crucial in cell'sdifferentiation of astrocytes, and functional to grade classification of gliomas, the most common of primary malignant brain tumors. Theefficiency of the nano-immuno sensing was tested by obtaining the immobilization of purified recombinant GFAP protein at differentconcentration in a standard solution then in a cellular lysate. A comparison of sensitivity between our technique and conventional ELISAassays is provided at the end of the paper.© 2014 Published by Elsevier Inc.
Key words: Nanodiagnostics; Cancer biomarkers; Atomic force microscopy; DNA-directed immobilization
31
32
33
34
35
36
37
CO
RBackground
Based upon the latest genetic and proteomic insights intocancer biology, which opened new avenues for novel appliedclinical research, trends in oncology highlight that molecularcharacterization of the tumorigenesis process will be essential intomorrow's clinical practice both to predict prognosis and to
UN 38
39
40
41
42
43
44
45
46
47
Financial support: This work was supported by FIRB 2011 grant“Nanotechnological approaches toward tumor theragnostic” to P.P. and AP.B.,ERC grant “Monalisa Quidproquo” to A.P., D.C. and G.S., and a grant fromthe Associazione Italiana per la Ricerca sul Cancro (AIRC 5 per mille SpecialProgram No. 12214) to A.B. and L.C.
⁎Corresponding author at: NanoInnovation Lab, Elettra SincrotroneTrieste S.C.p.A., 34149 Basovizza, Trieste, Italy.
E-mail address: [email protected] (L. Casalis).1 These authors equally contributed to the present article.
http://dx.doi.org/10.1016/j.nano.2014.04.0061549-9634/© 2014 Published by Elsevier Inc.
Please cite this article as: Ganau M., et al., A DNA-based nano-immunoassay foNanomedicine: NBM 2014;xx:1-8, http://dx.doi.org/10.1016/j.nano.2014.04.00
guide therapy.1 Noteworthy, the promise of individualizedmolecular medicine, that is particularly relevant to oncology,where even similarly classified tumors can follow quite differentclinical outcomes, could be realized by identifying moleculartargets for therapy and by measuring tangible response orregression in clinical trials.2
Specific patterns of protein expression in tumors and matchednormal tissues can now be reliably analyzed using quantitativeproteomic techniques. Among them the most effective areenzyme-linked immunosorbent assay (ELISA), two-dimensionalgel electrophoresis (2DGE) and matrix-assisted laser desorptionionization time-of-flight (MALDI-TOF) mass spectrometry,which allow to simultaneously identify and characterizedifferentially expressed proteins. Nevertheless, when it comesto the analysis of small cell's volumes (μl to pl), as in case ofcirculating tumor cells or tumor stem cells, only miniaturizedarrays of proteins may boost the detection of key biomolecules,
r the label-free detection of glial fibrillary acidic protein in multicell lysates.6
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
2 M. Ganau et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx
UNCO
RREC
eventually responsible for cellular–cellular or surface–cellularinteractions.2 In fact, one of the main limits to knowledgeadvancement in oncology is the sensitivity required to identifyspecific tumor (stem) cells protein patterns and to correlate themwith cell behavior.3,4 Such caveat suggests that new quantitativeapproaches, as the one based on nano-immunoarrays technology,could be highly effective in enabling a precise, and ultimately,low cost analysis of tumor cells' protein content (down to the fewcells level). More in general, nanotechnology-based approachescould be (and also are currently being) explored to discover,identify and quantify clinically useful molecular signatures forearly detection, diagnosis and prognosis of several tumors.5–8
Focusing our attention to neuro-oncology we identified theremarkable need for more sophisticated diagnostic tools withpotential capability of unprecedented pathological screening andsubtyping of glial tumors. These latter are themost common primarymalignant brain tumors, and, among them, themost aggressive ones,namely flioblastomas (WHOGrade IV) and anaplastic astrocytomas(WHO Grade III), present an incidence of 3.5-2.8 and 0.3-1.2 newcases per 100,000 per year respectively.9,10 Gliomas are character-ized by rapid growth, high level of cellular heterogeneity due togenetic alteration, and invasive behavior. Complete microscopicexcision followed by adjuvant radio- or chemotherapy represents thestandard of care, nevertheless tumor recurrence generally occurswithin fewmonths due to thewidespread neoplastic infiltration.11 Infact, primary brain tumors invade widely spreading single cellsanywhere within the brain parenchyma, through infiltration of bloodvessel walls, subpial glial spaces, or white matter tracts. Thesemechanisms lead to the development of tumor satellites, probablysustained by glial stem cells escaping resection and treatment,eventually serving as reservoirs for tumor recurrences. Despitecontinuous refinements in therapeutic strategies,12 the prognosis forpatients with high-grade gliomas remains dismal, and less than 5%survive more than 5 years despite aggressive therapies.11,13 As aresult, the median survival of glioblastoma is 14 months forpatients undergoing surgery followed by radiotherapy andadjuvant temozolomide14 and 22 months for those treated withchemotherapy in the presence of MGMT (O6-methylguanine-DNA methyltransferase) promoter methylation.15,16
There are several challenges to overcome in the successfultreatment of glioblastoma. Among others, these include thecharacteristic tumor heterogeneity of glioblastoma, and thepossible difficulty in overcoming resistant cancer stem cells(CSCs).17 In fact, despite some controversy concerning theirexact role and characterization, CSCs are believed to play acrucial role in malignant glioma tumor initiation, progression,angiogenesis and both drug- and radio-resistance.18 Therefore,the characterization of the protein content of glioma CSCs canhelp in developing drugs specifically targeting this extremelyrare tumor subpopulation and may improve the molecularclassification of brain tumors.17 As mentioned, this possibilitywould be made possible only by the development of innovativenano-immuno-arrays.
A systematic review of multiple independent proteomicanalyses of gliomas published on PubMed since 2008 hasdemonstrated alterations of almost 100 different proteins; amongthem we have chosen the glial fibrillary acidic protein (GFAP), abiomarker belonging to the family of intermediate filaments for
RO
OF
our proof of concept. GFAP is expressed by numerous celltypes of the central nervous system (CNS), especially astrocytesand is crucial in cell's differentiation, so that only well-differentiated cells retain the ability to express it, while themost aggressive ones lose this typical feature at a certain point intheir dedifferentiation.
As a proof of concept, we designed specific nano-immunoarrays capable to detect the presence of GFAP expressionin small volumes and envisage the possibility to integrate them to afacing array of microfabricated wells (see Figure 1) hosting livingastrocytes (ideally, one cell per microwell).
To realize the nano-immunoassay for the detection of GFAPwe used a combination of DNA barcoding, one of the leadingtechnology for multiplexing bioassays and DNA-directedimmobilization (DDI)19–23 of DNA–protein conjugates. Usingnanografting, an AFM-assisted nanolithographic technique24–26
we confined DNA molecules at the nanoscale in a nanoarrayformat, within a self-assembled monolayer (SAM) of biorepel-lent alkanethiols. By nanografting we then produced aminiaturized device, with the control of DNA probe density foroptimal device sensitivity.26,27
TED PMethods
Microwell fabrication
Arrays of microwells with different geometries (lateral dimen-sion 50-100 μm, depth 25 μm)were produced in three steps. First, asilicon-SU8 master was produced by standard photolithographyobtaining a pillar-like structure. Second, a polydimethilsiloxane(PDMS, Sylgard 184, Silicon elastomer, Dow Corning) mold wascreated casting a drop of PDMS (1:10 curing agent to prepolymer,as suggested by the manufacturer) on the master and treated at60 °C for 1 hour to promote the polymerization. Third, themembrane was peeled off and stuck on a glass slide.
Surface functionalization
In order to permit the neuron adhesion and growth on PDMSsurface of microwells we functionalized them with an invertedmicrocontact printing method. First of all the PDMS membranewas exposed to air plasma for 5 minutes to create negativecharges promoting hydrophilicity of the surface and theabsorption of positively charged molecule. A polyornithinesolution (0.01% polyornithine plus 5% Lucifer yellow for thevisualization, Sigma-Aldrich) was incubated on the surface for1 hour and then dried under a nitrogen stream. The excess ofpolyornithine outside each microwell was removed invertingimmediately the sample on a glass slide overnight and a weightwas placed on top during the stamping period.
Nano-assay
Multiple nanografting assembled monolayers (NAM) ofthiol-modified single-strand DNA (ssDNA) were prepared byserial AFM-based nanografting inside a self-assembled mono-layer of a top oligo-ethylene glycol-terminated alkanethiol,TOEG ((1-mercaptoundec-11-yl)hexa(ethyleneglycol), HS-(CH2)11-(OCH2CH2)6-OH from Sigma-Aldrich) on ultraflat
TED P
RO
OF
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
Figure 1. Nano-immuno detection platform for few cells proteomic screening. (A) Cartoon showing the hosting of one cell per well. (B) Biosensor topped wellcontaining a single cell secreting a protein of interest (antigen). (C) Representative AFM topographic image of the area containing of the DNA-based nano-immunosensors. (D) Cartoon showing the elements of one nano-immunosensor (patch). (E) Scheme for the realization of the nano-immuno-patch.
3M. Ganau et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx
UNCO
RREC
gold surfaces28 following the standard protocol reportedearlier.29 The DNA patches were obtained promoting thereplacement of the TOEG molecules with the oligonucleotidesby scanning an area of 1 μm × 1 μm with the AFM tip applyinga large force (about 100 nN) in the presence of a solution ofthiolated DNA sequences (5 μM in TE buffer 1 M NaCl) with ascan rate of 2 Hz. Several patches of two different sequences(cD1-SH and cD2-SH as control sequence) have been produced.
DDI strategy is used to immobilize anti-GFAP antibodies. Firsta solution of DNA–streptavidin conjugate (D1-STV) 500 nM inTE buffer 1 MNaCl is placed in contact with the sample for 1 hourat room temperature, promoting the hybridization of the DNAstrands. Subsequently, a solution of 100 μl containing biotinylatedmonoclonal GFAP antibodies (Synaptic Systems, dilution 1:100)in blocking buffer (3% bovine serum albumin (Sigma-Aldrich),0.05% Tween 20 in TBS) was incubated over the SAM for 1 hourat room temperature. The patches were than washed 3 times withTBST (0.05%Tween 20 in TBS) for 5 minutes. For every step, thetopographic images of the patcheswere successively collectedwithanAFM in contactmode and in liquid environment (300 μl of TBSbuffer) at a minimum force (always at the limit of AFM tipengaging) and constant 1-Hz speed.
Finally, produced nano-immunoarrays were tested toward therecognition of GFAP (Synaptic Systems), either added to theblocking buffer (BB) at a final concentration ranging from 0.2to100 nM, or to a cell lysate, at a final concentration of 40 nM.Celllysates were obtained by lysing 106 glioblastoma U87 cells in100 μl of RIPA buffer. In both experiments samples were
incubated for 1 hour at room temperature. The subsequent AFMimaging was performed as usual in TBS after three washes withTBST (5 minutes each).
AFM nanoassay data analysis
Results of the AFM topography analysis were expressed asmean ± standard deviation obtained from at least three indepen-dent patches. Paired one-tailed t-test has been used to evaluateheight increase upon immobilization on the nanopatches. Pairedtwo-tailed t-test has been used to compare the roughness on thebiorepellent SAM and on the nanopatches. For each test the P-value has been evaluated; P is the probability to commit an errorwhile rejecting the null hypothesis (no height variation in the firstcase or no roughness variation in the second) and values less than0.05 were considered significant.
ELISA assay
Microtiter plates were coated with 200 ng of GFAP anti-mouse IgG (Synaptic Systems) in 0.1 M sodium carbonate(pH 9.6) overnight at 4 °C and were blocked with 3% bovineserum albumin (Sigma) in phosphate-buffered saline containing0.05% Tween 20 (PBST) for 12 hours. After three washes withPBST, plates were transferred at room temperature and incubatedwith 100 μl cellular lysates and different concentrations ofGFAP protein (Synaptic Systems) diluted in 100 μl blockingsolution for 2 hours. Plates were washed three times with PBSTand incubated 1 hour with a 1:1000 dilution of the detector
F
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
Figure 2. Functionalization of PDMS microwells with polyornithine stainedin green by lucifer yellow. Scale bar: 100 μm.
4 M. Ganau et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx
EC
antibody (GFAP biotinylated antibody, Synaptic Systems). Thiswas followed by three washes with PBST, and an incubation of1 hour with the horseradish peroxidase (HRP)-conjugatedstreptavidin (Millipore), in a 1:5000 dilution in blocking buffer.Finally, plates were washed three times with PBST anddeveloped with TMB reagent (Sigma) for 5 minutes before thereaction was stopped by the addition of 1 M H2SO4. Absorbanceat 450 nm was measured in a Titertek Multiskan MCC/340plate reader.
Cell culture and immunofluorescence
U251-MG cells (ATCC) were cultured in DMEM added with10% fetal bovine serum (FBS). Glioma stem cells were isolatedand cultured applying, with minor modifications, a protocoloptimized for culturing glioma stem cells in adhesion.30 Briefly,glioma fragments were mechanically enzymatically dissociatedand cells less than 40 μm in diameter were cultured in expansionmedium on laminin-coated dishes (10 μg/mL, Sigma-Aldrich) asin Ref.30 The expression, on cultured glioma cells, of the stemcell marker CD133 (Miltenyi) was assessed by FACS (FacsAriaIII, BD Biosciences) (data not shown).
U251-MG cells and tumoral cells isolated from humanglioma biopsies (2 × 104 cells/mL) were seeded on a membraneof PDMS coated with 0.001% fibronectin (Sigma-Aldrich). Thenext day, cells were gently fixed with paraformaldehyde (PFA,Sigma-Aldrich). PFA of 4% was added into the cell medium to afinal concentration of 2% paraformaldehyde. Upon perme-abilization with 0.1% Triton X-100 for 10 minutes, cells weresubjected to immunofluorescence staining with anti-GFAPantibody (Dako, 1:100 in PBS 1X) for 2 hours at 37 °C. Thecells were then washed 3 times in PBS 1X and incubated withAlexa-555 labeled anti-rabbit antibody (Invitrogen, 1:800 in PBS1X) at room temperature for 1 hour. DNA was stained with 4′,6-diamidino-2-phenylindole (DAPI, Sigma). Epifluorescence andphase contrast images were collected by using the Leicamicroscope DMI6000B; 40× objective was employed.
R283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
UNCO
RResults
Fabrication and functionalization of microwells
The fabrication and functionalization of microwells areoperationally straightforward, require no special equipment andcan be carried out in a conventional laboratory on an inexpensiveoptically transparent polymeric support. The main advantage ofthe PDMS substrates is that they allow to pattern cells withinmicrowells of the desired diameters, so that, according to thepurposes of astrocyte analysis, our microwells were fabricatedwith a size of 100 μm in diameter × 25 μm in depth. Moreoverthe use of lucifer yellow to mark the presence of polyornithinewithin the microwells nicely demonstrated their effectivefunctionalization, as indicated by the fluorescent signal only inthe well's floor and walls (Figure 2). Finally, to test the ability ofcells to adhere and survive within the functionalized microwellswe conducted a series of trials (imaged via inverted opticalmicroscope) demonstrating that the use of adhesion substratessuch as polyornithine and fibronectin is mandatory to host living
TED P
RO
O
neural cells into the PDMS microwells (Figure 3). In fact, inorder to test the biocompatibility of our micropatterned PDMSwells, we decided to grow a human glioblastoma commercial cellline (U251-MG) as well as CD133-positive stem cells isolatedfrom glioblastoma biopsies on a fibronectin-coated PDMSmembrane. Cells entered into the PDMS microwells and adheredto the floor, where they survived for few days. Both U251 andglioma-isolated stem cells express high levels of GFAP, so theirpresence and viability within the microwells were revealed andconfirmed by anti-GFAP staining (Figure 3).
Preparation of nanopatches via DDI of biotinylated antibodiesand recognition of GFAP with and without cell lysate
The preparation of nano-immunoarrays was carried out with theaim to optimize protocols for future proteomicsmultiplexing analyses.In this regard, on one hand nanografting has unique capabilities forcontrolling and optimize density31 and conformation32 of patternedbiomolecules at the nanometer scale; on the other, the intrinsiccharacteristics of DDI, such as high efficiency of adsorption, siteselectivity and reversibility, allow for selective immobilization of aspecific protein of interest on the generated DNA patterns usingsemisynthetic protein–DNA conjugates.22
The suitability of our nano-immunoarrays to selectivelyrecognize the protein of interest not only in the presence ofrecombinant GFAP alone but also in the whole cell extracts,which contains several different proteins potentially prone tobind unspecifically over the NAMs and/or the surrounding SAM,was finally tested by incubating nanopatches with 100 μl of celllysate (obtained by lysing 106 glioblastoma U87 cells in 100 μlof RIPA buffer) with and without GFAP at a knownconcentration of 40 nM.
The three preliminary steps (grafting of ssDNA, hybridizationof DNA-STV conjugate, immobilization of biotinylated GFAPAb) needed to realize the nano-immunoarrays above described,along with the subsequent immobilization of GFAP in standardconditions and in the cell lysate were confirmed by topographicanalysis performed by AFM.
On average the DNA's NAM's height increased of 7.5 ±2.5 nm upon DNA-STV conjugate immobilization (n = 18,P ≪ 0.01) and 6.5 ± 2.3 nm after GFAP Abs (n = 17,P ≪ 0.01). This progressive NAMs' height increase indicates
TED P
RO
OF
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
Figure 3. Glioblastoma cells cultured into functionalized PDMS microwells and stained for GFAP. Tumoral cells isolated from a glioblastoma biopsy (left) andU251 cells (right) were seeded on a micropatterned PDMS membrane. GFAP expression was revealed by immunofluorescence (red fluorescence) and nucleiwere depicted by the blue fluorescence of the DAPI staining. Fluorescence and phase contrast images were overlaid by using Adobe Photoshop software.
Table 1 t1:1Roughness analysis of the nanopatches before and after incubation with thecell lysate. t1:2
t1:3Before incubation(Time 0)
After incubation(1 hour)
t1:4SAMt1:5cDNA + STV 0.2 ± 0.01 nm 0.2 ± 0.01 nmt1:6cDNA + GFAP Ab 0.28 ± 0.01 nm 0.28 ± 0.01 nmt1:7NAMt1:8cDNA + STV 1.3 ± 0.5 nm 1.3 ± 0.7 nmt1:9cDNA + GFAP Ab 1.9 ± 0.5 nm 1.9 ± 0.8 nm
5M. Ganau et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx
REC
that stepwise addition of the ligands (first DNA-STV, thenGFAP Ab) occurs.
The roughness of both NAMs and SAMs, as showed inTable 1, did not change significantly before and after incubationwith a GFAP-free cell lysate (P ≪ 0.01 in all comparison)meaning that the biomolecules contained in the lysate do notattach to the surface of the nanopatches and onto the surroundingbiorepellent SAM.
Conversely, as shown in Figure 4, there was a heightincrement of the nanopatches upon successful recognition of40 nM GFAP, added either to BB buffer or to cell lysate. In fact,the analysis of the topographic images showed that only in thepresence of GFAP, significant increment (P b 0.05) in the heightof the nanopatches (Table 2) could be observed.
A 3D image of the nanopatches after the recognition ofGFAP in cell lysate is performed, and also confirms the absenceof non-specifically bound proteins to the area surrounding eachNAM (Figure 5).
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
UNCO
R
The ELISA benchmark
Once GFAP detection within a cell lysate was optimized, wewanted to investigate whether it is possible to quantitativelydetermine the concentration of GFAP in the sample. For thisreason we obtained the affinity calibration curve for ournanobiosensor assay by measuring the variation in height ofthe nanopatches functionalized with the antibodies for GFAPafter the incubation with blocking buffer containing a pre-determined concentration of GFAP ranging from 200 pM to100 nM. As shown in Figure 6, the changes in height detectedfor an increase of GFAP concentration showed a sigmoidalbehavior, as expected for the binding of an antigen to anantibody. In order to quantitatively describe the recognition ofGFAP via our nanobiosensor assay, we used the Hill equation.From the fit, we obtained a dissociation constant in thenanomolar range (KD = 6.6 ± 3.8 nM) and a slightly coopera-tive effect (n N 1). However, the large error in the Hill coefficientcould mask the fact that the system also has a negative
cooperative effect meaning that once an antigen binds to anantibody (Ab), then the affinity of the Ab for other antigensdecreases. In fact, in case of our biosensor, we expect that thesteric hindrance between proximal Ab on the patches couldcontribute to a decrease of cooperativity.
In order to compare the binding affinity curve of ournanobiosensor assay with a standard technique, we also obtainedan ELISA assay for GFAPmeasurements using the same antibodiesand we analogously fitted the values of absorbance with the Hillequation. The value of n = 1.51 ± 0.26 indicated that a cooperativeeffect of the recognition event is present and a dissociation constantequal to KD = 8.30 ± 1.04 nM is perfectly in agreement with thevalue found for the nanosensor assay showing a potential of ournanodevice to be an alternative to ELISA.
Discussion
Molecular cancer diagnostics is rapidly moving beyondgenomics to proteomics. The ultimate goal of determining theprotein content of a cell is to characterize the flow of informationwithin the cells and the intercellular protein circuitry thatregulates the extracellular microenvironment.33
Due to its simplicity, low cost, easy reading, acceptability andsafety,34,35 ELISA is widely used for detecting pathogen and
F
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
Table 2 t2:1
Height variation of nanopatches topped with GFAP antibodies after a 1-hourincubation with BB solution and cell lysate with or without recombinantGFAP (40 nM). t2:2
t2:3Height variation (nm)
t2:4Incubation with BB solutiont2:5Without GFAP −0.8 ± 0.9 P = 0.13, N = 11t2:6With GFAP (40 nM) 3.9 ± 1.1 P = 0.02, N = 7* Q2t2:7Incubation with cell lysatet2:8Without GFAP 0.4 ± 0.5 P = 0.14, N = 6t2:9With GFAP (40 nM) 2.9 ± 1.5 P = 0.03, N = 4*
BB = blocking buffer. t2:10
Figure 4. (A) Height profiles of a nanopatch before (black) and after (red)incubation with a BB solution containing 40 nM of recombinant GFAP.(B) Height profile of a patch before (black) and after (red) incubation with acell lysate containing 40 nM of recombinant GFAP.
6 M. Ganau et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx
UNCO
RREC
protein markers of cancer and other various diseases, with adetection limit ranging from 2 pM to 20 nM (about 0.1 ng/mL to1 μg/mL), markedly depending on the affinity of the antibodiesused.34,36 While ELISA sensitivity and dynamic range areeffective in detecting biomarkers in big volume samples (asblood, urine, etc.), because of its large active area (about0.4 cm2) this technique can suffer from analyzing small bio-specimens, such as those obtained from stereotaxic surgery, orfrom tumor stem cells (i.e. CD133-positive glioma cells), freshlyisolated by cell sorting from neoplastic specimens. The demandfor parallel, multiplex analysis of protein biomarkers from thesesmall bio-specimens is an up and coming approach both forstudies of fundamental biology and in clinical diagnostics. Theneed for higher marker detection sensitivity while searching forkey proteins in small sample volumes is, therefore, the maindriving force behind the development of innovative methods forsingle-cell analyses.
Our approach, that combines nanotechnological principles suchas nanografting and DDI, capitalizes on the chemical robustness ofDNA oligomer strands and on the reliable assembly of DNA-labeled molecular binders via complementary hybridization.37 OurDNA immunoassays are miniaturized reaching an active area ofabout 1 μm2 per spot. Noteworthy, this strategy for realizing nano-immunoarrays is particularly advantageous in terms ofmultiplexinganalysis since the limiting factor for the number of protein to betested is only the number of different DNA strands and protein-binding antibody used. Moreover, while ELISA assays necessitateof at least two different antibodies per protein of interest, mappingdifferent epitopes, while our nanoarrays require only one. Thisresults in a smaller number of immobilization steps and widens theanalysis to biomarkers for which only one antibody is available.This approach has already been explored to build a protein array,involving up to three different DNA–protein conjugates, and usinga complex biological mixture such as standardized human serum.22
Herewith we have demonstrated the ability of antibodies to beimmobilized on the top of our nano-immunoarrays and to preserve
TED P
RO
Otheir binding capability and specificity toward the antigen bothwithand without the presence of a cell lysate. This consideration couldeasily open new routes for future clinical applications of thediagnostic approach herein described. Especially because theaccuracy achieved so far is a promising starting point to eventuallyreach the detection limits of ELISA protocols, while yielding asensitivity maximization in terms of sample quantities that arehardly accessible with conventional proteomic techniques.
The purpose of the study of single-cell biology of glialtumors, as valid for many other cancers, is to increase theknowledge base and understanding at the level of individualcells, as the average data obtained from bulk biological samplescan disguise specific patterns and even be misleading in terms ofdiagnosis and prognosis.38 This consideration is particularlyrelevant not only when the heterogeneity of cancer cells isaddressed, especially in terms of drug resistance, but also tounderstand the clinical history of gliomas and their tendency torecurrence. Namely, it is well known that individual cells differfrom each other in many aspects and the way certain cells, suchas stem cells, process input signals not only decides the fate ofthose particular cells but more importantly influences thebehavior of the whole tumor.
Envisioning the possibility of our nano-immunoassay to beexploited for the protein profiling at single cell level, we designed inparallel a system of wells for hosting from few up to one single cellper well. As a proof of concept, we demonstrate that inside ourmicrofabricated polyornithine-coated PDMSwells both commercialGFAP expressing cell lines and cells sorted from glioma tissuesgrow and proliferate while still expressing the protein of interest.
The proof of concept described in this paper and the resultsattained reflect some potentialities for an effective translationfrom the research laboratories into a proper clinical environment,however to match the requirements of standard clinical setting, adhoc AFM setup for differential height variation measurements,simplified with respect to the commercially available machines,should be conceived. This could eventually lead to a significantreduction in terms of both cost per machine and cost per exam.
Conclusion
The next generation of nanotechology-based devices holdsthe promise for improved multiplexed sensing and therefore for a
ORRECTED P
RO
OF
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452Q7
453
454
455
456
457
458
459
460
461
462
Figure 6. Binding curves obtained from the height variation (left) and absorbance obtained using ELISA (right) vs GFAP concentration. Experimental dataare shown in black and the fitting curves obtained by Hill equation in red. Values of hmin = 0.78 ± 0.28 nm, hmax = 5.07 ± 0.75 nm, n = 1.43 ± 0.82 andKD = 6.6 ± 3.8 nM were obtained for nanoassay and Amin = 0.00 ± 0.01, Amax = 1.55 ± 0.10, n = 1.51 ± 0.26 and KD = 8.30 ± 1.04 nM for ELISA assay.
Figure 5. AFM topographic 3D representation of successful recognition and immobilization of GFAP added to cell lysate (white patches) and the absence of non-specific binding of protein cell extract proteins (surrounding area).
7M. Ganau et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx
UNC
better diagnostic accuracy meant to eventually improve theoutcome of tumor treatment. The results herein representan initial step to test the sensitivity and specificity ofnano-immunoarrays designed for the detection of pivotalbiomarkers. To this regard the characterization of this DDI strategywill soon allow for the concomitant detection of several proteins ofinterest, making this technique a valid label-free alternative for thecurrent benchmark of ELISA-based proteomics.
Although further studies are warranted to better characterizethis technique of nano-immuno detection, the promising resultsobtained in a cell lysate confirm the suitability of this diagnosticmethod to effectively work in a complex environment withoutbeing influenced by the unspecific bindings of many elementsincluded in the cell's proteome.
Acknowledgments
We are grateful to Dr. Denis Scaini (University of Trieste,Trieste, Italy) for useful discussion.
References
1. Stegh AH. Toward personalized cancer nanomedicine—past, present,and future. Integr Biol 2013;5(1):48-65.
2. Heath JR, Davis ME. Nanotechnology and cancer. Annu Rev Med2008;59:251-65.
3. Rafat N, Beck GCh, Schulte J, Tuettenberg J, Vajkoczy P. Circulatingendothelial progenitor cells in malignant gliomas. J Neurosurg 2010;112(1):43-9.
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
564
8 M. Ganau et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx
ORREC
4. Ahmed AU, Auffinger B, Lesniak MS. Understanding glioma stem cells:rationale, clinical relevance and therapeutic strategies. Expert RevNeurother 2013;13(5):545-55.
5. Johnson CJ, Zhukovski N, Cass AEG, Nagy JM. Proteomics, nanotech-nology and molecular diagnostics. Proteomics 2008;8:715-30.
6. Rao J. Shedding light on tumors using nanoparticles (2008). ACS Nano2008;2(10):1984-6.
7. Ray S, Chandra H, Srivastava S. Nanotechniques in proteomics: currentstatus, promises and challenges. Biosens Bioelectron 2010;25:2389-401.
8. Sengupta S, Kulkarni A. Design principles of clinical efficacy of cancernanomedicine: a look into the basics. ACS Nano 2013;7(4):2878-82.
9. Ohgaki H, Kleihues P. Population based studies on incidence, survivalrates, and genetic alterations in astrocytic and oligodendroglial gliomas.J Neuropathol Exp Neurol 2005;64:479-89.
10. Deorah S, Lynch CF, Sibenaller ZA, Ryken TC. Trends in brain cancerincidence and survival in the United States: Surveillance, Epidemiology, andEnd Results program, 1973 to 2001. Neurosurg Focus 2006;20(4):E1.
11. Smoll NR, Schaller K, Gautschi OP. The cure fraction of glioblastomamultiforme. Neuroepidemiology 2012;39:63-9.
12. MrugalaMM. Advances and challenges in the treatment of glioblastoma: aclinician's perspective. Discov Med 2013;15(83):221-30.
13. CBTRUS (Central Brain Tumor Registry of the United States). CBTRUSStatistical Report: Primary Brain and Central Nervous System TumorsDiagnosed in the United States in 2004–2008. Hinsdale, Illinois, USA:Central Brain Tumor Registry of the United States; 2012.
14. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plusconcomitant and adjuvant temozolomide for glioblastoma. N Engl J Med2005;352(10):987-96.
15. Hegi ME, Diserens AC, Gorlia T, et al. MGMT gene silencing andbenefit from temozolomide in glioblastoma. N Engl J Med 2005;352(10):997-1003.
16. Wick W, Platten M, Meisner C, et al. Temozolomide chemotherapy aloneversus radiotherapy alone for malignant astrocytoma in the elderly: theNOA-08 randomised, phase 3 trial. Lancet Oncol 2012;13(7):707-15.
17. Kesari S. Understanding glioblastoma tumor biology: the potential toimprove current diagnosis and treatments. Semin Oncol 2011;38(Suppl4):S2-S10.
18. Chen J, McKay RM, Parada LF. Malignant glioma: lessons fromgenomics, mouse models, and stem cells. Cell 2012;149(1):36-47.
19. Niemeyer CM, Sano T, Smith CL, Cantor CR. Oligonucleotide-directedself-assembly of proteins: semisynthetic DNA–streptavidin hybridmolecules as connectors for the generation of macroscopic arrays andthe construction of supramolecular bioconjugates. Nucleic Acids Res1994;22(25):5530-9.
20. Niemeyer CM, Boldt L, Ceyhan B, Blohm D. DNA-Directedimmobilization: efficient, reversible, and site-selective surface bindingof proteins by means of covalent DNA-streptavidin conjugates. AnnBiochem 1999;268(1):54-63.
21. Niemeyer CM. The developments of semisynthetic DNA–proteinconjugates. Trends Biotechnol 2002;20(9):395-401.
UN
563
TED P
RO
OF
22. Bano F, Fruk L, Sanavio B, GlettenbergM, Casalis L, Niemeyer CM, ScolesG. Toward multiprotein nanoarrays using nanografting and DNA directedimmobilization of proteins. Nano Lett 2009;9(7):2614-8.
23. Niemeyer CM. Semisynthetic DNA–protein conjugates for biosensingand nanofabrication. Angew Chem Int Ed Engl 2010;49(7):1200-16.
24. Liu M, Amro NA, Chow CS, Gy Liu. Production of nanostructures ofDNA on surfaces. Nano Lett 2002;2(8):863-7.
25. Liu M, Amro NA, Liu Gy. Nanografting for surface physical chemistry.Annu Rev Phys Chem 2008;59:367-86.
26. Mirmomtaz E, Castronovo M, Grunwald C, Bano F, Scaini D, EnsafiAA, Scoles G, Casalis L. Quantitative study of the effect of coverage onthe hybridization efficiency of surface-bound DNA nanostructures.Nano Lett 2008;8(12):4134-9.
27. CastronovoM, Lucesoli A, Parisse P, KurnikovaA,Malhotra A, GrassiM,Grassi G, Scaggiante B, Casalis L, Scoles G. Two-dimensional enzymediffusion in laterally confined DNA monolayers. Nat Commun2011;2:297.
28. Gupta P, Loos K, Korniakov A, Spagnoli C, Cowman M, Ulman A.Facile route to ultraflat SAM-protected gold surfaces by “amphiphilesplitting”. Angew Chem Int Ed 2004;43:520-3.
29. Castronovo M, Scaini D. The atomic force microscopy as a lithographictool: nanografting of DNA nanostructures for biosensing applications.Methods Mol Biol 2011;749:209-21.
30. Pollard SM, Yoshikawa K, Clarke ID, et al. Glioma stem cell linesexpanded in adherent culture have tumor-specific phenotypes and aresuitable for chemical and genetic screens.Cell StemCell 2009;4(6):568-80.
31. Bosco A, Bano F, Parisse P, Casalis L, deSimone A, Micheletti C.Hybridization in nanostructured DNA monolayers probed by AFM:theory versus experiment. Nanoscale 2012;4:1734-41.
32. SanavioB, ScainiD,GrunwaldC,LegnameG, ScolesG,Casalis L.Orientedimmobilization of prion protein demonstrated via precise interfacialnanostructure measurements. ACS Nano 2010;4(11):6607-16.
33. Petricoin EF, Zoon KC, Kohn EC, Barrett JC, Liotta LA. Clinicalproteomics: translating benchside promise into bedside reality. Nat RevDrug Discov 2002;1(9):683-95.
34. Engvall E, Perlmann P. Enzyme-linked immunosorbent assay (ELISA).Quantitative assay of immunoglobulin G. Immunochemistry 1971;8(9):871-4.
35. Lequin RM. Enzyme immunoassay (EIA)/enzyme-linked immunosor-bent assay (ELISA). Clin Chem 2005;51(12):2415-8.
36. Jia CP, Zhong XQ, Hua B, Liu MY, Jing FX, Lou XH, Yao SH, XiangJQ, Jin QH, Zhao JL. Nano-ELISA for highly sensitive protein detection.Biosens Bioelectron 2009;24(9):2836-41.
37. Shin YS, Ahmad H, Shi Q, Kim H, Pascal TA, Fan R, Goddard III WA,Heath JR. Chemistries for patterning robust DNA microbarcodesenable multiplex assays of cytoplasm proteins from single cancer cells.ChemPhysChem 2010;11(14):3063-9.
38. Lindstrom S, Andersson-Svahn H. Miniaturization of biological assays—overview on microwell devices for single-cell analyses. Biochim BiophysActa 2011;1810:308-16.
CUNCO
RRECTED P
RO
OF
1 Graphical Abstract
2 Nanomedicine: Nanotechnology, Biology, and Medicine xxx (2014) xxx
4
5 A DNA-based nano-immunoassay for the label-free detection of6 glial fibrillary acidic protein in multicell lysates7
8
9 M. Ganau, MD PhD a,b, A. Bosco, PhD b, A. Palma, PhD c,d, S. Corvaglia a,b, P. Parisse, PhD b,e, L. Fruk, PhD f, A.P. Beltrami, MD c, D. Cesselli, MD c,d,10 L. Casalis, PhD b,⁎, G. Scoles, PhD b,c,d
1112
aUniversity of Trieste, Trieste, Italy13
bNanoInnovation Lab–Elettra Sincrotrone Trieste S.C.p.A, Trieste, Italy14
cDepartment of Medical and Biological Sciences, University of Udine, Udine, Italy15
dInstitute of Pathology, University Udine Hospital, Udine, Italy16
eINSTM, Trieste ST Unit, Trieste Italy17
fKarlsruhe Institute of Technology DFG, Centre for Functional Nanostructures, Karlsruhe, Germany
1819 We have developed a quantitative approach to eventually enable precise and multiplexing protein analysis of very small systems, down to a single or a few cells. As a proof of20 concept we realized a nano-immunoassay for glial fibrillary acidic protein (GFAP), a biomarker for glioma, the most common of primary malignant brain tumors. DNA nanoarrays21 were fabricated by means of AFM-based nanografting. Antibodies were then immobilized on the DNA nanostructures using DNA-directed immobilization of DNA–protein22 conjugates. We tested the efficiency of our assay in standard solution and in cellular lysate and we compared it with conventional ELISA assay.
2324
25
26
Nanomedicine: Nanotechnology, Biology, and Medicinexx (2014) xxx
nanomedjournal.com