Nested PCR amplification is an unlucky choice for detection of XMRV

Antoinette C. van der Kuyl,* Ben Berkhout, J Formos Med Assoc 2011;110(5):273–274:

Laboratory of Experimental Virology, Department of Medical Microbiology, Centre for Infection and Immunity Amsterdam
(CINIMA), Academic Medical Centre of the University of Amsterdam.
*Correspondence to: Dr Antoinette C. van der Kuyl, Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands.
E-mail: a.c.vanderkuyl@amc.uva.nl

The best evidence for XMRV replicating in
human cells is the detection of proviral integrations
flanked by human genome sequences in
prostate tissue from eleven patients.6,7

The flanks are human and not mouse sequences, and the
insertion sites differ for each clinical sample,
thus ruling out a mouse-contamination and demonstrating independent infection cases.

However, a recent paper suggested that the XMRV integration
sites in the patient material could have
resulted from laboratory contamination with artificial
XMRV cell lined-derived integration sites.8

In situ fluorescent hybridization of prostate cancer
tissue has detected XMRV proviral genomes
that correlated with neutralizing antibody reactivity
and/or PCR results in patients.9

These findings indicate that even though it is likely that
many positive cases may have to be dismissed as
representing contamination with mouse DNA,
genuine human infections with XMRV seem to
exist at a relatively low prevalence.

In retrospect, the method used in most studies
for detection of XMRV, almost exclusively
nested PCR amplification, has been a rather unlucky
choice. In clinical virology, a putative virus
infection is usually probed with a serological
assay, although real time PCR assays are currently
in use for infections in which a high copy number
of viral nucleic acid (RNA or DNA) is known
to correlate with disease. A simple detection PCR
is rarely the first method of choice for several reasons,
including the danger of getting false positives
due to contamination. Serological assays for
XMRV are currently being developed, and will be
much needed to distinguish genuine infected patients
from contamination cases.

Serology will be more informative than running a control PCR
for mouse DNA as contamination by XMRV particles
(e.g. from protein preparations or contaminated
cell cultures) will not show mouse DNA,
other than the viral genome. Read more>>