G-quadruplex stabilization provokes DNA breaks in human PKD1, revealing a second hit mechanism for ADPKD

Parsons, Agata M.; Byrne, Seth; Kooistra, Jesse; Dewey, John; Zebolsky, Aaron L.; Alvarado, Gloria; Bouma, Gerrit J.; Vanden Heuvel, Gregory B.; Larson, Erik D.

doi:10.1038/s41467-024-55684-y

Download PDF

Article
Open access
Published: 02 January 2025

G-quadruplex stabilization provokes DNA breaks in human PKD1, revealing a second hit mechanism for ADPKD

Nature Communications volume 16, Article number: 121 (2025) Cite this article

5256 Accesses
1 Citations
6 Altmetric
Metrics details

Subjects

Abstract

The “secondhit” pathway is responsible for biallelic inactivation of many tumor suppressors, where a pathogenic germline allele is joined by somatic mutation of the remaining functional allele. The mechanisms are unresolved, but the human PKD1 tumor suppressor is a good experimental model for identifying the molecular determinants. Inactivation of PKD1 results in autosomal dominant polycystic kidney disease, a very common disorder characterized by the accumulation of fluid-filled cysts and end-stage renal disease. Since human PKD1 follows second hit and mouse Pkd1 heterozygotes do not, we reasoned that there is likely a molecular difference that explains the elevated mutagenesis of the human gene. Here we demonstrate that guanine quadruplex DNA structures are abundant throughout human, but not mouse, PKD1 where they activate the DNA damage response. Our results suggest that guanine quadruplex DNAs provoke DNA breaks in PKD1, providing a potential mechanism for cystogenesis in autosomal dominant polycystic kidney disease specifically and for the inactivation of guanine quadruplex-rich tumor suppressors generally.

Renal plasticity revealed through reversal of polycystic kidney disease in mice

Article 11 October 2021

Identification of pathological transcription in autosomal dominant polycystic kidney disease epithelia

Article Open access 23 July 2021

Extracellular vesicles and exosomes generated from cystic renal epithelial cells promote cyst growth in autosomal dominant polycystic kidney disease

Article Open access 27 July 2021

Introduction

The tumor-blocking phenotypes for most tumor suppressor genes are only revealed once a germline pathogenic allele is accompanied by inactivation of the remaining functional allele in a somatic cell, a pathway known as “secondhit”¹. It is unknown why many tumor suppressors follow this pathway, or why orthologous genes in animal models are often genetically stable, but the human PKD1 (hPKD1) gene presents a unique experimental opportunity to uncover a mechanism. PKD1 encodes polycystin-1, a multi-functional transmembrane protein involved in cell proliferation, differentiation, apoptosis, cell adhesion, and fluid secretion. In ADPKD the single functional PKD1 allele undergoes somatic second hit inactivation^2,3,4,5,6,7, initiating cysts and cyst progression due to loss of polycystin-1 activity^8,9,10. Although mice have served as valuable animal models, mouse Pkd1 heterozygotes do not faithfully recapitulate human cystogenesis and second hit^2,3,4,5,6,7. This implies that an intrinsic molecular feature influences hPKD1’s risk of DNA breaks, mutagenesis, and subsequent inactivation.

Several observations suggest hPKD1 is mutation-prone compared to mouse Pkd1 (mPkd1). First, of the hundreds of renal cysts containing inactivated hPKD1 alleles in an affected kidney, each somatic inactivation is derived from an independent hPKD1 loss of heterozygosity or mutation event^3,4,6,7,11. Furthermore, hPKD1 is itself polymorphic and was partially duplicated in the human genome to produce 6 nearby pseudogenes^3,4,6. One clue to the sources of hPKD1 instability was the identification of a ∼2 kb intronic sequence repeat in intron 21 that inhibits replication^{12,13,14,15,16}, sequences that are notably absent in mPkd1¹⁷. This raises the possibility that DNA damage in hPKD1 arises because of non-canonical DNA structure formation^{12,13,14,15,16}. Consistent with that hypothesis, an 88 nucleotide repeat from intron 21 (IVS21) was reported to form guanine-quadruplex (G4) DNA^16,18, a four stranded DNA conformation (Fig. 1a). G4 DNAs are physiological structures found concentrated at regulatory domains in the genome, but if left unresolved they promote oncogenesis by inhibiting replication or repair to create DNA breaks that then lead to recombination and mutations^19,20,21.

Results

G4 sequences in PKD1 orthologs

Based on the potential for hPKD1 to adopt G4 DNA in intron 21 (IVS21)¹⁶, we quantitated G4 DNA sequences in the entire PKD1 gene using a G4-prediction program (QGRS mapper²²). Using a conservative G4 motif definition of at least three tandem guanines repeated four times (Fig. 1a), we find high G4 DNA content in hPKD1 (124) and low G4 DNA abundance in mPkd1 (13) and rat rPkd1 (10) (Fig. 1b). The predicted G4 DNA-bias in hPKD1 compared to mPkd1 (Fig. 1b) agrees with experimentally validated cellular G4 DNAs (from G4-ChIPseqs) curated by the EndoQuad database²³. Mapping of predicted G4 DNA onto hPKD1 shows a broad distribution, with some introns showing clusters of G4 DNAs, including 16 in IVS1, 38 in IVS21, 7 in IVS22, and 5 in IVS42 (Fig. 1c). This contrasts with mPkd1 and rPkd1, which have very few G4 motifs overall (Fig. 1b, c). The intronic G4 clusters found in hPKD1 are composed of poly-purine repeats. A G4 motif from the 16 tandem repeats in intron 1 formed G4 in vitro as detected by dot blot (Fig. 2a) and shows a characteristic G4 spectrum by circular dichroism spectroscopy with a peak at 264 and a dip at 240^24,25, which shifted when the guanine repeats were disrupted by substitution (Fig. 2b). CD scans of similar guanine-rich repeats derived from IVS21, 22 and 42 also show G4 spectra (Supplementary Fig. 1), as expected. We conclude that G4-forming sequences are abundant and distributed throughout hPKD1 but are comparatively rare in mPkd1 and rPkd1.

G4 DNA formation at the hPKD1 locus

If G4 DNAs are relevant to ADPKD pathophysiology we would expect to find the structures in renal cell nuclei and at the hPKD1 locus. The BG4 antibody has been well characterized for the visualization of quadruplexes in cells²⁶, and the newer SG4 camelid nanobody has low nanomolar affinity for G4s with a control version (R105A) available that is mutated for G4 DNA specificity²⁷. In our hands, all three antibodies performed as expected for G4 detection (Supplementary Fig. 2). To visualize the hPKD1 locus in cells we employed CASFISH²⁸, since the denaturing conditions for Fluorescence In Situ Hybridization would likely disrupt DNA structures. In CASFISH, a catalytically inactivated CAS9, dCAS9, is fluorescently labeled and targeted to a locus with sgRNAs. Using a labeled dCAS9 paired with sgRNAs directed to the 3’ end of hPKD1 and combined with SG4 labeling, we found overlapping foci in normal (Fig. 3A, B), and ADPKD tissue (Fig. 3C, D). Substituting SG4 for the mutant nanobody (R105A) eliminated G4 signals (Fig. 3E, F) and omitting the sgRNAs ablated PKD1 foci (Fig. 3G, H), confirming that the quadruplex signals are due to SG4 and that the PKD1 signals are dependent on the sgRNAs. Thus, the G4 sequences in the hPKD1 gene may adopt G4 structures.

**Fig. 3: G4 DNA at the human *PKD1* locus in normal and ADPKD tissue.**

We next used a more quantitative approach to compare G4 DNA formation in human and mouse Pkd1 by employing chromatin immunoprecipitations (ChIPs) with the BG4 antibody, anticipating enrichment of human but not mouse Pkd1. Fragmented crosslinked chromatin from human embryonic kidney (293T) or mouse mIMDC3 was precipitated with BG4 followed by qPCR to detect template enrichment. For HEK293T, we compared a region adjacent to IVS21 to another intron, IVS34. Exon/Intron 34 has 2 potential G4 motifs compared to the 38 in intron 21 (Fig. 1c), making this region a good negative control. For mIMCD3 chromatin, we used primers specific to nucleotide positions comparable to hPKD1, which in mPkd1 are near IVS21 and IVS37. BG4-IP enriched for the G4-dense human IVS21 locus by a factor of 1.7 compared to IVS34 (Fig. 4). Addition of the G4 DNA-stabilizing ligand Phen-DC3 to HEK293T significantly (P < 0.001) increased IVS21 enrichment compared to vehicle by 3 times (16.6/5.6), and to IVS34 by 4.4 times (16.6/3.8) (Fig. 4). Repeating the experiment with a chemically unrelated G4-ligand, CX-5461, yielded similar results (Fig. 4). None of the mPkd1 loci were significantly enriched for either ligand (P > 0.05)(Fig. 4). Since a G4-rich control locus in mIMCD3 was enriched when either ligand was present (Supplementary Fig. 3), both ligands function in mIMCD3 and BG4-IP results are consistent with the lack of G4 DNAs in mPkd1 (Fig. 1b, c). Collectively, the data suggest that G4 DNA forms within hPKD1.

Influence of G4 DNA on PKD1 expression

The abundant G4 DNAs distributed throughout hPKD1 (Fig. 1c) implies a functional role for the structures. Therefore, we asked if enforced G4 stabilization with a G4-specific ligand would alter PKD1 expression. Both HEK293T and mouse mIMCD3 cells were treated with 10 μM Phen-DC3, cDNA was collected over time, and resulting expression quantified by qPCR. Results were normalized to expression of a non-G4 control, β-actin. hPKD1 mRNA abundance was significantly reduced by more than half compared to mPkd1 after 2 days and by 3.25 times after a week (Fig. 5) (P < 0.01). Polycystin-1 protein was also reduced over this time as judged by Western (Supplementary Fig. 4). We cannot exclude the possibility that hPKD1 transcription factors could be downregulated by G4 stabilization, but identification of those factors would be needed a priori. Even so, the abundance of G4 sequences in hPKD1 (Fig. 1b) is consistent with G4-specific regulation in cis. While Phen-DC3 reduced polycystin-1 mRNA (Fig. 5) it should be noted that naturally folded G4s may impact various pathways, such as alternative splicing, so further studies on G4-based regulation of hPKD1 are required to uncover the mechanisms.

**Fig. 5: G4 DNA impacts the expression of hPKD1.**

G4-induced DNA breaks in hPKD1

We reasoned that if G4 DNAs increase the risk of PKD1 inactivation then G4 DNA stabilization should provoke genotoxic lesions. Upon DNA break formation, histone H2AX becomes phosphorylated at serine 139 (called γH2AX), generating up to a megabase-sized marker for signaling the DNA damage response²⁹. Therefore, we used an anti-γH2AX antibody to assess G4 DNA-induced breaks by ChIP after Phen-DC3 treatment. Since γH2AX signals can be large²⁹, we used a locus on another chromosome, PCNA, as a non-G4 control²³. γH2AX ChIP enriched for hPKD1 (IVS21) compared to hPCNA (1.9/1), which increased to 3.7 times (5.2/1.4) (P < 0.001) when Phen-DC3 was added to cells (Fig. 6a). Neither mPkd1 nor mPcna was enriched by γH2AX ChIP (Fig. 6a), consistent with a lack of G4 DNA at those loci. Experimental repeats with CX-5461 yielded similar results (Fig. 6a), indicating that the DNA breaks are a product of G4 formation and not due to the ligand itself. It is feasible that the γH2AX signals originated at regional G4s located adjacent to hPKD1, since hPKD1 resides in a CpG-rich area of human chromosome 16³⁰, but ChIPs for RAD51, an essential recombination repair protein, resulted in a similar G4-dependent enrichment of hPKD1 with no enrichment of hPCNA observed (Fig. 6b). Thus, G4 DNA folding causes DNA breaks in hPKD1 and recombination repair activities respond to those lesions. We conclude that the differences in hPKD1 and mPkd1 stability may be explained by the presence of G4 DNAs in the former and absence of the structure in the latter. Since DNA breaks are precursor lesions for known PKD1 gene inactivation mechanisms, such as loss of heterozygosity^5,6,7 and gene conversion³¹, formation of G4 DNAs in hPKD1 provide a molecular rationale for second hit mutagenesis in ADPKD (Fig. 7).

Fig. 6: G4 stabilization results in activation of the DNA damage response at *PKD1.*

**Fig. 7: Model for G4 DNA-induced second hit mutations in hPKD1.**

Discussion

G4 DNA is a biologically active structure that is well-known to regulate programmed recombination and gene expression^19,20,21,32, so it is likely that the poly-purine repeats in hPKD1 have regulatory roles. Reduction of polycystin-1 below a certain threshold results in cysts³³, making it important to understand how G4 DNA formation influences polycystin-1 levels and its isoforms. The various mechanisms proposed for G4 DNA-modulated transcription include the recruitment of chromatin remodelers, alteration of methylation status, juxtaposition of distant elements via loop formation, or nucleation of liquid-liquid phase separation²⁰, and so G4 DNA motifs throughout hPKD1 may have been retained as structural features for modulating more than one potential regulatory pathway. The effects may also be positive in nature, as G4 DNA folding in the c-MYC promoter has been shown to upregulate expression³⁴. Thus, we propose that the G/C-rich sequences in hPKD1 impart regulatory benefits to the gene that hinge on G4 DNA, R-loop³⁵ and/or H-DNA^14,15 structures, rather than the sequences themselves, but that it comes at some cost to gene stability^19,20,21,32.

Based on the model (Fig. 7), one might also predict increased renal cystogenesis in patients with deficiencies in G4-specific helicases, like BLM or FANCJ, however loss of those G4 DNA resolution activities results in severe phenotypes that may preclude identification of clinically significant renal cysts. Yet, given the essential nature of polycystin-1 and its diverse functions, deficiencies in G4 resolution activities could have developmental consequences via PKD1 deregulation that merits investigation. Regardless, identification of G4 DNA in human but not mouse PKD1 provides a mechanism for second hit gene inactivation that helps explain the autosomal dominant inheritance pattern for ADPKD.

The model that hPKD1 second hit mutagenesis derives from G4 DNA-induced DNA breaks (Fig. 7) enjoys support from prior research on intron 21 showing that the poly-purine repeats from that region promote mutagenesis^4,14,15. In addition, replication assays using an 88 nt. ectopic repeat from IVS21 caused replication-dependent DNA breaks and genetic instability^14,16,18,36, with the G4-binder telomestatin increasing deletions and hypermutation^16,18. Considering that the DNA damage response is activated at hPKD1 upon Phen-DC3 or CX-5461 exposure (Fig. 6), stabilization of G4 DNAs is likely responsible for causing the double-strand breaks. Since the responding repair pathways may initiate strand excision for repair, mutations that arise as a result can be distant from the G4-induced break site, making it difficult to identify the responsible G4 motif. Plus, G4 sequences are numerous and widely distributed in hPKD1 (Fig. 1c). Still, that observation may explain the absence of mutation hot spots^7,11,37, and individual somatic second hit mutations identified in cyst sequencing studies^7,11 are just as widely distributed in hPKD1 as the G4 DNA motifs (Supplementary Fig. 5). Despite that, and interestingly, there are similar independent somatic mutations that flank G4 DNAs in hPKD1, particularly surrounding intron 42^7,11 (Supplementary Fig. 5). Conclusions on the casual relationships and which G4 DNAs are the most mutagenic awaits development of a tractable model for measuring G4-induced second hits.

Stabilization of G4 DNAs has been a strategy for cancer drug development, with CX-5461 showing efficacy in patients with homologous-recombination deficient tumors^38,39 and increased mutagenesis in cultured cells⁴⁰. Based on the potential for G4 DNA to provoke PKD1 inactivation events, an approach that instead destabilizes G4 DNA could feasibly prevent or delay ADPKD onset. There are indeed small molecules identified that can disrupt G4 folding^41,42. The treatment implication for hPKD1 is that destabilization of G4 DNAs within PKD1 would decrease the risk of inactivating mutagenesis and thus limit cystogenesis for at-risk individuals. This approach may have utility beyond ADPKD if G4 DNAs also influence the stability of other tumor suppressors. For instance, mice heterozygous for pathogenic Brca1 mutations are not prone to spontaneous tumors, contrasting with humans who show BRCA1 second hits and cancer predisposition. Correspondingly, G4 DNA appears to be more abundant in human BRCA1 compared to mouse Brca1²³. Therefore, it is possible that disease risk for individuals inheriting a pathogenic and G4-rich tumor suppressor allele could be mitigated through G4 destabilization strategies that lower the likelihood of second hit mutagenesis.

Methods

Ethics

Frozen non-fixed ADPKD and normal human kidney tissue were generous gifts from the University of Kansas Medical Center provided by the NIDDK sponsored (NIH DK126126) Polycystic Kidney Disease Research Resource Consortium. Tissue was de-identified and affirmed to be non-human subjects research by the University of Kansas Medical Center and Western Michigan University Homer Stryker MD School of Medicine Human Research Protection Programs.

Sequences, programs, statistics

Predicted non-overlapping G4 DNA motifs for hPKD1 and mPkd1 were quantitated using QGRS mapper https://bioinformatics.ramapo.edu/QGRS/index.php²² using a 45-nucleotide window, 3 tandem repeats minimum and 8 nucleotides gap between repeats. Returned G4 DNA motifs had a G-score²² > 64. Both strands were queried between the start and stop codons of each gene. Experimentally detected G4 DNAs were curated by Endoquad https://endoquad.chenzxlab.cn/#/group-g4²³. Primers were designed with NCBI primer blast and synthesized by Integrated DNA technologies. Sequences for PCR are included in Supplementary Table 1, Supplementary Table 2, and Supplementary Table 3. Drawings were created using Biorender.com. Significance was calculated and graphs generated using Prism software. Unless otherwise stated, standard error is shown in graphs and significance was calculated from at least six technical replicates from two independent experiments.

Reagents and cells

Phen-DC3 (Sigma, SML2298) and CX-5461 (MedChemExpress, HY-13323) were dissolved in DMSO and used at a working concentration of 10 µM and 100 nM, respectively. BG4 antibody was purchased from Millipore (MABE917), and anti-γH2AX and anti-RAD51 from Novus Biologicals (NB 100-74435, NB 100-148). PC-1 antibody (7E12) (1/200, Western) was graciously provided by Dr. Chris Ward, University of Kansas Medical Center. SG4 and SG4-R105A plasmids (pHEN2-SG4 and pHEN2-SG4 R105A) were gifts from Dr. Shankar Balasubramanian²⁷(Addgene plasmids 196071 and 196072). SG4 nanobodies were purified essentially as described²⁷. Briefly, BL21 (DE3) E. coli with each plasmid were induced at an OD₆₀₀ 0.4, with 0.5 mM IPTG, cells collected after overnight incubation at 28 °C and protein purified by Nickel chromatography. Protein was eluted with imidazole (200 mM) and dialyzed in PBS at 4 °C. Protein purity (>95%) was judged by SDS-PAGE and Coomassie staining. Proteins were brought to 5% glycerol and stored at −80 °C. HEK293T were a gift from Tom Rothstein (Western Michigan University Homer Stryker MD School of Medicine) and mIMCD3 was purchased from the ATCC (cat# CRL-2123). HEK293T cells were cultured in opti-MEM (Gibco, 31985-070) with 10% FBS and 1% penicillin-streptomycin (Corning, 30-001-CI); mIMCD3 cells were cultured in DMEM (Gibco, 11995-065) with 10% FBS and 1% penicillin-streptomycin.

G4 DNA detection in vitro

G4 DNA oligonucleotide for the dot blot (Fig. 2a) corresponds to a repeat in intron 1, 5′-TTTTTAGAGGTGGGAGGGGCTGGCAGGGAGGGAGAGGT, except for the additional 5’ thymines. The GT oligo, 5′-TTTTTAGAGGTGTGAGTTGCTGGCAGTGAGTGAGAGGT is the same as the G4 DNA repeat except where guanines were interrupted with thymine (bolded). Each oligo was suspended in TE with 100 mM KCl and folded by incubation in a 98 °C water bath that was allowed to slowly come to room temperature. 200 pM of oligo, diluted 1:1 in the same buffer, was applied by a dot blot apparatus to Hybond nylon membrane (Amersham, PN 303 N), cross-linked by exposure to UV transilluminator for 2 min, and then blocked in 5% milk in TBS (50 mM Tris, 50 mM KCl), followed by incubation with SG4 or SG4-R105A (1/10 dilution) overnight, SG4-R105A is mutant for G4 binding²⁷. Membranes were washed 3 times with TBS-tween, followed by 1 h incubation with 1/800 dilution of a rabbit anti-FLAG antibody (Cell Signaling Technology cat# 14793S). The membrane was then washed 3 times with TBS-T, followed by incubation for 1 h with HRP-conjugated anti-rabbit IgG (Thermo Scientific, 65-6120) diluted 1/3000 in TBS. Unbound antibody was removed with 3 washes of TBS-T and luminescence detected after 5 min incubation with WesternSure premium chemiluminescent substrate (Li-COR, 926-95000) using a Li-COR Western Blot imager. The same blot was stained with SYBR gold (Invitrogen, S11494) and imaged to show oligonucleotide loading. A representative dot blot is shown for no less than four independent experiments. CD spectra were collected as previously described⁴³; briefly, the G4 oligonucleotide is derived from an IVS1 repeat 5′- CTGGCAGGGAGGGAGAGGTGGGAGGGGCTGGCA, and the GT control contains 6 substitutions (underlined) to decrease G4 folding capability 5′- CTGGCAGTGAGTGAGAGGAGTGAGTGTCTGGCA. Both oligonucleotides were subjected to G4 DNA folding conditions in a solution of TE containing 100 mM KCl incubated in a 98 °C water bath that was allowed to slowly come to room temperature. An Aviv model 215 spectrometer was used at 37 °C with a 1 cm path length cuvette. The spectra for parallel G4 structures peaks at 260, and dips at 240 nm^24,25. CD results are shown for the average of three scans from one of two independent assays.

PKD1 RT-qPCR

HEK293T and mIMCD3 cells were incubated with DMSO alone or with 10 µM Phen-DC3 in DMSO. Cells were collected, washed with PBS, and mRNA prepared using a total RNA Miniprep kit (New England Biolabs, T2010S) at the indicated time points. For the zero timepoint, cells were collected after <1 h of DMSO or Phen-DC3 addition and RNA extracted using a Total RNA Miniprep kit (NEB, T2010S) and 500 ng was converted to cDNA with a Protoscript cDNA kit (New England Biolabs, E6560S). One microliter of cDNA was used in qPCR with primers specific to hPKD1 and mPkd1. The amount of PKD1 amplification was normalized to β-actin qPCR. qPCR reactions were completed in triplicate using ThermoScientific QuantStudio qPCR machine. Reactions used 1x PowerUp™ SYBR™ green (Applied Biosystems, A25742). The amount of PKD1 amplification was normalized to β-actin by subtracting to generate DeltaCt values, and amplification results were displayed as 2^-DeltaCt⁴⁴.

Immunofluorescence and Immunohistochemistry

HEK293T were fixed with 100% ice cold methanol, washed 3 times in PBS, followed by incubation with PBS 1% Triton at 37 °C for 30 min for permeabilization. Cells were blocked with 10 % normal goat serum (Vector labs, S-1000), treated with 500 ng/µl RNase A (New England Biolabs, #T3018) and incubated overnight at 4 °C with BG4 (1:50) primary antibody. The following day cells were washed 3 times with PBS-T and incubated with 1:800 Rabbit anti-FLAG (Cell Signaling Technology, cat# 14793S) in PBS-T + 1% goat serum for 1 h at 37 °C. After washing 3 times with PBS-T, anti-rabbit Alexa Fluor 568 (Invitrogen, cat# A11011) 1:1000 was added and incubated 1 h at room temperature. After 3 washes with PBS, antifade mounting medium with DAPI (VECTASHIELD, H-1500) were added to mounted slides (blue). Images were collected using a Nikon A1R+ confocal microscope.

ADPKD tissue was taken from a minimally cystic region from one 56-year-old male with a BUN of 45 mg/dL and creatinine of 7.03 mg/dL. Normal human kidney tissue was derived from an aged-matched male. The tissue blocks were embedded in Tissue Tek Optimal Cutting Temperature compound (Sakura Finetek, 4583) and sectioned. Five micron-thick kidney sections were fixed by incubation in ice-cold methanol for 10 min at −20 °C, then washed in PBS, blocked with 10% normal goat serum (NGS) for 1 h at room temperature, then treated with SG4 or control (SG4-R105A)²⁷ primary antibodies (1:50) overnight at 4 °C. The following day tissue sections were washed three times with PBS at RT and incubated with rabbit anti-FLAG (1:800, Cell Signaling Technology, cat# 14793S) secondary antibody for 1 h at RT. Tissue sections were washed three times in PBS at RT and incubated with FITC goat anti-rabbit (1:400, Vector FL-1000) for 1 h at RT. Following washing three times in PBS, sections were mounted with VECTASHIELD antifade with DAPI and slides were viewed on a Zeiss Axioskop fluorescence microscope and images captured with a SPOT RT sCMOS digital camera.

CASFISH imaging followed methods described by Deng et al.²⁸. with the following modifications; SNAP-tagged dCAS9, TMR-snap ligand, and sgRNA synthesis kit (New England Biolabs, #M0652, #S9105S, #E3322V) were used as reagents, and since tween disrupted tissue morphology PBS alone was use in washes. 94 sgRNAs were designed using ChopChop, https://chopchop.cbu.uib.no/ and targeted to a ∼12 kb region encompassing the 3’ end of PKD1 starting at exon 35 and extending into the 3’ tail of adjoining TSC2, with the intent to limit sgRNA reactivity with PKD1 pseudogenes. Oligonucleotides were o-pool DNAs generated by Integrated DNA technologies. Sequences for sgRNA synthesis are available in Supplementary Data 1. Following labeling of normal human or ADPKD kidney tissue sections with SG4 or R105, as described above, sections were incubated with Oregon Green (NEB, cat# S9104S) or TMR-STAR (NEB, cat# 9105S)-labeled dCAS9 assembled with hPKD1 sgRNAs, described above, at 37 °C for 30 min. Following washing three times with PBS, sections were mounted with Vectashield medium with DAPI (Vector) and images captured using a Nikon A1R+ confocal microscope. For controls, sections were also incubated with SG4 nanobody, followed by Fluorescein or Texas Red labeled dCAS9, without hPKD1 sgRNAs.

Statistics and reproducibility

Images selected for normal human or ADPKD tissue labeling were representative and from one of four independent labeling experiments. For qPCR of cDNA, results were from three independent experiments and RM one-way ANOVA was used to calculate significance. ChIP experiments were repeatable, and data is shown for two or more independent experiments with three technical replicates each. Assays for each ChIP experimental endpoint (i.e., G4 formation (Fig. 4) or DNA damage response (Fig. 6)) are represented by four independent experiments using two different G4-specific ligands (Phen-DC3 and CX-5461). Significance was calculated with Prism using two-way ANOVA.

BG4 IP and chromatin IPs

HEK293T or mIMCD3 cells were plated at a density of 1.5 × 10⁶ cells/plate and allowed to come to 90% confluence prior to treatment with 10 µM Phen-DC3, 100 nM CX-5461, or DMSO alone for 5 h, then crosslinked with 1% formaldehyde for 10 min. Crosslinking was halted by addition of 125 mM glycine. PBS-washed cells were collected by centrifugation and the cell pellet resuspended in lysis buffer (50 mM HEPES, 140 mM NaCl, 1 mM EDTA, 1% triton X-100, 0.1% sodium deoxycholate, 0.1% SDS, and protease inhibitor) then sonicated to produce fragments averaging 1–2 kb. BG4 antibody (1/40), anti-γH2AX (1/100), and anti-RAD51 (1/100) were used for IP experiments where indicated. Anti-FLAG magnetic beads (Sigma, cat# M8823) washed with calf thymus DNA were used to IP BG4-associated DNAs, and protein A resin beads (GenScript, L00210S) were used to IP γH2AX-associated DNAs. Crosslinks from precipitated chromatin were reversed at 65 °C overnight. DNA was purified with Monarch PCR & DNA cleanup kit (New England Biolabs, T1030S). DNA enrichment was quantified by qPCR using hPKD1 or mPkd1-specific and hPCNA or mPcna-specific primers. PCR primers were designed to amplify a region 267 nt upstream of a G4 DNA-dense (IVS21) intron and near a G4 DNA-sparse (IVS34) intron within hPKD1, allowing comparisons of relative enrichments for two regions within the same gene. For mPkd1, primer sets were selected to capture template enrichments of comparable nucleotide positions (IVS21 and IVS37). Amplicons were verified by agarose electrophoresis and DNA sequencing. Input DNAs were equal and verified by qPCR. Following real time PCR, raw target Ct values were collected from chromatin incubated with beads only (no antibody, background) as well as chromatin incubated with beads plus primary antibody. Data was normalized (delta Ct) by subtracting the raw Ct value of beads only background from the raw Ct value of beads and antibody IP (i.e., Ct_{antibody+beads} – Ct_beads). Results displayed are from two or more independent assays and are representative of no less than four independent experiments.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Data is available within the paper, supplementary figures, and the source data file. Source data are provided with this paper.

References

Knudson, A. G. Mutation and cancer: statistical study of retinoblastoma. Proc. Natl. Acad. Sci. USA 68, 820–823 (1971).
Article ADS PubMed PubMed Central MATH Google Scholar
Qiu, J., Germino, G. G. & Menezes, L. F. Mechanisms of cyst development in polycystic kidney disease. Adv. Kidney Dis. Health 30, 209–219 (2023).
Article PubMed PubMed Central Google Scholar
Watnick, T. J. et al. Somatic mutation in individual liver cysts supports a two-hit model of cystogenesis in autosomal dominant polycystic kidney disease one possible explanation for the focal nature of renal cyst formation. Mol. Cell 2, 247–251 (1998).
Article CAS PubMed MATH Google Scholar
Koptides, M. et al. Loss of heterozygosity in polycystic kidney disease with a missense mutation in the repeated region of PKD1. Hum. Genet. 103, 709–717 (1998).
Article CAS PubMed Google Scholar
Brasier, J. L. & Henske, E. P. Loss of the polycystic kidney disease (PKD1) region of chromosome 16p13 in renal cyst cells supports a loss-of-function model for cyst pathogenesis. J. Clin. Investig. 99, 194–199 (1997).
Article CAS PubMed PubMed Central Google Scholar
Qian, F., Watnick, T. J., Onuchic, L. F. & Germino, G. The molecular basis of focal cyst formation in human autosomal dominant polycystic kidney disease type I. Cell 87, 979–987 (1996).
Article CAS PubMed Google Scholar
Zhang, Z. et al. Detection of PKD1 and PKD2 somatic variants in autosomal dominant polycystic kidney cyst epithelial cells by whole-genome sequencing. J. Am. Soc. Nephrol. 32, 3114–3129 (2021).
Article PubMed PubMed Central MATH Google Scholar
Paul, B. M. & Vanden Heuvel, G. B. Kidney: polycystic kidney disease. Wiley Interdiscip. Rev. Dev. Biol. 3, 465–487 (2014).
Article CAS PubMed PubMed Central MATH Google Scholar
Grantham, J. J. & Calvet, J. P. Polycystic kidney disease: in danger of being X-rated? Proc. Natl. Acad. Sci. USA 98, 790–792 (2001).
Article ADS CAS PubMed PubMed Central MATH Google Scholar
Igarashi, P. & Somlo, S. Genetics and pathogenesis of polycystic kidney disease. J. Am. Soc. Nephrol. 13, 2384–2398 (2002).
Article CAS PubMed MATH Google Scholar
Tan, A. Y. et al. Somatic mutations in renal cyst epithelium in autosomal dominant polycystic kidney disease. J. Am. Soc. Nephrol. 29, 2139–2156 (2018).
Article CAS PubMed PubMed Central MATH Google Scholar
Bacolla, A., Jaworski, A., Connors, T. D. & Wells, R. D. PKD1 unusual DNA conformations are recognized by nucleotide excision repair. J. Biol. Chem. 276, 18597–18604 (2001).
Article CAS PubMed Google Scholar
Blaszak, R. T., Potaman, V., Sinden, R. R. & Bissler, J. J. DNA structural transitions within the PKD1 gene. Nucleic Acids Res. 27, 2610–2617 (1999).
Article CAS PubMed PubMed Central Google Scholar
Patel, H. P., Lu, L., Blaszak, R. T. & Bissler, J. J. PKD1 intron 21: triplex DNA formation and effect on replication. Nucleic Acids Res. 32, 1460–1468 (2004).
Article CAS PubMed PubMed Central MATH Google Scholar
Van Raay, T. J. et al. A 2.5 kb polypyrimidine tract in the PKD1 gene contains at least 23 H-DNA-forming sequences. Micro. Comp. Genom. 1, 317–327 (1996).
MATH Google Scholar
Rider, S. D. et al. Stable G-quadruplex DNA structures promote replication-dependent genome instability. J. Biol. Chem. 298, 101947 (2022).
Article CAS PubMed PubMed Central Google Scholar
Piontek, K. B. & Germino, G. G. Murine Pkd1 introns 21 and 22 lack the extreme polypyrimidine bias present in human PKD1. Mamm. Genome 2, 194–196 (1999).
Article Google Scholar
Gadgil, R. Y. et al. Replication stress at microsatellites causes DNA double-strand breaks and break-induced replication. J. Biol. Chem. 295, 15378–15397 (2020).
Article CAS PubMed PubMed Central MATH Google Scholar
Sato, K. & Knipscheer, P. G-quadruplex resolution: from molecular mechanisms to physiological relevance. DNA Repair 130, 103552 (2023).
Article CAS PubMed Google Scholar
Robinson, J., Raguseo, F., Nuccio, S. P., Liano, D. & Di Antonio, M. G-quadruplex structures: more than simple roadblocks to transcription? Nucleic Acids Res. 49, 8419–8431 (2021).
Article CAS PubMed PubMed Central Google Scholar
Spiegel, J., Adhikari, S. & Balasubramanian, S. The structure and function of DNA G-quadruplexes. Trends Chem. 2, 123–136 (2020).
Article CAS PubMed PubMed Central MATH Google Scholar
Kikin, O., D’Antonio, L. & Bagga, P. S. QGRS mapper: a web-based server for predicting G-quadruplexes in nucleotide sequences. Nucleic Acids Res. 34, W676–W682 (2006).
Qian, S. H. et al. EndoQuad: a comprehensive genome-wide experimentally validated endogenous G-quadruplex database. Nucleic Acids Res. 52, 72–80 (2024).
Article MATH Google Scholar
Kypr, J., Kejnovská, I., Renčiuk, D., & Vorlíčková, M. Circular dichroism and conformational polymorphism of DNA. Nucleic Acids Res. 37, 1713–1725 (2009).
Article CAS PubMed PubMed Central Google Scholar
del Villar-Guerra, R., Trent, J. O. & Chaires, J. B. G-Quadruplex secondary structure obtained from circular dichroism spectroscopy. Angew. Chem. Int. Ed. Engl. 57, 7171–7175 (2018).
Biffi, G., Tannahill, D., Mccafferty, J. & Balasubramanian, S. Quantitative visualization of DNA G-quadruplex structures in human cells. Nat. Chem. 5, 182–186 (2013).
Article CAS PubMed PubMed Central Google Scholar
Galli, S. et al. DNA G-quadruplex recognition in vitro and in live cells by a structure-specific nanobody. J. Am. Chem. Soc. 144, 23096–23103 (2022).
Article CAS PubMed PubMed Central MATH Google Scholar
Deng, W., Shi, X., Tjian, R., Lionnet, T. & Singer, R. H. CASFISH: CRISPR/Cas9-mediated in situ labeling of genomic loci in fixed cells. Proc. Natl. Acad. Sci. USA 112, 11870–11875 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
Rogakou, E. P., Pilch, D. R., Orr, A. H., Ivanova, V. S. & Bonner, W. M. DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. J. Biol. Chem. 273, 5858–5868 (1998).
Article CAS PubMed Google Scholar
Germino, G. G. et al. The gene for autosomal dominant polycystic kidney disease lies in a 750-kb CpG-rich region. Genomics 13, 144–151 (1992).
Article CAS PubMed Google Scholar
Watnick, T. J., Gandolph, M. A., Weber, H., Neumann, H. P. H. & Germino, G. G. Gene conversion is a likely cause of mutation in PKD1. Hum. Mol. Genet. 7, 1239–1243 (1998).
Dell’Oca, M. C. et al. Spotlight on G-quadruplexes: from structure and modulation to physiological and pathological roles. Int. J. Mol. Sci. 25, 3162 (2024).
Article PubMed PubMed Central MATH Google Scholar
Ong, A. C. M. & Harris, P. C. A polycystin-centric view of cyst formation and disease: the polycystins revisited. Kidney Int. 88, 699–710 (2015).
Article CAS PubMed PubMed Central MATH Google Scholar
Esain-Garcia, I. et al. G-quadruplex DNA structure is a positive regulator of MYC transcription. Proc. Natl. Acad. Sci. USA 121, e2320240121 (2024).
Article CAS PubMed PubMed Central Google Scholar
Parsons, A. M. et al. Human PKD1 sequences form R-loop structures in vitro. Micropubl. Biol. 2024, https://doi.org/10.17912/micropub.biology.001058 (2024).
Liu, G. et al. Replication fork stalling and checkpoint activation by a PKD1 locus mirror repeat polypurine-polypyrimidine (Pu-Py) tract. J. Biol. Chem. 287, 33412–33423 (2012).
Article CAS PubMed PubMed Central Google Scholar
Kozlowski, P., Bissler, J., Pei, Y. & Kwiatkowski, D. J. Analysis of PKD1 for genomic deletion by multiplex ligation-dependent probe assay: absence of hot spots. Genomics 91, 203–208 (2008).
Article CAS PubMed MATH Google Scholar
Hilton, J. et al. Results of the phase I CCTG IND.231 trial of CX-5461 in patients with advanced solid tumors enriched for DNA-repair deficiencies. Nat. Commun. 13, 3607 (2022).
Article ADS CAS PubMed PubMed Central MATH Google Scholar
Xu, H. et al. CX-5461 is a DNA G-quadruplex stabilizer with selective lethality in BRCA1/2 deficient tumours. Nat. Commun. 8, 14432 (2017).
Article ADS CAS PubMed PubMed Central Google Scholar
Koh, G. C. C. et al. The chemotherapeutic drug CX-5461 is a potent mutagen in cultured human cells. Nat. Genet. 56, 23–26 (2024).
Article CAS PubMed Google Scholar
Fracchioni, G. et al. Structural unfolding of G-quadruplexes: from small molecules to antisense strategies. Molecules 29, 3488 (2024).
Article CAS PubMed PubMed Central MATH Google Scholar
Mitteaux, J. et al. Identifying G-quadruplex-DNA-disrupting small molecules. J. Am. Chem. Soc. 143, 12567–12577 (2021).
Article CAS PubMed Google Scholar
Williams, J. D., Fleetwood, S., Berroyer, A., Kim, N. & Larson, E. D. Sites of instability in the human TCF3 (E2A) gene adopt G-quadruplex DNA structures in vitro. Front. Genet. 6, 177 (2015).
Article PubMed PubMed Central Google Scholar
Schmittgen, T. D. & Livak, K. J. Analyzing real-time PCR data by the comparative CT method. Nat. Protoc. 3, 1101–1108 (2008).
Article CAS PubMed MATH Google Scholar

Download references

Acknowledgements

Research was supported by NIH/NIDDK R15DK119864 to E.L. and by Western Michigan University Homer Stryker MD School of Medicine. Studies utilized resources provided by the Kansas PKD center (NIH DK126126) and NIDDK sponsored Polycystic Kidney Disease Research Resource Consortium. The authors wish to thank Kristi Bailey, WMed histology, for assistance with tissue sectioning and Michael Clemente, WMed Imaging Core, for assistance with confocal microscopy. Drawings were created with BioRender.com.

Author information

Authors and Affiliations

Department of Biomedical Sciences, Western Michigan University Homer Stryker MD School of Medicine, Kalamazoo, MI, USA
Agata M. Parsons, Seth Byrne, Jesse Kooistra, John Dewey, Aaron L. Zebolsky, Gloria Alvarado, Gerrit J. Bouma, Gregory B. Vanden Heuvel & Erik D. Larson

Authors

Agata M. Parsons
View author publications
Search author on:PubMed Google Scholar
Seth Byrne
View author publications
Search author on:PubMed Google Scholar
Jesse Kooistra
View author publications
Search author on:PubMed Google Scholar
John Dewey
View author publications
Search author on:PubMed Google Scholar
Aaron L. Zebolsky
View author publications
Search author on:PubMed Google Scholar
Gloria Alvarado
View author publications
Search author on:PubMed Google Scholar
Gerrit J. Bouma
View author publications
Search author on:PubMed Google Scholar
Gregory B. Vanden Heuvel
View author publications
Search author on:PubMed Google Scholar
Erik D. Larson
View author publications
Search author on:PubMed Google Scholar

Contributions

E.L. directed and supervised the research. E.L., G.J.B, and G.B.V.H conceived experiments and edited the manuscript. J.D. and A.L.Z. analyzed G4 sequences in PKD1. G.A. performed CD scans. A.M.P performed ChIPs, dot blot, Westerns, qPCR, cell culture, and created diagrams with G.J.B. cDNA analysis was performed by A.M.P and S.B., G.B.V.H and J.K. performed IF microscopy and CASFISH. All authors reviewed, provided comments, and approved the manuscript.

Corresponding author

Correspondence to Erik D. Larson.

Ethics declarations

Competing interests

A pending U.S. Provisional Patent Application No. 63/693,922 (inventor E.L.) directed to G4 quadruplex structures in the PKD1 gene has been filed on behalf of the applicant, Western Michigan University Homer Stryker M.D. School of Medicine. The authors declare no other competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Description of Additional Supplementary Files

Supplementary Data 1

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Parsons, A.M., Byrne, S., Kooistra, J. et al. G-quadruplex stabilization provokes DNA breaks in human PKD1, revealing a second hit mechanism for ADPKD. Nat Commun 16, 121 (2025). https://doi.org/10.1038/s41467-024-55684-y

Download citation

Received: 27 March 2024
Accepted: 19 December 2024
Published: 02 January 2025
DOI: https://doi.org/10.1038/s41467-024-55684-y