Proximal tubule bicarbonate reabsorption involves several interconnected steps ( Fig. 9.1 ). First, H ⁺ is secreted into the luminal fluid. Apical Na ⁺ /H ⁺ exchanger (NHE)-isoform 3 (NHE3) mediates 60% to 70% of H ⁺ secretion and H ⁺ -ATPase the remainder. In the neonatal kidney, the Na ⁺ /H ⁺ exchanger, NHE8, substitutes for NHE3. Secreted H ⁺ combines with luminal HCO ₃ ^– to form carbonic acid (H ₂ CO ₃ ), which dissociates into H ₂ O and CO ₂ . This reaction is catalyzed by carbonic anhydrase IV (CA IV), which is an apical membrane-bound protein. Luminal CO ₂ then moves across the apical plasma membrane into the cell through both protein-mediated mechanisms and lipid diffusion. Cytosolic CO ₂ is then hydrated to form carbonic acid through a process accelerated by cytosolic carbonic anhydrase II (CA II). Carbonic acid rapidly dissociates to H ⁺ and HCO ₃ ^– , replenishing the H ⁺ secreted across the apical plasma membrane. In the S1 and S2 segments, the primary base exit mechanism involves the A splice variant of sodium-coupled, electrogenic bicarbonate cotransporter, NBCe-1A, ^, and a secondary mechanism is the B splice variant, NBCe1-B. ^, In the S3 segment, a Na ⁺ -dependent, Cl ^– /HCO ₃ ^– exchanger is the primary basolateral HCO ₃ ^– transport mechanism.

Bicarbonate reabsorption in the proximal tubule.

Protons are secreted into the luminal fluid by the Na ⁺ /H ⁺ exchanger, NHE3, and H ⁺ -ATPase, and titrate luminal HCO ₃ ^– to H ₂ CO ₃ . Approximately two-thirds of H⁺ secretion is mediated by NHE3 and the remainder by H⁺-ATPase. Luminal H ₂ CO ₃ dehydration to H ₂ O and CO ₂ is accelerated by apical CA IV-mediated carbonic anhydrase activity. CO ₂ enters the cell via AQP1 and passive lipid-phase diffusion, where cytoplasmic CA II accelerates its hydration to H ₂ CO ₃ . H ₂ CO ₃ dissociates to H ⁺ and HCO ₃ ^– , thereby “providing” cytosolic H ⁺ for continued apical secretion. Two HCO₃ ^– convert to H ₂ CO ₃ and CO ₃ ^2– , and the CO ₃ ^2– exits via basolateral NBCe1 proteins, of which NBCe1-A is the predominant splice variant in the cortex and NBCe1-B is the only splice variant in the PST in the outer medulla. Basolateral CA IV generates H ₂ CO ₃ , which dissociates to HCO₃– and H⁺, and the latter converts interstitial CO ₃ ^2– to HCO₃–, completing the process of luminal HCO₃⁻ reabsorption. Na⁺ that entered via apical NHE3 in the process of HCO₃– reabsorption exits either via basolateral NBCe1-A/B (∼75%) or Na⁺-K⁺-ATPase (∼25%). Exit via Na⁺-K⁺-ATPase maintains the low intracellular Na⁺ concentration necessary for continued apical NHE3 activity. In the PST, a basolateral Cl ^– /HCO ₃ ^– exchange activity (not shown) is the primary basolateral HCO ₃ ^– exit mechanism.

The proximal tubule also exhibits passive bicarbonate secretion, termed “backleak.” Backleak involves both paracellular and transcellular components ^, and is inhibited by AngII. Bicarbonate backleak contributes to the proximal tubule’s inability to lower luminal bicarbonate below ∼6 mmol/L.

Whether adaptive changes in TAL bicarbonate transport contribute to acid-base homeostasis is controversial. In one study, NH ₄ Cl loading (to induce metabolic acidosis), NaCl loading, and NaHCO ₃ loading each increased TAL bicarbonate reabsorption similarly, suggesting the stimulus was solute loading, not acid-base homeostasis. In another study, NH ₄ Cl-induced metabolic acidosis, but not NaCl loading, increased TAL NHE3 expression.

Several plasma membrane proteins either directly or indirectly alter bicarbonate reabsorption. Inhibiting the apical Na ⁺ -K ⁺ -2Cl ^– cotransporter, NKCC2, such as with a loop diuretic, increases bicarbonate reabsorption. This may reflect decreased NKCC2-mediated apical Na ⁺ entry decreasing intracellular Na ⁺ , thus increasing the gradient for apical NHE3 activity, and thereby increasing bicarbonate reabsorption. Inhibiting basolateral Na ⁺ /H ⁺ exchange activity decreases bicarbonate reabsorption through cytoskeletal alterations that decrease apical NHE3 expression.

Several hormones regulate bicarbonate reabsorption. AngII stimulates TAL bicarbonate reabsorption, likely through activation of AT ₁ receptors. ^, Glucocorticoid receptors are present in the TAL, and glucocorticoid hormones are necessary for basal bicarbonate reabsorption. Mineralocorticoids, at high concentrations, stimulate bicarbonate reabsorption, but their absence does not alter basal transport. Both AVP and PTH inhibit bicarbonate reabsorption through inhibition of apical Na ⁺ /H ⁺ exchange activity. ^, Proinflammatory cytokines regulate bicarbonate transport. Lipopolysaccharide (LPS) inhibits transport through the innate immune system receptor, TLR4. Luminal LPS involves the mTOR pathway, whereas peritubular LPS functions through the MEK/ERK pathway. ^,

Medullary osmolality is another crucial regulatory factor. Increased tonicity inhibits and decreased tonicity stimulates bicarbonate reabsorption; this occurs through phosphatidylinositol 3-kinase-mediated changes in apical Na ⁺ /H ⁺ exchange activity. ^,

The major transporters are discussed earlier in relation to the proximal tubule, and they are not repeated here. Please see the relevant section earlier.

The A-cell reabsorbs luminal HCO ₃ ^– ( Fig. 9.3 ). Apical H⁺ secretion involves vacuolar H ⁺ -ATPase and the P-type H ⁺ -K ⁺ -ATPase. A Na⁺-independent anion exchanger that is a truncated isoform of the erythrocyte anion exchanger, termed kAE1, mediates basolateral bicarbonate exit. Cl ^– then exits through the combined actions of the KCl cotransporter, KCC4, ^, and the basolateral Cl ^– channel, ClC-K2. ^, Cytoplasmic CA II enables intracellular H ⁺ and HCO ₃ ^– generation. Apical membrane-associated CA IV and basolateral CA XII are present.

Bicarbonate transport by the A-cell and B-cell.

Top panel, a model of acid-base transport by the type A intercalated cell. Two families of H ⁺ transporters, H ⁺ -ATPase and H ⁺ -K ⁺ -ATPase, are present in the apical plasma membrane. Secreted H ⁺ titrates luminal HCO ₃ ^– to form H ₂ CO ₃ , which dehydrates to water (H ₂ O) and CO ₂ . Cytosolic H ⁺ and HCO ₃ ^– are formed from CA II–accelerated hydration of CO ₂ to form H ₂ CO ₃ , which rapidly dissociates to H⁺ and HCO₃⁻. Cytosolic HCO ₃ ^– exits across the basolateral plasma membrane via the anion exchanger, kAE1. Cl ^– that enters via kAE1 recycles via a basolateral Cl ^– channel. K ⁺ that enters via apical H ⁺ -K ⁺ -ATPase can either recycle via an apical, Ba ⁺ -sensitive K ⁺ channel or the K-Cl cotransporter, KCC4. Bottom panel, H⁺/HCO₃– transport by the type B intercalated cell. Apical pendrin is the primary mechanism of bicarbonate secretion. Chloride enters the cell via pendrin and exits across a basolateral chloride channel, ClC-K2/ClC-Kb, resulting in Cl⁻ reabsorption coupled to HCO₃– secretion; recycles via apical CFTR, resulting in HCO₃– secretion without net Cl⁻ reabsorption; or couples with apical NDCBE to mediate net NaCl reabsorption without net HCO₃– transport. Basolateral H ⁺ -ATPase extrudes protons into the peritubular compartment. Cytoplasmic bicarbonate and protons are produced from CO ₂ and water in a CA II–catalyzed reaction. In addition, an apical H ⁺ -K ⁺ -ATPase is present. Basolateral AE4 is present and regulates net HCO₃– secretion via the regulation of apical pendrin.

The type B intercalated cell secretes HCO₃– and reabsorbs Cl⁻ ( Figs. 9.2 and 9.3 ). H ⁺ -ATPase is present in the basolateral plasma membrane and subapical vesicles but not in the apical plasma membrane, and it secretes H⁺ across the basolateral plasma membrane. HCO₃ ^– is secreted by the apical electroneutral Cl ^– /HCO₃ ^– exchanger, pendrin. Cl– that enters via pendrin can either recycle across the apical plasma membrane via the apical Cl ^– channel, CFTR, or the apical K⁺-Cl ^– cotransporter, KCC3A, or can exit the cell via the basolateral Cl ^– channel ClCK2/Kb. Cytoplasmic CA II facilitates H⁺ and HCO₃ ^– generation. The coupling of pendrin (see later) with these different Cl⁻ transporters enables various modes of net ion transport. Coupling to CFTR results in electrogenic HCO₃ ^– secretion. Coupling to ClCK2 results in HCO₃ ^– secretion with Cl ^– reabsorption, which may be beneficial in states of Cl ^– -depletion metabolic alkalosis. Pendrin may also be coupled to a Na⁺-dependent Cl ^– /HCO₃ ^– exchanger (NDCBE), in which two cycles of pendrin-mediated Cl–/HCO₃– exchange are coupled to one cycle of NDCBE and result in net NaCl reabsorption but no net acid-base transport.

The non-A, non-B cell is present in significant numbers only in the CNT and ICT. ^, It is characterized by the apical plasma membrane pendrin and H ⁺ -ATPase expression and the absence of basolateral plasma membrane H ⁺ -ATPase and AE1 (see Fig. 9.2 ). During development, the non-A, non-B cell arises from different foci than the B-cell, suggesting they are fundamentally different cell types. ^,

Principal cells contribute to bicarbonate reabsorption through indirect and direct roles. Principal cell Na ⁺ reabsorption leads to luminal electronegativity, which facilitates H ⁺ secretion by the electrogenic H ⁺ -ATPase. In rodent models, OMCDi principal cells have apical H ⁺ secretory and basolateral Cl ^– /HCO ₃ ^– exchange activities, and they express H ⁺ -ATPase and CA II, ^, which suggests they may contribute to bicarbonate reabsorption.

The IMCD cell is a distinct cell type and is the predominant cell present in the terminal IMCD. It exhibits carbonic anhydrase activity, both H ⁺ -ATPase and H ⁺ -K ⁺ -ATPase activity, ^, and basolateral Cl ^– /HCO ₃ ^– exchange activity. However, the expression of each of these is significantly less than in CCD and OMCD intercalated cells and generally cannot be detected by routine immunohistochemistry techniques.

The CNT and ICT contain type A and type B intercalated cell types and non-A, non-B cells, enabling them to contribute to acid-base homeostasis. Under basal conditions, the CNT secretes bicarbonate.

Unlike the OMCD and IMCD, which only reabsorbs bicarbonate, the CCD simultaneously reabsorbs and secretes bicarbonate. Net transport correlates with systemic acid-base status, with metabolic acidosis stimulating net reabsorption and metabolic alkalosis stimulating net secretion. The ability to secrete bicarbonate, which is not found in the OMCD or IMCD, correlates with the presence of B-cells in the CCD but not in the OMCD or IMCD.

The OMCD is responsible for approximately 40% to 50% of the net acid secretion in the collecting duct. Both intercalated cells and principal cells contribute to acid secretion, but intercalated cells are the primary cells responsible for OMCD acid secretion. ^,

The IMCD reabsorbs luminal bicarbonate, but absolute rates are substantially lower than in other collecting duct segments. This parallels the smaller number of A-cells in the IMCD.

The vacuolar H ⁺ -ATPase is an assembly of multiple subunits that form two main domains: the V ₁ domain, which is extramembranous and hydrolyzes ATP, and the V ₀ domain, which is transmembranous and transports protons. Distinct isoforms and splice variants have been identified for many of these H ⁺ -ATPase subunits, and their cell-specific distribution may contribute to cell-specific regulation of proton and bicarbonate transport. Normal H⁺-ATPase function appears to require the Atp6ap2/(Pro)renin Receptor. This is a type 1 transmembrane protein that is an accessory subunit of H⁺-ATPase. Although it is capable of binding prorenin, prorenin binding does not alter H⁺-ATPase activity.

A second mechanism of collecting duct H ⁺ secretion involves electroneutral H ⁺ -K ⁺ exchange coupled to ATP hydrolysis. The active protein is a heterodimer composed of α- and β-subunits. The α-subunit is an integral membrane protein with multiple membrane-spanning domains and contains the catalytic portion of the enzyme. Two α-subunit isoforms, HKα1 and HKα2, are present. HKα ₁ forms heterodimers with its specific β-subunit, HKβ. The β-subunit has only a single membrane-spanning region and is necessary for targeting of the α-subunit to the plasma membrane and for transport function. HKα ₂ forms heterodimers with the β ₁ -subunit of Na ⁺ -K ⁺ -ATPase.

The collecting duct is the final site controlling renal acid-base regulation. It responds quickly to physiologic conditions to increase acid or bicarbonate excretion as needed to maintain systemic acid-base homeostasis.

A second adaptive response mechanism used by the collecting duct also involves changes in the number of intercalated cells. Multiple conditions including metabolic acidosis, hypokalemia, and chronic lithium or acetazolamide administration are associated with an increased number of acid-secreting type A intercalated cells and decreased type B intercalated cells.

Increases in intercalated cell numbers could result from intercalated cell proliferation or principal cell proliferation, followed by conversion into intercalated cells. Metabolic acidosis, hypokalemia, and lithium administration are all associated with increased proliferation of collecting duct cells. ^{, ,} Some studies increased A-cell proliferation ^, while others show the proliferating cells are principal cells. ^{, , ,}

One major pathway that appears to mediate these changes involves endogenous Notch signaling. Multiple studies show a crucial role for Notch signaling in the interconversion of principal cells and intercalated cells. The transcription factors Elf5 and TFCP2L1 appear involved in Notch pathway–dependent principal cell–intercalated cell interconversion. ^,

Ammonia metabolism, a term used to indicate the combination of ammonia generation and transport, is fundamental to acid-base homeostasis. Ammonia is generated in the kidney through a process that produces an equal amount of bicarbonate. Ammonia transport along the nephron and collecting duct ( Fig. 9.5 ) determines the amount of ammonia excreted in the urine as opposed to entering the peritubular capillaries and being returned via the renal veins. Ammonia added to the renal vein is metabolized in the liver and skeletal muscle in an equimolar bicarbonate-consuming reaction and therefore does not contribute to acid-base homeostasis.

An integrated overview of renal ammonia metabolism.

Renal ammoniagenesis occurs primarily in the proximal tubule and generates NH₄⁺ and bicarbonate. NH₄⁺ is preferentially secreted into the luminal fluid. Ammonia reabsorption in the thick ascending limb, involving apical NKCC2-mediated uptake, results in medullary ammonia accumulation. Medullary sulfatides (highlighted in green) reversibly bind NH ₄ ⁺ , contributing to medullary accumulation. Ammonia is secreted in the collecting duct via parallel H ⁺ and NH ₃ secretion. The numbers in blue represent the proportion of total excreted ammonia at each location. Gsc, Galactosylceramide backbone.

Ammonia exists in two molecular forms, NH ₃ and NH ₄ ⁺ . NH ₃ , although uncharged, has a significant molecular polarity that limits NH ₃ movement across lipid bilayers. NH ₄ ⁺ has nearly identical biophysical characteristics as K⁺, which enables K⁺ transporters also to transport NH₄⁺.

Renal ammoniagenesis derives almost exclusively from glutamine metabolism. This process involves both apical and basolateral glutamine transport, mitochondrial glutamine metabolism, cytoplasmic malate metabolism, and differential entry of phosphoenolpyruvate into either the tricarboxylic acid (TCA) cycle or utilization for gluconeogenesis. Fig. 9.6 summarizes these processes.

Proximal tubule ammoniagenesis.

Glutamine uptake involves apical transport via B ^o AT1 and basolateral uptake via SN1. Apical glutamine uptake that is not used for ammoniagenesis exits via basolateral LAT2, enabling net glutamine reabsorption under basal conditions. Mitochondrial glutamine is then metabolized by PDG and GDH, yielding 2-oxoglutarate and the release of two NH₄⁺. 2-Oxoglutarate is converted to malate in the TCA cycle and then exits the mitochondria. Malate is metabolized to phosphoenolpyruvate (PEP) through a PEPCK-dependent pathway and then either converted to pyruvate via pyruvate kinase (PK) or 2P-glycerate via enolase. Pyruvate enters mitochondria, where its metabolism through the TCA cycle leads to HCO₃⁻ and ATP generation. 2P-glycerate is metabolized through the gluconeogenic pathway, which also generates HCO₃– and ATP, but the amount of ATP generated is less than that generated through the pyruvate-dependent TCA cycle.

Glutamine used for ammoniagenesis derives from the uptake of both luminal and peritubular glutamine. The primary apical transporter is B ⁰ AT1 (SLC6A19), with a lesser contribution from B ⁰ AT3 (SLC6A18). Under basal acid-base conditions, luminal glutamine reabsorption is greater than needed for ammoniagenesis and the remainder exits across the basolateral plasma membrane, leading to net reabsorption. During acidosis, total glutamine uptake can exceed luminal glutamine, and basolateral glutamine uptake mechanisms are stimulated. This basolateral uptake occurs through the Na⁺-coupled, neutral amino acid transporter, SN1 (SLC38A3). Basolateral SN1 expression is found only in the S3 segment under basal conditions. In conditions that increase ammoniagenesis, S3 segment expression increases and expression is induced in the S2 segment. ^, Glutamine is then transported into mitochondria, where the initial ammoniagenic steps occur.

Glutamine metabolism then leads to the production of 2 NH₄⁺ and 2 HCO₃ ^– molecules from each glutamine. The initial step involves mitochondrial deamidation and deamination by glutaminase (GLS1) and glutamate dehydrogenase (GLDH), respectively, leading to 2-oxoglutarate formation and release of two NH₄⁺ from each glutamine. 2-oxoglutarate is metabolized in the TCA cycle to malate, which exits mitochondria via the mitochondrial dicarboxylate transporter, DIC (SLC25A10). Cytoplasmic malate is then converted to phospho enol pyruvate (PEP) by phospho enol pyruvate carboxykinase (PEPCK). PEP is then a substrate for the TCA cycle and gluconeogenesis. Both pathways generate 2 HCO₃⁻ molecules per PEP. Because PEPCK is found only in the proximal tubule, glutamine-dependent, ammonia-associated bicarbonate generation is limited to this segment. PEP metabolism through the gluconeogenic pathway enables the kidneys to be an important source of glucose during starvation and during hypokalemia. However, ATP generation is substantially greater through the TCA cycle, ∼54 ATP/glutamine, than in the gluconeogenic pathway, ∼14 ATP/glutamine. There is also a minor component of ammonia recycling, which involves the conversion of glutamate back to glutamine via the enzyme glutamine synthetase.

Ammonia produced in the kidney undergoes both reabsorption and secretion in the nephron and collecting duct. In the proximal tubule, it is secreted preferentially into the tubule lumen, which is followed by reabsorption in the TAL. Proximal tubule apical secretion results from NHE3-mediated NH ₄ ⁺ secretion and from luminal acidification, which, by decreasing luminal NH₃ concentration, increases the apical gradient for NH₃ secretion. ^, The TAL reabsorbs luminal ammonia via the substitution of NH₄⁺ for K⁺ and reabsorption via the apical Na ⁺ -K ⁺ -2Cl ^– cotransporter, NKCC2. Basolateral ammonia exit involves both NHE4-mediated Na ⁺ /NH ₄ ⁺ exchange activity and dissociation of NH ₄ ⁺ into NH ₃ and H ⁺ , with NH₃ transport across the basolateral plasma membrane, and basolateral HCO₃⁻ entry via NBCn1 that buffers the released H⁺. Some ammonia absorbed by the TAL is then secreted by the thin descending limb of the loop of Henle. The net effect of this transport in the proximal tubule and loop of Henle is counter-current amplification and the development of an axial ammonia gradient. The net effect of ammonia transport in the loop of Henle is that ammonia exiting the loop of Henle accounts for only 20% to 40% of urinary ammonia, with the remainder deriving ammonia secretion in more distal segments.

There may be a small component of ammonia transport in the DCT, CNT, and ICT. Studies in rat show low rates of ammonia secretion. ^, Ammonia reabsorption may also occur under certain circumstances. Acute increases in ammonia delivery to the DCT and CNT increase blood ammonia levels, suggesting concentration-dependent ammonia reabsorption occurs.

Collecting duct ammonia secretion is responsible for the majority of urinary ammonia. This ammonia secretion involves several transporters ( Fig. 9.7 ). Basolateral uptake in the CCD and OMCD primarily involves the Rhesus glycoproteins, Rhbg and Rhcg, ^, although the HCN2 channel may also contribute. In the IMCD, but not the CCD, basolateral Na ⁺ -K ⁺ -ATPase contributes to basolateral NH ₄ ⁺ uptake. Apical ammonia secretion involves parallel NH ₃ and H ⁺ transport. H ⁺ secretion likely involves both H ⁺ -ATPase and H ⁺ -K ⁺ -ATPase. Apical Rhcg mediates the majority of apical NH ₃ secretion, and there may also be a smaller component of diffusive transport.

Mechanisms of collecting duct ammonia secretion.

Rhbg and Rhcg likely both contribute to basolateral ammonia uptake. Rhbg appears to transport ammonia in both molecular forms, NH ₄ ⁺ or NH ₃ , both of which enable ammonia uptake across the basolateral membrane. Rhcg mediates electroneutral NH ₃ uptake across the basolateral plasma membrane. NH ₃ is then secreted across the apical plasma membrane down its electrochemical gradient through processes involving both apical Rhcg transport and a separate mechanism that may reflect lipid-phase diffusion. H ⁺ ions are secreted across the apical plasma membrane by both H ⁺ -ATPase and H ⁺ -K ⁺ -ATPase. Cytoplasmic H ⁺ are supplied by either dissociation of NH ₄ ⁺ or via a carbonic anhydrase II (CA II)-dependent bicarbonate shuttle mechanism involving basolateral chloride-bicarbonate exchange and basolateral chloride channel-mediated bicarbonate shuttling. Also shown is NH ₄ ⁺ uptake by basolateral Na ⁺ -K ⁺ -ATPase, which contributes to ammonia secretion in the IMCD, where most cells do not express Rh glycoproteins.