J Appl Physiol 119: 55–60, 2015. First published May 14, 2015; doi:10.1152/japplphysiol.00894.2014.
Seasonal variation in natural abundance of 2H and rural Nigeria
18
O in urine samples from
Justin E. Harbison,1 Lara R. Dugas,1 William Brieger,2 Bamidele O. Tayo,1 Tunrayo Alabi,3 Dale A. Schoeller,4 and Amy Luke1 1
Division of Epidemiology, Department of Public Health Sciences, Loyola University School of Medicine, Maywood Illinois; Department of International Health, Bloomberg School of Public Health, John Hopkins University, Baltimore Maryland; 3 Geospatial Laboratory, Research for Development, International Institute of Tropical Agriculture, Ibadan, Nigeria; 4 Department of Nutritional Sciences, University of Wisconsin, Madison, Wisconsin 2
Submitted 10 October 2014; accepted in final form 7 May 2015
Harbison JE, Dugas LR, Brieger W, Tayo BO, Alabi T, Schoeller DA, Luke A. Seasonal variation in natural abundance of 2H and 18O in urine samples from rural Nigeria. J Appl Physiol 119: 55–60, 2015. First published May 14, 2015; doi:10.1152/japplphysiol.00894.2014.—The doubly labeled water (DLW) method is used to measure free-living energy expenditure in humans. Inherent to this technique is the assumption that natural abundances of stable isotopes 2H and 18O in body water remain constant over the course of the measurement period and after elimination of the loading dose of DLW will return to the same predose level. To determine variability in the natural abundances of 2 H and 18O in humans living in a region with seasonal shifts in rain patterns and sources of drinking water, over the course of 12 mo we collected weekly urine samples from four individuals living in southwest Nigeria as well as samples of their drinking water. From ongoing regional studies of hypertension, obesity, and energy expenditure, we estimated average water turnover rate, urine volumes, and sodium and potassium excretion. Results suggest that 2H and 18O in urine, mean concentrations of urinary sodium and potassium, urine volume, and total body turnover differed significantly from dry to rainy season. Additionally, seasonal weather variables (mean monthly maximum temperatures, total monthly rainfall, and minimum relative humidity) were all significantly associated with natural abundances in urine. No seasonal difference was observed in drinking water samples. Findings suggest that natural abundances in urine may not remain constant as assumed, and studies incorporating DLW measurements across the transition of seasons should interpret results with caution unless appropriate doses of the tracers are used. doubly labeled water; seasonal variation; Nigeria THE DOUBLY LABELED WATER (DLW) method, developed by Lifson in the late 1940s, has been utilized for over 60 years to measure free-living total energy expenditure (TEE) in animals, with early studies focusing on small mammals, pigeons, reptiles, and locusts (15, 17, 21, 22). Several decades later, it was adapted for use in humans by Schoeller and colleagues (7, 18, 26, 29 –32), providing objective measurement of energy expenditure and energy requirements. The basic principle behind the method is that the stable isotopes of oxygen in body water are in isotopic equilibrium with the oxygen in respiratory CO2. Thus, after a loading dose of DLW, i.e., enriched 2H218O, there is differential elimination of the isotopes. Deuterium is eliminated solely through water, e.g., urine, transcutaneous water, and respiratory water, whereas 18O is eliminated both via water and via respiratory CO2. Thus the differential elimination of
Address for reprint requests and other correspondence: J. Harbison, Dept. of Public Health Sciences, Loyola Univ. Chicago, 2160 South First Ave., Maywood, IL 60153 (e-mail: [email protected]). http://www.jappl.org
18
O is a measure of CO2 flux, thereby allowing for the calculation of TEE (30). One of the key assumptions of the DLW method is that the natural abundances, i.e., baseline concentrations, of the stable isotopes 2H and 18O are constant in an animal’s body throughout the measurement period, i.e., typically 1–2 wk for human studies, and that body water will return to the same predose abundances after the elimination of the loading dose (22). In reality, natural abundances of the isotopes in a body vary with the source and amount of water and food consumed, as well as with changes in the proportions of oxygen and hydrogen lost through CO2 production and the ratio of fractionated and nonfractionated water effluxes. Increases or decreases in isotope abundances can occur when water intake varies or when there is a change in water source. Changes in natural abundances may introduce error into the calculation of TEE (12). Of concern for human studies is when the natural abundance of one or both of the isotopes changes during the course of a measurement period. A stark example is that of measuring energy requirements during space flight. It is well established that astronauts’ drinking water has unusual isotopic abundances, and those changes in isotopic baseline must be made before or during the calculation of TEE (14, 33). There is also the potential for changing isotopic baseline as a result of dramatic seasonal effects, resulting in sudden changes in sources of drinking water, such as those that can occur in regions experiencing clearly demarcated rainy and dry seasons. We sought to determine whether the season of DLW measurement in the tropics, i.e., rainy or dry, affects the natural abundances of deuterium and 18O and whether or not changes in water source could explain any identified effect. Using data collected over the course of two research studies conducted in rural southwestern Nigeria, we documented natural abundance of the isotopes in urine over the course of a year, average water turnover rates (8, 19), and average urine volumes by season (34). METHODS
In conjunction with a concurrent study of TEE and weight change among rural Nigerians (8), weekly drinking water and urine samples were collected over the course of a year by four volunteers residing in the same community. The TEE and weight change study also allowed the calculation of water turnover rates by season. Urine volume and electrolyte concentrations were obtained from a study on sodium and blood pressure conducted in the same community (34). The studies were approved by the Institutional Review Board of Loyola University Chicago, Stritch School of Medicine, and the Ethics Committee of
8750-7587/15 Copyright © 2015 the American Physiological Society
55
56
Seasonal Variation of Deuterium and
the University College Hospital, University of Ibadan (Nigeria); written, informed consent was obtained from all participants. The studies were conducted in a village in rural southwest Nigeria, Igbo-Ora, located at latitude 7°N, where Loyola University Chicago Department of Public Health Sciences maintained a free-standing research site from the mid-1990s to 2009 (20). Drinking water for residents of the village is obtained from three primary sources: wells, boreholes, and collected rainwater. The source of drinking water tends to vary seasonally, with wells and collected rainwater being predominant during the rainy season and boreholes used more during the dry season. Nigeria has a tropical climate marked by consistently high temperatures (34/19°C, max/min) and wet and dry seasons associated with the Intertropical Convergence Zone. Whereas much of the country will have one rainy and dry season annually, in the southwestern part of the country (including the study area), a short decline in rainfall and intensity can occur in late July and August because of the Intertropical Convergence Zone (1, 23). The result is a long rainy season approximately from March to early July, a short dry period (or “little dry season”) in late July and August, a shorter rainy period in September and October, and a long dry season from November to February. For the year of data collection, there was no observable little dry season. Four participants from the two studies collected weekly samples of both their drinking water and urine between October 1, 2001 and September 16, 2002; samples were not collected during the month of December 2001 and half of January 2002, as the research staff was on holiday, resulting in a total of 40 wk of sample collection. Participants also recorded the source of the drinking water sample weekly, i.e., well, borehole, or collected rainwater. The first urine void on Monday mornings was collected along with a sample of the current source of drinking water. Duplicate 5-ml aliquots of each sample were prepared and stored in cryovials with Teflon O-rings. Samples were frozen at ⫺4°C and stored at the University College Hospital in Ibadan until being shipped to the University of Wisconsin, Madison for analysis in October 2002. Drinking water and urine samples from every second week were analyzed for abundance of deuterium and 18O at the Stable Isotope Core Laboratory in the Department of Nutritional Sciences, University of Wisconsin, Madison. Both urine and water samples were treated with 200 mg carbon black and filtered using 0.45-m sterile filters. Isotopes were analyzed using isotope ratio mass spectrometry as previously published (12). Daily temperatures and rainfall estimates during the same time period were recorded by two separate observers in the villages. These data were verified, and monthly averages of rainfall, maximum temperature, minimum relative humid-
18
Oxygen
•
Harbison JE et al.
ity, as well as the dates for the initiation and cessation of the annual rains were obtained from the Statistics Section of the International Institute of Tropical Agriculture, Ibadan, Nigeria, located ⬃60 km from Igbo-Ora. The average rate of water turnover by season was estimated using data from the parent project in which TEE was measured using DLW in 149 women (8). Total water turnover was calculated (kd ⫻ Nd/Fc) as the product of deuterium elimination rate (kd) and the deuterium dilution space (Nd), and corrected for isotope fractionation (Fc). The average value of Fc (9) was 0.98. Volume of urine produced and total urinary sodium (Na⫹) and potassium (K⫹) excretion were obtained during a separate project on hypertension and electrolyte excretion (34) in 756 adults from the same village. Mean volumes and electrolyte excretion of three 24-h urine collections were used in the present analyses. Data were analyzed using STATA (ver 12.0; College Station, TX). Average deuterium and 18O abundances for both urine and drinking water samples were calculated by season and expressed relative to standard mean ocean water (6). Covariation of the isotopic abundances over the course of the year was defined as having an intraindividual Pearson correlation coefficient ⱖ0.75 (12). Student’s t-tests were used to compare isotopic abundances, water turnover rates, urine volumes, and electrolyte concentrations between the rainy and dry seasons. RESULTS
The four participants (3 men and 1 woman), were 57.5 ⫾ 6.5 yr old and all resided in the village for the entire study period. For each participant, 21 urine and 21 drinking samples were analyzed. The natural abundances of deuterium and 18O of these urine samples are presented in Fig. 1. Although it appears that there are two outliers (one for subject 1 and the other for subject 4), there was no known physiological or environmental explanation, and they were included in all analyses. On the basis of daily rainfall patterns as recorded by local volunteers, March 12, 2002 was designated as the start of the rainy season; thus samples collected between October 1, 2001 and March 11, 2002 were determined to be from the dry season, and those from March 12, 2002 to the end of the study were from the rainy season. There was a marked difference in rainfall pattern indicating the occurrence of one short dry season (Fig. 2); therefore, a single dry and single rainy season were defined for analyses (October 1 to March 11 and March 12 to September
Subject 2 Subject 4
0 -10 -20 -30
Fig. 1. Deuterium and 18O abundances in urine from 4 adults in rural southwest Nigeria sampled between October 1, 2001 and September 16, 2002. The slope coefficients ⫾ SE for participants 1 to 4 were 6.00 ⫾ 0.29, 3.84 ⫾ 0.22, 5.52 ⫾ 0.26, and 5.35 ⫾ 0.50, respectively. per mil, parts per thousand.
Deuterium (per mil)
10
Subject 1 Subject 3
-6
-4
-2 18-Oxygen (per mil)
J Appl Physiol • doi:10.1152/japplphysiol.00894.2014 • www.jappl.org
0
Seasonal Variation of Deuterium and
A4
Temperature
Rainfall
Oxygen-18
45
Deuterium
40
3
35
2
25 0 20
cm and °C
Del SMOW
30 1
-1 15 -2
10
-3
5
-4
0 Oct
Nov
Dec
2001
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
2002
Oxygen-18
Deuterium
80
Minimum humidity
3
70
2
60
1
50
0
40
-1
30
-2
20
-3
10
-4
percent humidity
Del SMOW
B4
Jan
0 Oct
Nov
Dec
2001
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
2002
C4
Oxygen-18
Deuterium
200
Evaporation
180
3
160 2
120 100
0
mm per month
Del SMOW
140 1
80
-1
60 -2 40 -3
20 0
-4 Oct 2001
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
2002
Fig. 2. Mean (⫾ SE) monthly natural abundance of deuterium and 18O in urine from 4 adults in rural southwest Nigeria including median monthly temperature and total monthly rainfall (A), average monthly relative humidity (B), and total monthly evaporation (C) sampled between October 1, 2001 and September 16, 2002. Del SMOW, ␦ standard mean ocean water.
18
Oxygen
•
Harbison JE et al.
57
30, respectively). The average monthly maximum temperature was 33.7°C and 29.8°C during the dry and rainy seasons, respectively, while mean rainfall was 1.65 and 20.3 cm/mo, and minimum relative humidity was 43.8% and 64.6%. The mean abundances for both deuterium and 18O for the urine and two drinking water sources are presented by season in Table 1. Both isotopes were significantly more enriched in the urine samples during the dry compared with the rainy season (P ⬍ 0.001). The greatest variance in the urinary isotopes appeared to occur from January to May when the dry season was ending and the rains began to increase (Fig. 2). Mean monthly abundances of both deuterium and 18O in urine were also positively associated with mean monthly maximum temperatures (r ⫽ 0.55, P ⫽ 0.08 and r ⫽ 0.86, P ⬍ 0.001 respectively) and significantly inversely associated with rainfall (⫺0.74 and ⫺0.79; P ⬍ 0.001) and minimum relative humidity (⫺0.87 and ⫺0.91; P ⬍ 0.001). The majority of drinking water samples were taken from local wells during both the dry and rainy seasons (n ⫽ 57), followed by borehole samples (n ⫽ 21), and there were no significant differences in isotope abundances for either the well water or borehole water samples by season. Rain water samples (n ⫽ 6) were all collected during the rainy season. The mean rate of total body water turnover was significantly higher (P ⬍ 0.001) in the dry season (0.13 ⫾ 0.02 pools/day) than in the rainy season (0.11 ⫾ 0.02 pools/day; P ⬍ 0.001), as was absolute volume of daily water turnover, 3.8 ⫾ 0.8 vs. 3.5 ⫾ 0.7 l (P ⬍ 0.02). Average total urine volumes were significantly lower in the dry (1.34 ⫾ 0.50 l/day) compared with the rainy season (1.53 ⫾ 0.50 l/day; P ⬍ 0.001). Whereas urinary sodium concentrations were not different between dry and rainy seasons (117 ⫾ 4 vs. 121 ⫾ 2 meq/day, respectively), urinary potassium was significantly higher in the dry season, 49.9 ⫾ 1.8 vs. 43.1 ⫾ 0.9 meq/day (P ⬍ 0.001). Both the higher urinary potassium concentrations and a lower urine volume/water turnover rate indicate more concentrated urine being observed during the dry season. In Fig. 3, the difference between the abundance of the isotopes in urine and drinking water is plotted against the rate of water turnover. Although the rate of water turnover only explains 25% of the variation in deuterium abundance and 7% of the 18O abundance, these data support the concept that baseline urinary isotope abundances reflect water turnover. DISCUSSION
The natural abundances of deuterium and 18O in humans, as measured in urine, do vary significantly between dry and rainy seasons in this tropical region, with both being more enriched during the dry season. On the basis of our findings, the observed urinary differences in abundance between the seasons are not due to changes in the isotopic abundance of drinking water but rather likely due to changes in water turnover that are potentially influenced by environmental factors, including temperature and relative humidity (28). The DLW method has been widely used to measure TEE in multiple populations around the world; it has also been used to establish energy intake requirements for individuals of all ages and phase of life. Because the technique is based on the assumption that natural abundances within the body remain constant throughout the measurement period, knowing if and
J Appl Physiol • doi:10.1152/japplphysiol.00894.2014 • www.jappl.org
58
Seasonal Variation of Deuterium and
Table 1. Mean abundances for deuterium and
18
Oxygen
•
Harbison JE et al.
18
O for urine and drinking water sources by season
2
18
Hydrogen
Urine Well water Borehole water Rain water
Oxygen
Dry
Rainy
P Value
Dry
Rainy
P Value
0.47 (2.24) ⫺11.79 (4.00) ⫺12.71 (1.87) —
⫺2.19 (2.29) ⫺12.61 (4.83) ⫺11.56 (3.89) ⫺6.33 (6.92)
⬍0.001 0.49 0.39 —
⫺0.56 (0.46) ⫺2.96 (0.35) ⫺3.38 (0.50) —
⫺0.93 (0.49) ⫺2.90 (0.53) ⫺3.22 (0.52) ⫺2.18 (0.83)
⬍0.001 0.65 0.48 —
Applicable values are presented as means (SE). Number of samples analyzed (by season): urine (dry) n ⫽ 35, urine (rainy) n ⫽ 49; well water (dry) n ⫽ 23, well water (rainy) n ⫽ 34; borehole water (dry) n ⫽ 11, borehole water (rainy) n ⫽ 10; rain water (rainy) n ⫽ 6. Rain water was not used as drinking water source during dry season.
possibly when baseline abundances vary could better inform the use of this method in the tropics. Rural southwest Nigeria is a compelling area to evaluate natural abundances of stable isotopes because of its relatively warm year-round temperatures and distinct seasonal changes in rainfall and sources of drinking water. As hypothesized, measured natural abundances of deuterium and 18O did indeed differ significantly by season, however, only in urine and not well and borehole drinking water sources. It is unknown whether the addition of rainwater as a drinking water source during the rainy season may have played a role in the observed seasonal difference urine isotopes, as so few samples of rainwater were collected in this study. In studies of water movement and budget, the isotopic variance in rainwater is used to identify water from a precipitation event as opposed to ground water (11, 16). This suggests that, when used as a primary water source during only one season, rainwater might be associated with a shift in a body’s natural abundances of deuterium and 18O. During the study period, it appeared as if the greatest variance in urinary isotopes from the four participants occurred during the transition from dry to rainy seasons when rains began to increase in intensity. The weather-related variables of mean monthly maximum temperatures, rainfall, and minimum relative humidity were all significantly associated with natural abundance in urine and provide additional evidence that,
Fig. 3. Relationship between water turnover, presented as the natural log (Ln) of daily deuterium flux and the abundance of deuterium and 18O in urine relative to drinking water. permil, parts per thousand. D/H, deuterium/hydrogen.
within the climatic region of the study, seasonal weather patterns over the year can affect isotope abundance within the body. Physiologically, weather can influence daily obligatory water loss by increasing transcutaneous evaporation during extended periods of high temperatures and low humidity and as a result may impact deuterium and 18O abundances in the body. This would also explain the seasonal differences in turnover rate, urine volume, and K⫹ concentrations observed in the study. In studies of the isotopic compositions of water, the most significant fractionations are associated with phase changes (i.e., condensation and evaporation) (16), and it would thus follow that such variations could similarly occur as those phase changes manifest from the human body. Obviously a larger number of study participants could help clarify the relationship between our measured environmental variables and natural abundances, but the observed seasonal variances of our measured variables do appear to be associated with changes in water turnover rates (28). If mean monthly maximum temperatures, rainfall, and minimum relative humidity are indeed associated with fluctuations of natural abundances within the body, then other regions of the world, and particularly those with distinct seasons (e.g., higher-latitude climates with cold winters), may experience a similar or more pronounced isotopic variation. For example, increasing latitudes in the Northern Hemisphere have increasing ranges of monthly temperatures, which, as our results indicate, could potentially affect or add variance to natural abundances within participants of studies performed in those areas. This may particularly be the case during times of seasonal transition (months of September to October, March to May, etc.) and in regions farther away from larger bodies of water (e.g., inland vs. coastal areas) when temperatures and precipitation may fluctuate more. Likewise a subject’s daily exposure to environmental temperatures, humidity, and rainfall may also influence natural abundances. The four study participants lived in and worked in a rural area of Nigeria where air conditioning was uncommon, and these subjects were likely more directly exposed to environmental variables associated with weather than individuals residing and working in more urbanized climate-controlled environments. Such differences in local environmental exposures (occupational, rural vs. urban, etc.) may also introduce more variance into natural abundance measurements. However, potential error resulting from natural fluctuations in abundance can be reduced when deuterium and 18O are dosed in a manner such that enrichments have a ratio similar to the ratio of the anticipated abundance changes in urine. Natural changes in abundances create similar errors in the elimination
J Appl Physiol • doi:10.1152/japplphysiol.00894.2014 • www.jappl.org
Seasonal Variation of Deuterium and
rates of both isotopes, and thus these errors cancel when the difference of the two elimination rates is calculated. Anyone performing studies in which samples are collected weeks after the predose sample should be careful not to underestimate the scale and effect of natural isotopic variations over time, which may be significant, as the results of this study suggest. Ultimately such an underestimation may lead to error as a great as 30% (27). The doses of 18O- and 2H-enriched water should yield a postdose enrichment in parts per thousand units of 8 to 1 of 18O to 2H enrichment, respectively (4), although there is very little effect for doses in the ratio of between 6 and 12 to 1 (12). In our laboratory, we aim for a postdose 18O enrichment of 90 parts. With the use of 10 atom percent (AP) 18O water and 99.9 AP deuterium, respectively, this corresponds to a dose weight of 1.8 g of 10 AP 18O water per kilogram of total body water and 0.13 g of 99.9 AP 2H2O per kilogram total body water. Although it appears from this study that natural abundances of 2H and 18O in urine may change over the scope of the year, how this information will affect the results of other studies requires further inquiry. In general, energy expenditure studies will utilize the DLW method for a much smaller 1– 4-wk measurement period (35), which may or may not be too short a time period for significant isotopic variance to occur. Often, however, the exact dates of DLW measurements are not reported, so critical consideration of any potential effect of seasonality becomes difficult. Such information would appear to be particularly relevant for TEE studies in areas such as parts of Cameroon (2, 3), Brazil (25), India (13, 5), Peru (10), and Kenya (24) that experience similar wet and dry seasonality as our study site. Overall, our results suggest that, at least in southwest Nigeria, the assumption that natural abundances will return to predose measurements during DLW measurements may not be correct, and the accuracy of this method may depend on the timing of measurements, particularly during the transition between dry and rainy seasons. Results of energy expenditure studies that utilize DLW measurements for human TEE calculations during a period of seasonal transition or over a longer period of time should be interpreted with caution. The potential error, however, can be minimized by dosing the two stable isotopes in an appropriate ratio such that errors in the 2H and 18O elimination rates cancel when the difference between the rates is calculated. ACKNOWLEDGMENTS We thank Amanda Romsa for the mass spectrometric analysis of the samples. GRANTS This research was supported in part by grants from the National Institutes of Health (DK56780 and DK30031). DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the authors. AUTHOR CONTRIBUTIONS Author contributions: J.E.H., L.D., and B.T. analyzed data; J.E.H. and D.A.S. interpreted results of experiments; J.E.H. and L.D. prepared figures; J.E.H. drafted manuscript; J.E.H., L.D., W.B., B.T., T.A., D.A.S., and A.L. edited and revised manuscript; J.E.H., L.D., W.B., B.T., T.A., D.A.S., and A.L. approved final version of manuscript; L.D., W.B., B.T., T.A., D.A.S., and A.L. conception and design of research; L.D., W.B., B.T., T.A., and A.L. performed experiments.
18
Oxygen
•
Harbison JE et al.
59
REFERENCES 1. Adejuwon JO, Odekunle TO. Variability and the severity of the “Little Dry Season” in southwestern Nigeria. J Climate 19: 483–493, 2006. 2. Assah FK, Ekelund U, Brage S, Mbanya JC, Wareham NJ. Free-living physical activity energy expenditure is strongly related to glucose intolerance in Cameroonian adults independently of obesity. Diabetes Care 32: 367–369, 2009. 3. Assah FK, Ekelund U, Brage S, Corder K, Wright A, Mbanya JC, Wareham NJ. Predicting physical activity energy expenditure using accelerometry in adults from sub-Sahara Africa. Obesity (Silver Spring) 17: 1588 –1595, 2009. 4. Berman ES, Melanson EL, Swibas T, Snaith SP, Speakman JR. Interand intraindividual correlations of background abundances of 2H, 18O and 17 O in human urine and implications for DLW measurements. Eur J Clin Nutr. In press. 5. Corder K, Brage S, Wright A, Ramachandran A, Snehalatha C, Yamuna A, Wareham NJ, Ekelund U. Physical activity energy expenditure of adolescents in India. Obesity (Silver Spring) 18: 2212–2219, 2010. 6. Craig H. Isotope variations in meteoric waters. Science 133: 1702–1703, 1961. 7. Dugas LR, Harders R, Merrill S, Ebersole K, Shoham DA, Rush EC, Assah FK, Forrester T, Durazo-Arvizu RA, Luke A. Energy expenditure in adults living in developing compared with industrialized countries: a meta-analysis of doubly labeled water studies. Am J Clin Nutr 93: 427–441, 2011. 8. Ebersole KE, Dugas LR, Durazo-Arvizut RA, Adeyemo AA, Tayo BO, Omotade OO, Brieger WR, Schoeller DA, Cooper RS, Luke AH. Energy expenditure and adiposity in Nigerian and African-American women. Obesity (Silver Spring) 16: 2148 –2154, 2008. 9. Fjeld CR, Brown KH, Schoeller DA. Validation of the deuterium oxide method for measuring average daily milk intake in infants. Am J Clin Nutr 48: 671–679, 1988. 10. Fjeld CR, Schoeller DA, Brown KH. A new model for predicting energy requirements of children during catch-up growth developed using doubly labeled water. Pediatr Res 25: 503–508, 1989. 11. Granger SJ, Bol R, Meier-Augenstein W, Leng MJ, Kemp HF, Heaton TH, White SM. The hydrological response of heavy clay grassland soils to rainfall in south-west England using ␦2H. Rapid Commun Mass Spectrom 24: 475–482, 2010. 12. Horvitz MA, Schoeller DA. Natural abundance deuterium and 18-oxygen effects on the precision of the doubly labeled water method. Am J Physiol Endocrinol Metab 280: E965–E972, 2001. 13. Krishnaveni GV, Veena SR, Kuriyan R, Kishore RP, Wills AK, Nalinakshi M, Kehoe S, Fall CH, Kurpad AV. Relationship between physical activity measured using accelerometers and energy expenditure measured using doubly labelled water in Indian children. Eur J Clin Nutr 63: 1313–1319, 2009. 14. Lane HW, Gretebeck RJ, Schoeller DA, Davis-Street J, Socki RA, Gibson EK. Comparison of ground-based and space flight energy expenditure and water turnover in middle-aged healthy male US astronauts. Am J Clin Nutr 65: 4 –12, 1997. 15. Lee JS, Lifson N. Measurement of total energy and material balance in rats by means of doubly labeled water. Am J Physiol 199: 238 –242, 1960. 16. Lewicka-Szczebak D, Je˛drysek MO. Tracing and quantifying lake water and groundwater fluxes in the area under mining dewatering pressure using coupled O and H stable isotope approach. Isotopes Environ Health Stud 49: 9 –28, 2013. 17. Lifson N, Gordon GB, Visccher MB, Nier AO. The fate of utilized molecular oxygen and the source of the respiratory carbon dioxide, studied with the aid of heavy oxygen. J Biol Chem 180: 804, 1949. 18. Luke A, Adeyemo A, Kramer H, Forrester T, Cooper RS. Association between blood pressure and resting energy expenditure independent of body size. Hypertension 43: 555–560, 2004. 19. Luke A, Dugas LR, Ebersole K, Durazo-Arvizu RA, Cao G, Schoeller DA, Adeyemo A, Brieger WR, Cooper RS. Energy expenditure does not predict weight change in either Nigerian or African American women. Am J Clin Nutr 89: 169 –176, 2009. 20. Luke A, Durazo-Arvizu R, Rotimi C, Prewitt TE, Forrester T, Wilks R, Ogunbiyi OJ, Schoeller DA, McGee D, Cooper RS. Relation between body mass index and body fat in black population samples from Nigeria, Jamaica, and the United States. Am J Epidemiol 145: 620 –628, 1997.
J Appl Physiol • doi:10.1152/japplphysiol.00894.2014 • www.jappl.org
60
Seasonal Variation of Deuterium and
21. McClintock R, Lifson N. Applicability of the D2O18 method output of obese mice the total carbon dioxide output. J Biol Chem 226: 153–156, 1957. 22. Nagy KA. CO2 production in animals: analysis of potential errors in the doubly labeled water method. Am J Physiol Regul Integr Comp Physiol 238: R466 –R473, 1980. 23. Odekunle TO. An assessment of the influence of the inter-tropical discontinuity on inter-annual rainfall characteristics in Nigeria. Geograph Res 48: 314 –326, 2010. 24. Ojiambo R, Gibson AR, Konstabel K, Lieberman DE, Speakman JR, Reilly JJ, Pitsiladis YP. Free-living physical activity and energy expenditure of rural children and adolescents in the Nandi region of Kenya. Ann Hum Biol 40: 318 –323, 2013. 25. Scagliusi FB, Ferriolli E, Pfrimer K, Laureano C, Cunha CS, Gualano B, Lourenço BH, Lancha AH Jr. Characteristics of women who frequently under report their energy intake: a doubly labelled water study. Eur J Clin Nutr 63: 1192–1199, 2009. 26. Schoeller DA. Energy expenditure from doubly labeled water: some fundamental considerations in humans. Am J Clin Nutr 38: 999–1005, 1983. 27. Schoeller DA. Measurement of energy expenditure in free-living humans by using doubly labeled water. J Nutr 118: 1278 –1289, 1988. 28. Schoeller DA, Leitch CA, Brown C. Doubly labeled water method: in vivo oxygen and hydrogen isotope fractionation. Am J Physiol 251: R1137–R1143, 1986.
18
Oxygen
•
Harbison JE et al.
29. Schoeller DA, Ravussin E, Schutz Y, Acheson KJ, Baertschi P, Jéquier E. Energy expenditure by doubly labeled water: validation in humans and proposed calculation. Am J Physiol 250: R823–R830, 1986. 30. Schoeller DA, van Santen E. Measurement of energy expenditure in humans by doubly labeled water method. J Appl Physiol 53: 955–959, 1982. 31. Schoeller DA, van Santen E, Peterson DW, Dietz W, Jaspan J, Klein PD. Total body water measurement in humans with 18O and 2H labeled water. Am J Clin Nutr 33: 2686 –2693, 1980. 32. Schoeller DA, Webb P. Five-day comparison of the doubly labeled water method with respiratory gas exchange. Am J Clin Nutr 40: 153–158, 1984. 33. Siniak IuE, Skuratov VM, Gaı˘dadymov VB, Ivanova SM, Pokrovskiı˘ BG, Grigor’ev AI. [Investigation into fractionating of hydrogen and oxygen stable isotopes aboard the international space station]. Aviakosm Ekolog Med 39: 43–47, 2005. 34. Tayo BO, Luke A, McKenzie CA, Kramer H, Cao G, Durazo-Arvizu R, Forrester T, Adeyemo AA, Cooper RS. Patterns of sodium and potassium excretion and blood pressure in the African Diaspora. J Hum Hypertens 26: 315–324, 2012. 35. Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol 4: 90, 2013.
J Appl Physiol • doi:10.1152/japplphysiol.00894.2014 • www.jappl.org
Seasonal variation in natural abundance of 2H and rural Nigeria
18
O in urine samples from
Justin E. Harbison,1 Lara R. Dugas,1 William Brieger,2 Bamidele O. Tayo,1 Tunrayo Alabi,3 Dale A. Schoeller,4 and Amy Luke1 1
Division of Epidemiology, Department of Public Health Sciences, Loyola University School of Medicine, Maywood Illinois; Department of International Health, Bloomberg School of Public Health, John Hopkins University, Baltimore Maryland; 3 Geospatial Laboratory, Research for Development, International Institute of Tropical Agriculture, Ibadan, Nigeria; 4 Department of Nutritional Sciences, University of Wisconsin, Madison, Wisconsin 2
Submitted 10 October 2014; accepted in final form 7 May 2015
Harbison JE, Dugas LR, Brieger W, Tayo BO, Alabi T, Schoeller DA, Luke A. Seasonal variation in natural abundance of 2H and 18O in urine samples from rural Nigeria. J Appl Physiol 119: 55–60, 2015. First published May 14, 2015; doi:10.1152/japplphysiol.00894.2014.—The doubly labeled water (DLW) method is used to measure free-living energy expenditure in humans. Inherent to this technique is the assumption that natural abundances of stable isotopes 2H and 18O in body water remain constant over the course of the measurement period and after elimination of the loading dose of DLW will return to the same predose level. To determine variability in the natural abundances of 2 H and 18O in humans living in a region with seasonal shifts in rain patterns and sources of drinking water, over the course of 12 mo we collected weekly urine samples from four individuals living in southwest Nigeria as well as samples of their drinking water. From ongoing regional studies of hypertension, obesity, and energy expenditure, we estimated average water turnover rate, urine volumes, and sodium and potassium excretion. Results suggest that 2H and 18O in urine, mean concentrations of urinary sodium and potassium, urine volume, and total body turnover differed significantly from dry to rainy season. Additionally, seasonal weather variables (mean monthly maximum temperatures, total monthly rainfall, and minimum relative humidity) were all significantly associated with natural abundances in urine. No seasonal difference was observed in drinking water samples. Findings suggest that natural abundances in urine may not remain constant as assumed, and studies incorporating DLW measurements across the transition of seasons should interpret results with caution unless appropriate doses of the tracers are used. doubly labeled water; seasonal variation; Nigeria THE DOUBLY LABELED WATER (DLW) method, developed by Lifson in the late 1940s, has been utilized for over 60 years to measure free-living total energy expenditure (TEE) in animals, with early studies focusing on small mammals, pigeons, reptiles, and locusts (15, 17, 21, 22). Several decades later, it was adapted for use in humans by Schoeller and colleagues (7, 18, 26, 29 –32), providing objective measurement of energy expenditure and energy requirements. The basic principle behind the method is that the stable isotopes of oxygen in body water are in isotopic equilibrium with the oxygen in respiratory CO2. Thus, after a loading dose of DLW, i.e., enriched 2H218O, there is differential elimination of the isotopes. Deuterium is eliminated solely through water, e.g., urine, transcutaneous water, and respiratory water, whereas 18O is eliminated both via water and via respiratory CO2. Thus the differential elimination of
Address for reprint requests and other correspondence: J. Harbison, Dept. of Public Health Sciences, Loyola Univ. Chicago, 2160 South First Ave., Maywood, IL 60153 (e-mail: [email protected]). http://www.jappl.org
18
O is a measure of CO2 flux, thereby allowing for the calculation of TEE (30). One of the key assumptions of the DLW method is that the natural abundances, i.e., baseline concentrations, of the stable isotopes 2H and 18O are constant in an animal’s body throughout the measurement period, i.e., typically 1–2 wk for human studies, and that body water will return to the same predose abundances after the elimination of the loading dose (22). In reality, natural abundances of the isotopes in a body vary with the source and amount of water and food consumed, as well as with changes in the proportions of oxygen and hydrogen lost through CO2 production and the ratio of fractionated and nonfractionated water effluxes. Increases or decreases in isotope abundances can occur when water intake varies or when there is a change in water source. Changes in natural abundances may introduce error into the calculation of TEE (12). Of concern for human studies is when the natural abundance of one or both of the isotopes changes during the course of a measurement period. A stark example is that of measuring energy requirements during space flight. It is well established that astronauts’ drinking water has unusual isotopic abundances, and those changes in isotopic baseline must be made before or during the calculation of TEE (14, 33). There is also the potential for changing isotopic baseline as a result of dramatic seasonal effects, resulting in sudden changes in sources of drinking water, such as those that can occur in regions experiencing clearly demarcated rainy and dry seasons. We sought to determine whether the season of DLW measurement in the tropics, i.e., rainy or dry, affects the natural abundances of deuterium and 18O and whether or not changes in water source could explain any identified effect. Using data collected over the course of two research studies conducted in rural southwestern Nigeria, we documented natural abundance of the isotopes in urine over the course of a year, average water turnover rates (8, 19), and average urine volumes by season (34). METHODS
In conjunction with a concurrent study of TEE and weight change among rural Nigerians (8), weekly drinking water and urine samples were collected over the course of a year by four volunteers residing in the same community. The TEE and weight change study also allowed the calculation of water turnover rates by season. Urine volume and electrolyte concentrations were obtained from a study on sodium and blood pressure conducted in the same community (34). The studies were approved by the Institutional Review Board of Loyola University Chicago, Stritch School of Medicine, and the Ethics Committee of
8750-7587/15 Copyright © 2015 the American Physiological Society
55
56
Seasonal Variation of Deuterium and
the University College Hospital, University of Ibadan (Nigeria); written, informed consent was obtained from all participants. The studies were conducted in a village in rural southwest Nigeria, Igbo-Ora, located at latitude 7°N, where Loyola University Chicago Department of Public Health Sciences maintained a free-standing research site from the mid-1990s to 2009 (20). Drinking water for residents of the village is obtained from three primary sources: wells, boreholes, and collected rainwater. The source of drinking water tends to vary seasonally, with wells and collected rainwater being predominant during the rainy season and boreholes used more during the dry season. Nigeria has a tropical climate marked by consistently high temperatures (34/19°C, max/min) and wet and dry seasons associated with the Intertropical Convergence Zone. Whereas much of the country will have one rainy and dry season annually, in the southwestern part of the country (including the study area), a short decline in rainfall and intensity can occur in late July and August because of the Intertropical Convergence Zone (1, 23). The result is a long rainy season approximately from March to early July, a short dry period (or “little dry season”) in late July and August, a shorter rainy period in September and October, and a long dry season from November to February. For the year of data collection, there was no observable little dry season. Four participants from the two studies collected weekly samples of both their drinking water and urine between October 1, 2001 and September 16, 2002; samples were not collected during the month of December 2001 and half of January 2002, as the research staff was on holiday, resulting in a total of 40 wk of sample collection. Participants also recorded the source of the drinking water sample weekly, i.e., well, borehole, or collected rainwater. The first urine void on Monday mornings was collected along with a sample of the current source of drinking water. Duplicate 5-ml aliquots of each sample were prepared and stored in cryovials with Teflon O-rings. Samples were frozen at ⫺4°C and stored at the University College Hospital in Ibadan until being shipped to the University of Wisconsin, Madison for analysis in October 2002. Drinking water and urine samples from every second week were analyzed for abundance of deuterium and 18O at the Stable Isotope Core Laboratory in the Department of Nutritional Sciences, University of Wisconsin, Madison. Both urine and water samples were treated with 200 mg carbon black and filtered using 0.45-m sterile filters. Isotopes were analyzed using isotope ratio mass spectrometry as previously published (12). Daily temperatures and rainfall estimates during the same time period were recorded by two separate observers in the villages. These data were verified, and monthly averages of rainfall, maximum temperature, minimum relative humid-
18
Oxygen
•
Harbison JE et al.
ity, as well as the dates for the initiation and cessation of the annual rains were obtained from the Statistics Section of the International Institute of Tropical Agriculture, Ibadan, Nigeria, located ⬃60 km from Igbo-Ora. The average rate of water turnover by season was estimated using data from the parent project in which TEE was measured using DLW in 149 women (8). Total water turnover was calculated (kd ⫻ Nd/Fc) as the product of deuterium elimination rate (kd) and the deuterium dilution space (Nd), and corrected for isotope fractionation (Fc). The average value of Fc (9) was 0.98. Volume of urine produced and total urinary sodium (Na⫹) and potassium (K⫹) excretion were obtained during a separate project on hypertension and electrolyte excretion (34) in 756 adults from the same village. Mean volumes and electrolyte excretion of three 24-h urine collections were used in the present analyses. Data were analyzed using STATA (ver 12.0; College Station, TX). Average deuterium and 18O abundances for both urine and drinking water samples were calculated by season and expressed relative to standard mean ocean water (6). Covariation of the isotopic abundances over the course of the year was defined as having an intraindividual Pearson correlation coefficient ⱖ0.75 (12). Student’s t-tests were used to compare isotopic abundances, water turnover rates, urine volumes, and electrolyte concentrations between the rainy and dry seasons. RESULTS
The four participants (3 men and 1 woman), were 57.5 ⫾ 6.5 yr old and all resided in the village for the entire study period. For each participant, 21 urine and 21 drinking samples were analyzed. The natural abundances of deuterium and 18O of these urine samples are presented in Fig. 1. Although it appears that there are two outliers (one for subject 1 and the other for subject 4), there was no known physiological or environmental explanation, and they were included in all analyses. On the basis of daily rainfall patterns as recorded by local volunteers, March 12, 2002 was designated as the start of the rainy season; thus samples collected between October 1, 2001 and March 11, 2002 were determined to be from the dry season, and those from March 12, 2002 to the end of the study were from the rainy season. There was a marked difference in rainfall pattern indicating the occurrence of one short dry season (Fig. 2); therefore, a single dry and single rainy season were defined for analyses (October 1 to March 11 and March 12 to September
Subject 2 Subject 4
0 -10 -20 -30
Fig. 1. Deuterium and 18O abundances in urine from 4 adults in rural southwest Nigeria sampled between October 1, 2001 and September 16, 2002. The slope coefficients ⫾ SE for participants 1 to 4 were 6.00 ⫾ 0.29, 3.84 ⫾ 0.22, 5.52 ⫾ 0.26, and 5.35 ⫾ 0.50, respectively. per mil, parts per thousand.
Deuterium (per mil)
10
Subject 1 Subject 3
-6
-4
-2 18-Oxygen (per mil)
J Appl Physiol • doi:10.1152/japplphysiol.00894.2014 • www.jappl.org
0
Seasonal Variation of Deuterium and
A4
Temperature
Rainfall
Oxygen-18
45
Deuterium
40
3
35
2
25 0 20
cm and °C
Del SMOW
30 1
-1 15 -2
10
-3
5
-4
0 Oct
Nov
Dec
2001
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
2002
Oxygen-18
Deuterium
80
Minimum humidity
3
70
2
60
1
50
0
40
-1
30
-2
20
-3
10
-4
percent humidity
Del SMOW
B4
Jan
0 Oct
Nov
Dec
2001
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
2002
C4
Oxygen-18
Deuterium
200
Evaporation
180
3
160 2
120 100
0
mm per month
Del SMOW
140 1
80
-1
60 -2 40 -3
20 0
-4 Oct 2001
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
2002
Fig. 2. Mean (⫾ SE) monthly natural abundance of deuterium and 18O in urine from 4 adults in rural southwest Nigeria including median monthly temperature and total monthly rainfall (A), average monthly relative humidity (B), and total monthly evaporation (C) sampled between October 1, 2001 and September 16, 2002. Del SMOW, ␦ standard mean ocean water.
18
Oxygen
•
Harbison JE et al.
57
30, respectively). The average monthly maximum temperature was 33.7°C and 29.8°C during the dry and rainy seasons, respectively, while mean rainfall was 1.65 and 20.3 cm/mo, and minimum relative humidity was 43.8% and 64.6%. The mean abundances for both deuterium and 18O for the urine and two drinking water sources are presented by season in Table 1. Both isotopes were significantly more enriched in the urine samples during the dry compared with the rainy season (P ⬍ 0.001). The greatest variance in the urinary isotopes appeared to occur from January to May when the dry season was ending and the rains began to increase (Fig. 2). Mean monthly abundances of both deuterium and 18O in urine were also positively associated with mean monthly maximum temperatures (r ⫽ 0.55, P ⫽ 0.08 and r ⫽ 0.86, P ⬍ 0.001 respectively) and significantly inversely associated with rainfall (⫺0.74 and ⫺0.79; P ⬍ 0.001) and minimum relative humidity (⫺0.87 and ⫺0.91; P ⬍ 0.001). The majority of drinking water samples were taken from local wells during both the dry and rainy seasons (n ⫽ 57), followed by borehole samples (n ⫽ 21), and there were no significant differences in isotope abundances for either the well water or borehole water samples by season. Rain water samples (n ⫽ 6) were all collected during the rainy season. The mean rate of total body water turnover was significantly higher (P ⬍ 0.001) in the dry season (0.13 ⫾ 0.02 pools/day) than in the rainy season (0.11 ⫾ 0.02 pools/day; P ⬍ 0.001), as was absolute volume of daily water turnover, 3.8 ⫾ 0.8 vs. 3.5 ⫾ 0.7 l (P ⬍ 0.02). Average total urine volumes were significantly lower in the dry (1.34 ⫾ 0.50 l/day) compared with the rainy season (1.53 ⫾ 0.50 l/day; P ⬍ 0.001). Whereas urinary sodium concentrations were not different between dry and rainy seasons (117 ⫾ 4 vs. 121 ⫾ 2 meq/day, respectively), urinary potassium was significantly higher in the dry season, 49.9 ⫾ 1.8 vs. 43.1 ⫾ 0.9 meq/day (P ⬍ 0.001). Both the higher urinary potassium concentrations and a lower urine volume/water turnover rate indicate more concentrated urine being observed during the dry season. In Fig. 3, the difference between the abundance of the isotopes in urine and drinking water is plotted against the rate of water turnover. Although the rate of water turnover only explains 25% of the variation in deuterium abundance and 7% of the 18O abundance, these data support the concept that baseline urinary isotope abundances reflect water turnover. DISCUSSION
The natural abundances of deuterium and 18O in humans, as measured in urine, do vary significantly between dry and rainy seasons in this tropical region, with both being more enriched during the dry season. On the basis of our findings, the observed urinary differences in abundance between the seasons are not due to changes in the isotopic abundance of drinking water but rather likely due to changes in water turnover that are potentially influenced by environmental factors, including temperature and relative humidity (28). The DLW method has been widely used to measure TEE in multiple populations around the world; it has also been used to establish energy intake requirements for individuals of all ages and phase of life. Because the technique is based on the assumption that natural abundances within the body remain constant throughout the measurement period, knowing if and
J Appl Physiol • doi:10.1152/japplphysiol.00894.2014 • www.jappl.org
58
Seasonal Variation of Deuterium and
Table 1. Mean abundances for deuterium and
18
Oxygen
•
Harbison JE et al.
18
O for urine and drinking water sources by season
2
18
Hydrogen
Urine Well water Borehole water Rain water
Oxygen
Dry
Rainy
P Value
Dry
Rainy
P Value
0.47 (2.24) ⫺11.79 (4.00) ⫺12.71 (1.87) —
⫺2.19 (2.29) ⫺12.61 (4.83) ⫺11.56 (3.89) ⫺6.33 (6.92)
⬍0.001 0.49 0.39 —
⫺0.56 (0.46) ⫺2.96 (0.35) ⫺3.38 (0.50) —
⫺0.93 (0.49) ⫺2.90 (0.53) ⫺3.22 (0.52) ⫺2.18 (0.83)
⬍0.001 0.65 0.48 —
Applicable values are presented as means (SE). Number of samples analyzed (by season): urine (dry) n ⫽ 35, urine (rainy) n ⫽ 49; well water (dry) n ⫽ 23, well water (rainy) n ⫽ 34; borehole water (dry) n ⫽ 11, borehole water (rainy) n ⫽ 10; rain water (rainy) n ⫽ 6. Rain water was not used as drinking water source during dry season.
possibly when baseline abundances vary could better inform the use of this method in the tropics. Rural southwest Nigeria is a compelling area to evaluate natural abundances of stable isotopes because of its relatively warm year-round temperatures and distinct seasonal changes in rainfall and sources of drinking water. As hypothesized, measured natural abundances of deuterium and 18O did indeed differ significantly by season, however, only in urine and not well and borehole drinking water sources. It is unknown whether the addition of rainwater as a drinking water source during the rainy season may have played a role in the observed seasonal difference urine isotopes, as so few samples of rainwater were collected in this study. In studies of water movement and budget, the isotopic variance in rainwater is used to identify water from a precipitation event as opposed to ground water (11, 16). This suggests that, when used as a primary water source during only one season, rainwater might be associated with a shift in a body’s natural abundances of deuterium and 18O. During the study period, it appeared as if the greatest variance in urinary isotopes from the four participants occurred during the transition from dry to rainy seasons when rains began to increase in intensity. The weather-related variables of mean monthly maximum temperatures, rainfall, and minimum relative humidity were all significantly associated with natural abundance in urine and provide additional evidence that,
Fig. 3. Relationship between water turnover, presented as the natural log (Ln) of daily deuterium flux and the abundance of deuterium and 18O in urine relative to drinking water. permil, parts per thousand. D/H, deuterium/hydrogen.
within the climatic region of the study, seasonal weather patterns over the year can affect isotope abundance within the body. Physiologically, weather can influence daily obligatory water loss by increasing transcutaneous evaporation during extended periods of high temperatures and low humidity and as a result may impact deuterium and 18O abundances in the body. This would also explain the seasonal differences in turnover rate, urine volume, and K⫹ concentrations observed in the study. In studies of the isotopic compositions of water, the most significant fractionations are associated with phase changes (i.e., condensation and evaporation) (16), and it would thus follow that such variations could similarly occur as those phase changes manifest from the human body. Obviously a larger number of study participants could help clarify the relationship between our measured environmental variables and natural abundances, but the observed seasonal variances of our measured variables do appear to be associated with changes in water turnover rates (28). If mean monthly maximum temperatures, rainfall, and minimum relative humidity are indeed associated with fluctuations of natural abundances within the body, then other regions of the world, and particularly those with distinct seasons (e.g., higher-latitude climates with cold winters), may experience a similar or more pronounced isotopic variation. For example, increasing latitudes in the Northern Hemisphere have increasing ranges of monthly temperatures, which, as our results indicate, could potentially affect or add variance to natural abundances within participants of studies performed in those areas. This may particularly be the case during times of seasonal transition (months of September to October, March to May, etc.) and in regions farther away from larger bodies of water (e.g., inland vs. coastal areas) when temperatures and precipitation may fluctuate more. Likewise a subject’s daily exposure to environmental temperatures, humidity, and rainfall may also influence natural abundances. The four study participants lived in and worked in a rural area of Nigeria where air conditioning was uncommon, and these subjects were likely more directly exposed to environmental variables associated with weather than individuals residing and working in more urbanized climate-controlled environments. Such differences in local environmental exposures (occupational, rural vs. urban, etc.) may also introduce more variance into natural abundance measurements. However, potential error resulting from natural fluctuations in abundance can be reduced when deuterium and 18O are dosed in a manner such that enrichments have a ratio similar to the ratio of the anticipated abundance changes in urine. Natural changes in abundances create similar errors in the elimination
J Appl Physiol • doi:10.1152/japplphysiol.00894.2014 • www.jappl.org
Seasonal Variation of Deuterium and
rates of both isotopes, and thus these errors cancel when the difference of the two elimination rates is calculated. Anyone performing studies in which samples are collected weeks after the predose sample should be careful not to underestimate the scale and effect of natural isotopic variations over time, which may be significant, as the results of this study suggest. Ultimately such an underestimation may lead to error as a great as 30% (27). The doses of 18O- and 2H-enriched water should yield a postdose enrichment in parts per thousand units of 8 to 1 of 18O to 2H enrichment, respectively (4), although there is very little effect for doses in the ratio of between 6 and 12 to 1 (12). In our laboratory, we aim for a postdose 18O enrichment of 90 parts. With the use of 10 atom percent (AP) 18O water and 99.9 AP deuterium, respectively, this corresponds to a dose weight of 1.8 g of 10 AP 18O water per kilogram of total body water and 0.13 g of 99.9 AP 2H2O per kilogram total body water. Although it appears from this study that natural abundances of 2H and 18O in urine may change over the scope of the year, how this information will affect the results of other studies requires further inquiry. In general, energy expenditure studies will utilize the DLW method for a much smaller 1– 4-wk measurement period (35), which may or may not be too short a time period for significant isotopic variance to occur. Often, however, the exact dates of DLW measurements are not reported, so critical consideration of any potential effect of seasonality becomes difficult. Such information would appear to be particularly relevant for TEE studies in areas such as parts of Cameroon (2, 3), Brazil (25), India (13, 5), Peru (10), and Kenya (24) that experience similar wet and dry seasonality as our study site. Overall, our results suggest that, at least in southwest Nigeria, the assumption that natural abundances will return to predose measurements during DLW measurements may not be correct, and the accuracy of this method may depend on the timing of measurements, particularly during the transition between dry and rainy seasons. Results of energy expenditure studies that utilize DLW measurements for human TEE calculations during a period of seasonal transition or over a longer period of time should be interpreted with caution. The potential error, however, can be minimized by dosing the two stable isotopes in an appropriate ratio such that errors in the 2H and 18O elimination rates cancel when the difference between the rates is calculated. ACKNOWLEDGMENTS We thank Amanda Romsa for the mass spectrometric analysis of the samples. GRANTS This research was supported in part by grants from the National Institutes of Health (DK56780 and DK30031). DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the authors. AUTHOR CONTRIBUTIONS Author contributions: J.E.H., L.D., and B.T. analyzed data; J.E.H. and D.A.S. interpreted results of experiments; J.E.H. and L.D. prepared figures; J.E.H. drafted manuscript; J.E.H., L.D., W.B., B.T., T.A., D.A.S., and A.L. edited and revised manuscript; J.E.H., L.D., W.B., B.T., T.A., D.A.S., and A.L. approved final version of manuscript; L.D., W.B., B.T., T.A., D.A.S., and A.L. conception and design of research; L.D., W.B., B.T., T.A., and A.L. performed experiments.
18
Oxygen
•
Harbison JE et al.
59
REFERENCES 1. Adejuwon JO, Odekunle TO. Variability and the severity of the “Little Dry Season” in southwestern Nigeria. J Climate 19: 483–493, 2006. 2. Assah FK, Ekelund U, Brage S, Mbanya JC, Wareham NJ. Free-living physical activity energy expenditure is strongly related to glucose intolerance in Cameroonian adults independently of obesity. Diabetes Care 32: 367–369, 2009. 3. Assah FK, Ekelund U, Brage S, Corder K, Wright A, Mbanya JC, Wareham NJ. Predicting physical activity energy expenditure using accelerometry in adults from sub-Sahara Africa. Obesity (Silver Spring) 17: 1588 –1595, 2009. 4. Berman ES, Melanson EL, Swibas T, Snaith SP, Speakman JR. Interand intraindividual correlations of background abundances of 2H, 18O and 17 O in human urine and implications for DLW measurements. Eur J Clin Nutr. In press. 5. Corder K, Brage S, Wright A, Ramachandran A, Snehalatha C, Yamuna A, Wareham NJ, Ekelund U. Physical activity energy expenditure of adolescents in India. Obesity (Silver Spring) 18: 2212–2219, 2010. 6. Craig H. Isotope variations in meteoric waters. Science 133: 1702–1703, 1961. 7. Dugas LR, Harders R, Merrill S, Ebersole K, Shoham DA, Rush EC, Assah FK, Forrester T, Durazo-Arvizu RA, Luke A. Energy expenditure in adults living in developing compared with industrialized countries: a meta-analysis of doubly labeled water studies. Am J Clin Nutr 93: 427–441, 2011. 8. Ebersole KE, Dugas LR, Durazo-Arvizut RA, Adeyemo AA, Tayo BO, Omotade OO, Brieger WR, Schoeller DA, Cooper RS, Luke AH. Energy expenditure and adiposity in Nigerian and African-American women. Obesity (Silver Spring) 16: 2148 –2154, 2008. 9. Fjeld CR, Brown KH, Schoeller DA. Validation of the deuterium oxide method for measuring average daily milk intake in infants. Am J Clin Nutr 48: 671–679, 1988. 10. Fjeld CR, Schoeller DA, Brown KH. A new model for predicting energy requirements of children during catch-up growth developed using doubly labeled water. Pediatr Res 25: 503–508, 1989. 11. Granger SJ, Bol R, Meier-Augenstein W, Leng MJ, Kemp HF, Heaton TH, White SM. The hydrological response of heavy clay grassland soils to rainfall in south-west England using ␦2H. Rapid Commun Mass Spectrom 24: 475–482, 2010. 12. Horvitz MA, Schoeller DA. Natural abundance deuterium and 18-oxygen effects on the precision of the doubly labeled water method. Am J Physiol Endocrinol Metab 280: E965–E972, 2001. 13. Krishnaveni GV, Veena SR, Kuriyan R, Kishore RP, Wills AK, Nalinakshi M, Kehoe S, Fall CH, Kurpad AV. Relationship between physical activity measured using accelerometers and energy expenditure measured using doubly labelled water in Indian children. Eur J Clin Nutr 63: 1313–1319, 2009. 14. Lane HW, Gretebeck RJ, Schoeller DA, Davis-Street J, Socki RA, Gibson EK. Comparison of ground-based and space flight energy expenditure and water turnover in middle-aged healthy male US astronauts. Am J Clin Nutr 65: 4 –12, 1997. 15. Lee JS, Lifson N. Measurement of total energy and material balance in rats by means of doubly labeled water. Am J Physiol 199: 238 –242, 1960. 16. Lewicka-Szczebak D, Je˛drysek MO. Tracing and quantifying lake water and groundwater fluxes in the area under mining dewatering pressure using coupled O and H stable isotope approach. Isotopes Environ Health Stud 49: 9 –28, 2013. 17. Lifson N, Gordon GB, Visccher MB, Nier AO. The fate of utilized molecular oxygen and the source of the respiratory carbon dioxide, studied with the aid of heavy oxygen. J Biol Chem 180: 804, 1949. 18. Luke A, Adeyemo A, Kramer H, Forrester T, Cooper RS. Association between blood pressure and resting energy expenditure independent of body size. Hypertension 43: 555–560, 2004. 19. Luke A, Dugas LR, Ebersole K, Durazo-Arvizu RA, Cao G, Schoeller DA, Adeyemo A, Brieger WR, Cooper RS. Energy expenditure does not predict weight change in either Nigerian or African American women. Am J Clin Nutr 89: 169 –176, 2009. 20. Luke A, Durazo-Arvizu R, Rotimi C, Prewitt TE, Forrester T, Wilks R, Ogunbiyi OJ, Schoeller DA, McGee D, Cooper RS. Relation between body mass index and body fat in black population samples from Nigeria, Jamaica, and the United States. Am J Epidemiol 145: 620 –628, 1997.
J Appl Physiol • doi:10.1152/japplphysiol.00894.2014 • www.jappl.org
60
Seasonal Variation of Deuterium and
21. McClintock R, Lifson N. Applicability of the D2O18 method output of obese mice the total carbon dioxide output. J Biol Chem 226: 153–156, 1957. 22. Nagy KA. CO2 production in animals: analysis of potential errors in the doubly labeled water method. Am J Physiol Regul Integr Comp Physiol 238: R466 –R473, 1980. 23. Odekunle TO. An assessment of the influence of the inter-tropical discontinuity on inter-annual rainfall characteristics in Nigeria. Geograph Res 48: 314 –326, 2010. 24. Ojiambo R, Gibson AR, Konstabel K, Lieberman DE, Speakman JR, Reilly JJ, Pitsiladis YP. Free-living physical activity and energy expenditure of rural children and adolescents in the Nandi region of Kenya. Ann Hum Biol 40: 318 –323, 2013. 25. Scagliusi FB, Ferriolli E, Pfrimer K, Laureano C, Cunha CS, Gualano B, Lourenço BH, Lancha AH Jr. Characteristics of women who frequently under report their energy intake: a doubly labelled water study. Eur J Clin Nutr 63: 1192–1199, 2009. 26. Schoeller DA. Energy expenditure from doubly labeled water: some fundamental considerations in humans. Am J Clin Nutr 38: 999–1005, 1983. 27. Schoeller DA. Measurement of energy expenditure in free-living humans by using doubly labeled water. J Nutr 118: 1278 –1289, 1988. 28. Schoeller DA, Leitch CA, Brown C. Doubly labeled water method: in vivo oxygen and hydrogen isotope fractionation. Am J Physiol 251: R1137–R1143, 1986.
18
Oxygen
•
Harbison JE et al.
29. Schoeller DA, Ravussin E, Schutz Y, Acheson KJ, Baertschi P, Jéquier E. Energy expenditure by doubly labeled water: validation in humans and proposed calculation. Am J Physiol 250: R823–R830, 1986. 30. Schoeller DA, van Santen E. Measurement of energy expenditure in humans by doubly labeled water method. J Appl Physiol 53: 955–959, 1982. 31. Schoeller DA, van Santen E, Peterson DW, Dietz W, Jaspan J, Klein PD. Total body water measurement in humans with 18O and 2H labeled water. Am J Clin Nutr 33: 2686 –2693, 1980. 32. Schoeller DA, Webb P. Five-day comparison of the doubly labeled water method with respiratory gas exchange. Am J Clin Nutr 40: 153–158, 1984. 33. Siniak IuE, Skuratov VM, Gaı˘dadymov VB, Ivanova SM, Pokrovskiı˘ BG, Grigor’ev AI. [Investigation into fractionating of hydrogen and oxygen stable isotopes aboard the international space station]. Aviakosm Ekolog Med 39: 43–47, 2005. 34. Tayo BO, Luke A, McKenzie CA, Kramer H, Cao G, Durazo-Arvizu R, Forrester T, Adeyemo AA, Cooper RS. Patterns of sodium and potassium excretion and blood pressure in the African Diaspora. J Hum Hypertens 26: 315–324, 2012. 35. Westerterp KR. Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects. Front Physiol 4: 90, 2013.
J Appl Physiol • doi:10.1152/japplphysiol.00894.2014 • www.jappl.org
