Graduation Year


Document Type




Degree Granting Department

Marine Science

Major Professor

Mya Breitbart, Ph.D.

Committee Member

John Paul, Ph.D.

Committee Member

Gary Mitchum, Ph.D.

Committee Member

Kathleen Scott, Ph.D.

Committee Member

Craig Carlson, Ph.D.


Marine viruses are the most numerous biological entities in the ocean, with an estimated abundance of 4 x 1030. They merit study not only because of their sheer abundance, but also because of the role they play in the Earth's biogeochemical cycles. Viral lysis of bacteria redirects the flow of nutrients among marine microbes, which ultimately affects the efficiency of the biological pump. Viral diversity is important because most viruses are host-specific. In preying on a certain type of bacteria, viruses affect the diversity and structure of the bacterial community, leading to changes in carbon and nutrient flows. In turn, such variations can alter the amount of carbon dioxide in the Earth's atmosphere. However, studying viral diversity presents challenges. Morphological similarities among many types of viruses make it preferable to use genetic methods of investigation, but the absence of a single gene common to all families of viruses hampers the identification of viruses in environmental samples. Nonetheless, some genes are shared within phage families, and those shared ("signature") genes can be used as markers to identify members of a family. In addition, community profiling methods can fingerprint the diversity of a viral community.

Most previous studies of marine viral communities consist of a single glimpse—a representation of the community at a single time and place, or at a few depths sampled at one time. While the resources required to collect marine samples often make broader or repeated sampling impracticable, without studies conducted over greater time and spatial ranges, our knowledge of marine viral dynamics will remain limited. To gain strides in understanding spatial and temporal variability in marine viral diversity, this dissertation focused on a detailed examination of viral diversity at a single site in the Sargasso Sea. Time and depth intervals for sampling were kept as uniform as possible in order to strengthen the conclusions to be drawn from the research.

The Sargasso Sea is a seasonally oligotrophic portion of the North Atlantic Ocean, characterized by deep convective winter mixing and summer stratification of the water column. A tremendous amount of oceanographic research has been conducted in the Sargasso Sea because it is home to the Bermuda Atlantic Time-series Study (BATS), one of the world's longest-running ocean time series studies. Because of the core monthly measurements made at the BATS site and the vast amount of ancillary research that uses BATS as a platform, the site is an excellent place to study viral diversity. Using a variety of techniques, this research aimed to expand our knowledge of viral dynamics by analyzing the viral community of the Sargasso Sea over a several-year period, through different seasons, and at different depths.

The first chapter developed phoH as a new signature gene for assessing marine viral diversity. The phoH gene is disproportionately present in fully-sequenced marine phage, as opposed to phage isolated from non-marine environments, and is widespread in the marine environment. Diversity of the phoH gene was high, and most of the sequences recovered belonged to phylogenetic groups that did not contain any cultured representatives, indicating that cultured phage isolates do not adequately represent the diversity found in marine environments. Composition of the phoH communities at each sampled location and depth was distinguishable according to phylogenetic clustering, although most phoH clusters were recovered from multiple sites. These factors demonstrate that phoH will be useful for studying marine phage diversity worldwide.

Chapter 2 analyzed the viral diversity of a depth profile at BATS by amplifying and deep sequencing the phoH gene. This comprehensive study of the gene's diversity over three different years, several seasons, and a range of depths from the surface to 1000 m revealed that the viruses at BATS contain a large pool of phoH sequences, but that most of those sequences are rare. The phoH sequences were dominated by just a few operational taxonomic units (OTUs). Rarefaction analysis showed that the sequencing was sufficient to capture the diversity of the gene at BATS, and in fact no new phylogenetic clusters were identified that were not seen in the small amount of Sanger sequencing performed for the initial phoH study in Chapter 1. Some of the more abundant phoH OTUs recurred every season and every year, in varying degrees, although similar depths and seasons clustered together. Overall, the phoH gene revealed depth-based, seasonal, and interannual differences in the diversity of the viral community at BATS.

Chapter 3 continued the extensive examination of viral diversity at BATS by using several signature genes and a fingerprinting technique to assess changes between winter and summer viral communities over two depths in three different years. This chapter investigated whether the annually recurring subsurface peak in viral abundance corresponded to recurring changes in composition of the viral community in the vicinity of the peak. Clustering analysis was used to determine which samples were most similar. The results demonstrated that the viral communities at the surface and at 100 m depth were more similar to each other in winter (March), regardless of the year, than they were in summer (September), when the water column is stratified as opposed to well-mixed. These findings may stem from physical factors such as UV irradiation of viral particles during stratification, as well as seasonal and depth-related differences in host communities associated with the depth of the mixed layer.

This dissertation provides substantial advances to the field of microbial ecology. First, the development of phoH as a signature gene is an important addition to the limited set of tools available for studying marine viral diversity. This research also constitutes the first deep sequencing of a signature gene for marine viruses, providing a guide for the depth of sequencing needed to capture the diversity of a marine viral community and a benchmark for the level of viral diversity to expect in an oligotrophic marine system. Finally, the dissertation expands our knowledge of the viral community at BATS by examining the community based on four different measures of composition, rather than abundance. The research presented here also suggests several avenues of future investigation, including redesigning the phoH primers to expand their scope, sampling the viral community at BATS at the precise depth of the peak in abundance, working to identify the hosts of aquatic gokushoviruses, and culturing and sequencing additional marine viruses in order to improve the reflection of natural environmental communities in genomic databases.