Journal of the American College of Nutrition, Vol. 24, No. 5, 334-341 (2005)
Published by the American College of Nutrition
Factors Influencing Zinc Status of Apparently Healthy Indians
Vaishali V. Agte, PhD,
Shashi A. Chiplonkar, PhD and
Kirtan V. Tarwadi, MSc
Agharkar Research Institute, Pune, INDIA
Address reprint requests to: Vaishali V Agte, PhD, Agharkar Research Institute, G.G. Agarkar Road, Pune 411 004, INDIA. E-mail: vaishaliagte{at}hotmail.com
 |
ABSTRACT
|
|---|
Objectives: To identify dietary, environmental and socio-economic factors associated with mild zinc deficiency, three zinc status indices; erythrocyte membrane zinc (RBCMZn), plasma zinc and super oxide dismutase (SOD) were assessed in free living and apparently healthy Indian population.
Methods: Dietary patterns of 232 men and 223 women (20–65 yr) from rural, industrial and urban regions of Western India were evaluated by food frequency questionnaire. RBCMZn was estimated using atomic absorption spectrometry, hemoglobin and serum ceruloplasmin by spectrophotometer. On a sub sample (48 men and 51 women) plasma zinc and SOD were also assessed.
Results: Mean RBCMZn was 0.5 ± 0.1 µmols/g protein with 46% individuals showing zinc deficiency. Mean plasma zinc was 0.98 ± 0.12 µg/mL with 25% men and 2.5% women having values below normal range. Mean SOD was 0.97 ± 0.1 (u/mL cells). A significant positive correlation was observed between intakes of green leafy vegetables, other vegetables and milk products with RBCMZn status (p < 0.05). But these were not correlated with plasma zinc (p > 0.2). Cereal and legume intakes were negatively correlated with RBCMZn (p < 0.05) but positively correlated with plasma zinc (p < 0.05) and not correlated with SOD (p > 0.2). Fruit and other vegetable intake were positively correlated with SOD (p < 0.05) alone. Logistic regression analyses revealed that RBCMZn was positively associated with intakes of ß-carotene, zinc and environmental conditions and negatively associated with family size (p < 0.05). Plasma zinc indicated positive association with zinc, thiamin and riboflavin intakes (p < 0.05) and SOD showed negative association with iron and family size.
Conclusion: RBCMZn was a more sensitive indicator of long-term zinc status than plasma zinc and SOD. Prominent determinants of zinc status were intakes of beta-carotene and zinc along with environmental conditions and family size.
Key words: erythrocyte membrane zinc, vegetarian diets, plasma zinc, super oxide dismutase, socio-environmental factors
|
INTRODUCTION
|
|---|
Prevalence of nutritional zinc deficiency is predicted to be high in developing countries [1–2]. Micronutrient deficiencies have also been reported in adult Indian populations [3–5]. However studies reporting status of zinc and factors affecting zinc status are scanty. Dietary factors such as intakes of zinc, calcium, phytate and non-dietary factors like socio-economic status, environmental conditions may be responsible for causing zinc deficiency. Indian diets have higher phytate:zinc molar ratios which inhibit trace element absorption. However in Indian cooking processes, main inhibitory factor of zinc bioavailability, viz. phytate, gets partially degraded and may not remain as a strong inhibitor [6]. The impact of these factors on zinc status of Indians needs to be examined.
The question about the most appropriate indicator of blood zinc status is still unresolved since plasma zinc is not considered as a good biomarker of long-term zinc status [7]. Among tissue level indicators of zinc status, plasma zinc, nail and hair zinc are easily accessible and measurable, but not considered so specific for evaluating long-term zinc status [8]. Erythrocyte membrane zinc (RBCMZn) has been used as a biomarker of zinc status during prospective zinc depletion and repletion trial on human subjects [9]. As compared to total erythrocyte zinc, the erythrocyte membrane zinc has been stated as better indicator of zinc status [10]. Comparison of RBCMZn, plasma zinc and super oxide dismutase (SOD) was undertaken from the point of view of sensitivity of these indices to dietary, socio economic and environmental factors.
|
METHODS AND MATERIALS
|
|---|
Subjects
Healthy men (n = 232) and women (n = 223) (20–59 yr) were surveyed through health camps in rural and urban communities in Western India, for anthropometry, dietary intake of food and nutrients, blood micronutrient profile, socio-economic status, and environmental conditions. The subjects participated in the study voluntarily and gave an informed written consent. Although some subjects consumed animal foods, the intake was small and frequency was low. Therefore subjects could be considered as vegetarian population.
Subjects underwent a routine medical check-up by a medical doctor, which included physical examination, blood pressure, and past and present health complaints. Subjects having medical fitness were considered as apparently healthy and included in the study population. Subjects with chronic complaints like high blood pressure, diabetes mellitus, asthma and pregnant and lactating women were excluded from the study.
Information on age, environment and socio-economic status were collected through a separate questionnaire by interview method. Socioeconomic information was collected through questions on education, occupation, total monthly income of the family, and family size. Per capita daily income was calculated and classified into three categories of income viz.; lower (LSE), middle (MSE), higher (HSE) as described earlier [11].
Environmental Conditions
Environmental factors such as type of housing, ventilation, cleanliness of surroundings, overcrowding, water supply, sewage system, traffic pollution and smoking exposure were examined. Each of these factors was given a score on a 10-point scale. Scores of all the variables were added, maximum score being 100. Pooled score above 70 was termed as Good, between 50 to 70 Fair, 35 to 50 as Bad and less than 35 as Poor.
Dietary Nutrient Intakes
A specially designed semi-quantitative food frequency questionnaire (FFQ) for evaluating habitual micronutrient intakes was used to estimate average daily dietary intake of food groups and nutrients. The FFQ covered over 278 food items consumed throughout the year. The subjects were asked how often on an average they had consumed each food item during the year, the frequency of consumption per day, and amount in terms of standard measures like cup, bowl, spoon etc. Weights of the portion units were decided on the basis of average weight of each food item for that portion size collected from different households representing the sample. Daily food consumption under eight food groups was computed for every individual using the information in FFQ about frequency and quantity. Details of the FFQ are given elsewhere [11].
The average daily nutrient intakes, viz.; energy, protein, fat, riboflavin, folic acid, ß-carotene, vitamin C, sodium, potassium, calcium, magnesium, copper, zinc and iron were derived using food consumption data from FFQ. The nutritive values were taken from the database generated in our laboratory for a variety of cooked foods and the food tables [12].
Biochemical Analysis
Fasting venous blood (10 mL) was collected for every subject in plain and EDTA-coated sterile bulbs. Hemoglobin was estimated by cyanomethemoglobin method (Qualigens Diagnostics, Glaxo, India Ltd.). Erythrocyte membrane zinc (RBCMZn) was estimated by atomic absorption spectrophotometry (UNICHEM) as per the method of Ruz et al [10]. Reference serum (RANDOX) was used as a reference standard for all blood estimations. Plasma zinc was estimated by atomic absorption spectrophotometry. Super oxide dismutase (SOD) was estimated by the method of Beauchamp and Friedowich modified by Vidyawati and Krishnamurthy [13]. Serum ceruloplasmin was measured using paraphenylene diamine oxidation using spectrophotometry.
Statistical Methods
Statistical analyses were carried out using SPSS version 11.0 under Windows. Normality of the data was tested by Kolmogorov-Smirnov test. In case of non-normal data, median was used to describe the data. One-way ANOVA and Dunnet’s T3 multiple comparison test was used to test the differences between various parameters between rural and urban classes as well as gender differences. Spearman’s rank correlations were obtained to study associations between intakes of cereal, legume, fruit, vegetables, animal foods, as well as socioeconomic and environmental variables. Logistic regression analyses were performed by making erythrocyte membrane zinc or plasma zinc or SOD as a dichotomous response variable one at a time and nutrient intakes, socioeconomic and environmental conditions as explanatory variables.
|
RESULTS
|
|---|
Tables 1a and 1b describe general characteristics, socio-economic and environmental status of the study population. Average family size was 6 in rural, 5 in industrial, 3 in tribal, and 4 in urban regions. Education level was high in HSE, and low in LSE and tribal groups, the difference being statistically significant (p < 0.01). Environmental conditions were better in HSE and MSE groups than other regions (p < 0.045). Table 1c gives RBCMZn, plasma zinc, SOD, serum ceruloplasmin and hemoglobin levels. With multiple comparison test, it was found that: RBCMZn of urban HSE and LSE men were significantly higher than men from other regions (p = 0.001) while RBCMZn of urban HSE, MSE and LSE women were higher than the other regions (p = 0.001). Hemoglobin of LSE men was significantly lower than other regions (p = 0.0001) except HSE (p > 0.5). No significant differences were seen in hemoglobin of all women with the averages ranging from 12.1 to 12.9 g/dL.
Distribution of RBCMZn of both men and women was normal (p = 0.134 and p = 0.067 respectively) by Kolmogorov-Smirnov test. The difference between mean RBCMZn of men (0.491 ± 0.107) and women (0.498 ± 0.114) was not statistically significant (p = 0.505). RBCMZn in normal individuals has been reported as 0.72 ± 0.21 [10]. A value one S.D. below mean, which is 0.51 mg/g protein, was taken as cut off for normal individuals. Based upon this 46.6% men and 45.7% women had their RBCMZn below the cut off indicating zinc deficiency even in these apparently healthy adults. Plasma zinc was below normal range (0.8–1.1 µg/mL) in 25% men but only 2.5% in women.
Dietary intakes of green leafy vegetables (GLV) were lowest (5.4 ± 4.7 g/day) in rural women and highest in urban higher and lower socioeconomic classes (35.4 ± 26.2 g/day). Average fruit intakes were lowest in rural women (24.2 ± 20 g/day) and highest in urban HSE class (81.7 ± 67.2 g/day). Cereal and legume intakes were inadequate in women from rural and LSE groups. Table 2 describes median nutrient intakes of the subjects. Zinc intakes of tribals were significantly lower than other regions (p = 0.001) while zinc intakes of industrial workers were higher than LSE (p = 0.001). Beta-carotene intakes were higher in HSE and MSE subjects than other regions (p < 0.05).
When associations of food intakes and zinc status were examined, statistically significant positive rank correlations amongst RBCMZn and intakes of leafy vegetable, fruit and milk were observed (R = 0.16, 0.14, 0.17 respectively with p < 0.05). Plasma zinc was associated with legume intake (r = 0.255, p = 0.023) and SOD was associated with fruit intake (R = 0.245, p = 0.031). Regarding the non-dietary factors, RBCMZn was negatively associated with environmental conditions and family size, and both plasma zinc and SOD were negatively associated with family size alone (r = 0.165, 0.20, p < 0.05 respectively).
RBCMZn was negatively correlated with hemoglobin (r = –0.19, p = 0.001) and positively correlated with serum ceruloplasmin (r = 0.21, p = 0.001). Plasma zinc and SOD were neither correlated with hemoglobin nor with ceruloplasmin (p > 0.5). RBCMZn and plasma zinc were positively and significantly correlated with plasma retinol (r = 0.27, 0.268, p = 0.001 & 0.017 respectively). RBCMZn was weakly negatively correlated with plasma zinc (r = –0.215, p = 0.062) but not with SOD. There was no significant association between plasma zinc and SOD, except in case of women, plasma zinc and SOD were significantly positively correlated (r = 0.498, p = 0.01).
To investigate interrelationships of the factors affecting zinc status indicators, logistic regression analyses were carried out. In the univariate analyses BMI, sex were found to be statistically nonsignificant (p > 0.25). Hence in the multivariate model these factors were not included. Multiple logistic regression for RBCMZn with food intakes, and other non-dietary factors as explanatory variables indicated that legume, fruit, milk intakes and environment, income and family size as the significant influencing factors (Table 3). For plasma, cereal intake was the only influencing factor while for SOD, fruit, milk and animal food intakes and family size were the associated factors.
Multiple logistic regressions for RBCMZn for linkages of nutrient intakes and other factors resulted in significant odds ratios (OR) for beta-carotene intake, zinc intake, environmental conditions and family size (Table 4). Similarly risk of lower plasma zinc was 1.7 times with lower zinc intakes and equal with lower riboflavin and thiamine intakes (Table 4). In case of SOD, larger family size increases the risk of lower SOD by 1.56 times and by 1.18 times with lower iron intakes.
|
DISCUSSION
|
|---|
It has been stipulated that nearly half of the world’s population is at the risk of inadequate zinc intake suggesting that public health programmes are urgently needed to control zinc deficiency [1]. The deficiency of zinc, an essential micronutrient for human growth, development, and immune function, impairs overall immune function and resistance to infection [14,15]. In India, sub-clinical zinc deficiencies have been found in Northern population [4]. For want of proper indicator, it is still difficult to judge zinc status and its linkage with health disorders. In our earlier studies zinc status was found to be lower in hypertension and cataract patients than normal individuals [16,17].
Due to the universal nature of zinc metabolism in humans, there are special challenges in assessing human zinc status, and thus in identifying individuals and populations who are at risk for zinc deficiency. The cereal proteins consumed by individuals in developing countries contain large quantities of phytates, organic compounds that bind dietary zinc and render it unavailable for absorption. It was therefore necessary to investigate the prevalence of zinc deficiency in communities using cereal-based diets [15,18].
Zinc, an essential trace element, has antioxidant functions, stabilizes membranes, and plays a role in the activity of a host of Zn metalloenzymes. Erythrocytes and leukocytes carry considerable zinc, most of it rather tightly bound. In erythrocytes, zinc is bound as part of the enzymes Cu, Zn-SOD and carbonic anhydrase. The cell membrane also carries zinc and will lose it in cases of mild zinc insufficiency. Zinc binding to RBC membranes was reported to be rapid and readily reversible in a dynamic equilibrium with its binding sites [19]. Assessment of erythrocyte membrane zinc was therefore done as an indicator of long-term zinc status. Zn deficiency has been shown to increase erythrocyte fragility, decrease the Zn content of the erythrocyte membrane, and alter erythrocyte membrane fluidity [20]. The osmotic fragility of the erythrocytes depends partially on the status of Zn [21]. The data obtained by McClain et al [22] provide evidence that the erythrocyte membrane contains an intrinsic blocking mechanism whose efficacy can be synergistically enhanced by Zn2+ ions protecting the cells against the lytic effect of melittin. Although erythrocyte fragility has not been reported in the present paper, we found erythrocyte zinc uptake and osmotic fragility to be promising biomarkers of long-term zinc status [23]. These observations coupled with present data support use of RBC membranes as screening material for zinc status.
Essential trace elements such as zinc, copper, chromium and selenium take part in a variety of enzymatic processes as cofactors at cellular level. Ignoring these important biochemical roles, trace element concentrations determined in whole blood or plasma very often lead to conclusions contrary to the actual intracellular concentration. Significantly more information about trace element status can be obtained by investigating concentrations in blood cells instead of only evaluating the concentrations in plasma [24]. Secondly main functions of zinc are intracellular, therefore, like other minerals; serum and plasma levels do not reflect total body stores. Fever, infections, stress, pregnancy, and restriction on food intake and protein depletion are known to affect plasma zinc. RBCMZn showed a negative correlation with hemoglobin and positive correlation with ceruloplasmin reaffirming antagonism between zinc and iron. But plasma zinc and SOD were neither correlated with hemoglobin nor with ceruloplasmin indicating that RBCMZn to be a more sensitive indicator of zinc status than plasma zinc and SOD.
Reported value of erythrocyte membrane zinc lie in the range of 0.72 ± 0.21 µmols/g protein [10] in young adults. Our values are on the lower side than this range. In fact, 46.6% men and 45.7% women were zinc deficient even in these apparently healthy adults. On the other hand low plasma zinc values were less prevalent in spite of low zinc intakes. This fact further supports the greater sensitivity of RBCMZN than plasma zinc. RBCMZn in healthy men at the baseline was reported to be 120 ng/mg protein with S.D. of about 27 ng/mg protein by Woodhouse et al [9]. In addition to being traditionally vegetarians, the age of our study population was in the range 20–65 years, higher than those previously reported [9].
Among dietary factors, though intakes of phytates, fiber, other micronutrients like vitamin C, protein were considered as the variable in the model, only beta-carotene, zinc and phytate intake emerged as influential dietary factors. Apart from known risk factors like phytate, association of beta-carotene on zinc status is a new finding. Beta-carotene emerged as a factor influencing erythrocyte membrane zinc in addition to dietary zinc level probably arising due to low intakes of legumes, fruits and milk. In our recent study on ileostomy subjects and young healthy adults a significant association of zinc and beta-carotene was obtained [25]. Beta-carotene being a lipophilic antioxidant, may have a role in stabilizing the lipid-protein bilayer of erythrocyte membrane or in maintaining the zinc domain. Vitamin A and beta-carotene can improve nonheme iron absorption from rice, wheat, and corn by humans [26]. Further, experiments are necessary to confirm findings on beta-carotene and zinc status and whether the relationship is causal or arising due to some common root cause.
Dietary intakes of zinc, riboflavin and thiamine were found to be sole influencing factors for plasma zinc and no non-dietary factor was present in the final equation. This suggests that plasma zinc is more affected by concomitant intakes of other micronutrients rather than non-dietary factors. For SOD only iron intake and family size were significantly associated. Vijayalakshmi et al and RB Singh, et al, have reported mean plasma zinc (0.98 ± 0.04, 0.957 ± 0.115 µg/mL respectively) in Indians [27,28], which is of the same order of magnitude as in the present study.
Environment and family size were significant risk factors for RBCMZn status amongst non-dietary factors and indicates zinc status to have socio economic and environmental impact. Better environmental conditions reduce morbidity due to various infections. As a result of the interplay of host, dietary and environmental factors zinc deficiency is not uncommon on global basis. [29]. Higher family size is usually seen in lower income groups. Socioeconomic status affects cereal intake and in turn zinc intake in vegetarian populations like India. That is why higher odds were observed for family size against RBCMZn and SOD in present data. In conclusion, dietary intakes of zinc, beta carotene, riboflavin, thiamin, iron and phytate were major determinants of zinc status and RBCMZn was a better indicator of zinc status.
|
ACKNOWLEDGMENTS
|
|---|
Authors thank Dr. K.M. Paknikar, Geomicrobiology unit for help in trace metal analysis by atomic absorption spectrometry. This work is part of the project funded by Department of Science and technology, New Delhi, India (Project No: SP/SO/B39/94).
|
FOOTNOTES
|
|---|
This paper was presented at the conference [Zinc and Human Health] organized by International Zinc Association held in Sweden, June 2002.
Received May 21, 2004.
Accepted May 19, 2005.
|
REFERENCES
|
|---|
- Brown KH, Wuehler SE, Peerson JM: The importance of zinc in human nutrition and estimation of the global prevalence of zinc deficiency. Food Nutr Bull22
:113
–125,2002
.
- Sandstead HH, Darnell LS, Alcock NW: Association of iron and zinc nutrition. FASEB J5
:A1291
,1991
.
- Gopalan C: Micronutrient deficiencies. NFI Bulletin,1996
.
- Chakravarty I, Sinha RK: Prevalence of micronutrient deficiency based on results obtained from the national pilot program on control of micronutrient malnutrition. Nutr Rev60
:S53
–S58,2002
.[Medline]
- Chiplonkar SA, Mengale SS, Agte VV: Relative importance of micronutrient deficiencies in iron deficiency anemia. Nutr Res23
:1355
–1367,2003
.
- Agte VV, Tarwadi KV, Chiplonkar SA: Phytate degradation during traditional cooking: significance of the phytic acid profile in cereal-based vegetarian meals. J Food Comp Anal12
:161
–167,1999
.
- World Health Organization: Trace Elements in Human Nutrition and Health. Geneva: WHO, pp75
,1996
.
- Sandstrom B: Bioavailability of zinc. Euro J Clin Nutr51
:S17
–S19,1997
.
- Woodhouse LR, Bonnel E, Sutherland B, King JC: The effect of marginal zinc intake on measures of cellular membrane function in men. 11th International Symposium on Trace Elements in Man and Animals Abstracts [Abstract 193a], Berkley, California, June 2–6,2002
.
- Ruz M, Cavan KR, Bettger WJ, Gibson RS: Erythrocytes, erythrocyte membranes, neutrophils and platelets as biopsy materials for the assessment of zinc status in humans. Br J Nutr68
:515
–27,1992
.[Medline]
- Chiplonkar SA, Agte VV, Mengale SS, Tarwadi KV: Are lifestyle factors good predictors of retinol and vitamin C deficiency in apparently healthy adults? E J Clin Nutr56
:96
–1042002
.
- Gopalan C, Ramasastri BV, Balasubramanian SG: Revised and updated by Rao BSN, Deosthale YB, Pant KC: [Nutritive Value of Indian Foods.] Hyderabad: National Institute of Nutrition,1993
.
- Vidyawathi U, Krishnamurthy S: Purification and specificity of cuprozinc super oxide dismutase from Dolichos biflorus. Medical Nutr Res Communic1
:33
–39,1993
.
- Walker CF, Black RE: Zinc and the risk for infectious disease. Ann Rev Nutr24
:255
–275,2004
.[Medline]
- Prasad AS: Clinical spectrum of human zinc deficiency. In Prasad AS, ed. [Biochemistry of Zinc.] New York: Plenum:219
–258,1993
.
- Chiplonkar SA, Agte VV, Tarwadi KV, Paknikar KM, Diwate UP: Micronutrient deficiencies as predisposing factors for hypertension in lacto-vegetarian Indian adults. J Am Coll Nutr23
:239
–247,2004
.[Abstract/Free Full Text]
- Tarwadi K, Agte V: Linkages of antioxidant, micronutrient and socio-economic status with the degree of oxidative stress & lens opacity in Indian cataract patients. Nutrition20
:261
–267,2004
.[Medline]
- Prasad AS: Zinc deficiency in women, infants and children. J Am Coll Nutr15
:113
–120,1996
.[Abstract]
- Brandao-Neto J, Bell WR: Characteristics of zinc binding to human red blood cell membranes Am J Hematol45
:1
–9,1994
.[Medline]
- McClain C, Morris P, Hennig B: The effect of dietary fats on the hemolysis resistance of the erythrocyte membrane during alimentary zinc and calcium deficiency in rats. Z Ernahrungswiss30
:98
–108,1991
.[Medline]
- Rudenko SV, Nipot EE, Pavliuk OM: Zinc and endothelial function. Nutrition11(1 Suppl)
:117
–120,1995
.
- Roth HP, Kirchgessner M: The effect of Zn ions on erythrocyte hemolysis induced by melittin [Russian]. Biokhimiia60
:723
–733,1995
.[Medline]
- Agte VV, Nagmote RV and Chiplonkar SA: Role of vitamin-zinc interactions on in vitro zinc uptake by human erythrocytes. Biol Trace Elements Res98
:1
–13,2004
.[Medline]
- Kruse-Jarres JD, Rukgauer M: Trace elements in diabetes mellitus. Peculiarities and clinical validity of determinations in blood cells. J Trace Elem Med Biol.14
:21
–27,2000
.
- Agte VV, Jahagirdar M, Chiplonkar SA: Apparent absorption of eight micronutrients and phytic acid from vegetarian meals in human ileostomy volunteers. Nutrition (in press).
- Garcia-Casal MN, Layrisse M, Salana L: Vitamin A and beta-carotene can improve non-heme iron absorption from rice, wheat, and corn by humans. J Nutr128
:646
–650,1998
.[Abstract/Free Full Text]
- Vijaylakshmi P, Rema SR: Biochemical profile of selected normal healthy adult men and women. Indian J Nutr Diet31
:277
–284,1994
.
- Singh RB, Gupta UC, Mittal N, Niaz MA, Ghosh S, Rastogi V: Epidemiologic study of trace elements and magnesium on risk of coronary artery disease in rural and urban Indian population. J Am College Nutr16
:62
–67,1997
.[Abstract]
- Crebs NI: Overview of zinc absorption and excretion in human gastrointestinal tract. J Nutr130
:1374S
–1377S,2000
.[Abstract/Free Full Text]
TOP
ABSTRACT
INTRODUCTION
