Novel Polycystin-2 regulation of ezrin in renal epithelia reveals insights into ADPKD cystogenesis
Good and bad consequences of proximal tubular dysfunction: lessons from human genetics
The Walz group currently focuses on two projects:
1) polycystic kidney disease (ciliopathies), and
2) acute kidney injury.
Polycystic kidney disease - Ciliopathies
Autosomal dominant polycystic kidney (ADPKD) represents a common human genetic disease. With an estimated frequency of 1:500 to 1:2000, approximately 80.000 patients are affected in Germany with half of them progressing to end-stage renal disease due to massive cyst growth and renal failure; occasionally, the volume of both kidneys with exceed 10 liters. ADPKD is caused predominantly by mutations in PKD1 (85%) and PKD2 (15%). In addition to the autosomal dominant version, there are many other autosomal recessive cystic kidney diseases that are categorized based on their extra-renal manifestations in different syndromes (e.g. Nephronophthisis (NPHP), Bardet-Biedl syndrome (BBS), etc.).
Most gene products that are mutated in human cystic kidney disease localize to the cilium or affect the function of this microtubular organelle that is attached to most body cells. Although a dysfunction of the cilium has been postulated as the underlying cause of cystic kidney disease (hence the term “ciliopathy”), precise molecular mechanisms of the more than 100 gene products identified so far are still missing.
The Xenopus embryo as a model system to study cystic kidney disease
Figure: The Xenopus epidermis as a model system for multi-ciliated cells. Scanning electron microscopy (SEM) of the Xenopus epidermis at state 32, approximately 40 hours post fertilization (hfp). Higher magnification demonstrates isolated multi-ciliated cells that express more than 100 motile cilia.
To understand the function of cilia-associated proteins, we use the epidermis of Xenopus embryos as a model system. Within the first 24-48 hours after fertilization, the epidermis of the Xenopus embryos develops so called “multi-ciliated cells” (MCCs) that are surrounded by small secretory as well as ion and mucus-secreting cells. MCCs contain more than 100 motile cilia that generate a directional fluid flow from the anterior to the posterior end of the embryo.
Gene expression in the Xenopus embryo can be manipulated by (micro-) injecting antisense morpholino oligonucleotides (MOs) at very early developmental stages. Targeting cilia-associated molecules by MOs, the multi-ciliated cells of the Xenopus epidermis therefore represent an ideal model system to study the function of “ciliopathy” genes (e.g. NPHP, BBS, etc.).
Manipulation of cilia-associated molecules affects the actin cytoskeleton
An intact apical actin cytoskeleton is essential for ciliogenesis. The apical actin cytoskeleton of the multi-ciliated cells of the Xenopus epidermis consists of two layers, the apical and sub-apical actin layer. We found that NPHP family members are essential for formation and maintenance of the sub-apical actin layer.
Figure: NPHP family members are required for an intact actin cytoskeleton in multi-ciliated cells. Depletion of nephrocystin-4 with morpholino oligonucleotides (nphp4 MO) reduces the density ofthe apical actin cytoskeleton. The sub-apical actin layer is particularly affected by the reduction of nphp4 expression.
During early stages of development, the actin cytoskeleton in multi-ciliated cells is highly dynamic, and undergoes rapid assembly and disassembly. After basal body docking to the apical plasma membrane and subsequent cilia formation, the apical actin cytoskeleton becomes progressively more static.
Figure: Dynamics of the apical actin cytoskeleton. The apical actin cytoskeleton, visualized with RFP-Utrophin, is highly dynamic and undergoes rapid changes during early (stage 21) development, but becomes increasingly static at later stages (stage 28) of development (Quantification: white depicts highly dynamic filaments, red depicts static filaments).
We are currently trying to understand how gene products involved in cystic kidney disease (for example NPHP family members such as NPHP4) control the organization of the actin cytoskeleton. One hypothesis that we are currently pursuing is the physical interaction of NPHP family members with actin modifying protein.
Figure: Localization of Cordon-bleu (Cobl) in multi-ciliated cells of the Xenopus epidermis. Cobl (GFP, green) localizes the ends of actin filaments (RFP-Utrophin, red) in the apical actin cytoskeleton.
Acute Kidney Injury (AKI)
Acute kidney injury (AKI) remains a frequent complication of severe disease. Despite progress in supportive care, the overall mortality and long-term consequences of this complication have not changed significantly. Although complex genetic models and cell lineage tracking have identified subsets of cells that play a central role during the recovery of renal function, the processes that occur immediately after an injury remain unknown because it is virtually impossible to assess these events in mammalian models. Thus, the identification of strategies to prevent AKI and strategies to accelerate recovery has remained an unmet challenge.
We established an approach to elucidate early adaptive changes after kidney damage at the molecular level. Using a workflow, encompassing laser-induced zebrafish kidney injury, high-resolution video-microscopy, single-cell isolation and transcriptional profiling, genetic manipulation of repair genes, and kidney-specific gene knockout in mouse models, we can now analyze early up-stream signaling events and delineate the down-stream metabolic switches occurring in response to kidney injury. Extracting key regulators that orchestrate adaptive responses, we are translating experimental observations into clinical applications. Our strategy aims to change the grim prognosis of AKI by providing novel molecular insight into repair mechanism, while delivering new targets for therapeutic interventions.
Identification of gene products participating in the repair of injured kidney tubules
Figure: Zebrafish pronephros injury model. (A) Starting at 1 day post fertilization, the proximal segments of the pronephros are shaped by a posterior-to-anterior collective cell migration. The fluorescent image was taken from a time-lapse movie before the fusion of the two glomeruli. (B) Laser-induced cell ablation depicted in the lateral view of the pronephros. (C) The injury is repaired within few hours by a migratory response. Note that pronephric duct cells on the anterior side reverse their direction to close the gap. Vector analysis also reveals that cells increase their migration speed up to 10-fold.
It is unknown how tubular epithelial cells recognize tissue damage, and how this information is communicated to nearby cells to orchestrate directed cell migration, proliferation and differentiation, aiming to repair the endured injury. To address this question, we developed a zebrafish AKI model involving laser-mediated ablation of pronephric duct cells followed by high-resolution video microscopy. Since zebrafish embryos acquire the ability to repair a pronephros injury 30 hours post fertilization (hpf), we compared transcriptional profiles of zebrafish embryos before and after 30 hpf, and identified several crucial components of the repair process.
To validate the zebrafish results in a mammalian model of AKI, we generate doxycycline-inducible, kidney-specific knockout mice, using floxed alleles of candidate genes. Previously, we could confirm that deletion of either Cxcl12 or Myc worsened renal failure after ischemia/reperfusion (IR) injury, confirming the results obtained in the zebrafish AKI model.
Figure: Mammalian AKI model. (A) Floxed alleles of either Cxcl12 or Myc were used to generate a doxycycline-inducible, kidney (nephron)-specific knockout of these genes. Doxycycline permits activation of the TetO promoter by rtTA, expressed under the control of Pax8. TetO drives the expression of the CRE recombinase (indicated as scissors), which excises the floxed gene segments. (B) Mice with the Pax8rt *TetOCre genotype are treated for 3 weeks with doxycycline, followed by a washout period of one week to avoid doxycycline-dependent effects. Analysis was performed 12 hours after an ischemia/reperfusion injury to assess early changes caused by the nephron-specific removal of candidate genes.
Single-cell RNA sequencing (RNAseq) of pronephros tubular epithelial cells
To identify the transcriptional changes that occur in response to an injury, we isolate cells directly adjacent to the laser-induced injury.
Figure: Single-cell RNA sequencing of zebrafish pronephros cells. (A) To determine the transcriptional profile in tubular epithelial cells immediately involved in the repair response (i.e. adjacent to an injury), the pronephric tubules of cdh17:gfp transgenic zebrafish were injured, and compared to uninjured control embryos. (B) GFP-positive tubular epithelial cells were sorted by FACS to eliminate contaminating cells. Repairing cells were subsequently identified by up-regulation of marker genes (e.g. EpCAM).
Although cell ablation by a laser pulse is quite dissimilar to a mammalian IR injury, the cascade of signaling responses that are triggered in surviving cells to initiate the repair process need to entail similar programs: injured cells release signaling molecules to activate programs that allow cells to switch from homeostatic functions to repair programs, including the ability to migrate to the site of the injury. To capture the transcriptional changes occurring in tubular epithelial cells executing the repair, we established single-cell RNA sequencing (scRNAseq) of pronephric duct cells. Transgenic chd17:gfp zebrafish embryos are digested after unilateral pronephros injuries and individual GFP-positive cells rapidly sorted into 384-well plates. RNA from each cell was then extracted, and amplified with bar-coded primers to determine their transcriptome (in collaboration with D. Grün and Sagar at the Max Planck Institute, Freiburg).
1) Immediate (early) responses to injury. The proliferative response, occurring 24-48 hours after injury, has been extensively studied. However, the immediate mechanisms initiating the repair process remain poorly understood, mainly because this process cannot be easily studied in mammalian AKI. Thus, alternative models are needed to identify molecules that orchestrate early adaptive responses after injury.
2) Overall capacity of tubular regeneration. Despite the ability to regenerate, the limits of the regenerative capacity of the kidney remain poorly defined. Although unproven, it is believed the tubular basement membrane (TBM) as guidance cue for regenerating cells needs to remain intact to support the recovery from injury. A better understanding of the role of the TBM may not only help to define the “point of no return” in renal regeneration, but identify new approaches to support the integrity of the TBM.
3) Metabolic adaptation to ischemia. Cells suddenly deprived of oxygen need to switch from aerobic to anaerobic energy production. Although the biochemical pathways are well understood, how this switch is accomplished in vivo is unknown. Manipulating this switch either if AKI is anticipated or after an IR injury has occurred may facilitate recovery from injury.