Age and copper intake do not affect copper absorption, measured with the use of 65Cu as a tracer

Manuel Olivares,Bo Lönnerdal,Steve A Abrams,Fernando Pizarro,and Ricardo Uauy

ABSTRACT

controlled copper absorption is in the first months of life is crucialBackground:Copper homeostasis involves a high degree of reg-to determine whether normal infants are at risk of copper excessulation in which changes in absorption and biliary excretion arewithin the range of acceptable intakes (6,7). The aim of this studythe main mechanisms. Whether neonates and small infants canwas to evaluate copper absorption in infants during the first 3 mo ofmake these changes efficiently is unknown.

life and the effects of age and copper intake on copper absorption.

Objective:We evaluated the effect of age and copper intake oncopper absorption in infants during the first 3 mo of life.

Design:Thirty-nine healthy infants (19 infants aged 1 mo and 20SUBJECTS AND METHODS

infants aged 3 mo) were selected. One-half of the subjects wereThirty-ninehealthyinfants(19infantsaged1moand20infantsrandomly assigned to receive oral supplementation of 80mg Cuaged3mo)withabirthweight>2500gwereselected.One-halfof(as copper sulfate)·kg body wtϪ1·dϪ1for 15 d. At the end of thethesubjectswererandomlyassignedtoreceiveoralsupplementa-trial,copper absorption was measured by using orally adminis-tionof80mgCu(ascoppersulfatesolution)·kgbodywtϪ1·dϪ1tered 65Cu as a tracer and fecal monitoring of recovered 65Cu.(1.2␮mol·kgbodywtϪ1·dϪ1)for15d.Adailydoseof0.3mLResults:Mean (±SD) copper absorption at 1 mo of age wassolution/kgbodywt(commonly1–1.5mL)wasadministeredto83.6±5.8% and 74.8±9.1% for the unsupplemented and sup-theinfantsbetweenfeedingsbytheirmothers:a3mLsyringewasplemented infants,respectively. The corresponding figures at 3 moplacedatthebackoftheinfant’stonguebecausethetasteoftheof age were 77.6±15.2% and 77.7±11.3%. A two-way analysiscoppersolutionwasnotmasked.Toassesscompliancewiththeof variance showed that age,copper supplementation,and theinstructionsprovidedbytheinvestigators,afieldworkervisitedinteraction between age and copper supplementation did not haveeachinfant’shometwiceaweekatarandomtimetorecordthevol-a significant effect on copper absorption. There was an inverseumeofthecoppersolutionremaininginthebottle.Inaddition,thecorrelation between total fecal copper and the percentage of 65Cufieldworkerconductedonefood-frequencyquestionnaireforeachabsorption (r= Ϫ0.50,P < 0.003).

infanttoobtainthevolumeofcowmilkorformulareceived.Conclusion:Copper absorption in young infants is high but doesBecausetheconcentrationofcopperinbreastmilkisnotaffectednot respond to copper intake within the range tested. Am Jbymaternaldietorethnicfactors(8),wecalculatedthecopperClin Nutr2002;76:1–5.

intakeofbreast-fedinfantsonthebasisofdatapublishedbyButteetal(9)forhumanmilkcompositionandvolumeofintakebyage.KEY WORDSCopper absorption,human milk,infants,Beforetheinfantswereincludedinthestudy,informedconsentcopper supplementation,65Cu

wasobtainedfromtheirparents,andthestudyprotocolwasapprovedbytheEthicsonHumanResearchCommitteeoftheInsti-tuteofNutritionandFoodTechnologyoftheUniversityofChile.INTRODUCTION

The stable 65Cu isotope,which was 99.7% enriched (Oak RidgeA considerable amount of copper is transferred from the mother toNational Laboratory,Oak Ridge,TN),was obtained in the formthe fetus. Copper is accumulated in the fetus mainly at the end of theof copper wire,which was weighed,dissolved in a minimum of

gestation period,and a substantial portion of the accumulated copperis retained in the liver of the fetus (1). The substantial stores of copperin the liver of the term fetus may aid in preventing copper deficiency1

From the Institute of Nutrition and Food Technology,University of Chile,San-during the early months of life. After birth,liver copper concentrationtiago (MO,FP,and RU); the Department of Nutrition,University of California,decreases and serum copper and ceruloplasmin concentrations increaseDavis (BL); and the US Department of Agriculture/Agricultural Research Serviceand become more similar to those in adult serum (2,3).

Children’s Nutrition Research Center,Baylor College of Medicine,Houston (SAA).Copper homeostasis is believed to involve changes in both intes-2

Supported by the Copper Risk Assessment Research Program in Chile,whichtinal absorption and biliary excretion (4). Studies in human adultsis managed by the Chilean Center for Mining & Metallurgy Research,and thein which stable isotopes were used showed an inverse relationInternational Copper Association in the form of an unrestricted research grant.3

between copper intake and copper absorption (5). However,thereAddress reprint requests to M Olivares,Institute of Nutrition and Foodis insufficient information to ascertain whether the efficiency of cop-Technology (INTA),Macul 50,Santiago 11,Chile. E-mail:molivare@per absorption in infants is up- and down- regulated at low and highuec.inta.uchile.cl.

Received May 11,2001.

copper intakes,respectively. The answer to the question of how well

Accepted for publication November 2,2001.

Am J Clin Nutr2002;76:1–5. Printed in USA. ©2002 American Society for Clinical Nutrition

1

Downloaded from ajcn.nutrition.org by guest on November 11, 20132OLIVARES ET AL

ultrapure nitric acid and hydrochloric acid (Merck,Darmstadt,

Germany),and diluted with distilled,deionized water. The feed-ing solutions were made in one batch and diluted so that each dosehad 30 or 200mg 65Cu. Single doses were weighed in test tubesand stored frozen until the day of use.

Atrainedfieldworkeradministeredtheisotopetoeachinfantinasmallfeedingbottle.Onday1,eachinfantreceivedoneoraldoseof200mg(3.1␮mol)65Cu(supplementedgroup)or30mg(0.5␮mol)65Cu(unsupplementedgroup)dissolvedin10mLdistilled,deionizedwater.Thebottlewasthenrefilledandrinsedwith5mLadditionaldistilled,deionizedwater,whichwasagainadministeredtotheinfant.Theisotopewasgivenimmediatelyafterthesecondfeedingoftheday(1000to1100).

On day 1 and 72h later,infants received in a feeding bottle 150mgBrilliant Blue FCF marker (Foodsafe,Santiago,Chile) in 20 mL dis-tilled,deionized water to mark the beginning and end of the stool col-lection period. To ensure the quality of sampling and to avoid the lossof all or part of the samples,mothers were carefully trained beforethe study in how to collect the fecal samples from their infants. Whenthe mothers’ability to properly collect fecal samples was tested at theend of the training period,no differences were observed between themothers in the 4 study groups (ie,1-mo-old supplemented infants,1-mo-old unsupplemented infants,3-mo-old supplemented infants,and3-mo-old unsupplemented infants). Mothers were provided daily withdiapers that were labeled with an indelible code to document thesequence of diaper use. The field worker visited each infant’s homedaily to pick up the used diapers and check the sequence to assurethat all stools were collected. Stool was extracted from each diaperwith a wooden spatula in the laboratory,and the stained liner was cutout. The Brilliant Blue and the diaper liner were checked for coppercontent to verify negligible amounts of copper. Daily pools of stoolsand liners were prepared. Glassware was acid-washed before use toreduce the risk of copper contamination,and random tests of theglassware confirmed that it was free of trace elements at the time ofuse. Fecal pools were weighed in porcelain crucibles,heated in a muf-fle furnace for 7h at 90ЊC,and then ashed by heating for 12h at600ЊC. Later,50 mL of a 5-mol HNO3/L solution (Merck) was addedto the crucibles,and each crucible was covered with a watch glass.The liquid was evaporated on a hot plate in a clear plastic box undera hood. Samples were returned to the muffle furnace for an additional12h at 600ЊC,removed,covered,and cooled. The resulting ashwas dissolved in 10 mL of a 2-mol HNO3/L solution (Merck) forlater analysis.

A glass chromatography column (1 ϫ20cm),with a 40-mL reser-voir in the top,was filled with cation exchange resin (AG 50W X8,100–200 mesh; Bio-Rad Laboratories,Richmond,CA),which waswashed with 30 mL of a 6-mol ultraclean HCl/L solution and with20 mL distilled,deionized water and was reconditioned with 30 mL60% (by vol) acetone. Samples were loaded onto the column with 4mL 60% acetone and passed through the column,which was thenwashed first with 30 mL 60% acetone and then with 30 mL 75% ace-tone. Copper was extracted from the column with 30 mL 90% ace-tone. The solution collected was dried on a hot plate at a subboil-ing temperature,resuspended,and redissolved in 1–2 mL of a6-mol HCl/L solution to be used as the sample for the next anionexchange procedure. A polyethylene column (8cm ϫ0.4cm)with a 4-mL reservoir in the top was filled with anion exchange resin(AG-1 X8,100–200 mesh; Bio-Rad Laboratories). The resin in thecolumn was cleaned with 4 mL of a 6-mol ultraclean HCl/L solu-tion and with 4 mL distilled,deionized water. The column was then

TABLE 1Characteristics of the study groupsSupplementedUnsupplemented1 mo of agen109Initial age (d)29.6±1.5133.3±4.92SexMale63Female46Birth weight (g)3501±38039±532Birth weight (zscore)30.5±1.00.7±1.1Weight (g)4209±4824451±521Weight-for-age (zscore)30.1±0.70.6±0.73 mo of agen1010Initial age (d)90.5±7.7.6±12.6SexMale85Female2Birth weight (g)3598±4093475±522Birth weight (zscore)30.9±1.00.4±1.0Weight (g)6346±7286375±1045Weight-for-age (zscore)30.8±0.90.6±1.11–x±SD.2Significantly different from the supplemented group,P< 0.04.3Calculated from National Center for Health Statistics standards.reconditioned in 2 mL of a 6-mol HCl/L solution before the samplesolution was loaded. The sample was loaded with 1 mL of a 6-molHCl/L solution. After the sample solution had passed through the col-umn,it was washed with 6 mL of a 6-mol HCl/L solution before cop-per was extracted from the column with 4 mL of a 3-mol HCl/L solu-tion. The solution collected was dried on a hot plate at a subboilingtemperature,resuspended in 0.03 mol HNO3,and loaded onto the fil-ament for mass spectrometric analysis.

Copper content was measured by atomic absorption spectrom-etry (model 2280; Perkin Elmer,Norwalk,CT),and the ratio of65

Cu to 63Cu was measured as described by Turnlund et al (10) withthe use of a magnetic-sector thermal-ionization mass spectrometer(MAT 261; Finnigan,Bremen,Germany). Measurement precisionfor the ratio of 65Cu to 63Cu was 0.1%. Apparent absorption resultsare based on changes in fecal copper isotope ratios. Indexes ofmass balance were not derived because copper intake was notdirectly measured. In the first 4 infants who were recruited,theratio of 65Cu to 63Cu was measured but the total stool copper wasnot,because the samples for these latter measurements were lost.Values in the tables and the text are given as means±SDs. Statisti-cal analysis included two-way analysis of variance,Student’sttest,stepwise multiple regression,linear regression,and Pearson corre-lation. When analysis of variance was statistically significant,iden-tification of significant differences between groups was based onScheffe’s post hoc test. Statistical analyses were performed by usingSTATISTICA for WINDOWS (release 4.5; StatSoft Inc,Tulsa,OK).RESULTS

Except for initial age,the characteristics of the subjects inthe supplemented and unsupplemented groups were not signifi-cantly different (Table1). Twenty-eight of the 39 infants wereexclusively breast-fed,9 were partially breast-fed,and 2 weretotally weaned. Total derived copper intakes in the unsupplementedand supplemented infants at 1 mo of age were 60.3±0.2␮g·kgbody wtϪ1·dϪ1(268.2±30.2␮g/d; 0.9±0.0␮mol·kg body

Downloaded from ajcn.nutrition.org by guest on November 11, 2013COPPER ABSORPTION IN YOUNG INFANTS

3

TABLE 2Fecal copper and apparent 65Cu absorption in copper-supplemented andunsupplemented young infants1SupplementedUnsupplementedFecal copper21 mo of age(␮g/3 d)743.2±328.0 [9]333.5±113.2 [9](␮mol/3 d)11.7±5.2 [9]5.2±1.8 [9]3 mo of age(␮g/3 d)710.4±288.0 [9]507.1±294.1 [8](␮mol/3 d)11.2±4.5 [9]8.0±4.6 [8]Apparent 65Cu absorption (%)31 mo of age74.8±9.1 [10]83.6±5.8 [9]3 mo of age77.7±11.3 [10]77.6±15.2 [10]1–x±SD.2Significantly affected by copper supplementation but not by age or theinteraction between copper supplementation and age,P< 0.05 (two-wayANOVA).3Not significantly affected by copper supplementation,age,or the inter-action between copper supplementation and age (two-way ANOVA).wtϪ1·dϪ1; 4.2±0.5␮mol/d) and 86.3±1.9␮g·kg bodywtϪ1·dϪ1(362.4±33.8␮g/d; 1.4±0.0␮mol·kg body wtϪ1·dϪ1;5.7±0.5␮mol/d),respectively. The corresponding figures at 3 moof age were 35.6±0.3␮g·kg body wtϪ1·dϪ1(226.8±35.5␮g/d;0.6±0.0␮mol·kg body wtϪ1·dϪ1; 3.6±0.6␮mol/d) and55.9±1.7␮g·kg body wtϪ1·dϪ1(353.9±32.0␮g/d; 0.9±0.0␮mol·kgbody wtϪ1·dϪ1; 5.6±0.5␮mol/d),respectively.

The effect of age and of copper supplementation on totalfecal copper and apparent 65Cu absorption was tested by two-way analysis of variance (Table2),which showed a signifi-cant effect of copper supplementation on total fecal copper.However,apparent 65Cu absorption was not significantly dif-ferent by age group or copper supplementation group. In theentire group of infants,fecal copper and apparent 65Cuabsorption for the 3 d ranged from 178 to 1347mg Cu (from2.8 to 21.2␮mol Cu) and from 46% to 95%,respectively.Copper absorption in exclusively breast-fed,partially breast-fed,and totally weaned infants was 79.7±10.2%,78.5±13.3%,and 58.6% (no SD because there were only 2 infants in thiscategory),respectively.

Copperintakewasnotsignificantlycorrelatedwiththeper-centageofcopperabsorption.AsshowninFigure1,therewasaninverse,simplecorrelationbetweenfecalcopperandthepercent-ageof65Cuabsorption(r=Ϫ0.50,P<0.003).Whenfecalcop-per,typeoffeeding,copperintake,andagewereincludedasinde-pendentvariables,stepwisemultipleregressionforthepercentageofcopperabsorptionindicatedR2=0.29(P<0.02).Ofthevari-ablestested,onlyfecalcoppermetthesignificancelevelforentryintothemodel(P<0.02).

DISCUSSION

Copper homeostasis is believed to involve a high degree ofregulation in which modifications in absorption and endoge-nous excretion are the main mechanisms. Whether neonatesand small infants can make these modifications efficiently isunclear.

Composition of the diet,copper intake,and copper nutritionalstatus influence copper absorption (5,10–12). The type of feed-

FIGURE 1.Correlation between fecal copper and the percentage of65

Cu absorption in young breast-fed (᭹),partially breast-fed (᭺),andweaned (᭞) infants. r= Ϫ0.50.

ing and the amount of copper supplied affect copper balance inearly life. We observed a high apparent copper absorption(Ϸ80%) in infants aged 1–3 mo. This effect may be the conse-quence of breast-feeding or a developmental phenomenon.Breast-fed infants absorb more copper,perhaps because of thelower casein content of human milk or because of factors asso-ciated with human milk that enhance copper absorption (13). Inthe present study,copper absorption in breast-fed and partiallybreast-fed infants was higher (79.7±10.2% and 78.5±13.3%)than in 2 totally weaned infants who received only cow milk(58.6%). There is a basis for attributing an absorption-enhanc-ing effect to breast milk that could account for the absorptionof exogenous 65Cu as well. Although the isotope in the presentstudy was administered after the meal,the high absorption couldalso have been due,in part,to the isotope being administered inwater rather than in milk. Using the suckling rat pup model,Lönnerdal et al (13) found a higher copper absorption fromhuman milk than from cow milk. In chemical balance studiesand absorption studies in which stable isotopes were used,infants had higher retention and absorption of copper fromhuman milk than from cow milk formula (14,15). 65Cu absorp-tion from human milk in premature infants aged Ϸ1 mo wassignificantly greater (69.8±14.0%) than 65Cu absorption fromformula in premature infants (39.6±21.6%) and full-terminfants (26.5±6.9%) (14). Studies in rats showed that copperabsorption is very high during the neonatal period but that itdecreases by the time of weaning (13). It is possible that thisalso occurs in humans,but studies by Turnlund et al (5,10)showed that copper absorption is high in adults as well,evenwhen copper intakes are very low. We found that the percent-age of apparent copper absorption at 1 mo of age was modestly,but nonsignificantly lower in the infants who were supple-mented with copper than in the unsupplemented infants,but thispattern was not observed at 3 mo of age. The dose of 65Cu givento the supplemented infants was Ϸ7-fold that given to theunsupplemented infants; hence,the apparent copper absorptionin the supplemented group may have been even lower.

Thelackofamajordifferenceinthepercentageofcopperabsorptionobservedbetweenthesupplementedandunsupple-mentedinfantsmaybeexplainedindifferentways.Oneexpla-nationisthatcopperintakewasnothighenoughtotriggerhome-ostaticadaptationofintestinalabsorption.Infact,Turnlundetal(5)

Downloaded from ajcn.nutrition.org by guest on November 11, 20134OLIVARES ET AL

observedthatwhenthecopperintakeofadultswas10-foldthatofthepreviousintake,thepercentageofcopperabsorptiondecreased,whereasMilneetal(11)showednosucheffectwhenthecopperintakeofadultswas1.5-foldthatofthepreviousintake.Thecopperintakeofoursupplementedsubjectswas1.4-to1.5-foldthatoftheunsupplementedinfants.Anotherexplanationisthatthelatencyincopperabsorptionadaptationmaybe>15d.Athirdexplanationistheinabilityofyounginfantstoregulatecopperabsorption.Studiesinanimalsandhumanssupporttheexistenceofdevelopmentalchangesincopperabsorption.Inkineticstudies,Varadaetal(16)foundthatcopperabsorptionwassaturableonlyinadolescentrats,whereascopperabsorptionwaslinearandnonsaturableinsucklingandweanlinganimals.Asimilarefficiencyofcopperabsorptionwasobservedinratswhenthecoppercontentofhumanmilkwasincreasedto10timesthenormalcontentbytheadditionofacoppersalt(13).Furthermore,Dörneretal(15)foundalinearrelationbetweencopperintakeandcopperretentioninbalancestudiesininfants,supportingthesuggestionfromstudiesinratsthatcopperabsorp-tionisnonsaturableduringearlyinfancy.Weobservedaninverselinearrelationbetweenfecalcopper(anindirectmeasureofcop-perintake)andthepercentageofcopperabsorption.Itcouldbespeculatedthatinfants,giventhesmallertotalabsorptivesurfaceoftheirintestines,mayhaveasubstantialresidualabsorptivecapacityandthusmaybeatagreaterriskofcopperoverloadthanareolderindividuals.

Copper homeostasis is mediated not only by absorption butalso by modifications in the endogenous secretion of copper.However,the methodology used in our study did not allow esti-mations of the endogenous secretion of copper or its adaptation.The high apparent absorption may also be explained by a sys-tematic error in the balance technique,because more completestool collections will necessarily be associated with higher 65Curecovery from the feces and higher apparent absorption. We tookall possible measures to assure complete stool collection. How-ever,although Brilliant Blue is a good marker for stool collec-tion,it is not truly quantitative. Continuous markers are better butare harder to detect.

Our study was conducted in the subject’s homes and thus mayhave been affected by confounders that are better controlled instudies conducted in metabolic wards under rigidly controlledconditions. The ethics standards at our institutions restrict bal-ance studies in the metabolic ward to those that are absolutelyessential for the medical care of infants and do not allow balancestudies in healthy infants. Thus,the possibility of incompletestool collection is a limitation of this study. However,becausethis study was essentially a paired study,there is no reason tothink that incomplete stool collections affected one group morethan the other.

The results of our study show that copper absorption was highlyvariable (ranging from 46% to 95%) under the conditions of thisstudy and that absorption is not down-regulated within the rangeof copper intakes tested. These intakes were 2- to 3-fold the mostrecent adequate intake defined for 1-mo-old breast-fed infants bythe Food and Nutrition Board of the National Academy of Sci-ences and 1.2- to 2-fold the adequate intake defined for 3-mo-oldbreast-fed infants (7). It may be possible that down-regulation ofabsorption does occur at higher intakes,which would protectagainst excess. On the contrary,if absorption remains high,toxi-city could be possible even at moderately high exposures. Our invitro results suggest that copper absorption at the apical surface

is not down-regulated by chronic exposure to high copper con-centrations. We have shown that Caco-2 cells respond to copper byincreasing basolateral copper efflux. Thus,homeostatic control isnot dependent on regulation of copper uptake but rather on therelease from the enterocyte to the portal vein (17). This leads usto speculate that intestinal cell desquamation over time may con-tribute to whole-body copper homeostasis as copper stored inenterocytes is lost. The high percentage of apparent absorptionobserved in our study may be explained by trapping of copper inthe intestine and an incomplete release through desquamation overthe study period. Our stool collection period was 72h,whereasenterocyte turnover in humans takes 3–5 d (18). The experimen-tal design of the present study may be considered inadequatebecause intakes were not sufficiently high; yet,because of thepotential for increased susceptibility to excess copper in veryyoung infants,the ethical imperative of “do no harm”precluded usfrom testing higher intakes. Further research is required to eluci-date the full range of homeostatic regulation and the contributionof copper absorption to this process during the first months of life.Such elucidation is crucial to determine the safe range of copperexposure in young infants.

REFERENCES

1.Widdowson EM. Trace elements in foetal and early postnatal devel-opment. Proc Nutr Soc 1974;33:275–84.

2.Srai SKS,Burroughs AK,Wood B,Epstein O. The ontogeny of livercopper metabolism in the guinea pig:clues to the etiology of Wil-son’s disease. Hepatology 1986;6:427–32.

3.Klein D,Scholz P,Drash GA,Müller-Höcker J,Summer KH. Metal-lothionein,copper and zinc in fetal and neonatal human liver:changesduring development. Toxicol Lett 1991;56:61–7.

4.Linder MC,Hazegh-Azam M. Copper biochemistry and molecularbiology. Am J Clin Nutr 1996;63(suppl):797S–811S.

5.Turnlund JR,Keyes WR,Anderson HL,Accord LL. Copper absorp-tion and retention in young men at three levels of dietary copper byuse of stable isotope 65Cu. Am J Clin Nutr 19;49:870–8.

6.FAO/WHO/IAEA. Trace elements in human nutrition and health.Geneva:WHO,1996:123–43.

7.Food and Nutrition Board,Institute of Medicine. Dietary referenceintakes for vitamin A,vitamin K,arsenic,boron,chromium,copper,iodine,iron,manganese,molybdenum,nickel,silicon,vanadium,andzinc. Washington,DC:National Academy Press,2001.

8.Lönnerdal B. Effects of maternal dietary intake on human milk com-position. J Nutr 1986;116:499–513.

9.Butte NF,Garza C,O’Brian Smith E,Wills C,Nichols BL. Macro-and trace-mineral intakes of exclusively breast-fed infants. Am J ClinNutr 1987;45:42–8.

10.Turnlund JR,Keyes WR,Peiffer GL,Scott KC. Copper absorption,

excretion,and retention by young men consuming low dietary copperdetermined by using the stable isotope 65Cu. Am J Clin Nutr 1998;67:1219–25.

11.Milne DB,Johnson PE,Klevay LM,Sandstead H. Effect of copper

intake on balance,absorption,and status indices of copper in man.Nutr Res 1990;10:975–86.

12.Lönnerdal B. Copper nutrition during infancy and childhood. Am J

Clin Nutr 1998;67(suppl):1046S–53S.

13.Lönnerdal B,Bell JG,Keen CL. Copper absorption from human milk,

cow’s milk,and infant formulas using a suckling rat model. Am J ClinNutr 1985;42:836–44.

14.Ehrenkranz RA,Gettner PA,Nelli CM,et al. Zinc and copper nutri-tional studies in very low birth weight infants:comparison of stableisotopic extrinsic tag and chemical balance methods. Pediatr Res19;26:298–307.

Downloaded from ajcn.nutrition.org by guest on November 11, 2013COPPER ABSORPTION IN YOUNG INFANTS

5

15.DörnerK,DziadzkaS,HohnA,etal.Longitudinalmanganeseand

17.Arredondo M,Uauy R,González M. Regulation of copper uptake and

copperbalancesinyounginfantsandpreterminfantsfedontransport in intestinal cell monolayers by acute and chronic exposure.breastmilkandadaptedcow’smilkformulas.BrJNutr19;61:Biochim Biophys Acta 2000;1474:169–76.

559–72.

18.LipkinM.Proliferationanddifferentiationofgastrointestinal

16.VaradaKR,HarperRG,WapnirRA.Developmentofcopperintes-cellsinnormalanddiseasestates.In:JohnsonLR,ed.Physiologytinalabsorptionintherat.BiochemMedMetabBiol1993;50:ofthegastrointestinaltract.NewYork:RavenPress,1981:277–83.

145–68.

Downloaded from ajcn.nutrition.org by guest on November 11, 2013