Zinc Absorption from Milk Is Affected by Dilution but ...

The Journal of Nutrition. First published ahead of print April 19, 2017 as doi: 10.3945/jn.116.244426. The Journal of Nutrition Nutrient Physiology, Metabolism, and Nutrient-Nutrient Interactions

Zinc Absorption from Milk Is Affected by Dilution but Not by Thermal Processing, and Milk Enhances Absorption of Zinc from High-Phytate Rice in Young Dutch Women1,2 Elise F Talsma,3,6,9 Diego Moretti,7,9 Sou Chheng Ly,3 Renske Dekkers,8 Ellen GHM van den Heuvel,8 Aditia Fitri,8 Esther Boelsma,4 Tjeerd Jan Stomph,5 Christophe Zeder,7 and Alida Melse-Boonstra3* 3 Division of Human Nutrition, 4Wageningen Food and Biobased Research, and 5Center for Crop Systems Analysis, Wageningen University and Research, Wageningen, Netherlands; 6HarvestPlus, International Center for Tropical Agriculture (CIAT), Cali, Colombia; 7 Laboratory for Human Nutrition, Swiss Federal Institute of Technology, Zurich, Switzerland; and 8Friesland-Campina, Amersfoort, Netherlands

Downloaded from jn.nutrition.org at ASN Member on May 2, 2017

Abstract Background: Milk has been suggested to increase zinc absorption. The effect of processing and the ability of milk to enhance zinc absorption from other foods has not been measured directly in humans. Objective: We aimed to assess zinc absorption from 1) milk undergoing various processing and preparatory steps and 2) from intrinsically labeled high-phytate rice consumed with milk or water. Methods: Two randomized crossover studies were conducted in healthy young women [age:18–25 y; body mass index (in kg/m2): 20–25]: 1) a milk study (n = 19) comparing the consumption of 800 mL full-fat ultra-high temperature (UHT) milk [heat-treated milk (HTM)], full-fat UHT milk diluted 1:1 with water [heat-treated milk and water (MW)], water, or unprocessed (raw) milk (UM), each extrinsically labeled with 67Zn, and 2) a rice study (n = 18) comparing the consumption of 90 g intrinsically 67Zn-labeled rice with 600 mL of water [rice and water (RW)] or fullfat UHT milk [rice and milk (RM)]. The fractional absorption of zinc (FAZ) was measured with the double-isotope tracer ratio method. In vitro, we assessed zinc extraction from rice blended into water, UM, or HTM with or without phytate. Results: FAZ from HTM was 25.5% (95% CI: 21.6%, 29.4%) and was not different from UM (27.8%; 95% CI: 24.2%, 31.4%). FAZ from water was higher (72.3%; 95% CI: 68.7%, 75.9%), whereas FAZ from MW was lower (19.7%; 95% CI: 17.5%, 21.9%) than HTM and UM (both P < 0.01). FAZ from RM (20.7%; 95% CI: 18.8%, 22.7%) was significantly higher than from RW (12.8%; 95% CI: 10.8%, 14.6%; P < 0.01). In vitro, HTM and UM showed several orders of magnitude higher extraction of zinc from rice with HTM than from rice with water at various phytate concentrations. Conclusions: Milk enhanced human FAZ from high-phytate rice by 62% compared with water. Diluting milk with water decreases its absorption-enhancing proprieties, whereas UHT processing does not. This trial was registered at the Dutch trial registry as NTR4267 (http://www.trialregister.nl/trialreg/admin/rctview.asp?TC=4267). J Nutr doi: 10.3945/jn.116.244426.

Keywords: zinc, absorption, milk, rice, phytate, food processing, thermal processing, human, isotope dilution

Introduction Zinc deficiency is widespread globally (1), with estimates indicating that 20% of the total world population has some 1

Supported by Friesland-Campina, Amersfoort, Netherlands. Author disclosures: EF Talsma, D Moretti, SC Ly, E Boelsma, TJ Stomph, C Zeder, and A Melse-Boonstra, no conflicts of interest. R Dekkers and EGHM van den Heuvel are employees at Friesland-Campina. A Fitri conducted the in vitro studies as an internship student at Friesland-Campina. 9 These authors contributed equally to this work and share first authorship. *To whom correspondence should be addressed. E-mail: alida.melse@wur. nl. 2

degree of zinc deficiency (2). Zinc deficiency increases the risk of infections and restricts physical growth in children (3). Increased requirements, malabsorption, increased losses, and impaired utilization of zinc can all lead to inadequate zinc status, but the primary cause of zinc deficiency is low dietary intake of absorbable zinc (4). Zinc absorption is hampered by phytate, which is present in large amounts in plant-based foods, such as whole-meal cereals, legumes, and vegetables (5). Increasing the amount of absorbable zinc intake from the food supply is needed to combat zinc deficiency.

ã 2017 American Society for Nutrition. Manuscript received November 16, 2016. Initial review completed December 6, 2016. Revision accepted March 20, 2017. doi: 10.3945/jn.116.244426.

Copyright (C) 2017 by the American Society for Nutrition

1 of 8

Methods Human studies Study design. We conducted 2 separate open-label, single-blinded human absorption studies, both with a randomized crossover design including a 1-mo washout period: a rice study and a milk study. The milk study consisted of 4 different treatments with 800 mL of 1) full-cream UHT milk (HTM; LangLekker; Friesland-Campina), 2) UHT milk diluted 1:1 with water (MW), 3) demineralized water, or 4) raw milk [unprocessed milk (UM)]. All UHT milk used in both studies was obtained from the same batch. The rice study comprised 2 treatments: 90 g steamed rice that was intrinsically labeled with 67Zn (75% enrichment), consumed with either 600 mL demineralized water [rice and water (RW)] or 600 mL full-cream UHT milk [rice and milk (RM)]. In both studies the zinc content of all of the administered meals was matched to facilitate comparisons of zinc absorption. Sample sizes of the 2 studies were based on earlier studies that aimed to assess differences in the fractional absorption of zinc (FAZ) from foods, and a difference in FAZ of 15% was considered meaningful for the milk study, whereas a difference of 25% was considered meaningful for the rice study, generating required sample sizes of 15 and 14 for both studies, respectively. Treatment orders were allocated to participants by the study coordinator (SCL) by using randomized blocks, except for the UM arm in the milk study. Due to the short conservation period and the Abbreviations used: AGP, a1-acid glycoprotein; CRP, C-reactive protein; DAT, days after transplanting; FAZ, fractional absorption of zinc; HTM, heat-treated (ultra-high temperature) milk; ICP, inductively coupled plasma; LMM, linear mixed-effects model; MW, heat-treated (ultra-high temperature) milk and water (1:1); RM, rice and ultra-high temperature milk; RW, rice and water; UHT, ultra-high temperature; UM, unprocessed (raw) milk.

10

2 of 8

Talsma et al.

potential bias introduced by differences in raw-milk composition between batches, this treatment was assigned to participants as the last of the 4 treatments in the milk study and only to those who had fully completed the randomized scheme of the other 3 treatments. Study procedures. Participants were instructed to consume a similar evening meal low in zinc before each test day. They arrived fasted at the test facilities at the Division of Human Nutrition, Wageningen University, with a baseline spot urine sample collected that morning to be able to account for any residual enrichment from the previous administration (except on the first exposure day when they had not been previously exposed to labeled zinc). An intravenous catheter was inserted, and a blood sample was drawn into a trace element–free tube with a clot activator (article no. 368380; BD Diagnostics). Blood samples were centrifuged after 30 min (1200 3 g; 10 min; 4°C), and serum aliquots were stored at 280°C until analysis. Participants consumed the experimental meals containing the 67Zn label in 2 portions 1 h apart. All drinks were served in coded nontransparent beakers with a closed lid and consumed through a straw, thereby blinding the allocated experimental meals. Within 10 min after the first portion of the meal was consumed, participants received an intravenous infusion dose of 0.2 mg 70Zn in saline, as described elsewhere (13, 14). Subsequently, participants were monitored for 3 h to make sure that no foods or liquids (except for demineralized water) were consumed. Four days after exposure to each of the experimental meals, a spot urine sample was collected for isotopic ratio analysis (15, 16). Participants of the rice study had 2 test days and participants of the milk study had 4 test days. On each test day, participants filled out an electronically based FFQ based on an existing FFQ (17), extended with questions to assess zinc and phytate intake. The Estimated Average Requirement of zinc is 6 mg/d for women $19 y of age with a mixed or refined vegetarian diet, assuming a physiologic daily requirement of 1.86 mg absorbed zinc and 34% absorption from the diet (4). Body weight and height were measured on the first day to the nearest 0.1 kg and 0.1 cm by using a mechanical floor scale (Seca 703) and a portable stadiometer, and BMI was calculated as kg/m2. A pregnancy test was performed before participants were exposed to the raw-milk treatment. Participants. Thirty-seven female participants (nonpregnant, nonlactating, and apparently healthy) aged between 18 and 25 y with a normal BMI (20–25) were recruited in the vicinity of Wageningen, Netherlands, to participate in 1 of the 2 absorption studies. All of the participants had no history of illnesses, were not taking any long-term medication (except for contraceptives), and reported to tolerate milk and rice well. None of the participants were taking nutritional supplements or had donated blood from 2 wk before the start of the study until the end of the study. Data were collected from April to September 2014. A written informed consent was obtained from all participants before the start of the study. The study protocol was approved by the medical ethics committee of Wageningen University and was registered with the Dutch trial registry (NTR4267). Intrinsically labeled rice. Intrinsically labeled rice (cv 90B290) was grown hydroponically as previously described (14, 18) in a climatecontrolled greenhouse in Wageningen between June and October 2013. The carbon dioxide level was ;370 mmol/mol, and the relative humidity was 60–80%. Seeds were germinated and grown to the second leaf stage on quartz sand and transplanted to water containers (14). Until 36 d after transplanting (DAT) the rice received a slightly amended half Hoagland nutrition solution (14) and zinc was given as 0.05 ppm ZnSO4 with natural isotope abundance (12.6 mL 0.01-M ZnSO4.7H2O per container). At 36 DAT (just before panicle initiation) the Hoagland solution was fully replaced and zinc was given as 67Zn (isotope enrichment: 89.6%; Isoflex), 0.1 ppm 67Zn (24.6 mL 0.01-M 67ZnSO4.7H2O per container). At 54 DAT and 68 DAT (= 50% flowering) extra 67Zn was added (24.6 mL 0.01-M 67ZnSO4.7H2O per container). Extra Hoagland solution was added 54 and 72 DAT when electrical conductivity was reduced by ;33% to bring the nutrient concentration back to half the Hoagland strength (electrical conductivity = ;1.3 S/m). The rice was harvested at full maturity at 117 DAT. After harvesting, the rice was hulled


Rosado et al. (6) found that the addition of ultra-high temperature (UHT)10–treated skimmed milk or yogurt increased zinc absorption by 50% and 68%, respectively, from a typical plant-based rural Mexican diet rich in phytate by using an isotopic dilution protocol with extrinsic labeling of meals (6). Extrinsic zinc labels have been reported to exchange with native zinc in milk, which suggests that this is a valid method to determine zinc absorption from milk-based diets (7). However, extrinsically added zinc isotopes may not fully exchange with endogenous zinc in many other foods (8). Therefore, it cannot readily be concluded from the Mexican study if the increased amount of absorbed zinc was derived from the solid food or from the dairy source provided in the study. Human studies on zinc absorption from regular dairy products are scarce. The enhancing factors in milk on zinc absorption have not been fully elucidated (9), and the effect of food-processing methods such as UHT processing as well as dilution and concentration of milk solids have not been investigated. It has been suggested that digestive products of casein, such as caseinophosphopeptides, enhance zinc absorption, but their addition when derived from pasteurized skim milk did not lead to higher zinc bioavailability (10, 11). Unheated casein, however, was found to lead to higher zinc uptake in a Caco-2 cell model compared with heated casein (12), which suggests increased zinc absorption from raw milk. Furthermore, heuristically, if milk constituents have an enhancing effect on zinc absorption then diluting milk with water would reduce it. Therefore, our aims were as follows: 1) to assess zinc absorption from milk in humans compared with water, 2) to evaluate the effects of the dose of milk solids and thermal processing on zinc absorption from the milk matrix, and 3) to assess the potential enhancing properties of milk on zinc absorption from intrinsically labeled high-phytate rice compared with water.

manually and polished by using a Pearlest laboratory scale grain polisher (Kett Electric Laboratory). Grains were polished in portions of 20 g at a time for 90 s in order to artificially keep the phytate-to-zinc molar ratio high. Rice was harvested, polished, and stored dry and cool until the experiment was conducted in May and June 2014.

Preparation of intravenous doses. Sterile 70ZnCl2 doses for intravenous administration were prepared by dissolving 70ZnO (isotope enrichment: 98.5%; Chemgas) in diluted HCl (4 M). NaHCO3 (8.4%) was added until a pH of 5.7 (5.5–6.0) was reached, and the solution was further diluted by adding NaCl (0.9%) to a final concentration of 22 mg 70 Zn/mL. Individual doses of 9.5 mL were transferred to sealed injection vials, and the sterility of the batch was assessed by determining the colony count. Intravenous doses were prepared and released by the pharmacy of the Erasmus University according to Good Manufacturing Practice. The amount of intravenously injected 70Zn was calculated for each individual by exactly weighing the syringe on an analytical balance before and after administration. The zinc concentration of the intravenous solution was measured by atomic absorption spectroscopy in unused vials (n = 6); moreover, the zinc concentration of the solution remaining in the vials after administration was also measured to assess any possible change in zinc concentration over the duration of the study. Analytical methods. Total zinc concentrations and the 67Zn enrichment of raw rice were measured by isotopic dilution MS [MC–inductively coupled plasma (ICP)–multicollector, Neptune; Thermo-Finnigan] at ETH Zurich, Switzerland, and total zinc in cooked rice was measured by flame atomic absorption spectroscopy after microwave digestion at Wageningen University, Netherlands. The phytic acid concentration in raw rice samples was measured by the modified Makower method (14, 21). Milk samples were analyzed for total zinc concentration by ICP–atomic emission spectroscopy at the laboratory of FrieslandCampina, Leeuwarden, Netherlands, and fortified water and milk samples by ICP-MS at the Soil Science Laboratory, Wageningen University, Netherlands. To assess zinc status, serum zinc concentrations were measured at the Meander Medical Hospital, Amersfoort, Netherlands, on a Beckman Coulter UniCel DxC 880i analyzer according to the manufacturerÕs instructions, with a cutoff for zinc deficiency of 10.7 mmol/L (4). Because circulating zinc concentrations are known to decrease under inflammatory conditions (22), we also measured C-reactive

In vitro experiments Polished rice was purchased from a local shop and raw milk was purchased from a local farm. Full-cream UHT milk was derived from Friesland Campina (LangLekker). All chemicals and enzymes were purchased from Sigma-Aldrich unless stated otherwise. All of the experiments were conducted in triplicate. Fifty grams of rice was steamed and then ground (Braun Turbowatt 600 W, level 15) and blended for 240 s at 37°C in a water bath with 250 g UHT milk, raw milk, or water. To investigate the effect of phytate on zinc extracted from rice in the presence of milk, we added phytic acid from rice (phytic acid sodium salt hydrate; Sigma Aldrich P8810, molecular weight 660,04; Sigma Aldrich) exogenously to the mixture at amounts such that the final phytate-to-zinc molar ratios would be 4, 15, and 30. Phytate-to-zinc molar ratios of 15 and 30 are known to inhibit zinc bioavailability (4). Extraction was performed based on the modified method of Hedrén et al. (24), but without the addition of digestive enzymes. Moreover, pH and salt composition remained unaltered. Instead, the flask containing the food mixture was incubated by shaking it for 4 h at 37°C under subdued light on a rocking table (375 rpm). The suspension was centrifuged for 15 min at 10,000 3 g at ambient temperature (Avanti J301, rotor JLA 10.500; Beckman Coulter). The liquid phase, sediment phase, and starting materials were analyzed for zinc concentration by ICP–atomic emission spectroscopy analysis in the laboratory of Friesland Campina, Leeuwarden, Netherlands. The phytate content of the rice was analyzed with a colorimetric method as described previously (25). Statistical analysis The fractional absorption of the 67Zn dose from either rice or milk was calculated by using the oral-to-intravenous tracer ratio method as described by Friel et al. (15). Baseline urine samples for the second, third, and fourth exposures were used in the analysis as the new natural abundance concentrations. FAZ values are presented as means and SDs. A linear mixed-effects model (LMM) was fitted to each study separately with log-transformed FAZ as the dependent variable, meals as the fixed factor, and participants as the random factor (intercept). In the milk study, when an overall effect of meal was found, post hoc tests were conducted and P values reported with the Bonferroni correction. A pooled LMM analysis with data from both studies was performed with log-transformed FAZ as a dependent variable and meal, CRP, AGP, and plasma zinc as fixed factors and participants as the random factor (intercept). Results are presented as fold changes in FAZ with 95% CIs estimated by the LMM model. Pearson and SpearmanÕs correlation coefficients, where appropriate, were calculated between FAZ, serum zinc, and inflammatory markers on pooled data in each study and with dietary zinc intake on the same day. Results were considered significant when P # 0.05. Data were analyzed by using SPSS software (version 20; IBM SPSS Statistics) and GraphPad Prism version 5.04 for Windows (GraphPad Software). For the in vitro experiments, the amount of zinc extracted from the rice samples was calculated as the percentage extracted from white rice, after correcting the zinc in the liquid phase for the amount of zinc present in water and/or milk. Data were analyzed within each phytate-to-zinc molar ratio by using 1-factor ANOVA with Bonferroni post hoc test. In case of unequal variances (identified with LeveneÕs test) the TamhaneÕs T2 post hoc test was applied. All values are presented as means 6 SEs, with P < 0.05 indicating significance for comparisons with water as the control. Zinc absorption from milk and rice

3 of 8


Experimental meals. All of the rice was cooked in portions of 45 g (raw weight) by steaming it with distilled water (Aqua B Braun) in a rice cooker, stored at 218°C, kept frozen until the day before consumption, and was then thawed overnight in the refrigerator. Samples were reheated by microwave just before consumption. For the milk study, 67 ZnO was purchased (isotope enrichment: 90.59%; Chemgas) and dissolved in stoichiometric amounts of 2-M H2SO4, and the resulting 67 ZnSO4 solution was diluted to a zinc concentration of 0.5 mg 67Zn/g. All of the meals in the milk study were extrinsically labeled by adding 2 g of the 67ZnSO4 solution (1.0 mg 67Zn) on the day before exposure to permit exchange of endogenous zinc with labeled zinc within the protein fraction (19). Because the FAZ is influenced by the total amount of zinc given in a meal (20), zinc was added to the RW meal by adding 2 g ZnSO4 solution (1.2 mg Zn/g; zinc sulfate monohydrate FCC 8; Dr. Paul Lohmann GmbH). Similarly, total zinc contents of meals in the milk study were adjusted by adding 2 g ZnSO4 solution (0.8 mg Zn/g) to the MW treatment and 2 g ZnSO4 solution (1.6 mg Zn/g) to the water treatment. After consuming any of the meals, empty bowls (with spoon and lid) and cups (with lid) were rinsed twice with 10 mL demineralized water and the water was also consumed. Raw milk was collected 3 d before exposure from 1 farmer and stored at 4°C until use. This milk adhered to all production and quality-control parameters of untreated milk as marketed in Germany (Vorzugsmilk) before use in the intervention. Energy, protein, and calcium contents of the experimental meals were calculated on the basis of the Dutch food-composition table (NEVO 2013/4.0).

protein (CRP) and a1-acid glycoprotein (AGP) concentrations by immunoturbidimetric assay (Cobas; Roche Diagnostics). The presence of inflammation was assessed by using a cutoff for CRP of >5 mg/L and for AGP of >1 g/L (23). Spot urine samples, collected in polyethylene storage containers, were stored at 220°C until analysis. The samples were freeze dried and mineralized by microwave digestion, and zinc was isolated by ion-exchange chromatography at the Division of Human Nutrition in Wageningen, Netherlands (13). Isotopic ratios of 66 Zn:70Zn and 66Zn:67Zn were determined with a high-resolution double-focusing magnetic sector field multicollector ICP-MS (Neptune; Thermo-Finnigan) at the Human Nutrition Laboratory, ETH Zurich, Switzerland.

Results

TABLE 1 Characteristics of the healthy young Dutch women participating in the milk and rice studies1

Age, y Anthropometric indicators Weight, kg Height, cm BMI, kg/m2 Serum analytes Zinc, μmol/L C-reactive protein, mg/L a1-Acid glycoprotein, g/L Dietary intake Energy, MJ/d Protein, g/d Fat, g/d Carbohydrate, g/d Fiber, g/d Alcohol, g/d Calcium, mg/d Zinc, mg/d 1 2

Rice study (n = 18)

Milk study (n = 19)

21.6 6 2.0

22.1 6 2.1

64.4 6 3.2 172 6 7.3 22.0 6 1.0

64.7 6 5.9 170 6 6.4 22.4 6 1.9

11.7 6 0.77 1.43 (0.64, 3.35)2 0.68 6 0.14

12.9 6 2.1 1.30 (0.67, 3.43) 0.62 6 0.10

8.7 6 71.7 6 73.4 6 246 6 23 6 15.5 6 889 6 9.4 6

2.8 18.7 18.0 70.7 5.2 17.7 335 2.5

Values are means 6 SDs unless otherwise indicated. Median; 25th, 75th percentiles in parentheses (all such values).

4 of 8

Talsma et al.

8.5 6 67.5 6 78.3 6 238 6 22 6 10.6 6 828 6 9.0 6

2.3 17.5 24.7 72.3 7.2 9.20 264 2.3

Milk study HTM

MW

W

Rice study UM

RM

RW

Energy, kJ 2064 1032 0 2360 2898 1350 Protein, g 26.4 13.2 0 27.2 26.6 6.8 Calcium, mg 976 488 32 984 741 33 Zinc, mg 4.04 4.14 4.22 3.98 3.63 3.77 From milk, mg 3.04 1.52 0 2.96 2.28 0 From rice, mg NA NA NA NA 1.35 1.35 Added zinc, mg 0 1.61 3.21 0 0 2.42 Added 67Zn, mg 1.00 1.01 1.01 1.02 NA NA 67 Zn enrichment meal, % 25 24 24 26 28 27 67 Zn enrichment rice, % NA NA NA NA 75 75 Phytate, mg NA NA NA NA 297 297 Phytate-to-zinc molar ratio (rice) NA NA NA NA 22 22 1 HTM, heat-treated (ultra-high temperature) milk; MW, heat-treated (ultra-high temperature) milk and water (1:1); NA, not applicable; RM, rice and milk meal; RW, rice and water meal; UM, unprocessed (raw) milk; W, water.

(95% CI: 2.58-, 3.25-fold) higher than from HTM and different from all other treatments (P < 0.01). In contrast, zinc from the MW treatment (1:1) was only 0.77 times (95% CI: 0.69, 0.86 times; P < 0.01) as well absorbed as zinc from HTM or UM treatments (both P < 0.01) (Figure 1, Table 3). Zinc absorption from UM did not differ from HTM. The consumption of milk with rice resulted in a 1.62-fold increase (95% CI: 1.28-, 2.23-fold) in zinc absorption from rice (Figure 2, Table 3), corresponding to a 62% increase in FAZ. When pooling the data from both studies, the meal factor was a significant predictor of FAZ, but CPR, AGP, and plasma zinc were not significant predictors. FAZ did not correlate with inflammatory markers, nor with zinc intake. However, in a Spearman correlation analysis, FAZ was significantly correlated with plasma zinc (r = 0.24, P < 0.05). FAZ from any of the meals was also not significantly different between participants with low serum zinc concentration, low zinc intake, or inflammation and participants with normal intake or status.

FIGURE 1 FAZ from UHT milk, UHT milk diluted with water, water, and raw milk in 19 healthy young women (milk study). Participants acted as their own controls and received all test meals in random order, except for raw milk, which was tested in a subgroup (n = 10). Bars represent means. By linear mixed models, meal was a significant determinant of FAZ, P , 0.01. *Bracketed means without a common letter differ, P , 0.05. FAZ, fractional absorption of zinc; UHT, ultrahigh temperature.


In total, from the rice study, samples from 17 participants were available for data analysis, whereas from the milk study there were 16 participants with complete data over the full randomization scheme. Only 10 of the participants received the UM exposure. Subclinical inflammation had a low prevalence in this population: in the milk study, it affected 4 distinct participants at 3 time points (n = 1, n = 2, and n = 1 at HTM, MW, and UM administration, respectively), resulting in an incidence of subclinical inflammation of 5.8%. Furthermore, 1 participant had low plasma zinc (200 is required for hampering zinc absorption (32, 34). In contrast, an algorithm of zinc-absorption studies has shown that dietary calcium may have a modest enhancing effect on zinc absorption (35). In our experiment, the phytate 3 Ca:Zn molar ratio was ;145 for the RM meal. If any inhibition by calcium occurred in our experiment, this apparently was counteracted by other components, such as protein. Our study shows that zinc absorption from milk was substantially lower than that from water, but is comparable to the 28% absorption from pasteurized cow milk (3% fat) reported in an earlier human study (36). Consistent with previous reports, we found that the bioavailability of zinc was very high (72.3%) from an aqueous solution (13, 37–40). Zinc absorption from a food matrix is known to be much lower than

from an aqueous solution, even if the food is a fluid (9). This may be attributed to the food matrix itself as well as to differences in digestive processes, such as the release of gastric acid, intragastric pH, and transit time (41–43). Interestingly, diluting the milk with water resulted in lower zinc bioavailability than with undiluted milk, which may be explained by the lower amount of protein present in the meal. In a study in Bangladeshi preschool children, an FAZ of 20% was found from a rice-based meal with a total zinc content of 3.8 mg and a phytate-to-zinc molar ratio of 20 (44), which is higher than the 12.8% that we found for the rice meal with water but similar to the 20.8% for the rice meal consumed with milk. The rice meal in the Bangladeshi study was provided with lentil soup, papaya, and banana; had a somewhat lower phytate-tozinc ratio; contained a different variety of rice; was extrinsically labeled; and was conducted in growing children, which limits a direct comparison with our study in women (45). In a recent study that used the same rice variety as was used in the present study, we found the FAZ from an intrinsically labeled agronomically biofortified rice meal containing 1.1 mg Zn and with a phytate-to-zinc ratio of 12.3 to be 25% among Swiss adults aged 18–45 y (46). The higher FAZ in that study can be explained by the lower amount of zinc and phytate-to-zinc molar ratio in the meal. In agreement with the human study, our in vitro experiments suggest that milk increases zinc extraction from rice. Because no digestive enzymes or gastric acid was added in the experiments, the overall amphiphilic nature of milk or amphiphilic compounds or amino acids present in milk may have facilitated this increased solubility of zinc from the cereal matrix. It is interesting that specifically raw milk showed a strong effect on zinc extraction with increasing phytate-to-zinc molar ratios, but these results were not confirmed in our human study. Our studies have several strengths. First, intrinsic labeling of zinc in the rice allowed us to state that the increase in FAZ with milk can be attributed to higher zinc absorption from the rice 6 of 8

Talsma et al.

Acknowledgments We thank Henrie¨tte Fick for overseeing the human studies and Miranda Vlag for preparing all urine samples for analysis. We thank Rolf Bos of Friesland-Campina and Michael Zimmermann of ETH Zu¨rich for the initial discussions on the trial design. The authorsÕ responsibilities were as follows—DM, RD, EGHMvdH, and AM-B: conceived the research idea and designed the studies; EFT and SCL: were responsible for the conduction of the human studies; AF: was responsible for the conduction of the in vitro study; EB: was responsible for the ethical review of the research protocol; TJS: produced the intrinsically labelled rice; CZ: conducted the biochemical analyses of isotope-labeled samples


FIGURE 3 Percentage of zinc extracted from steamed white rice by the addition of water, UHT full-fat milk, or raw milk with different phytate-to-zinc molar ratios. All values are means 6 SEs; n = 3. *Within molar ratio, different from water (P # 0.05). UHT, ultra-high temperature.

matrix. In addition, absolute absorption values are better predicted by studies in which foods are labeled intrinsically (47, 48). For milk, extrinsic labeling has previously been shown to be a valid approach for determining zinc absorption from milk-based foods (7). Second, careful balancing of the amount of zinc in the meals allowed us to make true comparisons of FAZ between treatments. A drawback of this may be that the absorption of added zinc from the water fraction in the rice study was so high (as shown in the milk study) that it reduced zinc absorption from the rice fraction. However, this is unlikely because the rice and water were consumed together, and zinc from different foods typically readily exchanges in the gastrointestinal tract (39, 49). In addition, a recent study showed that zinc absorption from fortified water consumed with maize porridge was similarly low compared with zinc absorption from maize porridge with unfortified water (40). Third, the randomization of treatments over time points accounted for differences in effect over time, with the exception of the UM treatment. Only 10 of the 19 individuals participating in the milk study received the UM treatment. Therefore, the small nonsignificant difference in zinc bioavailability from raw milk compared with UHT milk can only be regarded as indicative. Zinc is one of the many essential nutrients found in milk. With a concentration of ;0.4 mg Zn/100 g semiskimmed milk, it forms an important source of zinc in dairy-consuming populations such as in The Netherlands (50). Milk consumption is on the increase in Asian countries where rice is the main staple food. In some of those countries, zinc deficiency is of public health concern (51). The combined consumption of rice with milk can hypothetically increase the total amount of zinc that is absorbed from rice. For example, Chinese women ingest, on average, 4.1 mg Zn from rice (52). When consumed with milk, the total absorbed zinc from rice would increase from 0.53 to 0.85 mg when applying the FAZ values found in the present study. This amount does not yet include the absorbed zinc from the milk itself and it does not take into account that FAZ could be lower due to higher inflammation rates in a different population, as our results suggest. Therefore, it remains to be explored whether this could be an acceptable and effective strategy to improve the supply of absorbable zinc in the diet. In conclusion, we have shown that the combination of milk with a phytate-rich rice meal enhances zinc solubility in vitro and increases the FAZ from rice by 62% in humans. Furthermore, we found that dilution of UHT milk with water decreases the FAZ, indicating that the enhancement of zinc absorption by milk components is concentration dependent and suggests that the addition of milk solids may enhance zinc absorption. Furthermore, zinc absorption from milk seems unaffected by UHT thermal processing, and the consumption of raw milk instead of UHT milk for the enhancement of zinc absorption does not seem justified.

and calculated FAZ values; AM-B: performed the data analysis and had primary responsibility for the final content of the manuscript; EFT: wrote the first draft of the manuscript; EFT, DM, SCL, RD, EGHMvdH, AF, EB, TJS, CZ, and AM-B: contributed to follow-up versions of the manuscript, in particular DM; and all authors: read and approved the final manuscript.

References 1.

2. 3.

4.

6.

7.

8.

9. 10.

11. 12.

13.

14.

15.

16.

17.

18. 19.

Zinc absorption from milk and rice

7 of 8


5.

Black RE, Victora CG, Walker SP, Bhutta ZA, Christian P, de Onis M, Ezzati M, Grantham-McGregor S, Katz J, Martorell R, et al. Maternal and child undernutrition and overweight in low-income and middleincome countries. Lancet 2013;382:427–51. Borwankar R, Sanghvi T, Houston R. What is the extent of vitamin and mineral deficiencies? Food Nutr Bull 2007;28:S174–81. Brown KH, Peerson JM, Rivera J, Allen LH. Effect of supplemental zinc on the growth and serum zinc concentrations of prepubertal children: a meta-analysis of randomized controlled trials. Am J Clin Nutr 2002;75:1062–71. Brown KH, Rivera JA, Bhutta Z, Gibson RS, King JC, Lo¨nnerdal B, Ruel MT, Sandtro¨m B, Wasantwisut E, Hotz C; International Zinc Nutrition Consultative Group. Assessment of the risk of zinc deficiency in populations and options for its control. Food Nutr Bull 2004;25:S99–203. Gibson RS, Bailey KB, Gibbs M, Ferguson EL. A review of phytate, iron, zinc, and calcium concentrations in plant-based complementary foods used in low-income countries and implications for bioavailability. Food Nutr Bull 2010;31:S134–46. Rosado JL, Diaz M, Gonzalez K, Griffin I, Abrams SA, Preciado R. The addition of milk or yogurt to a plant-based diet increases zinc bioavailability but does not affect iron bioavailability in women. J Nutr 2005;135:465–468. Egan CB, Smith FG, Houk RS, Serfass RE. Zinc absorption in women: comparison of intrinsic and extrinsic stable-isotope labels. Am J Clin Nutr 1991;53:547–53. Fairweather-Tait SJ, Fox TE, Wharf SG, Eagles J, Crews HM, Massey R. Apparent zinc absorption by rats from foods labelled intrinsically and extrinsically with 67Zn. Br J Nutr 1991;66:65–71. Lo¨nnerdal B. Dietary factors influencing zinc absorption. J Nutr 2000;130(Suppl):1378S–83S. Garc´ıa-Nebot MJ, Alegr´ıa A, Barbera´ R, Clemente G, Romero F. Addition of milk or caseinophosphopeptides to fruit beverages to improve iron bioavailability? Food Chem 2010;119:141–8. Miquel E, Farré R. Effects and future trends of casein phosphopeptides on zinc bioavailability. Trends Food Sci Technol 2007;18:139–43. Seiquer I. Valverde A, Delgado-Andrade C, Navarro MP. Influence of heat treatment of casein in presence of reducing sugars on Zn solubility and Zn uptake by Caco-2 cells after in vitro digestion. J Physiol Biochem 2000;56:237–46. Wegmu¨ller R, Tay F, Zeder C, Brnic MHR. Zinc absorption by young adults from supplemental zinc citrate is comparable with that from zinc gluconate and higher than from. J Nutr 2014;144:132–6. Brnić M, Wegmu¨ller R, Zeder C, Senti G, Hurrell RF. Influence of phytase, EDTA, and polyphenols on zinc absorption in adults from porridges fortified with zinc sulphate or zinc oxide. J Nutr 2014; 144:1467–73. Friel JK, Naake VL, Miller LV, Fennessey PV, Hambidge KM. The analysis of stable isotopes in urine to determine the fractional absorption of zinc. Am J Clin Nutr 1992;55:473–7. Lowe NM, Woodhouse LR, Matel JS, King JC. Comparison of estimates of zinc absorption in humans by using 4 stable isotopic tracer methods and compartmental analysis. Am J Clin Nutr 2000;71:523–9. Streppel MT, de Vries JHM, Meijboom S, Beekman M, de Craen AJ, Slagboom PE, Feskens EJ. Relative validity of the food frequency questionnaire used to assess dietary intake in the Leiden Longevity Study. Nutr J 2013;12:75. Stomph TJ, Jiang W, van der Putten PEL, Struik PC. Zinc allocation and re-allocation in rice. Front Plant Sci 2014;5:8. Pabon ´ ML, Lo¨nnerdal B. Bioavailability of zinc and its binding to casein in milks and formulas. J Trace Elem Med Biol 2000;14:146–53.

20. Krebs NE, Hambidge KM. Zinc metabolism and homeostasis: the application of tracer techniques to human zinc physiology. Biometals 2001;14:397–412. 21. Makower R. Extraction and determination of phytic acid in beans (Phaseolus vulgaris). Cereal Chem 1970;47:523–9. 22. Gabay C, Kushner I. Acute-phase proteins and other systemic responses to inflammation. N Engl J Med 1999;340:448–54. 23. Thurnham DI, McCabe LD, Haldar S, Wieringa FT. NorthropClewes CA, McCabe GP. Adjusting plasma ferritin concentrations to remove the effects of subclinical inflammation in the assessment of iron deficiency: a meta-analysis. Am J Clin Nutr 2010;92:546–55. 24. Hedrén E, Diaz V, Svanberg U. Estimation of carotenoid accessibility from carrots determined by an in vitro digestion method. Eur J Clin Nutr 2002;56:425–30. 25. Haug W, Lantzsch H-J. Sensitive method for the rapid determination of phytate in cereals and cereal products. J Sci Food Agric 1983; 34:1423–6. 26. Sandstro¨m B, Arvidsson B, Cederblad A, Bjorn-Rasmussen E. Zinc absorption from composite meals. I. The significance of wheat extraction rate, zinc, calcium, and protein content in meals based on bread. Am J Clin Nutr 1980;33:739–45. 27. Sandstro¨m B, Cederblad A. Zinc absorption from composite meals II. Influence of the main protein source. Am J Clin Nutr 1980;33 (8):1778–83. 28. Hansen M, Sandstro¨m B, Jensen M, Sørensen SS. Effect of casein phosphopeptides on zinc and calcium absorption from bread meals. J Trace Elem Med Biol 1997;11:143–9. 29. Hansen M, Sandstro¨m B, Jensen MSS. Casein phosphopeptides improve zinc and calcium absorption from rice-based but not from whole-grain infant cereal. J Pediatr Gastroenterol Nutr 1997;24:56–62. 30. Garc´ıa-Nebot MJ, Alegr´ıa A, Barbera´ R, Clemente G, Romero F. Does the addition of caseinophosphopeptides or milk improve zinc in vitro bioavailability in fruit beverages? Food Res Int 2009;42:1475–82. 31. Scho¨lmerich J, Freudemann A, Kottgen E, Wietholtz H, Steiert B, Lo¨hle E, Haüssinger D, Gerok W. Bioavailability of zinc from zinchistidine complexes: comparison with zinc-sulfate in healthy men. Am J Clin Nutr 1987;45:1480–6. 32. Forbes RM, Erdman JW, Parker HM, Kondo H, Ketelsen SM. Bioavailability of zinc in coagulated soy protein (tofu) to rats and effect of dietary calcium at a constant phytate:zinc ratio. J Nutr 1983;113: 205–10. 33. Etcheverry P, Grusak MA, Fleige LE. Application of in vitro bioaccessibility and bioavailability methods for calcium, carotenoids, folate, iron, magnesium, polyphenols, zinc, and vitamins B6, B12, D, and E. Front Physiol 2012;3:317–22. 34. Bosscher D, Lu Z, Janssens G, Van Caillie-Bertrand M, Robberecht H, De Rycke H, De Wilde R, Deelstra H. In vitro availability of zinc from infant foods with increasing phytic acid contents. Br J Nutr 2001; 86:241–7. 35. Miller LV, Krebs NF, Hambidge KM. Mathematical model of zinc absorption: effects of dietary calcium, protein and iron on zinc absorption. Br J Nutr 2013;109:695–700. 36. Sandstro¨m B, Cederblad ALB. Zinc absorption from human milk, cowÕs milk, and infant formulas. Am J Dis Child 1983;137:726–9. 37. Valberg LS, Flanagan PR, Chamberlain MJ. Effects of iron, tin, and copper on zinc absorption in humans. Am J Clin Nutr 1984;40:536–41. 38. Sandstro¨m B, Cederblad A. Effect of ascorbic acid on the absorption of zinc and calcium in man. Int J Vitam Nutr Res 1987;57:87–90. 39. Sandstro¨m B, Davidsson L, Cederblad A, Lo¨nnerdal B. Oral iron, dietary ligands and zinc absorption. J Nutr 1985;115:411–4. 40. Galetti V, Kujinga P, Mitchikpe` CES, Zeder C, Tay F, Tossou F, Hounhouigan JD, Zimmermann MB, Moretti D. Efficacy of highly bioavailable zinc from fortified water: a randomized controlled trial in rural Beninese children. Am J Clin Nutr 2015;102:1238–48. 41. Drago SR, Binaghi MJ, Valencia ME. Effect of gastric digestion pH on iron, zinc, and calcium dialyzability from preterm and term starting infant formulas. J Food Sci 2005;70:S107–12. 42. Collins PJ, Horowitz M, Cook DJ, Harding PE, Shearman DJ. Gastric emptying in normal subjects—a reproducible technique using a single scintillation camera and computer system. Gut 1983;24:1117–25. 43. Ziessman HA, Chander A, Clarke JO, Ramos A, Wahl RL. The added diagnostic value of liquid gastric emptying compared with solid emptying alone. J Nucl Med 2009;50:726–31.

44. Islam MM, Woodhouse LR, Hossain MB, Ahmed T, Huda MN, Ahmed T, Peerson JM, Hotz C, Brown KH. Total zinc absorption from a diet containing either conventional rice or higher-zinc rice does not differ among Bangladeshi preschool children. J Nutr 2013;143:519–25. 45. Miller LV, Hambidge KM, Krebs NF. Zinc absorption is not related to dietary phytate intake in infants and young children based on modeling combined data from multiple studies. J Nutr 2015;145:1763–9. 46. Brnihttps://www.ncbi.nlm.nih.gov/pubmed/?term=Brni%C4%87%20M %5BAuthor%5Dcauthor=truecauthor_uid=26674764 M, Wegmu¨ller R, Melse-Boonstra A, Stomph TJ, Zeder C, Tay FM, Hurrell RF. Zinc absorption by adults is similar from intrinsically labeled zinc-biofortified rice and from rice fortified with labeled zinc sulfate. J Nutr 2016;146:76–80. 47. Boza JJ, Fox TE, Eagles J, Wilson PDG, Fairweathertait SJ. The validity of extrinsic stable isotopic labeling for mineral absorption studies in rats. J Nutr 1995;125:1611–6.

48. Fox TE, Fairweather-Tait SJ, Eagles J, Wharf SG. Intrinsic labelling of different foods with stable isotope of zinc (67Zn) for use in bioavailability studies. Br J Nutr 1991;66:57–63. 49. Sandstro¨m B. Dose dependence of zinc and manganese absorption in man. Proc Nutr Soc 1992;51:211–8. 50. Vissers PAJ, Streppel MT, Feskens EJM, de Groot LCPGM. The contribution of dairy products to micronutrient intake in The Netherlands. J Am Coll Nutr 2011;30:415S–21S. 51. Wessells KR, Brown KH. Estimating the global prevalence of zinc deficiency: results based on zinc availability in national food supplies and the prevalence of stunting. PLoS One 2012;7:e50568. 52. Qin Y, Melse-Boonstra A, Yuan B, Pan X, Dai Y, Zhou M, Wegmu¨ller R, Zhao J, Kok FJ, Shi Z. Zinc biofortification of rice in China: a simulation of zinc intake with different dietary patterns. Nutrients 2012;4:517–28.


8 of 8

Talsma et al.