Absence of XMRV retrovirus and other murine leukemia virus-related viruses in patients with chronic fatigue syndrome

Abstract

Chronic fatigue syndrome (CFS) is a multisystem disorder characterized by prolonged and severe fatigue that is not relieved by rest. Attempts to treat CFS have been largely ineffective primarily because the etiology of the disorder is unknown. Recently, CFS has been associated with xenotropic murine leukemia virus-related virus (XMRV) as well as other murine leukemia virus (MLV)-related viruses, though not all studies have found these associations. We collected blood samples from 100 CFS patients and 200 self-reported healthy volunteers from the same geographical area. We analyzed these in a blind manner using molecular, serological, and viral replication assays. We also analyzed samples from patients in the original study that reported XMRV in CFS patients. We did not find XMRV or related MLVs either as viral sequences or infectious viruses, nor did we find antibodies to these viruses in any of the patient samples, including those from the original study. We show that at least some of the discrepancy with previous studies is due to the presence of trace amounts of mouse DNA in the Taq polymerase enzymes used in these previous studies. Our findings do not support an association between CFS and MLV-related viruses, including XMRV, and the off-label use of antiretrovirals for the treatment of CFS does not seem justified at present.

Fig. 1.

Study scheme for collection, processing, and analysis of blood samples from CFS patients and healthy volunteers.

Fig. 2.

Defining qPCR assay characteristics. (A) Reproducibility of all XMRV qPCR assays (pol, env, LTR, gag) performed in triplicate with pXMRV1 template amounts of 500, 50, and 5 copies. R² values show high reproducibility for each assay. (B) Sensitivity of XMRV LTR qPCR assay with a 6-carboxyfluorescein (FAM) reporter showing that the assay was linear with as low as 5 copies of pXMRV1 template added in a background of 400 ng of human placental DNA. R_n is the difference in fluorescence between the FAM reporter and the standard reference ROX dye. (C) Testing for cross-reactivity (or specificity) of XMRV qPCR assays against other common human pathogenic viruses. A positive clinical sample or plasmid DNA from a variety of common pathogenic viruses was amplified using the LTR and pol (shIN) qPCR assays. No significant cross-reactivity was seen with any of the following: BK virus, cytomegalovirus (CMV), Epstein-Barr virus (EBV), enterovirus (EV), human herpesvirus 6 variant A (HHV6-A), human herpesvirus 6 variant B (HHV6-B), human immunodeficiency virus (HIV), human metapneumovirus (HMPV), influenza A virus (FLUA), and influenza B virus (FLUB). Each sample was extracted with an exogenous internal control (IC) plasmid IC2, containing the Caenorhabditis elegans pax1/9 gene fused to green fluorescent protein (GFP), added to each aliquot of whole blood prior to sample extraction. This internal control plasmid was coamplified with each sample to identify potential inhibitors of PCR and to monitor extraction efficiency. Extraction was efficient, as shown by the IC C_T range of 33.5 to 35.9.

Fig. 3.

ELISA to measure reactivity of human sera against recombinant XMRV SU protein. Sera from CFS patients, healthy males, and healthy females in reaction with gp70 Env (SU) fragment were tested in three replicates. Arrow indicates average OD values using XMRV antiserum at 1:10,000 (22).

Fig. 4.

(A) Western blot analysis of a subset of CFS patient (P) and control (C) samples 6 weeks after spin inoculation of plasma onto cultured LNCaP cells. Cell lysates from XMRV-infected cells probed with rabbit anti-XMRV antisera are shown in the lane labeled “+.” The lane labeled “−” represents cell lysates from uninfected cells probed with rabbit anti-XMRV antisera. M, molecular weight marker. (B) To indicate loading amounts, the same gel as that used for panel A was probed with antitubulin antibody.

Fig. 5.

Nested-PCR assay. (A) MluI qPCR assay to detect pXMRV1 contamination. The 5′ end of the probe spans the MluI restriction site that was introduced to create pXMRV1. pAO-H4, which does not have the MluI restriction site, has lower peak fluorescence as well as a delay in C_Ts for the same copy numbers of plasmid. (B) Sensitivity of the IAP qPCR assay for different amounts of C57BL/6 mouse DNA ranging from 62.5 ng to 625 ag, all in the presence of 400 ng of human placental DNA. (C) Nested-PCR assay of a small set of samples, showing ∼5% of the samples to be positive for MLV gag sequences using NP116/NP117 (11). LNCaP genomic DNA is shown in lanes G. The lane labeled “−” represents the negative control. Lane l shows the 100-bp ladder. (D) Detection of mouse DNA in Platinum Taq (IP Taq, Invitrogen) and Recombinant Taq (IR Taq, Invitrogen) but not in AmpliTaq Gold (AAG Taq, Applied Biosystems). For each qPCR assay, the left column shows the number of replicates that are positive, and the right column shows the average C_T at which positivity occurred. The more-XMRV-specific pol qPCR assay (in triplicate) was consistently negative, but IAP and gag assays (eight replicates each) were both positive, as more Platinum or Recombinant Invitrogen Taq was used as a template.