Abstract
Background
The renin-angiotensin system (RAS) is highly conserved across vertebrates, including zebrafish, which possess orthologous genes coding for RAS proteins, and specialised mural cells of the kidney arterioles, capable of synthesising and secreting renin.
Methods
We generated zebrafish with CRISPR-Cas9-targeted knockout of renin (ren -/-) to investigate renin function in a low blood pressure environment. We used single cell (10X) RNA sequencing analysis to compare the transcriptome profiles of renin lineage cells from mesonephric kidneys of ren -/- with ren +/+ zebrafish, and with the metanephric kidneys of Ren1c -/- and Ren1c +/+ mice,
Results
The ren -/- larvae exhibited delays in larval growth, glomerular fusion and appearance of a swim bladder, but were viable and withstood low salinity during early larval stages. Optogenetic ablation of renin-expressing cells, located at the anterior mesenteric artery of 3-day-old larvae, caused a loss of tone, due to diminished contractility.
The ren -/- mesonephric kidney exhibited vacuolated cells in the proximal tubule, which were also observed in Ren1c -/- mouse kidney. Fluorescent reporters for renin and smooth muscle actin (tg(ren:LifeAct-RFP; acta2:EGFP)), revealed a dramatic recruitment of renin lineage cells along the renal vasculature of adult ren -/- fish, suggesting a continued requirement for renin, in the absence of detectable angiotensin metabolites, as seen in the Ren1YFP Ren1c -/- mouse. Both phenotypes were rescued by alleles lacking the potential for glycosylation at exon 2, suggesting that glycosylation is not essential for normal physiological function.
Conclusions
Phenotypic similarities and transcriptional variations between mouse and zebrafish renin knockouts suggests evolution of renin cell function with terrestrial survival.
Keywords: zebrafish RAS, renin knockout, renin assay, renin-cell ablation, scRNAseq analyses
Introduction
The renin-angiotensin system (RAS) is responsible for blood pressure, sodium homeostasis and water regulation. The rate limiting enzyme renin is predominantly expressed and synthesised in specialised renal, perivascular, mural cells 1-3 , or renal lineage cells (RLCs), which are involved in renal development 4 and have been implicated in angiogenesis of the renal vasculature 5-7 . Mice lacking renin require neonatal subcutaneous saline injections to survive, show recruitment of RLCs, have no detectable angiotensinogen metabolites (AngI or AngII) and show severe defects in renal development 8-10 .
Studies on the RAS have predominantly focused on mammalian systems. However, the optically clear zebrafish larva contains multiple genes orthologous to the human RAS genes - angiotensin converting enzyme (ace, ace2), angiotensinogen (agt), angiotensin receptors (agtr1a, agtr1b, agtr2), renin receptor (atp6ap2), mineralocorticoid receptor (nr3c2), and renin (ren) - lacking only an orthologue to the Mas receptor 11 . Renin has been linked to teleost survival in water of fluctuating osmolarity 12 . Placing zebrafish larvae in dilute (1/20) conditioned water (CW; 60mg/l sea salts), increases renin expression and circulating AngII levels 13-15 , analogous to low salt diet administration in mammals. Despite a functional RAS, zebrafish larvae are unable to survive reduced salinity in the presence of the Ace inhibitor, Captopril 16 . As in mammals, RLCs in the adult zebrafish mesonephros are present along afferent arterioles in a pre-glomerular position, express smooth muscle and pericyte markers, acta2 and pdgfrb, respectively, and are recruited under low salinity or Captopril challenge 17-19 .
Electron microscopy of perivascular RLCs or labelling of acidic organelles with fluorescent dyes, have revealed that the zebrafish RLCs contain dense core, acidic granules, suggestive of renin synthesis, storage, and processing 17 . In mammals, it is thought that glycosylation sites are involved in the processing of renin and sequences encoding these highly conserved glycosylation sites are found in exons 2 and 4 of the zebrafish renin gene. However, the functional role of glycosylation in zebrafish renin remain to be elucidated.
The zebrafish ren transcript is first expressed in larvae at 24hpf and has been implicated in the development of the pronephros, which is an active filtration organ from 3dpf (days post fertilization) 16, 20 . At the larval stage, renin expression is largely limited to the anterior mesenteric artery, which supplies the swim bladder. This buoyancy aid typically becomes visible from 4 to 5dpf, when the zebrafish larvae reach 3.4-3.7mm in length 21 , unless they are exposed to low salinity 13 .
Zebrafish kidney development terminates with the mesonephros and despite lacking the structural complexity of a metanephric kidney, nephron structure and tubular segmentation is highly conserved and easily accessible 22 . Furthermore, the zebrafish mesonephric kidney retains the ability to regenerate and restore damaged nephrons, making it an ideal model for studying renal injury and repair 23, 24 .
In this study we generated an allelic series of ren mutations using CRISPR/Cas9-targeting in zebrafish. Crossing of the ren -/- zebrafish to a double fluorescent reporter strain (tg(ren:LifeAct-RFP; acta2:EGFP)), enabled high resolution imaging of the mesonephric renal vasculature, and the generation of informative scRNAseq 10X datasets. The transcriptional profiles of RLCs from ren -/- and ren +/+ mesonephros were compared with those of equivalent mouse metanephric RLCs.
Methods
The authors declare that all supporting data are available within the article [and its online supplementary files]. Zebrafish scRNAseq libraries have been submitted to ArrayExpress (E-MTAB-11079). Mouse scRNAseq data have been submitted to GEO public repository (GSE180873); [NCBI tracking system #22225156].
Detailed methods are available online in Data Supplement All zebrafish (Danio rerio) 25 experiments were approved by the local ethics committee and conducted in accordance with the Animals (Scientific Procedures) Act 1986 in a United Kingdom Home Office approved establishment.
Statistical analysis
Statistical analyses were performed (GraphPad Prism version 8.4.3 for Mac; GraphPad Software, San Diego, California USA), by ANOVA and post hoc Sidak’s multiple comparisons test. Values are reported as means ±SD and P<0.05 was considered statistically significant.
Results
Viability of ren-/- larvae and the pronephric kidney
CRISPR-Cas9 targeting of G0 embryos produced chimaeric founders, with a range of indels at the target site. These were resolved by backcrossing to the wild-type strain, WIK, and subsequent genotyping of F1 fish. Indels were verified by sequencing of PCR products 26 spanning the target site.
In one fish (ren -/-), removal of 8bp near the glycosylation site in exon 2 was predicted to cause a non-sense mutation, bringing a stop codon into frame in the third exon, and thus truncating the protein product (Supplementary Fig. S1A). F2 fish showed a Mendelian ratio of 20:38:15 (WT: Het: Hom) indicating that homozygous knockout fish were viable in water of normal salinity (CW) and suggesting that an active RAS is not vital during early stages of development (Fig. S2F).
A second knockout line, with a 9bp deletion spanning the glycosylation site was also generated (ren Δ9/Δ9), together with a knock-in line (ren KI/KI), generated using an antisense ss-oligo (Supplementary Fig. S1B-C). One out of 40 fish screened carried the knock-in allele, which replaced amino acids encoding the glycosylation site, rendering it disabled. Fish homozygous for both the 9bp deletion (ren Δ9/Δ9) and the 9bp substitution (renKI/KI ) were viable.
Though viability was unaffected by loss of renin expression, a significant delay in somatic growth (length) and appearance of the swim bladder in ren -/- zebrafish, was observed using the Vertebrate Automated Screening Technology (VAST) system (Fig. S2A). Zebrafish length was significantly reduced at 4dpf (ren -/- - 3.58 +/- 0.03 mm (n=45); ren +/+ - 3.79 +/- 0.02 mm (n=29); p<0.0001), and 5dpf (ren -/- - 3.76 +/- 0.03 mm (n=37); ren +/+ - 3.89 +/- 0.03 mm (n=29); p=0.004; Fig. S2B). At 5dpf, ren -/- zebrafish had a dramatically reduced swim bladder size (0.055 +/- 0.006 mm 2 , n=80) and in 42.5% cases lacked a swim bladder completely, compared to ren +/+ zebrafish, (0.109 +/- 0.002 mm 2 ; n=61) which all had a swim bladder. By 8dpf, there was still a significant difference in swim bladder area but not length of ren -/- larvae compared to controls. Despite this, no obvious behavioural difference was noted.
We investigated pronephric development in ren +/+ and ren -/- zebrafish crossed to the transgenic reporter line tg(wt1b:EGFP), where Wt1b is localised to the proximal tubule and the glomerulus 27 . The pronephros of ren+/+ and ren-/- tg(wt1b:EGFP) zebrafish were imaged at 3-,4- and 5dpf, and the anterior glomerular distance (AGD) and posterior glomerular distance (PGD) were measured (Fig. S2C). AGD and PGD were significantly increased in ren-/- tg(wt1b:EGFP) at all three stages of development suggesting a delay in glomerular fusion in ren -/- larvae (Fig. S2D-E). No significant difference in viability was observed when exposing ren -/- zebrafish to either 1/20CW (3mg/L sea salt), or Captopril in CW compared to ren +/+ treated zebrafish. However, no ren -/- zebrafish survived in 1/20CW combined with Captopril, compared to 12.5% survival of ren +/+ zebrafish (Fig. S2F).
Larval RLC Ablation
In larvae, renin expression is limited to the anterior mesenteric artery (AMA), which supplies the swim bladder 16 . Despite the delay in development, no significant difference was observed in the area of AMA fluorescence in the ren -/-; Tg(ren:LifeAct-RFP;acta2:EGFP) line compared to ren +/+;Tg(ren:LifeAct-RFP;acta2:EGFP) 16, 28 at 5dpf, (Fig. 1A-C; n=5 per group), suggesting that renin expression is not essential in conditioned water.
Figure 1. AMA analysis.
A) Quantification of the LifeActRFP fluorescent signal at the AMA in Tg(ren:LifeAct-RFP;acta2:EGFP) 5dpf larvae on a ren +/+ or ren -/- background (n=5 per group); B) SPIM microscopy of pronephric AMA showing LifeActRFP (Ren-expressing cells) and EGFP (Acta2-expressing cells) fluorescence and C) representative greyscale image for mean area analysis. Laser ablation of renin-expressing cells in the AMA of Tg(ren:mem-KillerRed) 3dpf larvae showing kdrl:GFP signals at D) t = 0min and E) t = 60mins of the ablation protocol. The dorsal aorta (DA) can be seen with the anterior mesenteric artery (AMA) budding off it. Images represent single axial planes. Scale bars represent 30μm; AMA diameter was measured at the same two locations for each fish (dotted lines); F) An Estimation plot and mean of differences are shown; (Paired t-test*: p = 0.0147)
To determine whether or not RLCs in the AMA exhibit contractile ‘pericyte’ functions, we specifically ablated the RLCs in 3dpf Tg(ren:KillerRed) larvae, using a Bessel beam on the SPIM microscope as previously described 29 . We recorded time-lapse images and measured the maximum luminal width of the AMA blood vessel, accounting for its pulsatility with each heartbeat. We demonstrated that the width significantly increased (from 19.208+/-1.307μm to 21.678+/-0.615μm; n=5; p=0.01) following ablation of the RLCs (Fig. 1D-F and supplementary movie S1 30 ).
The ren-/- Mesonephric Kidney: Renin assay
AngI and AngII metabolites were predicted by comparison of zebrafish angiotensinogen protein sequence with that of higher mammals (Fig. S1E), synthesized and used as standards for the indirect kidney renin assay. AngI and AngII were measured in pooled mesonephric kidney samples isolated from ren +/+, ren-/-, ren Δ9/Δ9 and renKI/KI fish. Results are given in Table 1. Almost no angiotensin metabolites were detected in the ren -/- kidneys, confirming that renin activity is absent in these fish. Neither disabling (renKI/KI ) nor removal (ren Δ9/Δ9) of the glycosylation site in exon 2 affected the levels of AngI and AngII in their respective mesonephric kidneys.
Table 1. Angiotensin metabolites in mesonephric kidneys isolated from zebrafish homozygous for mutated renin alleles.
| Genotype | Ang 1-8 [pg/g] | Ang 1-10 [pg/g] |
|---|---|---|
| ren +/+ | 1159.0 | 6790.1 |
| ren -/- | <5 | <5 |
| renΔ9/Δ9 | 966.1 | 5496.0 |
| renKI/KI | 1626.9 | 8844.9 |
Histology
Periodic Acid Schiff (PAS) stain 31 was used to visualise renal structures in adult zebrafish mesonephros (Fig. S3A-B). Kidneys of ren -/- zebrafish exhibited widespread, severe vacuolation of the proximal tubular epithelial cells compared to very rare, mild vacuolation in ren+/+ controls (n=3 per group). Absence of PAS staining suggests that the vacuoles do not contain polysaccharides or glycoproteins. No differences were observed in the distal tubule, which lacks a brush border. Vacuolation of proximal tubular epithelial cells was also observed in the Ren1cYFP Ren1c -/- mouse kidney, compared to the Ren1c +/+ control (Fig. S3C-D).
Spinning disc confocal microscopy 32 of excised mesonephric kidneys isolated from ren+/+ and ren” tg(ren:LifeAct-RFP, acta2:EGFP) fish revealed that RLCs are located intermittently along the afferent arterioles in ren +/+ fish giving a banded or striped pattern (Fig. 2A). However, RFP labelling in ren -/- mesonephric kidneys showed extensive continuous expression and the virtual absence of stripedlbanded patterning, whilst EGFP fluorescence, which marked smooth muscle cells, was significantly reduced (Fig. 2B). This mirrors the RLC recruitment seen in mice 33
Figure 2.
Spinning disc microscopy showing the expression of fluorescent reporters ren:LifeAct-RFP and acta2:EGFP in mesonephric kidney squashes from A) ren +/+, B) ren -/- or the obligate heterozygous fish C) ren -/Δ9, and D) ren -/KI. Panels show separate red, green and merged channels. Scale bar = 50μm.
Renin Complementation
To determine whether the ren -/- knockout phenotype could be rescued, either by the deletion mutant ren Δ9/Δ9 or the knock-in mutant, renKI/KI , each was crossed with ren -/- Tg(ren:LifeAct-RFP; acta2:EGFP) to generate obligate heterozygotes: ren -/Δ9; Tg(ren:LifeAct-RFP +/-; acta2:EGFP +/-) and ren -/KI ;(ren:LifeAct-RFP+/-; acta2:EGFP+/- ) respectively. Kidney squashes from these were assessed for complementation of the banding pattern in afferent arterioles (Fig. 2C-D). Both substitution and removal of the glycosylation site rescued the knock-out phenotype.
scRNAseq analysis
10X scRNAseq libraries were made from ren:LifeAct-RFP+- and acta2:EGFP+-expressing cells, FAC sorted from ren +/+ and ren -/- double-transgenic fish. Dimensionality reduction (t-distributed Stochastic Neighbour Embedding; tSNE) 34 was performed (using resolution 0.1 to set the granularity of the clustering) on the merged libraries following principal component analysis (Fig. S4) and similar cells were grouped into five clusters (Fig. S4A). Violin plots and feature plots were used to explore the transcription of key genes of interest (Fig. S4B-C). Cluster 2 contained the majority of renin- and LifeActRFP- expressing cells while clusters 0, 1 and 2 contained acta2- and EGFP-expressing cells. RLCs from the control library (ZF1) represented 11.9% of the FAC sorted cells, whilst 36.7% of cells from the RenKO library (ZF2) resided in this cluster. This probably reflects the proportion of juxtaglomerular cells from ZF1 versus the recruited RLCs from ZF2. Lists of genes differentially expressed in each cluster were generated, and several cluster 2-specific transcription factors were identified (Fig. S4C) including cnot4b, twist1a, nr2f5, nkx3.2, cited4a, and sox6. Other transcripts enriched in RLCs included signalling proteins such as angptl3 and rgs5b, and also hmox1a and tfpia.
Pseudotime analysis 35 was projected onto the seurat-generated clusters and confirmed the relationship between smooth muscle cells and RLCs (Fig. S4D). Subsequently, cluster 2 was separated into sub-clusters, to identify genes that distinguish JG-cells from recruited cells (Fig. S5). Using a resolution of 0.5, three clusters were resolved (Fig. S5A). The distribution of cells from the two libraries indicated that cluster C2 contains cells from the control library (ZF1) and therefore represents JG cells, while cells from the ren -/- library (ZF2), representing recruited RLCs, could be divided into clusters C0 and C1. Genes differentiating the sub-clusters were interrogated further (Table 2). Transcription of ren was 4-fold higher in JG cells (C2) than RLCs, while transcription factors cnot4b and twist1a were up-regulated in cluster 1 and rbp4 was more highly expressed in C0 of the recruited cells (Fig. S5B-C). The top genes listed for each cluster were assessed by gene ontology analysis (http://geneontology.org). C0 showed a 16-fold enrichment for transcripts related to actomyosin structural organisation (including tagln, acta2, lmod1b, csrp1a), C1 showed enrichment for the vegf pathway (vegfaa, pgfb and gng2) and lysosomal and lytic vacuoles, but no significant enrichment was seen in C2 (data not shown). Pseudotime mapping suggested that cluster 0 may represent a ‘transition’ stage between JG and RLCs.
Table 2. Factors involved in transcription control, signalling pathways, actin filament organisation and ion transport, which are differentially expressed between JGs and RLCs (separated by subcluster analysis).
| Cluster 0 (RLC) | Cluster 1 (RLC) | Cluster 2 (JG) |
|---|---|---|
| rbp4 | twistla | cited4a |
| hmgxb4a | cnot4b | nkx3.2 |
| nr2f5 | idl | |
| acta2 | cremb | |
| EGFP | fosl2 | ren |
| myl9a | vegfaa | diol |
| tagln | angptl3 | tnmd |
| tpm1 | rgs5b | tnfaip8l3 |
| pfn2 | rgll | tnfsfl2 |
| tfpia | Pgfb | rergla |
| mmp2 | gng2 | rasll2 |
| lmod1b | nprlb | nrarpa |
| csrp1a | gprI37ba | kctd12.2 |
| tpm4b | alkali | |
| sgkl | ||
| igfbp5b | ||
| cdl64 | ||
| socs3b | ||
| ackr4b | ||
| adora2aa | ||
| cnpyl | ||
| psap | ||
| ptn | ||
| nocta | ||
| agtrap | ||
| kcne4 | ||
| clcn5b | ||
| ednraa |
The control and knockout libraries were further analysed using canonical correlation analysis (Fig. S6; resolution 0.2 gave eight clusters, cluster 1 containing most RLCs). This confirmed multiple genes with altered transcription in response to renin knockout, including transcription factors cnot4b, and nr2f5, and signalling factors such as angptl3, rgl1, nedd9, and npr1b. Gene Ontology analysis suggested an 8.6-fold enrichment in markers of lysosomes or lytic vacuoles in response to renin knockout.
Comparison of RLCs from zebrafish and mouse
scRNAseq expression matrices from RLCs of the ren +/+ and ren -/- zebrafish libraries were compared with expression matrices generated using C1 Fluidigm methodology 36-39 on Ren1cYFP FAC sorted cells isolated from Ren1c +/+ and Ren1c -/- mice respectively 40, 41 , using CCA analyses 34 . Cluster analysis of the two wildtype libraries gave two clusters at a resolution of 0.5 (Fig. 3A). Since cluster 0 included the majority of mouse and zebrafish cells, these were designated as JG like. By analogy, at a resolution of 0.6 following CCA analysis of the mouse and zebrafish renin knockout libraries, cluster 0 was designated as recruited RLCs (Fig. 3B). Dot plots show a number of transcripts, irrespective of genotype, which were mouse RLC-specific, including a number of integrins. The mouse Ren1c +/+ and Ren1c -/- libraries both showed significant enrichment of gene transcripts related to thyroid receptor binding (5.7-fold), regulation of focal adhesion assembly (5.7-fold) and integrin binding (4.0-fold), as assessed by Gene Ontology analysis, compared to the respective zebrafish libraries. There was a 4.3-fold enrichment of gene transcripts associated with positive regulation of blood pressure in the mouse libraries, while both zebrafish libraries showed a 4-fold enrichment of gene transcripts associated with sprouting angiogenesis (Table 3).
Figure 3.
Merged expression matrices from A) wild-type mouse and zebrafish JG cells and B) renin knockout mouse and zebrafish renin lineage cells following CCA analysis (resolution 0.5 and 0.6 respectively). Associated tables give contributing cell numbers from respective matrices; C) and D) genes differentially expressed between respective mouse and zebrafish libraries, irrespective of genotype, are shown in dot plots.
Table 3. Gene Ontology analyses giving fold enrichment of differentially expressed gene transcripts between wild type or renin knockout mouse and zebrafish libraries. (Genes in bold are only transcribed in mouse or zebrafish respectively).
| GO term | Mus Ren1 +/+ | ZF ren +/+ | Mus Ren1 -/- | ZF ren -/- | Associated genes | |
|---|---|---|---|---|---|---|
| Thyroid receptor binding | 6.38 | - | 5.07 | - | Trip12, Arid5a , Thrap3, Tacc1, Brd8 | |
| Integrin binding | 3.78 | - | 4.29 | - | Utrn , Cd81, Cd9, Emp2, Dmd, Lamb2 , Mfge8 , Nisch, Gsk3b, S1pr3 | |
| Integrin complex | 3.74 | - | 4.6 | - | Itgb1 , Itgb5, Itgav, 1tga7 | |
| Regulation of focal adhesion assembly | 5.59 | - | 5.8 | - | Rac1, Nrp1, Vegfa, Iqgap1, Limch1, Clasp2, Pten, Lrp1, Actg1, Ptk2, Dusp3, Macf1, S100a10, Map4k4 | |
| +ve regulation of blood pressure | 4.3 | - | 4.33 | - | Glpl1 , Wnk1, Agtr1a, id2, Nr2f2, Atp5j, Rarres2 | |
| Actin cap | 22.3 | - | 15.23 | - | Cald1 , Gsn, Actr2 | |
| Actin filament depolymerisation | 9.93 | - | 8.45 | - | Wdr1 , Gsn, Mical3, Dstn | |
| Actin filament polymerisation | - | 5.5 | - | 5.8 | capza1b, pfn2, pfn1, pfn2l, tmsb4x, Imod1b | |
| relaxation of smooth muscle | 16.55 | - | 9.02 | - | Rgs2, Prkg1, Slc8a1 , Gucy1a1, Mrvi1 | |
| Sprouting angiogenesis | - | 4.16 | - | 4.82 | pgfb, pkma, cxcl12b, vegfaa, rtn4a, atf4b, crema,b, lgals2a, twist1a |
Discussion
Using the CRISPR/Cas9 system we targeted zebrafish ren exon 2 and generated multiple allelic variants. The survival of ren -/- fish might reflect their aquatic environment, since homozygous Ren1 c-/- mice require daily administration of saline solution in order to prevent neonatal death 9 . This suggests that transition to land created additional homeostatic challenges, possibly leading to the development of the macula densa in the metanephric kidney and baroreceptors.
Although ren-/- larvae were completely viable, the delay in growth, late appearance of the swim bladder, and delayed glomerular fusion most likely resulted from an adverse effect on salt handling. Low salt in combination with Captopril resulted in complete loss of viability of ren -/- larvae (compared to reduced survival of wild-type larvae in low salt and 0.1mM Captopril 16 ). Renin is increased in ion-poor fresh water 13 , implicating renin functionality in salt absorption, but zebrafish larvae rely predominantly on the five types of ionocytes in their gills and skin, which provide alternative routes for establishing salt homeostasis, until the kidneys are fully developed 42 . The Na+Cl- co-transporter-rich NCC ionocytes are thought to be responsive to AngII 43 , though this does not explain the loss of viability of ren -/- larvae (which presumably lack AngII), in low salinity plus Captopril.
In larvae, renin expression is limited to the AMA region, which supplies the swim bladder. We saw no increase in renin reporter expression in the AMA of ren -/- larvae grown in conditioned water. However, optogenetic ablation of the RLCs, caused an increase in arterial diameter, suggesting loss of tone. The observed increase in lumen diameter would correspond to approximately 22% increase in lumen cross-sectional area. To our knowledge, this is the first direct demonstration of contractile functionality in RLCs, in vivo, though it does not suggest that renin or AngII are involved per se, rather that the cells have dual functionality. This should be explored in the mesonephric and metanephric kidney.
To interrogate renin functionality in our adult mutants, our collaborators (Attoquant Diagnostic GmbH, Vienna, Austria) developed an indirect renin assay and demonstrated a complete absence of AngI and AngII in the mesonephric kidneys of adult ren -/- zebrafish. Normal levels of the metabolites were found in mesonephric kidneys from ren Δ9/Δ9 and renKI/KI confirming that neither removal nor alteration of this glycosylation site dramatically affects protein folding or activity of renin. The glycosylation sites are located towards the surface of the renin protein, which may explain the apparent flexibility in protein conformation.
Loss of a functional ren gene resulted in alterations to mesonephric kidney morphology – specifically cytoplasmic vacuolation of proximal tubular epithelium, but not renal degeneration 44 . Vacuolation of proximal tubule cells often indicates an osmotic imbalance due to increased solute transport across the cells 45 . However, an almost complete absence of detectable AngII may have adverse effects on the proximal tubule where it normally stimulates sodium and water reabsorption 46 . Since vacuolation of proximal cells was also observed in the Ren1c -/- mouse, this suggests a conserved response to the lack of renin, reflecting a common action of AngII on proximal tubule ion/water transport across species and deserving further investigation. This contrasts with the absence of such a mechanism in the aglomerular teleost 47 .
During early mammalian kidney development, renin cells are expressed throughout the renal vasculature - only after birth are renin-expressing cells spatially restricted to the JGA 48 . However, RLCs retain the ability to switch to an endocrine renin phenotype in response to physiological challenge 18, 33, 49, 50 . We questioned whether the lack of functional renin might simulate such a challenge and crossed the double transgenic reporter line Tg(ren:RFP-LifeAct, acta2:EGFP) with ren +/+ and ren -/- fish to assess the extent of RLCs in the mesonephric kidney. There was a dramatic recruitment of RLCs along the renal vasculature with concomitant decrease of EGFP expression in the mesonephric kidney of the ren -/- zebrafish, as seen in the mouse 48 . RLCs may be recruited along the arterioles in an attempt to control ion concentration and/or blood volume. Obligate heterozygous fish carrying deletion or substitution of the exon 2 glycosylation site, in combination with the knockout allele complemented for loss of renin, as shown by the restoration of the banding pattern of afferent arterioles. This, together with the biochemical data, confirms that renin glycosylation in exon 2, though highly conserved, is not essential for renin activity. FAC sorting of vascular and renin fluorescent reporter-expressing cells on ren +/+ or ren -/- backgrounds proved to be a very effective method for isolating JG cells from the former and RLCs from the latter, where over three-fold more cells expressed the RFP reporter. Cluster analysis of the merged zebrafish libraries clearly distinguished RLCs from smooth muscle cells, identifying numerous up-regulated renin cell-specific transcription factors, including cnot4b, twist1a, nr2f5, nkx3.2, and sox6, or down-regulated transcription factors (cited4a), regulators of signalling pathways including angptl3 and rgs5b, and also hmox1a, tnmd and tfpia. Several of these have been recognised previously as RLC-specific in the mouse 18 .
Sub-cluster analysis allowed us to identify differentially expressed genes distinguishing JG cells from RLCs. Increased expression of cited4a, id1, dio1, nrarpa nkx3.2, and tnfaip8l3 was seen in JG cells relative to RLCs, while RLCs exhibited increased twist1a, cnot4b, cremb and fosl12 transcription-related factors and a large group of signalling factors including igfbp5b, cnpy1, alkal1, rgl1, angptl3, gpr137ba, vegfaa, pgfb and gng2. Pseudotime mapping indicated that cluster 0 may represent a ‘transition’ stage between JG and RLCs. This suggests a possible de-differentiation and re-differentiation of RLC-derived smooth muscle cells as they become recruited. The overall response to renin knockout was confirmed by canonical correlation analysis of the merged libraries.
Comparison with mouse expression matrices allowed us to look for conserved and divergent expression profiles from respective Ren +/+ and Ren -/- cells of the zebrafish and mouse. Irrespective of genotype, JG and RLCs from the zebrafish mesonephric kidney libraries expressed snai1a, snai2, twist1a and sox6, all of which enable DNA-binding of transcription factors, but none of these were transcribed to a significant extent in the mouse libraries. Equally irrespective of genotype, JG and RLCs from the mouse metanephric kidney libraries expressed transcripts associated with integrins and regulation of focal adhesion. This is highly suggestive that such functions have evolved in RLCs of higher mammals, along with their exposure to increasing blood pressure 51 . Transcripts from the recently identified nuclear mechano-transducer, lamin A (Lmna), associated with the mouse JG cell baroreceptor 52 , were also noticeably absent from zebrafish-derived RLCs, suggesting an additional control in land-based mammals.
In conclusion, despite the control of ion balance afforded to the zebrafish through ionocytes of the gills and skin, these studies reveal conservation of functions such as RLC recruitment and RAS involvement in proximal tubule function between the zebrafish and the mouse, while other functions, related to increased blood pressure, have evolved. The zebrafish is proving a valuable and tractable, vertebrate model for exploring mechanisms in the RAS and renin cell biology.
Perspective
Ablation of renin lineage cells in the anterior mesenteric artery of zebrafish larvae reveals dual functionality of the cells – both renin expression and contractility. This should be explored in the mesonephric and metanephric kidney. Targeted knockout of renin function is not lethal in zebrafish larvae, because sodium homeostasis is achieved by ionocytes in the gills. Following renin knockout, adult zebrafish show vacuolation of the proximal tubule and renin lineage cell recruitment in the mesonephric vasculature, both of which are seen in the metanephric kidney of the mouse, suggesting an ongoing requirement for renin in both species. Transcriptome analysis of renin lineage cells from zebrafish and mice reveal significant differences however – with mouse RLCs expressing transcripts associated with integrins and focal adhesion regulation. This suggests that such functions have evolved with transition to land, which created additional homeostatic challenges including increased blood pressure.
Supplementary Material
Pathophysiologic Novelty and Significance.
What is new
Renin knockout delays development of the swim bladder in zebrafish larvae
Mice and zebrafish show vacuolation of the proximal tubule in the absence of functional renin
Both mouse metanephric and zebrafish mesonephric kidneys recruit renin lineage cells in the absence of functional renin
What is relevant
zebrafish renin lineage cells share multiple functions and comparison reveals novel mechanistic insights relating to variation in homeostatic challenges
What are the pathophysiological implications
the zebrafish provides important information about the evolution of renin lineage cell function. Our results suggest a continued, but non-critical requirement for renin in adult mouse and zebrafish
Acknowledgements
We wish to thank Dr John Wilson-Kanamori for assistance with bioinformatics analysis, and Dr Alessandro Brombin for assistance with construction of the zebrafish reference genome. We thank the QMRI aquarium staff for technical help; Jason Early and the UK Zebrafish Screening Facility; Xiuyin Liang for technical help; Attoquant Diagnostic GmbH, Vienna, Austria for angiotensin metabolite assays.
Source of Funding
CB, CBB, SR, and AA were supported by the British Heart Foundation Centre of Research Excellence Award (RE/08/001/23904); SH by MRC/EPSRC DTA OPTIMA EP/L016559/1; LM by the BHF CoRE and Kidney Research UK (RP_026_20180305). RAG and MLSSL are supported by National Institutes of Health Grants (DK 096373, DK 116718, DK 116196, and HL 148044).
Footnotes
Disclosures
None of the authors has declared any conflict of interest
References
- 1.Minuth M, Hackenthal E, Poulsen K, Rix E, Taugner R. Renin immunocytochemistry of the differentiating juxtaglomerular apparatus. Anat Embryol (Berl) 1981;162:173–181. doi: 10.1007/BF00306489. [DOI] [PubMed] [Google Scholar]
- 2.Celio MR, Groscurth P, Inagami T. Ontogeny of renin immunoreactive cells in the human kidney. Anat Embryol (Berl) 1985;173:149–155. doi: 10.1007/BF00316297. [DOI] [PubMed] [Google Scholar]
- 3.Richoux JP, Amsaguine S, Grignon G, Bouhnik J, Menard J, Corvol P. Earliest renin containing cell differentiation during ontogenesis in the rat. An immunocytochemical study. Histochemistry. 1987;88:41–46. doi: 10.1007/BF00490165. [DOI] [PubMed] [Google Scholar]
- 4.Gomez RA, Norwood VF. Developmental consequences of the renin-angiotensin system. Am J Kidney Dis. 1995;26:409–431. doi: 10.1016/0272-6386(95)90487-5. [DOI] [PubMed] [Google Scholar]
- 5.Reddi V, Zaglul A, Pentz ES, Gomez RA. Renin-expressing cells are associated with branching of the developing kidney vasculature. J Am Soc Nephrol. 1998;9:63–71. doi: 10.1681/ASN.V9163. [DOI] [PubMed] [Google Scholar]
- 6.Guron G, Friberg P. An intact renin-angiotensin system is a prerequisite for normal renal development. J Hypertens. 2000;18:123–137. doi: 10.1097/00004872-200018020-00001. [DOI] [PubMed] [Google Scholar]
- 7.Tufro-McReddie A, Romano LM, Harris JM, Ferder L, Gomez RA. Angiotensin ii regulates nephrogenesis and renal vascular development. Am J Physiol. 1995;269:F110–115. doi: 10.1152/ajprenal.1995.269.1.F110. [DOI] [PubMed] [Google Scholar]
- 8.Pentz ES, Moyano MA, Thornhill BA, Sequeira Lopez ML, Gomez RA. Ablation of renin-expressing juxtaglomerular cells results in a distinct kidney phenotype. Am J Physiol Regul Integr Comp Physiol. 2004;286:R474–483. doi: 10.1152/ajpregu.00426.2003. [DOI] [PubMed] [Google Scholar]
- 9.Takahashi N, Lopez ML, Cowhig JE, Taylor MA, Hatada T, Riggs E, Lee G, Gomez RA, Kim HS, Smithies O. Ren1c homozygous null mice are hypotensive and polyuric, but heterozygotes are indistinguishable from wild-type. J Am Soc Nephrol. 2005;16:125–132. doi: 10.1681/ASN.2004060490. [DOI] [PubMed] [Google Scholar]
- 10.Yanai K, Saito T, Kakinuma Y, Kon Y, Hirota K, Taniguchi-Yanai K, Nishijo N, Shigematsu Y, Horiguchi H, Kasuya Y, Sugiyama F, et al. Renin-dependent cardiovascular functions and renin-independent blood-brain barrier functions revealed by renin-deficient mice. J Biol Chem. 2000;275:5–8. doi: 10.1074/jbc.275.1.5. [DOI] [PubMed] [Google Scholar]
- 11.Fournier D, Luft FC, Bader M, Ganten D, Andrade-Navarro MA. Emergence and evolution of the renin-angiotensin-aldosterone system. J Mol Med (Berl) 2012;90:495–508. doi: 10.1007/s00109-012-0894-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Armesto P, Cousin X, Salas-Leiton E, Asensio E, Manchado M, Infante C. Molecular characterization and transcriptional regulation of the renin-angiotensin system genes in senegalese sole (solea senegalensis kaup, 1858): Differential gene regulation by salinity. Comp Biochem Physiol A Mol Integr Physiol. 2015;184:6–19. doi: 10.1016/j.cbpa.2015.01.021. [DOI] [PubMed] [Google Scholar]
- 13.Hoshijima K, Hirose S. Expression of endocrine genes in zebrafish larvae in response to environmental salinity. J Endocrinol. 2007;193:481–491. doi: 10.1677/JOE-07-0003. [DOI] [PubMed] [Google Scholar]
- 14.Kumai Y, Perry SF. Mechanisms and regulation of na(+) uptake by freshwater fish. Respir Physiol Neurobiol. 2012;184:249–256. doi: 10.1016/j.resp.2012.06.009. [DOI] [PubMed] [Google Scholar]
- 15.Kumai Y, Bernier NJ, Perry SF. Angiotensin-ii promotes na+ uptake in larval zebrafish, danio rerio, in acidic and ion-poor water. J Endocrinol. 2014;220:195–205. doi: 10.1530/JOE-13-0374. [DOI] [PubMed] [Google Scholar]
- 16.Rider SA, Mullins LJ, Verdon RF, MacRae CA, Mullins JJ. Renin expression in developing zebrafish is associated with angiogenesis and requires the notch pathway and endothelium. Am J Physiol Renal Physiol. 2015;309:F531–539. doi: 10.1152/ajprenal.00247.2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Rider SA, Christian HC, Mullins LJ, Howarth AR, MacRae CA, Mullins JJ. Zebrafish mesonephric renin cells are functionally conserved and comprise two distinct morphological populations. Am J Physiol Renal Physiol. 2017;312:F778–F790. doi: 10.1152/ajprenal.00608.2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Brunskill EW, Sequeira-Lopez ML, Pentz ES, Lin E, Yu J, Aronow BJ, Potter SS, Gomez RA. Genes that confer the identity of the renin cell. J Am Soc Nephrol. 2011;22:2213–2225. doi: 10.1681/ASN.2011040401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lin EE, Sequeira-Lopez ML, Gomez RA. Rbp-j in foxd1+ renal stromal progenitors is crucial for the proper development and assembly of the kidney vasculature and glomerular mesangial cells. Am J Physiol Renal Physiol. 2014;306:F249–258. doi: 10.1152/ajprenal.00313.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Liang P, Jones CA, Bisgrove BW, Song L, Glenn ST, Yost HJ, Gross KW. Genomic characterization and expression analysis of the first nonmammalian renin genes from zebrafish and pufferfish. Physiol Genomics. 2004;16:314–322. doi: 10.1152/physiolgenomics.00012.2003. [DOI] [PubMed] [Google Scholar]
- 21.Parichy DM, Elizondo MR, Mills MG, Gordon TN, Engeszer RE. Normal table of postembryonic zebrafish development: Staging by externally visible anatomy of the living fish. Dev Dyn. 2009;238:2975–3015. doi: 10.1002/dvdy.22113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Diep CQ, Peng Z, Ukah TK, Kelly PM, Daigle RV, Davidson AJ. Development of the zebrafish mesonephros. Genesis. 2015;53:257–269. doi: 10.1002/dvg.22846. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Zhou W, Boucher RC, Bollig F, Englert C, Hildebrandt F. Characterization of mesonephric development and regeneration using transgenic zebrafish. Am J Physiol Renal Physiol. 2010;299:F1040–1047. doi: 10.1152/ajprenal.00394.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Diep CQ, Ma D, Deo RC, Holm TM, Naylor RW, Arora N, Wingert RA, Bollig F, Djordjevic G, Lichman B, Zhu H, et al. Identification of adult nephron progenitors capable of kidney regeneration in zebrafish. Nature. 2011;470:95–100. doi: 10.1038/nature09669. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Westerfield M. A guide for the laboratory use of zebrafish danio (brachydanio) rerio. University of Oregon Press; 2007. [Google Scholar]
- 26.Brinkman EK, Chen T, Amendola M, van Steensel B. Easy quantitative assessment of genome editing by sequence trace decomposition. Nucleic Acids Res. 2014;42:e168. doi: 10.1093/nar/gku936. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Perner B, Englert C, Bollig F. The wilms tumor genes wt1a and wt1b control different steps during formation of the zebrafish pronephros. Dev Biol. 2007;309:87–96. doi: 10.1016/j.ydbio.2007.06.022. [DOI] [PubMed] [Google Scholar]
- 28.Whitesell TR, Kennedy RM, Carter AD, Rollins EL, Georgijevic S, Santoro MM, Childs SJ. An α-smooth muscle actin (acta2/αsma) zebrafish transgenic line marking vascular mural cells and visceral smooth muscle cells. PLoS One. 2014;9:e90590. doi: 10.1371/journal.pone.0090590. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Buckley C, Carvalho MT, Young LK, Rider SA, McFadden C, Berlage C, Verdon RF, Taylor JM, Girkin JM, Mullins JJ. Precise spatio-temporal control of rapid optogenetic cell ablation with mem-killerred in zebrafish. Scientific Reports. 2017;7:5096. doi: 10.1038/s41598-017-05028-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Lawton PF, Lee MD, Saunter CD, Girkin JM, McCarron JG, Wilson C. Vasotracker, a low-cost and open source pressure myograph system for vascular physiology. Front Physiol. 2019;10:99. doi: 10.3389/fphys.2019.00099. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.McCampbell KK, Springer KN, Wingert RA. Atlas of cellular dynamics during zebrafish adult kidney regeneration. Stem Cells Int. 2015;2015:547636. doi: 10.1155/2015/547636. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, et al. Fiji: An open-source platform for biological-image analysis. Nat Methods. 2012;9:676–682. doi: 10.1038/nmeth.2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Gomez RA, Belyea B, Medrano S, Pentz ES, Sequeira-Lopez ML. Fate and plasticity of renin precursors in development and disease. Pediatr Nephrol. 2014;29:721–726. doi: 10.1007/s00467-013-2688-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Butler A, Hoffman P, Smibert P, Papalexi E, Satija R. Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nature Biotechnology. 2018;36:411–420. doi: 10.1038/nbt.4096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Trapnell C, Cacchiarelli D, Grimsby J, Pokharel P, Li S, Morse M, Lennon NJ, Livak KJ, Mikkelsen TS, Rinn JL. The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells. Nature Biotechnology. 2014;32:381–386. doi: 10.1038/nbt.2859. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Bolger AM, Lohse M, Usadel B. Trimmomatic: A flexible trimmer for illumina sequence data. Bioinformatics. 2014;30:2114–2120. doi: 10.1093/bioinformatics/btu170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C. Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods. 2017;14:417–419. doi: 10.1038/nmeth.4197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Soneson C, Love MI, Robinson MD. Differential analyses for rna-seq: Transcript-level estimates improve gene-level inferences. F1000Res. 2015;4:1521. doi: 10.12688/f1000research.7563.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.DeLaughter DM. The use of the fluidigm c1 for rna expression analyses of single cells. Curr Protoc Mol Biol. 2018;122:e55. doi: 10.1002/cpmb.55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Pentz ES, Lopez ML, Cordaillat M, Gomez RA. Identity of the renin cell is mediated by camp and chromatin remodeling: An in vitro model for studying cell recruitment and plasticity. Am J Physiol Heart Circ Physiol. 2008;294:H699–707. doi: 10.1152/ajpheart.01152.2007. [DOI] [PubMed] [Google Scholar]
- 41.Watanabe H, Martini AG, Brown EA, Liang X, Medrano S, Goto S, Narita I, Arend LJ, Sequeira-Lopez MLS, Gomez RA. Inhibition of the renin-angiotensin system causes concentric hypertrophy of renal arterioles in mice and humans. JCI Insight. 2021;6:e154337. doi: 10.1172/jci.insight.154337. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Pittman K, Yufera M, Pavlidis M, Geffen A, Koven W, Ribeiro L, Zambonino-Infante J, Tandler A. Fantastically plastic: Fish larvae equipped for a new world. Reviews Aquaculture. 2013;5:S224–267. [Google Scholar]
- 43.Guh YJ, Hwang PP. Insights into molecular and cellular mechanisms of hormonal actions on fish ion regulation derived from the zebrafish model. Gen Comp Endocrinol. 2017;251:12–20. doi: 10.1016/j.ygcen.2016.08.009. [DOI] [PubMed] [Google Scholar]
- 44.Bendele A, Seely J, Richey C, Sennello G, Shopp G. Short communication: Renal tubular vacuolation in animals treated with polyethylene-glycol-conjugated proteins. Toxicol Sci. 1998;42:152–157. doi: 10.1006/toxs.1997.2396. [DOI] [PubMed] [Google Scholar]
- 45.Dickenmann M, Oettl T, Mihatsch MJ. Osmotic nephrosis: Acute kidney injury with accumulation of proximal tubular lysosomes due to administration of exogenous solutes. Am J Kidney Dis. 2008;51:491–503. doi: 10.1053/j.ajkd.2007.10.044. [DOI] [PubMed] [Google Scholar]
- 46.Horita S, Nakamura M, Shirai A, Yamazaki O, Satoh N, Suzuki M, Seki G. Regulatory roles of nitric oxide and angiotensin ii on renal tubular transport. World J Nephrol. 2014;3:295–301. doi: 10.5527/wjn.v3.i4.295. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Zucker A, Nishimura H. Renal responses to vasoactive hormones in the aglomerular toadfish, opsanus tau. Gen Comp Endocrinol. 1981;43:1–9. doi: 10.1016/0016-6480(81)90024-1. [DOI] [PubMed] [Google Scholar]
- 48.Sauter A, Machura K, Neubauer B, Kurtz A, Wagner C. Development of renin expression in the mouse kidney. Kidney Int. 2008;73:43–51. doi: 10.1038/sj.ki.5002571. [DOI] [PubMed] [Google Scholar]
- 49.Lopez ML, Gomez RA. The renin phenotype: Roles and regulation in the kidney. Curr Opin Nephrol Hypertens. 2010;19:366–371. doi: 10.1097/MNH.0b013e32833aff32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Sequeira López ML, Pentz ES, Nomasa T, Smithies O, Gomez RA. Renin cells are precursors for multiple cell types that switch to the renin phenotype when homeostasis is threatened. Dev Cell. 2004;6:719–728. doi: 10.1016/s1534-5807(04)00134-0. [DOI] [PubMed] [Google Scholar]
- 51.Mohamed TH, Watanabe H, Kaur R, Belyea BC, Walker PD, Gomez RA, Sequeira-Lopez MLS. Renin-expressing cells require β1-integrin for survival and for development and maintenance of the renal vasculature. Hypertension. 2020;76:458–467. doi: 10.1161/HYPERTENSIONAHA.120.14959. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Watanabe H, Belyea BC, Paxton RL, Li M, Dzamba BJ, DeSimone DW, Gomez RA, Sequeira-Lopez MLS. Renin cell baroreceptor, a nuclear mechanotransducer central for homeostasis. Circ Res. 2021;129:262–276. doi: 10.1161/CIRCRESAHA.120.318711. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.