Transgenic expression of Theiler's murine encephalomyelitis virus genes in H-2(b) mice inhibits resistance to virus-induced demyelination

Abstract

We investigated the role of the immune system in protecting against virus-induced demyelination by generating lines of transgenic B10 (H-2(b)) congenic mice expressing three independent contiguous coding regions of the Theiler's murine encephalomyelitis virus (TMEV) under the control of a class I major histocompatibility complex (MHC) promoter. TMEV infection of normally resistant B10 mice results in virus clearance and development of inflammatory demyelination in the spinal cord. Transgenic expression of the viral capsid genes resulted in inactivation of virus-specific CD8(+) T lymphocytes (class I MHC immune function) directed against the relevant peptides, but it did not affect production of virus capsid-specific antibodies or lymphocyte proliferation to the virus antigen (class II MHC immune functions). Following intracerebral infection with TMEV, all three lines of mice survived the acute encephalitis but transgenic mice expressing VP1 (or the cluster of virus capsid proteins [VP4, VP2, and VP3] mapping to the left of VP1 in the TMEV genome) developed virus persistence and subsequent demyelination in spinal cord white matter. Transgenic mice expressing noncapsid proteins mapping to the right of VP1 (2A, 2B, 2C, 3A, 3B, 3C, and 3D) cleared the virus and did not develop demyelination. These results are consistent with the hypothesis that virus capsid gene products of TMEV stimulate class I-restricted CD8(+) T-cell immune responses, which are important for virus clearance and for protection against myelin destruction. Presented within the context of self-antigens, inactivation of these cells by ubiquitous expression of relevant virus capsid peptides partially inhibited resistance to virus-induced demyelination.

FIG. 1.

Construction of TMEV transgenes and detection of mice containing transgenes. (A) The TMEV genome was divided into three regions. Region I (LP) is the coding sequence mapping 5′ of VP1 (L, VP4, VP2, and VP3). Region II (VP1) is the entire VP1 coding block. Region III (RP) is the coding sequence mapping 3′ of VP1 (2A, 2B, 2C, 3A, 3C, and 3D). Region I (B), region II (C), and region III (D) nucleotide sequences were isolated by PCR from full-length cDNAs of the TMEV genome and cloned into vector pTKgPtF1s, adding a promoter and enhancer of H-2K^b and a 3′ untranslated fragment of H-2L^d, which came from plasmid 5A7. The promoter region, signal peptide, and first intron of class I MHC were retained at the 5′ end of the construct. Each transgene was added as a single exon. The termination codon, 3′ noncoding region, and poly(A) signal of the class I gene were fused to the end that meets the TMEV coding region. The recombinant constructs were amplified in E. coli and sequenced, and their ability to be transcribed was confirmed by the transfection of C57SV cells. The transgenes were injected into CBA × B10.M mouse embryos, and transgene-positive founders were backcrossed to achieve a resistant B10 genotype. Southern blots from two mice containing VP1 (E), three mice containing LP (F), and two mice containing RP (G) transgenes are shown.

FIG. 2.

TMEV coding region mRNA expression in transgenic mice. (A) Expression of RP transgene after stimulation with gamma interferon, as detected by RT-PCR followed by Southern blotting, was low but detectable. Examples of expression for the RP transgene are shown for both a transgene-positive B10 mouse (second lane from left) and transfected C57SV cells (fourth lane from left). (B and C) In situ hybridization to detect K^b-LP mRNA was performed on spinal cords of LP⁺ transgenic mice (B) and spinal cords of LP⁻ littermate control mice (C). (D and E) In situ hybridization to detect K^b-VP1 mRNA was performed on spinal cords of VP1⁺ transgenic mice (D) and spinal cords of VP1⁻ littermate control mice (E). To demonstrate the stringency by which the K^b-VP1 probe hybridizes to mRNA expressed by the K^b-VP1 transgene alone, the in situ experiment was repeated on the brains of 7-day-infected nontransgenic C57BL/10 mice. (E and F) Using anti-TMEV immunostaining, the hippocampus of 7-day-infected C57BL/10 mice stained positive for detection of TMEV infection (F) but negative for in situ detection of K^b-VP1 mRNA (E). Note that, as shown in panel E, the K^b-VP1 probe used to detect K^b-VP1 mRNA did not hybridize with K^b or TMEV mRNA.

FIG. 2.

TMEV coding region mRNA expression in transgenic mice. (A) Expression of RP transgene after stimulation with gamma interferon, as detected by RT-PCR followed by Southern blotting, was low but detectable. Examples of expression for the RP transgene are shown for both a transgene-positive B10 mouse (second lane from left) and transfected C57SV cells (fourth lane from left). (B and C) In situ hybridization to detect K^b-LP mRNA was performed on spinal cords of LP⁺ transgenic mice (B) and spinal cords of LP⁻ littermate control mice (C). (D and E) In situ hybridization to detect K^b-VP1 mRNA was performed on spinal cords of VP1⁺ transgenic mice (D) and spinal cords of VP1⁻ littermate control mice (E). To demonstrate the stringency by which the K^b-VP1 probe hybridizes to mRNA expressed by the K^b-VP1 transgene alone, the in situ experiment was repeated on the brains of 7-day-infected nontransgenic C57BL/10 mice. (E and F) Using anti-TMEV immunostaining, the hippocampus of 7-day-infected C57BL/10 mice stained positive for detection of TMEV infection (F) but negative for in situ detection of K^b-VP1 mRNA (E). Note that, as shown in panel E, the K^b-VP1 probe used to detect K^b-VP1 mRNA did not hybridize with K^b or TMEV mRNA.

FIG. 3.

VP1- and VP2-specific CTL assay. VP1⁺ transgenic and littermate nontransgenic mice were infected intracerebrally with TMEV and CNS-ILs harvested after 7 days for the CTL assay. The CNS-ILs were incubated with ⁵¹Cr-labeled C57SV cells transfected with either the VP1 or VP2 transgene, and lysis was determined by the level of chromium release. (A) CNS-ILs from TMEV-infected VP1⁺ transgenic mice did not lyse VP1-transfected C57SV cells, whereas those from nontransgenic littermates did. (B) In contrast, CNS-ILs from both VP1 transgenic and VP1 nontransgenic littermate mice infected with TMEV lysed VP2-transfected cells. Data shown are means of triplicate samples using pooled lymphocytes from the CNSs of four or five mice per experimental group.

FIG. 4.

Lymphocyte proliferation assay following in vitro stimulation with ConA or VP1. Splenocytes were obtained from VP1⁺ transgenic mice and from control mice previously immunized with recombinant VP1 protein in combination with complete Freund's adjuvant. Splenocytes were stimulated in vitro with ConA or recombinant VP1 peptide, and stimulation indices were determined using a [³H]thymidine uptake assay. Splenocytes from VP1⁺ and VP1⁻ mice incubated with ConA (100 ng/well) proliferated 8- to 14-fold above the response with media alone. VP1⁺ and VP1⁻ mice showed similar proliferative responses against recombinant VP1 protein (5 ng/well).

FIG. 5.

Detection of VP1- and VP2-specific IgG in transgenic mice infected with TMEV. LP⁺ transgenic, VP1⁺ transgenic, and littermate control nontransgenic mice were infected intracerebrally with 2 × 10⁶ PFU of TMEV, and sera were collected 45 days later for antibody analysis. VP1- and VP2-specifc IgG levels were determined by ELISA (A and C) or Western blotting (B and D) using recombinant virus capsid antigens. Diluted sera from individual mice were added to the coated plates or blots, and bound IgGs were detected using an alkaline phosphatase system. TMEV-infected VP1⁺ transgenic mice generated antibodies against the VP1 protein (A and B). Similarly, TMEV-infected LP⁺ transgenic mice generated antibodies against the VP2 protein, which is encompassed within the LP region of the genome (C and D).

FIG. 6.

Demyelination in the spinal cord white matter of transgenic mice following TMEV infection. VP1⁺ (A), LP⁺ (B), RP⁺ (C), VP1⁻ (D), LP⁻ (E), and RP⁻ (F) mice were infected with TMEV for 45 days, and spinal cord blocks were then embedded in glycol-methacrylate plastic. Sections were stained with erichrome-cresyl violet stain to detect demyelination and inflammation. Focal areas of demyelination are shown for LP1⁺ (A) and LP⁺ (B) transgenic mice. No demyelination was observed for RP⁺ (C), VP1⁻ (D), LP⁻ (E), or RP⁻ (F) mice.

FIG. 7.

(A) Electron microscopy results, showing multiple demyelinated axons in the spinal cord of a VP1⁺ transgenic mouse which had been infected with TMEV for 45 days. Axons are well preserved and without ultrastructural abnormalities. (B) No demyelination was observed for a VP1⁻ littermate control mouse infected with TMEV.

FIG. 8.

Detection of TMEV RNA persistence in spinal cord white matter of TMEV transgenic mice. Using paraffin, spinal cord sections were prepared from mice intracerebrally infected with TMEV for 45 days. TMEV RNA was localized by hybridization with a ³⁵S-labeled probe corresponding to the VP1 region of TMEV. Viral RNA was detected in VP1⁺ (A), LP⁺ (B), and RP⁺ (C) transgenic mice but not in VP1⁻ (D), LP⁻ (E), and RP⁻ (F) nontransgenic littermate controls. Sections were counterstained with Mayer's hematoxylin. Magnification, ×600.

FIG. 9.

Severity of brain disease in VP1⁺, VP1⁻, LP⁺, and LP⁻ mice infected with TMEV for 45 days. Each symbol represents an individual mouse, graded in a blinded fashion for each area of the brain according to the scale detailed in Materials and Methods.