SEARCH

SEARCH BY CITATION

Keywords:

  • pronephros;
  • kidney regeneration;
  • renal repair;
  • wound healing;
  • Mmp-9;
  • apoptosis;
  • Xenopus laevis

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Background: While the renal system is critical for maintaining homeostatic equilibrium within the body, it is also susceptible to various kinds of damage. Tubule dysfunction in particular contributes to acute renal injury and chronic kidney disease in millions of patients worldwide. Because current treatments are highly invasive and often unavailable, gaining a better understanding of the regenerative capacity of renal structures is vital. Although the effects of various types of acute damage have been previously studied, the ability of the excretory system to repair itself after dramatic tissue loss due to mechanical damage is less well characterized. Results: A novel unilateral nephrectomy technique was developed to excise pronephric proximal tubules from Xenopus laevis tadpoles to study tubule repair after injury. Immunohistochemical detection of protein expression and renal uptake assays demonstrated that X. laevis larvae have the capacity to regenerate functional proximal tubules following resection. Conclusions: We have validated the renal identity of the restored tubules and demonstrated their ability to functional normally providing the first evidence of regeneration of renal tissue in an amphibian system. Importantly, this tubule restoration occurs by means of a process involving an early apoptotic event and the biphasic expression of the matrix metalloproteinase, Xmmp-9. Developmental Dynamics 242:219–229, 2013. © 2012 Wiley Periodicals, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Because the kidney is necessary for the maintenance of vertebrate homeostasis, many organisms possess the ability to repair minor damage to nephric structures (Menè et al., 2003; Nony and Schnellmann, 2003; Bonventre, 2010; Humphreys, 2011). Despite this, injuries that result in a reduction of nephron number in mammals, lead to a permanent renal deficit, as they are unable to replace lost or severely damaged nephrons. In contrast, some vertebrates, most notably fish, are able to undergo neonephrogenesis as adults, replacing damaged nephric tissue throughout their lifetime (Reimschuessel and Williams, 1995; Augusto et al., 1996; Salice et al., 2001; Elger et al., 2003; Watanabe et al., 2009; Zhou et al., 2010). Although the renal systems of vertebrates differ in complexity (metanephroi in mammals, reptiles and birds; mesonephroi in adult amphibians and fish; pronephroi in juvenile amphibians and fish; Saxén, 1987), comparative studies provide unique opportunities to identify and characterize the underlying mechanisms that facilitate the restoration of a severely damaged excretory system.

Although studies have demonstrated that the mammalian metanephric kidney has remarkable regenerative capacity after experiencing acute ischemic and/or toxic injury, this repair is primarily restricted to repopulation of the denuded basement membrane of injured nephrons by means of local proliferation and migration of adjacent surviving epithelial cells (Nonclercq et al., 1992; Bonventre, 2003; Nony and Schnellmann, 2003). Compensatory hypertrophism also occurs in both the injured kidney (Wesson, 1989), and the contralateral kidney (Sheridan and Bonventre, 2000; Menè et al., 2003). In contrast, teleost (Reimschuessel and Williams, 1995; Reimschuessel, 2001; Salice et al., 2001; Liu et al., 2002) and elasmobranch fish (Elger et al., 2003) have demonstrated a greater capacity to restore damaged renal structures. Additional research has yielded important information regarding repair processes that occur after acute damage resulting from exposure to nephrotoxic agents, such as the aminoglycoside antibiotic gentamicin (Augusto et al., 1996; Hentschel et al., 2005; Watanabe et al., 2009; Diep et al., 2011). However, while these chemotoxic studies have been extensively investigated, very little is known about the renal response to mechanical injury.

The ability to study the renal repair process after severe damage, such as that seen in patients with acute kidney injury (AKI) or chronic kidney disease (CKD), has proven to be more challenging. While these disorders are diagnosed based on loss of renal function, oftentimes the conditions are the result of obstructive injury and/or irreversible destruction of nephric tissue integrity. Experimentally inducing damage of this magnitude in metanephric and mesonephric kidneys can be difficult due to their complexity and inaccessibility within the animal. The embryonic structure, the pronephros, provides a simpler and more easily accessible organ system with which to conduct these types of studies. Although laser ablation of pronephric tubules has recently been optimized in zebrafish as an alternative to chemotoxic damage (Johnson et al., 2011), no other mechanical injury models in pronephric kidneys currently exist. With this study, we offer a novel technique for inducing mechanical damage in renal tissue in the amphibian model organism, Xenopus laevis. While excision of fish pronephroi is complicated by a fused glomus, the tadpole has a separate glomus for each pronephros, enabling the partial or complete excision of one kidney. Additionally, the tadpole pronephros is located dorsolaterally, just beneath the skin, making it accessible for this type of manipulation. Furthermore, pronephric developmental is very well characterized (Vize et al., 1997; Brändli, 1999; Carroll et al., 1999a, 1999b), and as such may provide useful insight into repair mechanisms in differentiated tissue. Although aspects of kidney repair have been previously studied in mammals (Boti et al., 1982; Nonclercq et al., 1992; Imgrund et al., 1999) and most recently in fish (Watanabe et al., 2009; Zhou et al., 2010; Diep et al., 2011), little is known about the underlying mechanisms that mediate these events.

In this study, we provide the first evidence of pronephric kidney regeneration after severe mechanical damage in an amphibian system. Additionally, our initial characterization of the cellular and molecular basis of renal repair revealed both an early apoptotic event that occurs within hours of renal injury as well as a biphasic, regeneration-specific pattern of matrix metalloproteinase-9 (Xmmp-9) expression. Moving forward, a greater understanding of events that facilitate tissue repair after mechanical injury will be pivotal to help clarify whether similar processes can be re-established in mammalian metanephric renal systems.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Unilateral Nephrectomy Is a Tractable and Reproducible Model of Mechanical Renal Damage

Pronephric kidneys in X. laevis are functional from 3 days postfertilization (at room temperature, Nieuwkoop and Faber stage 37/38), and degenerate during premetamorphosis (beginning at NF stage 53), at which time the mesonephros starts to function (Nieuwkoop and Faber, 1994). To allow the maximum amount of time for the excised proximal tubules to regenerate, nephrectomies were performed at the onset of kidney function, 3 weeks before this natural degeneration process. NF stage 37/38 pronephric kidneys are cephalic organs located ventral to trunk somites I and II, in small opaque bundles on either side of the tadpole. The proximal tubules lie immediately adjacent to the skin, and can be identified by means of light microscopy as a thin tubular network once the skin has been pulled back. Careful excision with fine forceps was performed and excised tubules were visually identified upon removal.

In the majority of nephrectomized tadpoles, the proximal tubules were partially or completely removed as intended, as confirmed by the partial (Fig. 1A) or complete (Fig. 1B) ablation of 3G8 protein expression on the operated side (Fig. 1H). The vast majority of damage control tadpoles retained undamaged proximal tubules, expressing the normal 3G8 pattern (Fig. 1C,H). The 3G8 expression detected in the proximal tubules of unoperated sibling control tadpoles was always normal (Fig. 1D,H).

image

Figure 1. Proximal tubules are successfully excised during partial nephrectomy in NF stage 37/38 Xenopus laevis tadpoles. A–D: The 3G8-positive tissue was often partially (A) or completely ablated (B; left side) at 1 day postnephrectomy (dpn; 315 tadpoles), but occurred in the expected (phenotypically normal) pattern in the vast majority of damage control tadpoles (C; 52 tadpoles) and all unoperated sibling control tadpoles (D; 291 tadpoles). E–H: Damage in adjacent tissues was also examined by means of gene expression at 1 dpn. Normal expression of the alpha 5 subunit of Na, K-ATPase (E; 22 tadpoles), nephrin (F; 51 tadpoles), and 12/101 (G; 58 tadpoles) are retained in the pronephric duct, glomus, and skeletal muscle, respectively, on the operated side of the majority of nephrectomized tadpoles (H). A,C–G: All whole-mount photographs are lateral views of representative tadpoles, with anterior to the left. Black circles highlight the renal area. Black arrow marks the pronephric duct. White arrow indicates somitic muscle and white arrowhead points to hypaxial muscle. Photograph of 10 -μm section of paraffin-embedded/eosin counter-stained 1 dpn tadpole (B), with 3G8-positive proximal tubules (broken white outlines) present on the unoperated contralateral side but absent on the operated side. Scale bar = 100 μm; c, nephric coelom; g, gut; m, melanocytes (black arrowheads); n, notochord. Graph (H) illustrates the proportion of nephrectomized tadpoles with phenotypically normal, partially ablated, and completely ablated gene expression (in parentheses) in several tissues on the operated side.

Download figure to PowerPoint

Although nephrectomy undoubtedly disrupted the pronephric sinus (capillary system) surrounding the tubules, the surgery typically precluded damage to other nephric components as was demonstrated by examining nontubule components of the pronephric kidney 1 day postnephrectomy (dpn). Undamaged pronephric ducts, which express the alpha 5 subunit of the sodium-potassium adenosine triphosphate pump (Na,K-ATPase), were always observed at 1 dpn (Fig. 1E,H). The glomus, which is situated medially to the tubules below trunk somites I and II, was undamaged in the majority of nephrectomized tadpoles, displaying normal expression of the podocyte marker, nephrin (Fig. 1F,H). Additionally, the integrity of adjacent skeletal muscle was confirmed by examining 12/101 protein expression. Damage was rarely observed in either the somites or the hypaxial muscles, which flank the pronephros dorsally and ventrally, respectively (Fig. 1G,H). These results demonstrate that unilateral nephrectomy can be used to create mechanical damage to a specific component of the pronephric kidney in Xenopus laevis tadpoles, while maintaining the integrity of neighboring tissues.

Functional Proximal Tubule Tissue Is Restored After Partial Nephrectomy

After establishing a technique that allows for the removal of a specific renal structure, the ability to restore proximal tubules was examined. The presence of differentiated proximal tubule tissue was confirmed by means of detection of 3G8 expression on the ipsilateral (operated) side of tadpoles at 1 dpn and 21 dpn, as well as in the kidneys of unoperated sibling controls (Fig. 2). Only a few nephrectomized tadpoles (4%) retained normal 3G8 expression 1 day after nephrectomy. In contrast, at 21 dpn, significantly more nephrectomized tadpoles possessed coiled proximal tubules expressing 3G8 normally on the operated side (17%; P = 0.00012, Student's t-test; Fig. 2A). These regenerates were phenotypically comparable to unoperated control kidneys (compare Fig. 2B,C). This demonstrates that a portion of the X. laevis tadpole population can successfully restore differentiated proximal tubules after extensive resection of this tissue. The remaining 21 dpn tadpoles either had no detectable 3G8-positive cells or had tubule tissue in varying stages of repair (“partial 3G8”; Fig. 2A).

image

Figure 2. Xenopus laevis tadpoles are able to regenerate proximal tubules 21 days after partial nephrectomy. A: The majority of nephrectomized tadpoles showed partial or no 3G8 expression at 1 day postnephrectomy (dpn). At 21 dpn, significantly more tadpoles expressed 3G8 normally in coiled proximal tubules on the operated side. Conversely, significantly fewer 21 dpn tadpoles displayed partial 3G8 expression. The proportion of 21 dpn tadpoles with no 3G8-positive ipsilateral tissue, is comparable to the percentage of the 1 dpn population with completely excised 3G8-positive tubules. A total of 315 animals was assayed at 1 dpn. A total of 458 animals was assayed at 21 dpn. Error bars indicate standard error among 19 replicates/group. Asterisks denote two-tailed Student's t-test between 1 dpn and 21 dpn groups: *P = 0.03; **P = 0.0001. B,C: All 21 dpn tadpoles with normal ipsilateral 3G8 expression in regenerated kidneys, displayed gene expression and tubule morphology similar to those observed in the unoperated sibling control tadpoles (compare B and C). Photographs are magnified lateral views of proximal tubules in representative tadpoles, with anterior to the left.

Download figure to PowerPoint

To determine if regenerated tubules regained renal function, we performed renal uptake assays at 23 days postnephrectomy, as previously described (Zhou and Vize, 2004). Twenty-four hours after injection, the location of fluorescently tagged bovine serum albumin (BSA-AF488) was examined. All unoperated sibling control kidneys (Fig. 3A,B) and all contralateral kidneys in nephrectomized tadpoles (Fig. 3G,H), showed reabsorption of BSA into proximal tubule cells. Pronephric regenerates were also able to reabsorb BSA (Fig. 3D–E). As expected (Zhou and Vize, 2004), no fluorescence was observed in the distal tubules of any tadpoles (Fig. 3B,E,H). All injected tadpoles were also examined for 3G8 protein expression to confirm proximal tubule identity of the regenerates (compare Fig. 3C, 3F, and 3I). These results demonstrate that 3 weeks after severe mechanical injury, restored proximal tubules are able to function normally, contributing to homeostatic equilibrium of the tadpoles. This provides the first evidence of regeneration of functional nephric structures in an amphibian system after a severe renal insult.

image

Figure 3. Regenerated pronephric proximal tubules are functional 3 weeks after excision. A–C: All unoperated control kidneys (115 animals) were able to reabsorb fluorescently tagged bovine serum albumin (BSA) into proximal tubules (A,B), as confirmed by the expression of the proximal tubule-specific protein, 3G8 (C). D–F: Of the 109 nephrectomized tadpoles assayed, 10 individuals (9%) restored ipsilateral coiled structures capable of both reabsorbing fluorescently-tagged-BSA (D,E) and expressing 3G8 (F). G–I: Comparable coiled morphology, BSA re-uptake capability and 3G8 expression were also seen in all unoperated contralateral proximal tubules. Photographs are lateral views of proximal tubules in representative tadpoles, with anterior to the left in A–F, and to the right in G–I. BF, brightfield. Broken outlines indicate proximal tubules. Solid outlines indicate distal tubules.

Download figure to PowerPoint

While only a subset of the nephrectomized population was capable of successfully regenerating a fully coiled, functional pronephric organ, other individuals possessed smaller, uncoiled tubules that were also capable of reabsorbing BSA-AF488. It is likely that these shorter tubules had restored integration with the nephric coelom and so were able to receive filtered albumin. Presumably, these tubules would continue to elongate if allowed to repair for more than 3 weeks. These observations indicate that a single population of nephrectomized tadpoles contains individuals in many stages of tubule restoration at any given time after injury, and further suggests that there is some stochasticity in the regenerative process used by amphibian larvae after nephrectomy.

An Early Apoptotic Event Occurs During Proximal Tubule Regeneration

We assessed levels of apoptosis at the site of injury by examining the expression of active Caspase 3 protein. As expected for any developing organism, low levels of apoptosis were detected in various tissues of all tadpoles examined, most notably in the gut and brain. In the renal region, zero to three apoptotic cells could be found in unoperated control kidneys and contralateral kidneys in most nephrectomized tadpoles. For this reason, an increase in active Caspase 3 at the site of nephrectomy was determined by observing more active Caspase 3-positive cells on the ipsilateral side relative to the contralateral side. Damage control tadpoles and unoperated tadpoles were scored using the same criterion.

A dramatic increase of active Caspase 3-expressing cells was detected in the operated renal area in only the nephrectomized tadpole treatment group (Fig. 4). Apoptotic cells were observed on the ipsilateral side of nephrectomized tadpoles at 3 hr postnephrectomy (hpn) and remained detectable in the majority of this population through the first 24 hr following tubule excision (Fig. 4A). The maximum number of active Capase-3 positive cells was observed at 12 hpn (Fig. 4B–D). Levels remained significantly elevated in nephrectomized tadpoles at 2 dpn, but returned to basal levels by 7 dpn. Strikingly, almost no active Caspase 3 protein was found in damage control tadpoles at any time-point examined (Fig. 4A), suggesting that injury alone is not sufficient to induce apoptosis in this system. Apoptosis was also undetectable in unoperated control tadpoles. Thus, this early apoptotic event is part of a specific response to pronephric tubule excision.

image

Figure 4. Nephrectomized tadpoles have increased levels of apoptosis during the first 24 hr postsurgery. A: A significant increase in the number of nephrectomized tadpoles expressing active Caspase 3 protein (as compared to both damage controls and unoperated sibling controls) was observed throughout the first day postsurgery. This number decreased significantly by the second day, and remained at low levels throughout the remainder of the first week postsurgery. A total of 25–70 animals from each treatment group were assayed at each time-point examined. Error bars indicate standard error among 2–3 replicates/group. Asterisks denote single-factor analyses of variance tests among different groups at the same time-point: *P < 0.05; **P < 0.001. B–D: Photographs are lateral views (anterior to the left) of an unoperated sibling control (B), damage control (C), and nephrectomized tadpole (D), at 12 hr postsurgery. White circle indicates renal area with active Caspase 3-expressing cells.

Download figure to PowerPoint

Xmmp-9 Is Expressed in a Biphasic Pattern After Unilateral Nephrectomy

In addition to apoptosis, we also examined expression of the extracellular matrix (ECM) remodeler, matrix metalloproteinase 9 (Xmmp-9) during the first week after nephrectomy. As reported previously, small numbers of Xmmp-9-expressing cells are routinely observed in numerous tissues throughout development including the epidermis, intestine, cloaca, tail fin, and facial structures (Carinato et al., 2000). However, this baseline Xmmp-9 expression was punctate, and was not concentrated in any one location. In contrast, up-regulated Xmmp-9 expression was observed in the ipsilateral renal area of nephrectomized tadpoles in a biphasic pattern.

The first phase of increased Xmmp-9 expression began at 3 hpn and was sustained for the first 24 hr, with the vast majority of tadpoles displaying this phenotype between 12 hpn and 24 hpn (Fig. 5A). Notably, the number of Xmmp-9 positive cells in nephrectomized tadpoles also increased dramatically by 12 hpn (Fig. 5D) and remained elevated through 24 hpn. As predicted by the role of Xmmp-9 in wound healing (Carinato et al., 2000), damage control tadpoles also experienced up-regulation of transcript levels on the ipsilateral side during the first 24 hr. However, between 12 and 24 hr postinjury, significantly fewer damage control tadpoles than nephrectomized tadpoles had an increase in Xmmp-9 at the site of renal injury (Fig. 5A). Additionally, the level of transcript expression was also noticeably less in damage control tadpoles (compare Fig. 5C and 5D). Unoperated sibling control tadpoles had no Xmmp-9 up-regulation during this first 24-hr phase (Fig. 5A,B). Ipsilateral Xmmp-9 expression in all groups declined by 2 days postnephrectomy, and remained low for the following 48 hr.

image

Figure 5. Xenopus laevis tadpoles express Xmmp-9 in a biphasic pattern after unilateral nephrectomy. A: In situ hybridization analysis demonstrated that Xmmp-9 expression was dramatically up-regulated in both nephrectomized and damage control groups within 3 hr of injury. Significantly more nephrectomized tadpoles displayed this phenotype between 12 and 24 hr postsurgery. This number decreased by 2 days postnephrectomy (dpn), and remained low until 5 dpn, at which time a second phase of Xmmp-9 up-regulation began on the ipsilateral side of only nephrectomized tadpoles. This phase persisted through 7 dpn. A total of 25–70 animals from each treatment group were assayed at time-points examined. Error bars indicate standard error among three to five replicates/group. Asterisks denote single-factor ANOVA tests among different groups at the same time-point: *P = 0.05; **P < 0.005. B–G: Photographs are lateral views (anterior to the left) of unoperated sibling controls (B,E), damage controls (C,F) and nephrectomized tadpoles (D,G), at 12 hr (B–D) or 7 days (E–G) postsurgery. Black circles indicate renal area in each tadpole.

Download figure to PowerPoint

A second phase of ipsilateral transcript up-regulation began at 5 dpn, but only in nephrectomized tadpoles. Both the proportion of the population re-expressing Xmmp-9, and the number of Xmmp-9-positive cells in each individual, increased over the next 48 hr, peaking at 7 dpn (Fig. 5A,G). Strikingly, transcript up-regulation was not detected on the ipsilateral side of damage control tadpoles during this phase (Fig. 5A,F). Unoperated sibling control kidneys also displayed no increase in transcript levels (Fig. 5A,E).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

All nephric systems examined to date have been found to possess some form of reparative mechanism to offset the effects of acute damage. Regeneration of severely damaged or lost nephrons is less commonly observed, and as such is poorly understood. We are interested in understanding the renal repair response after this type of dramatic tissue loss. To accomplish this, we first needed to establish an injury model that would reproducibly create damage of this severity. In this study we describe a novel unilateral nephrectomy technique during which proximal tubules are preferentially excised from X. laevis pronephroi. We have demonstrated by means of morphological and molecular data that tadpoles have the capacity to regenerate functional proximal tubules, thus providing the first evidence of renal regeneration in an amphibian system. We also show that tubule restoration occurs by means of a process involving an early apoptotic event and the biphasic expression of the matrix metalloproteinase, Xmmp-9.

Using protein and transcript expression in various cell types of both renal and nonrenal organs, we confirmed that proximal tubules were successfully targeted during surgery, while nontarget tissues remained largely unaffected in nephrectomized tadpoles (Fig. 1). The 3G8 protein expression analyzed in 19 cohorts of nephrectomized tadpoles indicated that proximal tubule cells were either completely or partially removed in the majority of the population (Fig. 1H). Conversely, normal proximal tubule 3G8 expression was observed in almost all damage control tadpoles (Fig. 1C,H). In addition, although a few of the damage control sibling animals contained atypical, “partially ablated” 3G8 expression that resembled tubule structures, at 1 dpn no discernible tubules were observed in nephrectomized tadpoles that contained “partial 3G8 expression” (Fig. 1A). These results clearly demonstrate that this nephrectomy technique is effective at specifically creating severe loss of pronephric proximal tubules.

It is important to note, however, that a small percentage (4%) of nephrectomized tadpoles examined at 1 dpn, appear to have normal 3G8 expression (Fig. 1H). It is likely that the damage experienced by these few individuals is more subtle and thus not detectable by the macro-examination of protein expression. However, because the 3G8 pattern in these individuals is indistinguishable from that of unoperated control kidneys (Fig. 1D), we must consider the possibility that these individuals will retain relatively normal morphology and physiology in both pronephroi throughout the 3-week repair period. Nonetheless, a significantly larger percentage of nephrectomized tadpoles was able to undergo complete tubule restoration by 21 dpn (Fig. 2A), and these regenerates re-express genes associated with terminally differentiated kidneys (Fig. 2C). The 3G8 protein is specifically expressed in the apical membrane of proximal tubule cells (Vize et al., 1995), and so re-expression of this protein suggests that restored tubules regained not only morphological structure, but also cellular polarity.

Additionally, the ability of regenerates to reuptake albumin (BSA-AF488; Fig. 3E) indicates that these tissues are comprised of lumenized cells. Albumin reuptake from the glomal/glomerular filtrate occurs by means of an active process involving multi-ligand endocytic receptors specifically localized to the apical plasma membrane of proximal tubules segments of both pronephroi (Zhou and Vize, 2004; Christensen et al., 2008), as well as mesonephric and metanephric kidneys (Lauriola et al., 1986; Birn et al., 2000; Verroust et al., 2002; Lazzara and Deen, 2007). The reuptake of BSA-AF488 by restored proximal tubules indicates that these regenerates have regained luminal flow of glomal filtrate and are re-expressing crucial endocytic proteins.

Of interest, the proportion of 21 dpn tadpoles that had no detectable 3G8-positive tissue on the operated side is comparable to the percentage of the 1 dpn population lacking 3G8-positive tubules (Fig. 2A). This correlation suggests that tissue renewal may rely on the presence of surviving proximal tubule cells. Similar conclusions have been previously made after examining renal responses to both nephrotoxic injury (Cuppage et al., 1972; Kovacs et al., 1982; Wallin et al., 1992; Kays and Schnellmann, 1995) and ischemia/reperfusion (Venkatachalam et al., 1978; Witzgall et al., 1994; Humphreys et al., 2008) in metanephric nephrons (reviewed in Bonventre, 2003; Nony and Schnellmann, 2003; Humphreys et al., 2011). Surviving proximal tubules cells have also be shown to be important in repair events during human acute tubular necrosis (Nadasdy et al., 1995). It is likely that pronephric proximal tubule repair is similar to these metanephric systems, and thus also requires the survival of some proximal cells in order for tubule regeneration to occur. We postulate that the tubule regeneration observed in this study occurred in individuals that experienced partial nephrectomy of proximal tubules (Fig. 2A, 1 dpn grey bar).

Although metanephric kidneys are able to repopulate denuded basement membranes that remain intact after injury, these kidneys are unable to recover after dramatic loss of renal mass. This study, however, provides the first evidence of pronephric kidneys regenerating functional tubule architecture after dramatic loss of renal tissue. It is likely that the less complex nature of the pronephric kidney makes it amenable to this extensive restoration, thereby making it a suitable organ to investigate kidney repair. Moreover, because severe kidney damage in humans is often accompanied by an intense fibroproliferative response (Liu, 2011), which is absent in nephrectomized tadpoles, elucidating the mechanisms that contribute to this pronephric regeneration will have important implications in innovating alternative treatments for AKI and CKD. For this reason, we examined important repair events that occur upon nephrectomy in this pronephric system.

Apoptosis or programmed cell death (PCD) is a nonnecrotic mode of cellular disposal that is critical in amphibian development and is especially prevalent during metamorphosis (Nakajima et al., 2005). Notably, apoptosis has also been observed during many amphibian regenerative processes (Lo et al., 1993; Gardiner et al., 1999; Carlson et al., 2001; Flink, 2002; Bettencourt-Dias et al., 2003; Gargioli and Slack, 2004; Tseng et al., 2007; Yoshii et al., 2007). We have also detected an increase in apoptosis that is restricted to the first 24 hr after surgery in the majority of nephrectomized tadpoles (Fig. 4). In contrast, there was no significant increase in apoptosis in damage control tadpoles indicating that PCD is not a general response to injury. Of interest, apoptosis was also observed in tubule epithelia during the first 24 hr after induced regeneration of noninfarcted atrophic rat kidneys (Gobé et al., 1995). It is likely that this activity is a critical precursor to kidney regeneration as has been found in other systems. For example, Tseng et al. (2007) demonstrated that tail regeneration in X. laevis tadpoles cannot proceed if apoptosis is inhibited during the first 24 hr after tail amputation.

The timing of apoptosis after nephrectomy overlaps with the wound healing process, which may indicate a role for PCD in creating a permissive environment before regeneration. Although this permissive environment model is well-established in other regenerative systems (Price and Allen, 2004; Tucker et al., 2008; Zukor et al., 2011), the underlying role of apoptosis during early regenerative events is still unclear. It is possible that cell death is critical for eliminating damaged tissue before stimulation of renal restoration, or as suggested by Tseng and colleagues (2007), that apoptosis serves to destroy a subpopulation of cells that may normally inhibit regeneration. A third possibility is that PCD may be critical in limiting inflammatory damage, with a lack of apoptosis resulting in scar formation and perpetuation of renal disease (Gobe and Buttyan, 2002). This possibility is especially attractive considering that regenerative success is often thought to be the alternative to scarring (Poss et al., 2002; Harty et al., 2003; Price and Allen, 2004).

In mammalian metanephric kidneys, which have no neonephrogenic properties, injury typically results in extensive inflammatory activity leading to extensive renal fibrosis (Hewitt et al., 2008; Wynn, 2008; Liu, 2011). However, the immune system found in mammals is more highly evolved than that of anuran larvae, as well as adult urodeles (Robert and Cohen, 1998), both of which are well-established model systems for regeneration. Because the observed apoptotic event coincides with the onset of inflammation after injury, it is likely that apoptosis may modulate the inflammatory response to allow for tissue regeneration instead of fibrosis.

Establishing a fibrosis-free permissive environment for regeneration also involves remodeling of the ECM to facilitate cellular migration and cell–cell signaling, as well as to provide a scaffold for tissue reconstruction. Many of these ECM modifications are performed by the MMPs, a multi-member family of zinc-dependent endopeptidases (Matrisian, 1992; Johnson et al., 1998). MMP-9 in particular has been implicated in regeneration, with levels up-regulated both at the onset of wound healing (Salo et al., 1994; Fini et al., 1998; Yang et al., 1999; Carinato et al., 2000; Liu et al., 2009) and after the completion of wound healing (Yang et al., 1999). We speculate that the first 12 hr of Xmmp-9 expression observed after injury in both nephrectomized and damage control tadpoles mediates wound healing, which also coincides with the length of time for skin wound closure after injury. Strikingly, from 12–24 hr Xmmp-9 continues to be robustly expressed in nephrectomized tadpoles alone (Fig. 5A), suggesting that MMP-9 up-regulation during this period may be a specific response to loss of tubule tissue. Previous work demonstrated that suppressed MMP-9 activity in damaged metanephric kidneys results in excessive collagen deposition (González-Avila et al., 1998; Chromek et al., 2003; Chromek et al., 2004; Bengatta et al., 2009). Consistent with our findings, it is possible that MMP-9 plays a similar role in the pronephros and serves to minimize fibroproliferation, thereby facilitating the creation of the proper environment for regeneration to occur. This is even more likely given the ability of MMP-9 to cleave interstitial Type-I collagen (Bigg et al., 2007), a predominant protein found in renal scar tissue (Huang et al., 2011).

The second phase of Xmmp-9 expression (5–7 dpn), which was also only observed in nephrectomized tadpoles (Fig. 5A), indicates that this up-regulation is involved in tubule restoration, and is not a general response to damage. A similar biphasic pattern of Mmp-9 was observed in axolotl limb regeneration, both during wound healing and then later during the creation of new cartilage (Yang et al., 1999). In the renal system, MMP-9 may be playing a similar role in the reconstruction of tubule tissue. This is an especially appealing possibility given that MMP-9 is critical for tubulogenesis and branching during development of the kidney (Lelongt et al., 1997; Arnould et al., 2009). Over a decade ago, researchers began to examine molecular profiles of renal repair, and detected similarities in gene expression during mammalian metanephric development and epithelial renewal in injured adult kidneys (Reimschuessel, 2001). Because the developmental program of kidney organogenesis is well conserved among vertebrates (Vize et al., 1997; Drummond et al., 1998; Kuure et al., 2000; Dressler, 2006), then this may also be true of the regenerative process.

Of interest, murine MMP-9 has been shown to stimulate branching morphogenesis during kidney development by protecting cells from apoptosis (Arnould et al., 2009). Similarly, previous work on acute kidney injury in mice suggests that MMP-9 is critical to reduce the incidence of apoptosis during later stages of renal repair (Bengatta et al., 2009). Because the up-regulation of both MMP-9 and apoptosis overlap in this injury model, it is possible that they are intricately linked in a similar manner during this restorative process. Further investigation into this possibility is necessary to elucidate the roles that MMP-9 and apoptosis play in this pronephric kidney regeneration process following nephrectomy. Ultimately, this line of inquiry has important implications for clinical research aimed at attenuating the effects of renal disease in humans.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Tadpoles

All experiments conducted were performed in accordance with the Guide for Care and Use of Laboratory Animals and were approved by means of the Institutional Animal Care and Use Committee (IACUC) at Tufts University. Adult female Xenopus laevis frogs were induced to ovulate by means of chorionic gonadotropin hormone (Chorulon) injection. Adult male wild-type X. laevis frogs were killed by means of intraperitoneal injection with tricaine (MS-222, Acros Organics) and the testes removed and stored in 1X modified Barth's saline (pH 7.5). Eggs were fertilized in vitro, and resultant embryos were dejellied in a 2% cysteine solution (pH 8) and reared in 0.1X Marc's Modified Ringer's solution [MMR; 10 mM NaCl, 0.2 mM KCl, 0.1 mM CaCl, 0.2 mM MgCl2, 0.5 mM HEPES, 1 μM EDTA, pH 7.4 (modified from Kay and Peng, 1991)] at 14–25 °C. 0.1X MMR solution was changed daily. Embryos and tadpoles were staged according to Nieuwkoop and Faber (NF, 1994). Feeding-stage tadpoles were fed Nasco frog brittle or Sera Micron powdered growth food.

Unilateral Pronephrectomy

NF stage 37/38 tadpoles (Nieuwkoop and Faber, 1994) were randomly assigned to one of three treatment groups: nephrectomies, damage controls, or unoperated sibling controls. Tadpoles in all groups were first individually anaesthetized with 0.04% tricaine in 0.1X MMR. During nephrectomies, fine forceps were used to make a small lateral incision adjacent to the left gill flap, and to peel back the skin covering the pronephros. Proximal tubules were identified by means of light microscopy as a thin tubular network, and were promptly excised and rinsed off the forceps by submersion and gentle agitation in 0.1X MMR. Nephrectomized tadpoles were immediately transferred to fresh 0.1X MMR to recover from anesthesia. Damage control tadpoles were poked multiple times through the skin into the left-side renal region. Tadpoles were placed in 14 °C incubators overnight. After 24 hr, tadpoles were transferred to 18 °C where they were maintained for the entirety of each experiment. At the desired time-point postsurgery, tadpoles were euthanized with tricaine, fixed for 1 hr at room temperature in MEMFA (0.1 M MOPS, pH 7.4, 2 mM EGTA, 1 mM MgSO4, 3.7% formaldehyde), rinsed in 1X phosphate buffered saline (PBS), and gradually dehydrated to 100% methanol for storage at −20 °C.

Immunohistochemistry (IHC)

To examine localization of proteins, whole-mount IHC was performed on tadpoles (NF stages 37–45 and decapitated/tail-amputated NF stages 48–50). Briefly, tadpoles were permeabilized by washes in PBTr (1X PBS with 0.1 % Triton and 2 mg/ml bovine serum albumin [BSA]), blocked with 20% heat-inactivated goat serum in PBTr, and incubated overnight with primary antibody. Tadpoles were then rinsed in PBTr for 4–5 hr, blocked and incubated overnight in either goat anti-mouse or goat anti-rabbit IgG secondary antibodies conjugated to either alkaline phosphatase (AP; 1:1,500) or Alexa Fluor 555 (AF-555; 1:300; Invitrogen, Sigma, or SouthernBiotech). Protein expression was detected after 5 hr of PBTr washes, and incubation in chromogenic buffer (100 mM Tris, 100 mM NaCl, 50 mM MgCl2, 0.1% Tween-20) with 5-bromo,4-chloro,3-indolylphosphate/nitroblue tetrazolium (1:600 BCIP/1:600 NBT; Roche Diagnostics) or by means of fluorescence. Tadpoles were then rinsed in 1X PBS, post-fixed in MEMFA for 1–3 hr at room temperature or overnight at 4 °C, dehydrated to 100% methanol and stored at −20 °C. The following tissue types/markers were assessed through IHC: proximal tubule/3G8 (gift from Elizabeth Jones), pronephric duct/Alpha 5 subunit of Na, K-ATPase (Developmental Studies Hybridoma Bank, University of Iowa), skeletal muscle/12/101 (Developmental Studies Hybridoma Bank, University of Iowa), and apoptotic cells/active Caspase 3 (Abcam and BD Biosciences Pharminogen).

Histology

After IHC, dehydrated tadpoles were gradually transferred to ethanol through a series of washes of increasing gradient (70%, 90%, 100%) then rinsed twice in absolute ethanol. Specimens were cleared in three successive 20-min xylene incubations and infiltrated twice in paraffin wax at 60 °C for 45 min, then wax-embedded in molds overnight. Ten micrometer sections were cut with a Leica 2255 rotary microtome, and mounted on Superfrost Plus glass slides. Sections were dewaxed in a series of xylene washes, counterstained with eosin (Fisher Scientific) and mounted with Permount (Fisher Scientific). Sections were then examined with an Olympus BX40 light microscope and photographed using DP Controller software.

In Situ Hybridization (ISH)

To examine localization of mRNA transcripts, whole-mount ISH was performed on tadpoles (NF stages 37–45 and decapitated/tail-amputated NF stages 48–50). Tadpoles were hybridized with digoxigenin-labeled antisense riboprobes as described previously (Harland, 1991). mRNA expression was detected by means of an anti-digoxigenin antibody conjugated to AP (1:1,500; Roche Diagnostics) and a blue BCIP/NBT precipitate. Tadpoles were post-fixed as described for IHC. The following transcripts were detected through ISH: Xmmp-9 (Carinato et al., 2000) and nephrin (Barletta et al., 2003).

Analysis of Gene Expression

All tadpoles were examined for specific protein or mRNA expression under light/fluorescent microscopy, and representative individuals were photographed with a Nikon SMZ1500 stereomicroscope and Spot Insight Color digital camera and software. Area of gene expression in ipsilaterally nephrectomized and damage control kidneys was qualitatively compared with respective contralateral kidneys for all proteins and transcripts. For analysis of apoptosis, the number of cells expressing active Caspase 3 in the nephric region of all tadpoles was counted. An increase in active Caspase 3 at the site of nephrectomy was determined by observing more Caspase-positive cells on the ipsilateral side relative to the contralateral side. Damage control tadpoles and unoperated tadpoles were scored using the same criteria.

Renal Uptake Assay

BSA conjugated to Alexa Fluor 488 (BSA-AF488; 66 kDa; Molecular Probes), was injected into nephrectomized and unoperated sibling control tadpoles at 23 dpn as described previously (Zhou and Vize, 2004). All tadpoles were observed in real-time under light/fluorescent microscopy and representative individuals were photographed with a Nikon SMZ1500 stereomicroscope and Spot Insight Color digital camera and software before fixation.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

The authors thank Dr. Jonathan Henry for kindly providing the Xmmp-9 construct; Dr. Elizabeth Jones for kindly sharing the 3G8 antibody; Dr. Sara Lewis for assistance with statistical analysis; and Dr. Peter Vize for technical assistance with the renal uptake assays and helpful suggestions. The authors are also grateful to the members of the McLaughlin lab, the Levin lab and the regeneration community for many useful discussions. K.A.M. was funded by the NSF.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES