Effect of Time Delay after Necropsy on Analysis of Simian Varicella-Zoster Virus Expression in Latently Infected Ganglia of Rhesus Macaques

Primary infection with varicella-zoster virus (VZV) causes chickenpox, after which virus becomes latent in ganglia along the entire neuraxis. Reactivation decades later produces zoster and other serious neurological complications (2). Like VZV, simian varicella-zoster virus (SVV) becomes latent in ganglionic neurons of primates (4) and reactivates to produce zoster (7). Studies of viral gene expression during latency rely on fresh human tissue obtained at autopsy, but concerns have been raised regarding possible virus reactivation immediately after death that might confound analyses. Here, we examined the effect of time delay between euthanasia and necropsy on the extent of virus gene transcription and translation in a simian varicella-zoster virus model.

Two SVV-seronegative rhesus macaques (GL82 and GI39) were inoculated intrabronchially with 10³ to 10⁵ PFU of SVV propagated in rhesus fibroblasts. All procedures were performed following guidelines and protocols approved by the Institutional Animal Care and Use Committee of the Tulane National Primate Research Center. Both monkeys developed a vesicular rash 7 to 10 days postinfection (d.p.i.) that cleared after 3 weeks. In both monkeys, SVV open reading frame (ORF) 61-specific DNA was detected by real-time PCR in blood mononuclear cells (MNCs) beginning at 4 d.p.i., peaking at 7 d.p.i., and disappearing by 21 d.p.i. (Fig. 1), confirming earlier findings (8). Although viremia peaked at the same time in both monkeys, the levels of SVV DNA were 3-fold higher in monkey GL82 than in monkey GI39. In both monkeys, SVV-neutralizing antibodies appeared at 9 d.p.i. (1:20). At 17 d.p.i., the levels of SVV-specific antibody were higher in monkey GL82 (1:640) than in monkey GI39 (1:160). The antibody levels remained high (≥1:640) in monkey GL82 for the rest of its life. In monkey GI39, the antibody titer was 1:80 at 70 d.p.i., perhaps reflecting the reduced viremia seen during primary infection (Fig. 1).

Both monkeys were euthanized at 117 d.p.i. Multiple ganglia and the lung and liver were harvested immediately from monkey GL82, whereas monkey GI39 was kept at 4°C for 30 h before the same tissues were removed. In both monkeys, pooled ganglia from each segment of the neuraxis (trigeminal, cervical, thoracic, lumbar, and sacral) were obtained. Ganglia from one side of the neuraxis were snap-frozen, and ganglia from the other side were fixed in 4% paraformaldehyde and embedded in paraffin. A small portion of frozen tissues was used for DNA extraction as described previously (8). In both monkeys, quantitative real-time PCR detected glyceraldehyde-3-phosphate dehydrogenase (GAPDH) DNA in all tissues (data not shown); SVV DNA was found in ganglia but not in lung or liver (Table 1), indicating latent infection. In monkey GL82, SVV DNA (45 to 70 copies) was detected in thoracic, lumbar, and sacral ganglia but not in trigeminal or cervical ganglia. In monkey GI39, SVV DNA was found only in trigeminal ganglia (3 copies), consistent with reduced viremia (Fig. 1), although degradation of virus DNA during the 30-h interval from euthanasia to removal of ganglia from monkey GI39 cannot be ruled out.

RNA was extracted from snap-frozen ganglia from both monkeys, treated with DNase to remove contaminating virus DNA, and analyzed using quantitative real-time reverse transcription-PCR with primers and probes corresponding to SVV ORFs 21, 29, 40, 61, 62, 63, and 66 as described previously (8) (Table 1). The copy numbers of SVV ORF-specific transcripts per 500 ng of cDNA were determined. In monkey GL82, SVV ORF 61-specific transcripts were detected (361 to 4,000 copies). In cervical ganglia, SVV ORF 61 RNA was found in the absence of detectable SVV DNA, probably a reflection of robust transcription from SVV ORF 61 RNA in the context of low levels of virus DNA. The second-most-prevalent transcript corresponded to SVV ORF 63, of which 3 to 13 copies were detected in 3 of 4 ganglia. Low levels of transcripts corresponding to SVV ORFs 29 and 66 (3 and 6 copies, respectively) were found in lumbar ganglia. SVV ORFs 21 and 40 were not detected in any ganglia. No SVV transcripts were detected when reverse transcriptase was omitted during preparation of cDNA, confirming the removal of all SVV DNA.

In monkey GI39, SVV ORF 61 transcripts (4 to 3,600 copies) were detected in 4 of 5 dorsal root ganglia. As found in the cervical ganglia of monkey GL82, the thoracic, lumbar, and sacral ganglia of monkey GI39 contained significant quantities of SVV ORF 61-specific RNA in the absence of detectable virus DNA. No significant difference in the abundance of SVV ORF 61-specific transcripts was found in the two monkeys. The second-most-prevalent transcript corresponded to SVV ORF 62, which was detected in both thoracic and lumbar ganglia. In pooled thoracic ganglia, SVV ORF 62 (6 copies) was detected in the absence of reverse transcriptase during cDNA preparation, indicating incomplete digestion of virus DNA. SVV ORF 63 was detected in trigeminal, thoracic, and lumbar ganglia (6, 5, and 5 copies, respectively). SVV ORFs 21, 29, and 40 were not detected in any ganglia. Cervical ganglia were devoid of SVV DNA and RNA. The sensitivity of PCR with all primer sets was 1 copy per 500 ng of DNA or cDNA.

ORF 63 protein of both VZV and SVV localizes in the cytoplasm of neurons in latently infected human and monkey ganglia, respectively (4-8). To determine whether time delay after euthanasia alters the prevalence or localization of SVV ORF 63 protein in ganglionic neurons, we analyzed multiple ganglia from monkeys GL82 and GI39 from the side of the neuraxis opposite to that used for DNA and RNA analysis by immunohistochemistry with antibodies directed against SVV ORF 63 and SVV glycoproteins gL and gH (1) as described previously (6). Rabbit polyclonal antibody raised against VZV ORF 63 protein detected SVV ORF 63 protein in the cytoplasm of neurons in lumbar ganglia of both monkey GL82 (Fig. 2A) and GI39 (Fig. 2E). No staining was seen in adjacent sections of lumbar ganglia from monkey GL82 (Fig. 2B) or GI39 (Fig. 2F) stained with normal rabbit serum or in an adjacent section of lumbar ganglia from GI39 stained with anti-SVV gH and gL antibodies (Fig. 2G). Microscopic examination of over 4,000 neurons from monkey GL82 and 2,900 neurons from monkey GI39 in sections of multiple latently infected ganglia revealed that 6.6% ± 2.1% (mean ± standard deviation) and 8.6% ± 1.8% of neurons, respectively, were positive for SVV ORF 63 protein (Table 2). At a 95% confidence interval, we found no significant difference between the two monkeys in the number of latently infected neurons.

To test the possibility that our detection of SVV ORF 63 in latently infected monkey ganglia was due to neuromelanin, which can confound immunohistochemical detection of VZV proteins in human ganglia (9), we stained sections of lumbar ganglia from monkey GI39 (which contained SVV ORF 63 protein) with Nile blue solution as described previously (9). Nile blue is known to bind to neuromelanin in tissue sections, resulting in a dark green stain that is resistant to acetone extraction, whereas binding to cell nuclei and lipofuscin produces blue staining. Indeed, dark blue staining of neuronal nuclei and occasional lipofuscin (arrow in Fig. 2C) were seen, and acetone extraction resulted in only rare faint green neuronal staining due to neuromelanin (arrowhead in Fig. 2D). Thus, the detection of SVV ORF 63 was not due to neuromelanin in latently infected monkey ganglia.

Despite the limited sample size, our analysis of latently infected monkey ganglia immediately after death and 30 h later revealed no substantial difference in virus transcription or translation. After the death of humans, cadavers are placed in cold chambers at 4°C, which is likely to help maintain the integrity of nucleic acids and proteins; of course, changes in the patterns of transcription or translation are likely to occur 48 h or more after death. In one other study addressing the effect of time delay on varicella-zoster virus transcription in VZV-infected rats, the number of ganglia positive for VZV ORF 63 RNA was the same at the time of death and 48 h later; furthermore, VZV ORF 40 transcripts were detected rarely immediately after death but never 48 h later, indicating lack of virus reactivation after death (3). Overall, the findings indicate that human ganglia obtained at autopsy 30 h postmortem can be used for analysis of the expression of latent VZV RNA and protein.

• Open in new tab
• Download powerpoint

FIG. 1.

Detection of viremia in rhesus macaques inoculated with SVV. DNA extracted from MNCs isolated from blood samples obtained at multiple time points after inoculation was used for analysis by quantitative real-time PCR using primers specific for SVV ORF 61. Viremia peaked at 7 d.p.i. and reached background levels around 21 d.p.i.

• Open in new tab
• Download powerpoint

FIG. 2.

Detection of SVV ORF 63 protein in monkey ganglia latently infected with SVV. Paraformaldehyde-fixed, paraffin-embedded sections (5 μm) of lumbar ganglia from monkeys GL82 (A to D) and GI39 (E to G) were analyzed by immunohistochemistry. Note the detection of SVV ORF 63 protein in the cytoplasm of neurons using a 1:800 dilution of rabbit anti-VZV ORF 63 (A and E) but not with a 1:800 dilution of normal rabbit serum (B and F). Adjacent sections of lumbar ganglia from monkey GL82 were stained with Nile blue solution (C) and extracted with acetone (D); Nile blue distinguishes lipofuscin (dark blue staining indicated by arrow in C) from neuromelanin (faint green staining indicated by arrowhead in D), which is resistant to acetone extraction. Ganglia from monkey GL82 latently infected with SVV do not contain significant amounts of neuromelanin to interfere with detection of SVV ORF 63. No staining was seen in adjacent sections of lumbar ganglia from monkey GI39 after incubation with a 1:2,000 dilution of rabbit anti-SVV glycoprotein gH or gL polyclonal antiserum (G).

• Copyright © 2010 American Society for Microbiology