An Endostatin-Lentivirus (ES-LV)-EPC Gene Therapy Agent for Suppression of Neovascularization in Oxygen-Induced Retinopathy Rat Model

doi:10.21203/rs.2.16501/v3

Background: Transplantation of gene transfected endothelial progenitor cells (EPCs) has provided novel methods for tumor neovascularization therapy but not for ocular disease therapy. This study aimed to investigate the efficacy of endostatin transfected EPCs in retinal neovascularization therapy.

Results: Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) showed the high expression of endostatin in endostatin-lentivirus-EPCs. The neovascularization leakage area and the number of preretinal neovascular cell nuclei were significantly decreased in the endostatin-lentivirus and endostatin-lentivirus-EPC groups, and the effects of these two treatments on inhibiting retinal neovascularization were almost the same. These two groups also showed the greater retinal distribution of endostatin. Intravitreal injections of endostatin-lentivirus-EPCs inhibited retinal neovascularization, vascular endothelial growth factor (VEGF) and CD31 expression, and increased endostatin expression in vivo. Endostatin-lentivirus-EPCs targeted and prevented pathologic retinal neovascularization.

Conclusions: Gene-combined EPCs represent a potential new therapeutic agent for the treatment of neovascular eye diseases.

General Cell Biology & Physiology

Molecular Biology

Retinal neovascularization

Endothelial progenitor cells

Gene therapy

Endostatin

Retinal neovascularization is a severe complication in most types of retinopathy, such as proliferative diabetic retinopathy, retinopathy of prematurity, and age-related macular degeneration [1]. It is the first leading cause of vision impairment and irreversible blindness today. Although conventional therapies for retinal neovascularization include surgical vitrectomy, laser photocoagulation, photodynamic therapy, and intravitreal injection of anti-VEGF factors, which are widely available. The prognosis for retinal neovascularization remains extremely poor, and retinal neovascularization recurrence rates remain high; only a few patients achieve good vision without recurrences [1-3]. This outcome can be traced back to the finding that neovascular tissue has a distinct ability to secrete growth factors such as vascular endothelial growth factor (VEGF) and to infiltrate the blood-retinal barrier or to disrupt the extracellular matrix [4]. Thus, it eventually leads to retinal neovascularization recurrence following multiple initial treatments [4-5]. Therefore, novel therapeutic methods for patients with retinal neovascularization are required for improving therapy outcomes.

Previous studies indicated that anti-VEGF agents can inhibit ocular neovascularization through intravitreal injections [6-9]. Bevacizumab and ranibizumab have been reported to decrease optic disc edema or to have an anti-inflammatory and anti-neovascular effect on neovascular age-related macular degeneration [6-8]. However, expensive charges, frequent office visits, and multiple injections are added to the burden of patients with this condition. The injections are also associated with a low risk of an increase in intraocular pressure, and the incidence of vitreous hemorrhage, uveitis, vascular occlusion, or the retinal detachment is elevated [9-11]. Worse, neovascularization may reappear when the therapy is over because the effect of a single injection of anti-VEGF agents is temporary [12].

Continuous suppression of neovascularization may be more efficacious than monthly injections of anti-VEGF [13]. Moreover, there is a big difference in patients’ responses to treatment. Approximately 10% of patients do not respond to anti-VEGF therapy in spite of receiving monthly intravitreal injection therapy for 2 years [8, 12]. Therefore, based on the advantages of gene therapy, the local and sustained delivery of anti-angiogenic molecules is feasible, and it has been confirmed in animal models that neovascularization can be efficiently suppressed by gene therapy [13-15]. In the present study, the endostatin gene was selected due to its profound effects on angiogenesis. Firstly, endostatin is an angiogenesis inhibitor that inhibits endothelial cell (EC) proliferation, migration or invasion, blocks the formation of new blood vessels, and decreases retinal VEGF expression [16]; secondly, endostatin can be secreted by many cells but has no effect on the blood vessels around normal tissue [17]. Endostatin is non-toxic and has no drug resistance [16-17]. Although the amount of endogenous endostatin increases in proliferative diabetic retinopathy, this increase is not enough to inhibit retinal neovascularization [3]. Therefore, it is important for the amount of endostatin expression to be increased in vivo, and gene therapy enables this.

Endothelial progenitor cells (EPCs), which exist primarily in bone marrow, can migrate from blood circulation to ischemic or neovascular sites, have a high proliferative rate, and differentiate into ECs [18]. Under normal physiological conditions, healthy EPCs can be applied to repair ischemic vascular damage [19, 20]. However, under the pathological condition of neovascularization, the amount of EPCs increases, but their biological function is not improved; hence, the EPCs cannot repair the vascular endothelium. The mobilized EPCs may promote the formation of new blood vessels [21-23], so the transplantation of healthy EPCs is essential [24].

EPCs represent an effective delivery vehicle for gene therapy against neovascular formation by virtue of their mobilization [24-25]. The potential for EPCs to serve as cellular vehicles for molecular therapy against neovascularization depends on efficient and specific gene transfer and the ability to stably deliver therapeutic loads through the blood to the intended target [25]. It has been reported that the angiogenic gene (VEGF) transfected EPCs migrate to and increase the blood supply to sites of vascular injury [26]. Therefore, it can be inferred that the transplantation of endostatin-lentivirus -EPCs to retinal tissue may not only provide a sufficient number of healthy EPCs to the target area but also may inhibit the VEGF expression level and the migration of ECs via endostatin expression of endostatin-lentivirus-EPCs. The high VEGF level in the neovascular regions could be fundamentally solved through this strategy. Thus, the development of retinal neovascularization could be improved or reversed.

The aims of the present study were as follows: to establish a novel therapeutic modality using EPCs as vehicles to target neovascular vessels and to secrete endostatin; to investigate whether intravitreal injection of endostatin-lentivirus-EPC is effective for treating retinal angiogenesis in an animal model of oxygen-induced retinopathy (OIR) and to test our hypothesis; and to evaluate the feasibility of this technique in clinical therapy for retinal neovascularization.

Fluorescent images of EPCs

There was no difference between EPCs transduced with endostatin-lentivirus-GFP and EPCs transduced with lentivirus-GFP. No fluorescence was observed in non-transduced EPCs (Figure 1).

Endostatin expression in EPCs

The total RNA was extracted from EPCs using TRIzol® reagent (Invitrogen; Carlsbad, CA, USA), according to the manufacturer’s instructions. The straps were cleared at 28S rRNA and 18S rRNA on agarose gels. The value of optical density (260: 280 nm) was 1.9-2.2. EPCs transduced with a lentiviral vector encoding endostatin-GFP resulted in endostatin overexpression. The endostatin mRNA expression in the endostatin overexpression group increased significantly (P<0.001), as compared with that in the NC group. However, there was no difference in mRNA expression between the blank control and NC groups (Figure 2).

Fluorescent images of EPCs

There was no difference between EPCs transduced with endostatin-lentivirus-GFP and EPCs transduced with lentivirus-GFP. No fluorescence was observed in non-transduced EPCs (Figure 1).

Endostatin expression in EPCs

The total RNA was extracted from EPCs using TRIzol® reagent (Invitrogen; Carlsbad, CA, USA), according to the manufacturer’s instructions. The straps were cleared at 28S rRNA and 18S rRNA on agarose gels. The value of optical density (260: 280 nm) was 1.9-2.2. EPCs transduced with a lentiviral vector encoding endostatin-GFP resulted in endostatin overexpression. The endostatin mRNA expression in the endostatin overexpression group increased significantly (P<0.001), as compared with that in the NC group. However, there was no difference in mRNA expression between the blank control and NC groups (Figure 2).

Inhibition effects of endostatin-lentivirus on retinal neovascularization

Fundus fluorescein angiography results: A capillary non-perfusion zone and neovascularization leakage were observed in the experimental group but not in the control group (Figure 3A).

Compared with the blank control group (age-matched rats kept in normoxia with non-intravitreal injection), the amount of retinal neovascularization leakage significantly increased in the OIR+NC group (empty-lentivirus injection) on days 1, 3 (P<0.01), and 5 (P<0.001) but did not increase in the OIR +endostatin-lentivirus group (endostatin-lentivirus injection) at the above three time points. Compared with the OIR+NC group, the amount of retinal neovascularization leakage significantly reduced on days 3 (P<0.01) and 5 (P<0.001; Figure 3B, 4). Therefore, it can be inferred that retinal neovascularization leakage was inhibited in the OIR+ endostatin-lentivirus group and that endostatin-lentivirus played a role in inhibiting retinal neovascularization leakage.

Hematoxylin-eosin (HE) results: In comparison with the blank control group, the nuclei in the vascular endothelium increased significantly in the OIR+NC group (P<0.01) but not in the OIR+ endostatin-lentivirus group. Compared with the OIR+NC group, the nuclei in the vascular endothelium decreased significantly in the OIR+ endostatin-lentivirus group (P<0.05; Figure 6, 7A), so it can be inferred that retinal neovascularization was inhibited in the OIR+ endostatin-lentivirus group. The HE results agreed with the fundus fluorescein angiography results that endostatin-lentivirus helped in inhibiting retinal neovascularization leakage.

Effects of simple EPCs on retinal neovascularization

Fundus fluorescein angiography results: The amount of retinal neovascularization leakage significantly increased at each time point (P<0.05) in the OIR+EPC group (EPCs injected), as compared with the blank control group; but there were no significant differences at different time points (hour 1 and days 1, 3, and 5), as compared with the OIR+NC group (Figure 3B, 4). Therefore, it can be inferred that simple EPCs have no effect on inhibiting retinal neovascularization leakage or promoting neovascularization.

HE results: Compared with the blank control group, the nuclei in the vascular endothelium increased significantly in the OIR+EPC group (P<0.01); however, there was no significant difference between the NC and OIR+EPC groups (Figure 6, 7A), so it can be inferred that simple EPCs have no effect on inhibiting retinal neovascularization leakage. In addition, simple EPCs could not promote neovascularization.

Inhibition effects of endostatin-lentivirus-EPCs on retinal neovascularization

Fundus fluorescein angiography results: In comparison with the blank control group, the amount of retinal neovascularization leakage significantly increased in the OIR+NC group (empty-lentivirus injection) on days 1, 3 (P<0.01), and 5 (P<0.001) but did not increase significantly in the OIR +endostatin- lentivirus-EPC group (endostatin-lentivirus-EPCs injection) at each time point (hour 1 and days 1, 3, and 5); as compared with the OIR+NC group, the amount of retinal neovascularization leakage significantly decreased on day 5 (P<0.001; Figure 3B, 4). Hence, it can be inferred that retinal neovascularization leakage was inhibited in the OIR+ endostatin-lentivirus-EPC group. Although endostatin-lentivirus-EPCs play a role in inhibiting retinal neovascularization leakage, inhibition occurs a little later than it does for endostatin-lentivirus.

HE results: Compared with the blank control group, the nuclei in the vascular endothelium increased significantly in the OIR+NC group (P<0.01) but did not increase in the OIR +endostatin-lentivirus-EPC group. The nuclei in the vascular endothelium decreased significantly in the OIR+ endostatin-lentivirus-EPC group (P<0.05; Figure 6, 7A) in comparison with the OIR+NC group, so it can be inferred that retinal neovascularization was inhibited in the OIR +endostatin-lentivirus-EPC group. The HE and fundus fluorescein angiography results were in agreement.

Comparison between endostatin-lentivirus and endostatin- lentivirus-EPCs in inhibition effects on retinal neovascularization

Fundus fluorescein angiography results: The neovascularization leakage area significantly reduced in the OIR +endostatin-lentivirus and OIR+ endostatin-lentivirus-EPC groups on day 5 (P<0.001; Figure 4). The size of this leakage area and the inhibiting effects on retinal neovascularization were similar between the two groups (Figure 5), so it can be inferred that retinal neovascularization significantly reduced in both groups.

HE results: The nuclei in the vascular endothelium significantly reduced and were similar between the OIR+ endostatin-lentivirus and OIR+ endostatin-lentivirus-EPC groups (P<0.05; Figure 7A, 7B). The results of the HE method complied with those of fundus fluorescein angiography. It can be surmised

that the inhibiting effects on retinal neovascularization were similar between endostatin-lentivirus and endostatin-lentivirus-EPC groups.

Immunohistochemistry (IHC) results of endostatin, VEGF, and CD31 expression

IHC revealed that the expression of endostatin presented mainly on the retinal nerve fiber layer (RNFL), ganglion cell layer (GCL) and inner plexiform layer (IPL); the positive staining of this method is brown color (Figure 8). As compared with the blank control group, the level of endostatin expression decreased significantly in the OIR+NC group (P<0.05); as compared with the OIR+NC group, the endostatin expression level increased significantly in the OIR+ endostatin-lentivirus and OIR+ endostatin-lentivirus-EPC groups (P<0.01; Figure 9), but there were no significant differences between the OIR+ endostatin-lentivirus and OIR+ endostatin-lentivirus-EPC groups (Figure 10). The results showed that endostatin was overexpressed in the OIR+ endostatin-lentivirus and OIR+ endostatin-lentivirus-EPC groups, so it seemed that the overexpression of endostatin promoted the inhibition of retinal neovascularization. The results accorded with the fundus fluorescein angiography and HE results that retinal neovascularization was inhibited in these two groups.

IHC results (Figure 8) showed that the VEGF expression existed mainly on RNFL, GCL, IPL, the inner nuclear layer (INL), outer plexiform layer (OPL) and the outer nuclear layer (ONL). Compared with the blank control group, VEGF expression significantly increased in the OIR+NC group (P<0.05), but no significant differences were seen among the OIR+ endostatin-lentivirus, OIR+EPC, or OIR+ endostatin-lentivirus-EPC groups (Figure 9). As compared with the OIR+ endostatin-lentivirus group, no differences were seen in the OIR+EPC or OIR+ endostatin-lentivirus-EPC groups (Figure 10). Based on the results, VEGF expression did not increase in the OIR+ endostatin-lentivirus, OIR+EPC or OIR+ endostatin-lentivirus-EPC groups, so it appears that endostatin overexpression in the OIR+ endostatin-lentivirus and OIR+ endostatin-lentivirus-EPC groups may have decreased the VEGF expression. At the same time, the simple EPCs may have played a role in repairing the unhealthy neovascular tissues, which would explain why VEGF expression did not increase in the OIR+EPC group.

According to the IHC results, CD31 expression existed mainly on the RNFL, GCL, IPL, as well as the INL, OPL and ONL, which accorded with the VEGF expression findings (Figure 8). Compared with the blank control group, CD31 expression significantly increased in the OIR+NC and OIR+EPC groups (P<0.05), and no significant differences were seen in the OIR+ endostatin-lentivirus and OIR+ endostatin-lentivirus-EPC groups. However, CD31 expression decreased significantly in the OIR+ endostatin-lentivirus group (P<0.05; Figure 9), as compared with the OIR+NC group. Compared with the OIR+ endostatin-lentivirus group, no significant differences were observed in the OIR+ endostatin-lentivirus-EPC group (Figure 10). CD31 expression did not increase in the OIR+ endostatin-lentivirus or OIR+ endostatin-lentivirus-EPC groups, which was similar to the findings for VEGF expression. These IHC results complied with the fundus fluorescein angiography and HE results, and CD31 served as a biomarker of ECs (as a part of neovascularization), it can be inferred that retinal neovascularization was inhibited in the OIR+ endostatin-lentivirus and OIR+ endostatin-lentivirus-EPC groups.