Effect of the shear rate and residence time on the lysis of AC16 human cardiomyocyte cells via surface acoustic waves

Abstract

The efficient breakage of one cell or a concentration of cells for releasing intracellular material such as DNA, without damaging it, is the first step for several diagnostics or treatment processes. As the cell membrane is easy to bend but resistant to stretching, the exposure of the cell to a shear rate during a short period of time can be sufficient to damage the membrane and facilitate the extraction of DNA. However, how to induce high shear stresses on cells in small microliter volumes samples has remained an elusive problem. Surface acoustic waves operating at high frequencies can induce acoustic streaming leading to shear rates sufficient to cell lysis. Lysis induced by acoustic streaming in sessile droplets has been investigated in the past from the lysis efficiency point of view. However, the effects of the velocity field and shear rate induced by acoustic streaming on the lysis process remain unexplored. Here, we study the lysis of AC16 human cardiomyocytes in microliter droplets under the effect of the shear rate induced by acoustic streaming. It is identified that for a given shear rate, the extracted DNA is also affected by the actuation period which can be attributed to a cycling process that leads to an accumulation of damage on the cell membrane.

I. INTRODUCTION

The efficient breakage or lysis of the cell membrane for the realize of intracellular material is the starting point for molecular diagnostics of pathogens, immunoassays for point of care diagnostics, and down streaming processes such as protein purification for studying protein function and structure, cancer diagnostics, drug screening, etc. Chemical lysis is one of the most widely adopted approaches but the contamination of the sample by the added chemicals is one drawback. In this case, the breakage of the cell membrane is caused by specific chemicals that disrupt the lipid cell wall or by a hypotonic solution that causes the cells to take on water, swell, and subsequently burst. So although chemical lysis is considered the gold standard for cell lysis at the laboratory scale, there is growing interest in the development of chemical-free lysis^1,2 methods based on different approaches such as laser-based lysis for single cells,^3,4 electrical,^5–7 mechanical,⁸ and acoustics.^9,10

As the cell membrane is easy to bend but resistant to stretching with a lysis tension in the range of millinewtons per meter, the exposure of the cell to uniaxial tension of in the order of 580 mN/m (or 29 mN/m for biaxial stress) is a sufficient criterion for lysis.¹¹ However, the critical shear stress value for causing the damage to the cell membrane depends on the particular cell line as shown in Fig. 1. Then, to be able to effectively break the cell membrane, it is necessary to induce high stress rates at the micro-size range in small liquid samples which remains a challenge. Acoustic streaming has shown the potential of inducing high shear rates in the fluid and, thus, leading to the damage and breakage of the cell membrane motivating studies on the application of acoustic streaming for cell lysis in the past years. Rebound et al. (2012) demonstrated a SAW-based lysis of blood cells using a 9.5 MHz device achieving more than 98% lysis efficiency on a sessile droplet and low concentration of cells.¹² However, at high concentrations of cells, the lysis efficiency was affected, and the cause was attributed to the formation of cell clusters that disrupted the flow and prevented efficient streaming. Similarly, the lysis efficiency for smaller droplets of 5  $\mu$L was drastically reduced and a similar cause was attributed. Shilton et al.¹³ has shown that by reducing the size of the droplet, standing waves are observed, and for recovering the flow streaming, the frequency of the SAW needs to be increased. Then, as the droplet size is reduced, the flow streaming might lead to a deterioration of the lysis process. To improve the efficiency of the SAW-based lysis, 10  $\mu$m particles were added to the droplet¹⁴ which resulted in the disruption of both cell cytoplasms and nuclei. These results demonstrated the potential for isolating cell nuclei in a chemical-free fashion. Wang et al.¹⁵ further investigated the lysis induced by particles colliding with cells with an IDT working at 30 MHz in a range of 5–20  $\mu$l. Although the particle mediated lysis showed high lysis efficiency, the particles lead to contamination and add complexity to removing them from the lysate. With this contamination, it can be difficult to use the intracellular contents for gene therapy or other DNA operations. To overcome this mentioned issue, Farooq et al.¹⁶ showed the use of magnetic Ag-nanowires which can be separated from the sample for devices working at 20 and 30 MHz and a concentration of nanowires in the range of 0.5–2.5 mg/ml. Most of the mentioned studies have evaluated the performance of the concept by measuring the lysis efficiency in terms of the cell viability, and eventually, the quality of the lysate was evaluated, for example, by performing a PCR study. Lyford et al.¹⁷ analyzed the lysate using qPCR to determine whether specific genomic DNA sequences were not altered by the SAW processing. Salehi-Reyhani et al.¹⁴ determine the amount of p53 protein in the SAW lysates. Despite the research related to acoustic based cell lysis, the effect of the shear rate and the efficiency of the lysis process remain unclear. This can be attributed to the fact that most research has been focused on quantifying the lysate while limited research has been done in correlating the lysis performance to the flow characteristics. In particular, due to the thermal dissipation of the acoustic device and its potential damage to the cells, minimizing the required time and, thus, the heating effects are of great importance. Therefore, the effect of the exposure time of cells to the level of shear rate induced but the acoustic streaming on the lysis efficiency needs to be understood.

FIG. 1.

Shear stress threshold of different cell lines: RBCa,¹⁸ RBCb,¹⁹ RBCc,²⁰ Helaa,²¹ Helab,⁹ Endothelial,²² and MCF-7.²³

Studies the shown red blood cells, RBCs, either lyse or at least experience membrane damage when exposed to a prolonged high shear stress, or conversely the deformability of the RBC is improved as they are exposed to prolonged low shear stress corresponding to the physiological range. In addition, studies have shown evidence of the existence of a subhemolytic threshold for red blood cells (RBCs) at which deformation is impaired prior to the hemolysis.²⁴ Baskurt et al.²⁵ reported that cell death increased notably for cells previously exposed to a shear stress of 100 Pa suggesting a kind of memory effect on the cell membrane. Another study has observed that the shear threshold for hemolysis was 40% lower for cells that were not previously exposed to a shear.²⁵ Further, it was shown that a short-term and repeated exposure of RBCs to a supra-physiological shear stress of 100 Pa caused a progressive decrease in RBC deformability, and the accumulated RBC damage is significant after 30 duty-cycles.²⁰

In this work, we investigate the effect of the velocity field and shear rate induced by a leaking surface acoustic wave on the lysis of AC16 human cardiomyocytes in microliter droplets. The lysis efficiency is quantified by measuring the extracted DNA and correlated to the shear rate and actuation time.

II. MATERIALS AND METHODS

A. IDT device design and fabrication

The device consists of a single straight gold interdigitated transducer, IDT, deposited onto a ${128}^{°}$ Y–X lithium niobate ( ${\mathrm{LiNbO}}_3$) wafer [Figs. 2(a) and 2(b)]. The IDT consists of 15 connected and parallel finger pairs with overlapping width (aperture) of 40  $\mu$m comparable to the size of the AC16 human cardiomyocytes. A design frequency of 120 MHz was selected leading to finger spacing of 16.67  $\mu$m, corresponding to SAW wavelengths of 33  $\mu$m, and then the attenuation of the sound wave in the fluid and solid is given by the acoustic damping ${\beta }^{{-}^1}=3.1$ mm and ${\alpha }^{{-}^1}=0.39$ mm with a peak velocity location $\mathrm{\Xi }=1\mathrm{mm}$ for 120 MHz.²⁶ Then, most of the surface acoustic wave is attenuated in the base of the droplet. Given the ratio of the propagation of the sound speed in the media, the flow streaming presents an angle ${\theta }_R={22}^{°}$. Each set of interdigitated electrodes is interconnected by bus bars of 20  $\mu$m connected to square contact pads to which we apply external voltages. The IDTs are fabricated by photolithography on 4– ${128}^{°}$ Y–X lithium niobate double side polished wafers of 500 mm of thickness. The substrates are rinsed with acetone and IPA and treated with oxygen plasma for 5 min at 50% ${\mathrm{O}}_2$ (Diener Electronics) for cleaning. Then, they are heated for 5 min at 115  ${}^{°}$C for the remaining humidity removal of the substrate.

FIG. 2.

(a) Schematic of the acoustofluidic lysis device and the main design parameters. (b) Schematic of the mechanism of the acoustofluidic lysis device showing the characteristics and parameters of the induce flow streaming. (c) Bright field images of the droplet containing the AC16 cells and the relative size of the IDT. (d) Bright field images of the AC16 cells before, during, and after the applied SAW.

For photolithography, SPR-700 photoresist is spined at 4000 rpm with a ramp of 1000 rpm for 45 s, obtaining a final layer of 900 nm thickness. The sample is then heated during 1 min at 105  ${}^{°}$C for soft-baking. The exposure is done using a MLA150 (Heidelberg) and 110  $\mathrm{mJ}/{\mathrm{cm}}^2$. The post-exposure bake is done by heating the sample for 1 min at 115  ${}^{°}$C. The development is done using MF-26A for 1 min, followed by rinsing the sample with water and drying using ${\mathrm{N}}_2$. The metal deposition is done using e-beam evaporation with the AJA Sputter and Evaporator (AJA International Inc). For this, a 10 nm Ti layer followed by a 80 nm Au layer are deposited. As the final step, a lift-off for 1 min in an ultrasonic bath is performed for photoresist removal, leaving only the IDT pattern. For this step, the sample is introduced in a beaker with acetone. Then, the sample is rinsed with IPA and dried with ${\mathrm{N}}_2$.

B. Lysis device and experimental setup

The device consisting of the piezoelectric substrate with the IDT is mounted onto a 3D printed support. The bus bar of the IDT is connected by two pogo pins to a BELEKTRONIG SAW generator BSG by standard RF adaptors. The power and time of the signal sent to the IDT is controlled by a computer using a duty cycle of 100%. The device is then placed in an Eclipse Ti2-E florescence microscope with a high speed camera Photron UX-100 and a FLIR Blackfly S camera (FLIR, USA) installed at the output port.

C. Cell culture

AC16 human cardiomyocyte cell line (Merck-Millipore) was selected for this study which represents adherent cells. The AC16 human cardiomyocytes were grown in a six-well plate in DMEM/F-12, FBS 12:5% (Thermo Scientific) culture media. The Penicillin–streptomycin solution (Gibco, Life Technologies, MA, USA) is added to the culture medium to obtain a final concentration of 1%. Cells were cultured in a MCO-50AIC-PE (PHCBI) at 37  ${}^{°}$C with a ${\mathrm{CO}}_2$ level of 5%. Cells were harvested when growing to 90% confluency. For cell detachment, first, the culture medium is removed from the six-well plate by pipetting, and the well is risen four times with 3 ml of PBS. Then, 2 ml Trypsin-EDTA is added to the six-well and incubated at 37  ${}^{°}$C for 5 min. The cells are transferred to a centrifuge tube and centrifuged at 300 RPM for 5 min to pellet the cells and carefully discard the supernatant. Then, 1 ml of DMEM is added to resuspend the cells. The final cell concentration is measured using a Countess 3 (Thermo Fisher Scientific) leading to a concentration range of $1-2\times {10}^6$. The DNA concentration is measured using a Qubit 3 (ThermoFisher, USA).

D. Chemical-based cell lysis

The first part of the QIAamp DNA Mini Kit Qiagen is used as a reference case in this study, i.e., only the cell lysis stage of the kit, following the product protocol from the company. A 20  $\mu$l media containing the cells is placed into a 1.5 ml microcentrifuge tube together with 80  $\mu$l of ATL buffer and 100  $\mu$l AL buffer followed by 15 s of the pulsed vortex. Then, it is incubated for 10 min at 56  ${}^{°}$C and followed by 2 min to reach ambient temperature. The availability of intracellular material in the solution after cell lysis was quantified using a Qubit 3 using the dsDNA quantitation, high sensitivity kit.

E. SAW-based cell lysis

Before each experiment, the surface of the substrate is cleaned using ethanol 70% and isopropanol followed by blow dry with N2. A droplet of 20  $\mu$l of cell solution is deposited at a position W = 2D relative to the IDT. The SAW is applied for a period from 20 to 60 s with a duty of 100% and a power from 100 to 1000 mW. After lysis, the droplet is collected by pipetting and the DNA content of the sample is measured using Qubit 3 fluorometer. The experiments are repeated from three to five times for each condition.

F. Quantification of extracted DNA

Due to the variation of the concentrations of cells among different experiments, an individual control group was established for each experiment to compare the performance between the chemical and the SAW-based lysis. For each experimental study, a DNA control value, $DN{A}_{control}$, was defined as the amount of DNA present in the sample before a lysis step. This value will define the 0% efficiency of the process. Then, assuming that the chemical-based lysis will extract all available DNA in the sample, $DN{A}_{chem}$, this value will define the 100% efficiency of the process. Then, the lysis efficiency for an individual experiment, $DNA$, is defined as

$$DN{A}^{\ast }=\frac{DNA-DN{A}_{control}}{DN{A}_{chem}-DN{A}_{control}}.$$ (1)

III. RESULTS AND DISCUSSIONS

A. Acoustic streaming

The first step in this study is to characterize the flow inside the droplet where the cells will be lysed. Surface acoustic waves, Rayleigh waves, are generated in the piezoelectric substrate excited by an interdigitated transducer deposited on it.²⁷ These waves propagate, on the surface of the substrate and when the wave meets with a fluid above, energy is radiated into the fluid in the form of acoustic waves that induce a mean flow. The flow structure inside the droplet is three-dimensional and complex depending on the droplet size, working frequency, and fluid properties.^28,29 For simplifying the analysis, the velocity field and the shear rate at the base of the droplet are considered. The velocity field can be controlled by varying the applied power to the IDT and, thus, controlling the shear rate. The induced acoustic streaming and the shear rate inside the droplet are studied by visualizing the motion of 1  $\mu \mathrm{m}$ particles under different applied powers. The recorded images were processed using the software DAVIS 10.0 from Lavision from where the velocity field is computed. The value of the shear stress is estimated as

$$|\tau |=\sqrt{{\left(\mu \frac{\mathrm{\partial }u}{\mathrm{\partial }y}\right)}^2+{\left(\mu \frac{\mathrm{\partial }v}{\mathrm{\partial }x}\right)}^2},$$ (2)

where $\mu$ is the dynamic viscosity of the fluid and $u$ and $v$ are the x and y components of the velocity, respectively. The value of the viscosity of the culture media was measured to be $1.3\pm 0.1$ mPa s using a AR-G2 magnetic bearing rheometer (TA Instruments, UK).

Figures 3(a) and 3(b) show maximum velocity and shear rate in terms of the applied power. The trends show a clear increase with the applied power, but the deviation can be attributed to changes in the three-dimensional structures as the power is increased. The detail velocity field and the shear rate are shown in Fig. 3(d) for the case of 1000 mW. It is possible to observe the formation of a liquid jet crossing the plane with an approximated width of 500  $\mu$m. For the characterized flow, the study of the cell lysis consisted is measured in the extracted DNA for these conditions as shown in Fig. 3(c).

FIG. 3.

(a) Velocity and (b) shear share induced by the acoustic streaming. (c) Extracted DNA in terms of the applied power. The value control corresponds to the DNA measure in the sample before a lysis step, and the chemical lysis corresponds to the value of extracted DNA using chemical lysis. (d) Region of the microparticle image velocimetry study under acoustic streaming. (e) and (f) The results of the micro-PIV study.

B. Cell lysis

The velocity inside the liquid jet approaches 250 mm/s inside the droplet of about 2 mm diameter as shown in Figs. 3(e) and 3(f). Then, the lysis of the cell can be attributed to the larger vortices that bring the cells into the region of the liquid jet where the high shear rate and acceleration during a given time interval where the cell membrane is deformed by the combined action of fluid inertia and viscous forces. The elastic tension of the membrane needs to balance these forces. Previous studies that shown that the shear rate on a cell can induce transient pore formation in the cell membrane due to mechanoporation.^30,31 This induced porous can facilitate not only the uptake of small molecules,³¹ but also the release of intracellular material even if the cell membrane is not totally damaged. When liquid containing cells are approaching the region of high velocity, the flow accelerates leading to a deformation of the cell and tension build up in the membrane. In this low shear mode, cell lysis occurs if the areal strain $\mathrm{\Delta }A/A$ exceeds a critical value. The critical areal strain is estimated in the order of 1 $-$5 of rupture tensions of 1–25 mN/m for RBC cells.³² When the viscous forces are dominant, the cell lysis occurs if tangential viscous stresses $T_{visc}$ exceeds a critical tension $T_c$. Figure 4 shows that the DNA recovered increases as increasing the power which corresponds to higher velocity and shear rate. The figure also shows the value corresponding to the control tests and the one corresponding to the chemical lysis. The chemical lysis is assumed to represent the maximum ideal value of recovering DNA, and it is possible to observe that at 1000 mW, the recovered rate from the SAW-based lysis approaches the chemical lysis.

FIG. 4.

(a) Effect of the shear rate on the lysate. (b) Effect of actuation period on the lysate. (c) Effect of the combined shear rate level and actuation period on the lysis.

C. Effect of the applied time

The effect of the shear rate in terms of the extracted $DNA$ is shown in Fig. 4. No noticeable effect is observed for the threshold value of the shear rates below 2 kPa. The value is in the same order as previously reported shear rate thresholds as shown in Fig. 1. Above the threshold value, the extracted DNA gets proportional to the shear rate. The lysis mechanism can be attributed to the exposure of the cells to the shear rate that leads to the deformation of the cellular membrane. The increase of the shear rate is accompanied by an increase in the velocity field. Then, the cells are also reducing the traveling time inside the droplet, i.e., implying that the cells will cross the region of the higher shear rate most frequently. At the same time, the traveling time in this region is also reduced. These two processes might lead to a dominance of the shear rate magnitude.

To further investigate the mentioned process of the exposure of the cells to the shear rate, in the next experiment, the shear rate was kept constant while the period applied was varied as shown in Fig. 4. It is observed that the extracted DNA increases with the time that the cells are exposed to the shear rate. However, there is reaching a plateau lower than the ideal maximum one. This could be attributed to death recirculation zones that keep cells away from the high shear rate regions, but also related to the relative error of the chemical lysis in the order of 30% that defines the maximum value. The increase of the extracted lysis can be then attributed not only to an increase of time of exposure time, but also to the accumulative effect on the cells that facilitate the damage in the membrane. This effect can explain the fast increase of DNA extraction with time before reaching a limiting value. This observation would agree with the results shown in previous studies. This suggests that increasing the volume of the droplet for a similar condition will result in a reduction of extracted DNA as the exposure time is reduced as shown in Fig. 5. The reduction is significant when doubling the volume from 5 to 10  $\mu$l, but when the volume is doubled again the reduction is marginal. This can be attributed to the increase in the death regions inside the droplet that affect the overall process efficiency. Figures 3(e) and 3(f) confirm this assumption where it is possible to see that the shear stress is similar for the cases.

FIG. 5.

Effect of the droplet size on the lysate.

IV. CONCLUSIONS

In this work, the effect of shear rate induced by a leaking surface acoustic wave on the lysis of AC16 human cardiomyocytes was studied. It has been shown that the extracted DNS depends on the exposure time of the cells to a shear rate above a threshold. This can be attributed to the memory effects of the cells that facilitate the damage of the cell’s membrane. Then, to enhance the lysis process, both the shear rate and the exposure time need to be maximized.

SUPPLEMENTARY MATERIAL

See the supplementary material for further details of the device fabrication and facilities.

AUTHOR DECLARATIONS

Conflict of Interest

The authors have no conflicts to disclose.

Author Contributions

G. Almanza: Conceptualization (equal); Data curation (lead); Formal analysis (equal); Investigation (lead); Methodology (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). R. M. Trujillo: Conceptualization (equal); Investigation (supporting); Methodology (equal); Writing – review & editing (equal). D. Sanchez-Saldaña: Data curation (supporting); Investigation (supporting); Writing – review & editing (equal). Ø. Rosand: Investigation (supporting); Methodology (supporting); Writing – review & editing (equal). M. Høydal: Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Supervision (equal); Writing – review & editing (equal). M. Fernandino: Investigation (equal); Methodology (equal); Resources (equal); Supervision (equal); Writing – review & editing (equal). C. A. Dorao: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Supervision (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal).

DATA AVAILABILITY

The data that support the findings of this study are available from the corresponding author upon reasonable request.