Abstract
Primary cilia are sensory organelles built and maintained by intraflagellar transport (IFT) multi-protein complexes. Deletion of different IFT-B genes attenuates polycystic kidney disease (PKD) severity in juvenile and adult Autosomal Dominant (AD) PKD mouse models, while deletion of an IFT-A adaptor, Tulp3, attenuates PKD severity in adult mice only. These studies indicate that dysfunction of specific cilia components has potential therapeutic value. To broaden our understanding of cilia dysfunction and its therapeutic potential, here we investigate the impact of global deletion of an IFT-A gene, Thm1, in juvenile and adult ADPKD mouse models. Both juvenile and adult models exhibited increased kidney weight:body weight (KW/BW) ratios, renal cysts, inflammation, lengthened renal cilia, and increased levels of the nutrient sensor, O-linked β-N-acetylglucosamine (O-GlcNAc). Thm1 deletion in juvenile ADPKD mice reduced KW/BW ratios and cortical collecting duct cystogenesis, but increased proximal tubular and glomerular dilations and did not reduce inflammation, cilia lengths, and O-GlcNAc signaling. In contrast, Thm1 deletion in adult ADPKD mice markedly attenuated renal cystogenesis, inflammation, cilia lengths, and O-GlcNAc. Thus, unlike IFT-B genes, the role of Thm1 deletion in ADPKD mouse models is development-specific. Unlike an IFT-A adaptor gene, deleting Thm1 in juvenile ADPKD mice is partially ameliorative. Our studies suggest that different microenvironmental factors found in distinct nephron segments and between developing and mature kidneys modify ciliary homeostasis and ADPKD pathobiology. Further, elevated levels of O-GlcNAc, which regulates cellular metabolism and ciliogenesis, may be a novel feature and critical regulator of certain key ADPKD pathological processes.
Introduction
Autosomal Dominant Polycystic Kidney Disease (ADPKD) is among the most common, fatal monogenetic diseases, affecting approximately 1:1000 individuals worldwide1. ADPKD is characterized by the growth of large fluid-filled renal cysts, which cause injury and fibrosis and can lead to end-stage renal disease by the 6th decade of life. Tolvaptan is the only FDA-approved therapy, but has variable effectiveness and aquaresis side effects2, 3. Thus, the need to discover additional underlying disease mechanisms and design new therapeutic strategies continues.
Primary cilia are small, antenna-like sensory organelles that play an important role in ADPKD pathobiology via mechanisms that remain unclear. ADPKD is caused by mutations in PKD1 or PKD2, which encode polycystin 1 (PC1) and polycystin 2 (PC2), respectively4. PC1 and PC2 form an ion-channel receptor complex that functions at the primary cilium. While PC1 and PC2 also localize to other subcellular compartments, analyses of human ADPKD primary renal epithelial cells, of mouse models harboring human ADPKD mutations, and of an ethylnitrosourea (ENU)-induced Pkd2 mouse mutation that causes ciliary exclusion of PC2, indicate that deficiency of polycystins from the cilium is sufficient to cause ADPKD5–7.
Primary cilia are synthesized and maintained via intraflagellar transport (IFT), which is the bi-directional transport of protein cargo along a microtubular axoneme. Two multiprotein complexes mediate IFT. The IFT-B complex interacts with the kinesin motor and mediates anterograde IFT, while the IFT-A complex together with cytoplasmic dynein mediates retrograde IFT. IFT-A proteins are also required for ciliary import of membrane and signaling molecules8–10. Additionally, an IFT-A adaptor, TULP3, binds to the IFT-A complex and brings in certain G-protein signaling molecules. In mice, deletion of Ift-A or -B genes or of Tulp3, either perinatally or in the embryonic kidney results in renal cystic disease11–13. However, these mutants differ from ADPKD models in manifesting generally smaller renal cysts and greater fibrosis relative to cyst size14, 15. Additionally, Ift-A and –B mutants differ in cilia phenotype – often shortened and absent cilia, respectively - and can also show opposing signaling phenotypes, reflecting the differing functions of IFT-A and -B12, 16–18. Intriguingly, deletion of Ift-B genes, Kif3a, Ift20 and Ift88 in juvenile and adult Pkd1 or Pkd2 conditional knock-out (cko) mice reduces PKD severity19–22, while deletion of Tulp3 attenuates PKD in adult mice only. Collectively, these studies indicate that a component of cilia dysfunction has potential therapeutic value.
The role of IFT-A deficiency in ADPKD has not been reported. THM1 (TPR-containing Hedgehog modulator 1; also termed TTTC21B) is an ortholog of Chlamydomonas IFT139, an IFT-A gene16. Causative and modifying mutations in THM1 have been identified in patients with nephronophthisis, Bardet Biedl syndrome, Meckel syndrome, Jeune syndrome and renal agenesis14. Characteristic of IFT-A mutation, deletion of Thm1 impairs retrograde IFT, causing accumulation of proteins in bulb-like distal tips of shortened primary cilia16. Thm1 loss also impairs cilia entry of membrane-associated proteins, delays and reduces ciliogenesis, and promotes serum-induced cilia loss23. In mice, Thm1 deletion recapitulates many of the clinical manifestations of ciliopathies16, 24, 25. Perinatal global deletion of Thm1 results in renal cystic disease24, while deletion of Thm1 in adult mice does not result in a renal phenotype by 3 months of age, consistent with the developmental time-frame that determines whether loss of a cystogenic gene will cause rapid- or slow-progressing renal cystic disease26. To expand on the role of ciliary dysfunction in ADPKD, here we investigate the role of Thm1/ IFT-A deficiency in juvenile and adult ADPKD mouse models.
Results
Perinatal deletion of Thm1 in Pkd2 conditional knock-out mice reduces cortical collecting duct cystogenesis, but does not improve kidney function
To examine the effect of IFT-A deficiency in a rapidly progressing ADPKD mouse model, we deleted Thm1 together with Pkd2 at postnatal day (P) 0, and examined the renal phenotypes of control, Thm1 conditional knock-out (cko), Pkd2 cko and Pkd2;Thm1 double knock-out (dko) mice at P21. At this early stage, Thm1 cko mice on a pure C57BL6/J background showed kidneys with some cortical tubular dilations, reduced kidney weight/body weight (KW/BW) ratios, and elevated blood urea nitrogen (BUN) levels (Figures 1A-1C). Pkd2 cko mice showed cysts in both renal cortex and medulla, as well as increased KW/BW ratios and BUN levels. Relative to Pkd2 cko mice, Pkd2;Thm1 dko mice showed decreased cystogenesis specifically in the cortex (Figures 1D-1F), reduced KW/BW ratios, but similar BUN levels, indicating kidney dysfunction was not improved.
We examined the tubular origin of the renal cortical dilations and cysts. Thm1 cko renal cortical dilations were mostly LTL+, indicating proximal tubules, while fewer were THP+ or DBA+, marking loop of Henle and collecting duct, respectively (Figure 1G). Pkd2 cko renal cortices showed LTL+ dilations, THP+ cysts, and multiple, large DBA+ cysts. Relative to Thm1 cko and Pkd2 cko kidneys, Pkd2; Thm1 dko cortices showed greater LTL+ dilations (Figures 1G, 1H). Relative to Pkd2 cko kidneys, Pkd2; Thm1 dko cortices showed similar THP+ cystogenesis (Figures 1G, S1A), but decreased DBA+ cysts (Figures 1G, 1I). Glomerular dilations were also present in Pkd2 cko kidneys, and increased in Pkd2;Thm1 dko kidneys (Figures 1J, 1K). These data reveal nephron segment-specific effects of Thm1 deletion on a Pkd2 cko background.
Pkd2 deletion increases cilia length on renal epithelia
In control kidneys, ciliary axonemes marked by acetylated α-tubulin were 3.0μm and 2.1μm for LTL+ and DBA+ cortical tubular epithelial cells, respectively (Figure 2). We also noted qualitative differences between LTL+ and DBA+ primary cilia, with the former cilia being thinner and longer, and the latter being thicker and more rod-like. Cilia lengths of LTL+ and DBA+ cells were increased in Pkd2 cko cortices. However, relative to Pkd2 cko cells, cilia lengths of Pkd2;Thm1 dko LTL+ cells were further increased, but similar for DBA+ cells. These differences reveal tubular-specific effects on cilia phenotype.
ARL13B is a ciliary membrane protein essential for ciliogenesis27, and in cultured Thm1-null mouse embryonic fibroblasts, ARL13B in cilia is reduced23. We therefore examined whether ARL13B may have a role in the renal ciliary phenotypes. However, in DBA+ cortical renal epithelial cells, ARL13B intensities were similar across the mutant genotypes (Figure S2).
We examined whether Thm1 expression differed in proximal tubules versus cortical collecting duct. However, X-gal staining on kidneys of mice harboring a Thm1-lacZ allele showed relatively ubiquitous expression in the cortex and present in both LTL+ and cortical DBA+ tubules (Figure S3).
Thm1 deletion increases inflammation
To further evaluate the effect of Thm1 deletion on disease severity, we examined proliferation. Nuclear staining of proliferating cell nuclear antigen (PCNA) of LTL+ tubules was similar across control, Thm1 cko and Pkd2 cko, but slightly elevated in Pkd2;Thm1 dko kidneys (Figures 3A, 3B). In contrast, PCNA+ nuclei were increased in Pkd2 cko and Pkd2;Thm1 dko DBA+ non-cystic tubules, and further increased in Pkd2 cko and Pkd2;Thm1 dko DBA+ cysts (Figure 3C). These data support that increased proliferation is an early driver of ADPKD renal cystogenesis.
Cyst growth compresses surrounding parenchyma, leading to injury, inflammation, and fibrosis. We immunostained kidney sections for alpha smooth muscle actin (αSMA) and F4/80 to examine the presence of myofibroblasts and macrophages, respectively, which contribute to pro-inflammatory and pro-fibrotic processes. While control kidneys showed αSMA around blood vessels and a few F4/80+ cells, Thm1 cko kidneys showed increased αSMA around glomeruli and tubular dilations, and Pkd2 cko and Pkd2;Thm1 dko kidney showed even greater αSMA+ and increased F4/80+ labelling surrounding glomeruli, tubular dilations and cysts (Figure 3D-3F). Additionally, transcripts of inflammatory molecules, Tnfα and Tgfβ, were increased in Thm1 cko renal extracts, while those of Il6, Tnfα, Tgfβ, Ccl2 and Gli1 were increased in Pkd2 cko and Pkd2;Thm1 dko kidney extracts (Figures 3G-3K). Thus, deletion of Thm1 alone increases inflammatory processes, and its deletion on a Pkd2 cko background results in similar or slightly increased ADPKD inflammation at P21.
Perinatal deletion of Thm1 increases ERK and STAT3 signaling
Increased ERK and STAT3 signaling promote disease progression in ADPKD28–31. ERK activation promotes cell proliferation30, 32 and also acts upstream of mTOR and AMPK pathways regulating cellular metabolism33. STAT3 activation can have proliferative and inflammatory roles31, 34, 35. At P21, P-ERK did not localize with LTL, but localized with DBA (Figure 4A). While a similar number of P-ERK+ tubules was observed across control and the mutant genotypes (Figure 4B), P-ERK intensity was increased in dilated tubules of Thm1 cko mice, and further increased in cyst-lining cells of Pkd2 cko and Pkd2;Thm1 dko mice (Figure 4C). These data support that ERK activation is a driver of renal cyst growth. Using immunohistochemistry, P-STAT3 was revealed to be increased in epithelial cells lining cortical dilations in Thm1 cko kidneys and to be increased further in cyst-lining cells of Pkd2 cko and Pkd2;Thm1 dko kidneys, both in the cortex and medulla (Figure 4E). Western blot analyses also reveal increased ERK and STAT signaling in Pkd2 cko and Pkd2;Thm1 dko renal extracts (Figures 4F, 4G). Thus, Thm1 loss as well as Pkd2 cystic disease increase ERK and STAT3 signaling.
O-GlcNAc is increased in dilated tubules and cysts
Altered cellular metabolism has emerged as another component of ADPKD pathobiology33, 36–38. One of these alterations includes the Warburg effect33, whereby cells preferentially convert the product of glycolysis, pyruvate, into lactate, even in the presence of oxygen, at the expense of pyruvate entering mitochondria to enable oxidative phosphorylation. The nutrient sensor, O-linked β-N-acetylglucosamine (O-GlcNAc) regulates the balance between glycolysis and oxidative phosphorylation39, as well as ciliary homeostasis40, 41, both of which are altered in ADPKD33, 42. We therefore hypothesized that O-GlcNAc signaling is misregulated in PKD. Relative to control, Thm1 cko kidneys showed increased O-GlcNAc expression in nuclei of cells lining dilations in the cortex, while Pkd2 cko and Pkd2;Thm1 dko kidneys showed increased OGlcNAc staining in cyst-lining cells in both the cortex and medulla (Figures 5A, S4). O-GlcNAC levels are regulated by two enzymes, O-GlcNAc transferase (OGT) and O-GlcNAcase (OGA), which transfers and removes the O-GlcNAc moiety on protein substrates, respectively. In Thm1 cko kidneys, OGT was increased in the cytoplasm and nuclei of cells lining cortical dilations. In Pkd2 cko and in Pkd2;Thm1 dko mice, intense OGT staining was present in cyst-lining cells in the cortex and medulla. In control and Thm1 cko kidneys, light OGA staining was present in cytoplasm of cortical and medullary tubules. In Pkd2 cko and in Pkd2;Thm1 dko mice, OGA was increased in the nuclei and cytoplasm of cyst-lining cells in both the cortex and medulla. Western blot analyses showed increased O-GlcNAc levels in mutant renal extracts, with highest levels in Pkd2 cko renal extracts (Figures 5B, 5C). These data suggest that increased O-GlcNAc signaling is a feature of ADPKD.
Deletion of Thm1 in adult Pkd2 or Pkd1 conditional knock-out mice markedly attenuates ADPKD renal cystogenesis
We next examined the role of IFT-A deficiency in slowly progressive adult ADPKD mouse models. We deleted Thm1 together with Pkd1 at P35 and examined renal phenotypes at 6 months of age. Thm1 cko mice showed normal kidney morphology and BUN levels (Figure S5A). In Pkd1 cko mice, renal cysts were mostly in the cortex, with the largest and most abundant cysts being DBA+, fewer cysts being THP+ and dilations being LTL+ (Figure 6A). Notably, all these features were reduced in Pkd1;Thm1 dko kidneys. Similarly, KW/BW ratios and BUN levels were elevated in Pkd1 cko mice, and corrected in Pkd1;Thm1 dko mice (Figures 6B, 6C).
Since PKD2 mutations result in less severe PKD, we deleted Thm1 together with Pkd2 at P28 and examined renal phenotypes at 6 months of age. Thm1 cko kidney morphology and BUN levels resembled control (Figure S5B). Pkd2 cko mice show renal cysts mostly in the cortex, with the largest cysts being DBA+, and smaller cysts being LTL+ or THP+ (Figure 6D). In contrast, in Pkd2;Thm1 dko mice, the Pkd2 cko cystic phenotype is largely corrected morphologically. In Pkd2 cko mice, KW/BW ratios were similar to control and BUN levels showed a trend toward a slight elevation, reflecting the mild disease induced in adulthood. In Pkd2;Thm1 dko mice, KW/BW ratios were slightly reduced, due to increased body weight caused by global deletion of Thm125, and BUN levels showed a slightly decreasing pattern relative to Pkd2 cko mice.
Deletion of Thm1 in adult ADPKD mouse models reduces cilia lengths of cortical collecting duct renal epithelia
Similar to juvenile ADPKD models, cilia lengths of Pkd1 cko and Pkd2 cko DBA+ adult renal epithelial cells were increased (Figures 6G, 6H, 6J, 6K). However, in contrast to juvenile models, cilia lengths of Pkd1;Thm1 and Pkd2;Thm1 dko DBA+ epithelia were reduced relative to those of Pkd1 cko and Pkd2 cko epithelia and similar to control. Additionally, human ADPKD sections had longer renal epithelial cilia than normal human kidney (NHK) sections (Figures 6I, 6L), supporting that increased cilia length is also a feature of the human disease22.
Deletion of Thm1 in adult Pkd1 conditional knock-out mice reduces proliferation, inflammation, P-ERK, P-STAT3, and O-GlcNAc
We examined the extent of ADPKD attenuation by Thm1 deletion. Pkd1 cko kidneys showed increased PCNA in cyst-lining cells, as well as increased αSMA and F4/80 around cysts and glomeruli, which were markedly reduced in Pkd1;Thm1 dko kidneys (Figures 7A-D). Transcripts of pro-inflammatory molecules, Ccl2, Il6, Tnfα, Tgfβ, Ccl2 and Gli1, were elevated in Pkd1 cko kidneys, but reduced in Pkd1;Thm1 dko kidneys (Figure 7E). Similarly, P-ERK and P-STAT3 immunostaining was increased in Pkd1 cko kidneys and reduced in Pkd1;Thm1 dko kidneys (Figures 7J, 7K). Thus, proliferative and pro-inflammatory pathways are attenuated in late-onset ADPKD by deletion of Thm1.
Finally, we examined O-GlcNAcylation. In Pkd1 cko kidneys, increased O-GlcNAc was present in cyst-lining epithelia of the cortex and in proteinaceous substances within some cysts, and in tubules of the medulla (Figures 8A, S6). OGT and OGA were also increased in cyst-lining cells of the cortex and in some tubules of the medulla. In Pkd1;Thm1 dko kidneys, O-GlcNAc, OGT and OGA staining were reduced. Western blot showed increased O-GlcNAc levels in Pkd1 cko renal extracts (Figure 8B), and levels were reduced in Pkd1;Thm1 dko extracts. These data suggest that increased O-GlcNAcylation is a feature also of slowly progressive ADPKD mouse models. Further, deletion of Thm1 in adult ADPKD mouse models attenuates perturbation of this metabolic regulator.
Discussion
To expand on the ciliary components whose dysfunction in ADPKD mouse models can attenuate disease severity19, 21, 43, we examined the role of deletion of IFT-A gene, Thm1. Distinct from IFT-B, deletion of Thm1 does not improve kidney function in a juvenile ADPKD mouse model, but lessens most disease aspects in an adult model (Figure 9). Distinct from IFT-A adaptor, Tulp3, deletion of Thm1 in a juvenile ADPKD mouse model had partially ameliorative effects, reducing KW/BW ratios and cortical collecting duct cystogenesis. Our data also reveal that the effects of Thm1 deletion in juvenile ADPKD mice are nephron-specific, with no effect on loop of Henle-derived cysts, and exacerbation of proximal tubular and glomerular dilations. This suggests varying microenvironments between nephron segments modify cilia dysfunction. The differential effects of Tulp3 or Thm1 deletion in juvenile versus adult ADPKD mouse models further suggest distinct microenvironments between developing versus mature kidneys that influence not just renal cystogenesis, but also inflammation.
Embryonic or perinatal mutation of Kif3a, Ift88, Tulp3 or Thm1 causes renal cystic disease24, 21, 44, 45, indicating that all these genes are required for kidney development. Thus, their differential effects in juvenile ADPKD mouse models may reflect differences in their cellular functions, such as cilia length regulation/ciliary homeostasis, ciliary entry or exit of different protein cargo, and regulation of signaling pathways. While IFT-B and -A complexes co-operate in ciliary homeostasis, these complexes also have distinct roles. IFT-B genes mediate anterograde IFT, while IFT-A genes mediate retrograde IFT as well as ciliary entry of membrane and signaling proteins. In contrast, Tulp3 does not mediate IFT, but is required for ciliary entry of certain membrane-associated and signaling molecules43. Additionally, IFT-B and -A mutants can have opposing signaling phenotypes. Any of these functions can be explored to account for the differing juvenile dko phenotypes.
Increased cilia lengths on renal epithelia of several ADPKD mouse models, PKD1RC/RC, Pkd1 and Pkd2 cko mice46, 47, and recently, on human ADPKD tissue22, have been reported. Our data substantiate that polycystin dysfunction causes increased cilia length in mouse models as well as in human ADPKD, suggesting ciliary mechanisms are likely conserved between mouse and human. Further, our data show that in addition to genotype, cilia structure varies by renal tubule segment and maturation, suggesting that factors within a tubule’s microenvironment affect cilia length. Indeed, multiple factors including intracellular Ca2+ and cAMP, oxidative stress, cytokines, and fluid flow, affect cilia length of renal epithelial cells48–50, indicating that cilia length may be finely regulated in order to maintain renal tubular structure and function. In support of an ameliorative effect of reduced cilia length in ADPKD, inhibition of cilia disassembly in Pkd1 cko mice increased renal cilia length and exacerbated ADPKD51, while in the jck non-orthologous PKD mouse model, which also has increased renal cilia lengths, pharmacological shortening of primary cilia was associated with attenuated PKD52.
Reduced Ccl2 and Wnt signaling in Pkd; Ift-B dko mice20, 22, and reduced P53 enhancing cilia disassembly in Pkd; Ift-B dko cells42 are potential mechanisms by which ablation or shortening of primary cilia attenuate ADPKD. In addition to detecting ligands, primary cilia also detect mechanical cues. Although mechanosensing by primary cilia and the polycystins has been controversial, recent studies have renewed interest in a potential mechanosensory role for the polycystins, particularly regarding tissue microenvironment stiffness53, 54. If sensing of physical forces in the tissue microenvironment is essential to maintaining renal tubular function, then other mechanical cues that would change with cyst growth include shear stress and intraluminal pressure. Cilia length itself could then also be a possible contributing factor in PKD severity. Supporting a role for cilia length, a recent study has shown that primary cilia of proximal tubule epithelial cells transduce shear stress into metabolic pathways that culminate in oxidative phosphorylation55. Finally, by extrapolating findings of cilia studies from the cancer field56, cilia of not only renal tubular epithelial cells, but of interstitial cells might also affect signaling and disease severity.
Our data are the first to reveal increased O-GlcNAcylation in cyst-lining renal epithelial cells of both juvenile and adult ADPKD mouse models, suggesting increased O-GlcNAcylation may be a feature of ADPKD. Increased O-GlcNAcylation is a pathologic feature of diabetic nephropathy57, 58, and in rodent models, has shown to promote various aspects of chronic kidney disease59, 60 and also renal fibrosis61. In contrast, in a mouse model of contrast-induced acute kidney disease, an acute increase in O-GlcNAcylation was protective62, emphasizing important differences between chronic and acute increases in O-GlcNAcylation. While acute changes are adaptive and necessary to maintain cellular health and metabolism, chronic changes are likely to contribute to pathology63. Chronic elevation of O-GlcNAcylation in cancer cells promotes tumor growth, and cancer cells show the Warburg effect, suggesting similar cellular metabolic alterations between ADPKD and cancer. A recent study has demonstrated that O-GlcNAcylation of Phosphoglycerate kinase 1 (PGK1), which catalyzes the first ATP molecule in glycolysis, activates PGK1 to enhance lactate production and reduce mitochondrial oxidative phosphorylation, promoting the Warburg effect39. Perhaps a similar mechanism may occur in ADPKD cells.
OGT localizes to the pericentriolar region during the early phases of ciliogenesis56, and perturbation of O-GlcNAcylation affects cilia length40, 41. Further, the ciliary structural defects caused by Ogt deletion suggest impaired centriole formation and IFT56. Chronic hyper-O-GlcNAcylation in diabetic tissues results in ciliary defects64, demonstrating a causative link between misregulation of O-GlcNAcylation and defective ciliary homeostasis in a disease context. Since increased renal cilia lengths are associated with ADPKD renal cystogenesis46, 47, the regulation of O-GlcNAcylation on ciliogenesis could be another mechanism by which altered O-GlcNAcylation can affect ADPKD. Elucidating the mechanisms by which O-GlcNAc is upregulated in ADPKD and alters cellular metabolism and ciliogenesis could reveal novel mechanisms and therapeutic targets.
In summary, our data demonstrate for the first time the role of IFT-A deficiency in an ADPKD context, revealing differential effects between nephron segments and between developing and mature renal microenvironments. Our findings also reveal for the first time that O-GlcNAcylation is increased in ADPKD. We propose that as a regulator of ciliary homeostasis and of the balance between glycolysis and oxidative phosphorylation, increased O-GlcNAc may drive certain key ADPKD pathological processes.
Materials and Methods
Generation of mice
Pkd1flox/flox, Pkd2flox/flox and ROSA26-Cre mice were obtained from the Jackson Laboratories (Stock numbers 010671, 017292 and 004847, respectively). Generation of Thm1 cko mice has been described previously24: Thm1aln/+; ROSA26CreERT+ male mice were mated to Thm1flox/flox females. Pkd1 floxed alleles were introduced into the colony to generate Thm1flox/flox;Pkd1flox/flox or Thm1flox/flox;Pkd1flox/+ females and Pkd1flox/flox; Thm1aln/+, ROSA26-CreERT/+ males, which were mated. Similarly, Pkd2 floxed alleles were introduced into the colony to generate Thm1flox/flox;Pkd2flox/flox or Thm1flox/flox;Pkd2flox/+ females and Pkd2flox/flox; Thm1aln/+, ROSA26-CreERT/+ males. To generate early-onset Pkd2 models, Thm1flox/flox;Pkd2flox/flox or Thm1flox/flox;Pkd2flox/+ nursing mothers mated to Pkd2flox/flox; Thm1aln/+, ROSA26-CreERT/+ males were injected intraperitoneally with tamoxifen (8mg/40g; Sigma) at postnatal day 0 (P0) to induce gene deletion. Offspring were analyzed at P21. To generate late-onset Pkd1 models, offspring from matings between Thm1flox/flox;Pkd1flox/flox or Thm1flox/flox;Pkd1flox/+ females and Pkd2flox/flox; Thm1aln/+, ROSA26-CreERT/+ males were injected intraperitoneally with tamoxifen (8mg/40g) at P35. To generate late-onset Pkd2 models, offspring from matings between Thm1flox/flox;Pkd2flox/flox or Thm1flox/flox;Pkd2flox/+ females and Pkd2flox/flox; Thm1aln/+, ROSA26-CreERT/+ males were injected intraperitoneally with tamoxifen (8mg/40g) at P28. Mice were analyzed at 6 months of age. All mouse lines were maintained on a pure C57BL6/J background (backcrossed 10 generations). All animal procedures were conducted in accordance with KUMC-IACUC and AAALAC rules and regulations.
Blood Urea Nitrogen Measurements
Mouse trunk blood was collected in Microvette CB 300 Blood Collection System tubes (Kent Scientific), and centrifuged at 1800g at room temperature for 10 minutes to collect serum. BUN was measured using the QuantiChrom Urea Assay Kit (BioAssay Systems) according to the manufacturer’s protocol.
Histology
Kidneys were bisected transversely, fixed in 10% formalin for several days, then processed in a tissue processor and embedded in paraffin. Tissue sections (7μm) were obtained with a microtome. Sections were deparaffinized, rehydrated through a series of ethanol washes, stained with hematoxylin and eosin (H&E), and mounted in Permount (ThermoFisher). Images were taken with a Nikon 80i microscope equipped with a Nikon DS-Fi1 camera. Cystic areas of H&E-stained sections were quantified using ImageJ.
Immunofluorescence
Following deparaffinization and rehydration, tissue sections were subjected to antigen retrieval. Tissue sections were steamed for 15 minutes in Sodium Citrate Buffer (10 mM Sodium Citrate, 0.05% Tween 20, pH 6.0), returned to room temperature, rinsed 10 times in distilled water, washed 5 minutes in PBS, incubated for 5 minutes in 1% SDS in PBS based on a method by Brown et al., 199665, then washed 3 times in PBS. Sections were blocked with 1% BSA in PBS for 1 hour at room temperature, then incubated with primary antibodies alone or together with lectins (Table S1) overnight at 4°C. Sections were washed three times in PBS, then incubated with secondary antibodies (Table S2) for 1 hour. After three washes of PBS, sections were mounted with DAPI Fluoromount-G (Electron Microscopy Sciences). Staining was imaged using a Nikon 80i microscope with a photometrics camera or a Nikon Eclipse TiE attached to an A1R-SHR confocal, with an A1-DU4 detector, and LU4 laser launch.
Immunohistochemistry
Immunohistochemistry was performed as described66 using the following primary and secondary antibodies (Tables S1, S2). Once desired color was obtained, sections were counterstained with haemotoxylin. Staining was imaged using a Zeiss A1 microscope with a Axiocam 105 color camera.
X-gal staining
Kidneys of P0 mice harboring the Thm1-lacZ allele (KOMP Repository) were dissected, fixed for 5 min in 10% formalin, washed in lacZ wash buffer, then stained in X-gal solution at 37°C overnight as described67. Kidneys were fixed in 10% formalin at 4°C overnight, processed, embedded in paraffin, and sectioned. Sections were rehydrated and either mounted in Permount or labelled with DBA or LTL lectin and mounted in DAPI Fluoromount-G. Sections were imaged using brightfield and confocal microscopy as described68.
Western blot
Kidney pieces (1/3 of kidney) were homogenized in Passive Lysis Buffer (Promega) containing protease inhibitors (Thermofisher). BCA assays (Pierce, ThermoFisher) were performed according to manufacturer’s instructions. Western blots were performed as described66, using primary and secondary antibodies (Tables S1 and S2).
Scanning electron microscopy
SEM on renal tissue was performed as described69. Anesthetized mice were perfused with cold 2% glutaraldehyde in 0.1M cacodylate buffer, pH range 6.8-7.4. Kidney cortices were dissected into small pieces and fixed in 2% glutaraldehyde in 0.1M cacodylate buffer at 4°C overnight, then washed in 0.1 M Na-cacodylate, pH 7.4. Samples were post-fixed with 1% OsO4 in 0.1 M Na-cacodylate buffer for 30 minutes, then dehydrated in an ethanol series, followed by hexamethyldisilazane (HMDS; Electron Microscopy Sciences). Samples were mounted onto metal stubs and sputter coated with gold. Samples were viewed and imaged using a Hitachi S-2700 Scanning Electron Microscope equipped with a Quartz PCI digital capture.
qPCR
RNA was extracted using Trizol (Life Technologies), then reverse transcribed into cDNA using Quanta Biosciences qScript cDNA mix (VWR International). qPCR was performed in duplicate using Quanta Biosciences Perfecta qPCR Supermix (VWR International) in a BioRad CFX Connect Real-Time PCR Detection System with the following primers (Table S3)24, 70.
ADPKD and normal human kidney sections
ADPKD (K386, K408, K423) and normal human kidney (NHK; K357, K402, K419) sections were obtained from the PKD Biomarkers, Biomaterials, and Cellular Models Core in the Kansas PKD Center. The protocol for the use of discarded human tissues complied with federal regulations and was approved by the Institutional Review Board at KUMC66.
Statistics
GraphPad Prism 8 software was used to perform statistical analyses. For analyses of more than two groups, ANOVA or Kruskal-Wallis tests were used to determine statistical significance (p<0.05) of data with or without a normal distribution, respectively. For analysis of two groups, an unpaired t-test was used.
Competing interests
The authors declare no conflict of interest.
Contributions
WW, LMS, MAK, HHW, TSP, BAA, DTJ, RD, JTC, AC, MTP, MS, DPW, and PVT performed experiments. WW, LMS, MAK, HHW, TSP, BAA, DTJ, RD, JTC, AC, MTP, MS, CS, DPW, JPC and PVT analyzed and interpreted data. WW, LMS, BAA, and PVT designed research. WW, LMS, and PVT wrote the manuscript. All authors revised and approved the final manuscript.
Acknowledgements
We thank members of the KUMC Department of Anatomy and Cell Biology and the Jared Grantham Kidney Institute for helpful discussions. We thank Jing Huang of the KUMC Histology Core, which is supported by NIH U54HD090216 and NIH P30GM122731. We also thank Pat St. John and Larysa Stroganova of the KUMC Electron Microscopy Research Laboratory, which is supported by NIH P20GM104936. This work was also supported by a K-INBRE Summer Student Award to JTC (K-INBRE P20GM103418), the PKD Biomaterials and Biomarkers Repository Core in the Kansas PKD Research and Translational Core Center (NIH P30DK106912 to JPC), R01DK108433 to MS, and R01DK103033 to PVT.