Summary Report: The Jackson Laboratory DOES NOT Have MHV
JAX® Bulletin #3, Addendum 3
May 22, 2000
A COMMUNICATION FROM THE JACKSON LABORATORY ANIMAL HEALTH STAFF: Peggy J. Danneman, VMD, MS, dip ACLAM (Chief of Veterinary Services), James R. Fahey, PhD, DVM, dip ACVM (Chief of Diagnostic Services), Terrie Cunliffe-Beamer, DVM, MS, dip ACLAM (Head of Clinical Laboratory Animal Medicine), John P. Sundberg, DVM, PhD, dip ACVP (Head of Pathology, Senior Staff Scientist)
It came to our attention in the summer of 1999 that several institutions throughout the United States had experienced as yet unexplained outbreaks of mouse hepatitis virus (MHV) infection. Several of our customers called us requesting our help in determining whether JAX® Mice might be the source of the infection.
Although our rigorous routine monitoring program has shown that we have had no MHV at The Jackson Laboratory for over 18 years, we immediately began an intensive investigation into the possibility that we might have an undetected nidus of infection within our mouse colonies. We have now completed that investigation, which involved all of our Production mouse rooms. As you will see from the information below, we have extensive and overwhelming evidence that JAX® Mice do NOT have MHV.
MHV is a comprehensive term used to refer to a large group of single-stranded RNA viruses of the genus Coronavirus. At least 25 distinct strains of MHV have been identified, and it is likely that there are others, as yet unidentified. These strains vary widely in pathogenicity, and pathogenicity is further influenced by the genotype, age, and other characteristics of the host mouse. The strain of MHV associated with the recent outbreak in the United States appears to be a particularly nonpathogenic virus. All known strains of MHV are highly contagious and are readily transmitted by direct contact, aerosol, and fomites. MHV is also a common contaminant of transplantable cell lines.
Infection in immunocompetent mice typically runs its course within 3 weeks, although this too appears to vary with both virus strain and host genotype. Immunocompromised mice, such as athymic nude (Hfh11nu) mice, tend to develop chronic infections. The tissues infected also vary, but virus replication most commonly occurs in the nasal passages, lungs, and intestine. Spread to other tissues such as liver, brain, and lymph nodes is variable. During the phase of active viral replication, virus is shed in the feces and by aerosol.
The diagnostic tests most commonly used to detect MHV infection in immunocompetent mice are the enzyme-linked immunosorbent assay (ELISA) and indirect immunofluorescence assay (IFA). With proper performance and expert interpretation of results, these tests can provide accurate information about previous infection of immunocompetent mice with MHV. A commonly encountered difficulty with these tests is the occurrence of nonspecific reactions. For example, with the ELISA, the serum of mice from many strains (e.g., C57BL/6J and mice with certain abnormalities of the immune system) can react "positively" with both the uninfected cell extract control and the MHV infected cell extract.
A "positive" reaction with the uninfected control indicates that the serum has antibodies that bind to uninfected assay wells in addition to wells containing MHV. The presence of these antibodies confounds interpretation of the test, although the results are clearly different from true positives (in which the serum reacts only with the MHV infected extracts). A similar problem may occur with IFA, in which the cell culture controls and MHV infected cells may both bind nonspecific antibodies. The recent outbreaks of MHV in the U.S. have been detected primarily by ELISA or IFA.
Reverse transcriptase polymerase chain reaction (RT-PCR) assays have recently been developed to detect the presence of viral RNA in the feces and tissues of actively infected mice. At this time, RT-PCR is not a well-standardized or consistently reproducible assay among different laboratories testing for MHV; both false positive and false negative results sometimes occur.
Our investigation involved extensive serologic screening of all of our Production mouse colonies, along with more in depth investigations of autoimmune mouse colonies in two Jackson Laboratory animal rooms, AX-1 and AX-8. These autoimmune colonies were targeted for a more comprehensive investigation for two reasons: 1) calls from customers suggested that mice from these colonies might have been associated with outbreaks of MHV in their institutions; and 2) mice prone to the development of autoimmune disorders have a high rate of nonspecific reactivity in serologic tests, including ELISA and IFA tests for MHV.
General Serologic Screening Results Blood samples from 1,489 mice from 18 rooms were tested serologically (ELISA) for antibodies to MHV. Most of the samples (1,473) were negative, whereas 16 samples gave nonspecific results. No positive results were found upon IFA retesting of any samples that originally gave nonspecific results by ELISA. In addition, aliquots of serum from 25 of the original 1,489 samples were sent to Charles River Laboratories (Wilmington, MA) for confirmatory IFA analysis. All results were negative.
Data from Autoimmune Mice in AX-1
Two strains of mice in AX-1, B6.MRL-Faslpr (lpr/lpr) and B6Smn.C3H-Faslgld (gld/gld), were potentially implicated in MHV outbreaks in customers' institutions. An intensive investigation involving serologic, histologic, epidemiologic, and molecular techniques was undertaken to determine whether these allegations could be correct.
1. As with all mice that are prone to the development of autoimmune disease, lpr/lpr and gld/gld homozygous mice often present problems for serologic testing. These mice have high levels of serum antibodies that interfere with the interpretation of serologic tests such as ELISA. On initial serologic testing (ELISA), three of eight samples from lpr/lpr mice and two of eight samples from gld/gld mice were negative for MHV. The other samples showed nonspecific reactions.
2. Because lpr/lpr and gld/gld mice are so difficult to test directly, immunologically normal sentinelmice were placed in cages with lpr/lpr and gld/gld weanlings (the age at which active virus infections have most commonly been reported to occur in colonies with endemic MHV). DBA/2J mice were chosen as sentinels because this strain is known to be particularly sensitive to MHV and will mount a robust, serologically detectable immune response on infection. DBA/2J mice were put in each of 21 cages holding 10-18 lpr/lpr or gld/gld weanlings. Our plan was to maximize the likelihood of detecting any MHV-related histological lesions by removing half the sentinels after 3 weeks, and to maximize the likelihood of finding any antibodies to MHV by removing the remaining sentinels after 6 weeks.
Blood and fecal samples were taken from all sentinels after 3 weeks, and half of the sentinels were sent to The Jackson Laboratory's diagnostic laboratory for necropsy. Nineteen serum samples were also sent to Charles River Laboratories.
Blood and fecal samples were taken from the remaining 20sentinels after 6 weeks, and all were then necropsied.
Serology Results
Serum samples obtained from 40 sentinels after 3 weeks were serologically (ELISA)negative for MHV when tested in The Jackson Laboratory's diagnostic laboratory. The 19 samples sent to Charles River Laboratories were also negative (ELISA). Samples obtained from 20 sentinels after 6 weeks were also serologically (ELISA) negative for MHV.
RT-PCR Results
All 39 fecal samples taken from the sentinels after 3 weeks, and all 20 samples taken from sentinels after 6 weeks, were negative for MHV RNA by RT-PCR. Fecal samples were also taken from 11 lpr/lpr weanlings and 7 gld/gld weanlings after 3 weeks; these mice all came from cages holding sentinels. They were all negative for MHV RNA by RT-PCR.
Necropsy Results
No lesions other than common, well characterized, strain specific abnormalities were found in any of the 60 mice necropsied.
3. We performed serologic (ELISA) testing on immunologically normal mice from 70 cages located adjacent to the lpr/lpr and gld/gld colonies. These mice are handled by the same caretakers who care for the lpr/lpr and gld/gld mice. All mice in this room are housed in non-ventilated cages and are changed on open changing tables. Because MHV is so highly infectious, we would expect the infection to spread rapidly in such an environment if it was present. Nonetheless, all tests were negative for MHV.
4. Preweanling mice are often the most susceptible to MHV infection. For this reason, we took intestinal tissue from three 10-21 day old lpr/lpr mice and three 10-21 day old gld/gld mice and sent the samples to an outside laboratory, Anmed Biosafe, Inc. (Rockville, MD), for RT-PCR analysis. We also sent intestinal tissue from three 10-21 day old mice of a third autoimmune disease-prone strain, MRL/MpJ-Faslpr mice. In Anmed Biosafe's assay, all samples were negative for MHV.
5. Pooled fecal samples from each of 13 cages holding a total of 135 weanling lpr/lpr mice and 67 weanling gld/gld mice were sent to Anmed Biosafe, Inc. for RT-PCR testing. Of the 13 samples analyzed, 1 sample from each strain gave a PCR product with the MHV primers used. It is likely that these results were false positives based on the overwhelmingly negative results obtained from these colonies by serology, histology, PCR, and electron microscopy (above). Also, when the original samples were reanalyzed by RT-PCR in The Jackson Laboratory's diagnostic laboratory using a different primer set that recognizes a different MHV nucleotide sequence, the results were negative. Furthermore, mice that were still available from one of the cages that yielded questionable results on RT-PCR had not become serologically positive (ELISA) for MHV when tested 1 month later. Had these animals truly been exposed to MHV, they should have seroconverted within this time.
Data from Autoimmune Mice in AX-8
Very early in our investigation, pooled fecal samples from each of 6 cages holding a total of 50 MRL/MpJ-Faslpr (MRL) mice from AX-8 were sent to Anmed Biosafe, Inc., for RT-PCR testing.
Like the lpr/lpr and gld/gld strains, MRL mice are prone to the development of autoimmune disease. Five samples yielded negative results and 1 sample gave a PCR product with the MHV primers used.
1. As follow up to this single positive sample, the original sample was re-analyzed by TPCR in The Jackson Laboratory's diagnostic laboratory using a different primer set that recognizes a different MHV nucleotide sequence. The results were negative.
2. Blood samples from 768 mice in AX-8 were tested by ELISA for antibody to MHV. 746 samples were serologically negative for MHV and 22 samples gave nonspecific reactions. Fecal samples from the 22 animals whose sera gave nonspecific reactions were tested for MHV RNA using RT-PCR; all were negative.
3. A sentinel program was initiated to test for the presence of MHV among MRL mice in AX-8. Each week for 6 weeks, weanling (3 wk old) MRL mice were placed in a "sentinel" cage with one immunocompetent (various strains, including BALB/cJ, CBA/J, and FVB/NJ) and one immunodeficient (C57BL/6J-Prkdcscid/SzJ or C3HSmn.C-Prkdcscid/J) weanling mouse. Each "sentinel" cage contained 5-8 MRL weanlings from different litters in addition to the 2 sentinels. This approach allowed us to test progeny from all currently productive breeding cages in the MRL colony. Fecal samples from all sentinels were tested by RT-PCR 2, 4, 6, and 8 weeks after placement in the "sentinel" cages, and samples from the MRL weanlings in these cages were tested at 2 wk. Serum samples from all sentinels in the "sentinel" cages were tested by ELISA after 4 and 8 weeks. At the end of the study, 26 immunodeficient sentinels were sent to The Jackson Laboratory's diagnostic laboratory for necropsy.
Serology Results
All 67 serum samples from the MRL weanlings and immunocompetent sentinels were negative (ELISA) for antibodies to MHV.
RT-PCR and Electron Microscopy Results
A total of 327 fecal samples from immunocompetent sentinels, immunodeficient sentinels, and MRL weanlings from AX-8 were tested by RT-PCR using 1, 2, or 3 primer sets, each of which recognizes a non-overlapping region of the MHV genome. All samples were negative using two of the primer sets - our primary set which recognizes an MHV nucleotide sequence and a second set which recognizes an MHV matrix sequence. Fourteen samples were positive using a third primer set that recognizes another MHV nucleotide sequence. Fresh fecal samples were collected from these 14 putatively infected mice, some of which were severely immunodeficient, and tested again by RT-PCR using all three primer sets. All results were negative. Fecal samples from these mice were also examined by negative stain electron microscopy. No MHV like virus particles were found. All of the putative-infected mice that were immunocompetent were also retested serologically (ELISA). All results were negative.
Necropsy Results
No lesions other than common, well characterized, strain specific abnormalities were found in any of the 26 immunodeficient mice necropsied at the end of the study.
In summary, this overwhelming evidence confirms that JAX® Mice do NOT have MHV. All of the serologic, epidemiologic, and histologic data we have gathered from 2,450 mice indicate that JAX® mice are free of MHV. In addition, of 414 samples evaluated by RT-PCR, the vast majority were negative. Although 17 positive results were obtained by RT-PCR, it must be concluded that these results were false positives. This conclusion in based on the fact that all of the putatively-infected mice were negative for MHV RNA when retested by RT-PCR. In addition, fecal samples from 14 of the mice were negative for MHV virus particles when examined by electron microscopy. Finally, all immunocompetent mice that yielded positive results by RT-PCR were negative for antibodies to MHV when tested and, in most cases, retested, by ELISA.
The Jackson Laboratory has a long history of supplying genetically diverse mice of the highest quality. Our comprehensive health monitoring program is designed to assure the users of JAX® Mice that their animals will be free of pathogens such as MHV. We want to take this opportunity to reassure the scientific community of our commitment in this regard. If you have any questions regarding our MHV investigation or any other aspect of animal health at The Jackson Laboratory, please call us at 800-422-6423.
