No evidence of XMRV in prostate cancer cohorts in the Midwestern United States

Abstract

Background: Xenotropic murine leukemia virus (MLV)-related virus (XMRV) was initially identified in prostate cancer (PCa) tissue, particularly in the prostatic stromal fibroblasts, of patients homozygous for the RNASEL R462Q mutation. A subsequent study reported XMRV antigens in malignant prostatic epithelium and association of XMRV infection with PCa, especially higher-grade tumors, independently of the RNASEL polymorphism. Further studies showed high prevalence of XMRV or related MLV sequences in chronic fatigue syndrome patients (CFS), while others found no, or low, prevalence of XMRV in a variety of diseases including PCa or CFS. Thus, the etiological link between XMRV and human disease remains elusive. To address the association between XMRV infection and PCa, we have tested prostate tissues and human sera for the presence of viral DNA, viral antigens and anti-XMRV antibodies.

Results: Real-time PCR analysis of 110 PCa (Gleason scores >4) and 40 benign and normal prostate tissues identified six positive samples (5 PCa and 1 non-PCa). No statistical link was observed between the presence of proviral DNA and PCa, PCa grades, and the RNASEL R462Q mutation. The amplified viral sequences were distantly related to XMRV, but nearly identical to endogenous MLV sequences in mice. The PCR positive samples were also positive for mouse mitochondrial DNA by nested PCR, suggesting contamination of the samples with mouse DNA. Immuno-histochemistry (IHC) with an anti-XMRV antibody, but not an anti-MLV antibody that recognizes XMRV, sporadically identified antigen-positive cells in prostatic epithelium, irrespectively of the status of viral DNA detection. No serum (159 PCa and 201 age-matched controls) showed strong neutralization of XMRV infection at 1:10 dilution.

Conclusion: The lack of XMRV sequences or strong anti-XMRV neutralizing antibodies indicates no or very low prevalence of XMRV in our cohorts. We conclude that real-time PCR- and IHC-positive samples were due to laboratory contamination and non-specific immune reactions, respectively.

Figure 1

Bayesian maximum clade credibility phylogeny of endogenous murine MLV sequences, 22Rv1 cell line and patient derived MLV gag gene sequences. Sequences derived from PCa samples in this study are colored red. Sequences from [16] are colored blue. The tree is rooted against the Moloney MLV sequence. Bayesian posterior probabilities above 0.50 are indicated on the corresponding branches. The scale bar represents the number of nucleotide substitutions per site.

Figure 2

PCR for mouse mitochondrial DNA. qPCR positive samples (#15, 51, 52, 53, 103, 112) were PCR amplified for mouse mitochondrial DNA. Positive samples yielded PCR products at 153 bp [16]. Water was used as a control.

Figure 3

Detection of XMRV in prostate cancer tissues. (A) Specificity of anti-XMRV antiserum and anti-MLV antibody. 293T cells transfected with XMRV infectious plasmid (pcDNA3.1(-)/VP62) were stained with either rabbit anti-XMRV or goat anti-MLV. No positive staining was observed in control uninfected 293T cells. (B) Serial tissue sections from qPCR positive samples, including #51 (Gleason score (GS) 8) and #103 (GS 6) were immunostained with either anti-XMRV or anti-MLV antibody. H&E staining from each sample is also shown. (C) Serial tissue sections from qPCR positive (#112, GS 6) and negative (#47, GS 8) samples were immunostained with anti-MLV antibody, followed by TexasRed-conjugated donkey anti-rabbit antibody (Jackson ImmunoResearch Laboratories, Inc., 1:200).

Figure 4

Neutralization activity of patient sera. (A) XMRV infected 293T cells (XMRV+ control) and XMRV-infected and treated with anti-XMRV sera at a dilution of 1:10 [14] are shown. (B) Data from non-XMRV infected 293T cells is shown as a control. Patients samples which did not show positive neutralization reaction (Patients with non-prostate cancer (NonPCa)-1, Patients with prostate cancer (PCa)-1) are shown. (C) Two samples that showed positive reaction from patients with prostate cancer (PCa-2 and -3) are shown. (D) Six samples that showed positive neutralization reaction from patients with non-prostate cancer (NonPCa-2 to -6) are shown. 1:10 dilution of sera were applied for all the experiments. Percent GFP positive and percent neutralization are indicated within the gated areas and below the flow data, respectively. The percent neutralization was calculated as the reciprocal of infectivity, with a maximum infectivity being determined by incubation of the virus with an uninfected mouse serum. n.a., not applicable.