Reappraisal of bicarbonate secretion by the human oesophagus

Ussing Chamber Methods to Study the Esophageal Epithelial Barrier

The Ussing chamber was developed in 1949 by Hans Ussing and quickly became a powerful tool to study ion and solute transport in epithelia. The chamber has two compartments strictly separating the apical and basolateral sides of the tissue under study. The two sides of the tissue are connected via electrodes to a modified electrometer/pulse generator that allows measurement of electrical parameters, namely, transepithelial voltage, current, and resistance. Simultaneously, permeability of the tissue to specific solutes or markers can be monitored by using tracers or isotopes to measure transport from one side of the tissue to the other. In this chapter, we will describe the use of the Ussing chamber to study the barrier properties of the mouse esophageal epithelium. We will also briefly describe the use of the modified Ussing chamber to simultaneously study transepithelial and cellular electrophysiology in the rabbit esophageal epithelium. Lastly, we will cover the use of the Ussing chamber to study bicarbonate secretion in the pig esophagus. These examples highlight the versatility of the Ussing chamber technique in investigating the physiology and pathophysiology of epithelia including human biopsies.

Oesophageal sensation in response to high PCO(2) and acidic solutions in nonerosive reflux disease

BACKGROUND

Heartburn is commonly associated with the presence of acid in the oesophageal lumen. However, in patients with nonerosive reflux disease (NERD), the mechanism by which acid traverses the mucosa is not clear. We hypothesized that the luminal acid signal traverses the oesophageal epithelium in the form of the highly permeant gas CO(2) , which then is reconverted to H(+) in the submucosa.

MATERIALS AND METHODS

  Ten patients with heartburn, normal upper endoscopy and increased oesophageal acid exposure (NERD patients) and 10 healthy subjects were enrolled. Perceptual responses to intraoesophageal acid (0·1 N HCl solution) and a high PCO(2) solution were determined using a randomized cross over design. Stimulus-response functions to perfusions were quantified by three parameters: lag time to symptom perception, intensity rating and perfusion sensitivity score.

RESULTS

 In NERD patients, the difference in lag time to typical symptom perception, intensity rating and perfusion sensitivity score between high PCO(2) and acid perfusions was statistically significant (P = 0·02, 0·01 and 0·02, respectively). However, the difference in the same perfusion parameters between acid and high PCO(2) perfusions was nonsignificant in healthy controls. When NERD and controls were compared, the difference between the different perfusion variables was nonsignificant (adjusted to age).

CONCLUSIONS

In NERD subjects, acid perfusion reliably evoked heartburn symptoms of greater intensity than in healthy controls. Nevertheless, a high PCO(2) perfusion failed to produce symptoms in either group.

Effect of various test meals on gastric and jejunal carbon dioxide: A study in healthy subjects

OBJECTIVE

The normal pattern of carbon dioxide (CO2) levels in the human stomach and small bowel after meals is unknown. The intraluminal carbon dioxide level is a sensitive and early marker for organ mucosal ischemia. CO2 levels in both the stomach and small bowel are influenced by multiple factors other than adequacy of perfusion. Gastric acid production, salivary bicarbonate and CO2 produced or absorbed by meals are the disturbing variables. Prolonged gastric (and jejunal) tonometry after meals can be of additional value in the work-up of patients suspected of (chronic) gastrointestinal ischemia. The purpose of this study was to challenge these problems using in vitro tested meals and a rigid acid-suppression regimen in a group of healthy subjects.

MATERIAL AND METHODS

Standard meals were tested in vitro on the ability to produce and buffer CO2. Meals with the least CO2 variations were subsequently used in healthy subjects. Tonometry of the stomach and jejunum was performed for 24 h, with optimal and controlled acid suppression.

RESULTS

Ten subjects were enrolled in the study. Acid production was sufficiently suppressed. The gastric PCO2 baseline (fasting) was 6.5 (1.0), and significantly lower than the jejunum PCO2 baseline of 7.6 (0.9) kPa. The gastric baseline during the day was 6.9 (1.6), and significantly lower than the gastric baseline during the night of 8.0 (1.8), suggesting a diurnal variation of PCO2. Increases in PCO2 levels were seen in all subjects, after meals and between meals.

CONCLUSIONS

Prolonged gastric and jejunal tonometry is feasible in humans. PCO2 levels were seen to peak after, but also in-between, most meals. The diurnal variation in PCO2 might reflect reversible gastric mucosal ischemia.

HCO3- secretion in the esophageal submucosal glands

The mammalian esophagus has the capacity to secrete a HCO(3)(-) and mucin-rich fluid in the esophageal lumen. These secretions originate from the submucosal glands (SMG) and can contribute to esophageal protection against refluxed gastric acid. The cellular mechanisms by which glandular cells achieve these secretions are largely unknown. To study this phenomenon, we used the pH-stat technique to measure luminal alkali secretion in an isolated, perfused pig esophagus preparation. Immunohistochemistry was used to localize receptors and transporters involved in HCO(3)(-) transport. The SMG-bearing esophagus was found to have significant basal alkali secretion, predominantly HCO(3)(-), which averaged 0.21 +/- 0.04 microeq.h(-1).cm(-2). This basal secretion was doubled when stimulated by carbachol but abolished by HCO(3)(-) or Cl(-) removal. Basal- and carbachol-stimulated secretions were also blocked by serosal application of atropine, pirenzipine, DIDS, methazolamide, and ethoxzolamide. The membrane-impermeable carbonic anhydrase inhibitor benzolamide, applied to the serosal bath, partially inhibited basal HCO(3)(-) secretion and blocked the stimulation by carbachol. Immunohistochemistry using antibodies to M(1) cholinergic receptor or carbonic anhydrase-II enzyme showed intense labeling of duct cells and serous demilunes but no labeling of mucous cells. Labeling with an antibody to Na(+)-(HCO(3)(-))(n) (rat kidney NBC) was positive in ducts and serous cells, whereas labeling for Cl(-)/HCO(3)(-) exchanger (AE2) was positive in duct cells but less pronounced in serous cells. These data indicate that duct cells and serous demilunes of SMG play a role in HCO(3)(-) secretion, a process that involves M(1) cholinergic receptor stimulation. HCO(3)(-) transport in these cells is dependent on cytosolic and serosal membrane-bound carbonic anhydrase. HCO(3)(-) secretion is also dependent on serosal Cl(-) and is mediated by DIDS-sensitive transporters, possibly NBC and AE2.

Mechanisms of gastroesophageal reflux and gastroesophageal reflux disease

Gastroesophageal reflux is a physiological phenomenon, occurring with different severity and duration in different individuals. Reflux occurs when this normal event results in the occurrence of symptoms/signs or complications. The pathophysiology of gastroesophageal reflux is complex and diverse, since it is influenced by factors that are genetic, environmental (e.g., diet smoking), anatomic, hormonal, and neurogenic. However, many mechanisms remain incompletely understood. Future research should focus on a better understanding of the physiology of the upper and lower esophageal sphincters, and of gastric motility. The afferent and efferent neural pathways and neuropharmacologic mediators of transient lower esophageal sphincter relaxations and gastric dysmotility require further study. The role of anatomic malformations such as hiatal hernia in children has been underestimated. While therapeutic possibilities are greater in number and largely improved, the outcomes of some treatments are far from satisfactory in many cases. In addition to development of new forms of treatment, research should address better use of currently available medical and surgical treatments.

Esophageal mucin: an adherent mucus gel barrier is absent in the normal esophagus but present in columnar-lined Barrett's esophagus

OBJECTIVES

The presence of a protective adherent mucus gel barrier against gastric reflux in the healthy esophagus is uncertain. The aim was to characterize the surface mucin composition and determine the extent of any adherent mucus gel layer on the normal esophagus, and compare this with that in Barrett's esophagus.

METHODS

Isolated surface mucins were characterized by density centrifugation, gel filtration chromatography, and chemical composition. Adherent surface mucus was visualized in situ on unfixed and cryostat sections of mucosa and biopsies using a method that preserves mucus layer thickness.

RESULTS

There was a complete absence of adherent mucus gel layers on normal human, pig, and rat esophagi. This was in contrast to the thick adherent mucous layer (median thickness = 100-200 microm) seen on the corresponding gastric mucosa. Small quantities of glycoprotein with a composition characteristic of a secretory mucin were isolated from the pig esophagus surface. The mucin, density range between 1.44 and 1.48 g x ml(-1), contained 80% carbohydrate and was rich in serine, threonine, and proline. The mucin fragmented into smaller glycoprotein units on proteolysis and partially on reduction. Cryostat sections from columnar-lined esophageal biopsies had a substantial adherent surface mucous layer (median thickness = 90 microm, interquartile range = 84-94 microm) staining for neutral mucins (gastric-type epithelium) and acidic mucins (intestinal metaplasia).

CONCLUSIONS

A secretory mucin, with an analysis distinct from that of gastric or salivary mucin, is present in very small quantities on the esophageal mucosa and in amounts insufficient to form an adherent gel layer. It is unlikely that mucus has a role in protecting the normal esophagus against reflux. However, an adherent mucous layer was observed over columnar-lined esophagus, and this may protect against reflux.

Response of the lungs to aspiration

Aspiration of acid from the stomach and water from the mouth can cause significant lung injury. Animal experiments suggest that acid entering the lungs is normally neutralized by bicarbonate derived from the plasma. It is hypothesized that this process may be impaired in patients with cystic fibrosis and that some of the airway injury that they experience may be related to this defect. This disease is characterized by abnormalities in the cystic fibrosis transmembrane conductance regulator, which normally conducts bicarbonate and chloride exchange. Evidence is discussed regarding the role of water channels (aquaporins) in transporting water from the airspaces into the vasculature.