IgA antibodies in saliva as a noninvasive readout for intestinal immune responses

Enterotoksigene Escherichia coli (ETEC) utgjør en vesentlig årsak til diare hos barn i land med lav og middel s inntekt. ETEC bindes til epitelet i tarmen ved hjelp av koloniseringsfaktorer og forårsaker diare ved å frigi enteretoksiner. Det er behov for et ikke-invasivt korrelat som mål for beskyttende immunitet i tarmen når man skal teste effekten av intestinale vaksiner. Hensikten med dette prosjektet er å evaluere om IgA i spytt kan være et mål for immunitet i tarmen. Vi utviklet en flowcytometrisk metode for å måle slike responser. Vi fant at frivillige unge personer som hadde drukket ETEC bakterier hadde en kraftig immunrespons mot ETEC bakteriene i serum, i lymfocytt-supernatant (ALS), i tarmskyllevæske og i spytt. Disse antistoffresponsene var spesifikke mot test-stammen, og reagerte ikke med en heterolog ETEC-stamme. Dette tyder på at mye av responsen varrettet mot stammespesifikke lipopolysakkarider, dvs. O antigener. Resultatene viste også at spytt samlet under tunga (vesentlig fra submandibulære/sublingual e spyttkjertler) gjenspeilte tarmens immunrespons bedre enn spytt produsert I parotiskjertelen. Dette understøtter vår opprinnelige hypotese om at en subpopulasjon av B-celler stimulerte i tarmen vandrer spsielt til de submandibulære/sublingual e spyttkjertler. Studien viser også at ETEC-infeksjon gir en økning av total IgA i spytt (såkalt «bystander stimulation»); dette må tas med i betraktning når man vurderer om en skal normalisere IgA-responsen i spytt mot total IgA, noe som ofte gjøres. En vitenskapelig rapport som beskriver disse resultatene er publisert i oktoberutgaven (2015) av Mucosal Immunity.

Investigators at academic institutions in Oslo and Bergen propose research work to answer whether salivary specimens can be used to measure relevant IgA immune responses following intestinal immunization. Non-parotid salivary secretions have been proposed to reflect immunity induced in the gut-associated lymphoid tissue. The original idea was that this project should be closely integrated with WP6 of the GLOBVAC-funded project EntVac, in cooperation with the Norwegian Institute of Public Health (NIPH), wh ere the various immunoassays are being developed and performed. Appropriate enzyme-linked immunoassays (ELISAs) have been established for measurements of total secretory IgA (SIgA) and IgA as well as specific SIgA and IgA antibodies to the subunit B of Es cherichia coli heat-labile enterotoxin (LTB) and a recombinantly expressed surface antigen, EtpA. Extensive comparisons of different variables have led to the conclusion that the assays are more robust for both total IgA and IgA antibodies than for the SI gA counterparts. Analyses of salivary specimens obtained from ETEC-infected volunteers suggest that our original hypothesis indeed may be true because the specific IgA titres in sublingual/submandibular (SL) secretions of the infected individuals were rel atively high after day 7 compared to the mean titres in parotid secretions, whereas rechallenged individuals showed only a retarded response in SL and, on average, no response in parotid secretion. We found it puzzling, however, that when the IgA antibody titres of SL fluid were normalized against the total IgA level in individual samples, the response pattern disappeared. This result could be explained if ETEC-specific IgA constitutes a major portion of total IgA in SL fluid after enteric infection, or t here might still be an unexplained problem with nonspecific binding of IgA to the ELISA plates in some of the samples. The wide scatter among the ETEC titres favours the latter possibility, and all the problems reported by other researchers who have attem pted to use saliva IgA antibody as a proxy for gut immunity point to the fact that considerable pitfalls exist in ELISA-based measurements of antibodies in saliva. As an alternative method, we used heat-inactivated E. coli and examined the antibody respon se in serum by flow cytometry; a distinct IgG response was regularly observed and usually also a distinct IgA response. We believe that the antibody titres seen by this method are mainly directed against lipopolysaccharide (LPS, endotoxin) in the cell wal l of E. coli, but we cannot exclude that other surface antigens are involved. Our flow-cytometric method proved sensitive and showed that the orally infected volunteers indeed had responded immunologically after exposure to ETEC, although most of them rem ained clinically quite well. Based on these results, we would like to exploit flow cytometry further to establish a highly sensitive and versatile method to measure immune responses against purified E. coli antigens, both in serum and SL saliva specimens as well as in gut lavage fluid and ALS (antibodies in lymphocyte secretions). The best analytical method we could choose is the multiplex assay where reactivity for several antigens can be tested at the same time. With this particle-based assay nonspecifi c staining should be a negligible problem in contrast to our experience with salivary specimens on ELISA plates. We believe that this will be the most productive approach to establish a flexible and robust method for readouts of gut immune responses such as in serum, gut lavage and ALS, as well as in salivary specimens. We plan to purify LPS from the ETEC challenge strains and, if resources and time permit, also test other relevant E. coli antigens.