Previous Article | Next Article ![]()
Journal of Virology, January 2004, p. 520-523, Vol. 78, No. 1
0022-538X/04/$08.00+0 DOI: 10.1128/JVI.78.1.520-523.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Department of Microbiology and Immunology, College of Medicine, National Cheng Kung University, Tainan, Taiwan 70101, Republic of China,1 Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, Massachusetts 021152
Received 7 July 2003/ Accepted 17 September 2003
|
|
|---|
|
|
|---|
Truly TK- mutants derived from standard laboratory strains ordinarily do not reactivate from explanted murine ganglia (4, 30, 31) (although some derived from clinical isolates can reactivate inefficiently [7, 11]). This has generally been ascribed to a requirement for TK during reactivation (4, 6, 19, 24, 29-31). However, in the case of wt HSV and certain mutants, reactivation frequencies correlate with the numbers of viral genomes or latently infected neurons in ganglia or both (20, 21, 25, 26). Indeed, establishment of latency by TK- mutants is less efficient than that of wt virus, as evidenced by a 6- to 50-fold reduction in total viral genomes, viral genome-positive neurons, total LATs, or LAT-positive neurons (1, 4, 14-16, 27, 33). Accordingly, the idea that TK is primarily required for reactivation rather than for establishment of latency has been called into question (33).
To address this issue, we first investigated whether the number of viral genomes per latent ganglion (latent viral load) established by a TK- mutant is lower than the minimum latent viral load that permits wt virus to reactivate. Seven- to eight-week-old male CD-1 (Charles River Laboratories) or CD-1-derived Hsd:ICR (Harlan Sprague-Dawley) mice (similar results were obtained with mice from both suppliers) were anesthetized and inoculated on the eye via corneal scarification as previously described (18), with various doses of the wt strain KOS, ranging from 2 x 101 to 2 x 106 PFU, or with 2 x 106 PFU of dlsptk (4), a KOS-derived TK deletion mutant. Titers of virus at the site of inoculation and ganglia were assayed by swabbing eyes 1 day postinfection (p.i.) and by excising and homogenizing ganglia 3 days p.i. As shown in Fig. 1B, KOS did not detectably replicate on the eye when the inoculum was 2 x 101 or 2 x 102 PFU per eye. Viral replication was observed in 5 of 9 mouse eyes (55%) and 8 of 15 ganglia (53%) at the inoculum of 2 x 103 PFU per eye. At higher inoculating doses, KOS replicated in >90% of mouse eyes and ganglia. These data are similar to those obtained previously with KOS (9, 18). The TK- mutant, dlsptk, replicated to wt titers at 1 day p.i. in the mouse eye but was severely impaired for replication in ganglia, as previously observed (1, 4, 11, 16, 31).
![]() View larger version (35K): [in a new window] |
FIG. 1. Relationship between inoculum dose, reactivation, acute HSV replication, latent viral load, and LAT expression. Mice were infected with the doses of KOS, dlsptk, or tkLTRZ1 as indicated on the x axis. (A) Virus reactivation was assayed 30 days p.i., and the number of reactivations/number of ganglia assayed are shown. The reactivation percentages are shown in parentheses. (B) Titers of virus in tear film (open triangle) and ganglia (open diamond) were assayed in samples obtained by swabbing eyes 1 day p.i. and by excising and homogenizing ganglia 3 days p.i. Each point on each scattergram represents an individual sample. (C) Thirty days p.i., ganglia were harvested and subjected to quantitative PCR assays to determine the numbers of viral genomes per ganglion. Each circle on the scattergram represents log10 viral DNA per ganglion of an individual sample. (D) Selected ganglia samples were assayed for LATs using quantitative reverse transcription-PCR. Each square on the scattergram represents log10 LATs per ganglion value of an individual sample. In panels B and C, bars represent mean values for each group. In panels C and D, the number of samples assayed for each group is shown in parentheses and the dashed line delineates the detection limit for each assay. N.D., not done.
|
Because our assay for HSV DNA had a limit of detection of 10 viral genomes per ganglion (due to our sampling only 10% of the ganglion), we sought to increase sensitivity of detection of latent HSV by measuring LATs, since there are
105 LATs per viral genome in latently infected ganglia (2, 3, 16, 17). Using a quantitative reverse transcription-PCR assay for LATs as described previously (2, 3, 16, 17), we detected LATs only in ganglia in which we also detected viral DNA, even when six more cycles of PCR amplification were applied to the LAT assay (data not shown). Thus, we detected latent DNA and LATs in the same proportion of ganglia at each inoculating dose (Fig. 1C and D). At doses of 2 x 103 and above, the proportion of ganglia from which virus reactivated was similar to that in which latent HSV was detected molecularly. At a dose of 2 x 102, although there were a few ganglia in which we could detect HSV molecularly, no reactivation was observed. This difference is most likely due to chance (P = 0.162, Fisher's exact test). Similar to the viral load, the amount of LATs in latently infected ganglia increased with the dose of virus in inocula until the dose of 2 x 105 PFU, at which dose a plateau was reached. (The surprisingly high level of LATs in the ganglia latently infected with 2 x 105 PFU was unexpected, and the reason is unknown).
In contrast, following inoculation with dlsptk at 2 x 106 PFU per eye, although no reactivation was observed in 30 ganglia, 16 of 16 (100%) ganglia contained viral DNA and LATs (Fig. 1C and D). Moreover, the average latent viral load was 12-fold higher in these ganglia than in ganglia latently infected with KOS at 2 x 103 PFU per eye, in which reactivation was detected at a frequency of 50%. Therefore, the failure of dlsptk to reactivate is not due to its failure to establish latency as measured by the latent viral load.
We also investigated whether increasing the latent viral load of a TK- mutant to levels similar to that where wt virus reactivates efficiently would overcome the block to reactivation. In an initial experiment, we used the immunosuppression protocol of Halford and Schaffer (8, 9), which enabled HSV ICP0 mutants both to attain wt latent viral loads and to increase their reactivation frequency in mouse ganglia. However, this method increased neither the acute replication of dlsptk in six ganglia we titrated, the reactivation in 12 ganglia we assayed (similar to the results of a previous study) (24), nor the latent viral load in 14 ganglia we analyzed (data not shown). We therefore turned to inoculation with high doses of virus, which has previously resulted in wt levels of spontaneous reactivation of an ICP34.5 null mutant in rabbits (23) and low levels of reactivation from explanted mouse ganglia by a TK deletion mutant derived from a clinical isolate (11). We inoculated mice with 2 x 108 PFU of dlsptk or tkLTRZ1, a KOS-derived mutant with the Escherichia coli lacZ gene inserted into the tk coding region (5), per eye. We used this mutant, in part, because dlsptk has recently been shown to express lower levels of the UL24 protein, which is encoded mainly by long transcripts whose start sites lie within the deletion of this virus (22). Although these lower levels of UL24 protein evidently do not affect viral replication in cell culture or in the mouse eye, they could conceivably affect replication or reactivation in mouse ganglia. The insertion in tkLTRZ1 is well upstream of the start site for this transcript. We tested the expression of the UL24 protein by tkLTRZ1 in Vero cells using Western blot analysis with a rat polyclonal antiserum as previously described (22), and it was very similar to that of the wt at both 5 and 10 h p.i. (Fig. 2, top panel). As a control the blot was probed with an anti-glycoprotein D (gD) monoclonal antibody (Fitzgerald), which verified that similar levels of infection were attained and similar amounts of protein were loaded and transferred (bottom panel).
![]() View larger version (56K): [in a new window] |
FIG. 2. Expression of UL24 protein from tkLTRZ1 virus. Vero cells were either mock infected (M) (lane 1) or infected at a MOI of 10 with KOS (lanes 2 and 3) or tkLTRZ1 (lanes 4 and 5). Total cell lysates were harvested at the indicated hours postinfection, resolved on a sodium dodecyl sulfate-polyacrylamide gel, and analyzed by Western blotting with anti-UL24 serum (top panel). As a loading control the membrane was subsequently stripped and incubated with a monoclonal antibody directed against gD (bottom panel). The positions of UL24 and gD are indicated to the right of the panels.
|
Our having achieved relatively high latent viral loads with TK- mutants is akin to results in a previous report (33) that showed that one such mutant could achieve a wt profile of latent genome copy numbers per neuron. There were, however, several differences between the two studies: (i) the earlier study did not examine reactivation; (ii) we used high-dose inoculation on the cornea rather than combined snout and corneal inoculation; and (iii) we used a TK- mutant that does not exhibit reduced UL24 expression.
In this study, we found that TK- mutants did not reactivate from latency even when their latent viral loads were comparable to those that permit relatively efficient reactivation of wt virus. This contrasts with, for example, ICP0 mutants that can reactivate from latency, even when latent viral loads are severalfold lower than those of the wt (8). For tkLTRZ1, the inability to reactivate is clearly due to the tk mutation, because rescue of the mutation restores reactivation (12) and there are no mutations other than the engineered lacZ insertion in the BamHI fragment corresponding to the fragment used in marker rescue (A. Griffiths and D. M. Coen, unpublished results; GenBank accession no. AY326452 and AY326453). Moreover, the mutation does not affect expression of UL24 protein (Fig. 2). Thus, we infer that the inability of TK- mutants to reactivate is largely due to the requirement for TK in reactivation rather than in boosting the efficiency of establishment of latent infection.
Acyclovir-resistant mutants, mostly tk mutants, can arise during drug therapy and can pose a significant problem in immunocompromised patients. In at least one instance, such mutants have been shown to establish and reactivate from latency in a patient (11). How these mutants cause human disease, given their substantial attenuation in animal models as illustrated in this report, is an interesting question that requires further understanding of how TK is required for reactivation and how certain clinical isolates (7, 11) can circumvent this requirement.
This research was supported by grant NSC 91-2320-B-006-093 from the National Science Council in Taiwan (S.-H.C.) and AI-24010 and NS-35138 from the National Institutes of Health in the United States (D.M.C.).
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»