fiber intake and intestine

Carbohydrate and Fiber

Intestinal Structural Changes in African Green Monkeysafter Long Term Psyllium or Cellulose Feeding1'2

INGE PAULINI, TARA MEHTA ANDANN HARGIS

Department of Food Science and Human Nutrition and Department of Veterinary Microbiology andPathology, Washington State University, Pullman, WA 99164-2032

ABSTRACT Intestinal structure of male adult African Green monkeys (Cercopithecus aethiops ssp vervets) was studied after3'/2 yr of consuming diets containing 10% psyllium husk or cellulose. Scanning electron microscopy (SEM) identified mild

damage (cellular swelling and disarray, and microvillar denudation and disarray) at villous tips throughout the small intestine inthe psyllium-fed monkeys. The cellulose group had similar duodenal damage. Differences were not found in colons by SEM. Bylight microscopy, jejunum had shorter vUli with psyllium feeding, based upon villous height (P < 0.05), and length around asectioned villus (P< 0.1), but not based upon the number of enterocytes per villus. Jejuna! and Â¡lealcircular and longitudinalmuscle layer thicknesses were increased in psyllium-fed monkeys. Colonie mucosa! height was significantly (P < 0.05) reducedand muscle layer thickness was mildly reduced in the psyllium-fed monkeys. Group differences were not found in intestinal weightor length or in the weight of small intestinal mucosal scrapings. Psyllium husk may cause epithelial cell loss and muscle layerhypertrophy in the jejunum and ileum and thinning of the colonie wall after prolonged feeding. J. Nutr. 117: 253â€”266, 1987.

INDEXING KEY WORDS:

â€¢intestinal morphology â€¢nonhuman primates â€¢dietary fiber â€¢long term

Dietary fiber has been used more frequently in thetreatment of various disorders, i.e., diabetes, hypercho-

lesterolemia, diverticulitis, due to its effect on gastrointestinal function (1, 2). Relatively little is knownabout the potential harmful effects of long-term dietary

fiber supplementation to humans.Recent investigations in rats showed structural al

terations in the small intestinal morphology afteringestion of diets containing high concentrations ofwater-soluble fermentable fibers (3-6). Pectin, rather

than fiberless or cellulose diets, was associated withincreased cellular swelling (3) and microvillar disarrayof the jejunum (3, 5) as seen by scanning electron microscopy (SEM). Histological sections showed bluntedand fused ileal villi in one study (6) and greater cryptdepth and muscle layer thickness of jejunum and ileumin another study (4) compared to fiber-free diets. Smallintestines and mucosal scraping were also heavier afterpectin feeding (4). Increased mucosal crypt cell proliferation and epithelial cell turnover in the jejunum ofrats were documented after feeding diets containingguar gum (7, 8), and increased ileal thymidine kinaseactivity was reported in rats fed guar gum and psylliumhusk (9). Changes in small and large intestines afterfeeding non-water-soluble fiber, cellulose, largely re

sembled those seen after feeding fiber-free diets (3, 4,

10, 11).The present study investigated the effects of long-

term dietary supplementation with psyllium husk andcellulose on the intestinal structure of monkeys. Psyllium husk and cellulose were the chosen fibers becauseof the differences in their chemical and physical properties and physiological effects. Psyllium husk, alsoknown commercially as Isogel, is mostly a mucilaginous, highly branched arabino-xylan-polysaccharide (12).It has a high water-holding (13) and gelling capacity

(14), similar to guar gum and pectin. Psyllium is highlyfermentable in the human intestine (15). It is used asthe active ingredient of many bulk laxatives in the UnitedStates and in western Europe (16) and is the laxative ofchoice for long-term ingestion (17). In contrast, cellulose, a particulate fiber with low water-holding capacity, largely resists microbial fermentation (18) and is a

'Scientific Paper No. 7237. College of Agriculture & Home Eco

nomics Research Center, Washington State University, Pullman, WA99164-6240.

2Presented in part at the International Nutrition Congress, Brigh

ton, England, August 19, 1985.

0022-3166/87 $3.00 Â©1987 American Institute of Nutrition. Received: 26 August 1985. Accepted 8 September 1986.

253

at Weber S

tate Univ on O

ctober 29, 2008 jn.nutrition.org

Dow

nloaded from

http://jn.nutrition.org

254 PAULINI, MEHTA AND HARGIS

normal component of the human diet, representing 20-80% of the total dietary fiber in common foods (19).

MATERIALS AND METHODS

Animals and diets. Ten male adult African Greenmonkeys (Cercopithecus aethiops, ssp vervets) weighing 3.5-5.5 kg were used. They were housed individually in squeeze-back cages with stainless steel traysin rooms controlled for temperature, humidity and airexchange, and maintained on a 12-h light-dark cycle.The vivarium was an accredited laboratory animal carefacility (American Association for Accreditation of Laboratory Animal Care) and all animals were handled inaccordance with the National Institutes of Health policies of good laboratory animal practice. All monkeyswere examined routinely by a veterinarian for any clinical problems. Water was available at all times, and thediets were fed twice a day for 3'/2 yr. The semipurified

diets supplied 42% of total calories as carbohydrate,43% as fat, 15% as protein, and contained all macro-

and micronutrients (20, 21) and differed only in thesource of dietary fiber (Table 1). Added fibers were 9.7%

TABLE1Composition of experimental diets for vervets

Ingredients Amount

g/100 g of diets

9.005.00

31.0019.269.71

18.001.504.841.000.290.40

Casein'Lactalbumin'

White wheat flour (unbleached)SucroseFiber (cellulose or psyllium husk)2-'

LardCorn Oil4Salt mixture5Vitamin mixture6Cholesterol7Choline chloride2

'United States Biochemical Corporation, Cleveland, OH.2Cellulose (Cellufil), United States Biochemical Corporation.3Psyllium husk, Main Food Center, Ltd., Vancouver, British Co

lumbia.4DL-a-Tocopherol acetate (1 mg = 1 lu), Menadione, and retinyl

palmitate (1,000,000 USP units/g) were dissolved in the corn oil andwere present in the diet at the following levels; 5 mg/100 g diet, 1.25mg/100 g diet and 0.5 mg/100 g diet.

5The mineral mix contained (in g/kg): calcium carbonate, 290.5;

calcium phosphate (dibasic), 72.6; potassium phosphate (dibasic, 327.9;sodium chloride, 162.4; magnesium sulfate, 98.7; magnesium oxide,32.0; ferric citrate, 13.3; manganese sulfate, 1.2; zinc chloride, 0.915;cupric sulfate, 0.290; potassium iodide, 0.077; chromium acetate,0.044; sodium fluoride, 0.023; sodium selenite, 0.004. Teklad Mills,Madison, WI (mineral mix #T79033).

The vitamin mix contained (in g/kg]: myoinositol, 75.0:; L-ascor-bic acid, 60.0; cholesterol, 29.24; niacin, 8.0; D-calcium pantothenate,3.0; riboflavin, 0.88; thiamin HC1, 0.8; pyridoxine HC1, 0.8; folieacid, 0.4; cholecalciferol, 0.4; 5,000,000 USP units of vitamin D2 pergram; biotin, 0.02; cyanocobalamin, 0.02 (0.1% trituration with man-

nitol); made to 1 kg with 821.44 g sucrose.7Sigma Chemical Company, St. Louis, MO.

purified cellulose or 9.7% psyllium husk on dry-weight

basis.Experimental design. Four monkeys consumed the

psyllium diet, and six monkeys received the cellulosediet for 3'A yr. One month prior to the termination ofthe experiment, two of the cellulose-fed monkeys were

switched to a nonpurified diet (Purina Monkey Chow,Ralston Purina Co., St. Louis, MO) containing 2.5%crude fiber. These two monkeys were treated separatelyfrom the cellulose group.

Specimens for microscopy. After a 20-h fast, monkeys were anesthetized with 10-12 mg/kg body wtVetalar (ketamine HC1, Parke-Davis, Morris Plains, NJ)and were then killed by exsanguination from the femoral artery. The peritoneal cavity was opened rapidly,and the length of duodenum, jejunum, ileum and colonwas measured on a horizontal surface. Measurementswere made by the same investigator in an attempt tokeep the degree of tissue stretching consistent, and noadditional efforts were made to standardize the procedure. The most proximal and distal 40-50 cm of thesmall intestine were designated duodenum and ileum,respectively, and the middle portion as jejunum. Sections 2 cm long were taken from the proximal duodenum, proximal, mid- and distal jejunum and distalileum. The colon was divided into three segments ofequal length, and sections were taken from the proximal part of the ascending, transverse and descendingcolon.

The rest of the intestinal segments that were not usedfor microscopy were immediately placed on ice, cleanedof fat and mesentery, flushed with chilled saline andopened longitudinally to scrape the mucosa with glassslides. The muscles and scrapings were weighed andstored at - 10Â°Cuntil analyzed.

Tissue preparation for scanning electron microscopy. All sections for microscopy were placed immediately in chilled phosphate buffer (0.1 M, pH 7.35) before being opened longitudinally and pinned flat on dentalwax. In each instance, the mucosal side was kept uppermost, under slight tension. The degree of tensionwas kept as consistent as possible from section to section. Mucus and debris were gently removed from theintestines with a #1 camel-hair brush, and the specimens were fixed initially in 4% formaldehyde, 1% glu-

taraldehyde, in 0.1 M phosphate buffer, pH 7.3, 1000mosmol/kg, for 1 h. After this, all sections except thetransverse and descending colon, were immersed in freshfixative for 5-6 h. Specimens were washed three timesin phosphate buffer, postfixed in 2% osmium tetroxidefor an additional 1-2 h, washed three times in distilledwater, and dehydrated in an ethanol series (30, 50, 70,95, 3 x 100% EtOH). Tissues were critical-point driedin Teflon baskets with liquid carbon dioxide in a SPC1500 critical point device (Bomar Corp., ME), mountedon aluminum stubs with double-stick tape, and goldcoated to a thickness of 30 nm with a Hummer V sputtering device (Technics, San Jose, Ã‡A).A minimum of

at Weber S

tate Univ on O


Dow

nloaded from


DIETARY FIBER AND MONKEY INTESTINAL STRUCTURE 255

50 villi per section were viewed on an ETEC Autoscanelectron microscope (Hayward, CA) operated at an accelerating voltage of 20 kV. Micrographs were takenwith Polaroid type 55 film.

Tissue preparation for light microscopy. Specimensfor light microscopy were handled similarly to thosefor SEM,including cleaning the mucosa with a #1 camel-hair brush. All specimens were then immersed into10% buffered neutral formaldehyde solution. Specimens were processed in an automatic tissue processor,embedded in paraffin, sectioned at 6 (im and stainedwith hematoxylin and eosin.

Histology measurements. Only intact areas fromperpendicular cuts were measured by light microscopy.Measurements of mucosal and muscular thickness wereobtained with a Bioquant II image analyzer, an Apple2e-based system (R & M Biometrics, Nashville, TN)utilizing a Hipad measuring device. Measurements obtained from the image analyzer for small intestinal segments were as follows: length around the villus (lengthfrom the top of one crypt up to the luminal tip of thevillus and back down to the top of the second crypt);villous height (top of crypt to the tip of the villus, 25measurements per section); and thickness of circularand longitudinal layers of the tunica muscularis externa, hereafter referred to as the circular and longitudinal muscle layers (10 measurements per section).The number of enterocytes and goblet cells were countedin 10 villi per section. In the colon, 10 measurements,made with the image analyzer, of mucosal height (lumen to muscularis mucosa) and the thickness of bothmuscular layers were taken in each section.

Sections were screened for group differences in thenumber of mitotic figures by counting mitotic figuresin the five deepest crypts in mid-jejunum and distalileum. The number of inflammatory cells was determined by counting cells that were not endothelial, en-terocytic, fibroblastic, myocytic or necrotic in the fivetallest villi in the mid-jejunum and distal ileum, andin the lamina propria between two crypts in the ascending colon. These numbers were divided by the lengtharound the villus for jejunum and ileum and by mucosalheight for the colon to minimize differences in size ofthe counting area that had been introduced by groupdifferences in villous and mucosal height. Necrotic nuclei and parts of necrotic nuclei greater than 1 jim werecounted and evaluated as were the inflammatory cells.

Statistical analysis. Statistical analyses were performed using Student's one-tailed Ã®-test(22), and allvalues are given as means Â±so.

RESULTS

Mean body weights for cellulose- and psyllium-fedmonkeys were 4.13 Â±0.54 kg and 4.41 Â±0.38 kg,respectively, at the beginning of the study, and 3.86 Â±

0.58 kg and 4.73 Â±0.98 kg, respectively, on termination. These differences were not significant.

Scanning electron microscopy (SEM). Villous shapein segments viewed at low magnification (x 120) SEMdid not differ among the groups. Proximal duodenumand proximal and mid-jejunum had villous shapes varying from narrow, finger-like villi to broad, tongue-likevilli that were broader at their base than at their tips(Fig. 1). All shapes were interspersed in all specimens.Rarely, extremely broad ridgelike structures were present. At the more distal end of the small intestine, villiwere narrower, and no broad, tongue-like structureswere seen (Fig. 2).

Mucus was interspersed in the intervillous clefts andon the villous surface, especially in the distal jejunumand ileum. Due to mucosal cleaning during tissue preparation, differences in amount of mucus could not bequanti tated.

In all small intestinal specimens examined, highmagnification (> x 1000)SEMs revealed a mucosal surface consisting of densely packed microvilli that covered enterocytes (Fig.3). Goblet cells were seen as ovalor round indentations in the surface. Mucous plugs wereat the orifices of goblet cells. In most samples, bordersof epithelial cells were clear and hexagonal (Fig. 3).Linear areas of cellular disarray along villous crests wereseen in all segments of small intestine from psyllium-fed monkeys. These linear areas consisted of swollencells, mostly with microvillar disarray and denudation(Figs. 4 and 5). Cells along the sides of the villi wereusually intact.

Less mucosal damage in the small intestine was seenin cellulose-fed monkeys. In the duodenum, three offour cellulose-fed monkeys had linear areas along thevillar crests consisting of swollen cells at villous tipsand microvillar disarray and denudation. Two of fourcellulose-fed monkeys had some denuded microvilli inthe proximal jejunum, and one monkey had swollencells in the distal ileum.

Group differences in the ascending colon were notseen by SEM. Evaluation of low magnification (x 220)SEMs revealed a surface consisting of large smooth foldsinterrupted by the regularly arranged orifices of cryptsof Lieberkuhn (Fig.6). Evaluation of high magnification(x 2450) SEMs of the colonie mucosa revealed that goblet cells, surrounded by microvilli, were at orifices ofthe crypts. The microvilli, however, were either long,fuzzy or irregularly arranged (Fig.7), or they were short,closely packed and similar to the microvilli of the smallintestine (Fig. 8).

SEMshowed that the two cellulose-fed monkeys thathad been switched to nonpurified diet for 1 mo hadsimilar intestinal characteristics to monkeys fed thecellulose diet continually.

Light microscopy. Histologically, the small intestine had villi covered by an intact layer of epithelialcells with intact crypts ; circular and longitudinal mus-

at Weber S

tate Univ on O


Dow

nloaded from



FIGURE I Villous shapes on the proximal small intestine seen in all monkeys. Note the variation in shape from narrowand finger-like to broad and tongue-like villi. This micrograph was taken from a psyllium-fed monkey ( x 135).

FIGURE 2 Narrower villous shapes in the distal small intestine of all monkeys. Finger-like villi were more prevalent; broad,tongue-like structures were not seen. This micrograph depicts the distal jejunum of a psyllium-fed monkey ( x 120).

at Weber S

tate Univ on O


Dow

nloaded from



FIGURE 3 High magnification of the smooth undamaged appearance of the mucosal surface in the small intestine aftercellulose feeding. Note the densely packed microvilli (M), clearly demarcated cell borders (arrowhead) and goblet cells (G)(X2200).

cle layers were present (Fig. 9). Intestinal folds or lym-

phoid follicles were sectioned occasionally.Villous height decreased gradually in the more distal

parts of the small intestine in both dietary groups. Villous height in the proximal jejunum was approximatelytwice that in the distal ileum in both dietary groups.Villous width was not measured, but qualitative differences were not detected between groups. Measurements of villous height and muscle layer thickness inthe small intestine are summarized in Table 2. Measurements of the length around each villus (as definedearlier) and the villous height indicated that jejunal villiwere shorter in psyllium-fed monkeys compared to cellulose-fed monkeys. Measurements of the 10 tallestvilli per section gave the same results. However, thenumber of enterocytes per villus, which is generallyconsidered to be a more accurate and reproduciblemeasure of the villous height, was not significantlydifferent between the groups. Psyllium-fed monkeyshad a thicker (P < 0.05) circular muscle layer in theileum (Fig. 9, e and /) and a thicker (P < 0.05) longitudinal muscle layer in the proximal and mid-jejunum.

The colonie mucosa was sometimes convoluted andalways covered by an intact layer of enterocytes. Themucosa was significantly thinner in ascending(P < 0.001), transverse (P<0.05) and descending(P < 0.05) colon sections (Table 3) in the psyllium-fedmonkeys compared with cellulose-fed monkeys (Fig.10). The circular muscle layer was significantly (P < 0.05)

thinner in the psyllium-fed group in transverse and de

scending colon.Within the small intestine, obvious group differences

were not seen in the number of mitotic figures, all ofwhich were confined to the crypts. Numbers of intestinal folds and lymphoid follicles did not differ betweengroups. Differences in the number of inflammatory cellswere not present between groups. All monkeys had anunusually high number of pyknotic cells (necrotic cells)in the laminia propria of the small and large intestinecompared with normal domestic and laboratory animals. However, the psyllium group had many morenecrotic cells per area than the cellulose group (Fig. 11).Results for stock-diet-fed monkeys resembled those forcellulose-fed monkeys.

Lengths, weights of intestinal segments per lengthand weight of mucosal scrapings per length are listedin Table 4. None of the values were significantly different between groups. Colons from cellulose-fed monkeys tended to be heavier than those from psyllium-

fed monkeys. .

DISCUSSION

Controlled studies of long-term dietary fiber ingestion and intestinal structure are impractical in humans.The rodent intestine, with its large cecum, differs from

at Weber S

tate Univ on O


Dow

nloaded from



FIGURE 4 Micrographs in a and b are of small intestinal mucosal damage at the villous tips after psyllium feeding. Thedamage observed was cell swelling (S), microvillar disarray (D), microvillar denudation (*) and cellular disarray as depicted in

Fig 4b. Magnification: x 2200 (a), x 2500 (b).

at Weber S

tate Univ on O


Dow

nloaded from



FIGURE 5 The damage depicted in Fig. 4 was usually arranged linearly along top of the villi ( x 500).

FIGURE 6 The ascending colonie mucosa at a fold; note crypts of Lieberkuhn (double arrowheads) and goblet cells (arrowhead)( x 220).

at Weber S

tate Univ on O


Dow

nloaded from



â€¢- - *x.

FIGURE 7 High magnifi(X2450).

3T< I-* â€¢Ã*4.

3Â£

the colon with long, fuzzy and irregularly arranged microvilli and goblet cells (G)

FIGURE 8 High magnification of the colon with short, regularly arranged, and tightly packed microvilli, goblet cells (G) andmucous strands (M). Note the sparsely distributed microvilli on the goblet cells ( x3450).

at Weber S

tate Univ on O


Dow

nloaded from



FIGURE 9 LM micrographs of proximal jejunum (a,b], distal jejunum (c,d), and distal ileum (e,f) from monkeys fed cellulose(a,c,e,) and psyllium (b,d,f). Note the shorter villi and thicker muscles in the psyllium group ( x45).

at Weber S

tate Univ on O


Dow

nloaded from



TABLE 2

Villous size and muscle layer thickness in monkey smallintestines after cellulose and psyllium feeding1

Measure Cellulose Psyllium

Length around a villus2,|xmProximal

duodenumProximaljejunumMid-jejunumDistal

jejunumDistalileumVillous

height,\anProximalduodenumProximaljejunumMid-jejunumDistal

jejunumDistalileumNumber

of enterocytes pervillusProximalduodenumProximaljejunumDistaljejunumDistal

ileumCircularmuscle layer thickness,\.Proximal

duodenumProximaljejunumMid-jejunumDistal

jejunumDistalileum1441

Â±2551671Â±3221529Â±167b1250

Â±144902Â±153605

Â±57690Â±46663Â±75'511

Â±46398Â±93256

Â±58277Â±51239Â±22173Â±26on206

Â±42112Â±70137Â±26"142

Â±73205Â± 25-1578

Â±1041409Â±1431108Â±307999Â±194784Â±52686

Â±57588Â±47465Â±117461Â±63357Â±33250

Â±50235Â±43216Â±60160Â±25204

Â±22170Â±33172Â±24166Â±48253Â±23Longitudinal

muscle layer thickness,\uuProximalduodenumProximaljejunumMid-jejunumDistal

jejunumDistalileum76

Â±1748Â±31Â»51Â±6Â»51

Â±1165Â± 1083

127956812571072

10

'Values are means Â±so for 4 monkeys per group, with 25 villousmeasurements and 10 muscle measurements in each section. â€¢-bVal

ues are significantly different from the respective psyllium valueswith P < 0.05 for a and P < 0.1 for b.

2Length from top of one crypt around the villus to the top of the

second crypt.

human intestine; therefore, monkeys were chosen forthis study.

Animals remained healthy on both cellulose- andpsyllium-supplemented diets. Although psyllium-fedanimals were heavier throughout the study, the differences were not significant at any time. Group differences were also not seen in food consumption, whichwas monitored throughout the study.

The apparent structure of monkey intestines as shownby SEM has not been documented completely. To ourknowledge, the only SEM study on monkey intestineis that of Burke and Holland (23) on Macaca arctoides(speciosa), in which the age of the monkeys and thedietary composition were not reported. The basic structure of the small and large intestine of the African Greenmonkeys in our study is similar to findings of Burkeand Holland. The monkeys had broader villi proxi-

mally, with the villi becoming narrower distally, whichhas also been seen in humans. Ridges, convolutions orbranched villi were not seen. However, the monkeys

in our study had two different types of microvilli in theascending colon, which had not been reported previously. Our findings for LM were normal for what hasbeen reported for human beings and monkeys (24-28).The monkey intestinal structure seen in this study moreclosely resembled that of humans in developed countries, rather than tropical climates (27-29). Colonstructure as viewed by LM also appeared to be normal(3, 24, 25) and comparable with specimens from humans and rats.

Psyllium husk induced changes in structure throughout the small intestine that were similar to those seenin the jejunum of rats on pectin-supplemented diets (3,

5). Similar changes, though less marked, were seen inthe duodenum and to some extent in the jejunum andileum of cellulose-supplemented monkeys. However,

rats fed a 15% cellulose diet for 6 wk (3) or 10% cellulose diet for 12 wk (5) had normal jejunal structure;changes in the duodenum were not studied. Differencesin the length of the study, species differences in bileflow and microbial metabolism might be responsiblefor the observations made in our study on cellulosesupplementation. Aging in monkeys might have beenan additional factor.

There was evidence of increased cell loss caused bypsyllium husk, in view of the significantly shorter villiof the jejunum and the apical mucosal damage seen onSEMs as cellular disarray and swelling. However, villous height, when evaluated as the number of cells pervillus, was not different between the groups in anysection. These results agree with other studies done inrats with gel-forming degradable fibers (4, 7, 11).

The significant decrease in the mucosal height in thecolon, especially in the ascending colon (P < 0.001), inpsyllium-fed monkeys compared with cellulose-fed

TABLE 3

Mucosal height and muscle layer thickness in monkey largeintestines after cellulose and psyllium feeding1

Measure Cellulose Psyllium

Mucosal height, \unAscending colon 355 Â± 3V 233 Â±14Transverse colon 409 Â±119b 235 Â±30Descending colon 348 Â± 48C 254 Â±17

Circular muscle layer thickness, \imAscending colon 423 Â±53 403 Â±64Transverse colon 449 Â± 74d 331 Â±62Descending colon 461 Â± 42b 360 Â±69

Longitudinal muscle layer thickness, junAscending colon 110 Â± 30 116 Â±20Transverse colon 113 Â± 36 90 Â±14Descending colon 121 Â± 30 127 Â±30

'Values are means Â±so for 4 monkeys per group and 10 mea

surements for each parameter in all sections.a,b,c,dyajuesare significantly different from the respective psyllium

values with P < 0.001 for a; P < 0.05 for b; P < 0.02 for c, P < 0.1ford.

at Weber S

tate Univ on O


Dow

nloaded from



fthi/ÃA;. ;rÂ¿"

f

â€¢â€¢â€¢_--- - Vâ€¢

"-"^' *

FIGURE 10 LM micrographs of ascending (a,b), transverse (c,d) and descending (e,f) colon of animals fed cellulose (a,c,e) andpsyllium (b.d.f). Note the thinner mucosa in psyIlium-fed monkeys ( x48).

at Weber S

tate Univ on O


Dow

nloaded from



FIGURE 11 LM micrographs contain the increase in darkly stained pyknotic cells (arrowheads) in the mucosa of psyIlium(b,d,f) as compared to cellulose-fed animals (a.c.e) in mid jejunum [a.b], distal ileum (c,d) and ascending colon (e,f)Â¡X465).

at Weber S

tate Univ on O


Dow

nloaded from



TABLE 4

Intestinal weight and length and weight of mucosa! scrapings inmonkeys fed cellulose and psyllium'

MeasureLength,

cmSmallintestineColonWeight

per length, mg/cmDuodenumJejunum

IleumColonMucosal

scrapings per length,DuodenumJejunum

IleumCellulose162

Â±3058Â±7194

Â±37191Â±21

190 Â±63731Â±345mg/cm

60.2 Â±41.541.6Â±19.7

47.4 Â±31.1Psyllium144

Â±657Â±10218

Â±78178Â±40

191 Â±57499Â±17756.7

Â±14.863.0Â±27.9

41.5 Â±15.8

'Values are means Â±so for 4 animals per group. None of the values

is significantly different from another.

monkeys might be related to the increased microbialmetabolism and viable microbial counts seen in psyl-lium-fed monkeys (30). Bile acid flow, concentration

and pattern might also be affected by psyllium husk(unpublished) and might have affected mucosal cellturnover.

The increase in muscle layer thickness observed inmonkey jejunum and ileum after psyllium feeding hasalso been reported in rats after pectin feeding (4) andmay be due to an adaptation of the muscular layers topropel through the intestine the increased bulk causedby gel-forming fibers. On the other hand, in the descending and transverse colon, the circular muscle layerwas significantly thicker in the cellulose-fed monkeys.

An increase in the total muscle thickness in the distalcolon has also been noted by Jacobs and Schneeman(31) in rats fed 20% wheat bran compared with thosefed a no-fiber diet. Because most of the psyllium isfermented in the ascending colon, and cellulose is lessfermented, the bulk of colon contents would be expected to be greater in cellulose-fed animals. The excessbulk may increase the workload on the musculature,thereby increasing muscle thickness in the cellulose-fed group. Higher dry fecal output and similar freshfecal output was found in cellulose-fed monkeys relative to psyllium after 3 yr of dietary treatment (32). Thebulk hypothesis is supported by findings of increasedintraluminal pressure in the colon of monkeys fed somefiber-containing diets (33). However, the increase inbulk probably is not sufficient to explain the changesin muscle hyperplasia. In some studies, more complexchanges, such as the presence of different muscle cellsizes, were induced by different types of dietary fibersand suggest a different pathogenesis (34). The musclelayer thickness differences could be caused by the differential effects of fibers on microbial metabolism andtrophic hormones.

Based on the screening techniques used in this study,the density of mitotic figures in crypts was not differentbetween dietary groups. In rats, cellulose, as well aswater-soluble fibers like pectin and guar gum, havecaused greater proliferative activity than no-fiber and

wheat bran diets (7, 8, 31). However, a comparison wasnot made between the effects of cellulose and pectin.Increases in the number of mucosal neutrophils with5, 10 and 20% dietary cellulose after 3 wk of feedinghas been seen in rats (35). Comparable data with water-

soluble fiber is not available. We found an abnormaldegree of necrotic cells in the lamina propria in monkeys of both groups. The cells were more numerous inthe psyllium-fed group. The fibers themselves or bac

terial endotoxins could have caused necrosis in thelamina propria that might not have been observed inprevious studies of relatively short duration.

The absolute and relative values of wet weight ofsmall intestinal mucosal scraping were not differentbetween groups in our study, suggesting that the cryptsin psyllium-fed animals were elongated to compensatefor the decreased villous height or that the psylliumadhered to the mucosa and introduced a weighing error(36). In the case of a weighing error, the mucosa wouldhave actually been lighter (thinner) after psyllium feeding compared with cellulose feeding. Jacobs (7) foundthat 10% guar gum caused increased mucosal wet weight,RNA and DNA. Farness and Schneeman (10) also foundheavier mucosae after pectin feeding. However, the intestines of their pectin-fed rats were significantly longer

than those of other treatment groups. The investigatorsin the latter studies did not base the mucosal weighton intestinal length, which might have eliminated thedifferences.

Other studies have reported considerable increase inthe length of the small intestine after pectin and guargum feeding in rats (4, 8, 17). This increase in lengthwas suggested to be an adaptation process to compensate for delayed absorption due to the gel-forming fi

bers. In the present study, there were no differences inintestinal length between the two dietary groups. Several explanations are possible: Ã•)elongation of the intestine might have taken place at the beginning of ourstudy, but after long-term adaptation, the length returned to normal; 2) chronic ingestion of 10% psylliumhusk does not require such an adaptation, perhaps dueto the prolonged transit and longer contact time; 3) theelongation of the intestines in rats could be due to thetrophic effects of fiber breakdown products by bacteria.

Our studies suggest that psyllium husk may causeincreased epithelial cell loss and jejunal ileal musclelayer hypertrophy and thinning of the colonie wall afterprolonged intake, relative to cellulose. Both dietary fibers caused abnormal number of necrotic cells in thelamina propria. At present, the functional significanceof such injuries is uncertain. The injuries did not causeany obvious deleterious effects, as all animals werehealthy and had no diarrhea or abnormal feces. Ab-

at Weber S

tate Univ on O


Dow

nloaded from



sorption of iron (37)or zinc and copper (32)was withinnormal range in these monkeys. However, in view ofthe mucosal damage seen in monkeys, care should betaken in recommending long-term ingestion of gel-forming fibers.

ACKNOWLEDGMENT

The authors thank Dr. Charles W. Leathers for performing physical examinations on the monkeys, forperforming necropsies, and for assistance in manuscriptpreparation.

LITERATURECITED

1. ANDERSON,J. W. & CHEN, W.-J. L. (1979) Plant fiber carbohydrate and lipid metabolism. Am. /. Clin. Nutr. 32: 346-363.

2. LEEDS,A. R. (1982) Modification of intestinal absorption bydietary fibers and liver components. In: Dietary Fiber in Healthand Disease (Vahouny G. V. and Kritchevsky, D., eds.), pp. 53-71, Plenum Press, New York.

3. CASSIDY,M. M., LICHTFOOT,F. G., GRAU, L. E., STORY,J. A.,KRITCHEVSKY,D. & VAHOUNY,G. V. (1981) Effect of chronicintake of dietary fibers on the ultrastructural topography of ratjejunum and colon: a scanning electron microscopy study. Am.]. Clin. Nutr. 34: 218-228.

4. BROWN,R. C., KELLEHER,I. & LOSOWSKY,M. S. (1979) The effectof pectin on the structure and function of the rat small intestine.Br. /. Nutr. 42: 357-365.

5. TASMAN-JONES,C, OWEN, R. L. & JONES, A. L. (1982)Semipurified dietary fiber and small-bowel morphology in rats.

Dig. Dis. Sci. 27: 519-524.6. SCHEMEL,M. S. & MEHTA,T. (1985) Effect of dietary fiber on

bioavailability of zinc and copper and histology in rats. Nutr.Res. 5: 81-93.

7. JACOBS,L. R. (1983) Effects of dietary fiber on mucosal growthand cell proliferation in the small intestine of the rat: a comparison of oat bran, pectin, and guar with total fiber deprivation.Am. /. Clin. Nutr. 37: 954-960.

8. JOHNSON,I. T., GEE,J. M. & MAHONEY,R. R. (1984) Effect ofdietary supplements of guar gum and cellulose on intestinal cellproliferation, enzyme levels and sugar transport in the rat. Br. ].Nutr. 52: 477-487.

9. CALVERT,R. et al. (1985) Dietary fiber and intestinal adaptation: effects on intestinal and pancreatic digestive enzyme activities. Am. J. Clin. Nutr. 41: 1249-1256.

10. FARNESS,P. L. & SCHNEEMAN,B. O. (1982) Effects of dietarycellulose, pectin and oat bran on the small intestine in the rat./.Nutr. 112: 1315-1319.

11. SIGLEO,S., JACKSON,M. J. & VAHOUNY,G. V. (1984) Effects ofdietary fiber constituents on intestinal morphology and nutrienttransport. Am. /. Physio!. 246: G34-G39.

12. SANDHU,J. S., JUDSON,G. J. & KENNEDY,J. F. (1981) The gelnature and structure of the carbohydrate of Isphagula husk exPlantago ovata Forsk. Carbohydr. Res. 93: 247-259.

13. STEPHEN,A. M. & CUMMINGS,J. H. (1979) Water-holding by

dietary fiber in vitro and its relationship to fecal output in man.Gut 20: 722-729.

14. WHISTLER,R. L. & HYMOWITZ,I. (1979) Guar: Agronomy, Production, Industrial Use and Nutrition. West Lafayette, PurdueUniversity Press.

15. PRYNNE,C. J. & SOUTHGATE,D. A. T. (1979) The effects of a

supplement of dietary fiber on fecal excretion by human subjects.Br. J. Nutr. 41: 495-503.

16. FINGL,E. (1970) Cathartics and laxatives. In: PharmacologicalBasis of Therapeutics (Goodman, L. S. & Gilman, A., Ã©d.),pp.1020-1026, Macmillan, New York.

17. TEDESCO,F. J. et al. (1985) Laxative use in constipation. AmÂ¡.Gastroenterol. 80: 303-309.

18. HOLLOWAY,W. D., TASMAN-JONES,C. & LEE,S. P. (1978) Digestionof certain fractions of dietary fiber in humans. Am. f. Clin. Nutr.31: 927-930.

19. SOUTHGATE,D. A. T., BAILEY,B., COLLINSON,E. & WALKER,A. F.(1976] A guide to calculating intakes of dietary fibre. /. Hum.Nutr. 30: 303-313.

20. NATIONALRESOURCECOUNCIL (1978) Nutrient requirementsof non-human primates, hi: Nutrient Requirements of Domestic

Animals, No. 14. National Academy of Sciences, Washington,DC.

21. NICOLOSI,R. J. & HUNT, T. D. (1979) Dietary allowances fornutrients in non-human primates. In: Primates in NutritionalResearch (Hayes, K. C., Ã©d.),pp. 11-37, New York, Academic

Press.22. STEEL,R. G. D. & TORRIE,f. H., eds. (1960) Principles and

Procedures of Statistics. McGraw Hill Co., Inc., New York.23. BURKE,J. A. fin.HOLLAND,P. (1976) The epithelial surface of

the monkey gastrointestinal tract. A scanning electron-microscopic study. Digestion 14: 68-76.

24. RHODIN,f. A. G., ed. (1975) An Atlas of Histology, pp. 318-328, Oxford University Press, New York.

25. REITH,E. J. & Ross, M. H., eds. (1970) Atlas of DescriptiveHistology, 2nd Ã©d.,pp. 120-131, Harper &. Row Publishers Inc.,

New York.26. PFEIFFER,C. J. (1968) Intestinal villous morphology. Postgrad.

Med. 431: 215-221.

27. BAKER,S. J. (1973) Geographical variations in the morphologyof the small intestinal mucosa in apparently healthy individuals.Pathol. Microbio!. 39: 222-237.

28. CHACKO,C. J. G., PAULSON,K. A., MATHAN,V. I. & BAKER,S.J. (1969) The villous architecture of the small intestine in thetropics: a necropsy study. /. Pathoi. 98: 146-151.

29. OWEN,R. L. & BRANDBORG,L. L. (1977) Jejunal morphologicalconsequences of vegetarian diet in humans. Gastroenterology72: 1111 (abs.).

30. COSTA,M., MALES,J. R. & MEHTA,T. (1985) Effect of Cellulose, Psyllium Husk and Cholesterol Level on Fecal Output andMicrobial Activity of Monkeys. Presented at XIII InternationalNutr. Congress, Brighton, UK.

31. JACOBS,L. R. &. SCHNEEMAN,B. O. (1981) Effects of dietarywheat bran on rat colonie structure and mucosal cell growth. /.Nutr. Ill: 798-803.

32. PAULINI,I. & MEHTA,T. (1985) Zinc and copper bioavailabilityin monkeys after longterm cellulose and psyllium feeding. Fed.Proc. 44: abs. 2034.

33. BRODRIBB,A. J. M., CONDON,R. E., COWLES,V. & DECOSSE,J. J.(1979) Effect of dietary fiber on intraluminal pressure and my-oelectrical activity of left colon in monkeys. Gastroenterology77: 70-74.

34. JACOBS,R. L. (1985) Differential effects of dietary fibers on ratintestinal circular muscle cell size. Dig. Dis. Sci. 30: 247-252.

35. GORDON,D. T., BESCH-WILLIFORD,C. & ELLERSÅ’CK,M. (1983) The

action of cellulose on the intestinal mucosa and element absorption by the rat. /. Nutr 113: 2545-2556.

36. FORMAN,L. P. & SCHNEEMAN,B. O. (1980) Effects of dietarypectin and fat on the small intestinal contents and exocrinepancreas of rats. /. Nutr. 110: 584-590.

37. BUTH,M. & MEHTA,T. (1983) Effect of psyllium husk on ironbioavailability in monkeys. Nutr. Rep. Int. 28: 743-752.

at Weber S

tate Univ on O


Dow

nloaded from


fiber intake and intestine

Documents