Access Type

Open Access Dissertation

Date of Award

January 2020

Degree Type


Degree Name



Immunology and Microbiology

First Advisor

Philip E. Pellett


Human cytomegalovirus (HCMV) is an enveloped, single segment, double-stranded DNA virus. HCMV infection causes disease in immunocompromised (HIV patients, transplant recipients) and immunodeficient (fetuses, neonates) populations. Current treatments are effective but are either limited in use or can lead to organ damage and/or antiviral resistance, and no vaccines are available. Additional antiviral targets are needed. HCMV pUL103 is a potential antiviral target. pUL103 is a conserved herpesvirus protein present in the tegument, layer of proteins and RNA between the envelope and capsid of HCMV virions. pUL103 helps reorganize cellular secretory machinery (Golgi, endosomes) to form the cytoplasmic virion assembly compartment (cVAC) that is hypothesized to facilitate efficient virion assembly and egress. pUL103 is also important for cell-to-cell spread and virion maturation. The structure, function, and mechanisms of action of pUL103 are unknown.

To identify amino acids important for pUL103 functions, I aligned the sequences of 14 pUL103 homologs using MUSCLE (Multiple Sequence Comparison by Log-Expectation). My data showed 12 conserved amino acids outside of the conserved herpesvirus domain. Within the N terminus there are 13 conserved residues, which shows the importance of this region and these residues in pUL103 function.

To guide biological studies of pUL103, I used I-TASSER (Iterative Threading ASSEmbly Refinement) to predict structures of pUL103 and its homologs HSV-1 pUL7 (pUL7) and EBV BBRF2 (BBRF2) using their amino acid sequences. My data showed (a) their predicted secondary structures are similar, (b) the predicted structure of BBRF2 differs from pUL103 and pUL7, which are similar, and (c) the predicted pUL103 structure aligned with the non-catalytic LIM domain of human terminal uridylyl transferase 4 (TUT4), a nucleotidyl transferase domain lacking catalytic residues. This data suggests pUL103 may bind nucleic acids; experiments can be designed to test this hypothesis.

To identify when pUL103 mechanistic studies should be performed, I determined when pUL103 is necessary for maximal cVAC abundance. During 120-hour infections, I showed pUL103 must be present between 72 and 84 hpi. Such studies might include analysis of virion assembly, intra- and extracellular viral titers, and deep temporal and spatial proteomics analysis of viral and cellular proteins during this process.