IMEA's Letter to the editor of Science regarding serious flaws in the methodology of Paprotka et al.

by XMRV Global Advocacy on Tuesday, June 28, 2011:

Dear Dr Ash

We are a patient organisation and would like to make the following observations regarding the study entitled “Recombinant Origin of the Retrovirus XMRV”, in which John Coffin and his team claim to supply proof that XMRV arose during the passage of a xenograft through strains of mice that could potentially have been used in the formation of the 22Rv1 cell line and thus is a harmless contaminant. Should you choose to reply to this letter we undertake not to make the contents public without your express written permission.


This review first highlights the quote below.


The claim is that XMRV was created as a result of recombination of 2 proviral sequences dubbed PreXMRV-1 and PreXMRV-2.


“The complete sequence of PreXMRV-1 was determined from the early passage xenografts, the NU/NU and Hsd strains, and the CWR-R1 cell line.”


PreXMRV-1 is therefore a metaphor. There is no evidence that it exists as an independent entity. A synthetic clone was constructed from sequences amplified from NU/NU, Hsd and the CWR-R1 cell line. All three are known to contain one or more XMRV specific sequences. The sequences amplified in the early xenografts had a sequence homology to 3.5 kb of XMRV, in a region not unique to XMRV. The fact that the other sequences, used to construct the synthetic clone called PreXMRV-1, could actually have been XMRV specific sequences, must be obvious. It is difficult to understand how this escaped the attention of a competent peer reviewer.


The determination of XMRV concentration in 22Rv1, CWR-R1, NU/NU and Hsd nude mice.


“To quantify the amount of XMRV DNA in the CWR22 xenografts, we developed a real-time PCR primer-probe set that specifically detected XMRV env and excluded murine endogenous proviruses present in BALB/c and NIH3T3 genomic DNA (Fig. 1C). We used quantitative PCR of 22Rv1 DNA to estimate 20 proviruses/cell and used the 22Rv1 DNA to generate a standard curve. The CWR22 xenografts had significantly fewer copies of XMRV env (<1–3 copies/100 cells) compared to the 22Rv1 cells (2000 copies/100 cells). The CWR-R1 cell line had ~3000 copies/100 cells, and the NU/NU and Hsd nude mice, thought to have been used to passage the CWR22 xenograft, had 58 and 68 copies/100 cells, respectively.”


These primers were not used to search for XMRV in the other strains of wild and lab mice despite being able to detect such a low copy number of XMRV, nor were they used to examine the early xenografts for the presence of XMRV despite demonstrating such a high level of clinical sensitivity.


In order to examine the early xenografts the authors created 8 different primer sets. The CWR22 xenograft referred to above was not examined using the new primer sets below:


“We used the same XMRV-specific primer sets to amplify and sequence DNA from early passage xenografts (736, 777, 8L, 8R, 16R, and 18R; Fig. 2B); the results showed that XMRV env, but not gag sequences were present (sequencing coverage summarized in fig. S3), indicating that the early xenografts did not contain XMRV.”


This statement is an opinion and not objective fact. An examination of the primers used to examine the early xenografts is enlightening. The data can be referred to in Fig. S3 (A) of the supporting material. 8fsa-U5rsa can amplify env sequences when XMRV and only XMRV, is present. 18f-13r, 8fsa-Ursa and 8f-U3r, can amplify gag-pol sequences and env sequences when XMRV gag, pol and env sequences are known to be present at high concentrations in the late stage xenografts. There is no evidence however, that 8fsa-U5rsa and 8f-U3r produce amplicons in any situation other than when XMRV specific env sequences are present.


Hence, the presence of amplicons produced via the use of that 8fsa-U5rsa and 8f-U3r primers, means that the presence of XMRV in the early xenografts certainly cannot be ruled out. Indeed the amplicon produced using 18f-13r primers, which amplifies gag pol regions in XMRV, produces further evidence for the existence of XMRV in the early xenografts. This would not be expected to amplify the gag pol region of a proviral sequence PreXMRV-1, with only some 90% homology to the corresponding region in XMRV.


Taken together, these results strongly suggest that that the early xenografts did contain XMRV. This is especially so, when the fact that the existence of PreXMRV-1, as a real in vivo entity, has not been demonstrated. The conclusion, that the absence of gag sequences invalidates this finding, is clearly erroneous. It fails to take into the account that gag primers are some 10 fold less sensitive when dealing with such copy numbers of XMRV. Indeed, Danielson et al. demonstrated that their nested PCR, using input DNA of 600ng, could detect 1 XMRV provirus in 100,000 cells. Yet, these authors were quite unable to detect gag sequences in patients where env sequences were detected. Whereas Lombardi et al. only detected gag sequences in 7% of patients tested, using single round PCR, such as used in this study. They were however able to detect gag sequences in 67% of people with nested RT-PCR.


The concentration in the CWR22 xenograft was established as being less than 1 copy per 100 cells by the authors themselves, thus we would be dealing with concentrations even lower than that because we are looking at a time before the xenografts were fully formed. It is worthy of note that the primers and cycling conditions used to determine the concentration of XMRV in the CWR22 xenograft, were not at any time used to examine the early passage xenografts, nor indeed in the attempt to detect XMRV in other strains of wild and laboratory mice. The authors also make no attempt to determine the concentration of the proviral DNA sequence in the early xenografts.


Although MuLVs (henceforth called XMRV like viruses) can induce tumours by inserting into the promoter regions of tumour related genes, this is by no means the only mechanism by which XMRV like viruses can induce the formation of tumours, as the example below demonstrates. LP-BM5 is a mix of XMRV like viruses.


LP-BM5 murine leukemia retrovirus induces the excessive oxidative stress and immune dysfunction leading to B cell leukemia and murine AIDS with cytokine dysfunction. The cytokines upregulated in this investigation were Il-4 Il-6 and TNf alpha. The cytokines downregulated were INF gamma and IL-2(1).


Il-8 is a cytokine commonly elevated and deemed to be a causative factor in many cancers (2). Some information regarding the role of IL8 is given below:


"Our data show that IL-8 signaling increases AR expression and promotes ligand-independent activation of this receptor in two androgen-dependent cell lines, describing two mechanisms by which this chemokine may assist in promoting the transition of CaP to the androgen-independent state."(3)


Serum 1L-8 is elevated in men with prostate cancer (4).


"Aalinkeel et al. found that IL-8 was significantly higher in the more metastatic PC-3 and DU-145 prostate cancer cell lines, when compared to the poorly metastatic LnCAP cells. The results of our study of IL-8 in men with prostate cancer support the findings of Aalinkeel et al"


XMRV induces IL-8 expression in prostate cancer cell lines (5).


93% Xmrv infected people with ME have elevated il-8 levels (6).


Immune dysregulation and the production of oxidative stress in a host in response of a XMRV like virus infection is well documented. Thus, the claim that XMRV would have to infect many cells in prostate tissue to induce prostate cancer, is an opinion only. The cells shown to be directly infected may well have been transformed by insertional mutagenesis, but the two mechanisms are not mutually exclusive.



Summary


A PCR assay(1) with primers A and reagents B, using cycling conditions C, was able to detect XMRV specific env sequences when the copy number was as low as 1-3 provirus copies per 100 cells.


This PCR assay played no further part in the study.


Instead a new PCR assay(2) with different reagents and primer combinations of D E F G H I J K were used to assay cell lines of the CWR-R1 xenograft, the 22Rv1 cell line and the NU/NU and Hsd nude mice. Primers A B and C were able to detect XMRV sequences when only XMRV was present at a copy number of greater than 2000 proviral copies per 100 cells. Primer J was used to amplify an XMRV sequence from 100 ng of 22Rv1 cell DNA at an unknown level of proviral concentration. The use of primer J however, was not able to amplify any XMRV specific sequences in the investigation cited above. Primer J was the primer chosen to search the DNA of multiple mice species for the presence of XMRV.


Hence, it is impossible to say that XMRV was not present in the early xenografts or indeed the original prostate cancer tissue. It is also not possible to determine whether gag and env sequences were present in the multiple mouse species examined.


The explanation that XMRV entered the human population as a result of recombination of two proviruses would at least require that both proviruses did actually exist as in vivo entities. The existence, of the proviral sequence dubbed as PreXMRV-1 as such an entity, has not been established. The sequences are a construct from sources known to contain XMRV.


We are thus left with at least two competing explanations.


On the one hand we have the explanation that a PCR assay, of unknown sensitivity below the level of 2000 XMRV proviral copies per 100 cells, was simply unable to locate very low copy numbers of XMRV. This is supported by the fact that the use of primers J were unable to amplify XMRV sequences when proviral copies were as high as 3000 proviral copies per cell. Alternatively, we have the authors preferred explanation, that XMRV was formed by a recombination event which is so rare that it could only have happened once and the odds of it happening at any other time (by the authors own admission) are over a billion to one against. I leave the reader to judge what is the most parsimonious, conciliate hypothesis.



References

1) Lee, J.M.; Dehydroepiandrosterone Sulfate Inhibited Immune Dysfunction Induced by LP-BM5 Leukemia Retrovirus Infection through Regulating Th1/Th2 Type Cytokine mRNA Expression and Oxidative Stress in Murine AIDS Model; Journal of The Korean Society of Food Science and Nutrition (Dec 2006)

2) Yuan A, Chen JJ, Yao PL, Yang PC. The role of interleukin-8 in cancer cells and microenvironment interaction. Front Biosci. 2005 Jan 1;10:853-65. Print 2005 Jan 1.

3) Angela Seaton, Paula Scullin, Pamela J. Maxwell, Catherine Wilson, Johanna Pettigrew, Rebecca Gallagher, Joe M. O'Sullivan, Patrick G. Johnston and David J. J. Waugh: Interleukin-8 signaling promotes androgen-independent proliferation of prostate cancer cells via induction of androgen receptor expression and activation; Carcinogenesis (2008) 29 (6): 1148-1156.

4) Lehrer S, Diamond EJ, Mamkine B, Stone NN, Stock RG.; Serum interleukin-8 is elevated in men with prostate cancer and bone metastases.; Technol Cancer Res Treat. 2004 Oct;3(5):411.

5) Robert H. Silverman, Carvell Nguyen, Christopher J. Weight & Eric A. Klein; The human retrovirus XMRV in prostate cancer and chronic fatigue syndrome; Nature Reviews Urology 7, 392-402 (July 2010)

6) V.C. Lombardi, K. S. Hagen, K. W. Hunter, J. W. Diamond, J. Smith-Gagen, W. Yang And J. A. Mikovits; Xenotropic Murine Leukemia Virus-related Virus-associated Chronic Fatigue Syndrome Reveals a Distinct Inflammatory Signature; in vivo 25: 307-314 (2011)



These following extracts from the paper clearly show that the argument rests on highly subjective interpretations. One could easily reverse the conclusions of the authors and the whole study would fall.


"(B) PCR and sequencing of PreXMRV‐1. The complete PreXMRV‐1 genome was cloned and sequenced from the indicated sources using primers that specifically amplify XMRV or PreXMRV‐1 but exclude known endogenous MLV sequences (Fig. S2). We amplified PreXMRV‐1 from the CWR‐R1 cell line, but not the 22Rv1 cell line, indicating the absence of PreXMRV‐1 from these cells. Partial PreXMRV‐1 (env divergent region) was also amplified from xenografts 2524 and 2274, showing that both XMRV and PreXMRV‐1 are present in these samples."


"We used the same XMRV-specific primer sets to amplify and sequence DNA from early passage xenografts (736, 777, 8L, 8R, 16R, and 18R; Fig. 2B); the results showed that XMRV env, but not gag sequences were present (sequencing coverage summarized in fig. S3), indicating that the early xenografts did not contain XMRV."


Due to the serious implications that this paper will have on further research into this retrovirus, it is therefore necessary to amend these flaws, so as to allow your readership the opportunity to judge the data in their usual manner.

Yours sincerely

Louise Gunn

CEO
IMEA

27 June 2011