Search

The induction of autoimmunity by viruses has been hypothesized to occur by a number of mechanisms. Coxsackievirus B4 (CB4) induces hyperglycemia in SJL mice resembling diabetes in humans. While virus is effectively cleared within 2 weeks, hyperglycemia does not appear until about 8-12 weeks postinfection at a time when replicative virus is no longer detectable. In SJL mice, reinfection with CB4 enhanced the development of hyperglycemia. As predicted, the immune system responded more rapidly to the second infection and virus was cleared more swiftly. However, while infiltrating T cells were found within the pancreas, depletion of the CD4 T cell population prior to secondary infection or use of CD8 knock-out mice had no effect on the development of virus-mediated hyperglycemia. In conclusion, enhanced hyperglycemia induced by CB4 occurs independent of the T cell response.

Coxsackieviral infections have been linked etiologically to multiple diseases. The serotype CB4 is associated with acute pancreatitis and autoimmune type 1 diabetes. To delineate the mechanisms of host survival after an acute infection with CB4 (strain E2), we have investigated the role of nitric oxide (NO), generated by the inducible form of nitric oxide synthase (NOS2), in viral clearance and pancreatic beta-cell maintenance. Mice deficient in NOS2 (NOS2-/- mice) and their wild-type (wt) counterparts were injected with CB4, after which both groups developed severe pancreatitis, hepatitis, and hypoglycemia within 3 days. Within 4 to 7 days postinfection (p.i.), most of the NOS2-/- mice died and at a strikingly higher mortality rate than wt mice. Histological examination of pancreata from both infected NOS2-/- and infected wt mice revealed early and complete destruction of the pancreatic acinar tissue, but intact, insulin-stained islets. When examined up to 8 weeks p.i., neither surviving NOS2-/-mice nor surviving wt mice developed hyperglycemia. However, the clearance of infectious CB4 was different between the mice. The spleens of NOS2-/- survivors were cleared of infectious virus with kinetics similar to that of wt mice, but the livers, pancreata, kidneys, and hearts of the NOS2-/- groups cleared virus more slowly than those of the wt group. This delayed clearance was particularly prominent in the livers of infected NOS2-/- mice, which also showed prolonged histopathological features of viral hepatitis. Taken together, this outcome suggests that NOS2 (and NO) is not required for the prevention of pancreatic beta-cell depletion after CB4 infection. Instead the critical actions of NOS2 apparently occur early in the host immune response, allowing mice to survive and clear virus. Moreover, the data support the existence of an organ-specific dependency on NO for a rapid clearance of CB4., (Copyright 2001 Academic Press.)

Previous studies have shown that mouse adenovirus type-1 (MAV-1) caused a fatal hemorrhagic encephalitis in certain strains of mice. C57BI/6 mice exhibited 100% mortality when given as little 10(3) plaque-forming units (PFU) of MAV, in contrast to BALB/c mice which were resistant to as many as 10(6) PFU. Susceptible animals died with a flaccid paralysis on the 3rd or 4th day after inoculation. The brains and spinal cords of these animals displayed numerous petechial hemorrhages that were found in virtually all areas of the brain, but were more numerous in white matter. In this paper, immunohistochemistry and electron microscopy were used to identify the viral target of replication within the CNS of susceptible mice. These studies showed that the CNS vascular endothelial cell was the primary site of viral replication within the CNS of mice infected with MAV-1. Characterization of cytokine mRNA levels and disease course in immunodeficient mice revealed that the host immune response played little, if any, role in the pathogenesis of MAV-1 disease in susceptible mice and was not responsible for the resistance of BALB/c mice. These results support the conclusion that disease course and outcome in susceptible and resistant strains of mice were determined primarily by the ability of the virus to replicate within the CNS vascular endothelium.

Adenovirus type 35 (Ad35) is a member of Ad subgroup B, DNA homology cluster B2. The B2 Ads are unique in that they are isolated most frequently from immunosuppressed individuals such as AIDS patients and bone marrow transplant recipients and in that they have a tropism for the urinary tract. One region of the Ad genome which may influence serotype specific pathology is early region 3 (E3). E3 of subgroup C Ad2 and Ad5 has been shown to encode proteins which counteract the immune response to Ad infection. While a great deal is known about gene expression of the subgroup C Ad E3s, little is known about the E3 gene expression from the subgroup B Ads. Although some E3 open reading frames (ORFs) are shared between subgroups B and C, there are additional ORFs that appear in subgroup B. This paper demonstrates the results of an analysis of gene expression from the Ad35 E3 and describes differences in splicing and polyadenylation between the Ad35 and Ad2 E3s. RT-PCR, cDNA sequencing, RNase protection, 3'RACE, and Northern blotting techniques were utilized to identify, quantify, and determine the structure of six Ad35 E3 mRNAs predicted to encode at least seven proteins. A common intron that is removed during splicing of the subgroup C E3 mRNAs is not removed from Ad35 E3 mRNAs, and only one E3 polyadenylation signal is present in the Ad35 E3 while two polyadenylation signals are used in the formation only one E3 polyadenylation signal is present in the Ad35 E3 while two polyadenylation signals are used in the formation of subgroup C E3 mRNAs. The quantity of individual mRNAs encoding homologous proteins for Ad35 and Ad2 also differ substantially, presumably because of the absence in Ad35 of cis-acting signals which have been shown to be important for regulation of Ad2 E3 pre-mRNA processing. Such information should contribute to an understanding of the role the E3 plays in determining subgroup B Ad pathogenesis in general and Ad35 pathogenesis in particular.

The adenovirus type 2 (Ad2) early region 3 (E3) codes for a 19-kDa glycoprotein (gp19) that associates with the class I major histocompatibility (MHC) heavy chain in the endoplasmic reticulum (ER) and prevents the transport of class I MHC protein products to the cell surface. It has been shown previously in tissue culture that this reduction in class I MHC expression allows infected cells to escape detection by class I MHC restricted CD8+ cytotoxic T-lymphocytes (CTL). We now report the results of studies on the effects of Ad2/gp19 expression on virulence in vivo. Since we wanted to isolate the effect of Ad2/gp19 from the effects of other Ad E3 region gene products and human Ads do not replicate in the mouse, we cloned the Ad2/gp19 open reading frame (ORF) into the HindIII C region of WR vaccinia virus (VV). Two VV recombinants were constructed by inserting the Ad2/gp19 ORF in either an expressing (V-e19(+)) or a non-expressing (V-e19(-)) orientation under control of the VV P7.5 promoter. The V-e19(+)recombinant expressed Ad2/gp19 in infected tissue and could be co-precipitated with an antibody to the class I MHC antigen Kd. However, intracerebral or intranasal infections of BALB/c (H-2d), BALB.G (H-2g), or C3H (H-2k) mice showed that Ad2/gp19 expression by V-e19(+) had no significant effect either on viral lethality or on its ability to replicate in vivo when compared to V-e19(-). Furthermore, the nature of the CD8+ CTL response to a V-e19(+)-induced pneumonia in (H-2d) mice was unchanged by Ad2/gp19 expression. Modulating the CD8+ CTL response, by interfering with infected target presentation, may not be important in the control of VV replication or virulence in vivo when other aspects of the immune response to viral infection are not altered. However, the two VV recombinants V-e19(+) and V-e19(-) were both equally attenuated (10-fold) when compared to wild-type VV. This attenuation, which has been reported previously for an intracerebral infection, is believed to be caused by the disruption of a 37-kDa ORF in the VV HindIII C region. Interestingly, our studies showed that the attenuation is not accompanied by a reduction in viral titers in infected tissue.

Mutational analysis within the amino-terminal (N-t) domain of the adenovirus DNA-binding protein (DBP) defined a region (aa 2-38) important for DBP function. Several viruses carrying lesions in this region of DBP showed reduced accumulation of viral DNA and infectious virions. Characterization of one of these mutants, H5in800, indicated that the N-t domain affects viral DNA synthesis in vivo. The reduction in DNA synthesis was not due to a change in the amount or nuclear location of the H5in800 DBP. Expression of other early genes in H5in800-infected cells was similar to that seen in wild-type Ad5-infected cells, suggesting that the depression of DNA synthesis was not due to disruption of DBP's role in early gene expression. The H5in800 and wild-type DBP also had comparable affinities for single-stranded DNA and functioned with similar efficiencies in two DNA elongation assays. Prior studies have shown that the carboxyl-terminal (C-t) domain of DBP was responsible for these two activities. Together these results suggest that DBP has at least two separable functions in viral DNA replication in vivo and that both domains of the protein are necessary for full activity. The intragenic complementation between the N-t mutant H5in800 and the C-t mutant H5in804 supports this model.

The functional importance of the conserved region I (YGDTDSLF) found in several prokaryotic, eukaryotic, and viral DNA polymerases has been probed by site-directed mutagenesis of the adenovirus DNA polymerase (Ad Pol). Three different adenovirus-specific assays have been used to measure the in vitro activity of region I mutants of Ad Pol expressed in transiently transfected CMT-4 cells. In general, both conservative and nonconservative changes generally showed a greater than 5- to 10-fold reduction in activity in three different assays for activity. However, several replacements at the glycine residue showed activities closer to wild-type levels. For example, replacements of this glycine with cysteine (found in bacteriophage phi 29, another protein primed replication system), with serine, or with methionine had little effect on the activity observed in adenovirus-specific assays, such as initiation and elongation. These studies confirm the importance of this region of Ad Pol in specific initiation and elongation reactions on Ad DNA templates.

A mutant of human adenovirus type 34 (Ad34) has been isolated which contains DNA molecules with tandem reiterations of from two to eight copies of a 131-bp sequence within the right-sided inverted terminal repetition. Terminal heterogeneity was not eliminated by repeated plaque purifications indicating that the population of DNA molecules with various numbers of reiterations could rapidly evolve from the DNA of a single virus particle. These enlarged DNA molecules were capable of replication both in vivo and in vitro. The nucleotide sequence of the mutant Ad34 inverted terminal repetitions contained most of the essential features of the Ad origin of DNA replication. These features include the ATAATATACC sequence which is present between the highly conserved bases 9-18 in all human adenoviruses, as well as the consensus sequences for the binding of nuclear factor I and nuclear factor III. However, the reiterated sequences lacked a dG appropriately placed on the template strand to serve as a potential site for internal initiation. It appears that the rapid amplification of two to eight copies of the reiterated terminal sequences does not arise from internal initiation during replication but probably from homologous recombination.

Genetic and biochemical studies of adenovirus (Ad) DNA synthesis in vitro demonstrate that the Ad DNA binding protein (Ad DBP) is not necessary for the initiation of Ad DNA synthesis but is required for chain elongation. The DBP, which enhances early elongation to the 26th deoxynucleotide by approximately two- to fourfold, is absolutely required as chain elongation proceeds further. Ad DNA synthesis was assayed in a system requiring Ad DNA covalently linked at each 5' terminus to a protein (Ad DNA-pro), various fractions of Ad-infected cytoplasm, and an extract of uninfected Hela nuclei. Initiation of Ad DNA replication was measured by the formation of a covalent complex between the 80,000 dalton preterminal protein (pTP) and 5' dCMP. DNA binding proteins from two ts mutants, H5ts125 and H5ts107, have been purified and shown to be functional at 30 degrees but inactive at 38 degrees in an in vitro elongation system dependent on purified proteins. Chymotryptic cleavage of the 72K wild-type Ad2 DBP produces a 34K carboxyl terminal fragment which retains full activity in the in vitro elongation of Ad DNA.

Replication of human adenovirus (Ad) DNA requires three virus-encoded proteins that are coordinately transcribed from a single promoter at early times after infection. The mRNAs for two of these proteins, the preterminal protein (pTP) and the Ad DNA polymerase (Ad Pol), share several exons, including one encoded near Ad genome coordinate 39. Plasmids containing the putative exons that encode Ad Pol mRNA were constructed to determine if enzymatically active Ad Pol protein could be synthesized. An Ad Pol of 140 kDa was detected by immunoprecipitation with specific antibody and its enzymatic activity was confirmed by complementation of Ad DNA replication in vitro. In addition to an Ad2 DNA fragment from 24.7 to 9.2 map units which contains an open reading frame for a protein of 120 kDa, the HindIII-J fragment that encodes the exon at genome coordinate 39 can be shown to be essential for production of full-length (140 kDa), enzymatically active Ad Pol.

An open reading frame which encodes at least 90% of the adenovirus type 2 DNA polymerase gene was cloned behind the SP6 promoter and transcribed in vitro using the SP6 RNA polymerase. The resultant RNA was translated in a rabbit reticulocyte cell free system. In addition to the translation of a 120-kDa protein corresponding to the size of the complete open reading frame, the synthesis of a 62-kDa polypeptide was demonstrated. Data is presented to show that the synthesis of the 62-kDa polypeptide resulted from internal initiation of translation in frame in the middle of the message at the 11th or 12th AUG. Capping of the mRNA resulted in an increase in synthesis of the 120-kDa protein and a concordant decrease of the internally initiated polypeptide. We propose that there may be competition between the binding of the translational preinitiation complex at or near the 5' end of the mRNA and at the internal initiation site. Because of inhibition of synthesis of the 120-kDa but not the 62-kDa polypeptide by hybrid arrested translation using DNA complementary to approximately one third of the 5' Ad Pol mRNA sequences, scanning of the ribosome from the 5' end of the mRNA to the internal initiation site seemed unlikely. The sequence proximal to the 12th AUG is ACCCACCCCAUG which is similar to a noncontinuous sequence 5'AUCCACC(X)nAUG complementary to the 3' end of the 18 S rRNA. This sequence is a favored ribosome binding site based on the observation that it is the most commonly observed one at or near the 5' end of 162 mRNA's analyzed (D. R. Sargan, S. P. Gregory, and P. H. W. Butterworth, 1982, FEBS Lett. 147, 133-136).

The 34-kDa fragment of the carboxyl end of the adenovirus (Ad) DNA binding protein (DBP) binds to single-stranded (ss) DNA and is able to replace the intact 72-kDa DBP needed for Ad DNA replication in vitro. A similar fragment prepared from the temperature-sensitive (ts) mutant, H5ts107, which has a single amino acid change in the carboxyl end of the DBP, is temperature sensitive for DNA replication and defective in binding to ssDNA. However, in 20 mM NaCl which is the salt concentration during Ad DNA replication in vitro, the intact 72-kDa H5ts107 DBP is defective only in replication but not binding to DNA at nonpermissive temperatures. These observations indicate that the amino domain of the H5ts107 DBP can stabilize the binding of its carboxyl end to DNA.

The effects of the adenovirus type 2 (Ad2) structural proteins on Ad DNA synthesis in vitro have been examined. Both of the viral core proteins, polypeptides V and VII were shown to inhibit Ad2 DNA synthesis in vitro; however, only the major core protein, polypeptide VII, inhibited DNA synthesis at a ratio of protein to DNA proportional to the number of polypeptide VII molecules associated with the Ad2 DNA in the mature virion. In addition, fractions containing the precursor to polypeptide VII, pVII, were capable of inhibiting Ad2 DNA replication in vitro to the same extent as polypeptide VII. Purified polypeptide VII bound to double-stranded DNA with no apparent sequence specificity. In addition, polypeptide VII protected Ad2 DNA from digestion with micrococcal nuclease. The binding of polypeptide VII was probably responsible for the inhibition of Ad2 DNA synthesis in vitro by virtue of rendering the DNA inaccessible to viral replication proteins. These results suggest that the core proteins must be removed from the Ad2 genome before the template can function in genome replication and that assembly of pVII on Ad2 DNA can terminate the replication process.