243 The effects of phytate on intestinal absorption and secretion of zinc, and whole-body retention of Zn, copper, iron and manganese in rats BY N. T. DAVIES AND R. NIGHTINGALE Rowett Research Institute, Bucksburn, Aberdeen A232 9S3 (Received 18 November 1974 - Accepted 14 January 1975) I. The inclusion of phytate ( I o g(kg) in a purified diet containing zinc (I 5 mg/kg) fed to young male rats significantly reduced growth rate and food intake, and promoted a cyclic pattern of food intake characteristic of an uncomplicated Zn deficiency. The decreased growth rate could be accounted for by the reduced food consumption. 2. Rats maintained on a Zn-deficient diet (0.5 mg Zn/kg) were found to have a cyclic pattern of food intake and a very slight weight gain. The addition of phytate (10 gjkg) to the Zn-deficient diet promoted a net Ioss of mean body-weight. 3. Rats maintained on the Zn-supplemented diet containing phytate excreted significantly more Zn in their faeces than either pair-fed or ad Zib.-fed control rats. Rats given the Zn- deficient diet supplemented with phytate excreted more Zn in their faeces than Zn-deficient control rats. 4. Dietary phytate significantly reduced the average daily accumulation (pg/d) and whole- body retention (relative to dietary intake) of iron, copper, manganese and Zn, whether or not the diet was supplemented with Zn. 5. The addition of phytate to the lumen fluid of ligated loops of rat duodenum maintained in situ significantly inhibited 66Zn absorption, compared with the control systems without added phytate, 6. Other studies using ligated duodenal and ileal loops indicated that Zn is secreted into the gut lumen and approximately one-third of this is normally reabsorbed. Recycling of endo- genous Zn may be a significant process in the over-all body economy of this trace element. 7. The absorption of 65Zn added to the diet was significantly reduced by dietary phytate. I h post-admini- Dietary phytate also reduced the biological half-life of body 65Zn from 91 to 21 stration, possibly by inhibiting reabsorption of endogenous 65Zn and thus promoting a more rapid loss from the body. O’Dell & Savage (1960) were the first workers to suggest that naturally occurring phytic acid (myo-inositol 1,2,3,4,5,6-hexakis dihydrogen phosphate) in plant protein might reduce the availability of dietary zinc. This proposal seems amply justified in view of the numerous reports of impairment of Zn utilization when phytate-rich diets have been fed to a number of different species, e.g. chicks (Likuski & Forbes, 1964), pigs (Oberleas, Muhrer & O’Dell, 1962) and rats (Oberleas, Muhrer & O’Dell, b). A plausible mechanism suggested by Oberleas et aZ. (1966a, b) proposes 1966~2, that Zn is removed from solution in the intestinal lumen by co-precipitation with calcium and phytate as an unavailable Zn-Ca-phytate complex, although in few studies so far reported have direct effects of phytate on Zn absorption been found in vivo. In view of these findings, the work reported here was directed towards studying the action of phytate on Zn absorption both in experiments using ligated loops of rat duodenum in situ, and from an assessment of percentage absorption of tracer doses of 65Zn in whole-body retention studies, using the method of Heth & Hoekstra (1965). 244 N. T. DAVIES AND R. NIGHTINGALE I975 The results of recent studies of the intestinal handling of Zn by rats indicate that it is both rapidly absorbed (Davies, 1973; Methfessel & Spencer, 1973a) and secreted (Methfessel & Spencer, 1[973b) and that the mechanism of secretion and reabsorption may play an active part in the over-all absorption of Zn. Results of experiments re- ported in the present paper, involving injection of a metal-binding resin into ligated segments of rat intestine, may be interpreted by the suggestion that Zn is continually secreted and reabsorbed across the intestinal mucosa. The possibility that dietary phytate might affect over-all Zn utilization both by affecting the reabsorption of endogenously secreted Z;n and by direct action on the absorption of dietary Zn was thus studied. EXPERIMENTAL Animals In all experiments male Hooded Lister rats of the Rowett Institute strain were used. Throughout the treatment periods rats were housed individually in cages constructed of polypropylene and stainless steel. Diets In the dietary experiments, rats weighing 66-5-75.5 g were transferred from a commercial rat cube diet to a semi-purified basal diet similar in composition to that of Williams & Mills (1970) containing (g/kg) : spray-dried egg albumin 200, sucrose 660, arachis oil 100, plus vitamins and minerals, except that the Ca content was increased to 13 g/kg by addition of extra CaCO,. This basal diet supplied trace metals iron 50, manganese 52. in the following concentrations (mg/kg): Zn 0.5, copper 25, When phytate was added to the diet, the amount of sodium phytate (Sigma Chemical Corp., St Louis, USA) included gave a dietary phytic acid content of 10 g/kg, and for the Zn-supplemented diets, ZnSO, was added to give a dietary Zn content of 15 mg/kg. In experiments in which 65Zn was added to diets, a weighed portion of diet (3 g) was ml water containing 2 pCi 65Zn (specific activity (SA) thoroughly mixed with 0.5 1.0 mCi/mg ; Radiochemical Centre, Amersham, Bucks.). The resulting mixture was freeze-dried and ground to a consistency similar to that of the original diet. Dietary experiment Groups of either five or six rats were given five dietary treatments: (I) Zn-supple- Zn- and phytate-supplemented, fed ad lib.; (3) Zn-supple- mented, fed ad lib.; (2) mented, pair-fed to treatment 2; (4) Zn-deficient, fed ad lib. ; (5) Zn-deficient, phytate-supplemented, fed ad lib. Rats were maintained on these treatments for 21 d and body-weights and food intakes were measured daily. d, oven-dried (24 h at IIO'), crushed to Faeces were collected daily for the first 14 fine powder, and duplicate samples (approximately 50 mg) wet-ashed in conc. H,SO,- conc. HClO,conc. HNO, (0.5: 1.0: 5.0, by vol.). After dilution, the Zn content of the samples was estimated by atomic absorption spectrophotometry (Techtron AA5 ; Varian-Techtron Pty, Melbourne, Australia). VOl. 34 Phytate and trace metal utilization 245 Estimates of the average daily apparent absorption of Zn were obtained over a 10 d period from the differences between dietary intakes of Zn from day 4 to 13 and the faecal losses of Zn from day 5 to 14. At the end of the dietary treatment period rats were killed under diethyl ether anaesthesia, the entire intestinal contents flushed out and the whole bodies oven-dried (24 h at 110'). After an initial charring the whole bodies were dry-ashed at 450' for 24 h and the ash residue gently heated with 5 In1 conc. HNO,-H,O, (100 vol.) (50: 50, v/v) and evaporated to dryness before a second ashing at 450' for 4 h. The resulting white residue was dissolved in I M-HCl and diluted to I 1 for Zn, Cu, Fe and Mn determinations using atomic absorption spectrophotometry. Whole-body accumulation of each of these elements was estimated from their whole-body contents at slaughter and from the mean metal contents of a group of six rats of similar size and age which were killed at the beginning of the dietary experi- ment. From these values, whole-body retention of each element was calculated as a percentage of the dietary intake. For statistical treatment of results the groups maintained on Zn-supplemented diets and those on Zn-deficient diets were considered separately. Zn absorption from ligated intestinal loops Male rats weighing 150 g were anaesthetized with Nembutal (Abbot Laboratories mg/kg body-weight) Ltd, Queenborough, Kent) (administered intraperitoneally, 45 and the peritoneal cavity was opened by mid-line incision. The duodenum (150 mm segment distal to the pyloric sphincter) was flushed with 5 ml warm saline solution (9 g NaC1/1) and the fluid and solid material in the loop squeezed into the lower regions of the intestine by gentle finger pressure. Double ligatures were placed quickly and as close as possible to the duodenal wall without restricting blood flow to the intestinal loop. In order to assess the effect of phytate on Zn absorption, the isolated loops were filled with 0.8 ml saline or saline containing 5-07 pmol phytic acid (as sodium phytate, adjusted to pH 7.0 with HCl) using a plastic syringe with a blunt needle, inserted through the distal ligature. The test dose of 5 pg Zn (as ZnSO, containing 0.1 pCi 65Zn) in 0.2 ml saline containing 109 pmol Ca2+ was subsequently injected into the loop using a fine-bore needle, and the abdominal incision closed with sutures. The effect of ion-exchange resin on Zn absorption was studied using a similar technique. Isolated duodenal loops in situ were first filled with 0.8 ml of a slurry of Chelex-loo (100-200 mesh, sodium form; Bio Rad Laboratories, Richmond, Cali- saline (50: 50, v/v) and subsequently injected with 0-2 ml saline fornia, USA) - containing 5 pg Zn (as ZnSO, containing 0.1 pCi 65Zn). After 15 min the loops were rapidly excised, cut open and the contents flushed out with 5 ml warm saline. The loop tissue and loop contents and washings were sepa- rately assayed for 65Zn content using a gamma-well counter (Tracerlab, Instruments Division, Hersham, Surrey). 'Zn absorption' from loop to whole body was assessed from the loss of the injected radioactivity (injected radioactivity minus recovered radioactivity in loop contents and 246 N,, T. DAVIES AND R. NIGHTINGALE I975 tissue after flushing with saline). ‘Zn bound’ refers to the radioactivity remaining in the loop tissue after the inucosa was flushed with saline. ‘Mucosal Zn uptake’ refers to total uptake of Zn from the lumen fluid and was calculated as the sum of Zn bound and Zn absorbed. All results were converted to pg Zn using the initial SA values. Zinc secretion and reabsorption Duodenal or ileal loops (150 mm segment proximal to the ileo-caecal junction) were isolated in situ as described previously and filled with I ml warm saline or I ml of a slurry of Chelex-Ioo (100-zoo mesh, sodium form) - saline (50: 50, v/v). After I h the loops were removed and their contents and the saline washings were evaporated to dryness in an oven maintained at 110’. The resulting residues were wet-ashed and assayed for Zn using atomic absorption spectrophotometry. Since Zn was shown to be poorly absorbed when bound to Chelex resin, the dif- ference between the Zn accumulated in the Chelex-filled loops and the saline-filled loops was taken to represent the amount of secreted Zn available for reabsorption. EfSect of dietary phytate on 65Zn absoiytion and turnover Twenty male rats in the weight range 65-70 g were transferred from a stock cube diet offered ad lib. to the Zn-supplemented, semi-purified diet to which they were allowed restricted access for I h periods twice daily. After 10 d all rats were eating at least 4 g food/meal and I o g food daily compared with intakes of I 1-13 g for rats of similar age and weight allowed unrestricted access to the same diet. Following this initial training period the rats were randomized into four groups of five rats each to receive the following treatments. Group I. Rats were offered a meal of 3 g Zn- and phytate-supplemented diet, to which was added z pCi 65Zn (SA 1-0 mCi/kg). They were subsequently allowed un- restricted access to the same diet (without 65Zn) for 10 d. Group 2. Rats were offered a meal of 3 g Zn- and phytate-supplemented diet without added 65Zn. These animals were injected (intraperitoneally) with z pCi 65Zn (SA 1-0 mCi/mg) in 0.2 rnl saline and subsequently allowed unrestricted access to the same diet. Group 3. On the day following the administration of 65Zn to groups I and 2, rats were offered a meal of 3 g Zn-supplemented diet, to which was added 2pCi 65Zn mCi/mg), and subsequently maintained on this diet (without added ‘j5Zn) and (SA 1.0 pair-fed with group I. Group 4. On the same day as group 3, rats were offered a meal of 3 g Zn-supple- mented diet before the injection of z pCi 65Zn (SA 1-0 mCi/mg) in 0-2 ml saline. For the remainder of the experiment these animals were pair-fed to those of group 2. Immediately after either dietary dosing or injection of the radioactive Zn the amount of radioactivity in the rats was measured, using a whole-body counter, to determine the administered dose, and thereafter was measured daily for 9 d. From the 65Zn retention curves obtained for these two methods of administration the amount of the dose that was absorbed and biological half-lives of body 65Zn were cal- culated by the methods of Heth & Hoekstra (1965). VOl. 34 Phytate and trace metal utilization 247 140 I I ‘.-I / 3 6 / 4 2 20 0 9 12 15 18 21 -1 Period on diet (d) L Fig. I. Mean weight gains for groups of rats maintained on a semi-purified diet with or without zinc and sodium phytate supplements. (a) Zn- and phytate-supplemented diet offered ad lib. (treatment z) (0-0); Zn-supplemented diet, rats pair-fed to treatment z rats (A-A); Zn- lib. (0-0); (b) phytate-supplemented, Zn-deficient diet offered supplemented diet offered ad ad lib. (0-0); Zn-deficient diet offered ad lib. (0-0). For details of diets, see p. 24. Initial body-weights for all rats were between 66-5 and 74.5 g. The no. of rats/treatment is indicated in parentheses; the standard errors are represented by vertical bars. RESULTS Injuence of dietary phytate on growth and food intake (I) Zn-supplemented diets. The growth rates of rats maintained on the Zn-supple- mented diets (treatments I, 2 and 3) are shown in Fig. I a. The rats given the Zn- supplemented diet ad lib. grew at an approximately linear rate of 6.5 0.3 g/d (mean k SE) throughout the experiment. No significant differences in growth rates between the three groups were found for the first 3 d, but from day 4 to the end of the experiment the phytate-fed animals showed a marked reduction in daily weight gain (P < 0.001) compared with the ad lib.-fed control rats. As there was a comparable decrease in weight gain for the pair-fed control group, this can be accounted for mainly by a decrease in mean daily food intake. The average daily food intakes for days 1-3, 4-13 and 14-21 of the experimental period are shown in Table r. During the first 3 d of the experiment no significant differences in daily food intake were apparent between the phytate-supplemented and control (Zn-supplemented) animals fed ad lib. However, the intake for the phytate group decreased after 3 d and thereafter, while eating significantly less food than the ad lib.-fed controls, they established a cyclic pattern of intake similar to that reported for Zn-deficient rats by Williams & Mills (1970) (Fig. 2). The greater variability of food intake for the phytate-fed rats was confirmed in statistical analyses of the day-to- day variance of the food intakes of the phytate-fed rats and ad lib.-fed controls. The 248 N. T. DAVIES AND R. NIGHTINGALE I975 Table I. Average food intakes (gld) for groups of rats receiving ad lib. a serni-pur$ed diet+ with OY without zinc and sodium phytate supplements (Mean values with their standard errors for six rats/group) Average intake (g/d) Dietary treatment Zn-supplemented z Zn- and phytate-supplemented 4 Zn-deficient 5 Zn-deficient, phytate-supplemented I Days 1-3 Days 4-13 Days 14-21 16.5 fo.61 6.0 If: 0.19**# 8.3 f0 .16 12.6 0.49 5.8fO.I3\"** 8.3 f 0.22~' 6.4 If: 0.20 4.8 f 0.14 6.7 f0 .26~~ 4.6 0.10~' 4'7 0'12 4'4 O'IqNs The two Zn-supplemented groups (treatments I and 2) and the two Zn-deficient groups (treatments 4 and 5) were considered separately, for each time period, for statistical analysis by Student's t test. > 0.05); The statistical significance of t:he difference between mean values was: NS, not significant (P *** P < 0'001. t For details of diets, see p. 244. residual variation in food intake from day to day over the 21 d period, after removing any linear trend, was measured for each rat. All six values for the ad lib.-fed, Zn- supplemented group were substantially greater than the maximum of the six values for the ad lib.-fed, Zn- and phytate-supplemented group. This treatment difference < 0-001 was obtained with a was clearly significant, and a statistical significance of P t test done using log residual variances (ad lib.-fed, Zn-supplemented 0.17 & 0.08, ad lib.-fed, Zn- and phytate-supplemented 0.59 rf: 0.04). (2) Zn-dejicient diets. The growth rates of the rats offered the Zn-deficient, phytate- supplemented, and Zn-deficient diets are shown in Fig. I b. After the 3rd day on the diets for both groups the pattern of daily weight gain changed. The rats offered the Zn-deficient diet maintained a slight daily weight gain whereas the phytate-fed animals suffered a net loss in weight. There was a voluntary restriction of food intake for both these groups of rats on the 4th day of the experiment and for the remainder of the treatment cyclic patterns of food intake were recorded. The presence or absence of phytate in the Zn-deficient diets had no significant effect on the average daily food intakes throughout the experi- ment (Table I) and did not noticeably affect either the frequency or amplitude of the food-intake cycles (Fig. ::). Injbence of phytate on faecal Zn excretion The faecal excretion of Zn on the 1st day of treatment exceeded that for all sub- sequent days irrespective of treatment. Since this probably originated from residues of the pre-experiment stock cube diet, results for day I were excluded from the cumulative results for Zn output (Fig. 3). (I) Effect of adding phytate to Zn-supplemented diets. The mean cumulative faecal for the three groups of rats offered Zn-supplemented diets is Zn output for days 2-14 shown in Fig. 3a. The phytate-fed rats had a significantly higher daily faecal Zn excretion than either the ad lib.-fed or the pair-fed control rats (P < 0.001). No significant differences were found between the two control groups (P > 0.05). VOl. 34 Phytate and trace metal utilization 249 10 (4 1 3 5 7 9 I1 13 15 17 19 21 23 Period on diet (d) Fig. 2. Examples of daily food intakes for individual rats receiving ad lib. a semi-purified Zn-supplemented diet, (b) Zn- diet with or without zinc and sodium phytate supplements. (a) Zn-deficient, phytate-supplemented and phytate-supplemented diet, (c) Zn-deficient diet, (d) diet. For details of diets, see p. 244. The mean ( & SE) apparent Zn absorption values (pg Zn absorbed/d) calculated as the difference between dietary Zn intake (days 4-13) and faecal Zn output (days 5-14), for the six phytate-fed rats, six ad lib.-fed control rats and five pair-fed control rats were: 25.3 k 3-9, 149.0 k 7.4, 53.2 + 4.4 respectively. Dietary phytate significantly (P < 0.001) reduced the apparent absorption of dietary Zn compared with either the ad lib.-fed or the pair-fed control groups. (2) Effect of adding phytate to Zn-deJcient diets. The mean cumulative faecal Zn output for days 2-14 for the two Zn-deficient dietary treatments is shown in Fig. 3 6. The inclusion of phytate in the diet almost doubled faecal Zn output (P < 0.01) even though the dietary Zn intakes for these two groups were not significantly different. Calculation of apparent absorption of Zn from the difference between dietary Zn showed that while both intake for days 4-13 and faecal Zn output for days 5-14 groups suffered a net loss of Zn this effect was considerably greater (P < 0.001) for the phytate-fed rats. The results expressed as pg Zn loss/d (mean SE) were, for the +_ 0.28 and for the six Zn-deficient, phytate-fed rats six Zn-deficient control rats 0.87 7.6 k 0.08. 250 N. T. DAVIES AND R. NIGHTINGALE I975 0 .- 2 CI C 600 Period on diet (d) Fig. 3. Cumulative faecal zinc excretion for rats maintained on a semi-purified diet with or Zn- and phytate-supplemented diet offered without Zn and sodium phytate supplements. (a) adlib. (treatment 2) (0-0); Zn-supplemented diet, rats pair-fed to treatment z rats (A-A); Zn-supplemented diet offered ad lib. (.-a); (b) phytate-supplemented, Zn-deficient diet Zn-deficient diet offered ad lib. (@-0). For details of diets, see offered ad lib. (0-0); p. 244. Each point represents the mean value for five rats, with standard errors represented by vertical bars. Effects oj dietary phytate on whole-body contents and retention of Zn, Cu, Fe and Mn The estimated contents of Zn, Cu, Fe and Mn at the beginning (day 0) and end of the dietary treatments (day 21) are shown in Table 2. These values were used to calculate the average daily accumulation (,ug/d) and retention (relative to dietary intake) of these trace metals shown in Table 3. (I) Zn-supplemented diets. Phytate significantly reduced the average daily accumu- lation of Zn, Cu, Fe and Mn compared with that for both the ad &.-fed and the pair- fed control groups. When differences in dietary intake of these trace metals were taken into account and results were expressed as retention relative to intake, phytate- fed rats again had a significantly reduced retention of all trace metals studied. (2) Zn-deJicient diets. For the Zn-deficient animals, phytate significantly reduced both the average daily accumulation and the proportions of the dietary intake of the four trace metals which were retained (Table 3). Taking the results for the Zn-supplemented and Zn-deficient groups together it is clear that dietary phytate reduces the availability of Cu, Fe and Mn as well as Zn. The results of these studies of apparent absorption and whole-body retention indicate net losses of body Zn that cannot be attributed solely to an effect of phytate in reducing the availability of dietary Zn. A likely explanation for these findings is that dietary phytate may render endogenously secreted Zn unavailable for reabsorption. after the 21 d feeding plements Mn > * Day o Day 21 SE +7 - SE Mean Mean 0.59 0.95 41.9 41.8 150.4 35.8 24.9 1.05 0.30 0.52 41.1 42.1 62.8 46.1 2.11 3.75 ch were killed on the 1st day of tion of dietary intake) of Zn and sodium phytate ion cu Mn A-r7- 1 * Mean SE SE 0.0026 0.0075 0.0016 o~oo13tft 0~0009 o.ooozttf o.oo31f1 o'oozo 0.001111 0.0018 0.0026* 0.0008 0.0041 0.0003 o.oco7** istical analysis. For comparisons Fo r treatments 1-3, where there lues for treatment group I were treatment group 3 : 11P < 0.01. ndupIhia swte1 y 0)ta'902 0189 wt dahy223.6.7.nd a ( p8 1 8 ge6.6.7.82 aing\"umdi- SE 141171715aril msos) A 2 ayDn eaM ucsi f beginnd seheSE90 08 ont1.85682.0.1.tse anrathx t Zn repa5 0 8 0 si t n 8.8.5.9'of ag)ou its*o ayDn eac M13131313 up(pithra-f ros w o6 gnt o. n7 SE11977 47 46 a2ofte s onh ors; itor-ayn Dea54 2006 76 ntte5. 44262523on60se c wrr e-M c0. net+rde iedaF aln ga-SE29 421525etMm, 0and dansto ier n 0 ea2.miJei Z1530 mu d urthhayDn ea21SZZO21e C, ani-ph MIZZit-thd 65v A6 1an30.pem wseeslu- SE 11431222o e opa va2y F, 0 c n Iayn da2n,oneaDea28 66 97 18 n 0.rod (M 43161313 o~n-AMtsn , ineZ raZncair 0 e : o SE19 28117tht)zint f -w odyy ai od mPay Dn ea9994 7204tsI314w-btsM1313y-g bolera) ) ) ) odg/ hoor(6(6(6(6 b(pe heer w tw244. , fedodoms atrid,d,frue p.teted aleeimpeenenve v ss,stt mmriseetEledehedi end le2.tmteen pp pp e T ert. of t edsuedleeamedsudeen wenlsai-fe--fab trle-fe--frypplib.ath cisemet ytlii filib.atheri yt lib.pe dTtasuPhmiedeadPhadTDn-adI n-* exorFe t Z2Z4 5 tht atvar ountn teea91 167359 st01.n foithrer y M030176 0.01020.0.0.000.fo0,0eae w y < Mostff el ). th or0 f 4t4f76atPstar**teom ith0.01010087 oo0.0.000.o.ep*, 1U fr s nt0w od-b leSEho7 We F-- ed0.eyres)--n 4 s r-ea167 1 13 90 er feid
80% of the 65Zn remaining in the loop contents. These in vivo findings are consistent with the b), based on in vitro studies, that phytate reduces conclusion of Oberleas el' al. (1966a, Zn availability by precipitation of the Zn in the intestinal lumen as a Zn-Ca-phytate complex. Zn secretion and reabsorption in ligated loops In order to assess the possible quantitative significance of secretion and reabsorption of Zn across the intestinal mucosa, studies were done involving measurement of the recovery of stable Zn in ligated loops of duodenum and ileum filled with either saline alone or a saline-Chelex cation-exchange resin slurry. In a preliminary experiment the presence of the cation-exchange resin in isolated duodenal loops virtually stopped the absorption of Zn from a subsequently injected dose of 5 pg Zn labelled with 65Zn. Thus from four saline-filled loops 0.88 k 0.17 pg Zn was absorbed in 15 imin compared with 0.09 k 0.08 pg Zn absorbed from corre- sponding loops containing the ion-exchange resin (P < 0-001). In subsequent studies, isolated duodenal or ileal loops were filled with either saline alone or a saline-Cheler; resin slurry, and after I h the loops were excised and the contents and washings analysed for Zn. Previous work had indicated that 65Zn2+ retained by this resin could not be absorbed, and therefore it was considered reason- able to assume that the Z,n in the lumen contents of loops containing resin represented the total Zn (as either the free ion or cationic-Zn complexes) present in endogenous secretions produced during the I h experiment. The difference between this value and that for Zn present in loops from which resin was omitted was taken to represent VOl. 34 Phytate and trace metal utilization 253 Table 5. The recovery of zinc in duodenal and ileal loops of rat intestine in situ I h after they were$lled with either I ml saline (9 g NaClI1) OY I ml of a slurry of cation-exchange resin (CheZex)-saline (50: 50, v/v) (Mean values with their standard errors; no. of animals in parentheses) Recovery of Zn (,%/loop per h) Fraction reabsorbedt Duodenal loops Saline Chelex-filled Saline-filled Ileal loops Saline Chelex-filled Saline-filled + 1.68 & 0.10 (6)\"\" 1.07k0.15 (6) 1'33 k0-14 (5)* 0.88 0.13 (4) 0.36 + 0.34 * P < 0.05, *** P < 0'001. t For explanation, see p. 252. The statistical significance of the differences between mean values, analysed by Student's t test, were: the component of endogenous Zn secretion that had been reabsorbed. Under these experimental conditions it would seem that in both regions of the intestine about one- third of the endogenously secreted Zn is normally reabsorbed (Table s). Inj?uence of dietary phytate on absorption and retention of tracer doses of 6sZn Rats previously trained to consume their food in two I h periods were dosed with G5Zn either by addition to their food or by intraperitoneal injection. Trained, meal-fed rats were used to ensure complete ingestion of the oral dose of 65Zn. The 65Zn retention curves for the phytate-fed rats dosed either orally or by injection are shown in Fig. 4a, and those for pair-fed control rats not receiving phytate are shown in Fig. 4 b. Pair-fed control rats were used in order to eliminate any possible changes in absorption and retention of Zn resulting from the reduction in food intake and low growth rates of rats maintained on the phytate-containing diet (see Table I and Fig. I a). For each dietary treatment the linear portion of the semi-logarithmic retention curves between 91 and 211 h post-administration had the same slope whether the 65Zn was administered orally or by injection, indicating that these tracer doses of 65Zn, once inside the body, were handled in the same way whether absorbed or injected. to 21 I h post-administration for both injected Therefore, G5Zn retention results from 91 and orally dosed rats were pooled within each dietary treatment group for calculation of the biological half-life of the body 65Zn. The true proportion of dietary Zn absorbed as by phytate-fed and pair-fed control groups was derived from these values (Fig. 4) described by Heth & Hoekstra (1965). In brief, the method assumes that extrapolation of the injected-dose retention curve to the ordinate ( Yl) indicates the proportion of the original injected dose present in the linear component under study. As the injected and absorbed doses of Zn are handled in the same way, the intercept on the ordinate must of the extrapolated linear component of the oral-dose retention curve ( Yz) represent the same fraction of the total 65Zn initially absorbed (see Fig. 4). The true Yl. This analysis proportion of 'j5Zn absorbed can thus be calculated from Yz+ 254 - 1 - 0.75 N. T. DAVIES AND R. NIGHTINGALE '975 0.50 - yy:-- 0.25 - a- I 1 I I I I Fig. 4. Retention curves for a tracer dose of g5Zn administered in the diet (a-0) or by intra- peritoneal injection (O--O) to rats maintained on a Zn-supplemented, semi-purified diet, with or without sodium phytate supplements ; (a) Zn- and phytate-supplemented diet offered ad lib. ; (b) Zn-supplemented diet pair-fed to rats offered the phytate-supplemented diet. Each point represents the mean value for five rats. The slopes and intercepts for the ( Yl and Yz respec- extrapolated retention curves for the injected and orally administered 65Zn to 21 I h tively) were calculated from the regression equations for the linear components from 91 post-administration. Table 6. The effect of dietary phytate on the absorption of a tracer dose of s5Zn (2 pCi, spec@ activity 1.0 mCilkg), either added to the diet (treatment groups I and 3) or administered intraperitoneally in saline (9 g NaClIl) (treatment groups 2 and 4), and biological half-life 91-21 I h after the dose, for meal-fed rats given, with the tracer dose, 3 g Zn-supplemented, semi-puriJed diet? with or without sodium phytate, and subsequently either offered the phytate-supplemented diet (without 65Zn) ad lib. (treatment groups I and 2), or pair-fed the same diet without phytate (or 65Zn) (treatment groups 3 and 4) (Mean values with their standard errors; no. of rats in parentheses) Proportion of \"Zn absorbed Biological half-life 91-21 I h post-administration (h) and 2 Phytate- supplemented, ad lib.-fed I . Treatment group . Dietary treatment .. . . I 3 3 and 4 Phytate- supplemented, ad lib.-fed Control (no phytate), pair-fed CoAtrol (no phytate), pair-fed o.3.~4+0.017 (5) #**0.882+0.035 (5) 1oo1+156 (10) * 1732+289 (10) * P < 0.05, *** P < 0.001. The statistical significance of the differences between mean values, analysed by Student's t test, were : .f. For details of diets, see p. 246. VOl. 34 Phytate and trace metal utilization 25 5 corrects for the absorption and re-excretion of G5Zn into the gut at short periods after administration. The calculated proportion of G5Zn absorbed and biological half-life for body 65Zn are shown in Table 6. The results indicated that dietary phytate reduced Zn absorp- tion and decreased the biological half-life for body 65Zn. These findings may be explained by the suggestion that as well as inhibiting the absorption of dietary Zn, phytate interferes with the absorption of endogenously secreted Zn, thus promoting a more rapid removal of 65Zn from body stores. DISCUSSION Previous studies have shown that the inclusion of phytate in diets fed to many species will result in poor growth rates. Since growth can be improved by increasing the dietary Zn concentration, it has been concluded that this effect results from a reduction in absorption of dietary Zn (reviewed by Becker & Hoekstra, 1971 ; Ober- leas, 1973). The dietary experiments reported in this paper are in full agreement with these findings. Dietary phytate significantly reduced growth rates and food intake of rats given a diet normally adequate with respect to Zn content, and caused a cyclic pattern of food intake characteristic of an uncomplicated Zn-deficiency (Williams & Mills, 1970)- Furthermore, the reduction in body-weight of the rats given the Zn- deficient diet supplemented with phytate indicated that these animals suffered a more severe Zn deficiency than the control animals maintained on the Zn-deficient diet without added phytate. The findings that the addition of phytate to the Zn-supplemented diet resulted in both a substantial increase in faecal Zn excretion, and a reduction in Zn retention assessed both from faecal Zn balance and whole-body analysis, confirmed that phytate had reduced Zn utilization. These results are in full agreement with those of Likuski & Forbes (1965), who found that rats maintained on diets supplying phytic acid (20 g/kg) and Ca (8 g/kg) had reduced growth rates and impaired Zn retention compared with ad lib.-fed control animals. Similarly Reinhold, Nasr, Lahimgarzadeh & Hedayati (1973) have recently reported a reduction in Zn (and Ca) retention in human subjects maintained on phytate-rich diets. A possible mechanism to account for these effects of phytate on the utilization of dietary Zn has been proposed by Byrd & Matrone (1965) and Oberleas et al. (1966a, b). These workers suggest that Zn combines with phytate and Ca to form an insoluble Zn-Ca-phytate complex from which Zn is unavailable for absorption. In support of this suggestion Oberleas et al. (1966a, b) found that, within the pH range found in the intestinal lumen, Zn, phytate and Ca in the molar ratio I : 2: I respectively, form a precipitate in vitro containing 77% of the added Zn. However, these molar propor- tions are considerably different from those found in practical diets or experimental diets of the type used in this present study, where Ca would be in a 500-1000-fold molar excess and phytate in a 30-100-fold molar excess over Zn. These workers also found that the uptake of G5Zn by strips of jejunal tissue incubated in vitro was reduced by the presence of Ca and phytate in the incubation medium. 9 NUT 34 256 N. T. DAVIES AND R. NIGHTINGALE I975 The results of studies of Zn absorption from ligated loops of rat duodenum in situ reported here fully support this hypothesis. In these experiments, in which the relative molar ratio of Zn, phytate and Ca present in the loops was the same as their ratio in the phytate- and Zn-supplemented diets, phytate markedly reduced mucosal uptake, binding and absorption of 65Zn (Table 4). The finding that > 80% of the 65Zn remaining in the lumen contents of the loops injected with phytate was in a suspended particulate fctrm supports the suggestion that Zn is rendered unavailable for absorption by co-precipitation with Ca and phytate. Further evidence in support of a direct effect of dietary phytate on Zn absorption was obtained in the more physiological measurement of absorption derived from the retention curves subsequent to oral and injected dosing of rats with a tracer amount of 65Zn. In these experiments phytate reduced by 63 yo the absorption of 65Zn added to the diet. In contrast to these findings Heth & Hoekstra (1965), using the same tech- nique, reported that they were unable to find a direct effect of phytate on Zn absorp- tion when phytate was added to diets containing either protein hydrolysates or animal protein. However, the percentage absorption of dietary Zn has been found to be reduced by increasing the dietary Zn concentration (Becker & Hoekstra, 1971), and the effects of phytate in increasing dietary Zn requirements of rats may be influenced both by the Ca and Zn contents of the diet (Oberleas et al. 19664 b) and the phytate concentration (Likuski & Forbes, 1965). Since no details were given by Heth & Hoekstra (1965) of either the Zn, Ca or phytate contents of the diets used, it is difficult to speculate on why their findings are at variance with those reported here. From the results of this present study it is evident that, while a reduction in absorp- tion of dietary Zn may explain the effects of phytate on faecal Zn output and Zn retention in the rats fed Zn-supplemented diets, it cannot fully account for the net loss of Zn from the rats maintained on the Zn-deficient diet. One possible explanation for these findings is that dietary phytate forms complexes with endogenously secreted Zn and prevents its reabsorption. This proposal seems likely in view of the results of ligated loop experiments in which the recovery of Zn in both duodenal and ileal loops containing cation-exchange resin (Chelex) was 35 yo higher than the recovery in loops filled with saline alone. Since Zn2+ was shown to be unavailable for absorption in the presence of this resin, this extra amount of recovered Zn, amounting to 0-45-060 pg Zn/loop per h, may represent a proportion of secreted Zn normally reabsorbed. If the assumptions are made that these rates of secretion and reabsorption are maintained throughout the day, and along the entire length of the intestine, the total amount of Zn reabsorbed daily would correspond to about 90-100,ug. When this amount is compared with the mean daily accumulation of 139.5 & 6.9 pg Zn/d for the six rats maintained on the Zn-supplemented diet fed ad lib. (Table 3), it is clear that recycling of Zn across the intestine may be a significant factor in the over-all body economy of this trace metal. An alternative explanation for these results is that the presence of a strong Zn- complexing agent in the lumen of the gut, either ion-exchange resin or phytate and Ca, extracts Zn either from the intestinal mucosal cells or across the mucosal tissue from the blood and tissues. Which of these two explanations is correct cannot be VOl. 34 Phytate and trace metal utilization 257 ascertained from these experiments although either can adequately explain how phytate added to a Zn-deficient diet promotes a net loss of carcass Zn. The same explanations can be advanced to explain the decrease in half-life (in- creased rate of turnover) of the body 65Zn due to dietary phytate; namely, a more rapid loss of 65Zn due to an inhibited reabsorption of secreted 65Zn or extraction of 65Zn either from or through the intestinal mucosa. It is over 30 years since Widdowson & McCance (1942) first reported that Fe in phytate-rich diets is relatively unavailable. Since then numerous workers have con- firmed this report and have found that dietary phytate reduces Fe absorption. In a recent review on phytate, Oberleas (1973) notes that as ferric phytate is least soluble in dilute acid, but at the pH encountered in the duodenum this complex dissociates, the site of action of phytate in decreasing Fe absorption is probably the stomach, where it may prevent Fe binding to gastroferrin. The finding in the present study of a reduction in carcass retention of Fe due to phytate addition to both the Zn-supplemented and Zn-deficient diets is surprising in view of the high levels of Fe in these diets (50 mg/kg) in relation to the recommended dietary Fe requirement for the rat ((US) National Research Council, 1972). Even more surprising is the finding that phytate reduced the retention of Cu when the concentration of this trace metal in the diets, at 25 mg/kg, was four to five times the recommended minimum dietary level ((US) National Research Council, 1972). To the authors' knowledge no studies hitherto reported have shown a reduction in Cu or Mn availability due to the presence of phytate in the diets. However, Davis, Norris & Kratzer (1961) reported that diets containing an isolated soya-bean protein reduced the availability of both trace metals for the chick, and in view of the high phytate content of soya-bean meal (Oberleas et al. 1966a) it would seem likely that phytic acid was the agent responsible. The mechanism by which phytate reduces the availability of Cu and Mn probably results from the ability of these trace metals to form metal-phytate complexes which are stable within the intestinal tract (Maddaiah, Kurnick & Reid, 1964; Vohra, Gray & Kratzer, 1965). Whether or not a role for Ca can be implicated in these dietary interactions with phytate, as has been found for Zn (Oberleas et al. 1966a, b) has yet to be determined. In this context the finding by Oberleas (1973), that the amount of precipitate formed at pH 6-7 when equimolar proportions of Cu and Ca were added to a solution of phytate was more than double the amount formed on addition of Cu alone, may be of relevance. In addition to its effects on the retention of Zn, Fe, Cu and Mn, dietary phytate has and magnesium also been shown to reduce the availability of Ca (Bruce & Callow, 1934) (Likuski & Forbes, 1965; Seelig, 1964). In view of these findings, the nutritional significance of dietary phytate in relation to the essential trace metals needs further study. In agreement with the speculations of Oberleas (1973), it seems likely that this agent, under the appropriate dietary conditions, may promote a nutritional deficiency of any of these trace metals, depending upon which first becomes limiting in the diet 258 N. T. DAVIES AND R. NIGHTINGALE I975 The authors gratefully acknowledge the expert technical assistance of Mr A. Flett. R.N. was a Temporary Research Assistant. REFERENCES Becker, W. M. & Hoekstra, W. G. (1971). In Intestinal Absorption of Metal Ions, Trace Elements and Radionuclides, p. 229 [S. C. Skoryna and D. Waldron-Edward, editors]. Oxford: Pergamon Press. Bruce, H. M. & Callow, R. K. (1934). Bi0chem.J. 28, 517. Byrd, C. A. & Matrone, G. (1965). Proc. SOC. exp. Biol. Med. 119, 347. Davies, N. T. (1973). J. Physiol., Lond. 229, 46 P. Davis, P. N., Norris, L. C. & Kratzer, F. H. (1961). J. Nutr. 77, 217. Heth, D. A. & Hoekstra, W. G. (1965). J. Nutr. 85, 367. Likuski, H. J. A. & Forbes, R. M. (1964). J. Nutr. 84, 145. Likuski, H. J. A. & Forbes, R.. M. (1965). J. Nutr. 85, 230. Maddaiah, V. T., Kumick, A. A. & Reid, B. L. (1964). Proc. SOC. exp. Biol. Med. 1x5, 391. Methfessel, A. H. & Spencer, H. (1973a). J. appl. Physiol. 34, 58. Methfessel, A. H. & Spencer, H. (1973b). J. appl. Physiol. 3, 63. National Research Council (1972). Nutrient Requirements of Domestic Animals No. 10, Nutrient Re- quirements of Laboratory Aizimals, p. 56. Washington, DC : National Research Council. Oberleas, D. (1973). Toxikagzts Occurling Natural& in Foods, p. 363. Washington, DC: National Academy of Sciences. 57. Oberleas, D., Muhrer, M. E. & ODell, B. L. (1962). J. Anim. Scd. 21, Oberleas, D., Muhrer, M. E. & O’Dell, B. L. (1966a). In Zinc Metabolism, p. 225 [A. S. Prasad, editor]. Springfield, Ill. : Charles C , Thomas. Oberleas, D., Muhrer, M. E. & O’Dell, B. L. (1966b). J. Nutr. 90, 56. O’Dell, B. L. & Savage, J. E. (1960). Proc. SOL. exp. Biol. Med. 103, 304. Reinhold, J. G., Nasr, K., Lahimgarzadeh, A. & Hedayati, H. (1973). Lancet i, 283. Seelig, M. S. (1964). Am. J. din. Nutr. 14, 342. Vohra, P., Gray, G. A. & Kratzer, F. H. (1965). Proc. SOC. exp. Biol. Med. 120, 447. Widdowson, E. M. & McCance, R. A. (1942). Lancet i, 588. Williams, R. B. & Mills, C. F. (1970). BY. J. Nutr. q, 989. Printed in Great Britain