Monitoring of systemic and local immune signatures in response to HCMV reactivation during lactation

Abstract

The human herpesvirus cytomegalovirus (HCMV) persists in the host in latent form for a lifetime and may periodically reactivate. During lactation, reactivation of HCMV occurs locally in the mammary gland without detectable systemic DNAemia. Preterm infants are at risk of infection through ingesting breast milk and may develop life-threatening disease with sepsis-like symptoms. In this thesis, the modulation of the immune signature in response to HCMV reactivation in the mammary gland, investigated non-invasively by analyzing breast milk, was compared to corresponding simultaneously drawn blood samples of HCMV-seropositive, as well as HCMV-seronegative mothers. This “BlooMil” study was performed on longitudinal samples taken from birth through 60 days postpartum over four time ranges. The viral load in breast milk whey of HCMV-seropositive mothers showed a unimodal course and revealed high differences in peak viral loads between individuals (104 – 106 copies/ml). Breast milk whey of only one of 18 HCMV-seropositive mothers did not contain reactivated virus during the observation period. The humoral immune response against HCMV assessed as IgG in breast milk was very low. However, six of 18 mothers showed an increase of HCMV-specific IgGs after peak viral load. Five of 18 mothers had no measurable HCMV-IgG concentrations in breast milk using an electro-chemiluminescence immunoassay. Nevertheless, recomLine blots for six prominent HCMV antigens (IE1; CM2, fusion protein of pUL44 and UL57; p150; p65 and gB glycoprotein 1 and 2) detected anti-p150-IgG reactivity with intensities at least around the cut-off level in all breast milk samples. Anti-gB1 IgG was detected in all plasma samples, but only in 5 of 18 milk samples. Plasma HCMV-IgG titers increased over the analyzed time range. Immunomonitoring of mothers revealed a significant increase of CD3+ T cells in breast milk of HCMV-seropositive but not seronegative mothers over time. The CD4+/CD8+ T cell ratio was significantly lower in breast milk of HCMV-seropositive than seronegative mothers. Additionally, most T cells were activated (HLA-DR+) memory T cells, which increased over time. Naïve T cells were present only in very limited numbers in breast milk. HCMV-specific T cells were measured by quantifying CD8+ T cells binding to pp65 and IE1 peptides using MHC class I tetramers. Most breast milk samples displayed slightly higher frequencies of HCMV-specific T cells than the corresponding blood samples. The analysis of 92 inflammatory cytokines revealed elevated levels of CC- and CXC- chemokines in HCMV-seropositive compared to -seronegative mothers’ breast milk. The humoral IgG immune response in breast milk of most HCMV-seropositive mothers resulted in no evidence of interrelation to the observed decreasing viral load in the milk during the unimodal course of reactivation. However, a compartmentalization in the mammary gland regarding immune cells, such as CD3+ T cells and activated (HLA-DR+) CD8+ T cells, was observed. These immune cells showed increasing frequencies in breast milk over time compared to corresponding blood samples or seronegative mothers’ breast milk and might therefore contribute to the decrease of HCMV loads in breast milk after the peak.