Surface acoustic wave (SAW)-driven device for dynamic cell cultures

May 23, 2018 - In the last decades new types of cell cultures have been introduced to provide better cell survival and development, and ...
0 downloads 0 Views 1MB Size
Subscriber access provided by University of Massachusetts Amherst Libraries

Article

Surface acoustic wave (SAW)-driven device for dynamic cell cultures Gina Greco, Matteo Agostini, Ilaria Tonazzini, Damiano Sallemi, Stefano Barone, and Marco Cecchini Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b00972 • Publication Date (Web): 23 May 2018 Downloaded from http://pubs.acs.org on May 24, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Surface acoustic wave (SAW)-driven device for dynamic cell cultures Gina Greco1, Matteo Agostini1,2, Ilaria Tonazzini1, Damiano Sallemi1, Stefano Barone3 and Marco Cecchini1, * 1

NEST, Istituto Nanoscienze-CNR and Scuola Normale Superiore, Piazza San Silvestro 12, 56127 Pisa, Italy 2 Center for Nanotechnology Innovation@NEST, Istituto Italiano di Tecnologia, Piazza San Silvestro 12, 56127 Pisa, Italy; 3 Centro Procreazione Assistita-Ospedale Versilia-USL Toscana Nordovest, 55043 Viareggio, Italy * Corresponding author: [email protected] ABSTRACT In the last decades new types of cell cultures have been introduced to provide better cell survival and development, and micro/nano-environmental physico-chemical conditions aimed to mimic those present in vivo. However, despite the efforts made, the systems available to date are often difficult to replicate and use. Here, an easy-to-use surface acoustic wave (SAW)-based platform is presented for realizing dynamic cell cultures that is compatible with standard optical microscopes, incubators and cell culture dishes. The SAW chip is coupled to a standard Petri-dish via a polydimethylsiloxane (PDMS) disc and consists of a lithium niobate (LN) substrate on which gold interdigital transducers (IDTs) are patterned to generate the SAWs and induce acoustic streaming in the dish. Excitation of SAWs is verified and characterized by laser Doppler vibrometry and the fluid dynamics is studied by micro particle image velocimetry (μPIV). Heating is measured by an infrared (IR) thermal camera. We finally test this device with the U-937 monocyte cell line for viability/proliferation and cell morphological analysis. Data demonstrate that it is possible to induce significant fluid recirculation within the Petri dish while maintaining negligible heating. Remarkably, cell proliferation in this condition results enhanced by the (36±12) % with respect to standard static cultures. Finally, we show that cell death is not increased and cell morphology is not altered in presence of SAWs. This device is the first demonstration that SAW-induced streaming can mechanically improve cell proliferation, and further supports the great versatility and biocompatibility of the SAW technology for cell manipulation. INTRODUCTION Standard cell culture systems are indispensable for the biology research community, offering versatile, fast and relatively inexpensive tools for a plethora of in vitro studies1. In order to obtain better cell survival and development, new types of cell cultures have been introduced in the last decades based on environmental physico-chemical conditions more similar to those present in vivo2–5. ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 18

Advanced nanomaterials such as nano-textured films, nano-tubes, and nano-fibers were engineered to provide surfaces with nanotopographies recalling the features of the extracellular matrix (ECM). These enriched cell-culture substrates could enhance tissue regeneration 6, regulate stem cell fate 7 and were useful as in vitro models for the study of human apparatuses (the so-called organ-on-achip8), e.g.. the gastrointestinal tract 9. A large number of studies also demonstrated the advantages of 3D cultures over the standard 2D ones, which poorly represent the anatomy or physiology of a tissue 10; a further improvement was achieved by adding co-cultures in order to investigate multicellular interactions in vitro 11. Microfluidic spatial and temporal chemical-gradient generators also played an important role in many biological assays such as in the analysis of wound healing, inflammation and cancer metastasis 12. More in general, they are of interest for finely tailoring the chemical environment of the culture 13. This technology is of commercial interest and several companies are in the market with biochips specifically designed to expose cells to chemical gradients (i.e., Hypoxygen 14, Gradientech 15, Ibidi 16, to name but a few). Microfluidics can also be used to add a fluid flow to facilitate nutrient supply and waste removal. Interestingly, dynamic cultures, i.e. those where the medium is in motion and shear stress (SS) is applied to cells, could enhance cell differentiation 17 and metabolism 18 or alter cell functions 19,20. In more detail, the Datta group showed that osteoblastic differentiation in a 3D dynamic culture with in vitro generated ECM can be enhanced by fluid SS, and Esch group proved that a fluid flow can elevate the metabolic activity of a dynamic 3D co-culture of human primary liver cells. Pisano et al. 13 exploited a flow chamber which recapitulated the lymphatic capillary microenvironment to demonstrate that luminal flow upregulates the tumor cell transmigration rate whereas transmural flow further increases intravasation. Rashidi 14 showed that SS can modulate hepatocyte-like cell function, stressing the importance of adding mechanical stimuli to improve somatic cell-phenotype in vitro. Also embryo development could benefit from SS or micro-vibrations application 21, suggesting their use for improving in vitro fertilization protocols 21. Very elaborate systems were finally explored, like the four-compartment 3D dynamic perfusion device introduced by Miki et al. 22 that could promote hepatic differentiation of human embryonic stem cells. It is clear from this short review that dynamic cultures are desirable. However, despite the efforts made, the systems available to date are often difficult to replicate and use. The widespread use of microfluidics in this context is indeed limited by the difficulty to adapt standard tissue culture protocols to these miniaturized biochips and because they are still typically based on bulky external ancillary equipment for pumping liquids (e.g., air pressurized lines, peristaltic or syringe pumps). There is therefore a need of actuating fluids in a controlled and easy way, possibly maintaining the cell-culture protocols as similar as the standard ones. One appealing approach for integrated fluid handling exploits the interaction between surface acoustic waves (SAWs) and liquids23–25. When a SAW impinges on a fluid, the acoustic energy diffracts into the liquid at the Rayleigh angle generating a longitudinal pressure wave. This wave induces acoustic streaming, a net fluid motion that can lead to fast mixing, even in case of low Reynolds number regime where fluid flow is essentially laminar. SAWs were utilized to mix 26–28 and actuate fluids29–32, manipulate microparticles33–35, and actuate or modulate droplets36,37 on mm2-cm2 portable battery-operated chips. Moreover, the SAW technology offers significant advantages in terms of label-free manipulation of cells and biocompatibility 24,38–41. For

ACS Paragon Plus Environment

Page 3 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

example, high-efficient cell sorting 42, high purity cell washing scaffolds 44 have been achieved with SAWs.

43

, as long as fast cell seeding in

Here, we present an easy-to-use SAW-based device for dynamic cell cultures that is compatible with standard optical microscopes, incubators and cell culture dishes. This platform comprises a holder in which the SAW chip is mounted and connected to a printed circuit board (PCB) for electrical coupling. The SAW chip is coupled to a standard Petri-dish via a polydimethylsiloxane (PDMS) disc and consists of a lithium niobate (LN) substrate on which gold interdigital transducers (IDTs) are patterned to generate the SAWs and induce acoustic streaming in the dish. Excitation of SAWs was verified and characterized by laser Doppler vibrometry. The fluid dynamics induced by acoustic streaming in the culture medium was studied by micro particle image velocimetry (μPIV) and the heating measured by an infrared (IR) thermal camera. We finally verified this device with the U-937 monocyte cell line for viability/proliferation and cell morphological analysis.

EXPERIMENTAL SECTION Device Design, Fabrication and Assembly The SAW chip (Figure 1a) fabrication started with the dicing (1.6 x 4.6 cm2) of a 128° YX LN wafer, and cleaning of the obtained LN substrates [acetone (ACE)/ isopropyl alcohol (IPA) and O2 plasma at 100 W for 5 min (FEMTO, Diener, Germany)]. LN was chosen owing to its very high piezoelectric coupling constant, which enables efficient SAW generation. The substrates were metallized in a thermal evaporator (Kurt J. Lesker, Nano 38.) with 15 nm of Ti as an adhesion layer and 100 nm of Au for IDT fabrication. The desired SAW frequency was ≈50 MHz, thus 20-μm-wide fingers (= λ/4, where λ is the SAW wavelength) were fabricated. The IDTs have 5 mm acoustic aperture and 19 finger pairs, in order to have 50 Ω resistance for optimal electrical matching. Negative UV resist (AR-N-4340, Allresist GmbH) was spun at 6000 rpm for 1 min and baked at 90 °C for 1 min on the substrates. Subsequently, the chips were exposed to a dose of 71 mJ/cm2 in a UV lithography machine (MJB4 SUSS Microtech). They were then pre-baked at 85°C for 3.5 min, developed in AR-300-475 (Allresist GmbH) for 1 min and stopped in deionized (DI) water. Next, Au was etched for 3 min with I2:KI:H2O (1 : 4 : 40) diluted 1:10 in water and stopped in DI water, and subsequently post-baked at 90°C for 1 min. The remaining Ti metallization was finally removed by HF:H2O 1:30 solution for 70 s, and stopped in DI water. The resist was then stripped by ACE, IPA and O2 plasma cleaning (100 W, 5 min). A disk of polydimethylsiloxane (PDMS), SYLGARD® 184, of controlled thickness was used as an acoustic matching layer. It was fabricated by mixing the precursor PDMS polymer with the curing agent in a 5:1 ratio. After degassing, it was spun on a hydrophobized silicon wafer (Silanization Solution 1- Sigma Aldrich- for 20 min, n-hexane for 20 min and 1-octanol for 10 min) at 200 rpm, then baked at 80°C for 2.5 h in a convection oven. By bonding two layers of PDMS (O2 plasma activation at 25 W for 1 min and baking after contact at 60°C for 1 h), a 700-μm thick layer was obtained. The PDMS matching disk was cut with a 5 mm diameter puncher, visually aligned in the center of the chip and put in contact with the SAW chip. A plastic holder was manufactured for mounting the SAW chip and culture dish (Figure 1b). The cell culture was performed in a PDMS well (7 mm high, internal and external diameters of 1 cm and 2.5 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 18

cm, respectively) bonded on the bottom of a standard culture dish (Falcon, #353001). The bonding was realized by prior activating the PDMS surface with O2 plasma (60 W, 1 min), followed by functionalization with (3-aminopropyl)triethoxysilane (APTES, Sigma-Aldrich), 1% in DI water for 20 min. The culture dish surface was activated with O2 plasma (60 W, 1 min), then the functionalized PDMS was put in contact with the activated culture dish surface and the bonding was completed after 90 min in a convection oven at 60°C. SAW Excitation/Detection All the fabricated chips were characterized electrically by a VNA (ENA Series Network Analyzer, E5071C, Agilent Technologies.) and mechanically by an ultra-high-frequency laser Doppler vibrometer (LDV, UHF-120, Polytec, Germany). Only one IDT was activated to generate the SAWs during the experiments. Fluid Dynamics and Thermal Characterization SAWs with 750 pm amplitude were generated at a frequency of 48.8 MHz. A droplet of glycerol (volume of 3.5 μl, 5 mm radius) was used as a matching layer between the chip and the cell culture dish. We injected Milli-Q water containing 500-nm latex beads (L3280, Sigma-Aldrich) at a 8 concentration of 7.6 x 10 particles per mL in the culture dish for fluid flow visualization. We acquired three 60 fps videos (one for each chosen region of interest - ROI) by using a brightfield inverted microscope (Eclipse TI, Nikon, Japan, Tokyo) equipped with a 4X objective and a complementary metal oxide semiconductor (CMOS) camera (A602-f, Basler, Germany). The contrast of each frame was normalized, the time-averaged image subtracted to remove static objects, and all the frames superimposed, as in Ref. 45. In order to quantify the fluid velocity fields, data were analyzed with a micro particle image velocimetry (μPIV) code (Prana PIV, Virginia Polytechnic Institute and State University, Blacksburg, Virginia, USA) using MATLAB©; the resulting velocity fields  refers to the plane of focus. Therefore, the velocity module || was calculated by taking the refraction angle of the acoustic wave inside the fluid (θ) into account: || = . The angle (θ ≈ 22°) was calculated by using   simple refraction laws. For the experiments with cells, PDMS was used instead of glycerol. Glycerol is indeed very hygroscopic, and for this reason not suitable to be used in a standard, humified cell incubator. Given the short path that the acoustic waves travel through the matching layer (700 μm), the results here presented are not significantly affected by the matching layer used 46–48. In our specific case of Glycerol and PDMS, the wave amplitudes after travelling through matching layers differ of only the 6%. For this reason, glycerol and PDMS can be considered as equivalent for an estimation of the SS applied by SAW-induced streaming on cells. SAWs with amplitudes of 1.1 nm and 1.7 nm were excited during the experiments. In order to obtain the equivalent μPIV for these cases the following scaling laws of SAW-induced streaming49,50 were used:  ∝  ∝   , where  is the acoustic streaming velocity,  is the SAW intensity and  is the SAW amplitude. The SS applied by a fluid to a cell in suspension, assumed it to be a sphere of radius r ≈ 10 μm, is calculated as 21: ACS Paragon Plus Environment

Page 5 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

3 μ v 2  where μ is the dynamic viscosity of water at 37°C and v is the relative velocity.  =

An infrared (IR) camera (FLIR A655sc with macroscopic lens) was used to measure and monitor the temperature of the culture medium inside the PDMS well upon SAW generation. Cell Culture, Biological Assays and Statistical Analysis Human monocytic tumor cell line U-937 (ATCC® CRL-1593.2) was cultured in suspension in RPMI-1640 medium supplemented with 2 mM L-glutamine, 100 IU/ml penicillin, 100 µg/ml streptomycin and 10% fetal bovine serum (FBS), in standard conditions at 37°C in humidified atmosphere (95% humidity, 5% CO2). U-937 cells were seeded at a density of 3x104 cells/chip (area 0.8 cm2, in a total volume of 220 μl) and cultured for 48 h, under different conditions. Cell proliferation was assayed by the 2-(2-methoxy-4nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium-monosodium salt (WST-8) assay, according to instructions (Sigma, #96992). After 48 h from seeding, U-937 cells were incubated in a 10% WST-8 solution (in medium) for 2 h; afterwards, the supernatant was carefully aspirated, transferred to a 96-well plate, and the absorbance of each well was observed by a plate reader (GloMax multiplate reader, Promega) at a wavelength of 450 nm. The absorbance of formazan produced is directly proportional to the number of living cells; cell proliferation was reported in % with respect to control samples. After the assay, cells were collected in a tube, fixed for 10 min in 4% paraformaldehyde in phosphate buffered solution (PBS) at room temperature and processed as previously reported51 with few modifications52. Cells were stained with phalloidin-Alexa Fluor 647 (Lifetech A22287, 1:40; to stain actin fibers) in GDB buffer (0.2% BSA, 0.8 M NaCl, 0.5% Triton X-100, 30 × 10−3 M phosphate buffer, pH 7.4) for 2 h, then washed in PBS and mounted using mounting medium with 4',6-diamidino-2-phenylindole (DAPI) to stain nuclei (Fluorashield, Sigma). Fluorescence images were acquired using a laser scanning confocal microscope TCS SP2 (Leica Microsystems, Germany) with a 63x oil objective by using UV (405 nm) and argon (633 nm) lasers. Each reported confocal image was obtained from a z-series (stack-depth around 10 µm; steps = 1 µm). The resulting z-stack was processed by ImageJ software (NIH, USA) into a single image using the z-project and Max intensity options. Cell viability was also quantified by means of Trypan Blue 0.4% solution (Thermofisher, USA) exclusion test, performed by counting the number of viable (transparent) and non-viable (blue) cells using a Countess Automated Cell Counter (Thermofisher, USA). The number of dead/dying cells was reported as % over the total cell number. Moreover, cell death was also confirmed by the staining with the molecular probe Propidium Iodide (PI) as necrotic marker that measure the plasma membrane integrity 53. Cells were incubated with PI (8 μg/ml; P4864, Sigma Aldrich) and the nucleic acid stain Hoechst 33342 (5 μg/ml; Themrofisher, H3570). Samples were then imaged by bright-field/fluorescence microscopy Leica DMI 4000 B (Leica Microsystems, Wetzlar, Germany), within 15 min of treatment. Pictures (20x) were taken in 3

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 18

different and arbitrary sites for each sample, in transmission and fluorescence, and showed results consistent with the Trypan blue exclusion test. Data were reported as the average of the means calculated from 3 independent experiments, and standard error of the mean (SEM). Data were statistically analyzed by using the commercial software GraphPad Prism (San Diego, CA, USA). Student t-test (unpaired) analysis was used to compare samples. Statistical significance refers to results where P < 0.05 was obtained.

RESULTS AND DISCUSSION SAW chip and holder design The SAW platform is composed by a microfabricated SAW-exciting chip, a chip/Petri holder and PCBs for electrical connections (Figure 1b). The plastic holder, the PCB pogoes and the two clamps maintain the SAW-chip and the culture dish in a fixed position, ensuring the correct contact for optimal acoustic matching. The working principle can be outlined as follows (Figure 1c): the 48.8-MHz SAW that is generated by the IDT travels down to the PDMS matching-disk and scatters through it into the polystyrene (PS) Petri dish; this scattered wave becomes a longitudinal pressure wave in the culture medium after having crossed the culture dish bottom, giving rise to acoustic streaming. Given the acoustic attenuation coefficient of PDMS29,54 and PS55, the wave that reaches the liquid has an amplitude that is the 34% of the amplitude of the input wave. The net fluid motion inside the Petri dish is expected to lead to more efficient mixing and recirculation, and to beneficial effects for the cell cultures. The addition of the PDMS well in the culture dish (Figure 1b) allowed the reduction of the liquid volume for the cell culture and the confinement of the cells in the area overlying the matching disk. The liquid volume in the PDMS well was 220 μL. The whole SAW platform is compact (it covers an area of about 10 cm2) and can be easily placed inside standard cell incubators. Moreover, it was designed to be compatible with standard 35mm-diameter cell-culture dishes and standard inverted microscopes. Indeed, the optical window in the holder (Figure 1b) is compatible with optical imaging of the cells without unmounting the culture dish from the plastic holder.

ACS Paragon Plus Environment

Page 7 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 1. SAW platform schemes and working principle. (a) SAW chip. The IDTs generate SAWs that travel down to the center of the chip where a PDMS matching disk is placed. (b) Exploded scheme of the SAW platform. (c) Schematics of the working principle. The SAW scatters through the PDMS matching disk (dark blue rectangle) into the culture medium (light blue rectangle) generating a longitudinal pressure wave that induces acoustic streaming.

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 18

Electrical and Mechanical Characterization Figure 2 shows the electro-mechanical characterization of a representative device. The IDTs were electrically tested by measuring the RF power reflection spectra by a VNA. An LDV was used to map the SAW-amplitude spatial distribution and to measure the SAW-amplitude vs RF input power characteristics. Reflection spectra were quantitatively consistent among the different tested devices, showing a ≈ 10 dB dip centered at ≈ 48.8 MHz, corresponding to the expected SAW generation frequency (inset of Figure 2b). A representative 2D amplitude map of the area in front of the IDT is reported in Figure 2a for an excitation power of 21.5 dBm. In general, devices showed good SAW homogeneity in the scanned area, where the standard deviation (SD) of the SAW-amplitude resulted ≤ 15% of the mean value. Finally, the power law between the SAW-amplitude and RF input power is verified in Figure 2b. Specifically, the fit resulted in a 1.12-exponent (R2 ≥ 0.998), as expected.

Figure 2. Electro-mechanical characterization of the IDTs. (a) 2D map scan of the SAW amplitude performed with an LDV. The IDT was fed with 21.5 dBm at 48.8 MHz. (b) SAW mean amplitude vs. RF power; the power law fit resulted in an exponential (R2 ≥ 0.998). In the inset a typical power reflection spectrum of a fabricated IDT, measured with a VNA. SAW amplitude and duty cycle choice We chose to perform the experiments at two SAW amplitudes: low- and high- amplitude SAWs. We injected a solution of Milli-Q water containing 500-nm latex beads to visualize the fluid flow with an optical inverted microscope. We named SAW1, which corresponded to 1.1 nm amplitude, the minimum SAW amplitude at which streaming was detectable by visual observation. SAW2, found to have an amplitude of 1.7 nm, was chosen as the maximum SAW amplitude not leading to chip damage. The duty cycle D = 2.5 % (500 ms SAW every 20 s) was then chosen such that in presence of

ACS Paragon Plus Environment

Page 9 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

SAW2 (and consequently SAW1) the heating of the culture medium was below 0.5°C (see Thermal Characterization section for details and Video S1 in Supporting Information). Thermal Characterization and RF-excitation selection. Owing to mechanical damping and Joule effect56, it is known that the SAW can significantly increase the temperature of the substrate and the material inside which it scatters47,57–60. Therefore, we monitored the temperature of the culture medium upon SAW excitation, at different RF input-power and duty-cycle, with the aim to find a range of excitation parameters for which the temperature did not significantly elevate. To this end, the culture medium temperature was measured by an IR camera for several minutes after having switched the SAW on (Supporting Information, Video S2 ). As a result, the maximum RF power that we could apply to the IDT without increasing the timeaveraged temperature of the culture medium more than 0.5 °C was 26.7 dBm (corresponding to a SAW-amplitude of 1.7 nm), with 500 ms pulse and 20 s period. The plot of the heating over time in this case is shown in Figure 3. This excitation signal was thus chosen as the maximal perturbation to be applied for biological experiments.

Figure 3. Heating in the presence of high-amplitude SAW-induced streaming (RF input power = 26.7 dBm, with 500 ms pulse and 20 s period). The black line represents the temperature change of the culture medium vs. time (SAW activation at t = 0). Error bars are shown in grey. The red line represents an exponential fit (R2 = 0.82). The inset shows an image acquired with the IR camera at t= 15 min. The black dashed circle marks the perimeter of the cell culture dish; the white dashed circle is the region where the culture medium was confined by the PDMS well. Color bar ranges from 22.5 °C to 35.0 °C. Moreover, in order to test the SAW effect also for a very mild excitation condition, we determined the minimum RF-excitation power that could induce acoustic streaming in the Petri dish (named SAW1). To this end, we injected a solution of Milli-Q water containing 500-nm latex beads in the PDMS well and visualized the fluid flow with an optical inverted microscope upon SAW activation. We visually found that the minimum SAW amplitude for recirculation activation was 1.1 nm. The duty cycle and period were maintained the same as in case of SAW2 (500ms SAW pulse, 20s period). In this way

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 18

indeed, the culture medium temperature increase was negligible (Supporting Information, Video S2), and only one parameter of the excitation pattern (i.e. the SAW amplitude) was different between SAW1 and SAW2.

Fluid Dynamics Characterization A μPIV analysis was performed to map the fluid velocity field induced by acoustic streaming; these data were obtained for a SAW-amplitude of 750 pm and rescaled for the SAW amplitudes exploited during the experiments with cells, as described in section Fluid Dynamics and Thermal Characterization. We initially qualitatively evaluated the fluid flow that, as schematized in the top panel of Figure 4, was dominated by the presence of two main vortices in the planes parallel and perpendicular to the bottom of the culture dish. They were generated from a central jetting zone propagating from the bottom of the dish towards the surface of the culture medium. Representative velocity fields for three regions overlying the matching layer are reported in Figure 4abc. Here, jetting and vortices were present, and well tracked by the μPIV. In particular, the calculated mean velocity amplitude in the plane of acquisition was 600 ± 250 μm/s. These values allowed us to calculate the velocity fields for SAW1 and SAW2 and estimate the relative upper limit of mean shear stress (SS) applied to the cells (see Materials and Methods for details). Results of this analysis are summarized in Table 1. The resulting SS, considering a cell in suspension as a sphere of radius 10 μm, was 120 ± 50 mN/m2 and 280 ± 120 mN/m2 for SAW1 and SAW2, respectively.

Figure 4 (a-c) μPIV analysis of the ROIs a, b and c, arbitrary chosen on the plane of focus of the bottom of the culture dish (the depth of correlation is 3.5 μm61). The SAW amplitude was 750pm. The colorbar represents the logarithm of the fluid velocity and the arrows are the projections of the velocity vectors on the plane parallel to the bottom of the culture dish. Velocity amplitudes ACS Paragon Plus Environment

Page 11 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

range from 4 to 8100 μm/s. Scale bar = 300 μm.

Table 1. Fluid mean velocity (V) and its standard deviation (SDv); mean shear stress applied to cells (SS) and its standard deviation (SDss). The values were calculated for SAW1 and SAW2 starting for μPIV measurements. V (μm/s) SDv (μm/s) SS (mN/m2) SDSS (mN/m2) SAW1 1140 500 120 50 SAW2 2700 1200 280 120 Cell viability tests In order to test the performance of our SAW-enhanced dynamic device on cell proliferation and vitality, we chose a cell model growing in suspension. U-937 monocyte cells were seeded on different devices and cultured for 48 h under standard static condition (control) or in presence of the described SAW excitations: “low-power” SAWs in case of SAW1 and “high-power” SAWs in case of SAW2. The results of these cell culture experiments (n = 3) are plotted in Figure 5, normalized to the control condition. Following SAW application, cell proliferation did not show statistically significant variations for SAW1, while we found enhanced proliferation in case of SAW2 (136 % ± 12 %; P < 0.05 SAW2 vs. Control, Student t-test), with respect to the control static conditions. The cells were also monitored with DAPI/propidium iodide staining, which confirmed the results of the proliferation assay and further provided insights about cell viability status. Cellular viability is in fact correlated to the evolution of the cell death level, which can be estimated as the percentage of dying and necrotic cells (Trypan blue or PI positive cells) over the total cell number. The measured cell death level was similar for SAW1, SAW2 and the control static culture (Supporting information, Figure S1), confirming that the SAW streaming had not any detrimental effect on U-937 cells. Finally, bright-field and fluorescence images of the cells stained for nuclei (blue) and actin (red) (Figure 5 a-f) were acquired in order to evaluate possible cell morphological differences among the three different culture conditions. U-937 cells cultured in presence of SAW-induced streaming did not show any marked morphological difference with respect to the control cells. Altogether, these data show that SAW dynamic devices are well suitable for cell culturing and in particular the SAW2 condition has beneficial effects on cell proliferation of U-937 monocytes.

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 18

Figure 5. On the left, results of the cell proliferation assay performed on U937 cells after 48 h from seeding in case of U-937 cells cultured without SAW (red column, control), and in presence of SAWs ( SAW1-green column- and SAW2-blue column-); P < 0.05 Control vs. SAW2, Student t-test. On the right, representative optical and fluorescence images of the cells cultured under standard control condition (a and b, respectively) and in case of SAW1 and SAW2 induced streaming (c,d and e,f, respectively).

Discussion A new SAW-based device was here presented for realizing dynamic cell cultures under highlycontrolled thermal and fluid-dynamics conditions, demonstrating that SAWs can be fully biocompatible and exploited to enhance cell cultures. Remarkably, the device design was chosen to allow an easy coupling to standard cell incubators, optical microscopes and cell culture dishes. To the best of our knowledge, only few studies were previously published investigating the effect of SAWs on cell viability. Li et al.44,62 demonstrated first of all that cell invasion in a polycaprolactone 3D scaffold can be much faster in presence of 20 MHz SAWs. They also showed a characterization of the effect of SAW power, frequency and exposure time on cell viability, but could not find any condition for which viability was increased. Conversely, exposure times greater than 40 s (@20MHz and 380 mW) or powers greater than 260 mW (@10 MHz for 10 s) turned out to negatively affect viability. The mechanism leading to this effect was not either investigated or proposed in this paper. In another publication Collins et al.63 reported that 126 MHz SAWs at 320 mW can be detrimental for cells, whereas viability was not significantly affected in case of 220 mW SAWs until 2 h of exposure. Also in this case the origin of the observed cell death was not experimentally characterized. Given the similarity of the driving powers of the two cited experiments with the one used by us, we can hypothesize important SAW-induced heating originating from continuous wave operation. Indeed, temperature is a fundamental parameter to be considered for an optimal cell culture. In order to choose the SAW amplitudes for our experiments, we determined the maximum driving power that ACS Paragon Plus Environment

Page 13 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

resulted in an acceptable heating of the culture medium. The maximum tolerable temperature increase was chosen to be less than the typical temperature oscillations in a standard incubator upon normal operation (± 1 °C), in line with the safety definition used for ultrasonic medical imaging 38. The high-amplitude SAW protocol that we implemented (SAW2 case) induced a maximum heating of the culture medium of only 0.5 ± 0.2 °C. Beyond heating, the other important aspect to be discussed is the effect of SAW-driven acoustic streaming on cells 28. This fluid recirculation in the Petri dish leads to SS applied to the cells, that is a mechanical stimulus that can cause cell damaging 38 or, like in our case, turn into a positive factor. Our data on proliferation suggest a threshold effect: while a decreased or unmodified proliferation might be consistent with the SAW1 case, statistically significant enhancement was measured for the case of SAW2. Interestingly, a study was published where very low SS (