New WPI and CDC XMRV sequences in genbank

[Bob]

I have been updating this list as new XMRV sequences have been published, since 6th April 2011


A search in genbank shows up a number of new entries for XMRV.
http://www.ncbi.nlm.nih.gov/nuccore?term=xmrv


Hamon Center for Therapeutic Oncology Research, The University of Texas Southwestern Medical Center
Frequent detection of infectious xenotropic murine leukemia virus (XMLV) in human cultures established from mouse xenografts
Zhang,Y.-A., Maitra,A., Hsieh,J.-T., Rudin,C.M., Peacock,C., Karikari,C., Brekken,R.A., Stastny,V., Gao,B., Girard,L., Wistuba,I., Frenkel,E., Minna,J.D. and Gazdar,A.F.
29-JUN-2011

Xenotropic MuLV-related virus RKO, complete genome.
JF274252
http://www.ncbi.nlm.nih.gov/nuccore/JF274252.1


INSDC. Bulavaite A., Institute of Biotechnology, Vilnius University, V.A.Graiciuno 8, Vilnius, LT-02241, LITHUANIA
Xenotropic murine leukemia virus-related viruses in Lithuanian prostate cancer patient population
Sabaliauskaite,R., Bulavaite,A. and Lazutka,J.R.
10-June-2011

XMRV complete proviral genome, isolate S-162.
FR872816
http://www.ncbi.nlm.nih.gov/nuccore/FR872816.1


Whittemore Peterson Institute
Lombardi,V.C. and Khaiboullina,S.F.
21st May 2011

Xenotropic MuLV-related virus isolate WPI-CI-1303 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907633.1

Xenotropic MuLV-related virus isolate WPI-CI-1304 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907634.1

Xenotropic MuLV-related virus isolate WPI-CI-1310 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907641.1

Xenotropic MuLV-related virus isolate WPI-CI-1313 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907643.1

Xenotropic MuLV-related virus isolate WPI-CI-1314T gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907644.1

Xenotropic MuLV-related virus isolate WPI-CI-1315 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907645.1

Xenotropic MuLV-related virus isolate WPI-CI-1316 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907646.1


Whittemore Peterson Institute
Lombardi,V.C. and Khaiboullina,S.F.
13th May 2011

Xenotropic MuLV-related virus isolate WPI-CI-1301 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907631.1

Xenotropic MuLV-related virus isolate WPI-CI-1302 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907632.1

Xenotropic MuLV-related virus isolate WPI-CI-1305 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907635.1

Xenotropic MuLV-related virus isolate WPI-CI-1306 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907637.1

Xenotropic MuLV-related virus isolate WPI-CI-1307 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907638.1

Xenotropic MuLV-related virus isolate WPI-CI-1308 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907639.1

Xenotropic MuLV-related virus isolate WPI-CI-1309 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907640.1

Xenotropic MuLV-related virus isolate WPI-CI-1311 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907642.1

Xenotropic MuLV-related virus isolate WPI-CI-1317 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907647.1

Xenotropic MuLV-related virus isolate WPI-CI-1318 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907648.1

Xenotropic MuLV-related virus isolate WPI-CI-1327 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/JF907636.1



Division of HIV/AIDS Prevention, Ceners for Disease Control and Prevention.
No association of xenotropic murine leukemia virus-related viruses with prostate cancer
Switzer,W.M., Jia,H., Zheng,H., Tang,S. and Heneine,W.
10th May 2011

Xenotropic MuLV-related virus isolate 5935 polymerase (pol) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/HQ116790.1

Xenotropic MuLV-related virus isolate 5956 gag protein (gag) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/HM003612.1

Xenotropic MuLV-related virus isolate 6203 envelope (env) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/HM003611.1

Xenotropic MuLV-related virus isolate 5956 envelope (env) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/HM003610.1

Xenotropic MuLV-related virus isolate 6203 polymerase (pol) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/HM003609.1

Xenotropic MuLV-related virus isolate 5956 polymerase (pol) gene, partial cds
http://www.ncbi.nlm.nih.gov/nuccore/HM003608.1



Human Biology Division, Fred Hutchinson Cancer Research Center.
Mendoza,R. and Miller,A.D.
6th April 2011

XMRV-like mouse endogenous retrovirus mERV-XL, complete sequence
http://www.ncbi.nlm.nih.gov/nuccore/JF714652.1




Run a BLAST here:
http://blast.ncbi.nlm.nih.gov/Blast...astSearch&SHOW_DEFAULTS=on&LINK_LOC=blasthome

 

[Bob]

Switzer said that he had found 3 entirely new strains of XMRV in prostate cancer patients, so I imagine that these entries must be for those 3 strains, for the various gag, pol and env genes. In his paper, Switzer concludes that these sequences have enough genetic variability to suggest normal human infection, and he concludes that he has detected wild human XMRV virus in his prostate cancer samples.

No Association of Xenotropic Murine Leukemia Virus-Related Viruses with Prostate Cancer
William M. Switzer, Hongwei Jia, HaoQiang Zheng, Shaohua Tang, Walid Heneine
May 4, 2011
PLoS ONE
http://www.plosone.org/article/info:doi/10.1371/journal.pone.0019065#abstract0
(Results: XMRV in prostate cancer = 1.9%)




Have the WPI entered 11 entirely new strains?

 

[Jemal]

Nice find Bob, thanks!

 

[Sasha]

I think Cort writes about this in the XMRV Buzz of 16th May.

 

[Bob]

I think Cort writes about this in the XMRV Buzz of 16th May.

Thanks Sasha, here's the Buzz entry:
http://phoenixrising.me/?p=5568

 

[Mark]

Comparing the translation of Switzer's gag sequence with the WPI-CI-1327 and WPI-CI-1301 gag sequences, only 4 letters differ...

Switzer 5956
WPI-CI-1327
WPI-CI-1301

VTTPLSLTLQHWGDVQRIASNQSVDVRKRRWVTFCSAEWPTFNVGWPQDGTFNL...
_______LQHWGDVQRIASNQPVDVKKRRWVTFCSAEWPTFNVGWPQDGTFNL...
________QHWGSVQRIASNQSVDVKKRRWVTFCSAEWPTFNVGWPQDGTFNL...

...GIISQVKSRVFCPGPHGHPDQVPYIVTWEALAYDPPPWVKPFVSPKPPPLPTAPVLPPGPSAQPPSRSAL_______
...GVISQVKSRVFCPGPHGHPDQVPYIVTWEALAYDPPPWVKPFVSPKPPPLPTAPVLPPGPSAQPPSRSALYPALTLT
...GVISQVKSRVFCPGPHGHPDQVPYIVTWEALAYDPPPWVKPFVSPKPPPLPTAPVLPPGPSAQPPSRSALYPALTHS


Not sure quite what this implies, except that (if I've understood correctly):
- sequence variability is a key part of the necessary evidence for an actively replicating/infectious retrovirus;
- the supposed lack of sequence variability in detected XMRV is a key plank of the 'contamination' theory;
- the variability of XMRV sequences reported by the WPI appears to be confirmed here by Switzer;

I don't understand how this data can be consistent with the contamination theory, can anybody explain how such sequence variability could still be explained as laboratory contamination?

 

[Mark]

OK, I see now...from the XMRV Buzz article:

Hues genetic analysis indicated that the WPIs strains were not distinct from VP62/22RV1 and in an LA Times article Dr. Coffin disagreed with their conclusions stating the differences amount to a few base pairs in 8,000. Thats about the amount of difference I would expect to accumulate during one or two tissue culture passages, wrote Tufts University retrovirologist John Coffin.

The problem, of course, is that any gene sequence that appears to be associated with 22RV1 is now suspected (but not proved) of being a contaminant. If the WPI can prove that they have distinctive strains of XMRV that have been altered by their passage through the human body they can put a big dent in the contamination theory.

So: even sequence variability will have a counter-argument that such variability can be introduced during the culturing process.

 

[ukxmrv]

I don't understand how this data can be consistent with the contamination theory, can anybody explain how such sequence variability could still be explained as laboratory contamination?

If the contamination was from the lab straight to each patient using an unknown vector

 

[Jemal]

If the contamination was from the lab straight to each patient using an unknown vector

Which basically is pretty unbelievable / far-fetched?

 

[Bob]

OK, I see now...from the XMRV Buzz article:

Hues genetic analysis indicated that the WPIs strains were not distinct from VP62/22RV1 and in an LA Times article Dr. Coffin disagreed with their conclusions stating the differences amount to a few base pairs in 8,000. Thats about the amount of difference I would expect to accumulate during one or two tissue culture passages, wrote Tufts University retrovirologist John Coffin.

So: even sequence variability will have a counter-argument that such variability can be introduced during the culturing process.

A certain type of variability is expected to be seen in viruses that are created in the cell line (the cell line that manufactures XMRV - I can't remember it's name), which is different from variability expected to be found in wild virus. Also, XMRV detected due to contamination is said to be likely to have a very low variability.

Wild viruses will have a greater degree of variability than changes seen in the lab due to passage in cultures.

So, from the genetic variability, it can be worked out if a virus is a wild virus or a cell line contaminant, or any other lab contaminant.

I'm not exactly sure what viruses, of the WPI's, Coffin was referring to... I think that he was talking about the WPI's original strains, rather than the ones just published in genbank. But I'm a bit confused about it.

 

[Cort]

So: even sequence variability will have a counter-argument that such variability can be introduced during the culturing process.

It sounds like Coffin has an idea of how much variability culturing will induce. The human body presents so many more challenges to a virus that it should be much more distinct. (But what if it doesn't replicate much???? Then you have naturally low levels of distinctness?).

I think if it's distinct enough it will knock it out of the water. Now we get to hear about how distinct is distinct enough .

 

[Mark]

If the contamination was from the lab straight to each patient using an unknown vector

The patient samples were collected from across the US and sent to the WPI, so the lab can't really have infected ('contaminated') the patients, surely?

If the suggestion is that patients' blood is being 'contaminated' by some lab product, at some stage before the blood is collected, then that's "contamination, Jim, but not as we know it...".

But the contamination argument, so far, does seem to me to be just that: "Look: the XMRV retrovirus was created in the lab by mistake, it's infected cell lines and thus spread through labs all over the world, that sort of contamination is very common and happens all the time, there's so much contamination like that around that it's very hard to control for, we've never fully got to the bottom of all these discrepant results that keep cropping up, but the scientific consensus is that all this contamination with novel human retroviruses that are unintentionally created - while we accept that these retroviruses do infect humans sometimes - is really not that important and we shouldn't be wasting so much time and money investigating it."

As reassurances go, it's not doing much for me...

 

[Mark]

I'm not exactly sure what viruses, of the WPI's, Coffin was referring to... I think that he was talking about the WPI's original strains, rather than the ones just published in genbank. But I'm a bit confused about it.

I think that's right Bob, his comments there looked familiar to me and the publication of sequences in genbank is more recent.

 

[ukxmrv]

It's the "contaminated patients" theory and not the "contaminated blood".

The virus comes from somewhere (like a lab) and is introduced into patients by an unknown method. It's the Gkiks Margiorkinis theory from his Lancet letter

(snip from his letter)

Once a virus is endogenised, it is forced to follow the evolutionary rate of the host. Since XMRV is integrated in cell-lines the virus evolution is restricted to the host's pace of evolution, and viral descendants have none or minimum sequence diversity. Thus, if a contaminated product, previously cultured in cell-lines, is administered to people then the infections would provide the evolutionary patterns reported by Hue and colleagues.4 If the immunological data reported by Lombardi and colleagues5 are correct, then we need to trace the common source of XMRV and not through human contact.

(end of snip)

http://forums.phoenixrising.me/showthread.php?10813-Lancet-Article-by-Gkikas-Magiorkinis-April-2011

 

[Mark]

It sounds like Coffin has an idea of how much variability culturing will induce. The human body presents so many more challenges to a virus that it should be much more distinct. (But what if it doesn't replicate much???? Then you have naturally low levels of distinctness?).

I think if it's distinct enough it will knock it out of the water. Now we get to hear about how distinct is distinct enough .

Indeed...and on past evidence, I guess that this level of detail will be as murky and inconclusive as all the others before it, because even if there's not as much variability as one might normally expect, it could still be the case that XMRV replicates less, or conserves sequences more, than the typical points of comparison...its resistance to editing by APOBEC3 and its short sequence length at 8000bp tend to suggest that possibility to me...

I'm wondering what sort of degree of variability the new WPI sequences represent. The partial sequences I highlighted above seem to have 4 changes out of about 350 bp (if I read that right), which would be around 1%, and quite a bit more than "a few base pairs out of 8000" - but then the question will be whether there's as much variability in the rest of the sequences as there is in those partial sequences, or whether those 4 changes are the only ones in the full sequence...so I presume that statistical analysis is going to be just as confusing as everything else has been...

 

[Otis]

I'm not sure what to make of the new WPI sequences except to say that I was hoping for complete seeing complete (~8100 bps) sequences vs. these partials (325-374 bps) from the WPI. Hope some are coming...

Interestingly, I found this (nearly complete) sequence a while back: http://www.ncbi.nlm.nih.gov/nuccore/FR837946.1 but no sign of a publication yet from this Lithuanian group. It's 93% similar to VP62 whereas the full-length sequences from the original Science paper are 99%. A good argument for significant sequence variability but we need the paper....

 

[Mark]

It's the "contaminated patients" theory and not the "contaminated blood".

The virus comes from somewhere (like a lab) and is introduced into patients by an unknown method. It's the Gkiks Margiorkinis theory from his Lancet letter

(snip from his letter)

Once a virus is endogenised, it is forced to follow the evolutionary rate of the host. Since XMRV is integrated in cell-lines the virus evolution is restricted to the host's pace of evolution, and viral descendants have none or minimum sequence diversity. Thus, if a contaminated product, previously cultured in cell-lines, is administered to people then the infections would provide the evolutionary patterns reported by Hue and colleagues.4 If the immunological data reported by Lombardi and colleagues5 are correct, then we need to trace the common source of XMRV and not through human contact.

(end of snip)

http://forums.phoenixrising.me/showthread.php?10813-Lancet-Article-by-Gkikas-Magiorkinis-April-2011

Thanks much for this!

I've highlighted it because it seemed important to me at the time and I'd forgotten that point, which will become important soon I imagine.

So Margiorkinis' contention here is that sequence diversity is not necessarily expected when a virus is transmitted to humans in a consistent form, due to its origin in a contaminated product cultured in cell lines.

I imagine it will be important to confirm and emphasise that contention if the sequence diversity still proves to be relatively low, because such possible scenarios seem to be routinely missed out of the logic somehow...

 

[heapsreal]

I think the whole contamination thing falls over when the healthy controls arent testing positive, everybody should be testing positive if its contamination. Also i would like to see them marry up positive and negative patients with other immune tests like nk function as well.

 

[VillageLife]

So the CDCs xmrv is also similar, this rules contamination out then!

 

[Kati]

Well the cdc found it but they're adamant it's not in CFS patients...:headache: