Measles virus infection diminishes preexisting antibodies that offer protection from other pathogens

Abstract

Measles virus is directly responsible for more than 100,000 deaths yearly. Epidemiological studies have associated measles with increased morbidity and mortality for years after infection, but the reasons why are poorly understood. Measles virus infects immune cells, causing acute immune suppression. To identify and quantify long-term effects of measles on the immune system, we used VirScan, an assay that tracks antibodies to thousands of pathogen epitopes in blood. We studied 77 unvaccinated children before and 2 months after natural measles virus infection. Measles caused elimination of 11 to 73% of the antibody repertoire across individuals. Recovery of antibodies was detected after natural reexposure to pathogens. Notably, these immune system effects were not observed in infants vaccinated against MMR (measles, mumps, and rubella), but were confirmed in measles-infected macaques. The reduction in humoral immune memory after measles infection generates potential vulnerability to future infections, underscoring the need for widespread vaccination.

Fig. 1.. Measles virus infections reduce antibody diversity.

(A) Total epitopes recognized at time 1, left, and time 2, right, per cohort. For comparison across cohorts, values are standardized across all samples per cohort to a mean of 0 and standard deviation of 1. Each gray line indicates a paired sample from an individual, and black connecting lines indicate mean change from zero. Boxplots indicate interquartile range and median. Asterisks indicate paired t test P values. (B) Fold change of total antibody diversity (i.e., number of total epitope hits) at time 2 versus time 1. Each point represents one paired sample from an individual. Boxplots indicate interquartile range and median. Asterisks indicate significant differences relative to control A (Cntl A), based on student’s t test P values. (C) Number of measles epitope hits per sample at time 1 and time 2. Thin lines indicate paired samples. Black lines indicate cohort averages. (D) As in (A), but for individual viruses. Bonferroni-corrected P values in (A to D): *P < 0.05, **P < 0.001, ***P < 0.001, ****P < 0.0001. (E) Heatmap indicating the change in the total number of epitope hits per species between time 1 and time 2. Each column represents an individual paired sample and each row a pathogen. Cohorts are indicated by the solid bars at top and bottom and are in the same order (left to right) as in (A). Yellow cells indicate that pathogen was not assayed in those samples. Ab, antibody.

Fig. 2.. Measles eliminates preexisting immune memory.

(A) The proportion of total epitopes detected at time 1 that were retained at time 2. One point represents one child. Bonferroni-adjusted student’s t test P values for significant differences relative to control A are shown (****P < 0.0001). (B) Probability of retaining initial antibodies at time 2 for individual pathogens per child. Each point represents a single pathogen per child. Probabilities are obtained by fitting a binomial random effects model controlling for interval duration (materials and methods). Each boxplot represents the interquartile range and median retention probabilities calculated across pathogens for a single child. Points indicate pathogen-specific retentions that are outside of the interquartile range per child. Children are rank ordered along the x axis by the median (black dot per boxplot) retention probabilities. (C) Density distribution of probabilities shown in (B), derived by collapsing all points in (B) by cohort. Bonferroni-adjusted Wilcoxon signed rank test P values for differences relative to control A are shown (***P < 0.001).

Fig. 3.. Measles virus infections are associated with diminished epitope binding signal.

Epitope binding signal (EBS) was measured for each epitope. (A) Each point represents one pathogen species per sample pair. Changes in EBS for all epitopes recognized per pathogen per sample pair were compared using a Wilcoxon matched-pairs signed rank test. Fdr-adjusted P values (adjusted for a fdr of 5%) are indicated by color of each symbol. The total number of epitopes included in each paired test (i.e., the number of epitopes recognized for each species per paired sample) are indicated by the symbol shape. For each pathogen, the geometric mean (gMean) EBS was calculated at each time and the fold change at time 2 versus time 1 was calculated and indicated by the position of each point along the y axis. Density distributions adjacent to the points reflect the distribution of points that were significantly changed on the basis of a fdr P value < 0.05. Only pathogens with antibodies targeting ≥5 epitopes were included. (B) Change in gMean EBS per epitope, across all paired samples per cohort. Format as in (A), except here each point represents a single epitope, and the gMean EBS is calculated across all paired samples with the epitope detected. The color of each point represents the fdr-adjusted Wilcoxon matched-pairs signed rank test P value, and shapes indicate the number of paired samples per cohort with antibodies against the particular epitope). Only epitopes recognized in >6 paired samples were included. (C) Scatter plot showing each epitope (point) shown in (B), but instead of plotting the fold change, the gMean EBS at time 1 (x axis) versus time 2 (y axis) is plotted. The P values (colors) and samples sizes (shapes) are indicated. MV-specific epitopes were included and highlighted in green. Epitopes significantly changed in a positive direction are labeled with pathogen species name. NS, not significant; PIV-4, parainfluenza virus-4.

Fig. 4.. Increases in antibody epitope binding signal cluster within households and in postal units.

We tested whether individuals with increased EBS indicate new exposures due to pathogen transmission by looking for evidence of spatial clustering among children with increases in EBS. In (A to C), each child (node) is connected by an edge to their postal code (indicated by a central node with an assigned value, 2 through 7). Children shown as single nodes were the only children studied from their postal code. Children from the same household are connected by an edge and encircled by dashed gray ovals. Each of the three viruses where increases in EBS were most common are shown: (A) adenovirus C, (B) influenza A virus, and (C) respiratory syncytial virus. Children with significantly increased EBS for the pathogen (indicating exposure during the interval) are highlighted with red circles and those with decreased or unchanged EBS with gray circles.

Fig. 5.. Measles virus infection in macaques deletes preexisting immune memory.

Four rhesus macaques (14Y, 31Y, 46Y, and 50Y) were infected with the Bilthoven strain of wild-type MV [detailed in (34)], and plasma samples were collected before and 5.1 months after infection. (A) Heatmap showing the percentage change in total epitopes recognized per pathogen in each monkey from time 1 to time 2. Pathogens with >10 unique epitopes recognized at either time point per macaque are shown. The color and text indicate percent change in total epitopes recognized per pathogen species before versus after infection. (B) Heatmap showing the signal strength of anti-measles antibodies (EBS) for each epitope for which at least one monkey developed an antibody. Colors and text of each cell represent the respective EBS values. Letters indicate the MV protein (F, fusion protein; H, hemagglutinin; N, nucleoprotein; and P, phosphoprotein), and the numbers indicate the position along the protein of the first amino acid of the 56-amino-acid peptide. (C) The fractions of epitopes recognized before MV infection that remained 5 months after are shown (0 months after infection is baseline and thus set to 1 for all pathogens). Each gray line indicates a different pathogen, and the dark black line indicates the average across the pathogens. The boxplot summarizes the interquartile range and median fraction retained across the pathogens.