JVI Figure table search 04
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow An author's correction has been published
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Thompson, R. L.
Right arrow Articles by Sawtell, N. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Thompson, R. L.
Right arrow Articles by Sawtell, N. M.

 Previous Article  |  Next Article 

Journal of Virology, January 2000, p. 965-974, Vol. 74, No. 2
0022-538X/00/$04.00+0
Copyright © 2000, American Society for Microbiology. All rights reserved.

Replication of Herpes Simplex Virus Type 1 within Trigeminal Ganglia Is Required for High Frequency but Not High Viral Genome Copy Number Latency

Richard L. Thompson1,* and N. M. Sawtell2,*

Department of Molecular Genetics, Microbiology, and Biochemistry, University of Cincinnati Medical Center, Cincinnati, Ohio 45267-0524,1 and Division of Infectious Diseases, Children's Hospital Medical Center, Cincinnati, Ohio 45229-30392

Received 12 May 1999/Accepted 7 October 1999


    ABSTRACT
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

The replication properties of a thymidine kinase-negative (TK-) mutant of herpes simplex virus type 1 (HSV-1) were exploited to examine the relative contributions of replication at the body surface and within trigeminal ganglia (TG) on the establishment of latent infections. The replication of a TK- mutant, 17/tBTK-, was reduced by ~12-fold on the mouse cornea compared to the rescued isolate 17/tBRTK+, and no replication of 17/tBTK- in the TG of these mice was detected. About 1.8% of the TG neurons of mice infected with 17/tBTK- harbored the latent viral genome compared to 23% of those infected with 17/tBRTK+. In addition, the latent sites established by the TK- mutant contained fewer copies of the HSV-1 genome (average, 2.3/neuron versus 28/neuron). On the snout, sustained robust replication of 17tBTK- in the absence of significant replication within the TG resulted in a modest increase in the number of latent sites. Importantly, these latently infected neurons displayed a wild-type latent-genome copy number profile, with some neurons containing hundreds of copies of the TK- mutant genome. As expected, the replication of the TK- mutant appeared to be blocked prior to DNA replication in most ganglionic neurons in that (i) virus replication was severely restricted in ganglia, (ii) the number of neurons expressing HSV proteins was reduced 30-fold compared to the rescued isolate, (iii) cell-to-cell spread of virus was not detected within ganglia, and (iv) the proportion of infected neurons expressing late proteins was reduced by 89% compared to the rescued strain. These results demonstrate that the viral TK gene is required for the efficient establishment of latency. This requirement appears to be primarily for efficient replication within the ganglion, which leads to a sixfold increase in the number of latent sites established. Further, latent sites with high genome copy number can be established in the absence of significant virus genome replication in neurons. This suggests that neurons can be infected by many HSV virions and still enter the latent state.


    INTRODUCTION
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

Experimental herpes simplex virus type 1 (HSV-1) infection in mice parallels that which occurs in humans (13, 53). During primary infection, viral replication at the surface is detectable for a 10- to 12-day period. Within 24 h postinfection (p.i.) virus is transported to the innervating sensory ganglia, where peak titers are achieved around 3 to 4 days p.i., and infectious virus becomes undetectable within the ganglia by 7 to 9 days p.i. The establishment of latent infections occurs during this replicative phase. Once latency is established, the ~75 viral genes expressed during the lytic phase are extremely repressed and only the latency-associated transcripts (LATs) are abundantly transcribed (1, 25, 26, 52, 54, 63).

Viral replication is not an absolute prerequisite for the establishment of latency, as replication-deficient or -incompetent viral mutants can establish latent infections (3, 4, 8-11, 14, 20, 22, 27, 33, 34, 51). Indeed, lethal mutations that preclude most viral gene expression do not completely prevent the establishment of latency (8, 46). From these observations it has been hypothesized that the lytic and latent pathways must diverge at a very early stage in the virus replication cycle (8, 46). If the pathways diverge very early, i.e., prior to DNA replication, it would follow that DNA replication enzymes, such as the viral thymidine kinase (TK), would have no effect on establishment. Indeed, TK null mutants do establish latent infections (4, 28, 56). Additional studies have demonstrated that following footpad inoculation the number of neurons expressing LAT RNAs was only slightly reduced (28), and LAT-positive neurons were also detected following corneal inoculation (4). These findings led to the current hypothesis that the viral TK (e.g., viral DNA replication in neurons) is required primarily for reactivation from the latent state and not for the establishment or maintenance of latency (for a review, see reference 55). The work of Maroglis et al. supports this hypothesis. When mice were infected by direct injection of virus into the sciatic nerve trunk, treatment with antiviral nucleoside analogs did not significantly reduce the amount of latent DNA detected in dorsal root ganglia (DRG) (29). However, the opposite conclusion was reached by Slobedman et al., who compared the establishment of latency by the HSV-1 strain SC16 and a TK-negative (TK-) derivative, TKDM21 (50). Following infection of the mouse flank, a comparison was made of the number of LAT+ neurons detected and the total amount of HSV-1 DNA that was present in the innervating DRG. The authors concluded that both input (unreplicated) and progeny (replicated) viral genomes contributed to the number of LAT-positive sites in DRG. Although not directly measured, their data suggested that viral DNA replication within neurons was required to establish latent sites that contain a high number of HSV-1 genomes (50). Thus, viral replication (or viral DNA replication) within ganglia may increase the number of latent sites established or increase the number of HSV-1 genomes contained within latently infected neurons.

The importance of this issue resides in the fact that current antiviral therapies for HSV target the viral TK and/or DNA polymerase (DNA Pol) and are based on the disruption of viral DNA replication. Drug-resistant mutants are rapidly selected in vivo but are thought to be at a disadvantage in the wild. It is not yet clear whether the increased use of such drugs will contribute to a significant increase of drug-resistant mutants within the human population. Understanding more fully the impact of the loss of TK function on the ability of the virus to establish latent infections and to reactivate will be of predictive value. Several recent advances suggested to us that the role of the viral TK, and viral DNA replication in ganglia, on the establishment of latency should be revisited. First, Horsburgh et al. demonstrated that TK-negative mutants are reactivation competent when the mutation resides in a clinical isolate (18). Thus, drug-resistant variants of HSV may establish latency, subsequently reactivate, and spread through the population. Second, assays to directly determine the number of neurons containing the viral genome and the viral genome copy number in these neurons have recently been developed (40, 41, 43, 60). Since the earlier studies relied on indirect measurements of establishment, a direct measure could provide additional insight. Third, although it has been recognized that the replication capacity of TK- mutants can vary depending on the site of inoculation (11, 12), the effect of variable surface replicative capacity on the establishment of latency and copy number profile has not previously been examined.

Mice were inoculated on either the snout (permissive for TK null mutants) or the cornea (less permissive) with 17/tBTK- or a genomically rescued variant, 17/tBRTK+. The number of latent infections established in the trigeminal ganglia (TG), and the HSV-1 genome copy number profile within those neurons, was determined. We conclude that while the increased viral replication at the body surface is correlated with a modestly increased frequency of latent sites, viral replication in sensory ganglia significantly contributes to the number of latent infections established. The latent infections that were established in mice infected on the cornea with the TK- mutant contained a universally low HSV-1 genome copy number. In contrast, latently infected neurons in mice infected on the snout contained a wild-type HSV genome copy profile, with some latent sites containing hundreds of HSV-1 genomes. It therefore appears that individual neurons can be infected by hundreds of virions originating at the body surface and still enter the latent state.


    MATERIALS AND METHODS
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

Cells and viruses. Rabbit skin cells (RSC) were cultured as previously described (61). The wild-type HSV-1 strain 17syn+ was originally obtained from John H. Subak-Sharpe at the Medical Research Council, Virology Unit, in Glasgow, Scotland, and subsequently plaque purified as described previously (61, 62). The TK- mutant 17/tBTK- was generated by inserting a simian virus 40 promoter-Escherichia coli beta-galactosidase (beta -Gal) minigene into the SnaBI site present in the TK gene at bp 47560 on the viral genome as previously detailed (36). HSV-1 genome base pair designations are as described by McGeoch and colleagues (30, 35). This insertion totally obliterated enzyme function as measured by antiviral drug resistance, plaque autoradiography, and a sensitive in vitro enzyme assay (36). The mutant 17/tBTK- was restored genomically and phenotypically to wild type by recombination with sequences spanning bp 45055 to 48634 on the viral genome to generate the rescued variant 17/tBRTK+ as described previously (59). The genomic structures of these virus isolates are shown schematically in Fig. 1. Virus stocks were generated by routine propagation in RSC monolayers. Infected cells were harvested and sonicated, and the titer of the stock was determined by serial-dilution plaque assay on RSC monolayers as previously described (61).


View larger version (30K):
[in this window]
[in a new window]
  		FIG. 1.   Genomic structures and transcriptional analysis. (Top) A schematic representation of the HSV-1 genome is shown with the long terminal and internal repeats (TRL and IRL) and internal and terminal short repeats (IRS and TRS) shown as shaded bars. Below, the region containing the TK gene is expanded with the ORFs for glycoprotein H (gH), TK, UL24, and UL25 denoted by open arrows. The TK ORF was disrupted by insertion of a beta -Gal minigene construct (36), indicated by the hatched arrow. (Bottom) Representative RNA blots. The blot on the left was hybridized with 32P-labeled sequences specific for the TK mRNA, as described in Materials and Methods. The similar blot on the right was hybridized to a probe specific for UL24 mRNA.

Transcription of TK and UL24 mRNA. Slightly subconfluent RSC monolayers in 10-cm-diameter dishes (Falcon) were infected at a multiplicity of infection (MOI) of 10 PFU/cell. Following adsorption for 1 h, the dishes were rinsed, fed with 10 ml of minimal essential medium supplemented with 5% newborn calf serum (Gibco-BRL), and incubated at 37°C for 5 h. Total RNA was isolated with Ultraspec (Biotex) according to the manufacturer's protocol. Ten micrograms of RNA was glyoxylated, electrophoresed, transferred to nylon membranes (GeneScreen), and probed as described previously (39).

Inoculation of Mice. Male outbred Swiss Webster mice (Harlan Laboratories) 4 to 5 weeks of age were used throughout these studies. The animals were housed in American Association for Laboratory Animal Care-approved quarters with unlimited access to food and water. The mice were anesthetized by intraperitoneal injection of sodium pentobarbital (Nembutal [50 mg/kg of body weight]). Both corneas were scarified or both sides of the snout were shaved and abraded to expose a 5-mm2 surface area as described previously (44, 45). A total inoculation of 2 × 105 PFU of 17/tBTK- or 17/tBRTK+ was applied to the cornea or snout.

Replication kinetics in vitro and in vivo. Single- and multistep replication kinetic analysis was performed on slightly subconfluent RSC monolayers following infection at a high (10 PFU/cell) or low (0.001 PFU/cell) MOI. The cells and media were harvested from triplicate cultures at 4, 8, 12, 18, and 24 h p.i. (high MOI) or 4, 24, 48, and 72 h p.i. (low MOI) and subjected to three freeze-thaw cycles. Virus titers were determined on RSC monolayers.

A total of 80 mice (20 per group) were inoculated as described above. At each of the indicated times p.i., three mice from each of the four infected groups were euthanized and the appropriate tissues were removed, snap frozen, and stored at -80°C. Each group of three pairs of eyes, snouts, or TG was homogenized in 1 ml of minimal essential medium supplemented with 5% newborn calf serum (Gibco-BRL), centrifuged for 5 min at 5,000 × g, and assayed for infectious virus titer on RSC monolayers. The remaining mice were maintained for latency studies as described below.

In an additional experiment, 50 mice were inoculated (10 to 15 per group). On day 4 p.i., the mice were euthanized and the relevant tissues were harvested from each mouse and placed in separate tubes. The tissues from each of five mice from each group were homogenized, and the viral titer was determined for each in order to confirm the replication profiles and assess the extent of variability of the individual mice within a group. The tissues from an additional five mice per group were used to assess the amount of viral DNA by PCR as described below. The remaining 10 mice were perfusion fixed for immunohistochemistry as described below.

Quantitative PCR detection of HSV DNA. Quantitative PCR was carried out as described previously (22). Briefly, mice were infected as described above, and DNA was extracted from the relevant tissues (snout whisker pads, eyes, or TG). An average of 20 µg (range, 12 to 27 µg) of DNA was obtained from each TG pair, 50 µg (range, 23 to 62 µg) from each eye pair, and 100 µg (range, 64 to 135 µg) from the snouts. Aliquots of DNA were assessed for intactness and concentration by 0.8% agarose gel electrophoresis and quantified by comparison to commercial lambda DNA standards (Gibco-BRL).

One hundred nanograms of sample DNA or, as a standard, 100 ng of mouse liver DNA spiked with known concentrations of the cloned HSV-1 EcoRI N fragment, was analyzed by PCR with two primer pairs. One primer set was specific for the HSV-1 TK gene, and the second set was specific for the single-copy mouse adipsin gene as described previously (22). Only 10 ng of sample DNA was analyzed for the 17/tBRTK+ samples and the 17/tBTK- snout samples, as the amount of HSV-1 DNA present in 100 ng was beyond the linear range of the assay. The reaction products were electrophoresed on 12% polyacrylamide gels, transferred to nylon filters (GeneScreen; NEN), and probed with 32P-labeled oligonucleotides internal to the amplification primers. The resulting blots were scanned in a Molecular Dynamics PhosphorImager for quantification with ImageQuant software (Molecular Dynamics). The total amount of HSV-1 DNA in each tissue was calculated by determining the number of HSV genomes in the tested aliquot and normalizing this number to the average amount of DNA obtained from that tissue.

Immunohistochemical detection of viral proteins. Groups of mice were inoculated on the snout with 2 × 105 PFU of either 17/tBTK- or 17/tBRTK+ as described above. On day 4 p.i., the animals were anesthetized by intraperitoneal injection of a 2.5-mg dose of sodium pentobarbital. The deeply anesthetized animals were perfusion fixed with 0.5% paraformaldehyde, and the TG were removed and postfixed for 30 min in a solution of 4% paraformaldehyde. The tissue was dehydrated through graded ethanols and embedded in paraffin following routine procedures. Eight-micron sections were cut with a rotary microtome, placed on negatively charged glass slides (Superfrost Plus; Sigma), and incubated in a dry oven for 12 h at 60°C. Paraffin was removed by incubation in xylene followed by a graded ethanol series.

Immunohistochemical localization of HSV proteins was carried out with a standard three-step biotin-avidin peroxidase assay as detailed previously (42). Endogenous peroxidase activity was blocked by a 20-min incubation in methanol containing 0.65% hydrogen peroxide. Sections were rinsed in phosphate-buffered saline and incubated in 5% nonfat dry milk and 0.25% NP-40 for 45 min and again rinsed in phosphate-buffered saline. The M.O.M. immunodetection kit (Vector Laboratories, Burlingame, Calif.) was used to reduce the background staining associated with the use of mouse monoclonal antibodies on mouse tissue. Sections incubated with monoclonal antibodies were processed with kit reagents according to the protocol provided. Monoclonal antibodies directed against ICP4 and ICP5 (VP5) were obtained from ABI (Columbia, Maryland) and utilized at a 1:200 dilution. Monoclonal antibody LP-1 (kindly provided by A. Minson) recognizes VP16, a late viral protein, and was used at a 1:400 dilution. A rabbit polyclonal antibody that recognizes multiple HSV proteins (Accurate) was utilized at a dilution of 1:2,000. Secondary reagents used with this polyclonal antibody included a biotinylated anti-rabbit antibody and an avidin-horseradish peroxidase conjugate (Vector). Visualization of antibody-biotin-avidin-horseradish peroxidase complexes deposited in the tissue was carried out by incubating rinsed slides in a solution of 0.1 M Tris (pH 8.0) containing 0.25 mg of diaminobenzidine (Aldrich)/ml and 0.04% hydrogen peroxide. The reaction was stopped by rinsing the slides in distilled water, and the slides were subsequently dehydrated and mounted with Permount (Sigma). The slides were viewed and photographed with an Olympus BX40 photomicroscope outfitted with an Olympus DP10 digital camera system.

Tissue-sampling design. Two experiments were performed. In the first, 10 ganglia from 17/tBTK--infected mice or 8 ganglia from 17/tBRTK+-infected mice were randomly oriented in a single paraffin block. Each block was serially sectioned, and five consecutive sections were placed on a single slide until all sections from the block were collected. This resulted in a total of 25 to 30 slides per block. Six nonconsecutive slides spanning the 17/tBRTK+ tissue block were examined; five were treated with the anti-HSV antibody, and a sixth slide was used as the no-primary-antibody control slide to assess background staining. Seven slides were examined in an identical manner from the 17/tBTK- tissue block. Thus, 40 and 60 ganglion profiles per slide were examined for 17/tBRTK+ and 17/tBTK-, respectively. The number of positively stained neurons (based on morphology) in each ganglion profile in a section from each of the slides was determined.

In a second experiment, blocks of paraffin-embedded ganglia (from 17/tBRTK+- and 17/tBTK--infected mice on day 4 p.i.) were again serially sectioned. A group of serial sections were placed in consecutive order on each of five slides such that slide 1 contained sections 1, 6, and 11, slide 2 contained sections 2, 7, and 12, etc. Four and five sets of these five-slide serially sectioned ganglia were generated for analysis of 17/tBRTK+- and 17/tBTK--infected ganglia, respectively. These sets of slides were used so that slides 1, 2, 3, 4, and 5 of each set were incubated with an anti-HSV, anti-ICP4, anti-ICP5, and LP-1 (anti-VP16) antibodies and no primary control, respectively. The staining of each antibody could then be examined on multiple serial sections from different regions of multiple ganglia. The total number of positively stained neurons was determined for each antibody on profiles of the same ganglia to avoid any bias of the regional variations in the number of infected cells. For 17/tBTK-, 60 ganglion profiles were counted for each antibody. Because the number of positive neurons was much greater in the 17/tBRTK+ ganglia, fewer profiles were counted.

Contextual analysis of latency. Mice were infected as described above and maintained for at least 30 days p.i. Two sets of three mice from each group were analyzed. Enriched neuron populations were obtained exactly as described previously (40, 41, 43, 60). Briefly, animals were anesthetized with sodium pentobarbital and perfused with Streck's tissue fixative. Fixed TG were dissociated into single-cell suspensions with collagenase, and the neurons were purified on Percoll (Pharmacia) gradients. The neurons were stained with Ponceau S and aliquoted into 200-µl PCR tubes. The tube contents were visualized microscopically; thus, the actual number of neurons analyzed in each sample is known. For the 17/tBRTK+ samples, only tubes containing a single neuron were employed. Preliminary analysis of the 17/tBTK- samples revealed a low frequency of positive neurons. Therefore, samples containing 10 neurons were employed as described previously (40). Following treatment with immobilized DNase (Mobitec), intracellular DNA was liberated with proteinase K and analyzed for the presence of the HSV-1 genome by the quantitative PCR assay developed by Katz et al. (22). The products were electrophoresed, blotted, probed with an internal 32P-labeled oligonucleotide, and quantified on a PhosphorImager with ImageQuant software as detailed elsewhere (40).


    RESULTS
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

Characterization of 17/tBTK- and 17/tBRTK+. The generation and characterization of 17/tBTK- was reported previously (36). In brief, a simian virus 40 promoter-beta -Gal reporter minigene was inserted into the SnaBI site at bp 47560 in the viral genome. This insertion disrupted the TK open reading frame (ORF) that spans bp 47802 to 46674 and eliminated all detectable viral TK activity as assayed by antiviral drug resistance, plaque autoradiography, and a sensitive in vitro enzymatic assay for the viral TK enzyme (36). A schematic representation of the genomic structure of 17/tBTK- is shown in Fig. 1 (top). 17/tBTK- was rescued to wild type by recombination with wild-type sequences spanning bp 45055 to 48634 on the viral genome to generate 17/tBRTK+ as described in Materials and Methods. This rescued variant was wild type in plaque morphology, replication in vitro and in vivo in neural tissues, and sensitivity to antiviral drugs (see below and data not shown).

The beta -Gal minigene in 17/tBTK- was inserted upstream of UL24, a viral gene that is partially antisense to TK and is required for efficient viral replication (6, 19, 21). UL24 mRNA is transcribed from three different promoters (23, 37), the most distal of which was disrupted by the insertion. To determine the effect of the insertion on UL24 mRNA expression, RSC monolayers were infected with 17/tBTK- or 17/tBRTK+ and employed as a source of mRNA for RNA blot analysis (Fig. 1, bottom). In the left panel, a probe specific for the TK mRNA was employed. No signal above background was detected in the 17/tBTK- lane. In the right panel, a similar blot was hybridized to a probe specific for UL24 mRNA. As predicted by the genomic structure of 17/tBTK-, the 1.4-kb UL24 mRNA was not produced by the mutant. This form of the UL24 mRNA is transcribed from an upstream promoter that was disrupted by the insertion in the TK gene. Normal levels of the 1.2-kb UL24 mRNA were produced by 17/tBTK-. This mRNA species, transcribed from the second of the three UL24 promoters, represents a minor portion of the total transcription from this locus (15). The 0.9-kb UL24 mRNA was also readily detectable in the 17/tBTK- RNA samples at levels similar to those of the wild type.

The total amount of UL24 mRNA produced by 17/tBTK- in RSC cells at 5 h p.i. was reduced by the loss of the 1.4-kb mRNA form. This could result in a reduced amount of UL24 protein. Null mutants at the UL24 locus form small plaques and replicate poorly in cultured cells (21). As detailed below, 17/tBTK- plaqued normally and replicated with wild-type efficiency in RSC monolayers and therefore did not display a UL24 null phenotype. This suggests that transcription of the 1.2-kb mRNA that encodes the entire UL24 ORF and/or the 0.9-kb mRNA (which produces a smaller version of UL24) was adequate to provide UL24 function.

Replication properties in vitro and in vivo. Replication kinetic experiments were performed under low- and high-MOI conditions in RSC monolayers as described previously (61). Both 17/tBTK- and 17/tBRTK+ replicated as efficiently as the parental strain, 17syn+, in the cells following low- or high-multiplicity infection (data not shown). To examine the replication properties of 17/tBTK- and 17/tBRTK+ in vivo, groups of mice were infected on the scarified cornea or the shaved and abraded snout as described above (44, 45).

Both 17/tBTK- and 17/tBRTK+ replicated to higher titers on the snout than on the eye (Fig. 2A). On the snout, the replication of the two isolates was equivalent for the first 4 days p.i., after which the titer of the TK- mutant dropped rapidly. 17/tBTK- was detectable through day 7 p.i. and was cleared by 9 days p.i. The titer of the TK+ isolate did not drop significantly through day 7 p.i. and was still about 104 PFU/snout on day 9 p.i. Integration of the areas under the curves revealed a two-fold decrease in the total PFU produced by 17/tBTK- through day 9 p.i.


View larger version (30K):
[in this window]
[in a new window]
  		FIG. 2.   (A) Replication kinetics in vivo. Groups of mice were infected with 2 × 105 PFU of either 17/tBTK- or 17/tBRTK+ on scarified corneas or abraded snouts as described in Materials and Methods. At the indicated times p.i., relevant tissues were harvested from three mice in each group and assayed for virus content. The total number of PFU present in the tissues is plotted. The areas under the curves were calculated with Prism statistical software. (B) Comparison of total PFU (squares) and total viral DNA (circles) of 17/tBRTK+ (solid symbols) and 17/tBTK- (open symbols) in eyes, snouts, and TG on day 4 p.i. Each symbol represents tissue from an individual animal. The viral titers were determined by standard plaque assay, and total viral DNA was determined by quantitative PCR as described in Materials and Methods.

On the eye, the titer of 17/tBTK- was reduced by ~12-fold by day 4 p.i., and no virus was detected on day 7 p.i. The TK+ variant replicated efficiently through day 9 p.i. A comparison of the areas under the curves revealed a >12-fold decrease in the replication of the mutant through day 9 p.i. The decreased replication efficiency of TK-negative mutants on the cornea has been previously documented (4, 57). Some reports have suggested that TK-negative viruses replicate with wild-type efficiency on the mouse eye (24, 25). It should be noted that these studies examined replication only on day 2 p.i. At day 2 p.i., the titers of 17/tBTK- and 17/tBRTK+ were equivalent, but the titer present in 17/tBTK--infected animals rapidly diminished after this time.

No infectious 17/tBTK- was detected in the TG following infection of the cornea. The TK+ rescued isolate replicated to 105 PFU in these ganglia (Fig. 2A). Replication of the TK- mutant in TG was detected following infection via the snout. In these ganglia, less than 50 PFU of 17/tBTK- (<8 PFU/ganglion tested) was present at day 4 p.i. It should be noted that to detect this virus it was necessary to plate the entire undiluted tissue homogenate on an RSC monolayer in a single 60-mm-diameter tissue culture plate. In other similar experiments, the replication of 17/tBTK- was not detected in ganglia following snout inoculation (data not shown). It was therefore unclear whether the replication detected in this experiment represented a limited amount of replication in all ganglia or productive infection in a subset of the ganglia. To address this question, additional animals were infected with the mutant or rescued virus. Total infectious virus or total viral DNA content was determined at day 4 p.i. in the relevant tissues of five animals per group. The results of this analysis are shown in Fig. 2B.

As can be seen in Fig. 2B, the variation in virus titer was not great in any tissue examined regardless of the infecting strain. Therefore, the infection of these animals was highly reproducible. It can also be seen that in most cases the amount of viral DNA in the tissue was directly correlated with the amount of infectious virus present. In the eyes and snouts, the ratio of HSV-1 genomes to PFU was 103 for both 17/tBTK- and 17/tBRTK+. As expected, on the eye, where the titer of infectious virus was reduced about 15-fold in mice infected with the TK- mutant, the amount of HSV-1 DNA detected was also reduced about 15-fold. Likewise, on the snout, where 17/tBTK- replicated about twofold less well than the wild-type, the amount of DNA present was also reduced about twofold.

The HSV DNA/PFU ratio was ~102 in TG infected with the wild type regardless of the infection route. This correlation between the infectious virus titer and the HSV-1 genome content was not seen in the TG infected with 17/tBTK-. No infectious virus was found in any of the 10 ganglia from mice infected on the cornea. These ganglia contained between 103 and 5 × 103 HSV-1 genomes. Likewise, only one of five pairs of ganglia from mice infected with 17/tBTK- by the snout contained detectable virus (total, 3 PFU). However, all of these ganglia contained significantly more viral DNA than those infected on the cornea (1.4 × 104 to 8.0 × 104 HSV genomes).

Analysis of viral protein expression in neurons. The preceding analysis demonstrated that there was about 10-fold more viral DNA present in the ganglia of mice infected by the snout with 17/tBTK- than in ganglia of mice infected on the cornea. Since viral replication was not detected in the majority of these TG, the origin of these additional genomes was not clear. The increased levels of viral DNA may have come from limited viral genome replication within the ganglia or from the increased transport and uptake of viral genomes from the snout, where far more virus was produced (Fig. 2). It is not currently practical to quantify viral genome replication on a cellular basis. Therefore, an analysis of the expression of viral proteins of defined kinetic classes was employed to investigate whether the genome of 17/tBTK- was replicated in a significant number of neurons.

Mice were infected with 17/tBTK- or 17/tBRTK+ on their abraded snouts, and on day 4 p.i. TG were harvested and processed for the immunohistochemical detection of viral proteins as described in Materials and Methods. As shown in Fig. 3A and B, there was a clear difference between the numbers of HSV antigen-expressing cells in 17/tBRTK+ ganglia and in the 17/tBTK- group. In ganglia from mice infected with 17/tBRTK+, viral-antigen-positive neurons and some associated satellite cells were widespread and frequently clustered (Fig. 3A). The 17/tBTK- ganglia were characterized by rare isolated antigen-positive neurons (Fig. 3B). The number of HSV antigen-expressing neurons was determined as described in Materials and Methods. There were 30-fold-fewer anti-HSV-positive neurons per ganglion profile in the 17/tBTK--infected ganglia than in the 17/tBRTK+-infected ganglia, an average of 1.4 ± 0.1 and 44 ± 2.7, respectively (P < 0.0001; unpaired t test) (Table 1).


View larger version (121K):
[in this window]
[in a new window]
  		FIG. 3.   Immunohistochemical detection of HSV-1 proteins in TG. Mice were infected with 2 × 105 PFU of 17/tBTK- or 17/tBRTK+ on their abraded snouts. Shown are representative sets of serial sections of 17/tBRTK+ (left-hand panels)- and 17/tBTK- (right-hand panels)-infected ganglia on day 4 p.i. stained with antibodies directed against HSV proteins (see Materials and Methods). (A and B) Low-power photomicrographs of anti-HSV-1 antibody-stained TG sections. The insets show higher magnification of the boxed areas. Also shown are the serial sections analyzed with monoclonal antibodies directed against ICP4 (C and D), ICP5 (E and F), and VP16 (G and H). While the sections shown in panels F and H contained no positive neurons, the insets show examples of the rare positive neurons in 17/tBTK--infected ganglia staining for ICP5 (F) and VP16 (H). (I and J) No primary antibody controls.

                              
View this table:
[in this window]
[in a new window]
  		TABLE 1.   Anti-HSV antibody positively staining neurons in 17/tBRTK+- and 17/tBTK--infected ganglia on day 4 p.i.

Serial sections of ganglia on day 4 p.i. were analyzed with the anti-HSV antibody and monoclonal antibodies directed against ICP4, ICP5, and VP16. Photomicrographs of sections representative of the results are shown in Fig. 3. The total number of neurons positively stained with each of the antibodies was determined on serial sections as detailed in Materials and Methods. These numbers are presented in Table 2. In both 17/tBRTK+- and 17/tBTK--infected ganglia, most of the neurons staining with the anti-HSV antibody also stained for ICP4 (Fig. 3C and D). In the 17/tBRTK+-infected ganglia, 44 and 35% of the anti-HSV-staining neurons expressed ICP5 and VP16, respectively (Table 2) (Fig. 3E and G). In contrast, while the number of ICP4 expressing neurons in the 17/tBTK--infected ganglia was equal to the number positive with anti-HSV, only extremely rare neurons expressed either ICP5 or VP16 (5 and 2%, respectively) (Table 2) (Fig. 3F and H). The differences in the relative percentages of positive neurons expressing ICP5 and VP16 in 17/tBRTK+ and 17/tBTK- were significant (P < 0.001; Fisher's exact test). The numbers presented in Table 2 were generated by counting all of the positive neurons on serial sections of the same ganglia for each antibody. This was done to avoid any bias that would be introduced by the regional variations within the ganglia of infected cells. Thus, the results are not dependent on the identical neurons being present in all of the serial sections of a given ganglion. Note that while sections shown in Fig. 3F and H contain no positive neurons, the insets show examples of the rare positive neurons in 17/tBTK--infected ganglia staining for ICP5 (Fig. 3F) and VP16 (Fig. 3H).

                              
View this table:
[in this window]
[in a new window]
  		TABLE 2.   Comparison of the relative frequencies of neurons expressing immediate-early; early; and late viral proteins in 17/tBRTK+- and 17/tBTK--infected ganglia on day 4 p.i.

Both ICP5 and VP16 are components of the HSV-1 virion and are expressed with leaky late kinetics. Very limited ICP5 and VP16 protein is produced in the absence of viral DNA replication (17, 66). These results indicate that the normal pattern of viral protein expression was disrupted in most 17/tBTK--infected neurons. Consistent with previous reports of replication of TK- mutants in neural tissue, these findings suggest that in most neurons the infection did not progress past the early stage of lytic viral gene expression (11, 24, 25).

Effect of the viral TK gene on percentage of neurons in which latency was established. A comparison was made of the efficiency of the establishment of latency following inoculation of 17/tBTK- or 17/tBRTK+ on the snout or on the cornea. Additional mice from the groups infected above were maintained for at least 30 days p.i. and processed for the quantification of latently infected neurons by utilizing contextual analysis of DNA (CXA-D) as described previously (40, 41, 43, 60). Representative blots are shown in Fig. 4. Two groups of three mice were analyzed for each virus and inoculation route. The numbers shown in Fig. 4 are totals from the individual groups analyzed. In all cases there was very close agreement among groups: eye inoculations with 17/tBTK-, 1.7% (9 neurons positive of 531 tested) and 1.9% (12/635) latently infected; eye inoculations with 17/tBRTK+, 22.5% (25 of 111) and 23.5% (32 of 136); snout inoculations with 17/tBTK-, 5.2% (15 of 291) and 4.8% (15 of 311); snout inoculations with 17/tBRTK+, 27.4% (25 of 91) and 31.8% (14 of 44).


View larger version (38K):
[in this window]
[in a new window]
  		FIG. 4.   Analysis of latent infections at the single-cell level. (Top) Flow diagram of CXA-D. (Bottom) Representative southern blots of the PCR products probed with a [32P]ATP-labeled oligonucleotide internal to the amplification primers. The first four lanes are standards containing known amounts of HSV-1 DNA, representing the indicated genome equivalents. Lane 5 is a buffer blank performed on cell-free supernatant from the purified neuron preparations. As expected, the majority of the neurons tested did not contain any viral genomes. The number of neurons positive per number tested and the percent infected neurons (PIN) are shown on the right.

Following infection on the cornea, 17/tBTK- established latency in 1.8% of the sensory neurons within the TG. Of the 1,166 neurons examined, only 21 were positive for the viral genome. The establishment of latency by the rescued virus 17/tBRTK+ was much greater, with 23.1% of the neurons harboring latent viral genome (57 neurons positive of 247 tested; P < 0.0001; Fisher's exact test). The frequency of latent infections established by 17/tBRTK+ was indistinguishable from that established by the wild-type strain, 17syn+, under similar conditions (41, 60).

The number of latently infected neurons was significantly greater in mice infected with 17/tBTK- on the snout than in those infected on the eye (P = 0.0003) but still less than the number established in mice infected with 17/tBRTK+. In 17/tBTK--infected animals, 5% of the TG neurons harbored latent virus (30 positive of 602 tested). In the 17/tBRTK+-infected animals, 29% of the TG neurons were latently infected (P < 0.0001). The replication of 17/tBTK- on the snout was reduced only about twofold compared to that of 17/tBRTK+, and the percentage of latently infected neurons was ~6-fold reduced. This suggests that viral replication within the TG contributes significantly to the pool of latent neurons. In addition, these data suggest that the extent of replication at the body surface also impacts the number of latent sites established within the TG.

Analysis of the HSV-1 genome copy number profile within individual latently infected neurons. With the CXA-D approach the number of viral genomes present in individual latently infected neurons can be determined to generate a profile of the latent HSV-1 genome copy number (40, 43, 60). The amount of radioactivity present in each positive sample discussed above was determined with ImageQuant software and compared to that of a set of standard reactions run simultaneously as described previously (40). The results are presented as a scattergram in Fig. 5. Following corneal inoculation with 17/tBTK-, all positive neurons contained very few HSV-1 genomes (average, 2.3 genomes/neuron; range, 1 to 9). By comparison, those animals infected with 17/tBRTK+ displayed a wild-type HSV-1 genome copy number profile (average, 28.4 genomes/neuron; P < 0.0001; Mann-Whitney test).


View larger version (15K):
[in this window]
[in a new window]
  		FIG. 5.   Scattergram of the HSV-1 genome copy number in individual latently infected neurons. The positive signals from CXA-D analysis of latently infected neurons were quantified with ImageQuant software and compared to the signals from the standards on each blot to determine the number of HSV-1 genomes detected. Each point on the scattergram represents the number of HSV-1 genomes in an individual latently infected neuron. The horizontal line in each data column is the mean value of the points in the column.

Of interest, the copy number profiles in mice infected via the snout with 17/tBTK- or 17/tBRTK+ were indistinguishable. In 17/tBTK--infected mice, latent sites contained between 1 and 250 latent genomes, with an average of 35.9 genomes/neuron. The genome copy numbers in 17/tBRTK+ latently infected neurons ranged between 1 and 300, with an average of 33.8 (P = 0.8224; Mann-Whitney test). TK- mutants do not efficiently replicate their genomes in ganglia (25). While limited 17/tBTK- genome replication within neurons that survive cannot be absolutely ruled out, the immunohistochemical analysis presented above demonstrated that if this occurred it was a rare event. Even so, viral DNA replication in rare neurons could not result in a normal latent HSV-1 genome copy number profile. Therefore, the multiple copies of the 17/tBTK- genome found in most of these latent sites most likely were transported to the neurons from the surface. This suggests that the efficiency of viral replication at the body surface has a major impact on the latent HSV-1 genome copy number, at least in the context of a TK null mutant.


    DISCUSSION
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

Our results support the following conclusions. First, the viral TK gene is required for the establishment of latent infections at wild-type frequency. There were 10-fold-fewer latent infections in the TG of mice infected with 17/tBTK- on the cornea compared to the number established in the ganglia of mice infected with an equivalent titer of the rescued isolate, 17/tBRTK+. In mice infected on the snout, the TK-negative mutant established latency in sixfold-fewer neurons. Second, viral replication within the TG is required to establish latent infections at wild-type frequency. While 17/tBTK- replicated significantly less well on the mouse cornea, its replication on the mouse snout was nearly equivalent to that of 17/tBRTK+. This increased surface replication resulted in a 2.5-fold increase in the number of latent sites. This number was still significantly less than the number in ganglia infected with 17/tBRTK+, which replicated in the TG as well as on the surface. Third, viral replication within ganglia, and in particular viral DNA replication within sensory neurons, is not required to establish latent sites that contain many copies of the HSV-1 genome. Following infection on the snout, 17/tBTK- established a latent pool with a normal HSV-1 DNA genome copy number profile. Some of these neurons contained hundreds of copies of the HSV-1 genome. However, an analysis of viral replication within the TG, and of viral protein expression in neurons, demonstrated that DNA replication did not occur in most of the cells that expressed viral protein. In addition, the number of cells that expressed any detectable viral proteins at the peak of lytic infection was several orders of magnitude less than the number of cells in which latency was established.

The viral TK gene is required for the efficient establishment of latency. From the earliest studies with viral TK-negative or -deficient mutants it was apparent that TK is required for reactivation from latency. An early hypothesis suggested that TK was absolutely required for the establishment of latent infections (55). However shortly after the LAT RNAs were discovered, several reports proved that TK-negative mutants could establish latent infections as measured by the detection of LATs by in situ hybridization (4, 28, 56). Work by Leist and colleagues suggested that TK-negative mutants could establish latent infections with near-wild-type efficiency in mouse DRG (28). Coen et al. demonstrated that TK null mutants could establish latent infections in mice following corneal inoculation (4).

Based on these observations it has been hypothesized that the role of TK in latency is primarily to facilitate reactivation and not to enhance the establishment of latency (for a review, see reference 55). However, our finding that TK is required for the efficient establishment of latency is consistent with several other studies. Leist et al. found a reduced number of LAT-positive DRG neurons in mice infected with a TK mutant (28), as did Slobedman et al. (50). In mouse TG, Katz and colleagues have also demonstrated a 100-fold decrease in the amount of latent HSV DNA present in mice latently infected with the TK-negative mutant (22). This is consistent with the 10-fold reduction in the frequency of latent sites multiplied by the 10-fold reduction in genome copy number we report here.

Viral replication on the surface and its relationship to the number of latent sites established. These results are best viewed in the context of the physical relationship between the inoculation site and the site of latency. The axonal endings of sensory neurons project to the body surface, where many display significant dendritic branching (5, 7, 58). This network of neuronal endings is the portal for HSV to the neuronal cell bodies, where latency is established (for a review, see references 13 and 64). Therefore, it is likely that parameters that determine the number of infected neurons within the ganglia include (i) the total area of viral replication at the body surface, (ii) the efficiency of virion production, and (iii) the innervation density. Thus, the correlation between surface replication and the number of latent infections established is not surprising if more efficient replication results in a greater area of virus production and/or an increased density of virions at the site of infection. However, it should be noted that with the rescued isolate, 17/tBRTK+, the 100-fold increase in virion production on the snout compared to that on the cornea resulted in only a modest increase in the percentage of latently infected cells (from 23 to 28%). It seems most likely that replication within the TG is a dominant factor in determining the number of latent sites established. It is also possible that virus entry into the nervous system is not the same at these two inoculation sites.

Viral replication within ganglia and its relationship to the number of latent sites established. The mechanism underlying the correlation between viral replication in the ganglia and the increased number of latent sites is not straightforward. It is possible that lateral spread of virus within the ganglia increases the number of latent sites. While there is evidence that satellite and support cells are relatively nonpermissive and may limit spread within ganglia (16, 65), the focal distribution of acutely infected neurons in ganglia (Fig. 3) is consistent with such spread. This is a likely potential mechanism for an increase in the number of infected neurons.

Ganglionic viral replication may also serve to increase the area of the body surface that supports virus replication. It is readily apparent from the zosteriform spread model that lytic replication within neurons results in anterograde transport (32, 38, 47, 48). The arborizing nature of sensory innervation at the surface could result in the deposition of virions at a surface site near, but spatially distinct from, the initial infection site (5, 7, 58). Alternatively, lateral spread to new neurons within ganglia and subsequent anterograde transport would result in expansion of the area of surface to include additional surface regions innervated by the infected ganglia. These new sites of infection can be spatially quite distant from the initial infection site. For example, evidence has been presented that following infection of the mouse snout, virus is transported through the nervous system and infects the eye (2, 31). Replication at such new sites of infection and subsequent cell-to-cell spread at the surface would lead to viral uptake into the neurons that innervate the extended surface area. Repetitive cycles of neuronal infection and spread at the surface would be expected until specific immune mechanisms become inhibitory (31, 47). Thus, it is likely that an increase in the infected surface area mediated by ganglionic viral replication plays a major role in increasing the number of latent sites established. In support of this we have noted that increasing the area of the inoculation site at the surface increases the number of both lytically and latently infected neurons within TG (unpublished observations).

Viral thymidine kinase-mediated DNA replication and its relationship to HSV latent genome copy number. In the cornea infection model, the mutant 17/tBTK- established latent infections of universally low HSV-1 genome copy number (<10 genomes/latently infected neuron), with an average of 2.3 genomes/latent site. In contrast, 17/tBRTK+ established latent sites that contained between 1 and 1,000 HSV genome copies, with an average of 28.4. Importantly, the HSV-1 genome copy number profile of 17/tBTK- and 17/tBRTK+ were indistinguishable following snout inoculation. Consistent with previous reports (11, 24, 25), infectious virus was not detected in most ganglia of 17/tBTK--infected mice and when detected it was just a few PFU. In addition, 1,000-fold less viral DNA was present in these ganglia during the acute phase of infection (25) (Fig. 2B). The possibility that some very limited replication of the 17/tBTK- genome occurred in neurons that subsequently entered the latent state can not be absolutely ruled out. Indirect evidence that some neurons which express detectable viral lytic-phase proteins may survive to enter latency has been presented (49). However, our analysis of viral protein expression in neurons strongly suggested that the 17/tBTK- genome was not replicated in all but a very few neurons. HSV viral-protein-positive neurons were comparatively rare, and very few neurons positive for ICP5 or VP16 were detected (Fig. 3 and Tables 1 and 2). Therefore, the most likely origin of the multiple genome copies present in neurons latently infected with 17/tBTK- is that they were generated at the surface and transported to neurons.

This conclusion is different from that drawn by Slobedman et al. (50), who concluded that DNA replication within the DRG was required to establish latent sites with high genome copy numbers. The number of HSV-1 genomes in latent sites was not actually measured in their study, but rather an average number was estimated based on the number of LAT-positive sites and the total amount of viral DNA detected in the ganglia (50).

The difference in replication on the eye of 17/tBTK- and 17/tBRTK+ occurred after day 2 p.i. This suggests that viral replication at the surface after this time is important to increase the number of latent sites established and, importantly, to increase the HSV-1 copy number within these latent sites. A hypothesis as to when and how high-genome-copy-number latent sites are established and/or qualitatively modified in terms of viral DNA content must therefore account for the importance of surface viral replication after day 2 p.i.

One factor that could regulate the outcome of a neuron's encounter with HSV is the MOI. Although unproven, it seems likely those neurons infected by a large number of virions would be prone to enter lytic replication, whereas those receiving one or a few virus particles might enter the latent state. The HSV-1 LAT gene has been shown to be important for the establishment of latency (44, 60) and may function to downregulate viral gene expression (1, 44). Neurons that initially enter the latent pathway with a few viral genome copies may begin to express the LAT locus. The entry of virus into a given neuron can be protracted over many days. Thus, these LAT-positive (and therefore less permissive) neurons may be superinfected, resulting in an increased number of latent genomes. Since the replication of 17/tBTK- on the mouse eye rapidly diminished after day 2, the probability of superinfection was reduced. The more productive and prolonged replication of this mutant on the snout may have permitted superinfection and a resulting increase in latent genome copies. This hypothesis predicts that LAT null mutants would establish latent infections that contain reduced viral genome copy numbers. Our preliminary findings support this hypothesis.


    ACKNOWLEDGMENTS

This work was supported by Public Health Service grant AI32121 from the National Institute of Allergy and Infectious Diseases.

We thank Anthony Minson for providing antibody LP-1 and C. Tansky and S. Goins for technical assistance.


    FOOTNOTES

* Corresponding author. Mailing address for R. L. Thompson: University of Cincinnati Medical Center, Department of Molecular Genetics, Microbiology, and Biochemistry, 231 Bethesda Ave., Cincinnati, OH 45267-0524. Phone: (513) 558-0063. Fax: (513) 558-8474. E-mail: Richard.Thompson{at}UC.edu. Mailing address for N. M. Sawtell: Division of Infectious Diseases, Children's Hospital Medical Center, 3333 Burnet Ave., Cincinnati, Ohio 45229-3039. Phone: (513) 636-7880. Fax: (513) 636-7655. E-mail: Sawtn0{at}CHMCC.org.


    REFERENCES
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

1. Chen, S. H., M. F. Kramer, P. A. Schaffer, and D. M. Coen. 1997. A viral function represses accumulation of transcripts from productive-cycle genes in mouse ganglia latently infected with herpes simplex virus. J. Virol. 71:5878-5884[Abstract].
2. Claoue, C. 1986. A new look at experimental herpes simplex eye disease: preliminary results of clinical disease in the NIH mouse after zosteriform spread. Trans. Ophthalmol. Soc. U.K. 105:401-403.
3. Clements, G. B., and N. D. Stow. 1989. A herpes simplex virus type 1 mutant containing a deletion within immediate early gene 1 is latency-competent in mice. J. Gen. Virol. 70:2501-2506[Abstract/Free Full Text].
4. Coen, D. M., M. Kosz-Vnenchak, J. G. Jacobson, D. A. Leib, C. L. Bogard, P. A. Schaffer, K. L. Tyler, and D. M. Knipe. 1989. Thymidine kinase-negative herpes simplex virus mutants establish latency in mouse trigeminal ganglia but do not reactivate. Proc. Natl. Acad. Sci. USA 86:4736-4740[Abstract/Free Full Text].
5. Coggeshall, R. E., and S. M. Carlton. 1998. Ultrastructural analysis of NMDA, AMPA, and kainate receptors on unmyelinated and myelinated axons in the periphery. J. Comp. Neurol. 391:78-86[CrossRef][Medline].
6. Cook, W. J., K. K. Wobbe, J. Boni, and D. M. Coen. 1996. Regulation of neighboring gene expression by the herpes simplex virus type 1 thymidine kinase gene. Virology 218:193-203[CrossRef][Medline].
7. Cruz, M. C., D. Jeanmonod, K. Meier, and H. Van der Loos. 1984. A silver and gold technique for axons and axon-bundles in formalin-fixed central and peripheral nervous tissue. J. Neurosci. Methods 10:1-8[CrossRef][Medline].
8. Dobson, A. T., T. P. Margolis, F. Sedarati, J. G. Stevens, and L. T. Feldman. 1990. A latent, nonpathogenic HSV-1-derived vector stably expresses beta-galactosidase in mouse neurons. Neuron 5:353-360[CrossRef][Medline].
9. Ecob-Prince, M. S., C. M. Preston, F. J. Rixon, K. Hassan, and P. G. Kennedy. 1993. Neurons containing latency-associated transcripts are numerous and widespread in dorsal root ganglia following footpad inoculation of mice with herpes simplex type 1 mutant in 1814. J. Gen. Virol. 74:985-994[Abstract/Free Full Text].
10. Ecob-Prince, M. S., F. J. Rixon, C. M. Preston, K. Hassan, and P. G. Kennedy. 1993. Reactivation in vivo and in vitro of herpes simplex virus from mouse dorsal root ganglia which contain different levels of latency-associated transcripts. J. Gen. Virol. 74:995-1002[Abstract/Free Full Text].
11. Efstathiou, S., S. Kemp, G. Darby, and A. C. Minson. 1989. The role of herpes simplex virus type 1 thymidine kinase in pathogenesis. J. Gen. Virol. 70:869-879[Abstract/Free Full Text].
12. Field, H. J., J. R. Anderson, and P. Wildy. 1982. Atypical patterns of neural infection produced in mice by drug- resistant strains of herpes simplex virus. J. Gen. Virol. 59:91-99[Abstract/Free Full Text].
13. Fields, B. N., D. M. Knipe, and P. M. Howley. 1996. Fields virology, 3rd ed. Lippincott-Raven Publishers, Philadelphia, Pa.
14. Friedrich, A., J. P. Kleim, and K. E. Schneweis. 1990. Detection of latent thymidine kinase-deficient herpes simplex virus in trigeminal ganglia of mice using the polymerase chain reaction. Arch. Virol. 113:107-113[CrossRef][Medline].
15. Hann, L. E., W. J. Cook, S. L. Uprichard, D. M. Knipe, and D. M. Coen. 1998. The role of herpes simplex virus ICP27 in the regulation of UL24 gene expression by differential polyadenylation. J. Virol. 72:7709-7714[Abstract/Free Full Text].
16. Hill, T. J., and H. J. Field. 1973. The interaction of herpes simplex virus with cultures of peripheral nervous tissue: an electron microscopic study. J. Gen. Virol. 21:123-133[Abstract/Free Full Text].
17. Honess, R. W., and B. Roizman. 1974. Regulation of herpesvirus macromolecular synthesis. I. Cascade regulation of the synthesis of three groups of viral proteins. J. Virol. 14:8-19[Abstract/Free Full Text].
18. Horsburgh, B. C., S. H. Chen, A. Hu, G. B. Mulamba, W. H. Burns, and D. M. Coen. 1998. Recurrent acyclovir-resistant herpes simplex in an immunocompromised patient: can strain differences compensate for loss of thymidine kinase in pathogenesis? J. Infect. Dis. 178:618-625[Medline].
19. Jacobson, J. G., S. H. Chen, W. J. Cook, M. F. Kramer, and D. M. Coen. 1998. Importance of the herpes simplex virus UL24 gene for productive ganglionic infection in mice. Virology 242:161-169[CrossRef][Medline].
20. Jacobson, J. G., D. A. Leib, D. J. Goldstein, C. L. Bogard, P. A. Schaffer, S. K. Weller, and D. M. Coen. 1989. A herpes simplex virus ribonucleotide reductase deletion mutant is defective for productive acute and reactivatable latent infections of mice and for replication in mouse cells. Virology 173:276-283[CrossRef][Medline].
21. Jacobson, J. G., S. L. Martin, and D. M. Coen. 1989. A conserved open reading frame that overlaps the herpes simplex virus thymidine kinase gene is important for viral growth in cell culture. J. Virol. 63:1839-1843[Abstract/Free Full Text].
22. Katz, J. P., E. T. Bodin, and D. M. Coen. 1990. Quantitative polymerase chain reaction analysis of herpes simplex virus DNA in ganglia of mice infected with replication-incompetent mutants. J. Virol. 64:4288-4295[Abstract/Free Full Text].
23. Kibler, P. K., J. Duncan, B. D. Keith, T. Hupel, and J. R. Smiley. 1991. Regulation of herpes simplex virus true late gene expression: sequences downstream from the US11 TATA box inhibit expression from an unreplicated template. J. Virol. 65:6749-6760[Abstract/Free Full Text].
24. Kosz-Vnenchak, M., D. M. Coen, and D. M. Knipe. 1990. Restricted expression of herpes simplex virus lytic genes during establishment of latent infection by thymidine kinase-negative mutant viruses. J. Virol. 64:5396-5402[Abstract/Free Full Text].
25. Kramer, M. F., S. H. Chen, D. M. Knipe, and D. M. Coen. 1998. Accumulation of viral transcripts and DNA during establishment of latency by herpes simplex virus. J. Virol. 72:1177-1185[Abstract/Free Full Text].
26. Kramer, M. F., and D. M. Coen. 1995. Quantification of transcripts from the ICP4 and thymidine kinase genes in mouse ganglia latently infected with herpes simplex virus. J. Virol. 69:1389-1399[Abstract].
27. Leib, D. A., D. M. Coen, C. L. Bogard, K. A. Hicks, D. R. Yager, D. M. Knipe, K. L. Tyler, and P. A. Schaffer. 1989. Immediate-early regulatory gene mutants define different stages in the establishment and reactivation of herpes simplex virus latency. J. Virol. 63:759-768[Abstract/Free Full Text].
28. Leist, T. P., R. M. Sandri-Goldin, and J. G. Stevens. 1989. Latent infections in spinal ganglia with thymidine kinase-deficient herpes simplex virus. J. Virol. 63:4976-4978[Abstract/Free Full Text].
29. Margolis, T. P., F. Sedarati, A. T. Dobson, L. T. Feldman, and J. G. Stevens. 1992. Pathways of viral gene expression during acute neuronal infection with HSV-1. Virology 189:150-160[CrossRef][Medline].
30. McGeoch, D. J., M. A. Dalrymple, A. J. Davison, A. Dolan, M. C. Frame, D. McNab, L. J. Perry, J. E. Scott, and P. Taylor. 1988. The complete DNA sequence of the long unique region in the genome of herpes simplex virus type 1. J. Gen. Virol. 69:1531-1574[Abstract/Free Full Text].
31. Mester, J. C., J. C. Glorioso, and B. T. Rouse. 1991. Protection against zosteriform spread of herpes simplex virus by monoclonal antibodies. J. Infect. Dis. 163:263-269[Medline].
32. Nicholls, S. M., and W. A. Blyth. 1989. Quantification of herpes simplex virus infection in cervical ganglia of mice. J. Gen. Virol. 70:1779-1788[Abstract/Free Full Text].
33. Perng, G. C., K. Chokephaibulkit, R. L. Thompson, N. M. Sawtell, S. M. Slanina, H. Ghiasi, A. B. Nesburn, and S. L. Wechsler. 1996. The region of the herpes simplex virus type 1 LAT gene that is colinear with the ICP34.5 gene is not involved in spontaneous reactivation. J. Virol. 70:282-291[Abstract].
34. Perng, G. C., R. L. Thompson, N. M. Sawtell, W. E. Taylor, S. M. Slanina, H. Ghiasi, R. Kaiwar, A. B. Nesburn, and S. L. Wechsler. 1995. An avirulent ICP34.5 deletion mutant of herpes simplex virus type 1 is capable of in vivo spontaneous reactivation. J. Virol. 69:3033-3041[Abstract].
35. Perry, L. J., and D. J. McGeoch. 1988. The DNA sequences of the long repeat region and adjoining parts of the long unique region in the genome of herpes simplex virus type 1. J. Gen. Virol. 69:2831-2846[Abstract/Free Full Text].
36. Pyles, R. B., and R. L. Thompson. 1994. Mutations in accessory DNA replicating functions alter the relative mutation frequency of herpes simplex virus type 1 strains in cultured murine cells. J. Virol. 68:4514-4524[Abstract/Free Full Text].
37. Read, G. S., J. A. Sharp, and W. C. Summers. 1984. In vitro and in vivo transcription initiation sites on the TK-encoding BamHI Q fragment of HSV-1 DNA. Virology 138:368-372[CrossRef][Medline].
38. Robinson, T. W., and J. R. Dover. 1972. Experimental zosteriform herpes simplex virus infection in mouse skin. Br. J. Dermatol. 86:40-48[Medline].
39. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
40. Sawtell, N. M. 1997. Comprehensive quantification of herpes simplex virus latency at the single-cell level. J. Virol. 71:5423-5431[Abstract].
41. Sawtell, N. M. 1998. The probability of in vivo reactivation of herpes simplex virus type 1 increases with the number of latently infected neurons in the ganglia. J. Virol. 72:6888-6892[Abstract/Free Full Text].
42. Sawtell, N. M., and J. L. Lessard. 1989. Cellular distribution of smooth muscle actins during mammalian embryogenesis: expression of the alpha-vasculor but not the gamma-enteric isoform in differentiating striated myocytes. J. Cell Biol. 109:2929-2937[Abstract/Free Full Text].
43. Sawtell, N. M., D. K. Poon, C. S. Tansky, and R. L. Thompson. 1998. The latent herpes simplex virus type 1 genome copy number in individual neurons is virus strain specific and correlates with reactivation. J. Virol. 72:5343-5350[Abstract/Free Full Text].