The erythrocytes and renal tubular cells generate HCO₃⁻, the buffer base in the bicarbonate system, from CO₂. Under physiological conditions:

Bicarbonate is produced following the dissociation of carbonic acid formed from CO₂ and H₂O. This is catalysed by carbonate dehydratase (CD; carbonic anhydrase), which is present in high concentrations in erythrocytes and renal tubular cells:

Erythrocytes and renal tubular cells have high concentrations of CD, but also have means of removing one of the products, H⁺; thus both reactions continue to the right, and HCO₃⁻ is formed. One of the reactants (water) is freely available and one of the products (H⁺) is removed. Bicarbonate ion generation is therefore accelerated if the concentration of:

Therefore, an increase in intracellular PCO₂ or a decrease in intracellular [HCO₃⁻] in the erythrocytes and renal tubular cells maintains the extracellular [HCO₃⁻] by accelerating the production of HCO₃⁻. This minimizes changes in the ratio of [HCO₂⁻] to PCO₂ and therefore changes in pH.

In the normal subject, at a plasma PCO₂ of 5.3 kPa (a [CO₂] of about 1.2 mmol/L), erythrocytes and renal tubular cells maintain the extracellular HCO₃⁻ at about 25 mmol/L. The extracellular ratio of [HCO₃⁻] to [CO₂] (both in mmol/L) is just over 20:1. It can be calculated from the Henderson-Hasselbalch equation, and with a pK_a of 6.1 this represents a pH very near 7.4. An increase of intracellular PCO₂, or a decrease in intracellular [HCO₃⁻], accelerates HCO₃⁻ production and minimizes changes in the ratio and therefore in pH.

Haemoglobin is an important blood buffer. However, it only works effectively in cooperation with the bicarbonate system:

Erythrocytes produce little CO₂ as they lack aerobic pathways. Plasma CO₂ diffuses along a concentration gradient into erythrocytes, where CD catalyses its reaction with water to form carbonic acid (H₂CO₃), which then dissociates. Much of the H⁺ is buffered by Hb, and the HCO₃⁻ diffuses out into the extracellular fluid along a concentration gradient.

Electrochemical neutrality is maintained by diffusion of Cl⁻ in the opposite direction into cells. This movement of ions is known as the ‘chloride shift’.

Under normal circumstances the higher PCO₂ in the blood leaving tissues stimulates erythrocyte HCO₃⁻ production; consequently the arteriovenous difference in the ratio of [HCO₃⁻] to [CO₂], and therefore the pH, is kept relatively constant.

Extracellular and intracellular buffers other than HCO₃⁻ and Hb do not contribute significantly to blood buffering. They include:

Normal urine is nearly HCO₃⁻ free. An amount equivalent to that filtered by the glomeruli is returned to the body by the tubular cells. The luminal surfaces of renal tubular cells are impermeable to HCO₃⁻. Thus, HCO₃⁻ can only be returned to the body if first converted to CO₂ in the tubular lumina, and an equivalent amount of CO₂ is converted to HCO₃⁻ within tubular cells. The mechanism depends on the
action of CD, both in the brush border on the luminal surfaces and within tubular cells, and on H⁺ secreted into the lumina in exchange for sodium (Na+). This occurs predominantly in the proximal tubules but also in the first part of the distal tubules:

This self-perpetuating cycle reclaims buffering capacity that would otherwise have been lost from the body by glomerular filtration. The secreted H⁺ is derived from cellular water, and is incorporated into water in the lumina. Because there is no net change in H⁺ balance and no net gain of HCO₃⁻, this mechanism cannot correct an acidosis but can maintain a steady state.

The mechanism in renal tubular cells for generating HCO₃⁻ is identical to that of HCO₃⁻ reabsorption, but there is net loss of H⁺ from the body as well as a net gain of HCO₃⁻. Therefore this mechanism is well suited to correcting any type of acidosis.

Within tubular cells, CD may be stimulated by the following.

Normally almost all the filtered HCO₃⁻ is reabsorbed. Once the luminal fluid is HCO₃⁻ free, continued secretion of H⁺ and the intracellular generation of HCO₃⁻ depend on the presence of other filtered buffer bases (B⁻). In their absence, the luminal acidity would increase so much that further H⁺ secretion would be inhibited. These buffers, unlike HCO₃⁻, do not form compounds capable of diffusing back into tubular cells, nor is H⁺ incorporated into water.

There is net loss of H⁺ in urine as HB. The HCO₃⁻ formed in the cell is derived from cellular CO₂ and therefore represents a net gain in HCO₃⁻. Whenever 1 mmol of H⁺ is secreted into the tubular lumen, 1 mmol of HCO₃⁻ passes into the ECF with Na+. The mechanism of HCO₃⁻ generation is very similar to that in erythrocytes. However, unlike red cells, renal tubular cells are exposed to a relatively constant PCO₂. Bicarbonate generation, coupled with H⁺ secretion, becomes very important in acidosis, when it is stimulated by either a fall in extracellular [HCO₃⁻] (metabolic acidosis) or a rise in extracellular PCO₂ (respiratory acidosis).

At pH 7.4, most of the phosphate in plasma, and also in the glomerular filtrate, is monohydrogen phosphate (HPO₄²−), which can accept H⁺ to become dihydrogen phosphate (H₂PO₄⁻). Bicarbonate can continue to be generated within tubular cells, with H⁺, and to be returned to the body after all that in the filtrate has been reabsorbed. Therefore it can help to replace that used in extracellular buffering. The pK_a of this buffer pair is about 6.8:

Phosphate is normally the most important buffer in the urine because its pKa is relatively close to the pH of the glomerular filtrate and because the concentration of phosphate increases 20-fold, to nearly 25 mmol/L, as water is reabsorbed from the tubular lumen.

Even in a mild acidosis, more phosphate ions are released from bone than at normal pH; the need for increased urinary H⁺ secretion is linked with increased buffering capacity in the glomerular filtrate owing to the increase of phosphate. At a urinary pH below 5.5, most of the filtered phosphate is converted to dihydrogen phosphate. Therefore, at low pH, urinary phosphate cannot maintain the essential buffering of continued H⁺ secretion. The predominant urinary anion is Cl⁻, but, because hydrochloric acid is almost completely ionized in aqueous solution, it cannot act as a buffer.

As the urine becomes more acid, it can be shown to contain increasing amounts of ammonium ion (NH₄ +). Urinary ammonia probably allows H⁺ secretion, and therefore HCO₃⁻ formation, to continue after other buffers have been depleted.

Ammonia, produced by hepatic deamination of amino acids, is rapidly incorporated into urea, with a net production of H⁺. However, as the systemic [H⁺] increases, there is some shift from urea to glutamine (GluCONH₂) synthesis, with a slight fall in hepatic H⁺ production.

Glutamine is taken up by renal tubular cells, where it is hydrolysed by glutaminase to glutamate (GluCOO⁻) and NH₄+:

Ammonia and NH₄ + form a buffer pair with a pK_a of about 9.8:

Because of the very high pK_a, at pH 7.4 and below the equilibrium is overwhelmingly in favour of NH₄⁺ production. Ammonia can diffuse out of the cell into the tubular lumen much more rapidly than NH₄⁺. If the luminal fluid is acidic, NH₃ will be retained within the lumen by avid combination with H⁺ derived from the CD mechanism. This allows H⁺, produced in the kidneys, to be excreted as ammonium chloride (NH₄ +Cl⁻); thus, in severe acidosis, HCO₃⁻ formation can continue even when phosphate buffering power has been exhausted. There is a net gain of HCO₃⁻. However, H⁺ as well as NH₃ is released in renal tubular cells when NH₄⁺ dissociates; this is maintained by passive diffusion of NH₃ into the luminal fluid (Fig. 4.4).

On the face of it there seems no advantage in buffering one H⁺ secreted into the lumen if, at the same time, another is produced within the cell.

Explanations for the role of NH₃ in the correction of acidosis include the following:

The rate of gluconeogenesis and glutamine synthesis and glutaminase activity all increase in an acidosis.

The parietal cells of the stomach secrete H⁺ into the lumen together with Cl⁻. As H⁺Cl⁻ enters the gastric lumen, HCO₃⁻ diffuses into the ECF, thus accounting for the post-prandial ‘alkaline tide’. In the healthy subject this is rapidly corrected by HCO₃⁻ secretion, mainly by the pancreas, as food passes down the intestinal tract. This mechanism explains the metabolic alkalosis that occurs in pyloric stenosis.

Sodium bicarbonate secretion by the pancreatic and biliary cells in response to stimulation by secretin accounts for the alkalinization of the duodenal fluid and occurs by the reverse process of NaHCO₃ reabsorption in renal tubular cells. The pancreatic and biliary mechanisms are accelerated by the local rise in PCO₂ that results when H⁺ is pumped into the ECF and reacts with the HCO₃⁻ generated by gastric parietal cells. This is analogous to the stimulation of renal HCO₃⁻ formation by the rise in luminal PCO₂. Loss of large amounts of duodenal fluid may cause HCO₃⁻ depletion.

As fluid passes down the intestinal tract, HCO₃⁻ enters and Cl⁻ leaves the lumen by a reversal of the gastric mucosal mechanism. Therefore the gastric loss of Cl⁻ and the gain of HCO₃⁻ are finally corrected. Preferential reabsorption of urinary Cl⁻ by this mechanism after ureteric transplantation into the ileum, ileal loops or the colon explains the hyperchloraemic acidosis (normal anion-gap acidosis) associated with this operation (Fig. 4.5).

The primary disorder in the bicarbonate buffer system in a metabolic acidosis is a reduction in [HCO₃⁻], resulting in a fall in blood pH. The reduction in the HCO₃⁻ may be due to:

The causes of a metabolic acidosis are shown in Box 4.1 and can be divided into those with a high and those with a normal plasma anion gap.