Human cytomegalovirus is well recognized as a clinically relevant pathogen in those with immature or compromised immune systems, particularly congenitally infected fetus/newborns and immunosuppressed transplant recipients. Because of the pathogenic potential of HCMV in these and other populations, there have been four decades of calls to develop a vaccine that confers protective efficacy against HCMV infection and disease 
. Primary vaccine research has focused on those HCMV proteins encoding epitopes for either neutralizing antibodies (glycoprotein B-gB) or strong cellular immune responses (phosphoprotein 65 and immediate early-1), based on humoral and cellular immune correlates of protection against congenital infection and transplantation-associated HCMV disease, respectively 
. Two vaccine trials based on recombinant gB vaccination in seronegative pregnant women and seronegative renal transplant recipients demonstrated partial, but incomplete, protection. One possibility for boosting protective efficacy beyond that observed in the gB trials is to include additional HCMV antigens to broadly target vulnerable components of viral replication. Additionally, an important related issue is whether there is a need to direct the vaccine immune responses to mucosal sites to maximize the immediacy of antiviral responses that prevent dissemination of progeny virions beyond the site of challenge virus infection. Our current study in the rhesus macaque model of HCMV infection was initiated to begin to address these two issues of HCMV vaccine design based on the following considerations.
HCMV immune modulating proteins that disrupt host cell functions, including antigen presentation, signaling, trafficking, activation, and metabolism, have not been evaluated as vaccine candidates. A large percentage of the coding capacity of the HCMV genome is devoted to these proteins, and in vitro
studies strongly suggest that HCMV immune modulating proteins collectively could skew host immunity during critical junctures of HCMV infection in vivo
. Limited studies have observed cellular immune responses specific to presumptive HCMV immune modulating proteins, including those that disrupt MHC class I antigen presentation (US2, US3, US6, US11), and those that inhibit NK function (UL16, UL18) 
. Antibody responses have also been detected against UL111A (cmvIL-10) in some HCMV-infected individuals 
. One group profiling HCMV T cell responses noted that, based on the in vitro
phenotype of the immune modulating proteins and their potential role in vivo
, “Future investigations will need to assess the levels of protection afforded by responses directed toward these epitopes” 
Mucosal surfaces are the sites of both primary infection and reinfection. Analytic modeling of published congenital transmission rates has led to the conclusion that ~75% of congenital infections in the United States occur in women with preconceptional immunity to HCMV 
. The study did not determine whether maternal-to-fetal transmission was the result of maternal reactivation of preexisting virus or reinfection with a horizontally transmitted virus. Based on multiple studies demonstrating reinfection of seroimmune pregnant women, including in populations with universal seroprevalence, it is clear that prior immunity is only partially protective against reinfection 
. A salient point of these studies is that HCMV reinfection in these instances was the result of mucosal exposure to horizontally acquired virus, implying that mucosal antiviral immune responses were insufficient to contain viral reinfection locally.
With the long-term goal of testing in vivo
the concept of immune modulating proteins as vaccine candidates, immune responses to rhcmvIL-10 (RhUL111A) were evaluated in healthy RhCMV-infected rhesus macaques persistently infected with wild-type RhCMV. Studies have shown that rhcmvIL-10 plays a dynamic role in viral immune modulation, mimicking cellular IL-10 functions in vitro
and altering innate and adaptive immune responses to viral antigens in vivo
. Microbial manipulation of IL-10 appears to be a common feature of many persistent viruses (e.g.
, murine CMV, LCMV, hepatitis B and C viruses), bacteria (e.g.
, Chlamydia trachomatis, and Listeria monocytogenes
), and protozoa (e.g.
, Leishmania and Plasmodium
, and commensal bacteria 
. Thus, there is extensive precedent to focus on rhcmvIL-10 and cmvIL-10 as central players in primate CMV natural history. The potential viability of using rhcmvIL-10 in a vaccine was recently described for rhcmvIL-10 
. Structural biology was used to engineer biologically inactive mutants of rhcmvIL-10 that do not bind to the IL-10 high-affinity receptor and, therefore, lack wild-type functional activity. To provide a foundation for evaluating the immunogenicity of non-functional versions of rhcmvIL-10 in RhCMV-uninfected animals, peripheral and mucosal immune responses to wild-type rhcmvIL-10 were surveyed in RhCMV-infected juvenile and adult rhesus macaques.