Pancreatic acinar cell function and morphology in rats fed zinc-deficient and marginal zinc-deficient diets

GASTROENTEROLOGY

1986;90:946-57

Pancreatic Acinar Cell Function and Morphology in Rats Fed Zinc-Deficient and Marginal Zinc-Deficient Diets FRANCISCO PEREZ-JIMENEZ, MANJIT SINGH

DALE E. BOCKMAN,

Pancreatic Research Laboratory and Medical Research Service, Center (Downtown Division), and the Departments of Medicine of Georgia, Augusta, Georgia

The prevalence of marginal zinc nutriture in several populations of people in this country and the lack of reports on the effect of marginal zinc nutriture in experimental animals prompted us to look at pancreatic acinar cell function and morphology in rats fed a zinc-deficient diet ad libitum: 4 and 50 ppm zinc-supplemented diets in amounts isocaloric to a zinc-deficient diet and Rodent-Blox fed ad libitum for a period of 49 ? 1 (SEM) days. Because of a diminished rate of energy expenditure in zincdeficient rats, animals receiving 5’0 ppm zincsupplemented diets were ofered less food, resulting in decreased body weight and pancreas weight, DNA, RNA, total protein, lipase, amylase, and secretion of protein. Specific changes due to zinc deficiency included (a) further decrease in body weight and (b) increase in content, specific activity, and secretion of lipase. Both the size and volume fraction of zymogen granules were reduced in zinc deficiency. The lumina of acinar and small ducts were collapsed with paucity of secretion products. Zinc deficiency may therefore lead to a defect in discharge mechanism. A further reduction in volume fraction of zymogen granules in the 4 ppm zinc-supplemented group was associated with increased secretion of serine proteases (trypsinogen Received March 8, 1985. Accepted October 3, 1985. Address requests for reprints to: Manjit Singh, M.D., Medical Research Service, Veterans Administration Medical Center, Augusta, Georgia 30910. F. Perez-Jimenez’s permanent address is: Chief of Section of Internal Medicine, “Reina Sofia” Hospital, Medical School, University of Cordoba, Cordoba, Spain. This work was supported by the Fulbright Program and the Ministry of Education and Science, Spain and by the Medical Research Service of the Veterans Administration. The authors thank Marilyn M. LaSure for technical assistance and Patricia Story for secretarial assistance. 0 1986 by the American Gastroenterological Association 0016-5085/86/$3.50

and

Veterans Administration and Anatomy, Medical

Medical College

and chymotrypsinogen), which constitute -46% of total secretory protein in the pancreas under normal dietary conditions. This indicated an accelerated discharge due to an unknown mechanism. Changes in the secretion of digestive enzymes in the present study simulated ethanol-induced secretory alterations that were previously observed. Because abnormal zinc nutriture and chronic alcoholism are commonly associated, it is speculated that zinc deficiency may play a role in the ethanol-induced secretory alterations.

In severely zinc-deficient rats (l-4) and pigs (5), biochemical and ultrastructural changes in the pancreas and several other tissues were reported. Severe deficiency of zinc has also been reported in clinical conditions such as acrodermatitis enteropathica and protein calorie malnutrition in infants; and enteral and pare&era1 alimentation, intestinal malabsorption, and alcoholism in adults (6). However, marginal zinc nutriture is much more prevalent in the United States than we might assume. Populations at risk for poor zinc nutriture include children, preadolescent girls and college women, women on oral contraceptive agents, pregnant women, institutionalized persons, and patients with alcoholic liver disease (6). Inasmuch as severe zinc deficiency is of relatively limited occurrence and marginal deficiency is more prevalent in the humans (including alcoholics), the present study was performed under conditions of both severe and marginal deprivation of zinc to elucidate the role of zinc nutriture in structural and functional integrity of the pancreas in the rapidly growing young rats. Abbreviation

used

in this paper:

DNA, deoxyribonucleic

acid.

April 1986

ZIKC DEFICIENCY

Materials and Methods Diets and

Chemicals

The following were purchased: Wayne RodentBlox from Allied Mills, Chicago, Ill.; zinc-deficient diet (Bio-Mix #0935) from Bio-Serv, Inc., Frenchtown, N.J.; enterokinase (from porcine intestine), trypsin (from bovine pancreas, type III), N-a-benzoyl-oL-arginine-p-nitroanilide (50% HCl, benzoyl+tyrosine ethyl ester, lipase substrate olive oil, volivol), toluene, trition X, PPO (2,5-diphenyloxazole) and POPOP (1,4-bis[2-(5-phenyloxazolyl)]benzene), ribose and deoxyribose standards from Sigma Chemical Co., St. Louis, MO.; and [methyl“Clthymidine, [8-‘“Cladenine, and L-[14C]phenylalanine from New England Nuclear Corporation, Boston, Mass. All other chemicals were reagent grade.

Experimental

Design

Male Sprague-Dawley rats (64-85 g body wt) were obtained from Harlan Sprague Dawley, Indianapolis, Ind. They were housed individually in stainless steel cages and were maintained under the following conditions: a 6 AM to 6 PM light cycle, 74 k 2 (SEM) “F temperature, and 55% humidity. Animals were divided into 10 groups of 4 each. The first animal in each group was fed the Rodent-Blox ad libitum. Rodent-Blox is one of the commonly acceptable commercial diets recommended for the normal growth of rats, provides 30% of chemical energy as protein, 12% as fat, and 58% as carbohydrates (7), and contains 60 ppm of zinc. The second and third animals were fed a zincdeficient liquid diet; 18% of the chemical energy of the diet was made up of protein, consisting of a mixture of egg whites [containing L-cystine, uL-methionine, and supplemented with choline chloride), 35% was made up of neutral fat (a mixture of corn oil, olive oil, and safflower oil), and 47% was made up of carbohydrates (maltoseDextrins), supplemented with 4 and 50 ppm, respectively, of zinc. The fourth animal was fed the zinc-deficient diet ad libitum. Zinc deprivation is known to cause poor food consumption and the development of a characteristic cyclic pattern of eating when the intake is less than 6 ppm (8). This results in a failure of growth in the young rats due to the reduction in the rate of their energy expenditure (9). To allow for a distinction between the changes attributed to zinc deficiency or marginal zinc nutriture and those due to resultant inanition due to caloric deprivation (lo), animals on 4 ppm zinc and 50 ppm zinc diets were fed in amounts equal by volume to the recorded intake of zincdeficient animals. Liquid diets were served once daily in graduated feeding tubes (LDF-11 from Bio-Serv, Inc., Frenchtown, N.J.) that facilitated the recording of the daily food consumption for pair feeding. Diets were made fresh daily in a gallon-sized Waring Blender (Waring Products Division, Dynamics Corp. of America, New Hartford, Conn.) and allowed to stand for 24 h to reduce aeration. Body weights of the rats were determined once a week. Experiments were done after an average of 49 t 1 (SEM) days on these diets. Careful precautions were taken to minimize the environmental zinc contamination (4).

Biochemical

AN11 THE PANCREAS

947

Experiments

The animals were anesthetized, blood was collected for separation of plasma, and the pancreas was removed, trimmed of excess fat and connective tissue, and weighed. A small piece of tissue from the tail part of the pancreas was used for electron microscopy. Pancreatic: lobules were prepared according to the method described by Scheele and Palade (11). Incubation conditions for studies of DNA, RNA, and protein content, I.-[‘-‘C]phenylalanine incorporation in protein, [8-14C]adenine incorporation in RNA, and [methyl-‘“Clthymidine incorporation in DNA have been described previously (12). Assay

of Trypsinogen

ChymotrJIpsinogen

and Activit!i

The optimal conditions required for activation of two of the rat pancreatic zymogens were studied in preliminary experiments using the criteria specified b! Scheele and Palade (11). Comparative determinations of tryptic and chymotryptic activity derived by activation of the zymogens were not affected by presence of albumin or incubation in plastic or glass tubes at 37”C, although plastic tubes and pipettes (“Pipetman” supplied by Rainin Instrument Co., Boston, Mass.) were used in the present study for all incubations. In agreement with observations of others, activated enzymes were protected by a component of the assay system, which is probably the membrane (13). Chymotrypsinogen was activated with trypsin (from bovine pancreas, type III). A sample of tissue homogenate with the protein concentration ranging from 10 to 270 pg added to Tris-HCl buffer (pH 8.0) containing 0.01 mol . L ’ CaCl, and a specified amount of trypsin dissolved in 0.001 N HCl solution containing 0.01 M CaCI, were incubated at +3”C and f37”C for 1, 4, 8, 16, and 24 h (see Figure 1). Stability of the enzymes at the level of maximal activation was affected profoundly by the protein concentration of the sample. With lower protein concentration (10-30 pg), activation occurred rapidly and remained stable for a longer period of time. With higher protein concentration (90-270 wg), although the activation occurred in 1 h, loss of the activity was marked at 1 h and became extensive at later time points (Figure 1, upper panel). Activation of chymotrypsinogen at +3”C under similar conditions rovealed a stable plateau of enzymatic activity at +3”C for a period of up to 18 h, although the initial appearance of enzymatic activity was less (Figure 1, lower panel). In the present study, all samples of the tissue homogenates and incubation media were activated for chymotrypsin with a protein concentration ranging from 20 to 25 pg, as determined by BioRad assay. Considering our experimental conditions, activation conditions of 1 h at +37”C were preferred, and chymotrypsin was determined by the method described by Hummel (14). The units were expressed as micromoles of benzoyl-I.-tyrosine ethyl ester hydrolyzed per minute per milligram of protein. Activation of trypsinogen was done with enterokinase because it causes a negligible hydrolysis of the trypsin substrate N-cu-benzoyl-nl.-arginine-p-nitroanilide HCL. Without adding enterokinase, no activation of trypsinogen

948

GASTROENTEROLOGY

PEREZ-JIMENEZ ET AL.

micrograms trypsin.

of lrypsin

Assay

after

of Amylase

Vol. 90. No. 4

comparison

and

Lipase

with

standard

Activity

a-Amylase activity was assayed in the aliquots of the media and in the homogenized tissue by the method of Bernfeld (15) by using Lintner’s starch as the substrate. A unit of amylase activity was expressed as the amount that catalyzes the formation of 1 mg of maltose in 3 min at + 37°C. Lipase was assayed in the aliquots of media and in the homogenized tissue by the method of Tietz and Fiereck (16). The units were expressed as micromoles of NaOH neutralized per minute per milligram of protein. 1

z 2 2

4

8

Determination

24

16

Hours of Incubation

0.9 -

Plasma

0.8 -

proteins

(17) and

method

of Plasma

were

Proteins

were determined expressed

as grams

by the Lowry per deciliter.

P z

0.7-

. m

0.6 -

.z 3

0.5 -

s

0.4 -

Quantification electron

0" .E

0.3 -

c

1

4

18

Hours of Incubation

Figure

1. Effect of total protein concentration of the pancreatic homogenate on chymotrypsinogen activation at +3X (upper panel] and +3”C (lower panel). Pancreatic tissue was homogenized in 10 ml of 0.01N HCl containing 0.01M CaCl, at +4’C and diluted in 0.05 M Tris-HCI buffer (pH 8.0)containing 0.01M CaCI, to contain 10 pg (O---O ), 30 pg (n-a), 90 pg (A-A), and 270 Kg (O-O) per sample. Activation was initiated with addition of bovine trypsin -5% (wtiwt] to total homogenate protein [and also -20% (wtiwt) to secretory protein]. Activation of chymotrypsinogen was assayed by measurement of chymotrypsin activity with time at +37”C. No activation was seen without added trypsin.

was observed. When activation of trypsinogen was tested at protein concentrations ranging from 0.01 to 2.7 mg in the sample, the maximal levels of activation were observed after 18 h of incubation at +37”C with the secretory protein concentration of 270 pg. In the present study, aliquots of -270 pg of protein), 0.05 M the samples (containing Tris-HCl buffer (pH 8.0) containing 0.01 mol . L-l CaCl,, and 0.1% (wtiwt) partially purified enterokinase dissolved in 0.15 M NaCl were incubated at +37”C for 18 h, and a sample was used to determine the N-a-benzoyl-marginine-p-nitroanilide HCl hydrolysis. The amount of n-nitroaniline liberated was measured in a Coleman spectrophotometer at 383 nm. The units were expressed as

of Morphologic

Data

The size of zymogen granules was determined from micrographs with a final magnification of 9585.

The diameter of each granule was measured to the nearest millimeter and tabulated. A total of 200-329 granules was measured in each group. The volume fraction of zymogen granules in the different groups was determined by overlaying a transparent grid (containing 99 intersections) on electron micrographs with final magnifications of 7,425, 9,585, and 12,420. The proportion of intersections falling on zymogen granules was determined, and the scores were adjusted by subtracting the intersections that fell on extraacinar objects (connective tissue, blood vessels, lumina) or grid bars. Granules were counted on the same micrographs. Differences between the groups were tested using the Student’s t-test.

Statistical

Analysis

of Biochemical

Data

Two-way analysis of variance with replication (one time period x four treatments) was used to analyze the results (18). Group means were analyzed with the Tukey HSD (honestly significant difference) procedure for the 0.05 level. Values of p < 0.05 were considered to be statistically significant. The effect of feeding the hypocaloric diets was considered to be operative when the zinc-deficient group and the isocaloric controls fed a marginal zinc diet and the zinc-replenished diet were different from the Rodent-Blox animals. The effect of zinc deficiency or marginal zinc deficiency was considered to be operative when the analysis of variance showed the zinc-deficient or marginal zinc-deficient groups to be different from the Rodent-Blox group and other groups fed hypocaloric diets. Only in this way, by including foodrestricted controls, can alterations in the parameters due to zinc deprivation per se be distinguished from those due to resultant inanition and concomitant caloric deficit (10).

Results cant

The rats fed low-zinc diets developed anorexia and the onset of a cyclical

signifieating

April

1986

Table

Zlh’C DEFICIENCY

Effect of Varying Zinc Content of Diet on Body Weight, Pancreas of Protein, DNA, and RNA in Rat Pancreas In Vitro”

1.

Parameter

AND THk: PANCREAS

Weight, and Total Content

949

and Synthesis

Lab-Blox

Supplemented

276 + 9

mg!organ mgilO0 pg DNA mgig body wt Total protein mgiorgan

1029 + 55 15 + 2 3.5 2 0.14

mgilO0 pg DNA L-[‘“Clphenylalanine incorporation (dpmimg protein) DNA (pg/organ) (Methyl-“C]thymidine incorporation (dpm/loO kg DNA] RNA ( ygiorgan) [E-‘“Cladenine incorporation (dpmi100 ~g RNA)

2.3 -1- 0.3 1911 t 189

3.0 5 0.3 1786 t 100

2.4 +- 0.3 1923 +- 282

2.3 i 0.5 1652 t 211

7210 r+_610 53 2 6

3369 2 319” 94 k 35

4195 t 466” 94 ir 23

3097 + 340” 52 + 7

4013 4636

2393 t 181” 4041 +_ 422

2644 4393

2584 3247

Body weight (g) Pancreas weight

189 2 8

Marginal

Deficient

90 + 3”

91 2 5”

77 + 3’

544 t 14” 17 -+ 2 6.0 k 0.25”

549 c 13” 14 -” 1 6.0 2 0.28”

511 + 14” 18 ? 2 7.0 + 0.17”

91 t 3”

* 369 t 1619

91 + 6”

5 204” 5 545

DNA, deoxyribonucleic acid; dpm, disintegrations per minute. I’Results are mean k SEM of 5-10 animals ’Zinc deficiency effect (p < 0.05) in analysis diet effect (p < 0.05) in analysis of variance (see Methods).

pattern within 3-4 days after beginning the feeding. Average daily food intake of the rats fed Rodent-Blox and of the zinc-deficient rats was 10.1 + 0.5 g (SEM) The energy and 26 k 0.7 ml (SEM), respectively. density of Rodent-Blox was -14 kJ . g-l and that of the liquid diets was -4.2 kJ . ml-l. Characteristic stigmata of zinc deficiency, including stunting of the growth and poor hair coat, as described by others (4) were evident in the zinc-deficient group when they were killed. Animals whose diets were supplemented with 4 and 50 ppm zinc were stunted in their growth but did not develop other changes. Plasma protein levels in the zinc-deficient rats, in rats supplemented with 4 and 50 ppm zinc, and in the Rodent-Blox-fed rats (expressed as grams per deciliter) were 1.50 2 0.23 (57% of controls, p < 0.05), 1.81 I? 0.40 (68% of controls,

Table

Lipase

content

4,491

content

Trypsinogen content Chymotrypsinogen content Lipase (Uimg protein) Amylase (Lrimg protein) Protein secretion Basal

(mg/organ)

Stimulated (mgiorgan) I>-[“C]phenylalanine secretion Basal

(dpmiorgnn)

Stimulated

Lab-Blox

(dpmlorgan)

+ 232

65,970

? 4,866

10,291 965 24 353

+ + -t +

1,244 169 1 31

t 458” i- 473

in each group. ” Hypocaloric of variance (see Methods].

of controls, NS), and 2.65 + 0.33, respectively. Results of body weight, pancreas weight, DNA, RNA,

protein, [methyl-14C]thymidine incorporation in DNA, [8-14C]adenine incorporation in RNA, and I.[14C]phenylalanine incorporation in the proteins are shown in Table 1. Compared with the Rodent-Blox group, body weight, pancreas weight (expressed as milligrams per pancreas or milligrams per gram of body weight], DNA, RNA, and the protein content were decreased in the animals fed the liquid diets. Body weight was significantly decreased further in the zinc-deficient group compared with the marginal and supplemented groups. There was no significant difference between the Rodent-Blox-fed group and the animals fed liquid diets in pancreas weight, expressed as milligrams per 100 pg DNA; I.-[‘“Clphenylalanine incorporation in total protein, expressed as

2. Effect of Varying Zinc Content of Diet on Lipase, Amylase, Trypsinogen, Total Protein and Newly Labeled Protein in Rat Pancreas In Vitro” Parameter

Amylase

NS), 2.38 ? 0.48 (89%

85 t- 3”

Supplemented 1,842 13,990 7,277 813 20 152

t 89” t 2.046” t t t t

488 66 1 21”

and Chymotrypsinogen Marginal 1,756 14,338 6.667 714 19 152

oj

Deficient

+ 188” -+ 3.598” -c i rt t

Secretion

887 100 1 33”

2,662 15.845 8.920 728 32 193

5 I707\ t

1,110”

t + 2 t

1.639 95 3 16”

14.0 + 1.3

5.5 t 0.8”

6.0 -t 1 .O”

5.0 + 0.8”

19.0 + 0.6

8.0 + 0.4”

8.0 k 0.7”

8.0 i 1.0”

177,644

+ 50,922

34,239

t 4,678”

33,857

? 4,092”

30.841

+ 2,735”

192,722

t 50,996

42,308

t

44,095

-+ 1,422”

38,783

t

5,504”

dpm. disintegrations per minute. ” Results are mean 2 SEM of 5-10 animals in each group. ” Hypocaloric of variance (see Methods). ’Zinc deficiency effect (p < 0.05) in analysis of variance (see Methods).

3,843”

dirt effect (p < 0.05) in analysis

GASTROENTEROLOGYVol. 90,No.4

950 PEREZ-JIMENEZETAL.

disintegrations per minute per milligram of protein; [methyl-14C]thymidine incorporation in DNA: and [8-14C]adenine incorporation in RNA. The total content of lipase, amylase, trypsinogen, chymotrypsinogen, specific activities of amylase and lipase, and secretion of total and newly labeled protein are shown in Table 2. Compared with the Rodent-Blox group, a significant decrease in the total content and the specific activity of lipase and amylase occurred in all the groups fed the liquid diets. Additionally, a statistically significant increase of lipase activity was found in the zincdeficient rats. Total content and the specific activity of trypsinogen and chymotrypsinogen were not changed. Secretion of total proteins did not show a change, whereas the secretion of newly labeled proteins was lower in all animals fed the liquid diets. Both basal and bethanechol-stimulated secretion of lipase, trypsinogen, and chymotrypsinogen are shown in Figure 2. Lipase secretion after bethanechol stimulation was increased in the zincdeficient group, whereas during basal state, the increase in secretion did not reach statistical significance. Basal secretion of trypsinogen and chymotrypsinogen was increased in the marginal zincdeficient group. Secretion of amylase was similar in all groups fed the liquid diets (data not shown). Electron microscopic investigation revealed marked changes in size, volume fraction, and the distribution of the zymogen granules. The mean diameter of granules in the zinc-deficient group [0.5 t 0.009 pm (SEM)] was significantly less (p < 0.0051 than granules in the Rodent-Blox-fed group (0.78-Iwith 4 and 50 0.001 pm). The groups supplemented ppm zinc had granules with mean diameters (0.60 4 0.13 pm and 0.51 -+ 0.008 pm, respectively) similar to the zinc-deficient group and significantly smaller (p < 0.005) than the Rodent-Blox-fed group. None of the granules in the zinc-deficient group exceeded 0.9 pm in diameter, whereas 24% of the granules in the Rodent-Blox-fed group exceeded that size. The groups supplemented with 4 and 50 ppm zinc had granules with diameters >0.9 pm, although they had fewer granules than the controls. The 4-ppm group and the Rodent-Blox-fed group had maximum granule diameters of 1.25 pm. Size differences are shown clearly in Figures 3 and 4, which are the same magnifications. Stereology revealed that the volume fraction of zymogen granules in the deficient group [19.7 * 2.83 (SEMI] was significantly smaller (p < 0.005) than that of the Rodent-Blox-fed group (32.7 * 3.15). Furthermore, the zinc-supplemented groups had volume fractions (4.3 ? 1.45 for 4 ppm, 5.1 t 1.46 for 50 ppm) significantly smaller (p < 0.005) than the

i

I

A

-

1 B

Figure

C

I

2. Secretion of lipase, trypsinogen, and chymc )trYF lsinogen in vitro after a 60-min incubation expressed as percentage of total in the tissue at the beginning of incubation in animals fed Rodent-Blox ad libitum (A], liquid diet supplemented with 50 ppm of zinc (B), liquid diet containing 4 ppm of zinc (C), and zincdeficient diet (D] under basal (solid bars) and bethanechol-stimulated (open bars] conditions. *, p < 0.05 in analysis of variance with zinc-deficient or marginal zinc-deficient animals were compared with the other three dietary groups (see Methods).

zinc-deficient unit area in (17.8 5 10.1 smaller (p < The number significantly

group. The number of granules per the 50 ppm zinc-supplemented group granules per 100 pm’) was significantly 0.025) than in the zinc-deficient group. of granules in the 4-ppm group was not different from the deficient group.

April 1986

Figure 3. Electron micrograph showing zymogen granules from the pancreas of a rat fed Rodent-Blox ad libitum. Compare their size to those in Figure 4 (x9500).

The distribution of granules was distinct in the zinc-supplemented groups. The granules tended to be concentrated in a narrow zone along acinar lumina (Figures 5 and 6). Therefore, large areas of acini were devoid of granules, and the rough endoplasmic reticulum appeared quite prominent. After zinc deficiency, the lumina of acini (Figure 4) and small ducts (Figure 7) frequently appeared collapsed, with paucity of secretion product, while those from the Rodent-Blox-fed group (Figure 3) and the zinc-supplemented groups (Figures 5 and 6) had obvious secretion products in the lumina that were of greater diameter.

Discussion Although the marginal zinc nutriture in animals is not well defined, it is accepted that animals fed a diet containing 4 ppm of zinc do not develop overt clinical signs of zinc deficiency. The specific nutritional deficit, however, is reflected in a variety of parameters previously shown to be sensitive to

zinc deprivation (19). Zinc content of Wayne Rodent-Blox, which is a commonly acceptable commercial diet recommended for normal growth of rats, is 60 ppm. Hurley (20) reported that 60-100 ppm of zinc content in the diet, based on isolated soybean protein, was adequate for normal growth. As the availability of zinc is lower in the diets containing isolated soybean protein due to the presence of phytate, Lo et al. (21) recommended 50 ppm zinc as adequate for growth when the dietary protein source was egg white. As the zinc deficiency results in anorexia and cyclical eating in the animals fed a zinc-deficient diet (9), pair feeding was essential in planning this study. To that end, we included a zinc-supplemented group (with the diet providing zinc in an amount equal to Rodent-Blox and exceeding the level of 30 ppm recommended by the American Institute of Nutrition for laboratory animals; see Reference 22) pair fed with the marginal deficient group and the deficient group and a group fed Rodent-Blox ad libitum. The voluntary caloric intake of zinc-defi-

952

PEREZ-JIMENEZ ET AL.

GASTROENTEROLOGY Vol. 90, No. 4

Figure 4. The pancreas from a zinc-deficient animal. Although there are numerous zymogen granules, profiles of large granules are missing when the same magnification as in Figure 3 is used. Acinar lumina (arrows) are occupied mainly by microvilli and have little secretion product (x9500).

cient rats fell to -70% of that of Rodent-Blox, in agreement with observations of others (23). In addition to growth retardation, deficiency signs consisting of immature hair coats, fissures at the corners of the mouth, scaly feet, and alopecia occurred in the animals fed a zinc-deficient diet. A decrease in the plasma protein level was also observed in the animals fed low-zinc diets, in agreement with previous observations (20). Changes that could be ascribed to decreased caloric intake in the rats fed 50 ppm zincand 4 ppm zinc-supplemented diets due to pair feeding with zinc-deficient rats were observed in several parameters including the body weight, pancreas weight, total protein, L-[‘4C]phenylalanine incorporation in total protein, DNA, RNA, amylase, lipase, and secretion of proteins. As DNA content of the individual cells is relatively constant, parallel decreases in pancreatic weight and DNA content indicated diminished pancreatic cell numbers in proportion to the decrease in pancreatic weight. Furthermore, the ratio between protein and DNA

remained unchanged, suggesting that the cell size was not modified. These findings suggest stunting of pancreatic growth. A look at the ratio between pancreas weight and body weight shows that stunting of body growth, which occurred in all of the animals fed liquid diets, was out of proportion to the pancreatic dystrophy. Animals fed 4 and 50 ppm zinc in the diet, however, did not show signs of zinc deficiency other than stunting of body growth. In addition to the aforementioned changes due to decreased caloric intake, specific changes were observed in the zinc-deficient group. The loss of body weight was significantly higher in the zinc-deficient group than in the marginal zinc and the zincsupplemented groups, probably through an inefficient protein metabolism. The disturbed protein and amino acid metabolism in zinc deficiency may depend on several mechanisms. Zinc acts as a cofactor for RNA polymerase I and II in transcription, stabilizes the attachment of the ribosomes to the rough endoplasmic reticulum in translation, serves as a

April 1986

Figure

5. Zymogen containing

ZIKC DIIFICIENCY

granules from a marginally zinc-deficient secretion product (arrow). Abundant

animal are few, heterogeneous rough endoplasmic reticulum

cofactor for protein elongation factor, stabilizes intracellular membranes, and promotes normal glycolytic enzyme activity (24-28). Our failure to show a decrease in L-[‘4C]phenylalanine incorporation into the proteins was not surprising inasmuch as incorporation of the labeled amino acids in vivo and in vitro are different (29,SO). In the zinc-deficient rats, [methyl-14C]thymidine incorporation into DNA was not decreased although this was accompanied by a reduction in the total content of DNA. An alteration in nucleic acid metabolism has been pre-

AND THE PANCKEAS

in size, and distributed (KEK) fills the cytoplasm

along acinar (X.7500).

953

lumina

viously reported, including a diminished thymidine incorporation into DNA (31) and a reduction of thymidine kinase and DNA polymerase in the tissues from zinc-deficient animals (32). Lack of a parallel decrease in [methyl-14C]thymidine incorporation in DNA and DNA content suggested that an increase in DNA catabolism took place. In their elegant studies, Schick et al. (33) reported that the acinar cell directed its protein synthesis to elaborate increased protease zymogens to the extent of 95%) of total exocrine proteins synthesized after 12

954

PEREZ-JIMENEZ ET AL.

GASTROENTEROLOGY Vol. 90, No. 4

Figure 6. The pancreas from a pair-fed animal given zinc supplement. Some large profiles of zymogen granules are present. The secretion product fills the acinar lumen (arrow). Abundant rough endoplasmic reticulum (RER) is somewhat dilated. Same magnification as Figure 5 (X 7500).

days of feeding a protein-free diet in contrast to 45.6% of total exocrine proteins synthesized under normal laboratory dietary conditions (22% protein). During adaptation to the high protein content (30% 82%) and, correspondingly, decreased levels of the carbohydrates in the diet, amylase and the majority of proteases were synthesized in direct proportion to nutritional substrates (33). Therefore, in the present study, across-the-board changes in the content and specific activities of the digestive enzymes in animals fed a liquid diet supplemented with 50 ppm zinc compared to the Rodent-Blox-fed group appear to be due to decreased caloric intake (rather than a lack of utilization of carbohydrate or protein alone) and are in agreement with observations of Hatch et

al. (34). The changes in relative content of digestive enzymes in animals fed liquid diets deficient in zinc or supplemented with 4 ppm zinc were due to changes’in the zinc content of the diet per se. Total content and specific activity of lipase was increased in the zinc-deficient group; total content and specific activities of trypsinogen and chymotrypsinogen were not changed. Robinson and Hurley (3) demonstrated that the maternal zinc deficiency lowered the chymotrypsinogen content of fetal rat pancreas. Based on the present study on rapidly growing young rats, the role played by the zinc seemed to be more complex, suggesting that the effect on the digestive enzymes may vary depending on the age of the animal.

April

Figure

1986

7. Small duct from a zinc-deficient material (X12,500).

ZINC DEFICIENCY AND THE PANCKEAS

rat. The lumen

(arrow]

Because the tissue content of an enzyme is not always parallel with its secretion, and because the secretion may be a more sensitive parameter of cellular injury (35,36), basal and bethanecholstimulated pancreatic secretion in vitro was studied. Lipase secretion displayed a normal basal response, but a significantly increased secretion of lipase was observed after stimulation in the deficient rats. These findings suggest a perturbation in the regulatory mechanism of the acinar cell secretion related to zinc deprivation, Sullivan et al. (5) had reported a decreased secretion of proteases and amylase in vivo in the zinc-deficient pigs after stimulation with secretin in combination with one of the following: vagal stimulation, caerulein, or gastrin. It is hard to compare their findings with our results because action of effecters on the pancreatic secretion are not always the same in vivo and in vitro (373, and they did not include lipase in their assays and had used a different species. The present investigation has established a definite pattern to changes in zymogen granules in

is small,

occupied

almost

entirely

by microvilli

and other

955

nonsecretory

response to zinc deficiency and dietary reduction. Koo and Turk (4) noted a reduction in the number of zymogen granules, and an “aberrant size and shape,” due to zinc deficiency, but did not mention similar changes in the pair-fed controls. Weisblum et al. (38) reported a reduction in the number of zymogen granules after protein deprivation in rats, describing alterations in density, but not size. Blackburn and Vinijchaikul(39) described the greatly reduced number of zymogen granules associated with kwashiorkor as tending to be small and randomly distributed rather than clustering in the apical cytoplasm. None of these exactly parallels our observations. It seems likely that although zinc deficiency and dietary deprivation share effects on the pancreas, not all effects of zinc deficiency are due to diminished food intake. The volume fraction of zymogens was reduced significantly under zinc deficiency and was reduced significantly more in the marginally deficient and the supplemented animals. The presence of some profiles of large zymogen granules in dietary-deprived animals, and not in zinc-deficient

956

PEREZ-JIMENEZ

GASTROENTEROLOGY

ET AL.

animals, indicates a difference in the effect on the “packaging” mechanism or on the secretory products available for packaging. The quantity of zymogens in an acinar cell represents a balance between synthesis and packaging of secretory proteins and discharge of zymogen granules. The morphologic data suggest that zinc deficiency may lead to a defect in the discharge mechanism. When the zinc-deficient group is compared with the 50 ppm zinc-supplemented group, it has significantly more granules, a greater volume fraction of the zymogens, and contracted lumina deficient in secretory material. The zinc-supplemented groups seem to have a defect in synthesis or an acceleration of discharge, or both; secretory material is in the lumen, but the volume fraction of zymogens is low and the granules tend to be limited to a margin adjacent to lumina. The biochemical findings, including secretory changes of enhanced basal discharge of protease zymogens with marginal deficiency and depressed discharge of proteases and increased discharge of lipase with zinc deficiency, correlate well with this interpretation. Trypsinogen of the seand chymotrypsinogen represent -45.6% cretory product of the rat under normal dietary conditions (33), providing a better indication of bulk secretory protein than lipase, which represents -4.5% of total secretory protein. Although speculative, it is interesting to question whether nonparallel effects on pancreatic enzymes, as observed biochemically, are mirrored in the morphologic observations. The functional change observed in the present study is of a similar quality as observed after longterm feeding of ethanol, resulting in an increased basal secretion of trypsinogen, chymotrypsinogen, and lipase, and a diminished secretory response after bethanechol compared with controls (36). Basal secretion of trypsinogen and chymotrypsinogen was higher in the marginal deficient group compared to the pair-fed rats, whereas in this regard, the deficient rats responded as the supplemented and the RodentBlox-fed groups, suggesting a specific disorder in the marginal zinc-deficiency group. Similar behavior of both trypsinogen and chymotrypsinogen in the marginal deficient group, the presence of a small standard error in results, changes specific to the zymogens that provide the bulk of pancreatic secretory proteins, and the difference in morphologic appearances in the zinc-deficient and marginal zincdeficient group that correlate with the secretory changes make these results interesting. Further work is in progress to examine the effect of a marginal zinc-supplemented diet fed ad libitum, isocaloric amounts of 50 ppm zinc-supplemented diet, and Rodent-Blox fed ad libitum to rats in order to determine the effect of marginal zinc nutriture per se on

Vol. 90. No. 4

the pancreas, independent of the caloric deficit imposed by the design of experiments in the present study. Low extracellular concentration of zinc was reported to impair cell membrane function and induce a generalized membrane pathology as a primary component of zinc deficiency (40); therefore, altered secretion in the marginal deficient rats could probably be due to membrane perturbations, as suggested for alcoholic rats (36). Considering that marginal zinc deficiency and chronic alcoholism are frequently associated conditions (411, zinc deficiency may be playing a significant role in the alcohol-induced cell damage reported previously (35,36).

References 1. Prasad

2.

3.

4.

5.

6. 7. 8. 9. 10.

11.

12. 13.

14.

15.

16. 17.

AS, Oberleas D, Wolf P, Horwitz JP. Studies on zinc deficiency: changes in trace elements and enzyme activities in tissues of zinc-deficient rats. J Clin Nutr 1967;46:549-57. Mills CF, Quarterman J. Williams RB, Dalgarno AC. The effects of zinc deficiency on pancreatic carboxypeptidase activity and protein digestion and absorption in the rat. Biochem J 1967;102:712-8. Robinson LK, Hurley LS. Effect of maternal zinc deficiency or food restriction on rat fetal pancreas. I. Procarboxypeptidase A and chymotrypsinogen. J Nutr 1981;111:858-68. Koo SI, Turk DE. Effect of zinc deficiency on the ultrastructure of the pancreatic acinar cell and intestinal epithelium in the rat. J Nutr 1977;107:896-908. Sullivan JF, Burch RE, Quigley HJ, Magee DF. Zinc deficiency and decreased pancreatic secretory response. Am J Physiol 1974;227:105-8. Sandstead HH. Zinc nutrition in the United States. Am J Clin Nutr 1973;26:1251-60. Elseweidy M, Singh M. Folate deficiency and pancreatic acinar cell function. Proc Sot Exp Biol Med 1984;177:247-52. Williams RB, Mills CF. The experimental production of zinc deficiency in the rat. Br J Nutr 1970;24:989-1003. Chesters JK, Will M. Some factors controlling food intake by zinc-deficient rats. Br J Nutr 1973:30:555-66. Beach RS, Gershwin ME, Hurley LS. Growth and development in postnatally zinc-deprived mice. J Nutr 198O;llO: 201-11. Scheele GA, Palade GE. Studies on the guinea pig pancreas. Parallel discharge of exocrine enzyme activities. J Biol Chem 1975;250:2660-70. Singh M. Effect of vitamin B, deficiency on pancreatic acinar cell function. Life Sci 1980;26:715-24. Preiser H, Schmitz J, Maestracci D, Crane RK. Modification of an assay for trypsin and its application for the estimation of enteropeptidase. Clin Chim Acta 1975;59:169-75. Hummel BCW. A modified spectrophotometric determination of chymotrypsin, trypsin, and thrombin. Can J Biochem Physiol 1959;37:1393-9. Bernfeld P. Amylases cy and p. In: Colowick SP, Kaplan NO, eds. Methods in enzymology. New York: Academic, 1977: 149-59. Tietz NW, Fiereck EA. A specific method for serum lipase determination. Clin Chim Acta 1966;13:352-8. Lowry OH, Kosebrough NJ, Farr AL. Randall RJ. Protein measurement with the folin phenol reagent. J Biol Chem 1951;193:265-75.

April

18.

19.

20. 21.

22.

23. 24.

25. 26.

27.

28. 29.

30.

1986

Kim JO, Kohout JF. Analysis of variance and covariance. In: Nie NJ, Hull CH, Jenkins JG, Steinbrenner K, Bent DH, eds. Statistical package for social sciences. New York: McGrawHill, 1975:398-433. Golub MS, Gershwin ME, Hurley LS, Hendrickx AG. Baly DL. Induction of marginal zinc deficiency in female rhesus monkeys. Am 1 Primatol 1982;3:299-305. Hurley LS. Zinc deficiency in the developing rat. Am J Clin Nutr 1969:22:1332-9. Lo GS, Settle SL, Steinke FH, Hopkins DT. Effect of phytate: zinc molar ratio and isolated soybean protein on zinc bioavailability. J Nutr 1981;111:2223-35. Report of the American Institute of Nutrition Ad Hoc Committee on Standards for Nutritional Studies. J Nutr 1977; 107:1340-8. Chesters JK, Quarterman J. Effects of zinc deficiency on food intake and feeding patterns of rats. Br J Nutr 1970;24:1061-9. Valenzuela P, Morris RW, Faras A, Levinson W, Rutter WJ. Are all nucleotidyl transferases metalloenzymes? Biochem Biophys Res Commun 1973;53:1036-41. Tal M. On the role of Zn” and Ni2+ in ribosome structure. Biochim Biophys Acta 1968;169:.564-5. Kotsiopoulos PS, Mohr SC. Protein synthesis elongation factor 1 from rat liver: a zinc metalloenzyme. Biochem Biophys Res Commun 1975:67:979-87. Chvapil M. New aspects in the biological role of zinc: a stabilizer of macromolecules and biological membranes. Life Sci 1973;13:1041-9. Im MJC, Hsu JM, Hoopes JE. Enzyme activities in the epidermis of zinc-deficient rats. J Nutr 1975;105:1391-4. Morisset JA, Webster PD. Effects of fasting and feeding on protein synthesis by the rat pancreas. J Clin Invest 1972; 51:1-8. Couture Y, Mongeau R, Dunnigan J, Morisset J. Evidence that

ZINC DEFICIENCY

31.

32.

33.

34.

35.

malnutrition

on pancreatic

J Clin Nutr

1979;32:1224-30.

Singh

M, LaSure

Singh

M. Effect

enzyme

38.

zymogen

MM, Bockman

and morphology

diet. Gastroenterology

37.

957

protein synthesis can be increased in vitro following cholinergic stimulation. Can J Physiol Pharmacol 1972;50:874-82. Prasad AS. Oberleas D. Koniuch D. Thymidine kinase activity and incorporation of thymidine into DN,4 in zinc-deficient tissue. J Lab Clin Med 1974:83:634-Q. Duncan JR, Hurley LS. Thymidine kinase and DNA polymerase activity in normal and zinc deficient developing rat embryos. Proc Sot Exp Biol Med 1978;159:39-43. Schick J. Verspohl R, Kern H, Scheele G. Two distinct adaptive responses in the synthesis of exocrine pancreatic enzymes to inverse changes in protein and carbohydrate in the diet. Am J Physiol 1984;247:G611-6. Hatch TJ. Lebenthal E, Krasner J. Branski D. Effect of postnatal

function 36.

AND THE: PANCREAS

in the rat. Am

DE. Pancreatic

in rats chronically

acinar

ethanol

in rats in vitro.

feeding

on

pancreatic

Dig Dis Sci 1983;28:117-23.

Sing11 M. Effect of endorphins

on amylase

pancreas

Dig Dis Sci 1985;30:257-62.

in vivo and in vitro.

Weisblum acinar

B, Herman

cells

cell

fed an ethanol

1982:82:425-34.

of chronic

secretion

enzymes

during

L, Fitzgerald protein

secretion

PJ. Changes

deprivation.

from rat

in pancreatic

J Cell

Biol

1962;

12:313-22. 39.

Blackburn

WR. Vinijchaikul

An electron 40.

Bettger the

microscopic

K. The pancreas study.

Lab Invest

WJ, O’Dell BL. A critical

structure

and

function

in kwashiorkor. 1969;20:305-18.

physiological of

role of zinc in

biomembranes.

Life

Sci

1981;28:1425-38, 41.

McClain

CJ, Su

review.

Alcoholism:

1983;7:5-10.

L-C.

Zinc

deficiency

Clinical

and

in the Experimental

alcoholic: Research

a