2.63" x 3.5"

Slide 1

Dear Colleagues, First, I would like to thank our host Dr. Timothy Wang for the opportunity to present my work. We study inflammation in the eye, particularly in the cornea and the limbus (slide 1), which is a joint region of the cornea and the sclera. We are more interested in the initiation of the inflammatory response since downregulation of the inflammatory response at the early step may prevent its further development. I would like start my presentation from a short background and please do not hesitate to interrupt me if you have questions. The healthy cornea is avascular. Physical damage or infection provokes the inflammatory response. It appears in angiogenesis, edema and conjunctival vasodilation. Cornea neovascularization is not completely reversible. While new blood vessels rapidly loose their filling, they never degrade completely. Some of them become ?ghost vessels? and can be filled up again as well as extended to new capillary network. Since the cornea has to stay penetrative to the light, presence of blood vessels seriously impairs the vision.

Slide 2

The inflammatory response in the cornea is tissue-specific. While it is regulated by a variety of mediators including eicosanoids, platelet-activated factor, cytokines and serine proteases only eicosanoids were found in the inflamed eye shortly after the damage. Our studies (slide 2) demonstrated presence of two major eicosanoids, 12(R)-HETE and 12-(R)-HETrE in the inflamed cornea. 12(R)-HETE is an inhibitor of Na/K-ATPase, it causes relaxation of blood vessels and lowers intraocular pressure (IOP). 12(R)-HETrE is a powerful proinflammatory and proangiogenic factor. Its presence and appearance in the cornea straightly correlated with the severity of inflammatory response. Moreover, both eicosanoids contribute to the recruitment of the immune cells. All these facts point 12-hydroxyeicosanoids as obvious candidates for initiation of the inflammatory response.

Slide 3

The arachidonic acid-derived eicosanoids can be synthesized in three enzymatic pathways (slide 3). Cyclooxygenases convert arachidonic acid to prostaglandins (PGs), prostacyclin (PGI2) and thromboxane (TxA2). Lipoxygenases make leukotrienes (LTs), hydroxyeicosatri- and tetraenoic acids (HETEs and HETrEs). Cytochromes P450 (CYP) monooxygenases synthesize epoxyeicosatrienoic acids (EET), dihydroxyeicosatrienoic acids (DHETs), HETEs and HETrEs. Since the production of 12-hydroxyeicosanoids depends on the presence of NADPH and their stereospecificity is different from lipoxygenase- derived HETEs, we proposed that those compounds are synthesized by Cytochrome P450.

Slide 4

We hypothesized (slide 4) that injury to the cornea that caused a damage of cell membrane leads to a release of free arachidonic acid. That induces the expression and activity of a Cytochrome P450 and elevates levels of 12-HETE and 12-HETrE. After secretion from the cell, 12-hydroxyeicosanoids spread around the eye with tear fluid and appear in the corneosclera where they interact with vascular endothelium of the adjacent limbal vessels and initiate the inflammatory response causing edema, vasodilation and neovascularization.

Slide 5
With this project, we wanted to achieve the following specific aims:

Slide 6

At work, we usually follow few simple principles: "Multiple approaches" means that this work has been done at three different kinds of objects: in vitro, ex vivo and in vivo models and involved multiple techniques starting from cell culturing and finishing by monitoring of transfected animals (slide 6). "Closed cycle"- cell cultures that we used in this work were our source of preliminary results. Then we tested the obtained results in vivo and ex vivo. Besides, animals served us as a source of cells. "Follow up strategy"- we studied the phenomenon of ocular surface inflammation in its development following it from the step of initiation to wound healing. "Dynamic performance" means step-by-step approach the optimal conditions in combination with three time repeat of each principal experiment.

Slide 7

We isolated limbal microvascular cells from the rabbit eye (slide 7). Briefly, we cut the cornea and isolated the rim by removing the central part of the cornea. Then we placed rims on the Matrigel surrounding hydron pellet with absorbed 12(R)-HETrE by each rim. After polymerization of the gel, we incubated rims for 5 days until they developed intensive pseudocapillary network. Then, we scooped pseudocapillaries and dissolved Matrigel with dispase. We washed the cells, spin them down and selected with the specific medium.

Slide 7a

The experimental design included starvation of semiconfluent cells in basal medium followed by the introduction 12(R)-HETrE of the desired concentration for a certain period of time. Incubation of cells with inhibitors usually preceded the exposure to 12(R)-HETrE. After the treatment, we collected the medium, prepared cell homogenates or isolated total RNA depending on the goal of the particular experiment.

Slide 8
We found was that 12(R)-HETrE induced expression of VEGF in the time- and concentration-dependent manner. I would pay your attention that maximum of the stimulation was achieved in 30-45 min of the treatment at very low concentration of 12(R)-HETrE, approximately 0.1 nM. These very low numbers pointed 12(R)-HETrE as one of the most powerful signaling molecules.

Slide 9
We checked if treatment with 12(R)-HETrE had affected RNA stability and incubated the cells in presence of 12(R)-HETrE and actinomycin D (slide 9). In the following slide, you can see that 12(R)-HETrE had a moderate stabilizing effect on VEGF mRNA increasing its half-life from 9 h to 14h.

Slide 10

VEGF induction by 12(R)-HETrE occurred due to an activation of MAPK kinases. (slide 10). 12(R)-HETrE transiently activated, both isoforms of Erk: Erk1 and Erk2 (panel A). Levels of phosphorylated Erks peaked at 5 min (panel B) and then, it gradually decreased to the control levels. Immunoprecipitation with the specific antibody (panels D and E) also showed phosphorylation of Elk, the specific substrate of Erk. Besides, level of phosphorylated Elk paralleled the increase in phosphorylation of Erk. The observed effect was sensitive to the inhibition of Erk (panel C). The preincubation of cells with ERK specific inhibitor, PD98,059, significantly abolished the induction of VEGF. Lane 1 in this panel was our positive control, the recombinant phosphorylated Elk. Lane 2 was a negative control, the cells treated with vehicle; At lane 3, cells were activated with 12(R)-HETrE and preincubation with PD abolished VEGF induction.

Slide 11
Induction of VEGF gene followed by the synthesis of VEGF protein that we found in the media of treated cells (slide 11). We observed a progressive secretion of VEGF in a time- and dose ?dependent manner with maximum of the secretion at 0.1-1.0 nM.

Slide 12
Secretion of VEGF dramatically affected the behavior of endothelial cells. (slide 12). Unlike cells treated with vehicle, cells treated with 12(R)-HETrE at the concentration as low as 1 pM demonstrated a strong ability to form a pseudocapillaries on extracellular Matrix. The response was concentration- depended with the maximal stimulation at 1 nM.

Slide 13
Importantly, the addition of anti-VEGF antibodies to the incubation medium greatly attenuated pseudocapillary formation (slide 13). That indicated that 12(R)-HETrE effect on the cells was mediated, at least in part, by VEGF.

Slide 14
That is probably the best way to summarize the data (slide 14): As you see:

Slide 15
In the other words (slide 15):

Slide 16

After the discovery of RLME cells as the cells targeted by 12(R)-HETrE and 12(R)-HETrE signaling, we wanted to identify the enzyme responsible for the synthesis of 12(R)-hydroxyeicosanoids and prove its role in the inflammatory response. Previously we reported of CYP 4A1- like mRNA, which level increased after exposure to the hypoxia. Surprisingly, our study led to the isolation of the DNA fragment with 98.8% sequence homology to the rabbit lung CYP 4B1. In this slide, you can see our strategy for isolation of the promoter from the genomic DNA. Briefly, we digested genomic DNA by four different restrictases and ligated each DNA library with provided commercial adapter. Sequence of adapter included sequence of adapter primer (AP1) that we used for the primary PCR. We performed PCR with the original gene-specific primer (GSP1) and AP1. Since AP1 binding site appeared in PCR products after the first cycle of amplification, it improved our chances for a specific amplification. Then we performed secondary PCR with nested primers (AP2 and GSP2), which were located inside the amplified PCR product. Again, one of the primers, GSP2 was a gene-specific primer. Another primer, AP2 had a binding site inside the adaptor. Then we cloned the amplified PCR products in pBluescript and sent them for sequencing.

Slide 17
For the full-length cDNA, we isolated total RNA from rabbit corneas to hypoxia. After ligation with GeneRacer oligonucleotide, we performed the RT-reaction and amplified the obtained cDNA with GeneRacer primer and gene-specific primer. Then, we cloned the amplified PCR products to pBluescript and sent them for sequencing.

Slide 18
For the experiments with the transfection, we cloned CYP 4B1 cDNA into pIRES2 vector that already carried EGFP gene. Unlike non-transfected cells (slide 18), the corneal epithelial cells transfected with EGFP and CYP 4B1 emit green fluorescence. We estimate the efficiency of cell transfection as 90%.

Slide 20
In 24h hours after transfection, we collected the media and extracted the lipids with ethyl acetate. Then, we purified the extracted samples with HPLC and analyzed them with GC/MS. The GC/MS-analysis (slide 20) confirmed the presence of 12(R)-HETrE in cell culture medium after the transfection with EGFP/CYP 4B1. Cells transfected with EGFP only had sufficiently less 12(R)-HETrE in the medium.

Slide 19

At the same time point, we also analyzed the cells for their ability to produce 12(R)-HETrE. First, we incubated the cells in the presence of NADPH and ^14C-labeled arachidonic acid and collected the medium. We extracted the lipids with ethyl acetate and analyzed the samples for a presence of radioactive arachidonic acid metabolites. We found (slide 19) that unlike other cells only cells transfected with EGFP and CYP 4B1 gained the ability to metabolize ^14-C-arachidonic acid to a radioactive compound with the retention time on the chromatogram close to 12(R)-HETrE.

Slide 21

From the experiments on the cells we switched to experiments on animals and performed a transfection of rabbit eye (slide 21). Rabbits were transfected into the left eye with pIRES2 vector carrying EGFP or EGFP plus CYP4B1. The right, non-transfected eye we used as a control. Briefly, we dipped needle into plasmid solution and repeatedly inserted it into the limbus over 360°. This followed by topical application of the remaining plasmid solution. The development of the inflammatory response was monitored by slit lamp microscopy for up to 7 days after the transfection. Some rabbits were sacrificed in 4 days after the transfection. Corneas were dissected in four pieces and placed on Matrigel for ex vivo angiogenic activity assay. In parallel, medium was collected and level of 12-HETrE was assessed by GC/MS analysis.

Slide 22
After transfection with the EGFP and 4B1 (slide 22), the ocular surface becomes progressively inflamed over a 7-day period. This inflammatory response was characterized by dilation of limbal vessels and beginning of neovascularization. In contrast, the eyes transfected with EGFP only showed a little if any limbal dilation or neovascularization sprouting.

Slide 23

Using fluorescent microscopy (slide 23) we clearly showed that our method of the transfection effectively transduced EGFP expression into the cornea. Moreover, EGFP expression was primarily associated with the corneal epithelial cell layer and was sustained in the cornea for 6 days after the transfection. The EGFP fluorescence (slide 24) was also evident in corneas from eyes transfected with EGFP and 4B1.

Slide 24

Corneal explants (slide 24) taken from eyes transfected with EGFP and 4B1 demonstrated a significantly higher angiogenic activity compared to corneal explants from the sham transfected eyes and from the eyes transfected with the control EGFP plasmid. To examine the origin of pseudocapillary cells (slide 25), we performed a binding assay with lectin I from Grifonia simpliconica labeled with rhodamine. In this slide you can see that rhodamine conjugated lectin I specifically bound to pseudocapillary cells indicating that the capillary network is composed of endothelial cells.

Slide 25

In addition to significant inflammatory response in vivo, CYP4B1 transfection markedly increased the angiogenic activity of corneal-limbal explants. As you can see in this slide (slide 25), angiogenic activity measured as total capillary length was about 3 fold higher in explants obtained after transfection with 4B1 than the control explants. Moreover, transfection with CYP4B1 also resulted in a marked production of 12(R)-HETrE by the corneal explants. In fact, 12(R)-HETrE level in samples from 4B1 transfected corneas significantly exceeded the levels of 12(R)-HETrE in the controls suggesting a relationship between the angiogenic activity and the production of 12-HETrE.

Slide 26

Amazingly, the ability of explants to form pseudocapillaries depended on CYP metabolism. In this slide (slide 26), three curves represent three groups of explants. All of them have been obtained from the animals transfected with EGFP/CYP 4B1 and cultured on the Matrigel. Addition of 17-ODYA, the specific CYP inhibitor sufficiently decreased the pseudocapillaries formation compare to control. However, if we removed the inhibitor on the day #3, the explants develop the network of same intensity as in control suggesting the cause relationship between the transfection, synthesis of12(R)-HETrE and pseudocapillaries development.

Slide 27

In summary;

READ THE SLIDE or SAY THIS:

Transfection of the cornea with EGFP or EGFP-4B1 plasmids resulted in expression of EGFP in corneal epithelial cells.

In the eyes transfected with plasmid containing the 4B1, transfection was accompanied by increased inflammation in vivo, as well as 3-fold increase in capillary growth ex vivo and neovessel sprouting composed of endothelial cells. This was associated with increased levels of 12(R)-HETrE, the angiogenic eicosanoid formed by CYP4B1.

The results implicate the corneal CYP4B1 as a component of the inflammatory cascade initiated by the injury of the ocular surface.

I would thank my mentor Dr. Schwartzman and all colleagues at NYMC for a help in my work.

.