Versatile multicharacterization platform involving

0 downloads 0 Views 2MB Size Report
Sep 13, 2016 - applications to single cell, multiplex assays of secreted proteins ..... diffraction angle); (c) typical XRD pattern coming from ERa66 sample (the red dotted .... F. Drummy, D. M. Phillips, M. O. Stone, B. L. Farmer, and R. R. Naik,.
Versatile multicharacterization platform involving tailored superhydrophobic SU-8 micropillars for the investigation of breast cancer estrogen receptor isoforms Angelo Accardo, Emmanuelle Trevisiol, Aline Cerf, Christophe Thibault, Henrik Laurell, Melissa Buscato, Françoise Lenfant, Jean-François Arnal, Coralie Fontaine, and Christophe Vieu Citation: Journal of Vacuum Science & Technology B 34, 06K201 (2016); doi: 10.1116/1.4962382 View online: http://dx.doi.org/10.1116/1.4962382 View Table of Contents: http://scitation.aip.org/content/avs/journal/jvstb/34/6?ver=pdfcov Published by the AVS: Science & Technology of Materials, Interfaces, and Processing Articles you may be interested in Simplified prototyping of perfusable polystyrene microfluidics Biomicrofluidics 8, 046501 (2014); 10.1063/1.4892035 Empirical chemosensitivity testing in a spheroid model of ovarian cancer using a microfluidics-based multiplex platform Biomicrofluidics 7, 011805 (2013); 10.1063/1.4774309 A robotics platform for automated batch fabrication of high density, microfluidics-based DNA microarrays, with applications to single cell, multiplex assays of secreted proteins Rev. Sci. Instrum. 82, 094301 (2011); 10.1063/1.3636077 Non-positional cell microarray prepared by shape-coded polymeric microboards: A new microarray format for multiplex and high throughput cell-based assays Biomicrofluidics 5, 032001 (2011); 10.1063/1.3608130 Development of microfluidic device and system for breast cancer cell fluorescence detection J. Vac. Sci. Technol. B 27, 1295 (2009); 10.1116/1.3049529

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. IP: 181.41.221.15 On: Tue, 20 Sep 2016 08:59:57

Versatile multicharacterization platform involving tailored superhydrophobic SU-8 micropillars for the investigation of breast cancer estrogen receptor isoforms Angelo Accardoa) INSERM U1048-I2MC-CHU Rangueil, BP 84225, 31432 Toulouse Cedex 4, France and LAAS-CNRS, Universit e de Toulouse, CNRS, 31400 Toulouse, France

Emmanuelle Trevisiol and Aline Cerf LAAS-CNRS, Universit e de Toulouse, CNRS, 31400 Toulouse, France

Christophe Thibault LAAS-CNRS, Universit e de Toulouse, CNRS, INSA, 31400 Toulouse, France

Henrik Laurell, Melissa Buscato, Franc¸oise Lenfant, Jean-Franc¸ois Arnal, and Coralie Fontaine INSERM U1048-I2MC-CHU Rangueil, BP 84225, 31432 Toulouse Cedex 4, France

Christophe Vieu LAAS-CNRS, Universit e de Toulouse, CNRS, INSA, 31400 Toulouse, France

(Received 23 June 2016; accepted 23 August 2016; published 13 September 2016) Here, the authors report the fabrication of lotus-leaf-like tailored SU8 micropillars and their application in the context of a multitechnique characterization protocol for the investigation of the structural properties of the two estrogen receptors (ERa66/ERa46). ER (a) expression is undoubtedly the most important biomarker in breast cancer, as it provides the index for sensitivity to endocrine treatment. Beside the well-characterized ERa66 isoform, a shorter one (ERa46) is also expressed in ERa positive breast cancers and breast cancer cell lines. The superhydrophobic supports were developed by using a two-step approach including an optical lithography process and a plasma reactive ion roughening one. Upon drying on the micropillars, the biological samples resulted in stretched fibers of different diameters which were then characterized by synchrotron xray diffraction (XRD), Raman and Fourier-transform infrared spectroscopy. The evidence of both different spectroscopic vibrational responses and XRD signatures in the two estrogen receptors suggests the presence of conformational changes between the two biomarkers. The SU8 micropillar platform therefore represents a valid tool to enhance the discrimination sensitivity of structural features of this class of biomarkers by exploiting a multitechnique in situ characterization approach. C 2016 American Vacuum Society. [http://dx.doi.org/10.1116/1.4962382] V

I. INTRODUCTION The use of microfluidic engineered devices for tuning the conformation and arrangement of biomedical compounds1–3 is widely extended in the nanobiotechnology community. From one side, the use of droplet-based digital microfluidics architectures triggers the interest of the researchers in the biological and chemical fields, due to the possibility of controlling complex processes by means of a programmable sequence of discrete steps and with volumes sensitively smaller than those used in continuous flow microfluidics.4–6 On the other hand, dedicated lab-on-a-chip devices in combination with advanced characterization techniques [e.g., Raman/FTIR spectroscopy and x-ray diffraction (XRD)] allow to extrapolate sensitive structural information on biomedical relevant subjects.7–10 In this work, we propose the fabrication of lotus-leaf-like tailored surfaces, following a biomimetic approach inspired by natural lotus leaves,11 as a tool to induce the formation of free-standing fibers coming

a)

Electronic addresses: [email protected]; [email protected]

06K201-1 J. Vac. Sci. Technol. B 34(6), Nov/Dec 2016

from two estrogen receptor (ER) isoforms, namely, ERa46 and ERa66. The estrogen receptor a (ERa) is holding a key position in the diagnosis of breast tumors in several aspects.12 ERa protein immunoreactivity in the nucleus of mammary epithelial cells is systematically evaluated and quantified during the anatomopathologic diagnosis, and 70% of breast cancers are initially described as ERa positive.12 ERa can be subdivided into six domains from A to F, and in addition to the full-length 66 kDa (ERa66), a 46 kDa ERaisoform (ERa46), lacking the N-terminal portion (domains A/B), can be expressed through either an alternative splicing13 or an internal entry site.14 ERa46 has been reported in breast cancer cell lines where it inhibits the proliferative response to estrogen mediated by ERa66 in the MCF7 breast cancer cell line.15 However, all studies conducted on breast cancer diagnosis focused on the full-length isoform ERa66, and very little information is available about the expression of ERa46 in breast tumors probably due to the absence of a reliable tool to characterize ERa46. From a structural point of view, the absence of the flexible A/B domain in ERa46 (Ref. 16) should translate in a different vibrational mode. In

2166-2746/2016/34(6)/06K201/6/$30.00

C 2016 American Vacuum Society V

06K201-1

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. IP: 181.41.221.15 On: Tue, 20 Sep 2016 08:59:57

06K201-2 Accardo et al.: Versatile multicharacterization platform

06K201-2

order to support this hypothesis, we combined the use of biomimetic superhydrophobic surfaces to Raman/FTIR spectroscopy and synchrotron x-ray microdiffraction. The Raman and FTIR Amide I/III bands9 can indeed provide sensitive information on the structural variations of the analyzed sample while, by using a synchrotron source x-ray microbeam, it is possible to detect different small- and wide-angle x-ray scattering signatures10,17 linked to the conformation of the solid residues formed upon evaporation. By exploiting the convective microfluidic flows provided by droplets in a superhydrophobic state and their highly homogeneous evaporation rate,17 it was indeed possible to obtain well aligned fiber structures which highlighted among the two estrogen receptors ERa46 and ERa66 different secondary structure conformations (in terms of a-helical and b-sheet composition) making this platform a useful tool for further characterization of this class of biomarkers.

B. Configuration of the Raman, FTIR, and synchrotron beamline setup

II. EXPERIMENT A. Fabrication of the SU8 micropillars supports

SU8 micropillars were developed by exploiting a twostep approach involving optical lithography and plasma reactive ion roughening. The microstructures were grown either on CaF2 windows (CrystranV, for Raman and FTIR measurements) or on 5  5 mm2 Si3N4 membranes (for XRD measurements). The Si3N4 membranes were fabricated exploiting the following protocol: starting with a silicon h100i wafer with a Si3N4 nitride thin film (thickness of 500 nm) on both sides, a S1813 positive tone resist (MicrochemV) was spin-coated on the wafer at 4000 rpm for 60 s and baked at 90  C for 3 min; after soft contact exposure of 15 s and development with MF-319 developer for 1 min to remove the resist from the square areas, the wafer was rinsed in water and baked for 1 min to dry it completely; plasma reactive ion etching using CHF3 [Fluoroform, standard cubic centimeters per minute (sccm) 70] and O2 (sccm 5), RF power ¼ 70 W, source power ¼ 200 W, and P ¼ 1 Pa was then applied and the resist was stripped with acetone (cold, 3 min); finally, a KOH wet etch [100 g of KOH (SigmaV) in 150 ml of H2O, 105  C] was performed on the wafer that was then rinsed in milliQ water. The SU8 circular micropillars (arranged in a hexagonal lattice with 10 lm diameter, 20 lm interpillar gap and 20 lm height) were fabricated using the following protocol: SU8–25 resist (Microchem) was spin-coated at 1500 rpm for 60 s and prebaked at 65  C for 5 min followed by 40 min at 95  C; after a hard contact exposure of 4.5 s the sample was postbaked at 65  C for 5 min followed by 10 min at 95  C; the sample was then developed in SU8 developer for 9 min and rinsed in isopropanol for 1 min; plasma reactive ion roughening (Trikon-SPTS) of the pillars head was performed using CF4 (Tetrafluoromethane, 5 sccm), O2 (15 sccm), P ¼ 9.06 Pa, RF power ¼ 50 W, and source power ¼ 100 W; finally, a Teflon deposition by plasma process was obtained using C4F8 (Octafluorocyclobutane, 75 sccm), P ¼ 3.6 Pa, source power ¼ 600 W, and RF power ¼ 1 W.

Raman measurements were performed with a Horiba Labram Raman setup, using a 632 nm laser line with a laser beam of 2  2 lm spot size, 6 s of exposure, 1% of laser intensity and 10 accumulations. FTIR characterizations were performed in transmission mode with a Bruker FTIR microscope exploiting a Globar Source in the mid-infrared range, an MCT detector, and a spot-size of 10  10 lm2. The OMNIC software was employed for the interpretation of the spectral data. Synchrotron XRD experiments were performed at the ID13 beamline of the European Synchrotron Radiation Facility in Grenoble. The wavelength of the monochromatic ˚ (E ¼ 12.99986 keV) x-ray beam was equal to 0.935735 A with a beam spot dimension of 2.3 lm (ver)  2.5 lm (hor) and a sample to detector distance of 185.047 mm calibrated with Al2O3 calibrant. Regions of interest of the sample were selected using an on axis optical Olympus microscope aligned with the focal spot of the microbeam. The expression of the scattering vector Q in the illustrated azimuthal averages is equal to Q ¼ 4p sin h/k, where h is the diffraction angle and k is the x-ray wavelength. The FIT2D software was used for data analysis of the XRD patterns.

R

R

R

C. Preparation of the estrogen receptor samples

The plasmids (pCR3.1) harboring the human ERa66 and ERa46 complementary deoxyribose nucleic acid were kindly provided by Dr. Gilles Flouriot. The empty pCR3.1 plasmid was obtained by EcoRI digestion to release the ERa66 insert followed by ligation of the vector after gel purification. For preparations of plasmid deoxyribose nucleic acid (DNA), the GenEluteTM HP Plasmid Maxiprep kit was used according to the protocol. In vitro expression of ERa66 and ERa46 was performed by using in vitro transcription and translationcoupled rabbit reticulocyte extracts according to manufacturer’s protocol (PromegaV). The samples were afterward incubated at 30  C for 90 min and water transferred by centrifuging them 5 times at 13 000 rpm for 5 min using Vivaspin 500 columns (Sartorius Stedim) equipped with 30 000 Da molecular weight cut off filters obtaining a final solution concentration of 20 mg/ml. Samples were then stored at 20  C before deposition on the micropillared supports. R

D. SEM investigation and contact angle setup

SEM imaging was performed on the samples metalcoated with 15 nm of sputtered gold by using a Hitachi S4800 setup with a 5 kV acceleration voltage. Contact angle measurements were acquired using a Digidrop system (GBX) and the imaging software IMAGE J. III. RESULTS AND DISCUSSION A. Morphological investigation and FTIR characterization of the ER samples dried on SU8 micropillars

Droplets of 5 ll in volume, containing either ERa46 or ERa66, were deposited on top of the SU8 micropillars

J. Vac. Sci. Technol. B, Vol. 34, No. 6, Nov/Dec 2016

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. IP: 181.41.221.15 On: Tue, 20 Sep 2016 08:59:57

06K201-3 Accardo et al.: Versatile multicharacterization platform

06K201-3

FIG. 1. (Color online) (a) Droplets of ERa solutions and MilliQ water sitting on a SU8 micropillared surface fabricated on top of a CaF2 window; (b) SEM micrograph of the SU8 micropillar morphology.

[Figs. 1(a) and 1(b)] reaching a contact angle of 158 . This superhydrophobic behavior was reached thanks to the combination of a well-defined topology, achieved by optical lithography and plasma roughening, with a Teflon layer, obtained by plasma deposition of C4F8, whose intrinsic contact angle is of 114 . The complete evaporation of the solution droplets was reached after 1 h. Upon drying, it was possible to observe the formation of free-standing fiber structures of different diameters suspended between the SU8 pillars (Fig. 2). We attribute the formation of such fibers to the stretching mechanism provoked by the shrinking of the contact line of the droplet while drying and displacing from one pillar to the adjacent one. The surface roughness obtained by the plasma reactive ion roughening process (Fig. S1, see the supplementary material)34 was indeed introduced in order to provide anchoring points to the estrogen receptor fibers obtained during the drying process, although it is known how the presence of nanoscale rugosity affects as well the hydrophobicity of the system and can be modeled rigorously.18 The fiber formation process of analytes diluted in a droplet suspended over micropillared surfaces is a phenomenon that has been observed in several bio-samples such as DNA (Ref. 19) and viruses.20 In this work, we exploited such approach in order to influence the aggregation of estrogen receptors in the form of dense fibers composed by enough material in order to be characterized by spectroscopic and x-ray diffraction techniques. Two of the most important properties of superhydrophobic surfaces are from one side the presence of air pockets under the droplet (which increases the air fraction leading to the suspension of the water droplet on top of the asperities) and, on the other hand, the fact that they hold a very low friction coefficient. In this way, the adhesion force between the micropillars and the aqueous solution is extremely low thus avoiding possible pinning of the droplet and allowing the formation of suspended structures. Indeed, although the formation of fibers can be ensured in the presence of large enough interpillars gaps, it is also true that, by increasing too much such distance, the droplet can easily pin the asperities. This can be explained by the analysis of the so-called Cassie–Baxter (not-pinned) and Wenzel (pinned) states and their roughness factors r and u (Refs. 21–23) that for circular pillars are equal to r ¼ 1 þ ½ðpahÞ=ða þ dÞ2  and u ¼ pa2 =4ða þ dÞ2 , where a is the

pillar diameter, h is the pillar height, and d is the gap between the two pillars. If the height tends to zero so will r tend to 1 and the Wenzel state can be easily reached. The same thing happens, for example, when the diameter of the pillars tends to be 0. In this case, u tends to be 0 and r tends to become 1, so the Wenzel state is predominant. In contrast, a Cassie state can be easily reached, for example, when the gap between the pillars tends to be 0 and is at the same time much larger than the diameter of the pillars. Superhydrophobicity is generally reached when u is small and r is great (with a height sufficiently elevated). Therefore, the morphology of the micropillars (height, gap, and diameter) plays a fundamental role to make the droplet achieve the superhydrophobic state and avoiding its pinning. Further, it has been demonstrated elsewhere24 how the peculiar convective flows induced in a superhydrophobic droplet17 can be modeled in order to predict the trajectory followed by the floating analytes (by solving the Langevin equation24), thus their preferential distribution within the droplet that contributes to the formation of the stretched fibers. Since the amide I band is sensitive to protein secondary structure, FTIR spectroscopy is frequently used to study protein misfolding and aggregation processes in vitro.25 Due to unique hydrogen bonding environments for the different secondary structure elements, shifts are observed in the frequency of the amide I band, and it is possible to detect several features of proteins’ secondary structure which allows discriminating among beta-turn (1665–1685 cm1), alpha-helix (1655 cm1), unordered (1640 cm1), and beta sheet structures (1610–1640 cm1).26 The FTIR characterization of the two estrogen receptor samples dried on top of SU8 micropillars sitting on top of CaF2 windows show how ERa46 and ERa66 seem to feature different molecular configurations as depicted in Fig. 3. In particular, ERa46 has a main peak in the Amide I band at 1665 cm1 (related to b-turn configuration) and two weaker contributions at 1649 and 1641 cm1 (related to unordered phases). ERa66, on the other hand, does not present the strong peak at 1665 cm1 but, instead, a main contribution at 1656 cm1 (related to a-helical phases). As a first hypothesis, this different secondary structure conformation could be ascribed to the absence of the flexible A/B domain in ERa46 (Ref. 16) which should translate in a different vibrational mode.

JVST B - Nanotechnology and Microelectronics: Materials, Processing, Measurement, and Phenomena

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. IP: 181.41.221.15 On: Tue, 20 Sep 2016 08:59:57

06K201-4 Accardo et al.: Versatile multicharacterization platform

06K201-4

FIG. 2. SEM micrographs of ERa fiber structures suspended between SU8 micropillars.

B. Micro-Raman characterization of the ER samples dried on SU8 micropillars

In support of the results obtained by FTIR measurements, we performed a parallel micro-Raman spectroscopy investigation (Fig. 4) as, with this characterization technique, conformational information can be obtained as well by analyzing the amide I and amide III bands.27,28 ERa46 sample evidenced a strong amide III contribution at 1250 cm1 (related to random coils and turns) and some weaker b-sheet components in the band 1229–1235 cm1 while the amide I band shows antiparallel b-sheets (1621 and 1611 cm1) together with a weaker shoulder related to unordered phase (1640 cm1). On the other hand, compared to ERa46, ERa66 shows some structural differences in both the amide I and amide III bands. In particular, in the amide III band, we could detect the copresence of a random coils/turns peak at 1250 cm1 and of an a-helical one29 at 1282 cm1. In the amide I band, the strong antiparallel b-sheet peak at 1611 cm1 disappeared while the weaker component

FIG. 3. (Color online) FTIR spectra of the ERa46 and ERa66 samples dried on superhydrophobic SU8 micropillars.

at 1621 cm1 can still be detected together with an a-helical contribution at 1655 cm1. Further, in the 1500–1300 cm1 band, the ERa66 sample highlights a strong peak centered at 1450 cm1 (while, although still present in ERa46, it is clearly less pronounced). This contribution is related to the C-H deformation band30 and can be tentatively ascribed to diversity of CH, CH2, and CH3 groups in the side chain. On the other hand, the slightly stronger contribution in the region 1300–1330 cm1 can be attributed to a Ca-H bending vibrational mode. C. Synchrotron x-ray diffraction characterization of the ER samples dried on SU8 micropillars

Finally, we characterized ERa samples dried on top of SU8 micropillars sitting on thin Si3N4 membranes using the x-ray microdiffraction ID13 beamline at the European Synchrotron Radiation Facility. In Figs. 5(a) and 5(b) we plotted, respectively, a typical XRD pattern coming from ERa46 sample and its azimuthal averaging. The sample ˚ (related, shows two evident d-spacings at 4.3 and 9.3 A respectively, to the distance between hydrogen bonded strands and to a b-sheet stacking) characteristic of a not oriented (isotropic) secondary b-sheet structure.31 On the other hand, the ERa66 sample showed a quite evident anisotropy of the XRD patterns [Fig. 5(c)] which was not shown by the patterns coming from ERa46. Indeed, it is possible to recognize a well-defined oriented quasicrystalline cross b-sheet ˚ ) contrasting the configuration (sharp peaks at 3.9 and 9.4 A unoriented powderlike b-type material obtained in the ERa46 sample. The pattern indeed resembles to a fiberlike one (the red dotted lines show the possible axes of two overlapping fiber structures) and the presence of the small peak

J. Vac. Sci. Technol. B, Vol. 34, No. 6, Nov/Dec 2016

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. IP: 181.41.221.15 On: Tue, 20 Sep 2016 08:59:57

06K201-5 Accardo et al.: Versatile multicharacterization platform

06K201-5

FIG. 4. (Color online) Micro-Raman spectra of the ERa46 and ERa66 samples dried on superhydrophobic SU8 micropillars (insets: laser beam impinging an ER fiber and optical micrograph of suspended ER fibers).

˚ [Fig. 5(d)] may be attributed to the copresence of a at 5.8 A a-helical conformation,32 which is in good accordance with the FTIR and Raman characterization results. IV. SUMMARY AND CONCLUSIONS In this work, we showed how, by exploiting the superhydrophobic properties of SU8 micropillared surfaces, it is

possible to obtain free-standing fibers coming from two biomedical relevant biomarkers, namely, ERa46 and ERa66. The ability to grow the microstructured topology on different bulk substrates (i.e., CaF2 windows and Si3N4 membranes) allowed to perform a multitechnique in situ characterization by exploiting Raman, FTIR, and XRD techniques with no overlapping background signals. The achieved fibrillary morphology showed peculiar structural

FIG. 5. (Color online) (a) Typical XRD pattern coming from ERa46 sample; (b) Azimuthal average of the XRD pattern in (a) (Q ¼ 4psin H/k, where k is the x-ray wavelength and H the diffraction angle); (c) typical XRD pattern coming from ERa66 sample (the red dotted lines indicate possible fiber axes); and (d) Azimuthal average of the XRD pattern in (c). JVST B - Nanotechnology and Microelectronics: Materials, Processing, Measurement, and Phenomena

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. IP: 181.41.221.15 On: Tue, 20 Sep 2016 08:59:57

06K201-6 Accardo et al.: Versatile multicharacterization platform

differences between the two estrogen receptors, whose presence is used as a biomarker in breast cancer prevention. In particular, the three performed characterizations seem to converge to the conclusion that, from a protein secondary structure point of view, while ERa46 features mainly b-sheet phases, ERa66 contains also some evident a-helical contributions which we attribute, as a tentative hypothesis, to the absence of the flexible A/B domain in ERa46.16 The detection of discrimination factors in the Raman, FTIR, and XRD signatures of these two proteins could pave the basis for the development of a novel spectroscopic tool aiming at the structural identification of biomarkers linked to breast cancer disease. Furthermore, this approach could be an interesting alternative to more conventional methods employed to detect ERa46 and ERa66. Indeed, since no antibody presently available is able to recognize specifically ERa46 in immunohistochemistry (as all epitopes present in ERa66 are also theoretically present in ERa46), the only technique allowing the detection of ERa46 is Western blotting, a semiquantitative, time consuming, and labor-intensive technique, which is not compatible with clinical routine to perform diagnosis. Finally, considering the current absence of a complete crystalline three-dimensional structure of the full ERa protein,33 in further works we will improve the synthesis of the estrogen receptors in order to reach a higher crystalline phase of the dried samples in the context of sittingdrop crystallization experiments involving lotus-leaf-like engineered supports. ACKNOWLEDGMENTS The authors acknowledge the support of Christian Riekel and Martin Rosenthal at the ID13 beamline of the European Synchrotron Radiation Facility (ESRF) in Grenoble during the XRD measurements. The authors acknowledge the support of the team of Azzedine Bousseksou of the Laboratoire de Chimie de Coordination in Toulouse, and, in particular, Corinne Routaboul for the help provided during Raman measurements. The work at the INSERM unit 1048 was supported by INSERM, Universite Toulouse III, Faculte de Medecine Toulouse-Rangueil, Agence Nationale de la Recherche ANR No. 14-CE12-0021-01, Conseil Regional Midi-Pyrenees (R-BIO), and the Ligue Regionale contre le Cancer of the Comite de Haute-Garonne. Finally, the authors acknowledge support from the French national nanofabrication RENATECH network.

06K201-6 1

H. Cayron, B. Berteloite, C. Vieu, V. Paveau, J.-C. Cau, and A. Cerf, Microelectron. Eng. 135, 1 (2015). A. Lagraulet, J. Foncy, B. Berteloite, A. Esteve, M.-C. Blatche, L. Malaquin, and C. Vieu, Nat. Methods 12, i (2015). 3 E. Miele, A. Accardo, A. Falqui, M. Marini, A. Giugni, M. Leoncini, F. De Angelis, R. Krahne, and E. Di Fabrizio, Small 11, 134 (2015). 4 J. Berthier, Microdrops and Digital Microfluidics (William Andrew, New York, 2008). 5 J. Heikenfeld and M. Dhindsa, J. Adhes. Sci. Technol. 22, 319 (2008). 6 V. Hessel, H. Loewe, A. Mueller, and G. Kolb, Chemical Micro Process Engineering (Wiley VCH GmbH, Weinheim, 2005). 7 I. H. Chou, M. Benford, H. T. Beier, and G. L. Cote, Nano Lett. 8, 1729 (2008). 8 A. Keller, M. Fritzsche, Y. P. Yu, Q. Liu, Y. M. Li, M. Dong, and F. Besenbacher, ACS Nano 5, 2770 (2011). 9 A. Accardo, V. Shalabaeva, M. Cotte, M. Burghammer, R. Krahne, C. Riekel, and S. Dante, Langmuir 30, 3191 (2014). 10 A. Accardo, V. Shalabaeva, E. Di Cola, M. Burghammer, R. Krahne, C. Riekel, and S. Dante, ACS Appl. Mater. Interfaces 7, 20875 (2015). 11 C. Neinhuis and W. Barthlott, Ann. Bot. 79, 667 (1997). 12 S. R. Johnston, Clin. Cancer Res. 16, 1979 (2010). 13 G. Flouriot, H. Brand, S. Denger, R. Metivier, M. Kos, G. Reid, V. Sonntag-Buck, and F. Gannon, Embo J. 19, 4688 (2000). 14 P. Barraille, P. Chinestra, F. Bayard, and J. C. Faye, Biochem. Biophys. Res. Commun. 257, 84 (1999). 15 G. Penot et al., Endocrinology 146, 5474 (2005). 16 G. Flouriot, H. Brand, B. Seraphin, and F. Gannon, J. Biol. Chem. 277, 26244 (2002). 17 A. Accardo, E. Di Fabrizio, T. Limongi, G. Marinaro, and C. Riekel, J. Synchrotron Radiat. 21, 643 (2014). 18 F. Gentile et al., Microelectron. Eng. 88, 2537 (2011). 19 F. Gentile et al., Nano Lett. 12, 6453 (2012). 20 G. Marinaro, M. Burghammer, L. Costa, T. Dane, F. De Angelis, E. Di Fabrizio, and C. Riekel, ACS Appl. Mater. Interfaces 7, 12373 (2015). 21 R. N. Wenzel, Ind. Eng. Chem. 28, 988 (1936). 22 A. B. D. Cassie and S. Baxter, Trans. Faraday Soc. 40, 546 (1944). 23 A. Accardo, F. Gentile, F. Mecarini, F. De Angelis, M. Burghammer, E. Di Fabrizio, and C. Riekel, Langmuir 26, 15057 (2010). 24 F. Gentile, L. Ferrara, M. Villani, M. Bettelli, S. Iannotta, A. Zappettini, M. Cesarelli, E. Di Fabrizio, and N. Coppede`, Sci. Rep. 6, 18992 (2016). 25 L. M. Miller, M. W. Bourassa, and R. J. Smith, Biochim. Biophys. Acta 1828, 2339 (2013). 26 D. Naumann, Infrared Spectroscopy in Microbiology, Encyclopedia of Analytical Chemistry, edited by R. A. Meyers (John Wiley & Sons Ltd., New York, 2000). 27 J. T. Pelton and L. R. McLean, Anal. Biochem. 277, 167 (2000). 28 J. Laane, Frontiers of Molecular Spectroscopy, 1st ed. (Elsevier Science, Amsterdam, 2009). 29 Z. Chi, X. G. Chen, J. S. Holtz, and S. A. Asher, Biochemistry 37, 2854 (1998). 30 A. Rygula, K. Majzner, K. M. Marzec, A. Kaczor, M. Pilarczyka, and M. Baranska, J. Raman Spectrosc. 44, 1061 (2013). 31 R. N. Rambaran and L. C. Serpell, Prion 2, 112 (2008). 32 L. F. Drummy, D. M. Phillips, M. O. Stone, B. L. Farmer, and R. R. Naik, Biomacromolecules 6, 3328 (2005). 33 R. Kumar, M. N. Zakharov, S. H. Khan, R. Miki, H. Jang, G. Toraldo, R. Singh, S. Bhasin, and R. Jasuja, J. Amino Acids 2011, 812540. 34 See supplementary material at http://dx.doi.org/10.1116/1.4962382 for further details about the surface roughness obtained by the plasma reactive ion roughening process. 2

J. Vac. Sci. Technol. B, Vol. 34, No. 6, Nov/Dec 2016

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. IP: 181.41.221.15 On: Tue, 20 Sep 2016 08:59:57