Increased immunoglobulin A levels in milk by over-expressing the murine polymeric immunoglobulin receptor gene in the mammary gland epithelial cells of transgenic mice

Abstract

The polymeric immunoglobulin receptor (pIgR) transports dimeric immunoglobulin A (dIgA) across the epithelial cell layers into the secretions of various mucosal and glandular surfaces of mammals. At these mucosal sites, such as the gastrointestinal tract, respiratory tract, urogenital tract and the mammary glands, dIgA protects the body against pathogens. The pIgR binds dIgA at the basolateral side and transports it via the complex mechanism of transcytosis to the apical side of the epithelial cells lining the mucosa. Here, the extracellular part of the receptor is cleaved to form the secretory component (SC), which remains associated to dIgA, thereby protecting it from degradation in the secretions. One pIgR molecule transports only one dIgA molecule (1 : 1 ratio) and the pIgR is not recycled after each round of transport. This implies that the amount of available receptor could be a rate-limiting factor determining both the rate and amount of IgA transported per cell and therefore determining the total IgA output into the lumen or, in case of the mammary gland, into the milk. In order to test this hypothesis, we set up an in vivo model system. We generated transgenic mice over-expressing the murine pIgR gene under lactogenic control, by using a milk gene promoter, rather than under immunological control. Mice over-expressing the pIgR protein, in mammary gland epithelial cells, from 60- up to 270-fold above normal pIgR protein levels showed total IgA levels in the milk to be 1.5-2-fold higher, respectively, compared with the IgA levels in the milk of non-transgenic mice. This indicates that the amount of pIgR produced is indeed a limiting factor in the transport of dIgA into the milk under normal non-inflammatory circumstances.

Figure 1

Glycosylation of the murine SC protein. Western blot analysis of milk samples from F₂ females of four transgenic lines. Milk samples (1 µl) were treated with (+) or without (−) N-glycosidase F (1 U) and separated on a 7·5% SDS–PAGE gel. The SC protein has a MW of 95 000–100 000; the deglycosylated form has a MW of 80 000–85 000. Control milk samples: -M₁, non-transgenic littermate (mouse 9726, 13 days lactation); -M₂, non-transgenic littermate (mouse 9714, 11 days lactation). Milk samples: (a) line 3642 (mouse 8189, 12 days lactation); line 3643 (mouse 4779, 14 days lactation). (b) line 3644 (mouse 9507, 11 days lactation); line 3646 (mouse 4769, 13 days lactation). Molecular weight markers are indicated on the left (in kDa).

Figure 2

Intracellular and extracellular localization of the pIgR protein in the mammary gland. Mammary gland tissue from mice of all transgenic lines was isolated at day 12 of lactation and tissue sections were incubated with a rabbit anti-human SC antibody. This antibody cross-reacts with the murine SC protein. (a) Mammary gland tissue of line 3646 incubated with PBS as a negative control. (b) A non-transgenic littermate to detect the endogenous pIgR expression in the epithelial cells of the mammary gland. The secretory epithelial cells of the alveoli show specific pIgR staining in the lines 3642 (c), 3643 (d), 3646 (e) and 3644 (f).

Figure 3

IgA and SC in the milk of transgenic mice. Western blot analysis of milk samples, collected from F₂ female mice of all transgenic lines. Samples were diluted in PBS as indicated below and 3 µl was fractionated on a 7·5% SDS–PAGE gel under reducing conditions (a) or non-reducing conditions (b). (a) The blot was incubated with a rabbit anti-human SC antibody and a rabbit anti-mouse IgA (α) antibody. Milk samples used: -M, non-transgenic littermate (13 days lactation); line 3642 (mouse 5967, 12 days lactation); line 3643 (mouse 8068, 13 days lactation); line 3644 (mouse 9507, 13 days lactation); line 3646 (mouse 5971, 14 days lactation). Dilutions used: line 3642, 1 : 10; line 3643, 1 : 50; line 3644, 1 : 500; line 3646, 1 : 200; non-transgenic littermate, 1 : 10. C1, purified human sIgA from colostrum (hsIgA, 20 ng SC). The 80 000 MW SC protein of the human sIgA molecule is visible and used as a reference. C2, mouse myeloma protein IgA (MOPC 315, 30 ng). The IgA murine heavy chain is shown as a band of 60 000 MW. The numbers on the left indicate the molecular weight of the protein standards (in kDa). mSC, murine secretory component (MW 95 000–100 000); mIgA (α), murine IgA heavy chain (MW 60 000). (b) The blot was incubated with a rabbit anti-mouse IgA (α) antibody. Milk samples used were diluted 1 : 10 in PBS. -M, non-transgenic littermate (14 days lactation); line 3642 (mouse 8189, 12 days lactation); line 3643 (mouse 4779, 14 days lactation; mouse 8068, 13 days lactation); line 3644 (mouse 9507, 13 days lactation; mouse 9287, 12 days lactation); line 3646 (mouse 5971, 14 days lactation; mouse 10695, 12 days lactation). C1, purified human sIgA from colostrum (415 000); msIgA, murine secretory IgA (MW 435 000).

Figure 4

Total IgA and IgG levels in the milk of pIgR transgenic mice. IgA (a) and IgG (b) levels were determined in milk samples of five mice per line during mid-lactation (12–15 days) by sandwich ELISA. Mean values are shown and the standard deviation is indicated. The P-value obtained via statistical analysis of the data (anova) is indicated. C, control mice; transgenic line numbers: 3642, 3643, 3646 and 3644.