Xenotropic murine leukaemia virus-related virus is not found in peripheral blood cells from treatment-naive human immunodeficiency virus-positive patients


Corresponding author: F. Maggi, Virology Unit, Pisa University Hospital, Azienda Ospedaliera Universitaria Pisana, Via San Zeno, 35–37, 56127 Pisa, Italy
E-mail: fabrizio.maggi63@gmail.com


Clin Microbiol Infect 2012; 18: 184–188


The human pathogen xenotropic murine leukaemia virus-related virus (XMRV) has been tentatively associated with prostate cancer and chronic fatigue syndrome. Unfortunately, subsequent studies failed to identify the virus in various clinical settings. To determine whether XMRV circulates in humans and the relationship with its host, we searched for the virus in 124 human immunodeficiency virus-infected patients who might have been exposed to XMRV, might be prone to infection as a result of progressive immunodeficiency, and had not yet been treated with antiretroviral drugs. Using nested PCR and single-step TaqMan real-time PCR, both designed on the XMRV gag gene, we could not find any positive samples. These findings add to the growing amount of scepticism regarding XMRV.


Xenotropic murine leukaemia virus-related virus (XMRV) is a novel gamma retrovirus of humans that was first discovered in 2006 [1]. Originally, the virus was isolated from biopsy specimens of prostate cancer patients, and was believed to play an aetiological role in that disease; however, subsequent studies failed to reveal the presence of XMRV in large numbers of patients with prostate cancer, possibly because of uneven geographical distribution of the infection and/or differences in the sensitivity of the PCR methods used for detection [2–8]. In the past 4 years, several reports have documented the presence of XMRV in blood and respiratory secretions of subjects with chronic fatigue syndrome, adults with respiratory illness and transplant patients at rates ranging from 2% to 86%, suggesting that the virus is associated with various human diseases or is an endogenous virus that may take advantage of chronic inflammation processes or other underlying diseases (Table 1) [9–17]. To date, however, the aetiological role of XMRV, if any, is far from being proven. Moreover, the finding that tissues and biological fluids of healthy subjects also carry XMRV has led to scepticism regarding the real pathogenicity of the virus [5,6,9–11,13,15,16,18].

Table 1.   Summary of studies carried out to detect xenotropic murine leukaemia virus-related virus in patients affected by various diseases and healthy controls; all studies were performed on peripheral blood leukocytes unless otherwise specified
DiseaseNo. of samplesPrevalence (%)MethodEthnicityReferences
  1. BAL; bronchoalveolar lavage; COPD, chronic obstructive pulmonary disease; HCV, hepatitis C virus; HIV, human immunodeficiency virus; PBMC, peripheral blood mononuclear cell; QQ, R462Q-homozygous; RQ, R462Q- heterozygous; TS, tracheal secretion.

HIV-positive1240Nested PCR and real-time PCR for gagItalyPresent study
Prostate cancer (biopsies)8000Real-time PCR for gag for and immunohistochemistry for Gag and Env cleavage productsUSAAloia et al. [2]
8640 (RNase L QQ)
1.5 (RNase L RQ)
Real-time PCR for gagUSAUrisman et al. [1]
744Real-time PCRThe NetherlandsVerhaegh et al. [8]
5890Nested PCR and RT-PCRGermanyHohn et al. [6]
3346–23Real-time PCR and immunohistochemistryUSASchlaberg et al. [7]
1051Nested PCRGermanyFischer et al. [5]
14422Nested PCR for envUSADanielson et al. [4]
Chronic fatigue syndrome39 Gag- and Env-ELISA assays; nested PCRGermanyHohn et al. [31]
1700Real-time PCRUKGroom et al. [13]
3786.5Nested PCR for gagUSALo et al. [10]
760Real-time PCR assay for integrase gene and/or a nested PCR for gagThe Netherlandsvan Kuppeveld et al. [14]
650Multiplex real-time PCR and RT-PCRChinaHong et al. [15]
1020Western blot, ELISA; gag and pol nested PCRUSASwitzer et al. [16]
10175  Mikovits et al. [12]
1860Nested PCRUKErlwein et al. [17]
320Nested PCR assay for gagUSAHenrich et al. [29]
10167Nested PCR assay for gagUSALombardi et al. [9]
Multiple sclerosis with fatigue symptoms1120Gag- and Env-ELISA assays; nested PCRGermanyHohn et al. [31]
Autistic spectrum disorders2300Real-time PCR and serologyItaly and South CarolinaSatterfield et al. [18]
HCV-positive670PCR for gag or envUnited KingdomBarnes et al. [30]
HIV-positive101 (acute)
133 (chronic)
0PCR for gag or envSwitzerland and SpainBarnes et al. [30]
430Nested PCR assay for gagUSAHenrich et al. [29]
199 plasmas
19 PBMCs
50 culture supernatants
0Nested PCR or RT-PCR and real-time PCR assaysCameroon and UgandaTang et al. [28]
54 (seminal plasma)0Nested PCR for gagThe NetherlandsCornelissen et al. [27]
Rheumatoid arthritis970Nested PCR assay for gagUSAHenrich et al. [29]
Spondyloarthritis190Nested PCR for envFranceJeziorski et al. [32]
Paediatric haematological, neurological or inflammatory pathologies620Nested PCR for envFranceJeziorski et al. [32]
Paediatric respiratory diseases (nasopharyngeal aspirates)800Nested PCR for envFranceJeziorski et al. [32]
Respiratory tract infection without underlying disease42 (sputum, nasal swab)2.3Nested PCR or real-time PCRGermanyFischer et al. [11]
Respiratory tract infection without underlying COPD31 (BAL)3.2Nested PCR or real-time PCRGermanyFischer et al. [11]
Haematopoietic stem cell or solid organ transplant161 (BAL, TS)9.9Nested PCR or real-time PCRGermanyFischer et al. [11]
260Nested PCR assay for gagUSAHenrich et al. [29]
Patients presenting for medical care950Nested PCR assay for gagUSAHenrich et al. [29]
Healthy controls400gag- and env-ELISA assays; nested PCRGermanyHohn et al. [31]
70 (prostate biopsy)1.3Nested RT-PCRGermanyFischer et al. [5]
62 (BAL, throat swab)3.2Nested PCR or real-time PCRGermanyFischer et al. [11]
970Western blot, ELISA; gag and pol nested PCRUSASwitzer et al. [16]
446.8Nested PCR for gagUSALo et al. [10]
2183.7Nested PCR for gagUSALombardi et al. [9]
3950Quantitative PCRUKGroom et al. [13]
2040Real-time PCR and serologyItaly and USASatterfield et al. [18]
650Multiplex real-time PCR and RT-PCRChinaHong et al. [15]

Many aspects of the natural history and pathogenesis of XMRV are still poorly understood [19]. These include the prevalence in the general population and in categories of subjects who may have greater chances of acquiring XMRV infection, and the relationship that the virus establishes with the infected host. Among these individuals, those with human immunodeficiency virus (HIV) infection are particularly at risk, as they have most likely been exposed to other infectious agents, and, if acquired, XMRV could have taken advantage of the progressive immunodeficiency in order to establish a persistent infection in the host.

We studied a cohort of 124 HIV-positive patients who had never received antiretroviral therapy. The presence of XMRV infection in peripheral blood was investigated by using two independent molecular methods. Furthermore, in light of the conflicting evidence on the appropriateness of investigated cohorts and methods used to look for the virus, we reviewed previous studies showing the advantages and disadvantages of the methodology used.

Materials and Methods

Patients and samples

The study group consisted of 124 randomly selected patients (97 males and 27 females; mean age 47 ± 14 years; range 18–80 years). All of them were HIV-seropositive as assessed by commercial immunoenzymatic and immunoblotting assays, and HIV-viraemic as assessed by commercial molecular assays. All patients were treatment-naive. After informed consent had been obtained, peripheral blood was sampled in EDTA tubes from each patient. CD4+ T-lymphocyte counts in peripheral blood at the time of sampling were recorded for each patient. Aliquots of whole blood and plasma were immediately prepared, stored at −80°C, and kept frozen until use.


XMRV infection was assessed by extracting DNA from 400 μL of whole blood with the Maxwell 16 System, according to the manufacturer’s instructions (Promega, Madison, WI, USA), using 500–750 ng of DNA as PCR template. The presence of virus was determined by using the protocol described in part by Lombardi [9]. This includes a nested PCR and single-step TaqMan real-time PCR, both designed on XMRV gag. In the nested PCR, a 413-bp fragment was amplified in the first PCR. This was carried out for 35 cycles with sense primer GAG-O-F (5′-CGCGTTCGATTTGTTTTGTT-3′) and antisense primer GAG-O-R (5′-CCGCCTCTTCTTCATTGTTC-3′), under the following conditions: melting at 94°C for 30 s, annealing at 52°C for 30 s, and extension at 72°C for 45 s. The product of this reaction was then reamplified for 35 cycles with internal primers GAG-I-F (5′-TCTCGAGATCATGGGACAGA-3′) and GAG-1-R (5′-AGAGGGTAAGGGCAGGGTAA-3′) under the same PCR conditions except for 54°C for annealing. The reactions were carried out in a 50-μL PCR mixture containing Taq DNA polymerase, each dNTP at a concentration of 12.5 mM, primers (20 pmol/μL each), and optimized buffer components. All samples were tested at least in duplicate and on different occasions. The amplified product was analysed by electrophoresis on an agarose gel after ethidium bromide staining. Amplicon size was compared with standard molecular size markers. XMRV presence was also evaluated by a single-step TaqMan real-time PCR, designed on a 83-nucleotide fragment of gag and recently developed in our laboratory. Briefly, the assay was performed in a 25-μL format containing 0.9 μM each primer (Q445T, 5′-GGACTTTTTGGAGTGGCTTTGTT-3′; Q528R, 5′-GCGTAAAACCGAAAGCAAAAAT-3′), 0.1 μM labelled probe (F480PRO: FAM-ACAGAGACACTTCCCGCCCCCG-BHQ), 5 μL of extracted viral DNA, and a Master Mix containing dNTPs and Taq DNA polymerase. The reaction was run in triplicate for each sample in a iQCYCLER real-time PCR detection instrument (Bio-Rad Laboratories, Hercules, CA, USA), with a previously standardized program (95°C for 10 min; 55 cycles of 95°C for 15 s, and 60°C for 60 s). PCR data were collected and analysed with the iQ5 optical system software, version 2.1, developed by Bio-Rad. To exclude carryover contamination, negative controls were added during the DNA extraction and PCR amplification steps. To validate the amplification process, positive controls obtained from supernatants of a constitutively XMRV-infected B-cell line were run in each PCR. The assay sensitivities were measured by testing quadruplicates of ten-fold dilutions of the XMRV-positive control. The real-time PCR was found to have a sensitivity (about 100 DNA copies per millilitre of whole blood) of at least a dilution higher than that of nested PCR.


HIV viraemia ranged between 21 and 5 620 000 RNA copies per millilitre of plasma, with a median of 253 log10, revealing that plasma HIV load was relatively low in the majority of patients. The median level of CD4+ T-lymphocytes was 331 cells/μL: 36 (29%) subjects showed CD4+ T-lymphocyte counts below 200 cells/μL, whereas in 28 patients, the absolute numbers of CD4+ T-lymphocytes were within normal values for people of the same age (e.g. 500–1500 cells/μL). XMRV infection was first examined by nested PCR. No patient had detectable viral DNA in the blood (Table 1). As the sensitivity of nested PCR is lower than that of real-time PCR, which became available after the study commenced, all patients were retested with the latter method. Again, no evidence of XMRV DNA could be found, suggesting that this novel retrovirus was not present in whole blood of study patients.


The existence and clinical significance of XMRV is still disputed [19]. Although the virus was initially isolated from patients with prostate cancer and CFS, subsequent findings have failed to unequivocally link XMRV to these diseases. To further complicate this picture, it has been recently suggested that XMRV detected in patient samples is probably the result of PCR contamination with mouse DNA, and that the tumour cell line 22Rv1, from which the original XMRV clone was isolated and characterized, was probably infected by murine XMRV during xenografting in mice [20,21]. Supporting this idea is the very high similarity between XMRV and its murine counterpart. As compared with the latter, however, XMRV has a 24-bp deletion within the gag leader sequence [22], which is the genomic region targeted by most molecular methods developed to detect the XMRV genome. This leaves room for alternative hypotheses [23]. Supporting the exogenous origin is the observation that a few human tissues do not to express the human innate antiviral resistance factor APOBEC3G, and may thus create favourable conditions for XMRV replication and dissemination in the host [24]. These facts and the lack of identification of XMRV or XMRV-related sequences in the overwhelming majority of studies published so far (Table 1) strengthen the hypothesis that XMRV is a zoonosis rather than a genuine human pathogen. Whatever the origin, it is necessary to determine cell types and organs that support viral replication and persistence in order to define the clinical significance of XMRV in humans. Although the prevalence of XMRV infection has been extensively investigated in various clinical settings, definition of target tissue(s) is far from being completed. In this study, the presence of XMRV DNA in peripheral blood samples of 124 patients with HIV infection was investigated with two molecular assays. All blood samples were XMRV DNA-negative, indicating that the virus is either absent from the peripheral circulating cells of HIV-infected individuals or is very rare. Although the presence of XMRV at levels below our detection threshold cannot be ruled out, this is very unlikely, as all samples were tested in duplicate or triplicate with two highly sensitive amplification methods (i.e. nested PCR and real-time PCR), both designed on the basis of highly conserved fragments of the viral genome. Moreover, none of the patients was treated with antiretroviral drugs such as azidothymidine, tenofovir, and raltegravir, which are known to inhibit XMRV replication and hinder detection of the virus [25,26]. In line with our findings, recent studies showed no evidence of XMRV in HIV-infected patients [27–30]. Thus, the major conclusion of our study is that either HIV infection (and inherent circumstances that exposed individuals to this and other viruses) does not increase the risk of acquiring XMRV infection, or that XMRV does not replicate or persist in peripheral blood cells of HIV patients and does not influence CD4+ T-lymphocyte counts in HIV-positive patients.

Given the long list of studies reporting no occurrence of the XMRV genome in peripheral blood cells (summarized in Table 1), it is unlikely that the virus may persist and replicate in this body compartment. However, the possibility cannot be excluded that XMRV infection causes sporadic local outbreaks and its presence could therefore be restricted to patients with an HIV-unrelated and still unidentified risk factor. Further investigations on different populations and larger number of samples, by using improved PCR assays targeted to different regions of XMRV genome and directed to detect viral and proviral forms of the virus, are needed to determine the real prevalence and distribution of this novel agent.

Transparency Declaration

We declare that we have no conflicts of interest related to this manuscript.