Reader Comments

Post a new comment on this article

What about phages?

Posted by zurdo on 25 Jun 2013 at 02:11 GMT

I think this is a very interesting and clear paper. Most cancer researchers disregard the interaction of human cells with microorganisms and this work demonstrate a clear contribution of this interaction in cell transformation.
Concerning the λ phage sequences, as far as I understood, the authors may have missed one aspect. Bacteriophage particle numbers is even higher than bacteria cells in the human microbiome. Moreover, phages are known active vehicles of LGT between bacteria by means of the integration of fragments of the host DNA in the phage genome. Thus, the bacteria DNA sequences found in human somatic cells may be, at least in part, "transported" in phages and prophages.
We have recently shown that phage genomes that use protein-primed replication mechanism have increased nuclear internalization efficiency in eukaryotic cells, due to the presence of nuclear localization signals (NLS) in the 5'-end attached terminal proteins (TP) (see Redrejo-Rodríguez et al. PNAS, 2012). This would provide one possible mechanisms of LGT. Moreover, several λ protein sequences contain putive NLS (some of them DNA binding proteins), that may function similarly. In conclussion, the phages' contribution to interkingdom LGT should be taking into account.

No competing interests declared.

RE: What about phages?

jdunninghotopp replied to zurdo on 03 Jul 2013 at 07:08 GMT

I think the idea of phage mediating interkingdom is an excellent one. I'm not sure I have seen strong evidence for that as yet. The idea has definitely been put forth as a mechanism for the abundant interdomain LGT involving italicWolbachiaitalic endosymbionts and their arthropod hosts by several researchers. italicWolbachiaitalic are known to have phage, making it a possibility. But I don't think the italicWolbachiaitalic-host transfers described thus far are consistent with phage transfer, and we don't see an overabundance of phage genes transferred which would suggest that might be the case.

The lambda integration signal described in this manuscript was strong enough to report, but not as convincing as we needed it to be to make any bold claims. The other transfers we described do not show evidence of being phage mediated, although lack of evidence does not preclude that as the mechanism. But keep in mind that the mere presence of a prophage transfer would not imply that it is more likely. These transfers, even the phage ones, may merely be random transfers. Given that phage are so numerous (and we didn't quantify how numerous in this study although we did not detect prophages as a major component of the microbiome, which you would expect), it may be that the increased number phages/prophages leads to increased signal. If so then phage may be no more likely than any other sequence to be integrated, other than to be more numerous in the potential donor sequences.

I think of greater concern, and tempering our enthusiasm, is that lambda is commonly used in molecular biology and genomics laboratories. We went to great lengths to try to figure out if this was an artifact from a technique used and came up empty. Yet, I still have lingering doubts as to the significance. I think follow up experimental studies are necessary to examine this further, which we are formulating now.

No competing interests declared.