Skip to main content
NCBI home page
The American Journal of Pathology logoLink to The American Journal of Pathology
. 2022 Apr;192(4):604–612. doi: 10.1016/j.ajpath.2022.01.004

AKT Signaling Downstream of KGF Is Necessary and Sufficient for Blocking Cyclophosphamide Bladder Injury

Sridhar T Narla ∗, Daniel S Bushnell ∗, Joanne L Duara †, Carlton M Bates ∗,‡,∗
PMCID: PMC8961277  PMID: 35063403

Abstract

Keratinocyte growth factor (KGF) drives phosphorylated (activated) AKT (pAKT) in bladder urothelium, which correlates with cytoprotection from cyclophosphamide. The current study determined whether: i) KGF modifies AKT targets [B-cell lymphoma protein 2–associated agonist of cell death (BAD) and mammalian target of rapamycin complex (mTORC)-1] that could block apoptosis; ii) AKT signaling is required for KGF cytoprotection; iii) direct AKT activation drives cytoprotection; iv) co-administration of KGF and an AKT inhibitor blocks urothelial cytoprotection and AKT and AKT-target activation; and v) an AKT agonist prevents cyclophosphamide-induced urothelial apoptosis. Mice were given KGF and cyclophosphamide (or sham injury), and pBAD (readout of BAD inhibition) or p-p70S6k (pS6, readout of mTORC1 signaling) was assessed. KGF induced pBAD urothelial staining and prevented cyclophosphamide-induced loss of urothelial pS6 staining (likely stabilizing mTORC1 activity). Co-administration of KGF and AKT inhibitor blocked KGF-driven urothelial cytoprotection from cyclophosphamide and prevented pAKT, pBAD, and pS6 urothelial expression. Conversely, systemic AKT agonist blocked cyclophosphamide-induced urothelial apoptosis and induced pAKT, pBAD, and pS6, similar to KGF. Thus, the KGF–AKT signaling axis appeared to phosphorylate (suppress) BAD and prevent cyclophosphamide-induced loss of mTORC1 signaling, both of which likely suppress apoptosis. Additionally, AKT signaling was required for KGF-driven cytoprotection, and direct AKT activation was sufficient for blocking apoptosis. Thus, AKT may be a therapeutic target for blocking urothelial apoptosis from cyclophosphamide.

Cyclophosphamide, an alkylating agent used widely for treating solid tumors, leukemias, lymphomas, and other nononcologic diseases, can be bladder-toxic.1 Acrolein, a metabolite of cyclophosphamide, is excreted in the urine, where it can damage bladder urothelium, leading to early loss of outer superficial cells, followed by a wave of apoptosis of deeper intermediate and basal cells.2 From a clinical perspective, cyclophosphamide-induced bladder toxicity can manifest as hemorrhagic cystitis, ranging from microscopic hematuria and/or painful lower urinary tract symptoms, to severe, life-threatening bleeding. Long-term complications include bladder fibrosis, contractures, and dose-dependent increases in the risk for urothelial cancer,3, 4, 5, 6 with one study finding a nearly 15-fold risk in lymphoma survivors given cumulative doses of 50 g or more.7 All of these complications can occur even with current therapies, such as mesna (sodium-2-mercaptoethanesulfonate), which binds urinary acrolein, or hydration, which dilutes urinary acrolein.1,8,9

In 1997, Ulich et al10 reported that systemic administration of fibroblast growth factor (FGF)-7, also known as keratinocyte growth factor (KGF), in rats, prior to cyclophosphamide administration, led to better findings on urothelial histologic examination than in vehicle-treated mice (although the reason for the better outcomes was unclear). Narla et al2 then reported that systemic KGF activated its receptor [FGF receptor (FGFR)-2] in the urothelium, which largely abrogated intermediate and basal bladder urothelial cell apoptosis. The cytoprotection, in turn, led to faster and greater fidelity repair of the bladder urothelium. Moreover, the anti-apoptotic effects of KGF were correlated with increased phosphorylated (activated) AKT (pAKT) expression in the urothelium. Given that AKT activation can block apoptosis in many systems (reviewed by Song et al11), the current study investigated potential downstream targets of the KGF–AKT signaling axis that may drive urothelial cytoprotection. It also sought to determine whether AKT signaling downstream of KGF is required for the cytoprotective effect. Finally, the study assessed whether direct (KGF-independent) AKT activation was sufficient for blocking urothelial cell apoptosis from cyclophosphamide.

Materials and Methods

Mice

Female FVB/NJ mice aged 2 to 3 months were used for all experiments. The protocols of all of the proposed mouse experiments were approved by the Institutional Animal Care and Use Committee, University of Pittsburgh (Pittsburgh, PA), in accordance with the guidelines of the Association for Assessment and Accreditation of Laboratory Animal Care International.

Drugs/Chemicals Given to Mice

Mice were given 5 mg/kg s.c. injections of KGF (catalog number 251-KG-010; R&D Systems, Minneapolis, MN) dissolved in phosphate-buffered saline (PBS), or of PBS alone (vehicle), 24 hours before cyclophosphamide, sham injury, or no injury. To inhibit AKT signaling, 40 mg/kg i.p. injections of LY294002 (catalog number S1105; Selleckchem, Houston, TX) dissolved in 2% dimethyl sulfoxide (DMSO; catalog number D2438; Sigma-Aldrich, St. Louis, MO) or 2% DMSO alone (vehicle) were given concurrently with KGF/PBS and then every 12 hours until termination of the experiments. To activate AKT, 80 mg/kg i.p. injections of SC-79 (catalog number S7853; Selleckchem) dissolved in 5% DMSO (catalog number D2438; Sigma-Aldrich) 45% PEG300 (catalog number 8.07484; Sigma-Aldrich) and 50% saline (catalog number 7210; Ricca Chemicals, Arlington, TX) or vehicle (5% DMSO, 45% PEG300, 50% saline) were given at the times indicated in Results. To induce bladder injury with KGF and/or LY294002, 150 mg/kg i.p. injections of cyclophosphamide (catalog number C7397; Sigma-Aldrich) dissolved in PBS, or PBS alone (sham injury), were given. To induce bladder injury with SC-79, 225 mg/kg i.p. injections of cyclophosphamide or PBS were administered (given that the vehicle for SC-79 led to a small degree of urothelial protection at the 150 mg/kg dose of cyclophosphamide).

Histologic Examination, Immunofluorescence, and TUNEL Assays

Bladders were isolated and fixed in 4% paraformaldehyde and processed; tissues were embedded in paraffin and serially sectioned at 6 μm. For general histologic examination, hematoxylin and eosin (H&E) staining was performed. For immunofluorescence (IF), paraffin-embedded sections were dewaxed and subjected to antigen retrieval in a pressure cooker for 15 minutes in Tris-EDTA pH 9.0 buffer. Samples were then blocked with normal donkey serum for 1 hour at room temperature. Sections were then incubated overnight at 4°C with the following primary antibodies: uroplakin (UPK)-3a (superficial cell and intermediate cell subset marker) at 1:200 [catalog number sc-33570, Research Resource Identifier (RRID):AB_2213486; Santa Cruz Biotechnology, Dallas, TX], pAKT at 1:100 (catalog number 4060, RRID:AB_2315049; Cell Signaling Technology, Beverly, MA), phosphorylated B-cell lymphoma protein (BCL)-2–associated agonist of cell death (pBAD) at 1:100 (catalog number AB28824, RRID:AB_725616; Abcam, Waltham, MA), and p-p70S6K (pS6) (catalog number 4858S, RRID:AB_916156; Cell Signaling Technology). After being washed in PBS, slides were incubated with the following secondary antibodies: Alexa Fluor 594 (catalog number A-21207, RRID:AB_141637; Thermo Fisher Scientific, Waltham, MA), Alexa Fluor 488 (catalog number A-21202, RRID:AB_141607; Thermo Fisher), all at 1:500 for 2 hours at room temperature, followed by washes. Nuclei were stained with DAPI (catalog number D1306; Sigma-Aldrich). To assay for apoptosis, terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assays were performed with the ApopTag Plus in situ Apoptosis Fluorescein Detection kit (catalog number S7111; EMD Millipore, Burlington, MA) according to the manufacturer's protocol. Slides were imaged with a Leica DM2500 fluorescence microscope (Leica Microsystems, Buffalo Grove, IL) or a Zeiss LSM 710 confocal microscope (Carl Zeiss, Thornwood, NY).

Results

KGF Treatment Appears To Modify Activity of BAD and mTORC1, Two AKT Targets that Regulate Apoptosis

Systemic KGF induces pAKT in bladder urothelium that is correlated with cytoprotection with cyclophosphamide.2 Therefore, the study assessed readouts of two known targets of AKT, BAD and mammalian target of rapamycin complex (mTORC)-1, that could be involved in the cytoprotective effects. AKT phosphorylates BAD (pBAD), which should suppress apoptosis,12,13 and phosphorylates tuberous sclerosis protein (TSC)-2, activating mTORC1 signaling, ultimately leading to phosphorylation (activation) of p70S6K (p-p70S6K or pS6) and suppression of apoptosis.14, 15, 16, 17, 18 Thus, if KGF administration induced urothelial pBAD and/or pS6 staining, this would likely indicate that KGF acts through AKT and targets BAD and mTORC1 to suppress apoptosis from cyclophosphamide.

First, PBS vehicle or KGF was administered by s.c. injection, sham injury was induced by i.p. injection of PBS 24 hours after PBS/KGF, and pBAD or pS6 was examined at 30, 48, and 96 hours after PBS/KGF administration (Figure 1A). Minimal cytoplasmic pBAD staining was noted in all PBS-treated mice at all time points (Figure 1B). With KGF, moderately increased staining was observed at 30 hours (not shown), robust expression was seen at 48 hours (Figure 1C), and limited staining was noted at 96 hours (not shown). Notably, this temporal pattern of cytoplasmic pBAD staining with KGF treatment mimicked previous findings on pAKT staining with KGF treatment.2 Regarding pS6, similar robust cytoplasmic staining was noted in uninjured urothelium with PBS or KGF treatment at all time points (Figure 1, D and E).

Figure 1.

Figure 1

Open in a new tab

Representative images showing effects of KGF on AKT targets, BAD and mTORC1, 48 hours after treatment in sham-injured and cyclophosphamide (CPP)-injured mice. A: Timeline of the experimental design. B and C: In sham-injured mice, IF for pBAD (red), a readout of BAD suppression, shows minimal cytoplasmic urothelial staining with PBS treatment (B, arrowhead), but robust cytoplasmic signal with KGF treatment (C, arrowhead). D and E: In sham-injured mice IF for pS6 (red), a readout of mTORC1 activity shows cytoplasmic urothelial staining that appears relatively equivalent to that with PBS treatment (D, arrowhead) and KGF treatment (E, arrowhead). F and G: In cyclophosphamide-injured mice, IF for pBAD (red) shows a pattern similar to that in sham-injured mice with minimal cytoplasmic urothelial staining with PBS treatment (F, arrowhead), but strong signal with KGF treatment (G, arrowhead). H and I: In cyclophosphamide-injured mice, IF for pS6 (red) shows a loss of urothelial staining with PBS treatment (H, arrowhead), but relatively preserved signal with KGF treatment (I, arrowhead). DAPI (blue). Dashed line indicates urothelial border with stroma. ∗The 48-hour time point is shown. Scale bars = 50 μm. L, lumen.

Next, mice were treated with KGF or PBS, injured with i.p. injections of cyclophosphamide 24 hours later, and assessed for pBAD and pS6 staining 48 hours after KGF/PBS treatment. Similar to findings in sham-injured mice, minimal cytoplasmic pBAD staining was noted with PBS treatment (Figure 1F) versus robust signal with KGF (Figure 1G). Interestingly, PBS-treated mice subjected to cyclophosphamide had a sharp decrease in cytoplasmic urothelial pS6 staining (Figure 1H), while KGF-treated mice had relative preservation of urothelial pS6 staining (Figure 1I). PBS-treated mice injured with cyclophosphamide had a loss of pS6 staining at 30 and 36 hours after treatment (6 and 12 hours after cyclophosphamide), whereas KGF-treated mice had preservation of staining at those time points (not shown). The cyclophosphamide-induced loss of pS6K in urothelium is consistent with a report that acrolein (cyclophosphamide toxic metabolite) led to a loss of pS6K in cultured endothelial cells at as early as 4 hours.19 Taken together, KGF/AKT appeared to drive urothelial pBAD expression (irrespective of injury) and prevent the loss of pS6 (mTORC1 activity) with injury, both of which would be expected to suppress apoptosis.

AKT Is Required for KGF-induced Urothelial Cytoprotection from Cyclophosphamide

To test whether AKT activation is necessary for KGF-driven cytoprotection from cyclophosphamide, the effects of co-treatment with LY294002, a potent AKT inhibitor (AKTi),20 with KGF, were tested in mice injured with cyclophosphamide. To first test the efficacy of AKT inhibition in uninjured mice, s.c. KGF or PBS was co-administered with AKTi, or its vehicle (2% DMSO) was administered by i.p. injection, after which additional AKTi or vehicle was administered by i.p. injection every 12 hours until bladders were harvested at 36 hours after KGF/PBS treatment (Supplemental Figure S1A). As expected, mice given PBS + vehicle (Supplemental Figure S1B) or PBS + AKTi (not shown) had virtually no pAKT urothelial staining by IF, while mice given KGF + vehicle had robust urothelial pAKT signal (Supplemental Figure S1C). Mice given KGF + AKTi, however, had a marked reduction in pAKT urothelial staining (Supplemental Figure S1D), consistent with the inhibitor largely suppressing KGF-induced AKT activation.

The effects of AKTi on KGF-driven cytoprotection from cyclophosphamide were assessed next. The same dosing strategy as in the previous paragraph was used for PBS/KGF and AKTi/vehicle, except that cyclophosphamide was also administered 24 hours after KGF/PBS, followed by bladder harvest 36 hours after KGF/PBS treatment (12 hours after cyclophosphamide treatment) (Figure 2A). As expected (based on published work with PBS alone2), injured mice given PBS + vehicle had significant urothelial injury with sloughing cells and denuding by H&E staining (Figure 2B). In addition, IF revealed minimal pAKT urothelial staining (Figure 2C), and TUNEL staining showed marked urothelial apoptosis (Figure 2D) in injured mice treated with PBS + vehicle. Finally, IF for pBAD showed virtually no urothelial cytoplasmic staining (Figure 2E), and IF for pS6 revealed loss of most urothelial expression (Figure 2F) in injured mice given PBS + vehicle [similar to the observations with PBS alone (Figure 1)]. The H&E, pAKT TUNEL, pBAD, and pS6 staining patterns were similar in injured mice given PBS + AKTi (not shown). As expected, injured mice given KGF + vehicle had largely intact urothelium by H&E staining (Figure 2G), induction of urothelial pAKT expression (Figure 2H), minimal urothelial apoptosis (Figure 2I), induction of urothelial pBAD staining (Figure 2J), and restoration of urothelial pS6 expression (Figure 2K), similar to the findings in injured mice given KGF alone (Figure 1) (and as reported by Narla et al2). Injured mice treated with KGF + AKTi, however, had evidence of more urothelial injury by H&E staining (Figure 2L) than did KGF + vehicle–treated mice, albeit less than with PBS + vehicle. KGF + vehicle–treated and injured mice had largely reduced (but not absent) pAKT staining by IF (Figure 2M) (similar to uninjured mice in Supplemental Figure S1) and breakthrough TUNEL+ urothelial cells (Figure 2N), albeit fewer than in injured mice given PBS + vehicle. Injured mice given KGF + AKTi also had a significant reduction in cytoplasmic urothelial pBAD expression versus KGF + vehicle–treated mice (albeit more staining than in PBS + vehicle–treated mice), consistent with partial blockade of KGF-induced pBAD expression (Figure 2O). Finally, injured mice treated with KGF + AKTi had a significant loss of pS6 urothelial staining, consistent with partial blockade of KGF-driven restoration of staining (Figure 2P). Together, these results suggest that KGF acts through AKT to alter the BAD and mTORC1 pathways to induce cytoprotection from cyclophosphamide.

Figure 2.

Figure 2

Open in a new tab

Representative images indicating the effect of the AKT inhibitor (AKTi, LY294002) on KGF-driven cytoprotection after cyclophosphamide (CPP) at 36 hours after treatment. A: Timeline of the experimental design. B–F: Images from injured mice treated with PBS or vehicle for AKTi (PBS + Veh). B: H&E-stained image showing significant urothelial injury (arrowhead) with denuding and sloughing. C: IF for pAKT (red) showing limited staining (arrowhead). D: TUNEL staining (green) reveals many apoptotic urothelial cells (arrowhead). E: IF for pBAD (red) showing limited urothelial staining (arrowhead). F: IF for pS6 (red) showing minimal cytoplasmic urothelial staining (arrowhead), consistent with loss of baseline mTORC1 activity. G–K: Images from injured mice treated with KGF and vehicle for AKTi (KGF + Veh). G: H&E-stained image showing largely intact urothelium (arrowhead). H: IF for pAKT (red) showing induction of urothelial staining (arrowhead). I: TUNEL staining (green) showing virtually no urothelial apoptosis (arrowhead). J: IF for pBAD (red) showing induction of cytoplasmic urothelial staining (arrowhead). K: IF for pS6 (red) revealing restoration of cytoplasmic urothelial expression (arrowhead). L–P: Images from injured mice treated with KGF and AKTi (KGF + AKTi). L: H&E-stained image showing urothelial injury with sloughing and denuding (arrowhead), which is not as severe as PBS-treated mice. M: IF for pAKT (red) showing largely diminished staining compared to KGF + Veh with some breakthrough staining (arrowhead). N: TUNEL staining (green) showing many apoptotic urothelial cells (arrowhead) relative to mice given KGF + Veh, but fewer than in mice treated with PBS + Veh. O: IF for pBAD (red) showing a significant reduction in overall expression with regions of ongoing staining (concave arrowhead) and others with no staining (arrowhead). P: IF for pS6 (red) revealing an overall decrease in expression versus KGF + Veh–treated mice, with some regions of urothelial staining (concave arrowhead) and others with no signal (arrowhead). DAPI (blue). Dashed line indicates urothelial border with stroma. ∗Images from PBS + AKTi are not shown. Scale bars = 50 μm. L, lumen; Sham, sham injury.

Direct AKT Activation Is Sufficient for Urothelial Cytoprotection against Cyclophosphamide

Given that the AKTi blunted the cytoprotective actions of KGF with cyclophosphamide, the study next assessed whether direct activation of AKT alone would be sufficient for blocking apoptosis. To that end, it was determined whether SC-79,21, 22, 23, 24 a potent AKT agonist (AKTa), would offer cytoprotection similar to that with KGF. First, the temporal actions of AKTa were assessed in uninjured mice by urothelial pAKT staining. AKTa 80 mg/kg or vehicle (5% DMSO, 45% PEG300, 50% saline) was administered by i.p. injection and bladders were taken at 6, 12, 24, and 48 hours later (Figure 3A). As expected, mice administered vehicle had no appreciable urothelial pAKT expression at any time point (Figure 3, B–E). Mice administered AKTa also had no obvious urothelial pAKT 6 hours after AKTa administration (Figure 3F); however, significant pAKT urothelial expression was observed at 12 hours (Figure 3G) and 24 hours (Figure 3H) after AKTa. By 48 hours after AKTa administration, pAKT urothelial staining was essentially back to that at baseline (Figure 3I). Thus AKTa induces robust urothelial pAKT from 12 to 24 hours after injection.

Figure 3.

Figure 3

Open in a new tab

Representative images showing the temporal expression of pAKT after treatment with the AKT agonist (AKTa, SC-79). A: Timeline of the experimental design (Veh, vehicle for AKTa). B–E: In mice administered vehicle, IF for pAKT (red) shows virtually no urothelial expression at 6 (B), 12, (C), 24 (D), or 48 (E) hours after administration. F–I: In mice administered AKTa, pAKT urothelial staining (red, arrowheads) was minimal at 6 hours (F), robust at 12 (G) and 24 (H) hours, and then nearly back to baseline at 48 hours (I) after administration. Note increased stromal pAKT staining apparent at 12 hours after administration (G, concavearrowhead). DAPI (blue). Dashed line indicates urothelial border with stroma. Scale bars = 50 μm. L, lumen.

The study next sought to determine whether the administration of AKTa was able to block cyclophosphamide-induced urothelial injury. To that end, mice were administered AKTa 80 mg/kg or vehicle by i.p. injection and then cyclophosphamide by i.p. injection 12 hours later, and bladders were harvested 24 hours after AKTa/vehicle (12 hours after cyclophosphamide) (Figure 4A). As expected, vehicle-treated mice had significant urothelial injury with denuding and sloughing cells by H&E staining (Figure 4B). Injured vehicle-treated mice had many TUNEL+ urothelial cells (Figure 4C), consistent with apoptosis. IF for Upk3 revealed regional loss of staining (Figure 4D) consistent with loss of superficial and/or intermediate cells. In contrast, injured AKTa-treated mice had largely intact urothelium, with few sloughing cells by H&E staining (Figure 4E), virtually no urothelial apoptosis by TUNEL staining (Figure 4F), and robust luminal uroplakin expression (Figure 4G), consistent with relative preservation of cell layers. Thus, systemic administration of AKTa appears sufficient for replicating the urothelial cytoprotective effects of KGF with cyclophosphamide.

Figure 4.

Figure 4

Open in a new tab

Representative images from cyclophosphamide (CPP)-injured mice showing urothelial cytoprotection and readouts of downstream targets of the AKT agonist (AKTa, SC-79) 24 hours after treatment. A: Timeline of the experimental design (Veh, vehicle for AKTa). B–G: Images from injured vehicle-treated mice (B–D) and injured AKTa-treated mice (E–G) showing injury patterns. B: H&E-stained image showing vehicle-treated mice with significant urothelial injury (arrowhead), and sloughing and denuding cells. C: TUNEL staining (green) showing many TUNEL+ apoptotic urothelial cells (arrowhead) in vehicle-treated mice. D: IF for Upk3 (green) showing regional losses of urothelial staining (arrowhead) in vehicle-treated mice. E: H&E-stained image showing AKTa-treated mice with largely intact urothelium (arrowhead). F: TUNEL staining (green) revealing virtually no urothelial apoptosis in AKTa-treated mice (arrowhead). G: IF for Upk3 (green) showing relatively intact expression in AKTa-treated mice (arrowhead), consistent with preservation of urothelial cell layers. H–M: Images from vehicle-treated mice (H–J) and injured AKTa-treated mice (K–M) revealing expression of pAKT and readouts of its targets (pBAD and pS6). H: IF for pAKT (red) showing limited urothelial pAKT staining (arrowhead) in vehicle-treated mice. I: IF for pBAD (red) revealing limited cytoplasmic urothelial expression (arrowhead) in injured mice treated with vehicle. J: IF for pS6 (red) showing diminished cytoplasmic urothelial staining (arrowhead) in injured mice treated with vehicle. K: IF for pAKT (red) showing induction of staining in the urothelium (arrowhead) and stroma (concave arrowhead) in AKTa-treated mice. L: IF for pBAD (red) also showing induction of expression (arrowhead) in injured AKTa-treated mice. M: IF for pS6 (red) revealing restoration of cytoplasmic urothelial expression (arrowhead) in injured AKTa-treated mice. DAPI (blue). Dashed line indicates urothelial border with stroma. Scale bars = 50 μm. L, lumen.

To determine whether AKTa acted in a manner similar to KGF on AKT and its targets in the urothelium with cyclophosphamide, IF for pAKT, pBAD, and pS6 was performed. As shown, injured vehicle-treated mice had minimal urothelial pAKT (Figure 4H) and pBAD expression (Figure 4I) and significant loss of urothelial cytoplasmic pS6 staining (Figure 4J). In contrast, injured AKTa-treated mice had significant induction of urothelial pAKT (Figure 4K) and pBAD staining (Figure 4L) expression, and restoration of urothelial pS6 expression (Figure 4M). Thus, the urothelial cytoprotective effects of AKTa appear to act through AKT that then modulates BAD and mTORC1 activity, similar to KGF.

Discussion

This study found that the KGF–AKT signaling axis in urothelium appears to modify signaling of at least two AKT targets, BAD and mTORC1. AKT phosphorylates BAD (pBAD), causing its redistribution to the cytosol from mitochondria, relieving BAD suppression of anti-apoptotic BCL-XL and BCL2 proteins.12,13 Moreover, KGF and AKTa (SC-79) were both found to induce urothelial pBAD staining in a temporal manner that mimics the induction of urothelial pAKT expression (see Narla et al2 and Direct AKT Activation Is Sufficient for Urothelial Cytoprotection against Cyclophosphamide). Thus, targeting BAD might provide another therapeutic alternative to KGF for blocking cyclophosphamide-induced urothelial apoptosis (although no small-molecule inhibitors are currently known to exist). AKT also has been reported to phosphorylate TSC2, thereby de-repressing mTORC1 signaling, and mTORC1 then stimulates the translation of the anti-apoptotic protein myeloid leukemia cell–differentiation protein (MCL)-1 and phosphorylates (activates) p70S6K (pS6), which, in turn, phosphorylates BAD.14, 15, 16, 17, 18 In the present study, while pS6 urothelial staining did not increase with KGF in uninjured mice, cyclophosphamide injury alone led to a loss of cytoplasmic urothelial staining, which was reversed by KGF and AKTa. As noted in KGF Treatment Appears To Modify Activity of BAD and mTORC1, Two AKT Targets that Regulate Apoptosis, the cyclophosphamide-induced loss of pS6K in urothelium was consistent with a report that acrolein (cyclophosphamide toxic metabolite) leads to a loss of pS6K in cultured endothelial cells as early as 4 hours after injury.19 Thus, the preservation of pS6 staining (mTORC1 signaling) also likely contributes to the cytoprotective effects of KGF and AKTa in the urothelium. While beyond the scope of this study, direct mTORC1 activators, such as MHY1485, have been shown to drive mTORC1 activity in mice25 and could provide another means of blocking cyclophosphamide-induced urothelial apoptosis. Taken together, KGF/AKTa-induced activation of pBAD and preservation of pS6 urothelial staining likely indicate that KGF/AKT is acting, at least in part, through the BAD and mTORC1 pathways to suppress cyclophosphamide-induced apoptosis.

The findings from this study also strongly support the concept that the KGF-driven urothelial cytoprotection is dependent on intact AKT signaling, as noted by the studies with the AKTi (LY294002). Moreover, co-administration of KGF and the AKTi led to abrogation of cytoprotection as well as loss of KGF-mediated urothelial pAKT, pBAD, and pS6 staining. Furthermore, the findings from these studies strongly support the concept that direct AKT activation (independent of KGF) with AKTa (SC-79) is sufficient for inducing urothelial cytoprotection from cyclophosphamide and mimics the KGF-driven changes in urothelial pAKT, pBAD, and pS6 staining. Many reports have demonstrated that AKT signaling can block apoptosis (see review by Song et al11), including downstream of FGF signaling in nonbladder tissues.26, 27, 28, 29 Interestingly, others have shown that cyclophosphamide alone leads to small increases in pAKT expression in bladder urothelium, stroma, and inflammatory cells within 24 hours, and that intravesicular administration of an AKTi improved bladder capacity, void volume, and intercontraction void interval, suggesting that increased pAKT signaling was harmful.30 It is possible that the benefits of blocking of AKT signaling are specific to blockade in the muscle or inflammatory cells instead of the urothelium. If that is the case, then the use of the AKTa (which appeared to also drive stromal pAKT staining in this study) may have some negative effects on bladder function. Overall, however, it appears that the significant up-regulation of AKT signaling in the urothelium via KGF or AKTa is beneficial in blocking cyclophosphamide-induced apoptotic urothelial cell injury.

With both SC-79 (AKTa) and KGF, the activation of AKT (as noted by urothelial pAKT expression) is transient. Moreover, the activation of urothelial AKT appears to occur slightly sooner after the administration of SC-79 and with a narrower window (12 to 24 hours after administration) as opposed to KGF (15 to 48 hours after administration2). The likely reason for the earlier AKT activation with the AKTa is that there are fewer steps to activating AKT with SC-79, which binds directly to the PH domain of AKT22 versus KGF, which binds to FGFR2 that triggers phosphorylation of FGFR substrate 2α that then leads to phosphorylation of AKT.2 The reason for the shorter window of activity of SC-79 versus KGF is unclear. While optimal timing and dosing of either compound in humans (compared to how the drugs have been used in mice here and as previously reported2) are unclear, the onset of hemorrhagic cystitis in humans reportedly occurs within 24 to 48 hours of cyclophosphamide administration,31 similar to findings in mice (see Results and Narla et al2). Moreover, another group performed cystoscopies and bladder biopsies in patients 24 hours after the administration of ifosfamide (metabolized to acrolein like cyclophosphamide), and noted urothelial injury in most patients.32 Thus the timing of urothelial injury after cyclophosphamide administration in mice appears similar to that seen in humans, which may be useful for informing the timing of KGF or AKTa administration if either is approved for use in humans.

The results with the AKTa in this study may be relevant to many types of epithelial injury, including the bladder and elsewhere. Patients with ataxia-telangiectasia have a high risk for lymphoma and, in turn, often develop severe hemorrhagic cystitis with cyclophosphamide33 and, as such, would be good potential candidates for AKTa treatment to mitigate bladder injury. In addition, patients subjected to high cumulative doses of cyclophosphamide have a relatively high risk for long-term urothelial cancer7 and, as such, could also potentially benefit from AKTa therapy. Preclinical and clinical studies have shown that KGF/FGFR2 signaling reduces injury and drives regeneration in many epithelial tissues after the introduction of noxious stimuli.10,34, 35, 36, 37, 38, 39 Moreover, exogenous KGF has been reported to ameliorate toxin- or radiation-mediated injury in oral, retinal, alveolar, and intestinal epithelia.34, 35, 36, 37, 38, 39 Thus, direct activation of AKT, such as with SC-79, could provide an alternative to KGF for blocking many types of injury in a wide array of epithelial tissues. One potential problem with systemic administration of an AKTa with cyclophosphamide is that the agonist would likely block cyclophosphamide-driven tumor apoptosis (this is not the case with KGF, as lymphoma cells do not express the epithelial isoform of FGFR2, which is the receptor for KGF). One means of avoiding off-target systemic effects could be to directly instill SC-79 into the bladder, as it is lipophilic and has a low molecular weight, similar to the AKTi given by direct bladder instillation by others.30 Another potential limitation of systemic administration of the AKTa is that it would likely need to be given prior to the administration of cyclophosphamide, given that the window of pAKT staining is 12 to 24 hours after systemic infusion, and that the cyclophosphamide-induced urothelial apoptosis starts at 4 hours and peaks at 24 hours after administration.2 Whether direct bladder infusion of AKTa would result in earlier pAKT urothelial staining (and thus potentially be useful after cyclophosphamide administration) remains to be seen.

In conclusion, the findings from this study support a key role for AKT and its targets in acting downstream of KGF to block cyclophosphamide-induced urothelial apoptosis. Moreover, direct activation of AKT may be a potential therapy alternative to KGF in blocking urothelial cyclophosphamide-induced injury, as well as in mitigating injury from a variety of noxious stimuli in many other epithelial tissues.

Acknowledgments

We thank Drs. Rannar Airik, Sunder Sims-Lucas, Jacqueline Ho, and Catherine Forster for helpful discussions about the experimental design and interpreting the results.

Footnotes

Supported by NIH grant R01 DK095748 (C.M.B.).

Disclosures: None declared.

Supplemental material for this article can be found at https://doi.org/10.1016/j.ajpath.2022.01.004.

Supplemental Data

Supplemental Figure S1.

Supplemental Figure S1

Open in a new tab

Representative images of the effects of AKT inhibitor (AKTi, LY294002) on KGF-driven urothelial pAKT expression in mice 36 hours after treatment. A: Timeline of the experimental design. B: IF for pAKT (red) shows no urothelial expression (arrowhead) in uninjured mice treated with PBS and vehicle for AKTi (PBS + Veh). C: IF for pAKT (red) shows robust induction of urothelial staining (arrowhead) in mice treated with KGF and vehicle for AKTi (KGF + Veh). D: IF for pAKT (red) shows a sharp reduction in pAKT staining (arrowhead) in mice co-treated with KGF + AKTi. DAPI (blue). Dashed line indicates urothelial border with stroma. ∗Images from PBS + AKTi are not shown. Scale bars = 50 μm. L = lumen.

References

  • 1.Korkmaz A., Topal T., Oter S. Pathophysiological aspects of cyclophosphamide and ifosfamide induced hemorrhagic cystitis; implication of reactive oxygen and nitrogen species as well as PARP activation. Cell Biol Toxicol. 2007;23:303–312. doi: 10.1007/s10565-006-0078-0. [DOI] [PubMed] [Google Scholar]
  • 2.Narla S.T., Bushnell D.S., Schaefer C.M., Nouraie M., Bates C.M. Keratinocyte growth factor reduces injury and leads to early recovery from cyclophosphamide bladder injury. Am J Pathol. 2020;190:108–124. doi: 10.1016/j.ajpath.2019.09.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Baker G.L., Kahl L.E., Zee B.C., Stolzer B.L., Agarwal A.K., Medsger T.A., Jr. Malignancy following treatment of rheumatoid arthritis with cyclophosphamide. Long-term case-control follow-up study. Am J Med. 1987;83:1–9. doi: 10.1016/0002-9343(87)90490-6. [DOI] [PubMed] [Google Scholar]
  • 4.Kaldor J.M., Day N.E., Kittelmann B., Pettersson F., Langmark F., Pedersen D., et al. Bladder tumours following chemotherapy and radiotherapy for ovarian cancer: a case-control study. Int J Cancer. 1995;63:1–6. doi: 10.1002/ijc.2910630102. [DOI] [PubMed] [Google Scholar]
  • 5.Shirai T. Etiology of bladder cancer. Semin Urol. 1993;11:113–126. [PubMed] [Google Scholar]
  • 6.Vlaovic P., Jewett M.A. Cyclophosphamide-induced bladder cancer. Can J Urol. 1999;6:745–748. [PubMed] [Google Scholar]
  • 7.Travis L.B., Curtis R.E., Glimelius B., Holowaty E.J., Van Leeuwen F.E., Lynch C.F., et al. Bladder and kidney cancer following cyclophosphamide therapy for non-Hodgkin's lymphoma. J Natl Cancer Inst. 1995;87:524–530. doi: 10.1093/jnci/87.7.524. [DOI] [PubMed] [Google Scholar]
  • 8.Moghe A., Ghare S., Lamoreau B., Mohammad M., Barve S., McClain C., Joshi-Barve S. Molecular mechanisms of acrolein toxicity: relevance to human disease. Toxicol Sci. 2015;143:242–255. doi: 10.1093/toxsci/kfu233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Matz E.L., Hsieh M.H. Review of advances in uroprotective agents for cyclophosphamide- and ifosfamide-induced hemorrhagic cystitis. Urology. 2017;100:16–19. doi: 10.1016/j.urology.2016.07.030. [DOI] [PubMed] [Google Scholar]
  • 10.Ulich T.R., Whitcomb L., Tang W., O'Conner Tressel P., Tarpley J., Yi E.S., Lacey D. Keratinocyte growth factor ameliorates cyclophosphamide-induced ulcerative hemorrhagic cystitis. Cancer Res. 1997;57:472–475. [PubMed] [Google Scholar]
  • 11.Song G., Ouyang G., Bao S. The activation of Akt/PKB signaling pathway and cell survival. J Cell Mol Med. 2005;9:59–71. doi: 10.1111/j.1582-4934.2005.tb00337.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Bui N.L., Pandey V., Zhu T., Ma L., Basappa, Lobie P.E. Bad phosphorylation as a target of inhibition in oncology. Cancer Lett. 2018;415:177–186. doi: 10.1016/j.canlet.2017.11.017. [DOI] [PubMed] [Google Scholar]
  • 13.Igney F.H., Krammer P.H. Death and anti-death: tumour resistance to apoptosis. Nat Rev Cancer. 2002;2:277–288. doi: 10.1038/nrc776. [DOI] [PubMed] [Google Scholar]
  • 14.Ban K., Kozar R.A. Protective role of p70S6K in intestinal ischemia/reperfusion injury in mice. PLoS One. 2012;7:e41584. doi: 10.1371/journal.pone.0041584. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Fulda S. Synthetic lethality by co-targeting mitochondrial apoptosis and PI3K/Akt/mTOR signaling. Mitochondrion. 2014;19 Pt A:85–87. doi: 10.1016/j.mito.2014.04.011. [DOI] [PubMed] [Google Scholar]
  • 16.Harada H., Andersen J.S., Mann M., Terada N., Korsmeyer S.J. p70S6 kinase signals cell survival as well as growth, inactivating the pro-apoptotic molecule BAD. Proc Natl Acad Sci U S A. 2001;98:9666–9670. doi: 10.1073/pnas.171301998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Hosoi H., Dilling M.B., Shikata T., Liu L.N., Shu L., Ashmun R.A., Germain G.S., Abraham R.T., Houghton P.J. Rapamycin causes poorly reversible inhibition of mTOR and induces p53-independent apoptosis in human rhabdomyosarcoma cells. Cancer Res. 1999;59:886–894. [PubMed] [Google Scholar]
  • 18.Mills J.R., Hippo Y., Robert F., Chen S.M., Malina A., Lin C.J., Trojahn U., Wendel H.G., Charest A., Bronson R.T., Kogan S.C., Nadon R., Housman D.E., Lowe S.W., Pelletier J. mTORC1 promotes survival through translational control of Mcl-1. Proc Natl Acad Sci U S A. 2008;105:10853–10858. doi: 10.1073/pnas.0804821105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Lemaitre V., Dabo A.J., D'Armiento J. Cigarette smoke components induce matrix metalloproteinase-1 in aortic endothelial cells through inhibition of mTOR signaling. Toxicol Sci. 2011;123:542–549. doi: 10.1093/toxsci/kfr181. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Azaro A., Rodon J., Calles A., Brana I., Hidalgo M., Lopez-Casas P.P., Munoz M., Westwood P., Miller J., Moser B.A., Ohnmacht U., Bumgardner W., Benhadji K.A., Calvo E. A first-in-human phase I trial of LY2780301, a dual p70 S6 kinase and Akt inhibitor, in patients with advanced or metastatic cancer. Invest New Drugs. 2015;33:710–719. doi: 10.1007/s10637-015-0241-7. [DOI] [PubMed] [Google Scholar]
  • 21.Jing Z.T., Liu W., Xue C.R., Wu S.X., Chen W.N., Lin X.J., Lin X. AKT activator SC79 protects hepatocytes from TNF-alpha-mediated apoptosis and alleviates d-Gal/LPS-induced liver injury. Am J Physiol Gastrointest Liver Physiol. 2019;316:G387–G396. doi: 10.1152/ajpgi.00350.2018. [DOI] [PubMed] [Google Scholar]
  • 22.Jo H., Mondal S., Tan D., Nagata E., Takizawa S., Sharma A.K., Hou Q., Shanmugasundaram K., Prasad A., Tung J.K., Tejeda A.O., Man H., Rigby A.C., Luo H.R. Small molecule-induced cytosolic activation of protein kinase Akt rescues ischemia-elicited neuronal death. Proc Natl Acad Sci U S A. 2012;109:10581–10586. doi: 10.1073/pnas.1202810109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Wang S.Q., Yang X.Y., Yu X.F., Cui S.X., Qu X.J. Knockdown of IGF-1R triggers viral RNA sensor MDA5- and RIG-I-mediated mitochondrial apoptosis in colonic cancer cells. Mol Ther Nucleic Acids. 2019;16:105–117. doi: 10.1016/j.omtn.2019.02.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Yu J., Luo Y., Wen Q. Nalbuphine suppresses breast cancer stem-like properties and epithelial-mesenchymal transition via the AKT-NFkappaB signaling pathway. J Exp Clin Cancer Res. 2019;38:197. doi: 10.1186/s13046-019-1184-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Chen Z., Li L., Wu W., Liu Z., Huang Y., Yang L., Luo Q., Chen J., Hou Y., Song G. Exercise protects proliferative muscle satellite cells against exhaustion via the Igfbp7-Akt-mTOR axis. Theranostics. 2020;10:6448–6466. doi: 10.7150/thno.43577. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Chang H.L., Sugimoto Y., Liu S., Wang L.S., Huang Y.W., Ye W., Lin Y.C. Keratinocyte growth factor (KGF) regulates estrogen receptor-alpha (ER-alpha) expression and cell apoptosis via phosphatidylinositol 3-kinase (PI3K)/Akt pathway in human breast cancer cells. Anticancer Res. 2009;29:3195–3205. [PubMed] [Google Scholar]
  • 27.Qiu W., Leibowitz B., Zhang L., Yu J. Growth factors protect intestinal stem cells from radiation-induced apoptosis by suppressing PUMA through the PI3K/AKT/p53 axis. Oncogene. 2010;29:1622–1632. doi: 10.1038/onc.2009.451. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Bao S., Wang Y., Sweeney P., Chaudhuri A., Doseff A.I., Marsh C.B., Knoell D.L. Keratinocyte growth factor induces Akt kinase activity and inhibits Fas-mediated apoptosis in A549 lung epithelial cells. Am J Physiol Lung Cell Mol Physiol. 2005;288:L36–L42. doi: 10.1152/ajplung.00309.2003. [DOI] [PubMed] [Google Scholar]
  • 29.Cai Y., Wang W., Qiu Y., Yu M., Yin J., Yang H., Mei J. KGF inhibits hypoxia-induced intestinal epithelial cell apoptosis by upregulating AKT/ERK pathway-dependent E-cadherin expression. Biomed Pharmacother. 2018;105:1318–1324. doi: 10.1016/j.biopha.2018.06.091. [DOI] [PubMed] [Google Scholar]
  • 30.Arms L., Vizzard M.A. Role for pAKT in rat urinary bladder with cyclophosphamide (CYP)-induced cystitis. Am J Physiol Ren Physiol. 2011;301:F252–F262. doi: 10.1152/ajprenal.00556.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.George P. Haemorrhagic cystitis and cyclophosphamide. Lancet. 1963;2:942. doi: 10.1016/s0140-6736(63)90653-6. [DOI] [PubMed] [Google Scholar]
  • 32.Lima M.V., Ferreira F.V., Macedo F.Y., de Castro Brito G.A., Ribeiro R.A. Histological changes in bladders of patients submitted to ifosfamide chemotherapy even with mesna prophylaxis. Cancer Chemother Pharmacol. 2007;59:643–650. doi: 10.1007/s00280-006-0307-5. [DOI] [PubMed] [Google Scholar]
  • 33.Kaymaz N.E., Kupeli S., Yalcin B., Buyukpamukcu M. Hemorrhagic cystitis in a child with Hodgkin lymphoma and ataxia-telangiectasia after cyclophosphamide. Pediatr Blood Cancer. 2009;53:516. doi: 10.1002/pbc.22072. [DOI] [PubMed] [Google Scholar]
  • 34.Dorr W., Spekl K., Farrell C.L. The effect of keratinocyte growth factor on healing of manifest radiation ulcers in mouse tongue epithelium. Cell Prolif. 2002;35(Suppl 1):86–92. doi: 10.1046/j.1365-2184.35.s1.9.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Farrell C.L., Rex K.L., Chen J.N., Bready J.V., DiPalma C.R., Kaufman S.A., Rattan A., Scully S., Lacey D.L. The effects of keratinocyte growth factor in preclinical models of mucositis. Cell Prolif. 2002;35(Suppl 1):78–85. doi: 10.1046/j.1365-2184.35.s1.8.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Hu H., Hao L., Tang C., Zhu Y., Jiang Q., Yao J. Activation of KGFR-Akt-mTOR-Nrf2 signaling protects human retinal pigment epithelium cells from ultra-violet. Biochem Biophys Res Commun. 2018;495:2171–2177. doi: 10.1016/j.bbrc.2017.12.078. [DOI] [PubMed] [Google Scholar]
  • 37.Khan W.B., Shui C., Ning S., Knox S.J. Enhancement of murine intestinal stem cell survival after irradiation by keratinocyte growth factor. Radiat Res. 1997;148:248–253. [PubMed] [Google Scholar]
  • 38.Wu K.I., Pollack N., Panos R.J., Sporn P.H., Kamp D.W. Keratinocyte growth factor promotes alveolar epithelial cell DNA repair after H2O2 exposure. Am J Physiol. 1998;275:L780–L787. doi: 10.1152/ajplung.1998.275.4.L780. [DOI] [PubMed] [Google Scholar]
  • 39.Takeoka M., Ward W.F., Pollack H., Kamp D.W., Panos R.J. KGF facilitates repair of radiation-induced DNA damage in alveolar epithelial cells. Am J Physiol. 1997;272:L1174–L1180. doi: 10.1152/ajplung.1997.272.6.L1174. [DOI] [PubMed] [Google Scholar]

Articles from The American Journal of Pathology are provided here courtesy of American Society for Investigative Pathology

RESOURCES