SLC26A1 is a major determinant of sulfate homeostasis in humans

Abstract

Sulfate plays a pivotal role in numerous physiological processes in the human body, including bone and cartilage health. A role of the anion transporter SLC26A1 (Sat1) for sulfate reabsorption in the kidney is supported by the observation of hyposulfatemia and hypersulfaturia in Slc26a1-knockout mice. The impact of SLC26A1 on sulfate homeostasis in humans remains to be defined. By combining clinical genetics, functional expression assays, and population exome analysis, we identify SLC26A1 as a sulfate transporter in humans and experimentally validate several loss-of-function alleles. Whole-exome sequencing from a patient presenting with painful perichondritis, hyposulfatemia, and renal sulfate wasting revealed a homozygous mutation in SLC26A1, which has not been previously described to the best of our knowledge. Whole-exome data analysis of more than 5,000 individuals confirmed that rare, putatively damaging SCL26A1 variants were significantly associated with lower plasma sulfate at the population level. Functional expression assays confirmed a substantial reduction in sulfate transport for the SLC26A1 mutation of our patient, which we consider to be novel, as well as for the additional variants detected in the population study. In conclusion, combined evidence from 3 complementary approaches supports SLC26A1 activity as a major determinant of sulfate homeostasis in humans. In view of recent evidence linking sulfate homeostasis with back pain and intervertebral disc disorder, our study identifies SLC26A1 as a potential target for modulation of musculoskeletal health.

Keywords: Cartilage; Epithelial transport of ions and water; Genetic diseases; Genetics; Nephrology.

Figure 1. MRI demonstrated perichondritis at the level of the third and fourth ribs.

Axial fat-saturated T1-weighted images after administration of intravenous contrast agent (gadolinium): Linear contrast enhancement of the costal cartilage on both sides at the level of the third and fourth ribs (white arrowheads), predominantly at the costochondral junction, consistent with perichondritis. Note also the subtle diffuse enhancement of the adjacent soft tissue on the right side.

Figure 2. Clinical workup revealed a family history for consanguinity.

The parents of the index patient were first cousins. Genetic testing was only available in 4 family members (patient, parents, and 1 sibling). Filled symbols indicate a clinically affected individual (chondropathy). Symbols with a central dot represent an individual with no reported symptoms, but who was heterozygous for SLC26A1 mutation in genetic testing. E+ marks individuals in whom at least 1 mutated allele (c.824T>C) was detected.

Figure 3. Mapping of SLC26A1 variant on a homology model.

(A) Based on the structure of Prestin EMD-23334 (43), a SLC26A1 homology model was constructed by the Swiss model server. The SLC26 family transporters are dimers; each monomer is indicated in a different color. The fourth transmembrane domain is highlighted, illustrating the position of Leu²⁷⁵. STAS domain, sulfate transporter and anti–sigma factor antagonist domain. (B) Partial alignment of SLC26A1 among different species points to a high level of conservation.

Figure 4. Reduced sulfate and oxalate transport in Xenopus laevis oocytes expressing mutant SLC26A1 (p.Leu275Pro).

(A) SLC26A1-mediated SO₄²⁻ and (B) oxalate uptake by oocytes previously water injected or injected with 10 ng of cRNA encoding WT SLC26A1 or Leu275Pro mutant SLC26A1. Uptake was carried out from a bath solution containing 1 mM SO₄²⁻ or 1 mM oxalate for 15 minutes. The SO₄²⁻ uptake experiments were performed with oocytes from 4 different frogs (and 3 different cRNA preparations) with a total number of 35 WT, 30 Leu275Pro, and 33 water-injected oocytes, and for oxalate, 6 WT, 7 Leu275Pro, and 7 water-injected oocytes. Data are presented as mean ± SEM. ***P < 0.001 by 1-way ANOVA with Bonferroni’s multiple-comparison test.

Figure 5. Reduced cell surface expression of mutant SLC26A1 (p.Leu275Pro) in Xenopus laevis oocytes.

(A) Quantification of the mean gray value intensity of immunofluorescence in nonpermeabilized oocytes, carrying HA as a tag on the second extracellular loop of WT SLC26A1 [WT(HA)] or mutant SLC26A1 [Leu275Pro(HA)]. Oocytes injected with SLC26A1 without HA were used as a control (WT); 10 ng of the respective cRNAs was injected. (B) Representative oocyte immunofluorescence pictures. Scale bars: 100 μm. WT n = 15, WT(HA) n = 17, and Leu275Pro(HA) n = 16; oocytes were obtained from 3 different frogs (3 different cRNA preparations). (C and D) Quantification of SLC26A1 protein expression by immunoblots of membranes from oocytes expressing HA epitope–tagged (C) or untagged (D) WT or Leu275Pro mutant SLC26A1. An antibody against the HA tag was used in C, and an anti-SLC26A1 antibody (C-terminus) in D. Western blots include lanes from WT SLC26A1 or water-injected oocytes as specificity controls. For these experiments, 10–15 oocytes injected with 10 ng of the corresponding cRNA were pooled and actin was used as a loading control. Western blots were repeated at least 3 times with oocytes from 3 different frogs (3 different cRNA preparations were used). *P < 0.05; **P < 0.01 by 1-way ANOVA with Bonferroni’s multiple-comparison test (A) or unpaired, 2-tailed t test (C and D).

Figure 6. Carriers of putative damaging SLC26A1 variants in the GCKD study have lower median plasma sulfate levels than noncarriers.

Plasma sulfate levels (y axis) displayed by SLC26A1 rare variant carrier status (x axis). The y axis represents sulfate levels after inverse normal transformation, with units corresponding to standard deviations. *Denotes that Metabolon is highly confident in metabolite identity but a standard for this metabolite has not been run. The symbol color indicates observed rare variant carrier status, and symbol shape variant consequence (triangle, frameshift; circle, missense). The boxes range from the 25th to the 75th percentile of sulfate levels, the median is indicated by a line, and whiskers end at the last observed value within 1.5 × (interquartile range) away from the box. ***P < 0.001 (P value from aggregate variant test = 3.01 × 10^–5). Metabolon measurements yield semiquantitative rather than absolute metabolite levels.

Figure 7. Qualifying SLC26A1 variants in the GCKD study and their localization, frequency, consequence, effect size, and effect direction.

All qualifying rare, coding variants included in the aggregate variant test are plotted at the corresponding amino acid position of SLC26A1 (Uniprot ID Q9H2B4) on the x axis. The protein domains are based on Pfam 35.0. The y axis represents the minor allele count of each variant among the 4,708 GCKD study participants. The shape of the variant’s lollipop corresponds to its predicted consequence. The position of our patient’s mutation (Leu275Pro) is noted as well. The size of the variant’s lollipop represents the absolute value of the effect size of a single variant test with inverse normal transformed plasma sulfate levels (Supplemental Table 2), and the color indicates the direction of the effect size: the darker blue represents negative effect sizes and the lighter blue positive ones. All variants with an effect size of less than –1 or with a P value of less than 0.05 are labeled with their predicted amino acid exchange, where the ones with a P value of less than 0.05 are labeled in red. STAS domain, sulfate transporter and anti–sigma factor antagonist domain.

Figure 8. Functional evaluation of SLC26A1 variants identified in the GCKD study.

SLC26A1-mediated SO₄²⁻ uptake by oocytes previously water injected or injected with 10 ng of cRNA encoding WT SLC26A1 or the indicated mutants (Thr185Met, Pro237Leu, Leu348Pro, and Ser358Leu). Uptakes were carried out from a bath solution containing 1 mM SO₄²⁻ for 15 minutes. The experiments were performed with oocytes from 4 different frogs (and 3 different cRNA preparations) with a total number of 36 WT, 23 Thr185Met, 41 Pro237Leu, 41 Leu348Pro, 33 Ser358Leu, and 40 water injected. Data are presented as mean ± SEM. ***P < 0.001 by 1-way ANOVA with Bonferroni’s multiple-comparison test.