The protective role of conjunctival goblet cell mucin sialylation

Abstract

Gel-forming mucins secreted by conjunctival goblet cells have been implicated in the clearance of allergens, pathogens, and debris. However, their roles remain incompletely understood. Here we show that human and mouse conjunctival goblet cell mucins have Alcian blue-detectable sialic acids, but not sulfates in the steady state. Interestingly, Balb/c mouse strain lacks this sialylation due to a point mutation in a sialyltransferase gene, St6galnac1, which is responsible for sialyl-Tn synthesis. Introduction of intact St6galnac1 to Balb/c restores the sialylation of conjunctival goblet cell mucus. Sialylated mucus efficiently captures and encapsulates the allergen particles in an impenetrable layer, leading to the protection of mice from the development of allergic conjunctivitis. Expression of ST6GALNAC1 and sialyl-Tn is upregulated in humans under conditions with chronic stimuli. These results indicate that the sialylated glycans on the ocular mucins play an essential role in maintaining the conjunctival mucosa by protecting from the incoming foreign bodies such as allergen particles.

Fig. 1. Conjunctival goblet cell mucus of humans and C57BL/6 J mice, but not that of Balb/c mice, is sialylated in the steady state.

a A schematic diagram of impression cytology. b Periodic acid–Schiff (PAS) and Alcian Blue (AB) (pH2.5) staining of the bulbar conjunctival goblet cells from healthy subjects with or without neuraminidase treatment. Bar, 20 μm. c High-iron diamine (HID) and AB staining of sulfomucin of the human conjunctival goblet cells. Staining control: mouse colon. The arrowhead indicates an example of HID-positive (brown) goblet cell. Bar, 20 μm. PAS and AB-PAS (d), or AB staining (e) of mouse conjunctiva. Bar 50 μm. Inset: Bar, 10 μm. Tissue sections were treated with neuraminidase (NA) or hyaluronidase (HA) before staining in (e). GC, goblet cells; MC, mast cells. f HID-AB staining of the mouse conjunctivas. Bar, 100 μm. Inset: Bar, 10 μm. g AB-PAS staining of membrane-transferred conjunctival swab proteins. Data are representative of at least two independent experiments (b–g). Source data are provided as a Source Data file.

Fig. 2. Aberrant splicing disrupts sialyl motif of St6galnac1 in Balb/c mice.

a Pooled total RNA from three eyes for each strain was subjected to microarray analysis. CMT-Sia, cytidine-5ʹ -monophospho-N-acetylneuraminic acid; ppGalNAcT, peptidyl-N-acetylgalactosaminyl-transferases; Gcnt, glucosaminyl (N-acetyl) transferase; β3gnt, β3-GlcNAc-transferase; galt, galactosyltransferase. b Schematic representation of St6galnac1 gene and its product. The catalytic domain has four conserved sialyl motifs (L, S, III, VS). Priming sites for PCR used in (c–e) are shown. TM, transmembrane domain. Colony PCR (c) and nested PCR (d) of cloned St6galnac1. Data are representative of two independent experiments. e Quantitative PCR for intact sialyl motif L. n = 6 for each strain. **p < 0.01 (p = 0.0022 for F3-R3 and p = 0.0022 for F4-R4) by two-tailed Mann–Whitney test. Data represents mean ± S.E.M. Data are representative of two independent experiments. f Schematic representation of splicing sites. g Strain comparison of 3ʹ splice site sequences available at Ensembl database. Blue, potential 3ʹ splice sites; Red, disabled 3ʹ splice site; Magenta, premature stop codon. The range of Sanger sequencing shown in (h) is circled with a dotted line. h Sanger sequencing of 3ʹ splice site of intron before exon 5. i Schematic representation of exon usages of cloned St6galnac1. Source data are provided as a Source Data file (c, d, e).

Fig. 3. Defective St6galnac1 activity in Balb/c goblet cells.

a Biosynthetic pathways of Tn, Core1, and sialyl-Tn. S/T, serine or threonine residue. b AB and Kernechtrot staining of differentiated HT29-MTX-E12 (E12) cell line. Bar, 20 μm. Polyclonal knockout (KO) of C1GALT1 or ST6GALNAC1 in E12 cells by genome editing affects sialyl-Tn expression assessed by FACS (c) and immunostaining (d). UEA-1, Ulex Europaeus Agglutinin I. Bar, 20 μm. e ST6GALNAC1-KO E12 cells were transduced with cloned St6galnac1 sequences. Expression levels of sialyl-Tn and GFP in transduced (blue histogram) and un-transduced (gray empty histogram) cells were evaluated by FACS. a.a, amino acids. f B6J and Balb/c conjunctivas were stained with a monoclonal antibody against sialyl-Tn (clone MLS132) with or without neuraminidase treatment. GC, goblet cells. Bar, 50 μm. g Swab extract of conjunctival sac (n = 2 for each condition) was subjected to neuraminidase or BSA (control) treatment before coating onto the plate. Sialyl-Tn was detected by anti-sialyl-Tn antibody. Source data are provided as a Source Data file. Data are representative of two independent experiments.

Fig. 4. Expression of intact St6galnac1 restores Sialyl-Tn expression in conjunctival goblet cell mucins in vivo.

a Schematic representation of backcrossing of B6J-derived intact St6galnac1 gene to Balb/c mice. b Chromosomal distribution of PCR-confirmed strain-specific markers in Ao mice. c Intact St6galnac1 expression in the conjunctiva. n = 7 for WT and n = 14 for Ao mice. ****p < 0.0001 by two-tailed Mann–Whitney test. Error bar represents S.E.M. d AB-PAS staining of the conjunctiva. Bar, 20 μm. e Immunostaining of sialyl-Tn in conjunctiva with or without neuraminidase treatment. WGA, wheat germ agglutinin Bar, 20 μm. f A schematic diagram of secreted mucus collection. Western blotting of the secreted mucus with (h, i) or without (g, i) neuraminidase treatment. The photographs of pollen shell amount left in the wells of polyacrylamide gels after the electrophoresis are also shown as loading controls (g–i). Data are representative of at least two independent experiments (d, e, g–i). Source data are provided as a Source Data file (c, g, h, i).

Fig. 5. Sialylated mucins encapsulate pollens with an impenetrable layer.

a Representative pictures of pollen shell-mucus aggregates 25 min after challenge. Bar, 1 mm. See also Supplementary Fig. 5 for more examples. b Immunostaining of pollen shell aggregates retrieved from the conjunctival sac. AF, autofluorescence. Bar, 50 μm. c The pollen shell aggregates from indicated strain were soaked in the fluorescent beads-suspended PBS. Parallel-view 3D stereogram of the side-view is also shown. Bar, 20 μm along the closest edge. d The bead distribution was enumerated at various focus levels. Bar, 20 μm. n = 6 for WT and n = 7 for Ao mice. *p < 0.05 (adjusted p = 0.0101, 0.0101, and 0.0233 for levels B, C, and D) by two-tailed Two-way ANOVA with Holm-Sidak’s multiple comparisons. Focus level A is excluded from the statistical analysis. n = 6 for WT and n = 7 for Ao mice. Error bar represents S.E.M. Source data are provided as a Source Data file. See also Supplementary Fig. 6. e The pollen shell aggregates were soaked in fluorescent beads-suspended PBS containing WGA. The detail of the white-squared area is also shown. The white dotted circle indicates a pollen shell. Bar, 50 μm along the closest edge; rep, representative pollen shell aggregate. Data are representative of at least two independent experiments (a–e).

Fig. 6. Sialylation is required for the efficient entrapment of pollen shells.

a Representative picture of pollen shell-mucus aggregates from one eye. Bar, 1 mm. b Frequencies of aggregates with indicated lengths. Pooled data from 8 eyes each. ***p < 0.001 (p = 0.0009) by two-tailed Fisher’s exact test for the ratio of fractions <1 mm and ≧1 mm. c The pollen shells were instilled into eyes with or without 5 U/mL neuraminidase. Bar, 1 mm. d Frequencies of aggregates with indicated lengths. Pooled data from 4 eyes each. *p < 0.05 (p = 0.0344) by two-tailed Fisher’s exact test for the ratio of fractions <1 mm and ≧1 mm. e The pollen shell amount in the aggregate was divided by the amount of mucus protein in the same aggregate. n = 7 for WT and n = 8 for Ao mice. ***p < 0.001 (p = 0.0004) by two-tailed Welch’s t test. Error bar represents S.E.M. Data are representative of at least three independent experiments (a–e). Source data are provided as a Source Data file (b, d, e).

Fig. 7. Sialyl-Tn or ST6GALNAC1 is upregulated in chronic stimulatory conditions in humans.

a Representative immunostaining of impression cytology specimens. GC, goblet cells. Bar, 20 μm. b Frequencies of sialyl-Tn-positive goblet cells in impression cytology specimens from control (n = 6) and AKC (n = 7) subjects. *p < 0.05 (p = 0.0443) by two-tailed Mann–Whitney test. Error bar represents S.E.M. c ST6GALNAC1 expression in an RNA-seq dataset (GSE155776) of pterygium (n = 8) and control (n = 8) biopsies. *p < 0.05 (p = 0.0103) by two-tailed Mann–Whitney test. Error bar represents S.E.M. Source data are provided as a Source Data file (a, b).

Fig. 8. St6galnac1 protects mice against allergic conjunctivitis.

a Clinical appearance of PBS- or ragweed pollen (RW)-challenged eyes. b Clinical scores. n = 2, 1, 7, and 7 for PBS-WT, PBS-Ao, RW-WT, and RW-Ao conditions, respectively (b, c, f). **p < 0.01 (p = 0.0093) by two-tailed Mann–Whitney test. c Scratch bouts. *p < 0.05 (p = 0.0449) by two-tailed Mann–Whitney test. d Giemsa staining of the conjunctiva. Arrowheads indicate eosinophils. Bar, 50 μm. Inset: Bar, 10 μm. e Eosinophil frequency among CD45 + cells in the conjunctiva. Eosinophil frequencies in live single cells after challenges of 100 mg/mL (f) or indicated amount (mg/mL) (g) of RW pollen or PBS. *p < 0.05 (p = 0.0379) by two-tailed Mann–Whitney test (f). **p < 0.01 (adjusted p = 0.0091, WT vs Ao in RW 50 mg/mL condition), ***p < 0.001 (adjusted p = 0.0006, WT vs Ao in RW 100 mg/mL condition) by two-tailed ANOVA with Holm-Sidak’s multiple comparisons test (g). h RW scores represent the amount of remaining pollen in the conjunctival sac 30 min after pollen challenges. n = 14 for each strain. *p < 0.05 (p = 0.0325) by two-tailed Mann–Whitney test. i, j Accumulation of migratory dendritic cells (CD11c⁺MHC class II high) to the draining lymph nodes 12 h after instillation of pollen shells and RW extract labeled with AF647. FACS gating for pooled data (i) and enumeration (j) are shown. N, PBS-instilled; S/RWe, pollen shell and AF647-labeled RW extract. PBS-instilled data were used for setting AF647 gate and therefore excluded from statistical analysis. n = 4, 7, 4, and 8 for WT-N, WT-S/RWe, Ao-N, and Ao-S/RWe conditions, respectively. **p < 0.01 (p = 0.0037) by two-tailed Mann–Whitney test. Data are representative of four (a–h) or three (i, j) independent experiments. Error bar represents S.E.M. (b, c, f, g, h, j) Source data are provided as a Source Data file (b, c, f, g, h, j).