Decreased Absorption of Calcium, Magnesium, Zincand Phosphorus by Humans due to Increased Fiberand Phosphorus Consumption as Wheat Bread1

JOHN G. REINHOLD,2 BAHRAM FARADJI,"PARICHEHR ABADI ANDFARAMARZ ISMAIL-BEIGI *

Medical Research Unit, Institute of Nuclear Medicine,Nemazee Hospital, and the Departments of Biochemistryand Medicine, School of Medicine, Pahlavi University,Shiraz, Iran

ABSTRACT During a 20 day period of high fiber consumption in theform of bread made partly from wheaten wholemeal, two men developednegative balances of calcium, magnesium, zinc and phosphorus due to increased fecal excretion of each element. The fecal losses correlated closelywith fecal dry matter and phosphorus. Fecal dry matter, in turn, wasdirectly proportional to fecal fiber excretion. Balances of nitrogen remainedpositive. Mineral elements were well-utilized by the same subjects duringa 20 day period of white bread consumption. J. Nutr. 106: 493-503, 1976.INDEXING KEY WORDS bread •calcium •fiber •magnesium•phosphorus •zinc

The staple food in Iran, as elsewhere inthe Middle East and in Northern India,is bread made from wheat flours of varying degrees of refinement. In Iranian cities,the breads consumed are made almost entirely from flours of 80% to 90% extraction rate into leavened flat breads, Bazari(or Taftoon) and Sangak. Outside of thecities and larger towns, locally groundwholemeals of nearly total extraction rateare made mainly into the paper-thin shepherds breads, Tanok or Lavosh. Leaven isoften omitted in their preparation, butwhen used, limited fermentation occurspartly because of the brief time allowedand also because of the resistance of wholemeals to the action of yeast (1). Kouhes-tani et al. (2) have described the preparation of various traditional Iranian breads.The nutritional properties of these andsimilar breads affect the well-being of avery large population.

All rural and most urban breads consumed in these regions contain substantialamounts of fiber. They also include significant amounts of phytate, averaging 0.7%

by weight in the rural and about half thisconcentration in urban breads. This difference results in part from the use ofsomewhat lower extraction flours by citybakers but mainly from destruction of phytate by sourdough organisms of leaven ( 3 ).Fiber concentrations of the various breadsdiffer less than those of phytate and fiberintakes are consistently high.5 Maleki (4)found at least 50% of the energy intake ofvillage schoolboys in Iran to be derivedfrom bread, but the proportion approaches75% or more in both village and urbanfamilies.

A suspicion that the high intakes ofbreads of these types were not entirelybeneficial, arose during investigations ofthe cause of the hypogonadal dwarfism

Received for publication September 10, 1975.*A preliminary report was given at the Sixth

Pahlavi Medical Congress held in Shiraz in April1975.

'Present address: Apdo. Postale 20, Ajijlc, Jalisco,Mexico.

»Present address: School of Public Health, UCLA,Los Angeles, Calif. 90024.

* To whom requests for reprints should be addressed.•Relnhold, J. O. & Gharhamani, P. (1975) Un

published observation.

493

by guest on August 23, 2015

jn.nutrition.orgD

ownloaded from

494 REINHOLD, FARADJI, ABADI AND ISMAIL-BEIGI

that is a manifestation of zinc deficiency.This syndrome is seen only among adolescent groups in various villages and consumption of wholemeal breads in largeamounts appeared to be causally important(5, 6). Although village breads and thevillage diet contain enough zinc, iron, andcalcium to fulfill daily requirements forthese metals (4, 7), their availability forintestinal absorption is poor. As a result,in addition to zinc deficiency, iron (7) andcalcium deficiencies (8) also exist. Thoseof zinc and iron respond to supplementation with these metals (7, 9). The response to calcium supplementation is notas well-documented. However, when villagers after a lifelong consumption ofwholemeal bread consumed diets that supplied ample amounts of calcium, zinc andphosphorus in available forms, they retained abnormally high proportions of eachelement, although not of nitrogen (10). Themagnitude of these retentions suggestedthat body stores of the mineral elementswere depleted, and that increased risk ofthe development of mineral deficienciesmay be postulated. Our previous studieshave been concerned with wholemealbread, however, a close similarity in composition made it necessary to obtain additional information about the effects of consumption of bread made from flour ofsomewhat lower extraction rates uponmineral metabolism. In this paper we describe the response of two human subjectsto the consumption of such a bread, Bazari,made from flour of 80% to 90% extractionrate with that of white bread.

METHODS

Subjects. Experiments were carried outon two men who were fully informed as tothe studies to be made and their objectives.Subject Rah had participated in two previous long-term metabolic experiments. Hewas a 35-year-old itinerant barber, andhis friend, subject Mor was a 24-year-oldunemployed laborer. Both sold blood occasionally to the Hospital Blood Bank, although neither had donated recently, andtheir hemoglobin concentrations were 13.0and 12.7 g/100 ml respectively. Theyweighed 56.0 and 60.3 kg. Body weightswere normal for heights (158 and 163 cm)and neither subject showed evidence of

malnutrition. Their diets before admissionconsisted predominantly of leavened flatbreads of the type under investigation.Both ate meat two or three times a weekand eggs about as often but consumedfruit daily in large amounts. Clinical andlaboratory examination disclosed no signsof disease.

The men were housed for the study in ametabolism ward along with other patientsand were assigned light duties in the wardand laboratory to provide exercise and decrease boredom. They were allowed toleave the hospital one afternoon each week.An enclosed garden adjoining the wardpermitted exposure to the sun about 30minutes daily. Three times each day theywere instructed to climb four flights ofstairs to further forestall effects of inactivity.

After admission, they received the general hospital diet for 1 week. The diet wasthen modified during the next 20 days sothat somewhat more than 50% of theirenergy intake was provided by white bread.A second 20-day period followed in whichBazari bread replaced the white bread.6Diets are described in table 1. Meals wereassembled by a trained assistant whoweighed the required amounts of foods. Aduplicate meal was prepared at the sametime and refrigerated for later analysis.Food quantities in the duplicate meal werethose required by subject Rah. The additional requirements of subject Mor basedon body weight were supplied by an extra100 g of bread daily. Corrections weremade for the infrequent occasions whena meal was not eaten completely.

Drinking water in central Iran is rich inminerals and metals and makes an appreciable contribution to total intakes. Measurements of water consumption proved tobe unreliable. Instead, intakes were esti-

«The Bazari bread was supplied by a local bakery.It was made from wheat flour»of 80% to 90% extraction lin- a sourdough leaven and is similar tothe Taf toon described by Kouhestani et al. (2).Samples of Bazari taken at random from that servedour subjects contained in air-dried samples : calcium,511 ±07 ; magnesium, 1.066 ±84 ; zinc. 11.1 ±3.2 ;phosphorus. 980 ±60 nig/kg (means ±so of six samples). Phytate concentrations averaged 0.35% andfiber 3.6%. Average loss of weight on drying in airat room temperature was 20.6%. The white breadserved was made in a small bakery by a double fermentation pan baking process from flour of about70% extraction rate. Bakers yeast was used as leaven,Analyes of six samples showed its composition to beclose to that tabulated for Western white bread.

by guest on August 23, 2015

jn.nutrition.orgD

ownloaded from

HIGH FIBER CONSUMPTON AND MINERAL ABSORPTION 495

mated from the 24 hour urine volume onthe assumption that water generated frommetabolism balanced insensible losses. Anadjustment was made for the 800 ml of teaand milk consumed daily. A single watercooler served as the source of drinkingwater which contained an average of 0.44mg of zinc, 40 mg of calcium and 42 mg ofmagnesium per liter.

Five hundred milligrams of chromicoxide were mixed into one dish of eachof the meals, starting 1 day before fecalcollections were started and continueddaily thereafter. The chromic oxide hadbeen purified by extraction with concentrated hydrochloric acid (11), pulverizedand passed through a 60 mesh per inchsieve.

Analytical procedures. The duplicatediets of 2 consecutive days were combinedand weighed. Distilled water (four timesthe food weight) was added and the mixture homogenized in a large blendor. Fecalcollections were made in widemouth polyethylene containers. Collections for 2 dayswere combined, weighed, diluted with anequal weight of distilled water and dispersed in a high speed vibrating paintmixer. Samples of diets and feces werepreserved with the aid of toluene and refrigeration in polyethylene containersshown by test to be free of leachable zinc.

Diet and feces samples were thoroughlydispersed before sampling with the aid ofa mechanical stirring apparatus. Using azinc-free polyethylene syringe, 4 to 6 dropswere transferred to tared acid washed digestion tubes or Kjeldahl flasks of 30 mlcapacity (for nitrogen). Tubes and flaskswere weighed and weight of samples calculated by the difference. HNO3 was addedto the contents of the digestion tubes together with acid-washed glass beads. Afterstanding overnight, the HNO:) was boiledoff and ashing completed with the aid of60% HC1O4. The HC1O4 in turn was fumedoff, leaving about 0.05 ml in the tube toinsure complete solubility in the 5.0 ml ofdeionized water added. After appropriatedilution with LaCl:j solution, concentrationsof Ca, Mg and Zn were measured by atomicabsorption spectrometry.7 Instrument settings used were those recommended bythe manufacturer ( 12 ). Plasma was diluted1:3 with doubly distilled water and urine

BreakfastBread,white,orBazariCheese,

whiteMilkSucroseNoon

mealBread,white,orBazariMutton,

leanRice,whiteDriedbeansCornoilI<ettuce,

etc.FruitSucroseEvening

mealBread,white,orBazariTurnips,

beets,carrotsoronionsPotatoes,

whit«Lettuce,etc.FruitSucroseTABLE

DietsWeightt15015050200151501505050601010010015200200100100100100101Pro

teing13141170131483110110171912110Fat01.50.510.07.001.50.54.00.30.710.00.20.102.00.70.30.10.20.10EnergykJ1,9001,8502101282501,9001,850460780710380802402502,5002,45018034080240167

TotalsSubjectRahWhite

bread periodBazari breadperiodSubjectMorWhite

bread periodBazari bread period90

9499

10433

3034

3010,775»

10,625«12,050'

11,860*

«58.3% of total energy supplied by bread.energy supplied by bread.

* 62.3% of total

acidified and diluted 1:10 or 1:20 with Lasolution. Serum phosphorus concentrationswere measured photometrically followingreduction of phosphomolybdate by meansof 2% ascorbic acid solution.

Samples for Cr analysis were ashed separately and Cr was measured as dichromateby the method of Bolin et al. (13) withmodifications recommended by Day ( 14 ) .The recovery of CrsOa in the feces averaged 1,425 mg during periods when nolosses of feces occurred. Fecal excretionsof the elements studied were corrected bymultiplying the factor : l,425/Cr2O3 whereCr.,0;! is the weight in mg of the Cr^O3found in a 24 hour collection. The loss ofsome Cr2O;i is attributed mainly to a smallresidue that escaped ingestion, however,others ( 10 ) have experienced small lossesof chromic oxide added to the diet as amarker.

TModel 303, The Perkin Elmer Corp., Norwulk,Conii., U.S.A.

by guest on August 23, 2015

jn.nutrition.orgD

ownloaded from

496 REINHOLD, FARADJI, ABADI AND ISMAIL-BEIGI

Fiber contents of diets and feces weremeasured by two methods. Originally, theWeende method (15) was used but thiswas replaced by the acid-detergent (A-D)method ( 16 ) when chemicals became available for the latter. The A-D method avoidsthe partial destruction of some fiber components caused by the treatment withalkali in the older method. It was foundthat the drying procedure was critical inapplication of the A-D method and thatlow values resulted when this was done at60°or above. Consequently, temperaturesduring drying were kept below 40°by aprocedure that included décantationof freewater after centrifugation and two successive extractions with acetone each involume roughly equal to that of the originalsample of diluted diet. The acetone alsowas removed by décantation after centrifugation, and finally by evaporation toconstant weight in a heating block at 40°.The weight of the dry residue provided anestimate of total solids. The residue wasextracted with diethyl ether for 4 hours ina Soxhlet. Ether was removed by evaporation and 1.0 g samples digested with 100ml of boiling cetyltrimethylammoniumbromide, 2% W/V in l N H2SO, for 1 hour.Filtration of fiber residue through sinteredglass was too slow to be practicable. Instead, the fiber was separated by centrifugation in tared glass centrifuge tubes. Itwas washed with boiling water, centri-fuged, and washed finally with acetone before being dried to constant weight at 60°.Fecal fiber was less susceptible to damageby heat and fecal samples could be driedat 60°without washing with acetone andwithout decrease in yield.

The Weende method was less troublesome in application. Filtration throughfritted glass caused no difficulty. Bothmethods demonstrated a marked increasein dietary and fecal fiber associated withthe consumption of Bazari bread. However, values yielded by the A-D methodwere nearly double those of the Weendemethod. Only the results obtained by theA-D method are presented.

Phytate concentrations were measuredaccording to Oberleas (17) except thatseparations were made by centrifugationinstead of filtration.

Ten milliliters of blood was collected

and half added to a polyethylene tube containing 1 drop of sodium citrate trihydrate,20% w/v. Plasma from this portion wasused for measurement of calcium, magnesium and zinc concentrations. The citrateprevents changes in zinc concentration thataccompany clotting and results in a higheryield of plasma with less likelihood ofhemolysis than does blood permitted toclot. The remainder of the blood was allowed to clot and its serum was used tomeasure alkaline phosphatase (18), albumin (19), total protein (20), and P concentrations. With the exception of P noneshowed significant changes during thestudy periods and the results are omitted.Iron intakes and excretions, serum iron andhemoglobin concentrations in blood werealso examined. These together with datafrom other experiments are to be presentedlater.8 Blood collections were made initiallyand at approximately weekly intervals.Nitrogen in diets and excreta was determined by a micro-Kjeldahl method usingmercury as a catalyst (22).

Statistical methods used arc described byKlugh (21).

RESULTS

Fiber intake and output. The mean fiberintake of subject Rah rose by 34% and thatof subject Mor by 54% when Bazari replaced white bread in their diets ( table 2 ).The increase in weight of fecal fiber wasless than that of dietary fiber intake. Discrepancies may be the result of partialdestruction of food fiber by the bacteria ofthe large gut. The extent of destructionvaries in different individuals (23).

Fecal output. The mass of feces increasedsubstantially in both subjects during theperiod of Bazari consumption as comparedwith that excreted during the white breadperiod (table 2). Wet and dry weightsboth rose substantially but the relationshipbetween the two differed in the two subjects. Although the increase in fecal drymatter was considerably larger than thatof fecal fiber, the two were closely related.

Zinc. Positive balances established during the period of white bread consumptionbecame negative when Bazari replaced the

»Faradji, B.. Abadi, P. & Reinhold. J. G. (1975)Impairment of iron absorption in man by diets withhlKh fiber and phosphorus contents. The action ofcellulose. In preparation.

by guest on August 23, 2015

jn.nutrition.orgD

ownloaded from

HIGH FIBER CONSUMPTON AND MINERAL ABSORPTION

TABLE 2Dietary and fecal fiber and fecal wet and dry weights

SubjectRahDal«Oct.

1618202224262830Nov.

13Mean±SKNov.

57911131517192123Mean±SK1

1.1fiberg/%4

hr181924272229201621.9±1.63024402424252631284529.7°±2.3Wetg/*4hr12516517519060195200235205240179±7.5280330290330325320305385320265318»±10.5FecesDryg¡Hhr141914261625182.")242424.4±1.52844323331333336443034.4«1±1.7Fiberg1*4hrLowfiberdiet2.63.93.66.23.05.13.74.86.17.04.6±1.52High

fiberdiet8.217.08.69.19.78.78.610.811.28.110.0»±0.84Dietfiberg/24

hr181924272229201621.9±1.6332743282S293037345134.0°±2.5SubjectWetg/24hr230240260310280350360370350340309±16.6470450510470450580480710290480489»±33MorFecesDryg/24hr59101821212118232417.0±2.12841514!)495945•

72324746.3»±3.8Fiberg/84

hr1.22.12.43.95.44.95.44.05.05.84.0±0.515.710.013.313.211.812.614.020.68.413.212.3»±1.24

Significance of difference between the two diets evaluated by the Student t test with adjustment for samplesize. •P<0.01. *P< 0.001.

white bread (figs. 1 and 2, table 3). Increased fecal excretion was responsible forthe change in balances. The mean balancesof zinc during the two periods differed significantly for both subjects. Plasma zincconcentrations were initially normal (87/<.g/100 ml in both). They had risen to 96and 94 /¿g/100ml in Rah and Mor respectively at the end of the study. Urinaryexcretion of zinc did not change.

Calcium. A significant increase in fecalexcretion of calcium occurred following thechange from white bread consumption toBazari in both subjects (figs. 1 and 2, table3). At the same time, excretion of calciumin urine decreased although not sufficiently to prevent a change in calciumbalances from positive to negative in sub

ject Rah and to more negative values insubject Mor. The changes took place despite moderately increased calcium intakes.Subject Mor's plasma calcium concentra

tions was 8.8 mg/100 ml at the start. Itrose to 9.6 mg/100 ml during the whitebread feeding then fell to 9.0 mg/100 mland finally to 8.8 mg/100 ml during thelast 2 weeks of high fiber intake. The lattervalue borders on the abnormal. Plasmacalcium concentration of subject Rah fluctuated between 8.8 and 9.0 mg/100 ml.

Magnesium. The high content of magnesium in Bazari led to a nearly doubledintake during its consumption comparedwith white bread. However, magnesiumexcretion in feces increased to a still greaterextent while that in urine also rose sub-

by guest on August 23, 2015

jn.nutrition.orgD

ownloaded from

498 REINHOLD, FARADJI, ABADI AND ISMAIL-BEIGI

stantially (figs. 1 and 2, table 3). Moderately negative magnesium balances resulted while Bazari was being consumed.

Phosphorus. Phosphorus intake increasedby approximately one-third with replacement of white bread by Bazari bread in thediet. This was accompanied by a threefoldrise in the excretion of phosphorus in the

feces (figs. 1 and 2, table 3). Excretion inthe urine by Subject Mor increased markedly and remained high when Bazari wasconsumed. Subject Rah did not respondwith increased phosphorus excretion inurine. Nevertheless, phosphorus balancesof both subjects changed from markedlypositive to negative with the change being

1000

20001-

Fig. 1 Metabolic balances of subject Rah during consumption of a mixed diet for 20 daysin which about 60% of the energy was supplied by white bread followed by a second 20-day-period when white bread was replaced in the same proportion by Bazari bread made fromwheat flour of high extraction rate. Each rectangle represents a 2-day interval. Units aremg/24 hour. Crosshatched areas show excretion in feces, clear areas in urine.

by guest on August 23, 2015

jn.nutrition.orgD

ownloaded from

HIGH FIBER CONSUMPTON AND MINERAL ABSORPTION 499

Fig. 2 Metabolic balances of subject Mor under the same conditions.

especially large in Subject Mor. In serum,Mor's phosphorus concentrations which

had risen from 3.0 to 3.5 mg/100 ml during white bread consumption, fell to 3.1mg/100 ml after 1 week of Bazari consumption and then rose again to 3.5 mg/100 ml. In Rah's serum, phosphorus con

centrations also rose from 3.5 to 4.1 mg/

100 ml during consumption of white bread.The change to Bazari feeding was followedby a decrease to 3.8 mg/100 ml after 1week and a later rise to 4.0 mg/100 ml atthe end of the study.

Nitrogen. Nitrogen excretion in feces increased by about 0.5 g daily in both subjects when Bazari replaced white bread in

by guest on August 23, 2015

jn.nutrition.orgD

ownloaded from

500 REINHOLD, FARADJI, ABADI AND ISMAIL-BEIGI

TABLE 3Metabolic behavior during low and high fiber consumption during ten 2-day periods

Subject Bread Intake Urine Feces Balance

Zincmg/24hrCalcium

mg/24hrMagnesium

mg/24hrPhosphorus

mg/24hrNitrogen

K 24 hrRahMorRahMorRahMorRahMorRahMorWhite

BazariWhite

BazariWhite

BazariWhite

BazariWhite

BazariWhite

BazariWhite

BazariWhite

BazariWhite

BazariWhite

Bazari18.1±

0.81'19.0±0.3619.0±

0.8119.4±0.41733

±19.7784±17.1»772

±15.7841±18.2"348

±12.0650±19.5*380

±10.6724 ±41.7Ò1288

±49.41740±58.0*-1400

±511940±58615.5±

0.316.0±0.3°17.0±

0.317.5± 0.30.7

0.70.7

0.8154

±10.599 ±4.6»187

± 7.9128 ±5.0k170

± 6.9210±10.5»185

± 4.4235 ±&.Qk764

±51654±69"887

±241240±7.')"12.5±

0.412.5±0.414.8±

0.514.2± 0.5lf).8±

0.62'20.7 ±0.76"18.3±

0.9420. 1±0.45522

± 51846 ±20»625

±35835±326188

± 11.6484 ±34»191

±13.4618±44.3»346

± 381118±104»287

± 15948 ±81"2.3±

0.12.8±O.I62.4±

0.23.0± 0.1»+

1.6-2.4*0.0-

1.5«+

57-161»-

40-122'-

10-44»+

4-129»+

178-32°+226

-248»+

0.7+0.7-

0.2+ 0.3

1Mean±SEM. " Significance of differences between the means of thefiber (Bazari bread) periods estimated by the < test. P ^ 0.01. *P

low fiber (white bread) and high0.001.

the diet (table 3). The change in the meanexcretion was significant in both subjects.A small gain in nitrogen balance by Morwas not statistically significant. No changesoccurred in total protein or albumin concentrations of serum in either subject andboth were within the limits found inhealthy well-nourished persons.

Body weight. Subject Rah gained 0.8 kgand Mor lost 1.0 kg during the 40 daystudy.

DISCUSSION

Bazari bread contains somewhat morezinc and calcium and much more magnesium and phosphorus than white bread,yet our findings indicate that it is nutritionally inferior to the latter as a source ofthese nutrients. Poor availability, as shownby increased fecal losses of zinc, calcium,magnesium and phosphorus that accompanied Bazari consumption is the reason.Because Bazari and breads of similar com

position are major food staples of manyeconomically-deprived families, a decreasein availability could lead to suboptimalmineral nutrition with overt deficiencyduring childhood, adolescence and pregnancy when requirements for the metalsincrease. The higher intakes associatedwith consumption of Bazari are deceptive.Although these intakes exceeded requirements as presently defined for humans, netlosses of each of the four elements occurred while Bazari was being consumed.

The nutritional significance of the increased fecal losses will depend upon theirduration. Some evidence of adaptation maybe seen in Rah, for example, whose zincbalances became less negative during thefinal 10 days of Bazari consumption andwhose phosphorus balances also improved.However, no improvement in calcium andmagnesium balances occurred in eithersubject. The decrease in Mor's plasma cal

cium concentration is further evidence that

by guest on August 23, 2015

jn.nutrition.orgD

ownloaded from

HIGH FIBER CONSUMPTION AND MINERAL ABSORPTION 501

the effects of Bazari consumption were persisting. Our impression derived from similar experiments extending over periods of2 months or more is that adaptation isselective and incomplete.9'10 This belief issupported by the balance studies ofIranian villagers previously mentioned(10).

Interference with absorption of bivalentmetals by cereal grains has been attributedto their high content of phytate. Poorlysoluble complexes of phytate with bivalentmetals exist in wheat (24). However,phytate is digestible by the rat " andalmost certainly in man, although the extent of its digestibility in man remains tobe determined. Nevertheless, the phosphorus released by digestion of phytatecan also form complexes with bivalentmetals and decrease their availability.

Several findings create doubt as towhether phytate is the sole or even themost important complexant of bivalentmetals in wheat flours of high extractionrates. An increase in solubility and availability of zinc in wholemeal bread as a result of the action of yeast leaven far exceeded expectations based upon observeddestruction of phytate (25). Studies invitro in which phytate was removed fromwholemeals or bran by extraction withacid or by action of phytase(wjeso-inosi-tol-hexaphosphate phosphohydrolase, EC3.1.3.8) brought about increased bindingof metals instead of the expected decrease(23). These and other observations led usto conclude that the fiber of wheat waslargely responsible for metal binding. Cellulose has been shown to interfere withabsorption of zinc (26).

The potent effects of Bazari consumption, a bread of moderate phytate content,upon mineral metabolism are similar tothose observed when Tanok, a villagebread of much higher phytate but similarfiber content, was consumed (27).

While fiber intake increased by one-third and one-half respectively in the twosubjects when Bazari replaced white breadin the diet, outputs of fiber in feces increased twofold and threefold. It appearsfrom this that fiber of wheat is more resistant to degradation by the intestinalflora than is that derived from other foods.However, it is also possible that an in

creased fiber mass in the presence of alimited capacity for degradation may explain these effects.

The increase in fecal wet weight thatfollowed the rise in fiber intake differed inthe two subjects, a difference that mayalso depend upon the extent of bacterialdegradation of fiber in the large intestine.Presumably, production of water-solublebut unabsorbable saccharide degradationproducts from fiber are important determinants of fecal wet weight. Polysaccha-rides of this type would escape measurement by the procedure used for fiberanalysis.

A correlation between fecal dry matterand excretion of zinc has been observed(28). The relationship is confirmed by ourdata. In addition, these show a similarclose relationship between dry matter andcalcium and magnesium excretion in feces.However, the content of dry matter infeces depends directly upon the fiber intakeand fiber content of feces. The fundamental relationship, therefore, appears tobe between metal and fiber. Highly significant correlations between fiber and metalexcretions are shown in table 4.

The change from white bread consumption to Bazari bread and the accompanying rise in phosphorus intake was followedby considerably increased fecal excretionof phosphate. Phosphorus, whether asphytate, its decomposition products, or inorganic phosphate readily forms poorlysoluble complexes with zinc at the pH ofthe intestinal contents (29) and presumably with calcium and magnesium. Correlations between fecal phosphorus and calcium, magnesium and zinc excretions, likethose of fiber, are also highly significant.One may conclude, therefore, that bothfiber and phosphorus complex metals in thesmall intestine and that phosphorus whichescapes absorption carries a quota of metalinto the large gut and the feces. Correlation between fiber and phosphorus in thefeces is not as close or as consistent as thatof either with the bivalent metals. Indeed,the coefficient of correlation became nega-

•Reinhold, J. G., Faradjl, B., Abadi, P. & Ismail-Belgi, F. (1974) Unpublished observation.

«Campbell, J., Relnhold, J. G., Cannell, J. &Nourmand, I. (1975) Pahlavi Med. J. (Shiraz,Iran). In press.

"Reinhold. J. G. & Faradji, B. (1975) Unpublished observation.

by guest on August 23, 2015

jn.nutrition.orgD

ownloaded from

502 REINHOLD, FARADJI, ABADI AND ISMAIL-BEIGI

TABLE 4

Relationships between components of feces

Subject

Rahr°

Mor

Dry matter vs.fiberDrymatter vs. wetweightCalcium

vs.fiberMagnesiumvs.fiberZinc

vs.fiberCalciumvs.zincCalciumvs. totalphosphorusMagnesium

vs. totalphosphorusZincvs. totalphosphorusTotal

phosphorus vs.fiberLowfiberperiodHighfiber period0.940.940.750.760.560.880.620.950.860.69-0.23<0.001<0.001<0.001<0.0010.005<0.001<

0.005<0.001<0.0010.001>0.100.990.930.660.890.430.960.850.950.540.360.38<

0.001<0.001<0.001<

0.0010.025<0.001<

0.001<0.001<0.005>0.050.05

1Coefficient of correlation. *Probability that the association is due to chance.

tive in subject Rah during the period ofBazari consumption.

Increased intakes of fiber as wheat flourare accompanied by a decrease in digestibility (30), and this association is alsoshown by the increased fecal nitrogen excretion that occurred. Highly significantcorrelations between fecal fiber and nitrogen excretion existed in both subjects.

ACKNOWLEDGMENTS

This study was assisted by grants fromthe Pahlavi University Research Counciland the Nutrition Foundation, Inc., NewYork.

We thank Mrs. S. Sahani, dietician, forformulating the diets and supervising theirpreparation. Ms. P. Ghahramani, Mr. I.Nourmand, Ms. F. Tadoyan, and Ms. V.Taleban provided able technical assistance.

LITERATURE CITED

1. Reinhold, J. G. (1975) Phytate destructionby yeast fermentation in whole meals. J.Am. Diet. A. 68, 38-41.

2. Kouhestani, A. H., Ghavifeker, H., Rahmanian,M., Mayurian, M. & Sarkissian, N. (1969)Composition and preparation of Iranian breads.J. Am. Diet. A. 55, 262-266.

3. Reinhold, T. G. (1972) Phytate concentrations of leavened and unleavened Iranianbreads. Ecol. Food. Nutr. 1, 187-192.

4. Maleki, M. ( 1972 ) Food consumption andnutritional status of 13-year-old village andcity schoolboys in Fars Province in Iran. Ecol.Food. Nutr. 2, 39-42.

5. Reinhold, J. G. (1971) High phytate concentrations of rural Iranian bread: a possible

cause of human zinc deficiency. Am. J. Clin.Nutr. 24, 1204-1208.

6. Ronaghy, H. A., Reinhold, J. G., Mahloudji,M., Ghavami, P., Spivey Fox, M. R. & Halsted,J. A. ( 1974 ) Zinc supplementation of malnourished schoolboys in Iran: increasedgrowth and other effects. Am. J. Clin. Nutr.27, 112-121.

7. Haghshenass, M., Mahloudji M., Reinhold,J. G. & Mohammadi, N. (1972) Iron deficiency anemia in an Iranian population associated with high intakes of iron. Am. J.Clin. Nutr. 25, 1143-1146.

8. Reinhold, J. G. (1975) Zinc and mineraldeficiencies in man: the phytate hypothesis.In: Proc. 9th Int. Cong. Nutr., Mexico 1972.(Chavez, A., Bourges, H. & Basta, S., eds.),1, 115-122.

9. Halsted, J. H., Ronaghy, H. A., Abadi, P.,Haghshenass, M., Amirhakimi, R. M., Barakat,R. M. & Reinhold, J. G. (1972) Zinc deficiency in man. The Shiraz experiment. Am. J.Med. 53, 277-284.

10. Reinhold, J. G., Hedayati, H., Lahimgarzadeh,A. & Nasr, K. (1973) Zinc, calcium, phos-Ehorns, and nitrogen balances of Iranian vil-

igers following change from phytate-rich tophytate-poor diets. Ecol. Food. Nutr. 2, 157-162.

11. Whitby, L. G. & Lang, D. (1960) Experience with the chromic oxide method of fecalmarking in metabolic balance investigationsin humans. J. Clin. Invest. 39, 854-863.

12. Atomic Absorption Spectrophotometry. ( 1968)The Perkin-Elmer Corp., Norwalk, Conn.,U.S.A.

13. Bolin, D. W., King, R. P. & Klosterman, E.W. (1952) A simplified method for determination of chromic oxide (Cr^O»)when usedas an index substance. Science. 116, 634-637.

14. Day, K. M. ( 1954) Source of error in determination of chromic oxide with perchloric-

by guest on August 23, 2015

jn.nutrition.orgD

ownloaded from

HIGH FIBER CON'SUMPTON AND MINERAL ABSORPTION .-,015

sulfuric acid digestion method. Science 120,717-718.

15. Official Methods of Analysis (1960) Assoc.Off. Anal. Chem. 9th ed. p. 288.

16. Supplement to the llth eel. Official Methodsof Analysis (1973) Assoc. Off. Anal. Chem.p. 463.

17. Oberleas, D. (1971) The determination ofphytate and inositol phosphates. Methods Bio-chem. Anal. 20, 87-101.

18. Bessey, O. A., Lowry, O. H. & Brock, M. J.(1945) Determination of alkaline phospha-tase activity in 5 cubic microliters of serum.J. Biol. Chem. 164, 321-329.

19. Doumas, B. T. & Biggs, H. G. (1973) Determination of serum albumin. In Stand.Methods Clin. Chem. 7, 175-188.

20. Kingsley, G. R. (1942) The direct biuretmethod for determination of serum proteinsas applied to photoelectric and visual colorim-etry. J. Lab. Clin. Med. 27, 840-845.

21. Archibald, R. (1958) Nitrogen by theKjeldahl Method. Stand. Methods. Clin. Chem.2, 91-99.

22. Klugh, H. E. (1970) Statistics: Essentialsfor Research. Wiley & Sons, New York, N.Y.

23. Sonthgate, D. H. T. & Durnin, J. U. G. A.(1970) Caloric conversion factors. An experimental reassessment of the factors used incalculation of energy values in human diets.Brit. J. Nutr. 24, 517-535.

24. Reinhold, J. G., Faradji, B. & Ismail-Beigi, F.(1975) Fiber vs. phytate as determinant ofthe availability of calcium zinc, and iron inbreadstuff«. Nutr. Kept. Int. 12, 75-85.

25. Reinhold, J. G., Parsa, A., Karimian, N., Ham-mick, J. W. & Ismail-Beigi, F. ( 1974) Availability of zinc in leavened and unleavenedwholemeal breads as measured by solubilityand uptake by rat intestine in vitro. J. Nutr.104, 976-982.

26. Becker, W. M. & Hoekstra, W. G. (1972)In: Intestinal Absorption of Metal Ions, TraceElements, and Nuclides. ( Skoryna, S. C. &Waldron-Edwards, D., eds.), Pergamon Press,Oxford, New York.

27. Reinhold, J. G., Nasr, K., Lahimgarzadeh, A.& Hedayati, H. (1973) Effects of purifiedphytate and phytate-rich bread upon metabolism of zinc, calcium, phosphorus, and nitrogen in man. Lancet 1, 283-288.

28. Schraer, K. K. & Calloway, D. H. (1974)Zinc balance in pregnant teenagers. Nutr.Metabol. 17, 205-212.

29. Vohra, P. & Kratzer, F. H. ( 1966) Influenceof various phosphates and other complexingagents on the availability of zinc for turkeypoults. J. Nutr. 59, 106-112.

30. Moran, T. & Pace, J. (1942) Digestibilityof high extraction wheat meals. Nature ISO,224.

by guest on August 23, 2015

jn.nutrition.orgD

ownloaded from