Absence of XMRV in Peripheral Blood Mononuclear Cells of ARV-Treatment Nave HIV-1..

[Firestormm]

Absence of XMRV in Peripheral Blood Mononuclear Cells of ARV-Treatment Nave HIV-1 Infected and HIV-1/HCV Coinfected Individuals and Blood Donors

Received: August 22, 2011; Accepted: January 6, 2012; Published: February 13, 2012

Full paper: http://www.plosone.org/article/info:doi/10.1371/journal.pone.0031398

Abstract

Background

Xenotropic murine leukemia virus-related virus (XMRV) has been found in the prostatic tissue of prostate cancer patients and in the blood of chronic fatigue syndrome patients. However, numerous studies have found little to no trace of XMRV in different human cohorts. Based on evidence suggesting common transmission routes between XMRV and HIV-1, HIV-1 infected individuals may represent a high-risk group for XMRV infection and spread.

Methodology/Principal Findings

DNA was isolated from the peripheral blood mononuclear cells (PBMCs) of 179 HIV-1 infected treatment nave patients, 86 of which were coinfected with HCV, and 54 healthy blood donors. DNA was screened for XMRV provirus with two sensitive, published PCR assays targeting XMRV gag and env and one sensitive, published nested PCR assay targeting env.

Detection of XMRV was confirmed by DNA sequencing. One of the 179 HIV-1 infected patients tested positive for gag by non-nested PCR whereas the two other assays did not detect XMRV in any specimen. All healthy blood donors were negative for XMRV proviral sequences. Sera from 23 HIV-1 infected patients (15 HCV+) and 12 healthy donors were screened for the presence of XMRV-reactive antibodies by Western blot.

Thirteen sera (57%) from HIV-1+ patients and 6 sera (50%) from healthy donors showed reactivity to XMRV-infected cell lysate.

Conclusions/Significance

The virtual absence of XMRV in PBMCs suggests that XMRV is not associated with HIV-1 infected or HIV-1/HCV coinfected patients, or blood donors. Although we noted isolated incidents of serum reactivity to XMRV, we are unable to verify the antibodies as XMRV specific.

 

[currer]

From the paper,

XMRV-reactive antibodies in sera
To further search for evidence of XMRV in the HIV-1 infected and HIV-1/HCV coinfected patients, we screened for the presence of XMRV-reactive antibodies in 23 of the 179 HIV-1+ and HIV-1+/HCV+ subjects and in 12 additional healthy blood donors
Five of eight sera from HIV-1+ patients were reactive to XMRV whereas 8 of 15 sera from HIV-1+/HCV+ patients were reactive (Table 2). Of the 13 XMRV-reactive sera, 12 contained capsid-reactive antibodies, and one contained Env-reactive antibodies (Table 2). Reactivity to XMRV capsid was observed for patient 103219 (Figure 4A). Similar to the HIV-1+ and HIV-1+/HCV+ patient sera, reactivity to XMRV was seen for 6 of the 12 healthy blood donors (Figure 4B and Figure S2). For the healthy blood donors, reactivity was only observed for XMRV capsid (Table 2). The greater ability to detect capsid-reactive antibodies compared to Env-reactive antibodies has been reported previously for plasma from healthy donors and prostate cancer and chronic fatigue syndrome (CFS) patients [5]. Although these data are suggestive of infection, without serum from a confirmed XMRV-infected individual, it is unclear whether reactivity from these 13 patient sera represents a true adaptive immune response against XMRV or is simply due to the presence of cross-reactive antibodies.

Notably, all but one [49] study searching for XMRV in HIV-1 infected cohorts to date have screened for the virus in the blood or in constituents of the blood. While the agreement in results among reports regarding the prevalence of XMRV in HIV-1 infected cohorts may indicate that XMRV is largely absent from this population as a whole, it is also possible that XMRV resides primarily in a cellular compartment other than blood. On this note, it is important to point out that all but a few [2], [5], [25] reports on screens for XMRV in the blood or in blood constituents were unable to detect the virus [3], [4], [12], [13], [16], [45][48], [50][52]. Comparatively, more studies detect XMRV, at least at a low prevalence [1], [8], [11], [14], [15], [36], [41], [53], [54], than those that do not [6], [9], [10], [49], [51], when non-blood tissue specimens are screened. As most reports on screens for XMRV in non-blood-derived specimens pertain to prostate cancer cohorts, it is unclear whether disease status or the type of tissue screened is the main determinant for detection of the virus. A clue may be provided in a recent report on the kinetics and dissemination of XMRV in Indian rhesus macaques after intravenous inoculation [26]. In that study, XMRV provirus became undetectable in macaque PBMCs after only one month post-inoculation, whereas provirus could be detected from other macaque tissues throughout the 291 day duration of the study [26]. If XMRV provirus is cleared from the blood one month after infection of humans, then the blood (or its constituents) is not a reliable tissue compartment to screen when attempting to establish the prevalence of the virus.

It is possible that the positive signals obtained in our immunoblots are due to the presence of cross-reactive antibodies to proteins encoded by human endogenous retroviruses (HERVs), a large group of which is similar to MLVs [57], [58]. Human IgG reactivity to MLV capsid has been reported previously [59], [60]. In one study, a higher frequency of individuals with MLV capsid-reactive IgG was seen with HIV-1 infection compared to HIV-1 negative controls, a trend we did not observe with this cohort [59]. Nonetheless, proteins encoded by HERVs represent a potential source of antigen that may give rise to antibodies that are cross-reactive with XMRV. Alternatively, the XMRV-reactive antibodies detected in the sera of the HIV-1+ and HIV-1+/HCV+ and healthy subjects may have been elicited by an infection with XMRV or another related exogenous virus that had been cleared from the PBMCs prior to the time of blood collection, suggesting a latent infection in a tissue compartment other than blood as previously found in experimental infection of rhesus macaques [26]. The lack of an antibody that has proven specificity for XMRV has led to inconclusive results when using antibody-based screening methods. For example, it was recently discovered that human T-cell leukemia virus (HTLV) infection can elicit antibodies that are cross-reactive with XMRV p15E due to a homologous region on HTLV gp21 [61]

 

[currer]

We know that the sufferers from ME are generally female. I do not know of any association between HIV infection and ME, and no association between groups at a high risk of sexually transmitted disease or blood borne infection (such as drug users) and ME, - in fact all the statistics from patient support groups do not show any such association. (This is the reason for the inclusion of Hepatitis C sufferers in this research)

Now there may be an association between ME and these groups - but I have never seen any data to suggest there is. Does anyone know of any?

This paper assumes that HIV infection could be an additional risk factor for XMRV infection - but we do not know this.

The antibody results are interesting, though., but 6/12 healthy blood donors showing reactivity does not match previous figures of 4% in the healthy population. It is way too high.


So what are we to make of such high levels of non-specific antibodies....? That they were caused by Hep C and HIV?...yet half the healthy blood donors had them.....

I have just noticed that the PBMCs were frozen at -80 Doesnt that lead to a problem in retrieving the retroviral sequences?

 

[Firestormm]

Hi Currer,

I haven't read all of this properly (too occupied elsewhere I am afraid). But the use of HIV is to see if XMRV would be more likely to be present in a known immune-compromised patient cohort, I believe. I think there has been a previous paper that used HIV infected patients similarly.

Though I could of course be wrong. But I think that's why they looked at such cohorts and not because of any assumed associated between ME and HIV.

Hang-on I've seen something about this recently....

Ah yes:

'Maybe it is inappropriate to look for XMRV in healthy blood donors. Maybe you only 'see' XMRV when a patients immune system is compromised, like in HIV/AIDS:

Prevalence of XMRV Nucleic Acid and Antibody in HIV-1-Infected Men and in Men at Risk for HIV-1 Infection: http://www.ncbi.nlm.nih.gov/pubmed/22282703

Xenotropic murine leukaemia virus-related virus is not found in peripheral blood cells from treatment-naive human immunodeficiency virus-positive patients: http://www.ncbi.nlm.nih.gov/pubmed/21672082

Yet again, XMRV is not found in immunocompromised individuals, like HIV/AIDS patients, even in the absence of anti-retrovirals.'

http://scienceblogs.com/erv/2012/02/xmrv_update_the_science.php

 

[currer]

The problem with this argumant as far as ME is concerned is that there is no evidence to suggest that ME is associated with HIV status.

We know that ME is primarily seen in women and nowadays also in children. By itself, this observation does not suggest that "XMRV" (or whatever the agent we are looking for is,) is readily sexually transmitted. What about epidemics of ME type disease in schools and hospitals? The mind boggles.....

 

[Jemal]

The problem with this argumant as far as ME is concerned is that there is no evidence to suggest that ME is associated with HIV status.

We know that ME is primarily seen in women and nowadays also in children. By itself, this observation does not suggest that "XMRV" (or whatever the agent we are looking for is,) is readily sexually transmitted. What about epidemics of ME type disease in schools and hospitals? The mind boggles.....

Off-topic: are we sure ME is primarily seen in women? I am meeting an awful lot of men on these forums... I am male myself.

Oh, and thanks for posting Firestormm!

 

[Jemal]

And on-topic: yes, I do think they are looking into HIV patients, because their immune systems are compromised and they are picking up all kinds of viruses. It does make some sense from that point of view.

I wanted to bring something else to everyone's attention. This project is still running I think:

Abstract: The global AIDS pandemic continues to expand despite significant advances in understanding HIV-1 pathogenesis and the development of powerful antiviral drugs. HIV-1 is transmitted primarily through sexual contact and more than 30 million people are currently infected worldwide. In some regions of the world such as southern Africa the prevalence of HIV-1 infection exceeds 20%. The devastating spread of HIV-1 in young women in these countries appears out of proportion to the overall risk of infection. Thus it is possible that a biological co-factor contributes to virus spread. The hypothesis of this proposal is that acquisition by HIV- 1 of the envelope glycoprotein of gammaretroviruses (murine leukemia virus-related viruses) in a process we call ""natural pseudotyping"" expands the cellular tropism of HIV-1 enabling it to directly infect vaginal epithelial cells thereby dramatically increasing the risk of infection during sexual intercourse. We propose a molecular epidemiology study consisting of four major aims. 1) To complete in vitro studies of gammaretrovirus/HIV-1 pseudotyping; 2) To demonstrate gammaretrovirus/HIV-1 co-infections in local donors; 3) To demonstrate the effect of gammaretrovirus pseudotyping on vaginal transmission in an animal model; 4) To determine the prevalence of gammaretrovirus infection and HIV-1 co-infection in southern Africa. Natural pseudotyping of HIV-1 is predicted to occur in individuals co-infected with HIV-1 and a gammaretrovirus (xmrv/MLV) since the cellular tropism of these two viruses overlap to include T cells. The formation of xmrv-pseudoptyped HIV-1 (HIV-1/gp70) (now with the cell tropism of xmrv) in blood or lymphoid tissue would result in HIV-1 infection of normally resistant cells in the urogenital tract. Prior or subsequent infection of these cells with xmrv would result in the release of HIV-1/gp70 into seminal fluid or vaginal secretions. The potential implications of natural pseudotyping of HIV-1 are profound.

http://projectreporter.nih.gov/project_info_details.cfm?aid=8144150&icde=11419429

This might be another reason why some researchers are looking into the XMRV status of HIV patients.

 

[Firestormm]

Interesting Jemal. Very interesting. Oh and I'm a bloke too don't you know But yes it has been established that the majority of those diagnosed are of the opposite gender. Reasons for this are - of course - unclear at this moment in time. Indeed I am not sure anyone has really bothered looking or has hypothesised why this might be so. It just is. Go figure.

 

[currer]

Yes, the implications would be profound! I had heard of the possibility of pseudotyping with HIV before, but did not take much notice of this idea - why? - because it sounds too catastrophic to take seriously. So it really is a possibility?

I too, had wondered why there were so many studies done on HIV patients.

I had heard that the epidemic of HIV in young african women was assumed to be because of pre-existing untreated STDs, breaching the integrity of the epithelial barrier - herpes infection for example could do this.

has anyone any idea why the antibody studies in this paper came up with so many positives?
Even in healthy participants? Other papers looking at antibodies find a much lower percentage.
http://www.retrovirology.com/content/7/1/57

I could understand why if it were just the HIV/HCV infected,....but so many healthy positives?

 

[natasa778]

This paper assumes that HIV infection could be an additional risk factor for XMRV infection

Why not the other way around?

 

[Firestormm]

From the introduction and just to further clarify:

''The HIV-1 infected host may provide an immunological environment propitious for XMRV replication and spread. Apart from the overall deterioration of the immune system resulting largely from the depletion of CD4+ T cells (reviewed in [28]), HIV-1 encodes accessory proteins that antagonize innate antiviral host proteins shown to restrict XMRV replication, such as several members of the APOBEC3s and tetherin/BST-2 [29][35].

Thus, HIV-1 infected persons may potentially accommodate for XMRV replication due to suppressed immunological defenses on both systemic and cellular levels.

Based on evidence suggesting common transmission routes between HIV-1 and XMRV, and the ability of HIV-1 to neutralize immune components shown to restrict XMRV replication, we hypothesized that the prevalence of XMRV among HIV-1 infected patients may be elevated compared to healthy blood donors.''

Hi Currer,

I haven't read all of this properly (too occupied elsewhere I am afraid). But the use of HIV is to see if XMRV would be more likely to be present in a known immune-compromised patient cohort, I believe. I think there has been a previous paper that used HIV infected patients similarly.

Though I could of course be wrong. But I think that's why they looked at such cohorts and not because of any assumed associated between ME and HIV.

Hang-on I've seen something about this recently....

Ah yes:

'Maybe it is inappropriate to look for XMRV in healthy blood donors. Maybe you only 'see' XMRV when a patients immune system is compromised, like in HIV/AIDS:

Prevalence of XMRV Nucleic Acid and Antibody in HIV-1-Infected Men and in Men at Risk for HIV-1 Infection: http://www.ncbi.nlm.nih.gov/pubmed/22282703

Xenotropic murine leukaemia virus-related virus is not found in peripheral blood cells from treatment-naive human immunodeficiency virus-positive patients: http://www.ncbi.nlm.nih.gov/pubmed/21672082

Yet again, XMRV is not found in immunocompromised individuals, like HIV/AIDS patients, even in the absence of anti-retrovirals.'

http://scienceblogs.com/erv/2012/02/xmrv_update_the_science.php

Though, of course, and:

'In conclusion, the results of our screen of HIV-1 infected, HIV-1/HCV coinfected, and uninfected subjects do not support an association between XMRV and HIV-1 or HCV infections. Our report adds to accumulating evidence from other studies conducted around the world, not only against an association between these viral infections, but also against the presence of XMRV in the blood.'

 

[currer]

For the non-nested PCR assays, however, it was reported that sensitivity for detecting XMRV in PBMCs could be increased if the PBMCs are stimulated with PHA and cultured in IL-2-containing media prior to PCR [42]. Despite stimulating and culturing the PBMCs of 10 HIV-1+ patients (5 HCV+) we found no evidence of XMRV infection upon PCR screening with any of the three assays.

It looks as if they only cultured and stimulated 15 PBMC samples.
That could be your reason. They relied only on PCR.

 

[currer]

"The PCR used in this study is not that from Lombardi et al. They are only single round PCR assays that have been modified - not the same. The assay from Danielson is also modified as it was used on blood not tissue as per the original design. The copy number in blood compared to tissue is much lower.

So why did the western blot produce almost equal results? because the antibody is not specific to MRVs and will cross react with loads of things. "

From v99.

 

[Bob]

Thanks for the interesting discussion everyone.

It's not a very helpful study, is it.


Here's some interesting snippets from the paper...
This is not news for us, but it's interesting to see what we've been saying (esp about the Lipkin study), published in another paper:

"...it is also possible that XMRV resides primarily in a cellular compartment other than blood. On this note... ...
Comparatively, more studies detect XMRV, at least at a low prevalence [1], [8], [11], [14], [15], [36], [41], [53], [54], than those that do not [6], [9], [10], [49], [51], when non-blood tissue specimens are screened.

...the blood (or its constituents) is not a reliable tissue compartment to screen when attempting to establish the prevalence of the virus."


And another interesting snippet...
I'm going to have a look at the references here, because i'm interested in exactly how similar HERVs are to MLVs...
Even if everyone is detecting HERVs, and not XMRV, then they could still be used as a biomarker (except not in this study, seeing as controls and patients samples had similar results), and HERVs could potentially be causal...

"It is possible that the positive signals obtained in our immunoblots are due to the presence of cross-reactive antibodies to proteins encoded by human endogenous retroviruses (HERVs), a large group of which is similar to MLVs [57], [58]."


Did we know about the antibody cross reactivity between HTLV and XMRV? I think I might vaguely remember having come across this before...

"For example, it was recently discovered that human T-cell leukemia virus (HTLV) infection can elicit antibodies that are cross-reactive with XMRV p15E due to a homologous region on HTLV gp21 [61]"