WO2017078751A1 - Micoluidic cell deomailiy assay for enabling rapid and efficient kinase screening via the crispr-cas9 system - Google Patents

Abstract

A method for identifying if a target gene may be a tumor suppressor gene, the method comprising: providing a sample of cells wherein some of the cells have been transformed to knock out the target gene and some of the cells have not been transformed to knock out the target gene; passing the sample of cells through a device which is configured such that cells which exhibit greater mechanical deformability are able to pass further along the device than cells which exhibit lesser mechanical deformability; and determining if the cells passing further along the device correlate to the cells which have been transformed to knock out the target gene.

Description

MICROFLUIDIC CELL DEFORMABILITY ASSAY FOR ENABLING

RAPID AND EFFICIENT KINASE SCREENING VIA THE CRISPR-

CAS9 SYSTEM

Applicant

The Methodist Hospital

Inventors

Lidong Qin

Xin Han

Zongbin Liu

Reference To Pending Prior Patent Application

This patent application claims benefit of pending prior U.S. Provisional Patent Application Serial No. 62/252,324, filed 11/06/2015 by The Methodist Hospital and Lidong Qin et al . for MICROFLUIDIC CELL

DEFORMABILITY ASSAY ENABLES RAPID AND EFFICIENT KINASE SCREENING VIA THE CRISPR-CAS9 SYSTEM (Attorney's

Docket No. METHODIST-30 PROV) , which patent

application is hereby incorporated herein by

reference .

Field Of The Invention

This invention relates to genetic screening in general, and more particularly to methods and

apparatus for identifying tumor suppressor genes. Background Of The Invention

Systematic loss-of-function genetic screening is an essential approach for identifying genes and pathways involved in many biological processes and diseases. See Berns, K. et al . Nature 428, 431-437

(2004); Rad, R. et al . Science 330, 1104-1107 (2010); and Carette, J.E. et al . Science 326, 1231-1235

(2009) . The CRISPR (clustered regularly interspaced short palindromic repeats ) -Cas9 system represents an efficient tool for such screening and has been successfully utilized to identify genes that regulate cell survival, confer drug resistance, and/or drive tumor metastasis. See Shalem, O. et al. Science 343, 84-87 (2014); Wang, T. et al . Science 343, 80-84 (2014); and Chen, S. et al . Cell 160, 1246-1260

(2015) . The CRISPR approach provides complete target deletion (i.e., gene knockout) and yields cells with representative phenotypes. See Sander, J.D. & Joung, J.K. Nat. Biotechnol. 32, 347-355 (2014). A wide range of screening methods can then be used to sort such phenotypes and search for potential biomarkers. See Hsu, P.O. et al . Cell 157, 1262-1278 (2014).

Current cancer biomarker screening methods typically rely on in vitro or in vivo cell

proliferation and metastasis assays on the knockout cell phenotypes. See Shalem, 0. et al . Science 343, 84-87 (2014); Wang, T. et al . Science 343, 80-84 (2014); Chen, S. et al . Cell 160, 1246-1260 (2015); Koike-Yusa, H. et al . Nat. Biotechnol . 32, 267-273 (2014); and Zhou, Y. et al. Nature 509, 487-491

(2014) . Such assays are often hindered by low

efficiency and generally require extended time and effort, due to limited throughput and time-consuming characterization and selection of cells.

Integrated microfluidic chips can be advantageous in improving the efficiency of cell screening. See Lecault, V. et al. Nat. Methods 8, 581-586 (2011). Such microfluidic chips have been designed with micro- and nanostructures to rapidly distinguish cell

morphology and dynamics. See Zhang, W. et al . Proc. Natl. Acad. Sci. USA 109, 18707-18712 (2012).

Cell deformability is a promising label-free biomarker that indicates changes in cytoskeletal or nuclear organization. Research on cancer cell

mechanical-phenotyping has consistently revealed that high deformability is associated with increased tumor- initiating capacity and metastatic potential, which suggests that cell deformability-based genetic

screening may allow for the discovery of new cancer biomarkers. See Zhang, W. et al . Proc. Natl. Acad. Scl. USA 109, 18707-18712 (2012); Byun, S. et al.

Proc. Natl. Acad. Sci. USA 110, 7580-7585 (2013);

Gossett, D.R. et al . Proc. Natl. Acad. Sci. USA 109,

7630-7635 (2012); Wirtz, D. et al . Nat. Rev. Cancer 11, 512-522 (2011); and Cross, S.E. et al . Nat.

Nanotechnol. 2, 780-783 (2007). U 2016/000096

Summary Of The Invention

The present invention combines a CRISPR-Cas9

knockout screen with cell mechanics-based, on-chip sorting in order to identify potential tumor

suppressor kinases. Compared with traditional

screening for cellular activities, cell mechanics- based sorting via a microfluidic chip is a label-free, high-throughput, cost-effective, and time-saving

approach which can accelerate the discovery of genes and pathways underlying key cellular processes. The novel method and apparatus utilized by the present invention comprises the first CRISPR screening example developed in the microfluidics biotechnology field.

The present invention comprises the provision and use of a novel microfluidic cell deformability assay for performing genome-wide loss-of-function screening. More particularly, the present invention comprises the use of the CRISPR-Cas9 system to "knock out"

individual kinases, and the use of a microfluidic chip to identify when the loss of certain kinases makes cells more deformable and hence potentially more

invasive. Such kinases cover well-reported tumor

suppressors, including the chk2, IKK- , p38 MAPKs and DAPK2 genes, as well as the less-studied MAST1 and

STK4 genes. In an experiment performed using the

novel methods and apparatus of the present invention, STK4's role in cell deformability and cancer progression was investigated and its important tumor suppressing function was discovered.

Thus, with the present invention, a CRISPR approach (i.e., a CRISPR-Cas9 system) may be used to "knock out" a gene of interest, and then the cells may be introduced into a microfluidic device designed and arrayed so as to sort the cells based on their level of deformability, with the most deformable cells progressing the furthest along the microfluidic device. Based on the distance to which a cell with a transfected gene travels, the gene which is "knocked out" may be identified as a high interest candidate for a tumor suppressor gene. Using this technique, 38 potential gene candidates have already been identified whose function may involve the regulation of cell deformability and invasion.

In one preferred form of the invention, there is provided a method for identifying if a target gene may be a tumor suppressor gene, the method comprising: providing a sample of cells wherein some of the cells have been transformed to knock out the target gene and some of the cells have not been transformed to knock out Lhe target gene;

passing the sample of cells through a device which is configured such that cells which exhibit greater mechanical deformability are able to pass further along the device than cells which exhibit lesser mechanical deformability ; and determining if the cells passing further along the device correlate to the cells which have been transformed to knock out the target gene.

In another preferred form of the invention, there is provided a method for identifying genes which may be tumor suppressor genes, the method comprising:

providing a sample of cells and providing a library of agents, wherein each agent is configured to knock out a different target gene, and selectively introducing the library of agents to some of the sample of cells so as to knock out different target genes in the sample of cells;

passing the sample of cells through a device which is configured such that cells which exhibit greater mechanical deformability are able to pass further along the device than cells which exhibit lesser mechanical deformability; and

determining which genes have been knocked out in those cells passing further along the device.

In another preferred form of the invention, there is provided a system for identifying if a target gene may be a tumor suppressor gene, the system comprising: a sample of cells wherein some of the cells have been transformed to knock out the target gene and some of the cells have not been transformed to knock out the target gene;

a device which is configured such that cells which exhibit greater mechanical deformability are able to pass further along the device than cells which exhibit lesser mechanical deformability ;

an introducer for delivering the sample of cells into the device; and

an analyzer for determining if the cells passing further along the device correlate to the cells which have been transformed to knock out the target gene.

In another preferred form of the invention, there is provided a system for identifying genes which may be tumor suppressor genes, the system comprising:

a sample of cells wherein different target genes have been knocked out in some of the cells;

a device which is configured such that cells which exhibit greater mechanical deformability are able to pass further along the device than cells which exhibit lesser mechanical deformability ;

an introducer for delivering the sample of cells into the device; and

an analyzer for determining which genes have been knocked out in those cells passing further along the device .

In another preferred form of the invention, there is provided novel apparatus for separating cells, the apparatus comprising a microfluidic device, wherein the microfluidic device comprises:

a housing having an inlet, a flow chamber

connected to the inlet, and an outlet connected to the flow chamber; and a plurality of structures spaced laterally across the flow chamber and serially along the flow chamber, wherein the laterally-spaced structures define a plurality of gaps therebetween, and further wherein the gaps between the laterally-spaced structures become smaller along the length of the flow chamber so that cells passing through the gaps are increasingly mechanically deformed along the length of the flow chamber .

Brief Description Of The Drawings And Tables

These and other objects and features of the present invention will be more fully disclosed or rendered obvious by the following detailed description of the preferred embodiments of the invention, which is to be considered together with the accompanying drawings and tables wherein like numbers refer to like parts, and further wherein:

Figs. 1 and 2A are schematic views of a novel system for performing kinase screening using (i) the

CRISPR-Cas9 system to selectively knock out target genes, and (ii) a microfluidic device for separating cells based on cell deformability;

Fig. 2B is a schematic view showing fluorescence images of DMSO- and Cytochalasin D- treated MDA-MB-231 cells within a microfluidic device;

Fig. 2C is a schematic view showing a statistical analysis of transport distance through a novel microfluidic device for CRISPR kinase-KO (knockout) cells ;

Figs. 2D and 2E are schematic views showing the results of an experiment performed in accordance with the present invention;

Fig. 3A is a schematic view showing Western blot analysis of wild-type and sgRNA-modified MDA-MB-231 cells one week after infection;

Fig. 3B is a schematic view showing fluorescence images of wild-type and sgRNA-modified MDA-MB-231 cells trapped within a microfluidic device;

Fig. 3C is a schematic view showing the ratio of wild-type and sgRNA-modified MDA-MB-231 cells after separation using a microfluidic device;

Figs. 3D-3I are schematic views showing the results of an experiment performed using the novel system of the present invention;

Fig. 4A is a schematic view of a novel

microfluidic device formed in accordance with the present invention;

Figs. 4B-4E are schematic views showing further aspects of the novel microfluidic device of Fig. 4A;

Figs. 5A and 5B are schematic views showing the results of an experiment performed using the novel system of the present invention;

Fig. 6A is a schematic view showing a statistical analysis of transport distance through a microfluidic device for DMSO- and Cytochalasin D- treated MDA-MB- 231 cells;

Figs . 6B and 6C are schematic views showing the results of an experiment performed using the novel system of the present invention;

Fig. 7A is a schematic view outlining a CRISPR- Cas9-based kinase KO (knockout) screen using a

microfluidic device in accordance with the present invention;

Figs. 7B-7D, 8A-8C, 9A, 9B, lOA-lOC, 11A-11D,

12A, 12B, 13A, 13B and 14 are schematic views showing the results of experiments performed using the novel system of the present invention;

Table 1 is a listing of at least two independent sgRNAs targeting the same gene with log2Foldchange >=7 and p-value <0.001;

Table 2 shows a functional analysis of top-ranked genes;

Table 3 shows top kinase tumor suppressor hits from screening; and

Table 4 shows primers used for nested PCR and Miseq 50 sequencing.

Detailed Description Of The Preferred Embodiments

With the present invention, a CRISPR approach

(i.e., a CRISPR-Cas9 system) may be used to "knock out" a gene of interest, and then the cells may be introduced into a microfluidic device designed and arrayed so as to sort the cells based on their level of deformability, with the most deformable cells progressing the furthest along the microfluidic device. Based on the distance to which a cell with a transfected gene travels, the gene which is "knocked out" may be identified as a high interest candidate for a tumor suppressor gene. Using this technique, 38 potential gene candidates have already been identified whose function may involve the regulation of cell deformability and invasion.

Among other things, the present invention

comprises the provision and use of a unique

microfluidic chip capable of assessing cell

deformaliuii at the Single-cell level, and sorting cells on the basis of deformability and hence invasive potential. With the "on-chip" cell sorting system of the present invention, flexible cells with high deformability and metastatic propensity easily pass through microbarriers (sometimes hereinafter referred Ιυ as "microposts" ) , whereas stiff cells remain trapped .

In addition, the present invention also comprises the provision and use of the microfluidic cell

deformation screen in conjunction with a CRISPR-based knockout approach to facilitate CRISPR-based kinase screening . See Fig. 1, which is a schematic view summarizing how kinase screening may be effected via a CRISPR-Cas9 system and microfluidic cell separation. The Microfluidic Device For Screening Cells On The

Basis Of Deformability

Looking now at Fig. 2A, there is shown a novel microfluidic device 5 which may be used to sort cells on the basis of cell deformability and which may facilitate CRISPR-Cas9-based kinase screening, as will hereinafter be discussed in further detail.

Microfluidic device 5 generally comprises a base chip 10 and a cover chip 15 that is spaced from base chip 10 so as to define a flow chamber 20

therebetween. At least one inlet 25 is fluidically connected to one end of flow chamber 20 and at least one outlet 30 is fluidically connected to the opposite end of flow chamber 20. A plurality of microposts 35 extend upwardly from base chip 10 into flow chamber 20, whereby to define a plurality of gaps 40 between adjacent microposts 35.

Microfluidic device 5 utilizes microposts 35 (i.e., artificial microbarriers ) to separate flexible cells from stiff cells by hydrodynamic force. In one preferred form of the present invention, the

microfluidic device 5 comprises approximately two million rectangular microposts 35, each approximately 30 μπι in height, arrayed so as to form gaps 40 between adjacent microposts 35, with the distances between adjacent microposts 35 (i.e., the widths of gaps 40) decreasing from approximately 15 im to approximately 6 ym when moving through flow chamber 20 from the end of flow chamber 20 closest to inlet 25 to the end of flow chamber 20 closest to outlet 30. In other words, the widths of gaps 40 narrow along the length of flow chamber 20.

As a result of this construction, when cells of different deformability are introduced into the at least one inlet 25 and flowed down flow chamber 20, the progressively more restrictive gaps 40 enable those cells which are more deformable to pass further along flow chamber 20 than those cells which are less deformable. Significantly, the distance that a cell travels along flow chamber 20 reflects the relative deformability of that cell. As will hereinafter be discussed, Figs. 2B-2E and 3A-3I illustrate how cells of different deformabilities pass along flow chamber 20.

It will be appreciated that various

configurations may be provided for microfluidic device 5.

In one preferred form of the present invention, and looking now at Fig. 4A, microfluidic device 5 comprises three cell deformation zones 45A, 45B, 45C arrayed serially along flow chamber 20 and extending laterally across the width of flow chamber 20. Cell deformation zone 45A is located closest to inlet 25, and comprises a plurality of microposts 35A separated by gaps 40A between adjacent microposts. Cell

deformation zone 45B is located "downstream" of cell deformation zone 45A (i.e., extending away from cell deformation zone 45A toward outlet (s) 30), and

comprises a plurality of microposts 35B separated by gaps 40B between adjacent microposts. Cell

deformation zone 45C is located "downstream" of cell deformation zone 45B (i.e., extending away from cell deformation zone 45B toward outlet (s) 30), and

comprises a plurality of microposts 35C separated by gaps 40C between adjacent microposts. It should be appreciated that in this form of the invention, gaps 40A are wider than gaps 40B, and gaps 40B are wider than gaps 40C.

It should also be appreciated that while three cell deformation zones 45A, 45B, 45C are shown in Fig. 4A, more or fewer cell deformation zones 45A, 45B, 45C may be provided in flow chamber 20 without departing from the scope of the present invention. By way of example but not limitation, a plurality of cell deformation zones 45A may be disposed laterally across the width of flow chamber 20 and longitudinally along a portion of the length of flow chamber 20, such that any cell entering flow chamber 20 must pass through gaps 40A of cell deformation zones 45A; and a

plurality of cell deformation zones 45B may be disposed laterally across the width of flow chamber 20 and longitudinally along a portion of the length of flow chamber 20, such that any cell entering flow chamber 20 must pass through gaps 40B of cell

deformation zones 45B; and a plurality of cell

deformation zones 45C may be disposed laterally across the width of flow chamber 20 and longitudinally along a portion of the length of flow chamber 20, such that any cell entering flow chamber 20 must pass through gaps 40C of cell deformation zones 45C.

It should also be appreciated that additional cell deformation zones, having different gaps 40, may be provided in flow chamber 20.

And it should be appreciated that inlet 25 may be branched so as to evenly distribute the cells passing through inlet 25 as the cells enter flow chamber 20, and a plurality of outlets 30 may be provided (e.g., 4 outlets 30, as shown in Fig. 4A) without departing from the scope of the present invention.

Figs. 4B-4E show further details of the

construction of microfluidic device 5.

Using The Microfluidic Device To Sort Cells Based On

Cell Deformability

In an experiment, a mixture of MDA-MB-231 cells

(human breast cancer cells) treated with

dimethylsulfoxide (DMSO) and the cytoskeleton- inhibiting drug Cytochalasin D were passed through flow chamber 20 of microfluidic device 5 (i.e., introduced into flow chamber 20 via inlet 25 and collected from flow chamber 20 via outlet 30) in order to validate the separation efficiency of microfluidic device 5. Consistent with previous research, the

Cytochalasin D treatment inhibited actin

polymerization, reduced F-actin bundling, and enhanced flexibility as demonstrated by on-chip staining of trapped cells (Figs. 5A and 5B) . See Otto, O. et al . Nat. Methods 12, 199-202 (2015). As a "proof-of- concept" study, Cytochalasin D- and DMSO-treated MDA- MB-231 cells were stained with different fluorescent dyes and then mixed equally to a final density of 1 * 10⁶ cells/mL. After perfusion of the cells through microfluidic device 5, trapped cells were imaged by fluorescence microscopy. The distribution of

Cytochalasin D-treated cells in the chip (i.e., along flow chamber 20) differed from the distribution of DMSO-treated cells (i.e., along flow chamber 20) in the microfluidic device. Compared to DMSO-treated cells, there were more Cytochalasin D-treated cells trapped in the smaller gaps (i.e., gaps 40C) further along flow chamber 20 (Fig. 2B) .

Statistical analysis of on-chip transport

distance (i.e., transport distance along flow chamber

20) versus cell diameter reveals distinct separation efficiencies for the two treatments (Fig. 6A) . The average transport distance (i.e., transport distance along flow chamber 20) of Cytochalasin D-treated cells were about a 1.71-fold increase compared with DMSO- treated cells. Mixed cells at inlet 25 and collected cells at outlet 30 were also characterized with fluorescent imaging (Fig. 6B) , showing the increment of Cytochalasin D-treated cells from 50% at the inlet to 88% at the outlet (Fig. 6C) .

It should be noted that cell heterogeneity, which includes characteristics such as cell size and cell cycle phases, affects separation efficiency.

Nevertheless, the Cytochalasin D-treated cells

transported farther along flow chamber 20 of

microfluidic device 5, and because no clear

correlation between cell diameter and transport distance has been established, these data indicate that changes in the cytoskeleton induced by

Cytochalasin D are responsible for the separation of Cytochalasin D-treated cells from DMSO-treated cells in microfluidic device 5.

Using A CRlS K-Casb) Knockout Procedure With The Microfluidic Device To Identify Potential Tumor

Suppressor Kinases

Because microfluidic device 5 more readily passes flexible cells to the end of the micropost array

(i.e., the outlet end of flowchamber 20), and because the mechanical property of a cell is correlated with its metastatic potential, a mechanical cell sorting approach may be combined with the CRISPR-Cas9 knockout (KO) technology to identify potential tumor suppressor kinases .

To this end, in another experiment, as an initial test, a single-guide RNA (sgRNA) library targeting 507 kinase genes was screened for potential genes involved in the regulation of cell deformability (Fig. 7A) .

First, a derivative of the MDA-MB-231 cell line that stably expresses FLAG-Cas9 under a doxycycline- inducible promoter was generated (Fig. 7B) . The Cas9 expressing cell line was transduced with a CRISPR kinase-KO lentivirus pool at a ratio of greater than 500 cells per lentiviral CRISPR construct. After in vitro culturing of the cells for 1 week, the

transduced cells were loaded into microfluidic device

5 for sorting (i.e., a slurry of the transduced cells was introduced into flow chamber 20 via inlet 25) .

Non-transduced cells expressing FLAG-Cas9 were also loaded into microfluidic device 5 (i.e., introduced into flow chamber 20 via inlet 25) as a control.

The statistical data generated by the experiment demonstrated that the CRISPR kinase-KO cells (i.e., the CRISPR kinase-knockout cells) were more

heterogeneous when characterized by the "on-chip transport distance" (i.e., the distance along flow chamber 20 that the cells passed) , but there was no obvious change in the diameter of the cells (Fig. 2C) . In other words, since the data indicated that there was no obvious change in the diameter of the cells collected as a result of transformation, the distance that the cells moved along flow chamber 20 was

determined solely by cell deformability , i.e., the cells which were more deformable were more likely to pass more easily through gaps 40 between adjacent microposts 35. Thus, more deformable cells arrived at (or came closer to) outlet 30 than less deformable cells, thereby permitting microfluidic device 5 to be used to mechanically separate cells on the basis of cell deformability .

As expected, a small portion of the transduced cells (~15%) was transported longer distances on the chip, and these cells with higher deformability were allowed to flow out of outlet 30 of microfluidic device 5. The sgRNA barcodes of the sorted flexible cells (collected from output 30) were sequenced, as well as the entire initial pool of cells (introduced into input 25) (Figs. 7C and 7D) . To identify gene hits, the difference in abundance in the output cell population and the input cell population for each selected sgRNA was evaluated by calculating fold- enrichment and performing the Chi-Square test. Top sgRNA hits were identified by using the cutoff of log2Fold-change >=7 and p-value <0.001 (Fig. 2D) . 38 potential candidate genes were identified with at least two independent sgRNA among the top hits, whose function may be involved in the regulation of cell deformability and invasion (Tables 1 and 2).

According to earlier studies, the loss of tumor suppressor genes would cause the cells Lo become more flexible (i.e., mechanically deformable) and invasive.

As expected, the foregoing experiment identified 15 known kinase tumor suppressors, including the chk2, IKK-a, p38 MAPKs, and DAPK2 genes (Table 3),

demonstrating that the novel screening approach of the present invention is effective at identifying kinase tumor suppressors. The remainder of the list provided in Supplementary Table 3 may be considered to be new potential tumor suppressing kinases. Applying CRISPR-Cas9 Knockout To MASTl And STK4

Genes To Validate The Relationship Between The MASTl

And STK4 Genes And Cell Deformability With increased cutoff criteria, two novel genes, MASTl and STK4, were selected for further study to validate the relationship between the MASTl and STK4 genes and cell deformability . Both the MASTl and STK4 genes showed three independent sgRNAs in the top hits (Fig. 2E) .

Therefore, in another experiment, two isogenic MDA-MB-231 cell lines were generated with sgRNAs against the MASTl and STK4 genes and, consistent with previously obtained screening data, the MASTl and STK4 KO (knockout) cells were transported longer distances along flow chamber 20 of microfluidic device 5 than the wild-type cells (Figs. 3A and 3B) . In other words, the MDA-MB-231 cell lines which were

transformed to knock out MAST1 and STK4 genes were rendered more deformable, and were therefore more readily passed through gaps 40 between adjacent microposts 35 for a greater distance along flow chamber 20 than non-transformed wild-type cells (which were less deformable and therefore mechanically hindered from passing through gaps 40 between adjacent microposts 35). After on-chip separation using microfluidic device 5, the percentages of MAST1 and STK4 KO cells in the output (i.e., collected at outlet 30) increased 1.63-fold and 2.45-fold, respectively (Figs. 3C, 8A, 8B and 8C) . The cells expressing solely FLAG-Cas9 were used as a control in order to rule out the effect of Cas9 expression on cell

transport along flow chamber 20 of microfluidic device 5. These data confirmed the reliability of the cell deformability-based kinase screening approach of the present invention.

Assessing Cell Deformability In STK4 Knockout Cells Cytoskeletal structure plays a major role in cell deformability and it is usually analyzed by measuring the expression of F-actin, cytokeratin 18, and

vimentin. See Eckes, B. et al . J. Cell Sci. Ill, 1897-1907 1998); Sell, M. et al. Nat. Cell Biol. 5, 803-811 (2003); and Gardel, M.L. et al. Science 304, 1301-1305 (2004).

In view of this, another experiment explored how STK4 regulates cell deformability by analyzing the expression and distribution of these three molecules

(i.e., F-actin, cytokeratin 18 and vimentin) . STK4 has been reported to suppress the development of hepatocellular carcinoma through inactivation of the Hippo mediator YAP1. See Zhou, D. et al . Cancer Cell 16, 425-438 (2009). In hematological cancers,

inhibition of STK4 genetically triggers YAPl-mediated apoptosis. See Cottini, F. et al . Nat. Med. 20, 599- 606 (2014).

Three different STK4 knockout breast cancer cell lines were generated, each containing one sgRNA against STK4. The cell lines were then analyzed using microfluidic device 5 (i.e., the cell lines were introduced into flow chamber 20 via inlet 25 and passed through flow chamber 20, with more deformable cells being able to pass more quickly though gaps 40 between microposts 35 so as to arrive at, or closer to, outlet 30 than less deformable cells) . It was found that all three STK4 knockout breast cancer cell lines had longer transport distances along flow chamber 20 of microfluidic device 5 and were present at higher percentages in the output collected from outlet 30 when compared to wild-type cells (Figs. 9A and 9B) . In addition, STK4 appeared to co-localize with F-actin according to confocal imaging (Figs. 10A, 10B and IOC) , and F-actin bundling was reduced in the STK4 depleted cells (Figs. 11A-11C, 12A and 12B) , suggesting that STK4's role in cell deformability may be regulated through the F-actin remodeling. At the same time, obvious co-localization of STK4 with cytokeratin 18 or vimentin was not observed, and depletion of STK4 did not cause obvious change of their distribution (Figs. lOA-lOC and 11A-11D) .

Furthermore, obvious changes in the F-actin,

cytokeratin 18, or vimentin expression was not

detected according to western blotting assay on wild type and STK4 KO cells (Fig. 11D) . Therefore, it can be concluded that the altered distribution and

assembly of the F-actin, rather than the amount of F- actin, are responsible for the increased cell

deformability in STK4 KO cells.

In another experiment, the function loss of STK4 in tumorigenicity was explored by using a non-tumor MCF-IOA breast cell line. Consistent with previous results from prior experiments, STK4 co-localized with F-actin in the MCF-IOA cells (Figs. 3D and 3E) . Upon STK4 depletion, the F-actin bundling also decreased (Figs. 3F and 3G) . Depletion of STK4 yielded an invasive phenotype of the MCF-IOA cells, which has been associated with the up-regulation of multiple genes involved in cancer cell motility and metastasis (Figs. 3H and 13A) . Mammosphere formation assays revealed a significant increase in the size and numbers of mammospheres in the STK4 KO MCF-IOA cells (Fig. 31) . Moreover, the ratio of the

CD44high/CD241ow stem cell-like subpopulation

significantly increased from 1.9% in MCF-IOA control cells to 18% in STK4 KO cells (Fig. 13B) . These results demonstrate that an STK4 deficiency enhances cell invasiveness.

Interestingly, the correlation between STK4 gene expression in breast tumor biopsy and patient survival was analyzed by mining a publicly available database established by Clynes and Bertucci (R2 Genomics

Analysis and Visualization Platform

(http : //r2. amc . nl) ) (Fig. 14). See Clarke, C. et al. Carcinogenesis 34, 2300-2308 (2013); and Sabatier, R. et al. Breast Cancer Res. Treat. 126, 407-420 (2011). Low expression of STK4 in breast cancer samples predicted poor overall patient survival rate,

indicating that STK4 may be a novel breast cancer tumor suppressor.

Conclusions

Thus it will be seen that, with the present invention, a CRISPR approach (i.e., a CRISPR-Cas9 system) may be used to "knock out" a gene of interest, and then the cells may be introduced into a

microfluidic device (e.g., microfluidic device 5) comprising a series of physical barriers (e.g., microposts 35) designed and arrayed so as to sort the cells based on their level of deformability, with the most deformable cells progressing the furthest into the series of barriers (e.g., along flow chamber 20) . Based on the distance to which a cell with a

transfected gene travels, the gene which is "knocked out" may be identified as a high interest candidate for a tumor suppressor gene. Using this technique, 38 potential gene candidates have already been identified whose function may involve the regulation of cell deformability and invasion.

In other words, the results obtained from

experiments performed using the novel methods and apparatus of the present invention show that combining cell mechanical properties-based microfluidic sorting systems (e.g., microfluidic device 5) with CRISPR-Cas9 technologies (i.e., CRISPR-Cas9 kinase knockout screening) is a novel genetic screening strategy that facilitates rapid identification of genes that play roles in mechanical phenotypes, as well as in

physiological and pathological processes. The present invention provides the first "lab-on-chip" rapid screen gene function based on the CRISPR knockout system and opens new avenues for large-scale

integration of on-chip cell function study and search for potential biomarkers.

Methodology 1. Design And Fabrication Of Microfluidic Device 5

The microfluidic pattern for microfluidic device 5 was designed using AutoCAD (Autodesk) . In one preferred form of the invention, the fabricated microfluidic device 5 has one inlet 25, four outlets 30, and a long flow chamber 20 (Fig. 4A) . The

dimensions of flow chamber 20 are 40 mm long, 35 mm wide and 30 μπ\ high. Inlet 25 is connected to a flow chamber 20 having microposts 35 that are 20 μιη in diameter and separated by gaps measuring 20 μπι. The cell separation area comprises 2,000,000 rectangular microposts 35 which are arranged so as to define gaps 40 having widths that decrease from 15 μιη to 6 μιη along the length of flow chamber 20. Microfluidic device 5 is fabricated using standard photolithography and soft lithography procedures. The negative

photoresist SU8-3025 (MicroChem) pattern on the silicon wafer is fabricated with a photomask. The silicon wafer is then silanized with

trimethylchlorosilane (Thermo Scientific) to

facilitate polydimethylsiloxane (PDMS) mold release. PDMS prepolymer (Dow Corning) is poured onto the silicon wafer and cured at 80°C for 1 h. Holes are punched in the PDMS, and oxygen plasma treatment is used to chemically bond the PDMS mold to a glass slide . 2. Operation Of Microfluidic Device 5

icrofluidic device 5 is pretreated with 10% Basal membrane extracts (B E) in phosphate-buffered saline (PBS) for 1 h. BME represents a more

physiological microenvironment than the native PDMS surface. The channels (i.e., flow chamber 20) are then washed with 0.5% bovine serum albumin (BSA) in PBS for 1 h and filled with 0.1% BSA in PBS. BSA coats the surface and prevents nonspecific adhesion of cells to PDMS. A cell slurry (i.e., a slurry of the cells which are to be passed through microfluidic device 5) are loaded into a plastic Tygon tube with a 5-mL syringe and the tube is connected to inlet 25 by a flat steel pin. During flow experiments, a syringe pump is used to control the rate of the fluid flow through microfluidic device 5. In one preferred form of the invention, two different flow rates are used during different experiments in order to assess the separation capability of microfluidic device 5. The lower flow rate of 25 L/min is applied for 15 min to trap cells in microfluidic device b (i.e., within flow chamber 20), and the entire chip is scanned on a Nikon Al confocal microscope with an image stitching

function. Diameters and transport distances of single cells on the chip can then be measured for statistical analysis. A higher flow rate of 75 pL/min is applied for 15 min to flow the flexible cells out of microfluidic device 5 (i.e., out outlet(s) 30). As used herein, the "input cells" and the "output cells" refer to the cell samples that are loaded into

microfluidic device 5 via inlet 25 and the cell samples that were collected from the outlet (s) 30 of microfluidic device 5, respectively. Fluorescence images of the cell mixtures before and after sorting were taken with an Olympus 1X81 inverted fluorescence microscope .

3. Cas9-MDA-MB-231 Generation

Cas9-MDA-MB-231 cells were generated by

lentiviral transduction of a doxycycline-inducible FLAG-Cas9 vector. After 2 days of selection with 2 μg/mL puromycin, single cells were collected with a handheld single-cell pipette. See Zhang, K. et al . J. Am. Chem. Soc . 136, 10858-10861 (2014) . FLAG-Cas9 expression was analyzed for several isolates by western blotting in the presence and absence of 1 μg/mL doxycycline. Subsequently, a single colony with the greatest fold-change in Cas9 expression was chosen for further studies.

4. Lentivirus Library Production And

Transduction

Human Lentiviral sgRNA Library-Kinases used in experiments performed was provided by Eric Lander & David Sabatini (Addgene plasmid # 51044) . The lentivirus library was produced by cotransfecting a pool of CRISPR kinase vectors with the Delta-VPR envelope and CMV VSV-G packaging plasmids into 293T cells. The media was changed 24 h after transfection . The virus-containing supernatant was collected at 48 and 72 h after transfection and passed through a 0.45- μιη filter to remove cells. The Cas9-MDA-MB-231 cell line was transduced at a multiplicity of infection (MOI) of -0.4 with the lentivirus supernatant. The lentiviral CRISPR kinase library contained 5070 sgRNAs . Transduction was performed such that there were at least 500 cells per lentiviral CRISPR

construct . 5. On-chip Screening Using Microfluidic Device 5

Two days after infection, transduced CRISPR kinase-KO cells were selected for screening by growing cells in the presence of blasticidin for 48-72 h.

Selected cells were cultured for 7 days and loaded into microfluidic device 5 for deformability based sorting. In this "on-chip" sorting model, flexible cells with high deformability and metastatic

propensity pass through microbarriers (i.e., through gaps 40 between adjacent microposts 35) and exit microfluidic device 5 by hydrodynamic forces, whereas stiff cells remain trapped. The collected output cells, the flexible ones with high deformability, were stored for three days prior to genomic DNA harvesting. Genomic DNA was also harvested from the input cells with the same operation procedure.

6. Deconvolution And Analysis Of The Screen sgRNAs from the input and the output cell samples were PCR amplified by nested PCR and the resulting PCR products were sequenced on a MiSeq 50 (Illumina) with a single-end 50 bp run. Primer sequences are given in Table 4. Sequencing reads were aligned with the sgRNA library, and the abundance of each sgRNA calculated. sgRNA abundance between the output cells and the input cells was and the fold-enrichment (log2 counts) of the sgRNAs in the flexible cells calculated. If a gene had at least two independent sgRNA hits, each sgRNA with log2Fold-change >=7 and adjust p-value <0.001, it was identified as a "top hit" gene. In the screening strategy used in the experiments, only the most flexible cells (less than 15%) were collected from outlet (s) 30 of microfluidic device 5 for deep

sequencing so that only the most convincing

candidates, which may not have been represented by numerous sgRNAs, would be listed.

7. Generation Of sgRNA-Modified Cell Lines

Individual sgRNA constructs targeting MASTl and

STK4 genes were cloned, lentivirus was produced, and target cells transduced as described above. Cells were selected with blasticidin and cultured in doxycycline for one week before further experimentation .

8. In vitro Matrigel Transwell Invasion Assay Cell invasion was assayed using transwell plates

(Corning) . lxlO⁵ cells in serum-free medium were seeded into the upper chambers of a transwell plate (Matrigel is on the upper surface of the chambers and BME is on the bottom surface of the chambers) , while a complete medium was added to the bottom chambers.

After 48 h, the cells on the upper surface of the membrane were carefully removed and the cells that Lraversed to the opposite side of the membrane were fixed and stained with crystal violet.

9. Mammosphere Formation Assay

Cells were placed into 6-well ultra-low

attachment plates at a density of 4000 viable cells/mL in mammary epithelial cell growth medium containing 2% fetal bovine serum (FBS; Lonza) . Mammospheres were cultured for two weeks and then the numbers of

mammospheres were counted. Mammosphere size was evaluated by optical imaging.

10. Immunostaining, RT-qPCR And Flow Cytometry Cells grown overnight on cover slips were fixed with 4% paraformaldehyde and then permeabilized with a solution of 0.5% Triton X-100 and 300 mM sucrose. Cells were then immune-stained and visualized with an Olympus FVIOOO confocal microscope. Total RNA was isolated using an RNeasy mini kit (Qiagen) . Reverse transcription was performed using an iScript Select cDNA synthesis kit (Bio-Rad) . Real-time PCR was performed using an ABI StepOnePlus real-time PCR system and SYBR Green Master Mix (Invitrogen) . The CD44high/CD241ow cell ratio was determined by labeling the cells with monoclonal anti-CD24-FITC and anti- CD44-PE antibodies (Invitrogen) and by separating them on a BD LSRFortessa cell analyzer.

Modifications Of The Preferred Embodiments

It should be understood that many additional changes in the details, materials, steps and

arrangements of parts, which have been herein

described and illustrated in order to explain the nature of the present invention, may be made by those skilled in the art while still remaining within the principles and scope of the invention.

Claims

What Is Claimed Is:

1. A method for identifying if a target gene may be a tumor suppressor gene, the method comprising: providing a sample of cells wherein some of the cells have been transformed to knock out the target gene and some of the cells have not been transformed to knock out the target gene;

passing the sample of cells through a device which is configured such that cells which exhibit greater mechanical deformability are able to pass further along the device than cells which exhibit lesser mechanical deformability ; and

determining if the cells passing further along the device correlate to the cells which have been transformed to knock out the target gene.

2. A method according to claim 1 wherein the cells which have been transformed are transformed using sgRNA.

3. A method according to claim 2 wherein the sgRNA is delivered into the cells with a CRISPR-Cas9 system.

4. A method according to claim 1 wherein the device comprises a microfluidic device, and further wherein the microfluidic device comprises: a housing having an inlet, a flow chamber

connected to the inlet, and an outlet connected to the flow chamber; and

a plurality of structures spaced laterally across the flow chamber and serially along the flow chamber, wherein the laterally-spaced structures define a plurality of gaps therebetween, and further wherein the gaps between the laterally-spaced structures become smaller along the length of the flow chamber so that cells passing through the gaps are increasingly mechanically deformed along the length of the flow chamber .

5. A method according to claim 4 wherein the structures have a substantially rectangular cross- sectional configuration.

6. A method according to claim 4 wherein the structures have a height of approximately 30 pm.

7. A method according to claim 4 wherein the gaps have a width of between approximately 15 pm to approximately 6 pm.

8. A method according to claim 4 wherein the flow chamber is approximately 40 mm long, 35 mm wide and 30 pm in height.

9. A method according to claim 4 wherein the device comprises a substrate and a cover mountable to the substrate .

10. A method according to claim 9 wherein the structures are formed on the substrate.

11. A method according to claim 1 wherein the target gene encodes a kinase.

12. A method according to claim 11 wherein the kinase comprises a tumor suppressor.

13. A method for identifying genes which may be tumor suppressor genes, the method comprising:

providing a sample of cells and providing a library of agents, wherein each agent is configured to knock out a different target gene, and selectively introducing the library of agents to some of the sample of cells so as to knock out different target genes in the sample of cells;

passing the sample of cells through a device which is configured such that cells which exhibit greater mechanical deformability are able to pass further along the device than cells which exhibit lesser mechanical deformability; and

determining which genes have been knocked out in those cells passing further along the device.

14. A method according to claim 13 wherein the agents comprise sgRNA.

15. A method according to claim 14 wherein the sgRNA is delivered into the cells with a CRISPR-Cas9 system.

16. A method according to claim 13 wherein the device comprises a microfluidic device, and further wherein the microfluidic device comprises:

a housing having an inlet, a flow chamber

connected to the inlet, and an outlet connected to the flow chamber; and

a plurality of structures spaced laterally across the flow chamber and serially along the flow chamber, wherein the laterally-spaced structures define a plurality of gaps therebetween, and further wherein the gaps between the laterally-spaced structures become smaller along the length of the flow chamber so that cells passing through the gaps are increasingly mechanically deformed along the length of the flow chamber .

17. A method according to claim 16 wherein the structures have a substantially rectangular cross- sectional configuration.

18. A method according to claim 16 wherein the structures have a height of approximately 30 pm.

19. A method according to claim 16 wherein the gaps have a width of between approximately 15 pm to dppruximately 6 pm.

20. A method according to claim 16 wherein the flow chamber is approximately 40 mm long, 35 mm wide and 30 pm in height.

21. A method according to claim 16 wherein the device comprises a substrate and a cover mountable to the substrate.

22. A method according to claim 21 wherein the structures are formed on the substrate .

23. A method according to claim 13 wherein the target genes encode kinases.

24. A method according to claim 23 wherein the kinases comprise tumor suppressors .

25. A system for identifying if a target gene may be a tumor suppressor gene, the system comprising: a sample of cells wherein some of the cells have been transformed to knock out the target gene and some of the cells have not been transformed to knock out the target gene;

a device which is configured such that cells which exhibit greater mechanical deformability are able to pass further along the device than cells which exhibit lesser mechanical deformability;

an introducer for delivering the sample of cells into the device; and

an analyzer for determining if the cells passing further along the device correlate to the cells which have been transformed to knock out the target gene.

26. A system according to claim 25 wherein the introducer comprises a syringe.

27. A system according to claim 25 wherein the analyzer comprises a DNA sequencer.

28. A system for identifying genes which may be tumor suppressor genes, the system comprising:

a sample of cells wherein different target genes have been knocked out in some of the cells;

a device which is configured such that cells which exhibit greater mechanical deformability are able to pass further along the device than cells which exhibit lesser mechanical deformability;

an introducer for delivering the sample of cells into the device; and an analyzer for determining which genes have been knocked out in those cells passing further along the device .

29. A system according to claim 28 wherein the introducer comprises a syringe.

30. A system according to claim 28 wherein the analyzer comprises a DNA sequencer.

31. Novel apparatus for separating cells, the apparatus comprising a microfluidic device, wherein the microfluidic device comprises:

a housing having an inlet, a flow chamber

connected to the inlet, and an outlet connected to the flow chamber; and

a plurality of structures spaced laterally across the flow chamber and serially along the flow chamber, wherein the laterally-spaced structures define a plurality of gaps therebetween, and further wherein the gaps between the laterally-spaced structures become smaller along the length of the flow chamber so that cells passing through the gaps are increasingly mechanically deformed along the length of the flow chamber.

32. Apparatus according to claim 31 wherein the structures have a substantially rectangular cross- sectional configuration.

33. Apparatus according to claim 31 wherein the structures have a height of approximately 30 pm.

34. Apparatus according to claim 31 wherein the gaps have a width of between approximately 15 pm to approximately 6 pm.

35. Apparatus according to claim 31 wherein the flow chamber is approximately 40 mm long, 35 mm wide and 30 pm in height.

36. Apparatus according to claim 31 wherein the device comprises a substrate and a cover mountable to the substrate .

37. Apparatus according to claim 36 wherein the structures are formed on the substrate.