Gut, 1967, 8, 548

A study of the effect of saliva on the concentrationof mucin in gastric juice and its possible relationship

to the aetiology of peptic ulcerS. L. MALHOTRA

From the Medical Department, Western Railway, Bombay, India

EDITORIAL COMMENT This interesting paper suggests that the consistency of food affects thequality of gastric mucus derived from swallowed saliva, foods needing much mastication stimu-lating a high salivary output. Salivary mucus, by augmenting gastric mucus, may help to protectagainst peptic ulceration.

The two most important protective barriers againstpeptic ulceration are the surface epithelial cells them-selves with their astounding capacity for regenerationand the film of mucus which covers them. Theremarkable regenerative capacity of the epithelialcells has been studied by various workers withidentical conclusions (Grant, 1945; Palmer, 1952;Grant, Grossman, and Ivy, 1953), from which it isevident that erosion and repair are repeatedly takingplace as a result of the repeated trauma of digestion,but the process of regeneration is carried on withsuch facility that this trauma does not have theinflammatory results which it might have elsewhere.Much work has also been done which shows that

the effects of mechanical irritation of the gastricmucosa as well as of eating are similar in that theseresult in an increased outpouring of mucus from theepithelium (Wolf and Wolff, 1948).

It was shown in an earlier report that differencesin the pattern of diet and eating produced variationsin the total quantities, the buffer capacity, and themucin content of saliva (Malhotra, 1965; Malhotra,Saigal, and Mody, 1965), and from these studies itwas obvious that, apart from influencing gastricacidity (Malhotra, 1967a), swallowing of saliva willalso influence, to a variable extent, the total amount ofmucus present in the gastric lumen, because not onlythe total quantities of saliva swallowed with food,but also the concentration of mucus in the saliva,vary with the pattern of diet and eating (Brown,Wright, and Limbacher, 1938; Dewar, 1952; Dewarand Parfitt, 1954; Malhotra, 1965; Malhotra et al.,1965). This paper attempts to study by a quantitativemethod the effect of saliva on mucus in the gastricjuice.

548

METHOD AND THE DESIGN OF EXPERIMENTS

In designing the experiments an attempt has been madeto 'exclude' saliva from the stomach in one series ofobservations and allow its admixture with food, to theextent that it is normal during eating, in the other. Thiswas accomplished by giving the meal by mouth in one setof experiments and through an intranasal Ryle's tube,positioned in the fundus of the stomach fluoroscopically,in the other. In the experiments in which it was desiredto 'exclude' saliva the subjects were instructed not toswallow saliva but to eliminate it by frequent spitting,whilst in the experiments designed to study the effect ofsaliva, they swallowed it with the food, as it is normallydone.

THE MEALS Both solid and fluid meals were used. Adescription of the solid 'sloppy' and 'masticatory' mealshas been given elsewhere (Malhotra et al., 1965). Theliquid meal was the South Indian appetiser called rasam.The pH of the various batches of rasam ranged between4-3 and 5-2, the majority being ofpH 4-6. To eliminateany possible bias due to pH variations, the same pool ofthe liquid meal was used for 'oral' and the 'tube'experiment.

THE SUBJECTS Thirty-two adult males, who includedambulant patients from the wards of the Jagjivan RamHospital, free from gastric or small bowel disease, andhealthy volunteers, picked randomly from among theclass IV staff of the Hospital, and ranging in age from 17to 53 years, were included in experiments by randomallocation into four groups as follows:

Group A This comprised 13 persons: to seven,boiling hot rasam was given sip by sip in small sips orally,to encourage admixture with saliva in four runs (a to d)of 100 ml. each and subsequently at room temperaturethrough the Ryle's tube in four runs (e to h) so that saliva

group.bmj.com on June 23, 2017 - Published by http://gut.bmj.com/Downloaded from

Effect of saliva on the concentration of mucin

was excluded; in another six the procedure was reversedto eliminate the effect if any, due to changes in stimulusintensity, viz., the first four runs were given through thetube and the last four runs orally.Group B This comprised seven persons: to three,

rasam at room temperature was given through the tube(a to d) and then orally as a 'long drink' (e to h) as it isnormally taken in South India; in the remaining four theorder was reversed.Group C This comprised six persons in whom the

masticatory' solid meal was used: in three orally on thefirst day and on the next day through the Ryle's tubemade into a paste. In the other three the order wasreversed.Group D Of six persons, the South Indian style sloppy

meal of boiled rice and lentil soup was used in the samemanner as the Punjabi masticatory food.

GASTRIC JUICE SAMPLES To obtain samples, the methodof Brooks, Erskine, Gephart, Swaim, and Moore (1950)was used for groups A and B.

In this, the stomach is completely emptied of its fastingcontents and 100 ml. of the fluid meal marked with a dye(to determine the secretory and emptying rates), asdescribed elsewhere (Malhotra, 1967b), is given orally orthrough the tube as the case may be. After allowing themeal to 'digest' for 10 minutes, the residue was recovered.Eight such successive runs were carried out and labelledalphabetically a to h and all specimens analysed. Speci-mens a and b and e and f have not been taken intoaccount, as basal conditions do not prevail until after thefirst two runs (Brooks et al., 1950) and, moreover, thesehelp to wash out any residual mucin left over from thefasting contents.

In the case of the solid meals, because of the difficultyof complete aspiration, the experiments were done on twodifferent days. Two equal and identical portions of themeals were used, one of which was eaten in the normalmanner on one day so that saliva mixed with the meal andthe other was made into a thin paste and was introducedthrough the tube, on the succeeding day, so that thesaliva was excluded from the meal. Two hundred ml. ofnormal saline was allowed for each portion, in one caseto make the paste from the meal, and in the other thesubject swallowed this by mouth, at the end of the meal.With this mode of investigation, each patient suppliedeight samples (excluding the fasting contents): four withsaliva and four with no saliva. When working with'sloppy' meals (group D) we overcame the difficulty ofresidual salivary mucus by preceding the experimentswith two saline runs of 100 ml., each which served as awashout. Unfortunately, due to an initial error inplanning this step was not taken in the case of the solid'masticatory' meals (group C).

ESTIMATION OF MUCIN

The Congo red and acid-alcohol (A.C.R.A.) method ofBurnet (1948), which is based on the fact that if a mucussolution is dropped into acid-alcohol a firm pellicle formsat the interface, maintaining the drop either in globular

or with weaker solution in annular form, has been usedpreviously for the quantitation of salivary mucus (Dewarand Parfitt, 1954; Malhotra et al., 1965). The shapes canbe rendered easily visible by adding an alcohol-insolubledye to the mucus solution. In the absence of mucus thedrop flattens out immediately and leaves only a cloudypatch without a definite margin. The method is veryquick and results are reproducible within the requiredorder of accuracy (Burnet, 1948).For the present study, the A.C.R.A. method of mucus

estimation has been adapted to gastric juice. If a solutioncontaining mucus and saliva is allowed to stand at roomtemperature, 30% ofmucus is destroyed within 30 minutesand 50% in two hours. But if the sample is placed on iceimmediately after collection, destruction is only slight(Dewar and Parfitt, 1954; Malhotra et al., 1965). In orderto obtain comparable results we had, therefore, collectedour samples in stoppered glass cylinders surrounded withice. Sampling was done in a room adjoining the workbench to facilitate immediate assay. On the completionof full gastric aspiration, the glass cylinderswere vigorous-ly shaken to homogenize the mucus and 0 5 ml. from thiswas immediately (within two minutes) transferred to atest tube for serial two-fold dilutions with saline bufferedwith borate at pH 7 1. To each tube 2 drops of 0-1 %Congo red solution in water were added. In cases wherethe first tube was highly acidic and the indicator turnedblue, one drop of 1 N NaOH was added. This step notonly clarified the end point but also helped to homogenizethe mucus by dissolving it. While making the serial dilu-tions homogenization and thorough mixing was securedin the tubes by rapid up-and-down movement of thesolution in a teat pipette with a wide opening.To record the results the original procedure of Burnet

(1948) was used, in which one drop from each tube isdropped gently into a shallow layer of acid-alcohol (about4 mm.) in a flat petri dish with transparent bottom. Theacid-alcohol is a 1% solution of strong HCI in 70%alcohol. The results are recorded as A.C.R.A. titres,employing identical end-points as in Burnet's originaltechnique of mucin assay. The great advantage of theA.C.R.A. method is the simplicity with which it can beperformed to yield reproducible results, and its sensitivity,as it can measure even very minute amounts of mucusin solution.The tests were done immediately on collection, using

a 'blind' technique in which the observer noting the endpoints was not aware of the origin of the samples. Serialdilutions of every batch of the fluidmeals, and, in the caseof the solid meals, similar dilutions of 20% aqueousextract of the meal paste were tested for the presence ofmucus. None of these contained mucus.

RESULTS

The results of the mucin content of the samples ofgastric juice have been expressed as A.C.R.A. titres.For the purpose of the analysis of the results, thetitres have been grouped in ranges as shown inTable I.

549

group.bmj.com on June 23, 2017 - Published by http://gut.bmj.com/Downloaded from

TABLE IDISTRIBUTION OF SAMPLES OF GASTRIC CONTENTS IN DIFFERENT A.C.R.A. TITRE RANGES OF SALIVARY MUCUS

S>udy Sample A.C.R.A. Titres' Total No.of Samples

No. of Samples No. of Samples ShowingPositive A.C.R.A. Titres in Different Tube RangesShowing NegativeA.C.R.A. Titrein First Tube

A

B

C

D

With salivaWith no salivaWith salivaWith no salivaWith salivaWith no salivaWith salivaWith no saliva

001827091315

1-3968610733

4-6603

s3554

7-962

08222

10-1240003

0

1300000000

2626141424242424

Study A: fluid mealComparison in 13 individuals, between sip by sip ingestion versus introduction of meal through Ryle's tube.Study B: fluid mealComparison in seven individuals, between ingestion as a 'long drink' versus introduction of meal through Ryle's tube.Study C: solid 'masticatory' mealComparison in six individuals, between ingestion by thorough mastication versus introduction of meal as a paste through Ryle's tube.Study D: solid 'sloppy' mealComparison of six individuals, between ingestion by bolting down versus introduction of meal as a paste through Ryle's tube.In studies A and B with fluid meals each person supplied four samples, two with saliva and two with no saliva.In studies C and D with solid meals each person supplied eight samples, four with saliva and four with no saliva.'The figures indicate tube numbers for dilution, e.g., 1st = 1: 1; 2nd = 1: 2; 3rd = 1: 4; 4th = 1: 8 and so on.

COMPARISON OF A.C.R.A. TITRES OF GASTRIC CONTENTSAFTER FLUID MEALS AND MEALS THROUGH RYLE'S TUBE

Mucin in high titres was invariably present in thefasting samples ranging from a titre of 1 :1 to 1:256(first to ninth tube).Mucin was completely absent in 18 samples out of

26 (69%) when saliva was 'excluded' from the meals,in contrast to nil out of 26 samples with saliva. Therewere more samples with high A.C.R.A. titres goingup to the thirteenth tube (1:4,096) when saliva wasadmixed with the meal, as compared with when it wasexcluded.

COMPARISON OF A.C.R.A. TITRES OF GASTRIC CONTENTS

AFTER FLUID MEALS AS A LONG DRINK VERSUS THROUGH

THE RYLE S TUBE All fasting samples showed a highconcentration of A.C.R.A. titre ranging from 1: 2 to1: 256 (second to ninth tube).Mucin was completely absent in seven out of 14

samples (50%.), when the meals were given throughthe Ryle's tube thus 'excluding' saliva. The majorityof the samples with saliva (90%) were distributedbetween the lower titres of 1 :1 to 1: 8 (first to fourthtube). When these results are compared with theresults of sample obtained in study A, in which thefluid meals were taken sip by sip, to encourage ad-mixture with saliva, it was seen that the concentra-tion of mucin in the aspirates was much higher withthe latter mode, namely, study A, as compared withthe former, namely, study B (Table I).

COMPARISON OF A.C.R.A. TITRES OF GASTRIC CONTENTS

AFTER MASTICATORY SOLID MEALS BY MOUTH VERSUS

A PASTE INJECTED THROUGH RYLE S TUBE With solidmasticatory meals mucin was present in all sampleswhether taken by mouth or through the tube but theA.C.R.A. titres were higher in samples obtained duringexperiments when saliva was swallowed than whenit was 'excluded'. We had made an initial mistake inour planning that with the solid 'masticatory' mealswe did not wash out the stomach with saline to bringabout 'basal' conditions and, therefore, some mucinmay have been left in the stomach from the residualfasting juice despite our attempts to empty it outwith the fasting contents. One possibility, therefore,arises that because of incomplete removal of mucusfrom the fasting contents, the presence of mucin insamples obtained when saliva was 'excluded' mayreally have been due to the presence, in the gastriclumen, of residual salivary mucus ordinarily presentin the fasting contents, due to the swallowing ofsaliva during sleep. Many particles of mucus adhereto the gastric wall and are not recovered through thetube. This is the general opinion of gastroscopistsand it has been observed directly through a gastricfistula by Wolf and Wolff (1948). This possibility isconfirmed from our experiments with 'sloppy' mealsin which a saline washout preceded the collection ofsamples and also when we repeated the procedureswith masticatory meals on two other patients, notincluded in this series, with a saline washout toprecede the experiments.

COMPARISON OF A.C.R.A. TITRES OF GASTRIC CONTENTSWITH SLOPPY SOLID MEALS BY MOUTH VERSUS AS APASTE INJECTED THROUGH RYLE S TUBE The samples

550 S. L. Malhotra

group.bmj.com on June 23, 2017 - Published by http://gut.bmj.com/Downloaded from

Effect of saliva on the concentration of mucin

with saliva and with no saliva were equally distri-buted between different A.C.R.A. ranges. Therewere more samples showing a complete absence ofmucus in this mode of investigation in contrast tothe masticatory meals (study C). This can be ex-plained by the fact that larger amounts of saliva areswallowed with masticatory meals as compared withsloppy food (Dewar and Parfitt, 1954; Malhotra etal., 1965).

ADEQUACY OF THE METHOD The A.C.R.A. methodof quantitating mucus has been previously used forthe salivary mucus and has been found to givereproducible results within the required order ofaccuracy (Dewar, 1952; Dewar and Parfitt, 1954;Knox and Still, 1953; Malhotra et al., 1965). It hasnot been previously used for quantitating the gastricmucus, and the question whether this method deter-mines only the salivary mucus and ignores the gastricmucus must therefore be carefully considered. Toinvestigate this possibility, we introduced a test mealof 100 ml. of 5% NaHCO3 in distilled water into thestomach through the Ryle's tube on the completionof the main experiment in six of our patients who hadshown no mucus in samples obtained after the tubefeeding. Ten minutes were allowed for this meal to'digest'. Although the A.C.R.A. titre in samplespreceding the NaHCO3 instillation were negative inall these patients, without exception, high A.C.R.A.titres were found after the alkali meal (in one case upto the twelfth tube). Since saliva had been carefullyexcluded from these samples, this clearly shows thatthe high A.C.R.A. titres in these samples must bedue to mucus of gastric origin, because with a fall inacidity the intracellular mucus from the gastric epi-thelium is removed and goes into solution. Ourresults cannot, therefore, be due to any inadequacyof the method to measure the concentration ofgastric mucus, and obviously the A.C.R.A. methodof quantitating mucus is applicable to both thesalivary as well as gastric components of mucus.

Is it possible that if we have not succeeded inaspirating the gastric mucus through the Ryle's tubeit is because mucus precipitates into thick strands oncoming into contact with acid and may have thusoffered resistance to aspiration? This appears un-likely as it should have equally affected the salivarymucus swallowed with the meals, which, on the otherhand, we were able to aspirate without difficulty.Moreover, in the studies of James and Pickering(1949), as in ours, using aspiration after consumptionof solid food, it was sometimes surprising how thickthe material was which came up the Ryle's tube, thesample occasionally being of a consistency such thatit could hardly be poured.A final source of error can be introduced in the

method if the mixing of the mucus in the gastric juiceby thorough shaking of the cylinder is inadequate.Glass and Boyd (1948) have shown that if gastricjuice containing saliva is centrifuged part of thesalivary mucus collects on the bottom and at the topof the centrifuge tube. We employed vigorousshaking in a glass stoppered jar to secure homo-genization of the mucus, so producing homogeneousdispersion of the salivary and gastric mucus com-ponents in the fluid. This is shown by the followingtest:

Test for mixing In six of our samples (fromdifferent individuals), after the assay of the mucus inthe usual manner, we added 1 N NaOH to bring thepH to 9 and repeated the assay. The samples withoutsaliva showed no mucus by both the methods. Thesamples with saliva showed similar results (within ±1 dilution) with both the methods. Since in an alka-line medium mucus goes into solution, it is clear thatour method of vigorous shaking had secured nearlycomplete mixing and homogenization of both themucus components, as otherwise the results fromthis test could not have been comparable with theresults of the main experiment.The inference of these results, therefore, is that

(1) larger amounts of mucus are present in thestomach when saliva is added as compared to whenit is not; (2) mucus in gastric aspirates obtained afterserial test meals is chiefly of salivary origin andhardly any from the gastric mucous membrane,because in the samples with no saliva mucus wasgenerally absent or in very low concentrations.

EFFECT OF THE CEPHALIC PHASE OF GASTRIC SECRETIONON GASTRIC MUCUS A legitimate criticism of ourmethod in which we gave the meal by mouth in oneset of experiments and by intubation in the other, isthat the differences in mucus concentration noted byus between the samples with saliva versus the sampleswith no saliva may really be due to the possibilitythat important secretory events associated with thecephalic phase of gastric secretion may be lacking inthesituation inwhichfoodwas passed directlythroughthe tube into the stomach. This has been tested inthe following manner:

Seventeen convalescent patients admitted to theJagjivan Ram Hospital for diseases unrelated to thegastrointestinal tract were included in the test. Froma common pool of the fluid meal, for each patientwe secured admixture of saliva with the meal in fouraliquots by asking the patients to sip the meal, keepeach sip in the mouth in order to encourage admix-ture with saliva, and then collect it in a markedreceptacle. To pair with these aliquots, equalvolumes of the meal from the same pools weretaken. Thus we had four pairs of test meals for each

551

group.bmj.com on June 23, 2017 - Published by http://gut.bmj.com/Downloaded from

S. L. Malhotra

patient, namely, four test meals with saliva and fourtest meals with no saliva, which were in this testadministered by intubation after the evacuation offasting contents and using two runs each of 100 ml.of distilled water as a stomach wash. The experi-ments in this test too were done in a room adjoiningthe work bench and the patients were instructed notto swallow saliva but to eliminate it by spitting, thisbeing carefully supervised. Thus the possibility ofimportant secretory events associated with thecephalic phase of gastric secretion and a slower rateof gastric filling affecting the results was eliminatedin this test, since all the test meals were given byintubation. On the completion of the runs of testmeals, we intubated 100 ml. of 5% NaHCO3,allowed it to 'digest' for 10 minutes, and completelywithdrew the contents for similar analysis.The results of the test in Table II show that

in 14 out of 17 patients the samples with no saliva

TABLE II

RESULTS OF TEST TO SEPARATE EFFECT OF REPEATED

SWALLOWING, TEMPERATURE OF THE MEAL, AND THECEPHALIC PHASE OF GASTRIC SECRETION

Patient AgeNo.

2345

67891011121314151617

4125303135182038215027314823255134

A.C.R.A. Titre;

Sample with Saliva Sample with noSaliva

F a b c d e f g h A

6283345

75525

4

55

3

4

3

55

232

5

26

3

4

9

2

48174313

214220

5

31340

213

75

45

70

5

4

34

133

2

7127

8185

3

0

0

0

0

0

2

0

0

0

2

0

0

2

0

0

0

0

0

0

9

0

0

0

0

0

8

0

0

3

0

0

0

0

0

0

30

0

0

0

0

0

90

0

20

0

0

0

0

0

0

0

0

0

0

8

0

0

0

0

Comparison of the A.C.R.A. titres between samples with saliva andsamples with no saliva in 17 individuals, the test meals being ad-ministered by intubation. The figures represent the tube numbers.F = fasting, a to h are the different runs of test meals: in patients

1 to 9 the runs with saliva preceded the runs with no saliva whereas inpatients 10 to 17 this was reversed. A = the run in which the testmeal was 100 ml. of 5% NaHCO,. The figures of the A.C.R.A. titrestand for the tube numbers, showing the end point.

were negative for mucus whereas the samples withsaliva showed the presence of mucus in varying con-centrations. This shows that the differences we notedare not due to the cephalic phase of gastric secretionor to its effect on the secretion ofmucus by the gastricmucus membrane. The fasting contents as well as thesamples in which the test meal was 5% NaHCO3

were invariably positive. Cases 6, 13, and 16 showedhigh A.C.R.A. titres in the samples with saliva aswell as in the samples with no saliva. All of thesepatients had a marked duodenal reflux with heavybile contamination of the samples, which may ex-plain the high A.C.R.A. titres in their samples withno saliva. Case 13 gave a history of chronic gastritisand a healed gastric ulcer.

DISCUSSION

The results of our studies with fluid as well as solidmeals show unequivocally that mucus was eithercompletely absent or present in extremely low con-centrations in samples from which saliva had been'excluded'. In contrast to this, mucus in varying con-centration was invariably present in samples with-drawn after eating meals with which saliva wasallowed to mix as well as in the fasting samples.Since the various foods employed as test meals hada negative A.C.R.A. titre, these results are inde-pendent of the type of food used and are obviouslydue to whether the sample of gastric contents con-tained saliva or not. Two different interpretations ofthese results are possible:

(1) There is a relative absence of mucus in thegastric juice unless saliva is added to the food, whichsuggests that the healthy stomach does not ordinarilysecrete mucus in the gastric juice, in response tonormal food. (2) Larger amounts of mucus arepresent in the gastric juice when saliva is added to thefood as compared to when it is not. This suggeststhat most of the mucus in gastric juice is of salivaryorigin.

In the case of the samples of fasting contents, theA.C.R.A. titres were invariably high (fifth to tenthtube). This can be explained because a lot of salivais normally swallowed during sleep. However, it isimpossible to be certain that the fasting juice has nogastric component of mucus.

TOTAL AMOUNT OF MUCUS A comparison of theA.C.R.A. titres between samples with saliva indifferent studies shows that the concentration ofmucus was higher in study A than in study B andhigher in study C than in study D. Since the averagevolumes of gastric juice aspirated were larger in theseries with saliva as compared with the series with

k no saliva (P < 0-01), the mean values being87-0 ml. ± 8-21 and 51-5 ml. ± 8-27 in the sampleswith saliva and the samples with no saliva respective-ly in study A (Table III), this indicates that ourresults were not due to dilution, and not only theconcentration but also the total amount of mucus ingastric aspirates was higher in the samples with

552

group.bmj.com on June 23, 2017 - Published by http://gut.bmj.com/Downloaded from

Effect of saliva on the concentration of mucin

TABLE IIIMEAN VALUES FOR VOLUME OF GASTRIC CONTENTS RE-COVERED FROM 13 INDIVIDUALS IN STUDY A (FLUID MEALSIP BY SIP AND BY INTUBATION) AND FROM SEVEN INDI-VIDUALS IN STUDY B (FLUID MEAL GULPED VERSUS

INTUBATION)Experiment Volume of Gastric Contents (ml.) Probability

With Saliva With No Saliva

Study AMeanS.E. ±

Study BMeanS.E. ±

87-08-21

67-577-17

51-58-27

55.575.37

saliva than the samples with no saliva, and thisclosely correlates with the amount of salivaswallowed with food and with the concentration ofmucus in the saliva, because in study A where fluidmeals were taken sip by sip in small sips, the amountof saliva swallowed would naturally be larger ascompared with study B where the meal was taken asa 'long drink' or in one 'gulp', a mode of ingestionwhich does not encourage much admixture of salivawith the meal. Similarly larger amounts of saliva ofa higher mucus content are swallowed with mastica-tory meals (study C) as compared with 'sloppy'meals used in study D, because the chewing move-ments of the jaw not only increase the total amountsof saliva secreted but also result in an increasedconcentration of mucus in saliva (Dewar, 1952;Dewar and Parfitt, 1954; Knox and Still, 1953;Malhotra et al., 1965). This can explain why therewas a higher concentration of mucus in samples withsaliva obtained after eating masticatory meals ascompared with sloppy meals. This also is probablyone of the main reasons why Glass and Boyd (1948)found 60% lower values of mucin in an in vitromixture of equal parts of gastric juice and saliva thanin the same uncontaminated gastric juice. Since theyhad used unstimulated saliva, which ordinarily isexpected to contain very little mucus (Dewar, 1952;Malhotra, 1965), and since the gastric juice in theirexperiment had not been obtained after a stomachwash, it was rich in mucus just like our fastingsamples, a situation was created in which the addi-tion of saliva to the sample of gastric juice resulted ina dilution of the concentration of mucus originallypresent in the gastric juice. Their result is thus inconsonance with our study B, but contrary to ourstudies A, C, and D in which the addition of salivato the gastric juice in vivo markedly increased theconcentration of mucin, because the mucus concen-tration greatly increases in stimulated saliva as com-pared with unstimulated saliva. Gastric mucus is,

however, present in the gastric juice under abnormalconditions such as gastritis or duodenal reflux.The absence of mucus from samples in studies in

which the meals were introduced through the Ryle'stube thus 'excluding' saliva is extraordinary in viewof the findings of others that the effects of mechanicalirritation of the gastric mucosa as well as of eatingare similar because these result in an increased out-pouring of mucus from the epithelium (Wolf andWolff, 1948). The explanation ofour results probablylies in the fact that the concept of a layer of gastricmucus smeared over the mucus membrane does notgive a true picture of the relation of gastric mucusto the gastric epithelium where the 'cell itself formsa part of the mucus layer', and, as pointed out byJames (1957), the intracellular gastric mucus 'isadherent to the surface epithelium and continuouswith plugs of mucus filling the foveolae, beingfunctionally an additional layer of the stomachlining' and, therefore, it did not appear in ouraspirates. The fact that in six of our subjects when100 ml. of 5% NaHCO3 was introduced the A.C.R.A.titres rose markedly even though the preceding runsof test meals had shown no mucus supports thispossibility. This is also in agreement with the obser-vations of several workers who have found that manyparticles of mucus adhere to the gastric mucousmembrane and are not recovered through the gastrictube (Glass and Boyd, 1948; Wolf and Wolff, 1948).

In our subjects we had three individuals whosesamples were heavily bile stained. In two of them,who showed a high A.C.R.A. titre (seventh totwelfth tube), the pH of the samples was between7 and 7-4. In the case of the third, despite heavy bilestaining, the pH was acidic and the A.C.R.A. titreswere low (third to fourth tube). This would indicatethat mere regurgitation of bile cannot dissolvemucus, nor does it affect the mucus concentrationdue to the mucus components of bile contaminatingthe gastric juice, but if there is a fall in acidity, theintracellular mucus is removed from the epithelialcells and can be aspirated through the tube. This isalso supported by the observations of Ball and James(1961) and the results of our experiments in whichwe instilled in the stomach 100 ml. of 5% NaHCO3.We also investigated two patients suffering from

chronic gastritis. Both of them showed higherA.R.C.A. titres in samples with no saliva than ournormal subjects, although the titres were not as highas in samples with saliva. It is very difficult to in-terpret these results, but we suppose that, although ina healthy state the stomach may not secrete mucus, itdoes so in the presence of chronic inflammation andin this respect may resemble the respiratory mucosawhich secretes large quantities of mucus in chronicbronchitis and hardly any in the healthy state.

553

group.bmj.com on June 23, 2017 - Published by http://gut.bmj.com/Downloaded from

554 S. L. Malhotra

RELATIONSHIP OF THE SALIVARY MUCUS TO PEPTICULCERATION The findings reported here showunequivocally that as a result of eating food twokinds of mucus may be simultaneously present in thegastric lumen: (1) salivary mucus which may, oncoming into contact with gastric acid, precipitate asa layer of mucus smeared over the surface of themucosa and the food, and (2) intracellular gastricmucus which is an integral part of the epitheliallining but ordinarily cannot be recovered throughthe tube unless it has been removed from the celldue to a fall in acidity. Furthermore, the amount ofthe salivary mucus in the stomach varies a great dealand depends upon the pattern of diet and eating,being more when masticatory food is eaten as com-paredwithsloppymeals not requiring muchchewing.

Since mucus has a protective role against pepticulceration, it is clear that a person who swallowslarger quantities of mucus-rich saliva with his foodwill have a better protection because of the addedprotective barrier of the salivary mucus. Apart fromthis, it has been shown that human gastric as well assalivary mucus contain an acid-inhibitory factorwhich causes a true reduction in acid output, whichis not due to an antacid or buffering action of themucin (Johnston, 1965; Brunschwig, Rasmussen,Camp, and Moe, 1942; Menguy and Smith, 1959).Another possible mode of this acid-inhibitory actionof the added salivary mucus might also be, as ex-plained by James (1957), that mucus is precipitatedby the gastric juice on the surface of the epitheliumand in this state might be less permeable to secre-togogues, which are thus prevented from reachingthe receptors. The results of ourpH determinations,which are being reported elsewhere (Malhotra,1967a), support this hypothesis, because in studies A,C, and D, although the test meals employed for'oral' and 'tube' feeding were identical for eachstudy, the mean pH values and the mean values forHCI were lower in the samples with saliva as com-pared with the samples with no saliva, indicatingthat the addition of saliva had reduced gastricacidity, probably due to its mucus content, in addi-tion to its well-known buffering capacity due to itsbicarbonate content.The pattern of diet and eating exercises a profound

influence on both the acid as well as the mucus,because, in general a masticatory mode of eatingprovides saliva of a higher buffer capacity and alarger mucus content which is competent to neutralizegastric acid on the one hand and reinforce themucosal defence against both acid and bile, on theother, as already discussed (Malhotra, 1965;Malhotra et al., 1965; Malhotra, 1966a and b,1967a). A 'sloppy' diet, on the contrary, has boththese defects.

SUMMARY

A new and simple method of quantitative assay ofgastric mucus based on the original A.C.R.A. tech-nique of Burnet has been adapted to measure gastricmucus.

Results of quantitative assays of mucin in samplesof gastric juice obtained after test meals employingboth fluid and solid meals revealed that free mucus,unless due to ingested saliva, is ordinarily absentfrom the stomach and even the entry of a meal intothe stomach does not result in the outpouring ofmucus from the stomach.

Free mucus in the gastric juice, which is either insolution or in suspension, is of salivary origin andis swallowed with food in varying quantities,depending upon the manner of eating.

Masticatory foods produced higher concentrationof mucus in the gastric juice as compared with solidfood which did not require much mastication.

There was a close correlation between the con-centration of mucin in the gastric juice and theamount of saliva swallowed.The possible protective role of the added salivary

mucus, by augmenting the gastric mucus defencebarrier, is discussed.

REFERENCES

Ball, P. A. J., and James, A. H. (1961). The histological backgroundof gastric ulcer. Lancet, 1, 1365-1367.

Brooks, J. R., Erskine, J. M., Gephart, T., Swaim, 0., and Moore,F. D. (1950). The chloride output rate of the human stomachin healthy subjects and ulcer patients: the effects of vagotomyand acetylcholine. Surg. Gynec. Obstet., 90, 155-170.

Brown, J. B., Wright, H. D., and Limbacher. H. P. (1938). Studieson the chemistry of mixed saliva. III. The composition ofsaliva produced at varying rates of secretion by the sameperson with special reference to calcium and phosphorus.J. dent. Res., 17, 191-196.

Brunschwig, A., Rasmussen, R. A., Camp, E. J., and Moe, R. (1942).Gastric secretory depressant in gastric juice. Surgery, 12,887-891.

Burnet, F. M. (1948). The mucinase of V. cholerae. Aust. J. exp.Biol. med. Sci., 26, 71-80.

Dewar, M. R. (1952). An investigation of the physical, chemical andbacterial properties of saliva, with special reference to mucincontent, and their coirelation with susceptibility to dentalcaries. Ph.D. Thesis, University of London. Quoted by D.Afonsky (1961). Saliva and Its Relation to Oral Health. Univ.Alabama Press.

, and Parfitt, G. J. (1954). Mucin content, physical properties ofsaliva and caries activity. J. dent. Res., 33, 751-756.

Glass, G. B. J., and Boyd, L. J. (1948). Studies on dissolved mucin(mucoprotein) of the gastric juice. I. Preliminary tests and therationale for a new colorimetric quantitative method for thedetermination of dissolved gastric mucin. Rev. Gastroent.,15, 396-414.

Grant, R. (1945). Rate of replacement of the surface epithelial cellsof the gastric mucosa. Anat. Rec., 91, 175-185.

, Grossman, M. L, and Ivy, A. C. (1953). Histological changesin the gastric mucosa during digestion and their relationshipto mucosal growth. Gastroenterology, 25, 218-231.

James, A. H. (1957). The Physiology ofGastric Digestion, pp. 128, 131.Arnold, London.

, and Pickering G. W. (1949). The role of gastric acidity in thepathogenesis of peptic ulcer. Clin. ScJ., 8, 181-210.

group.bmj.com on June 23, 2017 - Published by http://gut.bmj.com/Downloaded from

Effect of saliva on the concentration of mucin 555

Johnston, I. D. A. (1965). Reduction of gastric acid secretion withoutoperation. In Recent Advances in Gastroenterology, edited byJ. Badenoch and B. N. Brooke, pp. 110-111. Churchill,London.

Knox, K. W., and Still, J. L. (1953). Observations on salivary mucoids.J. dent. Res., 32, 379-385.

Malhotra, S. L. (1965). Peptic ulcer: role of saliva in its aetiology.J. Ass. Phycns India, 13, 907-914.

(1966a). Genesis of peptic ulcer (Abst.). J. Ass. Phycns India,14, 96.

(1966b). Pathogenesis of peptic ulcer. Proc. 3rd Wid Congr.Gastroent., Tokyo vol. II, 70-72.

(1967a). Protective action of saliva in peptic ulceration. A studyof its effect on gastric secretion by dye-dilution technique.Scand. J. Gastroent., 2, 95.

(1967b). Effect of saliva in gastric emptying. Amer. J. Physiol.,213, 169-173.

-, Saigal, 0. N., and Mody, G. D. (1965). Role of saliva in theaetiology of peptic ulcer. Brit. med. J., 1, 1220-1222.

Menguy, R., and Smith, W. 0. (1959). Inhibition of gastric secretionin the rat by normal human gastric juice. Proc. Soc. exp. Biol.(N. Y.), 102, 665-667.

Palmer, E. D. (1952). Histology of the normal gastric mucosa: aninvestigation into the state of normalcy of the stomachs ofpersons without upper gastrointestinal complaints. Gastro-enterology, 21, 12-23.

Wolf, S., and Wolff, H. G. (1948). Studies on mucus in the humanstomach, estimation of its protective action against corrosivechemicals applied to the gastric mucosa and attempts atquantitation of gastric mucin by two chemical methods.Gastroenterology, 10, 251-255.

group.bmj.com on June 23, 2017 - Published by http://gut.bmj.com/Downloaded from

peptic ulcer.relationship to the aetiology of gastric juice and its possiblethe concentration of mucin in A study of the effect of saliva on

S L Malhotra

doi: 10.1136/gut.8.6.5481967 8: 548-555 Gut 

http://gut.bmj.com/content/8/6/548.citationUpdated information and services can be found at:

These include:

serviceEmail alerting

online article. article. Sign up in the box at the top right corner of the Receive free email alerts when new articles cite this

CollectionsTopic

collections Articles on similar topics can be found in the following

(484)Ulcer (1689)Stomach and duodenum

Notes

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:

group.bmj.com on June 23, 2017 - Published by http://gut.bmj.com/Downloaded from