Effective Treatment of Ocular HSK with a Human Apolipoprotein E Mimetic Peptide in a Mouse Eye Model

Abstract

purpose. To evaluate efficacy of the small apolipoprotein E (apoE) mimetic dimer peptide (apoEdp) in the treatment of herpetic stromal keratitis in a mouse ocular model and determine its therapeutic effects against HSV-1–induced inflammatory cytokines.

methods. Female C57Bl/6 mice were corneally infected with HSV-1 strain KOS-GFP; topical treatment was initiated 24 hours after infection and continued for 10 consecutive days. Treatment groups were 1% apoEdp, 1% trifluorothymidine (TFT), and phosphate-buffered saline (PBS). The incidence and severity of stromal keratitis were monitored by slit lamp examination in a masked fashion. Infectious HSV-1 titer in eye swabs and alteration in inflammatory cytokines were determined in the early postinfection period by real-time RT-PCR.

results. One percent apoEdp treatment, which significantly reduced the incidence and severity of HSK, was equal to the effect of 1% TFT; both groups had significantly lower incidence and severity than the placebo treatment group. The in vivo mouse ocular model results of apoEdp therapy correlated with accelerated clearance of virus from eye swabs. Topical 1% apoEdp treatment in mice significantly downregulated gene expression of mouse proinflammatory cytokines.

conclusions. These results suggest that topical treatment with apoE peptide has efficacy against HSK through anti-HSV-1 and anti-inflammatory activities.

Herpetic stromal keratitis (HSK) is a chronic, immunoinflammatory disease of the cornea caused by HSV-1. The current antiviral and anti-inflammatory therapies fail in approximately 33% of patients with necrotizing HSK, ^{1 2} resulting in the necessity of corneal transplantation. Currently, topical trifluorothymidine (TFT, 1% wt/vol) is the most widely used and effective topical antiviral for treating HSV-1 corneal infections. However, TFT has been reported to be cytotoxic with long-term use. ³ The therapeutic agents that are currently available consist of a small number of nucleoside analogues. A nucleoside analogue, acyclovir (ACV) as ointment has demonstrated efficacy against superficial herpes keratitis but has not been approved in the United States because of side effects associated with ointment therapy. ⁴ Therefore, the development of new drugs with reduced toxicity and different mechanisms of action is needed. 

Human apolipoprotein E (APOE [gene]; apoE [protein]) is a 34-kDa glycoprotein. In humans, APOE is polymorphic with three major allelic types: ε2, ε3, and ε4. The most common allelic type is APOE ε3 (77%–78%) followed by APOE ε4 (14%–15%), and APOE ε2 (7%–8%), which is the least frequent genotype. ^{5 6} The apoE ε4 genotype has also been strongly implicated as a risk factor in Alzheimer’s disease ^{7 8 9 10 11} and in viral infections such as HSV-1 ¹² and HIV. ¹³ APOE ε4 carriers have an increased risk of cold sores caused by HSV-1 and genital herpes caused by HSV-2. ¹⁴ The molecular basis by which apoE exerts its isoform-specific effects on the outcome of infection is unknown. The potential mechanisms responsible for the influence of apoE allotype on HSV-1 infection are (1) the alteration of viral transport through hematogenous routes, ¹⁵ (2) inflammatory modulation during infection, ¹⁶ (3) repair of tissues after infection, ¹⁷ and (4) a direct antiviral effect in vitro. ¹⁸  

Evidence suggests that apoE ε4 gene carriers are more susceptible to ocular herpes and other viral infections and inflammatory diseases than carriers of the apoE ε3 isoform. ^{19 20} While ApoE ε4 is not a disease-causing gene, it has been shown to be a susceptibility gene for these inflammation-associated diseases. The hypothesis that the apoE ε3 allele could be anti-inflammatory, whereas apoE ε4 is proinflammatory, has resulted in a search for therapeutic agents that mimic the apoE role and act to reduce or block this increased inflammation, thereby reducing or blocking the course of the diseases. It has been well established that carriers of the apoE ε4 gene are more susceptible to viral infections and inflammation-related diseases than their apoE ε3-carrying counterparts. ^{12 13 14 19} Therefore, we reasoned that apoE ε3 gene products could manifest potent anti-inflammatory and antiviral action against HSK. 

Both HSV-1 and apoE can bind to the extracellular matrix (often heparan sulfate proteoglycans, HSPGs) ^{21 22 23} and then to specific cellular receptors. ²⁴ Both apoE and HSV have been hypothesized to compete for the same binding sites or other receptors. ²¹ One receptor binding region of apoE, which has been critical in its biological activity, is located at residues 142-147 (the heparin-binding domain) and mediates the attachment of apoE to cellular HSPGs (Fig. 1) . ^{24 25 26} A tandem-repeat dimer peptide, apoEdp, derived from the apoE residues 141-149, has been reported to exhibit antiviral activity against HSV-1, HSV-2, and HIV in vitro. ¹⁸ The tandem-repeat design could reflect an increased adaptation of an α-helical structure and better stability. ^{27 28} This peptide (apoEdp) has not been tested in vivo against any infectious disease. 

Materials and Methods

Mice

All experimental procedures were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the LSUHSC Institutional Animal Care and Use Committee. Female C57Bl/6 mice (Charles River Laboratories Inc., Wilmington, MA), 5 to 6 weeks of age, were used. 

Cells and Virus

CV-1 cells (American Type Culture Collection, Manassas, VA) were propagated in Eagle’s minimum essential medium (EMEM) containing 0.15% HCO₃ supplemented with 10% fetal bovine serum (FBS), penicillin G (100 U/mL), and streptomycin (100 mg/mL). HSV-1 strain KOS-GFP ²⁹ was used and titered in CV-1 cells. 

Ocular Infection

Before HSV-1 inoculation, mice were anesthetized by intraperitoneal administration of xylazine (6.6 mg/kg of body weight) and ketamine (100 mg/kg). The eyes were scarified in a 2 × 2 cross-hatch pattern and inoculated with 5 × 10⁵ plaque forming units (PFU) of virus (in a volume of 4 μL) in each eye. 

Quantification of Viral Titer from Eye Swabs

Plaque assays to quantify infectious HSV-1 in tears were performed using CV-1 cells as indicator cells. Briefly, eye swabs were collected using sterile filter paper strips and placed in 1 mL of cold EMEM containing 10% FBS. Swab samples were serially diluted and plated on CV-1 cells for 1 hour at 37°C. Finally, the medium was aspirated, EMEM was added to each well, and viral plaques in each well were quantified after 2 days. 

Treatment Groups

The apoE mimetic peptide was synthesized (Genemed, Arlington, TX) with a purity of greater than 95%. The 18 amino acid (Ac-LRKLRKRLLLRKLRKRLL-amide) tandem-repeat dimer peptide (apoEdp) was derived from the apoE receptor binding region between residues 141 and 149 (Fig. 1) . In vivo treatments started 24 hours after viral infection (24 hours PI) and continued for 10 consecutive days. Before starting topical treatment, all mice eyes were checked for corneal fluorescence (Fig. 2A)as a measure of HSV-1 replication. A subjective score of 0 to 4 per eye was recorded: 0 indicates no fluorescence; 1, approximately 1 quadrant of the corneal circle; 2 half circle; 3, three quadrants; and 4, full circle. Mice were placed into balanced treatment groups based on the corneal fluorescence score. Topical treatment consisted of a five times daily application of 1% apoEdp solution made in PBS applied as 1 drop of 5 μL per eye. Control groups received either 1% TFT (Falcon Pharmaceuticals, Fort Worth, TX) or placebo treatment with PBS. All treatments were applied in a blind fashion. Before the start of the experiment, we determined the safety and tolerance of the peptide in the eyes of naïve mice. We found that up to 5% of this peptide concentration applied five times a day for 10 consecutive days was well tolerated. No ocular toxicity was observed. 

Extraction of RNA, Reverse Transcription, and Real-Time PCR

Total cellular RNA was isolated from corneas (RNeasy Mini Kit; Qiagen, Santa Clara, CA), as specified by the manufacturer. For in vivo mRNA analysis, at 24 hours after treatment, three mice (six corneas) from each group were killed. The corneas were harvested and were placed immediately in preservative (RNA-later; Qiagen). The RNA sample was reverse transcribed into cDNA by using a high-capacity cDNA reverse-transcription kit (Applied Biosystems [ABI], Foster City, CA). Two microliters (100 ng) of each cDNA sample were added to 10 μL of 2× PCR reaction buffer (Applied Biosystems) and 50 ng each of sense and antisense primer in a final reaction volume of 20 μL. Quantitative real-time PCR was performed in the following conditions: 95°C for 3 minutes, followed by 35 cycles of 95°C for 10 seconds, 60°C for 30 seconds, and a melting curve at 60 to 95°C at a heating rate of 0.5°C per second followed by cooling. The primer sequences for target genes are given in Table 1 . 

Real-time PCR was performed (MyIQ; Bio-Rad, Hercules, CA) using SYBR Green I reagent (Bio-Rad) according to the manufacturer’s protocol. Relative mRNA quantitation was calculated using the 2^−ΔΔCt method ³⁰ by normalizing the value of the target gene for each sample to its endogenous housekeeping gene (mouse β-actin) and then normalizing these values to a baseline sample, called the calibrator. The average of naïve mice cornea (scarified but not infected) samples served as the calibrator. 

Slit Lamp Examination and Quantitation of Corneal Opacity and Neovascularization

After corneal HSV-1 infection, the eyes were monitored with a slit lamp microscope (Eye Cap; Haag-Streit International, Mason, OH) in a masked fashion. HSK severity was quantified by measuring two clinical parameters, corneal opacity, and neovascularization. Corneal opacity was graded on a scale as follows: 0, no opacity; 1, mild cloudiness with visible iris; 2, moderate cloudiness with obscured iris; 3, total corneal cloudiness with invisible iris; and 4, total opacity with no posterior view. Mice with corneal opacity scores of ≥0.5 and limbal sprouting with a neovessel length of ≥0.1 mm were deemed positive for incidence determinations. Aided by an eye piece reticule, we scored neovascularization as the area of neovascularization using the following formula A = (C × 0.4 × L × π)/2, where C is clock hours of neovascularization; 30° arc is 1 (360° circle with a total of 12); L is the length of the longest neovessel (length of longest neovessel of a mouse eye was 1.6 mm); and, A is the area of neovascularization. 

Statistics

(1) The therapeutic efficacy against HSV-1–induced corneal disease (opacity and neovascularization) and HSV-1 titer in the eyes were expressed as the mean ± SEM and considered significant at P values < 0.05, as determined by the Student’s t-test. (2) Results of gene expression analysis by quantitative real-time RT-PCR were reported as the mean ± SEM. The magnitude of differences between groups was evaluated by using a two-way analysis of variance (ANOVA). ³¹ The independent factors were the drug treatments and the genes being assayed; the interaction of these two terms was included in the model. Separation of treatment by gene interaction levels was performed with t-tests on the least-square means derived from the ANOVA model. ³² Alpha levels were adjusted for the number of comparisons by using a simulation-based method. ³³ The significance criterion for the adjusted α levels was the standard P < 0.05, which was the overall experiment-wise α level for the entire set of comparisons conducted. 

Results

Treatment of the corneal lesions with apoEdp began 24 hours PI and continued for 10 consecutive days, significantly reducing the incidence and severity of HSK. All mice received topical application of the assigned drug 1 day after ocular infection. Treatment continued five times daily through day 10 PI. Clinical evaluation of HSK severity was monitored for progression of corneal opacity and neovascularization (Figs. 2B 2C) . Corneal opacity was evident in PBS-treated mice starting on day 7 with a gradual progression until day 21 PI (Fig. 3A) . The peptide-treated mice had significantly less (P < 0.05) corneal opacity, mimicking 1% TFT-treated mice (Fig. 3A) . In both apoEdp- and TFT-treated groups, significant (P < 0.05) inhibition of neovascularization was detected throughout the experiment (Fig. 3B) . In the PBS-treated group, neovascularization continued up to day 21 PI. PBS-treated mice eyes had a 60% incidence of corneal opacity and a 90% incidence of corneal neovascularization compared with the 10% incidence (corneal opacity and neovascularization) in both apoEdp- and TFT-treated mice. Our results show that treating HSV-1 corneal lesions beginning 24 hours after infection and continuing for 10 consecutive days with apoEdp effectively blocked the incidence and severity of corneal opacity and neovascularization (Figs. 3A 3B) . 

A direct antiviral role of apoEdp was observed through topical application to the mouse cornea. To determine whether diminished severity of HSK correlated to better virus control and a decrease in viral shedding, we examined the titers from tear film in HSV-1–infected eyes. Viral titers of the tear film from apoEdp- and TFT-treated eyes were significantly (P < 0.05) lower than in mock-treated eyes examined on days 4 and 6 PI (Fig. 4) . 

Reduced corneal disease in apoEdp-treated mice correlates with downregulated expression of mouse proinflammatory cytokines. IL-1α, IL-1β, IL-6, TNF- α, IFN-γ, and VEGF produced as a consequence of corneal HSV infection serve as potential molecules to induce the early inflammatory process. Selected groups of treated mice were killed 24 hours after treatment (48 hours PI), and their corneas were analyzed for gene expression at mRNA specific to proinflammatory cytokines known to be upregulated in mouse corneas after HSV-1 infection (IL-1α, IL-1β, IL-6, TNF- α, IFN-γ, and VEGF). Treatment with apoEdp significantly reduced proinflammatory cytokine gene expression at 24 hours after treatment (48 h PI) compared with mock-treated eyes, as revealed by real-time reverse transcription-PCR, similar to the result of TFT treatment (Table 2) . 

Discussion

We reported the effect of human apolipoprotein E genotype on the pathogenesis of HSK. ²⁰ In that study, using transgenic mice, we found increased pathogenesis of ocular herpes in human apoE isoforms ε4 knockin compared with isoform ε3 knockin. We have also reported that mouse apoE affects survivability and HSV-1 DNA load in trigeminal ganglia in the highly neurovirulent HSV-1 strain 17Syn+ through an ocular route of infection. ³⁴ Mouse apoE did not appear to have a role in ocular virus titer and acute corneal disease of epithelial keratitis. ³⁴  

This study is the first to demonstrate the beneficial effect of human host-derived peptidomimetic apoE (apoEdp) in an in vivo mouse model of HSK. ApoEdp blocked the development and progression of HSK severity, as demonstrated by slit lamp, virological, and immunologic evaluations. 

From our observations, apoEdp has the potential for anti-viral effects because viral shedding in the tears of infected mice eyes was significantly reduced at 4 and 6 days PI in the apoEdp-treated eyes compared with mock-treated eyes. This result was similar to mice treated with 1% TFT. 

The mechanism by which apoEdp suppresses stromal keratitis is not clear. ApoE is known to compete for one of the receptors, cell-surface heparan sulfate; therefore, if the apoE occupied these sites, the result would be reduced HSV uptake (Fig. 5) . ^{22 23} One could speculate that the amount of apoE protein expressed at the surface of the corneal epithelium is insufficient to occupy the HSPG receptor and inhibit viral infection. This could be further supported by the fact that high concentrations (greater than 100 μmol) of apoEdp were needed to inhibit in vitro virucidal assay. ¹⁸ Dose-dependent antiviral effects in vitro suggest that a direct antiviral effect may be responsible and may depend on the α-helical structure of apoEdp. ¹⁸ ApoE 141-149 as a monomer does not form an α-helical structure. However, a tandem-repeat dimer (dimer of apoE 141-149) of 18 acids does form an α-helical structure similar to the native apoE 141-149 sequence within full-length apoE. 

ApoEdp has also previously been shown to interfere with the earliest stages of viral infection, preventing viral attachment and exerting a mild virucidal action in vitro as determined by plaque reduction and virucidal assay, respectively. ¹⁸ That report ¹⁸ of in vitro mild virucidal action mimics our in vivo results. A significant inhibition of virus titer at days 4 and 6 PI was observed in apoEdp-treated eyes compared with that in mock-treated eyes. However, 1% TFT showed faster clearance of virus than did apoEdp; no virus was detected in TFT-treated eyes at day 4 and beyond. Based on our results, the antiviral action of apoEdp could be explored further by using a combination therapeutic trial of apoEdp and TFT to determine a possible synergistic or additive action. 

An important in vivo effect of apoEdp treatment observed in our study was the reduction of the mRNA specific to inflammatory cytokines in the cornea. This observation is novel and may be related to the amino acid sequence of the heparin-binding domain (142-147) that overlaps another 17 amino acid-peptidomimetic apoE monomer (133-149) known to have anti-inflammatory activity. ³⁷ ApoE 133-149 is recognized as a macrophage receptor-binding site and has an anti-inflammatory role in closed head injuries and a traumatic brain injury model ³⁷ but has not been tested in any infectious disease model. Whether the decrease in the inflammatory response of the cornea observed by us in HSV-1–infected eyes treated by apoEdp is the result of inhibition of macrophages requires further investigation. However, our study of the proinflammatory cytokines IL-1 α, IL-1β, IL-6, TNF α, and IFN-γ, and one proangiogenic cytokine, VEGF, found a significant reduction in apoEdp-treated eyes compared with mock-treated eyes. A previous report suggests that antagonizing the effect of IL-1 by a specific receptor antagonist protein abrogates the cascade of events that culminate in HSK, ³⁸ supporting our findings of downregulated IL-1α and -1β expression related to attenuated corneal pathology. In the HSV-1–infected cornea, IFN-γ can regulate neutrophil invasion. ^{39 40} TNF-α induces corneal fibroblasts and epithelial cells to synthesize and secrete IL-6. ⁴¹ IL-6 is important in the recruitment of neutrophils into the HSV-1–infected cornea. ⁴² In this study, both IL-6 and TNF-α were significantly reduced after apoEdp and TFT treatment compared with mock treatment. From the perspective of corneal neovascularization, VEGF is a specific mitogen for vascular endothelial cells that is produced by a variety of cell types, including activated macrophages. ⁴³ In humans, alternate mRNA splicing of the VEGF gene product gives rise to four different VEGF isoforms ⁴⁴ including VEGF₁₆₅ (equivalent of mouse VEGF₁₆₄ isoform) that binds heparan sulfate. ⁴⁵ VEGF₁₆₅ must interact with HSPG at the cell surface for this proangiogenic factor to bind and signal through its respective receptor (Fig. 5) . ^{35 36} We suggest that blocking the HSPG receptor with exogenous apoEdp may have inhibited ocular angiogenesis in our study. 

Our in vivo results suggest that apoEdp could be exploited therapeutically, on its own or in combination with current conventional chemotherapy, to treat HSK. 

 Supported in part by National Eye Institute Grants NEI R01 EY06311 (JMH), F32EY016316 (DMN), NEI EY002672 (HEK), NIH NEI P30 EY002377 (LSU Eye Center Core Grant); a Research to Prevent Blindness (RPB) Senior Scientific Award (JMH); LSUHSC Translational Research Initiative Grant (PSB); and an unrestricted departmental grant from RPB.

 Submitted for publication March 26, 2008; revised May 13, 2008; accepted August 11, 2008.

 Disclosure: P.S. Bhattacharjee, None; D.M. Neumann, None; T.P. Foster, None; C. Clement, None; G. Singh, None; H.W. Thompson, None; H.E. Kaufman, None; J.M. Hill, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Human apoE consists of 299 amino acid residues. The heparin binding domain is located between residues 142 and 147 in the receptor binding region of apoE, which mediates its attachment to cellular HSPGs. ^{24 25 26} ApoE dimer tandem repeat peptide (apoEdp) is reported to contain anti-HSV-1 activity in vitro. ¹⁸

Figure 2.

 

Representative photographs of (A) HSV-1 KOS replication (GFP expression) in the cornea at 1 day PI. (B) ApoEdp-treated and (C) mock-treated eyes on day 7 PI.

Table 1.

 

Primer Sequences Used for Real-Time RT-PCR Assay

Gene	Primer	Sequence
Mouse IL-1α	Forward	5′-ATG GCC AAA GTT CCT GAC TTG TTT
	Reverse	3′-CCT TCA GCA ACA CGG GCT GGT C
Mouse IL-1β	Forward	5′-ATG GCA ACT GTT CCT GAA CTC ACC T
	Reverse	3′-CAG GAC AGG TAT AGA TTC TTT CCT TT
Mouse IL-6	Forward	5′-ATG AAG TTC CTC TCT GCA AGA GAC T
	Reverse	3′-CAC TAG GTT TGC CGA GTA GAT CTC
Mouse TNF-α	Forward	5′-TTC TGT CTA CTG AAC TTC GGG GTG ATC GGT CC
	Reverse	3′-GTA TGA GAT AGC AAA TCG GCT GAC GGT GTG GG
Mouse IFN-γ	Forward	5′-TGA ACG CTA CAC ACT GCA TCT TGG
	Reverse	3′-CGA CTC CTT TTC CGC TTC CTG AG
Mouse VEGF	Forward	5′-GCG GGC TGC CTC GCA GTC
	Reverse	3′-TCA CCG CCT TGG CTT GTC AC
Mouse β-actin	Forward	5′-GTG GGC CGC TCT AGG CAC CAA
	Reverse	3′-CTC TTT GAT GTC ACG CAC GAT TTC

Figure 3.

 

C57Bl/6 mice corneally infected with HSV-1 KOS were treated topically with 1% apoEdp; 1% TFT or PBS was used in the control eyes. Topical application was begun on day 1 PI and continued through day 10 PI, five times daily. Eyes were examined with a slit lamp to assess the therapeutic efficacy against HSV-1–induced corneal opacity and neovascularization. Results are expressed as the mean ± SEM and are considered significant, with P < 0.05 determined by Student’s t-test. (A) Corneal opacity. ApoEdp topical treatment of HSV-1–infected eyes completely blocked the development of corneal opacity. In the control groups, TFT-treated mice had mild corneal haze that was observed on day 7 PI and was not seen after that. In the mock-treated eyes, a gradual progression of corneal opacity was observed on days 7 to 21 PI (B). ApoEdp treatment significantly (P ≤ 0.05) reduced corneal neovascularization compared to mock treatment. The results are expressed as the mean ± SEM.

Figure 4.

 

Detection of infectious HSV-1 in the eyes after topical treatment of 1% apoEdp from 1 to 10 days PI. Five mice (per group) per time point were swabbed to collect tear fluid at days 2, 4, and 6 PI, and the infectious HSV-1 titer results were determined. On days 4 and 6 PI, a significant reduction (P ≤ 0.05) in the ocular HSV-1 titer was seen in the eyes treated with apoEdp compared with mock-treated eyes. Treatment with 1% TFT cleared virus faster than 1% apoEdp. On day 4 PI and beyond, no virus was detected in TFT-treated eyes.

Table 2.

 

Cytokine mRNA Changes in Mock-Treated Eyes and Those Treated with apoEdp and TFT

Cytokines	Normalized (Mouse β-actin) Cytokine mRNA Expression Relative to Naïve Cornea
		Mock-Treated	1% apoEdp-Treated	1% TFT-Treated
IL-1α	78 ± 15.54	1.53 ± 0.99	0.35 ± 0.1
IL-1β	14.65 ± 1.7	2.8 ± 1.32	0.65 ± 0.26
IL-6	12.88 ± 5	0.4 ± 0.16	0.18 ± 0.1
IFN-γ	11 ± 2.77	0.55 ± 0.27	0.22 ± 0.1
TNF-α	9.03 ± 1.39	0.33 ± 0.05	0.22 ± 0.1
mVEGF	2.45 ± 0.27	0.26 ± 0.09	0.01 ± 0

 Data are the mean multiples of change (x-fold) in mRNA ± SEM.

Figure 5.

 

Possible mechanism of interaction among apoE, HSV, and VEGF. ApoE, ^{24 26} HSV, ²³ and VEGF ^{35 36} compete for binding to the HSPG molecule on the cell surface. ApoEdp treatment may block the binding of VEGF and HSV.

The authors thank Maxine Simpson and Cheryl C. Vega for technical assistance.