All human studies were approved by the Institutional Review Board at the Children’s National Hospital or the University Hospital of Cologne. Human kidney samples were obtained from the NIDDK-funded UAB Childhood Cystic Kidney Disease Center Translational Resource at the University of Alabama at Birmingham and the University Hospital of Cologne. Kidney samples were obtained from patients ranging from 26 weeks of gestation age to 3 years of age (Supplementary Table S1).

All mouse experiments were approved by the Institutional Animal Care and Use Committees at Children’s National Research Institute, and experiments were carried out in accordance with relevant guidelines and regulations. Mouse colonies were maintained in the animal facility at Children’s National Research Institute. All mouse kidneys were harvested from 14-day-old, 10- and 12-month-old mutants and age-matched wild-type (WT) littermates. Genetic background information for all mouse lines used in this study is shown in Supplementary Table S2.

All antibodies used for this study are listed in the Supplementary Table S3, unless specified in the text.

Immunohistochemical staining for MYC was performed on formalin-fixed, paraffin embedded tissues using heat induced epitope retrieval solution (BOND Epitope Retrieval Solution 2, Leica Biosystems, Cat. No. AR9640) and an automated stainer (Bond-Max, Leica Biosystems). Tissues were incubated with anti-MYC antibody (Recombinant Anti-c-Myc antibody [Y69] - ChIP Grade, Abcam) at 1:25 dilution for 120 min.

Cultured cells and kidney tissues were collected, homogenized, and processed for immunoblotting as previously described (Wu et al., 2013; Dafinger et al., 2020). Immuno-reactive protein bands were visualized using SuperSignal West Femto Maximum Sensitivity Substrate (Thermo Fisher Scientific, Cat. No. 34095) and images were obtained with ChemiDoc Imaging System (Bio-Rad laboratory). Densitometry was performed using Image Lab software (Bio-Rad laboratory, Version 6.0).

Kidney tissue samples from 14-day-old male mice were snap frozen, transferred to a gentleMACS M tubes (Miltenyi Biotec, Cat. No. 130-093-236, RRID:SCR_020269) in Buffer RLT plus 2-Mercaptoethanol (as per RNeasy Mini Kit instructions, Qiagen, Cat. No. 74104), and homogenized using a gentleMACS Dissociator (Miltenyi Biotec) per the manufacturer’s program RNA-02. Homogenized samples were transferred to microcentrifuge tubes for total RNA extraction using the RNeasy Mini Kit according to the manufacturer’s instructions. Total RNA from both 5-day-postconfluent mIMCD-3 cells stably expressing FPC-CTD and hTERT-immortalized human renal epithelial cells (hTERT-HRE) transiently transfected with FPC-CTD were isolated using the RNeasy Mini Kit (Qiagen) according to the manufacturer’s instructions. Isolated total RNA was treated with RQ1 RNase-Free DNase (Promega, Cat. No. M6101), and then repurified using the RNeasy Mini kit. RNA samples were reverse-transcribed using SuperScript III First-Strand Synthesis SuperMix (Thermo Fisher Scientific, Cat. No. 18080400) and oligo dT primers as described in the manufacturer’s instructions.

Quantitative RT-PCR was performed on a QuantStudio 7 Flex Real-Time PCR System (Thermo Fisher Scientific) using the default program. The PCR was performed with cDNA templates using Power SYBR Green PCR Master Mix (Thermo Fisher Scientific, Cat. No. 4368706) and mouse primers specific for sequences of Myc (forward: 5′- GCC CCC AAG GTA GTG ATC CT -3’; reverse: 5′- GTG CTC GTC TGC TTG AAT GG -3′). Peptidylprolyl isomerase A (PPIA) was used for normalization (forward: 5′- AGC ACT GGA GAG AAA GGA TT -3’; reverse: 5′- ATT ATG GCG TGT AAA GTC ACC A-3′) (Arensdorf and Rutkowski, 2013). Overexpression of FPC-CTD in cell lines was confirmed with Pkhd1 exon 66–67 specific primers (forward: 5′-CCA GAA GAC ATA TCT GAA TCC CAG GC-3’; reverse: 5′-AGC AAG AGA TCC TGG AAC ACA GGT-3′). Results were analyzed using QuantStudio Real-Time PCR Software and the ΔΔCt method (Livak and Schmittgen, 2001).

Protein sequences of FPC were collected from NCBI protein database (https://www.ncbi.nlm.nih.gov/protein/) using an advanced search with gene name, Pkhd1 and taxonomic groups Mammalia, Aves, Reptilia, Amphibia, Caecilians, and Fish. This resulted in 102 FPC sequences from mammals, birds, reptiles, amphibians, and fish (Supplementary Table S4). FPC sequence from each species was verified by protein alignment with human FPC; the protein sequences that were significantly shorter than human FPC sequence were removed. WebLogo 3 (Crooks et al., 2004) was used to visualize FPC sequence alignment (Supplementary Figure S1) that was generated with Clustal Omega (Sievers et al., 2011). The conservation scores of FPC amino acids were extracted from the WebLogo 3 raw data (Supplementary Table S5). Conserved domains in FPC were mapped by Conserved Domain Database (CDD) (Lu et al., 2020).

Prediction of nuclear localization signals (NLSs) were performed with the human and mouse FPC-CTD construct sequences using SeqNLS (Lin et al., 2012) with 0.86 as the cut-off score.

pcDNA5/FRT/TO (pcDNA5) was obtained from Thermo Fisher Scientific (Cat. No. V652020).

pcDNA5/FRT/TO-mPkhd1-CTD-V5 (pcDNA5-mFPC-CTD): the cytoplasmic tail of mouse FPC (Follit et al., 2010) expression construct was generated from pcDNA5/FRT/TO-mPkhd1 (full length) -V5 (gift from Dr. Feng Qian) with site-directed mutagenesis (SDM) using 5′- GCT AAC TGG ACA TGA TGC TTT GCT GCT GGT TTA AGA AAA GC -3′ and 5′- GCT TTT CTT AAA CCA GCA GCA AAG CAT CAT GTC CAG TTA GC -3′ primer set.

pcDNA5/FRT/TO-mPkhd1-CTD ^delCTS -V5 (pcDNA5-mFPC-CTD ^delCTS ): the ciliary targeting sequence (Follit et al., 2010) deleted mFPC-CTD expression construct was made from mFPC-CTD by SDM with 5′- GCT AAC TGG ACA TGA TGC TTG ACA TAT CTG AAT CCC AGG CT -3′ and 5′- AGC CTG GGA TTC AGA TAT GTC AAG CAT CAT GTC CAG TTA GC -3′ primer set.

pcDNA5/FRT/TO-hPKHD1-CTD-V5 (pcDNA-hFPC-CTD): the expression construct containing the hFPC-CTD, comparable to mFPC-CTD (Follit et al., 2010), and fused to V5-tag was made by LifeSct LLC. The hFPC-CTD coding sequence was cloned between KpnI and NotI sites in pcDNA5/FRT/TO.

pcDNA5/FRT/TO-hPKHD1-CTD ^delCTS -V5 (pcDNA-hFPC-CTD ^delCTS ): the ciliary targeting sequence deleted hFPC-CTD expression construct was made from hFPC-CTD by SDM using 5′- CCG TGG ACA GAA TGA CTG CCG AGA TTC CTG AAT CCC AGA C -3′ and 5′- GTC TGG GAT TCA GGA ATC TCG GCA GTC ATT CTG TCC ACG G -3′ primer set.

pGL4.22 [luc2CP/Puro] vector was purchased from Promega (Cat. No. E6771).

pRL-TK vector was purchased from promega (Cat. No. E2241).

pGL4.22-mouse Myc P1 (pGL4.22-mMyc P1): the mouse Myc P1 promoter (chr8:127735983-127736125, GRCm38/mm10 mouse genome assembly) construct was described previously (Wu et al., 2013).

pGL4.22-human MYC P1 (pGL4.22-hMYC P1): the human MYC P1 promoter (chr15:61985298-61985433, GRCh38/hg38 human genome assembly), which is comparable to the mouse Myc P1 promoter, was amplified by PCR from HEK293 genomic DNA and cloned into pGL4.22 vector at XhoI and HindIII sites using 5′- CCG CTC GAG GAG GGC GTG GGG GAA AAG A-3′ and 5′- CCC AAG CTT AGC CAG GGA CGG CCG G -3′ primer set. Sequence alignment of human and mouse MYC/Myc P1 promoter shown in Supplementary Figure S2 was created by Clustal Omega (Sievers et al., 2011).

Mouse TERT immortalized cortical collecting duct (mTERT-CCD) cells (Steele et al., 2010) were cultured in TERT culture medium [DMEM/F-12 medium (Thermo Fisher Scientific, Cat. No. 11330057) containing 5% heat-inactivated fetal bovine serum (FBS) (Atlanta Biologicals, Cat. No. S11050H), 1% penicillin/streptomycin (Thermo Fisher Scientific, Cat. No. 15140163), 1x Insulin-Transferrin-Selenium solution (Thermo Fisher scientific, Cat. No. 41400045), 0.2 μg/mL dexamethasone (Sigma-Aldrich, Cat. No. D8893) and 10 nM 3,3′,5-Triiodo-L-thyronine sodium salt (Sigma-Aldrich, Cat. No. T6397)] at 37°C in 5% CO₂.

Mouse inner medullary collecting duct (mIMCD)-3 cells were purchased from American Type Culture Collection (ATCC, Cat. No. CRL-2123) and cultured in complete growth medium (CGM) [DMEM/F-12 medium (Thermo Fisher Scientific, Cat. No. 11330057) containing 10% heat-inactivated fetal bovine serum (Atlanta Biologicals, Cat. No. S11050H) and 1% penicillin/streptomycin (Thermo Fisher Scientific, Cat. No. 15140163)] at 37°C in 5% CO₂.

mIMCD-3 mFPC-CTD stable cell lines were generated by transfection of either pcDNA5-mFPC-CTD or pcDNA5 into mIMCD-3 cells with Lipofectamine2000 transfection reagent (Thermo Fisher Scientific, Cat. No. 11668019). At 48 hrs post-transfection, cells with spontaneously integrated plasmids were selected with hygromycin (1 mg/mL) (Thermo Fisher Scientific, Cat. No. 10687010) for 1 week. The mFPC-CTD stably overexpressing cells and the control empty vector cells were then maintained in CGM with hygromycin (200 μg/mL).

Human kidney sections were minced and immediately placed in 1% collagenase type I (Sigma, Cat. No. C0130-1G) in DMEM/F12 (Thermo Fisher, Cat No. 11330032) and incubated on a rotator for 30 min at room temperature (RT). Renal epithelial cells and tubule fragments were then transferred to a conical tube containing DMEM/F12 and centrifuged at 750 g for 10 min. The supernatant was removed, and the tissue was resuspended in a complete medium, containing 0.2 mg/mL dexamethasone (Sigma, Cat. No. D8893-1 MG), 5% heat inactivated FBS (Hyclone, Cat. No. SH30396-03), 2 mM glutamate (Thermo Fisher, Cat. No. 25030081), 1x insulin-transferrin-sodium selenite (ITS) (Thermo Fisher, Cat. No. 1400045), 100 U/mL penicillin/streptomycin (Thermo Fisher, Cat. No. 15140122), and 10 nM triiodothyronine (Sigma, Cat. no. T6397-100 MG) in DMEM/F12. No antibiotics were added to the complete medium in preparation for the transduction with the hTERT lentiviral expression construct. Cells were maintained in a 37°C humidified incubator with 5% CO₂.

Lenti-hTERT-Neo Virus (Cat. No. LV622), Lenti-p53 siRNA Virus (Cat. No. G219), and Polybrene (Cat. No. G062) were purchased from Applied Biological Materials (Richmond, BC, Canada). One day prior to transfection, primary human renal epithelial cells were plated at 20%–30% confluency in a 6 well plate. The next day, the complete medium was replaced with 1 mL of transfection medium, which contained 6 μg/mL of polybrene in the complete medium. Then, 1.54 × 10⁸ transducing units (TU)/ml of Lenti-hTERT-Neo Virus and 1 × 10⁶ TU/mL of Lenti-p53 siRNA Virus were added to the transfection medium at a multiplicity of infection (MOI) of 7. The plate was centrifuged at 200 g for 30 min and then placed back in the incubator at 37°C. After 24 h culture at 37°C, 1 mL of the complete medium was added to each 6 well and cultured for an additional 24 h. Cells immortalized with the hTERT gene then were selected at 48 h post transduction, using 800 μg/mL G418 in the complete medium.

Characterization of the hTERT-HRE cells was performed using immunoblotting of E-cadherin as an epithelial marker, ZO-1 as a tight junction marker, α-ENaC as a renal epithelial cell marker, AQP1 as a renal tubule cell marker, and Keratin 17/19 as an epithelial cell marker (Supplementary Figure S3A). In addition, RT-PCR was performed for PKHD1 as a renal epithelial marker (Supplementary Figure S3B).

For MYC immunofluorescence staining, cells stably overexpressing mFPC-CTD were seeded onto coverslips in 6-well plates and cultured until confluent. For FPC-CTD localization assay, mTERT-CCD or hTERT-HRE cells were transiently transfected with pcDNA-mFPC-CTD and pcDNA-mFPC-CTDd^elCTS, or pcDNA5-hFPC-CTD and pcDNA5-hFPC-CTD^delCTS, respectively. Forty-8 hrs. after transfection, cells were washed with PBS and fixed with 4% paraformaldehyde (PFA) for 10 min at RT and then permeabilized with 0.5% Triton X-100 in PBS for 5 min, followed by three washes with PBS before blocking with 1% BSA for 30 min. The cells were incubated with primary antibodies (anti-V5 or anti-Myc) overnight at 4°C followed by incubation with secondary antibody (Alexa Fluor 488 at 1:400 dilution) for 1 h at RT. The cells were then washed three times with PBS and mounted with ProLong Gold + DAPI (Life Technologies, Cat. No. P36935). Fluorescently labeled cells were analyzed on an Olympus FV1000 scanning laser confocal microscope configured with both an Argon Laser (5 mW, 488 nm), and a Green HeNe (10 mW, 543 nm) laser. Images were analyzed using Olympus FV10-ASW 3.0 Viewer software.

Cells were seeded in 24-well plate, grown to ∼90% confluence, and then transfected with pGL4.22-mMyc P1 or pGL4.22-hMYC P1): (0.3 µg/well) and pcDNA5-mFPC-CTD (0.6 μg or 1.2 μg/well) or pcDNA5-hFPC-CTD (0.6 µg/well) using Lipofectamine 2000. The differences in transfection efficiency were normalized by co-transfecting with 15 ng/well pRL plasmid that expressed Renilla luciferase (Promega) and adjusting the total amount of plasmid DNA to 1.5 ug/well by adding pcDNA5. The transfected cells were incubated for 48 h, lysed in 100 µL/well passive lysis buffer (Promega) and shaken for 20 min at RT. Firefly and Renilla luciferase activities were measured with Dual-Luciferase Reporter Assay System (Promega, Cat. No. E1910). The luminometer (FLUOstar OPTIMA, BMG LABTECH) was programmed using OPTIMA software to perform a 0 s delay, followed by a 5-s measurement period for each reporter assay. The 20 μL cell lysate was transferred into 96-well plate (Costar, Cat. No. 3912; white flat bottom), followed by the addition of 100 µL Luciferase Assay Reagent II and luminescence reading. After measurement of firefly luciferase activity, 100 µL of Stop&Glo reagent was added and quickly put back for reading of the Renilla luciferase activity. Data were collected from three independent transfections and processed using GraphPad Prism version 9.1.2 for Windows, GraphPad Software, San Diego, California United States of America, www.graphpad.com.

To determine the binding of FPC-CTD to the Myc P1 promoter, experiments were performed using Magna ChIP A/G Chromatin Immunoprecipitation Kit (MilliporeSigma, Cat. No. 17-10085), according to the manufacturer instructions and our previously published protocol (Wu et al., 2013). Because ChIP-grade anti-FPC-CTD antibodies were not available, we generated mIMCD-3 cells stably overexpressing mFPC-CTD-V5 and control cells stably transfected with empty vector and used ChIP-grade anti-V5 antibody (Abcam, Cat. No. ab15828) for immunoprecipitation. Cells were grown to 80%–90% confluence prior to experiments and processed according to the Magna ChIP A/G protocol. Following immunoprecipitation with anti-V5 antibody that recognized mFPC-CTD-V5 bound to the chromatin and subsequent protease digestion, we amplified Myc P1 using PCR primers specific to the full-length Myc P1 (forward primer: 5′- CGC TCG AGG AGA GAG GTG GGG AAG GGA GAA AG -3’; reverse primer: 5′- CCC AAG CTT AGT GAG GCG AGT CGG ACC CGG CA -3′) using the following PCR program: 94°C 3 min; 94°C 20 s, 62°C 20 s, 72°C 15 s, repeat for 40 cycles, 72°C 5 min, 10°C holding.

Gene expression profiles of PKHD1/Pkhd1, CYS1/Cys1, NDN/Ndn, and MYC/Myc across kidney developmental stages in humans and mice, were downloaded from Evo-devo mammalian organs portal (https://apps.kaessmannlab.org/evodevoapp/) (Cardoso-Moreira et al., 2019). RPKM (reads per kilo base of transcript per million mapped reads) values were normalized using the highest value as 1 and the lowest value as 0 for each gene.

Non-parametric Wilcoxon sign rank test was used for analysis of qRT-PCR data normalized to control samples (Figure 1C, and 2D). All other data were analyzed using either two-way Student’s t-test or nonparametric test with GraphPad Prism version 9.1.2 for Windows, GraphPad Software, San Diego, California United States, www.graphpad.com.

MYC overexpression is a signature feature of cystic renal epithelia in human ADPKD and various mouse PKD models (Trudel, 2015). However, MYC expression in ARPKD has not been reported. In the current study, we analyzed MYC expression in ARPKD kidneys by immunoblotting. While MYC expression was marginally detectable in adult kidneys and a kidney from an infant without kidney disease (Supplementary Figure S4), we observed higher MYC abundance in all kidneys from patients with defined pathogenic variants in PKHD1 (Figure 1A), with the highest MYC levels detected in kidneys from patients with PKHD1 truncating pathogenic variants (Supplementary Table S1), resulting in the loss of FPC-CTD (AR1, AR2, AR3, and AR8) (Figure 1A and Supplementary Figure S4).

We then evaluated MYC expression in mouse models of ARPKD. Using IHC, we confirmed increased nuclear expression of mouse MYC protein in dilated collecting ducts from cpk kidneys (Figure 1B). Quantitative analysis confirmed 11-fold higher levels of MYC protein and 1.8-fold higher levels of Myc mRNA in the kidneys from cpk mice compared to WT mice (Figure 1C). In contrast, MYC protein levels were not elevated in kidneys from four different Pkhd1 mutant mouse model lines (Pkhd1 ^cyli , Pkhd1 ^del3-4 , Pkhd1 ^del3-67 , or Pkhd1 ^del67 ) that did not exhibit a cystic kidney phenotype (Figure 1D). These data indicate an association between high MYC expression and the renal cystic phenotype in both human ARPKD and mice with an ARPKD-like kidney phenotype. Furthermore, the lack of enhanced MYC expression in Pkhd1 mutant mice without cystic kidney phenotype suggests differences in the function of mouse and human FPC-CTDs.

We note that in each of our mouse mutant lines, the predicted Pkhd1 translated products would be missing the FPC-CTD. Furthermore, while the mouse and human FPC sequences are 73% identical overall, the CTDs share only 55% identity (Nagasawa et al., 2002). Therefore, we analyzed phylogenetic conservation of FPC in vertebrates to better understand potential differences in the regulation of MYC expression in renal epithelia derived from patients with ARPKD and cpk and Pkhd1 mutant mice. We hypothesized that the low sequence conservation of the FPC-CTDs may contribute to functional differences among FPC orthologs; an important consideration given that mouse FPC undergoes Notch-like processing that releases the FPC-CTD, which can translocate to the nucleus.

To understand phylogenetic changes in Pkhd1, we queried the NCBI protein database for PKHD1 orthologs and collected FPC protein sequences of 66 mammalian, 27 bird, 5 reptile, 3 amphibian, and 1 fish (Supplementary Table S4). The large variance in the number of FPC proteins in each class may in part reflect the number of sequenced vertebrate genomes in the NCBI data base. However, while genomic data are available for multiple fish, a PKHD1 ortholog was identified only in the genome of Latimeria chalumnae (a coelacanth).

After alignment of 102 vertebrate FPC protein sequences, we used the WebLogo 3 entropy scores to evaluate FPC conservation across species (Supplementary Figure S1 and Supplementary Table S5 ). First, we compared the WebLogo 3 entropy scores for each of the conserved domains in the extracellular portion of FPC: five IPT domains, three TIG domains, two G8 domains, two PbH1 domains, and the TM domain with entropy scores from three regions of FPC that do not correspond to any conserved domains (Supplementary Figure S5B). Since the average WebLogo 3 entropy scores were similar (Supplementary Figure S5B), we then compared the WebLogo 3 entropy scores of FPC extracellular and cytoplasmic domains. Consistent with different protein sequence identities between human and mouse extracellular and cytoplasmic portions of FPC (Nagasawa et al., 2002) the WebLogo 3 entropy scores of vertebrate FPC extracellular domain were higher than entropy scores of FPC-cytoplasmic domains (Supplementary Figure S5C). This analysis supports the hypothesis that lower sequence conservation of mouse and human FPC cytoplasmic domain compared to extracellular domain may be functionally significant and potentially contribute to the phenotypic variability observed between PKHD1 vs Pkhd1 mutants.

Both human and mouse, FPC-CTD are encoded by PKHD1/Pkhd1 exons 65, 66 and 67 (Supplementary Figure S5D). Overall, human and mouse FPC-CTD share 55% sequence identity, (Nagasawa et al., 2002). However, the CTS, localized in the human and mouse FPC-CTDs, are highly conserved (Figure 2A, red and blue highlights). Prior studies have experimentally validated one NLS in the mouse FPC-CTD (Figure 2A, purple highlight) (Hiesberger et al., 2006). However, using the web-based NLS prediction tool, SeqNLS (Lin and Hu, 2013), we identified two NLSs in mouse FPC-CTD (score >0.86), one of which overlapped with the experimentally identified NLS (Figure 2A, red bold text for predicted and purple highlight for experimentally validated NLS) (Hiesberger et al., 2006). On the other hand, human FPC-CTD had only one predicted NLS (score >0.86) (Figure 2A, red bold text). Interestingly, the two mouse, and the one predicted human NLS showed low sequence similarity.

While it has been determined that the mouse FPC-CTD translocates to the nucleus (Hiesberger et al., 2006), the intracellular trafficking of human FPC-CTD has not been determined and the nuclear function of FPC-CTD is not fully understood. Therefore, we compared intracellular localization of human and mouse FPC-CTD and tested their functions in the nucleus, concentrating on MYC/Myc regulation. Sequences the human and mouse FPC-CTD included the V5 tag and both the mouse and human FPC-CTDs contained the intracellular portion of the CTS (Figure 2A, blue highlight). The NLS sequences are shown in Figure 2A (red bold text).

By immunofluorescence, the mFPC-CTD localized primarily to the nucleus and was essentially absent from the cytoplasm (Figure 2B, left column). In contrast, the hFPC-CTD was largely excluded from the nucleus, localized to the cytoplasm and decorated the cell membrane (Figure 2B, second column from left). We suspected that nuclear trafficking of the hFPC-CTD construct with only one NLS was confounded by the CTS. Therefore, we deleted the CTS from both human and mouse constructs to generate plasmids expressing hFPC-CTD^delCTS and mFPC-CTD^delCTS respectively. Overexpression of these proteins showed strong nuclear localization for both human and mouse FPC-CTDs, although a fraction of hFPC-CTD^delCTS was retained in the cytoplasm (Figure 2B, right two columns).

We then investigated whether mFPC-CTD can regulate Myc expression in mIMCD-3 cells stably expressing the intact mFPC-CTD mouse construct. mIMCD-3 cells stably expressing the empty vector (pcDNA5) served as a control. This experimental approach allowed evaluation of the effect of mFPC-CTD on Myc expression in non-proliferating cells, 5 days post-confluence. The intensity of MYC immunostaining was higher in mIMCD-3 cells expressing mFPC-CTD than in the control cells (Figure 2C). Both Myc mRNA and MYC protein levels were higher in mIMCD-3 cells stably expressing mFPC-CTD, compared to control cells (Figure 2D). Taken together, our data provide the first evidence that overexpression of mFPC-CTD enhances MYC expression in cultured renal epithelial cells.

Efforts to perform these experiments with hFPC were confounded by the inability to generate an hTERT-HRE cell line that stably overexpressed either hFPC-CTD or hFPC-CTD^delCTS. We observed that expression of hFPC-CTD was silenced after several passages of hTERT-HRE cells under selection, suggesting that stably expressed hFPC-CTD may be cytotoxic.

The mechanisms that govern Myc transcription are complex and involve multiple promoters (P0, P1, P2, and P3) and transcription start sites (Battey et al., 1983; Bentley and Groudine, 1986; Ray et al., 1987). The P1 and P2 promoters are the predominant Myc regulatory elements (Albert et al., 2001). In disease states, Myc overexpression primarily is driven from the P1 promoter (Wierstra and Alves, 2008). Therefore, we cloned the mouse Myc P1 and human MYC P1 promoter into a pGL4.22 reporter gene construct (Figure 3A). Co-transfection of mIMCD-3 cells with constructs expressing mFPC-CTD and Myc P1 promoter driven reporter gene showed dose-dependent activation of the Myc P1 promoter by mFPC-CTD (Figure 3B). Myc P1 promoter activation in mTERT-CCD cells was not affected by deletion of the CTS (Figure 3C). In comparison, co-transfection of hTERT-HRE cells with constructs expressing hFPC-CTD and MYC P1 reporter plasmid showed only minimal, though statistically significant activation of the MYC P1 promoter (Figure 3D). However, deletion of the CTS from the hFPC-CTD, which enhanced its nuclear localization, resulted in a 1.6-fold activation of the MYC P1 promoter (Figure 3D). These data demonstrate that both human and mouse FPC-CTDs can activate the Myc/MYC P1 promoter in cultured cells. Furthermore, we tested hFPC-CTD activation of MYC P1 promoter in mIMCD-3 cell line and found comparable activation in this mouse line as in the human cell line, hTERT-HRE (Supplementary Figure S6). However, we note that hFPC-CTD nuclear trafficking is regulated by the CTS.

To confirm the binding of mFPC-CTD to the Myc P1 promoter, we performed ChIP assays. We found that mFPC-CTD-V5 bound to the endogenous Myc P1 promoter in non-proliferating cells (Figure 3E, asterisk). As noted above, corresponding ChIP experiments with hFPC-CTD were confounded by our inability to generate an appropriate hTERT-HRE cell line.

Considering the observed differences in the nuclear trafficking and function of human and mouse FPC-CTDs, we sought to better understand how species-specific regulation of MYC/Myc expression may contribute to the divergent renal phenotypes in human ARPKD and the Pkhd1 mouse models. Therefore, we analyzed cystogene expression patterns in human and mouse kidneys during intrauterine and postnatal development using the Evo-devo mammalian organs portal (http://evodevoapp.kaessmannlab.org) (Cardoso-Moreira et al., 2019). We specifically focused on genes that are known to be mutated in human ARPKD (PKHD1/Pkhd1), in cpk mice (CYS1/Cys1), as well as necdin (NDN/Ndn), which we have previously shown regulates Myc expression (Wu et al., 2013). We normalized the kidney developmental stages of mouse and human during pre- and postnatal development. RPKM of PKHD1/Pkhd1, CYS1/Cys1, NDN/Ndn, and MYC/Myc genes were graphed at the corresponding developmental stages of human and mouse kidneys (Figure 4). These analyses demonstrated different timing of PKHD1/Pkhd1 and CYS1/Cys1 mRNA expression peaks during human and mouse kidney development (Figure 4, top panels). In the fetal human kidney, PKHD1 mRNA levels progressively increase and reach maximal expression prior to birth, whereas the progressive increase in CYS1 mRNA lags, reaching maximum expression in the post-natal period. Conversely, in the mouse kidney, maximum expression of Cys1 mRNA precedes the peak of Pkhd1 mRNA expression. Expression patterns of NDN/Ndn mRNA during human and mouse kidney development were similar (Figure 4, middle panels), with expression peaking in early developmental stages (6–8 weeks post conception and e12.5–14.5 in human and mouse kidney, respectively) and decreasing thereafter. Expression of MYC/Myc mRNA also peaked during early nephrogenesis and gradually decreased thereafter in both human and mouse kidneys (Figure 4, bottom panels).

With the assumption that mRNA expression serves as an appropriate proxy for protein levels, the species-specific differences in the PKHD1/Pkhd1 and CYS1/Cys1 developmental expression patterns and the observation that cystin may be protective of MYC/Myc activation suggest a mouse-specific renoprotective mechanism in mice lacking functional FPC. Therefore, we hypothesized that limiting cystin protein in kidneys from Pkhd1 mutant mice may evoke dilatation of renal tubules and/or collecting ducts. To address this hypothesis, we crossed the cpk allele into mice that are homozygous for the Pkhd1 ^cyli mutation (cyli), an indel in Pkhd1 exon 48 that causes premature termination of protein translation (Yang et al., 2023). Kidneys from 10- and 12- month-old cyli/cyli;cpk/+ mice on a mixed genetic background (D.B/11Ei; C57BL/6J, Supplementary Table S2) had mild tubular dilations that were absent in kidneys from age-matched cyli/cyli;+/+ mice (Supplementary Figure S7).