@Copyright 1987 by The Humana Press Inc. All rights of any nature, whatsoever, reserved. 0163-4984/87/1200~247502.20
Zinc, Copper, and Iron Nutrition Studied with Enriched Stable Isotopes JUDITH R. TURNLUND USDA, Western Human Nutrition Research Center, 800 Buchanan Street, Berkeley, CA 94710
ABSTRACT Enriched stable isotopes were used in nutrition studies of normal, healthy adults to measure zinc, copper, and iron absorption. After obtaining baseline values for zinc, copper, and iron absorption from diets adequate in all nutrients, the effects of age, pregnancy, and several dietary variables were studied. Stable isotopes of zinc, copper, and iron were incorporated into diets. Complete fecal samples were collected and the unabsorbed isotopes remaining in the samples were measured by thermal ionization mass spectrometry, the most precise analytical method for the determination of stable mineral isotopes. Stable isotopes were also infused in five young men to evaluate the potential of studying mineral utilization and kinetics with stable isotopes. The results of these studies demonstrate that a number of factors can affect mineral absorption, but the specific effects differ for different minerals. Isotopic enrichments could be measured in urine and blood, so kinetic studies of utilization of essential minerals are now feasible with enriched stable isotopes. Continued use of stable isotopes to determine mineral absorption, combined with stable-isotope studies of mineral utilization, balance data, and biochemical indicators of mineral status, should result in a better understanding of mineral requirements and metabolism under a variety of conditions. Index Entries: Stable isotopes; zinc; copper, iron; bioavailability; absorption; mass spectrometry; nutrition.
INTRODUCTION This paper reviews the approach used by the author and her collaborators to study zinc, copper, and iron absorption with enriched stable Biological Trace Element Research
247
Vol. 12, 1987
248
Turnlund
isotopes. Experimental design, critical features of experiments, and the analytical approach are summarized. Recent studies measuring isotopic enrichments in urine, blood, and excretion into the gastrointestinal tract are also included. Research using stable isotopes of minerals to answer nutritional questions in humans has expanded considerably during the 1980s. Progress is primarily the result of the keen interest of nutritionists in this valuable tool, since there are a number of drawbacks to mineral isotope techniques that would otherwise result in abandonment of their use. The primary drawbacks are the high cost of isotopes, difficulty and cost of analysis, and lack of access to suitable instrumentation. However, stable isotopes provide a means of labeling minerals in humans with no exposure to radioactivity. In addition, multiple isotope experiments can be conducted. Thus, mineral absorption, utilization, and interactions an be studied for several minerals simultaneously. When our stable-isotope work began, we selected the potentially most accurate analytical method, because we planned to conduct experiments that required the measurement of low levels of isotopic enrichment and, thus, a high degree of analytical precision. Upon investigation, it was apparent that the most precise approach was magnetic sector, thermal ionization mass spectrometry (TIMS). It is capable of precision of within 0.1% relative standard deviation for isotope ratio measurements. Zinc, copper, and iron are poorly absorbed compared to vitamins. Their dietary requirements depend heavily upon the efficiency of absorption, since only the absorbed fraction of the mineral in the diet can be utilized by the body to perform essential functions. For example, endogenous losses of zinc have been estimated to be 2.2 mg/d (1). An individual would need to consume 22 mg of dietary zinc/d to replace endogenous losses if only 10% of dietary zinc were absorbed, but only 5.5 mg/d if 40% were absorbed. Endogenous iron losses have been estimated to be 1.5 mg/d in premenopausal women. If only 3% of the dietary iron were absorbed, as is the case for nonheme iron (2), 50 mg of dietary iron would be required to replace endogenous losses. If 25% of the dietary iron were absorbed, as for heme iron, only 6 mg/d would be required. In the examples above, the higher level of zinc or iron could not be obtained in a normal diet without supplements. However, most diets contain the lower amounts of zinc and iron. Labels are required to accurately determine mineral absorption, since the minerals from a specific meal or day's diet must be distinguished from minerals consumed on other days and from endogenous minerals. Radioisotopes have been used as labels for most studies conducted to date. Use of stable isotopes as labels is far more difficult, expensive, and analysis is slower than for radioisotopes. Their supply is sometimes limited and at times some specific isotopes may not be available at all. Stable isotopes must be used in quantities larger than true traBiological Trace Element Research
Vol. 12, 1987
Zn, Cu, and Fe Nutrition~StableIsotopes
249
cer doses. However, there is no exposure to radioactivity with stable isotopes, thus, one of the primary drawbacks of radioisotopes is eliminated when stable isotopes are used as labels. Other advantages of stable isotopes include the ability to use an unlimited number of the available isotopes simultaneously. Since stable isotopes do not decay, samples can be stored for extended periods of time. When no suitable radioisotope is available, stable isotopes are the only choice for a label.
METHOD USED FOR STABLE-ISOTOPE STUDIES
Experimental Design of Human Studies The studies described here were conducted in volunteers who were confined to a metabolic unit, either at the University of California, Berkeley, or at the Western Human Nutrition Research Center in San Francisco (3-10). Volunteers were supervised at all times during the studies. Dietary intake was constant and controlled throughout the studies. The type of diet used in an experiment depended upon the specific objectives of the experiment. Diets were either semipurified formula diets, blenderized food diets, single-me-u food diets, or 3-7 d rotating menus of food diets. Energy intake required to maintain body weight was calculated for each subject and it was adjusted only if an individual gained or lost weight. Dietary staff were present at all meals and all food served was consumed. Stable isotopes were fed to subjects, then monitored in fecal collections following isotope feedings. The isotopes were incorporated into formula diets or added to a component of food diets. Whenever possible, the isotopes replaced equal amounts of minerals in the diet, but in some cases they were in addition to the usual dietary minerals. Polyethylene glycol (PEG), a fecal marker that is not absorbed, was added to the diets with the isotopes to determine transit time and to assure that fecal collections were long enough for complete collection of all unabsorbed isotopes. Transit time was variable and, in most cases, from 6 to 12 d were required to eliminate all PEG from the gastrointestinal tract. The fecal marker was used because incomplete collections would result in overestimation of mineral absorption.
Preparation of Samples for Analysis by TIIVlS Following collection, fecal samples were homogenized, lyophilized, and crushed in a plastic bag. This procedure was tested and found to result in excellent sample homogeneity, which is essential to obtain reliable experimental results. Subsamples were then weighed, all organic material destroyed, and the remaining minerals dissolved in 6N HC1. Highpurity reagents were used throughout all procedures. Samples were then applied to anion exchange columns, which had been thoroughly cleaned Biological Trace Element Research
Vol. 12, 1987
250
Turnlund
to remove all impurities or traces of minerals. Copper, iron, and zinc were eluted from the columns separately with decreasing concentrations of HC1. Each mineral collected was purified separately with additional anion exchange columns. Throughout sample preparation procedures, extreme care was taken to avoid contamination with trace elements, which would result in changes in isotopic ratios. It was also necessary to assure the samples were free from all traces of organic material, which would make samples impossible to analyze by TIMS.
Determination of Isotopic Ratios by TIIVlS Following sample preparation, isotopic ratios were determined by magnetic sector TIMS. The total amounts of minerals in the samples were determined either by isotope dilution with a known amount of another stable isotope of the same mineral or by atomic absorption analysis of multiple replicates. Zinc and copper were determined using a silica gel ionization enhancement technique, which permitted analysis of smaller samples. A single direction, focusing instrument with an electron multiplier detector, MS6, which was built by Dr. Maynard Michel and located in Lawrence Berkeley Laboratory, was used for these analyses. Iron was determined using an automated TIMS with a Faraday collector. A new TIMS is currently being installed in our laboratory, which should result in faster, even more precise determination of mineral isotopes. This instrument (Finnigan MAT 261, Bremen, Germany) is fully computer controlled and automated, with a variable, multiple Faraday collector system, which will provide simultaneous collection of multiple isotopes of a mineral.
Other Methods For Stable-Isotope Analysis Several analytical approaches have been used in recent years for the determination of stable isotopes of minerals. These methods include neutron activation analysis (NAA) (11), gas chromatography/mass spectrometry (GCMS) (12,13), electron impact mass spectrometry (EIMS) (14), and magnetic sector, TIMS. Each has advantages and disadvantages, but the best analytical precision and accuracy an be achieved with TIMS (15). Janghorbani and coworkers used NAA for table isotope analysis in a number of experiments (11). However, when they compared the precision of NAA with TIMS in measuring the ratio of 7~176 in blood plasma (16), they demonstrated the precision of TIMS was far better. The coefficient of variance (CV) of the ratios was 0.13% for TIMS and 8.3% for NAA. Although NAA has advantages, including less sample preparation, disadvantages have resulted in declining use of the method. Disadvantages include inability to measure several important isotopes, inadequate precision for some work, and requirement for access to an operating nuclear reactor.
Biological Trace Element Research
Vol. 12. 1987
Zn, Cu, and Fe Nutrition~Stable Isotopes
251
Hachey and Klein used GCMS to measure stable isotopes of a number of minerals (12). They have since compared the precision and dynamic range of GCMS and TIMS and concluded that only TIMS is capable of sufficient precision to carry out kinetic studies of mineral metabolism (17). The TIMS has 10-1000 times the precision of GCMS, depending upon the minerals and natural ratios of the isotopes measured. This results in a huge increase in the range of isotopic enrichments that can be accurately measured. In the early stages of our work with stable isotopes, we conducted a study in which both stable and radioisotopes of zinc and iron were used (18). Stable isotopes were determined using both TIMS and NAA. Results of TIMS agreed best with data obtained using radioisotopes and the precision of TIMS was considerably better than NAA. We proceeded to develop methods for TIMS analysis of our samples and applied the methods to the experiments described in the next section. Two new approaches to mineral isotope ratio measurement have recently been reported. These methods, fast atom bombardment mass spectrometry (FAB) (19) and inductively coupled plasma-mass spectrometry (ICP-MS) (20), appear to have the potential to be very useful in human mineral nutrition research, even though they lack the precision of TIMS.
RESULTS OF STABLE-ISOTOPE EXPERIMENTS Zinc Absorption Zinc absorption from food diets was determined in young men fed two levels of dietary zinc (3). Once during the experiment 67Zn was substituted for an equal amount of dietary zinc in diets containing 16.5 mg Zn/d. Twice during the experiment, 70Zn was substituted for an equal amount of dietary zinc in diets containing 5.5 mg of Zn/d. The results are summarized in Table 1. Fractional zinc absorption was significantly lower from the 16.5 mg Zn diet than from the 5.5-mg Zn diet, but about 1 mg more zinc was absorbed from the diet that contained more zinc. Zinc balance was maintained with both diets. Zinc absorption was studied in young men fed formula diets (4) with or without added phytate or s-cellulose, a fiber source (Table 1). Zinc absorption was significantly lower when phytate was added to the diets, but absorption was not affected by the fiber. Zinc balance was negative with the phytate treatment. The results suggested that large amounts of phytate in diets could have a negative effect on zinc status. Zinc absorption, retention, and blood levels were recently studied when young women were fed diets deficient in vitamin B-6 (5). Zinc absorption from the vitamin-B-6-deficient diets was increased and more zinc was retained, as shown in Table 1. However, serum zinc levels de-
Biological Trace Element Research
Vol. 12, 1987
252
Turnlund
t.~,l ,.~ oo +I +I +I
+I +I +I
+I +I +I
+I
+I
+I
+I
+I
+I
It'-.. '~D C~
o
u~
t'-,l
b~
~.:
=5 V
Ca ca
[--. Ca
C~
9
-~
~. ,
9 o o'-~ ~ ,o- , . ,o ~
Zinc, Copper, and Iron Nutrition Studied with Enriched Stable Isotopes JUDITH R. TURNLUND USDA, Western Human Nutrition Research Center, 800 Buchanan Street, Berkeley, CA 94710
ABSTRACT Enriched stable isotopes were used in nutrition studies of normal, healthy adults to measure zinc, copper, and iron absorption. After obtaining baseline values for zinc, copper, and iron absorption from diets adequate in all nutrients, the effects of age, pregnancy, and several dietary variables were studied. Stable isotopes of zinc, copper, and iron were incorporated into diets. Complete fecal samples were collected and the unabsorbed isotopes remaining in the samples were measured by thermal ionization mass spectrometry, the most precise analytical method for the determination of stable mineral isotopes. Stable isotopes were also infused in five young men to evaluate the potential of studying mineral utilization and kinetics with stable isotopes. The results of these studies demonstrate that a number of factors can affect mineral absorption, but the specific effects differ for different minerals. Isotopic enrichments could be measured in urine and blood, so kinetic studies of utilization of essential minerals are now feasible with enriched stable isotopes. Continued use of stable isotopes to determine mineral absorption, combined with stable-isotope studies of mineral utilization, balance data, and biochemical indicators of mineral status, should result in a better understanding of mineral requirements and metabolism under a variety of conditions. Index Entries: Stable isotopes; zinc; copper, iron; bioavailability; absorption; mass spectrometry; nutrition.
INTRODUCTION This paper reviews the approach used by the author and her collaborators to study zinc, copper, and iron absorption with enriched stable Biological Trace Element Research
247
Vol. 12, 1987
248
Turnlund
isotopes. Experimental design, critical features of experiments, and the analytical approach are summarized. Recent studies measuring isotopic enrichments in urine, blood, and excretion into the gastrointestinal tract are also included. Research using stable isotopes of minerals to answer nutritional questions in humans has expanded considerably during the 1980s. Progress is primarily the result of the keen interest of nutritionists in this valuable tool, since there are a number of drawbacks to mineral isotope techniques that would otherwise result in abandonment of their use. The primary drawbacks are the high cost of isotopes, difficulty and cost of analysis, and lack of access to suitable instrumentation. However, stable isotopes provide a means of labeling minerals in humans with no exposure to radioactivity. In addition, multiple isotope experiments can be conducted. Thus, mineral absorption, utilization, and interactions an be studied for several minerals simultaneously. When our stable-isotope work began, we selected the potentially most accurate analytical method, because we planned to conduct experiments that required the measurement of low levels of isotopic enrichment and, thus, a high degree of analytical precision. Upon investigation, it was apparent that the most precise approach was magnetic sector, thermal ionization mass spectrometry (TIMS). It is capable of precision of within 0.1% relative standard deviation for isotope ratio measurements. Zinc, copper, and iron are poorly absorbed compared to vitamins. Their dietary requirements depend heavily upon the efficiency of absorption, since only the absorbed fraction of the mineral in the diet can be utilized by the body to perform essential functions. For example, endogenous losses of zinc have been estimated to be 2.2 mg/d (1). An individual would need to consume 22 mg of dietary zinc/d to replace endogenous losses if only 10% of dietary zinc were absorbed, but only 5.5 mg/d if 40% were absorbed. Endogenous iron losses have been estimated to be 1.5 mg/d in premenopausal women. If only 3% of the dietary iron were absorbed, as is the case for nonheme iron (2), 50 mg of dietary iron would be required to replace endogenous losses. If 25% of the dietary iron were absorbed, as for heme iron, only 6 mg/d would be required. In the examples above, the higher level of zinc or iron could not be obtained in a normal diet without supplements. However, most diets contain the lower amounts of zinc and iron. Labels are required to accurately determine mineral absorption, since the minerals from a specific meal or day's diet must be distinguished from minerals consumed on other days and from endogenous minerals. Radioisotopes have been used as labels for most studies conducted to date. Use of stable isotopes as labels is far more difficult, expensive, and analysis is slower than for radioisotopes. Their supply is sometimes limited and at times some specific isotopes may not be available at all. Stable isotopes must be used in quantities larger than true traBiological Trace Element Research
Vol. 12, 1987
Zn, Cu, and Fe Nutrition~StableIsotopes
249
cer doses. However, there is no exposure to radioactivity with stable isotopes, thus, one of the primary drawbacks of radioisotopes is eliminated when stable isotopes are used as labels. Other advantages of stable isotopes include the ability to use an unlimited number of the available isotopes simultaneously. Since stable isotopes do not decay, samples can be stored for extended periods of time. When no suitable radioisotope is available, stable isotopes are the only choice for a label.
METHOD USED FOR STABLE-ISOTOPE STUDIES
Experimental Design of Human Studies The studies described here were conducted in volunteers who were confined to a metabolic unit, either at the University of California, Berkeley, or at the Western Human Nutrition Research Center in San Francisco (3-10). Volunteers were supervised at all times during the studies. Dietary intake was constant and controlled throughout the studies. The type of diet used in an experiment depended upon the specific objectives of the experiment. Diets were either semipurified formula diets, blenderized food diets, single-me-u food diets, or 3-7 d rotating menus of food diets. Energy intake required to maintain body weight was calculated for each subject and it was adjusted only if an individual gained or lost weight. Dietary staff were present at all meals and all food served was consumed. Stable isotopes were fed to subjects, then monitored in fecal collections following isotope feedings. The isotopes were incorporated into formula diets or added to a component of food diets. Whenever possible, the isotopes replaced equal amounts of minerals in the diet, but in some cases they were in addition to the usual dietary minerals. Polyethylene glycol (PEG), a fecal marker that is not absorbed, was added to the diets with the isotopes to determine transit time and to assure that fecal collections were long enough for complete collection of all unabsorbed isotopes. Transit time was variable and, in most cases, from 6 to 12 d were required to eliminate all PEG from the gastrointestinal tract. The fecal marker was used because incomplete collections would result in overestimation of mineral absorption.
Preparation of Samples for Analysis by TIIVlS Following collection, fecal samples were homogenized, lyophilized, and crushed in a plastic bag. This procedure was tested and found to result in excellent sample homogeneity, which is essential to obtain reliable experimental results. Subsamples were then weighed, all organic material destroyed, and the remaining minerals dissolved in 6N HC1. Highpurity reagents were used throughout all procedures. Samples were then applied to anion exchange columns, which had been thoroughly cleaned Biological Trace Element Research
Vol. 12, 1987
250
Turnlund
to remove all impurities or traces of minerals. Copper, iron, and zinc were eluted from the columns separately with decreasing concentrations of HC1. Each mineral collected was purified separately with additional anion exchange columns. Throughout sample preparation procedures, extreme care was taken to avoid contamination with trace elements, which would result in changes in isotopic ratios. It was also necessary to assure the samples were free from all traces of organic material, which would make samples impossible to analyze by TIMS.
Determination of Isotopic Ratios by TIIVlS Following sample preparation, isotopic ratios were determined by magnetic sector TIMS. The total amounts of minerals in the samples were determined either by isotope dilution with a known amount of another stable isotope of the same mineral or by atomic absorption analysis of multiple replicates. Zinc and copper were determined using a silica gel ionization enhancement technique, which permitted analysis of smaller samples. A single direction, focusing instrument with an electron multiplier detector, MS6, which was built by Dr. Maynard Michel and located in Lawrence Berkeley Laboratory, was used for these analyses. Iron was determined using an automated TIMS with a Faraday collector. A new TIMS is currently being installed in our laboratory, which should result in faster, even more precise determination of mineral isotopes. This instrument (Finnigan MAT 261, Bremen, Germany) is fully computer controlled and automated, with a variable, multiple Faraday collector system, which will provide simultaneous collection of multiple isotopes of a mineral.
Other Methods For Stable-Isotope Analysis Several analytical approaches have been used in recent years for the determination of stable isotopes of minerals. These methods include neutron activation analysis (NAA) (11), gas chromatography/mass spectrometry (GCMS) (12,13), electron impact mass spectrometry (EIMS) (14), and magnetic sector, TIMS. Each has advantages and disadvantages, but the best analytical precision and accuracy an be achieved with TIMS (15). Janghorbani and coworkers used NAA for table isotope analysis in a number of experiments (11). However, when they compared the precision of NAA with TIMS in measuring the ratio of 7~176 in blood plasma (16), they demonstrated the precision of TIMS was far better. The coefficient of variance (CV) of the ratios was 0.13% for TIMS and 8.3% for NAA. Although NAA has advantages, including less sample preparation, disadvantages have resulted in declining use of the method. Disadvantages include inability to measure several important isotopes, inadequate precision for some work, and requirement for access to an operating nuclear reactor.
Biological Trace Element Research
Vol. 12. 1987
Zn, Cu, and Fe Nutrition~Stable Isotopes
251
Hachey and Klein used GCMS to measure stable isotopes of a number of minerals (12). They have since compared the precision and dynamic range of GCMS and TIMS and concluded that only TIMS is capable of sufficient precision to carry out kinetic studies of mineral metabolism (17). The TIMS has 10-1000 times the precision of GCMS, depending upon the minerals and natural ratios of the isotopes measured. This results in a huge increase in the range of isotopic enrichments that can be accurately measured. In the early stages of our work with stable isotopes, we conducted a study in which both stable and radioisotopes of zinc and iron were used (18). Stable isotopes were determined using both TIMS and NAA. Results of TIMS agreed best with data obtained using radioisotopes and the precision of TIMS was considerably better than NAA. We proceeded to develop methods for TIMS analysis of our samples and applied the methods to the experiments described in the next section. Two new approaches to mineral isotope ratio measurement have recently been reported. These methods, fast atom bombardment mass spectrometry (FAB) (19) and inductively coupled plasma-mass spectrometry (ICP-MS) (20), appear to have the potential to be very useful in human mineral nutrition research, even though they lack the precision of TIMS.
RESULTS OF STABLE-ISOTOPE EXPERIMENTS Zinc Absorption Zinc absorption from food diets was determined in young men fed two levels of dietary zinc (3). Once during the experiment 67Zn was substituted for an equal amount of dietary zinc in diets containing 16.5 mg Zn/d. Twice during the experiment, 70Zn was substituted for an equal amount of dietary zinc in diets containing 5.5 mg of Zn/d. The results are summarized in Table 1. Fractional zinc absorption was significantly lower from the 16.5 mg Zn diet than from the 5.5-mg Zn diet, but about 1 mg more zinc was absorbed from the diet that contained more zinc. Zinc balance was maintained with both diets. Zinc absorption was studied in young men fed formula diets (4) with or without added phytate or s-cellulose, a fiber source (Table 1). Zinc absorption was significantly lower when phytate was added to the diets, but absorption was not affected by the fiber. Zinc balance was negative with the phytate treatment. The results suggested that large amounts of phytate in diets could have a negative effect on zinc status. Zinc absorption, retention, and blood levels were recently studied when young women were fed diets deficient in vitamin B-6 (5). Zinc absorption from the vitamin-B-6-deficient diets was increased and more zinc was retained, as shown in Table 1. However, serum zinc levels de-
Biological Trace Element Research
Vol. 12, 1987
252
Turnlund
t.~,l ,.~ oo +I +I +I
+I +I +I
+I +I +I
+I
+I
+I
+I
+I
+I
It'-.. '~D C~
o
u~
t'-,l
b~
~.:
=5 V
Ca ca
[--. Ca
C~
9
-~
~. ,
9 o o'-~ ~ ,o- , . ,o ~
