Mitochondrial organization in the developing proximal tubule is controlled by LRRK2

Abstract

The proximal tubule of the nephron performs energy-demanding functions such as resorption of water, amino acids and glucose. Formation of the energy-producing machinery is an essential step in proximal tubule epithelial cell differentiation, and this report asks how mitochondria are localized within these cells. We show that mitochondria move from the apical to basolateral side of the proximal tubule cell coincident with the initiation of lumen flow and that proximal tubules deficient in filtration maintain mitochondria in the apical position. Mitochondrial localization depends on the activity of LRRK2 and modeling fluid flow on cultured proximal tubule epithelial cells demonstrates that LRRK2 activity is regulated by fluid shear stress, explaining how onset of flow in the newly differentiated proximal tubule may trigger the apical-to-basolateral dissemination of mitochondria. These findings indicate that mitochondrial redistribution is one component of a cellular program in the nascent proximal tubule that drives function and that this process is triggered by flow.

Fig. 1. Mitochondrial mass and localization changes during PT differentiation.

A–H 18-20 week human fetal kidney stained with (A-F) PT markers (HNF4A, cyan; LTL, green) and mitochondrial marker (TOMM20, red). A Nephrogenic zone, cortex and medulla and location of nephrogenic zone-adjacent HNF4A + /LTL+ PTs (PT-early) with narrow lumens and deep cortical HNF4A + /LTL+ PTs (PT-diff) with distended lumens. B Schematic of nephron differentiation. C Renal vesicle. D S-shaped body with HNF4A+ presumptive PT. E PT-early with mainly apical mitochondria. F PT-diff with distended lumen and basolateral mitochondria. G PT-early and H PT-diff stained with mitochondrial marker citrate synthase (CS, red) and LTL (green). I–K P0 mouse kidney. I Nephrogenic zone, cortex and medulla and location of nephrogenic zone-adjacent HNF4A + /LTL+ PTs (PT-early) with narrow lumens and deep cortical HNF4A + /LTL+ PTs (PT-diff) with distended lumens. J PT-early with mainly apical mitochondria. K PT-diff with distended lumen and basolateral mitochondria. L Mean integrated density of TOMM20/cell in human fetal PTs (n = 16) with apical mitochondria versus PTs with basolateral mitochondria (n = 16). Results aggregated from 2 samples. Box shows first to third quartiles with mean (line), minimum–maximum values represented by whiskers and dots are individual datapoints; two-tailed Student’s t-test. M–R Transmission electron micrographs of P0 mouse kidney showing M–O PT-early with mainly apically located rounded mitochondria (blue arrows). P–R PT-diff with distended lumen and elongated basolateral mitochondria (blue arrow). S Apical to basolateral mitochondrial number in PT-early (n = 4 cross-sections) versus PT-diff (n = 4 cross-sections). T Length of apical and basolateral mitochondria in PT-early (n = 4 cross-sections) and PT-diff (n = 3 cross-sections). In S, T box shows first to third quartiles with mean (line), minimum–maximum values represented by whiskers and dots are individual datapoints; one way ANOVA with post-hoc Tukey HSD test. AP apical, BL basolateral, CD collecting duct, LU lumen, NPC nephron progenitor cell, N nucleus, NZ nephrogenic zone, pPT presumptive proximal tubule, PT proximal tubule, PTA pretubular aggregate, RV renal vesicle, SSB s-shaped body. Immunofluorescence counterstained with DAPI (blue). Scale bars in A & I 100 µm, other panels 5 µm. 1B adapted from with permission.

Fig. 2. Correlation between mitochondrial activity and mitochondrial localization.

A Schematic of approach for visualization of TMRM fluorescence in kidney following intravenous injection. B Representative CCCP-corrected heatmap from 8 individual measurements of red fluorescence in kidneys immersed in TMRM for 30 minutes. C TMRM staining in renal vesicles outlined with dashed lines. D TMRM staining in S-shaped body outlined with dashed line. E TMRM staining in nephron tubule. F Mitochondrial localization in renal vesicles outlined with dashed line G Mitochondrial localization in S-shaped body outlined with dashed line. H Newly formed cortical nephron tubule counterstained with LTL (green) and HNF4A (cyan) to identify PTs. AP apical, BL basolateral, CD collecting duct, NPC nephron progenitor cell, NZ nephrogenic zone, PT proximal tubule, RV renal vesicle, SSB s-shaped body. Immunofluorescence images F, G, H were counterstained with the nuclear dye DAPI (blue).

Fig. 3. Correlation between mitochondrial localization and onset of function in PTs.

A Approach for intravital labeling of functional PTs with Alexafluor-488-labeled 10 K dextran to distinguish between pre-functional and functional PTs. B Widefield image of dextran-labeled (green) P0 mouse kidney, stained with DAPI (nucleus, blue), TOMM20 (mitochondria, red), and HNF4A (PT, white). C, D Dextran-negative, pre-functional PTs. E, F Dextran-positive, functional PTs. G, H E15.5, I, J E14.5, K, L E13.5 kidneys from embryos intravenously injected with dextran (green) and Alexafluor-647-labeled IB4 (red). DAPI (nucleus, blue), TOMM20 (mitochondria, red), HNF4A (PT, white), dextran (green) labeling of M, N E15.5 tubules with apical dextran uptake; O, P E15.5 tubules without apical dextran uptake; Q, R E14.5 tubules with traces of apical dextran uptake; S, T E13.5 tubules without apical dextran uptake. AP apical, BL basolateral, Dx dextran, LU lumen, PT proximal tubule. All immunofluorescence images are counterstained with the nuclear stain DAPI (blue). Scale bar in B 100 µm, all other panels 5 µm.

Fig. 4. Mitochondria are apically localized in aglomerulur nephrons.

E18.5 kidneys from Six2cre;Numb^+/loxP;Numbl^+/loxp (control) and Six2cre;Numb^loxP/loxP;Numb^loxploxp (mutant) littermates. A, C Identification of glomeruli by staining for WT1 (podocyte marker, red) and IB4 (endothelial cell, green). B, D PTs labeled with HNF4A (white) and TOMM20 (mitochondria, red). E, F Quantification of glomeruli and PTs in triplicate sections of control kidneys vs triplicate sections of mutant kidneys (N = 3 individuals each). Data is shown as box-and-whisker, box shows data distribution from first to third quartiles with mean (middle line), minimum–maximum values are represented by whiskers and individual datapoints are shown as separate dots; two-tailed Student’s t-test. G–J Control and mutant kidneys stained with HNF4A (PT-white) and TOMM20 (mitochondria-red). Nephrogenic zone-adjacent PTs from control G and mutant I kidneys showing largely apical mitochondria. PTs from the deep cortex of control kidneys H showing largely basolateral mitochondria and from mutant kidneys J showing abundant apical mitochondria. K, L Quantification of PT localization in sections from control vs mutant (N = 3 individuals each) shown as box-and-whisker, box shows data distribution from first to third quartiles with mean (middle line), minimum–maximum values are represented by whiskers and individual datapoints are shown as separate dots; two-tailed Student’s t-test. PT proximal tubule, NZ nephrogenic zone. All sections were counterstained with the nuclear dye DAPI (blue). Scale bars in A–D 10 µm, and G–J 5 µm.

Fig. 5. Basolateral localization of mitochondria coincides with microtubule network establishment in PTs.

Kidney from P0 mouse pulsed with 10kD dextran-Alexa568 to distinguish between pre-functional, dextran negative PT-early vs functional dextran positive PT-diff. A-B’ α-tubulin (green) and HNF4A (white). C Comparison of number of junctions of α-tubulin in PT-early vs PT-diff. D-E’ Phalloidin (green) and LTL (white) (n = 4 cross-sections). Data are represented as bar graph, (mean value + /- SD). F Comparison of abundance of stress fibers in PT-early vs PT-diff. G-H’ NaK ATPase (green) and LTL (white) (n = 4 cross sections). Data are represented as bar graph, (mean value + /- SD). I Comparison of the abundance of NaK ATPase pump in PT-early vs PT-diff. J, K’ EdU (white) and LTL (PT) (n = 4 cross-sections). Data are represented as bar graph, (mean value + /- SD). For EdU labeling studies (n = 3 kidney), PT-early were classified as tubules with barely discernible LTL staining, while PT-diff were classified as tubules with strong LTL staining. L Comparison of proliferation in PT-early vs PT-diff. Data are represented as bar graph, (mean value + /- SD). M-N” Co-localization of α-tubulin (green), TOMM20 (red) and HNF4A (white) in dextran-negative M–M” versus dextran-positive N–N” tubules. O Schematic model for organelle redistribution in the PT-early to PT-diff transition that we propose is initiated by lumen flow. All bar graphs were tested with a two-tailed Student’s t-test. EdU 5-ethynyl deoxyuridine; PT proximal tubule. All sections were counterstained with the nuclear dye DAPI (blue). Scale bars in all panels 5 µm.

Fig. 6. Mitochondria in organoid PTs primarily localize apically.

A Kidney organoid differentiated from H9 hESCs at 16 days (9 + 7, 9 days differentiation in adherent cell culture followed by 7 days non-adherent culture of aggregated cells). HNF4A (PT, red); E-Cadherin (distal tubules, green); nephrin (podocytes, cyan). B Overlay of PT labeled with LTL (green) and the mitochondrial marker TOMM20 (red), and B’ TOMM20 alone. C Comparison of apical to basolateral TOMM20 in HNF4A+ early and PTs- diff from P0 mouse kidneys (n = 5) with HNF4A+ PTs from H9 organoids harvested after 9 + 7, 9 + 14 and 9 + 21 days of culture (n = 5 analyzed from each stage). Graph plotted is ratio of apical to basolateral mean fluorescence intensity of TOMM20 in PT. Data is shown as box-and-whisker, box shows data distribution from first to third quartiles with mean (middle line), minimum–maximum values are represented by whiskers and individual datapoints are shown as separate dots; one way ANOVA with post-hoc Tukey HSD test. D–F Fluorescent protein detection in organoids from AICS-0078-79 iPSCs that express a TOMM20-GFP fusion protein (green) counterstained with HNF4A (white). D, D’ Day 9 + 7, E, E’ Day 9 + 14, and F, F’ Day 9 + 21. (G-I) TOMM20 (green) and HNF4A (red) in PT-early of G P0 mouse PT-early, H HFK PT-early and I AICS-0078-79-derived kidney organoids on day 9 + 7 of culture. AP apical, DT distal tubule, pGlom presumptive glomerulus, PT proximal tubule. All immunofluorescence images are counterstained with the nuclear stain DAPI (blue). Scale bar in A 100 µM, all other panels 5 µM.

Fig. 7. Small molecule panel identifies LRRK2 as a mediator of apical localization of mitochondria in PTs.

A Ratio of apical to basolateral mitochondria in PTs of kidney organoids derived from AICS-0078-79 iPSCs following 4 hours compound treatment (n = 3 organoids), 3 PTs per organoid. Data are represented as bar graph, (mean value + /- SD). B-C’ D9 + 14 Kidney organoids derived from HNF4A-YFP iPSCs (YFP under HNF4A promoter, PTs (green)) immunostained for LRRK2 (red). D–F P0 mouse kidneys were stained for LTL (PT, green) and LRRK2 (red). D Widefield of P0 mouse kidney showing LRRK2 (red) expression restricted to PTs. E PT-early displaying cytoplasmic LRRK2. F PT-diff displaying cytoplasmic LRRK2. G–I Maximum intensity projections of PTs from AICS-0078-79 kidney organoids (D9 + 14) with eGFP tagged mitochondria (green) stained with HNF4A (PTs, white) following 4 hours treatment with G, G’ vehicle, H, H’ GSK2587215A, and I, I’ MLi-2. J Comparison of apical:basolateral ratio of mean fluorescence intensity of TOMM20/PT (n = 5 per condition). Data is shown as box-and-whisker, box shows data distribution from first to third quartiles with mean (middle line), minimum–maximum values are represented by whiskers and individual datapoints are shown as separate dots; one way ANOVA with post-hoc Tukey HSD test. AP apical, BL basolateral, T distal tubule, pGlom presumptive glomerulus, PT proximal tubule. All immunofluorescence images are counterstained with the nuclear stain DAPI (blue). Scale bar in D 100 µm, all other panels 5 µm.

Fig. 8. LRRK2 controls apical localization of mitochondria in the newborn kidney.

A Immunoblot of kidney lysates from P0 mice treated with vehicle (DMSO) or 30 mg/kg MLi-2 (MLi-2) blotted for active phosphorylated LRRK2 (pLRRK2, S935), total LRRK2 and loading control (β-tubulin). B Densitometric quantification of immunoblot plotted as bar graph; two-tailed, unpaired Student’s t-test. (Vehicle injected P0, n = 3 and MLi-2 injected P0, n = 5). Data are represented as bar graph, (mean value + /- SD). C, D’ PT-early from P0 mouse pulsed with 10 kD dextran-A647 stained with HNF4A (PT, green) and TOMM20 (mitochondria, red). C-C’ DMSO-treated mouse. D, D’ MLi-2 treated mouse. E Comparison of apical:basolateral ratio of mitochondria in PT-early from 3 kidneys per condition DMSO-treated (n = 4 PT/kidney) versus MLi-2-treated (n = 4 PT/kidney). Data is shown as box-and-, box-and-whisker, box shows data distribution from first to third quartiles with mean (middle line), minimum–maximum values are represented by whiskers and individual datapoints are shown as separate dots; two-tailed, paired Student’s t-test. F, G’ Representative wild type (WT) and Lrrk2 null P0 mouse littermate kidneys from analysis of 3/group pulsed with 10kD dextran-A647 and stained for HNF4A (PT, green) and TOMM20 (mitochondria, red). F, F’ PT-early from WT. G, G’ PT-early from Lrrk2 null. H Comparison of apical:basolateral ratio of mitochondria in PT-early in kidneys from 3 individuals per cohort of WT (n = 3PT/kidney) versus Lrrk2 null (n = 3PT/kidney). Data is shown as box-and-whisker, box-and-whisker, box shows data distribution from first to third quartiles with mean (middle line), minimum–maximum values are represented by whiskers and individual datapoints are shown as separate dots; two-tailed unpaired, Student’s t-test. I Immunoblot of hTert-RPTEC subjected to fluid shear stress (FSS) for 6 hours and blotted for phosphorylated LRRK2 (pLRRK2-S935), total LRRK2 and loading control (β-tubulin). J Densitometric quantification of immunoblot plotted as bar graph; two-tailed unpaired Student’s t-test. (n = 3 biological replicates of static and FSS cultures of hTert-RPTECs). AP apical, BL basolateral. PT proximal tubule. Immunofluorescence images C, E are counterstained with the nuclear stain DAPI (blue). Scale bars in all panels 5 µm.