Accurate estimates of viral production in natural environments are critical for assessing the impacts of viral lysis on bacterial mortality and dissolved organic matter release. Here, viral production was estimated using a tangential flow diafiltration (TFD) dilution method, which reduced viral abundance to about 25% of ambient while maintaining near ambient levels of bacterial abundance. In subsequent incubations, the rate of virus-like particle increase was measured and used to calculate viral production. TFD viral production estimates were compared to those from simultaneous incubations using a vacuum diafiltration procedure. At 4 stations in the Chesapeake and Delaware Bays, viral production averaged 4.8 ± 1.7 × 10 10 and 5.9 ± 4.4 × 10 10 viruses l -1 d -1 as assessed by the TFD and vacuum methods, respectively. The TFD procedure improved upon the vacuum-based method by recovering significantly more of the bacterial community and requiring less sample processing time. Optimization tests of the TFD procedure found that a 0.22 µm pore size filter with a flushing rate of 40 ml min -1 , and a flushing volume 4-fold the initial sample volume gave the best combination of bacterial recovery, viral dilution, and processing time. Based on TFD viral production estimates, viral lysis was responsible for the loss of 14 to 93% of the bacterial standing stock and the release of 22 to 47 µg C l -1 d -1 in the Chesapeake and Delaware Bays. These results indicate that viral lysis is a significant factor for microbial mortality and dissolved organic matter cycling within these estuaries.
KEY WORDS: Viral production · Tangential flow diafiltration · Viral-mediated mortality · Dilution method
Resale or republication not permitted without written consent of the publisherAquat Microb Ecol 41: [221][222][223][224][225][226][227][228][229][230][231][232] 2005 amount of C, N, and P released into a system by viral cell lysis (Fuhrman 1992, Steward et al. 1996, Gobler et al. 1997, Weinbauer & Höfle 1998, Wilhelm et al. 2002. Approaches to estimation of VP have been numerous and include electron microscopic observation of the frequency of visibly infected cells (FVIC) (Proctor et al. 1993, Steward et al. 1996, Binder 1999, Guixa-Boixereu et al. 1999, Hwang & Cho 2002, Middelboe et al. 2002, Choi et al. 2003, Weinbauer et al. 2003a, contact rates of viruses and bacteria (Murray & Jackson 1992, Suttle & Chan 1994, incorporation of 3 H-thymidine or 32 P into viral DNA (Steward et al. 1992a, Fuhrman & Noble 1995, Kepner et al. 1998), viral decay rates (Heldal & Bratbak 1991, Bratbak et al. 1992, Guixa-Boixereu et al. 1999, Tuomi & Kuuppo 1999, fluorescently labeled viruses (FLVs) as tracers of viral decay and VP (Noble & Fuhrman 2000, Helton et al. 2005, this issue), and dilution methods (Wilhelm et al. 2002, Winter et al. 2004a, Helton et al. 2005. The assumptions, advantages, and disadvantages of each of these approaches to measuring VP are summarized in Table 1. For many of these methods, 1 or more of the assumptions or...