Some Aspects of Protein Digestion and Absorption in Health and Disease C

POSTGRAD. MED. J. (1961), 37, 745 Postgrad Med J: first published as 10.1136/pgmj.37.434.745 on 1 December 1961. Downloaded from

SOME ASPECTS OF PROTEIN DIGESTION AND ABSORPTION IN HEALTH AND DISEASE C. W. CRANE, B.Sc., M.B., B.S., A.R.I.C. Senior Research Associate in Medicine and Hon. Lecturer in Medicine, Wellcome Research Wing, Department of Medicine, University of Birmingham and Queen Elizabeth Hospital, Birmingham, 15

THE study of protein digestion and absorption has logous serum protein by mouth in contrast to the occupied the attention and interest of physiologists high values found when normal dietary proteins and biochemists for over I50 years since the were given. In contrast Kekwick (1959) con- classical experiments of Spallanzani and the epoch- cluded that the evidence for the absorption of making discovery by Beaumont of the digestive antibodies by newborn babies from maternal properties of the gastric juice of his patient, colostrum was not very strong. However, Wilson Alexis St. Martin. It has been demonstrated by and Walzer (I935) from their work claimed that nitrogen balance experiments that the overall babies could absorb whole egg white and Gruskay mechanism in man is an efficient one. and Cooke (1955) reported that after bouts of

extremely Protected by copyright. Reifenstein, Albright and Wells (I945) fed human diarrhoea in young children an increase as much as subjects increasing amounts of protein (from 20 to fivefold in the absorption of this protein was 250 g. a day), but found that the daily faecal possible. More recently Parish, Barrett, Coombs, nitrogen excretion remained substantially the same Gunther and Camps (I960) have claimed that the level of between i and 2 g. Evidence regarding serum of babies fed cows' milk may contain anti- the origin of the faecal nitrogen from studies on bodies to the milk proteins. Thus it appears patients undergoing a prolonged fast or receiving certain that if whole protein is absorbed un- diets containing negligible quantities of protein changed then this is confined to special circum- (Lusk, 1928) strongly suggest, however, that a large stances either during the neonatal period of life proportion of the faecal nitrogen is derived from or in association with disease, but, in any case, endogenous sources, so that the efficiency of the amounts must be extremely small. It is not assimilation at levels of intake of protein of 250 g. yet certain, however, whether the conclusions apply per day is thus greater than 95%. in respect to homologous serum plasma protein. If Yet Fisher, writing in 1954, pointed out that proteins are not degraded completely to amino- precise information on a number of important acids prior to absorption and larger fragments are http://pmj.bmj.com/ details in these events was still lacking, among absorbed into the mucosa, evidence for the them the rates of digestion and absorption of pro- presence of large quantities of circulating peptides tein and the form in which the split products are in the blood in the post-absorptive phase after absorbed by the intestinal mucosa. The rate of giving protein by mouth has not yet been demon- breakdown of proteins in vivo has long been known strated. For example, Dent and Schilling (I949) to be much faster than occurs in vitro by the were unable to show any significant concentration successive actions of and en- of in the blood of their gastric pancreatic peptide portal experimental on September 30, 2021 by guest. zymes and ' erepsin ' and Fisher discusses at some dogs after protein feeding. length the evidence for concluding that proteins may not be degraded completely to single amino- Methods for Investigating Protein Digestion acids before absorption. There is some recent and Absorption evidence for believing that dietary proteins are Although nitrogen balance experiments may give absorbed from the gut unchanged. Bangham and useful information regarding the overall loss or gain Terry (I957) and Bangham, Ingram, Roy, Shillam of nitrogen by the body under a variety of con- and Terry (1958) have shown that plasma proteins ditions, such studies provide no measure of the may be absorbed unchanged from the gut in new- rates of digestion and absorption of protein. born rats and calves, while Dent and Schilling Clearly many of the experimental techniques used (1949) were unable to show any rise of amino- in animal work (e.g. Wright and Wynn, 1951;; acids in the portal blood of dogs receiving homo- Gupta, Dakroury and Harper, 1958; Geiger, 746 POSTGRADUATE MEDICAL JOURNAL December 1096 Postgrad Med J: first published as 10.1136/pgmj.37.434.745 on 1 December 1961. Downloaded from Human and Middleton, I958) cannot be applied creatic disorders, while Chinn, Lavik, Babb, to human subjects. The occurrence of patients Buckaloo, Stott and Abbott (I953) fed 311-labelled with intestinal fistula such as described by Kuroda albumin with a meal of gelatin to normal subjects. and Gimbel (I954) and suitable for absorption Freeark, Kozoel and Meyer (1957) and Polachek, studies is extremely rare, while results of experi- Cope, Williard and Barnes (1959) have made ments with patients having portal collaterals on the extensive studies on normal subjects and patients abdominal wall and which can be sampled at with pancreatic lesions using labelled albumin intervals after giving protein by mouth were found along with unlabelled gelatin. More recently by Bean, Franklin, Embrick and Daum (195i) 35S- and 32P-labelled proteins have been used to to be difficult to interpret. Techniques which study problems of digestion and absorption in have been used with some success in human sub- patients after gastrectomy by Nakayama, Ohtsuka, jects fall into two main groups: those which Kuvaishi, Koshibu, Arima, Fukushima, Naka- involve sampling the contents of the lumen of the gami and Fuse (I958). These workers have gut at given levels and those which measure attempted to simulate normal feeding in their changes in the concentration of constituents of the experiments by administering their test substances peripheral blood. McGee and Emery (1940) along with a meal of bread, butter and egg. studied the digestion and absorption of solutions Compared with radioactive tracers, the stable of casein and gelatin by sampling the contents of isotopes have, up to the present, not been used segments of the jejunum blocked distally by in- extensively in problems of clinical interest. As far flated balloons, while Nasset and Davenport (1954) as the stable isotope of nitrogen, 15N, is concerned, investigated the proteolysis of lactalbumin in the besides freedom from possible hazards of radiation, stomach by sampling the gastric contents. The its use would be expected to provide more direct most successful of the techniques so far has been infprmation regarding the digestion and absorption that devised by Blankenhorn, Hirsch and Ahrens of protein and metabolism of subsequent ingestedProtected by copyright. (I955), in which a fine polythene catheter is passed nitrogen by the body, and White and Parson (I950) along the entire gut, enabling sampling at a number and Sharpe, Lassen, Shankman, Gebhardt and of given levels. This method has been used most Hazlet (1956) have used yeast protein so labelled successfully by Borgstrom, Dahlqvist, Lundh and to study some aspects of protein metabolism in Sjovall (1957) to study the digestion and absorp- human subjects. Crane and Neuberger (i96oa, tion of carbohydrate, fat and protein from a I960b) have shown that the time to reach maxi- number of human subjects and to acquire a great mum labelling of both urinary ammonia and urea deal of useful information. A similar technique has in samples of urine passed at frequent intervals also been used by Lundh (1958), who investigated after administering the protein by mouth may be a series of patients with malabsorption following used to investigate the digestion and absorption partial gastrectomy. of protein in normal subjects and in patients suf- The other main type of study measures thanges fering from adult coeliac disease. in amino-acid concentration in the peripheral To some degree criticism can be levelled at all venous blood after the ingestion of a standard the methods of investigation that have been men- quantity of protein or the appearance in the blood tioned. It is difficult to be absolutely certain, for http://pmj.bmj.com/ of an isotopic tracer after giving a dose of labelled example, that the presence of a sampling tube lying protein. Free and Leonards (I944), using un- within the gut does not produce abnormal peri- labelled protein, compared the times to reach stalsis, while experiments involvingsampling of the maximum amino-acid concentration in normal contents from a segment of small bowel closed subjects ingesting meat, whole blood and casein distally by a balloon must be regarded as non- hydrolysate, while West, Wilson and Eyles (1946) physiological. In those studies in which followed the blood amino-nitrogen levels in normal in changes peripheral-blood amino-nitrogen are measured on September 30, 2021 by guest. children and those with deficient pancreatic func- after ingesting a quantity of test substance it is tion after giving gelatin and casein. Christiansen necessary to be sure that these bear some relation and Schwachman (I949) refined this technique and with respect to time to the changes occurring in the measured the changes in concentrations in the portal blood. In patients suffering from mal- plasma of a specific amino-acid, glycine, after nutrition these differences may be considerable. feeding gelatin to both normal persons and those The rate of digestion of a test protein will be with pancreatic fibrosis. dependent upon whether considerable denaturing Other workers have used the rates of appearance has occurred during its preparation and whether of radioactive tracers in the peripheral blood after it is fed together with a quantity of a denatured feeding isotopically-labelled protein. Lavik, Mat- protein like gelatin, or with an undenatured protein thews, Spector and Friedell (1952) administered like skim milk. The fact that these rates may be '31I-labelled casein to children suffering from pan- appreciably different has been used by Bocci (I961) December 1961 CRANE: Some Aspects of Protein Digestion and Absorption in Health and Disease 747 Postgrad Med J: first published as 10.1136/pgmj.37.434.745 on 1 December 1961. Downloaded from to determine the extent of denaturing of 131I- milk powder must be accepted with some reserve, labelled gamma globulins. Work with radioactive since there may be considerable differences in the tracers involves usually the measurement of activity rate of proteolysis of milk protein and the albumin. of samples of whole blood and no attempt is made Using cooked pig's heart as the test protein, to separate and identify specifically-labelled amino- Elsom, Chornock and Dickey (1942), on the other acids. Provided that one can be sure that no hand, found considerable proteolysis in the free iodide is present in the sample, the method is stomach of their subjects. Whether a proportion probably sufficiently accurate for work with 131L- of these large peptides may be further hydro- labelled proteins, but when 35S and 32p tracers are lysed and absorbed by the stomach mucosa is used allowance must be made for the radioactivity not at present definitely known, but it is of some of secondary products occurring in the plasma importance that Linderstrom-Lang and his col- from the metabolism of these substances within the leagues (1952) were able to find peptidases capable body. of hydrolysing small synthetic peptides in the The use of15N-labelled protein offers advantages fundal cells of the pig's stomach. The observation in that the isotope appearing in the urinary of Nasset and Davenport (I954) that after one ammonia and urea is almost certainly derived from hour the concentration of free amino-acid in the all the amino-acids entering the body and that stomach contents after feeding lactalbumin had small quantities can be given alone or mixed with risen tenfold is also of some interest, but these other proteins. The technique can also be used in authors were unable to exclude the regurgitation ill patients where intubation or large amounts of from the duodenum of powerful pancreatic en- protein by mouth would not readily be tolerated. zymes. However, it is unlikely that significant Furthermore, the 15N excreted in the urine and proteolysis could have occurred from the action faeces allows assessment to be made of the amount of these enzymes owing to the low pH of the gastric of tracer retained under different conditions. contents. the of the stomach

Following emptying Protected by copyright. contents in small portions over about four hours, The Digestion and Absorption of Protein in the Swedish workers found that a three- to five- Normal Subjects fold dilution took place in the duodenum and that Hydrolysis of dietary protein (warmed, masti- extensive digestion (50 to 60%) resulted from the cated and moistened with saliva) begins in the action of the pancreatic enzymes. The absorption stomach by the action of pepsin producing a com- of protein was almost complete during the passage plex mixture of high-molecular weight peptides through the proximal Ioo cm. of the jejunum often referred to as ' peptones '. The extent of this (maximum absorption occurring within the second initial proteolysis will depend upon a number of hour), but that about Io% of the test protein factors: the availability of hydrochloric acid and could be recovered unchanged from the lowest pepsin, the type of protein eaten, and the rate at portion of the small intestine. These workers which the stomach empties, and would therefore suggested, therefore, that some dietary protein is be expected to vary considerably from person to normally digested and absorbed in the large bowel. person and from meal to meal in the same person. The pH of the contents of various sections ofthe For example, if the consumption of a meal is rapid, gut were also measured. Over the first three hours http://pmj.bmj.com/ a greater dilution of enzymes and buffering of the the pH of the gastric contents were found to be available hydrochloric acid would be expected, between 3 and 4, while that of the small intestine while the presence of fat and a state of hunger was between 6 and 8, the latter values being would both influence the rate at which the stomach reached only in the ileum. In view of the rapid empties. Lastly, a protein denatured by thorough digestion of the protein in the upper jejunum it is cooking would be attacked more readily by pro- surprising that these values are so far removed from than one the determined for teolytic enzymes only partially denatured, pH optima experimentally pep- on September 30, 2021 by guest. for example, underdone meat. Thus any general sin and the pancreatic enzymes. statement regarding the extent of gastric proteo- Recently Taylor (I959a) has demonstrated that lysis must take into account these features. Cath- gastric juice from normal subjects contains pepsin cart (192I) suggested under normal conditions with maximum activities occurring between pH that the degree of gastric digestion was un- 1.6 to 2.4 and also between pH 3.3 to 4. Measure- important and Borgstrom, Dahlqvist, Lundh and ment of proteolysis in the lower ranges of pH Sjovall (1957) concluded that after feeding a meal alone therefore underestimates the true extent of of corn oil, dextrose and skim milk powder not breakdown of which this enzyme is capable. more than 15% of an admixed tracer dose of Estimates of the overall rates of digestion and 3I-labelled albumin was hydrolysed in the absorption of protein have been made by a number stomach of normal subjects. The assumption by of workers and, as would be expected, these show these authors that this figure applies also to the variations according to the type and quantities of 748 POSTGRADUATE MEDICAL JOURNAL December 1961 Postgrad Med J: first published as 10.1136/pgmj.37.434.745 on 1 December 1961. Downloaded from protein fed. Free and Leonards (I944) found that 07 the maximum amino-acid concentration occurred at three and four hours respectively after the administration of a litre of blood and 400 g. of meat to normal subjects. In studies of normal 06 children, West, Wilson and Eyles (1946) found that these maxima occurred at i hours after feeding either casein or gelatin in amounts 0-5 equivalent to 1.7 g./kg. body weight. Using radioactive techniques, Lavik and others (1952) reported maximum radioactivity of venous blood within three hours after a test meal of 0.5 g. of gelatin/kg. body weight admixed with 131I-labelled albumin, while Chinn and others (I953) have found similar results. Nakayama and others (1958), on the other hand, showed maximum concentrations of radioactivity in the peripheral blood were 0*2 delayed to between six and seven hours in normal Time (04r.) subjects receiving test doses of 35S- and 32P_ labelled proteins along with a meal consisting of 86 g. of carbohydrate, 30 g. of fat and 26 g. of 0I protein. Using much smaller quantities (circa 2 g.) of 15N-labelled yeast protein in fasting subjects (Fig. I). 0 1 2 3 4 5 6 7Protected by copyright. Time (hr.) 05 - FIG. 2.rThe isotopic labelling of urinary ammonia after giving whole yeast protein (-e-) and yeast protein hydrolysate (-o-) to a fasting normal 0*4 tionsubjectof wholein amountsproteinequivalentand theofto o.8 mg.ammonia15N/kg. body weight. Values areabllingplotted asurinaryfor Fig. I. after whoedgivinglabelle yeast protein -) andit was hydrolysis of the proteinin(--)the intea fastines retarded 0-3 - X equivalentsubjtheoverallct inamountabsorptionof bywater-solublealentthemost.8enzymicallymg. inutes body weight.Simialus re plotted as for Fig. i. 0 estimated that under the conditions of study the 00-2 administeringwhole protein and the abosrptionhydof an lysate and labelled yeastprouryein anditrper(was958), O'1 Janneywhoexaminmated(I9I5),that whoundermeasuredthe condisappearancetheurinaryof studyoutputproteinof http://pmj.bmj.com/ nitrogenandmixtrolysisamino-acidandof theglucoseproteinin phloridzinizedin the intestinesdogsrats.ardedof after hen,overall absorption by at thelledmoaterial was protein/kgg.lysate 2).and bySimilarbodyGupta,weightDakrourythereandwasHarperconsiderabled(1958),by I I I I I I proteinneyand amino-acid(915), whomixturesmelabellingofsuredfromtheintestinesurinary outputof rats.of 0 I 2 3 4 5 6 When, however, the 15N-labelled material was Time (hr.) mixednitrogetherwithwithmilk thglucosepowderiof equivalentBoridzinizedtodogs0.4o.4 g.afterof on September 30, 2021 by guest. FIG. i.-Comparison of the rate of change of 15N protein/kg. body weight there was considerable labelling of ammonia (-o-) and urea (-*-) in urine thatprodministein digestionand absorptionis averydro- after whole yeast protein to a fasting normal giving rapidexaminprocess the rates of of subject equivalent to o.8 mg. '5N/kg. body weight. monia and urea disappearanceat about andalimentarythreeprottheto Values are plotted at the mid-points of the collection occurring 24 intervals. four hours respectively, values which are in agree- mentcanal withandthethatthoseoverallalreadydigesratediscussed.ion and absorptionThese findings,of this Crane and Neuberger (Ig6oa) showed that diges- together maximumwithlaptying.those of Borgstr6m,the Dahlqvist, tion and absorption were extremely rapid, urinary protein,Lundh and SjovallSj6vall (I957),(1957),Tofallow of theurinaryconclusionam-olysis ammonia being maximally labelled at about 45 that protein digestion and absorption is a very minutes after administration and urinary urea at rapid process in the upper part of the alimentary about 75 minutes. Little difference could be canal and that the overall rate is controlled mainly detected between the rates of digestion and absorp- by gastric emptying. The very rapid hydrolysis December 1961 CRANE: Some Aspects of Protein Digestion and Absorption in Health and Disease 749 Postgrad Med J: first published as 10.1136/pgmj.37.434.745 on 1 December 1961. Downloaded from of the protein is likely to be due to the very high in lysine and tryptophan, when supplemented with concentrations of trypsin and chymotrypsin which pure amino-acids. From these studies it is difficult are found in the small intestine. Borgstr6m and to escape the conclusion that digestion and absorp- others (1957) found values between 200 to 800 tion of dietary proteins proceed at rates comparable tJg./ml. within a short time of giving a test meal. with the rates of absorption of pure amino-acids. When yeast protein was digested in vitro with A further point of interest in this work is the concentrations of these enzymes equivalent to contribution made by protein from endogenous about 400 Lg./ml., Crane and Neuberger (i96oa) sources. Nasset and his colleagues (see Nasset estimated that more than 50% of the protein was 1957 for review) in a series of experiments claimed converted into peptides soluble in trichloracetic that sufficient endogenous protein is secreted into acid within io minutes, and it is likely that the con- the gut during digestion to exert a marked in- version of protein to peptides is also a very fast fluence on the composition of the amino-acids process in vivo. It has already been mentioned that absorbed from the gut. Although some of their there is considerable uncertainty regarding the conclusions have been questioned by Geiger and his form in which the products of digestion of proteins colleagues (1958), it is certain that a consider- are absorbed by the mucosa. Experiments such as able quantity of endogenous protein is turned over those devised by Wright and Wynn (I95I), which in the intestine in the form of enzymes, mucus, demonstrate the presence of both free amino- shed mucosal cells and possibly plasma protein. acids and peptides in the upper intestine after Mention has already been made of the high con- feeding a protein meal to dogs, do not allow centrations of pancreatic enzymes which are conclusions to be reached regarding the amount of secreted into the small intestine, but Avakian protein already absorbed either in the form of (1961) found that trypsin was rapidly inactivated peptide or as amino-acids, or whether the peptides by incubation with intestinal juice at 37° C., are derived mainly from the dietary protein or although chymotrypsin was more stable. The from the breakdown of endogenous protein turnover of the mucosal cells in animals has beenProtected by copyright. secreted within the lumen of the gut. Newey and estimated at about 24 hours (Leblond and Stevens, Smyth (1959), in animal experiments, have shown 1948). If the figures for human mucosa are com- that simple peptides are hydrolysed inside the parable, the addition of endogenous protein to the mucosal cells. It is reasonable to assume that gut must be considerable and may well be of the larger peptides may be similarly degraded by a order of the 50 to Ioo g./day estimated by Nasset. mixture of peptidases which Wright, Jennings, Information concerning the breakdown and ab- Florey and Lium (I940) have shown to be essenti- sorption of protein in the large intestine is lacking. ally intracellular enzymes. Rapid digestion of Borgstr6m and others (i957) estimated from their dietary proteins which has been observed can studies that Io% of dietary protein remained un- therefore be regarded as mainly due to the action absorbed by the time a test meal had reached the of high concentrations of pancreatic enzymes, the large intestine and concluded that about Io% of chief products of which are peptides, which are in dietary protein is normally degraded and absorbed turn further hydrolysed to amino-acids after there. The work of Reifenstein and his colleagues absorption into the mucosal cells. There is further (I945) on the absorption of diets containing large http://pmj.bmj.com/ evidence for these conclusions: it is known from amounts of protein has already been mentioned. studies such as that of Hokin (i951) that for In those studies in which 250 g. of protein were protein synthesis to occur all the requisite amino- fed daily to human subjects a total of 25 g. of acids must be present at the same moment and that dietary protein would thus be expected to be the body is unable to store quantities of amino- digested and absorbed by the colon, a quantity acids for any length of time. Thus Geiger (1947) equal to between a quarter and one-third of the showed that supplementing a diet deficient in normal daily intake. This suggests that the esti- essential amino-acids by giving the missing amino- mate by the Swedish workers is too high. On the on September 30, 2021 by guest. acids several hours later failed to promote growth other hand, it is possible that some of the intestinal in animals. However, addition of pure amino- enzymes at least, reaching the large intestine, may acids to diets deficient in these substances which escape digestion, since Avakian (1961) records that cause cessation of growth led Rose to the classic '31I-labelled protein can be recovered unchanged discovery of the essential nature of a group of after 24 hours in the human rectum. amino-acids. More recently this type of experiment has been extended to human subjects. The Effect of Gastrectomy on Protein Gomez, Ramos-Galvan, Cravioto and Frenk (1958) Absorption and Scrimshaw, Brissani, Behar and Viteri (I958) The success of partial gastrectomy in the treat- have both reported increase in weight together ment of peptic ulcer in the last 40 years, which with nitrogen retention in children fed diets poor in general rarely produces catastrophic upsets in 750 POSTGRADUATE MEDICAL JOURNAL December 1961 Postgrad Med J: first published as 10.1136/pgmj.37.434.745 on 1 December 1961. Downloaded from nutrition, has been responsible no doubt for the was used. Lundh (1958), using the intubation general belief that protein digestion in the stomach techniques of Blankenhorn, Hirsch and Ahrens is not really essential. The amount of proteolysis (I955), studied a large series of patients after produced by pepsin on a variety of dietary pro- gastrectomy. In patients with Billroth I type of teins in vitro (Fisher, 1954) is not usually marked, operation he found gastric emptying to be more but Cathcart (1921), summarizing earlier views on rapid than with Billroth II, but in both groups this point, suggested that the initial action of there was a very rapid passage of a test meal to pepsin on dietary proteins was nevertheless im- the distal parts of the small intestine and also portant for the subsequent proteolysis by the evidence of defective digestion. Concentrations pancreatic enzymes. The recent work of Taylor of trypsin in the lumen of the gut were found to (I959a and I959b) shows that the true extent of be low and in a number of samples no activity gastric proteolysis is underestimated if measured was found in patients with the Billroth II type of only at pH 2, and has demonstrated that in gastrectomy. In Billroth I patients trypsin was patients with pernicious anemia, activity is present mixed with a test meal, but the poor mixing in at higher pH values. Billroth II patients was explained by the lack of When the details of a large series of patients co-ordination of pancreatic secretion following with adequate follow-up after partial gastrectomy gastric emptying. The conclusion that the poor were studied, it became clear that a large majority mixing of protein with pancreatic enzymes was a of the patients had lost weight and further studies more important factor than rapid intestinal passage (see Lundh, 1958, for review) showed that weight after gastrectomy was based upon the disappear- loss was greater in Billroth II operations than with ance of l31I-labelled albumin from the small Billroth I and that steatorrhcea and increased intestine. fecal nitrogen losses were greater in the former It has already been pointed out that the rates operation. In patients with total gastrectomy these of digestion of denatured and native proteins findings were shown to be more marked and differ markedly and the assumption therefore that Protected by copyright. more persistent. A large amount of work on the proteins of the normal diet behave in the gastric proteolysis following gastrectomy has same manner as albumin remains to be proved. shown that in general the extent of digestion is Only one patient with a sub-total gastrectomy impaired by the increased rates of emptying of has been investigated with 15N-labelled yeast the stomach remnant and the reduced amount of protein, so far. After feeding 15N-labelled yeast acid and pepsin available, but studies on pan- protein to a patient who underwent an operation creatic function following gastrectomy have led to three years previously for a three-quarter resection conflicting results. Under normal conditions of the stomach for suspected carcinoma and who secretin and pancreozymin which stimulate pan- presented with a steatorrhtea of an average of creatic secretion are released from the duodenum I gm. of fat together with 2 to 3 gm. of nitrogen by the passage of contents from the stomach. per day, it has been found that the rate of uptake Nakayama and his colleagues (1958), using 32P- of the protein was within normal limits. When and 3S-labelled proteins, have shown that the rate the experiment was repeated and an amount of of absorption is lower, and the faecal excretion of milk protein equivalent to 0.4 gm./kg. body weight http://pmj.bmj.com/ isotope greater, in patients after total gastrectomy added to the yeast protein, maximum concentra- than in normal subjects. The absorption of their tions of isotope in the urinary ammonia and urea test material was more effective after oesophageal- were reached more quickly than in normal sub- duodenal anastomosis compared with oesophageal- jects and this was considered to result from the jejunal anastomosis. more rapid entry of the test materials into the The changes in this mechanism resulting from small intestine. From the first study the amount gastrectomy (Annis and Hallenback, 1952) were of isotope eliminated in the urine during the first studied in dogs after Billroth I and II operations six hours was found to be as high as Io% of the on September 30, 2021 by guest. and results showed that there was marked diminu- amount ingested. In normal subjects this value tion of pancreatic secretion during the ingestion was reasonably constant between 6 and 7%. It of cooked meat although response to secretin was is suggested that these high figures indicate con- nevertheless normal. Using 311-labelled proteins siderable malutilization of the absorbed protein a number of investigations, for example those of since not less than 80% of the nitrogen of the Babb, Chinn, Stitt, Lavik, Levey, Kreiger and urine is contained in the urea fraction. Abbott (i953), have shown that there may be rapid increases in the radioactivity of the plasma The Effect of the Loss of Pancreatic after Billroth II operations, but a more slow rise Secretion on Protein Absorption and more prolonged activity in patients with Because of their high activity the loss of the Billroth I operations, when 131I-labelled albumin pancreatic enzymes to the body would be expected December 1961 CRANE: Some Aspects of Protein Digestion and Absorption in Health and Disease 751 Postgrad Med J: first published as 10.1136/pgmj.37.434.745 on 1 December 1961. Downloaded from to produce extremely serious nutritional conse- protein in this patient who is alive and well quences. Cystic disease of the pancreas in children ii years after his operation. After three days (Clarke and Hadfield, 1924) or surgical excision without supplements of pancreatin it has been of the pancreas may produce marked weight loss shown that this patient did absorb a considerable with the appearance of large quantities of fat and quantity of his test protein although the absorp- nitrogen in the faeces, but in disturbances of pan- tion was delayed as measured by the rates and creatic function following acute or chronic pan- extent of labelling of urinary ammonia and urea. creatitis all grades of disability can be encountered, These findings in a patient who has lost the though the finding of markedly increased faecal benefit of practically all his proteolytic enzymes, nitrogen and fat usually denotes quite severe thought essential for the digestion of dietary damage. It is possible that apart from the two protein, is of some interest. Nakayama and others conditions mentioned, there is never complete (I958) have recorded also that in ten patients loss of the external secretion of the pancreas. undergoing combined gastrectomy and pan- Some estimate of pancreatic insufficiency may be createctomy, digestion and absorption of labelled obtained by duodenal intubation of patients after protein, although incomplete, was neverthelsss stimulation of the pancreas by secretin or pan- possible. Handleman, Golden and Pratt (I934) creozymin and measuring the concentrations of have recorded that dogs deprived of pancreatic various enzymes. For trypsin and chymotrypsin, secretion can also absorb large quantities of food Lundh (I957) has recently introduced a very and that weight loss after operation is regained by sensitive method. excessive dietary intake but fxecal nitrogen losses West, Wilson and Eyles (1946) fed gelatin to rise accordingly. More recently Coles (I96I) children with deficient pancreatic function and reported that after the occlusion of the pancreatic showed that the rise of blood amino-acid nitrogen duct of young cats there was hypertrophy of the compared with normal children was extremely small intestinal mucosa and increased proteolytic depressed, but this situation could be improved activity was found but positive evidence on these Protected by copyright. by administering pancreatin. Test proteins tagged lines has not yet been obtained for man. usually with 131I have been used extensively to investigate pancreatic disease, and Chinn, Lavik, The Absorption of Protein in Coeliac Disease Stitt and Buckaloo (1952) fed 13I-labelled albumin There has not been much work reported on to patients with pancreatic disease and showed protein absorption in cceliac disease. In one of that the faecal excretion of the isotope was greater the few studies Angfanger and Heavenrich (I949) than normal. Freeark, Kozoel and Meyer (1957) investigated a series of children with the con- showed that the increase of radioactivity appearing dition using a test dose of gelatin, and concluded in the blood after giving a dose of '31I-labelled that digestion and uptake were within normal albumin was low compared with healthy subjects. limits. Prior to the publications of Taylor, Wol- Polachek and others (I959) carried out a similar laeger, Comfort and Power (1952) and Cooke, investigation on patients with severe pancreatic Thomas, Mangall and Cross (I953) it was gener- dysfunction and found identical results for the ally held that in adult coeliac disease there was plasma, but the faecal excretion in their patients little disturbance of nitrogen assimilation, since http://pmj.bmj.com/ was seldom abnormal. These studies are of diag- estimates of faecal nitrogen were reported as little nostic interest and confirm what would be ex- different from normal values. Further interest in pected from the loss of pancreatic enzymes by the the problem arose from studies by these two body, but are, nevertheless, subject to a criticism groups who showed that considerable losses of which has already been discussed. The test pro- nitrogen can nevertheless occur and the observa- tein used is a native one and hence may behave tions of Paulley (I954) and Shiner and Doniach in a different way from the cooked and denatured from that there was often (I958) biopsy specimens on September 30, 2021 by guest. dietary proteins that are eaten by man. gross pathological change in the jejunal mucosa The disturbance of protein metabolism is much with partial or complete atrophy of the villi. more marked following total pancreatectomy. Crane and Neuberger (i96ob and c) fed quantities Whitfield, Gourevitch and Thomas (1952) pub- of 15N-labelled yeast protein and protein hydro- lished a detailed metabolic study on a patient lysate to four adult patients with severe coeliac who had a partial gastrectomy and total pan- disease and found that the times for maximum createctomy. They showed that although fat labelling of both urinary ammonia and urea absorption improved with pancreatin the excre- (Fig. 3) when whole protein was used were tion of nitrogen was not greatly affected and reached between 60 and 90 minutes later than remained between 5 and 8 g. a day on a normal was found for normal subjects (Fig. i). When a diet. A preliminary communication has already hydrolysate was given there was again a delay in been given (Crane, 1961) on the fate of 15N-yeast maximum labelling (Fig. 4), but this was more 752 POSTGRADUATE MEDICAL JOURNAL December I96I Postgrad Med J: first published as 10.1136/pgmj.37.434.745 on 1 December 1961. Downloaded from 03 0'2

(O ~0-2 io.I

I I I I 0 1 2 3 4 5 Time (hr.) FIG. 3.-'5N content of urinary ammonia (-o-) and urea 0 1 2 3 4' 5 6 (-*-) after giving labelled whole yeast protein Time (hr.) (0.4 mg. :5N/kg. body weight) mixed with skim milk powder equivalent to 0.4 g. protein/kg. body FIG. 4.-Comparison between the 15N labelling of weight to a Values are as urinary ammonia after giving whole yeast protein fasting subject. plotted (-q-) and hydrolysate (-o-) to a fasting adult patient for Fig. i. with cceliac disease (0.4 mg. SN/kg. body weight). Values are plotted as for Fig. I. variable and considerably less than when whole was fed. These results were protein explained by Protected by copyright. assuming that while atrophy of the villi of the Experimental work in dogs (Flint, I9I2) has shown jejunum reduced its total absorbing surface and that up to 50% of the small bowel can be resected hence caused some delay in the uptake of the hy- with complete recovery, but in human beings the drolysate, the reduced total activity of intracellular disability appears to be more variable. This may peptidases necessitates a greater digestion of the be due to disease in the remaining portions of the protein by pancreatic enzymes within the lumen gut-for example, Crohn's disease-or formation of the gut. Analysis of the stool nitrogen for 15N of blind loops, but Jackson (1958) believes that in showed that there was little difference between the absence of complications the resection of less the amounts of isotope excreted when hydrolysate than two-thirds of the small bowel causes no was substituted for whole protein and that in each serious metabolic disturbances. After massive case about two-thirds was in a form insoluble in resection large losses of faecal nitrogen occur; these trichloacetic acid indicating a loss of whole pro- may be aa s 50% of the nitrogen intak and tein. It seemed possible, therefore, that this loss amino-acid tolerance tests are abnormal (Altha sen, of protein, which was larger than that found for Uyeyama and Simpson, 1949). Perhaps the most normal subjects, could contribute to the hypo- striking feature of reported cases is that within a http://pmj.bmj.com/ albuminaemia, muscle wasting and negative nitro- few months of massive resection, bowel move- gen balance often encountered in this disease. ments, initially 20 or so a day, return to normal In one patient who responded rapidly to with- and patients often regain their pre-operative body drawal of gluten from the diet it was found after weights. Flint (I9I2) showed that the remaining three weeks that the absorption of labelled proportion of the small bowel of his dogs underwent tein and the elimination of the isotope in the considerable hypertrophy with increased size of faces had returned to normal. In more recent the villi, and Jackson (1958), summarizing the on September 30, 2021 by guest. investigations from this laboratory, yet unpub- findings of a number of investigations of patients, lished, it has been found that faecal losses of reports considerable dilatation and hypertrophy of isotope may diminish before improvement in the remaining portion of the small bowel seen absorption and digestion of the labelled protein. either at autopsy or on radiological examination, but there was no convincing demonstration of The Absorption of Protein in Massive Resec- villous hypertrophy. However, from the clinical tion of the Small Intestine condition of some of the patients, a considerable Immediately following massive resection of the compensation in digestion and absorption of pro- small bowel for such conditions as mesenteric tein must take place. Perhaps the dilatation and thrombosis, injury, etc., the body is faced with a hypertrophy serve as a functional reservoir by sudden decrease in total absorbing area of bowel which protein, as has been suggested in adult and mucosal cells capable of elaborating enzymes. coeliac disease, may be more thoroughly attacked December 1961 CRANE: Some Aspects of Protein Digestion and Absorption in Health and Disease 753 Postgrad Med J: first published as 10.1136/pgmj.37.434.745 on 1 December 1961. Downloaded from by pancreatic enzymes, or a larger surface area of of absorbing surface of the intestine due to surgery the villi enables peptides to be more readily or disease. It is hoped that a combination and ex- absorbed and broken down, but further investiga- tension of some of the techniques described will tion on these patients is obviously required to help to eliminate some of the points on which elucidate the problem. there is at present still some doubt. It seems likely from the evidence that has been that the I wish to acknowledge with gratitude the support of discussed in these few selected topics Professor W. Melville Arnott and Dr. J. M. French of digestion and absorption of dietary protein by the Department of Medicine, University of Birmingham man is a more complicated and more rapid affair and of Professor A. Neuberger, Department of Chemical than would appear from the accounts usually given. Pathology, St Mary's Hospital, London, in whose In some conditions evidence has been forward Department the earlier 15N work reported here was put carried out. I also wish to thank the United Birming- suggesting that some degree of compensation takes ham Hospitals Endowment Research Fund and the Med- place for the loss of particular enzymes or the loss ical Research Council for financial help. REFERENCES ALTHAUSEN, T. L., UYEYAMA, K., and SIMPSON, R. G. (1949): Digestion and Absorption after Massive Resection of the Small Intestine, Gastroenterology, 12, 795. ANGFANGER, H., and HEAVENRICH, R. M. (I949): Aminoacid Tolerance in Children, Amer. J. dis. Child., 77, 425. ANNIS, D., and HALLENBECK, G. A. (1952): The Effects of Partial Gastrectomy on Canine External Pancreatic Secretion, Surgery, 31, 517. AVAKIAN, S. (I96I): Current Concepts in Therapy: Chymotrypsin and Trypsin, New Engl. J. Med., 264, 764. BABB, L. I., CHINN, A. B., STITT, R. M., LAVIK, P. S., LEVEY, S., KREIGER, H., and ABBOTT, W. E. (1953): Evaluation of Protein and Fat Metabolism in Postgastrectomy Patients, Arch. Surg., 67, 462. BANGHAM, D. R., INGRAM, P. L., ROY, J. H. B., SHILLAM, K. W. G., and TERRY, R. J. (1958): The Absorption of 131I- labelled Serum and Colostral Proteins from the Gut of the Young Calf, Proc. roy. Soc., B 149, 184. -, and TERRY, R. J. (1957): The Absorption of 131I-labelled Homologous and Heterologous Serum Proteins Fed Protected by copyright. Orally to Young Rats, Biochem. J., 66, 579. BEAN, W. B., FRANKLIN, M., EMBRICK, J. F., and DAUM, K. (195 ): Absorption Studies Using Portal Anastomotic Veins, J. clin. Invest., 30, 263. BLANKENHORN, D. H., HIRSCH, J., and AHRENS, E. H. (1955): Transintestinal Intubation Technique for Measurement of Gut Length and Physiologic Samping at Known Loci, Proc. soc. Exp. Biol. (N. Y.), 88, 356. BoccI, V. (1961): Proteolysis of 31I-labelled Gamma Globulin, Int. J. appl. Radiat., 10, 94. BORGSTROM, B., DAHLQVIST, A., LUNDH, G., and SJOVALL, J. (1957): Studies of Intestinal Digestion and Absorption in the Human, J. din. Invest., 36, 1521. CATHCART, E. P. (1921): ' The Physiology of Protein Metabolism '. London: Longmans, Green. CHINN, A. B., LAVIK, P. S., BABB, L. I., BUCKALOO, G. W., STITT, R. M., and ABBOTT, W. E. (1953): Blood Isotope Levels following a Test Meal of 13I-labelled Protein, J. Lab. clin. Med., 42, 377. , -, STITT, R. M., and BUCKALOO, G. W. (I952): Use of 13I-labelled Protein in the Diagnosis of Pancreatic Insufficiency, New Engl. J. Med., 247, 877. CHRISTIANSEN, H. N., and SHWACHMAN, H. (1949): Determination of the Plasma Glycine after Feeding Gelatin as a Diagnostic Procedure for Pancreatic Fibrosis, J. clin. Invest., 28, 319. CLARKE, C., and HADFIELD, G. (I924): Congenital Pancreatic Disease with Infantilism, Quart. J. Med., 17, 358. COLES, B. L. (1961): Report of Symposium on Small Intestinal Function, Lancet (1961), i, 1342. COOKE, W. T., THOMAS, G., MANGALL, D., and CROSS, H. (1953): Observations on the Faecal Excretion of Total Solids, http://pmj.bmj.com/ Nitrogen, Sodium, Potassium, Water and Fat in the Steatorrhoea Syndrome, Clin. Sci., 12, 223. CRANE, C. W. (I96I): Report on Symposium on Small Intestinal Function, Lancet, i, 1342. , and NEUBERGER, A. ( 96oa): The Digestion and Absorption of Protein by Normal Man, Biochem. J., 74, 3 3. (,(960b): Absorption and Elimination of "'N after Administration of Isotopically Labelled Yeast Protein and Yeast Protein Hydrolysate to Adult Patients with Cceliac Disease: I. Rate of Absorption of 5N Yeast Protein and Yeast Protein Hydrolysate, Brit. med. J., ii, 815. , (I96oc): Absorption and Elimination of 15N after Administration of Isotopically Labelled Yeast Protein and Yeast Protein Hydrolysate to Adult Patients with Cceliac Disease: II. Elimination of the Isotope in the Urine

and Faeces, Ibid., ii, 888. on September 30, 2021 by guest. DENT, C. E., and SCHILLING, J. A. (I949): Studies on the Absorption of Proteins: The Amino-acid Pattern in the Portal Blood, Biochem. J., 44, 318. ELSOM, K. A., CHORNOCK, F. W., and DICKEY, F. G. (I942): Intubation Studies of the Human Small Intestine, J. clin. Invest., 21, 795. FISHER, R. B. (1954): ' Protein Metabolism'. London: Methuen. FLINT, J. M. (1912): The Effect of Extensive Resection of the Small Intestine, Bull. Johns Hopk. Hosp., 23, 127. FREE, A. H., and LEONARDS, J. R. (1944): Studies on the Ingestion of Large Quantities of Protein and Aminoacids, J. Lab. clin. Med., 29, 963. FREEARK, R. J., KOZOEL, D. D., and MEYER, K. A. (1957): The use of 13I-labelled Albumin in the Diagnosis of Pan- creatic Disease, Surgery, 41, 268. GEIGER, E. (1947): Experiments with Delayed Supplementation of Incomplete Aminoacid Mixtures, J. Nutr., 34, 97. , HUMAN, L. E., and MIDDLETON, M. J. (1958): Nitrogen Content of Gastrointestinal Tracts of Rats during Absorp- tive Period, Proc. Soc. exp. Biol. Med., 97, 232. GOMEZ, F., RAMOS-GALVAN, R., CRAVIOTO, J., and FRENK, S. (1958): Prevention and Treatment of Chronic Severe Infantile Malnutrition (Kwashiorkor), Ann. N.Y. Acad. Sci., 69, 969. 754 POSTGRADUATE MEDICAL JOURNAL December 1961 Postgrad Med J: first published as 10.1136/pgmj.37.434.745 on 1 December 1961. Downloaded from GRUSKAY, F. L., and COOKE, R. E. (1955): The Gastrointestinal Absorption of Unaltered Protein in Normal Infants and in Infants Recovering from Diarrhcea, Peaiat., x6, 763. GUPTA, J. D., DAKROURY, A. M., and HARPER, A. E. (1958): Observations on Protein Digestion in vivo,J. Nutr., 64, 447. ·HANDLEMAN, M. B., GOLDEN, L. A., and PRATT, J. H. (1934): The Effect of Variations in the Diet on the Absorption of Food in the Absence of Pancreatic Digestion, Ibid., 8, 479. HOKIN, L. E. (1951): Aminoacid Requirements of Amylase Synthesis by Pigeon Pancreas Slices, Biochem. J., 50, 2x6. JACKSON, W. P. U. (1958): Massive Resection of the Small Intestine, in ' Modem Trends in Gastroenterology ', second series, ed. by Jones, F. A. London: Butterworth. JANNEY, N. W. (1915): A Note on the Rate of Metabolism of Protein and Aminoacids, J. biol. Chem., 22, I91. KEKWICK, R. A. (1959): The Serum Proteins of the Fetus and Young of Some Mammals, 'Advances in Protein Chemistry ', vol. 14. New York Academic Press. KURODA, Y., and GIMBEL, N. S. (1954): Selective Disappearance of Aminoacids from the Human Small Intestine, J. appl. Physiol., 7, 148. LAVIK, P. S., MATTHEWS, L. W., SPECTOR, S., and FRIEDELL, H. (1952): Utilization of 13I-labelled Casec in the Study of Protein Digestion and Absorption, Amer. J. dis. Child., 84, 736. LEBLOND, C. P., and STEVENS, C. E. (1948): The Constant Renewal of the Intestinal Epithelium in the Albino Rat, Anat. Rec., 100, 357. LINDERSTROM-LANG, K. U. (1952): ' Lane Medical Lectures: Proteins and Enzymes '. California: Stanford University Press. LUNDH, G. (1957): Determination of Trypsin and Chymotrypsin in Human Intestinal Contents, Scand. . clin. Lab. Invest., 9, 229. (1958): Intestinal Digestion and Absorption after Gastrectomy, Acta chir. scand., Suppl. 23I. LUSK, G. (1928): ' The Elements of the Science of Nutrition ', fourth edition. London: Saunders. McGEE, L. C., and EMERY, E. S. (1940): Rate of Absorption of Aminoacids from the Small Intestine of Man, Proc. Soc. exp. Biol. Med., 45, 475. NAKAYAMA, K., OHTSUKA, A., KUVAISHI, T., KOSHIBU, M., ARIMA, M., FUKUSHIMA, M., NAKAGAMI, T., and FUSE, S. (1958): 'Study of Digestion and Absorption with Radioactive Isotopes following Gastrointestinal Surgery: Isotopes in Medicine'. Geneva: United Nations. NASSET, E. S. (1957): The Role of-the Digestive Tract in the Utilization of Protein and Aminoacids, J. Amer. med. Ass.,

I64, 172. Protected by copyright. - , and DAVENPORT, A. (1954): Canine and Human Digestion of Proteins in vivo, J. appl. Physiol., 7, 447. NEWEY, H., and SMYTH, D. H. (1959): The Intestinal Absorption of Some Di-peptides, J. Physiol., 145, 48. PARISH, W. E., BARRETT, A. M., COOMBS, R. R. A., GUNTHER, M., and CAMPS, F. E. (1960): Hypersensitivity to Milk and Sudden Death in Infancy, Lancet, ii, I I6. PAULLEY, J. W. (I954): Observations on the iEtiology of Idiopathic Steatorrhcea, Brit. med. J., ii, 1318. POLACHEK, A. A., COPE, C. B., WILLIARD, R. F., and BARNES, F. W. (1959): 13'I-labelled Protein and Fat Meals in Patients with Chronic Pancreatitis, Gastroenterology, 37, 38. REIFENSTEIN, E. C., ALBRIGHT, F., and WELLS, S. L. (1945): The Accumulation, Interpretation and Presentation of Data Pertaining to Metabolic Balances, Notably Those of Calcium, Phosphorus and Nitrogen, J. clin. Endocr., 5, 367. SCRIMSHAW, N. S., BRISSANI, R., BEHAR, M., and VITERI, F. (1958): Supplementation of Cereal Proteins with Amino- acids, J. Nutr., 66, 485. SHARPE, G. S., LASSEN, S., SHANKMAN, S., GEBHART, A. F., and HAZLET, J. W. (1956): Studies of Protein Absorption Using Nitrogen 15 as a Tag, Ibid., 58, 443. SHINER, M., and DONIACH, I. (1958): Histopathologic Studies in Steatorrhcea, Proc. World Congr. of Gastroenterology, Washington, 1958, i, 586. Baltimore: Williams and Wilkins. TAYLOR, W. H. (1959a): Studies on Gastric Proteolysis, I, Biochem. J., 71, 73.. - (I959b): Gastric Proteolysis in Disease, I, J. clin. Path., 12, 2Io. http://pmj.bmj.com/ TAYLOR, A. B., WOLLAEGER, E. E., COMFORT, M. W., and POWER, M. H. (1952): The Effect of Cortisone on Non- tropical Sprue, Gastroenterology, 20, 203. WEST, C. D., WILSON, J. L., and EYLES, R. (1946): Changes in Blood Amino Nitrogen Levels following Ingestion of Protein and of a Protein Hydrolysate in Infants with Normal and with Deficient Pancreatic Function, Amer. J. dis. Child., 72, 25 . WHITFIELD, A. G. W., GOUREVITCH, A., and THOMAS, G. (1952): Metabolic Effects of Total Pancreatectomy in Man, Lancet, i, I8o. WHITE, A. G. C., and PARSON, W. (1950): Comparative Studies on Nitrogen Excretion, Arch. S. and M. of Unaltered Biochem., 26, 205. WILSON, J., WALZER, (1935): Absorption Egg Protein in Infants and Children, Amer. dis. on September 30, 2021 by guest. Child., 50, 49. J. WRIGHT, R. D., JENNINGS, M. A., FLOREY, H. W., and LIUM, R. (1940): The Influence of Nerves and Drugs on Secretion by the Small Intestine and an Investigation of the Enzymes in the Intestinal Juice, Quart.J. exp. Physiol., 30, 73- ,and WYNN, V. (I951): The Digestion of Protein in the Alimentary Canal, Aust. J. exp. Biol. med. Sci., 29, 281.