D-mannose reduces cellular senescence and NLRP3/GasderminD/IL-1β-driven pyroptotic uroepithelial cell shedding in the murine bladder

SUMMARY

Aging is a risk factor for disease via increased susceptibility to infection, decreased ability to maintain homeostasis, inefficiency in combatting stress, and decreased regenerative capacity. Multiple diseases including urinary tract infection (UTI), are more prevalent with age; however, the mechanisms underlying the impact of aging on the urinary tract mucosa and the correlation between aging and disease remain poorly understood. Here, we show that, relative to young (8-12 weeks) mice, the urothelium of aged (18-24 months) female mice accumulates large lysosomes with reduced acid phosphatase activity and decreased overall autophagic flux in the aged urothelium, indicative of compromised cellular homeostasis. Aged bladders also exhibit basal accumulation of reactive oxygen species (ROS) and a dampened redox response, implying heightened oxidative stress. Furthermore, we identify a canonical senescence-associated secretory phenotype (SASP) in the aged urothelium, along with continuous NLRP3-inflammasome- and Gasdermin D -dependent pyroptotic cell death. Consequently, aged mice chronically exfoliate urothelial cells, further exacerbating the age-related urothelial dysfunction. Upon infection with uropathogenic E. coli, aged mice harbor increased bacterial reservoirs and are more prone to spontaneous recurrent UTI. Finally, we discover that treatment with D-Mannose, a natural bioactive monosaccharide, rescues autophagy flux, reverses the SASP, mitigates ROS and NLRP3/Gasdermin/IL-1β -driven pyroptotic epithelial cell shedding in aged mice. Collectively, our results demonstrate that normal aging affects bladder physiology with aging alone increasing baseline cellular stress and susceptibility to infection and suggest that mannose supplementation could serve as a senotherapeutic to counter age-associated urothelial dysfunction.

Graphical Abstract

In brief (eTOC)

Joshi and Salazar et al., demonstrate that aging elevates risk of rUTIs and disrupts cellular homeostasis in the bladder. Urothelial cells exhibit heightened oxidative stress, senescence, and spontaneous cell shedding. Treatment with D-Mannose, a natural sugar, ameliorates age-associated bladder dysfunction and restores cellular homeostasis, supporting its potential as a serotherapeutic.

INTRODUCTION

The average age of the world’s population is increasing. By 2050, projections estimate there will be more than twice as many 65-year-olds than children less than 5¹. Aging impacts the organism at the cellular, tissue, and organ level. Further, a plethora of chronic and severe diseases arise as a consequence of the organism’s decreased ability to maintain homeostasis or combat stress, increased susceptibility to infection^2-5, and decreased regenerative capacity⁶. Aging alters the immune system’s ability to mount both adaptive and innate immune responses, and promotes chronic stimulation of low-grade inflammation (inflammaging)⁷. Aging also contributes significantly to increased risk for several diseases and conditions of the urinary tract, including urinary incontinence and urinary tract infections (UTIs), especially in postmenopausal women^8,9. UTIs are the most common bacterial infection with more than 404.6 million cases reported worldwide in 2019¹⁰ and recurrence of infection (rUTI) is a major complication¹¹. The main causative organisms of UTIs are uropathogenic Escherichia coli (UPEC)¹². UPEC pathogenesis events have been well defined in animal models and human bladder epithelial cells. Briefly, to invade the urothelium, UPEC employ filamentous adhesive organelles called type 1 fimbriae with a terminal protein, FimH – a C-type lectin with high affinity for Mannose residues on uroplakin receptors that line the urothelial cell surface^13-15. Following entry into fusiform vesicles from which they escape into the cytoplasm^16-18, UPEC establish intracellular bacterial communities (IBCs)¹⁹. Recognition of intracellular bacteria via host TLR-4/NLR receptors triggers the production of reactive oxygen species (ROS)²⁰ and activates autophagic processes to expel bacteria into the bladder lumen^21,22. thereby restricting further spread of UPEC. Urothelial cells can also be shed into the lumen to expel most of the UPEC-laden cells¹³. Shedding further facilitates the recruitment of innate immune cells, including polymorphonuclear leukocytes, monocytes, and macrophages to attack the bacteria^23-25. Immune infiltration, along with the inflammatory response, limits further invasion of UPEC. The bladder also activates antioxidant pathways via NRF2 to prevent high ROS build-up in the urothelium^20,26, as this balance is essential to avoid excessive cell damage. However, despite the innate cellular and molecular responses employed by the bladder immune milieu, UPEC form quiescent intracellular reservoirs (QIRs)²⁷, bacterial communities that establish within terminally-differentiated superficial cells and/or underlying transitional epithelial cells by co-opting autophagy and persisting within autophagosomes for months and can re-start the cycle of infection^22,26,28. Many of these elements have been shown to occur in women with rUTIs (e.g. ^3,29-35).

Recent work in mice and women has shown that aging alone initiates significant changes to the immune landscape of the aged female urinary tract including the formation of bladder tertiary lymphoid tissues (bTLT)^36-40. This raises important questions about how the urothelium is affected by aging and related immune changes, and what mechanistic role these changes might play in the development of UTIs and their recurrence. There are few previous studies of aging effects on key baseline aspects of the bladder’s ability to manage inflammation and respond to infection.

In this study, using young and aged female mice with and without induced UTI, we provide evidence of significant cellular and molecular alterations in the aged bladder epithelium at baseline. Cytological and immunological assays and transcriptomic analyses revealed substantial alteration in regulation of autophagy, quantity of lysosomes, and mitochondrial function, concomitant to a dampened antioxidative response and an exacerbated immune response. Most importantly, correlating with what is observed in human patients, aged mice showed the effects of having increased number of bacterial reservoirs, including frequent spontaneous recurrence of infection, and increased urothelial shedding We also show that the aged epithelium exhibits a senescence-associated secretory phenotype (SASP) and spontaneous, chronic pyroptotic cell death via activation of NLRP3 inflammasome and GsdmD. Treatment of aged mice with D Mannose reverses the SASP, promotes autophagy-lysosomal activity, ameliorates oxidative stress, and limits epithelial pyroptosis. Our data suggest the potential use of mannose supplementation as a senotherapeutic to treat age-associated urothelial dysfunction.

RESULTS

Aged bladders exhibit altered urothelial cellular architecture, and reduced autophagic flux and lysosomal function.

Bladder tissues from young (8-12 weeks old) and aged (15-18 months old) female mice were compared. Aged bladder epithelial cells showed accumulation of cytoplasmic vesicular structures (Figure 1A). Aged bladder tissue also showed, as expected, sub-epithelial follicular lymphocytic foci that we previously have characterized as age-induced tertiary lymphoid tissue (bTLT) (Figure 1A)³⁸. In aged compared to young mouse urothelium, transmission electron microscopy (TEM) revealed a 4-fold increase in the number and 8-fold increase in mean size of lysosomes (Figure 1B-D). The accumulation of lysosomes in aged urothelium was underscored by staining with LysoTracker, a fluorescent dye that concentrates in lysosomes (Figure 1E, Figure S3A). Similar lysosome congestion has been reported previously in aged rat bladder epithelia⁴¹. Lysosomal accumulation can indicate either increased lysosomal activity in the cell or decreased lysosomal hydrolase activity such that lysosomal function, and therefore, turnover is impaired. Thus, we probed lysosomal function using an assay to detect enzymatic activity of the lysosomal enzyme acid phosphatase (ACP). ACP activity was significantly lower in aged (~64% reduction) compared to young bladders (Figure 1F), consistent with lysosomal congestion due to impaired function.

Figure 1. Aged bladders exhibit altered urothelial cellular architecture, and reduced autophagic flux and lysosomal function.

(A) H&E staining shows sub-epithelial follicular foci (arrowheads) and bTLT (arrow). (B) TEM shows lysosome (white) and enlarged lysosome (black). (C) Number of lysosomes and (D) lysosome size in the urothelium. (n=9 images from 5 mice each group, p-values by 2-tailed unpaired t-test). (E) Lysosome staining shows accumulated lysosomes in aged urothelium. (F) ACP assay. (n=8 young and 9 aged, p-values by Mann-Whitney test). (G) WB quantitation of pS6 (n=5 each group, p-values by Mann-Whitney test). (H) WB quantitation of pAMPK (n=5 each group, ns=non-significant by Mann-Whitney test). (I) qRT-PCR analysis of AMPK alpha1 and AMPK alpha2. (n=4 each group, ns=non-significant by two-tailed unpaired t-test). (J) WB quantitation of LC3 I and II. (n=5 young and 4 aged, p-values by Mann-Whitney test). (K) WB quantitation of p62 (n=5 each group, p-values by Mann-Whitney test).

Lysosomal and mitochondrial responses and functions are tightly associated^42-45. Thus, we examined mitochondria in aged bladder cells. In almost all eukaryotes, mitochondrial range from 0.5 to 3μm, and their shape and number vary depending on cell types⁴⁶. MitoTracker staining revealed that the mitochondria in aged mouse urothelium were not only more abundant but were significantly larger (>8-fold) than those observed in a young bladder (Figure S1). Lysosomal accumulation is often associated with dysregulated autophagy^22,26,28, which, in turn, is regulated by the mTORC1-AMP Kinase (AMPK) axis⁴⁷. The mTOR complex 1 (mTORC1) in particular suppresses autophagy. Phosphorylation of the small ribosomal subunit 6 (phosphoS6) depends on mTORC1 phosphorylation of S6 kinase. Thus, phosphoS6 abundance in tissue is a faithful surrogate for mTORC1 activity^48,49. Aged bladders had ~2.5-fold increase in phosphoS6 (Figure 1G) relative to young bladders, suggesting that aging was associated with increased mTORC1 activity. Activation of AMPK by phosphorylation or changes in mRNA levels of AMPK alpha 1 and 2 (AMPK alpha catalytic subunit) was not consistently different between aged and young bladders (Figure 1H-I). To examine steady state autophagic flux, we measured the unlipidated (LC3I) and lipidated (LC3II) forms of LC3, the latter is a protein that is lipidated during autophagosome formation, and p62, a protein that targets cellular structures into autophagosomes. Aged bladders had increased LC3I and LC3II protein levels with a significant increase in LC3II/LC3I ratio (Figure 1J) and in p62 abundance (~60%) (Figure 1K). LC3II requires active autophagy to form but requires completion of autophagy via lysosomal fusion and lysosomal enzymatic activity to be degraded, and p62 degradation also requires autophagosome-lysosome fusion and activity. Thus, the results are consistent with an overall pattern of increased mTORC1 activity with a block in late-stage autophagy, increased formation of autophagosomes and autolysosomes with decreased turnover of those structures resulting in reduced autophagic flux.

Aged urothelium exhibits senescence-associated secretory phenotype and a blunted antioxidant response.

Next, we dissected the global status of transcripts in young and aged mice to evaluate the molecular signatures in the bladder in response to aging. RNA-Seq transcriptomic data revealed a robust and unique aged bladder signature (Figure S2A) with increased expression of 1,038 genes and decreased expression of 376 genes (Figure S2B). Gene set enrichment using the MSigDB Hallmark compendium revealed that in addition to changes in immune and inflammatory pathways, allograft cell/tissue rejection, cell death, oxidative stress, fatty acid synthesis, and mTOR/autophagy were identified as the top enriched pathways (Figure S2C). The elevation in immune inflammatory pathways was consistent with our previous report describing altered inflammatory profile and formation of tertiary lymphoid structures³⁸.

Inflammation, oxidative stress, as well as increased lysosomal and mitochondrial size all occur during the cell-based program known as senescence^50,51. We reasoned that the aged urothelium could be undergoing a cellular senescence program even at steady state. Senescent cells display distinctive features like growth arrest, unresolved DNA damage, high senescence-associated β-galactosidase (SA-β-gal) activity, and pro-inflammatory SASP, which is a component of inflammaging (i.e., chronic inflammation that develops with advanced age, and can help spread senescence to adjacent cells)⁵². Consistent with senescence chronically occurring in aged bladder cells, we found positive SA-β-gal staining in aged compared to young bladders (Figure 2A). Senescent cells also shed portions of chromosomes/DNA from the nucleus, forming cytoplasmic chromatin fragments (CCFs) that can be marked by cytoplasmic accumulation of the DNA damage marker, γ-H2AX^53,54. We observed that aged bladders exhibited high intensity γ-H2AX staining, indicative of CCFs further solidifying the presence of DNA damage response in aged bladder (Figure 2B). Both SA-β-gal and γ-H2AX staining are observed exclusively in aged bladders. There was also increased expression of transcripts for the cell cycle arrest/DNA damage response genes p16 (~80%), p21 (~60%), and p53 (~25%) in aged compared to young bladders (Figure 2C-E). Furthermore, multiplex analysis of urine samples from aged versus young mice showed a significant increase in cytokines, chemokines, and monocyte chemoattractant proteins, which recruit immune cells to clear senescent cells. We observed high expression of G-CSF, IL-1α, IL-6, IP-10, MCP-1, MIP-1α, and MIP-1β in aged urine samples compared with those from young mice (Figure 2F). Together, these findings demonstrate that the aged bladders display a senescence-associated secretory phenotype.

Figure 2. Aged urothelium exhibits senescence-associated secretory phenotype (SASP) and a blunted antioxidant response.

(A) SA-β-gal staining. (n=5 each group). (B) γ-H2AX staining. (n=5 each group). (C-E) qRT-PCR analysis of p16, p21, and p53. (n=4 each group, p-values by Mann-Whitney test). (F) Multiplex ELISA assay (n=5 each group, p-values by Mann-Whitney test). (G) DHE staining. (arrows=nuclear DHE). Quantitation of DHE. (n=7 images from 7 mice each group, p-values by two-tailed unpaired t-test). (H) qRT-PCR analysis of Nrf2, Nqo1, Gclc, and Hmox1. (n=5 each group, ns=non-significant by two-tailed unpaired t-test).

We previously showed that altered autophagy or lysosomal function leads to impaired antioxidant response resulting in the accumulation of ROS in the context of UTIs²⁰. To investigate the effect of aging on ROS production, we used dihydroethidium (DHE), a dye that hydrolyzes upon interaction with ROS and stains the nucleus. This revealed high signal intensity in the aged urothelium, and quantification of the signal indicated higher levels of ROS (~100% higher) in the aged urothelium relative to the low baseline levels in the young bladders (Figure 2G, Figure S3B). However, despite markedly elevated ROS, the normal cellular antioxidant response genes (Nrf2, Nqo1, Gclc, and Hmox1) were not increased relative to control levels (Figure 2H), indicating ROS may continuously accumulate in aged bladders because of a diminished antioxidant response. Overall, these findings demonstrate that aged bladder epithelial cells constitutively induce a cellular senescence program even in an otherwise unperturbed state.

Aging is associated with increased NLRP3/GsdmD/IL-1β axis in the urothelium and pyroptotic epithelial cell death.

Next, we sought to further characterize the mechanisms contributing to the hyper-inflammatory state and consequences of a senescent and oxidatively stressed bladder. The NLRP3 inflammasome is an important component of the innate immune system that modulates caspase-1 activation and the secretion of pro-inflammatory cytokines in response to microbial infection in epithelial cells and macrophages^55-59. Together with our whole bladder RNA Seq data showing multiple gene pathways involved in inflammatory responses and cell death, we reasoned that excessive ROS and SASP could trigger a baseline inflammatory and cell damage response. Indeed, we observed that aged bladders have significantly elevated levels of Nlrp3 (Figure 3A). Given that NLRP3 as part of the inflammasome that converts inactive pro-caspase 1 into its mature form, we investigated the levels of pro- and cleaved caspase-1 in young and aged bladders and found significant accumulation of the mature form in aged bladders (Figure 3B). Caspase-1 cleaves pro-IL-1β into IL-1β⁶⁰. Accordingly, we found that pro-IL-1β and cleaved IL-1β were both significantly higher in aged compared with young bladders (Figure 3C). Caspase-1 cleaves a member of the family of proteins called Gasdermins, namely Gasdermin D, into two fragments, which plays a pivotal role in releasing IL-1β by forming membrane pores that increase cell membrane permeability and induces pyroptosis⁵⁷. We detected increased expression of transcripts for Gasdermin family members, GsdmA, GsdmC, GsdmD, and GsdmE in aged bladders (Figure 3D). The Human Protein Atlas indicates that the predominant Gasdermin protein expressed in human bladder epithelium is GSDMD⁶¹. Thus, we further investigated protein expression of GsdmD in aged murine bladders and found significantly increased expression of cleaved GsdmD resulting in N-terminal (GsdmD-NT) fragments (Figure 3E) that form the membrane pore through which IL-1β is released. Thus, the aged bladder exhibits high levels of the NLRP3 inflammasome and a steady-state activated caspase-1/GsdmD/IL-1β axis (Figure 3F).

Figure 3. Aging is associated with increased NLRP3/GsdmD/IL-1β axis in the urothelium and pyroptotic epithelial cell death.

(A) qRT-PCR analysis of Nlrp3. (n=5 each group, p-values by Mann-Whitney test. (B) WB quantitation of cleaved Caspase-1. (n=5 each group, p-values by Mann-Whitney test). (C) WB quantitation of pro-IL-1β and mature IL-1β. (n=5 each group, p-values Mann-Whitney test). (D) qRT-PCR analysis of GsdmA, GsdmC, GsdmD, GsdmE. (n=5 each group, p-values by Mann-Whitney test). (E) WB quantitation of cleaved GsdmD. (n = 5 animals in each group, p-values by Mann-Whitney test. (F) Graphical representation of GsdmD-mediated pyroptosis. (G) Representative urine cytology images show exfoliated bladder epithelial cells. Quantitation of exfoliated/shed epithelial cells. (n=9 each group, p-values by two-way ANOVA Bonferroni's multiple comparisons test).

Next, we determined if the constitutively increased NLRP3/GsdmD/IL-1β axis in the urothelium would affect tissue homeostasis. Indeed, cytological examination of urines from young and aged mice revealed a significant level of spontaneously shed epithelial cells in aged urine samples (~80% increase) (Figure 3G). This finding is similar to what has been demonstrated in urine samples from aged/postmenopausal women³⁰. Incidentally, urines from these women exhibit high levels of the proinflammatory cytokine, Interleukin-6 (IL-6). In aged mouse bladder tissue, we also note a significant increase in IL-6 (Figure 2F). IL-6 has been shown to prime or sensitize cells for NLRP3 inflammasome activation as it can upregulate the expression of NLRP3 and its associated components, including pro-caspase-1, making the cells more responsive to NLRP3 stimuli⁶². Examination of urines of aged (15-month-old) IL-6^−/− female mice showed that in fact, epithelial shedding is significantly decreased (Figure S4). Together, our results suggest that in the aged bladder, high levels of IL-6 and NLRP3 inflammasome is associated with activated caspase-1/IL-1β/GsdmD cascade leading to pyroptotic cell death, which may be one of the contributors of the spontaneous and constitutive epithelial shedding.

Aging exacerbates UTI recurrence.

Postmenopausal/elderly women have a higher risk and frequency of UTIs than younger premenopausal women^8,9. We next evaluated whether the epithelial and immune changes we have delineated in our aged bladders would impact UTI susceptibility and/or pathogenesis in aged mice. Young and aged mice were infected with a clinical UTI isolate, UTI89, and infection course and outcomes determined. Initial urine titers between young and aged mice were similar (Figure 4A). However, infected aged bladders were robustly inflamed compared with infected young bladders 24 hours post infection (hpi) (Figure 4B) with increased influx of mononuclear cells observed in urine (~100% increase) (Figure 4C). We determined how aging-associated inflammatory response and cellular changes affects UPEC pathogenesis of UPEC, in the urothelium and found that infected aged bladders showed a 50% increase in IBC formation, detected as IBC-containing superficial cells shed in urine during the time 6 to 72h time window of an acute UTI (Figure 4D-E).

Figure 4. Aging exacerbates UTI recurrence.

(A) Bacterial load quantitation at 6hpi. (n=8 each group, ns=non-significant by two-tailed unpaired t-test). (B) Inflammation score at 24hpi. (n=12 each group, p-values by two-tailed unpaired t-test. (C) Number of monocytes at 24hpi. (n=7 each group, p-values by two-tailed unpaired t-test). (D) Representative urine cytology images at 24hpi show immune cells (arrowheads) and cells with IBC (arrow). (E) Number of IBC at different timepoints. (n=6 at 6hpi, n=12 at 24hpi, n=6 at 48hpi, n=6 at 72hpi, p-values by two-way ANOVA Bonferroni's multiple comparisons test). (F) IF localization of E. coli (green) forming QIR (arrow) in aged superficial cell. E-cadherin (red) marks cell membrane. Number of QIRs formed at 14dpi. (n=15 each group, p-values by two-tailed unpaired t-test). (G) Quantitation of mice with bacteriuria at different timepoints. (young group: 1dpi=38, 2dpi=38, 3dpi=39, 7dpi=19, 10dpi=13, 14dpi=13; for aged group: 1dpi=41, 2dpi=41, 3dpi=42, 7dpi=28, 10dpi=19, 14dpi=19, p-values by two-way ANOVA Bonferroni's multiple comparisons test). NOTE: At 7dpi, number of surviving mice decreased rapidly due to long-term exposure to infection thus affecting the sample size towards the end of the 14-day study period.

We have previously shown that a subset of UPEC can persist within epithelial cells in the intact epithelium in the form of QIRs²⁷. Quantification of QIRs at 14 days post infection (dpi) in infected mice using an E. coli-specific antibody revealed a significant increase in the number of QIRs in infected aged bladder relative to infected young bladders (Figure 4F). Finally, commensurate with increased QIRs that can seed recurrent infections, a significantly higher percentage of infected aged mice exhibited spontaneous recurrence of bacteriuria (Figure 4G). We noted that at 7dpi, the number of surviving aged mice decreased rapidly due to long-term exposure to infection suggesting increased mortality risk. Overall, our findings indicate that the uninfected aged bladder exhibits a significantly higher level of inflammation and infected aged bladders exhibit increased bacterial invasion and proliferation and persistence. These observations provide an explanation for the increased susceptibility of aged mice to spontaneously recurrent UTIs.

D-mannose treatment ameliorates age associated SASP, inflammation, and epithelial pyroptosis.

Several clinical investigations have described the benefits of D-mannose in patients suffering from rUTI^63-65. D-mannose is known to bind to UPEC adhesin protein FimH and thus prevent them from adhering to bladder epithelial cells (reviewed in ⁶⁶). However, D-mannose has also been shown to dampen inflammation⁶⁷. Here, we sought to test the hypothesis that D-mannose ameliorates the severity of aging associated decline in bladder structure and functionality. We treated aged mice with 1.1M D-mannose in drinking water for 14d , a dose adjusted according to that used in clinical studies⁶³. As expected from human studies, D-mannose administration did not have any adverse effects on weight or overall health of the mice. A subset of aged mice who were infected with UPEC for 14d and treated with D-mannose alone also did not show a change in urinary bacterial load, consistent with D-mannose not having bacteriostatic or bactericidal activity. However, in D-mannose treated aged mice, we not only observed a significant increase in the AMPKalpha2 subunit, indicative of autophagy activation (Figure 5A, Figure S5) but also a significant reduction in the LC3II/LC3I ratio when compared with untreated aged mice (Figure 5B) and a significant reduction in p62 (~65%), indicative of increased autophagy flux (Figure 5B). Next, we determined functionally whether increased flux would improve lysosomal activity and found that D-mannose treatment improved ACP enzymatic activity (~89%), indicating improved/rescued lysosomal activity (Figure 5C). Together, our results suggest that D-mannose administration reverses or ameliorates the defects in autophagic flux and lysosomal enzyme activity.

Figure 5. D-mannose treatment ameliorates age-associated SASP, inflammation, and epithelial pyroptosis.

(A) qRT-PCR analysis of AMPK alpha1 and AMPK alpha2. (n=4 each group, ns=non-significant and p-values by Mann-Whitney test). (B) WB quantitation of LC3, and p62. (n=4 untreated and 5 D-mannose-treated aged mice for LC3; n=5 each group in p62, p-values by Mann-Whitney test). (C) ACP assay. (n=9 each group, p-values by Mann-Whitney test). (D) WB quantitation of Beta-galactosidase. (n=5 each group, p-values by Mann-Whitney test. (E) qRT-PCR analysis of p16, p21, and p53. (n=4 each group, ns=non-significant and p-values by Mann-Whitney test). (F) WB quantitation of IL-10. (n=5 each group, ns=non-significant, p-values by ANOVA with Tukey’s multiple comparisons test). (G) qRT-PCR analysis of Gclc and Hmox1. (n=5 each group, p-values by Mann-Whitney test). (H) qRT-PCR analysis of Nlrp3. (n=5 each group, p-values by Mann-Whitney test). (I) WB quantitation of cleaved Caspase-1. (n=5 each group, p-values by Mann-Whitney test). (J) WB quantitation of mature IL-1β. (n=5 each group, p-values by Mann-Whitney test). (K) WB quantitation of cleaved GsdmD. (n=5 each group, p-values by Mann-Whitney test). (L) Representative urine cytology images from untreated and D-mannose-treated aged mice show epithelial cell shedding. (M) Quantitation of exfoliated/shed epithelial cell. (n=9 each group, p-values by two-way ANOVA Bonferroni's multiple comparisons test). (N) Inflammatory cell score. (n=7 each group, p-values by Mann-Whitney test).

We also determined whether the positive impact of D-mannose in restoring autophagy flux and lysosomal activity would be evident in young mice. Not unexpectedly, we did not observe a change in autophagy markers, LC3II/I and P62 with D-mannose administration as there are no reported baseline defects in autophagy flux in young healthy mice (Figures S6-7). Further, no significant difference was observed in the ACP enzymatic activity between young (untreated and treated) and aged (treated) bladders (Figure S8). Thus, D-mannose appears to be more effective as a therapeutic intervention than a preventative measure.

D-mannose administration also induced a significant decrease in the level of β-galactosidase in aged bladders (~55%) (Figure 5D) suggesting reduced senescence and SASP. Accordingly, we observed reduced transcript levels of p21 (~45%) and p53 (~65%) in the D-mannose-treated aged bladder compared with the untreated group although p16 levels remained unchanged (Figure 5E). Next, we determined whether D-mannose activated anti-inflammatory cytokines and indeed, we noted a significant decrease in IL-10, an anti-inflammatory cytokine which can limit release of IL-1β and IL-6, production in aged bladders treated with D-mannose (Figure 5F). In addition, significant increase in the levels of targets of Nrf2 such as Gclc and Hmox1 was noted in D-mannose-treated aged mice (Figure 5G). Significantly, D-mannose treatment of aged mice decreased NLRP3 expression from the high levels associated with age with a concomitant decrease in caspase 1 cleavage and GsdmD cleavage and IL-1β production (Figure 5H-K), Consequently, we observed a significant decrease in spontaneous epithelial cell shedding from uninfected D-mannose-treated aged mice urine compared with uninfected non-treated aged mice (~75% decrease) (Figure 5L-M). Finally, cytological quantification of inflammatory cells in urine samples revealed a lowered inflammation score in D-mannose-treated aged mice when compared with the untreated aged group (Figure 5N). Our findings suggest that D-mannose significantly improves age-associated dysfunction by rescuing the block in autophagic flux, SASP, inflammation, and urothelial cell pyroptosis.

DISCUSSION

The aging population is steadily increasing, creating increasing need to provide healthcare and treatments capable of meeting the increased susceptibility to infectious diseases and other conditions that occurs in aging. In the lower urinary tract, especially in women, aging combined with postmenopausal status exacerbates multiple conditions of the urinary tract including rUTIs. There is a great need to understand impact of aging and age-related changes in the urinary tract mucosa, and to delineate the potential mechanisms driving this increased susceptibility. In this study, we demonstrate that bladder urothelial cells undergo significant cellular and biochemical modifications simply because of normal aging alone, including the development of SASP, sustained oxidative stress, lysosome dysfunction, and decreased autophagic flux. We show that aging is also associated with increased inflammation, pyroptotic cell death, and susceptibility to rUTIs. We further demonstrate that oral administration of D-mannose, a natural bioactive monosaccharide, limits SASP and inflammatory death of urothelial cells.

Lysosomes, cellular organelles that contain enzymes to break down waste materials and cellular debris in post-mitotic cells progressively accumulate oxidatively modified macromolecules and defective organelles that are not fully degraded with age. Lysosomal dysfunction is commonly observed in age-related neurodegenerative diseases including Alzheimer’s and Parkinson’s and impaired lysosomal activity has been shown to play an important role in the development of these disorders^68,69. We found significant modifications in the lysosomal component of the uninfected aged urothelium, including low acid phosphatase activity, and increased number and size of lysosomes. Our findings align with previously reported work showing accumulation of an aging pigment, lipofuscin, in aged urothelial cells⁷⁰, decreased acidification, loss of cathepsin B activity, and expanded endolysosomal compartments in the aged rat urothelium⁴¹. A decrease in lysosomal degradation can lead to altered autophagy, including defective mitophagy, further leading to the accumulation of ROS and disrupted cellular function. Indeed, we noted an increase in the size and number of mitochondria in addition to lysosomes in the aged urothelium. Cellular respiration and mitochondrial bioenergetics in aged urothelial cell cultures have recently been shown to have a reduced mitochondrial Ψm, oxygen consumption rate, and ATP release⁷¹. Mitochondrial accumulation during high oxidative conditions is also attributed to autophagy dysfunction⁷².

We and others have shown that dysfunctional autophagy leads to downregulation of the antioxidant response^20,73. In agreement with previous findings⁷¹, we report here that the aged bladder has elevated levels of ROS. If mitochondrial accumulation leads to more ROS and more oxidative damage, one could expect that the antioxidative process would be activated to counteract high oxidative stress²⁰. However, the levels of antioxidant response-related genes were not activated in aged mice compared to young mice, suggesting an ineffective or a defective antioxidant response, which may lead to an unregulated increase of inflammation in aged bladders. Higher ROS activity that exacerbates oxidative damage could lead to apoptosis or senescence. Thus, aging of the bladder overall increases the initiation or activation of a constitutively stressed phenotype.

In this study, we further determined that the aged urothelium displays multiple hallmarks of cellular senescence including increased SA-β-galactosidase activity, and the acquisition of a pro-inflammatory SASP. Senescent cells can exhibit a compromised nuclear envelope allowing chromatin fragments to bud off and enter the cytoplasm to be degraded via an autophagosome-lysosome process⁵⁴ . Lysosomes play an important role in limiting generation of SASP, which is in part driven by the presence of cytoplasmic chromatin fragments (CCF) ^74-76. Reduced autophagic flux and lysosomal function that was evident in our aged bladder epithelial cells would suggest that chromatin fragments could accumulate in the cytoplasm. Indeed, we observed the presence of γ-H2AX positive staining in cytoplasm of aged urothelial cells but not in young cells, suggesting activation of the DNA stress pathway and the presence of CCF in senescent urothelial cells. There are also other factors that contribute to the induction of γ-H2AX foci in the cells such as oxidative and replication stress, which was evident in aged bladders. Thus, the aged urothelium is highly senescent.

Accumulation of DNA in the cytoplasm serves as a potent danger signal that can activate an innate immunity cytosolic DNA sensing pathway, leading to proinflammatory responses. In fact, SASP is associated with various growth-regulatory factors, cytokines, and chemokines⁷⁷. We found a significant increase in secreted soluble factors such as IL-1α, IL-6, and MIP-1α in aged mouse urines and tissues. The role of cytoplasmic DNA in age-associated inflammation in several diseases is beginning to be described. For example, premature aging syndrome such as ataxia telangiectasia (AT), a severe neurodegenerative syndrome, is characterized by accumulation of γ-H2AX-enriched cytoplasmic DNA along with the expression of inflammatory genes, including IL-6⁷⁵. Previous work from our group has shown that the aged urothelium has increased permeability characterized by exposure of underlying urothelial cells and the stromal compartment to urinary content³³, detrimental to the normal urothelial function of maintaining an impermeable barrier to waste products and cytotoxic factors in urines⁷⁸. We posit that chronic exposure to urinary cytotoxic irritants could trigger DNA damage secondary to accumulation of ROS and drive CCF formation, leading to an SASP-associated inflammatory response coupled with age-associated dysfunctional lysosomal function and mitophagy defects.

The outcomes of these age-associated changes appear to be inflammatory/lytic cell death of the urothelial cells via activation of an inflammasome/GsdmD/IL-1β axis triggered by the NLRP3 inflammasome. Activation of this lytic cell death pathway (pyroptosis) has been shown to lead to pro-inflammatory cytokine IL-1β release⁷⁹. We and others have previously shown that UPEC infection induces IL-1β mediated pyroptosis in bladders and in macrophages^58,59. Bone marrow derived macrophages isolated from Nlrp3^−/− mice demonstrated that the increased IL-1β production upon infection was dependent upon NLRP3 inflammasome and caspase-1 activation and inhibition of IL-1β signaling modified the UPEC infection response⁵⁸. Importantly, limiting NLRP3 (via Glyburide) or caspase-1 or IL-1β (via Anakinra), ameliorated the condition indicating a direct link. Our current study suggests that aging itself, as a condition, is sufficient to activate pyroptosis and facilitate the release of pro-inflammatory cytokines by increased activation of NLRP3, caspase-1, and IL-1β akin to an acute infection condition. Indeed, aged mice deficient for IL-6 which can prime the NLRP3 inflammasome, exhibit reduced shedding, providing further support for a direct association between inflammasome activation and pyroptosis in the aged bladder. The release of IL-1β during pyroptosis can further stimulate IL-6 production, creating a positive feedback loop (e.g., ⁸⁰). This loop would amplify the inflammatory response as we note in aged bladders. These findings support a direct link between IL-6 and NLRP3 and pyroptosis. Interestingly, spontaneous epithelial shedding in response to inflammatory triggers has been noted in multiple tissues and conditions including NLRP3 inflammasome-driven epithelial pyroptosis in allergic rhinitis⁸¹; in astrocyte pathological injury^82,83; the aging dry eye syndrome and more⁸⁴. Pyroptosis has been implicated in the etiology of nerve cell injury and death in neurodegenerative diseases including Alzheimer’s disease and Parkinson’s disease⁸⁵. The changes we have noted in the aging urothelium could constitute urothelial degeneration and underlie a number of bladder diseases. We suggest that mechanisms of senescence and pyroptosis may be common to multiple tissues with age and should perhaps be investigated in a pan-tissue manner using mice deficient in genes in the pyroptosis pathway.

The excessive shedding of urothelial cells captured in urines of aged female mice in the current study and noted previously in postmenopausal women³⁰ could be responsible in part for the dysregulated urothelial barrier function³³. The regenerative capacity of the urothelium in response to injury/infection/stress is notably robust⁸⁶. However, it is likely that the aged urothelium may be compromised in its capacity to regenerate new luminal cells to replenish the continuously shedding cells into the urine. Our findings that aged urothelial cells present with increased cell cycle arrest/DNA damage response genes (p16 and p21) supports the notion that urothelial cells undergo senescence and are likely unable to fully restore barrier damage. Interestingly, activated p21 has been shown to further induce ROS production, thus reinforcing the response. Given our findings that the antioxidant response via NRF2 and its cognate pathway components is compromised, the constitutive stress and shedding cycles could significantly impact barrier function. Indeed, NRF2 signaling is known to protect cells from pyroptosis by attenuating the activation of the NLRP3 inflammasome⁸⁷. Furthermore, we have demonstrated that Nrf2^−/− mice also exhibit spontaneous urothelial shedding in urine and the shed cells are highly ROS-positive²⁰. Thus, urothelial shedding in response to unmitigated ROS could represent an inability to repair damaged/stressed cells. This aging phenotype coupled with the inability to restore the lost cells could underlie multiple urological disease processes. If so, early interventions to block ROS, limit SASP, and promote urothelial repair in aging and other conditions are warranted.

Frequency of rUTIs is significantly higher in postmenopausal women⁸⁸. There are multiple hypotheses for the increase including reduction of systemic estrogen levels, reduced detrusor activity, and an altered urobiome, among others⁸⁹. Recent work from our group and others has identified a potentially critical contributor to an altered bladder mucosal environment with age that is responsible for inflammaging in mice and in women. In mice, our work comparing young and aged bladders indicated a significant change in the immune repertoire in the latter³⁸. Single-cell resolution, transcriptomic map of resident immune cells within bladder tissue of both young and aged mice revealed subsets of macrophages and dendritic cells (DCs), significant enrichment in B and T cells, which spontaneously formed organized bladder tertiary lymphoid tissue (bTLT) with bone fide germinal centers that arise in an age-dependent manner coincident with loss of fertility in female mice³⁸. Notably, bTLT only forms in a sex-dependent manner in female bladders³⁶. We also recently demonstrated that these structures occur in bladders of post-menopausal women and are strongly predictive of incidence/frequency of recurrent UTIs, shorter time intervals between recurrent episodes^37,40. These structures have been associated with bacteria including E. coli in postmenopausal women ^90-92.In support of these findings in post-menopausal women, we find that aged mice infected with UPEC exhibit a spontaneous rUTI phenotype. Further, we noted a significantly higher numbers of QIRs, which can spontaneously emerge to give rise to rUTIs. We saw an increased LC3II/I ratio with age and a concomitant block in autophagy flux, which suggests that there is accumulation of autophagosomes that could provide a greater number of protected niches for QIR formation. The mechanistic and direct links between our murine studies and clinical findings of bTLT and increased frequency of rUTIs need to be better understood.

Finally, we report on the utility of a simple sugar monosaccharide, D-mannose, as a surprisingly effective senotherapeutic in limiting pyroptotic epithelial cell death, reversing cell cycle arrest, limiting SASP, and activating and restoring autophagic flux that is otherwise notably decreased in non-treated aged mice. D-Mannose is a sugar that is considered clinically as an alternative treatment to UTIs^63,93 as it binds to the adhesion protein FimH on uropathogens and prevents their adherence to bladder epithelial cells ⁹⁴ and has been known to decrease secretion of inflammatory cytokines⁹⁵. Our results show that D-mannose has a cell intrinsic impact in addition to the known extrinsic roles, as treatment reduces accumulation of SA-β-galactosidase and restores the lysosomal activity and autophagy flux noted in aged bladders. D-mannose administration also dampened the inflammatory/pyroptotic phenotype with inducing a significant decrease in NLRP3 /GsdmD/IL-1β expression and increased expression of Nrf2 transcriptional targets such as Gclc and Hmox1), which correlated with decreased epithelial shedding in mice, consistent with human cohort study documenting D-mannose-treatment associated decrease in uroepithelial shedding³³. Thus, uncovering the inflammatory mechanism underlying epithelial shedding pointed us to a pharmacological approach that inhibits this pathway and limits shedding. We also note that D-mannose treatment is associated with a drop in IL-10 abundance. IL-10 suppresses the production and release of pro-inflammatory cytokines⁹⁶, such as IL-6, which is constitutively increased in age, potentially signifying an improvement in the inflammatory condition persisting in aged bladders and thereby reduced need to dampen inflammation. Interestingly, treatment also affects cell cycle regulators, p21and p53 although not p16. p21 is mainly activated early in the initiation of senescence, whereas p16 maintains cellular senescence⁹⁷, which together suggest reduced senescence in the D-mannose-treated relative to untreated aged bladders. Thus, D-mannose might block early steps in the process of senescence and thereby limit further progression. This would make D-mannose a highly effective seno-morphic as it modifies senescence risk.

Limitations of the study:

our study does not establish a causal link between IL-6/NLRP3 activation and epithelial cell shedding in aged bladders. However, our findings that D-mannose treatment significantly reduces the components of the NLRP3/GasderminD/IL-1β axis; and mice deficient in IL-6 exhibit reduced shedding provide indirect support. Future studies to delineate the mechanisms that causally link pyroptosis and epithelial shedding in the bladder are warranted not only in the context of aging but also in infection, inflammation, and other bladder-related disorders. Our bulk whole bladder transcriptomics data indicating increases in inflammatory response and cell death pathways and decreases in pathways associated with protein degradation and epithelial mesenchymal transition might provide further clues to understanding interdependent urothelial-immune crosstalk and implications for disease pathogenesis. Additionally, we do not fully understand the mechanism(s) of how D-mannose modifies urothelial senescence, autophagy and affects pyroptosis. However, recent studies suggest that D-mannose treatment in macrophages limits the increase in glycolytic rate induced by pro-inflammatory triggers⁹⁸. Whether altered glycolysis leads to altered Tricarboxylic acid cycle dynamics, which could limit ROS in the aged urothelium remains to be determined.

Nevertheless, multiple studies support that D-mannose treatment appears to be an effective intervention against multiple inflammatory conditions. For example, mannose has been reported to protect against intestinal barrier dysfunction in a (young) mouse model of colitis⁹⁹, ameliorate autoimmune phenotypes in a mouse model of Lupus¹⁰⁰, attenuate bone loss in a mouse model of Estrogen deficiency¹⁰¹, prevent bone loss under weightlessness¹⁰² and shown to suppress osteoarthritis development and delay IL-1β-induced degeneration by promoting autophagy¹⁰³. Moreover, we have recently reported that D-mannose prophylaxis led to a significant decrease in UTI incidence rate in patients with bTLT⁶³. Thus, D-mannose treatment might produce far-reaching urothelial cell intrinsic and extrinsic reversal of age-associated changes in addition to limiting rUTI frequency. As the decrease in shedding, cell death, and senescence point to rehabilitation or improvement in the overall cellular and molecular status of the aging bladder, future investigation is warranted to understand the ways D-mannose treatment contributes to these processes and examine the potential benefits of D-mannose beyond rUTI prevention and treatment. D-Mannose supplementation holds promise as a therapeutic strategy to counter age-related urothelial dysfunction, offering a potential avenue to enhance the quality of life for the increasing elderly population.

STAR METHODS

RESOURCE AVAILABILITY

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Indira U. Mysorekar (Indira.mysorekar@bcm.edu).

Materials availability

This study did not generate new unique reagents.

Data and code availability

• This study does not report original code.

• Data are available on the Gene Expression Omnibus under the following identifier: GSE149571.

EXPERIMENTAL MODEL AND STUDY PARTICIPANTS

Mice

C57B6/J female mice (8-12 weeks: young; and 15-18 months: aged) were obtained from the National Institute of Aging and maintained at Washington University School of Medicine and Baylor College of Medicine. Standard rodent chow and water were available ad libitum throughout the experiment. Mice were housed in groups of 4 in a temperature- (22 ± 1 °C) and humidity-controlled vivarium with lights maintained on a 12:12 light/dark cycle. All animal experimental procedures were approved by the Institutional Animal Care and Use Committee at Washington University School of Medicine (Animal Welfare Assurance #A-3381-01) and Baylor College of Medicine (Animal protocol number AN-8629). All mice were humanely euthanized at the end of each experiment.

METHOD DETAILS

UPEC strain and infection

A clinical cystitis isolate of uropathogenic E. coli, UTI89, was grown in Luria-Bertani broth overnight in static culture at 37°C incubator the day before infecting mice. Adult female mice (2-mon- to 18-mon-old) were anesthetized and transurethrally-inoculated with 10⁷ CFU of UTI89 suspended in PBS as previously described in Justice et al ¹⁹.

D-Mannose treatment

Aged female mice were treated with 200 ml of drinking water supplemented with 1.1M D-mannose for 14 days⁶³. Control cohort was provided with equal volumes of un-supplemented water. Weights of the mice were recorded, and urine samples were collected for cytological preparation and analysis. Water consumption was also recorded/calculated. The dosing was adjusted for murine studies based on dose/body weight considerations in human studies⁶³.

H&E staining

Bladder sections were deparaffinized in 100% Histoclear (3x, 5mins each) and rehydrated in ethanol (100%, 90%, 70%, 50%; 5mins/concentration). Nuclei were stained in Hematoxylin (3mins) followed by rinsing in running tap water. Sections were dipped (3times) in acid alcohol solution, soaked for 5 minutes in sodium bicarbonate solution, followed by rinsing in distilled water. Staining of the cytosol was done immediately by dipping the slides 3 times in Eosin solution, soaking in increasing concentrations of ethanol (50%, 70%, 90%, 100%) at 3 minutes each, followed by a final dip in Histoclear. Sections were applied with xylene-based mounting medium, covered with cover glass, and edges were sealed with clear nail polish. Image capture was performed with a Panoramic Midi microscope (3DHISTECH Ltd, Hungary).

Transmission Electron Microscopy (TEM)

Young and aged mouse whole bladders were processed as described previously 23 to examine lysosomes and lipid droplets. The bladders were fixed with fixative containing 2% glutaraldehyde and 3% PFA in 0.1M sodium cacodylate. Samples were rinsed three times in sodium cacodylate buffer and post-fixed in 1% osmium tetroxide for 1 hour, stained in 1% uranyl acetate for 1 hour, rinsed and dehydrated, and subjected to critical point drying. Samples were then gold-coated and analyzed on a JEOL-1200 EX II Transmission Electron Microscope (JEOL, USA). For count and size determination of lysosomes, 9 images of five mice in each group were analyzed using ImageJ analysis software.

Lysosome Staining

Fixed, fresh-frozen sections of young and aged bladders were utilized for staining lysosomes with Lysotracker Red (L7528, Life Technologies) following manufacturer’s protocol with modifications. Sections were soaked and incubated in 100nM Lysotracker Red probe solution for 30 mins at 37 °C, rinsed in 1X PBS, and counterstained with DAPI. Sections were applied with Xylene-based mounting medium, covered with cover glass, and sealed edges with clear nail polish. Images were captured using Zeiss Axio Imager M2 Plus Wide Field Fluorescence Microscope (Carl Zeiss Inc, Thornwood, NY).

Mitochondria Staining

Fixed, fresh-frozen sections of young and aged bladders were utilized for staining of mitochondria with Mitotracker dye (M22426, Thermo Fisher Scientific, USA) following manufacturer’s protocol with modifications. Sections were soaked and incubated in 100 nM Mitotracker dye solution for 5 minutes at 37°C, rinsed in 1X PBS, and counterstained with DAPI. Sections were applied with Prolong Gold Antifade mountant (P10144, Thermo Fisher Scientific, USA), covered with cover glass, and edges were sealed with clear nail polish. Images were captured using Zeiss Axio Imager M2 Plus Wide Field Fluorescence Microscope (Carl Zeiss Inc, Thornwood, NY). Quantitation of mitochondrial size was performed on 5 images from 5 mice in each group using ImageJ.

RNA-Seq Analysis

Tissue preparation and RNA sequencing protocols were adapted from Ligon et al³⁸. Briefly, snap-frozen mice bladders were homogenized to isolate RNA using RNeasy Mini Kit (74101, Qiagen) and RNase-free DNase digestion kit (79254, Qiagen) following manufacturer’s protocol. Libraries were prepared with Ribo-Zero rRNA depletion kit (Illumina) and sequenced on a HiSeq3000 (Illumina). Reads were aligned to the Ensembl GRCm38.76 top-level assembly with STAR version 2.0.4b and gene counts were derived from the number of uniquely aligned unambiguous reads by Subread: featureCount version 1.4.5. Sequencing performance was then assessed for total number of aligned reads, total number of uniquely aligned reads, genes detected, ribosomal fraction, known junction saturation, and read distribution over known gene models with RSeQC version 2.3 (Raw data available at NCBI Gene Expression Omnibus repository under the following identifier: GSE149571).

Gene expression expressed as counts was normalized using upper quartile normalization and RUVr¹⁰⁴, then differentially expressed genes were determined using the EdgeR R package¹⁰⁵. Significance was achieved for FDR-adjusted p-value<0.05 and fold change exceeding 1.5x. Enriched pathways were determined using over-representation analysis (ORA) as implemented by the MSigDB online platform¹⁰⁶, specifically using the hypergeometric distribution against the MSigDB Hallmark pathway collection¹⁰⁷, with significance achieved for FDR-adjusted p-value<0.05. Data was visualized using the R statistical system and GraphPad Prism (version 10.0.0 (153)).

Quantitative RT-PCR (qRT-PCR)

Bladders of 5 young and 5 aged mice were collected and homogenized in TRIzol^™ Reagent (15596026, Thermo Fisher Scientific, USA) to extract total RNA followed by DNase 1 treatment (18068-015, Thermo Fisher Scientific, USA) following manufacturer’s protocol. One microgram of total RNA was utilized to perform cDNA synthesis using SuperScript^™ II Reverse Transcriptase (18064-014, Thermo Fisher Scientific, USA) following manufacturer’s protocol. All cDNAs were diluted to 1:8 with RNase-free water prior thereafter. Primer designs (see Key Resources Table) and qRT-PCR setup was performed using SsoAdvanced Universal SYBR^® Green Supermix (1725274, Bio-Rad, USA) following manufacturer’s protocol in 10 μl reactions (5 μl Supermix; 1 μl each of Forward and Reverse Primers; 2 μl diluted cDNA; 1 μl RNase-free water), each reaction was done in triplicate. 18s rRNA was used as a housekeeping gene. The qRT-PCR reaction was run in QuantStudio^™3 Real-Time PCR System (Applied Biosystems^™, USA) using the following settings: 98 °C, 3 minutes (initial activation); 98 °C, 30 seconds (Denaturation); 58 °C, 30 seconds (Annealing/Extension); 40 cycles; and instrument default setting for melt-curve analysis. Raw quantitation values were used for calculating fold change based on the young bladder.

KEY RESOURCES TABLE

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Mouse anti-p62	Abcam	Cat# ab56416; RRID: AB_945626
Rabbit anti-LC3B	Novus Biologicals	Cat# NB600-1384; RRID: AB_669581
Rabbit Anti-AMPK-α, phospho (Thr172)	Cell Signaling Technology	Cat#2535S; RRID: AB_331250
Anti-S6 Ribosomal Protein, phospho (Ser240 / Ser244)	Cell Signaling Technology	Cat#2215S; RRID: AB_331682
Mouse anti-beta actin	Cell Signaling Technology	Cat#3700S; RRID: AB_2242334
Anti-β-galactosidase	Cell Signaling Technology	Cat# 27198; RRID: AB_2798940
Anti-γH2AX	Cell Signaling Technology	Cat#9718S; RRID: AB_2118009
Anti-E-Cadherin	BD Biosciences	Cat# 610181; RRID: AB_397580
Anti-IL-1β	Cell Signaling Technology	Cat#12703S; RRID: AB_2737350
Anti-IL-10	Proteintech	Cat#: 60269-1-Ig; RRID: AB_2881389
Anti-Caspase-1	Abcam	Cat# ab138483; RRID: AB_2888675
Anti-Gasdermin D	Thermo Fisher Scientific	Cat# PA5-115330; RRID: AB_2899966
Anti-Escherichia coli (E. coli) (AP)	US Biological	Cat#E3500-26-AP; RRID: N/A
Bacterial and virus strains
UTI89, clinical Isolate	Stemler et al., 201326	N/A
Chemicals, peptides, and recombinant proteins
Histoclear	Electron Microscopy Sciences	Cat#: 64110
Hematoxylin	Fisher Scientific	Cat#: 22-220-109
Eosin	Fisher Scientific	Cat#: 22-050-197
X-gal	Thermo Fisher Scientific	Cat#: R0404
Lysotracker Red	Life Technologies	Cat#: L-7528
Mitotracker dye	Thermo Fisher Scientific	Cat# M22426
Dihydro-ethidium	Thermo Fisher Scientific	Cat#D11347
Critical commercial assays
Papanicolau EA staining kit	Thermo Fisher Scientific	Cat#: 22-050-211
Bioplex cytokine assay	Bio-Rad	Cat#M60009RDPD
Creatinine assay	Millipore-Sigma	MAK080
Acid Phosphatase assay	Sigma-Aldrich	Cat#CS0740-1KT
Experimental models: organisms/strains
C57B6	National Institute of Aging	Young (8-12 weeks)
C57B6	National Institute of Aging	Aged (18-24 months)
Oligonucleotides
See Table S1 for the list of oligonucleotides designed and utilized in this study
Software and algorithms
Bio-Plex Manager Software 6.2	Bio-Rad	https://www.bio-rad.com/en-us/product/bio-plex-manager-software-standard-edition
Prism v.9.0.1	GraphPad Software	https://www.graphpad.com/scientific-software/prism/
Image Lab, version 6.1.0 build 7, standard edition	Life science research Bio-Rad	https://commerce.bio-rad.com/en-us/product/image-lab-software
Biorender licensed to Mysorekar Lab.	Biorender	https://app.biorender.com/

Multiplex ELISA (Cytokine Assay)

SASP-specific cytokines (G-CSF, GM-CSF, IL-1α, IL-6, IP-10, MCP-1, MIP-1α, MIP-1 β, and TNF-α) were quantitated using the Bioplex cytokine assay kit (Cat#M60009RDPD, Bio-Rad). Second void urine samples from 5 animals in each group were collected and processed the same day in duplicate following the manufacturer’s protocol. Microplate reading was performed using BIOPLEX 200 (Biorad, USA) at Baylor College of Medicine Antibody-Based Proteomics Core Laboratory. Intra-assay %CVs (coefficient of variation) were calculated for GM-CSF (8.03%), IL-1α (9.13%), IL-1β (7.17%), IL-6 (8.10%), MIP-1α (9.79%), and TNF- α (6.96%). In addition, creatinine assay (MAK080, MilliporeSigma, USA) was performed on the same urine samples (3μl) following the manufacturer’s protocol. Cytokine concentrations were normalized to the corresponding average creatinine level of young and aged mice.

SA-β-galactosidase Staining

Fresh-frozen 8 μM sections of young and aged bladders were utilized for SA-β-galactosidase staining. Briefly, sections were fixed in 1% formamide solution for 5 minutes at room temperature and washed twice in 1X PBS for 3 minutes each. Staining was done using a solution composed of the following: 40 mM citric acid/sodium phosphate buffer (pH 6.0), 150 mM NaCl, 2 mM MgCl2, 5 mM potassium hexacyano-ferrate II, 5 mM potassium hexacyano-ferrate III, and fresh 1 mg/mL 5-bromo-4-chloro-3-indolyl-β-D-galactosidase (X-gal) dissolved in dimethylformamide (DMF) at 37 °C (without CO2) overnight. Image capture was performed with a Panoramic Midi microscope (3DHISTECH Ltd, Hungary).

γ-H2AX Staining

Formalin-fixed paraffin-embedded (FFPE) bladder sections from young and aged mice were deparaffinized by soaking in 3 separate solutions of 100% Histoclear at 5 minutes each. Sections were rehydrated in decreasing concentrations of ethanol (100%, 90%, 70%, 50%) at 5 minutes per concentration and soaked in 1X PBS for 5 minutes. Sections were blocked in 1% bovine serum albumin for I hour at room temperature, followed by antibody staining with anti-γH2AX (1:500; 9718, Cell Signaling Technology) in blocking buffer plus 0.1% Tween-20 overnight at 4°C. Sections were rinsed in 1X PBS (3 times, 5 minutes each), and incubated in secondary antibody AF488 (A11034, Thermo Fisher Scientific, USA) for 1 hour at room temperature and rinsed in 1X PBS (3 times, 5 minutes each). Sections were applied with Prolong Gold Antifade reagent with DAPI (P36935, Thermo Fisher Scientific, USA), covered with cover glass, and edges were sealed with clear nail polish. Sections were imaged under ECLIPSE Ni Epi-fluorescence Upright Microscope (Nikon, USA).

Western Blotting (WB)

Bladders of 5 young and 5 aged mice were collected and homogenized in RIPA lysis buffer to extract total proteins. Total protein was quantitated using a BCA assay (Pierce^™ BCA Protein Assay Kit, 23225, Thermo Fisher Scientific, USA). 10 ug of protein was loaded onto precast gel (4561095, 4–20% Mini-PROTEAN^® TGX^™ Precast Protein Gels, Bio-Rad, USA) and resolved at 200 volts for 30-35 minutes. Proteins bands were immobilized through transfer to PVDF membrane (IPFL00010, Immobilon Transfer Membrane, Millipore, Ireland) for one hour at 110 volts in ice. The membrane was treated with blocking buffer (927-60001, Intercept^® (TBS) Blocking Buffer, LI-COR, USA) for 1 hour at room temperature with gentle agitation. Membranes were incubated with primary antibody solution in blocking buffer plus 0.1% Tween-20, overnight at 4°C with agitation. Beta-actin was used as a loading control. Membranes were washed in 1X TBS with 0.1% Tween-20, 5 times, for 5 minutes each at room temperature with agitation followed by treatment with appropriate secondary antibody solution in blocking buffer plus 0.1% Tween-20 for 1 hour at room temperature with agitation. Membranes were washed in 1X TBS with 0.1% Tween-20, 5 times, for 5 minutes each at room temperature with agitation followed by washing in 1x TBS 2 times, for 5 minutes each. Imaging was performed using the ChemiDocTM MP imAging system (Bio-Rad, USA). Densitometric analysis was done using Bio-Rad Image Lab software (6.0.1).

Acid Phosphatase Assay

Fresh-frozen sections of young and aged bladders were tested for acid phosphatase activity using an acid phosphatase assay kit (CS0740-1KT, Sigma-Aldrich) following the manufacturer’s protocol. The experiment was carried out in 8 young and 9 aged bladders. Values were normalized to the bladder weight.

In vivo ROS Assay and Imaging

OCT-embedded fresh young and aged bladders were cut into 10 μm thickness and mounted in glass slides for staining. Sections were soaked in ROS-sensing dye (Dihydro-Ethidium: DHE), incubated for 10 minutes at room temperature, washed in 1X PBS, and followed by counterstaining with DAPI. Images were captured using Zeiss Axio Imager M2 Plus Wide Field Fluorescence Microscope (Carl Zeiss Inc, Thornwood, NY). ROS signal in 7 images from 7 mice per groups was quantitated using ImageJ.

Urine analysis and Inflammation Score

Young and aged mouse urine samples were collected for sediment analysis (10 μl urine plus 40 μl 1X PBS) and subjected to cytospin3. Sediments on the microscope slides were fixed in acetic acid/alcohol for 15 minutes and subjected to Epredia^™ Papanicolaou EA Staining (22-050-211, Thermo Fisher Scientific, USA) following the manufacturer’s protocol. Stained urine sediments were examined using blind scoring under the light microscope. An inflammation score scaled from 0 to 4 was adopted, where 0 indicated <1 and 4 indicated >20 polymorphonuclear leukocytes per high-powered field (as previously described in Stemler et al., 2013²⁶). Exfoliated cells were also quantitated and expressed per 10 μl of urine, including the number of monocytes at specified hpi.

Urinary bacterial load from infected bladders of young and aged mice were quantified as described previously 23. Briefly, urine samples were collected at indicated time points and serially diluted in 1X PBS. LB plates were spotted with 5 μL of each dilution 6 times, and plates were incubated at 37°C overnight. Bacterial titers were calculated as log10 CFU/ml (6 hpi). Calculation of mice with bacteriuria was also performed.

Papanicolau Staining of IBCs in Urine

Urine from infected bladders of young and aged mice was fixed using acetic/alcohol fixative and dehydrated in 95% alcohol, followed by rehydration in water. Hematoxylin was incubated for 10 minutes and developed in running tap water. Sections were dehydrated in 95% alcohol and stained with brand OG-6 solution (from 22-050-211, Thermo Fisher Scientific, USA) for 2 minutes and washed with 95% alcohol. Next, EA-65 stain (from 22-050-211, Thermo Fisher Scientific, USA) was used for 10 minutes, followed by a dehydration step. Clearing was done in xylene and a xylene-based mounting media was used to mount the sides, followed by imaging under a Panoramic Midi microscope (3DHISTECH Ltd, Hungary).

Quantification of QIRs

Quantification of QIRs was done as described previously²². Briefly, young and aged mice were sacrificed at 14 days post infection (dpi) and bladders were collected and processed. Eight separate 5 μm step sections over a thickness of 300 μm were stained with antibodies against E. coli, and E-cadherin. Sections were imaged at 63S oil on a Zeiss Apotome microscope and processed using ImageJ. The total number of UPEC reservoirs were counted and reported per bladder (n=15 animals in each group).

QUANTIFICATION AND STATISTICAL ANALYSIS

All measured values were plotted using GraphPad Prism version 10.0.0 (153) (GraphPad Software, La Jolla, CA, USA). WB and qRT-PCR data were expressed as mean ± Standard Deviation (SD) of the sample size (n) (indicated in each figure). Unpaired Student's t-tests, Mann-Whitney, and ANOVA were used to determine statistical significance.

Supplementary Material

1

Table S1. Oligonucleotides designed and utilized in this study.

Highlights.

• Aging heightens the frequency of recurrent bladder infections

• Aging causes urothelial dysfunction, senescence, and oxidative stress

• Aged urothelium undergoes spontaneous pyroptotic cell death

• D-Mannose treatment reverses age-associated bladder dysfunction