Label-free analysis of the characteristics of a single cell trapped by acoustic tweezers

Single-cell analysis is essential to understand the physical and functional characteristics of cells. The basic knowledge of these characteristics is important to elucidate the unique features of various cells and causative factors of diseases and determine the most effective treatments for diseases. Recently, acoustic tweezers based on tightly focused ultrasound microbeam have attracted considerable attention owing to their capability to grab and separate a single cell from a heterogeneous cell sample and to measure its physical cell properties. However, the measurement cannot be performed while trapping the target cell, because the current method uses long ultrasound pulses for grabbing one cell and short pulses for interrogating the target cell. In this paper, we demonstrate that short ultrasound pulses can be used for generating acoustic trapping force comparable to that with long pulses by adjusting the pulse repetition frequency (PRF). This enables us to capture a single cell and measure its physical properties simultaneously. Furthermore, it is shown that short ultrasound pulses at a PRF of 167 kHz can trap and separate either one red blood cell or one prostate cancer cell and facilitate the simultaneous measurement of its integrated backscattering coefficient related to the cell size and mechanical properties.

(PRF) of 1 MHz. These pulses are amplified to 50 V p-p in the power laterally diffused metal oxide semiconductor (LDMOS) transistor [3]. Moreover, a diode-based expander and limiter were used for the protection of the transmitter and receiver circuitry [4], [5]. The expander blocks the echo signals from the ultrasound transducer to the transmitter, and the limiter blocks the high voltage generated in the transmitter to the receiver. The primary part of the receiver is a low noise amplifier (LNA) that enhances a signal-to-noise ratio (SNR).
Performance evaluation of tightly focused high-frequency ultrasound transducer and front-end system developed for this study. To examine the performance of an ultrasound transducer, a pulseecho response and an ultrasound image of a 2.5 µm diameter tungsten wire were acquired (Fig. S1). For this, a flat quartz target or a tungsten wire was immersed into a degassed deionized water container. The position of the transducer integrated with IMN was adjusted for locating the target at the focal length of the transducer, which was controlled by a 3D linear translation/rotation stage (ILS100HA, Newport, Irvine, CA) with a customized MATLAB program (MathWorks Inc., Natick, MA). The transducer was excited with an electrical pulse of energy 1 µJ, and the echoes were amplified by 20 dB before digitized by a 12-bit analog-to-digital card (CS122G1, Dynamics Signals LLC., Lockport, IL) at a sampling frequency of 1 GHz. From the pulse-echo response, the centre frequency and −6 dB bandwidth of the transducer were found to be 153 MHz and 12% (144-162 MHz), respectively.
Moreover, it was found from the wire target image that its axial and lateral resolutions were 28.5 and 8.6 μm, respectively. The performance evaluation of the developed high-frequency ultrasound frontend system was conducted. For this purpose, electrical pulses at the output of the developed pulse generator were measured using a digital phosphor oscilloscope (TDS 5052, Tektronix, Beaverton, Oregon), which were generated at a PRF of 167 kHz (Fig. S2). In this experiment, it was found that monocycle bipolar pulses generated by the front-end system had a centre frequency of 200 MHz, a −6 dB bandwidth of 90% (110 MHz-290 MHz), and a maximum peak-to-peak amplitude of 50 V.
Additionally, the pulses were successfully generated at a PRF of 167 kHz. The amplification gain of the receiver in the front-end system was 19.5 dB and its noise figure was measured to be 1.6 dB by using a spectrum analyser (E4402B, Agilent technologies, Santa Clara, CA).

Change in acoustical trapping force as a function of pulse repetition frequency (PRF).
Trapping forces from the acoustic tweezers were measured using a micropipette aspiration technique (MAT) [6].
The inner diameter of a micropipette was fabricated to be 3 μm by using a vertical micropipette puller (PC-10, Narishige, NY). The suction force was controlled by the control software (NBSC Controller, Neo biosystem, CA). One microsphere in the chamber filled with PBS was trapped by the respective ultrasound microbeams generated at PRFs of 33, 67, and 167 kHz. Note that the IB coefficient measurement was conducted under the same experimental conditions as the trapping force measurement. The suction force from a micropipette acting on the microsphere was equally balanced by the trapping force generated from the ultrasound microbeams. The microsphere gradually moved away from the centre of acoustic beam with increase in the suction force. Depending on the displacement of the trapped microsphere from the centre of ultrasonic beam area while trapping, the trapping force changes similar to an elastic spring [7]. At the fixed excitation voltages of 50 V p-p , the maximum trapping force was found to be 25.66 ± 2.81, 75.35 ± 4.61, and 122.44 ± 13.4 nN at PRFs of 33, 67, and 167 kHz, respectively.