Natural history of postnatal rhesus cytomegalovirus shedding by dams and acquisition by infant rhesus monkeys

Abstract

Background: Human infants frequently acquire human cytomegalovirus (HCMV) through breastfeeding, resulting in persistent high-level viral shedding in saliva and urine and infectivity to others, including pregnant women. Thus, vaccination to interrupt postnatal HCMV transmission is an attractive strategy to prevent HCMV spread and congenital infection. Rhesus CMV (RhCMV) in nonhuman primates is a valuable model for the study of immune strategies to prevent CMV transmission. Although rhesus monkeys typically acquire RhCMV before 1 year of age, the timing and mode of natural infant RhCMV transmission remain unknown.

Methods: We followed 5 RhCMV-seropositive dams and their infants from birth until weaning, approximately 6 months later. RhCMV DNA levels in plasma, breast milk, saliva, and urine were measured every 2 weeks by quantitative PCR. RhCMV-specific T cell responses in peripheral blood and breast milk were measured by interferon gamma ELISpot assays. Serum IgG antibody levels were measured by ELISA.

Results: Four of five postpartum RhCMV-seropositive mothers had intermittent, low-level RhCMV shedding in breast milk, whereas all had high-magnitude RhCMV shedding in saliva and urine. The kinetics of maternal blood RhCMV-specific T cell responses and viral shedding in urine and saliva did not strongly associate, though dams with consistently high systemic RhCMV-specific T cell responses tended to have undetectable RhCMV shedding in breast milk. All RhCMV-exposed infants had intermittent, low-level RhCMV shedding in saliva during the lactation period, with minimal systemic RhCMV-specific T cell responses.

Conclusions: Despite exposure to RhCMV shedding in breast milk and other maternal fluids, postnatal mother-to-child RhCMV transmission appears to be less efficient than that of HCMV. A greater understanding of the determinants of RhCMV transmission and its usefulness as a model of HCMV mucosal acquisition may provide insight into strategies to prevent HCMV infections in humans.

Fig 1. The frequency and magnitude of postpartum RhCMV shedding in mothers and infants.

Maternal breast milk, saliva, and urine and infant saliva were routinely assessed for RhCMV shedding by qPCR during the lactation period, until approximately 26 weeks postpartum, and for follow-up time points. The median and range of the positive qPCR replicates for each time point and sample type is reported as CMV DNA copies per μg of DNA for saliva and urine and as CMV DNA copies per mL of breast milk (BM). Samples with less than two positive qPCR replicates were considered not detectable (ND). The frequency of detectable samples across the assessed time points for each animal and sample type is denoted on each plot.

Fig 2. The kinetics of maternal systemic T cell responses and RhCMV shedding.

Maternal systemic T cell responses were measured by IFNγ ELISpot, using the immunodominant peptide pool for each dam (monkeys 337 and 222: pp65; monkeys 196 and 187: IE-1; monkey 205: IE-2). RhCMV virus load was measured by qPCR in maternal breast milk (A, reported as copies/mL) and in maternal saliva and urine (B-C, both reported as copies/μg of DNA). The median and range of the positive qPCR replicates are indicated when 2 or more replicates were positive.

Fig 3. The kinetics of RhCMV-specific T cell responses and RhCMV viral load in breast milk.

Maternal RhCMV-specific immunodominant T cell responses (IE1, IE2, or pp65) in breast milk were measured by cultured IFNγ ELISpot (grey line), using the immunodominant peptide pool for each dam (monkeys 337 and 222: pp65; monkeys 196 and 187: IE-1; monkey 205: IE-2). Breast milk virus load (green line) was measured by qPCR, and the median and range of the positive qPCR replicates are indicated when 2 or more replicates were positive. T cell responses were considered positive if they were greater than twice the background and 50 SFU/10⁶ cells (dotted line).

Fig 4. Infant RhCMV-specific T cell responses and virus shedding.

RhCMV-specific T cell responses against IE1, IE2, pp65, and IL-10 RhCMV homologs were measured in peripheral blood of each infant monkey by IFNγ ELISpot. A response was considered positive if it was greater than twice the background and greater than 50 SFU/10⁶ cells (dotted line). Infant saliva viral load was measured by qPCR and is plotted as the median and range of the positive qPCR replicates.

Fig 5. Infant RhCMV IgG responses during and after the period of lactation.

Infant RhCMV-binding IgG responses were measured by a whole virion RhCMV ELISA and are plotted as the mean and error of technical duplicates. Infant antibody responses through the 16-week postpartum time point (shaded region) represent congenitally-acquired maternal antibody. Infants were weaned and introduced to group housing approximately 26 weeks postpartum. Red arrows indicate infant seroconversion after introduction to group housing.