Fig 1: CUL3?403–459 displays increased auto-ubiquitylation and ubiquitylation of KLHL3A–D In vitro ubiquitylation assays with purified proteins. As ubiquitin is covalently attached to the substrate lysine residue, the appearance of higher molecular weight protein bands reflects the modification of the protein with ubiquitin. All assays contain purified UBE1, UBE2D3, ubiquitin, 0.1 mM ATP, 1 mM MgCl2 and were buffered in 50 mM HEPES, 150 mM NaCl and incubated at 30°C for the time indicated. Reactions were stopped by the addition of SDS–Laemmli buffer to a concentration of 1×. SDS–PAGE, staining with Coomassie blue, or detection with the indicated antibody following immunoblotting enabled the visualisation of ubiquitylation. (A) to determine the relative modification of KLHL3 by CUL3WT and CUL3?403–459, KLHL3 was included into the reaction at 2× molar concentration over CUL3WT and CUL3?403–459. (B, C) Reactions serve to determine basal auto-ubiquitylation of CUL3WT or CUL3?403–459 and do not contain KLHL3 or other potential substrate proteins. Lysine residues on CUL3WT or CUL3?403–459 act as the substrate. (B) High molecular weight bands reflect ubiquitin chain linkages on CUL3 or the multiple mono-ubiquitylation of a number of CUL3 lysine residues. (C) Activity assays contain methylated ubiquitin, a form of ubiquitin incapable of forming ubiquitin chains; as such, higher band shifts reflect the attachment of mono-ubiquitin to one more lysine residue on CUL3WT or CUL3?403–459, respectively. (D) Activity assay performed as in (C) with methyl-ubiquitin, the boxes shown on the gel are indicative of the gel pieces excised for mass spectrometry analysis in (F). E Schematic representation (to linear residue scale) highlighting the domains of CUL3 and schematic representation of the lysine residues of CUL3WT or CUL3?403–459, modified at 5 and 45 min, respectively, in the in vitro ubiquitylation assay shown in (D) and as determined by mass spectrometry. F Structural docking model of CUL3?403–459 based on the structure of full-length CUL1 (1LDK) using Chimera (see Materials and Methods). The NTD is coloured green, while the CTD is red, and the four panels sample possible positioning of the CTD relative to the NTD in CUL3?403–459, assuming full flexibility of the linker after deletion of three helices encoded by exon 9. Source data are available online for this figure.
Fig 2: CUL3 and KLHL3 are present in mouse and human aorta, and CUL3WT/?403–459 mice undergo aortic vascular remodellingWestern blot of tunica media-intima thoracic aorta lysates from mice culled after a minimum 4-h fast. Following exsanguination after surgery, mouse tissues were rapidly harvested and the tunica adventitia removed before storage, and later, the samples were homogenised, clarified and quantified prior to SDS–PAGE. Western blot analysis confirmed the expression of KLHL3 and CUL3. Similar to the kidney, the aorta of CUL3WT/?403–459 showed slightly lower levels of CUL3 compared to CUL3WT without any change in KLHL3 levels. Fresh frozen thoracic aorta tissues from donor cadavers were homogenised, clarified and quantified prior to SDS–PAGE. Western blot analysis confirmed the expression of KLHL3 and CUL3 in normal healthy human aorta. Representative maximum-intensity z projections of immunofluorescently stained thoracic aorta sections showing the distribution of CUL3 and KLHL3 between CUL3WT/?403–459 and CUL3WT mice at a minimum 4-h fasting baseline (n = 4 per genotype). CUL3 and KLHL3 localisation is comparable between genotypes. The highest levels were detected in the vascular smooth muscle cells and endothelium, with a minimal expression in the perivascular adipose tissue of the adventitia. Scale bar, 50 µm. Representative immunohistochemical staining of KLHL3 and CUL3 in human thoracic aorta sections (n = 6). Similar to the mouse staining seen in (C), vascular smooth muscle cells in the tunica media of the aortic wall are positive for KLHL3 and CUL3. Scale bar, 50 µm. Morphometric analysis of thoracic aortae reveals vascular remodelling in CUL3WT/?403–459 mice. There is an increase of ˜21% in the vessel wall intima-media thickness (demarcated by arrows) of CUL3WT/?403–459 compared to CUL3WT mice (***P = 0.0003). However, there is no change in the number of elastin laminae (P = 0.1458) and therefore no increase in the number of medial muscle layers between genotypes. Two-tailed unpaired Student’s t-test; data are mean ± SEM. Source data are available online for this figure.
Fig 3: CUL3WT/?403–459 mice recapitulate PHAII electrolyte imbalances due to over-activation of the renal WNK4-SPAK pathwayWestern blot of whole kidney lysates from mice culled after a minimum 4-h fast. Following exsanguination after surgery, mouse tissues were rapidly harvested and stored, samples were then homogenised, clarified and quantified prior to SDS–PAGE. Immunodetection with the antibodies shown highlights elevated signalling through the WNK kinase pathway in CUL3WT/?403–459 versus CUL3WT mice. The lowest panel is an anti-CUL3 immunoprecipitation of the kidney lysate samples, whereby the CUL3 antibody was cross-linked to Protein G sepharose and used to affinity purify CUL3WT and CUL3?403–459 from kidney lysates. The samples were incubated together overnight to allow deneddylation of CUL3 proteins to occur and maximise CUL3 binding to the anti-CUL3 resin. Samples were thoroughly washed and eluted from the resin prior to SDS–PAGE and immunodetection for CUL3. The IP highlights that CUL3?403–459 is indeed present within the kidney lysate. Fresh frozen kidney tissues from transplant patient and donor cadavers were homogenised, clarified and quantified prior to SDS–PAGE. Western blot analysis confirmed the expression of KLHL3 and CUL3 in normal healthy human kidneys. Plasma aldosterone after a minimum 4-h fast was calculated by HTRF (homogeneous time-resolved fluorescence) aldosterone assay. The average aldosterone level per mouse was calculated from duplicate samples run in parallel on the assay. Blood was rapidly harvested in heparin-coated plasma extraction tubes following exsanguination after surgery, and samples were snap-frozen for storage. A 58% increase in aldosterone was detected in CUL3WT/?403–459 versus CUL3WT mice (*P = 0.0245). Two-tailed unpaired Student’s t-test; data are mean ± SEM. Arterial blood biochemistries after a minimum 4-h fast. Under anaesthesia, the right carotid artery was cannulated to minimise atmospheric exposure of samples collected for iSTAT blood gas and electrolyte measurements. CUL3WT/?403–459 mice present with abnormal electrolyte homoeostasis compared to CUL3WT mice, exhibiting hyperkalaemia (***P = 0.0004) and hyperchloraemia (***P = 9.5 × 10-5) with a compensated metabolic acidosis (P = 0.7766), marked by a decrease in bicarbonate () (***P = 3.4 × 10-5), base excess (BE) (***P = 9.1 × 10-5) and partial pressure of carbon dioxide (pCO2) (***P = 0.0038). Two-tailed unpaired Student’s t-test; data are mean ± SEM. Source data are available online for this figure.
Fig 4: CUL3?403–459 is unable to ubiquitylate WNK1 or WNK4 kinases in an in vitro system, and this deficiency cannot be rescued by the presence of CUL3WTA–C In vitro ubiquitylation assays were performed as described in Figure 2, but with the addition of immunoprecipitated WNK1 in (A), or immunoprecipitated over-expressed FLAG-WNK4 in (B, C) (see Materials and Methods). The WNK kinases are modified by CUL3WT-KLHL3, with the higher molecular weight smear observed in anti-WNK1 and anti-FLAG panels representative of multiple ubiquitin molecules being covalently attached to the WNK protein. CUL3?403–459 is unable to modify WNKs. Samples from the same assay reactions were divided to allow immunodetection of the different protein components modified within the same assay reaction. (C) CUL3WT, CUL3?403–459 and an equimolar solution CUL3WT:CUL3?403–459 (1:1 Mix) were incubated with KLHL3 and immunoprecipitated FLAG-WNK4 in ubiquitylation reactions to determine the influence of CUL3?403–459 on the ubiquitylation activity of CUL3WT. Notably, the presence of CUL3?403–459 does not inhibit WNK ubiquitylation by CUL3WT.
Fig 5: WNK4 and SPAK specifically accumulate and form puncta in the distal convoluted tubule of CUL3WT/?403–459 miceRepresentative pseudocoloured maximum-intensity z projections of immunofluorescently stained kidney sections showing the distribution of the WNK/SPAK pathway components between CUL3WT/?403–459 and CUL3WT mice at a minimum 4-h fasting baseline (n = 4 per genotype). CUL3 and KLHL3 are comparable between genotypes, with significantly higher levels of CUL3 in the proximal convoluted tubule (PCT) compared to the weak staining in distal convoluted tubule (DCT) and thick ascending limb of the loop of Henle (TAL), while KLHL3 shows higher expression in the DCT/TAL cytosol with staining of the PCT confined to the apical membrane. Total (t) and phospho (p) NCC Thr44 and SPAK Thr243 show increased apical membrane expression in the parvalbumin (PVALB)-positive early and late (PVALB-negative) DCT of CUL3WT/?403–459 mice. Unexpectedly, the increased levels of WNK4 and SPAK resulted in the formation of discrete puncta specifically in the DCT of CUL3WT/?403–459 mice. It is possible that the autophagy–lysosomal system may attempt to compensate and degrade these excess proteins although the WNK4 puncta do not colocalise with the lysosomal marker LAMP1. Scale bar, 20 µm. Representative immunohistochemical staining of KLHL3 and CUL3 in human kidney sections (n = 6). KLHL3 shows preferential DCT/TAL cytosolic staining similar to the mouse despite no evidence of PCT apical staining as seen in (A). CUL3 exhibits preferential basolateral cytosolic staining of the PCT with similarly low levels of diffuse staining in the DCT/TAL to that of the mouse. Scale bar, 100 µm.
Supplier Page from Abcam for Anti-KLHL3 antibody - N-terminal