GASTROENTEROLOGY
1991:100:189-195
How to Protect Human Pancreatic Enzyme Activities in Frozen Duodenal Juice DARLENE Division
G. KELLY, BERIT STERNBY,
of Gastroenterology,
Mayo Clinic
We determined whether activity of pancreatic enzymes could be maintained in frozen duodenal juice by diluting the specimens or by adding nutrients or a chymotrypsin inhibitor. Human duodenal juice was obtained during cholecystokinin octapeptide IV administration. Trypsin, chymotrypsin, lipolytic, lipase, and colipase activities were measured in fresh undiluted or diluted (1:4 and 1:16 with saline and T-tube bile) duodenal juice as well as after adding CaCl,, casein, triolein, or a chymotrypsin inhibitor. Subsequently, the samples were frozen at -2O”C, and enzyme activities were measured at 1,2,3,7,14, 28, and 56 days. Activities of chymotrypsin and colipase did not change during freezer storage. Trypsin survival was variable in juice from different subjects. By contrast, in duodenal juice to which no nutrient or only CaCl, had been added, 90%, 65%, and 40% (P = 0.05 vs. undiluted) of lipolytic activity was lost by 56 days in undiluted and 1:4 or 1:16 diluted duodenal juice samples, respectively. The loss of lipolytic activity was prevented (P < 0.05) by adding casein or casein and triolein to undiluted and 1:4 diluted samples and turkey egg white to undiluted samples. The loss of lipolytic activity was strongly associated with loss of lipase activity (r = 0.97) but only weakly associated with loss of colipase activity (r = 0.49). In summary, chymotrypsin and colipase are well preserved in frozen duodenal juice and can be used to accurately assess concentrations of pancreatic enzymes after thawing frozen duodenal samples. If it is necessary to measure lipolytic activity after freezing samples, lipase can be maintained by adding casein or a chymotrypsin inhibitor to juice before freezing.
he concentrations of pancreatic enzymes in duodenal juice are frequently used to assess the status of pancreatic enzyme secretion in clinical, physiological, and pathophysiologic studies. Because it is common practice to freeze samples before assay, loss of enzyme activity during freezer storage may have an
T
and EUGENE P. DMAGNO
and Foundation,
Rochester,
Minnesota
unrecognized impact on these studies. Loss of enzyme activity during freezer storage and mechanisms of loss of enzyme activity under these conditions have not been extensively investigated. Proteolytic activity (trypsin and chymotrypsin) is more stable than lipolytic activity in duodenal juice during in vitro incubation (1) and during aboral transit (2). However, it has been reported that 50% of trypsin activity may be lost in duodenal juice that has been frozen after collection following CCK and secretin stimulation (3). Previous investigators showed that lipase is stable in frozen postprandial duodenal samples (3,~) and in fasting (hormonally stimulated) duodenal samples to which a homogenized meal was added, but not in hormonally stimulated juice without food present (3). In a preliminary study we showed that pure nutrients (casein and triolein) added to CCK-stimulated duodenal juice before freezing improve survival of lipolytic activity for 1-3 months (3, whereas starch did not prevent the loss of lipolytic activity. Thus, currently available data suggest that pancreatic enzyme activities may be lost during freezer storage. Lipolytic activity in human duodenal juice results from the activities of four pancreatic proteins: lipase, colipase, carboxylester lipase (cholesterol esterase), and phospholipase A,. The substrate used in the assay system for quantitation of “lipase” activity determines which of these components are measured. In most assay systems, the majority of the lipolytic activity is the result of hydrolysis of triglycerides by the combined action of lipase and colipase. However, carboxylester lipase may also participate in the digestion of physiological fat mixtures in vitro (6). Phospholipase A, is also included in measurements of some assay systems as a result of its hydrolysis of phospholipids Abbreviations used in this paper: ATEE, N-acetyl+tyrosine ethyl ester; CCK-OP, cholecystokinin octapeptide; TAME, p-tosyl-Larginine methyl ester HCl; TEW, turkey egg white. o 1991by the American Gastroenterological Association 0018-5085/91/$3.00
GASTROENTEROLOGYVol. 100, No. 1
190 KELLYET AL.
in the substrate (7). Thus, normal levels of enzyme activity must be determined for the specific system used by each diagnostic laboratory. In our system, lipolytic activity is measured by using Lipomul (Upjohn, Kalamazoo, MI) as the substrate; in this assay, lipolytic activity is composed of lipase, colipase, phospholipase A,, and carboxylester lipase activities. The loss of lipolytic activity in frozen duodenal juice may occur secondary to proteolytic digestion, because pure human lipolytic proteins are stable during freezer storage and after repeated thawing and refreezing (Sternby, unpublished data). Support for this hypothesis is derived from previous studies in which we have found that lipolytic activity in human duodenal juice decreases during incubation at 3 7°C in vitro (1)and during aboral transit in the human small bowel (2). The major factor affecting survival of lipolytic activity in these situations is the level of chymotryptic activity, as demonstrated by acceleration of the loss of lipolytic activity by the addition of exogenous chymotrypsin and by enhanced survival Similar when a chymotrypsin inhibitor is added (1,8). manipulation of tryptic activity does not alter lipolytic activity. In addition, incubation studies have shown that other proteases such as trypsin, elastase, bromelain, papain, thermolysin, and subtilisin are not capable of digesting the porcine lipase molecule (9). Chymotrypsin cleaves the native lipase molecule at or near its enzymatic locus (9). We have shown that the addition of casein or triolein to human duodenal juice during incubation in vitro prevents loss of lipolytic activity, whereas addition of starch does not (10). The aims of the present study were to determine if pancreatic protease and lipolytic activities are lost during freezer storage and if the rate of the loss of enzyme activity is affected by dilution, chymotryptic activity, CaCl, (which stabilizes trypsin and increases its proteolytic activity) (ll),and nutrients. Furthermore, because lipolytic activity is likely very fragile and is composed of several components, we determined if loss of lipolytic activity was caused by loss of the activity of lipase, colipase, or both.
Materials and Methods Experimental
Design
Duodenal samples were obtained from four healthy humans who were participating in an intestinal infusion study. This study was approved by our Institutional Review Board, and signed consent was obtained from all subjects. Each subject was intubated with a multilumen oroduodenal tube and received IV CCK-OP (60 ng . kg-’ . h-‘). The duodenum was infused with saline at the level of the ampulla of Vater, and duodenal juice was obtained from an aspiration port located 20 cm distally. All samples were collected on ice.
The duodenal sample from each subject was handled separately. Duodenal juice was used undiluted or diluted 1:4 and 1:16with a mixture of physiological saline and T-tube bile to maintain similar bile acid concentrations in all dilutions of juice obtained from a patient. Dilution of duodenal samples was performed by measuring the total bile acid concentration (12) of the undiluted duodenal juice and the T-tube bile (which was devoid of pancreatic enzyme activity) and subsequently calculating the amounts of saline and T-tube bile necessary to achieve the desired pancreatic enzyme concentration while maintaining the bile concentration at the original level. Eight replicate samples were prepared for each of the dilutions (undiluted, 1:4, and 1:16) and additives (total of 144 aliquots per subject). To l-mL aliquots of the undiluted and diluted juice we added no nutrients, casein (33.5 mg/mL), triolein (14.8 mg/mL), casein (16.8 mg/mL) and triolein (7.4 mg/mL), chymotrypsin inhibitor (turkey egg white [TEW]; 5 mg/mL), or CaCl, (0.5 mg/mL) .
Freezer Storage Samples were frozen at -20°C. Only one sample from each of the sets of the eight replicate samples (sets included each of the five additives and the control for undiluted samples and those diluted l:4 and 1:16)was thawed after 1, 2, 3, 7, 14,18,or 56 days of storage. After each sample was thawed and assayed, it was discarded. Thus, no sample was thawed, refrozen, and then used in a subsequent analysis.
Assays Samples were assayed for trypsin, chymotrypsin, lipolytic activity, lipase, and colipase. All enzymes were measured by automatic titration (pH Stat; Radiometer, Copenhagen, Denmark). Trypsin was assayed at pH 8.0and 37°C and chymotrypsin was assayed at pH 7.8 and 37°C with the substrate p-tosyl-L-arginine methyl ester HCl (TAME) and N-acetyl-Ltyrosine ethyl ester (ATEE), respectively (13). The assay system for lipolytic activity, lipase, and colipase consisted of 4 mL containing final concentrations of 10% Lipomul, 154 mmol/L NaCl, and 70 mmol/L Na cholate. This solution was maintained at 37°C at a pH of 7.4 (13). In preliminary experiments we ascertained that in this assay no lipolytic activity is detected in the presence of only pure lipase (i.e., free of colipase) or pure colipase (i.e., free of lipase). To measure lipolytic, lipase, and colipase activity, we performed two assays on each sample. During each procedure, the assay was used first to measure total lipolytic activity of duodenal juice (Figure 1A).To measure total lipase activity, an excess of colipase (20 pmol) was added to the system and titration was continued (14). In this case, lack of change of the titration rate indicated that the lipolytic activity and the lipase activity were identical and that there was adequate colipase in the duodenal juice for the quantity of lipase present (Figure 1B). However, if the titration rate accelerated following addition of surplus colipase, the quantity of lipase in the sample exceeded that
PANCREATIC ENZYMES IN FROZEN DUODENAL JUICE
January 1991
B
A No added
colipase
Added
C colipase
Added
colipase
191
Results Trypsin and Chymotrypsin Activity
Time
Time
Time
Figure 1. In our system, lipolytic activity is calculated as FEq NaOH required per minute to maintain the pH constant at 7.4. A. Lipolytic activity in duodenal
juice as measured by our assay (see text for details). Lipolytic activity in our assay consists of the concentration of lipase and colipase when present in equivalent amounts and small amounts of phospholipase A, and carboxylester lipase. B. A situation
in which lipase activity and colipase activity are identical (e.g., no change in the slope of the lipolytic activity titration curve after the addition of colipase, and therefore no excess lipase activity present in the samples).
C. Addition of colipase accelerates the titration curve, indicating that the sample contains lipase activity in excess of colipase activity. The lipase activity in this sample is calculated from the increased slope of lipolytic titration curve after the addition of colipase.
of lipolytic activity and the amount of lipase was calculated from the accelerated rate after the addition of colipase (Figure 1C). Likewise, to quantitate colipase activity, first lipolytic activity was measured. Then excess pure lipase (20 pmol) was added to the system and the titration was performed as described above. Total bile salt concentrations were measured using an enzymatic method ( 12).
In the undiluted (Figure 2) and diluted samples (data not shown), the mean activities of the proteases remained stable over the 56 days of freezer storage. Trypsin, however, decreased by 25, 20, 25, and 0% over 56 days in the four subjects. Chymotrypsin, by contrast, was quite stable since more than 92% of the original activity was present after 56 days of freezer storage in all four subjects’ juice. The absence or the addition of nutrient or CaCl, did not affect trypsin and chymotrypsin activity during freezer storage, because the areas under the curve (Figure 3, last two groups of columns) for these enzymes were similar among all the test solutions except that the chymotrypsin inhibitor, as expected, significantly reduced (80%) chymotrypsin activity.
Lipoly-tic Activity and Lipase
In undiluted samples containing no additives (control), lipolytic activity (Figure 4, top) rapidly decreased to 25% of original activity within 2 weeks of storage at -20°C. Thereafter, the rate of decrease slowed so that approximately 10% of the original activity remained after 8 weeks. The trend was similar in l:4 and 1:16 dilutions, although the magnitude of loss of lipolytic activity was less in these diluted
Data Analysis During this study we found that addition of nutrients, TEW and CaCl, did not affect enzyme activities in fresh duodenal juice except that TEW inhibited chymotrypsin activity. Thus, at each time (O-56 days) enzyme or coenzyme activities (units . ml-’ . min-‘) were expressed as the percent of initial activity (before addition of nutrient). Survival of activity over time was calculated as area under the curve using the trapezoid rule and expressed as the percent of maximal area under the curve (i.e., complete preservation over 56 days). Statistical comparisons were made on proportions of maximal area under the curve after transforming the data to arcsin in order to avoid the effects of unequal variability at the extremes (near 0 and 1) (15). Analysis of variance was used to test whether the means of the arcsins over all of the treatments of duodenal juice were significantly different. Dunnett’s procedure for multiple comparisons was used to test whether the survival of enzymes for each of the treatments was significantly different from the survival of enzyme activity in native untreated duodenal juice (control). Regression analysis was used to determine the relationship of lipolytic activity to lipase and colipase activities.
40
. ..-......................**._.._.._.._..
20 0
10
20
30
40
50
56
Days in freezer Figure 2. Effects of nutrients and chymotrypsin inhibitor (TEW) on survival of trypsin (top) and chymotrypsin (bottom) during storage of undiluted human duodenal juice at -20°C for l-56 days. Chymotrypsin is stable during freezer storage and is not significantly affected by the addition of nutrients. Trypsin, by contrast, is stable in juice from some subjects but somewhat labile in others. The lower panel also shows the effect of TEW on chymotrypsin activity (80% inhibited).
GASTROENTEROLOGY
192 KELLYETAL.
tZZi None 0 Casein Triolein
150
& Q, 2
m Cas + tri m CaCI, R§I TEW
*vs. none PcO.05 ANOVA
;
Vol. 100. No. 1
100
Lipolytic activity
Lipase
Colipase
Trypsin
Chymotrypsin
Figure 3. Effects of nutrients and chymotrypsin inhibitor (TEW) on survival of pancreatic enzyme activity storage in undiluted human duodenal juice (top) and duodenal juice diluted 1:4 (middle) and 1:16(bottom) during 56 days of freezer storage. The last group of bars (chymotrypsin activity) does not include values for TEW because addition of this chymotrypsin inhibitor abolished chymotrypsin activity (Figure 2). In undiluted juice, casein, casein plus triolein, and chymotrypsin inhibitor significantly preserved lipolytic activity and lipase compared with juice frozen without additives (P < 0.05, ANOVA). In juice diluted 1:4,casein and casein plus triolein significantly improved survival of lipolytic activity. These additives as well as chymotrypsin inhibitor significantly enhanced the none of the additives altered survival of enzymatic activity compared with no nutrient. preservation of lipase. In juice diluted 1:16,
samples; after 8 weeks 65% and 40% of lipolytic activity was lost in the 1:~ and 1:16 diluted samples, respectively (1:lvs.1:16, P = 0.05; 1:l vs. 1:4, NS). In both control and treated samples loss of lipolytic activity closely paralleled the loss of lipase activity (Figure 4, middle), whereas colipase was well preserved during freezing (Figure 4, bottom). The addition of nutrients and TEW did not affect the activities of the lipolytic enzymes before freezing (data not shown). During freezer storage, the presence of casein or casein and triolein in the undiluted and 1:4diluted samples significantly decreased the loss of lipolytic activity and lipase (P < 0.05, ANOVA, Figure 3). Turkey egg white also significantly decreased the loss of lipolytic activity and lipase in undiluted juice. In the 1:16 diluted samples none of the additives significantly affected survival of lipolytic activity. CaCl, and triolein did not significantly alter the loss of lipolytic activity or lipase, although with added triolein the survival of enzyme was intermediate between samples containing casein and those with no nutrient.
Relationships Colipase
of Lipase Activity to Lipase and
The survival of lipolytic activity expressed as percentage remaining correlated strongly with survival of lipase (r = 0.97; P < 0.02; Figure 5, top) but only weakly with the survival of colipase (r = 0.49; P < 0.05; Figure 5, bottom). The relationship between lipolytic activity and lipase suggests that loss of lipolytic activity is caused by loss of lipase, whereas the relationship between lipolytic activity and colipase indicates that the loss of lipolytic activity greatly exceeded the loss of colipase.
Discussion During freezer storage of duodenal juice obtained after CCK-OP stimulation, there was maintenance of chymotrypsin and colipase activities, variable loss of trypsin activity, and pronounced loss of lipolytic activity and lipase over 8 weeks. Survival of lipolytic activity and lipase during storage, however, was improved by diluting the samples 1:16 or by the
January 1991
PANCREATIC ENZYMESIN FROZENDUODENAL JUICE
120 l- Lipolytic activity
'F; 2
0
I
I
I
I
I
I
10
20
30
40
50
56
Days in freezer Figure 4. Effect of nutrients and chymotrypsin inhibitor (TEW) on survival of lipolytic activity (top), lipase (middle), and colipase (bottom) during storage of undiluted human duodenal juice at -20°C for l-56 days. In aliquots frozen after the addition of casein, casein plus triolein or chymotrypsin inhibitor, lipolytic activity, and lipase were almost totally preserved compared with those frozen with no additives or with CaCI,. Triolein alone was only partially protective. By contrast, colipase was stable during freezing regardless of additives.
addition of chymotrypsin inhibitor or of casein (alone or in combination with triolein) before freezing. In contrast, triolein alone and CaCl, had no effect on survival of lipolytic activity or lipase. Lipase was the major component of lipolytic activity which was lost during freezing as was indicated by the close correlation (r = 0.97) between lipolytic activity and lipase remaining. Colipase, which was almost completely preserved during freezer storage, was only weakly correlated with loss of lipolytic activity. The loss of up to 50% of tryptic activity in frozen CCK and secretin stimulated duodenal juice reported by others (3) was not observed in the present study. In that study (3) three samples decreased tryptic activity by more than 40% over 15-30 days, and a fourth specimen lost 30% activity after 50 days. The remaining five specimens lost less than 20% of activity after 15-40 days of storage. In contrast, the juice from our four subjects lost 25, 20, 15, and 0% of activity after 56 days of storage. After dilution 1:4,two samples lost no activity, and after 1:16 dilution all samples had 85%-100% of activity after 56 days of storage. Thus,
193
although it appears that trypsin may be stable to freezing in duodenal juice of some patients, there may be enough loss to prohibit its routine use in frozen samples. The reason for this variable loss of trypsin activity is unclear. Chymotrypsin, by contrast, was very stable with more than 92% of activity remaining after 56 days of freezer storage in undiluted samples. There was slightly more lability of chymotrypsin in the 1:4 dilution, but even at this concentration, 85% or more of activity remained after 56 days. In 1:16 dilutions, two samples lost 15%-20% of activity. A third lost about 10% of activity and a fourth was completely stable. These findings suggest that chymotryptic activity may be a better indicator than trypsin of actual activity of pancreatic enzymes if samples must be frozen before analysis. The irreversible loss of lipolytic activity and lipase in duodenal juice which occurs during freezer storage has been described previously (4,5) and is similar to that observed intraluminally (2) and during incubation at 37°C (1). Loss of activity seems to be the result of chymotryptic digestion, as the addition of the chymotrypsin inhibitor TEW blocks the loss of lipolytic activity and lipase during freezing just as it does in vivo (8) and during incubation (1). It has been shown
160
y=8.8+0.81x
120 - r=0.97 80
Lipase, % remaining
J 0
20
40
80
80
loo
120 140
160
Colipase, % remaining Figure 5. Relationships of percent of lipolytic activity remaining to percent of lipase activity remaining (top) and to percent of colipase remaining (bottom) after 56 days of freezer storage. The strong correlation between lipolytic activity and lipase suggests that lipase is the major factor destroyed during freezer storage. The slight deviation of the best fit line from the identity line could be the result of unmeasured components (e.g., phospholipase A, and carboxylester lipase) which may have also been destroyed during freezer storage.
194
GASTROENTEROLOGY
KELLY ET AL.
that chymotrypsin, but not other proteases, digests porcine lipase near its active site (9). Casein protected lipase during freezing, an effect similar to that observed when pureed chicken was added to duodenal juice before freezing (3). It is likely that these proteins provide an alternate substrate for chymotrypsin, replacing lipase in that role. Although triolein did not significantly improve survival of lipase, the lipase activity of samples containing triolein was intermediate between that of no nutrient and of casein, casein plus triolein, or TEW. Previous studies had shown that corn oil added to duodenal juice protected lipase during freezing (3). The differences between this observation and our study may be explained by the presence of several lipids in corn oil (24% oleic acid, 58% linoleic acid, and 11% stearic acid) which may protect lipase better than triolein alone. Although triglyceride provides an anchor for the lipase-colipase complex, this effect of triglyceride only partially protects lipase as surface denaturation of lipase may occur in the presence of triglyceride (16). The correlation between lipolytic activity and lipase suggests that digestion of the lipase is primarily responsible for the decline in lipolytic activity. In addition, this correlation suggests that lipase is the major component which is measured in the assay system using Lipomul. The lack of perfect correlation between lipolytic activity and lipase may, in part, be caused by phospholipase A, and carboxylester lipase which were not measured separately in this study. While these latter enzymes may also be destroyed during freezer storage, they are responsible for a small amount of lipolytic activity in duodenal juice as measured by our assay system because lipase activity accounted for nearly all of the total lipolytic activity in the samples. Colipase was stable during freezing. This cofactor is required for optimal activity and stability of lipase (14). Colipase in pancreatic extract is not inactivated by heating (17) but trypsin and bile may affect the activity of colipase. The incubation of trypsin with pure rat colipase (trypsin:colipase specific activities = 6.5:1) causes inactivation of colipase which is minimized by the addition of rat bile (18). The ratio of trypsin and colipase in the study of Lairon et al. (18), however, was much higher than in the normal human juice used in the present study (approximately 1:l). The low trypsin-colipase ratio and bile in typical intraluminal concentrations in our study may have protected the colipase in frozen duodenal juice. The dilution of duodenal juice used in the present study was intended to mimic pancreatic insufficiency. However, in pancreatic insufficiency, protease levels are relatively higher than lipase levels (19). Therefore, the magnitude of the loss of lipolytic activity which might be expected in juice from pa-
Vol. 100, No. 1
tients with pancreatic insufficiency would in all likelihood be greater than we demonstrated because the relative amounts of chymotrypsin-lipase would be greater in pancreatic insufficiency. Our observations are particularly important if duodenal juice from patients with suspected pancreatic insufficiency is frozen before assay. Samples which are obtained during hormonal stimulation will lose lipolytic activity as a result of storage which may lead to an erroneous diagnosis of pancreatic insufficiency, especially if samples are stored for more than a few days. Therefore, either lipolytic activity in duodenal juice obtained during CCK-OP stimulation should be assayed immediately, or if juice is frozen, casein (33 I5 mg/mL as a final concentration) or TEW (5 mg/mL as a final concentration) should be added to protect lipolytic activity (lipase) from chymotryptic digestion. Alternatively, colipase or chymotrypsin should be measured in previously frozen samples because they are stable components of pancreatic secretion. References 1 Thiruvengadam R, DiMagno EP. Inactivation of human lipase by proteases. Am J Physiol 1988;255:G476-G481, 2 Layer P, Go VLW, DiMagno EP. Fate of pancreatic enzymes during small intestinal aboral transit in humans. Am J Physiol 1986;251:G475-G480. enzyme activi3. Muller DPR, Ghale GK. Stability of pancreatic ties in duodenal juice after pancreatic stimulation by a test meal or exogenous hormones. Ann Clin Biochem 1982;19:8993. 4. Legg EF, Spencer AM. Studies on the stability of pancreatic enzymes in duodenal fluid to storage temperature and pH. Clin Chim Acta 1975;65:175-179. 5. Sternby B, Kelly DG, DiMagno EP. How do fat and protein preserve lipolytic, lipase (LIP) and colipase (COL) activity in frozen human duodenal samples? Physiologic and clinical implications (abstr). Pancreas 1988;3:619. 6. Lindstrom MB, Sternby B, Borgstrom B. Concerted action of human carboxylester lipase and pancreatic lipase during lipid digestion in vitro: importance of the physiochemical state of the substrate. Biochim Biophys Acta 1988;959:178-184. 7. Borgstrom B. Importance of phospholipids, pancreatic phospholipase A, and fatty acid for the digestion of dietary fat. In vitro experiments with porcine enzymes. Gastroenterology 1980;78: 954-962. 8. Thiruvengadam R, Zinsmeister AR, DiMagno EP. Can lipase activity be preserved during aboral intestinal transit in humans? (abstr). Gastroenterology 1986:90:1663. 9. Bousset-Risso M, Bonicel J, Rovery M. Limited proteolysis of porcine pancreatic lipase. Lability of the Phe 335-Ala 336 bond towards chymotrypsin. FEBS Lett 1985:182:323-326. 10. Kelly DG, Bentley KJ, Sandberg RJ, Zinsmeister AR, DiMagno EP. Do nutrients and bile in human duodenal juice affect the survival of lipase activity? Possible clinical implications (abstr). Gastroenterology 1988;94:A222. 11. Sipos T, Merkel JR. An effect of calcium ions on the activity, heat stability, and structure of trypsin. Biochemistry 1979;9: 2766-2775. 12. Talay P. Enzymatic analysis of steroid hormones. Methods Biochem Anal 1960;8:119-143. 13. Pelot D, Grossman MI. Distribution and fate of pancreatic
January
14. 15.
16.
17. 18.
19.
1991
enzymes in small intestine of the rat. Am J Physiol 1962;202: 285-288. Borgstrom B, Erlanson C. Pancreatic juice colipase: physiological importance. Biochim Biophys Acta 1971;242:509-513. Fleiss JL. The parallel groups design. In: The design and analysis of clinical experiments. New York: Wiley, 1986:4690. Borgstrom B. The temperature-dependent interfacial inactivation of porcine pancreatic lipase. Effect of colipase and bile salts. Biochim Biophys Acta 1982;712:490-497, Rosenheim 0. On pancreatic lipase. III. The separation of lipase from its co-enzyme. J Physiol1910;4O:xiv-xvi. Lairon D, Nalborne G, LaFont H, Domingo N, Hauton JC. Protective effect of biliary lipids on rat pancreatic lipase and colipase. Lipids 1977;13:211-216. DiMagno EP, Malagelada JR, Go VLW. Relationship between
PANCREATIC
ENZYMES
alcoholism and 1975;252:200-207.
IN FROZEN
pancreatic
DUODENAL
insufficiency.
Ann
JUICE
195
NY Acad
Sci
Received December 12, 1989. Accepted July 16, 1990. Address requests for reprints to: Eugene P. DiMagno, M.D., Tower Gastrointestinal Diagnostic Unit, Mayo Clinic, Rochester, Minnesota 55905. Supported by National Institutes of Health fellowship DK07198 (to D.K.); Henning and Johan Throne Holst Foundation, Stockholm, Albert Pahlssons Foundation, Malmo, and the Swedish Medical Association, Stockholm, Sweden (to B.S.); Clinical Research Center grant RR585; and the Mayo Foundation. The authors thank Karen J. Bentley, David Katzmann, and Richard Tucker for skillful technical assistance and Linda Bakken for preparing the manuscript.
1991:100:189-195
How to Protect Human Pancreatic Enzyme Activities in Frozen Duodenal Juice DARLENE Division
G. KELLY, BERIT STERNBY,
of Gastroenterology,
Mayo Clinic
We determined whether activity of pancreatic enzymes could be maintained in frozen duodenal juice by diluting the specimens or by adding nutrients or a chymotrypsin inhibitor. Human duodenal juice was obtained during cholecystokinin octapeptide IV administration. Trypsin, chymotrypsin, lipolytic, lipase, and colipase activities were measured in fresh undiluted or diluted (1:4 and 1:16 with saline and T-tube bile) duodenal juice as well as after adding CaCl,, casein, triolein, or a chymotrypsin inhibitor. Subsequently, the samples were frozen at -2O”C, and enzyme activities were measured at 1,2,3,7,14, 28, and 56 days. Activities of chymotrypsin and colipase did not change during freezer storage. Trypsin survival was variable in juice from different subjects. By contrast, in duodenal juice to which no nutrient or only CaCl, had been added, 90%, 65%, and 40% (P = 0.05 vs. undiluted) of lipolytic activity was lost by 56 days in undiluted and 1:4 or 1:16 diluted duodenal juice samples, respectively. The loss of lipolytic activity was prevented (P < 0.05) by adding casein or casein and triolein to undiluted and 1:4 diluted samples and turkey egg white to undiluted samples. The loss of lipolytic activity was strongly associated with loss of lipase activity (r = 0.97) but only weakly associated with loss of colipase activity (r = 0.49). In summary, chymotrypsin and colipase are well preserved in frozen duodenal juice and can be used to accurately assess concentrations of pancreatic enzymes after thawing frozen duodenal samples. If it is necessary to measure lipolytic activity after freezing samples, lipase can be maintained by adding casein or a chymotrypsin inhibitor to juice before freezing.
he concentrations of pancreatic enzymes in duodenal juice are frequently used to assess the status of pancreatic enzyme secretion in clinical, physiological, and pathophysiologic studies. Because it is common practice to freeze samples before assay, loss of enzyme activity during freezer storage may have an
T
and EUGENE P. DMAGNO
and Foundation,
Rochester,
Minnesota
unrecognized impact on these studies. Loss of enzyme activity during freezer storage and mechanisms of loss of enzyme activity under these conditions have not been extensively investigated. Proteolytic activity (trypsin and chymotrypsin) is more stable than lipolytic activity in duodenal juice during in vitro incubation (1) and during aboral transit (2). However, it has been reported that 50% of trypsin activity may be lost in duodenal juice that has been frozen after collection following CCK and secretin stimulation (3). Previous investigators showed that lipase is stable in frozen postprandial duodenal samples (3,~) and in fasting (hormonally stimulated) duodenal samples to which a homogenized meal was added, but not in hormonally stimulated juice without food present (3). In a preliminary study we showed that pure nutrients (casein and triolein) added to CCK-stimulated duodenal juice before freezing improve survival of lipolytic activity for 1-3 months (3, whereas starch did not prevent the loss of lipolytic activity. Thus, currently available data suggest that pancreatic enzyme activities may be lost during freezer storage. Lipolytic activity in human duodenal juice results from the activities of four pancreatic proteins: lipase, colipase, carboxylester lipase (cholesterol esterase), and phospholipase A,. The substrate used in the assay system for quantitation of “lipase” activity determines which of these components are measured. In most assay systems, the majority of the lipolytic activity is the result of hydrolysis of triglycerides by the combined action of lipase and colipase. However, carboxylester lipase may also participate in the digestion of physiological fat mixtures in vitro (6). Phospholipase A, is also included in measurements of some assay systems as a result of its hydrolysis of phospholipids Abbreviations used in this paper: ATEE, N-acetyl+tyrosine ethyl ester; CCK-OP, cholecystokinin octapeptide; TAME, p-tosyl-Larginine methyl ester HCl; TEW, turkey egg white. o 1991by the American Gastroenterological Association 0018-5085/91/$3.00
GASTROENTEROLOGYVol. 100, No. 1
190 KELLYET AL.
in the substrate (7). Thus, normal levels of enzyme activity must be determined for the specific system used by each diagnostic laboratory. In our system, lipolytic activity is measured by using Lipomul (Upjohn, Kalamazoo, MI) as the substrate; in this assay, lipolytic activity is composed of lipase, colipase, phospholipase A,, and carboxylester lipase activities. The loss of lipolytic activity in frozen duodenal juice may occur secondary to proteolytic digestion, because pure human lipolytic proteins are stable during freezer storage and after repeated thawing and refreezing (Sternby, unpublished data). Support for this hypothesis is derived from previous studies in which we have found that lipolytic activity in human duodenal juice decreases during incubation at 3 7°C in vitro (1)and during aboral transit in the human small bowel (2). The major factor affecting survival of lipolytic activity in these situations is the level of chymotryptic activity, as demonstrated by acceleration of the loss of lipolytic activity by the addition of exogenous chymotrypsin and by enhanced survival Similar when a chymotrypsin inhibitor is added (1,8). manipulation of tryptic activity does not alter lipolytic activity. In addition, incubation studies have shown that other proteases such as trypsin, elastase, bromelain, papain, thermolysin, and subtilisin are not capable of digesting the porcine lipase molecule (9). Chymotrypsin cleaves the native lipase molecule at or near its enzymatic locus (9). We have shown that the addition of casein or triolein to human duodenal juice during incubation in vitro prevents loss of lipolytic activity, whereas addition of starch does not (10). The aims of the present study were to determine if pancreatic protease and lipolytic activities are lost during freezer storage and if the rate of the loss of enzyme activity is affected by dilution, chymotryptic activity, CaCl, (which stabilizes trypsin and increases its proteolytic activity) (ll),and nutrients. Furthermore, because lipolytic activity is likely very fragile and is composed of several components, we determined if loss of lipolytic activity was caused by loss of the activity of lipase, colipase, or both.
Materials and Methods Experimental
Design
Duodenal samples were obtained from four healthy humans who were participating in an intestinal infusion study. This study was approved by our Institutional Review Board, and signed consent was obtained from all subjects. Each subject was intubated with a multilumen oroduodenal tube and received IV CCK-OP (60 ng . kg-’ . h-‘). The duodenum was infused with saline at the level of the ampulla of Vater, and duodenal juice was obtained from an aspiration port located 20 cm distally. All samples were collected on ice.
The duodenal sample from each subject was handled separately. Duodenal juice was used undiluted or diluted 1:4 and 1:16with a mixture of physiological saline and T-tube bile to maintain similar bile acid concentrations in all dilutions of juice obtained from a patient. Dilution of duodenal samples was performed by measuring the total bile acid concentration (12) of the undiluted duodenal juice and the T-tube bile (which was devoid of pancreatic enzyme activity) and subsequently calculating the amounts of saline and T-tube bile necessary to achieve the desired pancreatic enzyme concentration while maintaining the bile concentration at the original level. Eight replicate samples were prepared for each of the dilutions (undiluted, 1:4, and 1:16) and additives (total of 144 aliquots per subject). To l-mL aliquots of the undiluted and diluted juice we added no nutrients, casein (33.5 mg/mL), triolein (14.8 mg/mL), casein (16.8 mg/mL) and triolein (7.4 mg/mL), chymotrypsin inhibitor (turkey egg white [TEW]; 5 mg/mL), or CaCl, (0.5 mg/mL) .
Freezer Storage Samples were frozen at -20°C. Only one sample from each of the sets of the eight replicate samples (sets included each of the five additives and the control for undiluted samples and those diluted l:4 and 1:16)was thawed after 1, 2, 3, 7, 14,18,or 56 days of storage. After each sample was thawed and assayed, it was discarded. Thus, no sample was thawed, refrozen, and then used in a subsequent analysis.
Assays Samples were assayed for trypsin, chymotrypsin, lipolytic activity, lipase, and colipase. All enzymes were measured by automatic titration (pH Stat; Radiometer, Copenhagen, Denmark). Trypsin was assayed at pH 8.0and 37°C and chymotrypsin was assayed at pH 7.8 and 37°C with the substrate p-tosyl-L-arginine methyl ester HCl (TAME) and N-acetyl-Ltyrosine ethyl ester (ATEE), respectively (13). The assay system for lipolytic activity, lipase, and colipase consisted of 4 mL containing final concentrations of 10% Lipomul, 154 mmol/L NaCl, and 70 mmol/L Na cholate. This solution was maintained at 37°C at a pH of 7.4 (13). In preliminary experiments we ascertained that in this assay no lipolytic activity is detected in the presence of only pure lipase (i.e., free of colipase) or pure colipase (i.e., free of lipase). To measure lipolytic, lipase, and colipase activity, we performed two assays on each sample. During each procedure, the assay was used first to measure total lipolytic activity of duodenal juice (Figure 1A).To measure total lipase activity, an excess of colipase (20 pmol) was added to the system and titration was continued (14). In this case, lack of change of the titration rate indicated that the lipolytic activity and the lipase activity were identical and that there was adequate colipase in the duodenal juice for the quantity of lipase present (Figure 1B). However, if the titration rate accelerated following addition of surplus colipase, the quantity of lipase in the sample exceeded that
PANCREATIC ENZYMES IN FROZEN DUODENAL JUICE
January 1991
B
A No added
colipase
Added
C colipase
Added
colipase
191
Results Trypsin and Chymotrypsin Activity
Time
Time
Time
Figure 1. In our system, lipolytic activity is calculated as FEq NaOH required per minute to maintain the pH constant at 7.4. A. Lipolytic activity in duodenal
juice as measured by our assay (see text for details). Lipolytic activity in our assay consists of the concentration of lipase and colipase when present in equivalent amounts and small amounts of phospholipase A, and carboxylester lipase. B. A situation
in which lipase activity and colipase activity are identical (e.g., no change in the slope of the lipolytic activity titration curve after the addition of colipase, and therefore no excess lipase activity present in the samples).
C. Addition of colipase accelerates the titration curve, indicating that the sample contains lipase activity in excess of colipase activity. The lipase activity in this sample is calculated from the increased slope of lipolytic titration curve after the addition of colipase.
of lipolytic activity and the amount of lipase was calculated from the accelerated rate after the addition of colipase (Figure 1C). Likewise, to quantitate colipase activity, first lipolytic activity was measured. Then excess pure lipase (20 pmol) was added to the system and the titration was performed as described above. Total bile salt concentrations were measured using an enzymatic method ( 12).
In the undiluted (Figure 2) and diluted samples (data not shown), the mean activities of the proteases remained stable over the 56 days of freezer storage. Trypsin, however, decreased by 25, 20, 25, and 0% over 56 days in the four subjects. Chymotrypsin, by contrast, was quite stable since more than 92% of the original activity was present after 56 days of freezer storage in all four subjects’ juice. The absence or the addition of nutrient or CaCl, did not affect trypsin and chymotrypsin activity during freezer storage, because the areas under the curve (Figure 3, last two groups of columns) for these enzymes were similar among all the test solutions except that the chymotrypsin inhibitor, as expected, significantly reduced (80%) chymotrypsin activity.
Lipoly-tic Activity and Lipase
In undiluted samples containing no additives (control), lipolytic activity (Figure 4, top) rapidly decreased to 25% of original activity within 2 weeks of storage at -20°C. Thereafter, the rate of decrease slowed so that approximately 10% of the original activity remained after 8 weeks. The trend was similar in l:4 and 1:16 dilutions, although the magnitude of loss of lipolytic activity was less in these diluted
Data Analysis During this study we found that addition of nutrients, TEW and CaCl, did not affect enzyme activities in fresh duodenal juice except that TEW inhibited chymotrypsin activity. Thus, at each time (O-56 days) enzyme or coenzyme activities (units . ml-’ . min-‘) were expressed as the percent of initial activity (before addition of nutrient). Survival of activity over time was calculated as area under the curve using the trapezoid rule and expressed as the percent of maximal area under the curve (i.e., complete preservation over 56 days). Statistical comparisons were made on proportions of maximal area under the curve after transforming the data to arcsin in order to avoid the effects of unequal variability at the extremes (near 0 and 1) (15). Analysis of variance was used to test whether the means of the arcsins over all of the treatments of duodenal juice were significantly different. Dunnett’s procedure for multiple comparisons was used to test whether the survival of enzymes for each of the treatments was significantly different from the survival of enzyme activity in native untreated duodenal juice (control). Regression analysis was used to determine the relationship of lipolytic activity to lipase and colipase activities.
40
. ..-......................**._.._.._.._..
20 0
10
20
30
40
50
56
Days in freezer Figure 2. Effects of nutrients and chymotrypsin inhibitor (TEW) on survival of trypsin (top) and chymotrypsin (bottom) during storage of undiluted human duodenal juice at -20°C for l-56 days. Chymotrypsin is stable during freezer storage and is not significantly affected by the addition of nutrients. Trypsin, by contrast, is stable in juice from some subjects but somewhat labile in others. The lower panel also shows the effect of TEW on chymotrypsin activity (80% inhibited).
GASTROENTEROLOGY
192 KELLYETAL.
tZZi None 0 Casein Triolein
150
& Q, 2
m Cas + tri m CaCI, R§I TEW
*vs. none PcO.05 ANOVA
;
Vol. 100. No. 1
100
Lipolytic activity
Lipase
Colipase
Trypsin
Chymotrypsin
Figure 3. Effects of nutrients and chymotrypsin inhibitor (TEW) on survival of pancreatic enzyme activity storage in undiluted human duodenal juice (top) and duodenal juice diluted 1:4 (middle) and 1:16(bottom) during 56 days of freezer storage. The last group of bars (chymotrypsin activity) does not include values for TEW because addition of this chymotrypsin inhibitor abolished chymotrypsin activity (Figure 2). In undiluted juice, casein, casein plus triolein, and chymotrypsin inhibitor significantly preserved lipolytic activity and lipase compared with juice frozen without additives (P < 0.05, ANOVA). In juice diluted 1:4,casein and casein plus triolein significantly improved survival of lipolytic activity. These additives as well as chymotrypsin inhibitor significantly enhanced the none of the additives altered survival of enzymatic activity compared with no nutrient. preservation of lipase. In juice diluted 1:16,
samples; after 8 weeks 65% and 40% of lipolytic activity was lost in the 1:~ and 1:16 diluted samples, respectively (1:lvs.1:16, P = 0.05; 1:l vs. 1:4, NS). In both control and treated samples loss of lipolytic activity closely paralleled the loss of lipase activity (Figure 4, middle), whereas colipase was well preserved during freezing (Figure 4, bottom). The addition of nutrients and TEW did not affect the activities of the lipolytic enzymes before freezing (data not shown). During freezer storage, the presence of casein or casein and triolein in the undiluted and 1:4diluted samples significantly decreased the loss of lipolytic activity and lipase (P < 0.05, ANOVA, Figure 3). Turkey egg white also significantly decreased the loss of lipolytic activity and lipase in undiluted juice. In the 1:16 diluted samples none of the additives significantly affected survival of lipolytic activity. CaCl, and triolein did not significantly alter the loss of lipolytic activity or lipase, although with added triolein the survival of enzyme was intermediate between samples containing casein and those with no nutrient.
Relationships Colipase
of Lipase Activity to Lipase and
The survival of lipolytic activity expressed as percentage remaining correlated strongly with survival of lipase (r = 0.97; P < 0.02; Figure 5, top) but only weakly with the survival of colipase (r = 0.49; P < 0.05; Figure 5, bottom). The relationship between lipolytic activity and lipase suggests that loss of lipolytic activity is caused by loss of lipase, whereas the relationship between lipolytic activity and colipase indicates that the loss of lipolytic activity greatly exceeded the loss of colipase.
Discussion During freezer storage of duodenal juice obtained after CCK-OP stimulation, there was maintenance of chymotrypsin and colipase activities, variable loss of trypsin activity, and pronounced loss of lipolytic activity and lipase over 8 weeks. Survival of lipolytic activity and lipase during storage, however, was improved by diluting the samples 1:16 or by the
January 1991
PANCREATIC ENZYMESIN FROZENDUODENAL JUICE
120 l- Lipolytic activity
'F; 2
0
I
I
I
I
I
I
10
20
30
40
50
56
Days in freezer Figure 4. Effect of nutrients and chymotrypsin inhibitor (TEW) on survival of lipolytic activity (top), lipase (middle), and colipase (bottom) during storage of undiluted human duodenal juice at -20°C for l-56 days. In aliquots frozen after the addition of casein, casein plus triolein or chymotrypsin inhibitor, lipolytic activity, and lipase were almost totally preserved compared with those frozen with no additives or with CaCI,. Triolein alone was only partially protective. By contrast, colipase was stable during freezing regardless of additives.
addition of chymotrypsin inhibitor or of casein (alone or in combination with triolein) before freezing. In contrast, triolein alone and CaCl, had no effect on survival of lipolytic activity or lipase. Lipase was the major component of lipolytic activity which was lost during freezing as was indicated by the close correlation (r = 0.97) between lipolytic activity and lipase remaining. Colipase, which was almost completely preserved during freezer storage, was only weakly correlated with loss of lipolytic activity. The loss of up to 50% of tryptic activity in frozen CCK and secretin stimulated duodenal juice reported by others (3) was not observed in the present study. In that study (3) three samples decreased tryptic activity by more than 40% over 15-30 days, and a fourth specimen lost 30% activity after 50 days. The remaining five specimens lost less than 20% of activity after 15-40 days of storage. In contrast, the juice from our four subjects lost 25, 20, 15, and 0% of activity after 56 days of storage. After dilution 1:4,two samples lost no activity, and after 1:16 dilution all samples had 85%-100% of activity after 56 days of storage. Thus,
193
although it appears that trypsin may be stable to freezing in duodenal juice of some patients, there may be enough loss to prohibit its routine use in frozen samples. The reason for this variable loss of trypsin activity is unclear. Chymotrypsin, by contrast, was very stable with more than 92% of activity remaining after 56 days of freezer storage in undiluted samples. There was slightly more lability of chymotrypsin in the 1:4 dilution, but even at this concentration, 85% or more of activity remained after 56 days. In 1:16 dilutions, two samples lost 15%-20% of activity. A third lost about 10% of activity and a fourth was completely stable. These findings suggest that chymotryptic activity may be a better indicator than trypsin of actual activity of pancreatic enzymes if samples must be frozen before analysis. The irreversible loss of lipolytic activity and lipase in duodenal juice which occurs during freezer storage has been described previously (4,5) and is similar to that observed intraluminally (2) and during incubation at 37°C (1). Loss of activity seems to be the result of chymotryptic digestion, as the addition of the chymotrypsin inhibitor TEW blocks the loss of lipolytic activity and lipase during freezing just as it does in vivo (8) and during incubation (1). It has been shown
160
y=8.8+0.81x
120 - r=0.97 80
Lipase, % remaining
J 0
20
40
80
80
loo
120 140
160
Colipase, % remaining Figure 5. Relationships of percent of lipolytic activity remaining to percent of lipase activity remaining (top) and to percent of colipase remaining (bottom) after 56 days of freezer storage. The strong correlation between lipolytic activity and lipase suggests that lipase is the major factor destroyed during freezer storage. The slight deviation of the best fit line from the identity line could be the result of unmeasured components (e.g., phospholipase A, and carboxylester lipase) which may have also been destroyed during freezer storage.
194
GASTROENTEROLOGY
KELLY ET AL.
that chymotrypsin, but not other proteases, digests porcine lipase near its active site (9). Casein protected lipase during freezing, an effect similar to that observed when pureed chicken was added to duodenal juice before freezing (3). It is likely that these proteins provide an alternate substrate for chymotrypsin, replacing lipase in that role. Although triolein did not significantly improve survival of lipase, the lipase activity of samples containing triolein was intermediate between that of no nutrient and of casein, casein plus triolein, or TEW. Previous studies had shown that corn oil added to duodenal juice protected lipase during freezing (3). The differences between this observation and our study may be explained by the presence of several lipids in corn oil (24% oleic acid, 58% linoleic acid, and 11% stearic acid) which may protect lipase better than triolein alone. Although triglyceride provides an anchor for the lipase-colipase complex, this effect of triglyceride only partially protects lipase as surface denaturation of lipase may occur in the presence of triglyceride (16). The correlation between lipolytic activity and lipase suggests that digestion of the lipase is primarily responsible for the decline in lipolytic activity. In addition, this correlation suggests that lipase is the major component which is measured in the assay system using Lipomul. The lack of perfect correlation between lipolytic activity and lipase may, in part, be caused by phospholipase A, and carboxylester lipase which were not measured separately in this study. While these latter enzymes may also be destroyed during freezer storage, they are responsible for a small amount of lipolytic activity in duodenal juice as measured by our assay system because lipase activity accounted for nearly all of the total lipolytic activity in the samples. Colipase was stable during freezing. This cofactor is required for optimal activity and stability of lipase (14). Colipase in pancreatic extract is not inactivated by heating (17) but trypsin and bile may affect the activity of colipase. The incubation of trypsin with pure rat colipase (trypsin:colipase specific activities = 6.5:1) causes inactivation of colipase which is minimized by the addition of rat bile (18). The ratio of trypsin and colipase in the study of Lairon et al. (18), however, was much higher than in the normal human juice used in the present study (approximately 1:l). The low trypsin-colipase ratio and bile in typical intraluminal concentrations in our study may have protected the colipase in frozen duodenal juice. The dilution of duodenal juice used in the present study was intended to mimic pancreatic insufficiency. However, in pancreatic insufficiency, protease levels are relatively higher than lipase levels (19). Therefore, the magnitude of the loss of lipolytic activity which might be expected in juice from pa-
Vol. 100, No. 1
tients with pancreatic insufficiency would in all likelihood be greater than we demonstrated because the relative amounts of chymotrypsin-lipase would be greater in pancreatic insufficiency. Our observations are particularly important if duodenal juice from patients with suspected pancreatic insufficiency is frozen before assay. Samples which are obtained during hormonal stimulation will lose lipolytic activity as a result of storage which may lead to an erroneous diagnosis of pancreatic insufficiency, especially if samples are stored for more than a few days. Therefore, either lipolytic activity in duodenal juice obtained during CCK-OP stimulation should be assayed immediately, or if juice is frozen, casein (33 I5 mg/mL as a final concentration) or TEW (5 mg/mL as a final concentration) should be added to protect lipolytic activity (lipase) from chymotryptic digestion. Alternatively, colipase or chymotrypsin should be measured in previously frozen samples because they are stable components of pancreatic secretion. References 1 Thiruvengadam R, DiMagno EP. Inactivation of human lipase by proteases. Am J Physiol 1988;255:G476-G481, 2 Layer P, Go VLW, DiMagno EP. Fate of pancreatic enzymes during small intestinal aboral transit in humans. Am J Physiol 1986;251:G475-G480. enzyme activi3. Muller DPR, Ghale GK. Stability of pancreatic ties in duodenal juice after pancreatic stimulation by a test meal or exogenous hormones. Ann Clin Biochem 1982;19:8993. 4. Legg EF, Spencer AM. Studies on the stability of pancreatic enzymes in duodenal fluid to storage temperature and pH. Clin Chim Acta 1975;65:175-179. 5. Sternby B, Kelly DG, DiMagno EP. How do fat and protein preserve lipolytic, lipase (LIP) and colipase (COL) activity in frozen human duodenal samples? Physiologic and clinical implications (abstr). Pancreas 1988;3:619. 6. Lindstrom MB, Sternby B, Borgstrom B. Concerted action of human carboxylester lipase and pancreatic lipase during lipid digestion in vitro: importance of the physiochemical state of the substrate. Biochim Biophys Acta 1988;959:178-184. 7. Borgstrom B. Importance of phospholipids, pancreatic phospholipase A, and fatty acid for the digestion of dietary fat. In vitro experiments with porcine enzymes. Gastroenterology 1980;78: 954-962. 8. Thiruvengadam R, Zinsmeister AR, DiMagno EP. Can lipase activity be preserved during aboral intestinal transit in humans? (abstr). Gastroenterology 1986:90:1663. 9. Bousset-Risso M, Bonicel J, Rovery M. Limited proteolysis of porcine pancreatic lipase. Lability of the Phe 335-Ala 336 bond towards chymotrypsin. FEBS Lett 1985:182:323-326. 10. Kelly DG, Bentley KJ, Sandberg RJ, Zinsmeister AR, DiMagno EP. Do nutrients and bile in human duodenal juice affect the survival of lipase activity? Possible clinical implications (abstr). Gastroenterology 1988;94:A222. 11. Sipos T, Merkel JR. An effect of calcium ions on the activity, heat stability, and structure of trypsin. Biochemistry 1979;9: 2766-2775. 12. Talay P. Enzymatic analysis of steroid hormones. Methods Biochem Anal 1960;8:119-143. 13. Pelot D, Grossman MI. Distribution and fate of pancreatic
January
14. 15.
16.
17. 18.
19.
1991
enzymes in small intestine of the rat. Am J Physiol 1962;202: 285-288. Borgstrom B, Erlanson C. Pancreatic juice colipase: physiological importance. Biochim Biophys Acta 1971;242:509-513. Fleiss JL. The parallel groups design. In: The design and analysis of clinical experiments. New York: Wiley, 1986:4690. Borgstrom B. The temperature-dependent interfacial inactivation of porcine pancreatic lipase. Effect of colipase and bile salts. Biochim Biophys Acta 1982;712:490-497, Rosenheim 0. On pancreatic lipase. III. The separation of lipase from its co-enzyme. J Physiol1910;4O:xiv-xvi. Lairon D, Nalborne G, LaFont H, Domingo N, Hauton JC. Protective effect of biliary lipids on rat pancreatic lipase and colipase. Lipids 1977;13:211-216. DiMagno EP, Malagelada JR, Go VLW. Relationship between
PANCREATIC
ENZYMES
alcoholism and 1975;252:200-207.
IN FROZEN
pancreatic
DUODENAL
insufficiency.
Ann
JUICE
195
NY Acad
Sci
Received December 12, 1989. Accepted July 16, 1990. Address requests for reprints to: Eugene P. DiMagno, M.D., Tower Gastrointestinal Diagnostic Unit, Mayo Clinic, Rochester, Minnesota 55905. Supported by National Institutes of Health fellowship DK07198 (to D.K.); Henning and Johan Throne Holst Foundation, Stockholm, Albert Pahlssons Foundation, Malmo, and the Swedish Medical Association, Stockholm, Sweden (to B.S.); Clinical Research Center grant RR585; and the Mayo Foundation. The authors thank Karen J. Bentley, David Katzmann, and Richard Tucker for skillful technical assistance and Linda Bakken for preparing the manuscript.
