Method for selective quantification of immune and inflammatory cells in the cornea using flow cytometry

The cornea serves as a protective surface against the environment (i.e., allergens, pollutants, desiccation and microorganisms) and promotes vision, made possible by corneal transparency. This protocol describes corneal preparation for flow cytometry to assess cells localized in the cornea. Our model details the process, from determining how many corneas are needed in the experiment to corneal excision to digestion and staining of the cornea cells. The simplicity of the model allows for systematic analysis of different corneal mechanisms of immunity, inflammation, angiogenesis and wound healing. In corneal transplantation, residential immune and inflammatory cells are key to the mechanisms that underlie angiogenesis, opacity, and graft rejection. In addition, this model can also elucidate cellular mechanisms mediating corneal graft outcomes and wound healing. Lastly, this model can be used to analyze the efficacy of new medications such as instillation and subconjunctival injections and assess the potential of therapeutic molecules to enhance graft survival and wound healing in vivo.


BACKGROUND
The ocular surface is continuously exposed to environmental agents such as allergens, pollutants, desiccation and microorganisms [1]. The smooth wet surface of the cornea is the major refractive surface of the visual system [2], made possible by corneal transparency. When this environment collapses due to infection or inflammation, the normally transparent and avascular cornea develops opacification and neovascularization [3,4], resulting in decreased visual acuity. Dry eye disease induces ocular damage by the inflammatory cascade from innate and adaptive immune responses [5][6][7]. In the corneal wound healing process, opacity occurring due to corneal stromal fibroblasts and myofibroblasts is problematic [8]. Corneal transplantation is performed to address the corneal opacity. However, high-risk recipient corneas with angiogenesis experience acute rejection in greater than 40%-90% of cases [3,9].
The cornea can be observed optometrically and is an excellent choice for analyzing the efficacy of new medications such as instillation and subconjunctival injection [10,11]. Clinical and experimental studies are investigating the use of hyaluronic acid and fibronectin for wound healing [12][13][14], steroids and immunosuppressants [15,16], and even anti-VEGF drugs for neovascularization, inflammation and rejection [17][18][19][20]. In wound healing, immune cells migrate along neovessels in the transparent corneal tissue [6,21,22], so controlling the corneal environment is important. Immunostaining, PCR, and western blotting can be used to confirm immune cells and inflammatory cells localized in the cornea, but flow cytometry is the most effective method of quantification of inflammatory cells in the cornea. Much research on corneal flow cytometry has been done so far [9,23]. Previously, studies have shown that immune cells circulate in the cornea and cervical lymph nodes to control the immunocompetence of the cornea [24]. Furthermore, immune cells localized in the cornea affect rejection, angiogenesis, and wound healing [8,25]. However, since the

Cornea harvesting
1. Punch out the corneal limbus with a 30 G needle (Fig. 1A). In parallel to the iris, punch outside the limbus and reduce the sclera as much as possible.

2.
Cut the cornea circularly from the puncture wound at the corneal limbus with Vannas scissors (Fig. 1B).

Peel the iris from the back of the cornea under a microscope.
Cornea digesting 4. Add DNase І (0.002 g) and collagenase D (0.004 g) in 1 ml of RPMI-1640 to a 1.5 ml tube (ideally up to 4 corneas per tube) (Fig. 1C).

6.
Add 200 μl of FBS to the tube to stop the DNase І and collagenase D, and add the tube contents to a cell strainer in a 6-well plate with 5 ml RPMI for filtering (Fig. 1E).
Single cell suspension 7. Mash the corneas using the plunger of a 5 ml syringe (Fig. 1F).

8.
Add filtrate to the FACS tube (Fig. 1G). Wash the filter and dish with 5 ml RPMI-1640 and transfer to the FACS tube.
Flow cytometry 9. Centrifuge the FACS tube at 1300 RPM for 10 min. Pour and discard the supernatant.
10. Add 2 ml of FACS buffer (1% BSA with PBS) and centrifuge at 1300 RPM for 10 min. Pour and discard the supernatant.
POL Scientific protocol 11. Add 1 μl of Fc receptor blocking antibody to each sample and keep the samples at 4°C on the shaker for 30 min.

12.
Add the surface antibodies (i.e., CD45 and CD4) directly to each sample, and keep the samples at 4°C for 1-2 h.
13. Add 2 ml of FACS buffer to each sample and centrifuge at 1300 RPM for 10 min. Pour and discard the supernatant.

15.
Results were analyzed using FlowJo software.

17.
Dead cells should be excluded from further analysis.  Figure 2A shows the total number of cornea cells and passenger leukocytes obtained, counted using binocular microscope. When using between one and four corneas, the number of single suspended cells can be collected as expected. However, loss of single suspended cells was noted at eight corneas (See Troubleshooting). The frequencies of live/ dead cells were shown in Figure 2B. Among the cells obtained after doublets discrimination, over 90% were live cells between one and four corneas, however the dead cells were increased at eight corneas. Therefore, the ideal number of corneas to be used with the amount of lysis reagent in our protocol is four corneas. Based on this result, it is possible to set up an experiment plan corresponding to the required number of cells. POL Scientific protocol Our model allows for detailed analysis of immune pathways of interest, including the Th1 immune response, regulatory effect of T cells, pro-inflammatory and tolerogenic capacity of antigen presenting cells, and the role of innate and adaptive immunity in angiogenesis [9,23]. We also recommend this protocol for assessing the changes in cornea cells in the wound healing process. Moreover, this model can be used with knockout and transgenic mice to evaluate various molecular pathways participating in the immune response, angiogenesis and wound healing processes.

ANTICIPATED RESULTS
Other models for assessing cornea cells have been used previously and may complement the model proposed in this paper. Real-time PCR and western blot analysis can be used for quantification of corneal cells, but it is difficult to accurately measure the limited amount of RNA and protein in individual corneas, and the aforementioned methods are not effective for residential cells that require multiple staining. Figure 3 shows CD45 + leukocytes in corneal cells induced by inflammation [23,26]. Our protocol enables the detection of dynamic trends of localized cells in the cornea during wound healing.
In this model, you can detect dynamic changes in local corneal cells used in conjunction with other models such as the eye drop challenge, medical injections to the eye, wound healing and corneal transplantation model. Previously, we published the protocol for the high-risk corneal transplantation model [27]. Briefly, inflamed and neovascularized (highrisk) host beds (BALB/c) were created by three intrastromal sutures placed into the central cornea using 11-0 nylon sutures (AB-0550S, MANI, Tochigi, Japan) 14 days before corneal transplantation to induce host beds prone to reject the grafts. Mice with unmanipulated (low-risk) host beds served as controls. For allogeneic corneal transplantation, C57BL/6 corneas were grafted onto BALB/c host beds. Figure 4 shows CD4 + T cells in corneal cells post-transplantation. We can detect recruited CD4 + T cells post-transplantation and demonstrate significantly increased CD4 + T cells in high-risk inflamed corneal transplantations compared to low-risk corneal transplantations.
Overall, the simplicity, availability, and the cost-effectiveness of the described model provide a unique platform to study basic principles of immunity and angiogenesis for ophthalmic pathologies.  Here we list potential issues and provide troubleshooting instructions. If few cells can be collected, melting time may be too short or the mesh time for the single cell suspension may be insufficient. Consider increasing the dissolution time and mesh time. Figure 5 shows the number of cells obtained when the cell suspension was performed with only 1 ml of RPMI. If the RPMI is not sufficient for the single cell suspension, cells may remain in the mesh. Cell loss may occur, a problem that worsens as the number of corneas increases. Therefore, we recommend preparing 5 ml or more of RPMI when doing a single cell suspension in a 6 well plate.  If single cells cannot be isolated due to too many cornea samples, adjust the amount of DNase І and collagenase D as necessary. Figure  6 shows that the number of cells increases with the number of corneas up to four corneas. However, beyond four corneas, cell loss occurs.
Since the number of cells is falsely increased if iris samples are included (Fig. 6), peel off the iris as much as possible when collecting the cornea.
Possible problems and their troubleshooting solutions are listed in Table 1.