Two Cells, One Antibody: The Discovery of the Cellular Origins and Transport of Secretory IgA

Secretory Abs of the IgA and IgM isotypes are the foot soldiers of the mucosal immune system, serving as the first line of specific immunity against inhaled, ingested, and sexually transmitted Ags. Adult humans transport some 3 g/d of secretory (SIgA) into external secretions, more than the daily production of all other Ig isotypes combined. Studies in the 1960s and early 1970s revealed a novel structure for SIgA compared with that of serum-type IgA, comprising a dimer of two basic units of IgA (each containing two identical α H chains and two identical κ or λ L chains) connected by a small joining chain (J chain), as well as a polypeptide chain unique to secretory Igs, initially called secretory piece and later named secretory component (SC) (1–8). Immunohistochemical studies revealed that SIgA was the product of two cell types: J chain–containing IgA dimers were found to be secreted by local plasma cells, whereas SC was shown to be produced by epithelial cells lining glandular and mucosal surfaces (9). Several locations were proposed for the association of SC with the IgA dimer to form SIgA: in the extracellular space, at the surface of epithelial cells, or within epithelial cells during an as yet undefined transport process.

In the featured Pillars of Immunology article by Professor Per Brandtzaeg (10), published in 1974 in The Journal of Immunology, evidence was presented in support of a model in which SC acts as a membrane-bound receptor that mediates epithelial transcytosis of SIgA. The experiments in this study used previously characterized Abs (5) that distinguished free (i.e., uncomplexed) SC from bound SC (i.e., part of the SIgA complex). Immunohistochemical localization of IgA, IgG, free SC, and bound SC in fixed sections of human nasal and colonic mucosa revealed that epithelial cells contain membrane-associated IgA and SC but not IgG. Free SC alone was detected intracellularly in the Golgi region of epithelial cells, whereas IgA, as well as free and bound SC, was detected in the cell membranes and in the apical portion of the cytoplasm. Based on these findings, Brandtzaeg proposed the following stepwise process for the selective external transfer of IgA into mucosal and glandular secretions. Free SC is produced by serous-type epithelial cells and accumulates in the Golgi apparatus, from which it is distributed to the rest of the cytoplasm and to the cell membrane. Dimeric IgA with J chain is produced by local plasma cells adjacent to the secretory epithelium. Membrane-associated SC acts as an epithelial receptor with specific affinity for dimeric IgA. The SC–IgA complexes are delivered from the membrane to the cytoplasm outside the Golgi region by pinocytosis or facilitated diffusion; at some point during this process, enzymatically catalyzed interunit disulfide bonds stabilize the SIgA complex. Finally, the SIgA molecules are extruded into the glandular lumen along with an excess of free SC. Although consistent with all of the data in this and earlier studies, Brandtzaeg was careful to point out that the transport model was based solely on in vitro experiments with purified proteins and immunofluorescence of dead tissue. The article concluded with the prescient statement that “it should be possible in the future to test several of the proposed steps by kinetic studies on living cells.” Indeed, the major features of this model were confirmed and extended during the subsequent four decades by studies in living cells and, ultimately, in living animals.

Initial support for Brandtzaeg’s model of IgA transport came from studies with cultured epithelial cells that expressed surface SC and were found to internalize polymeric IgA and IgM (11–13). Further immunohistochemical studies in Brandtzaeg’s laboratory provided evidence for an integrated function of J chain and SC in the epithelial transport of polymeric IgA and IgM (14, 15). Mechanistic evidence for SC as a membrane receptor for IgA came with the discovery that SC is synthesized initially by epithelial cells as a transmembrane protein (16, 17). To better reflect its functional role, transmembrane SC was subsequently given the synonym polymeric IgR (pIgR). Molecular cloning of the PIGR gene from several species (18–23) confirmed the presence of a leader polypeptide that would deliver newly synthesized pIgR into the endoplasmic reticulum and Golgi complex, as well as a transmembrane segment that would anchor pIgR in cellular membranes. Transfection of pIgR cDNA into living cultures of MDCK cells confirmed that ectopic expression of pIgR alone conferred the ability of polarized epithelial cells to transport dimeric IgA in a basolateral-to-apical direction, mimicking the natural transport process in mucosal and glandular epithelia (24). This experimental approach also facilitated the identification of sequence motifs in the extracellular region of the pIgR polypeptide that were responsible for selective binding of J chain–containing polymeric IgA and IgM (25–28), as well as motifs in the cytoplasmic domain of the pIgR molecule that were responsible for its intracellular trafficking (29). The cell culture pIgR–IgA transcytosis model was also used to demonstrate that SIgA can effect intracellular neutralization of viruses and excretion of IgA-containing immune complexes (30, 31). The ultimate proof of Brandtzaeg’s transport model was the demonstration that external secretions of mice with targeted deletions of the genes encoding pIgR (32) or J chain (33) were devoid of SIgA.

A corollary of Brandtzaeg’s transport model is that the rate of transport of SIgA could be regulated by the level of pIgR expression in mucosal and glandular epithelial cells. Support for this concept came from the finding that mice with a targeted deletion of one Pigr allele (Pigr^+/− mice) had significantly reduced levels of fecal SIgA compared with wild-type (Pigr^+/+) mice (34). It could further be postulated that upregulation of pIgR expression would enhance the rate of SIgA transport. Sequence analyses of the human and murine PIGR genes revealed the presence of regulatory motifs that respond to signaling pathways initiated by host cytokines and microbe-associated molecular patterns (35–41). The biological significance of these findings was supported by studies demonstrating reduced pIgR expression and/or mucosal SIgA levels in mice with targeted deletions of genes encoding key factors in cytokine- and microbe-activated signaling pathways (42–44). These molecular studies provided a theoretical framework for observations made by Brandtzaeg some 30 y earlier that mucosa-associated bacteria are coated with IgA in vivo (45) and that, SIgA, in contrast to other Ig isotypes, can limit uptake of environmental pathogens and Ags and the formation of immune complexes within mucosal tissues without provoking damaging inflammatory responses (46). In addition to the long-known role of SIgA in preventing infections at mucosal surfaces, it is now appreciated that cooperativity among SIgA, pIgR, and the commensal microbiota promotes host–microbial mutualism (47–52), which is critical for maintaining optimal health in the gut and throughout the body. The model for SIgA transport proposed in this Pillars of Immunology article still stands after four decades and has been incorporated into immunology and cell biology textbooks as an illustration of epithelial transcytosis of Igs. Brandtzaeg’s seminal work on SIgA has inspired generations of mucosal immunologists to unravel the structure, transport, and function of our most abundant Ab isotype.