By Ben Butkus
A group led by researchers from the US Food and Drug Administration and National Institutes of Health has used nested PCR assays to detect gene sequences from a virus related to murine leukemia virus in patients diagnosed with chronic fatigue syndrome, a disorder that has no known cause or cure, according to a recently published paper.
The findings are in line with a previous study by different researchers that identified a similar virus called xenotropic MLV-related virus, or XMRV, in a high percentage of CFS sufferers, suggesting a link between the viruses and the disease, but not causality. However, the most recent research also conflicts with the results of several other research groups that failed to detect gene sequences from either virus in CFS patients.
The USDA and NIH researchers have identified differences in sample collection, processing methods, and PCR protocols and primers as potential reasons for the discrepant results, but also presented strong evidence that their own PCR methods were airtight and free of potential contamination.
Nevertheless, efforts are underway to compared different testing methods and develop standardized molecular testing for the viruses in an effort to further unravel the mysterious connection between various mouse leukemia retroviruses and chronic fatigue syndrome, the researchers said.
In the latest study, published online last week in the Proceedings of the National Academy of Science, the FDA and NIH researchers examined 41 peripheral blood mononuclear cell-derived DNA samples from 37 patients meeting accepted diagnostic criteria for CFS.
The researchers had originally obtained the serum and whole-blood samples from CFS patients in the mid-1990s to investigate possible mycoplasma infections, but saved 37 of the samples in frozen storage. Twenty-five of the patients were from an academic medical center and 12 were referred by community physicians. In addition, the researchers obtained repeat blood samples some 15 years later from eight of the academic medical center patients.
The research group developed nested PCR assays targeting the MLV-related virus gag gene using both previously described primer sets and an in-house-designed primer set with highly conserved sequences from different MLV-like viruses, as well as XMRVs, and tested DNA from the 37 blood samples for presence of the gag gene sequences. In parallel, the researchers tested DNA extracted from frozen PBMC samples of 44 healthy volunteer blood donors.
Their assays revealed MLV-like virus gag gene sequences in 32 of 37, or 86.5 percent, of samples, compared with only three of 44, or 6.8 percent, of samples from healthy volunteer blood donors. In addition, seven of the eight gag-positive patients again tested positive in a sample obtained 15 years later.
The researchers retrieved all PCR products and used Sanger sequencing to confirm that the sequences were in fact from MLV-related virus gag genes.
The results of the study seem to jive with data from a study conducted by researchers at the Whittemore Peterson Institute and published last year in Science. The authors of that study found that a high percentage of patients diagnosed with CFS were infected with XMRV, a closely related virus that was first identified in samples of human prostate cancer tissue about four years ago.
However, two subsequent studies failed to find an infectious MLV-related virus in prostate cancer patients from Germany; and four recent studies conducted in Europe and the US have failed when trying to use PCR to detect XMRV or MLV-related viral gene sequences in the blood of CFS patients.
In a conference call discussing the results of the most recent PNAS paper, corresponding author Harvey Alter, chief of clinical studies and associate director for research in the department of transfusion medicine at the NIH Clinical Center, noted that his group found that MLV-related virus had "a dramatic association with CFS, but that's all it is, an association. We have to emphasize, as we did in the paper, that we have not [identified] causality for this agent, and that there are many more pieces we have to fulfill. But we think basically it confirms the findings of the Whittemore Peterson group."
Alter also noted that other laboratories, including at the US Centers for Disease Control and Prevention, have not been able to find the same association.
"As with many things in science, it's controversial, and not everybody finds the same thing," Alter said. "Very good laboratories have come up with disparate results, and this is not totally explained. We think there are reasons for this, and we think it's in the patient populations, rather than the laboratory testing, but that hasn't completely been ruled out.
"So the dilemma at present is how to reconcile how some labs find this association and others do not," Alter added. "There are many possible explanations for that, and probably the most common voice is that CFS is a symptom complex. There is no specific tissue you can biopsy and no specific test to say 'this is a CFS patient.' It's probably a spectrum, some of which may be associated with XMRV or MLV, and others not at all."
In their paper, seemingly anticipating questions regarding potential reason for the differing results, the NIH and FDA researchers devoted an entire section to debunking the idea that their PCR results could have been false positives due to viral gene-specific primers nonspecifically amplifying nucleic acid sequences that differ from the target or to potential contamination from mouse genetic material.
Specifically, the researchers noted that they re-tested every one of their positive results for evidence of mouse DNA contamination using a semi-nested PCR assay exponentially more sensitive in detecting mouse mitochondrial DNA than MLV-related gag sequences, and found no hits. They also carefully reviewed how the clinical samples were collected and the conditions under which the PCR assays were performed, and found no evidence for contamination.
Lastly, they noted that they amplified at least six different MLV-related gag gene sequences from the CFS patients' blood; and that the observed sequences all had significant variations from previously reported exogenous MLVs or viral vectors.
"In fact, we have very good confidence that this cannot be a laboratory artifact or a PCR-contamination product, because the sequences vary so much from one patient to another," Shyh-Ching Lo, lead author on the paper and director of the Tissue Safety Laboratory Program at the FDA, said during the conference call. "Normally if you had PCR contamination, you would more or less anticipate that all the sequences would be the same."
The researchers also suggested in their paper that heterogeneity in gag gene sequences suggest that geographic differences in different MLV-related viruses may be considerable and could affect both the sensitivity and specificity of molecular amplification using standard primer sets, thereby accounting for the differing results of other research groups.
During the conference call, Lo also conceded that the diversity of the gene sequences are actually a reason that it has been and will continue to be difficult to develop a standardized molecular test for presence of either XMRV or MLV-related virus.
"Certainly the diversity of the sequence will make this test more difficult, particularly if we want to develop a standard PCR test and try to cover a broad spectrum of different sequences," Lo said. "But in the present study, the sequence[s] that we used are actually present for both the XMRV and the polytropic MLV; we tried to use the more conserved primer sequence."
Lo also admitted that this "does not really explain the present difference, because most of the laboratories are using similar primers; or the primer actually covers both XMRV and MLV. However, even that different primer … will have a different sensitivity or specificity than we are dealing with in the clinical sample. There is a much more complex human genome sequence there."
However, ideally the researchers would "like to be able to standardize the test so different laboratories can be doing the same thing, and seeing the same thing, and have consistent findings," Lo added. "This is still in a very early stage … but eventually I think a more standardized test will be developed."
In fact, such a study is already underway at the NIH's National Heart Lung and Blood Institute, according to Steve Monroe, director of the division of high-consequence pathogens and pathology at the CDC, and a co-author on one of the previous studies that did not find a link between XMRV or MLV-related virus and CFS.
This study, he said during the conference call, "is looking to address some of these questions by collecting samples from a small number of concerned cases and having those processed in different ways and tested by a number of different laboratories to compare different assay methods."
In the end, Monroe noted, the results of the FDA and NIH study "do raise as many questions as they answer. Clearly the different findings from different labs that have been reported suggest that there are a lot of things about this virus that we don't know. In particular there are a lot of things about the biology of the virus and how it interacts with humans — where it replicates in humans; what kind of disease it causes, if any; what are the best samples to use for diagnosis; how those samples should be collected and processed; and what tests should be used."