Molecular Vision 1999; 5:36 <>
Received 1 June 1999 | Accepted 7 December 1999 | Published 20 December 1999

Iontophoresis of lysophosphatidic acid into rabbit cornea induces HSV-1 reactivation: Evidence that neuronal signaling changes after infection

Rex E. Martin,1 Jeannette M. Loutsch,2 Hildegardo H. Garza, Jr.,3 Daniel J. Boedeker,1 James M. Hill2

1Department of Cell Biology and Oklahoma Center for Neurosciences, University of Oklahoma College of Medicine, Oklahoma City, OK; 2LSU Eye Center and Neuroscience Center, Louisiana State University Medical Center School of Medicine, New Orleans, LA; 3Department of Biology, Texas A & M University, Kingsville, TX

Correspondence to: Rex E. Martin, Ph.D., Room 517, 940 Stanton Young Boulevard, Department of Cell Biology and Oklahoma Center for Neurosciences, University of Oklahoma College of Medicine, Oklahoma City, OK, 73104; Phone: (405) 271-2377, ext 229; FAX: (405) 271-3548; email:


Purpose: Lysophosphatidic acid induces neurite retraction; it is also present in tears and aqueous humor. We determined whether lysophosphatidic acid induces HSV-1 reactivation in latently infected rabbits and whether the nerve growth associated protein GAP-43 undergoes posttranslational modification during the course of HSV-1 infection.

Methods: Rabbits were infected with HSV-1 and acute infection was documented by slit lamp examination. Corneas of latently infected rabbits were treated with lysophosphatidic acid or lysophosphatidylserine (structurally similar but lacking biological potency). For application to the cornea, these compounds were impregnated into collagen shields, applied as topical drops, or iontophoresed. In another experiment, corneas of latently infected rabbits were either untreated or treated iontophoretically with lysophosphatidic acid, lysophosphatidylserine, or saline. Ocular swabs detected shedding of infectious virus. Western blot and immunoprecipitation identified GAP-43 in corneal extracts and densitometry of silver-stained isoelectric focusing gels measured changes in GAP-43 isoform abundance.

Results: Iontophoresis of lysophosphatidic acid induced HSV-1 shedding more frequently than lysophosphatidylserine or saline. Viral shedding induced by collagen shield and topical drop administration was low and not significantly different for lysophosphatidic acid and lysophosphatidylserine. Five discrete GAP-43 isoforms predominated in the IEF gels. Most abundant were the pI 4.7 band in uninfected cornea and the pI 5.05 band in latently-infected cornea. Compared to latently-infected cornea, there was no significant change in isoform abundance 1 h after lysophosphatidic acid iontophoresis, but 24 and 72 h later, the pI 5.05 band was diminished.

Conclusions: Lysophosphatidic acid can induce HSV-1 reactivation and changes in GAP-43 pI suggest that posttranslational modifications, possibly related to phosphorylation and ADP-ribosylation, are occurring during HSV-1 latency and after LPA is iontophoretically applied to the cornea. How lysophosphatidic acid-induced signaling, HSV-1 reactivation, and GAP-43 pI are related remains to be determined.


Most clinically significant ocular herpes simplex virus type-1 (HSV-1) infections result from the reactivation of latent virus [1]. Systemic stresses or physical traumas that disrupt the corneal nerves induce HSV-1 reactivation [2]. Our intent is to learn how to impair or prevent HSV-1 reactivation by characterizing the biochemical changes that occur in neurons during infection. As a marker of these biochemical changes, we have chosen to study a nerve growth associated protein called GAP-43 [3] and we have shown that the concentration of GAP-43 increases in corneal nerves during HSV-1 latency [4,5]. A specific role for GAP-43 in HSV-1 infection has not been established.

GAP-43 participates in the remodeling and regeneration of synaptic terminals; it is also implicated in the constant remodeling of nerve endings that occurs in the densely innervated cornea [3-7]. GAP-43 can be palmitoylated or ADP-ribosylated and has at least four phosphorylation sites [8-11]. When phosphorylated in the appropriate site by protein kinase C, GAP-43 interacts with the cytoskeleton to potentiate actin polymerization and neurite outgrowth [12]. GAP-43 expression is induced by nerve growth factor [13,14]. This finding, plus the fact that systemic nerve growth factor depletion induces HSV-1 reactivation [15], suggests that neuronal systems that potentiate neurite elongation are involved in the establishment and maintenance of HSV-1 latency as well as the switch from the latent to the lytic phase. To test this concept, we hypothesized that if the latent HSV-1 in corneal nerves can be reactivated by physical trauma or systemic stress, then substances that perturb neurite homeostasis in in vitro culture could also induce latent HSV-1 to reactivate in vivo.

Lysophosphatidic acid (LPA) was selected to perturb neuronal homeostasis because it is present in fluids that bathe the cornea and it induces neurites to retract in culture [16,17]. LPA is a negatively charged, naturally occurring, receptor binding, lysophospholipid that is synthesized from phosphatidic acid by phospholipase A2 [18-20]. In non-pathologic conditions, LPA is present at concentrations of 2-20 µM in serum, 1 µM in lacrimal gland fluid, and 0.2 µM in aqueous humor [16,21]. In cryogenically-injured cornea, the LPA concentration increases 3-6 fold in aqueous humor but not in tear fluid [16]. The details of signaling events that relate to LPA-receptor binding and to posttranslational changes in GAP-43 are unknown. A direct connection between LPA and GAP-43 signaling is also unknown, but a compelling connection is suggested because LPA stimulates neurite retraction and GAP-43 supports neurite elongation [3,13,14,17]. Moreover, the LPA receptor functions through membrane-bound and soluble G proteins [18,20] and GAP-43 is associated with G proteins [22,23], and ADP-ribosylated [9,10].

We delivered LPA and another negatively charged lysophospholipid, lysophosphatidylserine, to the corneas. To determine the optimal way to deliver these compounds, topical drops, collagen shields, or iontophoresis were tested. We then determined frequency of HSV-1 reactivation after each procedure. We observed that the pI of GAP-43 changed during HSV-1 infection and after LPA treatment.


Virus and Rabbits

HSV-1 strain McKrae was propagated on primary rabbit kidney cells and titered on African Green Monkey cells (CV-1). The virus was aliquoted and stored at -70 °C. New Zealand White rabbits (2-3 kg) were inoculated bilaterally by applying 2 x 105 plaque forming units of HSV-1 in a 25-µl suspension into the cul-de-sac of each eye. The eye was closed and massaged for 30 seconds. The corneas were not scarified. Slit-lamp examination and ocular swabbing were done 3 days after inoculation to identify epithelial lesions and to verify a productive infection, respectively. HSV-1 was presumed to be latent when slit-lamp examination indicated no corneal defects and tear film swabs contained no infectious virus [24].

The reactivation studies were performed 5-6 weeks after virus inoculation. All eyes underwent membranectomies 1 week prior to treatment. A sample of the tear film from each eye of each rabbit was collected on Dacron swabs once daily for 7 days after the start of treatment. The swabs were placed in tubes containing monolayers of primary rabbit kidney cells and the cells were monitored for the development of cytopathic effect indicative of infectious virus. Rabbits used in studies of reactivation were sacrificed seven days after initiation of the first treatment. For densitometric analysis of GAP-43 isoforms, corneas from latently infected rabbits were taken 1, 24, and 72 h after iontophoretic LPA treatment. The rabbits were handled and maintained in accordance with the tenets of the Association for Research in Vision and Ophthalmology Resolution on the Use of Animals in Research.


LPA and lysophosphatidylserine were solubilized in deionized water at a final concentration of 40 µM (Sigma Chemical Co., St. Louis, MO). Lysophosphatidylserine was used as a control for the non-specific effects of LPA because it is structurally similar to LPA but lacks biological potency. The phospholipids were applied to both corneas of latently infected rabbits in topical drops (8 eyes), in collagen shields (8 eyes), or by iontophoresis (8 eyes). Three different methods were used to apply the phospholipids in the first experiment. Topical drops were applied every 30 minutes for 9 hours. Collagen shields (Bio-Cor, Bausch & Lomb; Clearwater, FL) were saturated with LPA or lysophosphatidylserine, applied to the corneal surface, and left in place for 8 h. Iontophoretic drug delivery was performed once daily for 3 consecutive days as previously described [24]. In a second experiment, LPA, lysophosphatidylserine, or saline was iontophoresed once daily in 10 eyes each for 3 consecutive days as in experiment 1; eight eyes were monitored as untreated controls.

Corneal Proteins

Following previously described procedures [5], corneas were dissected and immediately frozen in liquid nitrogen. The frozen tissue from individual corneas were ground to a powder in a liquid nitrogen-cooled mortar and pestle. The powder from each cornea was resuspended in 15 volumes of homogenization buffer, which consisted of 20 mM Tris-HCl (pH 7.5) containing 320 mM sucrose, 10 mM ethylene glycol-bis [ß-aminoethylether], 2 mM ethylenediaminetetraacetic acid (EDTA), 2 mM dithiothreitol, 0.05% leupeptin, and 0.2 mM phenylmethylsulfonyl fluoride. The suspension was centrifuged at 2000 x g for 10 min at 4 °C to remove the collagenous stromal matrix. The pellet was resuspended with sonication in five volumes of ice-cold homogenization buffer and centrifuged to remove the collagen and maximize extraction of non-collagenous proteins. Protein concentrations were determined using Bio-Rad (Richmond, CA) protein assay reagent and bovine serum albumin as a standard. All corneal protein procedures were done at 4 °C.

GAP-43 Identification and Densitometry

Three different approaches were used to identify discrete GAP-43 isoforms in rabbit corneal extracts (Figure 1). Two of these approaches used pI 4-6 horizontal IEF gels and an anti-GAP-43 monoclonal antibody that binds GAP-43 irrespective of its phosphorylation state [13]. In the third approach, proteins migrating at 43 kDa in sodium dodecyl sulfate (SDS)-polyacrylamide gels were electroeluted from the gel, concentrated, and then vertically electrophoresed in precast pI 3-7 IEF gels (Novex, San Diego, CA). All procedures were done at room temperature.

Figure 1, IP/IEF, shows the first approach taken to identify GAP-43 in IEF gels. GAP-43 was immunoprecipitated with staphylococcal protein G-coated agarose beads (Hyclone Laboratories Inc., Logan, UT) and monoclonal antibody 91E12 (Boehringer Mannheim, Indianapolis, IN) [25,26]. The extracts were first cleared of proteins that non-specifically bound protein G-coated agarose beads and then incubated overnight with gentle stirring and monoclonal antibody/protein G-coated agarose beads. The resulting immune complex was collected by centrifugation at 300 x g for 5 min. The supernatant was discarded and the immune complex was washed free of impurities with three cycles of centrifugation and resuspension in phosphate-buffered saline containing 5 mM EDTA and 1% IGEPAL CA-630 (NP-40, Sigma Chemical Co.). The immune complex was dissociated with brief sonication in 2 M urea and centrifuged at 300 x g. Proteins in the resulting supernatant were dialyzed against water, concentrated by lyophilization, resuspended in water, and applied to horizontal IEF gels with a 4-6 pI gradient (pI 3-10 and pI 4-6 ampholytes mixed at a ratio of 1:4). These 0.8 mm-thick gels were prefocused in a BiophoresisTM apparatus (Bio-Rad) for 15 min at 400 V (constant) before the samples were added. The samples (approximately 5 µg protein in 7-10 µl and containing 5% carrier ampholytes) were added to wells (depressions) which were cast in the gel. The gel was then run for an additional 90 min at 8 W (constant power). The immunoprecipitated GAP-43 isoforms were detected with the Bio-Rad Silver Stain Kit according to the manufacturer's protocol.

Western blotting [5,27] was the second approach taken to identify GAP-43 (Figure 1, W). Proteins were separated in pI 4-6 IEF gels and transferred to nitrocellulose membranes in 0.2% acetic acid at 20 V for 20 min using plate-type electrodes. The nitrocellulose was incubated for 60 min in phosphate-buffered saline blocking solution containing 2% non-fat dry milk and 0.1% polyoxyethylenesorbitan monolaurate (Tween 20, Bio-Rad). Monoclonal antibody 91E12 was diluted 1:1000 in blocking solution and incubated with the nitrocellulose membrane overnight at 4 °C. According to the manufacturer's protocol, the ECLTM kit of Amersham (Arlington Heights, IL) and a peroxidase-conjugated horse anti-mouse IgG (Vector Laboratories, Burlingame, CA) were used to identify the GAP-43 isoforms by chemiluminescence.

The third approach taken to identify GAP-43 required two electrophoretic steps (Figure 1, EE). Proteins migrating at 43 kDa in 12% SDS-polyacrylamide gels were excised and electroeluted at 10 mA for 4.5 h using a Bio-Rad model 422 electroelution apparatus. The electroeluted proteins were concentrated by lyophilization, resuspended in water, and electrophoresed in precast, pI 3-7 IEF gels (Novex, San Diego, CA). The gels were run vertically at a series of voltages (100 V for 1 h, 200 V for 2 h, and 500 V for 30 min) as per manufacturer recommendations. The GAP-43 isoforms were silver stained as described above.

The pI of GAP-43 corresponds to its degree of phosphorylation and ADP-ribosylation [9-11,28,29]. To quantify the pI of GAP-43 isoforms in rabbit cornea, extracts from corneas that were frozen in liquid nitrogen and prepared as described above were electrophoresed in precast, pI 3-7 IEF gels and silver stained [10,30]. Digital images of the gels were collected with a Hewlett-Packard ScanJet ADF scanner and the density (abundance) of each silver-stained GAP-43 isoform in the sample was measured using NIH Image software. To obtain a densitometric average for the abundance of each GAP-43 isoform, at least four separate corneal samples were analyzed independently. Because both phosphorylation and ADP-ribosylation add negative charges to GAP-43, the degree of GAP-43 posttranslational modification can be equated to increases in its acidity. This was measured by determining the percent contribution of each GAP-43 isoform to the total detectable GAP-43 in the sample (Figure 2).

Calculations and Statistics

Shedding of virus was evaluated in terms of frequency (eyes: number of positive eyes divided by total number of eyes; swabs: number of positive swabs divided by total number of swabs) and duration (number of positive swabs taken over 7 days following treatment divided by total number of eyes). In the analysis of LPA-induced HSV reactivation, the p values for statistical comparisons were determined using a two-tailed Fisher's exact test. One-way analysis of variance with post-hoc Newman-Keuls tests determined statistical significance of differences in GAP-43 isoform abundance between various treatment groups. P values less than 0.05 were considered significant.


LPA-Induced HSV-1 Ocular Shedding

The first experiment evaluated methods of applying LPA and lysophosphatidylserine to the cornea for their ability to reactivate HSV-1. In terms of swabs taken during the 7-day period after treatment was begun, a significantly higher frequency of HSV-1 shedding was seen in the eyes treated iontophoretically with LPA (32% for LPA vs 11% for lysophosphatidylserine; p<0.010; Table 1). When collagen shields or topical drops were used to administer the compounds, significant differences in shedding between LPA and lysophosphatidylserine were not evident (Table 1).

In a second experiment that compared no treatment, saline iontophoresis, and lysophosphatidylserine iontophoresis with LPA iontophoresis, a significantly higher frequency of shedding (positive swabs) was seen after LPA treatment (p<0.001 for all; Table 2). In terms of eyes positive for infectious virus, the frequency was significantly greater for LPA treatment, compared with saline (p=0.020) and no treatment (p=0.003), but not in comparison with lysophosphatidylserine treatment (p=0.057). The average duration of shedding was markedly greater for LPA-treated eyes than for any of the other groups (2.90 days for LPA, 1.10 days for lysophosphatidylserine, 0.4 days for saline and 0.12 days for no treatment; Table 2).

In both of these experiments, iontophoresis of LPA induced shedding of infectious virus in at least one eye of every rabbit tested (Table 1 and Table 2). The results demonstrate that HSV-1 reactivation and shedding are induced in the rabbit when a compound known to perturb neural homeostasis (LPA) is iontophoresed into the cornea.

Corneal Proteins

GAP-43 is abundant in neurites of the cornea [4,5]. It is an ADP-ribosylated and multiply-phosphorylated protein with isoelectric points ranging from 4.3 to 5.2 [9,10,28,29]. To correlate changes in neuronal homeostasis with the course of HSV-1 infection, IEF gels were used to evaluate changes in the endogenous pI of GAP-43. Irrespective of the experimental approach that was used to find GAP-43 in IEF gels, protein bands with isoelectric points in the reported range for GAP-43 (GAP-43 isoforms) were identified [28]. The most basic band had a pI of approximately 5.05 and the most acidic had a pI of 4.5. In some samples, doublet bands existed at pI 4.7 and 4.5 after immunoprecipitation (Figure 1, IP/IEF). The consistent focusing of the GAP-43 isoforms in IEF gels facilitated densitometric quantitation. The percent contribution of each band (A, B, C, D, and E) to their total was measured densitometrically (Figure 2). When doublet bands were evident at pI 4.7 (Figure 2), they were considered one band in these calculations. The results shown in Figure 2 indicated that there were statistically significant differences in the relative abundance of GAP-43 isoforms between uninfected (UI), latently-infected (LAT), and LPA-treated latently-infected rabbits (LPA).

In uninfected, untreated rabbits (Figure 2, UI), the most basic GAP-43 isoform, band A (pI 5.05) represented 19% of the five GAP-43 isoforms while band D (pI 4.7) represented 33% of the total GAP-43. Comparing uninfected rabbits (Figure 2, UI) to latently-infected rabbits (Figure 2, LAT), the abundance of band C (pI 4.8) was increased in the latently-infected rabbits and band D (pI 4.7) was less abundant. The statistically significant increase in band C and decrease in band D, relative to uninfected rabbits, was maintained in the LPA-treated rabbit corneas suggesting that the GAP-43 isoforms in latently infected corneas were collectively less acidic (modified). One hour after LPA was iontophoretically applied to latently-infected cornea there were no statistically significant differences in GAP-43 isoform abundance, but after 24 and 72 h, the LPA-treated corneas contained less band A (pI 5.05) (Figure 2, LPA 12 h & LPA 72 h). This decrease in band A abundance following LPA treatment demonstrates an increase in the collective acidity (posttranslational modification) of GAP-43 after LPA iontophoresis.


The effects of HSV-1 latency on neurons and the effects of host neurons on latent HSV-1 are not understood. Until we demonstrated that GAP-43 expression increased in the trigeminal ganglia and corneas of rabbits latently infected with HSV-1 [5], it was believed that latent HSV-1 infections had no prolonged effect on the host's neuronal proteins. We show here that when LPA is administered iontophoretically, latent HSV-1 can be induced to reactivate. Moreover, we show that when HSV-1 infects corneal neurons, the pI of the nerve growth associated protein GAP-43, changes.

In determining the best way to deliver LPA we demonstrated that significant reactivation of latent HSV-1 occurred when LPA was iontophoresed into the cornea. This result is attributed to the anatomy of the cornea and rapid metabolism of LPA by endogenous lysophospholipases [31,32]. The corneal nerves pass through the stroma superficial to the endothelium and the sensory terminals are located in epithelium [32]. LPA in the tear film (lacrimal gland fluid), in collagen shields, or in topical drops could make direct contact with the corneal nerve endings but it would be unlikely to diffuse into the epithelium or stroma without being metabolized. However, iontophoresis driving LPA though the epithelium and into the stroma would facilitate LPA binding to receptors on the nerve fiber (in addition to those on the nerve ending) before it could be metabolized [31]. Even though LPA concentrations increase after cryogenic injury [16], we do not know whether they ever reach 40 µM (40 times that of normal lacrimal gland fluid and 200 times that of normal aqueous humor), but these experiments do demonstrate that HSV-1 can be reactivated by a new and different pathway. Perhaps an understanding of LPA receptor-mediated signaling will provide more clues to the mechanisms that regulate HSV-1 latency.

GAP-43 has isoelectric points ranging from 4.3 to 5.2 This range can be attributed to phosphorylation or ADP ribosylation [9,10,28,29]. Proteins that are phosphorylated or ADP-ribosylated typically appear as a single band in SDS-polyacrylamide gels but they are resolved into multiple bands in IEF gels because these posttranslational modifications increase the acidity of the protein and decrease its pI. When GAP-43 was immunoprecipitated from rabbit corneal extracts and electrophoresed in SDS-polyacrylamide gels, it migrated as a single band at 43 kDa (data not shown). However, when these extracts were electrophoresed in IEF gels, multiple distinct GAP-43 isoforms were identified (Figure 1). The differing pI of these isoforms is likely due to phosphorylation and/or ADP ribosylation. Edgar et al. [11], showed that there are four (or more) sites for endogenous GAP-43 phosphorylation. If there was no ADP-ribosylation, the five bands seen in Figure 1 and Figure 2, therefore, could reflect increasing (successive) phosphorylation of GAP-43, with the most basic band (pI 5.05) being dephosphorylated GAP-43 and the most acidic band being GAP-43 with four phosphorylations. Finally, either of these posttranslational modifications could induce conformational changes that "cover-up" or "expose" charged amino acids and change GAP-43 pI. Whether the changes in pI reported in Figure 2 reflect changes in GAP-43 phosphorylation, ADP ribosylation, or protein confirmation, the changing abundance of specific GAP-43 isoforms is indicative of changes in neuronal signal transduction. These results support the hypothesis that substances that perturb neurite homeostasis also induce latent HSV-1 to reactivate.

In our previous experiments, increased concentrations of GAP-43 were observed while the corneal nerves were recovering from the lytic phase of infection This result was expected because GAP-43 concentrations are high when neurites are growing or regenerating or when terminals remodel [3,4,13,14]. We also found that 154 days post inoculation (during viral latency), the increased concentration of GAP-43 persisted [5]. This increased GAP-43 abundance during HSV-1 latency was not expected; it suggested that neuronal homeostasis was changed by HSV-1 infection. Our current study further supports the hypothesis that HSV-1 infection changes neuronal homeostasis because the abundance of a pI 4.8 GAP-43 isoform was increased and that of a pI 4.7 isoform was decreased when the rabbits were latently infected with HSV-1. These findings demonstrate that in addition to the increased expression of GAP-43 in latently infected rabbits [5], there are also biochemical differences in the corneal nerves of latently infected rabbits that likely involve the kinase(s) and/or phosphatase(s) that act on GAP-43 [33,34]. Moreover, the abundance of a pI 5.05 GAP-43 isoform was decreased after latently-infected corneas were treated with LPA. This change was not noted 1 hour after LPA iontophoresis but it was evident 24 and 72 h later. The altered pI of GAP-43 at 24 and 72 h after LPA iontophoresis could be due to three things: LPA-stimulated signaling, reactivation of latent HSV-1, or simply the iontophoretic treatment of rabbit corneas. The latter possibility seems unlikely because saline iontophoresis did not induce significant HSV-1 reactivation and 1 h after LPA treatment there was no difference in the pI of GAP-43 between latently-infected corneas and LPA-treated latently-infected corneas (Figure 2). More research will be necessary to demonstrate that neuronal signaling mechanisms affecting GAP-43 posttranslational modification mediate either LPA signalling or HSV-1 infection.

The nerve growth factor receptor [15], the glucocorticoid receptor/response element [35], and the cAMP response element [36] have been implicated in HSV-1 reactivation. We demonstrate that LPA, a lipid present in lacrimal gland fluid and aqueous humor [16], also stimulates HSV-1 reactivation and shedding. Significantly, another positively charged lysophospholipid, lysophosphatidylserine, has minimal effect on reactivation. This specificity of LPA in the induction of reactivation suggests that it could act through a receptor-mediated event [18,20]. In neurons of the cerebral cortex, LPA binds a 7-transmembrane receptor that activates two different signaling cascades. One cascade operates through a pertussis toxin-sensitive Gi G-protein and the other cascade, which is pertussis toxin-insensitive, operates through the small GTP-binding protein Rho [20]. Further experiments are necessary to demonstrate if either pathway is active in HSV-1-infected neurons.

We propose that the signal transduction cascades mediating GAP-43 expression and posttranslational modification are intertwined with those mediating the course of recurrent HSV-1 disease. We make this statement because GAP-43 posttranslational modification and expression change as HSV-1 becomes latent [5] and because neuronal stimulators (nerve growth factor, LPA, and glucocorticoids) manifest changes in cell signaling that affect GAP-43, HSV-1 reactivation and neurite elaboration [3,5,13,14,33-35]. Moreover, the reversible palmitoylation of GAP-43 mediates its association with membranes or its presence in the cytoplasm [37] and correspondingly, there are data confirming that GAP-43 interacts with both membrane and soluble G-proteins. In fibroblasts transfected with GAP-43, the membrane-bound protein stabilizes filopodia in a Rho GTPase-dependent manner [22]. When oocytes are transfected with GAP-43, G protein-coupled, serotonin-induced chloride current is augmented if GAP-43 is not palmitoylated [38]. Work with chromaffin cells has shown that secretory granules (which could be similar to vesicles secreting nascent HSV-1) are associated with Go and Rho A. In these cells, Go activation inhibits priming of exocytosis and when cytosolic (but not membrane derived) GAP-43 is incubated with these chromaffin granules, phosphatidylinositol 4-kinase activity is stimulated in a Rho A-dependent manner [39]. Perhaps future experiments manipulating GAP-43 posttranslational modification and LPA-induced signaling will increase our knowledge of events that occur during HSV-1 reactivation and improve our understanding of mechanisms underlying neuronal latency.


Supported in part by OCAST contract HN4-018; NIH grants EY06789, EY06311, EY02672, and EY02377 (Core grant), and an Incentive Grant from the LSU Neuroscience Center. The authors thank Drs. Herbert E. Kaufman and Nicolas G. Bazan for their encouragement, enthusiasm, and support for this project, Mrs. Maxine S. Evans for her expert technical support, Dr. Hilary W. Thompson for his expert assistance with the statistical analysis, and Dr. Robert E. Anderson for critical reading of the manuscript. Presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology, Fort Lauderdale, Florida, May 1998.


1. Stevens JG. Overview of herpesvirus latency. Seminars in Virology 1994; 5:191-6.

2. Fawl RL, Roizman B. The molecular basis of herpes simplex virus pathogenicity. Seminars in Virology 1994; 5:261-71.

3. Benowitz LI, Routtenberg A. GAP-43: an intrinsic determinant of neuronal development and plasticity. Trends Neurosci 1997; 20:84-91.

4. Martin RE, Bazan NG. Growth-associated protein GAP-43 and nerve cell adhesion molecule in sensory nerves of cornea. Exp Eye Res 1992; 55:307-14.

5. Martin RE, Henken DB, Hill JM. Altered expression and changing distribution of the nerve growth associated protein GAP-43 during ocular HSV-1 infection in the rabbit. J Neurovirol 1996; 2:127-35.

6. Harris LW, Purves D. Rapid remodeling of sensory endings in the corneas of living mice. J Neurosci 1989; 9:2210-4.

7. Klyce SD, Beuerman RW. Structure and function of the cornea. In: Kaufman HE, Barron BA, McDonald MB, editors. The cornea. 2nd ed. Boston, MA: Butterworth-Heinemann; 1998. p. 3-50.

8. Skene JH, Virag I. Posttranslational membrane attachment and dynamic fatty acylation of a neuronal growth cone protein, GAP-43. J Cell Biol 1989; 108:613-24.

9. Philibert K, Zwiers H. Evidence for multisite ADP-ribosylation of neuronal phosphoprotein B-50/GAP-43. Mol Cell Biochem 1995; 149-150:183-90.

10. Chao D, Severson DL, Zwiers H, Hollenberg MD. Radiolabelling of bovine myristoylated alanine-rich protein kinase C substrate (MARCKS) in an ADP-ribosylation reaction. Biochem Cell Biol 1994; 72:391-6.

11. Edgar MA, Pasinelli P, DeWit M, Anton B, Dokas LA, Pastorino L, DiLuca M, Cattabeni F, Gispen WH, De Graan PN. Phosphorylation of the casein kinase II domain of B-50 (GAP-43) in rat cortical growth cones. J Neurochem 1997; 69:2206-15.

12. He Q, Dent EW, Meiri KF. Modulation of actin filament behavior by GAP-43 (neuromodulin) is dependent on the phosphorylation status of serine 41, the protein kinase C site. J Neurosci 1997; 17:3515-24.

13. Jacobson RD, Virag I, Skene JHP. A protein associated with axon growth, GAP-43, is widely distributed and developmentally regulated in rat CNS. J Neurosci 1986; 6:843-55.

14. Jap Tjoen San ER, Schmidt-Michels M, Oestericher AB, Schotman P, Gispen WH. Dexamethasone-induced effects on B-50/GAP-43 expression and neurite outgrowth in PC12 cells. J Mol Neurosci 1992; 3:189-95.

15. Hill JM, Garza HH Jr, Helmy MF, Cook SD, Osborne PA, Johnson EM Jr, Thompson HW, Green LC, O'Callaghan RJ, Gebhardt BM. Nerve growth factor antibody stimulates reactivation of ocular herpes simplex virus type 1 in latently infected rabbits. J Neurovirol 1997; 3:206-11.

16. Liliom K, Guan Z, Tseng JL, Desidero DM, Tigyi G, Watsky MA. Growth factor-like phospholipids generated after corneal injury. Am J Physiol 1998; 274:C1065-74.

17. Smalheiser NR, Dissanayake S, Kapil A. Rapid regulation of neurite outgrowth and retraction by phospholipase A2-derived arachidonic acid and its metabolites. Brain Res 1996; 721:39-48.

18. Moolenaar WH, Kranenburg O, Postma FR, Zondag GC. Lysophosphatidic acid: G-protein signalling and cellular responses. Curr Opin Cell Biol 1997; 9:168-73.

19. Snitko Y, Yoon ET, Cho W. High specificity of human secretory class II phospholipase A2 for phosphatidic acid. Biochem J 1997; 321:737-41.

20. Hecht JH, Weiner JA, Post SR, Chun J. Ventricular zone gene-1 (vzg-1) encodes a lysophosphatidic acid receptor expressed in neurogenic regions of the developing cerebral cortex. J Cell Biol 1996; 135:1071-83.

21. Schulze C, Smales C, Rubin LL, Staddon JM. Lysophosphatidic acid increases tight junction permeability in cultured brain endothelial cells. J Neurochem 1997; 68:991-1000.

22. Aarts LH, Schrama LH, Hage WJ, Bos JL, Gispen WH, Schotman P. B-50/GAP-43-induced formation of filopodia depends on Rho-GTPase. Mol Biol Cell 1998; 9:1279-92.

23. Strittmatter SM, Valenzuela D, Kennedy TE, Neer EJ, Fishman MC. GO is a major growth cone protein subject to regulation by GAP-43. Nature 1990; 344:836-41.

24. Hill JM, Wen R, Halford WP. Pathogenesis and molecular biology of ocular HSV in the rabbit. In: Brown SM, MacLean AR, editors. Herpes Simplex Virus Protocols. Totowa (NJ): Humana Press; 1998. p. 238-63.

25. Springer TA. Monoclonal antibody analysis of complex biological systems. Combination of cell hybridization and immunoadsorbents in a novel cascade procedure and its application to the macrophage cell surface. J Biol Chem 1981; 256:3833-9.

26. Sastre L, Kishimoto TK, Gee C, Roberts T, Springer TA. The mouse leukocyte adhesion proteins Mac-1 and LFA-1: studies on mRNA translation and protein glycosylation with emphasis on Mac-1. J Immunol 1986; 137:1060-5.

27. Burnette WN. "Western blotting": electrophoretic transfer of proteins from sodium dodecyl sulfate--polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal Biochem 1981; 112:195-203.

28. McIntosh H, Parkinson D, Meiri K, Daw N, Willard M. A GAP-43-like protein in cat visual cortex. Vis Neurosci 1989; 2:583-91.

29. Zwiers H, Verhaagen J, van Dongen CJ, de Graan PN, Gispen WH. Resolution of rat brain synaptic phosphoprotein B-50 into multiple forms by two-dimensional electrophoresis: evidence for multisite phosphorylation. J Neurochem 1985; 44:1083-90.

30. Siegmund KD, Klink F. Introduction of additional charges as an aid in protein purification: isolation of elongation factor 2 from Sulfolobus acidocaldarius by preparative isoelectric focusing before and after ADP-ribosylation. Protein Expr Purif 1994; 5:553-8.

31. Wang A, Dennis EA. Mammalian lysophospholipases. Biochim Biophys Acta 1999; 1439:1-16.

32. Rozsa AJ, Beuerman RW. Density and organization of free nerve endings in the corneal epithelium of the rabbit. Pain 1982; 14:105-120.

33. Dokas LA, Ting S, Edgar MA, Oestreicher AB, Gispen WH, DeGraan PN. Regulation of in vitro phosphorylation of the casein kinase II sites in B-50 (GAP-43). Brain Res 1998; 781:320-8.

34. Liu Y, Storm DR. Dephosphorylation of neuromodulin by calcineurin. J Biol Chem 1989; 264:12800-4.

35. Hardwicke MA, Schaffer PA. Differential effects of nerve growth factor and dexamethasone on herpes simplex virus type 1 oriL- and oriS-dependent DNA replication in PC12 cells. J Virol 1997; 71:3580-7.

36. Bloom DC, Stevens JG, Hill JM, Tran RK. Mutagenesis of a cAMP response element within the latency-associated transcript promoter of HSV-1 reduces adrenergic reactivation. Virology 1997; 236:202-7.

37. Baker LP, Storm DR. Dynamic palmitoylation of neuromodulin (GAP-43) in cultured rat cerebellar neurons and mouse N1E-115 cells. Neurosci Lett 1997; 234:156-60.

38. Nakamura F, Strittmatter P, Strittmatter SM. GAP-43 augmentation of G protein-mediated signal transduction is regulated by both phosphorylation and palmitoylation. J Neurochem 1998; 70:983-92.

39. Gasman S, Chasserot-Golaz S, Hubert P, Aunis D, Bader MF. Identification of a potential effector pathway for the trimeric Go protein associated with secretory granules. Go stimulates a granule-bound phosphatidylinositol 4-kinase by activating RhoA in chromaffin cells. J Biol Chem 1998; 273:16913-20.

Martin, Mol Vis 1999; 5:36 <>
©1999 Molecular Vision <>
ISSN 1090-0535