Fetal human cytomegalovirus transmission correlates with delayed maternal antibodies to gH/gL/pUL128-130-131 complex during primary infection

Abstract

Primary human cytomegalovirus (HCMV) infections during pregnancy are associated with a high risk of virus transmission to the fetus. To identify correlates of intrauterine HCMV transmission, serial serum samples from HCMV transmitter and non-transmitter pregnant women with primary HCMV infection were analyzed for the presence of neutralizing antibodies against different glycoproteins and glycoprotein complexes, which are known to mediate entry into distinct types of host cells. Neutralizing activity was detected in the sera early after primary infection; absorption with a soluble pentameric complex formed by gH/gL/pUL128-131, but not with gH/gL dimer or with gB, abolished the capacity of sera to neutralize infection of epithelial cells. Importantly, an early, high antibody response to pentamer antigenic sites was associated with a significantly reduced risk of HCMV transmission to the fetus. This association is consistent with the high in vitro inhibition of HCMV infection of epithelial/endothelial cells as well as cell-to-cell spreading and virus transfer to leukocytes by anti-pentamer antibodies. Taken together, these findings indicate that the HCMV pentamer complex is a major target of the antibody-mediated maternal immunity.

Figure 1. SDS-PAGE and Western blot (WB) analysis of soluble HCMV glycoprotein complexes.

(A) Pentameric gH/gL/pUL128-131 complex, (B) gH/gL and (C) gB complexes were subjected to SDS-PAGE and Western blotting. Samples were stained with Coomassie blue (left panels) or probed with specific antibodies after WB (right panels). (B) A band of about 60 kDa on gL blot is apparently a co-purified contaminant recognized by anti-gL polyclonal serum raised in rabbits. (C) Soluble gB exists in full-length and truncated forms that associate in dimeric complex, as described in Carlson et al. (32). N-terminus of truncated gB form is not detected since gB was his-tagged on the C-terminus and the blot was probed with anti-histag antibody.

Figure 2. Kinetics of the antibody response to primary HCMV infection.

(A) Kinetics of the appearance of anti-pentamer (IgG-Pent), anti-gH/gL (IgG-gH/gL) and anti-gB (IgG-gB) antibodies in 46 subjects with primary HCMV infection. Shown are individual values and non-linear regression curves for the three specificities. P-values were calculated using the extra-sum-of square F test. (B) Kinetics of the mean HCMV serological response in the 46 patients’ sera. IgG-HCMV refers to HCMV lysate-specific IgG Arbitrary Units (AU); IgM-HCMV refers to HCMV lysate-specific IgM ratio; Nt-HELF and Nt-ARPE19 refer to neutralizing antibody (Neut Ab) titer on human embryonic lung fibroblasts (HELF) and epithelial (ARPE-19) cells, respectively; HCMV DNA refers to viral load in blood. (C) Correlation between IgG-pentamer antibody titers measured by ELISA and neutralizing antibody titers measured using ARPE-19. Values were obtained from sera collected <60 day post infection (filled diamonds) or >60 days post infection (empty diamonds). The regression coefficient (r²) is also shown. (D) Neutralizing antibody (Neut Ab) titers of sera from 3 patients (Pt.A, Pt.B, and Pt.C) were measured using ARPE-19. Sera were used either untreated or after absorption with the soluble pentameric complex, gH/gL dimer, or gB proteins from HCMV or with influenza A virus hemagglutinin (HA).

Figure 3. IgG antibody titers and DNA viral load at three time points of primary HCMV infection.

IgG antibody titers in 11 non-transmitter (NT) and 12 transmitter (T) mothers against gB (A), gH/gL (B), or pentamer (C) at three different time intervals (≤30 days, 30–60 days, >60 days) after onset of infection. (D) HCMV DNA copies/ml blood in the same sera samples at the three time intervals. Dotted lines represent the detection limit of the assays. P-values were calculated using the Mann-Whitney U-test (A-C) or the Fisher’s exact test (D).

Figure 4. Cumulative antibody response to the pentamer neutralization sites.

Patients’ antibody response to neutralizing sites of the pentamer estimated via inhibition of monoclonal antibody binding (IMAB) assay using a panel of 10 human monoclonal antibodies with defined specificity for different sites on the HCMV pentamer complex, as indicated. The cumulative incidence of site-specific antibodies in 23 pregnant women with primary HCMV infection during the entire follow-up period is shown.

Figure 5. IMAB₅₀ titers by site-specific antibodies present in human sera from primary HCMV infection.

IMAB₅₀ titers of neutralizing site-specific IgG antibodies in 12 transmitter (T) and 11 non-transmitter (NT) mothers at three different time intervals (≤30 days, 30–60 days, >60 days) after onset of infection. (A-J): Sites 1 to 10 as defined by the indicated human monoclonal antibodies to gH/gL/pUL128-130-131 . Dotted lines represent the detection limit of the assays. P-values were calculated using the Mann-Whitney U-test; n.d., not detected.

Figure 6. Overall number of neutralization sites of the pentamer reactive with human sera and viral load at three time points.

Number of neutralization sites recognized by sera from 11 non-transmitter (NT) and 12 transmitter (T) mothers at (A) less than 30 days, (C) 31–60 days, and (E) more than 60 days after onset of infection. (B, D, F): Number of HCMV-DNA positive (black bars) and HCMV-DNA negative (white bars) women in the NT and T groups at the same time intervals. P-values were calculated using the Mann-Whitney U-test (A, C, E) or the Fisher’s exact test (B, D, F).

Figure 7. Plaque formation and leukocyte transfer inhibition by human monoclonal and serum antibodies.

Dose-effect inhibition of (A) HCMV plaque formation (PFI) in ARPE-19 cells, and (B) HCMV transfer to leukocytes (LTI) from infected HUVEC cells by serial concentrations of human mAbs specific for different sites on the pentamer complex. An anti-gB mAb (7H3) was used as a control. (A) mAbs to gB and gH show a PFI₅₀ activity at a concentration about a thousand-fold higher in comparison to human mAbs directed to pUL128-131. (B) Comparable LTI₅₀ activity was displayed by human mAbs directed to pUL128-131, whereas mAbs to gH and gB again show a much weaker inhibitory effect. (C) Individual PFI₅₀ titers in sequential sera collected within the first 3 months after onset of infection in 5 transmitter (T) and 4 non-transmitter (NT) pregnant women. (D) PFI₅₀ titers at three different time intervals (≤30 days, 30–60 days, >60 days) after onset of infection are shown. P-values were calculated using the Mann-Whitney U-test. (E) Regression curves of PFI₅₀ titers in transmitter (T) and non-transmitter (NT) women. P-value was calculated using the extra-sum-of square F test.