Previous Article | Next Article ![]()
Journal of Virology, June 2003, p. 6660-6665, Vol. 77, No. 12
0022-538X/03/$08.00+0 DOI: 10.1128/JVI.77.12.6660-6665.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Departments of Neurology,1 Microbiology, University of Colorado Health Sciences Center, Denver, Colorado 80262,2 Department of Ophthalmology, University of Pittsburgh, Pittsburgh, Pennsylvania 15213,3 Department of Neurology, University of Glasgow, Glasgow G51 4TF, Scotland, United Kingdom4
Received 19 December 2002/ Accepted 24 March 2003
|
|
|---|
70 predicted VZV genes, only five (genes 4, 21, 29, 62, and 63) have been shown by multiple techniques to be transcribed during latency. IE62, the protein product of VZV gene 62, is the major immediate-early (IE) virus-encoded transactivator of viral gene transcription and plays a pivotal role in transactivating viral genes during lytic infection. The protein kinase (66-pk) encoded by VZV gene 66 phosphorylates IE62, resulting in cytoplasmic accumulation of IE62 that mitigates nuclear IE62-induced gene activation. Analysis of latently infected human trigeminal ganglia for 66-pk expression by reverse transcriptase-dependent nested PCR, including DNA sequence analysis, in situ hybridization, and immunohistochemistry, revealed VZV open reading frame 66 to be a previously unrecognized latently expressed virus gene and suggests that prevention of IE62 import to the nucleus by VZV 66-pk phosphorylation is one possible mechanism by which VZV latency is maintained. |
|
|---|
IE62, the essential immediate-early protein product of VZV gene 62, stimulates transcription from all VZV promoters tested (2, 14). IE62 increases the infectivity of transfected VZV DNA (23). IE62 localizes to the nucleus early in virus infection and later accumulates in the cytoplasm (10, 16). Interestingly, this protein is localized predominantly in the cytoplasm during latency (20). Phosphorylation of IE62 by the protein kinase encoded by VZV ORF 66 (66-pk) results in the cytoplasmic accumulation of IE62 (19). While cytoplasmic IE62 is incorporated into enveloped virions (17, 18), the restriction of IE62 from the nucleus also regulates the function of this potent activator of gene transcription (18). Thus, phosphorylation of IE62 by 66-pk appears to be one possible mechanism by which IE62 remains present within a cell without stimulating gene transcription.
We report here the detection of 66-pk expression in latently infected human TG by reverse transcriptase (RT)-PCR, by ISH, and by immunohistochemistry, raising the possibility that a block in nuclear import of IE62 by 66-pk is a crucial determinant of VZV latency.
|
|
|---|
Nucleic acid extraction, cDNA synthesis, and PCR analysis. Nucleic acids were extracted as described previously, using DNase treatment of all RNA preparations (3). Reverse transcription of 1 µg of RNA primed with either d(T) or anchor-d(T) (Table 1) was carried out as described elsewhere (6). Oligonucleotide primers for cDNA synthesis and PCR amplification (Table 1) were commercially prepared (IDT, Coralville, Iowa). PCR cycle times and temperatures have been described previously (6). All PCRs include a no-template (water) control, which was uniformly negative. When necessary, PCR products were resolved on 2% agarose gels, transferred to membranes (Zeta probe; Bio-Rad Laboratories, Hercules, Calif.), and probed with 32P-end-labeled DNA oligonucleotides (5). The DNA sequences of PCR products were determined (Macrogene, Seoul, Korea) after ultrafiltration (Micron PCR; Millipore Corp., Bedford, Mass.).
|
View this table: [in a new window] |
TABLE 1. Oligonucleotide primers for cDNA synthesis and PCR amplification
|
Antibody production. The VZV ORF 66 derived from the plasmid PKCMVHA66 (19) was cloned into the pMalCR1 vector (New England Biolabs, Beverly, Mass.), and the maltose binding protein-ORF 66 gene fusion was expressed and purified using amylose affinity chromatography. The purified protein was used to raise antibody to 66-pk in rabbits as described previously (15). The antibody reacts with a 55-kDa protein in Western immunoblotting and stains both the nucleus and cytoplasm of VZV-infected cells (P. R. Kinchington, unpublished data).
|
|
|---|
Verification of latent VZV ORF 66 transcription by DNA sequence analysis. To identify the 3'-terminal sequences of VZV gene 66 transcripts in latently infected human TG, cDNA was synthesized from RNA extracted from the pooled left and right TG of a single subject and analyzed by PCR. Priming cDNA synthesis with the anchor-d(T) primer resulted in modified cDNA, allowing subsequent PCR amplification across the 3' terminus, including the poly(A) tail. In preliminary experiments, PCR amplification of cDNA synthesized from VZV-infected MeWo cells produced DNA fragments migrating in ethidium bromide-stained agarose gels to the approximate size position expected with primers actin-F1 and anchor (386 bp), ORF 40-P2 and anchor (447 bp), and ORF 66-P2 and anchor (336 bp) (Fig. 1A, left panels, lane 3) and was verified by hybridization with the internally located 32P-end-labeled oligonucleotides (Fig. 1A, right panels, lane 3). In separate experiments, actin and VZV gene 66 transcripts were amplified from latently infected human TG cDNA (Fig. 1A, lane 1). In both instances, the amplification products depended on reverse transcription (Fig. 1A, lane 2). No amplification product was detected when VZV gene 40-specific primers were used (Fig. 1A, lane 1). VZV gene 40 encodes the major capsid protein of the virus, and the lack of gene 40 transcription suggests that VZV had not reactivated in the human ganglia removed at autopsy (5).
![]() View larger version (59K): [in a new window] |
FIG. 1. PCR amplification of VZV cDNA. (A) Total TG RNA extracted was reverse transcribed with (lane 1) or without (lane 2) RT. In a second reaction, VZV-infected MeWo cell RNA was reverse transcribed with (lane 3) or without (lane 4) RT. Each cDNA product was amplified with primers designed to yield products spanning the poly(A) tract. Amplified products were resolved by agarose gel electrophoresis and stained with ethidium bromide (left). VZV-specific amplification products were transferred to nylon membranes and probed with 32P-end-labeled, internally located oligonucleotides (right). Both TG and VZV-infected MeWo cell cDNA contained actin transcripts (top gels); VZV gene 40 transcripts were detected only in virus-infected MeWo cell cDNA (middle gels); and VZV gene 66 transcripts were detected in TG and VZV-infected MeWo cell cDNA (bottom gels). In all instances, amplification was RT dependent (lanes 1 and 3). No product was seen when RT (lanes 2 and 4) or template DNA (lane 5) was omitted from the PCR. M, double-stranded DNA size markers. (B) Primer efficiencies were determined by PCR amplification of serial log10 dilutions of VZV ORF 40 and ORF 66 artificial transcripts (AT). In the primary PCR, both ORF primer sets gave strong signals with 108 copies of AT. In nested PCR, the ORF 40 primer set detected 104 to 105 copies of AT, while ORF 66 primers detected 106 to 107 of added AT. ND, no AT template DNA added to the PCR.
|
85 nucleotides (nt) upstream from the location of the primers used to detect the transcripts (ORF 40- and 66-P1) and extended to the respective ORF termination codon. The artificial transcripts were also modified by incorporating the anchor primer site at the 3' end of the DNA fragments. Following PCR amplification, the artificial transcripts were cleaned and quantitated, and serial log10 dilutions were PCR amplified using the same primer sets as were employed to generate those shown in Fig. 1A. The amplification products were transferred to nylon-based membranes and probed with 32P-end-labeled ORF 40- or 66-P3. The results demonstrate that ORF 40 primer sets were
1 order of magnitude more efficient than the ORF 66 primer sets. In separate experiments, the 3'-terminal cDNA sequence was determined for gene 66 transcripts from VZV-infected MeWo cells and latently infected human TG (Table 2). Both lytic and latent gene 66 transcripts terminate at nearly identical sites: 114,238 nt (lytic) and 114,236 nt (latent). This very slight difference in poly(A) addition sites has been seen in latent gene 21 and 63 transcription (5) and most likely reflects inherent inaccuracies in poly(A) addition or slight virus strain differences, as opposed to a mechanism of latent gene transcriptional control. However, both lytic and latent gene 66 termination sites are located within the poly(A) stretch, which suggests that alternative signal sites to those of the proposed polyadenylation signal (8) are employed. Two such potential alternative poly(A) addition signaling sites are located within 30 bp of the 3' end of the transcript (Table 2).
|
View this table: [in a new window] |
TABLE 2. 3' cDNA sequence of ORF 66 transcripts
|
![]() View larger version (46K): [in a new window] |
FIG. 2. Detection of VZV transcripts by ISH. Adjacent sections of a latently infected human TG were hybridized to digoxigenin-labeled VZV-specific probes (66F1, 66R1, and 62-AS; Table 1). (A and B) No signal was detected after hybridization with a sense VZV gene 66 probe (A), whereas a strong signal was seen in the nuclei of two neurons (open arrows) and a weaker signal was seen in two neurons (dark arrows) after hybridization with an antisense VZV gene 66 probe (B). (C) A strong signal in three neurons (open arrows) was also detected after hybridization with an antisense probe to VZV gene 62. Magnification, x450.
|
![]() View larger version (108K): [in a new window] |
FIG. 3. Immunohistochemical detection of VZV ORF 66 protein. TG sections from two subjects were incubated with antibody to VZV gene 66 protein as described in Materials and Methods. (A and B) Strong labeling in the cytoplasm of one neuron (open arrow) and weaker labeling in another neuron (closed arrow) from the first subject was detected (A), whereas no signal was detected in the second subject (B). TG sections from the first subject but not the second subject were positive for VZV gene 66 transcripts by ISH (data not shown). (C and D) VZV gene 66 protein was present in VZV-infected MeWo cells (C), but not in uninfected MeWo cells (D). Images were captured at a magnification of x427.5; panel B was printed at a magnification of x541.5.
|
|
|
|---|
20 nt distal to the termination codon; ISH identified 66-pk transcripts in the nuclei of latently infected neurons; and immunohistochemistry detected 66-pk protein exclusively in the cytoplasm of latently infected neurons. The sequence analysis and cellular localization of 66-pk within human TG parallels previous analyses of VZV transcripts in latently infected human ganglia (5, 11, 12). Together, the results indicate that VZV gene 66 is both transcribed and translated in latently infected human ganglia. Given the presence of at least 70 VZV ORFs, the extent of virus transcription in latently infected human ganglia is obviously restricted. The presence of gene 62 transcripts (5, 22) and, by one account, VZV gene 62-encoded protein (20) during latency coupled with the function of IE62 in transcriptional activation (2, 4, 23) suggest that the functional activity of IE62 during latency is modulated. We recently showed that 21p and 29p, proteins encoded by VZV genes 21 and 29, do not affect IE62-induced transactivation of VZV gene 20, 21, 28, or 29 promoters in transfection studies (4). Since activation of VZV gene transcription by IE62 most likely occurs in the nucleus, nuclear import of IE62 may be a critical event in gene transactivation, and this can clearly be regulated. Kinchington et al. (16, 18, 19) showed that phosphorylation of IE62 by 66-pk leads to cytoplasmic accumulation of IE62. While the phosphorylation of IE62 late in VZV lytic infection appears to be a mechanism by which IE62 is incorporated into progeny virions (18), the cytoplasmic location of IE62 during latency that results from 66-pk phosphorylation might also modulate IE62-dependent gene transactivation.
IE62 is an essential, immediate-early protein pivotal to the production of infectious virus (10, 25). Thus, it is surprising that IE62 transcripts are detected during latency. In this report, VZV gene 66 transcripts were detected by independent means in two separate laboratories using autopsy samples collected from separate population pools. The known effect of 66-pk on the cellular location of IE62 and the detection of VZV gene 66 transcription during latent infection suggest that latent IE62 is functionally restricted through 66-pk phosphorylation.
Although our experiments were not designed to address the prevalence of ORF 66 transcription in the human population or the abundance of ORF 66 transcripts within individual ganglia, we did find that 2 of 6 (33%) individual TG were positive for VZV gene 66 transcripts by RT-PCR and that 5 of 27 (18%) individual ganglia were positive for VZV gene 66 transcription by ISH. The apparent twofold increase in the frequency of gene 66 detection by RT-PCR compared to ISH most likely reflects the different sensitivities of each technique. For example, RT-PCR uses RNA extracted from the entire powdered TG and samples from >3 x 1010 mRNA molecules per reaction mixture, while ISH samples individual sections of each TG; thus, only a portion of the entire TG is sampled in each ISH assay. Further, the focal and patchy distribution of ISH signal in TG sections makes it difficult to give a precise number for the percentage of neurons which are 66 and/or 62 positive. However, we estimate
2 to 5% of neurons are positive for gene 66 or gene 62 during latent infection. No obvious difference in the positivity rate between genes 66 and 62 was discerned. It was also difficult to determine if the same neuronal cell is positive for both gene 66 and 62 transcripts; however, our analysis of multiple serial sections within the same ganglion has indicated that both gene transcripts are present in the same ganglion. Also, we could not detect TG neurons in which VZV gene 66 protein was expressed in both the nucleus and the cytoplasm.
Finally, the identification of VZV gene 66 RNA and protein in latently infected human ganglia suggests a possible mechanism by which the gene-activating function of latently expressed IE62 is regulated. An interesting alternative interpretation is that 66-pk expression in neuronal TG is a means of damping reactivation of latent VZV. Feldman et al. (9) have documented rare occurrences of spontaneous subclinical herpes simplex virus type 1 reactivation in a small subset of latently infected mouse TG neuronal cells. If a population of VZV-latently infected neuronal cells undergoes reactivation, one would predict IE62 to be among the first virus genes expressed. Thus, induction of 66-pk transcription would limit virus reactivation. In this case, gene 66 transcription could be seen as a "smoking gun," indicating those neuronal cells which have undergone recent (and now abortive) reactivation. Although this hypothesis is difficult to test in humans, it is testable in developing animal models of VZV latency and reactivation (27).
Expert technical assistance was provided by John Smith, Carrie Essman, Jeannie Welsher, Stephanie E. Turse, and Michael Hurley. We also thank Marina Hoffman for editorial help and Cathy Allen for manuscript preparation.
|
|
|---|
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»