Pancreatic Lipolytic Enzymes in Human Duodenal Contents Radioimmunoassay Compared with Enzyme Activity

Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 01/05/15 For personal use only.

B. STERNBY, A. NILSSON, T. MELIN & B. BORGSTROM Dept. of Medical and Physiological Chemistry, University of Lund, and Dept. of Internal Medicine, University Hospital of Lund, Lund, Sweden Sternby B, Nilsson A, Melin T, Borgstrom B. Pancreatic lipolytic enzymes in human duodenal contents. Radioimmunoassay compared with enzyme activity. Scand J Gastroenterol 1991, 26, 859-866 The total pancreatic lipolytic capacity was determined in duodenal contents in healthy humans 10-120 min after a liquid test meal, by estimating the amount of pancreatic lipase, colipase, carboxyl ester lipase, and phospholipase A2 by means of radioimmunoassays and enzymatic assays. The molar concentrations of the different proteins were of the same order of magnitude. The relative specific activity (enzyme activity/milligram immunoreactive protein expressed as a percentage of the specific activity of the respective pure protein) amounted to 75120% for lipase, 4580% for colipase, 30-70% for carboxyl ester lipase, and 45120% for phospholipase A2.These vaned, and sometimes low values can be explained by the fact that the enzymes are inhibited or partly inactivated in the duodenal contents by surface denaturation, in which cases the products are still immunoreactive. Also, the proforms of colipase and phospholipase A2 may not always be completely activated. Furthermore, the specific activities of the pure enzymes (and thus the relative specific activities) are related to the methods used, which are not specific enough to distinguish completely the three enzymes and the cofactor in duodenal contents. Key words: Carboxyl ester lipase; colipase; enzymatic assay; human; lipase; pancreatic; phospholipase A*; radioimmunoassay

B. Sternby, Dept. of Medical and Physiological Chemistry 4, P.O. Box 94, S-221 00 Lund, Sweden

The levels of pancreatic lipolytic enzymes in human small-intestinal contents are generally determined by measuring enzyme activities (14). In recent years radioimmunoassays for the pancreatic lipolytic enzymes have also become available (5, 6). So far, these methods have not been used to estimate the mass concentrations of the lipolytic enzymes in normal human duodenal contents. In this study we present for the first time a comparison between the two kinds of methods for measuring the three pancreatic lipolytic enzymes lipase, carboxyl ester lipase, and phospholipase A2 and the cofactor colipase in human duodenal contents after a liquid test meal. By comparing the enzymatic activity per

milligram protein in the duodenal samples and the pure proteins, information might be obtained as to whether the enzymatic assay used is likely to underestimate widely the amount of enzyme protein secreted, owing to assay problems or a rapid inactivation of the enzyme. MATERIALS AND METHODS Eleven healthy subjects (26-46 years old, six women and five men) with no history of gastrointestinal disease participated. The protocol had been approved by the local ethical committee, and the subjects had given informed written consent.

Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 01/05/15 For personal use only.

860

B. Sternby et ul.

Experimental procedure The subjects fasted from midnight. At about 0800 h they were subjected to duodenal intubation with a one-lumen nasoduodenal polyvinyl tube with a diameter of 4mm. The final location of the tube (just proximal to the ligament of Treitz) was checked by roentgenography. The ending of the tube and the leader during the intubation procedure were made of radiopaque material. The test meal consisted of 300ml liquid formula (Semper sondnaring, Ilgviskos, Semper AB, Stockholm, Sweden), containing 4.5 g protein, 12 g carbohydrate, and 3.5 g fat (palm and sunflower) per 100ml, plus minerals and vitamins. Ten microcuries of the non-absorbable marker, 1,2-'4C-polyethyleneglycol(PEG) (New England Nuclear) were added to the test meal. The specific radioactivity of the PEG was 15 mCi/g. After the test meal intestinal contents were collected on ice by siphonage for a total period of 2 h, divided into eight intervals, as indicated in the figures. The samples were divided into aliquots and were kept frozen until analyzed. Storage for up to 10 months after collection, using these conditions for collection of intestinal contents obtained after a test meal, has been found not to affect the pancreatic enzyme activities significantly ( 3 , 7).

Analyses Enzyme activities (mmol/min 'I) and enzyme mass (mg/l) were measured on the same day. after storage in the freezer for about 2 months. After being thawed, the samples were kept on ice. Lipase and colipase activities were assayed with a modification of the method described by Hildebrand & Borgstrom (3). In brief, lipase and colipase were assayed by automatic titration using a pH-stat (Mettler, Switzerland). Glycerol tributyrate was used as substrate. Colipase 'activity' measures the ability of colipase to activate bile salt-inhibited pancreatic lipase. Thus an excess of pure human pancreatic lipase (20pmol) was present in the colipase assay, which also includes 4 mM sodium taurodeoxycholate (NaTDC). Lipase and colipase activities were determined at room temperature at pH7.0 when the rate of

hydrolysis was stable and are expressed in millimoles of free fatty acids (FFA) released per minute and liter. Carboxyl ester lipase activity was measured at room temperature by a spectrophotometric method using p-nitrophenylacetate as substrate (8). Phospholipase A 2 activity was measured at 40°C with egg yolk phospholipids as substrate (9). All determinations of enzyme activity were carried out at least in duplicate, using intestinal samples that had been diluted 100 times in the radioimmunoassay (RIA) buffer (see below). Polyclonal monospecificRIA of lipase, colipase, and phospholipase A2 was done in accordance with Sternby & Akerstrom (5). The RIA for carboxyl ester lipase was performed as described by Aho et al. (6), using monospecific polyclonal antiserum. The RIAs for phospholipase AZ, lipase, and colipase are specific for the respective lipolytic pancreatic proteins, whereas the RIA for carboxyl ester lipase shows a cross-reactivity with bile salt-stimulated lipase in human milk. Since the levels of pancreatic lipases in smallintestinal contents are high, the samples had to be diluted 1:10,000 before the RIAs. The dilutions of duodenal juice were made with RIA buffer (0.1 M phosphate buffer, pH 7.5, also containing 0.1% (w/v) bovine serum albumin and 0.02% (w/ v) sodium azide (NaN3).The RIA determinations were done at least in triplicate. The dilution of the test meal was measured by comparing I4CPEG per milliliter in the given meal and in the samples of duodenal contents. Aliquots of 50100 PI were added to 10 ml of Instagel :toluene (1:l) and I4C, and the determination was carried out in a Packard 460CD liquid scintillation counter equipped with computerized external standard for quench correction. The dilution factor was thus defined as the ratio of 14C-PEG per 1 ml test meal and 14C-PEGper 1 ml duodenal contents. The RIAs were also used to check whether any molecular fragments of the antigen contributed to the measured quantities. Duodenal contents from different patients and time periods were diluted as for standard curve determination to ascertain whether they were superimposable on the standard curve.

Radioimmunoassay versus Enzyme Activity Lipase

Lipase 6000

1000

25

0

Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 01/05/15 For personal use only.

861

50

loo

75

0

0

115

0

25

50

75

minutes

minutes

Colipase

Colipase

0

100

125

A

-.

600

-

aE

450

6000

A

300 150

0 5

minutes

minutes

Carboxyl ester lipase

Carboxyl ester lipase

40m

2000

-

2

:::

1500

1000

so0

0

I 25

50

75

100

125

minutes

minutes

Phospho I ipose A2

Phospholipase A 2 100

G

3

0

0

1

1

.C

E

1 '.

E"E

T

0

0

25

50

75

minutes

I

100

125

n

0

25

50

75

100

1 5

minutes

Fig. 1. Distribution of immunoreactivity (closed symbols) and enzymatic activity (open symbols) of the four lipolytic pancreatic proteins in human duodenal contents. Values were obtained from 11 healthy volunteers at eight time intervals. N o corrections for dilution of the test meal are included.

862

B.

al.

Sternby et

Lipase

Lipose 5

0

0

0

l

1

100

125

52 - '

Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 01/05/15 For personal use only.

minutes

minutes Colipase

Colipase

,

- 005

25

50

75

100

6000 - I

0

12s

25

Carboxyl ester lipase

75

Carboxyl ester lipase

1500 40

5

M

minutes

minutes

7

1000

E C

; 0

500

0

25-

25

50

75

100

125

minufes

minutes Phosphotipase A 2

Phospholipase A2 60 1

' 0 ° 250

-

7 - I I

I

0 0

25

50

75

minutes

100

I25

-0

25

So is minutes

too

1 5

Fig. 2. Mean levels and standard deviations of immunoreactivity (closed symbols) and enzymatic activity (open symbols). For further explanations, see legend to Fig. 1.

Radioimmunoassay versus Enzyme Activity

Calculations The enzyme activities were calculated in millimoles FFA released per minute and liter intestinal contents (mmol/min 1). The amount of the protein was expressed in milligrams protein per liter (mg/l). Molar concentrations of the different proteins were calculated by dividing the concentration of protein (mg/l) by the molecular weight (lipase, 48,000; colipase, 10,000; carboxyl ester lipase, 100,000; and phospholipase A*, 14,000). The relative specific activities of the lipolytic pancreatic proteins present in duodenal contents were calculated as one hundred times the ratio between enzymatic activity and immunoreactivity divided by the specific activity of the pure enzyme proteins. The relative specific activity was thus expressed as a percentage of that of the pure protein. The lipolytic proteins in duodenal contents could thereby be compared with those of the pure proteins. The specific activities used for the different enzymes have been 8000,25,000, 40, and 200 pmol/min/mg for lipase (lo), colipase ( l l ) , carboxyl ester lipase (12), and phospholipase A2 (13), respectively.

Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 01/05/15 For personal use only.

-

RESULTS

Of the eleven subjects intubated a complete collection at all eight time intervals was obtained from four. Fig. 1 shows the time courses for the mass concentration and the activity of each lipolytic pancreatic protein in intestinal contents obtained from each individual during the 2-h period after the test meal. The variation between the different individuals was large. The mean value and standard deviation for both masses and activities for each time period are shown in Fig. 2. The average mass concentration expressed as milligram protein per liter was highest for carboxyl ester lipase, the concentration of which was roughly twice that of lipase. The colipase and phospholipase A2 concentrations were less than 20% of that of carboxyl ester lipase. The molar concentrations of the four proteins were, however, of the same order of magnitude. Except for the first and last time intervals the

863

L

0

zU

9-

S

0 .+

.-3

5-

a _ .

0

25

50

75

100

125

minutes Fig. 3. Dilution factor of the test meal (mean values and standard deviations). The dilution factor was calculated as I4C-PEG per milliliter test meal divided by I4C-PEG per milliliter duodenal contents.

dilution factor obtained from the non-absorbable reference substance present in the meal was close to two (Fig. 3). This indicates that the concentration of enzymes in the pancreatic juice is rather constant over the observed time period and at least twice those measured in the intestinal contents. The relative specific activities of the different proteins were calculated and, presented as a percentage of the specific activity of the respective pure enzymes, are shown in Fig. 4. For lipase the highest relative specific activity was observed 20 min after the test meal and for colipase after 30-50 min. The relative specific activity for carboxyl ester lipase and phospholipase A2 showed no distinct time optimum. Table I shows the correlation coefficients for all values of immunoreactivity and enzymatic activity (with the dilution factor included) for all the four proteins. There was a good correlation between mass data obtained for the different proteins (part A). The correlations between lipase and colipase activities (part B) and between their mass and activity (part C) were also close to one. In contrast, only a weak or insignificant correlation was obtained between mass and activity of carboxyl ester lipase and phospholipase A*.

864

B . Sternby et al.

Relative s p e c i f i c activity 150 125 +

A. Correlation coefficients between protein masses (mg/l) LIP COL CEL PLA 1.00 0.90 0.85 0.94 LIP COL 0.90 1.00 0.80 0.96 CEL 0.85 0.80 1.00 0.75 PLA 0.94 0.96 0.75 1.00

100

S

a,

2

75

a Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 01/05/15 For personal use only.

50 25

0

Table I. Correlation coefficients for all immunoreactivity and enzymatic activity values (dilution factor included) for the four human lipolytic pancreatic proteins in duodenal contents*

1

I

25

50

75

100

1 5

minutes Fig. 4. Relative specific activity of the three lipases and colipase ( A ) , carboxyl ester lipase colipase. Lipase (a), (01, and phospholipase Az (0). Values are means f SD .

When diluted samples were superimposed on standard curves, no indication was obtained which supports the suspicion that the RIAs also measure molecular fragments.

DISCUSSION This is the first study in which the levels in normal duodenal contents during digestion of the three lipolytic enzymes lipase, carboxyl ester lipase, and phospholipase A2 and the lipase cofactor colipase have been determined both by measuring enzymatic activities and by measuring the immunoreactivity of the respective proteins. The relative specific activity (enzyme activity per milligram immunoreactive protein) could thereby be determined and compared with that of the respective pure proteins. As a general finding, the specific activity of the enzymes and of colipase was lower than those of the pure protein during most collection periods (Fig. 4). This may have several explanations. One is that colipase and phospholipase A2 are secreted in the pancreatic juice in the inactive but immunoreactive zymogen form and must be activated by trypsin in the lumen. In two of the volunteers the activity determined for phospholipase A 2 increased with

B. Correlation coefficients between protein activities (mmoI/min * I) LIP COL CEL PLA 1.00 0.91 0.53 0.51 LIP COL 0.91 1.00 0.58 0.59 CEL 0.53 0.58 1.00 0.10 PLA 0.51 0.59 0.10 1.00 C. Correlation coefficients between mass and activity (mg/I versus mmol/min- I) LIP COL CEL PLA 0.63 0.92 0.25 0.94

* LIP = lipase; COL = colipase; CEL = carboxyl ester lipase; PLA = phospholipase A*.

time in their first samples and reached steady state after 10 min, suggesting that in these cases a conversion of the proform to active enzyme may actually occur. The enzymatic colipase assay used is, however, independent of whether the protein is present in the procolipase o r colipase form when purified colipase is assayed but may be different when colipase is present in duodenal contents. Secondly, active enzyme molecules may be inactivated by surface denaturation. Pancreatic lipase is especially sensitive to surface inactivation when bound to its substrate (14), but carboxyl ester lipase is also reported to be inactivated in a similar manner (15). This possibility is difficult to evaluate but has to be considered particularly for lipase and carboxyl ester lipase when the test meal includes lipids. Thirdly, enzymes can be inactivated by proteoytic cleavage without losing all the immunologic determinants that are recognized in the RIA. This is most likely to occur in the case of lipase and carboxyl ester lipase, which are big and sensitive molecules compared with colipase and phospholipase A*. The latter two proteins are relatively small, rigid molecules containing five and seven disulfide bridges, re-

Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 01/05/15 For personal use only.

Radioimmunoassay versus Enzyme Activity

spectively. However, in none of the RIAs did we find indications that we determined protein fragments of antigens. Fourthly, the in vitro assays for the enzymes may not be completely specific in the estimation of the different lipases when present as a mixture in duodenal contents. Both in the lipase and colipase assays carboxyl ester lipase contributes to a small extent that is difficult to estimate. In addition, the possibility of contributions to lipolytic activity by gastric lipase cannot be excluded when using tributuryl glycerol as substrate (24). However, pH7.0 is not optimal for gastric lipase. Calculation of the relative specific activity of an enzyme requires an accurate determination of the specific activity (units per milligram protein) of the pure enzyme. This figure is influenced by the method used for determination of enzyme activity. The same method must therefore be used to assay the enzyme both when present in its pure form and when present in duodenal contents. For phospholipase A2 the conditions used deserve a special comment. The first and only time the specific activity of human pancreatic phospholipase A2 was determined it was reported to be 600 pmol/min.mg (13). The method used for its determination was a modification of the method of deHaas et al. (9) used for the phospholipase from other species. This modification includes a higher concentration of deoxycholate and a lower concentration of Ca++ and reports an increase with the human enzyme by a factor of around three compared with the original method (9). The method used by us (9) therefore was calculated to give a specific activity for the human enzyme of 200 pmol/min.mg. In this study the relative specific activity of lipase is higher than that of colipase, which may indicate an earlier underestimation of the specific activity of lipase. The enzymatic activity reached its maximum somewhat earlier than the mass concentration of the respective proteins (Fig. 2). The colipase concentration (both activity and mass) reached its maximum somewhat later than lipase, although our results give no support to the observation by Gaskin et al. (19) that colipase is secreted in a lower amount than lipase in healthy humans. An

865

explanation of the results of these authors is that they measured lipase at pH 8 in the presence of bile salt and colipase. This pH is not the optimal one for human lipase-colipase interaction or for interaction of human lipase-porcine colipase (10) since only about 65% of the activity observed at pH 7.0 is obtained. The relative specific activity is highest for lipase and colipase and lowest for carboxyl ester lipase-only around 50%. Carboxyl ester lipase was found in the highest amount (expressed as protein mass) of all the lipolytic enzymes in duodenal juice. This enzyme has a broad substrate specificity, but the specific activity towards different substrates in assays in vitro is low compared with lipase. In recent in vitro studies with the human enzyme and human lipase-colipase, gastric lipase, and phospholipase A2 a concerted action was shown (20,21). In other in vitro studies with chylomicrons labeled with arachidonic (22) or eicosapentaenoic acid (23) carboxyl ester lipase was found to be important for the hydrolysis of diglycerides containing these fatty acids. These findings, combined with the high concentration of carboxyl ester lipase found in human duodenal contents in the present study, suggest that this enzyme may have a more important role in acylglycerol digestion than has earlier been assumed. The large variation in data at any time point for the different subjects is most likely explained by individual variations in gastric emptying, pancreatic secretion, and dilution by the test meal. Measurements of digestive enzyme activities and masses in intestinal contents are of interest mainly for two reasons. It provides information on 1) the conditions of importance for an understanding of the basic mechanism of digestion, and 2) the functional capacity of the pancreas in secreting digestive enzymes in response to a test meal. In conclusion, determination of mass concentration of enzymesfproteins in duodenal contents is more accurate than enzymatic methods and enables comparison between patients and studies. For routine purposes wellstandardized enzymatic assays will, however, be adequate to reveal pancreatic insufficiency. This study provides information about the total amount of pancreatic lipolytic proteins in the

866

B. Sternby et al.

intestine in healthy humans. This information can be useful in the further development of treatment of fat maldigestion/absorption when traditional treatment does not give the expected result.

Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 01/05/15 For personal use only.

ACKNOWLEDGEMENTS This study was supported by grants from the Medical Research Council (3969), the Medical Faculty, University of Lund, and Albert Piihlssons Foundation, Malmo; the Crafoord Foundation, Lund; and h e Wibergs Foundation, Stockholm; Sweden. REFERENCES 1. Borgstrom B, Dahlquist A, Lundh G, Sjovall J. Studies of intestinal digestion and absorption in the human. J . Clin Invest 1957, 36, 1521-1536 2. Lundh G. Pancreatic exocrine function in neoplastic and inflammatory disease. A simple and reliable new test. Gastroenterology 1962, 42, 275-280 3. Borgstrom B, Hildebrand H. Lipase and colipase activities in human small intestinal contents after a liquid test meal. Scand J Gastroenterol 1975, 10, 585-591 4. Ihse I, Arnesjo B, Kugelberg C, Lilja P. Intestinal activities of trypsin, lipase and phospholipase after a test meal. Scand J Gastroenterol 1977, 12, 663668 5. Sternby B, Akerstrom B. Immunoreactive pancreatic- colipase, lipase and phospholipase A, in human plasma and urine from healthy individuals. Biochim Biophys Acta 1984, 789, 164-169 6 Aho HJ, Sternby B, Kallajoki H, Nevalainen T. Carboxyl ester lipase in human tissues and in acute pancreatitis. Int J Pancreatol 1989, 5, 123-134 7 Muller DPR, Ghale GK. Stability of pancreatic enzyme activities in duodenal juice after pancreatic stimulation by a test meal or exogenous hormones. Ann Clin Biochem 1982, 19, 89-93 8. Erlanson C. p-Nitrophenylacetate as a substrate for a carboxylester hydrolase in pancreatic juice and intestinal content. Scand J Gastroenterol 1970, 5, 333-336 9. de Haas GH, Postema NM, Nieuwenhuizen W, van Deenen LLM. Purification and properties of phospholipase A from porcine pancreas. Biochim Biophys Acta 1968, 159, 10?-117 10. Sternby B, Borgstrom B. Comparative studies on the ability of pancreatic colipases to restore activity of lipases from different species. Comp Biochem Physiol 1981, 68B, 15-18

Received 10 February 1990 Accepted 7 March 1991

11. Sternby B, Borgstrom B. Purification and characterization of human pancreatic colipase. Biochim Biophys Acta 1979, 572, 235-243 12. Lombard0 D, Guy 0, Figarella C. Purification and characterization of a carboxyl ester hydrolase from human pancreatic juice. Biochim Biophys Acta 1978, 527, 142-149 13. Grataroli R, De Car0 A , Guy 0,Armic J, Figarella C. Isolation and properties of prophospholipase A2 from human pancreatic juice. Biochemie 1981, 63, 677-684 14. Borgstrom B. The temperature-dependent interfacial inactivation of porcine pancreatic lipase. Effect of colipase and bile salts. Biochim Biophys Acta 1982, 712, 49G497 15. Tsujita T, Brockman HL. Regulation of carboxylester lipase adsorption to surfaces. I. Chemical specificity. Biochemistry 1987, 26, 8423-8429 16. Figarella C, Ribeiro T. The assay of human pancreatic phospholipase A in pancreatic juice and duodenal contents. Scand J Gastroenterol 1971, 6, 133-137 17. Ihse I, Arnesjo B. The phospholipase A, activity in human small intestinal contents. Acta Chem Scand 1973, 21,2749-2756 18. Grataroii R, Dijkman R , Dutilh CE, Van der Ouderaa F, de Haas GH, Figarella C. Studies on prophospholipase A, and its enzyme from human pancreatic juice. Catalytic properties and sequence of the N-terminal region. Eur J Biochem 1982, 122, 111-1 17 19. Gaskin KJ, Durie PR, Hill RE, Lee LM, Forstner GC. Colipase and maximally activated pancreatic lipase in normal subjects and patients with steatorrhea. J Clin Invest 1982, 69, 427-434 20. Lindstrom MB, Sternby B, Borgstrom B. Concerted action of human carboxyl ester lipase and pancreatic lipase during lipid digestion in vitro. Importance of the physicochemical state of the substrate. Biochem Biophys Acta 1988, 959, 178-184 21. Lindstrom MB, Persson J, Thurn L, Borgstrom B. Effect of pancreatic phospholipase A z and gastric lipase on the action of pancreatic carboxyl ester lipase against lipid substrates in vitro. Biochim Biophys Acta 1991 (in press) 22. Chen Q, Sternby B, Nilsson A. Hydrolysis of triacyglycerol arachidonic and linoleic acid ester bonds by human pancreatic lipase and carboxyl ester lipase. Biochim Biophys Acta 1989,1004,372385 23. Chen Q, Sternby B, Nilsson A. Effects of human pancreatic lipase-colipase and carboxyl ester lipase on eicosapentaenoic and arachidonic acid ester bonds of triacyglycerols rich in fish oil fatty acids. Biochim Biophys Acta 1990, 1044, 111-117 24. Gargouri Y, Pieroni G , Riviere C, et al. Kinetic assay of human gastric lipase on short- and longchain triacylglycerol emulsions. Gastroenterology 1986, 91, 919-925

Pancreatic lipolytic enzymes in human duodenal contents. Radioimmunoassay compared with enzyme activity.

The total pancreatic lipolytic capacity was determined in duodenal contents in healthy humans 10-120 min after a liquid test meal, by estimating the a...
548KB Sizes 0 Downloads 0 Views