Zebrafish Renal Pathology: Emerging Models of Acute Kidney Injury

The renal system is vital to maintain homeostasis in the body, where the kidneys contain nephron functional units that remove metabolic waste from the bloodstream, regulate fluids, and balance electrolytes. Severe organ damage from toxins or ischemia that occurs abruptly can cause acute kidney injury (AKI) in which there is a rapid, life-threatening loss of these activities. Humans have a limited but poorly understood ability to regenerate damaged nephrons after AKI. However, researchers studying AKI in vertebrate animal models such as mammals, and more recently the zebrafish, have documented robust regeneration within the nephron blood filter and tubule following injury. Further, zebrafish kidneys contain progenitors that create new nephrons after AKI. Here, we review investigations in zebrafish which have established a series of exciting renal pathology paradigms that complement existing AKI models and can be implemented to discover insights into kidney regeneration and the roles of stem cells.


Introduction
The kidneys perform a suite of physiological roles that are critical for organismal homeostasis. These tasks include the excretion of metabolites, which are filtered from the circulation to produce urine, as well as the conservation of essential nutrients, osmoregulation, and acid/base balance. The nephron functional units within the kidney are epithelial tubules comprised of a blood filter, specialized proximal and distal segments that modify filtrate composition through solute absorption and secretion, and a duct where water and solute levels are fine-tuned (Fig. 1). An interstitial populace of supporting cells and microvasculature surrounds these nephrons. Kidney tissues can incur damage from a variety of sources that include ischemia and toxins in the bloodstream. Depending on their scope, such injuries can lead to the rapid loss of renal functions over several hours or days, causing a condition termed acute kidney injury (AKI) [1]. AKI is defined clinically as a decreased glomerular filtration rate (GFR) and impaired clearance of waste products, which is diagnosed based on elevated serum creatinine and blood urea nitrogen (BUN) levels, and can be associated with reduced or intermittent urine production [2]. AKI is known to affect primarily the proximal epithelial segments within the nephron tubule, though damage to the glomerular filter, interstitial, and vascular compartments can also initiate an AKI episode [2]. Proximal tubule cells are especially sensitive to reductions in blood supply due to their reliance on aerobic respiration [3]. Further, as proximal tubule cells perform the bulk of solute reabsorption, they preferentially acquire toxins present in the filtrate, leading to regional apoptosis [2,4]. When nephron cells are destroyed, the resulting cellular debris and protein casts fill the lumen, leading to downstream tubular obstruction and the interruption of fluid flow [2]. Inflammation due to both innate and adaptive immune responses to injury can lead to additional direct insults and destruction of renal tissues [5]. Over time, these parameters can improve in AKI survivors, which historically supported the notion that AKI is a 'reversible' condition, at least partially due to regenerative processes in the kidney [5,6]. However, AKI can lead directly to endstage renal disease (ESRD), necessitating replacement therapy with dialysis, or an organ transplant to sustain life [7]. Despite the recovery of some patients, AKI has a high morbidity worldwide [7], and AKI-associated mortality in the critically ill patient population ranges between 50 and 80 % [8]. Further, there has been an increasing appreciation that the long-term consequences of AKI include fibrosis and an elevated risk for chronic kidney disease (CKD), in which patients have permanent loss of renal functions [9][10][11]. Reciprocally, there is a significant risk of AKI in patients with CKD [9-11]. These associations emphasize the significance of understanding the mechanisms of AKI, as well as the need to identify biomarkers and develop therapeutics for this diverse clinical syndrome.
Despite many advances in knowledge about the pathophysiology of AKI in recent years, there remain major unresolved questions about the molecular mechanisms at play during these conditions and how to enhance the native regenerative response of human kidney tissues [12, 13 • , 14]. To date, research using animal models has provided many valuable pathological insights into the sequence of events associated with AKI [15]. Mammalian experimental systems using rodents such as the mouse and rat have been shown to mimic various clinical features of human AKI but nevertheless involve some limitations [15]. In recent years, the translational research value of the zebrafish, Danio rerio, has been increasingly appreciated based on their high genetic conservation with humans and the development of sophisticated tools for advanced molecular studies [16,17]. Kidney research using the zebrafish is relevant to the understanding of many aspects of vertebrate renal development and disease based on the formative discovery that fundamental elements of nephron structure and function are conserved (Fig. 1)

Studying AKI Pathology in Embryonic Zebrafish
The zebrafish embryo offers several unique opportunities to study the molecular and cellular biology of AKI events, as well as undertake AKI-related drug discovery. Zebrafish embryos undergo rapid, external development and are optically transparent [24], which facilitates their use for organogenesis and disease modeling research. As previously noted, zebrafish nephrons display a conserved segmental composition with other vertebrates [25]. The zebrafish embryo forms an anatomically simple kidney that has two nephrons, which share a common blood filter and a single exit portal for urine at the cloaca [26 •• ]. This simplicity enables experimental access to the renal tissues, which allows for the ablation of specific compartments of the nephron as well as for the ability to observe in vivo changes in real time [25]. Renal organogenesis commences early, with the embryonic nephrons emerging during the first day of development when mesenchymal renal b Fig. 1  Subsequent research utilizing chemical genetics led to the discovery that histone deacetylase inhibitors (HDACi) were capable of ameliorating gentamicin-induced AKI in the zebrafish embryo and ischemia-reperfusion (IR)-induced AKI in the mouse [46]. HDACi were initially identified in a small molecule screen for compounds capable of expanding the renal progenitor field in the zebrafish embryo [47 •• ]. The researchers found that embryos treated with the HDACi 4-(phenylthio)butanoic acid (PTBA) or trichostatin A had an expansion of renal progenitors that expressed the genes lhx1a, pax2a, and pax8 [47 •• ]. Interestingly, the effect exhibited by these HDACi was similar to previously published work in Xenopus with retinoic acid (RA), showing that RA treatment expanded the kidney field [48]. In subsequent studies, treatment with a modified form of PTBA, m4PTB, was demonstrated to attenuate the detrimental effect of gentamicin exposure to the nephron [49 •• ]. Following gentamicin exposure in zebrafish embryos, treatment with m4PTB led to an increased survival rate [49 •• ]. Further, the cell proliferation of tubular cells doubled, while there was not a noticeable effect of m4PTB on levels of apoptosis, suggesting that increased survival rates were due to increased proliferation of renal cells [49 •• ]. Interestingly, m4PTB treatment was also sufficient to enhance AKI recovery in the adult mouse [49 •• ]. Animals that received the chemical displayed less postischemic fibrosis and showed an improved ability to clear creatinine [49 •• ]. The researchers also observed that m4PTB treatment decreased the number of renal tubular epithelial cells in the G2/M phase [49 •• ]. This led them to suggest that the reduction in cell cycle arrest at this stage may account for the observed reduction in fibrosis, a hypothesis that certainly warrants further investigation. Overall, these findings demonstrate the potential for translational drug discovery studies using the zebrafish kidney as a research model.

Mechanical Obstruction of the Pronephros
Several studies have examined the effects associated with mechanical obstruction of the pronephros tubules in the zebrafish embryo [50 • , 51]. Obstruction can be performed using fine tweezers to physically compress, and thus damage, a section of the pronephros tubules or their terminus at the cloaca [50 • ]. This type of occlusion injury results in the immediate loss of fluid flow within the pronephros and the rapid formation of cystic distension of the tubule within approximately 30 min [50 • ]. Interestingly, further work revealed that this mechanical obstruction was found to be associated with increased cilia beat rate and dramatic gene expression changes [51]. Namely, researchers discovered that foxj1a, which encodes a transcription factor that regulates ciliogenic genes, was upregulated after obstruction in nephron tubule regions where distension had occurred [51]. These findings provided new insights on the response of renal epithelial cells to damage that impacted tubular flow and corroborate increasing evidence that cilia-based mechanosensory signaling is a central component of nephron homeostasis [51].

Using Localized Laser Ablation to Trigger Focal AKI
While the above studies show the utility of nephrotoxins or mechanical injury to induce various types of AKI in the zebrafish embryo, the use of these techniques is irrevocably associated with lethality several days later in development-presumably because the AKI is too catastrophic. Alternative injury methods utilizing laser ablation have been established and have demonstrated that epithelial regeneration can occur robustly in the zebrafish pronephros when tubular damage is localized [52 •• , 53, 54]. As previously discussed, due to their optical transparency, it is possible to inject zebrafish embryos with fluorescently labeled dextran molecules that can be monitored as it circulates and becomes absorbed by proximal tubule cells through endocytosis [26,40]. This technique to fluorescently mark functional proximal tubule cells has been used in our laboratory, in conjunction with a pulsed micropoint laser system, to perform targeted epithelial cell ablation in one nephron [52 •• ]. As the pronephros comprised two nephrons, we adopted a methodology to ablate cells in just nephron, thereby leaving the contralateral side unperturbed to serve as a control as well as retaining a source of renal function for the embryo [52 •• ]. As shown by whole mount in situ hybridization with the proximal tubule marker slc20a1a, application of the laser was demonstrated to result in consistent ablation of the targeted area, and that the length of ablation can be well controlled [52 •• ]. Following proximal tubule laser ablation at 72 hpf, tubule integrity was restored by 168 hpf, and the new cells were capable of dextran uptake [52 •• ]. This injury method revealed for the first time that rapid and robust tubule epithelial regeneration can occur in the zebrafish embryo.
A similar ablation technique has been applied to transgenic lines in which a fluorescent reporter was placed under control of a tissue or segment-specific promoter, circumventing the need for dextran uptake-based labeling [53,54]. The ablation was performed with localized violet laser irradiation using a confocal microscope to assess the effects of nephron cell destruction in ET33d10 and ET-9 transgenic embryonic zebrafish expressing green fluorescent protein (GFP) in the proximal and distal segments of the tubule, respectively [53]. Characterization of this laser photoablation revealed that tubular cell death and proliferation were occurred by 36 h post-injury [53]. Interestingly, through in vivo imaging, collective cellular migration was found to be triggered before this time point such that surviving epithelia moved to fill in the damaged area [53]. Using a series of long (*90 lm) and short (*20 lm) ablation lengths, the researchers uncovered a correlation between the area of damage and collective movement, such that migration was increased when a longer segment of the tubule was destroyed [53]. Additionally, it was shown that the migrating cells likely remained differentiated based on an absence of vimentin staining, the presence of cilia, and intact cell-cell junctional complexes [53]. This study provided the first evidence that collective cell migration is a mechanism of tubule epithelial regeneration and demonstrated that in vivo video imaging can be used in conjunction with this AKI paradigm to capture cellular dynamics in real time.
Further, this photoablation technique has been recently applied to study nephron epithelial regeneration in zebrafish embryos in which researchers compromised formation of the exocyst complex, which mediates the docking and targeting of vesicles in polarized epithelial cells, and is vital for the trafficking of basolateral membrane proteins [54]. Researchers performed antisense knockdown of sec10, which encodes a key exocyst component, followed by photoablation tubule injury. They observed that sec10 knockdown embryos were more sensitive to ablation-induced AKI [54], leading to the proposition that exocyst activity impacts renal cell recovery. This study provides proof of principle that genetic manipulations can be used in conjunction with laser ablation to assess AKI with high resolution in vivo using the zebrafish.
Taken together, the above-mentioned studies illustrate the utility of embryonic zebrafish as a model in which to stimulate AKI through a variety of techniques that allow for detailed characterization of different aspects of tubule repair at the cellular and molecular level. The fact that zebrafish embryos contain only two nephron tubules allows for simple comparisons to be made between damaged and undamaged nephrons, and their optical transparency allows for observation of real time responses to injury. These attributes provide outstanding opportunities for future studies to elucidate new insights into AKI mechanisms through methods that include gene expression analysis [55], forward genetics [56], and the identification of therapeutics with continued chemical genetics [57]. Emergent methods for microscopy in living zebrafish [58] and quantitative renal physiology metric assessments [59,60] further emphasize the ability to implement novel AKI studies using zebrafish embryos and larvae. However, while young zebrafish embryos perform rapid regeneration of the epithelial nephron tubule, they lack a response seen in the adult kidney: the formation of new nephrons in response to damage or de novo nephrogenesis. The use of adult zebrafish to study nephrogenesis in response to AKI, as well as to model nephron epithelial regeneration in the adult environment, is discussed in the next part of this review.

The Regenerative Responses to AKI in the Adult Zebrafish Kidney
The adult zebrafish kidney is a single, flattened organ situated adjacent to the dorsal body wall [61]. To test the proliferative potential of these progenitors, the authors did a serial transplantation experiment and found that in secondary and tertiary transplanted fish, whole kidney marrow cells were still able to engraft and give rise to fluorescently labeled nephrons [67 •• ].
Next, to identify the putative renal progenitor cell populace, the authors examined lhx1a and wt1b, the zebrafish orthologs of Lhx1/Lim1 and Wt1, respectively, which are transcription factors expressed in developing mammalian nephron pre-tubular aggregates [ In sum, these studies identified a renal progenitor population in the adult zebrafish kidney that can be isolated based on lhx1a expression and is capable of producing nephrons following AKI.

Nephron Epithelial Regeneration in the Adult Zebrafish Kidney
Recent descriptive studies by our laboratory have chronicled the pathology of cellular events associated with nephron epithelial regeneration following gentamicin-induced proximal tubule injury in the adult zebrafish kidney, chronicling them with respect to the timing of neonephrogenesis (Fig. 1) [65 •• ]. This analysis was enabled by the adaptation of histological and other molecular labeling methods to distinguish tubule segments in zebrafish tissues, such as through their ultrastructural features and differential binding to lectins [64 •• , 65 •• ]. For example, we established that the zebrafish proximal tubule can be identified based on the presence of a brush border, which can be visualized by hematoxylin and eosin or periodic acid-Schiff staining, along with reactivity to Lotus tetragonolobus lectin (LTL) [65 •• ]. These same aspects are routinely used to distinguish the proximal tubule in mammalian kidneys [69] but simply had not been documented in zebrafish. Using these techniques in our study, we found that proximal tubular epithelial regeneration occurred robustly and rapidly, over just one week following the AKI insult. At 1-day post-injection (dpi) of gentamicin, tubular casts were apparent in distal tubules, and LTL? tubules contained approximately 30 % of dying cells based on staining for nuclear fragmentation with terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) [65 •• ]. Cell death continued at 3 and 5 dpi, with a peak of approximately 40 % of LTL? cells showing TUNEL colabeling at 3 dpi [65 •• ]. Concomitant with this wave of cell death, we discovered a partially overlapping wave of cell proliferation based on PCNA reactivity in LTL? proximal tubules, which began at 2 dpi, peaked to include approximately 60 % of cells at 5 dpi, and declined at 7 and 10 dpi [65 •• ]. Interestingly, we found that mesenchymal cells in regenerating nephrons expressed the Pax2 transcription factor [65 •• ]. This recapitulates a well-known feature of the regenerating mammalian nephron following AKI [70] and has been similarly observed in the zebrafish embryo following gentamicin injury [49 •• ]. Interestingly, while cellular restoration occurred in the first week subsequent to AKI, widespread restoration of proximal tubule function took substantially longer, with only a subset of nephrons showing dextran uptake at 7 dpi but most by 21 dpi [65 •• ]. Taken together, these observations provide a fundamental framework of the pathological events of zebrafish adult AKI and will be useful in future mechanistic studies.

Emerging Models of Cell-Type Specific Inducible Injury in Zebrafish
While traditional AKI studies in zebrafish have utilized gentamicin, inducible models of injury have also been developed and represent an exciting new tool to control damage in a temporal and spatial manner. Researchers have now formulated transgenic lines in which conditional, targeted cell ablation can be achieved by utilizing tissue specific expression of the bacterial Nitroreductase (Ntr) enzyme. The transgenic line is made such that the target tissue expresses Ntr [71][72][73]. To induce cell death, the transgenic fish are exposed to the prodrug metronidazole, which will be up taken by virtually all tissues of the organism [71][72][73]. In the presence of Ntr, the metronidazole is processed into a DNA cross-linking agent that causes cell destruction [71][72][73][74]. To date, nephrology researchers have used this transgenic technology to selectively ablate podocytes, which are specialized epithelial cells that form the glomerular filter, in order to assess podocyte regeneration [75]. Specifically, a Tg(podocin:ntr-mCherry) line was engineered in which damage could be induced specifically in the podocytes of the glomerulus [76]. Treatment with metronidazole led to severe edema, as well as apoptosis within the glomeruli, indicating a failure of renal function [76]. Furthermore, metronidazole-treated Tg(podocin:ntr-mCherry) fish exhibited leakage in which large proteins were capable of entering into the nephron tubule [76]. A similar line, Tg(podocin:ntr-GFP), was used in an independent study in which fish were treated with metronidazole for either 12 or 72 h [77]. At 12 h post-treatment, glomeruli displayed foot process disruption, as well as chromatin condensation [77]. This was exacerbated in fish treated for 72 h that exhibited a complete loss of foot processes and had significantly reduced functional podocytes; however, stopping treatment with metronidazole for 7 days resulted in the reappearance of GFP, suggesting glomerular repair [77]. These research reports illustrate the power of transgenic systems to manipulate cellular viability and hold significant promise for future AKI modeling. This method can be used to make models of cell injury across all nephron cell types, using identified promoters or taking advantage of the powers of recombineering with BACs, which does not necessitate promoter isolation [78], to further expand the types of AKI which can be modeled.

Conclusion
Despite a rich classical literature documenting the existence of nephron regeneration across vertebrates, the genetic and molecular mechanisms have remained elusive. The studies discussed here illustrate how models of AKI have been created in embryonic and adult zebrafish models in recent years. Taken together, these reports show that a wide variety of methods exists to induce an AKI-like phenotype in zebrafish including injection of nephrotoxins such as gentamicin and cisplatin, laser ablation, and the use of transgenic lines expressing the bacterial nitroreductase in specific tissues. These methods of damage induction have been used in tandem with chemical screens to identify molecules that can convey a pro-regenerative effect and enhance renal recovery, leading to the discovery that exposure to HDACi compounds can ameliorate kidney injury. The field has recently made significant strides in characterizing nephron anatomy and the architecture of the adult mesonephros. Recent work highlights that zebrafish adult nephrons undergo robust epithelial regeneration and the formation of nephrons de novo, the latter that sets zebrafish apart from mammals. Future studies are likely to employ these damage methods as well as new paradigms such as transgenic lines in which injury can be induced in the tubular cells via the nitroreductase-metronidazole system, to further characterize how individual populations of cells in the nephron respond to damage. For example, zebrafish continue to be used as a valuable organismal setting in which to identify nephrotoxins and evaluate their effects [79][80][81][82][83]. As many regenerative phenomena at least partly involve the recapitulation of developmental pathways [84,85], a complementary and powerful future approach to those noted here will be to explore the roles of nephrogenesis factors in renal regeneration. Additionally, exploration of how reprogramming might be invoked to elicit desirable cell behaviors in vivo or create in vitro sources of replacement cells [86]. The continued use of zebrafish to investigate AKI presents great promise for systematically unraveling the complexities of this diverse clinical condition, as well as identifying potent regenerative therapeutics to treat these conditions.  This study was the first to describe the zebrafish embryonic kidney, to establish methods to characterize essential nephron elements like the glomerular filter, and presented the first collected panel of mutations that disrupt renal formation in the zebrafish 27.