In vivo evidence for therapeutic applications of beclin 1 to promote recovery and inhibit fibrosis after acute kidney injury

Abstract

Autophagy regulator beclin 1 activity determines the severity of kidney damage induced by ischemia reperfusion injury, but its role in kidney recovery and fibrosis are unknown and its therapeutic potentials have not been tested. Here, we explored beclin 1 effects on kidney fibrosis in three models of acute kidney injury (AKI)- ischemia reperfusion injury; cisplatin kidney toxicity; and unilateral ureteric obstruction in mouse strains with three levels of beclin 1 function: normal (wild-type), low (heterozygous global deletion of beclin 1, Becn1^+/−); and high beclin 1 activity (knock-in gain-of-function mutant Becn1, Becn1^FA). Fourteen days after AKI induction, heterozygous mice had more, but knock-in mice had less kidney fibrosis than wild-type mice. One day after ischemia reperfusion injury, heterozygous pan-kidney tubular Becn1 null mice had more severe kidney damage than homozygous distal tubular Becn1 null mice, which was similar to the wild-type mice, implying that proximal tubular beclin 1 protects the kidney against ischemia reperfusion injury. By 14 days, both pan-kidney heterozygous Becn1 null and distal tubular homozygous Becn1 null mice had poorer kidney recovery than wild-type mice. Injection of beclin 1 peptides increased cell proliferation in kidney tubules in normal mice. Beclin 1 peptides injection either before or after (2–5 days) ischemia reperfusion injury protected the kidney from injury and suppressed kidney fibrosis. Thus, both endogenous beclin 1 protein expression in kidney tubules and exogenous beclin 1 peptides are kidney protective via attenuation of acute kidney damage, promotion of cell proliferation, and inhibition of kidney fibrosis, consequently improving kidney recovery post-AKI. Hence, exogenous beclin 1 peptide may be a potential new therapy for AKI.

Graphical Abstract

INTRODUCTION

Acute kidney injury (AKI) is a severe disease with high prevalence.^1,2 AKI is not a benight and self-limited disease, as patients have high rates of short and long-term morbidity and mortality,³ and a high risk of progression to chronic kidney disease (CKD) even after recovery from AKI.⁴ To date, the mechanisms behind kidney fibrosis are not completely understood, and strategies to suppress fibrosis are very limited.

Autophagy is a conserved cellular homeostatic mechanism.^5,6 Normal autophagy protects kidneys from acute damage.^7–10 Elevation of autophagy lessens kidney fibrosis in unilateral ureteric obstruction (UUO), ischemia-reperfusion injury (IRI), and nephrotoxicity models.^7–11 However, another study showed that sustained activation of autophagy in kidney tubules promotes kidney fibrosis in the UUO model,¹² indicating that unremitting high autophagy activity may also be a risk of CKD progression.

Beclin 1 is one key component in autophagy machinery and involved in autophagosome nucleation. Beclin 1 peptide, as an autophagy inducer,^13,14 has been shown to be effective in reducing virus-associated disease, and improving cardiac dysfunction in aged mice.^15,16 Similar to many autophagy core proteins,¹⁷ beclin 1’s cytoprotection may be both autophagy-dependent and independent.

We already showed that high beclin 1 function by genetic or pharmacologic modulation is protective against IRI-induced AKI through escalation of autophagy activity.⁸ To explore suppressive effect of beclin 1 on kidney fibrosis triggered by AKI, we performed three acute kidney disease models (IRI; CN; and UUO) covering a wide range of pathophysiology and translatability in mouse strains with low beclin 1 activity by haploinsufficiency (Becn1^+/−);¹⁸ normal level of beclin activity in wild type (WT), or high beclin 1 activity by a knock-in of gain-of-function mutant F121A beclin 1 (Becn1^FA) (Supplemental Table 1).^8,15,16 To examine the more spatially-delineated effect of kidney tubular beclin 1 specifically on kidney fibrosis after kidney damage, we induced AKI by IRI in two mouse lines: heterozygous beclin 1 deletion in all kidney tubules (KT-Becn1^+/−) and specifically deletion in distal tubules (DT-Becn1^−/−). Finally, to test the therapeutic efficacy of exogenous beclin 1 peptides in AKI mice, we treated IRI-AKI WT mice with a short form of beclin 1 peptide comprising of 11 amino acids (TB-11) derived from Tat-beclin 1, which has the same bioactivity as Tat-beclin 1, but higher cell permeability.¹⁶ These in vivo experiments showed that beclin 1 in kidney tubules is renoprotective. Beclin 1 peptide may be a novel and effective therapeutic agent to prevent AKI.

METHODS

Detailed experimental materials and methods are in supplementary material.

The mouse lines used in this study were presented in Supplemental Table 1.

Three kidney disease models were made. (1) Ischemic reperfusion injury (IRI): Bilateral IRI was induced with 45-minute ischemia by renal artery cross-clamp.^8,19 (2) Cisplatin nephrotoxicity (CN): Cisplatin-induced AKI was prepared with intraperitoneal injection of cisplatin (10 mg/Kg body weight) once.^20,21 (3) Unilateral ureteral obstruction (UUO) model: UUO in the right kidney was conducted.⁹

After having compared the effect of different ischemia times (32, 35, 38, 40, 43 and 45 minutes) on plasm Cr in Becn1^+/−, WT and Becn1^FA mice in our pilot experiment, we selected 45 minutes for Becn1^FA mice, 40 minutes for WT mice and 35 minutes for Becn1^+/− mice to induce similar kidney injury in three mouse lines. We evaluated acute kidney damage at Day 1 and kidney fibrosis at Day 14 respectively. We also titrated cisplatin doses and selected 10 mg/Kg for Becn1^FA mice, 8 mg/Kg for WT mice and 6 mg/Kg for Becn1^+/− mice to induce similar kidney injury in the three different mouse lines. We evaluated acute kidney damage at Day 4 and kidney fibrosis at Day 14 respectively.²⁰

To examine beclin 1 effect on the kidney, Tat-beclin 1 11 peptides (TB-11) (YGRKKRRQRRR-GG-VWNATFHIWHD) and Tat-Scrambled peptides (TB-SC) (YGRKKRRQRRR-GG-WNHADHTFVWI) at the dose of 2 mg/kg/day¹⁶ were intraperitoneally injected into LC3-GFP reporter mice and WT mice daily at 2 days before or after IRI or Sham operation through 7 days after the operation. We selected the dose of TB-11 based on our previous findings of TB-11 effect on autophagy activity.¹⁶

RESULTS

beclin 1 function determines kidney recovery after IRI

We first used an IRI model to test if beclin 1 function affects kidney outcome in the recovery phase. At baseline, the levels of plasma Cr and BUN among three genotypes were similar (Figures 1a and 1b). At 14 days after IRI, Becn1^FA mice had the lowest, and Becn1^+/− mice had the highest plasma Cr and BUN 14 days after IRI (Figures 1a and 1b).

Figure 1. Kidney function and morphology in mice 14 days after ischemia reperfusion injury (IRI).

Becn1^+/−, WT, and Becn1^FA mice were subjected to sham or 45-minute renal artery cross-clamp operations. Fourteen days after surgery, mice were sacrificed, blood collected for (a) plasma Cr and (b) BUN measurement, and kidneys harvested for evaluation of kidney fibrotic changes with TC staining and immunoblot. (c) Kidney fibrosis was evaluated with TC staining in the kidney sections. Left panel: presentative microscopic photos of TC-stained kidney sections from 6 independent mice in each group. Scale bar = 200 μm. Right panel: Summary of kidney fibrosis quantified with kidney fibrosis score. (d) Kidney fibrotic marker was evaluated with immunoblots. Left panel: representative immunoblots for α-SMA and β-actin in the kidneys from 6 independent mice in each group; right panel: summary of all immunoblots from each group. Quantitative data are expressed as means ± S.D. and statistical significance was evaluated by two-way ANOVA followed by Student-Newman-Keuls post hoc test and significance was accepted when *: P<0.05; **: P<0.01 between two groups. Sample number in each group is presented in the brackets underneath each corresponding bar.

Similar to our previous findings,^8,16 autophagy flux in the kidney was highest in Becn1^FA mice, followed by WT mice; and lowest in Becn1^+/− mice at baseline (Supplementary Figure S1A). IRI upregulated autophagy flux in the kidney of WT mice, which was amplified in Becn1^FA mice and attenuated in Becn1^+/− mice (Supplementary Figure S1A). There was no pathology in the kidney of three mouse lines after sham operation (Figure 1c and Supplementary Figure S1B). Fourteen days after IRI, there was more tubulointerstitial fibrosis in Becn1^+/− mice and less in Becn1^FA mice compared to WT mice (Figure 1c and Supplementary Figure S1B). Immunoblots showed higher elevation of fibrotic marker in Becn1^+/− mice and less in Becn1^FA mice compared to WT mice post-IRI (Figure 1d). Thus, higher autophagy flux from constitutively high beclin 1 function is associated with less fibrosis after IRI.

To examine if lower p62 protein in the kidneys results from lower p62 transcripts, p62 mRNA was quantified with qPCR (Supplementary Figure S2). The levels of p62 transcript were not significantly different among Becn1^FA, WT and Becn1^+/− mice before and after IRI. Moreover, IRI did not significantly change p62 mRNA levels. Therefore, the change in p62 protein is mainly regulated post-translationally.

Beclin 1 function determines kidney recovery after cisplatin injection

Next, we explored beclin 1 effect on CN, which was shown to progress to CKD in animals^21,22 and supported by clinical observations.²³ We showed that autophagy flux in the kidney was stimulated by cisplatin injection in WT mice. This response was blunted in Becn1^+/− mice, and enhanced in Becn1^FA mice (Supplementary Figure S3A).

Fourteen days after cisplatin injection, Becn1^+/− mice had higher, and Becn1^FA mice had lower levels of plasma Cr and BUN than WT mice (Figures 2a and 2b). Becn1^+/− mice had more fibrosis than WT mice, while Becn1^FA mice had less (Supplementary Figure S3B, and Figure 2c). Becn1^+/− mice also had higher whereas Becn1^FA mice had lower α-SMA expression than WT mice, (Figure 2d). Therefore, higher autophagy flux driven by increased beclin 1 function ends up with milder kidney fibrosis. But it is unknown whether better kidney outcome in mice with higher beclin 1 function is due to lesser acute kidney damage and/or quicker kidney recovery.

Figure 2. Kidney function and morphology in the mice after cisplatin injection.

Becn1^+/−, WT, and Becn1^FA mice were subjected to vehicle (normal saline) or cisplatin injection. Fourteen days after injection, the mice were sacrificed, blood collected for (a) plasma Cr and (b) BUN measurement, and kidneys harvested for evaluation of kidney fibrotic changes with TC staining and immunoblot. (c) Kidney fibrosis was evaluated with TC staining in the kidney sections. Left panel: representative microscopic photos of TC-stained kidney sections from 6 independent mice in each group. Scale bar = 200 μm. (d) Right panel: summary of kidney fibrosis quantified with kidney fibrosis score on TC-stained sections. Kidney fibrotic marker was quantified with immunoblots. Left panel: the presentative immunoblots for α-SMA and β-actin in the kidneys from 6 independent mice in each group; right panel: summary of all immunoblots from each group. Quantitative data are expressed as means ± S.D. and statistical significance was evaluated by two-way ANOVA followed by Student-Newman-Keuls post hoc test and significance was accepted when *: P<0.05; **: P<0.01 between two groups. Sample number in each group is presented in the brackets underneath each corresponding bar.

Beclin 1 inhibits kidney fibrosis independent of degree of injury

To explore if heightened autophagy reduces kidney fibrosis and promotes kidney recovery independently from the acute renoprotection, we induced similar peak kidney damage in Becn1^+/−, WT and Becn1^FA mice and found that 45-minute ischemia in Becn1^FA mice, 40-minute ischemia in WT mice, and 35-minute ischemia in Becn1^+/− mice induced comparable peak levels of plasma Cr (Figure 3a) and BUN (Supplementary Figure S4A), similar kidney damage assessed by H&E staining (Supplementary Figure S4B) and immunoblot for kidney NGAL and active caspase-3 expression (Supplementary Figure S4C) at Day 1. Interestingly, 14 days later, Becn1^+/− mice had less, and Becn1^FA mice had more decline in plasma Cr than WT mice (Figure 3a). Kidney fibrosis (Figure 3b) and α-SMA expression (Figure 3c) were more noted in Becn1^+/− mice and less in Becn1^FA mice when compared to WT mice. In the parallel nephrotoxic model, cisplatin injection at 10 mg/Kg in Becn1^FA mice, 8 mg/Kg in WT mice and 6 mg/Kg in Becn1^+/− mice induced similar peak levels of plasma Cr (Figure 3d) and BUN (Supplementary Figure S4D), similar kidney damage assessed by H&E staining (Supplementary Figure S4E) and immunoblot for kidney NGAL and active caspase-3 expression (Supplementary Figure S4F) at Day 4. Interestingly, 14 days after cisplatin injection, Becn1^+/− mice had less, and Becn1^FA mice had more decline in plasma Cr than WT mice in (Figure 3d). Becn1^+/− mice had more, and Becn1^FA mice had less kidney fibrosis in plasma Cr than WT mice (Figure 3e and 3f). Therefore, the slower kidney recovery in Becn1^+/− mice, and the quicker kidney recovery in Becn1^FA mice are likely due to beclin 1 activity and autophagy function.

Figure 3. Changes in plasma Cr in Becn1^+/−, WT, and Becn1^FA mice at early and later phase in both ischemia reperfusion injury (IRI) and cisplatin nephrotoxicity (CN) models.

(a) Ischemic kidney damage was induced with 35-minute ischemia in Becn1^+/−, 40-minute ischemia in WT, and 45-minute ischemia in Becn1^FA mice. Blood was collected for measurement of plasma Cr at Day 1 and Day 14. Kidneys were harvested at Day 14 for histology, immunohistochemistry, and immunoblot. Plasma Cr is plotted individually and also as averages at early and later phases. Solid circles depict deceased mice. (b) Kidney fibrosis was evaluated with Trichrome (TC) staining. Upper panel: representative microscopic photos of TC-stained kidney sections from 6 independent mice in each group. Scale bar = 200 μm; bottom panel: summary of kidney fibrosis quantified with kidney fibrosis score on TC-stained sections. (c) Kidney fibrotic marker α-SMA was quantified with immunoblots. Upper panel: the presentative immunoblots for α-SMA and β-actin in the kidneys from 6 independent mice in each group; bottom panel: summary of all immunoblots from each group. (d) CN was induced by injection of cisplatin at the dose of 6 mg/Kg in Becn1^+/−, 8 mg/Kg in WT, and 10 mg/kg in Becn1^FA mice. The blood was collected for the measurement of plasma Cr at Day 4 and Day 14, respectively. The kidneys were harvested at Day 14 for histology and immunohistochemistry, and immunoblot. Plasma Cr is presented at average levels and also plotted individually at early and later phases. Solid circles depict dead mice. (e) Kidney fibrosis was evaluated with TC staining in the kidney sections. Upper panel: representative microscopic photos of TC-stained kidney sections from 6 independent mice in each group. Scale bar = 200 μm; bottom panel: summary of kidney fibrosis quantified with kidney fibrosis score on TC-stained sections. (f) Kidney fibrotic marker α-smooth muscle actin (α-SMA) was quantified with immunoblots. Upper panel: the presentative immunoblots for α-SMA and β-actin in the kidneys from 6 independent mice in each group; bottom panel: summary of all immunoblots from each group. Statistical significance of plasma Cr among groups was evaluated by two-way ANOVA followed by Student-Newman-Keuls post hoc test and significance was accepted when *: P<0.05; **: P<0.01 between two groups. Sample number in each group is presented in the brackets underneath each corresponding bar.

Beclin 1 function determines kidney fibrosis after UUO

To explore whether beclin 1 inhibits kidney fibrosis, we used the well-established kidney fibrosis model, UUO. At 14 days after the operation, there was no detectable tubulointerstitial fibrosis in the contralateral non-ligated kidneys of UUO mice and the kidneys from sham mice (data not shown). Gross anatomical changes, and severity of hydronephrosis were similar in ligated kidneys among Becn1^+/−, WT and Becn1^FA mice, indicating that the degree of obstruction was likely comparable (Figure 4a). However, there was more kidney fibrosis in Becn1^+/−, while Becn1^FA mice had less compared to WT mice (Figure 4b). In addition, Becn1^FA mice had higher and Becn1^+/− mice had less upregulation of autophagy flux in response to ureteric ligation than WT mice (Supplementary Figure S5). When compared to UUO-WT mice, UUO-Becn1^+/− mice had more histologic fibrosis and higher expression of fibrotic markers, while UUO-Becn1^FA mice had the opposite changes (Figures 4b – d), implying that beclin 1 significantly suppresses kidney fibrosis triggered by UUO even when the degree of obstruction and hydronephrosis was similar. These results further support suppressive effect of beclin 1 on kidney fibrosis.

Figure 4. Kidney function and morphology in mice after unilateral ureteric obstruction (UUO).

Becn1^+/−, WT, and Becn1^FA mice were subjected to sham or UUO operations. Fourteen days after surgery, the mice were sacrificed, the kidneys harvested for evaluation of fibrotic changes with Trichrome (TC) and Picrosirius Red (PSR) staining, immunohistochemistry, and immunoblot. (a) Representative photos of kidney sections of TC and PSR staining were from 6 independent mice in each group. Scale bar = 2 mm. (b) Kidney fibrosis was assessed by TC staining. Upper panel: representative microscopic photos of TC-stained kidney sections from 6 independent mice in each group. Scale bar = 200 μm; bottom panel: kidney fibrosis was quantified with kidney fibrosis score on TC-stained sections. (c) Immunofluorescence photos for fibrotic markers in the kidneys of mice. Representative images of Col I (green) in kidney proximal tubules (blue, Lotus Tetragonolobus Lectin; LTL) and collecting ducts (red, Dolichos biflorus agglutinin; DBA) from 5 mice in each treatment. Scale bars = 100 μm. (d) Kidney injury marker evaluated with immunoblots. Left panel: representative immunoblots for α-smooth muscle actin (α-SMA) and α-actin in the kidneys from 6 independent mice in each group; right panel: summary of immunoblots from each group. Quantitative data are expressed as means ± S.D. and statistical significance was evaluated by two-way ANOVA followed by Student-Newman-Keuls post hoc test and significance was accepted when *: P<0.05; **: P<0.01 between two groups. Sample number in each group is presented in the brackets underneath each corresponding bar.

Normal expression of beclin 1 in kidney tubules protects the kidney from IRI

The in vivo results presented above using global genetic manipulation of beclin 1 activity, although encouraging, cannot prove whether renoprotection by beclin 1 depends on its expression in kidney tubules. To determine the necessity of tubular beclin 1, we need conditional deletion of beclin 1 in kidney tubules.

To characterize the recombinase specificity and activity in kidney tubules, we generated the reporters tdTomato;Six2-Cre⁺ (pan-tubular) and tdTomato;NCC-Cre⁺ (distal tubule) mice. As shown in Figure 5a, tdTomato red fluorescent signal was expressed in almost all kidney tubules but patchy in collecting duct epithelial cells in tdTomato;Six2-Cre⁺ mice, and only in distal tubules in tdTomato;NCC-Cre⁺ mice, confirming the pan-tubular and distal tubule-specific recombinase activity driven by Six-2 or NCC promoter respectively.

Figure 5. Normal beclin 1 expression in kidney tubules is required for renoprotection against ischemia reperfusion injury (IRI).

The tdTomato;Cre⁻, tdTomato;Six2-Cre⁺, and tdTomato;NCC-Cre⁺ mice were sacrificed and kidneys were harvested for mapping fluorophore expression in kidney tubules. (a) The presentative fluorescent microscopic photos sections were from 3 independent mice in each group. TdTomato (tdT, red), LTL (Lotus tetragonolobus lectin, proximal tubule marker, green), Na-Cl Cotransporter (NCC, distal tubule marker, blue), and Dolichos biflorus agglutinin (DBA, collecting duct marker, green). Scale bar = 200 μm. WT, KT-Becn1^+/−, and DT-Becn1^−/− mice were subjected to sham or 45-minute ischemia operations. (b) and 14 days after surgery, Mice were sacrificed, blood collected for plasma Cr measurement at Days 1, 7, and 14, after surgery, and the kidneys harvested for evaluation of autophagy activity and histology. (c) Time course of plasma Cr values. Plasma Cr levels were presented as an average from 10–12 mice. *: p<0.05, **: p<0.01 vs WT; #: p<0.05, ##: p<0.01 vs DT-Becn1^−/− within three genotyping IRI mice at same time point by one-way ANOVA followed by Student-Newman-Keuls post hoc test. (d) Kidney injury markers were evaluated with immunoblots in WT, KT-Becn1^+/−, and DT-Becn1^−/− mice at Day 1 after IRI or sham surgeries. Upper panel: representative immunoblots for NGAL, active caspase-3, and β-actin in the kidneys from 6 independent mice at each group; bottom panel: summary of all immunoblots from each group. (e) Kidney histological alterations in H&E stained sections. Left panel: representative microphotographs from 6 independent mice in each group, Scale bar = 50 μm. Arrows depict protein casts and detached cell debris in kidney tubules. Arrow heads depict loss of cell nuclei; right panel: pathologic kidney injury score. Fibrotic markers in the kidneys was evaluated with immunoblots in WT, KT-Becn1^+/−, and DT-Becn1^−/− mice at Day 14 after IRI or sham surgeries (f). Upper panel: the presentative immunoblots for α-smooth muscle actin (α-SMA) and β-actin in the kidneys from 6 independent mice at each group; bottom panel: summary of all immunoblots from each group. (g) Kidney histology was assessed with TC staining. Upper panel: representative microscopic photos of TC-stained kidney sections were from 6 independent mice in each group. Scale bar = 200 μm; bottom panel: kidney fibrosis was quantitatively assessed with kidney fibrosis score on TC-stained sections. Quantitative data are expressed as means ± S.D. and statistical significance was evaluated by two-way ANOVA followed by Student-Newman-Keuls post hoc test and significance was accepted when *: P<0.05; **: P<0.01 between two groups for Figures 5d – g. Sample number in each group is presented in the brackets underneath each corresponding bar.

We generated two murine lines of conditional deletion of beclin 1 in kidney tubules: KT-Becn1^+/−, and DT-Becn1^−/−. At baseline, KT-Becn1^+/− mice had lower beclin 1 expression and autophagy flux, compared to WT mice. DT-Becn1^−/− mice had intermittent levels of beclin 1 expression and autophagy flux between KT-Becn1^+/− and WT mice (Supplementary Figure S6), indicating that KT-Becn1^+/− mice have more significant impaired autophagy flux than DT-Becn1^−/−mice.

We next induced IRI in these two lines (Figure 5b). After the Sham procedure, plasma Cr levels were similar between KT-Becn1^+/−, DT-Becn1^−/− and WT mice (Figure 5c and Table 1), indicating that the decrease in kidney tubular beclin 1 does not modify baseline kidney function. At Day 1 after IRI, plasma Cr levels were higher in KT-Becn1^+/− mice than DT-Becn1^−/− and WT mice, while the latter two had similar plasma Cr levels (Figure 5c and Table 1). Therefore, the decrease in beclin 1 in all tubules, but not distal tubules, yielded more severe kidney damage induced by IRI.

Table 1.

Plasma Cr in mice of conditional knock-out of Becn1 in kidney tubules

		Day 0			Day 1			Day 7			Day 14
		WT	KT-Becn1+/−	DT-Becn1−/−	WT	KT-Becn1+/−	DT-Becn1−/−	WT	KT-Becn1+/−	DT-Becn1−/−	WT	KT-Becn1+/−	DT-Becn1−/−
Sham	Means±SD	0.094±0.012	0.093±0.012	0.093±0.009	0.092±0.013	0.094±0.0091	0.093±0.012	0.097±0.012	0.094±0.011	0.098±0.009	0.095±0.012	0.093±0.012	0.090±0.011
	N	21	21	21	7	7	7	7	7	7	7	7	7
IRI	Means±SD	0.096±0.011	0.09±0.012	0.09±0.008	1.160±0.204	2.366±0.356	1.177±0.176	0.453±0.097	1.164±0.208	0.643±0.085	0.123±0.029	0.406±0.062	0.310±0.054
	N	30	30	30	7	7	7	8	7	7	9	7	8
P value by unpaired t test (Sham vs IRI)		p>0.05	p>0.05	p>0.05	p<0.01	p<0.001	p<0.001	p<0.001	p<0.001	p<0.001	p<0.001	p<0.001	p<0.01

KT-Becn1^+/− mice had a slower decline in plasma Cr post-IRI compared to DT-Becn1^−/− mice and WT mice, and DT-Becn1^−/− mice had higher plasma Cr than WT mice at day 7 and 14 (Figure 5c and Table 1), indicating delayed kidney recovery in mice with low kidney tubular beclin 1. Immunoblots showed higher expression of kidney injury markers (NGAL and active caspase-3 protein) in KT-Becn1^+/− mice than in DT-Becn1^−/− mice and WT mice; but no differences were noted between DT-Becn1^−/− mice and WT mice (Figure 5d). There was more tubular damage such as tubular necrosis, brush border membrane detachment, casts in the tubular lumens, interstitial edema, and inflammatory infiltration, and higher pathologic scores in KT-Becn1^+/− mice than in DT-Becn1^−/− and WT mice (Figure 5e); but there was no significant difference in acute kidney damage between DT-Becn1^−/− mice and WT mice (Figure 5e). The α-SMA expression in the kidney was the highest in KT-Becn1^+/− mice among three genotypes (Figure 5f). Interestingly, there was higher α-SMA expression in DT-Becn1^−/− mice than in WT mice at 14 days after IRI (Figure 5f). Although KT-Becn1^+/− mice had more kidney fibrosis than other two (DT-Becn1^−/− mice and WT mice), WT mice had less fibrosis than DT-Becn1^−/− mice after IRI (Figure 5g). In addition, there was no significant difference in death rate between KT-Becn1^+/− (9/30, 30.0%) and DT-Becn1^−/− mice (7/30, 23.4%, p=0.371). However, the mortality in KT-Becn1^+/− mice was 10% higher than that in WT mice (6/30, 20%, p<0.05). Thus, tubular beclin 1 expression, outside distal tubules, is required for prevention of ischemic injury, suppression of kidney fibrosis and promotion of kidney recovery. Normal beclin 1 expression in distal tubules does not prevent ischemic kidney damage, but contributes to kidney recovery. The mechanism behind distal tubular beclin 1 effect on kidney recovery remains to be explored.

To define whether renoprotection by tubular beclin 1 is specifically mediated through modulation of autophagy flux, we decreased autophagy flux through a different way by generating an additional conditional knock-out tubular deletion of Atg5,²⁴ an important autophagy protein, (Supplementary Figures S7A and S7B). Similar to the findings in beclin 1 deleted mice, pan-tubular Atg5 deletion led to more severe kidney damage at acute phase and more kidney fibrosis at late phase (Supplementary Figures S7C – S7E). Absence of Atg5 protein in distal tubules did not change the acute response to IRI, but delayed kidney recovery with more fibrosis compared to WT mice (Supplementary Figures S7C – S7E). Taken together with the findings in the beclin 1 knock-out mice (Figure 5), the change in kidney response to Atg5 deletion supports that beclin 1 exerts kidney protection through modulation of autophagy flux in the kidney.

Pre-treatment with beclin 1 peptide prevents AKI induced by IRI

The important next step is to test the translatability of the model created thus far to eventual clinical application. Two-day of TB-11, beclin 1 peptide, daily injection increased autophagy flux in the kidney and 5-day injection further increased it (Figures 6a and 6b). Daily injection of TB-11 starting 2 days before IRI and continuation for another 7 days after IRI (Figure 6c) decreased kidney damage evaluated by plasma Cr measurement (Table 2 and Figure 6d), HE stain (Figure 6e) and immunoblots (Figure 6f) compared to TB-SC treatment. TB-11 pre-treatment accelerated decline in plasma Cr at 7 and 14 days (Figure 6f) and attenuated kidney fibrosis (Figures 7a – c) after IRI, supporting that the pre-treatment is prophylactic against IRI and improves kidney outcome.

Figure 6. Beclin 1 peptide protects the kidney from ischemia reperfusion injury (IRI).

Beclin 1 peptide (TB-11) or scrambled peptide (TB-SC) as control were intraperitoneally injected into LC3-GFP reporter mice daily for 2 and 5 days. Four hours prior to sacrifice, mice were treated intraperitoneally with chloroquine, and the kidneys were harvested for autophagy flux assessment. (a) Immunofluorescence microphotograph for autophagy activity. Upper panel: representative photos of GFP-LC3 immunofluorescence in kidney tubules. Scale bars = 50 μm. Bottom panel: summary of quantitation of GFP-LC3 punctas in kidney tubules of 6 mice from each treatment. (b) Immunoblots of total kidney lysates for autophagy markers. Upper panel: representative immunoblots for LC3, p62, and β-actin from each treatment. Bottom panel: summary of data. (c) TB-11 peptides or TB-SC were intraperitoneally injected into WT mice daily starting 2 days before and for up to 7 days after IRI. IRI for AKI induction or laparotomy for Sham were performed in WT mice with 45-minute ischemia, 10 mice from each group were sacrificed, blood collected, and kidneys harvested at 2, 7 and 14 days after the operation. (d) Plasma Cr changes from 0 – 14 days. Data are presented as average of plasma Cr from 10–12 mice at each time point in each group. *: p<0.05, **: p<0.01 vs. Pre-TB-11; ^#: p<0.05, ^##: p<0.01 vs. Post-TB-11 within IRI mice with three treatments at same time point by one-way ANOVA followed by Student-Newman-Keuls post hoc test. (e) Kidney damages were evaluated with H&E staining at 2 days. Upper panel: representative images of kidney damages, scale bar = 50 μm; bottom panel: kidney histological scores based on H&E-stained kidney sections from 6 mice in each group. (f) Kidney damage was also evaluated with immunoblots at 2 days after surgeries. Upper panel: representative immunoblots for neutrophil gelatinase-associated lipocalin (NGAL), active caspase-3, and β-actin in the kidneys from 6 independent mice from each group; bottom panel: summary of all immunoblots from each group. Quantitative data are expressed as means ± S.D. and statistical significance was evaluated by two-way ANOVA followed by Student-Newman-Keuls post hoc test and significance was accepted when *: P<0.05; **: P<0.01 between two groups. Sample number in each group is presented in the brackets underneath each corresponding bar.

Table 2.

The effect of beclin 1 peptides on plasma Cr of IRI-AKI mice

		Day 0			Day 2			Day 7			Day 14
		TB-SC	Pre-TB 11	Post-TB 11	TB-SC	Pre-TB 11	Post-TB 11	TB-SC	Pre-TB 11	Post-TB 11	TB-SC	Pre-TB 11	Post-TB 11
Sham	Means±SD	0.084±0.006	0.085±0.005	0.084±0.007	0.087±0.008	0.086±0.007	0.085±0.007	0.087±0.006	0.084±0.01	0.085±0.008	0.087±0.007	0.087±0.012	0.084±0.011
	N	22	23	22	10	10	10	10	10	11	12	13	12
IRI	Means±SD	0.083±0.007	0.084±0.006	0.085±0.005	0.960±0.112	0.396±0.046	0.977±0.121	0.395±0.043	0.124±0.014	0.214±0.025	0.201±0.021	0.095±0.013	0.143±0.015
	N	26	26	26	10	10	10	10	10	10	11	15	13
P value by unpaired t test (Sham vs IRI)		p>0.05	p>0.05	p>0.05	p<0.001	p<0.001	p<0.001	p<0.001	p<0.001	p<0.001	p<0.001	p>0.05	p<0.01

Figure 7. Beclin 1 peptide reduces the kidney fibrosis and promotes kidney recovery from ischemia reperfusion injury (IRI).

Kidney samples were from the experimental mice shown in Figure 5c. (a) Kidney fibrosis at 14 days was evaluated with trichrome staining. Upper panel: representative TC images were from 6 independent mice in each group. Scale bar = 200 μm; bottom panel: kidney fibrosis was quantified with kidney fibrosis score on TC-stained sections. (b) Collagen I accumulation in the tubulointerstitium of IRI mice. Representative immunofluorescent images for collagen I (green) co-stained with 4′,6-diamidino-2-phenylindole (DAPI, blue) from 3 independent mice in each group. Scale bar = 100 μm. (c) Fibrotic marker in the kidneys was assessed with immunoblots. Upper panel: representative immunoblots for α-smooth muscle actin (α-SMA) and β-actin in the kidneys from 6 independent mice in each group; bottom panel: summary of all immunoblots from each group. Quantitative data are expressed as means ± S.D. and statistical significance was evaluated by two-way ANOVA followed by Student-Newman-Keuls post hoc test and significance was accepted when *: P<0.05; **: P<0.01 between two groups. Sample number in each group is presented in the brackets underneath each corresponding bar.

Beclin 1 peptide treatment after IRI suppresses IRI-triggered kidney fibrosis

Although the in vivo results presented above are encouraging, they could not exclude the possibility that suppression of kidney fibrosis and promotion of kidney regeneration by beclin 1 pre-treatment may result from its prophylactic effect on the attenuation of kidney damage. Also, while prophylaxis of AKI is important, most clinical scenarios necessitates the ability of therapy to be given after the acute insult. We therefore tested whether injection of TB-11 after AKI onset still improves outcome and inhibits kidney fibrosis (Figure 6d). At 2 days after IRI when plasma Cr levels were similar between post-TB-11 group and TB-SC group (Table 2 and Figure 6d), TB-11 and TB-SC were given daily to IRI mice for 5 days. TB-11 post-treatment led to a much faster decline in plasma Cr and maintained their lower levels for up to 14 days compared to TB-SC treatment (Table 2 and Figure 6d). However, mice with post-TB-11 treatment still had higher levels of plasma Cr at day 14 than baseline indicating incomplete recovery (Table 2 and Figure 6d). TB-11 post-injection also attenuated fibrosis (Figure 7a) and reduced fibrotic markers in the kidneys (Figures 7b and c) compared to TB-SC-treated AKI mice. Obviously, TB-11 post-injection worked less efficiently than pre-injection (Figures 7a – b). Overall, pre- or post-TB-11 treatment significantly reduced mortality in AKI mice to 3.8% (1/26, p<0.05) and 11.5%, (3/26, p<0.05) respectively from TB-SC-treated AKI mice (4/26, 19.2%) throughout experiment, while pre-treatment led to lower mortality in AKI mice than post-treated AKI mice (p<0.05). Therefore, the earlier intervention with TB-11, the better outcome.

Beclin 1 peptide promotes cell proliferation in kidney epithelial cells

Because tubular cell proliferation capacity is associated with kidney recovery and tubular survival after AKI,^25,26 we examined the proliferating cell nuclear antigen, a cell proliferation marker²⁷ and found that TB-11 increased cell proliferation in the kidneys, and longer treatment was associated with more proliferating cells (Figure 8a). AKI mice had higher cell proliferation in the kidneys compared to Sham mice (Figure 8a). TB-11 pre-treatment attenuated the magnitude of increase in cell proliferation in AKI mice (Figure 8a), which might be associated with less kidney damage.

Figure 8. Beclin 1 peptide promotes kidney tubular epithelial proliferation after ischemia reperfusion injury (IRI).

Kidney samples were from experimental mice shown in Figure 5c. (a) Immunoblots for cell proliferation marker in total kidney lysates of mice at 7 days after the operation. Left panel: representative immunoblots for proliferating cell nuclear antigen (PCNA) and β-actin from each treatment; right panel: summary of data from all of blots from 6 independent experiments. (b) Proliferating cells in proximal tubules were evaluated with immunohistochemistry. Left panel: representative images of Ki67 (red) in kidney proximal tubules (green, Lotus tetragonolobus lectin; LTL) from 6 mice in each treatment. White arrows depict positive Ki67 cells in proximal tubules, white arrow heads depict positive Ki67 cells in non-proximal tubules, scale bars = 50 μm; right panel: summary of quantitation of positive Ki67 cells in LTL-positive kidney tubules. (c) Proliferating cells in distal tubules were evaluated with immunohistochemistry. Left panel: representative images of Ki67 (red) in kidney distal tubules (green, Na-Cl Cotransporter; NCC) from 6 mice in each treatment. White arrows depict positive Ki67 cells in distal tubules, white arrow heads depict positive Ki67 cells in non-distal tubules, Scale bars = 50 μm; right panel: summary of quantitation of Ki67-positive cells in NCC-positive kidney tubules. Quantitative data are expressed as means ± S.D. and statistical significance was evaluated by two-way ANOVA followed by Student-Newman-Keuls post hoc test and significance was accepted when *: P<0.05; **: P<0.01 between two groups. Sample number in each group is presented in the brackets underneath each corresponding bar.

Next, we localized proliferating cells in kidney tubules. There were more proliferating cells in kidney tubules of AKI than Sham mice (Figures 8a – c). Interestingly, proliferating cells were not only present in proximal and distal tubules, but also in endothelial cells and fibroblasts after IRI (Supplementary Figure S8). Compared to the number of proliferating cells in proximal tubules (8.27±1.21%), the number of proliferating cells in distal tubules (6.74±1.13%, P=0.0471) were slightly but statistically significantly lower 7 days after IRI.

Beclin 1 peptide significantly increased the number of proliferating cells in both proximal and distal tubules of Sham mice (Figures 8b and c). Increased cell proliferation was less in mice that received TB-11 after AKI onset. Pre-treatment further reduced the number of proliferating cells. One potential interpretation is that pre-treated AKI mice had less kidney damage.

Beclin 1 peptide protects NRK cells from H₂O₂-induced cytotoxicity

To test whether TB-11 exerts cytoprotection independently of autophagy, we suppressed autophagy activity in NRK cells with chloroquine, and treated them with TB-11. TB-11-upregulated autophagy activity was almost completely blocked by chloroquine (Figure 9a). Interestingly, chloroquine induced elevation of LDH release in the media when the cells were incubated in serum-free media, which was significantly reduced by TB-11 (Figure 9b). H₂O₂-induced LDH release was exacerbated when autophagy activity was suppressed by chloroquine; and significantly decreased, but not completely abolished by TB-11 (Figure 9b). There were similar changes in active capase-3 and NGAL protein expression in NRK cells as LDH release into culture media (Figure 9c). TB-11-induced reduction of active capase-3 and NGAL expression in the cells was significantly blunted, but not completed abolished by chloroquine. Therefore, low autophagy renders cells more susceptible to oxidative cytotoxicity, and TB-11’s cytoprotection is mediated by both autophagy-dependent and independent pathway.

Figure 9. Beclin 1 peptide protects cultured kidney cells in vitro.

Autophagy activity in normal rat kidney (NRK) cells was manipulated with chloroquine to examine if TB-11 (10 μM) still protects NRK cells from H₂O₂-induced cytotoxicity. (a) NRK cells seeded in petri dishes (10 cm) were treated with added chloroquine (50 μM) or sterile water companied by TB-11 or TB-SC for 4 hours. Cell lysates were immunoblotted for LC3 and p62. Upper panel: representative immunoblots for LC3, p62, and β-actin in cell lysates from 6 independent experiments; bottom panel: summary of all immunoblots from each group. (b) NRK cells were seeded in 12-well plates. Post-confluent cells were treated with chloroquine or sterile water companied by TB-11 or TB-SC at presence or absence of H₂O₂ (200 nm) for 24 hours. Lactate dehydrogenase (LDH) release into media from NRK cells were measured. The results were presented from 7 independent experiments. (c) NRK cells were treated with chloroquine or sterile water companied by TB-11 or TB-SC for 4 hours followed by treatment of H₂O₂ for 20 hours. Cell lysates were immunoblotted for neutrophil gelatinase-associated lipocalin (NGAL) and active caspase-3. Left panel: representative immunoblots for NGAL, active caspase-3, and β-actin in cell lysates from 6 independent experiments; right panel: summary of all immunoblots from each group. Quantitative data are expressed as means ± S.D. and statistical significance is evaluated by two-way ANOVA followed by Student-Newman-Keuls post hoc test and significance was accepted when *: P<0.05; **: P<0.01 between two groups. Sample number in each group is presented in the brackets underneath each corresponding bar. CQ: chloroquine

DISCUSSION

This is a proof-of-concept in vivo study confirming the renoprotective role of kidney tubular beclin 1 in three kidney injury models (IRI, CN, and UUO). We showed that 1) higher beclin 1 activity suppresses kidney fibrosis and promotes kidney recovery, 2) pan-tubular beclin 1 reduction exacerbates kidney damage, and accelerates kidney fibrosis, whereas low beclin 1 only in distal tubular increases kidney fibrosis in the IRI model, 3) beclin 1 peptide promotes cell proliferation in kidney tubules, 4) beclin 1 peptide given before AKI ameliorates ischemic kidney damage with better kidney recovery, and given after AKI significantly reduces kidney fibrosis and promotes kidney recovery.

Many experiments showed that higher autophagy is associated with better kidney protection and cytoprotection.^7,28 However, the milder kidney damage in the acute phase does not necessarily beget less kidney fibrosis that follows. Therefore, the study of autophagy in kidney fibrosis in later phase after kidney injury is deem required. The published data presented both pro- and anti-fibrosis roles of autophagy in the kidney.^6,11,29–31 These seemingly contradictory findings likely reflect technical and biologic variances such as different experimental protocols for preparation of animal models including the nature and the severity of kidney insults, different methods and timings for measurement of autophagy activity along kidney disease courses, and different cell signaling pathways involved in different animal models.

Persistent autophagy activation was reported to increase fibroblast proliferation and activation, induce tubular atrophy, and result in interstitial macrophage infiltration and over-production of pro-fibrotic factors in the UUO model.¹² UUO-induced kidney fibrosis was suppressed by autophagy inhibitors or by Atg7 knockout in proximal tubules.¹² Yan et al also showed that autophagy inhibitors reduced the expression of fibrotic markers and the number of tubular cell apoptosis through reduction of lipid accumulation in tubular cells in same model.³¹ On the other hand, our group and others proposed anti-fibrotic role of autophagy in kidney disease through enhancing cell’s capacity to remove collagen accumulation,^11,32 because high autophagy suppresses and low autophagy promotes kidney fibrosis. Li et al. showed that Atg5 deficiency in proximal tubules increased collagen I, and promoted kidney fibrosis.³³ Similarly, more kidney fibrosis and collagen deposition, and higher TGF-β were noted in mice with Atg7 deletion in distal tubules,²⁹ LC3 null, and Becn1^+/− after UUO.¹⁰ The inconsistent results of autophagy role in kidney fibrosis call for standardization of measurement of autophagy activity and animal protocols for kidney disease.

Monitoring autophagy activity requires determination of autophagy flux. Despite numerous experimental studies showing that combined measurement of LC3 and p62 is gold standard to evaluate autophagy activity, there continues to be confusion regarding acceptable methods.³⁴ Additionally, transcriptional regulation and post-translational degradation modulate the intracellular levels of p62.^35,36 We determined LC3 and p62 protein at steady state, and also measured p62 mRNA. No significant changes in p62 mRNA in the kidney between mice with different beclin 1 activity, and between before and after kidney injury indicate that p62 protein in the ischemic kidney is mainly regulated post-translationally.

Our study showed an inverse relationship between autophagy activity and kidney fibrosis in UUO. The concept of autophagy being anti-fibrotic is further supported by similar results in IRI and CN models. Moreover, poorer kidney recovery in Becn1^+/− mice and better kidney outcome in Becn1^FA mice than WT mice after similar acute kidney damage was induced strongly support that beclin 1 activity in kidney tubules determines kidney recovery independent of severity of kidney damage. However, it is important to note that current study does not provide evidence to support the concept that high beclin 1 directly suppresses kidney fibrosis after kidney damage. Furthermore, whether effect of beclin 1 on kidney fibrosis is autophagy-dependent or autophagy-independent warrants further exploration.

The role of the proximal tubule in kidney repair after kidney damage has been shown in several murine studies.^26,37–40 However, the role of distal tubules in AKI onset and recovery is inconclusive.⁴¹ Distal tubules may support proximal tubular survival and promote proximal tubular regeneration in a paracrine manner,⁴² because distal tubules do not dedifferentiate into proximal tubules after kidney injury.^43,44 We demonstrated that distal tubular beclin 1 deficiency did not render kidney more susceptible to ischemia as the proximal tubular beclin 1 does, but reduced cell proliferation in proximal tubules, and promoted fibrosis, indicating that normal beclin 1 function contributes to normal autophagy activity in distal tubules, and healthy distal tubules may consequently promote proximal tubular repair.⁴² However, we do not have direct evidence to support that beclin 1 promotes tubular regeneration after AKI.

Beclin 1 modulates autophagy activity at least in part through modulation of beclin 1-bcl2 complex formation.^15,16 TB-11, the short beclin 1 peptide (residues 269 –279 of human beclin 1 protein) is not within bcl2 binding domain.⁴⁵ Therefore, TB-11-upregulated autophagy activity is not dependent on the modulation of beclin 1-bcl2 complex. Many autophagy core proteins exert biologic actions through both autophagy-dependent and independent pathways.^17,46,47 Beclin 1 is not an exception.^48,49 Our in vitro results demonstrated that beclin 1 peptides significantly decreased both serum depletion and H₂O₂-induced cell injury, and the cytoproptection was dramatically blunted, but not completely abolished by autophagy inhibitor. Therefore, beclin 1 peptides exert cytoprotection probably through both autophagy-dependent and in-dependent pathway. However, the in vivo experiment in autophagy deficient mice such as Atg5 null in kidney tubules is required in the future to explore if TB-11 renoprotection is autophagy-dependent and/or -independent.

We are the first to test the therapeutic implication of the beclin 1 peptide in IRI-AKI model and to demonstrate that beclin 1 peptide is effective in both prevention and treatment of IRI-AKI. The current study not only shows that beclin 1 peptides modestly increased the number of proliferating cells in kidney tubules at baseline that may be attributable to better intracellular autophagy homeostasis to more efficiently recycle nutrient, remove damaged organelles, maintain the stem cell, and prevent cellular senescence.^50,51 The fewer proliferating cells in the kidney of TB-11-treated IRI mice compared to those in TB-SC-treated IRI mice probably resulted from less kidney damage, which reduces the stimulation of cell proliferation. The molecular mechanism of beclin 1 in regulation of cell proliferation is unknown.

It appears there is a paradox regarding beclin 1’s effect on cell proliferation between normal cells and tumor cells, because beclin 1, as a tumor suppressor, inhibits tumorigenesis,¹⁸ but does increase cell proliferation in normal cells.^50,51 Therefore, normal cells have different response to beclin 1 from tumor cells. Maintenance of autophagy activity is required to defend against intracellular and extracellular insults, and autophagy-suppressed senescence could help maintain certain levels of cell proliferation.^50,51 The mild increase in proliferating cells in kidney tubules might not induce tumorigenesis.

Taken together, normal beclin 1 expression in kidney tubules protects the kidney against a diverse group of insults, maintains cell proliferation, and inhibits kidney fibrosis. Distal tubular beclin 1 suppresses kidney fibrosis without protecting against acute kidney damage (Figure 10). To our knowledge, this is the first test and proposal that beclin 1 peptide is a promising therapeutic agent for prevention against AKI, suppression of kidney fibrosis, and promotion of kidney recovery (Figure 10). In the future, beclin 1 peptide effect on blocking or retarding CKD progression in CKD model deserves to be confirmed.

Figure 10. Schematic model for the effect of kidney tubular beclin 1 on kidney fibrosis and recovery after kidney damage.

Beclin 1 protein is expressed in the kidney tubules. Normal beclin 1 function in proximal tubules is required for renoprotection, inhibition of kidney fibrosis and promotion of tubular epithelial cell proliferation and kidney recovery after injury. Normal beclin 1 function in distal tubules does not contribute significantly to renoprotection, but participates in suppression of kidney fibrosis, and maintains tubular epithelial cell proliferation. Exogenous beclin 1 peptides can protect the kidney from acute injury, inhibit kidney fibrosis and promote kidney recovery.

Supplementary Material

1

Supplementary Methods. Detailed experimental materials and methods including Mice lines and Genotyping, Kidney Disease models, Autophagy Analyses, Blood, and Kidney Samples Collection, Kidney Histopathology, Antibodies, Immunohistochemistry, Immunoblotting, Quantification of Ki67 Positive Cells in Kidney Tubule Segments, Cell culture and cell lines, Statistical Analyses, and References for supplementary method.

Supplemental Table 1 A list of genetically manipulated mouse lines

Supplemental Figure S1. The changes in autophagy flux in the kidneys and kidney fibrosis of Becn1^+/−, WT, and Becn1^FA mice after IRI

Supplemental Figure S2. The changes in p62 mRNA in the kidneys of Becn1^+/−, WT, and Becn1^FA mice before and after IRI

Supplemental Figure S3. The changes in autophagy flux in the kidneys and kidney fibrosis of Becn1^+/−, WT, and Becn1^FA mice after cisplatin injection

Supplemental Figure S4. The changes in plasma levels of blood urea nitrogen (BUN) at early and later phase, and kidney morphology as well as kidney injury markers at early phase in both ischemia reperfusion injury (IRI) and cisplatin nephrotoxicity (CN) models

Supplemental Figure S5. The changes in autophagy flux in the kidneys of Becn1^+/−, WT, and Becn1^FA mice after UUO

Supplemental Figure S6. The changes in autophagy flux in the kidney of WT, KT-Becn1^+/−, and DT-Becn1^−/− mice after IRI

Supplemental Figure S7. Normal Atg5 expression in kidney tubules exerts renoprotection against ischemia reperfusion injury

Supplemental Figure S8. Localization of proliferating cells in the kidney of IRI mice at 14 days after IRI

TRANSLATIONAL STATEMENT.

Higher beclin 1 activity reduces acute kidney injury (AKI), but its role in accelerating recovery and reducing propensity to fibrosis is unknown. We showed that high beclin 1 activity in kidney tubules promotes recovery and suppresses fibrosis after AKI caused by ischemia-reperfusion, cisplatin nephrotoxicity and ureteric obstruction. Absence of beclin 1 in distal tubules promotes fibrosis and slows recovery after AKI. Administration of exogenous beclin 1 peptides reduces kidney damage, suppresses fibrosis, and promotes recovery post-AKI. Beclin 1 has an important role in the pathophysiology of AKI and beclin 1 peptides may be a novel effective therapeutic agent for AKI.