Herpes viruses are among the most well-studied microorganisms that cause chronic infection and are a well-described cause of ocular diseases (89,102). Ocular degeneration due to infection with herpes simplex type 1 (HSV-1) and herpes simplex type 2 (HSV-2) is a leading cause of blinding keratitis in industrialized countries (102,103). Analysis of ocular viral isolates demonstrated that HSV-1 is responsible for about 85% of ocular HSV infections (102,103). Ocular HSV-1 infection consists of an acute keratoconjunctivitis followed by a lifelong cycle of latency, reactivation, and recurrent infection (102,103). This section reviews research on herpetic eye infection and the use of iontophoresis to deliver therapeutic doses of antiviral agents and/or agents that cause experimental reactivation of latent HSV-1.
a. Antivirals for Treatment of Epithelial Keratitis Hill et al. (104) were the first to report the transdermal delivery of antiviral agents by iontophoresis. They demonstrated that the antiviral agents iododeoxyuridine (IDU), phosphonoacetic acid (PPA), and vidarabine monophosphate (Ara-AMP) could be iontophoresed across mouse epidermis with enough efficiency to achieve long-lasting and potentially therapeutic concentrations of drug.
A subsequent study (105) examined the pharmacokinetics of ionto-phoresed Ara-AMP. Ara-AMP was chosen because the parent compound, vidarabine, was shown to be effective in the treatment of HSV-1 infection after topical or systemic administration. Ara-AMP is phosphorylated and highly charged and thus an excellent candidate for iontophoresis. Tritiumlabeled Ara-AMP was applied either topically or iontophoresed into unin-fected rabbit eyes. Transcorneal iontophoresis was performed at 0.5 mA for 4 minutes with the cathode in contact with the drug solution. Iontophoresis resulted in drug concentrations in the cornea, aqueous humor, and iris which were 3-12 times higher than those obtained with topical application. No obvious corneal damage was observed. The results showed that iontophoresis increased corneal penetration of an antiviral agent in comparison to topical application.
Kwon et al. (106) were the first to demonstrate that iontophoresed Ara-AMP retained its antiviral properties. They studied the virucidal effects of iontophoretically applied Ara-AMP on experimental HSV keratitis in rabbits infected with HSV-1 McKrae strain. Transcorneal iontophoresis of Ara-AMP (0.3 M, 3.4%) was done in both eyes at 24, 48, and 72 hours postinoculation. A second group of infected rabbits served as a control for the procedure and were iontophoresed with a 0.9% sodium chloride solution. Additional groups of rabbits received either topical 10% Ara-AMP, 0.5% IDU, or 0.9% sodium chloride five times daily for 4 days. Slit-lamp examination was performed daily for 10 consecutive days after initiation of treatment and the severity of the disease was scored. Mean lesion scores for the eyes treated by iontophoresis of Ara-AMP were significantly lower (i.e., less severe disease of shorter duration) compared to the scores of eyes treated with iontophoresis of sodium chloride or topically with Ara-AMP, IDU, or sodium chloride.
b. Antivirals for Treatment of Stromal Disease Hill et al. (107) extended these studies by comparing the efficacy of transcorneal iontophoresis versus intravenous injection of either acyclovir or Ara-AMP (alone or in combination) for the treatment of HSV-1 stromal infection in rabbits. Herpetic infection was achieved by an intrasomal injection of purified HSV-1 McKrae strain. Treatments (iontophoresis of either Ara-AMP, acyclovir, or sodium chloride; or intravenous infusion of acyclovir or sodium chloride) were initiated 24 hours later. Two ophthalmologists performed slit-lamp examination daily for up to 22 days postinoculation and recorded the severity of the disease.
Iontophoresis itself did not have any effect on the severity of the disease. Iontophoresis of acyclovir or Ara-AMP significantly shortened the duration of the disease compared to iontophoresis of sodium chloride, and iontophoresed Ara-AMP was as effective as iontophoresed acyclovir. Acyclovir significantly reduced the severity of the disease whether it was delivered iontophoretically or by intravenous injection. Intravenous administration of acyclovir resulted in significantly higher levels of drug in ocular tissues than that obtained with iontophoresis, but iontophoresis delivered a dose that achieved the same therapeutic effect. These results suggested that iontophoresis of acyclovir or Ara-AMP (with or without intravenous supplementation) could be a valuable treatment option for patients with HSV-1 stromal keratitis.
c. Adrenergic Agents for Experimental Reactivation of HSV-1 in the Study of Ocular Herpetic Disease Primary HSV infection is an acute inflammatory response. While the acute infection is occurring, HSV travels along neuronal axons to regional sensory ganglia where it enters a state of dormancy (i.e., latency) (108-110). The exact stimulus that triggers HSV reactivation is unknown, but certain environmental factors (stress, irradiation, hypo- or hyperthermia) result in the appearance of infectious virus at the site of primary infection and a recurrence of the disease (110113). The rabbit eye model has allowed investigators to study the pathogenesis of HSV latency and reactivation (102,114). In this model, viral latency can be established by inoculation of the cornea with 17Syn + , McKrae, or other strains of HSV-1 (102,115,116). Viral reactivation can be induced by iontophoresis of adrenergic agents and monitored by viral shedding on the ocular surface (102,116,117).
Clinicians had noted for some time the relationship between physical stress and the appearance of fever blisters (cold sores) or genital herpetic lesions. This suggested a role for epinephrine in the process, and various investigators showed that topical or intravenous applications of epinephrine could cause reactivation of HSV-1 in rabbits. Researchers from Hill's group (118,119) were the first to report that transcorneal iontophoresis of this agent could induce viral shedding in rabbits harboring latent HSV-1. They inoculated rabbits with HSV-1 and 60 days later performed epinephrine iontophoresis on one eye from each rabbit. A 0.01% epinephrine solution was iontophoresed (0.8 mA for 8 min) once daily for 3 consecutive days. Viral shedding was verified by ocular swabbing and culturing the tear film in cell culture. Antigenic and/or nucleic acid analysis was used to confirm the identity of the viral isolate. Epinephrine iontophoresis resulted in HSV-1 viral shedding in all treated eyes (118). A follow-up report from this group (119) validated these results and demonstrated that the HSV-1 titers rose, peaked, and fell with time after iontophoresis. This reliable and efficient means of inducing viral shedding served as the basis for the development of other animal models for the study of herpetic disease (120-123).
These models have been used to study the kinetics, pathogenesis, and molecular biology of HSV-1 latency, reactivation, and recurrence of clinical disease.
A subsequent study from this group (124) characterized the appearance of HSV-1 in the nerves and ganglia after reactivation by epinephrine iontophoresis. Hill et al. (124) showed that the presence of infectious virus could be detected more rapidly from cultured neuronal tissue from rabbits that had undergone epinephrine iontophoresis than from their untreated counterparts.
Shimomura et al. (125) modified the epinephrine iontophoresis protocol so as to eliminate the need to anesthetize animals once daily for 3 consecutive days. They developed a method that was based on a single iontophoresis of 6-hydroxydopamine followed by topical application of epi-nephrine over several days. The rationale for this came from studies using iontophoresed 6-hydroxydopamine to treat glaucoma. As described above, 6-hydroxydopamine causes a selective and reversible degeneration of sympathetic nerve terminals in the anterior segment, after which the innervated structures of the eye are exquisitely sensitive to extremely dilute solutions of epinephrine. With this method, viral shedding was induced by a single iontophoresis of a 1% solution of 6-hydroxydopamine under various conditions (0.5-0.75 mA for 3-8 min). Two drops of 2% epinephrine were applied 6 hours after iontophoresis and twice daily for the next 4 days. All treated eyes (17/17; 100%) shed HSV-1, regardless of the iontophoretic conditions.
Hill et al. (115,126) used this modified procedure to further characterize ocular HSV-1 shedding induced by 6-hydroxydopamine iontophoresis plus topical epinephrine. They were the first to quantify the number of plaque-forming units of HSV-1 shed into the tear film following 6-hy-droxydopamine iontophoresis plus topical epinephrine-induced reactivation (126). Titers ranged up to 105 plaque-forming units/eye. A subsequent study (115) showed that dipivefrin hydrochloride could be used in place of epi-nephrine in this reactivation model. Dipivefrin hydrochloride is a prodrug of epinephrine, which has increased corneal penetration compared to epinephr-ine. In clinical studies, 0.1% dipivefrin hydrochloride was as effective as topical 1 or 2% epinephrine. Hill et al. (115) demonstrated that iontophor-esed 6-hydroxydopamine followed by topical application of 0.1% dipivefrin hydrochloride resulted in HSV-1 ocular shedding and recurrent HSV-1 corneal lesions. This was the first report showing that an adrenergic drug could induce both HSV-1 ocular shedding (reactivation) and HSV-1 corneal epithelial lesions (recurrence) in rabbits harboring latent virus.
Rivera et al. (110) used epinephrine iontophoresis to study the temporal relationship between viral reactivation and the presence of viral particles or virions in corneal nerves or the cornea, respectively. Transmission electron microscopy revealed viral particles (in low abundance) in unmyeli-nated axons, but no enveloped virions were found. This study suggests that ocular iontophoresis of epinephrine reactivates HSV-1 in the ganglia and that the virus is translocated from the sensory ganglia (trigeminal and/or superior cervical) to the cornea by anterograde axonal transport mechanisms.
Hill et al. (117) demonstrated that levo(-)epinephrine was significantly more potent than dextro( + )epinephrine for inducing HSV-1 ocular shedding. The data suggested that the mechanism of induction of HSV-1 ocular shedding by epinephrine is correlated with the receptor potency of levo(-)epinephrine. Epinephrine activates both a- and p-adrenergic receptors in the eye. If the signal for viral reactivation was transmitted exclusively through a p-adrenergic receptor, a p-adrenergic receptor antagonist should inhibit epinephrine-induced reactivation.
Studies from Hill's group (127-129) assessed the effect of topically applied timolol maleate, a nonspecific p1,p2-adenergic receptor blocking agent, on ocular HSV-1 reactivation in rabbit eyes. Timolol maleate was applied following iontophoresis of 6-hydroxydopamine or by direct trans-corneal iontophoresis. It was discovered that timolol iontophoresis for 3 consecutive days could cause corneal epithelial lesions (127). However, a single direct transcorneal iontophoresis of timolol induced ocular HSV-1 shedding (128). The data suggest that both timolol (a p-adrenergic receptor antagonist) and epinephrine (an a,p-adrenergic receptor agonist) could induce ocular HSV-1 shedding. The results of these studies demonstrated that specific receptor occupancy of epinephrine alone is not an exclusive signal for viral reactivation. We are unable to explain why iontophoresis of this agonist/antagonist pair of adrenergic agents induces viral reactivation in animals that harbor latent HSV-1 strain McKrae.
Studies with propranolol, also a p-adrenergic antagonist, have yielded paradoxical results. Kaufman et al. (130) showed that propranolol suppressed both ocular HSV-1 recurrence and severity following spontaneous reactivation in the rabbit. Gebhardt and Kaufman (131) demonstrated that propranolol blocked the HSV-1 reactivation following hyperthermic induction in latently infected mice. Garza and Hill (132) examined the effect of propranolol on HSV-1 reactivation in latently infected rabbits following induction either by epinephrine iontophoresis (stress pathway) or by systemic immunosuppression (nonstress pathway). Immunosuppression was produced by injection of cyclophosphamide and dexamethasone (133). Propranolol was given intramuscularly by injection in doses of 5, 20, or 200 mg/kg twice daily. Propranolol had no effect at any concentration on blocking HSV-1 induction, either by transcorneal iontophoresis of epineph-rine or by systemic immunosuppression due to cyclcophosphamide plus dexamethasone treatment. These results suggest that (a) propranolol can specifically block the induction pathway in the mouse model but not the rabbit model, or (b) propranolol is not a broad/potent enough inhibitor to prevent HSV-1 induction in the rabbit.
d. Epinephrine Iontophoresis and the Study of LATs Initial studies in murine models of HSV infection demonstrated the presence of novel HSV RNA transcripts in sensory neurons of mice harboring a latent HSV infection (134-136). These mRNA species resulted from transcription of the only region of the HSV genome that is transcriptionally active during the latent phase of HSV infection (134,137-139). Latency-associated transcripts (LATs RNA) have also been demonstrated in rabbits (140,141) and humans (135,142,143). Three "major" LATs (2, 1.55, and 1.45 kb) as well as a "minor" transcript (8.3 kb) have been detected in ganglia latently infected with HSV-1 (134,136,144-146). The "major" LATs are highly expressed and poly(A)_, whereas the "minor" LAT has low level expression and is poly(A) + . LATs are expressed at similar levels in sensory neurons of mice, rabbits, and humans (136,140,142). Although LAT is not essential for latency, it remains the only known molecular marker for latent HSV infection (134,139,147,148).
Epinephrine iontophoresis and/or the rabbit eye model have been used in conjunction with various mutant strains of HSV to examine the role of LATs in the process of establishment, maintenance, and reactivation of latent HSV infection. HSV constructs that contain genetic alterations in the LAT locus were compared to unmodified (wild-type) or marker rescued (LAT-positive) viruses for their ability to establish latency and/or reactivate from latency (134,136). Hill et al. (149,150) used mutant HSV-1 strains (X10-13 or 17ASty) that could not produce a latently associated transcript (LAT-negative) and genetically engineered LAT-positive mutants (XC-20 or 17ASty-Res) to demonstrate that the LAT-positive strains, which are similar to the parent HSV-1 strains 17Syn + , could be reactivated by iontophoresis with high efficiency. The LAT-negative strains showed very limited or significantly reduced reactivation. The results suggested a role for the latently associated transcript in viral reactivation. All the mutants (LAT-negative or LAT-positive) were able to establish and maintain latency in sensory ganglia (149,150).
Bloom et al. (151) demonstrated that HSV-1 mutants with deletions encompassing the LAT promoter area have significantly reduced ocular reactivation following transcorneally iontophoresed epinephrine induction. 17APst (LAT-negative) has a 202 bp deletion in the LAT promoter in an area encompassing the TATA box, cAMP response element (CRE), upstream stimulatory factor, and transcription start sites, which signifi-
cantly reduces LAT transcription (147,151). They demonstrated that although this virus established latency with the same kinetics as its parent, 17Syn+, spontaneous and adrenergically induced reactivation is significantly reduced. This evidence suggested that expression of LATs was important in regulating HSV-1 reactivation. Bloom et al. (152) also examined the LAT promoter region at a more precise level. HSV-1 mutants containing a mutated CRE-1 element were shown to be significantly impaired in their ability to reactivate either spontaneously or by induction following epi-nephrine iontophoresis in the rabbit eye model (152). This evidence suggested a role for CRE-1 (CRE-1 binding factors) (134,136,153) in the reactivation process.
Similar studies have provided evidence that viral strains exhibit varying sensitivities to genetic alterations within the LAT domain. Loutsch et al. (154) used mutant strains of McKrae and 17Syn+ that had identical 371-base-pair deletion mutations in the LAT genes to examine spontaneous in vivo reactivation kinetics. They were able to demonstrate that an identical deletion resulted in different in vivo reactivation phenotypes. The results suggest a difference in genetic background of McKrae and 17Syn+ resulting in different in vivo reactivation phenotypes.
Zheng et al. (155) recently investigated the ability of LAT-positive and LAT-negative strains of HSV-1 to establish latent infection in rabbit corneas following penetrating keratoplasty. 17APst (LAT-negative, low reactivation) and 17Pr (LAT-positive, high reactivation recusant of 17APst) were inoculated into rabbit corneas. Latently infected rabbits received corneal allografts from naive rabbits, and naive rabbits received grafts from latently infected rabbits. Penetrating keratoplasty of latently infected and naive rabbits made it possible to study the migration of HSV from a site of latency to other tissues. They demonstrated that corneas from latently infected rabbits contain HSV-1 DNA that can replicate after viral reactivation by transcorneal epinephrine iontophoresis. The LAT negative strain also had a significantly reduced ability to replicate. The data showed that HSV-1 in latently infected rabbit sensory ganglia could be induced by epinephrine iontophoresis and migrate into other uninfected tissues (i.e., the transplanted cornea of a naive rabbit). The LAT-negative strain had a reduced ability to migrate. The results suggest that HSV-1 can migrate in both anterograde and retrograde directions between the site of viral latency (the trigeminal ganglion in the rabbit) and corneal tissues following epi-nephrine-induced reactivation. This is the first report to provide evidence to support the concept of corneal HSV-1 latency and reactivation (156,157).
Numerous other studies have used iontophoresis to study viral reactivation, as well as the treatment and prevention of recurrent herpetic disease (158-164). In summary, these studies validate the accuracy, reliability, and reproducibility of iontophoresis as a noninvasive mode of drug delivery and suggest that the iontophoretic model may be beneficial in the development of new, designer drugs to replace or supplement current herpetic disease regimens. Experimentally, ocular iontophoresis and the rabbit eye model are mainstays in the study of HSV-1 latency and reactivation.
Was this article helpful?