Improved Laser Manipulation for On-chip Fabricated Microstructures Based on Solution Replacement and Its Application in Single Cell Analysis

Abstract

In this paper, we present the fabrication and assembly of microstructures inside a microfluidic device based on a photocrosslinkable resin and optical tweezers. We also report a method of solution replacement inside the microfluidic channel in order to improve the manipulation performance and apply the assembled microstructures for single cell cultivation. By the illumination of patterned ultraviolet (UV) through a microscope, microstructures of arbitrary shape were fabricated by the photocrosslinkable resin inside a microfluidic channel. Based on the microfluidic channel with both glass and polydimethylsiloxane (PDMS) surfaces, immovable and movable microstructures were fabricated and manipulated. The microstructures were fabricated at the desired places and manipulated by the optical tweezers. A rotational microstructure including a microgear and a rotation axis was assembled and rotated in demonstrating this technique. The improved laser manipulation of microstructures was achieved based on the on-chip solution replacement method. The manipulation speed of the microstructures increased when the viscosity of the solvent decreased. The movement efficiency of the fabricated microstructures inside the lower viscosity solvent was evaluated and compared with those microstructures inside the former high viscosity solvent. A novel cell cage was fabricated and the cultivation of a single yeast cell (w303) was demonstrated in the cell cage, inside the microfluidic device.

Keywords

Laser Manipulation On-chip Fabrication Microstructures Solution Replacement Single Cell Analysis

1. Introduction

Traditional group cell analysis has been widely used in microbiology and biotechnology. It studies information based on the averages of large cell populations. However, the information obtained by this method is often incomplete and misleading. It can report only the average values for the population. It is not capable of determining the characteristics or contributions of individual cells [1]. Also, group cell analysis requires many cell samples and a great deal of time. Recently, single cell analysis has received significant attention and it will in future contribute to more detailed information about the individual cell [2].

It is reported that microfluidic devices are effective in realizing a stable environment for single cell analysis. Microfluidic devices have many advantages, such as their low cost, high integration and high speed analysis, as in [3]. The essential technologies for on-chip single cell analysis are the cell sorter, which can divide and retrieve specified cells [4], the cell response measurement system [5] and the immobilization technique for cells in the desired space [6]. Additionally, advanced techniques have been established and used in studying the physical properties of single cells, such as stiffness measurement and local environment measurement, as in [7, 8]. However, as seen in [9], the functions of conventional microfluidic devices are poor and simple, because of the unchangeable of microfluidic channel. As seen in [10], 3D complicated shape microstructures are fabricated directly inside the microfluidic channel by photon microfabrication with about 100 nm resolution. However, expensive devices are needed and the fabrication process takes a long time. Besides, single cell analysis and local environment control have been investigated by using microstructures made of polystyrene beads or gel, as in [11, 12]. The shapes of these conventional microstructures are simple and they are inadequate for constructing microfluidic single cell analysis systems.

To solve this problem, we present the on-chip fabrication and assembly method for microstructures based on a photocrosslinkable resin and laser manipulation. The on-chip fabrication and manipulation of the microstructures are expected to be applied for single cell analysis, such as cell manipulation, cultivation and measurement. As shown in Figure 1, via the objective lens of a microscope, the patterned ultraviolet (UV) was illuminated through the mask to the photocrosslinkable resin, which was inside a microfluidic channel made of polydimethylsiloxane (PDMS). The microfluidic channel had both glass and PDMS surfaces for fabricating immovable and movable microstructures, respectively. The photocrosslinkable resin is poly(ethylene glycol) diacrylate (PEGDA). It is polymerized by UV and the arbitrarily shaped microstructures are fabricated at desired places inside the microfluidic channel, as seen in [13]. PEGDA has low toxicity and good biocompatibility [14]. Besides, cell viability has been positively confirmed inside PEGDA [15]. It is not necessary to prefabricate microstructures or to inject the microstructures into the channel from outside. It is possible to fabricate the required microstructures at the desired places. The fabricated microstructures can be manipulated by a laser manipulation system, as in [16–17]. As such, the microstructures can be assembled and applied as functional components of the microfluidic single cell analysis devices.

Figure 1.

A schematic drawing of the on-chip fabrication and laser manipulation for arbitrarily shaped microstructures

Beyond the microfluidic devices, cell samples also play a key role in single cell analysis. The selected cells with the desired characteristics, which are obtained by single cell diagnosis methods, are required to be cultured to obtain cell samples with identical characteristics. Single cell diagnosis methods, including the cell viability test, surface stiffness measurement and cell-cell adhesion force measurement, have been developed in our previous works, as seen in [18 –21]. There are numerous applications of functional microfluidic devices, and an important of these is single cell cultivation [22]. The main issues are immobilization and environment control. The most convenient methods for cell immobilization are based on those such as aspiration, channel structures and dielectrophoresis (DEP), as in [23]. The fixing force via aspiration is large but it damages the cells [24]. Cells can be immobilized inside microfluidic devices by channel structures. However, the immobilized cells are difficult to analyse further because of the fixed channel structures, as seen in [25]. The DEP force is non-contact and it involves low damage to cells. However, the DEP force is not sufficient to immobilize cells during changes in a medium with high flow velocity, as seen in [26 –28].

On-chip fabrication and assembly comprise a feasible way to conduct cell immobilization and cultivation. We also report on the on-chip solution replacement method inside the microfluidic channel. The solution replacement reduced the viscosity of the solvent and improved the manipulation performance of the microstructures. Higher speed laser manipulation was obtained and demonstrated by rotating a microgear. The solution replacement also provided the environment control method for the single cell cultivation inside the microfluidic channel. A novel cell cage was assembled with both movable and immovable microstructures and the cultivation of a single yeast cell (w303) was conducted in the cell cage, inside the microfluidic device. The entire system developed shows strong potential for application in single cell analysis.

2. On-chip laser manipulation and fabrication system

2.1 System hardware setup

The on-chip laser manipulation and fabrication systems are shown in Figure 2. The laser (ytterbium fibre laser, wavelength 1064 nm) for the optical tweezers was built into an inverted microscope (IX-71, Olympus). The X-Y stage and the height of the objective lens (Z-axis) were controlled for manipulating the objects. Several objective lenses were used, including the oil immersion objective UPTFTN 100XOI2 (Olympus, for laser manipulation and on-chip fabrication) and the air objective 40X (Olympus, for on-chip fabrication). The laser was scanned on the X-Y coordinate by controlling the angle of the galvanometer mirror (TSA-10A-30, Harmonic Drive Systems). UV was illuminated by the mercury lamp (USH-1030T, USHIO) and the exposure time was controlled by the shutter (BSH-RIX, Sigmakoki). The microfluidic device was set on the stage. The observation was performed by the CCD camera (XC-555, Sony).

Figure 2.

A schematic drawing of the on-chip laser manipulation and fabrication systems

2.2 On-chip fabrication method of microstructures

As shown in Figure 2, a mask made of PET (polyethylene terephthalate) was set in front of the shutter. The patterns on the mask were able to be shaped arbitrarily. UV was illuminated through the mask and the objective lens into the PEGDA (molecular weight 200, Polysciences) inside the channel of the PDMS (SITPOT 184 W/C, Dow Corning Toray) microfluidic devices. The UV was patterned by the masks. The photocrosslinkable resin was polymerized and the arbitrarily shaped microstructures were fabricated inside the solution.

The position of the microstructure was controlled by the X-Y stage. When the surface of the bottom in the channel is glass, the fabricated microstructure adheres on the glass surface. As seen in [29], when the surface of the channel bottom is PDMS, the fabricated microstructure is able to move in the non-polymerized resin freely without adhering. This is because PDMS has permeability to air and because it penetrates oxygen. Oxygen molecules inhibit the radical polymerization of PEGDA. The oxygen layer near the PDMS surface forms a non-polymerized PEGDA layer between the microstructure and the PDMS. The thickness of the non-polymerized PEGDA layer is about 2 µm. Therefore, the thickness of the microstructures depends on the thickness of the channel and PDMS surface. For example, inside a 40 µm deep channel with both PDMS top and bottom surfaces, the thickness of the fabricated microstructure is about 36 µm.

The fabricated microstructures are shown in Figure 3. They were fabricated by illuminating the UV-ray for 0.2 s via the 100X and 40X objective lenses. The microstructure fabricated using the 100X objective lens was about 1/100 in size compared with the pattern of the mask, while it was about 1/40 size using the 40X lens. The surface morphological analysis was conducted as shown in Figure 4. The surface of the microstructure was measured following the three lines, as shown in Figure 4(a), by an instrument (Surfcorder ET200, Kosaka Laboratory Ltd). The top surface of the microstructure was smooth and the thickness error was small, as shown in Figure 4(b-d). It shows the good uniformity of the UV illumination.

Figure 3.

The arbitrarily shaped on-chip fabricated microstructures

Figure 4.

The surface morphological analysis of a microstructure: (a) Three lines of the measurement paths; (b-d) Measurement results of the thickness along with the three lines

3. Assembly method for on-chip fabricated microstructures

3.1 Assembly method based on movable and immovable microstructures

As mentioned in Section 2.2, movable and immovable microstructures are able to be fabricated on different surfaces, such as glass and PDMS. A unique microfluidic device that has both glass and PDMS surfaces on the bottom of the channel was fabricated, as shown in Figure 5. The photo resist (AZ 5214-E, Clariant (Japan) K.K.) was used as the sacrificial layer. The patterned AZ layer on the glass was made by the photolithography method. The PDMS layer was spin-coated on the top. After the PDMS was solidified, the AZ layer was removed and only the PDMS on the glass remained. The precise PDMS coating area was constructed. The side view of the device is shown in Figure 6. The microstructures fabricated on the PDMS surface are movable and those fabricated on the glass are fixed. Therefore, a novel on-chip assembly method for microstructures was developed.

Figure 5.

(a) The fabrication method of partial PDMS coated glass, (b) The top view of the microfluidic channel, with both glass and PDMS surfaces on the bottom.

Figure 6.

The microfluidic device, which has both a glass surface and a PDMS surface on the bottom of the channel

The rotational microstructure, including a rotation axis and a microgear, was assembled as an example to demonstrate the assembly technique. A movable microstructure - a microgear - was fabricated at the PDMS area, and then relocated to the glass area by laser manipulation. Secondly, an immovable microstructure - a rotation axis - was fabricated in the hole of the microgear. The microgear and the rotation axis were fabricated by illuminating the UV for 0.2 s via the 100X oil immersion objective lens (Olympus). The fabricated rotation axis was fixed and the microgear was able to rotate on it. The designs of these microgears are shown in Figure 7. The assembled rotational structures are shown in Figure 8. Based on this assembly method, various functional structures for single cell analysis are able to be constructed, such as actuators, cell sorters and cell cages.

Figure 7.

The designed patterns of the microgears

Figure 8.

The on-chip assembly method and the assembly results of the microgears with eight and 14 teeth

3.2 Rotational speed evaluation of the assembled microstructures

In order to apply a microstructure to a microfluidic device, it is important to understand the manipulation performance of the microstructure. In this section, the microgear is evaluated by measuring the rotation speeds under different laser powers.

The rotation speed of a microgear depends on the laser power, the shape of microgear and the viscosity of the solvent. In the experiment, the dependence on the laser power is focused. It is known that the laser trapping force is proportional to the laser power. As seen in [30], the following equation shows this fact:

F = Q (n_{1} \frac{P}{c})

(1)

where F is the laser trapping force, Q is the laser trapping efficiency, n₁ is the refraction index of an object for the solvent, P is the laser power and c is the velocity of light. While the laser power decreases gradually, the laser trapping force decreases. As the fluid resistance acts on the microstructure, the laser trapping points with a constant rotation speed will depart from the trapped points of the microgear. Just before the moment of departure, the laser power was recorded and it was considered as the appropriate power level for the rotation speed used. Based on this result, we demonstrated the relationship between the rotation speed of microgear and the correct laser power.

The fluid resistance, which acts on the object in the direction of movement, is related with the viscosity. The viscosity - also called ‘dynamic viscosity' - represents the internal friction coefficient, which determines the resistance for the object moving inside the liquid [31]. The viscous resistance is expressed as the following equation:

F_{r} = η A \frac{d v}{d y}

(2)

where η is the viscosity of the solution, A is a constant that depends on the dimensions of the object, and (dv/dy) is the derivative of the fluid speed in the direction perpendicular to the object. The moment for rotating the microgear is calculated by:

M = r F - \int_{A} F_{r} d l

(3)

where r is the distance between the laser trapping point and the rotation centre, and ι is the distance between the viscous resistance acting point and the rotation centre. The above theoretical analysis is a simplified model for the qualitative analysis, which contributes to the following experiments for improving the laser manipulation by adjusting the fluid viscosity η.

The microgear with six teeth shown in Figure 9 was used in the experiment. The laser trapping field was formed into an arbitrary pattern by the high-speed scanning of a single laser with a galvanometer mirror [32]. Therefore, multiple points were able to be trapped. Here, a single laser was formed into three points and the fabricated microgear was rotated as shown in Figure 9. The rotation diameter was 15.8 µm. The laser trapping points were at the tip of the teeth and the rotational manipulation was performed by trapping these teeth. The experiment was done inside 100% PEGDA (molecular weight 200, with a 6% polymeric initiator, Polysciences/Shin-Nakamura chemical Co., Ltd). By releasing the negative pressure of the outlet after the PEGDA injection, all the experiments were done inside the solution at a flow-rate of zero. The experimental results show that the rotation speed is almost proportional to the laser power. The relationship between the rotation speed and laser power is shown in Section 4.3, in Figure 15 (the line represents the gear inside 100% PEGDA 200). The rotation speed was about 5 rpm when the laser power was 2 W. This rotation speed is low, and the reason for this is the high viscous resistance. If the resistance can be reduced, it is possible to achieve a motion with a higher speed in the microfluidic device, which is important for single cell handling and analysis.

Figure 9.

The rotational manipulation by trapping multiple points

Figure 10.

The microgears fabricated inside PEGDA solutions of different concentrations

Figure 11.

Rotational manipulation results of the fabricated microgear inside PEGDA solutions of different concentrations

Figure 12.

Schematics of the microstructure immobilization method for replacing 100% PEGDA (200) with 99.5% ethanol solvent inside a microfluidic channel without flushing away the microstructure

Figure 13.

Different observation results of the microgear under an optical microscope inside 100% PEGDA (200) and 99.5% ethanol solvent

Figure 14.

Different experimental images comparing the microgears inside 100% PEGDA (200) and 99.5% ethanol at the same laser power of 1 W

Figure 15.

The rotational speeds of the microgears inside 100% PEGDA (200) solution and 99.5% ethanol solvent

4. Solution replacement inside the microfluidic channel

4.1 Microgears fabricated inside different concentrations of PEGDA

We tried to reduce the viscosity of the solution inside the microfluidic channel. Based on Equation (2), a lower viscosity means a lower resistance for the movement of microstructures. Therefore, the improvement of the laser manipulation's performance would be expected with any decrease in the viscosity of the solution inside the microfluidic channel.

Two methods for reducing the viscosity of the solution are presented. The first one is to adjust the concentration of the PEGDA by mixing it with other low viscosity liquids, such as ethanol, and then fabricating the microstructures with this mixed solution. The second one is to replace the whole solution inside the channel with a low viscosity liquid after the fabrication of the microstructures. The biggest distinction between them is that the first method is performed before fabricating the microstructure while the second one is performed afterwards. This section describes the first method.

The viscosity of PEGDA (molecular weight 200, with 6% polymeric initiator, Polysciences /Shin-Nakamura chemical Co., Ltd) is 24 mPa · s/25°C. The viscosity of ethanol (99.5 Vol%, Amakasu chemical industries) is 1.09 mPa · s/25°C. The reason why we choose ethanol is that PEGDA (200) does not dissolve inside water but does dissolve in ethanol. By mixing 100% PEGDA with 99.5% ethanol, which has a much lower viscosity, the viscosity decreases. For this mixed solution, a lower concentration of PEGDA means a lower viscosity, which means a lower viscous resistance.

Based on this method, a series of microgears were fabricated inside different concentrations of PEGDA. The concentrations were based on volume. As shown in Figure 10, the shape of the microgear becomes more transparent as the concentration of PEGDA decreases. The reason for this is that the shape of the microstructure is determined by the amount of photocrosslinkable material inside the solvent. For the laser trapping force, the trap efficiency of the microstructure is determined by its shape and refractive index. Based on Equation (1), the shape of the microstructure influences the refraction index, which in turn influences the laser trapping force.

In terms of the theories above, while the viscous resistance decreases, the laser trapping force also decreases. Based on Equation (3), it is barely apparent that the manipulation performance (the rotation moment) can be improved. The experiment results of the microgear fabricated and rotated inside different concentrations of PEGDA are shown in Figure 11. They demonstrate that during a decrease in the concentration, the rotational speed of the microgear cannot increase invariably.

4.2 Method of solution replacement from 100% PEGDA to 99.5% ethanol solvent

Based on the second method mentioned in Section 4.1, we succeed in replacing the whole solution inside the microfluidic channel after the fabrication of the microgear. 100% PEGDA (200) was injected into the PDMS microfluidic channel. The microgear was fabricated by illuminating the UV for 0.2 s via the 100X objective lens on the PDMS surface and then transported to the glass surface, as presented in Section 3.1.

The procedure for replacing the solution is shown in Figure 12. Firstly, we adjusted the UV expose position where the microgear was in the centre and then exposed the UV through the mask with the square pattern.

Secondly, a square wall was fabricated and fixed on the glass surface with a movable microgear inside. Next, 99.5% ethanol was introduced through the inlet of the channel. With the negative pressure applied at the outlet, the PEGDA solvent inside the microchannel was continuously replaced by the ethanol solvent from the inlet. The side length of the wall was about 500 µm. As the top surface of the channel was PDMS, the square wall had an open top, which granted space for replacing the solvent inside the wall. The thickness of the space was about 2 µm. By calculating the volume of PEGDA and ethanol, we indicated that the solution inside the microchannel changed to 99.5% ethanol after a certain time. Figure 13 shows the observation results of the microgears through an optical microscope inside 100% PEGDA (200) and 99.5% ethanol solvent. These two images are obviously different because of the total replacement of the solution.

4.3 Performance evaluation the manipulation of a microgear inside a 99.5% ethanol solvent

In order to demonstrate the value of the solution replacement inside the microchannel, we evaluated the rotation speed of the microgears shown in Figure 12. Laser tweezers trapped three teeth of the microgear with three trapping points. By releasing the negative pressure of the outlet, all the experiments were done inside the solution at a flow-rate of zero. Figure 14 shows the different experiment images between the microgears inside 100% PEGDA (200) and 99.5% ethanol at the same laser power of 1 W. The rotational speed inside 100% PEGDA (200) was about 2.5 rpm while inside 99.5% ethanol it was about 32 rpm.

Both experimental results (inside PEGDA or ethanol) for the rotation speeds were almost proportional to the laser power, as shown in Figure 15. The approximation polynomial of the microgear inside 100% PEGDA (200) is expressed by the following equation:

y = 2.5 x

(4)

That of the microgear inside 99.5% ethanol is expressed by the following equation:

y = 28 x

(5)

From this result, the rotation speed of the microgear inside 99.5% ethanol is approximately 11 times as large as that inside 100% PEGDA (200). This improvement of manipulation is obvious, considering with the change in viscosity. The manipulation speed increased because the viscous resistance for the movement was weaker inside the lower viscosity solvent. The laser manipulation performance for the microgear was improved. However, there are other factors influencing the improvement of the manipulation, such as velocity and shape [31]. We will study this more in future works. Furthermore, this solution replacement method is important in applying the system for cells. It provides an approach of on-chip changing the PEGDA to water or other low-viscosity solutions which are also suitable for cell living and culture. The fabricated microstructures can be used for manipulating cells via the laser manipulation system.

5. Cell cages for single cell cultivation based on solution replacement

5.1 The concept and fabrication of cell cages

The on-chip solution replacement does not only realize the higher speed manipulation of microstructures, but also provides a novel technique to change and control the cell culture environment inside the microfluidic devices. There are many potential applications, such as long-term single cell cultivation, in applying this solution replacement method. Single cell cultivation is very useful for obtaining cell samples with identical characteristics. It is important and significant for single cell analysis.

For conducting single cell cultivation, a novel cell cage is presented, as shown in Figure 16. It contains both movable and immovable parts. The size of the cage is about 20 µm which is suitable for the yeast cells. Based on the assembly and fabrication method for movable and immovable microstructures - as mentioned in Section 3 - it is possible to manipulate the cells by movable microstructures and immobilize the cells by immovable microstructures. The assembled cages shown in this figure are able to immobilize cells for longer than 24 hours during which the culture medium has been changing inside the microfluidic channel. By changing the solution from PEGDA to water and a culture medium, it is possible to conduct long-term cultivation for an immobilized cell inside the cage.

Figure 16.

A schematic drawing of the cell cage assembled by on-chip fabricated microstructures

Before applying the movable parts of the cages for yeast cells, the evaluation of the laser manipulation of the movable parts was conducted, inside both PEGDA (200) and the YPD culture medium (5% YPD Broth, Becton Dickinson and Company). Figure 17 (a)(c) and (b)(d) show the moving and rotational manipulations, respectively. All the manipulation experiments were conducted using a laser power of 1 W. The moving speed inside the YPD was about three times as large as that inside the PEGDA, and the rotational speed was about 2.5 times as large. The results demonstrate the improved laser manipulation inside the YPD culture medium.

Figure 17.

The laser manipulation evaluation of the movable parts inside both the PEGDA and YPD culture mediums

5.2 Cultivation of a single yeast cell inside a cell cage

After the fabrication of the movable and immovable microstructures designed as shown in Fig 16, a surrounding wall similar as that mentioned in Section 4.2 was fabricated on the glass surface, with the microstructures inside. However, there are small gaps of about 7 µm in the wall. The cells are able to go inside the wall, while the microstructures cannot escape from the wall through these gaps during the change of the solution. Then solution inside the microfluidic channel was changed from 100% PEGDA (200) to 99.5% ethanol, first, and then water, and finally the culture medium of YPD liquid mixed with the cells. The viscosity of the pure water was 0.9 mPa·s/25°C. The microstructures inside the different solutions are shown in Figure 18 and Figure 19.

Figure 18.

Shapes of the microstructure fabricated by PEGDA inside different solutions

Figure 19.

Images of the division of yeast cells during 12 hours culturing inside the cell cages

The wild-type yeast cell (w303) was used for demonstrating single cell cultivation [18]. Inside the microfluidic channel, the solution was replaced by the YPD culture medium mixed with yeast cells. After immobilizing the cells inside the cages - by assembling the movable part with the immovable part - the YPD culture medium was continuously supplied from the inlet of the microfluidic channel in conducting the long-term cell culture. After 12 hours cultivation, divisions of the yeast cells were observed. Besides this, cell cages of different sizes were fabricated based on different masks. More cells were able to be cultured inside the bigger cell cage. As shown in Figure 19, the smaller cage (a1, b1) (diameter 15 µm) may not have enough space for four cells, while the bigger cage (a2, b2) (diameter 20 µm) has. Four cells were obtained with the bigger cell cage. Based on these results, long-term (12 hours) single cell cultivation and cell samples divided from one cell were demonstrated.

6. Conclusion

We presented the fabrication and assembly of microstructures inside a microfluidic device based on a photocrosslinkable resin and optical tweezers. We also reported a solution replacement method inside a microfluidic channel to improve the manipulation performance and apply the assembled microstructures for single cell cultivation. Microstructures of arbitrary shapes were fabricated by photocrosslinkable resin inside a microfluidic channel. By replacing the solution components, the manipulation speed of the rotational microstructure increased when the viscosity of the solvent decreased. The rotation speed of the microgear inside 99.5% ethanol was approximately 11 times greater than that inside 100% PEGDA. We demonstrated the availability of the microstructures for manipulation at a higher speed inside the microfluidic device. A novel cell cage was fabricated and the long-term cultivation of a single yeast cell (w303) was demonstrated in the cage, inside the microfluidic device. Such on-chip fabrication, assembly and solution replacement will contribute to the creation of microfluidic devices for single cell analysis with many more functions.

Footnotes

7. Acknowledgments

We thank Prof. Toshifumi Inada at Tohoku University for providing us with the yeast wild-type strain W303. This work was partially supported by MEXT KAKENPII (Hyper Bio Assembler for 3D Cellular Systems (BioAssembler)) and Nagoya University Global COE Programme (COE for Education and Research of Micro-Nano Mechatronics).

References

Krylov

Dovichi

(2000) Single-cell analysis using capillary electrophoresis: influence of surface support properties on cell injection into the capillary. Electrophoresis. 21: 767–773.

Inoue

Wakamoto

Moriguchi

Okano

Yasuda

(2001) On-chip culture system for observation of isolated individual cells. Lab Chip. 1: 50–55.

Wheeler

Throndset

Whelan

Leach

Zare

Liao

Farrell

Manger

Daridon

(2003) Microfluidic device for single-cell analysis. Anal. Chem. 75: 3581–3586.

Chou

Spence

Arnold

Quake

(2002) An integrated microfabricated cell sorter. Anal. Chem. 74: 2451–2457.

Tamaki

Sato

Tokeshi

Aihara

Kitamori

(2002) Single-cell analysis by a scanning thermal lens microscope with a microchip: direct monitoring of cytochrome c distribution during apoptosis process. Anal. Chem. 74: 1560–1564.

Yang

(2002) Cell docking and on-chip monitoring of cellular reactions with a controlled concentration gradient on a microfluidic device. Anal. Chem. 74: 3991–4001.

Stuart

Sweedler

(2003) Single-cell analysis by capillary electrophoresis. Anal. Bioanal. Chem. 375: 28–29.

Thoumine

Ott

Cardoso

Meister

(1999) Microplates: a new tool for manipulation and mechanical perturbation of individual cells. J. Biochem. Biophys. Methods. 39: 47–62.

Yue

Xue-Feng

(2006) Novel multi-depth microfluidic chip for single cell analysis. J. Chromatogr. A. 1117: 228–233.

10.

Mamo

Nakamura

Kawata

(1997) Three-dimensional microfabrication with two-photon-absorbed photopolymerization. Opt. Lett. 22: 132–134.

11.

Maruyama

Arai

Fukuda

(2007) Gel-tool sensor positioned by optical tweezers for local pH measurement in a microchip. IEEE Int. Conf. on Robotics and Automation, pp. 806–811.

12.

Arai

Endo

Maruyama

Fukuda

Shimizu

Kamiya

(2007) 3D Manipulation of lipid nanotubes using laser trapped functional gel microbeads. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, pp. 3125–3130.

13.

Ito

Nakajima

Maruyama

Fukuda

(2009) On-chip fabrication and assembly of rotational microstructures. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, pp. 1849–1854.

14.

Panda

Ali

Chung

Hatton

Khademhosseini

Doyle

(2008) Stop-flow lithography to generate cell-laden microgel particles. Lab Chip 8: 1056–1061.

15.

Chan

Zorlutuna

Jeong

Kong

Bashir

(2010) Three-dimensional photopatterning of hydrogels using stereolithography for long-term cell encapsulation. Lab Chip. 10: 2062–2070.

16.

Grier

(2003) A revolution in optical manipulation. Nature. 424: 810–816.

17.

Ashkin

(1970) Atomic-beam deflection by resonance-radiation pressure. Phys. Rev. Lett. 25: 1321.

18.

Shen

Nakajima

Ahmad

Kojima

Homma

Fukuda

(2011) Effect of ambient humidity on the strength of the adhesion force of single yeast cell inside environmental-SEM. Ultramicroscopy. 111: 1176–1183.

19.

Shen

Nakajima

Kojima

Homma

Fukuda

(2011) Study of the time effect on the strength of cell-cell adhesion force by a novel nano-picker. Biochem. Biophys. Res. Commun. 409: 160–165.

20.

Shen

Nakajima

Kojima

Homma

Kojima

Fukuda

(2011) Single cell adhesion force measurement for cell viability identification using an AFM cantilever-based micro putter. Meas. Sci. Technol. 22 (115802).

21.

Shen

Nakajima

Yang

Kojima

Homma

Fukuda

(2011) Design and characterization of nanoknife with buffering beam for in situ single-cell cutting. Nanotechnology. 22 (305701).

22.

Kortmann

Chasanis

Blank

Franzke

Kenig

Schmid

(2009) The Envirostat - a new bioreactor concept. Lab Chip. 9: 576–585.

23.

Tsutsui

Marquina

Valamehr

Wong

(2010) Efficient dielectrophoretic patterning of embryonic stem cells in energy landscapes defined by hydrogel geometries. Ann. Biomed. Eng. 38: 3777–3788.

24.

Tixier-Mita

Jun

Ostrovidov

Chiral

Frenea

Le Pioufle

Fujita

(2004) A silicon micro-system for parallel gene transfection into arrayed cells. Int. Conf. on Mniaturized Systems for Chemistry and Life Sciences, pp. 180–182.

25.

Di Carlo

Aghdam

Lee

(2006) Single-cell enzyme concentrations, kinetics, and inhibition analysis using high-density hydrodynamic cell isolation arrays. Anal. Chem. 78: 4925–4930.

26.

Jaeger

Uhlig

Schnelle

Mueller

(2008) Contact-free single-cell cultivation by negative dielectrophoresis. J. Phys. D-Appl. Phys. 41: 175502.

27.

Lombardini

Bocchi

Rambelli

Giulianelli

Guerrieri

(2010) Horizontal nDEP cages within open microwell arrays for precise positioning of cells and particles. Lab Chip. 10: 1204–1207.

28.

Thomas

RSW

Mtchell

Oreffo

ROC

Morgan

(2010) Trapping single human osteoblast-like cells from a heterogeneous population using a dielectrophoretic microfluidic device. Biomicrofluidics. 4 (2).

29.

Dendukuri

Pregibon

Collins

Hatton

Doyle

(2006) Continuous-flow lithography for high-throughput microparticle synthesis. Nat. Mater. 5: 365–369.

30.

Maruyama

Iitsuka

Onda

Arai

(2010) Massive parallel assembly of microbeads for fabrication of microtools having spherical structure and powerful laser manipulation. IEEE Int. Conf. on Robotics and Automation, pp. 482–487.

31.

Streeter

Wylie

Bedford

(1998) Fluid Mechanics. WCB/McGraw Hill, Boston.

32.

Arai

Yoshikawa

Sakami

Fukuda

(2004) Synchronized laser micromanipulation of multiple targets along each trajectory by single laser. Appl. Phys. Lett. 85: 4301–4303.