Urinary Tract Infections: Renal Intercalated Cells Protect against Pathogens

Abstract

Urinary tract infections affect more than 1 in 2 women during their lifetime. Among these, more than 10% of patients carry antibiotic-resistant bacterial strains, highlighting the urgent need to identify alternative treatments. While innate defense mechanisms are well-characterized in the lower urinary tract, it is becoming evident that the collecting duct (CD), the first renal segment encountered by invading uropathogenic bacteria, also contributes to bacterial clearance. However, the role of this segment is beginning to be understood. This review summarizes the current knowledge on CD intercalated cells in urinary tract bacterial clearance. Understanding the innate protective role of the uroepithelium and of the CD offers new opportunities for alternative therapeutic strategies.

Uropathogenic Escherichia coli Causes Urinary Tract Infections

Urinary tract infections (UTIs) can occur in the urethra (urethritis), bladder (cystitis), ureters, and kidneys (pyelonephritis)¹ and predominantly affect women.² Uropathogenic E. coli (UPEC) causes 70% of acute pyelonephritis in male patients and 80% of female patients.³ The remaining UTI cases are caused by other organisms, including Staphylococcus saprophyticus, Klebsiella, Enterobacter, Proteus species, and enterococci.^3,4

UTIs can be classified as uncomplicated or complicated, depending on the probability for recurrence or progression to more severe infections.^1,2 Uncomplicated UTIs occur in healthy individuals without urinary tract abnormalities and resolve with antibiotic treatment. Complicated UTIs can be due to abnormalities in the urinary tract or host defense. This review focuses on the most common cause of UTI, the UPEC.

Once UPEC reaches the bladder, its type 1 pilus adhesins will adhere to mannosylated uroplakins and integrin receptors expressed at the apical surface of the uroepithelium.⁵ UPEC can then be endocytosed and colonize the host cells, which may protect them from antibiotic treatments.

However, bladder cells have natural defense mechanisms to evade the infection. The UPEC internalization process triggers a Toll-Like Receptor 4 (TLR4)-dependent innate immune response and UPEC exocytosis. Alternatively, UPEC can escape to the cytoplasm where they can form quiescent, nondividing intracellular bacterial communities. UPEC also secretes toxins, such as α-hemolysis and proteases, that promote the release of cellular nutrients and siderophores to highjack the released cellular iron. α-Hemolysin also promotes epithelial exfoliation to allow further UPEC spreading. Cystitis pathogenesis has been recently reviewed in detail.^6,7 On overcoming bladder natural defenses, UPEC can continue their ascension to the kidneys, further attaching through adhesins or pyelonephritis-associated pili to globoside-containing glycolipids at the apical surface of renal cells. Eventually, UPEC can overcome the tubular epithelial barrier to enter the blood stream, causing bacteremia.⁵

Mechanistic Insights into Intercalated Cells and Their Role in Host Defense against UTIs

Intercalated Cells in the Collecting Ducts

As host defense mechanisms in the bladder are evaded, UPEC can reach the collecting duct (CD). The CD contains a salt-and-pepper, highly plastic epithelium⁸ that is composed of two major cell types: intercalated cells (ICs) and principal cells (PCs) (Figure 1). Aquaporin-2–expressing cells serve as precursors for PCs, and IC subtypes and their differentiation is regulated by forkhead box protein I1 (Foxi1) transcription factor and JAG1 and NOTCH1/2 signaling.⁸ In the CD, ciliated PCs mediate sodium and water reabsorption and potassium secretion.⁹ Also located in the CD, ICs are dispersed throughout the renal cortex and medulla.¹⁰ Although some antibacterial molecules synthesized by PCs contribute to the innate immune response as stated below,¹¹ this review focuses on the role of ICs in innate immunity.

Figure 1.

Diagram illustrating the various proteins in the IC involved in acid–base and salt homeostasis. (Top) A-ICs, (bottom) B-ICs. TJ claudins are represented in green between the two cells. TJ, tight junction; ATP, adenosine triphosphate; A-IC, type A-IC; B-IC, type B-IC; IC, intercalated cell; kAE1, kidney chloride/bicarbonate exchanger 1; MR, mineralocorticoid receptor; NDCBE, Na⁺-driven chloride/bicarbonate exchanger.

At least three types of ICs coexist: type A (A-IC), type B (B-IC), and non-A–non-B ICs. Although the role of non-A–non-B ICs remains unclear, recent RNA velocity studies showed that various IC subtypes and hybrid PCs–ICs coexist, suggesting that these cells may represent intermediate states between the two main CD cell types.¹² Performing scRNA-seq on a resected kidney showed that six IC clusters exist. Marker genes differentiated the ICs into three A-ICs (subtype A, B, and C), one B-IC, one non-A–non-B-IC, and one hybrid PC–IC subtype.¹² Although all A-ICs identified expressed SLC4A1, the A-IC subtypes differed in their expression of the heat shock protein HSPA1A and early growth response 1.

ICs Regulate Acid–Base Balance and Salt Homeostasis

Acid–Base Balance

Because natural defenses against UTIs involve the cell machinery responsible for acid–base homeostasis, we briefly outline the mechanisms involved in this process below.

In ICs, acidic conditions stimulate the diffusion of carbon dioxide across the membrane and the formation of carbonic acid, which is catalyzed by carbonic anhydrase II.¹⁰ Carbonic acid next ionizes into protons and bicarbonate. A-ICs excrete protons into the tubular lumen by the apical H⁺/K⁺-ATPase and the v-H⁺-ATPase (Figure 1).¹⁰ Heightened intracellular bicarbonate activates the bicarbonate-sensor soluble adenylate cyclase, which, in turn, results in Ser-175 phosphorylation and activation of the A subunit of the v-H⁺-ATPase.¹³

In the basolateral membrane, the kidney chloride/bicarbonate exchanger 1 and Slc26a11¹⁴ contribute to bicarbonate reabsorption into the interstitial fluid.¹⁰ During metabolic acidosis, both the v-H⁺-ATPase and the kidney chloride/bicarbonate exchanger 1 are upregulated to restore pH.¹⁵

A mirror mechanism occurs in type-B ICs during alkalosis by the now basolateral v-H⁺-ATPase and apical chloride/bicarbonate antiporter pendrin.¹⁶ Exclusively detected in the connecting tubule and initial CD, the role of non-A–non-B ICs remains unclear.¹⁷

Salt Homeostasis

The second main function of the CD is to fine-tune salt homeostasis, and recent evidence supports that the hyperosmolarity and, more specifically, medullary sodium chloride are essential in promoting an efficient immune response against bacterial infection.^18,19 The role of electrogenic epithelial sodium channel ENaC in maintaining salt homeostasis is well-established. ENaC activity contributes to a lumen-negative transepithelial voltage, thereby promoting proton secretion in the CD, although this theory has been recently challenged.²⁰

In B-ICs, the apical pendrin and the Na⁺-driven chloride/bicarbonate exchanger mediate thiazide-sensitive electroneutral NaCl reabsorption.²¹ Two cycles of pendrin exchange and one cycle of Na⁺-driven chloride/bicarbonate exchanger coordinated with basolateral AE4 result in the net uptake of one sodium and one chloride ion and secretion of two bicarbonate ions.²² However, the role of AE4 in salt homeostasis has been recently challenged.²³ The essential contribution of tight junctions in electrolyte homeostasis has not been detailed in this study.^24,25

ICs Sense and Trigger Inflammation

Recent studies have unveiled the role of ICs in innate immune reaction and in renal inflammation. The A-IC apical proinflammatory P2Y14 receptor (GPR105) triggers sterile inflammation.²⁶ After an acute kidney injury, damaged cells in the proximal tubule release luminal UDP-Glucose, which is a DAMP molecule. When it reaches the CD A-ICs, UDP-Glucose binds to the apical P2Y14 receptors, leading to the activation of MEK1/2-extracellular signal-regulated kinase (ERK)1/2 pathways resulting in the production of proinflammatory chemokines, such as IL-8, CXCL1, CXCL2, CCL2, and CCL3. This secretion, in turn, attracts neutrophils and monocytes to the renal medulla.^27,28

In Pyelonephritis, UPEC Preferentially Attaches to Desmoglein 2 in ICs and Activates TLR4

UPEC uses virulence factors to infiltrate and colonize the urinary system. Common UPEC cell surface factors include FimH and PapC, and common secreted factors include hemolysin (hly) and siderophores.^29,30 The most prevalent virulence factor associated with UPEC, FimH, is present in over half of UPEC strains. Although the prevalence of virulence factors in cystitis and pyelonephritis varies depending on the demographic, the observed frequency of virulence factors in pyelonephritis and cystitis UPEC strains is not significantly different.^29,30 In mice subjected to transurethral catheterization, UPEC uses its type 1 pili FimH adhesin to attach to the apical mannosylated desmoglein 2 receptor on the apical side of the renal medullary collecting duct (MCD).³¹ PapG adhesin of P pili also facilitates UPEC binding to kidney epithelial tissue.³² Although several TLRs, such as TLR2,³³ 5,³⁴ and 11,³⁵ have been implicated in the kidney response to UTIs, TLR4 is the one triggered in ICs.^36,37 Knocking out murine TLR4 expression results in a lack of activation of proinflammatory mediators and a failure to clear UPEC.

In mice, two nonhemolytic UPEC strains triggered both NF-kB and MyD88-dependent and independent pathways, both resulting in the secretion of leukocyte chemoattractant macrophage inflammatory protein 2 (Figure 2).³⁷ Noncytolytic apical UPEC was further shown to bind to inner medullary collecting duct (IMCD) cells through TLR4 and translocate transcellularly through caveolin-1 and clathrin-dependent pathways.³⁶

Figure 2.

Schematic diagram summarizing the sequence of events from UPEC attachment to the apical surface of ICs to the multiple lines of defense, leading to UPEC clearance. Luminal UPEC attaches to TLR4 in ICs (1) and triggers downstream MyD88-dependent and -independent signaling, resulting in synthesis and secretion of inflammatory molecules; LCN2; and AMP, such as ADM, cathelicidins, and defensins (2). By various mechanisms, LCN2 and these AMPs result in bacterial clearance (3). Urinary acidification contributes to bacterial clearance as well (4). ADM, adrenomedullin; AMP, antimicrobial peptide; IC, intercalated cell; MAP3K7, mitogen-activated protein kinase kinase kinase 7; TLR4, toll-like receptor 4; UPEC, uropathogenic E. coli.

Additional pathways and markers are triggered on UTI. Isolated human and mouse ICs exposed to UPEC upregulate the expression of the Mitogen-activated protein kinase kinase kinase 7, as shown by both RNAseq and immunofluorescence on mouse kidney sections.³⁸ Mitogen-activated protein kinase kinase kinase 7 is an essential component of the NF-kB signaling pathway, which results in the secretion of inflammatory molecules IL1, IL-6, and TNFα.³⁹ Its activation is triggered by multiple stimuli, including LPS-mediated TLR4 activation (Figure 2).

Cellular Changes in ICs Promote an Innate Immune Response on UTIs

The abovementioned roles of ICs in UTI were further confirmed with flow cytometry cell sorting enrichment, followed by single-cell RNA sequencing of human and murine ICs.¹² This study revealed the presence of six different subtypes of ICs, including two able to produce mature phagosomes in response to UTI. These cells also upregulate the v-H⁺-ATPase gene expression on UPEC exposure (Figure 2) and increase luminal acidification of phagosomes containing engulfed UPEC. In addition, a change in expression of genes involved in phagosome maturation was detected during UTI. IC subtypes A, B, and C showed that phagosome maturation becomes the predominant function after UPEC exposure. Thus, this groundbreaking study not only confirmed the role of ICs in UPEC internalization and degradation but also provided direct evidence of IC remodeling after UPEC exposure.

In Pyelonephritis, ICs Are Essential to Clear UPEC

Given the role of ICs in acid–base balance, the contribution of extracellular (luminal or interstitial) pH and the relationship between bacterial exposure and acid–base balance have been investigated. Indeed, studies in the medullary thick ascending limb support an ERK-mediated inhibitory effect of interstitial LPS on the sodium-proton exchanger 3, resulting in inhibition of bicarbonate absorption.^40,41 In the CD, early in vitro studies supported that urinary acidification limits bacterial growth,⁴² suggesting that this may be a physiological defense mechanism for bacterial clearance. Patients with UTI who drank cranberry juice presented an increased urinary secretion of the organic hippuric acid, which inhibited bacterial growth.⁴³ However, on the interstitial side, acidosis seems to have a detrimental effect on UPEC clearance. Despite an urine acidification, mice in metabolic acidosis and those exposed to UPEC display an increased accumulation of neutrophils and tissue resident macrophages; higher abundance of cytokines TNFα, IL-1β, and IL-6 as well as chemokines CXCL1, 2, and 5; and intensified pyelonephritis.⁴⁴ These studies support that acidic urine plays a minimal role in bacterial clearance while acidosis results in a poorer outcome.

Further reports supported an essential role of ICs themselves in bacterial clearance. Knocking out CAII in mice (Car2^−/− mice) results in metabolic acidosis, alkaline urine, and, importantly, a significant loss of ICs.⁴⁵ Interestingly, after UTI, Car2^−/− mice also have a heavier urinary bacterial load and are less able to clear UPEC bacteria compared with control littermates.⁴⁶ In fact, the loss of CAII-positive ICs in Car2^−/− mice is more detrimental to bacterial clearance than just the pharmacological inhibition of CAII activity.⁴⁷ Similarly, knocking out murine ICs by deleting Tcfcp2l1 transcription factor showed that, while the other nephron segments and PC develop normally, the renal v-ATPase immunostaining is lost, supporting a loss of IC.⁴² On UTI, the knockout mice are unable to acidify their urine and to clear the bacterial invasion in comparison with their WT littermates.

The physiological mechanisms behind the role of acidic pH on IC-based bacterial clearance are starting to be deciphered from further in vitro studies. Research in the cancer field has demonstrated that cytosolic acidification upregulates 2-hydroxyglutarate, which, in turn, stabilizes the transcriptional regulator hypoxia-inducible factor 1-α subunit (HIF-1a).⁴⁸ Growing immortalized CD cells M1 (that contain both PC and IC) in an acidic growth medium blunted UPEC growth because of HIF-1a upregulation, itself associated with a release of nitric oxide, and a TLR4-dependent upregulation of several antimicrobial peptides (AMPs), such as defensin b2 and cathelicidin (more details in the following sections).⁴⁹ Mimicking acidosis by pharmacological stabilization of short-lived HIF-1a with AKB-4924 results in reduced UPEC-mediated epithelial cell death, lower UTI, an improved urinary bacterial clearance, and reduced inflammation in a UTI mouse model.⁵⁰

Overall, these results support that metabolic acidosis or loss of ICs is detrimental to an efficient immune response and clearance of UPEC.

ICs Express Antimicrobial Proteins

Antimicrobial molecules are essential components of the urinary tract's response to UTIs. Like many epithelial cells, ICs express these molecules as eluded above. Flow cytometry cell sorting enrichment of isolated murine ICs has revealed that these cells endogenously express several types of antimicrobial molecules, including the bacteriostatic protein lipocalin 2 (LCN-2, also referred to as neutrophil gelatinase-associated lipocalin or NGAL); AMPs, such as adrenomedullin (ADM); and ribonucleases (RNAses).¹¹ Some defensins, such as Defensin b1, Defensin b26, and antimicrobial peptide RNAse4, are enriched in both PCs and ICs compared with cells from other nephron segments. Cathelicidin, Calgranulin A (S100a8), and NGAL are specifically enriched in ICs versus other cell types.⁵¹ Importantly, urine composition, pH, and osmolality can affect AMP activity. For instance, high extracellular glucose treatment lowers the abundance of psoriasin (S100A7), RNAse7, and Defensin b4 in uroepithelial cells and urine exfoliated cells from patients with diabetes.^52–54 In addition, peak performance of AMPs has been shown at urinary pH 5–6.5,⁵⁵ although pH has variable effects. While alkaline extracellular pH reduces RNAse7 activity, β defensin 1 remains unaffected.^56,57 Finally, UTIs typically cause a reduction in urine osmolality that may affect AMP activities.^58,59

Together, these findings support that the urinary tract and in particular both ICs and PCs are equipped with antimicrobial proteins, with ICs encoding a specific pool of molecules to fight UTIs. These secreted AMPs are further described in the next paragraphs, with a focus on IC-produced AMPs.

CD Cells Secrete Antibacterial Peptides to Clear UPEC

IC-Produced LCN2/NGAL Chelates Luminal Iron and Impairs Bacterial Growth

UPEC relies on fimbrial adhesions, toxins, and nutrient-acquisition strategies for survival in the harsh conditions of the urinary tract.⁶⁰ On TLR4 activation, UPEC-exposed A-ICs secrete the bacteriostatic protein LCN2 (Figure 2). LCN2 messenger levels increase by 27-fold in the mouse bladder and by six-fold in the kidney of mice and human urine on infection with Gram-negative UPEC.^42,61

LCN2 exists in two forms: apo-LCN2 and holo-LCN2.⁶² Iron-free (Apo) LCN2 depletes cytosolic levels of iron and results in apoptosis induction while iron-bound (Holo) LCN2 increases cytosolic iron concentration and protects from apoptosis.⁶² By inoculating a UTI-susceptible mouse model with GFP-transfected UPEC, Paragas and colleagues showed that A-ICs specifically express LCN2 and that UPEC bacteria associate with A-ICs.⁴² After UPEC inoculation, mice lacking ICs display a blunted urinary acidification and urinary LCN2 expression compared with control mice.

LCN2 bactericidal mode of action is well-understood. UPEC bacteria capture iron from the host transferrin by releasing an enterochelin (Ent) siderophore with a high affinity for iron. A-ICs secrete LCN2 in the iron-free apo form,⁶³ which also has a high affinity and is specific to the Ent:iron complex. The now holo-LCN2 dispels iron from bacterial use and reroutes it either for lysosomal degradation or for urinary excretion. In acidified lysosomes, the iron in the LCN2:Ent:iron complex is released and reduced while LCN2 is proteolytically cleaved.^61,64,65 LCN2 was found to be necessary and sufficient to suppress UPEC activity.^64,66

Cationic Host Defense Peptides Contribute to the Host Defense against UTI

Apart from starving bacteria from iron, the host organism secretes a variety of small cationic host defense peptides (CHDPs) that have microbicidal properties and/or enhance the host's immune response.⁶⁷ Two classes of CHDPs, the cathelicidins and defensins, work in synergy to maximize their microbicidal activity (Figure 2). CHDP mode of action consists in interacting with the negatively charged bacterial membrane by an electrostatic interaction, permeation of the membrane, release of bacterial content, and bacterial death. CHDP can also alter bacterial cell wall biosynthesis and cell processes, such as replication, transcription, and translation. Furthermore, CHDP can have an immune-regulatory role on the host defense. Once uptaken into the host cells, CHDP can interact with a variety of host effectors, alter intracellular signaling pathways, and result in activation of specific transcription factors that modulate the inflammatory and immune system, as summarized in 67.

ADM Contributes to UTI Clearance

ADM is a multifunctional 52 residue, cationic peptide produced in many tissues and cells, such as macrophages and renal parenchymal cells.⁶⁸ With ADM receptors' widespread expression, the peptide controls central body functions, including vascular tone regulation, fluid and electrolyte homeostasis, and inflammation.⁵¹

Children with acute pyelonephritis display lower plasma ADM but higher urine ADM concentrations compared with healthy participants.⁶⁹ In rat IMCD at the basal state, ADM is expressed by both ICs and PCs and is enriched 19.44-fold compared with non-IMCD cells.⁷⁰ Finally, in mice exposed to UPEC, ADM expression in both ICs and PCs is also upregulated.¹²

ADM disrupts the E. coli outer and cytoplasmic membrane through its conserved α-helical carboxyl terminus that carries the antimicrobial activity.⁷¹ The mechanism of action of ADM has been deciphered in other cell types. After LPS stimulation to rat macrophages, ADM gene expression and secretion is followed by the production of macrophage migration inhibitory factor, IL-1β, and IL-6.⁷² Opposingly, ADM also suppresses the secretion of the inflammatory mediator, TNFα.

In Humans, Ribonucleases (RNase) 7 and 4 Also Contribute to Bacterial Clearance

The RNase A superfamily encodes canonical peptides with antibacterial activity. Originally identified in the human epidermis,⁷³ the 14.5-kD ribonuclease 7 (RNase 7) is found in the mature human kidney, ureter, and bladder, where it is the highest, and in the ICs and is secreted in urine in response to acute pyelonephritis.⁷⁴ In keratinocytes, RNase 7 gene expression is triggered by TNFα, interferon gamma, and IL-1β⁷³ and is regulated by insulin through a PI3K/AKT signaling pathway.⁷⁵

Its broad-spectrum antimicrobial activity contributes to the sterility of the urinary tract against Gram-negative bacteria, Gram-positive bacteria, and yeast.⁷⁴ Independent of its RNase catalytic activity, RNase 7 exerts its broad-spectrum antimicrobial role by permeating and disrupting the bacterial cell membranes, causing membrane splitting, bleb formation, and significant microbial cell death.⁷⁶

Further supporting the role of RNase 7 in UPEC clearance, children and adolescent girls with a recurrent UTI history display a low urinary RNase 7 concentration.⁵³ Female mice overexpressing RNase-7 and exposed to UTI showed reduced intracellular bacterial communities and UPEC titers in urine and the bladder. Conversely, knocking down RNase 7 activity results in a significant increase in bacterial growth.⁷⁴

Expressed in the bladder, proximal tubule, ICs, and PCs, RNase 4 also seems to protect the urothelial barrier.⁷⁷ Women with a history of UTIs produced urine with a lower RNase 4 concentration than healthy women, and neutralizing RNase 4 in urine promotes UPEC growth. RNase 4 antibacterial activity occurs through agglutination, permeabilization, and leakage of bacterial membranes.

In Humans, Secretion of Defensins Correlates with a Lower Recurrence of UTI

In the CHDP family, defensins are classified as a, b, and q defensins.⁶⁷ Antimicrobial α-defensin 5 is a 32 amino acid peptide expressed in the human distal nephron, CD, and low urinary tract.⁷⁸ On UTI, α-defensin 5 gene expression and protein abundance are upregulated and the peptide becomes abundantly secreted into the urine. Another class of antimicrobial α-defensin called human neutrophil peptides 1-3 (HNP1-3, encoded by DEFA1A3 gene) is also protective against UTIs in humans. A large study on children with vesicoureteral reflux showed that higher the number of DEFA1A3 gene copies, lower the risk of recurrent UTIs in patients receiving antibiotic prophylaxis, but not in untreated patients.⁷⁹ More recently, mice expressing human DEFA1A3 defensin on UTI showed that DEFA1A3 is expressed by CD cells and is upregulated on UTI, and the mice display a better protection against kidney and bladder infection compared with control animals.⁸⁰ Finally, this study demonstrated a synergistic effect of HNP1-3 with cathelicidin against UPEC and antibiotic-resistant strains of E. coli. Similarly, human b-defensin 1 is upregulated three times in pyelonephritis,⁸¹ and mice lacking this protein have higher UPEC titers in their urine.⁸² However, in uncomplicated UTIs, its concentration was similar between noninfected patients and patients with UTI.⁸³ Whether this defensin is expressed in IC is unknown.

Cathelicidin is an α helical, small (23–37 amino acids) amphipathic peptide that is activated by serine protease cleavage.⁶⁷ LL-37 is the only human cathelicidin and is upregulated in the urine of patients with UTI compared with controls.⁸³ Cathelicidin is constitutively expressed in the epithelium of the urinary tract and of the kidney⁵¹ and prevents E. coli bacterial growth in UTIs, even in acidic urine conditions.⁴⁹ In the initial stages of infection, bacterial contact leads to a rapid increase in synthesis and secretion of cathelicidin by epithelial cells to prevent bacterial attachment.

AMP Synergistically Contribute to the Host Defense

Bacterial clearance involves several physiological weapons that can synergistically contribute to bacterial clearance. Indeed, innate immune processes originating from different segments of the nephron contribute to the fight. For example, uromodulin (also called Tamm–Horsfall protein) is secreted into the TAL and contributes to the innate defenses by binding to ascending UPEC, acting as decoys for mannose-rich apical membrane proteins in epithelial cells.⁸⁴ Hepcidin also contributes to UTI clearance by regulating iron availability, inflammation, and urinary acidification by a stimulatory action on Atp4a and Atp12a subunits.⁸⁵ Finally, cathelicidin was found to have a synergistic effect with HNP1-3 while it has an additive effect with RNase 7.⁸⁰ More synergistic effects may exist but have not been characterized yet.

Clinical Importance and Risk Factors

Chronic Kidney Diseases and Vesicourethral Reflux

As eluded above, the acid–base status of individuals may influence their UTI response. Children have a higher risk of developing acute pyelonephritis, in part, because of their high frequency of vesicourethral reflux (VUR).^86–89 Metabolic acidosis in a mouse model of VUR exacerbates UTI severity and risk of kidney injury.⁴⁴ Furthermore, type IV renal tubular acidosis, characterized by hyponatremia, hyperkalemia, and nonanion gap metabolic acidosis, can be a complication of UTIs in infants.⁹⁰ In addition, the presence of renal tumors⁹¹ can influence the patient's response to UTIs, although these do not specifically affect ICs. Mutations in the PKD1 and PKD2 genes can cause polycystic kidney disease, a condition that can cause a reduction in the number of ICs and as a result may affect the patients' response to UTIs.⁹² Patients with autosomal dominant polycystic kidney disease are more prone to UTIs than healthy individuals.^93–95

Diabetes and UTI

Type 2 diabetes is one well-understood example of a chronic disease that can influence the IC's innate immune response to UTIs. Patients with type 2 diabetes are at a higher risk of UTIs, pyelonephritis, and acute kidney injury,^96,97 independently of glucosemia or glucosuria⁹⁸; however, circulating levels of insulin regulate AMP production.⁷⁵ Expressed in ICs, the insulin receptor forms tetramers consisting of two extracellular α subunits and two transmembrane β subunits. On insulin binding, the β subunits change conformation, which activates the receptor's kinase activity and triggers downstream signaling pathways involving PI3K and AKT kinase, which are essential for AMP synthesis.⁹⁹ Type 2 diabetic and prediabetic mice both clear UTIs less efficiently than controls, have a decreased abundance of β subunits of insulin receptor, and have less AKT phosphorylation.⁹⁸ Mice knocked out for the IC-insulin receptor appropriately acidify their urine but display reduced phosphorylated AKT kinase and fail to activate the secretion of AMP, such as LCN2 and RNase 4.⁹⁸ Therefore, defective insulin receptor signaling in IC makes patients with diabetes more susceptible to UTIs.

Potential New Therapeutic Approaches

There is now evidence that ascending UPEC that have reached the CD is facing another layer of immune response predominantly triggered by ICs (Figure 2). Between sensing LPS, engaging in secretion of multiple AMPs, and triggering inflammation, these cells are clearly polyvalent. There is much work to do to fully understand the IC heterogeneity and specific function because no immortalized cell model fully recapitulates the physiology of CD cells. Mouse models are formidable tools to unveil the physiology of UTI defenses; however, they only partly recapitulate the human physiology. For example, RNase 7 is not present in the mouse genome. In addition, certain mouse strains are more susceptible to UTIs than others because of defective LPS signaling.¹⁰⁰ Finally, a large amount of data have been generated in nonrefluxing mice that are not prone to pyelonephritis, although VUR is a common condition that could contribute to UTIs in children. Thus, many questions remain, including: what is the contribution of the abovementioned antimicrobial proteins in UTI defense in humans? Why are some individuals prone to recurrent UTIs? Do the mouse models really mirror spontaneous UTIs occurring in humans?

Despite these limitations and remaining questions, supporting the innate immune response by enhancing the secretion of various AMPs seems to be a promising treatment option. Given their nonspecific effect on pathogens, they may be less likely to cause bacterial resistance. Further understanding the physiology associated with innate antibacterial processes will be essential to treat UTIs in the future.

Disclosures

All authors have nothing to disclose.

Funding

This work was supported by the Canadian Institutes of Health Research from PS 168871 and the Natural Sciences and Engineering Research Council of Canada from RGPIN-2017-06432 (E. Cordat).

Author Contributions

Conceptualization: Manav Batta, Forough Chelangarimiyandoab, Emmanuelle Cordat, Priyanka Mungara.

Data curation: Manav Batta, Forough Chelangarimiyandoab, Priyanka Mungara.

Resources: Emmanuelle Cordat.

Supervision: Emmanuelle Cordat.

Writing – original draft: Manav Batta, Forough Chelangarimiyandoab, Emmanuelle Cordat, Priyanka Mungara.

Writing – review & editing: Manav Batta, Forough Chelangarimiyandoab, Emmanuelle Cordat, Priyanka Mungara.