Inactivation of Invs/Nphp2 in renal epithelial cells drives infantile nephronophthisis like phenotypes in mouse

Nephronophthisis (NPHP) is a ciliopathy characterized by renal fibrosis and cyst formation, and accounts for a significant portion of end stage renal disease in children and young adults. Currently, no targeted therapy is available for this disease. INVS/NPHP2 is one of the over 25 NPHP genes identified to date. In mouse, global knockout of Invs leads to renal fibrosis and cysts. However, the precise contribution of different cell types and the relationship between epithelial cysts and interstitial fibrosis remains undefined. Here, we generated and characterized cell-type-specific knockout mouse models of Invs, investigated the impact of removing cilia genetically on phenotype severity in Invs mutants and evaluated the impact of the histone deacetylase inhibitor valproic acid (VPA) on Invs mutants. Epithelial-specific knockout of Invs in Invsflox/flox;Cdh16-Cre mutant mice resulted in renal cyst formation and severe stromal fibrosis, while Invsflox/flox;Foxd1-Cre mice, where Invs is deleted in stromal cells, displayed no observable phenotypes up to the young adult stage, highlighting a significant role of epithelial-stromal crosstalk. Further, increased cell proliferation and myofibroblast activation occurred early during disease progression and preceded detectable cyst formation in the Invsflox/flox;Cdh16-Cre kidney. Moreover, concomitant removal of cilia partially suppressed the phenotypes of the Invsflox/flox;Cdh16-Cre mutant kidney, supporting a significant interaction of cilia and Invs function in vivo. Finally, VPA reduced cyst burden, decreased cell proliferation and ameliorated kidney function decline in Invs mutant mice. Our results reveal the critical role of renal epithelial cilia in NPHP and suggest the possibility of repurposing VPA for NPHP treatment.

. Ciliary dysfunction is associated with an array of human disorders, collectively termed ciliopathies, that include obesity, mental retardation, retinal degeneration, and polycystic kidney disease (PKD), among many others (reviewed in Hildebrandt et al., 2009). In renal epithelial cells, cilia protrude into the lumen of tubules, exposed to both mechanical and chemical signals carried by extracellular fluid flow. Disruption of cilia biogenesis almost inevitably leads to the formation of epithelium-lined fluid filled kidney cysts in mouse (Pazour et al., 2000;Yoder et al., 2002).
Polycystin 1 and 2 (PC1 and PC2), encoded by the autosomal dominant PKD (ADPKD) genes Pkd1 and Pkd2, respectively, are targeted to cilia and this specific trafficking highly correlates with their in vivo function (Yoder et al., 2002;Barr and Sternberg, 1999;Cai et al., 2014;Pazour et al., 2002;Yoshiba et al., 2012). Interestingly, although both cilia biogenesis and polycystin mutants develop cystic kidney, concomitant removal of cilia ameliorates cyst progression triggered by polycystin inactivation, demonstrating that intact cilia are required for cyst growth in ADPKD models (Ma et al., 2013). Based on these results, it was postulated that polycystins function to inhibit a novel and uncharacterized cilia-dependent cyst activating (CDCA) signaling pathway (Ma et al., 2013). This hypothesis implies the existence of a separate cyst-inhibiting (CDCI) pathway mediated by cilia, inactivation of which, together with CDCA, underlies the milder renal cyst formation in cilia biogenesis mutants compared to polycystin mutants. Renal fibrosis is also frequently observed in cystic kidney diseases; but is often considered as secondary to cyst formation and the underlying mechanism remains poorly understood.
Nephronophthisis (NPHP) is a recessive renal cystic disease with cysts mostly confined in the corticomedullary junction region in the juvenile form of the disease, and more widely distributed across the kidney in the infantile form. Since multiple NPHP proteins are localized to cilia, basal bodies and/ or show functional involvement in cilia biogenesis and/or signaling, NPHP is considered as a ciliopathy (Fliegauf et al., 2006;Otto et al., 2003;Otto et al., 2008;Shiba et al., 2010;Shiba et al., 2009;Williams et al., 2010;Winkelbauer et al., 2005;Won et al., 2011). Compared with other types of renal ciliopathies, a prominent feature of NPHP is the severity of interstitial fibrosis. Genetically NPHP is highly heterogenous and at least 25 genes have been identified for the classic form of this disease (for a review, see Srivastava et al., 2017). However, mouse mutants of NPHP genes do not always show obvious renal phenotypes (Won et al., 2011;Jiang et al., 2008;Ronquillo et al., 2016). Invs, the homolog of which is associated with both infantile and juvenile NPHP in human, was discovered as a novel gene inactivated in a mouse mutant with reversed internal organ asymmetry along the eLife digest One of the most common causes of kidney failure in children and young adults is nephronophthisis. This genetic disease causes cysts and tissue scarring in the kidneys, leading to excessive urine production and extreme tiredness. Unfortunately, there is no targeted therapy available for this condition.
Scientists do not fully understand how genetic mutations lead to these symptoms. Previous research in mice showed that blocking the gene for a protein called INVS recreated signs similar to nephronophthisis. However, it is not clear how the different cell types in the kidneys are involved. Previous results suggest that cilia, the hair-like projections on the surface of cells, could be involved in developing cysts in nephronophthisis.
To understand how the disease is driven, Li, Xu et al. created a range of genetically modified mice with INVS missing in different cell types. When INVS was removed from cells that line the kidney tubules, the mice developed scarring and cysts. By contrast, there were no symptoms when connective tissue cells were lacking INVS. When Li, Xu et al. removed the cilia from the cells, it helped to reduce the negative impact of the loss of INVS. In addition, a drug called valproic acid reduced the cysts and tissue scarring, and slowed kidney decline in the mutant mice, suggesting the possibility of repurposing this drug for nephronophthisis treatment.
These results could help researchers to study other conditions that are influenced by the health of cilia. Future work on nephronophthisis will be needed to understand how INVS causes the disease and the mechanism for the benefits of valproic acid. Li left-right body axis and was originally named as Inversin (Invs) (Yokoyama et al., 1993;Mochizuki et al., 1998;Morgan et al., 1998). This whole-body knockout model additionally displays kidney cysts in both the medulla and cortex and renal fibrosis at the neonatal stage, resembling the renal phenotypes of infantile NPHP (Morgan et al., 1998;Phillips et al., 2004). However, the functional significance of Invs in different renal cell types is undefined. Moreover, the progression of renal fibrosis and its relationship with cyst formation remains to be characterized.
Here, to dissect tissue-specific function of Invs and the mechanism of interstitial fibrosis resulting from defective Invs, we generated two conditional mouse knockout models: (1) targeting Invs deletion in epithelial cells of the distal nephron (Invs flox/flox ;Cdh16-Cre mice), and (2) deleting Invs in stromal cells (Invs flox/flox ;Foxd1-Cre mice). We found that the infantile NPHP-like phenotypes in mutant kidneys arise from abnormal signaling specifically within and from renal epithelial cells. Further, we investigated the relationship between cyst formation and interstitial fibrosis. Our results suggest that the fibrotic response and abnormal increase of cell proliferation precede and therefore could initiate independently of cyst formation in this model. Mechanistically, we find that removal of cilia reduces the severity of Invs phenotypes, suggesting that, similar to polycystins, INVS inhibits a cilia-dependent cyst and fibrosis activating pathway. Finally, we show that the histone deacetylase inhibitor (HDACI) valproic acid (VPA) reduces cell proliferation, suppresses cyst expansion, and preserves kidney function in Invs mutant mice, suggesting VPA, and perhaps other HDACIs, as a potential candidate drug for treating NPHP.

Inactivation of Invs in renal epithelial cells of the distal nephron leads to progressive cystic kidney disease in mouse
To dissect tissue-specific function of Invs, we obtained embryonic stem cell clones generated by the European Conditional Mouse Mutagenesis Program (EUCOMM). The Invs tm1a 'knockout-first' allele contains an IRES:lacZ-neo selection cassette located within two FLPE recombinase (FRT) sites and two loxP sites flanking the third exon of mouse Invs ( Figure 1A). Deletion between the two loxP sites is predicted to lead to frame shift and a premature stop codon. The Invs tm1a/+ mice were crossed to a deleter strain expressing FLPe under the control of a ubiquitous actin promoter (Rodríguez et al., 2000) to generate a floxed allele of Invs. The carrier Invs flox/+ mice were backcrossed with C57BL/6 J for five generations and then crossed with carriers of Cdh16-Cre, which directs Cre expression in the distal nephron of the developing metanephros after E11.5 (Shao et al., 2002a;Shao et al., 2002b), to generate Invs flox/+ ;Cdh16-Cre mice. Crossing Invs flox/flox with Invs flox/+ ;Cdh16-Cre mice led to the generation of live Invs flox/flox ;Cdh16-Cre pups in Mendelian ratio. In mutants, neonatal pups showed progressive enlargement of the kidney: At postnatal day 7 (P7), the mutant kidney size was comparable to that in control animals and there was no significant difference in body weight and kidney to body weight (KBW) ratio between the two groups ( Figure 1B and C). At P14, the mutant kidneys were larger, and KBW ratios were significantly increased, although the body weight remained comparable ( Figure 1B and C). Mutant kidneys became increasingly enlarged at P21 and P28, with comparable body weight at P21 and reduced body weight in mutants at P28 ( Figure 1B and C). We additionally compared KBW ratios between male and female mice and no significant differences were detected between genders in either control or mutant groups at P14 and P21 ( Figure 1D).
To inspect for potential cyst formation, we performed Hematoxylin/eosin (HE) stainning on kidney sections and quantified cystic index as defined previously (Cao et al., 2009;Shibazaki et al., 2008). Consistent with the progression of the KBW ratio, no cyst was detected in the P7 mutant kidney and the cystic index remained comparable between the wild type and mutant kidney ( Figure 1E and F). At P14, small cysts were detected, and the cystic index was already significantly increased in the mutant kidney ( Figure 1E and F). At P21, cysts became larger and more numerous in mutant kidneys ( Figure 1E), and by P28 mutant kidneys were severely cystic with large cysts throughout the cortex and medullar region ( Figure 1E). The cystic index progressively increased in mutant kidneys during the same time window ( Figure 1F).
To understand the cellular origin of kidney cysts in mutant mice, we performed immunofluorescence staining, labeling different segments of the nephron in kidney sections. Consistent with histology, neither the collecting duct, labelled with Dolichos Biflorus Agglutinin (DBA), nor the proximal tubule, The online version of this article includes the following source data for figure 1: Source data 1. Related to Figure 1C, D, F and I. labelled with Lotus Tetragonolobus Lectin (LTL), showed enlargement at P7 ( Figure 1G). At P14, cysts were detected mostly in the DBA positive collecting duct ( Figure 1G). At P21 and P28, DBA positive cysts were enlarged and numerous ( Figure 1G). Large cysts were also detected in the thick ascending limb, labelled by anti-Tamm-Horsfall protein (THP) at both P21 and P28 ( Figure 1H). By contrast, the LTL positive proximal tubule was free from cysts at P28 ( Figure 1G). Overall, the cystic region was consistent with the expression domain of Cdh16-Cre.
To monitor kidney function, we assayed for the blood urea nitrogen (BUN) level. At P7, mutant and control mice were comparable ( Figure 1I). At P14, BUN levels became significantly higher in mutant mice ( Figure 1I), and BUN levels became more elevated in mutant mice at P21 and P28, demonstrating a progressive decline of kidney function ( Figure 1I).
Invs flox/flox ;Cdh16-Cre mice develop severe renal interstitial fibrosis As interstitial fibrosis is a significant phenotype in NPHP, we first used trichrome staining to detect collagen deposition in P21 kidney sections. In mutant kidneys, an increase of the blue staining of collagen was detected in the cortex region ( Figure 2A). Further, we performed immunostaining of smooth muscle actin (SMA), a marker of myofibroblast, to monitor the fibrotic response over time. In the wildtype kidney, SMA signal was confined to the peri-vascular region. SMA signal was increased in the cortex region of the mutant kidney at P14 and became dramatically elevated and widespread The online version of this article includes the following source data for figure 2: Source data 1. Related to Figure 2D and G.
in the mutant kidney at P21 and P28 ( Figure 2B). We additionally performed immunostaining of vimentin, a marker of activated fibroblasts that becomes upregulated early during injury responses in the kidney (Ó hAinmhire et al., 2019;Zhou et al., 2019). Vimentin signal was increased in the cortex region of the mutant kidney at P14 and P21 ( Figure 2C).

Fibrotic response precedes cyst formation in Invs flox/flox ;Cdh16-Cre mice
To further delineate the relationship between cyst formation and interstitial fibrosis, we analyzed mutant kidney at earlier stages. At P9, the KBW ratio was normal in mutants ( Figure 2D). By contrast, SMA and vimentin signal in the cortex region was already increased ( Figure 2E). However, we also detected isolated tubule dilation on mutant kidney sections ( Figure 2E). At the non-cystic P7 stage, SMA and vimentin signals were modestly increased in the cortex region ( Figure 2F). Reverse transcription and quantitative PCR (RT-qPCR) using kidney lysates also showed a significant increase of Sma and Vimentin mRNA level in the mutant at P7 ( Figure 2G). Combined, these results reveal an increase of fibrotic markers at P7, preceding detectable cyst formation.
Cell over-proliferation precedes cyst formation in the Invs flox/flox ;Cdh16-Cre kidney Since over-proliferation of epithelial cells is a driver of cyst progression (Nadasdy et al., 1995;Takakura et al., 2009;Zhang et al., 2021), we performed immunofluorescence staining of P7 kidney sections using anti-proliferating cell nuclear antigen (PCNA), a nuclear marker for cell turnover. PCNA positive cells increased in the collecting duct (DBA+) in the cortex region of the mutant kidney ( Figure 3A and B, upper panels). PCNA positive cells also increased outside of DBA positive regions. To investigate whether the number of proliferating interstitial cells were increased, we outlined epithelial tubules by co-labeling with anti-laminin. The number of proliferating interstitial cells, outside of epithelial regions encircled by laminin, was significantly increased in the mutant kidney ( Figure 3A and B, lower panels). Taking advantage of the distinct localization patterns of PCNA at different cell cycle stages, that is replication foci in S-phase and diffusive nuclear localization in G1 and G2 cells (Zerjatke et al., 2017), we further showed that the number of cells in interphase and S phase, respectively, was significantly increased in both the collecting duct and interstitium ( Figure 3A and B). Moreover, the number of mitotic cells, labelled by the mitosis marker phosphorylated histone H3 (pHH3), was significantly increased in both regions ( Figure 3C and D).

Invs flox/flox ;Foxd1-Cre mice show no observable renal phenotypes up to the young adult stage
As interstitial fibrosis is a prominent phenotype of NPHP, we investigated the role of Invs in the stroma by generating and characterizing Invs flox/flox ;Foxd1-Cre mice. Foxd1-Cre drives Cre expression in metanephric mesenchymal cells destined to be stromal cells, including interstitial cells, mesangial cells and pericytes; and Cre activity can be readily detected at E11.5, similar to the onset of Cdh16-Cre activity (Shao et al., 2002a;Shao et al., 2002b;Hum et al., 2014;Humphreys et al., 2010). The resultant mutant mice were viable. At P28, we did not observe any morphologic or functional phenotypes. No differences were seen in kidney size, KWB ratio, cyst formation, or collagen deposition in the mutant kidney at P28 ( Figure 4A-E). In addition, immunofluorescence staining of kidney sections with the connecting tubule and collecting duct marker aquaporin-2 (AQP2), the collecting duct maker DBA, the thick ascending limb marker THP and the proximal tubule marker LTL showed no difference in the mutant kidney ( Figure 4F). Moreover, immunostaining of SMA failed to detect any changes in the mutant kidney at P28 ( Figure 4G). In accordance, BUN levels were also normal in P28 mutants ( Figure 4H). To verify deletion of Invs in interstitial cells, we first analyzed the expression of eGFP-Cre by immunostaining of eGFP and detected wide expression of eGFP-Cre in the renal stroma in both the cortex and medulla region ( Figure 4I). We then cultured primary interstitial cells following a previously published protocol (Nakai et al., 2021) and performed immunostaining on cultured cells. We detected INVS in the proximal compartment of 85.01% of cilia in cells from control kidneys, consistent with the previously reported localization pattern of INVS (Shiba et al., 2009;Bennett et al., 2020), whereas    only 34.24% INVS-positive cilia were observed in cells of mutant kidneys ( Figure 4J and K). Similarly, ANKS6, encoded by another NPHP gene and is known to localize to the Inversin compartment in a INVS dependent manner (Bennett et al., 2020;Czarnecki et al., 2015), was detected in 87.78% of cilia in cells from the control kidney, but was reduced to 39.89% cilia in cells from the mutant kidney ( Figure 4J and K), validating the deletion of Invs in a significant fraction of interstitial cells.
Taken together, these results suggest that deletion of Invs in stromal cells fails to trigger renal fibrosis nor kidney cysts up to young adults.
We then investigated the impact of Ift88 deletion on interstitial phenotypes observed in P21 Invsflox/flox ;Cdh16-Cre mutant kidneys. Immunostaining of kidney sections revealed that the increase of of SMA and vimentin was more moderately in Ift88 flox/flox ;Cdh16-Cre mutants compared to Invs flox/ flox ;Cdh16-Cre mutants ( Figure 5I). In addition, Invs flox/flox ;Ift88 flox/flox ;Cdh16-Cre double mutants showed reduced signal of SMA and vimentin compared to Invs flox/flox ;Cdh16-Cre single mutants ( Figure 5I). Western Blot of total kidney lysates was then used to quantify the levels of SMA and Collagen I. While the levels of both proteins were significantly increased in Invs single mutants compared to controls, the increase in Ift88 single mutants was significant by t-test but not by one-way ANOVA analysis followed by Šidák's test ( Figure 5J). Moreover, the levels of SMA and Collagen I in double mutants were significantly reduced compared to Invs single mutants ( Figure 5J). Combined, these results suggest that both myofibroblast activation and fibrosis were reduced when Ift88 was deleted in Invs flox/ flox ;Cdh16-Cre mutants.
The online version of this article includes the following source data for figure 5: Source data 1. Related to Figure 5E, G, H and J. and B). In the Invs flox/flox ;Ift88 flox/flox ;Cdh16-Cre double mutant kidney, 13.4% of DBA positive cells were PCNA positive, a significant decrease from the Invs flox/flox ;Cdh16-Cre single mutant ( Figure 6A and  B). Further, Ift88 deletion reduced proliferating cells in the interstitium of the Invs flox/flox ;Cdh16-Cre kidney. In the cortex region of the control kidney, 2.1% of interstitial cells were PCNA positive. In the Ift88 flox/flox ;Cdh16-Cre mutant kidney, this number increased to 7.2% ( Figure 6A and B). 22.3% of interstitial cells were PCNA positive in the cortex of the Invs flox/flox ;Cdh16-Cre kidney, which reduced to 12.8% in the Invs flox/flox ;Ift88 flox/flox ;Cdh16-Cre double mutant kidney ( Figure 6A and B). To validate  the impact of Ift88 deletion on cell proliferation, we analyzed the level of multiple proteins involved in cell proliferation by performing Western blot using P21 kidney lysates. Results showed that the level of PCNA, CDK1, a kinase required for cell cycle progression, and cyclin D1, which is essential for cell proliferation and prevents cells enter quiescence, was increased in the Invs flox/flox ;Cdh16-Cre mutant kidney ( Figure 6C and D). In the Ift88 flox/flox ;Cdh16-Cre mutant kidney, the level of PCNA was increased significantly by t-test but did not reach statistical significance by one-way ANOVA analysis followed by Šidák's test ( Figure 6C and D). The level of cyclin D1, but not CDK1, was also increased ( Figure 6C  and D). Like the pattern seen by immunostaining of PCNA, the level of PCNA, CDK1 and cyclin D1 in the Invs flox/flox ;Ift88 flox/flox ;Cdh16-Cre double mutant kidney was significantly decreased in comparison to the Invs flox/flox ;Cdh16-Cre single mutant kidney ( Figure 6C and D).
To investigate whether INVS functions in ciliary localization of PC2, we used mouse inner medullary collecting duct (mIMCD3) cells and generated two clones, CR81 and CR84, of Invs -/cells via CRISPR ( Figure 6-figure supplement 1). Immunostaining with anti-PC2 and the cilia marker antiacetylated tubulin showed no difference in ciliary localization of PC2 between mutant and control cells ( Figure 6E and F).
Taken together, these results reveal that the phenotypes of Invs flox/flox ;Cdh16-Cre mutants are partially suppressed when cilia biogenesis is disrupted genetically, similar to Pkd2 mutants through a mechanism independent of the ciliary localization of PC2.
The HDAC inhibitor valproic acid suppresses disease progression in Invs flox/flox ;Pkhd1-Cre mice NPHP is the most frequent monogenic cause of end stage renal disease (ESRD) in children and young adults (Hildebrandt et al., 2009). Currently directed therapy is lacking for NPHP. In previous studies, we found that HDACIs could partially suppress kidney cyst formation and renal fibrosis caused by inactivation of polycystins or the cilia biogenesis gene Arl13b in model organisms (Cao et al., 2009;Li et al., 2016). To investigate whether HDACIs could be used to suppress the phenotypes of NPHP models, we treated Invs mutant mice with the pan HDACI valproic acid (VPA). Since the Invs flox/flox ; Cdh16-Cre model develops disease phenotypes very rapidly at the neonatal stage, posing technical challenges for drug administration, we generated Invs flox/flox ;Pkhd1-Cre mice. Pkhd1-Cre directs Cre expression in the collecting duct starting at E14.5, later than the onset of Cdh16-Cre expression (Ma et al., 2013). In Invs flox/flox ;Pkhd1-Cre mice, mutant kidneys were enlarged and cystic, with a significantly increased KBW ratio, cystic index and BUN level in both male and female mice and decreased body weight at P28 ( Figure 7A-E). We analyzed body weight change of Invs flox/flox ;Pkhd1-Cre mice in more detail and compared it to Invs flox/flox ;Cdh16-Cre mice. At P28, the reduction of body weight in Invs flox/flox-;Pkhd1-Cre mice in comparison to control mice was more moderate than that in Invs flox/flox ;Cdh16-Cre mice (Figure 7-figure supplement 1). Mutant and control mice were intraperitoneal (IP) injected with VPA from P10 to P28. Results showed that this regiment of VPA treatment decreased the size of the mutant kidney, but had no impact on body weight, in comparison to vehicle-treated mutants ( Figure 7A, B). In addition, the KBW ratio, cystic index of kidneys and the BUN level in mutant mice were reduced significantly by VPA treatment in comparison to vehicle-treated mutants ( Figure 7B-E). Moreover, VPA treatment led to a more moderate level of myofibroblast activation in the cortex region of the mutant kidney, as indicated by the level of SMA ( Figure 7F). By contrast, the kidney size, KBW ratio, cystic index and BUN level were not significantly affected by the same VPA treatment in control mice, suggesting that mutant mice are selectively sensitive to VPA treatment ( Figure 7A, B, D and E). However, the reduced body weight phenotype in mutant mice was not suppressed by VPA treatment ( Figure 7B). We cannot rule out the possibility that the side effects of VPA contributed to weight loss in treated mice. In addition, VPA may reduce body weight through inhibiting HDAC during the growth period: the HDACI Trichostatin A was shown to inhibit adipogenesis (Lv et al., 2021). We then investigated the impact of VPA treatment on cell proliferation. The number of PCNA positive cells in the collecting duct labeled by DBA and the interstitium outside of tubular regions outlined by anti-laminin signal was reduced significantly by VPA treatment (Figure 7G and H). This result was further validated by western blot using kidney lysates. The abnormally increased level of PCNA, CDK1, and cyclin D1 in mutant kidneys was decreased significantly by VPA treatment (Figure 7I and J).
Combined, these results show that the HDACI VPA ameliorates the progression of multiple phenotypes in Invs mutants.

Discussion
Similar to infantile NPHP, mutants of the ove allele of Invs, generated by transgenic insertion of plasmid DNA in mouse, show severe cystic kidney disease and fibrosis at the neonatal stage (Yokoyama et al., 1993;Mochizuki et al., 1998;Morgan et al., 1998). It is possible that defective Invs in epithelial and stromal cells caused epithelial cysts and interstitial fibrosis, respectively, in this model. Alternatively, abnormal epithelial-stromal cross-talks, resulting from defective tissue-specific Invs function, might have triggered the phenotypes observed in those previous studies.
In this study, we generated a floxed allele of Invs and used Cdh16-Cre to specifically inactivate Invs in epithelial cells in the distal nephron and Foxd1-Cre to drive Invs deletion in the renal stroma. Significantly, inactivation of Invs in the distal nephron leads to cyst formation throughout the cortex and medullar region by P21, accompanied by interstitial fibrosis, recapitulating the main phenotypes of infantile NPHP. By contrast, inactivation of Invs in the renal stroma caused no observable phenotypes in the kidney at least up to week 8, even though the onset of Foxd1-Cre expression in the developing metanephros occurs at the same time as that of Cdh16-Cre (Shao et al., 2002a;Shao et al., 2002b;Hum et al., 2014;Humphreys et al., 2010). Combined, our results suggest that defective epithelial cells are the main driver of NPHP phenotypes in Invs mutants, highlighting the significance of epithelial-stromal cross-talks in the development and progression of interstitial fibrosis in this model. However, our result does not rule out functional significance of interstitial cells once a pro-cystic and fibrotic response is triggered in mutant epithelial cells.
PKD is frequently accompanied by renal fibrosis. In our previous study, we found that inactivation of the cilia biogenesis gene Arl13b in mouse leads to kidney cysts and fibrosis (Li et al., 2016). However, since the onset of fibrosis and cyst formation overlap in the Arl13b flox/flox ;Cdh16-Cre model, it is difficult to address whether interstitial fibrosis is solely secondary to epithelial cyst formation through, for example, mechano-stress exerted by expanding cysts. We therefore performed a detailed time course analysis of cyst formation and the fibrotic response in the Invs flox/flox ;Cdh16-Cre kidney. Our results showed that SMA and vimentin is already upregulated at P7, before detectable cyst formation, suggesting that the fibrotic response is triggered before cyst formation by signals from defective epithelial cells in this model. Mechano-stress on stromal cells impinged by cysts, and additional signals from cystic epithelial cells could amplify the progression of fibrosis at later stages. Interestingly, we detected an abnormal increase of proliferating cells at the pre-cystic stage in both the collecting duct and interstitium of the Invs flox/flox ;Cdh16-Cre kidney. An attractive model is that INVS functions to signal terminal differentiation of renal epithelial cells and Invs mutant cells fail to exit cell cycle to enter the quiescent stage. Currently, the precise nature of the signals from mutant cilia and epithelial cells remains unknown. Previous studies on the fibrotic response triggered by kidney injury detected upregulation of multiple developmental pathways, including TGFβ, WNT, notch, and SHH (reviewed in Humphreys, 2018), providing ample candidate pathways for future analysis.
Previous work suggests that INVS is specifically localized to a proximal segment of the cilium dubbed the Inversin compartment (Shiba et al., 2009). Super-resolution imaging revealed that INVS is localized to a novel fibril-like structure in the Inversin compartment (Bennett et al., 2020). Moreover, INVS interacts with multiple infantile NPHP proteins, including NPHP3, ANKS6 and NEK8; and is required for concentrating these proteins to the Inversin compartment (Shiba et al., 2009;Czarnecki et al., 2015). On the other hand, INVS is dispensable for cilia biogenesis and is also found in the plasma membrane Nürnberger et al., 2002), mitotic spindle  and stress granule (Estrada Mallarino et al., 2020). It therefore remains possible that INVS functions outside of the cilium. To test the relationship between cilia and INVS, we removed cilia genetically in the Invs flox/flox ;Cdh16-Cre model by generating Ift88 flox/flox ; Invs flox/flox ;Cdh16-Cre double knockout mouse. Significantly, Ift88 inactivation partially reduced cell proliferation, suppressed the severity of renal cyst and fibrosis, and improved kidney function in Invs mutants. This result suggests that, similar to polycystins, INVS functions to inhibit a cilia-dependent pro-cystic and pro-fibrotic pathway ( Figure 7K). Although cilia-dependent, the downstream pathway could potentially operate outside of cilia and receive parallel signals from both ciliary activity and INVS. Interestingly, genes encoding Inversin compartment components seem to form a unique functional module among the many NPHP genes. In human, mutations in Inversin compartment genes are associated with the infantile form of NPHP. In mouse, Anks6, Invs, Nphp3, and Nek8 mutants develop kidney cysts, while mutants of several other NPHP genes show no obvious kidney phenotypes (Won et al., 2011;Jiang et al., 2008;Ronquillo et al., 2016;Morgan et al., 1998;Phillips et al., 2004;Czarnecki et al., 2015;Bergmann et al., 2008;Liu et al., 2002;Omran et al., 2001). Our results exclude the ciliary localization of PC2 as a potential mechanism for the INVS mediated pathway. However, whether this pathway is related or distinct from the polycystin-mediated CDCA pathway is currently unresolved ( Figure 7K). Understanding the Inversin compartment could provide critical clues to the molecular mechanisms linking cilia, polycystins, cyst formation, and fibrosis.
NPHP is a major cause of ESRD in children and young adults (Hildebrandt et al., 2009). Although the vasopressin V2 receptor antagonists OPC31260 and Tolvaptan have been shown to partially suppress phenotypic progression in Nphp3 pcy mutant mice, no directed therapy has been established for this disease (Aihara et al., 2014;Gattone et al., 2003). Epithelial cysts and macrophages have become the focus in efforts to identify novel treatment for cystic diseases. The early and severe fibrotic response in Invs mutants suggest that stromal cells might also serve as an important target. In this study, we show that the pan HDACI VPA reduces cell proliferation, slows down the progression of disease phenotypes and partially preserves renal function in Invs mutant mice. Although the expression of many genes is changed by HDACIs, clinical treatment window for these compounds exists. For example, at least four new HDACIs have been approved by FDA to treat lymphomas (Cappellacci et al., 2020). In addition, VPA is a drug that has been used to treat epilepsy with a known safety profile. Because of the broad target range of pan HDACIs, the precise target(s) contributing to the phenotypic improvement in Invs mutants remains unknown. It is attractive to speculate that a broad inhibition of multiple pathways, among them cell cycle control, by HDACIs might contribute to the effectiveness of this treatment. In the meantime, it will be informative to investigate whether the impact of HDACI treatment is HDAC type specific, thus narrowing down the list of potentially relevant targets. It is important to note that VPA could affect targets other than HDACs and testing newly approved HDACIs will provide useful insights.

Continued on next page
Key resources

Mouse breeding
Embryonic stem cell clones containing the Invs tm1a allele were purchased from EUCOMM and injected into C57BL/6 J blastocysts to generate chimeric mice by Yale Genome Editing Center. Chimeric animals were mated to C57BL/6 J mice to generate Invs tm1a/+ mice. The Invs flox allele was generated by crossing Invs tm1a/+ carrier mice with a deleter line expressing FLPe recombinase (Rodríguez et al., 2000) and backcrossed with C57BL/6 J for five generations. Cdh16-Cre and Pkhd1-Cre mice (Ma et al., 2013;Shao et al., 2002b)  Immunostaining Mouse kidneys were fixed by 4% PFA at 4 °C overnight, embedded in OCT and cryo-sectioned at 5 µm. Sections were incubated in blocking buffer (10% FBS in PBS with 0.1% Tween-20 (PBST)) for 1 hr at room temperature and then in primary antibody diluted with blocking overnight at 4 °C. Slides were then washed three time with PBST for 10 min each, incubated in secondary antibody diluted with blocking buffer for 1-2 hr at room temperature protected from light, followed by washing three times with PBST, mounted in ProLong Gold antifade reagent with DAPI (Molecular Probes by Life Technologies REF P36935) and imaged. The primary antibodies and lectins used are specified in Key Resources Table. Anti-PC2 (YCC2) has been described before (Cai et al., 1999).
Primary culture of mouse kidney interstitial cells A previously described protocol were followed (Nakai et al., 2021). Briefly, Invs flox/flox ;Foxd1Cre and sibling control mice were sacrificed at P12. Kidneys were harvested and the fibrous capsule was removed using forceps. Kidney tissue was minced with razor blade in 1 mL 0.25% trypsin (Gibco 25200-056) on a Petri dish at room temperature and incubated at 37 °C for 30-45 min. Trypsin activity was quenched by adding 4 mL DMEM (Gibco 11965-092) with Antibiotic-Antimycotic (Gibco 15240-062) and 10% FBS (Gibco 16140-071) per dish. Tissues were broken up by pipetting 10-20 times. Cell suspension was filtered through a 100 μm sterile cell strainer (Fisher Scientific #22363549) and 40 μm sterile cell strainer (Fisher Scientific #22363549). Cell suspension was centrifuged at 1000 g for 5 min and supernatant was removed. Cell pellet was resuspended in media. Cells were seeded on Microscope Cover Glasses (Glaswarenfabrik Kari Hecht GmbH& Co. 92100100030) placed in cell culture dishes (FALCON 353001). Cells were allowed to grow to confluency and subjected to serum starvation for 24 hr before fixation. Cells were fixed with MeOH at -20 °C for immunostaining of INVS and with 4% PFA for immunostaining of ANKS6.

Cystic index measurement
Cystic index was measured as previously published (Shibazaki et al., 2008). Briefly, two sections were analyzed for each animal. Hematoxylin/eosin-stained sections were scanned using the scan slide module in Metamorph v.7.1 acquisition software (Universal Imaging). Cystic and non-cystic areas were measured by the Region Statistics feature in Metamorph. Cystic index is defined as cystic area to total kidney area in each section.  (Jonassen et al., 2008).

Generation of Invs -/cells through CRISPR-Cas9
mIMCD3-cells were purchased from ATCC (CRL-2123) and used before passage 25. MycoAlert Mycoplamsma Detction Kit (Lonza, REF LT-07-118) was used to verify that these cells were negative for mycoplasma. RNA-seq validated that the cells were of murine origin. The morphology and karyotype of the cells were also consistent with known profiles of mIMCD3 cells (epithelial, near triploid). A pool of three plasmids for CRISPR/Cas9 knockout of Invs (SC-421144) was purchased from Santa Cruz Biotechnology and used following the manufacturer's protocol. These plasmids direct the expression of the reporter GFP and sgRNAs against the 5' side of Invs. mIMCD3 cells were electroporated and electroporation efficiency was monitored by GFP signal. Single colonies were selected by puromycin treatment. The target regions were PCR amplified and subjected to fragment analysis. Clones with changed size of target regions were further analyzed by RT-PCR using 5'-GAAC ACCA CTTA TGTA CTGT GTG-3' and 5'-GATG CTGC AAAA ACAC TTTG AC-3'. PCR products were cloned into the TA TOPO vector and sequenced. Clones CR81 and CR84 contains indels that lead to frameshift and premature stop codon after the first 105 and 90 amino acid of INVS, respectively.

Quantification of PCNA and pHH3-positive cells
At least three randomly selected field in the cortex region and at least 1000 DBA positive or interstitial cells outside of regions encircled by Laminin were evaluated per animal. Three animals were evaluated for each condition.

Protein extraction, SDS-PAGE, western blot, and quantification
Mouse kidney was homogenized in ice-cold Tris lysis buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1% NP40 and freshly added 0.1 mM DTT) with Complete EDTA-free Protease inhibitor (Roche, 11836170001) and PhosSTOP phosphatase inhibitor (Roche, 04906837001). Lysates were cleared by centrifugation at 12,000 g at 4 °C for 5 min. Protein concentration was measured using Protein Assay Dye Reagent Concentrate (Bio-Rad #500-0006). Equal amounts of total protein were loaded, separated on Mini-Protean TGX Precast Gels (Bio-Rad) and transferred to PVDF membranes. Membranes were incubated with 5% Non-fat milk in TBST (TBS +0.1% Tween-20) at room temperature for 1 hr and then incubated with primary antibodies at 4 °C overnight, followed by HRP-conjugated secondary antibodies (Jackson ImmunoResearch, 1:5000 in 5% Non-fat milk). Signal was detected using Super-Signal West Pico PLUS Chemiluminescent Substrate (34580, Thermo Scientific). Western blot signals from four groups of mouse kidneys were quantified using ImageJ IntDen software. Densitometric ratios are relative to HSP90 on the same blot. Fold change is shown relative to the mean of the ratio in the control samples on each blot, which is set to a value of 1.0. The antibodies used are specified in Key Resources Table. VPA reatment VPA was purchased from Acros Organics (CAS: 1069-66-5). Daily IP injection was performed from P10 to P28. The VPA treatment group received 200 mg/kg VPA in 15% DMSO and 85% saline, whereas the vehicle control group received the same volume of 15% DMSO and 85% saline without VPA.

Statistical analyses
Unpaired t-test between mutants and control animals (Figures 1-4, Figure 4-figure supplement 1 and Figure 7-figure supplement 1) and one-way ANOVA analysis followed by Šidák's test between selected pairs among multiple groups (Figures 5-7) were performed using GraphPad Prism 9.2.0. p=0.05 was used as the threshold for statistical significance. All data are presented as mean ± SEM.