Mutant Cullin 3 causes familial hyperkalemic hypertension via dominant effects
Mohammed Z. Ferdaus, … , Curt D. Sigmund, James A. McCormick
Mohammed Z. Ferdaus, … , Curt D. Sigmund, James A. McCormick
Published December 21, 2017
Citation Information: JCI Insight. 2017;2(24):e96700. https://doi.org/10.1172/jci.insight.96700.
View: Text | PDF
Research Article Nephrology

Mutant Cullin 3 causes familial hyperkalemic hypertension via dominant effects

Abstract

Mutations in the ubiquitin ligase scaffold protein Cullin 3 (CUL3) cause the disease familial hyperkalemic hypertension (FHHt). In the kidney, mutant CUL3 (CUL3-Δ9) increases abundance of With-No-Lysine [K] Kinase 4 (WNK4), with excessive activation of the downstream Sterile 20 (STE20)/SPS-1–related proline/alanine-rich kinase (SPAK) increasing phosphorylation of the Na+-Cl– cotransporter (NCC). CUL3-Δ9 promotes its own degradation via autoubiquitination, leading to the hypothesis that Cul3 haploinsufficiency causes FHHt. To directly test this, we generated Cul3 heterozygous mice (CUL3-Het), and Cul3 heterozygotes also expressing CUL3-Δ9 (CUL3-Het/Δ9), using an inducible renal epithelial–specific system. Endogenous CUL3 was reduced to 50% in both models, and consistent with autoubiquitination, CUL3-Δ9 protein was undetectable in CUL3-Het/Δ9 kidneys unless primary renal epithelia cells were cultured. Abundances of WNK4 and phosphorylated NCC did not differ between control and CUL3-Het mice, but they were elevated in CUL3-Het/Δ9 mice, which also displayed higher plasma [K+] and blood pressure. Abundance of phosphorylated Na+-K+-2Cl– cotransporter (NKCC2) was also increased, which may contribute to the severity of CUL3-Δ9–mediated FHHt. WNK4 and SPAK localized to puncta in NCC-positive segments but not in NKCC2-positive segments, suggesting differential effects of CUL3-Δ9. These results indicate that Cul3 haploinsufficiency does not cause FHHt, but dominant effects of CUL3-Δ9 are required.

Authors

Mohammed Z. Ferdaus, Lauren N. Miller, Larry N. Agbor, Turgay Saritas, Jeffrey D. Singer, Curt D. Sigmund, James A. McCormick

×

Figure 1

Generation of CUL3-Het and CUL3-Het/Δ9 mice.

Options: View larger image (or click on image) Download as PowerPoint
Generation of CUL3-Het and CUL3-Het/Δ9 mice. (A) The Pax8-rtTA-LC1 syste...
(A) The Pax8-rtTA-LC1 system (present in all mice) was used for doxycycline-induced Cre-mediated recombination at loxP sites specifically in renal epithelia. Cul3 heterozygous mice (CUL3-Het mice, left) carried 1 WT Cul3 allele and a Cul3 allele with exons 4–7 floxed. CUL3-Het/Δ9 mice (right) additionally carried a transgene fragment consisting of a strong synthetic promoter (CAG) promoter, a floxed transcription blocker (STOP), cDNA encoding Cul3 Δ403-459 (CUL3-Δ9), an internal ribosome entry site (IRES), and cDNA encoding the fluorophore tdTomato; the IRES permitted translation of CUL3-Δ9 and tdTomato separately from the same transcript. Doxycycline (Doxy, 2 mg/ml in 5% sucrose drinking water vehicle for 2 weeks) treatment induced recombination, disrupting 1 Cul3 allele in both CUL3-Het and CUL3-Het/Δ9 mice, and — in CUL3-Het/Δ9 mice only — initiating expression of CUL3-Δ9 and tdTomato. Control mice were genetically identical to either CUL3-Het or CUL3-Het/Δ9 mice and were given vehicle only. (B) To confirm recombination in CUL3-Het/Δ9 mice following Doxy treatment, PCR was performed on genomic DNA from several tissues. In control mice, a 1,300 bp product, consistent with no recombination, was detected in brain, heart, lung, liver, and kidney. In CUL3-Het/Δ9 mice, a 430 bp product derived from the recombined transgene was detected in kidney and liver, but not in brain, heart, or lung. Pax8-rtTA activity in liver has been reported previously (33). (C) tdTomato fluorescence was observed in renal cortex and medulla in CUL3-Het/Δ9 mice, but not in controls (or CUL3-Het mice; data not shown), representative of 8 independent experiments, and Western blot of whole kidney confirmed transgene activation following recombination. (D) Western blot of whole kidney lysate showed that both CUL3-Het (*P = 0.02, 2-tailed unpaired t test) and CUL3-Het/Δ9 (*P = 0.0001, 2-tailed unpaired t test) mice had significantly reduced WT Cullin 3 (CUL3) abundance compared with controls. For quantification, densitometric values were normalized using Coomassie stained gels (see Supplemental Figure 1 for details); values in parentheses represent n. (E) Direct comparison of CUL3-Het and CUL3-Het/Δ9 mice confirmed no difference in CUL3 abundance. (F) Western blot of cell lysates from primary renal epithelial cells infected with empty adenovirus (Ad-Null) or Cre adenovirus (Ad-Cre) confirmed Cre-mediated CUL3-Δ9 expression. Primary cells from uninduced CUL3-Het and CUL3-Het/Δ9 mice were then infected with virus for 48 hours. The short infection time was not sufficient to reduce endogenous CUL3 abundance. Lysates from HEK293 cells transfected with plasmids encoding the transgene (Δ9) and Cre confirmed the size of CUL3-Δ9 in primary cells.