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.

Primary cilia are widely distributed cell surface organelles that function as signaling hubs for almost all vertebrate cell types (Schneider et al., 2005; Rohatgi et al., 2007; Lancaster et al., 2011; Huangfu et al., 2003; Haycraft et al., 2005; Corbit et al., 2008). 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 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.

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 (Phillips et al., 2004; Nürnberger et al., 2002), mitotic spindle (Nürnberger et al., 2004) 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.

No large scale datasets generated. Data analyzed can be found in source data files for Figures 1–7, Figure 4—figure supplement 1 and Figure 7—figure supplement 1.

1. 1. Aihara M
   2. Fujiki H
   3. Mizuguchi H
   4. Hattori K
   5. Ohmoto K
   6. Ishikawa M
   7. Nagano K
   8. Yamamura Y

   (2014) Tolvaptan delays the onset of end-stage renal disease in a polycystic kidney disease model by suppressing increases in kidney volume and renal injury

   The Journal of Pharmacology and Experimental Therapeutics 349:258–267.

   https://doi.org/10.1124/jpet.114.213256

   • PubMed
   • Google Scholar
2. 1. Barr MM
   2. Sternberg PW

   (1999) A polycystic kidney-disease gene homologue required for male mating behaviour in C. elegans

   Nature 401:386–389.

   https://doi.org/10.1038/43913

   • PubMed
   • Google Scholar
3. 1. Bennett HW
   2. Gustavsson A-K
   3. Bayas CA
   4. Petrov PN
   5. Mooney N
   6. Moerner WE
   7. Jackson PK

   (2020) Novel fibrillar structure in the inversin compartment of primary cilia revealed by 3D single-molecule superresolution microscopy

   Molecular Biology of the Cell 31:619–639.

   https://doi.org/10.1091/mbc.E19-09-0499

   • PubMed
   • Google Scholar
4. 1. Bergmann C
   2. Fliegauf M
   3. Brüchle NO
   4. Frank V
   5. Olbrich H
   6. Kirschner J
   7. Schermer B
   8. Schmedding I
   9. Kispert A
   10. Kränzlin B
   11. Nürnberg G
   12. Becker C
   13. Grimm T
   14. Girschick G
   15. Lynch SA
   16. Kelehan P
   17. Senderek J
   18. Neuhaus TJ
   19. Stallmach T
   20. Zentgraf H
   21. Nürnberg P
   22. Gretz N
   23. Lo C
   24. Lienkamp S
   25. Schäfer T
   26. Walz G
   27. Benzing T
   28. Zerres K
   29. Omran H

   (2008) Loss of nephrocystin-3 function can cause embryonic lethality, meckel-gruber-like syndrome, situs inversus, and renal-hepatic-pancreatic dysplasia

   American Journal of Human Genetics 82:959–970.

   https://doi.org/10.1016/j.ajhg.2008.02.017

   • PubMed
   • Google Scholar
5. 1. Cai Y
   2. Maeda Y
   3. Cedzich A
   4. Torres VE
   5. Wu G
   6. Hayashi T
   7. Mochizuki T
   8. Park JH
   9. Witzgall R
   10. Somlo S

   (1999) Identification and characterization of polycystin-2, the PKD2 gene product

   The Journal of Biological Chemistry 274:28557–28565.

   https://doi.org/10.1074/jbc.274.40.28557

   • PubMed
   • Google Scholar
6. 1. Cai Y
   2. Fedeles SV
   3. Dong K
   4. Anyatonwu G
   5. Onoe T
   6. Mitobe M
   7. Gao J-D
   8. Okuhara D
   9. Tian X
   10. Gallagher A-R
   11. Tang Z
   12. Xie X
   13. Lalioti MD
   14. Lee A-H
   15. Ehrlich BE
   16. Somlo S

   (2014) Altered trafficking and stability of polycystins underlie polycystic kidney disease

   The Journal of Clinical Investigation 124:5129–5144.

   https://doi.org/10.1172/JCI67273

   • PubMed
   • Google Scholar
7. 1. Cao Y
   2. Semanchik N
   3. Lee SH
   4. Somlo S
   5. Barbano PE
   6. Coifman R
   7. Sun Z

   (2009) Chemical modifier screen identifies HDAC inhibitors as suppressors of PKD models

   PNAS 106:21819–21824.

   https://doi.org/10.1073/pnas.0911987106

   • PubMed
   • Google Scholar
8. 1. Cappellacci L
   2. Perinelli DR
   3. Maggi F
   4. Grifantini M
   5. Petrelli R

   (2020) Recent progress in histone deacetylase inhibitors as anticancer agents

   Current Medicinal Chemistry 27:2449–2493.

   https://doi.org/10.2174/0929867325666181016163110

   • PubMed
   • Google Scholar
9. 1. Corbit KC
   2. Shyer AE
   3. Dowdle WE
   4. Gaulden J
   5. Singla V
   6. Chen M-H
   7. Chuang P-T
   8. Reiter JF

   (2008) Kif3A constrains beta-catenin-dependent Wnt signalling through dual ciliary and non-ciliary mechanisms

   Nature Cell Biology 10:70–76.

   https://doi.org/10.1038/ncb1670

   • PubMed
   • Google Scholar
10. 1. Czarnecki PG
    2. Gabriel GC
    3. Manning DK
    4. Sergeev M
    5. Lemke K
    6. Klena NT
    7. Liu X
    8. Chen Y
    9. Li Y
    10. San Agustin JT
    11. Garnaas MK
    12. Francis RJ
    13. Tobita K
    14. Goessling W
    15. Pazour GJ
    16. Lo CW
    17. Beier DR
    18. Shah JV

    (2015) ANKS6 is the critical activator of Nek8 kinase in embryonic situs determination and organ patterning

    Nature Communications 6:6023.

    https://doi.org/10.1038/ncomms7023

    • PubMed
    • Google Scholar
11. 1. Estrada Mallarino L
    2. Engel C
    3. Ilık İA
    4. Maticzka D
    5. Heyl F
    6. Müller B
    7. Yakulov TA
    8. Dengjel J
    9. Backofen R
    10. Akhtar A
    11. Walz G

    (2020) Nephronophthisis gene products display RNA-binding properties and are recruited to stress granules

    Scientific Reports 10:15954.

    https://doi.org/10.1038/s41598-020-72905-8

    • PubMed
    • Google Scholar
12. 1. Fliegauf M
    2. Horvath J
    3. von Schnakenburg C
    4. Olbrich H
    5. Müller D
    6. Thumfart J
    7. Schermer B
    8. Pazour GJ
    9. Neumann HPH
    10. Zentgraf H
    11. Benzing T
    12. Omran H

    (2006) Nephrocystin specifically localizes to the transition zone of renal and respiratory cilia and photoreceptor connecting cilia

    Journal of the American Society of Nephrology 17:2424–2433.

    https://doi.org/10.1681/ASN.2005121351

    • PubMed
    • Google Scholar
13. 1. Gattone VH
    2. Wang X
    3. Harris PC
    4. Torres VE

    (2003) Inhibition of renal cystic disease development and progression by a vasopressin V2 receptor antagonist

    Nature Medicine 9:1323–1326.

    https://doi.org/10.1038/nm935

    • PubMed
    • Google Scholar
14. 1. Haycraft CJ
    2. Banizs B
    3. Aydin-Son Y
    4. Zhang Q
    5. Michaud EJ
    6. Yoder BK

    (2005) Gli2 and gli3 localize to cilia and require the intraflagellar transport protein polaris for processing and function

    PLOS Genetics 1:e53.

    https://doi.org/10.1371/journal.pgen.0010053

    • PubMed
    • Google Scholar
15. 1. Hildebrandt F
    2. Attanasio M
    3. Otto E

    (2009) Nephronophthisis: disease mechanisms of a ciliopathy

    Journal of the American Society of Nephrology 20:23–35.

    https://doi.org/10.1681/ASN.2008050456

    • PubMed
    • Google Scholar
16. 1. Huangfu D
    2. Liu A
    3. Rakeman AS
    4. Murcia NS
    5. Niswander L
    6. Anderson KV

    (2003) Hedgehog signalling in the mouse requires intraflagellar transport proteins

    Nature 426:83–87.

    https://doi.org/10.1038/nature02061

    • PubMed
    • Google Scholar
17. 1. Hum S
    2. Rymer C
    3. Schaefer C
    4. Bushnell D
    5. Sims-Lucas S

    (2014) Ablation of the renal stroma defines its critical role in nephron progenitor and vasculature patterning

    PLOS ONE 9:e88400.

    https://doi.org/10.1371/journal.pone.0088400

    • PubMed
    • Google Scholar
18. 1. Humphreys BD
    2. Lin S-L
    3. Kobayashi A
    4. Hudson TE
    5. Nowlin BT
    6. Bonventre JV
    7. Valerius MT
    8. McMahon AP
    9. Duffield JS

    (2010) Fate tracing reveals the pericyte and not epithelial origin of myofibroblasts in kidney fibrosis

    The American Journal of Pathology 176:85–97.

    https://doi.org/10.2353/ajpath.2010.090517

    • PubMed
    • Google Scholar
19. 1. Humphreys BD

    (2018) Mechanisms of renal fibrosis

    Annual Review of Physiology 80:309–326.

    https://doi.org/10.1146/annurev-physiol-022516-034227

    • PubMed
    • Google Scholar
20. 1. Jiang S-T
    2. Chiou Y-Y
    3. Wang E
    4. Lin H-K
    5. Lee S-P
    6. Lu H-Y
    7. Wang C-KL
    8. Tang M-J
    9. Li H

    (2008) Targeted disruption of Nphp1 causes male infertility due to defects in the later steps of sperm morphogenesis in mice

    Human Molecular Genetics 17:3368–3379.

    https://doi.org/10.1093/hmg/ddn231

    • PubMed
    • Google Scholar
21. 1. Jonassen JA
    2. San Agustin J
    3. Follit JA
    4. Pazour GJ

    (2008) Deletion of IFT20 in the mouse kidney causes misorientation of the mitotic spindle and cystic kidney disease

    The Journal of Cell Biology 183:377–384.

    https://doi.org/10.1083/jcb.200808137

    • PubMed
    • Google Scholar
22. 1. Lancaster MA
    2. Schroth J
    3. Gleeson JG

    (2011) Subcellular spatial regulation of canonical Wnt signalling at the primary cilium

    Nature Cell Biology 13:700–707.

    https://doi.org/10.1038/ncb2259

    • PubMed
    • Google Scholar
23. 1. Li Y
    2. Tian X
    3. Ma M
    4. Jerman S
    5. Kong S
    6. Somlo S
    7. Sun Z

    (2016) Deletion of ADP ribosylation factor-like GTPase 13b leads to kidney cysts

    Journal of the American Society of Nephrology 27:3628–3638.

    https://doi.org/10.1681/ASN.2015091004

    • PubMed
    • Google Scholar
24. 1. Liu S
    2. Lu W
    3. Obara T
    4. Kuida S
    5. Lehoczky J
    6. Dewar K
    7. Drummond IA
    8. Beier DR

    (2002) A defect in A novel nek-family kinase causes cystic kidney disease in the mouse and in zebrafish

    Development 129:5839–5846.

    https://doi.org/10.1242/dev.00173

    • PubMed
    • Google Scholar
25. 1. Lv X
    2. Qiu J
    3. Hao T
    4. Zhang H
    5. Jiang H
    6. Tan Y

    (2021) Hdac inhibitor trichostatin A suppresses adipogenesis in 3T3-L1 preadipocytes

    Aging 13:17489–17498.

    https://doi.org/10.18632/aging.203238

    • PubMed
    • Google Scholar
26. 1. Ma M
    2. Tian X
    3. Igarashi P
    4. Pazour GJ
    5. Somlo S

    (2013) Loss of cilia suppresses cyst growth in genetic models of autosomal dominant polycystic kidney disease

    Nature Genetics 45:1004–1012.

    https://doi.org/10.1038/ng.2715

    • PubMed
    • Google Scholar
27. 1. Mochizuki T
    2. Saijoh Y
    3. Tsuchiya K
    4. Shirayoshi Y
    5. Takai S
    6. Taya C
    7. Yonekawa H
    8. Yamada K
    9. Nihei H
    10. Nakatsuji N
    11. Overbeek PA
    12. Hamada H
    13. Yokoyama T

    (1998) Cloning of inv, a gene that controls left/right asymmetry and kidney development

    Nature 395:177–181.

    https://doi.org/10.1038/26006

    • PubMed
    • Google Scholar
28. 1. Morgan D
    2. Turnpenny L
    3. Goodship J
    4. Dai W
    5. Majumder K
    6. Matthews L
    7. Gardner A
    8. Schuster G
    9. Vien L
    10. Harrison W
    11. Elder FF
    12. Penman-Splitt M
    13. Overbeek P
    14. Strachan T

    (1998) Inversin, a novel gene in the vertebrate left-right axis pathway, is partially deleted in the inv mouse

    Nature Genetics 20:149–156.

    https://doi.org/10.1038/2450

    • PubMed
    • Google Scholar
29. 1. Nadasdy T
    2. Laszik Z
    3. Lajoie G
    4. Blick KE
    5. Wheeler DE
    6. Silva FG

    (1995) Proliferative activity of cyst epithelium in human renal cystic diseases

    Journal of the American Society of Nephrology 5:1462–1468.

    https://doi.org/10.1681/ASN.V571462

    • PubMed
    • Google Scholar
30. 1. Nakai T
    2. Iwamura Y
    3. Suzuki N

    (2021) Efficient isolation of interstitial fibroblasts directly from mouse kidneys or indirectly after ex vivo expansion

    STAR Protocols 2:100826.

    https://doi.org/10.1016/j.xpro.2021.100826

    • PubMed
    • Google Scholar
31. 1. Nürnberger J
    2. Bacallao RL
    3. Phillips CL

    (2002) Inversin forms a complex with catenins and N-cadherin in polarized epithelial cells

    Molecular Biology of the Cell 13:3096–3106.

    https://doi.org/10.1091/mbc.e02-04-0195

    • PubMed
    • Google Scholar
32. 1. Nürnberger J
    2. Kribben A
    3. Opazo Saez A
    4. Heusch G
    5. Philipp T
    6. Phillips CL

    (2004) The invs gene encodes a microtubule-associated protein

    Journal of the American Society of Nephrology 15:1700–1710.

    https://doi.org/10.1097/01.asn.0000128291.30249.d0

    • PubMed
    • Google Scholar
33. 1. Ó hAinmhire E
    2. Wu H
    3. Muto Y
    4. Donnelly EL
    5. Machado FG
    6. Fan LX
    7. Chang-Panesso M
    8. Humphreys BD

    (2019) A conditionally immortalized gli1-positive kidney mesenchymal cell line models myofibroblast transition

    American Journal of Physiology. Renal Physiology 316:F63–F75.

    https://doi.org/10.1152/ajprenal.00460.2018

    • PubMed
    • Google Scholar
34. 1. Omran H
    2. Häffner K
    3. Burth S
    4. Fernandez C
    5. Fargier B
    6. Villaquiran A
    7. Nothwang H-G
    8. Schnittger S
    9. Lehrach H
    10. Woo D
    11. Brandis M
    12. Sudbrak R
    13. Hildebrandt F

    (2001) Human adolescent nephronophthisis: gene locus synteny with polycystic kidney disease in pcy mice

    Journal of the American Society of Nephrology 12:107–113.

    https://doi.org/10.1681/ASN.V121107

    • PubMed
    • Google Scholar
35. 1. Otto EA
    2. Schermer B
    3. Obara T
    4. O’Toole JF
    5. Hiller KS
    6. Mueller AM
    7. Ruf RG
    8. Hoefele J
    9. Beekmann F
    10. Landau D
    11. Foreman JW
    12. Goodship JA
    13. Strachan T
    14. Kispert A
    15. Wolf MT
    16. Gagnadoux MF
    17. Nivet H
    18. Antignac C
    19. Walz G
    20. Drummond IA
    21. Benzing T
    22. Hildebrandt F

    (2003) Mutations in INVS encoding inversin cause nephronophthisis type 2, linking renal cystic disease to the function of primary cilia and left-right axis determination

    Nature Genetics 34:413–420.

    https://doi.org/10.1038/ng1217

    • PubMed
    • Google Scholar
36. 1. Otto EA
    2. Trapp ML
    3. Schultheiss UT
    4. Helou J
    5. Quarmby LM
    6. Hildebrandt F

    (2008) Nek8 mutations affect ciliary and centrosomal localization and may cause nephronophthisis

    Journal of the American Society of Nephrology 19:587–592.

    https://doi.org/10.1681/ASN.2007040490

    • PubMed
    • Google Scholar
37. 1. Pazour GJ
    2. Dickert BL
    3. Vucica Y
    4. Seeley ES
    5. Rosenbaum JL
    6. Witman GB
    7. Cole DG

    (2000) Chlamydomonas IFT88 and its mouse homologue, polycystic kidney disease gene Tg737, are required for assembly of cilia and flagella

    The Journal of Cell Biology 151:709–718.

    https://doi.org/10.1083/jcb.151.3.709

    • PubMed
    • Google Scholar
38. 1. Pazour GJ
    2. San Agustin JT
    3. Follit JA
    4. Rosenbaum JL
    5. Witman GB

    (2002) Polycystin-2 localizes to kidney cilia and the ciliary level is elevated in orpk mice with polycystic kidney disease

    Current Biology 12:R378–R380.

    https://doi.org/10.1016/s0960-9822(02)00877-1

    • PubMed
    • Google Scholar
39. 1. Phillips CL
    2. Miller KJ
    3. Filson AJ
    4. Nürnberger J
    5. Clendenon JL
    6. Cook GW
    7. Dunn KW
    8. Overbeek PA
    9. Gattone VH
    10. Bacallao RL

    (2004) Renal cysts of inv/inv mice resemble early infantile nephronophthisis

    Journal of the American Society of Nephrology 15:1744–1755.

    https://doi.org/10.1097/01.asn.0000131520.07008.b3

    • PubMed
    • Google Scholar
40. 1. Rodríguez CI
    2. Buchholz F
    3. Galloway J
    4. Sequerra R
    5. Kasper J
    6. Ayala R
    7. Stewart AF
    8. Dymecki SM

    (2000) High-Efficiency deleter mice show that flpe is an alternative to Cre-loxP

    Nature Genetics 25:139–140.

    https://doi.org/10.1038/75973

    • PubMed
    • Google Scholar
41. 1. Rohatgi R
    2. Milenkovic L
    3. Scott MP

    (2007) Patched1 regulates hedgehog signaling at the primary cilium

    Science 317:372–376.

    https://doi.org/10.1126/science.1139740

    • PubMed
    • Google Scholar
42. 1. Ronquillo CC
    2. Hanke-Gogokhia C
    3. Revelo MP
    4. Frederick JM
    5. Jiang L
    6. Baehr W

    (2016) Ciliopathy-Associated IQCB1/NPHP5 protein is required for mouse photoreceptor outer segment formation

    FASEB Journal 30:3400–3412.

    https://doi.org/10.1096/fj.201600511R

    • PubMed
    • Google Scholar
43. 1. Sang L
    2. Miller JJ
    3. Corbit KC
    4. Giles RH
    5. Brauer MJ
    6. Otto EA
    7. Baye LM
    8. Wen X
    9. Scales SJ
    10. Kwong M
    11. Huntzicker EG
    12. Sfakianos MK
    13. Sandoval W
    14. Bazan JF
    15. Kulkarni P
    16. Garcia-Gonzalo FR
    17. Seol AD
    18. O’Toole JF
    19. Held S
    20. Reutter HM
    21. Lane WS
    22. Rafiq MA
    23. Noor A
    24. Ansar M
    25. Devi ARR
    26. Sheffield VC
    27. Slusarski DC
    28. Vincent JB
    29. Doherty DA
    30. Hildebrandt F
    31. Reiter JF
    32. Jackson PK

    (2011) Mapping the NPHP-JBTS-MKS protein network reveals ciliopathy disease genes and pathways

    Cell 145:513–528.

    https://doi.org/10.1016/j.cell.2011.04.019

    • PubMed
    • Google Scholar
44. 1. Schneider L
    2. Clement CA
    3. Teilmann SC
    4. Pazour GJ
    5. Hoffmann EK
    6. Satir P
    7. Christensen ST

    (2005) PDGFRalphaalpha signaling is regulated through the primary cilium in fibroblasts

    Current Biology 15:1861–1866.

    https://doi.org/10.1016/j.cub.2005.09.012

    • PubMed
    • Google Scholar
45. 1. Shao X
    2. Johnson JE
    3. Richardson JA
    4. Hiesberger T
    5. Igarashi P

    (2002a) A minimal ksp-cadherin promoter linked to A green fluorescent protein reporter gene exhibits tissue-specific expression in the developing kidney and genitourinary tract

    Journal of the American Society of Nephrology 13:1824–1836.

    https://doi.org/10.1097/01.asn.0000016443.50138.cd

    • PubMed
    • Google Scholar
46. 1. Shao X
    2. Somlo S
    3. Igarashi P

    (2002b) Epithelial-specific cre/lox recombination in the developing kidney and genitourinary tract

    Journal of the American Society of Nephrology 13:1837–1846.

    https://doi.org/10.1097/01.asn.0000016444.90348.50

    • PubMed
    • Google Scholar
47. 1. Shiba D
    2. Yamaoka Y
    3. Hagiwara H
    4. Takamatsu T
    5. Hamada H
    6. Yokoyama T

    (2009) Localization of inv in a distinctive intraciliary compartment requires the C-terminal ninein-homolog-containing region

    Journal of Cell Science 122:44–54.

    https://doi.org/10.1242/jcs.037408

    • PubMed
    • Google Scholar
48. 1. Shiba D
    2. Manning DK
    3. Koga H
    4. Beier DR
    5. Yokoyama T

    (2010) Inv acts as a molecular anchor for nphp3 and NEK8 in the proximal segment of primary cilia

    Cytoskeleton 67:112–119.

    https://doi.org/10.1002/cm.20428

    • PubMed
    • Google Scholar
49. 1. Shibazaki S
    2. Yu Z
    3. Nishio S
    4. Tian X
    5. Thomson RB
    6. Mitobe M
    7. Louvi A
    8. Velazquez H
    9. Ishibe S
    10. Cantley LG
    11. Igarashi P
    12. Somlo S

    (2008) Cyst formation and activation of the extracellular regulated kinase pathway after kidney specific inactivation of PKD1

    Human Molecular Genetics 17:1505–1516.

    https://doi.org/10.1093/hmg/ddn039

    • PubMed
    • Google Scholar
50. 1. Srivastava S
    2. Molinari E
    3. Raman S
    4. Sayer JA

    (2017) Many genes-one disease? genetics of nephronophthisis (NPHP) and NPHP-associated disorders

    Frontiers in Pediatrics 5:287.

    https://doi.org/10.3389/fped.2017.00287

    • PubMed
    • Google Scholar
51. 1. Takakura A
    2. Contrino L
    3. Zhou X
    4. Bonventre JV
    5. Sun Y
    6. Humphreys BD
    7. Zhou J

    (2009) Renal injury is a third hit promoting rapid development of adult polycystic kidney disease

    Human Molecular Genetics 18:2523–2531.

    https://doi.org/10.1093/hmg/ddp147

    • PubMed
    • Google Scholar
52. 1. Williams CL
    2. Masyukova SV
    3. Yoder BK

    (2010) Normal ciliogenesis requires synergy between the cystic kidney disease genes MKS-3 and NPHP-4

    Journal of the American Society of Nephrology 21:782–793.

    https://doi.org/10.1681/ASN.2009060597

    • PubMed
    • Google Scholar
53. 1. Winkelbauer ME
    2. Schafer JC
    3. Haycraft CJ
    4. Swoboda P
    5. Yoder BK

    (2005) The C. elegans homologs of nephrocystin-1 and nephrocystin-4 are cilia transition zone proteins involved in chemosensory perception

    Journal of Cell Science 118:5575–5587.

    https://doi.org/10.1242/jcs.02665

    • PubMed
    • Google Scholar
54. 1. Won J
    2. Marín de Evsikova C
    3. Smith RS
    4. Hicks WL
    5. Edwards MM
    6. Longo-Guess C
    7. Li T
    8. Naggert JK
    9. Nishina PM

    (2011) NPHP4 is necessary for normal photoreceptor ribbon synapse maintenance and outer segment formation, and for sperm development

    Human Molecular Genetics 20:482–496.

    https://doi.org/10.1093/hmg/ddq494

    • PubMed
    • Google Scholar
55. 1. Yoder BK
    2. Hou X
    3. Guay-Woodford LM

    (2002) The polycystic kidney disease proteins, polycystin-1, polycystin-2, polaris, and cystin, are co-localized in renal cilia

    Journal of the American Society of Nephrology 13:2508–2516.

    https://doi.org/10.1097/01.asn.0000029587.47950.25

    • PubMed
    • Google Scholar
56. 1. Yokoyama T
    2. Copeland NG
    3. Jenkins NA
    4. Montgomery CA
    5. Elder FF
    6. Overbeek PA

    (1993) Reversal of left-right asymmetry: a situs inversus mutation

    Science 260:679–682.

    https://doi.org/10.1126/science.8480178

    • PubMed
    • Google Scholar
57. 1. Yoshiba S
    2. Shiratori H
    3. Kuo IY
    4. Kawasumi A
    5. Shinohara K
    6. Nonaka S
    7. Asai Y
    8. Sasaki G
    9. Belo JA
    10. Sasaki H
    11. Nakai J
    12. Dworniczak B
    13. Ehrlich BE
    14. Pennekamp P
    15. Hamada H

    (2012) Cilia at the node of mouse embryos sense fluid flow for left-right determination via PKD2

    Science 338:226–231.

    https://doi.org/10.1126/science.1222538

    • PubMed
    • Google Scholar
58. 1. Zerjatke T
    2. Gak IA
    3. Kirova D
    4. Fuhrmann M
    5. Daniel K
    6. Gonciarz M
    7. Müller D
    8. Glauche I
    9. Mansfeld J

    (2017) Quantitative cell cycle analysis based on an endogenous all-in-one reporter for cell tracking and classification

    Cell Reports 19:1953–1966.

    https://doi.org/10.1016/j.celrep.2017.05.022

    • PubMed
    • Google Scholar
59. 1. Zhang C
    2. Balbo B
    3. Ma M
    4. Zhao J
    5. Tian X
    6. Kluger Y
    7. Somlo S

    (2021) Cyclin-Dependent kinase 1 activity is a driver of cyst growth in polycystic kidney disease

    Journal of the American Society of Nephrology 32:41–51.

    https://doi.org/10.1681/ASN.2020040511

    • PubMed
    • Google Scholar
60. 1. Zhou D
    2. Fu H
    3. Liu S
    4. Zhang L
    5. Xiao L
    6. Bastacky SI
    7. Liu Y

    (2019) Early activation of fibroblasts is required for kidney repair and regeneration after injury

    FASEB Journal 33:12576–12587.

    https://doi.org/10.1096/fj.201900651RR

    • PubMed
    • Google Scholar

Decision letter after peer review:

Thank you for submitting your article "Inactivation of Nphp2 in renal epithelial cells drives infantile nephronophthisis like phenotypes in mouse" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by a Reviewing Editor and Martin Pollak as the Senior Editor. The reviewers have opted to remain anonymous.

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

1) Please provide an explanation for why the Nphp2 mutants in Figure 7, regardless of treatment status, have body weights that are significantly lower than the controls, with treated mutants even trending lower than their untreated mutant counterparts.

2) Please provide more details on how you controlled for possible genetic background effects as this factor could be an important confounder for the interpretation of your genetic interaction and valproic acid treatment studies.

3) Please acknowledge that the mechanism by which VPA is exerting its effects is not certain in this study as VPA has been reported to have numerous pharmacologic effects and targets and your study provides no direct evidence in support of any one specifically.

4) Please explain how your double knock-out studies "support a significant role of cilia in Nphp2 function in vivo" and can exclude that ciliary activity is operating in an Nphp2-independent, parallel fashion that modulates some common downstream pathways.

5) While most of the issues can either be addressed by more explanation or editing, the reviewers have asked for more data to support several of your conclusions. This could be in the form of additional data from timepoints later than 28 days to support your conclusions that Nphp2 loss in stromal cells is not important to cystic kidney disease or kidney fibrosis, since late cyst formation has been reported in mice with stromal inactivation of Pkd1. Alternatively, you can compare the effects of global activation of Nphp2 to epithelial cell-specific inactivation of Nphp2 to support your conclusion that stromal loss of Nphp2 has no effect on early, severe cystic disease.

6) Although this is not required, it would strengthen the paper to include more timepoints and data supporting their conclusion that fibrosis precedes cyst development in the Nphp2 epithelial cell ko model.

Reviewer #1 (Recommendations for the authors):

General comment: This is a solid study but its results and impact are quite modest. The study's impact would be greatly strengthened if it provided some insight into the nature of the mutant epithelial cell-"normal" stomal cell signaling, how impaired ciliogenesis ameliorates disease (beyond the reduction in cellular proliferation and fibrosis), or a more mechanistic understanding of the effects of VPA.

Reviewer #2 (Recommendations for the authors):

I have a few comments that might be addressed or discussed further to improve the interpretations of underlying mechanisms of Nphp2 function in renal homeostasis.

First, as the authors allude to the nature of the NPHP-regulated cilia-dependent pathway is related or distinct from the polycystin-mediated CDCA pathway is currently unresolved. The lack of PC2 trafficking in Nphp2 mutants suggests that the CDCA pathway might not at all be activated in the Nphp2 mutants. Rather NPHP2 could be functioning independently in regulating cystogenesis from PKD-regulated CDCA. Testing double mutants of Nphp2 with Pkd1/2 might resolve this issue.

Second, an independent interpretation of Nphp2 lack-induced cystogenesis being reduced partially from ciliary disruption is that the lack of Nphp2 or cilia regulates independent pathways with relation to fibrosis and epithelial regulation. Late assessment of phenotypes such as epithelial and stromal proliferation by the authors shows partial restorative effects. In fact, the authors show partial inhibition of stromal proliferation in Nphp2 loss from epithelial lack of cilia which is very interesting. As an early indicator of Nphp2 inactivation is stromal fibrosis and proliferation, does parallel cilia disruption prevent such early phenotypes? Does cilia disruption itself cause fibrosis at earlier stages? Such early assessment of phenotypes might be more insightful in the cilia single mutants and double mutants.

Third, as NPHP2 interacts with multiple infantile NPHP proteins, including NPHP3, ANKS6, and NEK8 in the inversin compartment, one way of ruling out if Nphp2 function is in cilia, would be to compare/discuss phenotypes with other mutants that disrupt these interacting genes. As NPHP2 is localized to a novel fibril-like structure in the inversin compartment in cilia, the elephant in the room is the lack of our current understanding of the inversin compartment's role in the ciliary composition. Such understanding could provide important mechanistic clues and might also provide insights into the role of inversin compartment in regulating L-R asymmetry.

Reviewer #3 (Recommendations for the authors):

Using a newly established floxed model of Nphp2, the authors show convincing data that Nphp2 loss in the kidney epithelium is critical to kidney cyst formation and fibrosis. In addition, the authors show compelling data that cilia loss in the setting of epithelial Nphp2 loss reduces kidney cyst severity. This cilia-dependent idea of PKD has been previously published in the setting of ADPKD, but remained unexplored for NPHP. No additional, or novel mechanistic insight is shown in the manuscript. The authors also suggest that Nphp2 loss in stromal cells is not important to cystic kidney disease or kidney fibrosis, but that data is underdeveloped. Lastly, the authors perform a preclinical trial using a broad-spectrum HDAC inhibitor, in their model, which shows efficacy in slowing disease. This experiment seems disconnected from the rest of the paper.

Ksp-cre model characterization: (1) It would be important to compare the Ksp-cre results to a global knockout of Nphp2 (CAGGc-cre) in terms of PKD severity. The authors stress the possible importance of epithelial-stromal communication. To better define this a comparison of epithelial versus global knock-out is needed. (2) It would be helpful to have a Kaplan-Meyer curve of the model. (3) Could the authors comment on what percentage of DBA or THP-positive tubules are cystic in the model? (4) The data supporting that fibrosis precedes cysts is underdeveloped. The SMA staining at P7 seems more intense than at P14 which is surprising (P14 and later is not quantified – cystic versus non-cystic regions), and vimentin is not analyzed at the later stages. (5) The data that epithelial Nphp2 loss results in higher proliferation in the interstitium are interesting; however, the authors provide no experiments/speculations on why that may be. Further, these analyses are done only at P7, and it is unclear how it pertains to stages when cysts develop/are present. And, it is unclear why for all other figures examining proliferation, the authors performed PCNA staining and western blotting of essential proteins, but not for this one.

Foxd1-cre model characterization: (1) The characterization of this model is underdeveloped. Animals need to be aged much longer to assure no phenotype is observed. (2) The experiments performed in Figure 4I are not intuitive. Based on the mouse model shown in Figure 1A it is unclear where eGFP expression should come from. (3) While this is intrinsically an issue with the Foxd1-cre model, the observation that was made that only 60% of all stromal cells lost Nphp2 (Figure 4J) is concerning to make such a strong conclusion that NPHP2 in the stromal compartment is not important to the NPHP phenotype (fibrosis or cysts).

NPHP2 and Cilia: (1) While the finding that the CDCA may be critical to NPHP, as well as ADPKD, is novel, the provided data is only descriptive. It would improve the paper drastically if the authors could provide data suggesting whether NPHP2 functions in the same pathway as PC1 or PC2 or a different one. (2) For their cell analyses, the authors should show successful loss of NPHP2 in their CRISPR clones as well as also analyze PC1. (3) Fibrosis was not analyzed.

HDAC preclinical trial: This data is not connected to anything else within the manuscript. It utilized a different cre than for all the other experiments and is not mechanistically linked to the story of whether NPHP2 function in epithelial or stromal cells is important to pathogenesis. Unless the authors can link the stories better or highlight why the HDAC function is important to the observations, this reviewer would suggest removing Figure 7.

Essential revisions:

1) Please provide an explanation for why the Nphp2 mutants in Figure 7, regardless of treatment status, have body weights that are significantly lower than the controls, with treated mutants even trending lower than their untreated mutant counterparts.

We examined the time course of body weight change more carefully and added Figure 7—figure supplement 1 to present the results. Although Invs^flox/flox;Pkhd1-Cre mice displayed reduced body weight at P28 in comparison to controls, this reduction was more moderate than that of Invs^flox/flox;Cdh16-Cre mice (Figure 7—figure supplement 1A). Notably, the trend of body weight difference started at around P21 in both Invs^flox/flox;Pkhd1-Cre and Invs^flox/flox;Cdh16-Cre mice, coinciding with weaning (Figure 7—figure supplement 1B). It is possible that mutants with compromised kidney function were less capable to thrive and gain weight at around this transition time. In terms of VPA treatment, body weight trended down in both wild type and mutant mice subjected to the treatment, although the difference did not reach statistical significance (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: the HDAC inhibitor Trichostatin A was shown to inhibit adipogenesis (PMID: 34232916) and 4-hexylresorcinol, another HDAC inhibitor, reduced body weight in treated rats (PMID: 34445640). To include the additional data and references, we added the following in the Results section:

"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)."

“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 (51)."

2) Please provide more details on how you controlled for possible genetic background effects as this factor could be an important confounder for the interpretation of your genetic interaction and valproic acid treatment studies.

In Figure 5 (genetic interaction), all mice are in C57BL/6J background. In this experiment, multiple genotypes (i.g. Invs^flox/flox;Cdh16-Cre, Invs^flox/flox;Ift88^flox/flox;Cdh16-Cre and Ift88^flox/flox;Cdh16-Cre) were analyzed. Although control and mutant littermates were used, nonlittermates had to be included in some cases because of the limited number of animals per litter and low yield of desired genotypes. Littermate status is now highlighted by colors in the data tables of Figure 5 source data. See Author response table 1 for KBW ratio.

In Figure 7 (VPA), all mice are in C57BL/6J background. Because of the limited number of animals per litter and the need to subject each genotype to VPA and vehicle treatment, nonlittermates had to be included in some cases. Littermate status is now labeled by highlight colors in the data tables of Figure 7 source data. The table for KBW ratio is pasted in Author response table 2.

We added more detailed description of genetic background in the Materials and methods section. Littermate status is now also indicated in figure legends.

3) Please acknowledge that the mechanism by which VPA is exerting its effects is not certain in this study as VPA has been reported to have numerous pharmacologic effects and targets and your study provides no direct evidence in support of any one specifically.

We agree that this is an important point to clarify. We added the following text to the Discussion section:

"It is important to note that VPA could affect targets other than HDACs and testing newly approved HDACIs will provide useful insight."

4) Please explain how your double knock-out studies "support a significant role of cilia in Nphp2 function in vivo" and can exclude that ciliary activity is operating in an Nphp2-independent, parallel fashion that modulates some common downstream pathways.

Our results and model do not exclude the possibility that INVS and ciliary activity feed into a common downstream pathway, i.e., a cilia-dependent cyst-activating pathway could operate outside of cilia. We added "Although cilia-dependent, the downstream pathway could potentially operate outside of cilia and receive parallel signals from both ciliary activity and Invs." to Discussion to clarify and reflect the results and model more accurately.

5) While most of the issues can either be addressed by more explanation or editing, the reviewers have asked for more data to support several of your conclusions. This could be in the form of additional data from timepoints later than 28 days to support your conclusions that Nphp2 loss in stromal cells is not important to cystic kidney disease or kidney fibrosis, since late cyst formation has been reported in mice with stromal inactivation of Pkd1. Alternatively, you can compare the effects of global activation of Nphp2 to epithelial cell-specific inactivation of Nphp2 to support your conclusion that stromal loss of Nphp2 has no effect on early, severe cystic disease.

Regarding comparing epithelial-specific and global knockout models, for an interpretable comparison, it is essential that the stage and knockout efficiency in epithelial cells are matched between the two models. However, Cdh16-Cre is expressed in the distal nephron specifically, sparing epithelial cells in other segments, while epithelial cells in all segments would be affected by Cagg-Cre. In addition, global knockout of Invs leads to peri-natal lethality. Inducible Cagg-Cre could potentially be used to bypass earlier functional requirements. But matching stage and knockout efficiency in renal epithelial cells between Cdh16-Cre and inducible CaggCre mediated knockout remains challenging. These factors make a direct comparison problematic. Finally, our results revealed the role of defective epithelial cells in triggering the phenotypes but did not rule out a role of interstitial cells once abnormal signaling is initiated in epithelial cells. To clarify this point, we added "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." to the Discussion section.

In addition, we followed the first suggested approach and expanded our analysis to 8-week-old mice. We now show that Invs^flox/flox;Foxd1-Cre mice displayed normal kidney weight, KBW ratio, kidney function and histology at P56. By comparison, Invs^flox/flox;Cdh16-Cre kidneys were already cystic at P14. The new P56 results are presented in Figure 4—figure supplement 1 and described in the Results section. We also added "up to the young adult stage" to make our conclusion regarding the lack of renal phenotypes in interstitial KO of Invs more precise.

6) Although this is not required, it would strengthen the paper to include more timepoints and data supporting their conclusion that fibrosis precedes cyst development in the Nphp2 epithelial cell ko model.

Results at P7, 9, 14 and later timepoints were presented in Figure 2. At P14, Invs^flox/flox;Cdh16Cre kidneys were already cystic. At P7, the increase of SMA and vimentin was modest. We therefore analyzed kidneys at P9, a timepoint between P7 and P14, in more detail. Results showed normal KBW ratio and normal kidney morphology at P9. However, vimentin signal was increased, in addition to the increase of SMA signal detected previously. A caveat of this result is that mild tubule dilation could be detected at P9. We added the new results in Figure 2 and the Results section.

Reviewer #2 (Recommendations for the authors):

I have a few comments that might be addressed or discussed further to improve the interpretations of underlying mechanisms of Nphp2 function in renal homeostasis.

First, as the authors allude to the nature of the NPHP-regulated cilia-dependent pathway is related or distinct from the polycystin-mediated CDCA pathway is currently unresolved. The lack of PC2 trafficking in Nphp2 mutants suggests that the CDCA pathway might not at all be activated in the Nphp2 mutants. Rather NPHP2 could be functioning independently in regulating cystogenesis from PKD-regulated CDCA. Testing double mutants of Nphp2 with Pkd1/2 might resolve this issue.

We agree with the reviewer that testing double mutants of Invs and Pkd1/2 will provide important insight. Although this is a research direction we are interested in pursuing, we feel it ultimately is outside the scope of this manuscript.

Second, an independent interpretation of Nphp2 lack-induced cystogenesis being reduced partially from ciliary disruption is that the lack of Nphp2 or cilia regulates independent pathways with relation to fibrosis and epithelial regulation. Late assessment of phenotypes such as epithelial and stromal proliferation by the authors shows partial restorative effects. In fact, the authors show partial inhibition of stromal proliferation in Nphp2 loss from epithelial lack of cilia which is very interesting. As an early indicator of Nphp2 inactivation is stromal fibrosis and proliferation, does parallel cilia disruption prevent such early phenotypes? Does cilia disruption itself cause fibrosis at earlier stages? Such early assessment of phenotypes might be more insightful in the cilia single mutants and double mutants.

We agree with the reviewer that analyzing double mutants at earlier timepoints potentially could lead to interesting findings. Technically, cilia abrogation due to the loss of IFT genes is a slow process (PMID: 23892607). Our current system, in which knockout of Invs and Ift88 was driven by the same Cdh16-Cre, was not particularly suitable for this analysis as cilia would persist at earlier timepoints. In the revision, we analyzed markers of myofibroblast activation and fibrosis via immunostaining and Western Blot at P21 and showed a reduction of these markers in double mutants. In addition, our results showed that at P21, the increase of SMA, vimentin and collagen I was more moderate in Ift88 mutants than in Invs mutants. The new results were added to Figure 5 and described in Results.

Third, as NPHP2 interacts with multiple infantile NPHP proteins, including NPHP3, ANKS6, and NEK8 in the inversin compartment, one way of ruling out if Nphp2 function is in cilia, would be to compare/discuss phenotypes with other mutants that disrupt these interacting genes. As NPHP2 is localized to a novel fibril-like structure in the inversin compartment in cilia, the elephant in the room is the lack of our current understanding of the inversin compartment's role in the ciliary composition. Such understanding could provide important mechanistic clues and might also provide insights into the role of inversin compartment in regulating L-R asymmetry.

Following reviewer suggestion, we added "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 developed kidney cysts, while mutants of several other NPHP genes show no obvious kidney phenotypes" to Discussion.

Reviewer #3 (Recommendations for the authors):

Using a newly established floxed model of Nphp2, the authors show convincing data that Nphp2 loss in the kidney epithelium is critical to kidney cyst formation and fibrosis. In addition, the authors show compelling data that cilia loss in the setting of epithelial Nphp2 loss reduces kidney cyst severity. This cilia-dependent idea of PKD has been previously published in the setting of ADPKD, but remained unexplored for NPHP. No additional, or novel mechanistic insight is shown in the manuscript. The authors also suggest that Nphp2 loss in stromal cells is not important to cystic kidney disease or kidney fibrosis, but that data is underdeveloped. Lastly, the authors perform a preclinical trial using a broad-spectrum HDAC inhibitor, in their model, which shows efficacy in slowing disease. This experiment seems disconnected from the rest of the paper.

Ksp-cre model characterization: (1) It would be important to compare the Ksp-cre results to a global knockout of Nphp2 (CAGGc-cre) in terms of PKD severity. The authors stress the possible importance of epithelial-stromal communication. To better define this a comparison of epithelial versus global knock-out is needed.

As discussed above, it is essential for an interpretable comparison but technically challenging to match knockout timing and efficiency in epithelial cells between a global knockout model and a Cdh16-Cre model.

(2) It would be helpful to have a Kaplan-Meyer curve of the model.

We agree that a Kaplan-Meyer curve could provide useful information. However, our current animal protocol does not allow a death as endpoint experiment.

(3) Could the authors comment on what percentage of DBA or THP-positive tubules are cystic in the model?

We counted close to 200 tubules for each marker on P21 kidney sections. 91.9% DBA positive tubules were cystic and 39.3% THP positive tubules were cystic.

(4) The data supporting that fibrosis precedes cysts is underdeveloped. The SMA staining at P7 seems more intense than at P14 which is surprising (P14 and later is not quantified – cystic versus non-cystic regions), and vimentin is not analyzed at the later stages.

The increase of SMA staining is more modest at P7 than at P14 (Figure 2, please compare B and F). We have now performed immunostaining for vimentin at more timepoints (P7, 9, 14 and 21) and results were added to Figure 2.

(5) The data that epithelial Nphp2 loss results in higher proliferation in the interstitium are interesting; however, the authors provide no experiments/speculations on why that may be. Further, these analyses are done only at P7, and it is unclear how it pertains to stages when cysts develop/are present. And, it is unclear why for all other figures examining proliferation, the authors performed PCNA staining and western blotting of essential proteins, but not for this one.

P7 is a precystic stage and we used it to detect defects that are not secondary to cyst formation. Increased proliferation at an earlier stage could potentially contribute to overt cyst at a later stage and increased proliferation was detected at P21, a cystic stage (Figure 6A, B). Regarding Western Blotting, we developed these assays for the analysis of double mutants and VPA treatment, after the completion of the characterization of P7 single mutants. Moreover, we feel the difference shown by immunostaining and quantification of PCNA and phosphohistone H3 positive cells was already convincing (Figure 3).

Foxd1-cre model characterization: (1) The characterization of this model is underdeveloped. Animals need to be aged much longer to assure no phenotype is observed.

We now show that Invs^flox/flox;Foxd1-Cre mice show normal kidney weight, kidney/body weight ratio, kidney function and histology at P56. These results were presented in Figure 4—figure supplement 1 and described in the Results section.

(2) The experiments performed in Figure 4I are not intuitive. Based on the mouse model shown in Figure 1A it is unclear where eGFP expression should come from.

We apologize for not making this clear. Cre is tagged with eGFP in Foxd1-Cre. We changed eGFP to "eGFP-Cre" in the label of Figure 4I.

(3) While this is intrinsically an issue with the Foxd1-cre model, the observation that was made that only 60% of all stromal cells lost Nphp2 (Figure 4J) is concerning to make such a strong conclusion that NPHP2 in the stromal compartment is not important to the NPHP phenotype (fibrosis or cysts).

We have expanded the analysis to P56 and the results are presented in Figure 4—figure supplement 1 and described in the Results section. In brief, Invs^flox/flox;Foxd1-Cre mice displayed normal kidney weight, KBW ratio, kidney function and histology at P56. By comparison, Invs^flox/flox;Cdh16-Cre kidneys were already cystic at P14. We also added "up to the young adult stage" to make our conclusion regarding the lack of renal phenotypes in interstitial KO of Invs more precise.

NPHP2 and Cilia: (1) While the finding that the CDCA may be critical to NPHP, as well as ADPKD, is novel, the provided data is only descriptive. It would improve the paper drastically if the authors could provide data suggesting whether NPHP2 functions in the same pathway as PC1 or PC2 or a different one.

We agree with the reviewer whether INVS functions in the same pathway as polycystins is an interestingly question. However, we feel it is out of the scope of this manuscript and would pursue this research direction in our future studies.

(2) For their cell analyses, the authors should show successful loss of NPHP2 in their CRISPR clones as well as also analyze PC1.

We added Figure 6—figure supplement 1 to provide more detailed characterization of the CRISPR clones.

We did not analyze PC1 for the lack of a reliable antibody. We also feel it outside of the scope of this manuscript.

(3) Fibrosis was not analyzed.

We have now analyzed markers of myofibroblast activation and fibrosis via immuno-staining and Western Blot at P21 and detected a reduction of these markers in double mutants. The new results were added to Figure 5 and described in Results.

HDAC preclinical trial: This data is not connected to anything else within the manuscript. It utilized a different cre than for all the other experiments and is not mechanistically linked to the story of whether NPHP2 function in epithelial or stromal cells is important to pathogenesis. Unless the authors can link the stories better or highlight why the HDAC function is important to the observations, this reviewer would suggest removing Figure 7.

Given the current lack of treatment for NPHP, we feel it important to communicate the results to the research community even though the molecular mechanism remains to be defined.

Author details

1. Yuanyuan Li

   Department of Genetics, Yale University School of Medicine, New Haven, United States

   Contribution

   Data curation, Formal analysis, Investigation, Methodology, Writing - original draft, Project administration, Writing - review and editing

   Contributed equally with

   Wenyan Xu

   Competing interests

   No competing interests declared

2. Wenyan Xu

   Department of Genetics, Yale University School of Medicine, New Haven, United States

   Contribution

   Data curation, Formal analysis, Investigation

   Contributed equally with

   Yuanyuan Li

   Competing interests

   No competing interests declared

3. Svetlana Makova

   Department of Pediatrics, Yale University School of Medicine, New Haven, United States

   Contribution

   Data curation, Formal analysis, Investigation

   Competing interests

   No competing interests declared

4. Martina Brueckner

   Department of Pediatrics, Yale University School of Medicine, New Haven, United States

   Contribution

   Conceptualization, Supervision, Funding acquisition, Project administration

   Competing interests

   No competing interests declared

5. Zhaoxia Sun

   Department of Genetics, Yale University School of Medicine, New Haven, United States

   Contribution

   Conceptualization, Formal analysis, Supervision, Funding acquisition, Writing - original draft, Project administration

   For correspondence

   zhaoxia.sun@yale.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2307-7719

• 494

  Page views

• 104

  Downloads

• 0

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.