The IgA in milk induced by SARS-CoV-2 infection is comprised of mainly secretory antibody that is neutralizing and highly durable over time

Abstract

Approximately 10% of infants infected with SARS-CoV-2 will experience COVID-19 illness requiring advanced care. A potential mechanism to protect this population is passive immunization via the milk of a previously infected person. We and others have reported on the presence of SARS-CoV-2-specific antibodies in human milk. We now report the prevalence of SARS-CoV-2 IgA in the milk of 74 COVID-19-recovered participants, and find that 89% of samples are positive for Spike-specific IgA. In a subset of these samples, 95% exhibited robust IgA activity as determined by endpoint binding titer, with 50% considered high-titer. These IgA-positive samples were also positive for Spike-specific secretory antibody. Levels of IgA antibodies and secretory antibodies were shown to be strongly positively correlated. The secretory IgA response was dominant among the milk samples tested compared to the IgG response, which was present in 75% of samples and found to be of high-titer in only 13% of cases. Our IgA durability analysis using 28 paired samples, obtained 4-6 weeks and 4-10 months after infection, found that all samples exhibited persistently significant Spike-specific IgA, with 43% of donors exhibiting increasing IgA titers over time. Finally, COVID-19 and pre-pandemic control milk samples were tested for the presence of neutralizing antibodies; 6 of 8 COVID-19 samples exhibited neutralization of Spike-pseudotyped VSV (IC50 range, 2.39-89.4ug/mL) compared to 1 of 8 controls. IgA binding and neutralization capacities were found to be strongly positively correlated. These data are highly relevant to public health, not only in terms of the protective capacity of these antibodies for breastfed infants, but also for the potential use of such antibodies as a COVID-19 therapeutic, given that secretory IgA is highly in all mucosal compartments.

Fig 1. A robust, Spike-specific IgA response in milk commonly occurs after SARS-CoV-2 infection.

(A) Screening of undiluted milk samples for specific IgA by ELISA against the full-length Spike trimer. COVID-19 group, N = 74; control group N = 20. Mean values with SEM are shown. Dotted line: positive cutoff value (mean OD of negative control milk samples + 2*SD). ****p<0.0001. Mann-Whitney U test (2-tailed) was used to compare grouped data with significance level set at p < 0.05. (B) Full titration against Spike of 40 milk samples found to be positive by the initial screening. (C) Endpoint dilution titers of the 40 titrated milk samples. Segmented line: positive cutoff value; dotted line: 5x positive endpoint cutoff value, designating samples as ‘high-titer’. Mean values with SEM are shown.

Fig 2. The dominant Spike-specific IgA response in milk after SARS-CoV-2 infection is strongly correlated with a robust secretory Ab response, while specific IgG activity is relatively modest.

Twenty samples assayed for Spike-specific IgA were also assessed for Spike-specific secretory Ab (by detecting for SC), and IgG. (A, B) Full titration against Spike, detecting (A) secretory Ab, and (B) IgG. NEG (i.e. negative)/segmented lines: pre-pandemic controls. COV/solid lines: milk from COVID-19-recovered donors. Dotted lines: positive cutoff values. (C, D) Endpoint titer values calculated for (A) secretory Ab, and (B) IgG. Segmented lines: positive cutoff values; dotted lines: 5x positive cutoff (high-titer cutoff). (E) IgA and secretory Ab binding OD values or endpoint titers were used in 2-tailed Spearman correlation tests. SC: secretory component.

Fig 3. The Spike-specific IgA response in milk after SARS-CoV-2 infection is highly durable over time.

(A) IgA endpoint titers determined from Spike ELISA for 28 pairs of milk samples obtained from COVID-19-recovered donors 4–6 weeks and 4–10 months after infection are shown. Mean endpoint values for each group are shown. Blue lines: >10% increase, red lines: >10% decrease, grey lines: <10% change. NS: not significant. A paired t-test (2-tailed) was used to assess significance. (B) IgA endpoint titers for a subset of 14 paired samples obtained 4–6 weeks and 7–10 months after infection. Mean with SEM is shown. Mean endpoint values for the 4–6 week and 7–10 month groups are indicated on the y-axis as green and pink ticks, respectively. Blue bars: >10% increase, red bars: >10% decrease, grey bars: <10% change.

Fig 4. Extracted milk IgA from COVID-19-recovered donors exhibits SARS-CoV-2 Spike-targeted neutralization potency that is highly correlated with IgA binding activity.

(A) Total IgA was purified from 8 COVID-19 and 8 control milk samples by conventional means using peptide M agarose. IgA was titrated and tested in a VSV-based SARS-CoV-2 pseudovirus neutralization assay. NEG/segmented lines: pre-pandemic controls. COV/solid lines: COVID-19-recovered milk samples. Segmented line: 50% neutralization cutoff value. (B) Percent neutralization achieved using 50ug/mL of total extracted milk IgA. Mean values with SEM are shown. **p = 0.0064. Mann-Whitney U test (2-tailed) was used to compare grouped data with significance level set at p < 0.05. (C) Neutralization IC50 values determined from IgA titration curves. (D) Endpoint titer values determined in Fig 1 and IC50 values were used in a 2-tailed Spearman correlation test.