Abstract

Recent theory suggests that viruses influence bacterial community composition by killing the winners of resource competition. We tested aspects of this theory by growing natural marine bacteria communities in seawater microcosms that had either significantly reduced or increased virus abundances and monitoring changes in the bacterial communities. Bacterial commu- nity composition was assayed by 2 whole-community fingerprinting techniques, terminal restriction fragment length polymorphisms (TRFLP) and automated ribosomal intergenic spacer analysis (ARISA), at the beginning and end of experiments to examine the effect of changes in viral abun- dance on bacterial community composition. Our results suggest that changes in viral abundances have mixed effects on microbial community fingerprints. Modest, but statistically significant, changes in community fingerprints were seen when most viruses were removed by filtration and bacteria sub- sequently grown over 5 d compared to growth at normal virus density. There were no significant dif- ferences between community fingerprints from microcosms grown with a normal versus 3-fold den- sity of viruses over 2 d (possibly because of slow growth rates); however, significant changes occurred over time, regardless of virus abundance, suggesting that manipulation and containment alone had a strong influence on community fingerprints, which may have masked some virus effects. Also, mod- erate natural variation in composition between replicate microcosms made it difficult to distinguish statistically significant virus effects. Given that relatively few significant changes were apparent in community fingerprints between virus treatments at the end of our experiments, it is possible that current models of virus infection and the possible roles of viruses in controlling community composi- tion need re-evaluation. The persistence of high viral abundance and apparent high turnover, in con- cert with our results, suggests that viruses and their hosts may have a more stable coexistence than is now currently thought. However, it is possible that the modest effects of viral infection observed in this short-term study could be significantly amplified over longer time-scales of weeks or months, resulting in viruses having a more substantial influence on bacterial community composition.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call