Modeling congenital kidney diseases in Xenopus laevis Open Access

Congenital anomalies of the kidney and urinary tract (CAKUT; see Box 1 for a glossary of terms) are a leading cause of pediatric kidney failure, accounting for 40-50% of pediatric chronic kidney disease (CKD; Box 1) worldwide (Vivante et al., 2014). CAKUT encompasses a wide range of structural malformations resulting from morphogenetic defects, including Wilms tumor and renal hypodysplasia (Box 1; Fig. 1) (Yosypiv, 2012). With an average 3- to 5-year wait time for a deceased donor kidney transplant [United Network for Organ Sharing (https://www.kidney.org/atoz/content/transplant-waitlist), accessed December 10, 2018], the need to find alternative treatments that preserve renal function is essential. Monogenic disease with strong genetic causality only accounts for 12% of CAKUT cases (Vivante and Hildebrandt, 2016), and polygenic causes are speculated to occur but are largely unknown (Sanna-Cherchi et al., 2018). Next-generation sequencing has helped to uncover novel causative genes of CAKUT, but a high-throughput strategy to test their function in the kidney is needed. To understand how CAKUT arises, it is crucial to understand how the kidneys and urinary tract develop to uncover the genetic mechanisms that coordinate these events. Mice and zebrafish have been the predominant models used in kidney research, while recent advances have made kidney organoids (Box 1) useful for nephrotoxicity screening as well as for modeling kidney diseases (Hurtado del Pozo et al., 2018). However, Xenopus laevis frogs (Box 1; hereafter referred to simply as Xenopus) possess many qualities that make them an effective in vivo model to study congenital kidney diseases.

Xenopus share a relatively close evolutionary history with mammals because frogs are tetrapods. Thus, this model has the advantage of rapid development, like zebrafish, but evolutionarily it lies closer to mammals. Additionally, Xenopus and human genomes have long stretches of gene collinearity, and 79% of identified human disease genes have a verified ortholog in Xenopus (Hellsten et al., 2010). The embryonic kidney of Xenopus has many characteristic features of a mature mammalian kidney (Fig. 2) (Zhou and Vize, 2004; Raciti et al., 2008). Thus, many genes and processes necessary for Xenopus kidney development are also important in mammalian kidney development. Furthermore, a limited number of kidney-disease-causing genes have been analyzed in Xenopus or are expressed in its embryonic kidney, introducing potential paths for future kidney research (Fig. 1 and Table 1). Although few in number, the Xenopus genes studied have been shown to function in kidney development, as their mammalian orthologs do in mammalian kidneys, and their disrupted expression results in similar nephron (Box 1) phenotypes to those seen in humans. Additionally, Xenopus possesses unique qualities that other in vivo model systems lack.

With a simple hormone injection, Xenopus produces large clutch sizes with hundreds of embryos that develop externally. Their kidneys can be easily visualized and imaged through their transparent epidermis, and they develop a fully functional kidney in ∼56 hours post-fertilization (hpf) (Vize et al., 2003). Additionally, cleavage-stage Xenopus embryos have been fate-mapped (Box 1), which allows for the tracking of specific blastomeres (Box 1) as they develop from an early-stage embryo into a differentiated body plan (Dale and Slack, 1987; Moody, 1987). This facilitates unilateral tissue-targeted injections that are specific to organs such as the kidney (DeLay et al., 2016) and permits using the uninjected side as an internal control. In addition, researchers have developed assays to study kidney function in Xenopus (Zhou and Vize, 2004; Tran et al., 2010).

Xenopus oocytes (Box 1) and early embryos can easily be injected with DNA, mRNA, protein and/or morpholinos (Box 1) to overexpress or knock down proteins. A common assay in Xenopus involves disrupting genes/proteins in both kidneys by injecting morpholinos, or CRISPR sgRNAs and the Cas9 protein, in both ventral cells of four-cell embryos. If the gene/protein is important for kidney formation, its loss will result in edema, characterized by swelling in the chest cavity due to fluid retention. This technique allows for disruption of the kidney while avoiding the heart and liver, two other common causes of edema, generating tissue-targeted knockdown or knockout embryos. Upon edema formation, researchers can then use an excretion assay (Box 1) that allows visualization of the passage of fluorescent molecules through the kidney and out through the cloaca, which can show whether the kidneys are still functioning (Zhou and Vize, 2004). Although not directly comparable to the Cre-loxP systems used in mouse studies, this edema assay, which is unique to Xenopus, allows for some evaluation of tissue specificity. Knockdown of Pkd2, a protein involved in autosomal-dominant polycystic kidney disease (ADPKD; Box 1), in Xenopus embryonic kidneys results in edema (Tran et al., 2010). Similar edema phenotypes occur due to the loss of other proteins implicated in CAKUT, such as Pax2 and Pax8 (Buisson et al., 2015). These established techniques demonstrate that Xenopus possesses distinctive qualities suitable for disease modeling, and new technologies continue to improve the experimental opportunities in this model system.

Genetic screens of the allotetraploid (Box 1) Xenopus are becoming more feasible with the development of the CRISPR/Cas9 system (Aslan et al., 2017), and the ability to target CRISPR-mediated genetic manipulation to the kidney is unique to Xenopus (DeLay et al., 2018). These advantages make Xenopus a valuable model for studying the kidney, can aid in our understanding of the mechanisms through which the kidney develops, and can help identify new genes important for renal function. This Review will highlight the Xenopus genes and pathways that are associated with human kidney disease, demonstrate the similarity in phenotypes upon gene or protein disruption, and establish Xenopus as a relevant model for CAKUT.

The kidney is an excretory organ that filters blood. It functions by reabsorbing nutrients and excreting waste and excess fluid in the form of urine. In humans, nephrogenesis starts at around 5 weeks of gestation and completes between weeks 34 and 36 (Black et al., 2012). Environmental factors, such as vitamin A deficiency or exposure to teratogenic substances such as angiotensin-converting enzyme (ACE) inhibitors, among others, can affect proper kidney development (Rosselot et al., 2010; Al-Maawali et al., 2012). Retinoic acid is derived from vitamin A and retinoic acid signaling (Box 1) is important for embryonic kidney patterning (Rosselot et al., 2010). Additionally, taking teratogenic ACE inhibitors during pregnancy can lead to CAKUT, as the kidney also plays an essential role in controlling blood pressure (Al-Maawali et al., 2012). The kidney is a unique organ in that it develops from successive forms that replace the previous structure (Davidson, 2008). These forms are the pronephros, mesonephros and metanephros (Box 1). The organizational complexity increases as each sequential form of the kidney is superseded by the next (Davidson, 2008). In mammals, both the pro- and mesonephros are replaced by the metanephros, which persists to function as the adult kidney (Little and McMahon, 2012). In contrast, the onset of metamorphosis in Xenopus marks the replacement of the pronephros with the mesonephros, which is the adult kidney of amphibians. It is important to note that, although the pronephros is non-functional in mammals, it is required for the subsequent formation of the mesonephros and metanephros.

Although the arrangement of kidneys differ, the structural and functional unit of the kidney, the nephron, remains the same. An adult human kidney contains approximately 1 million nephrons (Bertram et al., 2011), whereas an adult mouse kidney has over 10,000 (Short et al., 2014). Mammalian nephrons develop asynchronously, making them challenging to study (Sorokin et al., 1992). In contrast, Xenopus embryos contain only one functional nephron on either side of their body, which serves as the embryonic kidney. Therefore, the fully functional Xenopus pronephric nephron serves as a simplified model for mammalian meso- and metanephric nephron development (Fig. 2).

The pronephros forms from the intermediate mesoderm (Box 1) in both mammals and amphibians. In Xenopus, the nephric primordia (Box 1) is induced by signals from the surrounding tissues and forms during neurula (Box 1) stages 12-15 (∼16-20 hpf), and later generates the tubules of the pronephros (Fox, 1963). Pronephros specification in Xenopus occurs through the interactions between the transcription factors Osr1, Osr2, Pax8, Lhx1 and Hnf1B, among others (Vize et al., 2003; Tena et al., 2007; Drews et al., 2011; Boualia et al., 2013). Morphogenesis of the pronephros begins at the early-tailbud stage 21 (∼23 hpf) (http://www.xenbase.org/anatomy/alldev.do). In mice, the Osr1 gene is essential for the formation of renal structures and can be found in the intermediate mesoderm, while Osr2 is found later on in the mesonephros and is not essential (Tena et al., 2007). Consistent with Xenopus (Buisson et al., 2015; DeLay et al., 2018), mice that lack Lhx1 or both Pax2 and Pax8 have severe kidney defects (Pax2 and Pax8 seem to have redundant functions in mice) (Shawlot and Behringer, 1995; Bouchard et al., 2002). Finally, Hnf1B is expressed in the ureteric bud as well as in comma- and S-shaped bodies (Box 1), and mutations in this gene lead to kidney cyst formation in mice (Gong et al., 2009). In addition, Wnt (wingless/integrated; Box 1), FGF (fibroblast growth factor), BMP (bone morphogenetic protein) and GDNF (glial-cell-line-derived neurotrophic factor) signaling pathways are involved in both Xenopus and mammalian kidney development (Dressler, 2006; Costantini and Kopan, 2010).

Analogous to the mammalian nephron, the amphibian pronephros is segmented along its proximal-distal axis (Møbjerg et al., 2000). Mammals have a glomerulus to filter blood, proximal tubules, a loop of Henle and distal tubules for reabsorbing water and various nutrients, and a collecting duct that reabsorbs water and carries urine from the kidneys to the bladder (Little and McMahon, 2012). Similarly, Xenopus embryos have a glomus for filtering blood, proximal and distal tubules for reabsorbing nutrients and a connecting tubule, which opens out to the cloaca (Fig. 2). In addition, at the tips of the proximal tubules, the amphibian also has multi-ciliated cells called nephrostomes (Box 1), which cause fluid influx from the coelomic cavity (Box 1) into the proximal tubules (Wessely and Tran, 2011).

Each segment of the nephron has a specific function, which is reflected by distinct cell morphologies (Møbjerg et al., 2000) and gene expression signatures (Zhou and Vize, 2004; Raciti et al., 2008). Strikingly similar to mammals, the proximal tubules in Xenopus are responsible for reabsorbing ions, water, glucose and amino acids (Eid et al., 2002; Zhou and Vize, 2004, 2005; Christensen et al., 2008; Raciti et al., 2008). The most distal part of the Xenopus proximal tubule is analogous to the mammalian proximal straight tubule (Raciti et al., 2008). Because Xenopus is an aquatic freshwater frog, it is unlikely that the amphibian pronephros requires a true loop of Henle for concentrating urine. Within the literature, the Xenopus intermediate tubules have been defined using markers that are co-expressed in the proximal and distal tubules of the mouse kidney (Reggiani et al., 2007; Raciti et al., 2008). Thus, assessment of a marker that is exclusive to this region is necessary to confirm or refute the existence of this structure. The Xenopus distal tubule is analogous to the thick ascending limb of the loop of Henle and the distal convoluted tubule in mammals (Raciti et al., 2008); it functions to transport ammonium and reabsorb magnesium ions, and is important for urine acidification (Vize, 2003; Zhou and Vize, 2004, 2005; Raciti et al., 2008). The pronephric connecting tubule of Xenopus connects the distal tubules to the cloaca to excrete urine and is only analogous to mammals in the region neighboring the distal tubule (Raciti et al., 2008). The pronephros does not have a collecting duct to further concentrate urine or connect multiple nephrons to a secondary structure, as is necessary for the mammalian metanephros.

Recent research has shown that human and mouse nephron protein-expression signatures in the renal vesicles (Box 1) and S-shaped bodies are very similar (Lindström et al., 2018). Not surprisingly, Xenopus shares similar gene expression signatures to humans during early pronephric development in the kidney (Fig. 3). Although Fig. 3 only illustrates four gene expression signatures, it is important to note that, of the 29 expression signatures identified in the human embryonic kidney (Lindström et al., 2018), 18 can also be found in the Xenopus kidney, including mafb, slc3a1, pou3f3 and foxc2 (xenbase.org). Additionally, a comparison of gross expression signatures in the kidney demonstrated similarities between mouse and Xenopus in both overlapping transcription factors such as Pax8 and additional ones such as Foxc1 and Sall1 (Kaminski et al., 2016). Furthermore, some of the proteins analyzed in humans have not been studied in Xenopus, potentially leading to an underestimate of the similarities. This overall conservation on the molecular, genetic, structural and functional levels supports the use of the Xenopus pronephros as an appropriate model for the study of CAKUT.

Xenopus has been historically used to elucidate molecular mechanisms and signaling cascades involved in early developmental processes (Beck and Slack, 2001). More recently, Xenopus has played an important role in identifying the function of genes that are involved in CAKUT. The Wnt pathway has been heavily studied in Xenopus due to this pathway's significant contribution to various developmental processes. The canonical Wnt/β-catenin pathway plays a role in kidney development in both mice (Carroll et al., 2005; Iglesias et al., 2007) and Xenopus (Saulnier et al., 2002; Lyons et al., 2009). WNT4 is a well-known ligand that activates canonical Wnt signaling and is important for induction and mesenchyme-to-epithelial transition of the metanephric mesenchyme (Box 1) (Murugan et al., 2012). In humans, WNT4 loss-of-function mutations have been associated with renal hypodysplasia, a condition characterized by dysplastic kidneys with a reduced number of nephrons (Vivante et al., 2013). Studies in Xenopus demonstrate that Wnt4 protein controls the medio-lateral patterning of the pronephros (Naylor and Jones, 2009). Additionally, knockdown of Wnt4 protein in Xenopus results in complete loss of kidney tubules (Saulnier et al., 2002).

In addition to Wnt ligands, other proteins have been found to regulate Wnt signaling. In humans, Wilms tumor 1 (WT1) is a known regulator of canonical Wnt signaling. Mutations that affect WT1 result in nephroblastoma, more commonly known as Wilms tumor (Kim et al., 2010a,b). Knockdown of Wt1 protein expression in Xenopus embryos was shown to reduce Wnt4 expression in the prospective pronephros (Murugan et al., 2012). Mutations in the RNA-binding protein BICC1, a negative Wnt regulator, result in a cystic kidney phenotype in mice and renal anomalies in patients (Kraus et al., 2012). Prior studies in Xenopus indicate that Bicc1 inhibits the microRNA miR-17, preventing miR-17 from destabilizing the pkd2 transcript (Tran et al., 2010). This was demonstrated by a rescue of the mutant kidney phenotype when both miR-17 and Bicc1 were simultaneously knocked down in tadpoles (Tran et al., 2010). Interestingly, loss of the Pkd1 gene product, polycystin-1, downregulates BICC1 expression in mouse kidneys and mouse cell lines (Lian et al., 2014). The connection between BICC1 downregulation and polycystin-1 loss in mice and the stabilization of pkd2 mRNA by Bicc1 in Xenopus suggests that disruption of BICC1 may induce cystic phenotypes through polycystin signaling (Box 1). Similarly, targeted deletion of Hnf1b, a key transcription factor involved in kidney development, decreases Bicc1 and Pkd2 mRNA expression in the mouse kidney (Verdeguer et al., 2010). HNF1B in the mouse (Gresh et al., 2004; Verdeguer et al., 2010) and BICC1 in mice and Xenopus (Tran et al., 2010) regulate the Pkd2 transcript, and mutations in either gene can lead to kidney cysts in mouse models (Gong et al., 2009; Tran et al., 2010). Thus, it is likely that these genes act through similar pathways. Xenopus would be an ideal model for studying the components of this pathway because molecular strategies to assess them have already been established (Tran et al., 2007, 2010).

Another common congenital kidney disease whose pathway has been elucidated using Xenopus is nephronophthisis (NPHP; Box 1). NPHP is a ciliopathy (Box 1), resulting in either abnormal formation or function of cilia (Box 1). NPHP is a rare birth defect, but it is the most common cause of kidney failure, or end-stage renal disease (ESRD; Box 1), in the first three decades of life (Wolf and Hildebrandt, 2011). Studies in Xenopus linked several genes with NPHP or clarified components of the mechanism that causes the disease. First, targeted kidney knockdown of the Wnt signaling inhibitor Invs (inversin; also known as NPHP2) demonstrated that Invs is important for morphogenetic cell movements during tubule elongation (Lienkamp et al., 2010). This knockdown led to impaired ventral proximal pronephros extension and distal tubule differentiation (Lienkamp et al., 2010). Later, it was discovered that several genes that cause NPHP form a distinct complex that contributes to kidney development. In Xenopus, Anks6 was identified as an NPHP family member that assembles a protein complex of the NPHP-associated proteins Nek8, Invs and Nphp3 to regulate kidney development (Hoff et al., 2013). This study showed that Anks6 localizes to the base of the cilium and that knockdown of its protein expression results in kidney anomalies (Hoff et al., 2013). The study also linked ANKS6 mutations in patients with an NPHP-like clinical syndrome (Hoff et al., 2013). Additionally, knockdown of either Anks6 or Nphp3 protein in Xenopus embryos results in edema, which suggests that they may be important for kidney function (Hoff et al., 2013). Likewise, edema has been shown to occur due to depletion of Invs protein (Lienkamp et al., 2010).

Although Xenopus has helped uncover some of the causative genes for NPHP and their role in kidney development and function, only ∼40% of patients have a mutation in one of the 20 known NPHP-related genes (Stokman et al., 2016; Srivastava et al., 2018). A recent genome-wide association study (GWAS; Box 1) has uncovered epigenetic signatures of CKD (Box 1), including NPHP4 (Wing et al., 2014). Some NPHP patients may suffer from epigenetic alterations of NPHP genes rather than a mutation in the gene itself. Future research using Xenopus may lead to the identification of new NPHP-causing genes and may uncover further epigenetic regulation of known genes as they have been used to study epigenetics.

A notable CAKUT-causing gene encodes the transcription factor LMX1B, mutations in which can lead to nail-patella syndrome (NPS; Box 1), a rare cause of autosomal-dominant ESRD (Edwards et al., 2015). Additionally, mutations in LMX1B can result in glomerulopathies (Box 1). Mouse studies suggest that Lmx1b is necessary for the maintenance of podocytes, which are highly specialized cells that wrap around capillaries of the glomerulus and restrict the passing of macromolecules into the kidney (Miner et al., 2002). In Xenopus, lmx1b is expressed in the glomus, which is analogous to the mammalian glomerulus, and is also known to have a role in podocyte specification (Haldin et al., 2008; White et al., 2010). Morpholino knockdown of Lmx1b protein in Xenopus showed that the glomus was reduced in size and the development of the proximal tubules was limited (Haldin et al., 2008). Taken together, these studies demonstrate the usefulness of Xenopus in uncovering the role of proteins important for kidney development and function. Additionally, they show that Xenopus is capable of recapitulating specific kidney disease phenotypes found in humans.

In human studies, missense mutations are often analyzed via software programs such as PredictSNP (Bendl et al., 2014) and Meta-SNP (Capriotti et al., 2013). These programs predict the functional relevance of an amino acid change and suggest whether they are likely to be pathogenic. One drawback of this method is the surprising data from the 1000 Genomes Project indicating that people with predicted pathogenic variants do not suffer from the expected disease (Xue et al., 2012). Xenopus can be used to verify the pathogenicity and expressivity (Box 1) of potential disease-causing genetic variants in vivo.

Work in mouse models found that CLCNKB is important for concentrating urine (Matsumura et al., 1999). In humans, CLCNKB mutations cause a salt-losing tubulopathy (Box 1) known as Bartter syndrome type III (Simon et al., 1997). The first genotype-phenotype correlation for Bartter syndrome type III was validated by functional analysis of eight missense and two nonsense mutations in Xenopus. By performing voltage-clamp experiments in Xenopus oocytes, it was found that nine of the ten mutations significantly decreased normal conductance (Seys et al., 2017), indicating altered salt homeostasis. This work verified that the reported human genetic mutations were, in fact, Bartter syndrome type III-causing mutations (Seys et al., 2017). A similar study demonstrated that dysregulation of NRIP1-dependent retinoic acid signaling in both Xenopus and mouse disrupted kidney formation (Vivante et al., 2017). In Xenopus, knockdown of Nrip1 protein causes kidney anomalies that can be rescued with wild-type human NRIP1 mRNA but not with the truncated NRIP1 mRNA identified in affected individuals (Vivante et al., 2017). Xenopus can also serve to assess the distinct effects of different mutations in a single gene.

HNF1B mutations have been shown to manifest as distinct renal diseases in humans (Bohn et al., 2003). An attempt to model these differences in Xenopus was successfully executed using different methods in two separate studies. In one study, researchers generated transgenic Xenopus lines expressing two different human HNF1B mutants using a heat-shock-inducible (Box 1) Cre-loxP system (Sauert et al., 2012). One transgenic line expressed an insertion mutation, while the other line expressed a deletion. The deletion led to reduced pronephric development, while the insertion enlarged the pronephros, with both phenotypes primarily affecting the proximal tubules (Sauert et al., 2012). In another study, Bohn and colleagues compared nine different human HNF1B mutations, including indels, missense and nonsense mutations, by injecting mutant mRNA. The mutations led to distinct renal disease phenotypes in Xenopus, as they do in humans (Bohn et al., 2003). In Xenopus, six of the mutants resulted in an enlargement of the pronephric structures, while the other three mutations led to a reduction or loss of the tubules and the anterior part of the duct (Bohn et al., 2003). The range of kidney phenotypes seen in Xenopus recapitulate the kidney phenotypes observed in humans. This potentially allows for a correlation between an observed patient phenotype with a specific mutation in the HNF1B gene. Studies like these are an efficient way to identify pathogenic genetic variants and can represent the variation of expressivity seen in patients.

In Xenopus, many genes that are involved in kidney development have essential functions in other tissues. To bypass embryonic lethality due to the loss of these essential proteins, researchers have developed techniques that are not possible in other organisms. For example, tissue-targeted microinjection has been used to study ADPKD, the most common inherited human renal disease. ADPKD is caused by mutations in either PKD1 or PKD2, with PKD2 mutations causing a less severe form of ADPKD with a later onset than PKD1 mutations (Deltas, 2001). Homozygous mutations or deletions of either Pkd1 or Pkd2 in mouse models are embryonic lethal (Lu et al., 2001; Kim et al., 2009). This limitation can be resolved by tissue-targeted microinjections in Xenopus as described earlier, as both pkd1 (Burtey et al., 2005) and pkd2 (Tran et al., 2010) are expressed in the Xenopus embryonic kidney. Using this technique, Tran and colleagues elucidated some aspects of pkd2 post-transcriptional regulation (Tran et al., 2010). Although a variety of mouse models have since been developed to bypass the embryonic lethality of a complete knockout mouse, there is still a need for ADPKD models that more closely resemble the human disease. Xenopus has a lifespan that permits an extended survival that would add additional information of how ADPKD progresses over time. Additionally, elucidating the mechanism of the disease using Xenopus may be more cost and time effective.

Inducible systems have also been successfully used in Xenopus embryos, which allows for temporal or spatial control of gene expression. Four inducible systems were developed in Xenopus: the GAL4-UAS (Box 1) (Denayer et al., 2006), Tet-On (Box 1) (Rankin et al., 2011), heat-shock inducible (Roose et al., 2009) and the Dex-inducible (Box 1) strategy (Zhuo et al., 2013). In addition, knocking down a specific protein can be titrated, either by using morpholinos at varying doses to reduce lethality (Eisen and Smith, 2008) or by microinjecting less mRNA of an inhibitor of the protein of interest (Baker et al., 1999). Alternatively, the CRISPR/Cas9 system can be utilized in Xenopus to bypass early developmental lethality caused by knocking out essential genes such as lhx1 (DeLay et al., 2018). Human 17q12 deletions, which span two genes important for kidney development, HNF1B and LHX1, have been implicated in CAKUT (Nagamani et al., 2010). Although embryonically lethal in mice, Xenopus embryos survive lhx1 knockout and demonstrate abnormal kidney development (DeLay et al., 2018). CRISPR/Cas9 editing results in mosaic knockout in Xenopus, which may explain why the knockout embryos were able to bypass embryonic lethality (DeLay et al., 2018), making this mosaic expression a silver lining. Furthermore, in Xenopus tropicalis, transplanting the primordial germ cells of CRISPR-mutagenized F₀ embryos into a wild-type host, termed leapfrogging (Box 1), results in F₁ embryos bypassing the embryonic lethality of the loss of the homeobox protein goosecoid (Blitz et al., 2016). Taken together, there are a range of techniques available to bypass embryonic lethality in Xenopus.

Cilia are hair-like cellular protrusions that are found in most tissues in the body. Defects in genes important for ciliogenesis often lead to a vast array of phenotypes, including retinal degeneration, renal disease and congenital fibrocystic diseases of the liver (Waters and Beales, 2011). In addition to studying the ciliated A6 cell line derived from Xenopus and the primary cilia present in vivo in the pronephros, Xenopus embryos have a ciliated epidermis that has been useful for studying cilia function. WDPCP (Fritz) is one of 20 known genes that cause Bardet-Biedl syndrome (BBS; Box 1), all of which are involved in primary cilia function (Suspitsin and Imyanitov, 2016). Primary features of BBS include rod-cone dystrophy, genital abnormalities, learning difficulties, obesity and renal defects (Suspitsin and Imyanitov, 2016). The Xenopus epidermis has been used to study BBS because the cilia are easy to visualize. The authors found that Wdpcp localizes cytoskeletal proteins that control cell movement, called septins (Box 1), to the ciliary base (Kim et al., 2010a,b). Additionally, mutations in human WDPCP are linked to Meckel-Gruber (Box 1) syndrome (Kim et al., 2010a,b). This study demonstrates the versatility that the Xenopus model provides in understanding kidney diseases that extend beyond the pronephros.

Until now, many studies utilized morpholinos to knock down protein expression. Additionally, the allotetraploid genome and long generation time (1 year) of X. laevis have previously made it less attractive for genetic experiments than the diploid X. tropicalis (Beck and Slack, 2001). However, the field is moving towards using CRISPR/Cas9 technology to generate gene knockouts. CRISPR/Cas9 gene editing in Xenopus is a rapid and inexpensive means of studying genes involved in developmental processes (Bhattacharya et al., 2015), introducing the possibility of performing large-scale genetic screens. Recently, the CRISPR/Cas9 system has been used in Xenopus to knock out an important kidney gene, lhx1 (DeLay et al., 2018). Knock-in strategies using CRISPR/Cas9 have also been successfully utilized in X. tropicalis (Shi et al., 2015). Targeted CRISPR/Cas9 knock-in strategies facilitate the fusion of fluorescent tags to endogenous genes (Shi et al., 2015), allowing for more precise visualization of proteins in vivo. Furthermore, leapfrogging, which effectively circumvents a generation of breeding, allows researchers to generate transgenic lines quicker than with traditional breeding strategies (Blitz et al., 2016). These technologies are being favored over new criticisms concerning morpholino-driven phenotypes. Morpholinos have been recently shown to cause immune-response-related gene transcription and mistargeted splicing events. Although these effects cannot be completely ameliorated, researchers can take steps to reduce them, such as lowering the GC content and titrating the morpholino (Gentsch et al., 2018). Previous studies using morpholinos should not be discarded though, as our group's recent work suggests that they match CRISPR/Cas9 phenotypes (DeLay et al., 2019), but optimization to reduce unintended effects as well as proper controls should be utilized.

Distinctly, Xenopus has been used to study acute kidney injury and regeneration. Upon unilateral nephrectomy of the proximal tubules at stage 37/38 (∼53 hpf), Xenopus tadpoles are able to regenerate functional proximal tubules with no drug intervention (Caine and Mclaughlin, 2013; Suzuki et al., 2019). Additionally, Xenopus can even be used to generate kidney organoids that can replace the Xenopus kidney in vivo (Chan et al., 1999), which broadens the usefulness of this model organism for studying kidney disease. Because Xenopus oocytes are equipped with all the necessary machinery for development upon fertilization, they have long been used to study channel proteins. Given the large number of ion channels present in the kidney, this system can be readily exploited to study patient mutations in various channels and their effects on conductance. Additionally, two new transgenic frog lines have been developed in which GFP is expressed in the developing pronephros and mesonephros (Corkins et al., 2018), as well as in the pronephros alone (Ochi et al., 2012). This permits live imaging of the developing kidney, which can also be used to assess kidney disruption upon specific gene knockdown/out without the need for immunostaining or in situ hybridization. This new Xenopus line has the potential to streamline large-scale screens of kidney-disease-causing genes and of environmental factors that affect renal development. Also, bypassing the need to process embryos will expedite drug discovery screening in Xenopus (Schmitt et al., 2014), which may ultimately lead to clinical applications. The development of these new tools will undoubtedly advance our ability to study the genetic mechanisms underlying kidney development and disease.

Although Xenopus has many advantages that make it an attractive model for kidney development and disease, like all other models it possesses its own set of limitations. The most notable limitation may be that Xenopus is used to study the pronephros and has mesonephric nephrons that serve as its adult kidney instead of the metanephros of mammals. Although these differing nephrons maintain the same structure and function, this should be taken into consideration. Studying the pronephros may be different, but it offers a simplified method to study non-integrated individual nephrons. Many cases of CAKUT are associated with structural developmental defects such as vesicoureteral reflux, which cannot be visualized in Xenopus as they do not have a ureteric bud that forms a true collecting-duct system. Therefore, for malformations such as this, researchers should study Xenopus as a way to implicate new proteins in CAKUT and use in concurrence with another model system, such as mouse or kidney organoids, to study the anomalies more directly. Additionally, the Xenopus glomus does not have a Bowman's space, but instead blood filtrate goes directly into the coelomic cavity. Although this should be considered when conducting studies concerning the glomerulus, it may have a separate advantage. In mice, the glomerulus forms at the most proximal end of the S-shaped body, linking it to the development of the rest of the nephron. Xenopus however, can develop a glomus independently from the tubules, which may make understanding how the glomus forms more accessible (Urban et al., 2006). Lastly, Xenopus is an aquatic organism that does not require the resorption of water that is necessary in terrestrial animals.

The studies reviewed here demonstrate that Xenopus is well suited to model congenital kidney diseases. Xenopus shares many genes and proteins with the mammalian kidney that have similar expression patterns and functions. As described here, many of the renal phenotypes observed in humans can be recapitulated in frogs. Next-generation DNA sequencing strategies, such as whole-exome sequencing, has helped identify candidate CAKUT-causing genes. An ideal organism to functionally assay these genes in vivo would develop quickly and produce large numbers of embryos, qualities that Xenopus possesses.

CAKUT accounts for 40-50% of pediatric chronic kidney disease worldwide (Vivante et al., 2014). Although important advances have been made in basic nephrology research, there are still missing pieces in the molecular basis of kidney development. This failure to fully grasp how the kidney develops translates into low cure rates for many of the genetic disorders that cause kidney disease. Being able to uncover novel genes and pathways that are involved in nephrogenesis will give us a better understanding of kidney development as a whole. This may lead to better treatment for patients with CAKUT, and its causative mutations may ultimately be eliminated with the application of gene therapy in human embryos. Xenopus is a very promising organism that can get us closer to accomplishing these goals.

We thank Richard Behringer, Mark Corkins, Bridget DeLay and Vanja Krneta-Stankic for critical reading of the manuscript and thoughtful suggestions.