Robotic Cell Rotation Based on Optimal Poking Direction - MDPI

micromachines Article

Robotic Cell Rotation Based on Optimal Poking Direction Chunlin Zhao 1,2 , Yaowei Liu 1,2 , Mingzhu Sun 1,2, * and Xin Zhao 1,2 1

2

*

Institute of Robotics and Automatic Information System (IRAIS), Nankai University, Jinnan District, Tianjin 300350, China; [email protected] (C.Z.); [email protected] (Y.L); [email protected] (X.Z.) Tianjin Key Laboratory of Intelligent Robotics (TJKLIR), Nankai University, Jinnan District, Tianjin 300350, China Correspondence: [email protected]; Tel.: +86-22-23503960 (ext. 802)

Received: 31 January 2018; Accepted: 20 March 2018; Published: 22 March 2018







Abstract: It is essential to have three-dimensional orientation of cells under a microscope for biological manipulation. Conventional manual cell manipulation is highly dependent on the operator’s experience. It has some problems of low repeatability, low efficiency, and contamination. The current popular robotic method uses an injection micropipette to rotate cells. However, the optimal poking direction of the injection micropipette has not been established. In this paper, a strategy of robotic cell rotation based on optimal poking direction is proposed to move the specific structure of the cell to the desired orientation. First, analysis of the force applied to the cell during rotation was done to find the optimal poking direction, where we had the biggest moment of force. Then, the moving trajectory of the injection micropipette was designed to exert rotation force based on optimal poking direction. Finally, the strategy was applied to oocyte rotation in nuclear transfer. Experimental results show that the average completion time was up to 23.6 s and the success rate was 93.3% when the moving speed of the injection micropipette was 100 µm/s, which demonstrates that our strategy could overcome slippage effectively and with high efficiency. Keywords: robotic cell rotation; optimal poking direction; slip-resistance; high efficiency

1. Introduction It is necessary to adjust cells or move specific structures of cells to the desired orientation before biological manipulation, such as in nuclear transfer [1–3] or zebrafish embryo injection [4–6]. Taking nuclear transfer as an example, the polar body of the oocyte needs be rotated to the desired orientation, such as 2 o’clock or 4 o’clock, before enucleation. At present, most biological manipulations are done manually, and these manual manipulations demand a high professional level of operators. With the development of robotics and automatic techniques, a number of minimally invasive techniques have been used to adjust the position and orientation of the cell. Some biomanipulation methods based on hydrodynamic force have been proposed [7,8]. Yalikun put forward a method using hydrodynamic rotation for a single cell [9]. However, using this method, it is not convenient to hold the cell for nuclear transfer. Microfluidic technology has also attracted the interest of researchers. Leung applied microfluidic flow control to perform 3-D rotation of mouse oocytes [10]. However, this method is not suitable for manipulating batch oocytes and the efficiency is low. Wang demonstrated a technique that utilized the friction force exerted on the standard micropipette and microchannel to rotate the embryo to a desired orientation [11]. However, this method is not suitable for small cells. A method based on dielectrophoresis is widely used [12–15]. However, the increased electrical field may affect the properties of the cell membrane and lead to

Micromachines 2018, 9, 141; doi:10.3390/mi9040141

www.mdpi.com/journal/micromachines

Micromachines 2018, 9, 141 Micromachines 2018, 9, x

2 of 10 2 of 10

to potential damage when voltage the electricalfield fieldexceeds exceedsa acertain certainrange. range.The Theuse useof of optical optical potential damage when thethe voltage ofof the electrical tweezers is a technique that applies optical forces to manipulate the cell [16–20]. However, the tweezers is a technique that applies optical forces to manipulate the cell [16–20]. However, the accuracy accuracy opticalistweezers is limited the easy of anyparticles dielectric particles near theZhao laser of optical of tweezers limited due to the due easyto capture of capture any dielectric near the laser trap. trap. Zhao put forward an approach of robotic cell rotation based on minimum rotation force [21]. put forward an approach of robotic cell rotation based on minimum rotation force [21]. The rotation The rotation onrelated the celltowas to the of deformation of the whenmicropipette the injection force exerted force on theexerted cell was the related deformation the cell when the cell injection micropipette poked it. However, the efficiency was low because the optimal poking direction of the poked it. However, the efficiency was low because the optimal poking direction of the injection injection micropipette based on minimum rotation force was not used. micropipette based on minimum rotation force was not used. Tosolve solve these these problems, problems, we we calculated calculated the the optimal optimal poking poking direction direction of of the the injection injection micropipette micropipette To to have the biggest moment of force. The biggest moment of force leads to the highest efficiency when to have the biggest moment of force. The biggest moment of force leads to the highest efficiency when the same force is exerted on the cell by the injection micropipette. Furthermore, we propose a strategy the same force is exerted on the cell by the injection micropipette. Furthermore, we propose a strategy of robotic roboticcell cellrotation rotation based optimal poking direction the injection micropipette. We of based on on the the optimal poking direction of the of injection micropipette. We applied applied this strategy to oocyte rotation in nuclear transfer. The experimental results demonstrate the this strategy to oocyte rotation in nuclear transfer. The experimental results demonstrate the system’s system’s capability for fast oocyte orientation and high efficiency. capability for fast oocyte orientation and high efficiency. The rest of the paper is organized as follows: First, wewe analyze thethe force of cell rotation and and find The rest of the paper is organized as follows: First, analyze force of cell rotation the optimal poking direction of the injection micropipette in Section 2.1. Then, the strategy of robotic find the optimal poking direction of the injection micropipette in Section 2.1. Then, the strategy of cell rotation based based on optimal poking direction is introduced in Section 2.2. 2.2. TheThe experiments on robotic cell rotation on optimal poking direction is introduced in Section experiments porcine oocytes, designed to validate the feasibility of the strategy, are described in Sections 3.1 and on porcine oocytes, designed to validate the feasibility of the strategy, are described in Sections 3.1 3.2. After that, that, the experimental results are given in Section 3.3. Finally, some some discussions are given and 3.2. After the experimental results are given in Section 3.3. Finally, discussions are in Section 4. given in Section 4. 2. Materials and Methods 2.1. Derivation of Optimal Poking Direction Direction Figure 11 illustrates illustrates the the three-dimensional three-dimensional model model of of oocyte oocyte rotation. rotation. During the process of cell cell rotation, rotation, the the injection injection micropipette micropipette and and holding holding micropipette micropipette are are used to revolve and hold the oocyte, respectively. respectively. Because Because of of occlusion occlusion of of the the cytoplasm, cytoplasm, which which is is not not transparent transparent in in the the bright bright field field of of the the microscope, the polar body can be observed only when it is revolved into an area near the focal plane. microscope, the polar body can observed focal plane. This This area area was was defined defined as as the the focal focal zone. zone. Therefore, Therefore, on on the the basis basis of of whether whether the the polar polar body body could be be observed, observed, 3-D 3-D cell cell rotation rotation could could be be decoupled decoupled into into two plane motion processes, the revolving revolving course course and the the rotation rotation course. We defined O-XYZ as the cell coordinate frame, where O is located located at the the center center of of the the cell. cell. The X axis followed the central axis of the holding micropipette, and the the X-Y X-Y plane plane was was consistent consistent with with the the focal focal plane. plane. The revolving revolving course course (the (the first first stage) stage) was described described as as the process of moving the polar body to the focal zone (near the X-Y plane), rotating the cell around the Y axis. In In the the rotation rotation course course (the (the second second stage), stage), the the polar polar body body was was moved moved to to the the desired desired angular angular orientation, orientation, such such as as 11 11 o’clock, o’clock, and and the the cell cell was was rotated rotated around aroundthe theZZ axis. axis. Holding Micropipette

Z Y

X

Focal zone

oocyte Polar body

1. Three-dimensional model of oocyte oocyte rotation. rotation. Figure 1.

Through many many previous previous experiments, experiments, we we found found that that the the poking poking position position of of the the injection injection Through micropipette could not revolve the cell normally when the poking position was the intersection of micropipette could not revolve the cell normally when the poking position was the intersection the cell membrane and the X axis. To explain this phenomenon, we analyzed the forces of cell rotation in the rotation course. The rotation speed of the cell was very low (second level) and the cell could be

course where the injection micropipette runs in the X-Y plane and makes the oocyte rotate around the Z axis. Then the same analysis was applied to the revolving course. Figure 2 depicts the schematic of forces in the rotation course. The rotation force FI is exerted by the injection micropipette. FI can be divided into normal force FIn along the center of the cell and Micromachines 2018,F9, 141 3 of 10 tangential force It along the tangential direction. The inclined angle of FI and FIn is defined as β. Their relationship is shown as: of the cell membrane and the X axis. To explain this we analyzed the forces of(1) cell cos   F / Fphenomenon, In I rotation in the rotation course. The rotation speed of the cell was very low (second level) and the cell 2 could be seen in an equilibrium state. According of rigidization, the cell can be assumed (2) FI  to FInthe  principle FIt 2 to be in rigid body mode, and an analysis of static theory can be performed to calculate the optimal Theposition. effects exerted by thethe holding micropipette the holding pressure FH, the poking In addition, direction of rotationcan of be thedivided cell wasinto assumed to be counterclockwise. static As fiction force FS,above, which first is parallel to theout negative direction of the Y axis, and the supporting force mentioned we carried the corresponding mechanical analysis in the rotation Fcourse N along the X axis. The resisting force between the cell and the culture media was in accordance with where the injection micropipette runs in the X-Y plane and makes the oocyte rotate around the Newton’s lawthe of same inner analysis friction. was As the cell was inrevolving an equilibrium Z axis. Then applied to the course.state, the fluid velocity gradient du / dyFigure was2 depicts approximately equal to zero, according Newton’s law offorce inner the schematic of forces in the rotationto course. The rotation FI isfriction, exerted Fby the  A(injection du / dy) . micropipette. Therefore, theFIresisting force between the force cell and the culture media could be can be divided into normal FIn along the center of the cell and

ignored, and theFmain resisting force was the force between the cell and micropipette. tangential force the tangential direction. The inclined angle of FIthe andholding FIn is defined as β. Their It along relationship shown out as: the corresponding mechanical analysis in the revolving course, where the When weiscarried injection micropipette runs in the X-Z plane,cos in order balance the effects of gravity G and buoyancy β = Fto (1) In /F I p FF, the effects of the holding micropipette exerted on2the cell can also be divided into the force along F = FIn + FIt 2 (2) the X axis and the force along the Z axis. I

Y

Holding Micropipette

FH

FI '

FI 2 '

FN

FI 1' O

X

Z

Holding Micropipette

FI



FIt

FH

FF

FN

RH _ in

FIn

X

O G

FS

L

Figure2.2.Initial Initialpoking pokingdirection directionofofinjection injectionmicropipette micropipettein inthe thexxaxis. axis. Figure

On condition thatbythe is in the equilibriumcan state, the impact FI can be replaced byFaH , Thethe effects exerted thecell holding micropipette be divided intoofthe holding pressure ’ ’ ’ force F I and a rotation m at point O, where F I is equal to FI. Further, F I can be divided into the static fiction force Fmoment S , which is parallel to the negative direction of the Y axis, and the supporting ’I1 along the X axis and F’I2 along the Y axis. They have a relationship as follows: Fforce FN along the X axis. The resisting force between the cell and the culture media was in accordance with Newton’s law of inner friction. As the cell was in an equilibrium state, the fluid velocity gradient FI 1'  FH  FN  L * (G FF ) / RH _ in (3) du/dy was approximately equal to zero, according to Newton’s law of inner friction, F = µA(du/dy). Therefore, the resisting force between the cell and FI 2' the F culture media could be ignored, and the main (4) resisting force was the force between the cell and the Sholding micropipette. carried out theof corresponding mechanical and analysis in the revolving where the whereWhen RH_in iswe the inner radius the holding micropipette L is the distance fromcourse, the center of the injectionmicropipette micropipettetoruns in the can X-Zbe plane, in order holding O, which expressed as:to balance the effects of gravity G and buoyancy FF , the effects of the holding micropipette exerted on the cell can also be divided into the force along 2 2 (5) the X axis and the force along the Z axis. L  R  RH _ in On the condition that the cell is in the equilibrium state, the impact of FI can be replaced by a where R’ is the radius of the cell. force F I and a rotation moment m at point O, where F’ I is equal to FI . Further, F’ I can be divided into F’ I1 along the X axis and F’ I2 along the Y axis. They have a relationship as follows: F 0 I1 + FH = FN + L ∗ ( G − FF )/R H_in

(3)

F 0 I2 = FS

(4)

where RH_in is the inner radius of the holding micropipette and L is the distance from the center of the holding micropipette to O, which can be expressed as: L=

q

R2 − R H_in 2

(5)

As shown in Figure 3, F I can also be divided into F I1 along the X axis and F I2 along the Y axis. They are derived as:

Micromachines 2018, 9, 141

FI1'  FI ' * cos(   )

(6)

FI 2'  FI ' * sin(   )

4(7) of 10

where θ is the angular orientation of the contact point between the injection micropipette and the X where radius of the cell. m at O in the X-Y plane can be derived as: axis (  R is ).the The rotation moment As the cell is rotated counterclockwise, the static fiction force FS should be parallel to the positive  FI * DI conflicts FS * L  Fwith * ( Rthe  Dabove ) * sinassumption;   FS * L the injection micropipette (8) I In direction of the Y axis. Thismconclusion cannot revolve the cell normally in this poking position and direction. Therefore, the poking direction where DIn is the poking depth, and the relationship of DI and DIn is DI  ( R  DIn ) * sin  . Substituting of the injection micropipette is estimated to relate to the inclined angle β and be parallel to the X axis Equations (5), and (7) into (8), m can be calculated as: in the X-Y(4), plane. As shown in Figure 3, F’ I can also be divided into F’ I1 along the X axis and F’ I2 along the Y axis. m  FI * ( R  DIn ) * sin   FI * sin(   ) * R2  RH _ in 2 (9) They are derived as: F 0 I1rotation = F 0 I ∗ cos (θ − ) constant, the optimal initial poking (6) If the magnitude and direction of the force FI βare direction should make the rotation moment Since F 0 I2 = m F 0 Ilargest. ∗ sin(θ − β) FI, RH_in, DIn, R, β are constant, the (7) different poking positions of the injection micropipette cause different θ. The rotation moment m is where θ by is the angular orientation the contact point between the injection micropipette and the decided θ at this moment andofshould be optimized. Equation (10) is derived by taking theX axis (θ ≥ β). The rotation moment m at O in the X-Y plane can be derived as: derivative of Equation (9): m'I −FFSI ∗* L R=2 FIR∗H (_ inR2 − * cos(  ) β − FS ∗ L m = FI ∗ D D In) ∗ sin

(10)(8)

According Equation (10), when θ is equal to of β, D the rotation value. where DIn is thetopoking depth, and the relationship DIn ismoment D I = ( Rm−isDat ) ∗ maximum sin β. Substituting In its I and According direction of the injection micropipette is parallel Equations to (4),the (5),parallel and (7)line intotheorem, (8), m canthe be poking calculated as: to the X axis at this moment. It is the optimal poking direction, inq which the angular orientation of 2− 2 β. Meanwhile, the the contact point between injection micropipette and the cell θRis equal m = Fthe R H_into (9) I ∗ ( R − D In ) ∗ sin β − FI ∗ sin( θ − β ) ∗ rotation force FI should be parallel to the X axis.

Y

Holding Micropipette

FH

RH _ in

FS FN

m

FI 1' FI

'

O

L

FI

FIt FIn

X

 FI 2

FIt

FI



'

DI

DIn FIn



O

Figure3.3.Initial Initialpoking pokingposition positionofofinjection injectionmicropipette micropipetteperpendicular perpendiculartotothe theXXaxis. axis. Figure

We canmagnitude calculate the direction of the injection theoptimal revolving course as If the andoptimal direction of the rotation force FImicropipette are constant,inthe initial poking the same as the rotation course. Figure 4 depicts the schematic forces in the revolving course. Due direction should make the rotation moment m largest. Since FI , Rof , D , R, β are constant, the different In H_in topoking gravitypositions G and buoyancy FF alongmicropipette the center ofcause oocyte O, they θ. doThe not rotation produce moment the rotation of the injection different m ismoment decided mby atθOatinthis themoment X-Z plane. Therefore, the rotation moment m at O in the X-Z plane is thethe same as it is inof and should be optimized. Equation (10) is derived by taking derivative the X-Y plane. Equation (9): q m0 = − FI ∗

R2 − R H_in 2 ∗ cos(θ − β)

(10)

According to Equation (10), when θ is equal to β, the rotation moment m is at its maximum value. According to the parallel line theorem, the poking direction of the injection micropipette is parallel to the X axis at this moment. It is the optimal poking direction, in which the angular orientation of the contact point between the injection micropipette and the cell θ is equal to β. Meanwhile, the rotation force FI should be parallel to the X axis. We can calculate the optimal direction of the injection micropipette in the revolving course as the same as the rotation course. Figure 4 depicts the schematic of forces in the revolving course. Due to gravity G and buoyancy FF along the center of oocyte O, they do not produce the rotation moment m at O in the X-Z plane. Therefore, the rotation moment m at O in the X-Z plane is the same as it is in the X-Y plane.

Micromachines 2018, 9, x 9, 141 Micromachines 2018, Micromachines 2018, 9, x

5 of 105 of 10 5 of 10

Z

Holding Micropipette Holding Micropipette

FZ I

m

FIt F FInI

FIt

F FmI 1' 'F F X FIn F S FI 1 ' F X RH _ in FHFN ' O FOI 2 RH _ in FNFI FI 2 ' FI ' G L G L FH

FS

FIt

FI

 DI

DI 

O

FI



DIn

FIt FIn

DIn

FIn



O

Figure 4. Force analysis of oocyte in the X-Z plane. Figure4.4.Force Forceanalysis analysisofofoocyte oocyteininthe theX-Z X-Zplane. plane. Figure

When θ is equal to β, the rotation moment m is at its maximum value. It is the optimal poking θθ isisthe equal to rotation moment mmisisXat its When equal toβ,β,the the rotation moment at itsmaximum maximumvalue. value.ItItisisthe theoptimal optimalpoking poking direction,When in which rotation force FI is parallel to the axis. direction, in which the rotation force F I is parallel to the X axis. direction, in which the rotation force FI is parallel to the X axis. 2.2. The Strategy of Robotic Cell Rotation Based on Optimal Poking Direction 2.2.The TheStrategy StrategyofofRobotic RoboticCell CellRotation RotationBased Basedon onOptimal OptimalPoking PokingDirection Direction 2.2. The oocyte is required to be rotated to a desired orientation where the polar body is visible. The Theoocyte oocyteisisrequired required rotated atodesired a desired orientation wherepolar the polar body is visible. The to to bebe rotated body is visible. desired orientation can be determined by theto operator. It orientation is necessarywhere to usethe a strategy of robotic cell The The desired orientation can be determined byoperator. the operator. It is necessary toause a strategy of robotic desired orientation can be determined by the It is necessary to use strategy of robotic rotation based on the optimal poking direction of the injection micropipette. In order to overcome cell cell rotation based on optimal the optimal poking direction of the injection micropipette. ordertotoovercome overcome rotation based on the poking direction of oocyte, the injection micropipette. InInorder slippage between the injection micropipette and the the micropipette should maintain a slippage between the injection micropipette and the oocyte, the micropipette should maintain a slippage between injection micropipette and thethe oocyte, the micropipette maintain constant poking depth the in the revolving course. Therefore, revolving trajectory wasshould designed as a a constantpoking pokingdepth depthininthe therevolving revolvingcourse. course. Therefore,the therevolving revolvingtrajectory trajectorywas wasdesigned designed as a constant circular motion. Figure 5 illustrates the trajectory ofTherefore, the revolving course. In this course, the injection as a circular motion. Figure 5 illustrates the trajectory of the revolving course. In this course, the injection circular motion. 5 illustrates therotated trajectory of the course. In this course, the injection micropipette moved Figure in the X-Z plane and around therevolving Y axis. The injection micropipette could micropipettemoved movedin inthe theX-Z X-Zplane planeand androtated rotatedaround around theYYaxis. axis.The Theinjection injectionmicropipette micropipettecould could micropipette not revolve around the cell normally clockwise when it lay inthe the first quadrant (the first stage) [22], notrevolve revolvearound aroundthe thecell cellnormally normallyclockwise clockwisewhen whenititlay layininthe thefirst firstquadrant quadrant(the (thefirst firststage) stage)[22], not therefore, it moved counterclockwise. First, the injection micropipette poked to a certain depth and [22], therefore, moved counterclockwise. First, theinjection injection micropipette pokedtotoaacertain certaindepth depthand and therefore, moved the micropipette poked moved arounditit the trackcounterclockwise. in θ, shown in blueFirst, in Figure 5. Then, the injection micropipette moved along movedaround aroundthe theαtrack track θ,shown shown blue Figure the injection micropipette moved along ininθ, ininblue ininFigure 5.5.Then, the injection micropipette moved along in a moved circular motion in on the track shown in red in Figure 5,Then, until the polar body was observed; the inaacircular circularmotion motioninin α on the track shown in red in Figure 5, until the polar body was observed; in α on the track shown in red in Figure 5, until the polar body was observed; small amplitude strategy was applied in this process. The small amplitude strategy realized the the the small amplitude strategy was applied in this process. The small amplitudestrategy realizedthe small strategy was applied in this process. The small amplitude realized function ofamplitude real-time revolving, which divided the 60° amplitude into steps of 10°. Instrategy order to rotate the ◦ amplitude into steps of 10◦ . In order to rotate the function of real-time which divided the 60 function of real-time revolving, which divided the 60° amplitude into steps of 10°. In order to rotate the polar body to the focal plane exactly before biological manipulation, we applied the second polar body to the focal planeplane exactly beforebefore biological manipulation, we applied the second trajectory the polar body to the focal exactly biological manipulation, we applied the second trajectory to the revolving course, and the amplitude of each step was reduced to 5° when the polar ◦ to the revolving course, and the amplitude of each step was reduced 5 when the polar the body was trajectory to the revolving course, thethe amplitude each step was reduced tothe 5° when polar body was nearly observed. At the sameand time, injection of micropipette wasto moved to focal plane nearly observed. At the same time, the injection micropipette was moved to the focal plane and then body was nearly observed. At the same time, the injection micropipette was moved to the focal plane and then to the rotation course when the polar body was visible. to the rotation thewhen polarthe body wasbody visible. and then to the course rotationwhen course polar was visible. Injection Micropipette Injection Micropipette

Holding Micropipette Holding Micropipette

Z

Zα θ

X

α θ

X

Figure 5. Rotation strategy thestage first stage Figure 5. Rotation strategy of theof first (front(front view).view). Figure 5. Rotation strategy of the first stage (front view).

Figure 6 shows the motion trajectory of the rotation course. In this the injection Figure 6 shows the motion trajectory of the rotation course. In course, this course, the injection Figure 6 moved shows trajectory ofaround the rotation course. thisamplitude course, the injection micropipette moved in the X-Ymotion plane and rotated thethe Z Zaxis. small strategy micropipette inthe the X-Y plane and rotated around axis.The TheIn small amplitude strategy was micropipette moved in the X-Y plane and rotated around the Z axis. The small amplitude strategy was also exact for for the the rotation rotationcourse courseto torealize realizethe thesame samereal-time real-timefunction functionas the also designed designed to make it more exact was also designed make it injection more exact for the rotation course tocounterclockwise, realize the samealong real-time function as the revolving course.to First, micropipette was the track revolving course. First, the micropipette wasmoved movedcounterclockwise, along the track in as the revolving course. First, the injection micropipette was moved counterclockwise, along the in θ shown in yellow in Figure 6. Then the injection micropipette poked to a certain depth and movedtrack in θ shown in yellow in Figure 6. Then the injection micropipette poked to a certain depth and moved

Micromachines 2018, 9, 141 Micromachines 2018, 9, x Micromachines 2018, 9, x

6 of 10 6 of 10 6 of 10

along in in a circular motion alongthe the track shown in red in Figure until the polarand body was θ shown yellow in Figurein6.αThen injection micropipette poked to 6a certain depth moved along in a circular motion in α along the track shown in red in Figure 6 until the polar body was rotated desired position. improve precision inin thered rotation course, the the second trajectory of along intoa the circular motion in α To along the track shown in Figure 6 until polar body was rotated to the desired position. To improve precision in the rotation course, the second trajectory of the injection was To used, the same as thein revolving course. Thethe experiment reducedofeach rotated to themicropipette desired position. improve precision the rotation course, second trajectory the the injection micropipette was used, the same as the revolving course. The experiment reduced each step to 5°micropipette when the polar position (such 11 o’clock). injection wasbody used,was theclose sameto asthe thedesired revolving course. The as experiment reduced each step step◦to 5° when the polar body was close to the desired position (such as 11 o’clock). to 5 when the polar body was close to the desired position (such as 11 o’clock). Injection Micropipette Injection Micropipette

Holding Micropipette Holding Micropipette

Y Y

α α

θ θ

X X

Figure 6. Rotation strategy of the second stage (vertical view). Figure 6. Rotation strategy of the second stage (vertical view).

3. Results 3. Results Results 3. 3.1. Experimental Facilities 3.1. Experimental Facilities The robotic cell rotation experiment experiment was performed by the the NK-XMR160 NK-XMR160 micromanipulation The robotic cell rotation experiment was performed by the NK-XMR160 micromanipulation system (NK, is is shown in in Figure 7. It (NK, Tianjin, Tianjin, China), China), which whichwas wasdeveloped developedby byour ourlaboratory laboratoryand and shown Figure 7. system (NK, Tianjin, China), which was developed by our laboratory and is shown in Figure 7. It consists ofofthe It consists thefollowing followingfacilities: facilities:An Aninverted invertedmicroscope microscope(Olympus (OlympusCK-40, CK-40, Chroma Chroma Technology Technology consists of the following facilities: An inverted microscope (Olympus CK-40, Chroma Technology Corp, Bellows Falls, VT, USA) forms the basic experimental facility. A charge-coupled device (CCD) (CCD) Corp, Bellows Falls, VT, USA) forms the basic experimental facility. A charge-coupled device (CCD) (Panasonic W-V-460, Panasonic Corporation, Osaka, Japan) was installed on the experimental facility (Panasonic W-V-460, Panasonic Corporation, Osaka, Japan) was installed on the experimental facility to (Panasonic W-V-460, Panasonic Corporation, Osaka, Japan) was installed on the experimental facility to gather microscopic images atframes/s, 20 frames/s, toprocessed be processed by a host computer (i5, 2.8-GHz, 4GB gather microscopic images at 20at to beto by a host (i5, 2.8-GHz, 4GB RAM). to gather microscopic images 20 frames/s, be processed by a computer host computer (i5, 2.8-GHz, 4GB RAM). The motorized X-Y translational stage (withrange a travel range of×100 mm ×repeatability 100 mm, repeatability The motorized X-Y translational stage (with a travel of 100 mm 100 mm, of ± 1 µm, RAM). The motorized X-Y translational stage (with a travel range of 100 mm × 100 mm, repeatability of ±1 μm, and maximum speed of 2 mm/s) was installed to position the hybrid substrate, which and maximum speed of 2 mm/s) was installed to position the hybrid substrate, which is placed inisa of ±1 μm, and maximum speed of 2 mm/s) was installed to position the hybrid substrate, which is placed a 35 mm Petri A pair of 3-DOF micromanipulators are used toinjection position the injection 35 mm in Petri dish. pair dish. of 3-DOF micromanipulators are used to position placed in a 35 mmAPetri dish. A pair of 3-DOF micromanipulators are usedthe to position micropipette the injection micropipette (φ20 μm) and the holding micropipette (φ120 μm). Theof maximum speed of 3-DOF (φ20 µm) and the holding micropipette (φ120 µm). The maximum speed 3-DOF micromanipulators micropipette (φ20 μm) and the holding micropipette (φ120 μm). The maximum speed of 3-DOF micromanipulators is 1 mm/s, with repeatability of ±1 μm and travel range of 50 mm. A pneumatic is 1 mm/s, with a repeatability of ±aa1 repeatability µm and a travel range of 50aa mm. pneumatic syringe was fixed micromanipulators is 1 mm/s, with of ±1 μm and travelArange of 50 mm. A pneumatic syringe was fixed on the 3-DOF micromanipulators to provide pressure (atkPa a range on the 3-DOF micromanipulators to provide negative holdingnegative pressureholding (at a range of 0–3 and syringe was fixed on the 3-DOF micromanipulators to provide negative holding pressure (at a range of 0–3 kPa and resolution of 10 Pa) and positive injection pressure (at a range of 0–200 kPa and resolution 10 Pa) and positive pressure injection (at a range of 0–200(atkPa and resolution 10and Pa). of 0–3 kPa of and resolution of 10 injection Pa) and positive pressure a range of 0–200 of kPa resolution ofdata 10 acquisition Pa). The pressure data acquisition and motion control of the platform and The pressure and motion control of the platform and micromanipulators are processed resolution of 10 Pa). The pressure data acquisition and motion control of the platform and micromanipulators are processed by the host computer. by the host computer. micromanipulators are processed by the host computer.

Figure 7. The NK-XMR160 micromanipulation system. Figure 7. 7. The NK-XMR160 NK-XMR160 micromanipulation micromanipulation system. system. Figure The

3.2. Experimental Process 3.2. Experimental Process

Micromachines 2018, 9, 141

7 of 10

3.2. Experimental Process In order to verify the feasibility of the robotic cell rotation strategy proposed in this paper, we did the experiments on an NK-XMR160 micromanipulation system. The injection micropipette was assembled on the micromanipulator and moved by controlling the micromanipulator. The success rate of rotating the oocyte to the desired position, standard deviation, and average completion time were used to evaluate the strategy. Completion time was the total rotation time during which the cell was rotated from the initial position to the desired angular position. Standard deviation was calculated from total rotation time. A total of 30 mature porcine oocytes, divided into two groups (groups A and B), were used to perform robotic cell rotation with different moving speeds of the injection micropipette starting from the optimal poking direction. The injection micropipette started from the same initial position in all experiments of optimal poking direction. The goal of revolving was to adjust the cell to the desired orientation (11 o’clock). Through many previous experiments, we found that the poking position of the injection micropipette θ should equal 30◦ . In summary, the poking position of the contact point between the injection micropipette and the X axis θ is 30◦ and the injection micropipette should be parallel to the X axis. The experimental process was organized as follows: (1) (2) (3) (4)

(5)

(6)

Adjust the injection micropipette to the optimal direction. Set the radius of the circular motion according to the cell radius. Set θ to 30◦ (θ is alterable, and the operator can modify it according to the experimental requirements). Move the injection micropipette counterclockwise in the X-Z plane and rotate it around the Y axis. When the polar body is visible, the operator manually stops the first stage and proceeds to the second stage. The injection micropipette is moved counterclockwise in the X-Y plane and rotated around the Z axis. When the polar body is rotated to the desired orientation, the operator manually stops the second stage. The injection micropipette is automatically withdrawn to a suitable position and the experiment is over.

We did some experiments on nonoptimal poking direction to compare with the experiments on optimal poking direction. Another 30 mature porcine oocytes, also divided into two groups (groups C and D), were used to perform the same robotic cell rotation as in the experiment on optimal poking direction. The experimental process was similar, except for (1). In the experiments on nonoptimal poking direction, the injection micropipette was not parallel to the X axis. The injection micropipette started from the same initial position as in the experiments on nonoptimal poking direction. 3.3. Experimental Results Figure 8 shows the revolving course (first stage) and rotation course (second stage), in which the injection micropipette started from the optimal poking direction. As shown in Figure 8a–c, the injection micropipette was not in the focal zone when it moved in the red track shown in Figure 5. Figure 8d–f show the process of rotation when the polar body was in the focal zone (second stage), where it was required to be rotated to the desired orientation. Figure 9 shows the revolving course and rotation course, in which the injection micropipette started from the nonoptimal poking direction and was not parallel to the X axis. For group A of oocytes, the moving speed of the injection micropipette was 100 µm/s, and the system rotated 14 oocytes to the desired position successfully, with a success rate of 93.3%. The single failure was due to the obvious elliptical shape of the oocyte. Because the motion trajectory was standard circular motion, the radius of the oocyte was estimated to determine the injection micropipette motion. In order to avoid failure in future experiments, we should try our best to improve the circle

Micromachines 2018, 9, 141 Micromachines 2018, 9, x Micromachines 2018, 9, x

8 of 10 8 of 10 8 of 10

resistance force algorithm. between the the oocyte and the holding holding micropipette. Inthe order to choose choose good elastic elastic radius detection Foroocyte group and B of the oocytes, the moving speed of injection micropipette was resistance force between micropipette. In order to good cells, in future research we should try to add the step of measuring the elasticity of the cell membrane. 200 µm/s and 13 oocytes were successfully operated, hence the success rate was 86.7%. The reason for cells, in future research we should try to add the step of measuring the elasticity of the cell membrane. For group C ofcell oocytes, when the the moving moving speed of the thethe injection micropipette was 100 viscous μm/s, the the failure was thatC the membrane was too soft.speed The softer cell membrane is, the more it For group of oocytes, when of injection micropipette was 100 μm/s, system only rotated 13 oocytes to the desired position successfully, for a success rate of 86.7%. For becomes. The stickiness of the cell membrane increased the resistance force between the oocyte and system only rotated 13 oocytes to the desired position successfully, for a success rate of 86.7%. For group D of of micropipette. oocytes, when whenInthe the moving speed of the the injection micropipette was 200 200should μm/s, 12 12 oocytes the holding order to choose good elastic cells, micropipette in future research we tryoocytes to add group D oocytes, moving speed of injection was μm/s, were successfully operated, and the theofsuccess success rate was only only 80%. 80%. the step of measuring the elasticity the cellrate membrane. were successfully operated, and was

Figure 8.8. 8.Experiment Experiment on optimal poking direction, showing the process process of revolving. revolving. (a) The The Figure onon optimal poking direction, showing the process of revolving. (a) The strategy’s Figure Experiment optimal poking direction, showing the of (a) strategy’spoint, starting point, as by shown by the red track track in Figure Figure 5; (b) (b) the revolvingcourse coursein in progress; progress; starting as point, shownas the by redthe track in Figure 5; (b) thethe revolving strategy’s starting shown red in 5; revolving course in (c) the strategy’s endpoint, as shown by the red track in Figure 5, with the polar body visible. (d–f) (c) strategy’s endpoint, endpoint,asasshown shownbyby track in Figure 5, with the polar visible. (c) the strategy’s thethe redred track in Figure 5, with the polar body body visible. (d–f) The process of rotation when the polar body is in the focal zone (second stage). (d) The injection (d–f) The process of rotation when polar body thefocal focalzone zone(second (secondstage). stage). (d) (d) The injection The process of rotation when the the polar body is is in in the injection micropipette moves to the starting point of the rotation course; (e) the rotation course in progress; (f) micropipette moves to the starting point of the rotation course; (e) the rotation course in progress; (f) micropipette moves to the strategy’s strategy’s endpoint, endpoint,where wherethe thepolar polarbody bodyhas hasbeen beenrotated rotatedto tothe thedesired desiredposition. position. the the strategy’s endpoint, where the polar body has been rotated to the desired position.

Figure 9. 9. Experiment on on nonoptimal poking poking direction. (a–c) (a–c) The The process process of of revolving revolving (first (first stage). stage). (d– (d– Figure Figure 9. Experiment Experiment onnonoptimal nonoptimal pokingdirection. direction. (a–c) The process of revolving (first stage). f) The process of rotation when the polar body is in the focal zone (second stage). The injection f) TheThe process of of rotation when (second stage). stage). The The injection injection (d–f) process rotation whenthe thepolar polarbody bodyisisin in the the focal focal zone zone (second micropipette is not parallel to the X axis. micropipette is is not not parallel parallel to to the the X X axis. axis. micropipette

For groups groups A A and and B B (experiment (experiment on on optimal optimal poking poking direction), direction), when when the the moving moving speed speed of of the the For For group C of oocytes, when the moving speed of thetime injection micropipette was5 100 µm/s, injection micropipette was 100 μm/s, the average completion was 23.6 s, which was s less than injection micropipette was 100 μm/s, the average completion time was 23.6 s, which was 5 s less than the system only rotated moving 13 oocytes to the position successfully, a success rate of 86.7%. in [21]. [21]. With increased speed, thedesired average completion time was wasfor17.8 17.8 s. Slippage Slippage could be in With increased moving speed, the average completion time s. could be effectively avoided, avoided, and and the the success success rate rate of of rotating rotating the the oocyte oocyte was was up up to to 86.7%. 86.7%. When When the the moving moving effectively speed of the injection micropipette was 100 μm/s, standard deviation was 5.5. As the moving speed speed of the injection micropipette was 100 μm/s, standard deviation was 5.5. As the moving speed was increased increased to to 200 200 μm/s, μm/s, standard standard deviation deviation remained remained at at 5.7, 5.7, effectively effectively proving proving the the consistency consistency of of was

Micromachines 2018, 9, 141

9 of 10

For group D of oocytes, when the moving speed of the injection micropipette was 200 µm/s, 12 oocytes were successfully operated, and the success rate was only 80%. For groups A and B (experiment on optimal poking direction), when the moving speed of the injection micropipette was 100 µm/s, the average completion time was 23.6 s, which was 5 s less than in [21]. With increased moving speed, the average completion time was 17.8 s. Slippage could be effectively avoided, and the success rate of rotating the oocyte was up to 86.7%. When the moving speed of the injection micropipette was 100 µm/s, standard deviation was 5.5. As the moving speed was increased to 200 µm/s, standard deviation remained at 5.7, effectively proving the consistency of this proposed strategy. For groups C and D (experiment on nonoptimal poking direction), when the moving speed of the injection micropipette was 100 µm/s, average completion time was 30.4 s and standard deviation was 5.9. When the moving speed of the injection micropipette was 200 µm/s, average completion time was 22.9 s and standard deviation was 6.2. Comparing the results of the optimal poking experiment and the nonoptimal poking experiment strongly proves the effectiveness of the strategy, in which we had the biggest moment of force. The proposed strategy effectively reduced the completion time. The experimental results are shown in Table 1. Table 1. The relationship of average completion time, standard deviation, and success rate with the moving speed of the injection micropipette. Speed Index

100 µm/s

200 µm/s

Optimal poking direction

Average completion time Standard deviation Success rate

23.6 s 5.5 93.3%

17.8 s 5.7 86.7%

Nonoptimal poking direction

Average completion time Standard deviation Success rate

30.4 s 5.9 86.7%

22.9 s 6.2 80%

4. Discussion Three-dimensional orientation is needed for biological manipulation. In this paper, in order to find the optimal poking direction of the injection micropipette, we analyzed the force of cell rotation. A strategy of robotic cell rotation based on optimal poking direction was proposed, so that the specific structure of the cell can be moved to the desired orientation before biological manipulation. Experimental results show that when the speed of the injection micropipette was up to 100 µm/s, the average completion time was 23.6 s and the success rate was 93.3%. Meanwhile, the circular motion effectively overcame slippage between the injection micropipette and the oocyte. With increased moving speed of the injection micropipette, slippage could be effectively avoided and the success rate of the rotating oocyte was maintained up to 86.7%. Moreover, the experiment on nonoptimal poking direction was designed to compare with the experiment on optimal poking direction. The experimental results demonstrate that this strategy had the advantages of real-time operation, slip-resistance, and high efficiency. Future work will focus on automatic cell rotation, which adds a visual detection program to the proposed strategy, replacing the manual work of detecting the polar body. Acknowledgments: This work was supported by the National Natural Science Foundation of China (61327802 and U1613220) and in part by the Science and Technology Major Project of Tianjin (14ZCDZGX00801). Author Contributions: Chunlin Zhao proposed the method; Chunlin Zhao, Mingzhu Sun, and Xin Zhao conceived and designed the experiments; Chunlin Zhao developed the NK-XMR160 micromanipulation system; Yaowei Liu performed the experiments; Chunlin Zhao and Mingzhu Sun analyzed the data and wrote the paper. Conflicts of Interest: The authors declare no conflict of interest.

Micromachines 2018, 9, 141

10 of 10

References 1.

2. 3. 4. 5. 6. 7. 8.

9. 10. 11. 12.

13. 14. 15. 16. 17. 18. 19. 20. 21. 22.

Tachibana, M.; Amato, P.; Sparman, M.; Gutierrez, N.M.; Tippner-Hedges, R.; Ma, H.; Kang, E.; Fulati, A.; Lee, H.S.; Sritanaudomchai, H.; et al. Human embryonic stem cells derived by somatic cell nuclear transfer. Cell 2013, 153, 1228–1238. [CrossRef] [PubMed] Kwon, D.; Koo, O.; Kim, M.; Jang, G.; Lee, B.C. Nuclear-mitochondrial incompatibility in interorder rhesus monkey–cow embryos derived from somatic cell nuclear transfer. Primates 2016, 57, 471–478. [CrossRef] [PubMed] Shufaro, Y.; Reubinoff, B.E. Nuclear treatment and cell cycle synchronization for the purpose of mammalian and primate somatic cell nuclear transfer (SCNT). Methods Mol. Biol. 2017, 1524, 289–298. [PubMed] Sun, Y.; Nelson, B.J. Biological cell injection using an autonomous microrobotic system. Int. J. Robot. Res. 2002, 21, 861–868. [CrossRef] Wang, W.; Liu, X.; Gelinas, D.; Ciruna, B.; Sun, Y. A fully automated robotic system for microinjection of zebrafish embryos. PLoS ONE 2007. [CrossRef] [PubMed] Huang, H.B.; Sun, D.; Mills, J.K.; Cheng, S.H. Robotic cell injection system with position and force control: Toward automatic batch biomanipulation. IEEE Trans. Robot. 2009, 25, 727–737. [CrossRef] Torino, S.; Iodice, M.; Rendina, I.; Coppola, G.; Schonbrun, E. A Microfluidic approach for inducing cell rotation by means of hydrodynamic forces. Sensors 2016, 16, 1326. [CrossRef] [PubMed] Zhen, M.; Shan, J.W.; Lin, H.; Shreiber, D.I.; Zahn, J.D. Hydrodynamically controlled cell rotation in an electroporation microchip to circumferentially deliver molecules into single cells. In Microfluidics and Nanofluidics; Springer: Berlin/Heidelberg, Germany, 2016. Yalikun, Y.; Kanda, Y.; Morishima, K. Hydrodynamic vertical rotation method for a single cell in an open space. In Microfluidics and Nanofluidics; Springer: Berlin/Heidelberg, Germany, 2016. Leung, C.; Lu, Z.; Zhang, X.P.; Sun, Y. Three-dimensional rotation of mouse embryos. IEEE Trans. Biomed. 2012, 59, 1049–1056. [CrossRef] [PubMed] Wang, Z.; Feng, C.; Muruganandam, R.; Ang, W.T.; Tan, S.Y.M.; Latt, W.T. Three-dimensional cell rotation with fluidic flow-controlled cell manipulating device. IEEE/ASME Trans. Mechatron. 2016, 21, 1995–2003. [CrossRef] Han, S.I.; Joo, Y.D.; Han, K.H. An electrorotation technique for measuring the dielectric properties of cells with simultaneous use of negative quadrupolar dielectrophoresis and electrorotation. Analyst 2013, 138, 1529–1537. [CrossRef] [PubMed] Gray, D.S.; Tan, J.L.; Voldman, J.; Chen, C.S. Dielectrophoretic registration of living cells to a microelectrode array. Biosens. Bioelectron. 2004, 19, 1765–1774. [CrossRef] [PubMed] Chau, L.H.; Liang, W.; Cheung, F.W.K.; Liu, W.K.; Li, W.J.; Chen, S.C.; Lee, G.B. Self-rotation of cells in an irrotational AC E-field in an opto-electrokinetics chip. PLoS ONE 2013, 8, e51577. [CrossRef] [PubMed] Benhal, P.; Chase, G.J.; Gaynor, B.; Oback, B.; Wang, W. AC electric field induced dipole-based on-chip 3D cell rotation. Lab Chip 2014, 14, 2717–2727. [CrossRef] [PubMed] Zhang, H.; Liu, K.K. Optical Tweezers for single cells. J. R. Soc. Interface 2008, 5, 671–690. [CrossRef] [PubMed] Wu, M.C. Optoelectronic tweezers. Nat. Photon. 2011, 5, 322–324. [CrossRef] Kolb, T.; Albert, S.; Haug, M.; Whyte, G. Optofluidic rotation of living cells for single-cell tomography. Biophotonics 2015, 8, 239–246. [CrossRef] [PubMed] Chen, H.; Wang, C.; Lou, Y. Flocking multiple microparticleswith automatically controlled optical tweezers: Solutions and experiments. IEEE Trans. Biomed. Eng. 2013, 60, 1518–1527. [CrossRef] [PubMed] Cheah, C.C.; Li, X.; Yan, X.; Sun, D. Simple PD control scheme for robotic manipulation of biological cell. IEEE Trans. Autom. Control 2014, 60, 1427–1432. [CrossRef] Zhao, Q.; Sun, M.; Cui, M.; Yu, J.; Qin, Y.; Zhao, X. Robotic cell rotation based on the minimum rotation force. IEEE Trans. Autom. Sci. Eng. 2015, 12, 1504–1515. [CrossRef] Zhao, Q. Micro-Manipulation Methods Based on Cell Mechanical Properties; Nankai University: TianJin, China, 2014; pp. 69–71. © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).