Pinpointing Podocin Trafficking Defects in Kidney Organoids

Over 80 human genes have been associated with glomerular dysfunction,¹ and with large genetic sequencing studies, this list continues to expand. Genes that are essential for podocyte function were among the first discoveries and include NPHS1 encoding the slit diaphragm protein nephrin² and its interacting partner podocin, encoded by NPHS2.³ Defects in both genes cause congenital or early-onset nephrotic syndrome, which is unresponsive to corticosteroid treatment and referred to as steroid-resistant nephrotic syndrome (SRNS). To understand more about the mechanisms of podocyte injury in SRNS and to examine the functional effects of specific variants in podocyte genes, it is necessary to have appropriate experimental models.

Human kidney organoids have emerged as an attractive system for investigating kidney development and disease. They are generated by driving the differentiation of pluripotent cells into kidney progenitor populations that self-organize to form early nephron structures. The differentiation protocols for kidney organoids were first described in 2014,^4-6 and since then, organoids have been extensively characterized with single-cell transcriptomics and proteomic studies and they have been used to model genetic kidney disease. Where there is a relevant phenotype, such as cyst formation with variants in the genes that cause polycystic kidney disease, organoids can be used for investigating disease mechanisms and for screening potential therapies.⁷ Although kidney organoids contain clusters of cells expressing podocyte markers including NPHS1 and NPHS2⁸ and organoid podocytes form cell-cell junctions that may represent early slit diaphragm structures, it has been unclear whether the organoid system could be used to further understand the functional consequences: missense, nonsense, and splicing variants in podocyte genes.

To address this question, Dorison et al.⁹ used kidney organoids to model missense variants in NPHS2. The organoids were differentiated from gene-edited and patient-derived induced pluripotent stem cells (iPSC) to create an allelic series for investigating the functional consequences of variants in NPHS2 (Figure 1). Overall, the study highlighted several unappreciated effects of missense variants on podocin protein levels, interactions with nephrin, and subcellular podocin localization. Trafficking defects have been previously reported and linked to abnormalities in protein glycosylation, but these earlier studies used gene overexpression approaches in immortalized podocytes or nonpodocyte lines.¹⁰ This new study investigated variants of the endogenous NPHS2 gene in the organoid system and is therefore likely to be a greater reflection of the true trafficking itinerary of podocin in vivo.

Figure 1.

A workflow for modeling genetic variants in kidney organoids. Pluripotent cells with pathogenic variants can be generated by genome editing or from patient-derived somatic cells. These stem cells are then differentiated into kidney organoids to model the genetic defects. Validation is possible with patient samples and multiomics, and imaging permits deep phenotyping of the genetic variants. This illustration is created with BioRender.com.

The study focused on five NPHS2 pathogenic variants, namely p.G92C, p.P118L, p.R138Q, p.R168H, and p.R291W. These variants were selected due to their frequency and severity and their previously reported effects on podocin trafficking and localization. A key finding was a universal decrease in podocin protein levels in all of the variant lines examined (Figure 2). This could be due to less efficient transcription or translation or due to increased podocin degradation. The reduction in podocin levels may ultimately lead to podocyte apoptosis, which was also observed across all gene-edited and patient-derived iPSC variant lines. The podocyte apoptosis was independent of endoplasmic reticulum (ER) stress, and the occurrence as a common feature across all NPHS2 variants is suggestive of a common downstream mechanism, which requires further investigation.

Figure 2.

Missense variants in NPHS2 affect podocin and nephrin localization and podocyte apoptosis. Five pathogenic missense variants in NPHS2 caused a decrease in podocin abundance and ER stress–independent podocyte apoptosis. Podocin and nephrin mislocalization had variant-specific patterns. Both the R168H and R291W variants caused the retention of podocin in the Golgi apparatus.

Other findings were specific to individual NPHS2 variants. There were unique patterns of subcellular podocin localization and trafficking, and one interesting observation was the variant-specific accumulation of podocin in the Golgi apparatus. As a component of the podocyte slit diaphragm, podocin is translated and trafficked to the plasma membrane. Previous studies have shown that missense NPHS2 variants result in unique patterns of podocin trafficking and consistent with these previous reports. Dorison et al. found that the podocin proteins encoded by the P118L and R138Q variants were sequestered in the ER. However, with the R168H and R291W variants, podocin accumulated in the Golgi. The authors proposed that the change in podocin conformation with both of these variants may affect association with lipids in organelle membranes, and in turn, this could influence podocin transit through the Golgi. While the mechanisms of Golgi retention remain unclear, the organoid system could be used for further exploration of the post-translational modifications of podocin that influence protein trafficking.

The study further reports differential effects of the NPHS2 variants on nephrin localization in organoid podocytes. This key component of the slit diaphragm plays a central role in cell-cell adhesion, cell survival, and regulation of actin cytoskeleton. Nephrin localization is dependent on its cytosolic interaction with podocin, and the P118L, R168H, and R291W variants all showed a degree of nephrin sequestration in the ER or Golgi. This is likely to reduce nephrin transport to the plasma membrane and may also influence cell survival.

In addition to the targeted investigation of podocin trafficking, this study also included a global transcriptomic analysis of all variant lines at the day 7+14 time point through the differentiation protocol. Overall, there were surprisingly few changes in transcription between the variants. Possible explanations for the lack of differentially expressed genes include the selected time point, which may have been before or beyond peak transcriptional flux, or as the authors suggest, the predominant effect of the NPHS2 variants is at the protein level and with post-translational modifications.

In summary, this study is a further example of how gene-edited or patient-derived kidney organoids represent an appropriate experimental model for investigating genetic kidney disease. This system could be used to further explore underlying mechanisms of altered podocin trafficking and to screen and test potential gene or small-molecule therapies to rescue podocin function.

Disclosures

R. Lennon reports Consultancy: Travere Therapeutics, Caliditas Therapeutics, Purespring Therapeutics, River 3 Renal; Advisory or Leadership Role: Kidney Research UK grants panel; Scientific Advisory Research Network for the Alport Syndrome Foundation; and Other Interests or Relationships: Funding from The Wellcome Trust and Kidney Research UK, Trustee for Alport UK, and Trustee for Kidneys for Life.

Funding

R. Lennon was supported by a Wellcome Trust Senior Fellowship award (202860/Z/16/Z), and P. Tian was supported by a Kidney Research UK grant (RP52/2014).

Author Contributions

R. Lennon and P. Tian conceptualized the study, wrote the original draft, and reviewed and edited the manuscript.