Comparison of SARS-CoV-2 spike-specific IgA and IgG in nasal secretions, saliva and serum

Abstract

Introduction: Several novel vaccine platforms aim at mucosal immunity in the respiratory tract to block SARS-CoV-2 transmission. Standardized methods for mucosal sample collection and quantification of mucosal antibodies are therefore urgently needed for harmonized comparisons and interpretations across mucosal vaccine trials and real-world data.

Methods: Using commercial electrochemiluminescence antibody panels, we compared SARS-CoV-2 spike-specific IgA and IgG in paired saliva, nasal secretions, and serum from 1048 healthcare workers with and without prior infection.

Results: Spike-specific IgA correlated well in nasal secretions and saliva (r>0.65, p<0.0001), but the levels were more than three-fold higher in nasal secretions as compared to in saliva (p<0.01). Correlations between the total population of spike-specific IgA and spike-specific secretory IgA (SIgA) were significantly stronger (p<0.0001) in nasal secretions (r=0.96, p<0.0001) as opposed to in saliva (r=0.77, p<0.0001), and spike-specific IgA correlated stronger (p<0.0001) between serum and saliva (r=0.73, p<0.001) as opposed to between serum and nasal secretions (r=0.54, p<0.001), suggesting transudation of monomeric spike specific IgA from the circulation to saliva. Notably, spike-specific SIgA had a markedly higher SARS-CoV-2 variant cross-binding capacity as compared to the total population of spike specific IgA and IgG in both nasal secretions, saliva and serum, (all p<0.0001), which emphasizes the importance of taking potential serum derived monomeric IgA into consideration when investigating mucosal immune responses.

Discussion: Taken together, although spike-specific IgA can be reliably measured in both nasal secretions and saliva, our findings imply an advantage of higher levels and likely also a larger proportion of SIgA in nasal secretions as compared to in saliva. We further corroborate the superior variant cross-binding capacity of SIgA in mucosal secretions, highlighting the potential protective benefits of a vaccine targeting the upper respiratory tract.

Keywords: Covid-19; SARS-CoV-2; antibodies; mucosal immunity; nasal sampling; saliva sampling; secretory IgA; vaccines.

Figure 1

COMMUNITY cohort study timeline. Study participants in the COMMUNITY cohort (n=2149) were initially enrolled in April 2020 and have been followed regularly since then. This figure depicts all regular samplings in addition to the specific follow-up designed for this sub study. Participants enrolled in this sub study (n=1048) were sampled in November 2022 (n=72) and February 2023 (n=976). FU, Follow-up.

Figure 2

Comparison of total and spike-specific IgA and IgG levels in nasal secretions and saliva (passive drool). Total IgA and IgG-levels in nasal secretions (A) and in saliva (B). Correlation between spike-specific IgA in nasal secretions and saliva (C). Levels of spike-specific IgA in nasal secretions and saliva (D). Correlations between spike-specific SIgA and spike-specific IgA in nasal secretions (E) and saliva (F). Correlations between spike-specific IgA in serum and saliva (G) and serum and nasal secretions (H). Nasal secretion and saliva spike-specific IgA, SIgA and IgG titers were normalized to total IgA and IgG in the same sample. Dashed lines indicate cut-off levels for spike-specific IgA in nasal secretions (red) and saliva (blue). Solid black lines indicate median antibody levels. The percentage of samples above cut-off level is displayed over each scatter plot. **p < 0.01, ****p < 0.0001. r = Spearman’s rank correlation. SIgA, secretory IgA.

Figure 3

SARS-CoV-2 cross-variant binding capacity of spike-specific SIgA, IgA and IgG. Nasal secretion and saliva spike-specific SIgA, IgA and IgG titers were normalized to total IgA and IgG in the same sample. Green bars represent spike-specific SIgA and IgA and blue bars represent spike-specific IgG. Solid bars represent median ratios, error bars represent 95% confidence intervals. SIgA, secretory IgA; WT, Wild-type. *p < 0.05, ns, non-significant.

Figure 4

Nasal secretion, saliva (passive drool) and serum spike-specific IgA and IgG levels in participants stratified on prior SARS-CoV-2 infection. Nasal secretion, saliva, and serum spike-specific IgA levels stratified by prior infection (A–C). Nasal secretion, saliva (passive drool), and serum spike-specific IgG levels stratified by prior infection (D–F). Dashed lines indicate cut-off levels for detectable spike-specific Ig in nasal secretions (red), saliva (blue) and serum (black). Solid black lines indicate median antibody levels. The percentage of samples with spike-specific Ig above cut-off level is displayed over each scatter plot. Nasal secretion and saliva spike-specific IgA and IgG titers were normalized to total IgA and IgG in the same sample. Ns, non-significant; *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 5

Comparison of spike-specific IgA levels in saliva collected by different methods. Drool (n=16) - Participants were instructed to passively drool into a clean cup for 5 minutes which was then transferred using a pipette into a 2 mL tube. Swab (n=18) - A buccal swab was immersed in saliva accumulated on the tongue, minimizing contact with the oral mucosa and then transferred into a 15 mL Falcon tube with 0.5 mL PBS. Roll no chew (n=9) - An absorbent cotton roll was kept still between the cheek and gum for one minute, the roll was then transferred to a tube provided by the manufacturer and centrifuged for 5 minutes at 1000g, leaving saliva at the bottom of the tube. Roll chew (n=9) - An absorbent cotton roll was kept in the mouth for three minutes, during which saliva production was stimulated by chewing, the roll was then transferred to a tube provided by the manufacturer and centrifuged for 5 minutes at 1000g, leaving saliva at the bottom of the tube. Correlation coefficients (Spearman’s rank correlations) between sample types are indicated by numbers in the heatmap (A). Individual spike-specific IgA levels in saliva (n=18) (B). Coefficients of variation (CV) between sample types are indicated by percentages in the heatmap (C) Spike-specific IgA titers were normalized to total IgA in the same sample.

Figure 6

Comparison of SARS-CoV-2 spike-specific IgA levels in nasal secretions, saliva (collected by buccal swabbing) and serum. Correlation between spike-specific IgA in nasal secretions and saliva (A). Spike-specific IgA levels in nasal secretions and saliva (B). Nasal secretion spike-specific IgA levels stratified by prior infection (C). Saliva spike-specific IgA levels stratified by prior infection (D). Serum spike-specific IgG levels stratified by prior infection (E). Dashed lines indicate cut-off levels for detectable spike-specific Ig in nasal secretions (red), saliva (blue) and serum (black). Solid black lines indicate median antibody levels. The percentage of samples with spike-specific Ig above cut-off level is displayed over each scatter plot. Nasal secretion and saliva spike-specific IgA levels were normalized to total IgA in the same sample. Ns, Non-significant; *p < 0.05, ****p < 0.0001. r = Spearman’s rank correlation.