Background:  We previously used nested gag PCR amplification strategies to report the prevalence of xenotropic murine leukemia virus-related virus (XMRV) among 293 participants with chronic fatigue syndrome (CFS), HIV, rheumatoid arthritis (RA), organ transplantation, or a general cohort of patients presenting for medical care. XMRV DNA was not detected in any participants’ peripheral blood mononuclear cell (PBMC) samples. Mouse endogenous retrovirus (MERV) sequence was detected in one RA subject but was not reproducibly amplified on subsequent attempts. We have since enrolled 2 CFS participants with outside laboratory evidence of XMRV or gammaretrovirus infection. We used a combined env and gag amplification strategy to identify XMRV or other gammaretroviral DNA in these 2 participants and performed a clonal analysis of the previously reported MERV⁺ sample.

Methods:  CFS subjects fulfilled the Centers for Disease Control and Prevention case definition criteria. One participant tested positive for XMRV and a second participant reported a positive blood test for a gammaretrovirus; both tests were performed at the same outside laboratory. In our laboratory, DNA was extracted from 5 million fresh PBMC; validated primer sets were used in 1 nested env and 2 separate nested gag amplification strategies. A range of input DNA quantities (200 to 600 ng) was used with positive and negative XMRV controls for each amplification. Human beta-globin was amplified to verify DNA integrity. A standard clonal analysis of the prior amplified MERV gag sequence was performed; 6 clones were bi-directionally sequenced.

Results:  We could reliably detect 1 env copy and 10 gag copies of XMRV control plasmid. We did not detect XMRV or other gammaretroviruses from 2 CFS participants at any input DNA concentration for both gag and env PCR amplifications. The 6 MERV clones isolated from 1 participant had 90 to 91% sequence identity to recently described CFS-associated murine leukemia virus (MLV)-like sequences. A Bayesian inference of phylogeny demonstrated clustering of MERV sequences separate from MLV-like sequences and the original CFS-associated XMRV sequences.

Conclusions:  XMRV was not detected in 2 patients with previous positive test results. Sequence analysis suggested significant differences between a non-reproducibly amplified MERV sequence and CFS-associated MLV-like and XMRV sequences. A standardized, cross-validated amplification strategy for gammaretroviral detection in clinical samples is urgently needed.