Mechanisms of Metabolic Acidosis–Induced Kidney Injury in Chronic Kidney Disease

Abstract

Retrospective analyses and single-center prospective studies identify chronic metabolic acidosis as an independent and modifiable risk factor for progression of CKD. In patients with CKD, untreated chronic metabolic acidosis often leads to an accelerated reduction in GFR. Mechanisms responsible for this reduction include adaptive responses that increase acid excretion but lead to a decline in kidney function. Metabolic acidosis in CKD stimulates production of intrakidney paracrine hormones including angiotensin II, aldosterone, and endothelin-1 (ET-1) that mediate the immediate benefit of increased kidney acid excretion, but their chronic upregulation promotes inflammation and fibrosis. Chronic metabolic acidosis also stimulates ammoniagenesis that increases acid excretion but also leads to ammonia-induced complement activation and deposition of C3 and C5b-9 that can cause tubule-interstitial damage, further worsening disease progression. These effects, along with acid accumulation in kidney tissue, combine to accelerate progression of kidney disease. Treatment of chronic metabolic acidosis attenuates these adaptive responses; reduces levels of angiotensin II, aldosterone, and ET-1; reduces ammoniagenesis; and diminishes inflammation and fibrosis that may lead to slowing of CKD progression.

Arterial blood bicarbonate and pH are maintained within narrow ranges of 22–29 and 7.36–7.44 mEq/L, respectively.^1–3 Chronic metabolic acidosis, defined as a serum bicarbonate level <22 mEq/L in a patient with normal pulmonary function, is common in CKD and reflects reduced acid excretion in the diseased kidneys, leading to net acid accumulation.^2–8 Consequences of metabolic acidosis accompanying CKD include bone disease, increased muscle protein catabolism, decreased albumin synthesis, increased systemic and kidney inflammation, insulin resistance, increased risk of cardiovascular disease, and progressive GFR loss.^9–17

In this review, we discuss mechanisms by which the adaptive response to metabolic acidosis, which promotes excretion of accumulated acid, becomes increasingly maladaptive, leading to progressive kidney injury with gradual loss of kidney function.

Acid-Base Balance in Normal and Diseased Kidneys

Nonvolatile acids derive from endogenous metabolism and from metabolism of dietary proteins, phospholipids, nucleic acids, and incomplete oxidation of carbohydrates and fats.^3,18,19 The daily load of nonvolatile acids is approximately 0.7–1 mEq/kg body wt, or approximately 50 to >70 mEq/d in healthy adults eating acid-inducing diets typical in Western societies.^4,18,20–25 Individuals with normal kidney function quantitatively match acid excretion to production, avoiding acid accumulation, whereas those with depressed kidney function are less able to do so.^4,7,22

Protons entering the systemic circulation are first bound to intracellular and extracellular substances called buffers, which defend against lowering of blood pH, before being excreted.²⁶ Intracellular buffers include phosphates and various anionic proteins, whereas bicarbonate is the primary extracellular buffer. Additional buffers include extracellular hemoglobin and albumin, along with carbonates and phosphates in bone and proteins in muscle.^4,27

To maintain systemic acid-base balance, the kidney must first reclaim all filtered bicarbonate entering the proximal tubule.^3,28 Within proximal tubule cells, carbonic anhydrase enzymatically converts carbon dioxide and water to bicarbonate, which is transported into the blood, and a proton. This proton is secreted into proximal tubule fluid and combines with bicarbonate to generate carbon dioxide and water. Filtered bicarbonate is thus reclaimed; however, this process does not result in net acid excretion. To excrete acid and regenerate new bicarbonate, the kidney uses two mechanisms. First, it metabolizes the amino acid glutamine, primarily in proximal tubule cells, to form two ammonium ions and an α-ketoglutarate in a process known as ammoniagenesis.²⁹ Ammonia enters the proximal tubule lumen by either ammonia diffusion or ammonium exchange with sodium (Na⁺) through the Na⁺-hydrogen (H⁺) exchanger 3 (NHE3; SLC9A3).^3,30 Tubule lumen ammonia binds a proton to form ammonium that is excreted into the urine. The α-ketoglutarate is metabolized to two bicarbonate anions. The combination of two protons leaving the body in the form of ammonium and the metabolism of α-ketoglutarate to two bicarbonate anions leads to net acid excretion.²⁹ In addition, both proximal and distal nephron tubule cells generate and secrete protons into the tubule lumen that attach to nonbicarbonate buffers, forming titratable acids (principally phosphate) that are excreted in the urine. In the process, the kidney tubule epithelial cells generate new bicarbonate that is transported into the blood. Approximately 60% of endogenously produced acid is excreted as ammonium and 40% as titratable acids in those eating the typical acid-producing Western diet.^3,28,31 As glomerular and tubular function decline in patients with CKD, the kidney is unable to quantitatively excrete the daily endogenous acid production, leading to acid retention and a subsequent fall in systemic bicarbonate concentration.^4,18,19

Eubicarbonatemic Metabolic Acidosis in Patients with CKD

Responses to prevent acid accumulation in the early stages of CKD maintain serum bicarbonate and blood pH within normal ranges. Acid accumulation in this early stage of CKD is insufficient to reduce serum bicarbonate below the normal range (designated eubicarbonatemic metabolic acidosis) but can trigger pathologic consequences, such as bone loss, muscle protein breakdown, and kidney hypertrophy.^4,7,32–34 In rats with normal kidney function or reduced nephron mass, dietary acid increased blood and kidney cortical acid content and increased urine net acid excretion with no measurable decrease in plasma pH or bicarbonate concentration.^35–37 Patients with hypertension who were eubicarbonatemic and macroalbuminuric (urine albumin to creatinine ratio >200 mg/g) with moderately reduced eGFR (CKD stage 2), compared to those with normal eGFR (CKD stage 1)—both of whom were eating a Western, acid-inducing diet—retained acid and had elevated plasma and urine ET-1 and aldosterone levels.³⁸ An acute oral Na⁺ bicarbonate (NaHCO₃) bolus reduced urine net acid excretion less in patients with CKD stage 2 than in those with stage 1, consistent with greater acid retention in the patients with CKD stage 2. Thirty days of oral NaHCO₃ decreased plasma and urine levels of both ET-1 and aldosterone in patients with CKD stage 2. These studies support the hypothesis that patients with CKD stage 2 and no metabolic acidosis eating a Western, acid-inducing diet^18,25 nevertheless retain acid, as reflected by increased plasma and urine levels of ET-1 and aldosterone, each of which mediates progression of CKD in animal models.^39,40

Recent studies support that urine excretion of citrate, the most abundant organic base in the urine that can be metabolized to bicarbonate, has utility in identifying acid retention in patients with CKD who are eubicarbonatemic, and in assessing their response to alkali therapy.^33,34 Baseline acid retention was higher in patients with CKD 2 than CKD 1, and there was lower urine citrate excretion in patients with CKD 2 than those with CKD 1.³⁸ Thirty days of eating base-producing fruits and vegetables reduced acid retention in those with CKD 2 but not in those with CKD 1, yet overall acid retention remained higher and urine citrate excretion remained lower in those with CKD 2 than those with CKD 1. These data support that low urine citrate excretion can help identify acid retention in patients with CKD who are eubicarbonatemic.^33,41

Because eubicarbonatemic metabolic acidosis might contribute to CKD progression, other urine markers are being evaluated to help assess risk for progression to ESKD and mortality in patients with CKD and no overt acidosis.^42,43 Reduced urine ammonium excretion accompanies the development of metabolic acidosis in CKD.³² In a study of 1044 patients in the African American Study of Kidney Disease and Hypertension database, urine ammonium excretion decreased with lower measured GFR, as expected. Event-free survival (the composite end point of death and dialysis) was highest in the patients with CKD with the highest urine ammonium excretion rate and was significantly lower in those with compromised urine ammonium excretion (excretion <20 mEq/d). The observation that lower urine ammonium excretion signals higher mortality or ESKD suggests that early markers of kidney failure affecting acid-base balance may help identify at-risk individuals with normal serum bicarbonate.

Together, these studies suggest that acid accumulation—due in part to Western, acid-producing diets—can occur in early CKD and may not be reflected by serum acid-base parameters indicative of metabolic acidosis.^33,34,41

Metabolic Acidosis in Patients with CKD

Chronic metabolic acidosis activates the kidney paracrine hormones angiotensin II, aldosterone, and ET-1 that increase net acid excretion.^{3,4,35,38–40,44–48} These hormones upregulate H⁺ transporters in the proximal tubule (NHE3 and the proton-translocating ATPase [H⁺-ATPase]) and ascending limb of the loop of Henle, distal tubule, connecting tubule, and collecting duct (NHE3/NHE2, H⁺-ATPase, and the proton-potassium (K⁺) exchange ATPase [H⁺/K⁺-ATPase]). Chronic metabolic acidosis increases ammoniagenesis, providing ammonia that is titrated to ammonium, leading to net acid excretion.^{3,28,29,49–52} Urine acid excretion increases within hours to days, decreasing accumulated acid, which is reflected by an increase in serum bicarbonate.^3,22,31,53

Metabolic acidosis in patients with CKD is associated with more rapid loss of kidney function.^{12,32,54–56} Several single-center clinical trials demonstrate slowing of CKD progression in patients with CKD and metabolic acidosis treated with oral alkali or dietary acid reduction.^47,57–61 Some of these studies included kidney hormone and biomarker measurements that reveal underlying adaptive mechanisms to correct acidosis.^42,60–62 These studies support that direct damage to the kidney is the consequence of acid retention leading to sustained increased levels of angiotensin II, aldosterone, and ET-1 and resulting inflammation and fibrosis, which is further aggravated by continual ammoniagenesis that can damage kidney tissue through complement activation.^40,63 Figure 1 summarizes adaptive and maladaptive responses to chronic metabolic acidosis in patients with CKD.

Figure 1.

Adaptive and maladaptive responses to metabolic acidosis in CKD. The adaptive response to metabolic acidosis in CKD is centered on an increase in ammoniagenesis, ET-1 production, and intrakidney RAAS expression (angiotensin II and aldosterone), all of which are necessary to promote acid excretion. In the adaptive response, these components activate NHE3-mediated Na⁺/H⁺ exchange, along with H⁺-ATPase and H⁺/K⁺-ATPase expression, and increase ammonium excretion to effectively increase acid excretion per nephron, thereby decreasing acid retention. However, sustained expression of this adaptive response, as seen in chronic metabolic acidosis, can lead to damaging activation of the alternative complement pathway cascade due to increased ammoniagenesis, along with production of proinflammatory and profibrotic mediators. This maladaptive response injures tubule-interstitial tissue, leading to loss of nephron function and worsening CKD.

Molecular Response to Acid Retention: Sensors of the Acid-Base Environment and Response Components

The kidney has cellular and membrane sensors that monitor tubule fluid and coordinate responses to maintain acid-base homeostasis.⁶⁴ Molecular sensors expressed by tubule epithelial cells include various kinases (Pyk2,^65–68 ET receptor type B [ET_B],^68,69 extracellular signal–related kinase 1/2 [ERK1/2]^70–73), proton-sensing receptors (G protein–coupled receptor 4 [GPR4]),^74–77 the bicarbonate-stimulated soluble adenylyl cyclase,^78–82 and the V-ATPase transporter^83,84 that both senses intraluminal proton concentration and is part of a complex of proteins that excrete acid from the intracellular environment (Table 1). These sensors activate signal transduction pathways responsible for regulating expression of NHE3 (Pyk2/ET_B pathway), bicarbonate reabsorption (ErB1/2), and monitoring tubule acid-base balance (soluble adenylyl cyclase), each of which are essential components in kidney acid excretion for maintaining acid-base homeostasis.

Table 1.

Acid-base sensors in kidney tubule epithelial cells

Sensor	Signal	Target	Function	References
Pyk2	Low intracellular and intraluminal pH within the proximal tubule	NHE3, proton secretion	Pyk2 autophosphorylation and kinase activity are activated by low pH (<7.4), which stimulates NHE3 expression leading to proton secretion	65–68
ETB receptor pathway	Low intracellular and intraluminal pH within the proximal tubule	NHE3, proton secretion	ERK1/2, and c-fos transcriptional activation of ET-1 and ETB receptors leads to NHE3 membrane accumulation and increased proton secretion	68,69
Erb1/2 heterodimer, receptor protein tyrosine phosphatase-γ, angiotensin receptor AT1A	Increased basolateral CO2	Bicarbonate reabsorption in proximal tubules	Erb1/2 and AT1A receptor pathways sense basolateral CO2 and bicarbonate levels and activate appropriate level of bicarbonate reabsorption	70–73
GPR4	Low blood and interstitial pH, low urinary pH in collecting duct	Proton secretion mediated by cAMP and V-ATPase translocation in A-type intercalated cells	GPR4 increases intracellular cAMP and IP3 when complexed with proton	74–77
Soluble adenyl cyclase	Bicarbonate anion	Monitors extracellular acid-base balance in the kidney tubules	Soluble adenyl cyclase produces cAMP, which directly stimulates V-ATPase membrane accumulation and proton secretion; regulates V-ATPase–mediated proton transport in tubules	78–82
V-ATPase	pH in the proximal tubule	Sensor of intraluminal and intracellular pH	V-ATPase has both proton-sensing and transport functions; V-ATPase forms a complex with other apical proteins to transport proton into the tubule lumen; sensor component works in concert with soluble adenyl cyclase	83,84

Pyk2, nonreceptor tyrosine kinase; ERK1/2, protein-serine/threonine kinases that participate in the Ras/Raf/mitogen-activated protein kinase kinase/ERK signal transduction cascade; c-fos, a proto-oncogene that is the human homolog of the retroviral oncogene v-fos; Erb1/2, receptor tyrosine kinases which are structurally related to the epidermal growth factor receptor (EGFR); AT1A, fibroblast angiotensin II type 1a receptor; CO₂, carbon dioxide; V-ATPase, vacuolar (H⁺)-ATPase which is an ATP-dependent proton pump that functions to acidify intracellular compartments and transport protons across the plasma membrane of eukaryotic cells; IP3, inositol triphosphate.

Tubule epithelial cells alter proton secretion in response to perturbations in systemic pH. For example, decreased proximal tubule pH increases metabolism of glutamine with subsequent bicarbonate production and increased ammonium excretion.³ Increased ammoniagenesis is mediated by selective upregulation of several transporters and enzymes, including SNAT3, glutamine transporter, phosphate-dependent glutaminase (PDG), glutamate dehydrogenase, α-ketoglutarate dehydrogenase, and phosphoenolpyruvate carboxykinase. In experimental models, enzyme mRNA levels (e.g., PDG, phosphoenolpyruvate carboxykinase) increase within hours of a reduction in extracellular pH, reflecting enhanced transcription and stability of the cognate mRNAs.^85–87

Distal nephron and collecting duct cells are examples of specialized tubule epithelial cells that promote proton excretion in response to lowered systemic pH.^88–91 The kidney collecting duct comprises cortical, outer medullary, and inner medullary segments and, of these components, the outer medullary collecting duct (OMCD) exhibits the highest rate of acid secretion, mediated by type A acid-secreting intercalated cells.⁹² Given the critical role of type A intercalated cells in overall acid-base status, investigators have examined mechanisms involved in pH-sensing and -signaling pathways that mediate upregulation of both H⁺-ATPase and H⁺/K⁺-ATPase.⁸⁹ These ATPases are ion pumps that use the energy of ATP hydrolysis to transport H⁺ and, in the case of the H⁺/K⁺-ATPase, exchange K⁺ ions. Mouse OMCD cells cultured with interfering RNAs and inhibitors showed that Pyk2 functions as a pH sensor to increase H⁺-ATPase activation in response to acid via the ERK1/2-mediated signaling pathway. Also, chronic metabolic acidosis and GPR4 increased H⁺/K⁺-ATPase mRNA abundance in OMCD and inner medullary collecting duct cells, implicating a regulatory role of pH-activated GPR4 in the homeostatic regulation of H⁺/K⁺-ATPase and the integrated response to acidosis.⁹³

H⁺/K⁺-ATPases in the distal collecting tubule and collecting duct are responsible for approximately half of the intercalated H⁺ secretory capacity in the distal nephron, and thus constitute a pivotal H⁺ secretory response to metabolic acidosis.⁸⁸ Evaluation of enzymatic functions along the nephron and collecting duct and studies in knockout mice suggest that the H⁺/K⁺-ATPases function to transport ions other than H⁺ and K⁺, including Na⁺. These reports and recent studies in mice lacking H⁺/K⁺-ATPase isotypes suggest important roles for the kidney H⁺/K⁺-ATPases in acid-base balance as well as K⁺ and Na⁺ homeostasis.⁹⁴

Kidney signal transduction pathways induced by an acidic pH, which mediate the adaptive response of increasing acid excretion, often overlap pathways that activate vascular endothelial cell inflammatory genes. For example, a chronic acid environment that increases angiotensin-II expression has been implicated in vascular inflammation, fibrosis, and extracellular matrix formation.⁹⁵ Angiotensin II mediates these effects via mitogen-activated protein kinases (MAPK; ERK1/2, c-Jun N-terminal kinase, p38MAPK), receptor tyrosine kinases (PDGF, EGF receptor, insulin receptor substrate-1), and nontyrosine receptor kinases (Src, Jak/Stat, focal adhesion kinase) through activation of angiotensin II type 1 receptors on target tissues, including the kidney.^95,96 Angiotensin II type 1 receptor–mediated NADPH oxidase activation generates reactive oxygen species that can trigger inflammation and fibrosis.^97–100 Other examples include acidosis activation of GPR4 in the vascular endothelial cell inflammatory response.^101,102 Pyk2 regulates coupling of receptor proteins to MAPK signaling pathways, such as integrins, GPRs, vascular endothelial growth factor receptor, and EGF receptor.¹⁰³ Additionally, Pyk2 is involved in signaling pathways entailing both ET and angiotensin II, and the control of vasoconstriction and BP.¹⁰⁴ Sustained expression of these pathways resulting from reduced acid excretion in CKD might aggravate fibrosis, inflammation, and BP control, leading to further decline in kidney function.

Mediators of the Adaptive Response to Metabolic Acidosis

Acid retention during CKD leads to adaptive kidney responses to eliminate excess acid. Without complete correction of metabolic acidosis, these responses remain chronically stimulated and directly contribute to kidney damage, as summarized below and in Figures 2–4.

Figure 2.

Angiotensin II: benefits and consequences. Angiotensin II expression is increased in response to metabolic acidosis. Increased intrakidney angiotensin II stimulates NHE3-mediated Na⁺/H⁺ exchange by increasing the content of this transporter in tubule cells, thereby increasing acid excretion. This short-term response can become maladaptive with continual acid exposure, because sustained intrakidney angiotensinII production is linked with proteinuria, tubular atrophy, inflammation, and fibrosis, which together promote further tubule-interstitial injury and worsening CKD. ECM, extracellular matrix; ETS-1, transcription factor erythroblastosis virus E26 oncogen homolog-1; NF-κΒ, nuclear factor κ-light-chain-enhancer of activated B cells; TLR-4, Toll-like receptor 4.

Figure 4.

Ammonia: benefits and consequences. Metabolic acidosis increases ammoniagenesis, leading to increased acid excretion. Increased ammoniagenesis can increase local ammonia concentration surrounding the nephron, triggering activation of complement C3 and C5b-9 and the alternative complement pathway. Damage from this maladaptive complement activation includes increased production of inflammatory and profibrotic mediators, which can promote proteinuria, inflammation, and fibrosis, thus worsening tubule-interstitial injury and CKD. NH₃, ammonia; NH₄⁺, ammonium.

Angiotensin II

Angiotensin II, produced by the intrakidney renin-angiotensin-aldosterone system (RAAS) that is compartmentalized from systemic RAAS, is a central regulatory element controlling the short-term response to metabolic acidosis (Figure 2).^105–108 Within the kidney, proximal tubule cells actively secrete angiotensinogen into the filtrate, which is then converted to angiotensin II in the distal tubules where it promotes tubule acidification.^{106,109–112} Systemic inhibition of angiotensin-II formation by an angiotensin-converting enzyme inhibitor does not reduce intrakidney angiotensinII production.^107,113,114 In contrast, alkali supplementation in patients with CKD and metabolic acidosis—via base-producing food components or oral alkali—reduces urine angiotensinogen, consistent with a reduction in elevated intrakidney angiotensin II, and reduces urine aldosterone.^42,60,61 Elevated intrakidney angiotensin II stimulates proximal tubule acid secretion through upregulation of NHE3.^{45,49,112,115–119} A persistent increase in intrakidney angiotensin II, however, can trigger interstitial inflammation, fibrosis, tubular atrophy, and proteinuria that progressively reduces kidney function.^120–122 Table 2 summarizes the animal and human evidence linking sustained angiotensin-II production to kidney damage.

Table 2.

Effects of the angiotensin II response to metabolic acidosis

Model	Measurements	Observations	Conclusions	References
Rodent model, exposure to NH4Cl	RAAS components gene expression, including angiotensinogen, ACE, AT1R and AT2R	Angiotensinogen, ACE, AT1R, and AT2R were significantly increased (all P<0.05) after 1 wk of exposure to NH4Cl compared with control and NaCl-treated groups. Sustained AT1R expression was found in the NH4Cl group after 8 wk	Sustained expression of intrarenal AT1R is consistent with an increase in kidney angiotensin II accompanying acidosis	49,105
Subtotal nephrectomy in rats	Intrakidney angiotensin II, angiotensin II receptor, intrakidney acid load	Increased acid load and increased intrakidney levels of angiotensin II detected. Angiotensin II receptor antagonism (with valsartan) reduced distal nephron acidification to sham-operated control animal levels. Dietary alkali reduced intrakidney levels of angiotensin II and reduced distal nephron acidification (all P<0.05)	Subtotal nephrectomy increased kidney levels of angiotensin II mediated by acid retention associated with GFR reduction	39,40,112
Kidney physiology studies	Measurement of the effects of angiotensin II on glomerular filtration barrier and selectivity	Angiotensin II decreased the synthesis of negatively-charged proteoglycans and suppressed nephrin transcription, a transmembrane protein that maintains the podocyte structure. Angiotensin II–mediated suppression of nephrin resulted in podocyte apoptosis and overall reduction of GFR (all P values <0.05)	Angiotensin II exhibits direct effects on the integrity of the glomerular filtration barrier	108,123–125
Animal transfection studies	Animal transfection with viral vectors that overexpress renin and angiotensinogen in rat glomeruli	Seven days after transfection, extracellular matrix was expanded in rats with glomerular renin and angiotensinogen overexpression, leading to fibrosis and kidney scarring. Transfected animals that received 4 wk of irbesartan treatment had significantly lower levels of angiotensin II and tubulointerstitial fibrosis (P<0.005) compared with control animals	Angiotensin II is directly involved in kidney fibrosis; fibrosis is a contributing factor in progressive GFR decline in CKD	121,126
Acidemia induced in volunteers with NH4Cl	Measurements of plasma aldosterone concentration, urinary aldosterone secretion; cortisol, corticosterone, deoxycorticosterone, plasma renin, plasma K+, AT1R	Plasma aldosterone concentration and urinary aldosterone secretion was increased, without significant changes in cortisol, corticosterone, deoxycorticosterone, plasma renin activity, or serum K+. Indicates that plasma H+ concentration is a direct and separate regulator of aldosterone secretion, sensitive to net acid load in the body. AT1R blockade by losartan exacerbated the induced acidosis by blocking distal tubular acidification in response to the acid challenge (all P values <0.05)	Acidemia induced in human volunteers with NH4Cl increased RAAS activity, including angiotensin II and AT1R	127–129
Effect of dietary alkali on net acid excretion, ET-1, and aldosterone in subjects with CKD stage 2	Plasma ET-1, aldosterone, NAE in subjects with CKD stage 2	Baseline plasma ET-1 and aldosterone concentrations were each higher in the patients with CKD 2 than in normal control subjects (P<0.05). An oral bolus of NaHCO3 reduced urine NAE less in those with CKD 2 than in normal subjects, consistent with greater acid retention in the subjects with CKD 2 (P<0.05). Thirty days of oral NaHCO3 reduced acid retention in subjects with CKD 2, but not in normal subjects, and reduced plasma ET-1 and aldosterone in both groups but to levels that remained higher in the patients with CKD 2 (P<0.05)	Dietary alkali can reduce acid retained in patients with CKD stage 2 who are eubicarbonatemic, possibly through reduction of ET-1 and aldosterone	38
Effect of dietary intervention on UAGT excretion	UAGT excretion	Patients with reduced GFR and metabolic acidosis had increased urine excretion of angiotensinogen that was decreased after dietary acid reduction with NaHCO3 or base-producing fruits and vegetables (P<0.05)	UAGT reflects kidney levels of angiotensin II; dietary intervention with alkali (fruits and vegetables) can decrease UAGT and kidney angiotensin II levels	61,109,111

NH₄Cl, ammonium chloride; ACE, angiotensin-converting enzyme; AT1R, angiotensin II type 1 receptor; AT2R, angiotensin II type 2 receptor; NaCl, sodium chloride; NAE, net acid excretion; UAGT, urinary angiotensinogen.

Luminal and peritubular angiotensin II contributes to the regulation of ammoniagenesis. Stimulation by luminal angiotensin II increases ammonia entering the lumen, whereas more ammonia exits across the basolateral membrane with peritubular angiotensin II.^29,49,52 Because angiotensin II enhances ammoniagenesis and kidney production of ET-1 and aldosterone (see below), increased angiotensin II might accelerate progression of kidney dysfunction.^{49,120,123–135} Supporting this hypothesis are observations in both animals and humans that correction of metabolic acidosis reduces intrakidney angiotensin II levels and mitigates expression of the short- and long-term effects of this hormone (Table 2).^{39,40,45,49,61,105,112,136} Both bicarbonate supplementation and consumption of base-producing foods reduce angiotensin-II levels and slow the rate of eGFR decline in patients with CKD.^47,57,59,61

ET-1 and Aldosterone

Other mediators of kidney disease progression stemming from the adaptive response to metabolic acidosis include increased ET-1 and aldosterone (Figure 3). ET-1 is an endothelial-derived peptide that is crucial for normal kidney function through its effects on kidney blood flow; glomerular filtration; renin release; and regulation of Na⁺, water, H⁺, and bicarbonate transport.¹³⁷ These effects are exerted through receptor interactions in all parts of the kidney, including mesangial cells, podocytes, endothelium, vascular smooth muscle, and kidney nerves.^{38,122,137,138}

Figure 3.

ET-1: benefits and consequences. ET-1 is activated in response to metabolic acidosis. Increased intrakidney ET-1 stimulates NHE3-mediated Na⁺-H⁺ exchange and increases aldosterone activity in the kidney which increases acid excretion. This short-term response can become maladaptive with continual acid exposure, because sustained intrakidney ET-1 production is linked with proteinuria, inflammation, and fibrosis, which together promote further tubule-interstitial injury and worsening CKD. CTGF, connective tissue growth factor (CCN2); ECM, extracellular matrix; HCO₃⁻, bicarbonate; MCP-1, monocyte chemoattractant protein-1; NF-κΒ, nuclear factor κ-light-chain-enhancer of activated B cells.

Dietary acid intake stimulates ET-1 production, and kidney ET-1 increases proximal and distal tubule H⁺ secretion (both directly, through NHE3 and H⁺-ATPase, and through adrenal ET-1–induced increases in aldosterone production).^{6,25,45,46,139} Rats chronically eating an acid load increase ET-1 production. These sustained ET-1 levels are associated with increased tubule-interstitial damage, inflammation, and fibrosis associated with the synthesis of fibronectin and collagen as well as podocyte effacement, together leading to increased glomerular permeability and overall GFR decline.^{35,63,115,122,140,141} In a remnant kidney model, an acid-producing diet induced tubule-interstitial injury and GFR decline mediated by ET-1, and alkali supplementation ameliorated these effects.^63,140

Stimulation of the intrakidney RAAS in response to continual acid retention might generate additional aldosterone that increases distal nephron acidification, but also results in hemodynamic and profibrotic effects that promote kidney damage.^121,122,135 Chronic metabolic acidosis, induced in four normal subjects by oral administration of the acid precursor ammonium chloride, led to sustained stimulation of aldosterone secretion.¹²⁷ Additional aldosterone in the setting of reduced nephron mass contributed to hypertension, proteinuria, and glomerulosclerosis in the rat remnant model.^142,143 In the same model, alkali administration was associated with reduced kidney cortical aldosterone production.^40,46

Ammoniagenesis

Ammonia-induced complement activation is another mechanism by which metabolic acidosis is thought to cause kidney damage (Figure 4). Although total ammonium excretion decreases as CKD worsens, ammonia generation per functioning nephron increases, helping to maintain increased per-nephron acid secretion in damaged kidneys.^144,145 This adaptive response is deleterious to surviving nephrons, as ammonia activates the alternative complement pathway, increases inflammation, and increases tubulointerstitial damage.^146,147 In studies by Nath et al.,^146,147 nephrectomized rats were treated with NaHCO₃ or Na⁺ chloride. After 4–6 weeks of treatment, rats receiving NaHCO₃ demonstrated less impairment of tubule function (lower urinary protein), less tubulointerstitial damage (histology), less deposition of C3 and C5b-9 complement components, and lowered kidney venous total ammonia concentration compared with rats given Na⁺ chloride. The likely mechanism for these kidney-damaging effects is through ammonia (a nitrogen nucleophile) reacting with C3 to form a convertase that activates the alternative complement system, causing cellular inflammation and damage to kidney parenchyma.^146,147 Additional evidence suggests that elevated angiotensin-II expression associated with an animal model of angiotensin II–induced nephropathy also activates C3 complement. In this model, angiotensin-converting enzyme inhibition reduced cell infiltration and C3 immunoreactivity in 5/6 nephrectomized rats.¹⁴⁸

A cohort study of patients with CKD examined the association between various acid-base indices and the kidney profibrotic marker urine TGF-β1.¹⁴⁹ TGF-β1 was selected for this analysis because it is reduced by oral alkali treatment in patients with CKD, suggesting its linkage with acid-mediated organ injury.^47,150,151 The investigators found direct correlations between urine ammonium excretion and urine TGF-β1. A hypothesis for this association is that high concentrations of intrarenal ammonia and subsequent activation of the alternative complement pathway cause kidney fibrosis, reflected in elevated levels of TGF-β1.^146,152

Cellular Damage to the Kidney Caused by Chronic Metabolic Acidosis

Chronic metabolic acidosis leads to inflammation and fibrosis, resulting in damage to kidney cells (Figure 1). Kidney fibrosis observed in CKD is characterized by the abnormal accumulation of macrophages; the appearance of myofibroblasts recruited to the kidney or derived from cells resident in the kidney; and the deposition of a fibrotic interstitial matrix, including fibronectin and collagen. As fibrosis increases, tubular epithelial cells lose their resorptive capacity and the kidney decreases in size.¹⁵³

The molecular mechanisms of inflammation associated with CKD are beyond the scope of this review; however, a summary of both innate and adaptive inflammatory processes in CKD has recently been published.¹⁵⁴ Elements of the adaptive response to acidosis, particularly upregulation of angiotensin II, have been associated with kidney infiltration by leukocytes.^121,122,155 In vitro and in vivo evidence suggests that angiotensin II promotes cell proliferation and fibroblast activation, worsening abnormal accumulation of extracellular matrix in the damaged kidney and contributing to development of kidney fibrosis and inflammation.^108,156

Metabolic acidosis induces oxidative stress in the kidney that can stimulate further inflammation and fibrosis, extending the damage in the failing kidney.^157–159 CKD produces elevated oxidative stress, as evidenced by an increased concentration of reactive oxygen and nitrogen species, oxidized end products, and reduced levels of antioxidants in patients.^160,161 In cultured LLC-PK1 cells exposed to a chronic acid environment, total cellular glutathione and protein thiol content was reduced, along with an increase in glutathione peroxidase activity, ammonia generation, and the expression of heat shock proteins HSP70 and HSP60, all markers of oxidative stress.¹⁶² In Nox-deficient mouse models of diabetic nephropathy, Nox-family NADPH oxidases have been identified as mediators of kidney cell injury, responsible for diabetes-induced injury in glomerular and tubulointerstitial cells.¹⁶³ Finally, oxidative stress and a chronically elevated inflammatory state in CKD may contribute to accelerated atherosclerosis via direct endothelial injury and alteration in nitrogen handling.¹⁵⁸

In animal models, prolonged acid exposure directly damages kidney substructures, such as podocytes and the glomerular filtration apparatus, and disrupts the signaling within proximal and distal tubules that is needed to maintain acid-base balance.^36,37 Sustained acid exposure causes cortical and tubular necrosis, and treating metabolic acidosis with calcium citrate in 5/6 nephrectomized rats ameliorated glomerular and tubule-interstitial damage as measured by reduction of glomerular and tubular proliferating cell nuclear antigen, a sensitive marker of direct cellular damage in the kidney.^164–168 Oral NaHCO₃ treatment in patients who were acidemic reduced tubular protein catabolism and markers of tubular injury.¹⁶⁶ In these studies, kidney tubular catabolism of technetrium-99m–labeled aprotinin, measuring both metabolism and fractional excretion of this tubule-specific protein, was reduced after treatment; additionally, urinary N-acetyl-β-D-glucosaminidase, an index of kidney tubulointerstitial injury and glomerular filtration, was reduced, suggesting that treating acidosis has the potential to protect kidney tubule structure and function.

Investigators have recently used gene expression analysis to explore the effect of metabolic acidosis on kidney injury. Acid stress upregulated expression of genes in cultured MDCK cells (an immortalized cell line derived from the dog distal kidney tubule) that participate in the proinflammatory process, leading to glomerulosclerosis and fibrosis.¹⁶⁹ A comprehensive genomics study examining the transcriptome, proteome, and metabolome associated with the response to chronic metabolic acidosis in CKD is warranted.^85,102

Metabolic Damage Associated with Prolonged Metabolic Acidosis

Acid accumulation can affect kidney function in a variety of important metabolic processes, including sensitivity to insulin. As kidney function declines, metabolic acidosis and insulin resistance are common.^170–173 Because CKD and metabolic acidosis cause significant inflammation, regulatory molecules that control the extent of inflammation and also affect sensitivity to insulin (e.g., angiotensin II, the tyrosine phosphatase signal regulatory protein-α) may be disordered, causing insulin resistance. Experiments in human volunteers with normal kidney function using euglycemic and hyperglycemic clamps revealed that ammonium chloride–induced metabolic acidosis reduced tissue sensitivity to insulin.¹⁷⁴ Studies in patients with CKD demonstrate impaired glucose tolerance and insulin resistance, particularly before and after the initiation of hemodialysis.¹⁷ The defect in insulin sensitivity associated with metabolic acidosis results from alteration in downstream signaling and not an alteration in insulin receptor binding.^175,176

A recent study evaluated whether metabolic acidosis treated with NaHCO₃ reduced insulin use in people with type 2 diabetes and CKD.¹⁷⁷ The investigators prospectively explored the association of serum bicarbonate as a continuous variable and insulin resistance over a broad range of values of serum bicarbonate (i.e., from 18 to 31 mEq/L). The data showed that this association is nonlinear and insulin sensitivity decreased for values of serum bicarbonate <24 mEq/L and >28 mEq/L, with the optimum range for increased insulin sensitivity between 24 and 28 mEq/L, confirming a link between serum bicarbonate and insulin resistance.

Patients with CKD on average have shorter life spans due in part to higher rates of cardiovascular disease characterized by vascular injury with calcification.¹⁷⁸ As such, CKD might be viewed as a human model of accelerated aging.¹⁷⁹ An animal model of accelerated aging, the k1/k1 mouse model, with deficient expression of the klotho gene, is used to study the kidney’s role in aging and tissue calcification. Kidneys are the main source of klotho and klotho expression is significantly reduced in k1/k1 mice, as it is in patients with CKD.¹⁸⁰ In k1/k1 mice, marked atherosclerosis with extensive medial and tissue calcification is accompanied by increased levels of aldosterone and antidiuretic hormone. In this model, NaHCO₃ supplementation reduced serum aldosterone and antidiuretic hormone, while increasing phosphaturia, and also reduced vascular and tissue calcification.¹⁸¹ The authors concluded that NaHCO₃ delays tissue calcification and premature death in k1/k1 mice, partly attributable to reduction of aldosterone and antidiuretic hormone, as well as reduced intestinal phosphate absorption. If the reduced klotho expression associated with CKD contributes to progressive arteriosclerosis and vascular calcification, the effects of dietary acid reduction, using NaHCO₃ or dietary measures, might attenuate these effects while reducing the overall kidney damage promoted by acid retention.

In conclusion, metabolic acidosis associated with CKD exacerbates kidney-function decline, increasing the risk of progression to ESKD. The acid that accumulates in many patients with CKD increases production of intrakidney hormones angiotensin II, aldosterone, and ET-1, which acutely increase kidney acid excretion and thus ameliorate acid retention, but chronically promote inflammation and fibrosis leading to a decline in kidney function. Similarly, ammoniagenesis stimulated in response to metabolic acidosis increases net acid excretion but can cause complement-mediated damage with worsening proteinuria, inflammation, and fibrosis. Treatment of metabolic acidosis attenuates these adaptive responses and may help to preserve kidney function. As the design of future studies for the treatment of metabolic acidosis in CKD evolve, an appreciation of the varied mechanisms that cause kidney damage in patients with acid accumulation will help refine and focus these studies.

Disclosures

Dr. Bushinsky is a member of medical advisory boards at Tricida and reports consulting fees, stock, and stock options from Tricida during and outside this work. Dr. Bushinsky was the lead investigator for the phase 1/phase 2 study of veverimer (TRCA-101) sponsored by Tricida and is on the advisory board for the ongoing VALOR-CKD study sponsored by Tricida. Dr. Bushinsky also reports stock ownership in Amgen; consulting fees from Amgen, Fresenius/Relypsa/Vifor, and Sanofi/Genzyme; past stock ownership in Relypsa; personal fees as a medical advisory board member from Sanifit; and speaker fees from Sanofi/Genzyme, all outside this work. Dr. Bushinsky reports grant support from the National Institutes of Health and Renal Research Institute, both outside this work. Dr. Buysse reports consulting fees and stock and stock options from Tricida during and outside this work; Dr. Buysse is listed on granted and pending Tricida patents. Dr. Buysse also reports consulting fees from Orbimed Advisors, outside this work. Dr. Wesson is a member of a medical advisory board at Tricida and reports consulting fees from Tricida during and outside this work. Dr. Wesson also reports grant support from the National Institutes of Health, outside this work. Dr. Wesson was the lead investigator for the phase 3 studies of veverimer (TRCA-301 and TRCA-301E) sponsored by Tricida and is on the advisory board for the ongoing VALOR-CKD study sponsored by Tricida. This work was produced without direct financial support. Dr. Wesson receives salary support paid through his employing institution from Tricida, Inc. and from the National Institutes of Health (R21DK113440). Dr. Bushinsky receives grant support outside of the work from the National Institutes of Health (R01 DK 75462) and the Renal Research Institute.