Developmental variations in plasma leptin, leptin soluble receptor and their molar ratio in healthy infants
© Koo et al; licensee BioMed Central Ltd. 2007
Received: 07 November 2006
Accepted: 04 June 2007
Published: 04 June 2007
Leptin and its soluble receptor (sOB-R) are important to regulation of body composition but there are no data on the developmental variations in these plasma variables and their relationship with body composition measurements,
Weight, length, and body composition (bone, fat and lean mass) by dual energy absorptiometry, and plasma variables were measured in healthy infants at 2, 4, 8 and 12 months.
15 whites and 29 African Americans (21 males and 23 females) with mean birth weight 3357 +/- 45 (SEM) g and gestation of 39.3 +/- 0.17 weeks were studied. The overall Z score for weight, length and weight for length during the study were 0.00 +/- 0.15, -0.08 +/- 0.11 and 0.12 +/- 0.14 respectively. With increasing age, plasma leptin (1.0 to 18.2, median 5.5 ng/mL) and sOB-R:leptin molar ratio (10.1 to 247.4, median 59.9) were lowered (r = -0.47, p < 0.01; and r = -0.37, p < 0.05 respectively), best predicted by weight Z score and percentage of fat mass, and higher in African American and female. Presence of body composition measurements eliminated the race and gender effect on the plasma variables. Plasma sOB-R (49.5 to 173.9, median 81.3 ng/mL) did not change significantly with age and was correlated and predicted only by body composition measurements.
In healthy growing infants, plasma leptin but not sOB-R decreases with age. Gender, race and anthropometric measurements are additional physiological determinants predictive of plasma leptin and the receptor:ligand ratio. However, body composition is the only variable that can predict plasma leptin and its soluble receptor and the receptor: ligand ratio; and body composition measurements eliminated the race and gender effect on these plasma variables.
Hormonal responses from the gastrointestinal tract, brain and body tissues associated with dietary intake and regulation of nutritional status and body weight are well described in adults [1–3]. There are increasing data indicating that similar hormonal changes can occur in children and infants [4–9]. Understanding their relationship with body composition measurements as indicators of tissue accretion might shed light on the physiological basis to integrate nutrition support, growth and tissue accretion.
Concentration of circulating leptin, an adipocyte hormone, reflects the amount of energy stored in adipose tissue and is considered a marker of nutritional status [1–3]. Leptin is bound to the soluble leptin receptor (sOB-R) in the circulation, which modulates steady state leptin levels by preventing the hormone from degradation and clearance , and sOB-R:leptin ratio may be considered a marker of bioavailable leptin.
In the neonate, dramatic changes in circulating leptin and sOB-R with a decrease in leptin by as much as 80% [5–7] and increase in sOB-R by >5 folds  are reported. At birth, weight and body mass index are positively correlated with plasma leptin concentrations [5, 6, 8], and one report indicated that sOB-R is inversely related to birth weight . This pattern of changes in leptin and its sOB-R and their relationship with body weight and body mass index presumably reflect the loss of placental contribution for leptin and the physiologic adaptation to lower availability of free leptin, thereby allowing an increase in energy intake to initiate the phase of rapid postnatal growth.
Infancy is the period of most rapid postnatal growth and one report in healthy but relatively malnourished infants with average weight for age Z score of -1.9 at 52 weeks shows a decrease in plasma leptin from birth to 1 year. Changes in plasma sOB-R in infants beyond the immediate newborn period are not well defined. Furthermore, the relationship between body composition measurements as reflection of tissue accretion, with these plasma variables is not known. The aim of this study is to define the developmental variations in circulating leptin and sOB-R in healthy infants specifically the physiologic determinants including age, gender, race, and measures of growth or body composition on these plasma variables.
This is a cross sectional design with anthropometry, body composition and circulating leptin and sOB-R measurements performed on the same day for each subject. All subjects were singleton infants between 37 and 42 weeks gestation with appropriate birth weight for gestation  and studied at approximately 2, 4, 8 and 12 months. None of the infants had major malformation or medical or surgical conditions that may affect long term growth.
Weight, length, and head circumference were measured using standard methods [12, 13]. Infants were weighed in the nude to the nearest 5 g using an electronic scale (Seca, Toledo, OH) that was calibrated daily. Length was measured in duplicate to the nearest 0.1 cm with the infant in a recumbent position using O'Leary Lengthboard ™ (Ellard Instruments Ltd, Seattle, WA).
Body composition is indicated by total body bone mass as bone mineral content, and fat and lean mass were determined by fan beam dual energy X-ray absorptiometry (DXA) (QDR 4500A, Hologic Inc, Waltham, MA). Scan acquisition techniques have been reported elsewhere . Each infant was wrapped in cotton blanket for the scan. The use of a diaper with or without light undergarment for the infant was allowed prior to bundling the infant in the cotton blanket. However, all coverings were weighed with an electronic scale and the weight recorded. Scan analysis used software vKHS11 validated by carcass analysis [15, 16]. Only scans with no significant movement artifacts  were included in data analysis. In our laboratory, the precision error  from duplicate infant whole body scans for bone, fat and lean mass were 2.6, 7.1 and 2.5% respectively. Bone, fat and lean mass also were expressed as a percent of DXA measured total body mass.
Plasma samples for the measurement of leptin and sOB-R were kept at -70°C until measurement. Plasma leptin and sOB-R were measured using the commercial enzyme linked immunoassay kits from the same manufacturer (Diagnostic Systems Laboratories, Inc., Webster, TX). Recombinant human leptin and human soluble leptin receptor were used as standards and controls in the respective assays. In our laboratory, the coefficient variation of the leptin assay was 9% and for the sOB-R assay was 5%.
Ethical approval for the study was obtained from the Institutional Review Board for Human Investigations at Wayne State University, Detroit, MI. Written informed consent was obtained from the parent of each infant.
The absolute values of anthropometric measurements were normalized by expression as Z scores using the age and gender matched normative data from the National Center for Health Statistics . The absolute values for bone, fat and lean mass were transformed to a percentage of total weight. Plasma sOB-R:leptin molar ratio was calculated according to the molecular mass of 130 kD for sOB-R and 16 kD for leptin. All statistical analysis included the use of absolute and transformed measurements.
Pearson's correlation was used to determine the relationship for each plasma parameter (leptin, sOB-R and sOB-R:leptin ratio) with each anthropometric and each DXA parameter. Analysis of covariance was used to determine the relation of gender, race and age to plasma measurements. Stepwise regression analysis was used to determine the relative contribution from each of the physiologic variables in the prediction of plasma leptin, sOB-R, and sOB-R:leptin ratio. By design, the use of absolute and percentage of body composition measurement was mutually exclusive in regression analysis, as well as the use of anthropometric and DXA measurements, since body weight is the sum of body composition measurements; and both weight and length are predictive of various aspects of body composition [20–22]. Neither age nor gender was entered as independent variables with any analysis using Z scores since the Z scores were standardized to age and gender.
All values are mean +/- SEM. Statistical tests were performed with SPSS Version 13.5 for Windows (SPSS Inc., Chicago, IL) at an adopted significance level of 0.05 and were two-tailed.
There were 44 infants with mean birth weight 3357 +/- 45 g and gestation of 39.3 +/- 0.17 weeks, with 15 whites and 29 African Americans, and 21 (7 white) males and 23 (8 white) females. Anthropometric and DXA measurements and blood collection were performed at 56 +/- 0.8, 112 +/- 1.0, 240 +/- 1.5, and 366 +/- 2.7 days. Age was positively correlated with all absolute values of anthropometric and DXA measurements (p < 0.01 for all comparisons). Weight for age Z score (WAZ) decreased (r = -0.36, p = 0.02) but recumbent length for age (HAZ) and weight for length (WHZ) Z scores were not significantly different with age. The overall WAZ, HAZ and WHZ were 0.00 +/- 0.15 and -0.08 +/- 0.11, and 0.12 +/- 0.14 respectively. Bone mass as a percentage of total mass was increased (r = 0.58, p < 0.01) but the percentages of fat and lean mass did not change significantly with age. The overall percentages for bone, fat and lean mass were 2.5 +/- 0.04, 26.3 +/- 0.92, and 71.2 +/- 0.93 respectively.
Correlation of plamsa leptin, leptin soluble receptor (sOB-R), and sOB-R:Leptin ratio with anthopometric measurements and their Z scores in term infants
Leptin soluble receptor ng/mL
sOB-R:Leptin molar ratio
Weight for length
Correlation of plasma leptin, leptin soluble receptor (sOB-R), and sOB-R:Leptin ratio with body composition measurements as absolute values or as percentage of total weight in term infants
Leptin soluble receptor ng/mL
sOB-R:Leptin molar ratio
As percentage of total weight
Plasma sOB-R:leptin molar ratio varied from 10.1 to 247.4 (median 59.9) and decreased with age (r = -0.37, p = 0.05). Weight and WAZ were the only anthropometric variables significantly correlated with plasma sOB-R:leptin ratio. Whereas, plasma sOB-R:leptin ratio was positively correlated with absolute and percent fat mass, but negatively correlated with lean mass and percent lean and bone mass. The relationship was stronger based on percent lean or fat mass than the absolute mass (Table 2). There was no race or gender effect on plasma sOB-R:leptin ratio.
Prediction values of physiologic variables including age, race, gender, and absolute values (left 2 columns) or age and gender specific Z scores (right 2 columns) of anthropometric measurements for plasma concentrations of leptin and leptin soluble receptor*, and leptin soluble receptor:leptin molar ratio.
Dependent variable – plasma leptin (ng/mL)
Independent variables: R2 0.49
Independent variables: R2 0.46
Weight Z score
Length Z score
Gender (Female = 1, male = 0)
Dependent variable – plasma soluble leptin receptor : leptin molar ratio
Independent variables: R2 0.50
Independent variables: R2 0.44
Weight Z score
Gender (Female = 1, male = 0)
Prediction values of physiologic variables including age, race, gender, and absolute values (left 2 columns) or percent bone, lean and fat mass (right 2 columns) for plasma leptin and leptin soluble receptor concentrations, and leptin soluble receptor:leptin molar ratio
Dependent variable – plasma leptin (ng/mL)
Independent variables: R2 0.37
Independent variables: R2 0.37
Dependent variable – plasma soluble leptin receptor (ng/mL)
Independent variables: R2 0.22
Independent variables: R2 0.43
Dependent variable – plasma soluble leptin receptor : leptin ratio
Independent variables: R2 0.54
Independent variables: R2 0.59
Study of leptin and its regulation has demonstrated its importance as an integral part of homeostatic mechanism in the regulation of body weight [1–3]. However, it is not known whether this is applicable to all life stages and what changes it may have during growth in which weight gain and tissue accretion rather than maintenance of body weight is the physiologic norm.
To our knowledge, this is the first report of the relationship of the circulating leptin and sOB-R concentrations to various physiological variables of growth and body composition specifically, bone, fat and lean masses, during infancy. Age appears to be a major physiologic determinant of plasma leptin concentrations. It is decreased during infancy in those with normal age and gender specific weight and length Z scores as indicated by our data, and is also decreased in infants with poor postnatal growth during longitudinal measurement of plasma leptin from cord blood and at 8, 16 and 52 weeks . In this study, the initial measurement of circulating leptin and its soluble receptor at 2 months likely eliminated the confounding factors of placental leptin , the initial physiologic adaptation that occurs commonly with other endocrine systems , and the apparent transient increase in plasma leptin during the first weeks after birth observed by some [25, 26] but not by other  investigators. In the two reports on postnatal increase in plasma leptin concentration, one reported an increase in plasma leptin at 30 days which was significantly correlated with interval weight gain  but no data was available beyond 30 days. In the other report, plasma leptin concentration was higher in term versus preterm infants up to 30 days but there was no significant difference between groups at 90 days. The increase in plasma leptin was correlated with weight gain and increase in subcutaneous tissue . Whether this transient increase in plasma leptin is related to changes in leptin transport, metabolism or clearance is not known.
Plasma sampling in our study tend to correspond to ages when milk intake is the exclusive or dominant source of nutrient, namely at 2 and 4 months, and when mixed diet becomes increasingly established at 8 and 12 months respectively. Our preliminary data suggest that usual dietary intake in healthy and normally grown infants probably does not affect plasma leptin or its soluble receptor concentrations, although determination of the relationship between leptin and its receptor with details of nutritional intake was not the primary goal of this study and further studies are needed.
Anthropometric and body composition measurements are related to and predictive of plasma leptin. This is consistent with other reports that plasma leptin is correlated with actual  or gain [25, 26] in body weight, and actual [8, 9] and changes  in body mass index, and with indirect indicators of body fat such as subcutaneous skinfold thicknesses [7, 25, 26]. Plasma leptin was also found to discriminate both the long term and changes in energy status based on skinfold thickness .
The consistent relation between the plasma leptin particularly with fat mass is supportive of adipose tissue being the major source of circulating leptin. The negative correlation of percent bone mass and percent lean mass with plasma leptin is not surprising since an increase in the proportion of fat mass is generally correlated with decreased proportion of lean and bone mass. However, a direct relation between plasma leptin and other tissue mass may be possible since increasing numbers of non-adipose tissues including skeletal muscle , chondrocyte  and human osteoblast  are reported to synthesize leptin and may have cellular leptin receptor forms with physiological activity in experimental models. In any case, the exact role of leptin in the changes in skeletal muscle and bone in humans remain to be defined.
Our data show elevated sOB-R concentrations throughout infancy. Other investigators have reported persistently elevated sOB-R concentrations during early childhood . The positive relation between plasma sOB-R and fat mass may be indicative of the increased membrane-bound leptin receptor forms, the source of sOB-R. The negative correlation of percent lean mass with plasma sOB-R is consistent with the generally inverse relation between percent lean and fat mass.
The correlation between these plasma variables with anthropometric or body composition measurements were better with the use of Z scores rather than absolute measurements, with fat percent rather than absolute fat mass, and generally better with body composition, specifically fat mass, rather than anthropometric measurements. Thus the use of standardized rather than absolute measurements of anthropometry and body composition is indicated in future studies on the interplay of leptin and its receptor with different nutrition support in growing subjects. Furthermore, body composition measurements are probably more sensitive indicators of leptin production and bioactivity.
It is interesting that even with the limited sample size, our findings of higher leptin concentration in African American infants independent of fat mass is consistent with the report on adult males and females that non-Hispanic blacks have slightly higher values compared to non-Hispanic whites or Mexican Americans . Our data of higher plasma leptin concentration in female infants also are consistent with the presence of sexual dimorphism [9, 32]. Furthermore, our data show that race and gender effects were eliminated in the presence of body composition measurements, presumably body composition measurements more specifically reflect the source of leptin and its receptors. Our data indicate that variations in plasma sOB-R are independent of race, gender or age but are predicted by body composition measurements.
Complexes of leptin with sOB-R reflect a molecular ratio of 1:1  and reached a median value of >10 as early as 3 days after birth because of a decrease in circulating leptin with an accompanied increase in sOB-R . Our data indicated that plasma sOB-R:leptin ratios remained >10 throughout infancy. Limited data indicate the high plasma sOB-R:leptin ratios may persist until 2 to 3 years . It is possible that high circulating concentrations of sOB-R may block leptin function by its competition with the membrane receptor for the ligand, which in turn may be an important stimulus for energy uptake in the rapidly growing infant or in other conditions with a high energy demand. However, the decreasing sOB-R:leptin ratio during later infancy is presumably associated with increasing bioavailable leptin, and is consistent with slowing of growth  and tissue accretion [20–22].
Our report represents an exploratory step to determine the developmental variations of plasma leptin and its soluble receptor during the period of most rapid postnatal growth when the body weight and tissue accretion triples over a one year period. These data when coupled with the body composition measurements are critical to the design of future studies to determine the interplay of leptin and its receptors with nutrition support and the regulation of growth and tissue accretion.
We conclude that in healthy growing infants, plasma leptin and sOB-R:leptin ratio but not sOB-R decreases with age. Body composition is the only variable that can predict plasma leptin and its soluble receptor and the receptor:ligand ratio, and body composition measurements eliminate the race and gender effect on these plasma variables. Based on limited size of subgroups in this study, the race and gender effect on these plasma variables appears to be consistent with that for adults.
- Janeckova R: The role of leptin in human physiology and pathophysiology. Physiol Res. 2001, 50: 443-459.PubMedGoogle Scholar
- Jequier E: Leptin signaling, adiposity, and energy balance. Ann NY Acad Sci. 2002, 967: 379-388.View ArticlePubMedGoogle Scholar
- Wynne K, Stanley S, McGowan B, Bloom S: Appetite control. J Endocrinol. 2005, 184: 291-318. 10.1677/joe.1.05866.View ArticlePubMedGoogle Scholar
- Ellis KJ, Nicolson M: Leptin levels and body fatness in children: effects of gender, ethnicity, and sexual development. Pediatr Res. 1997, 42: 484-8. 10.1203/00006450-199710000-00010.View ArticlePubMedGoogle Scholar
- Marchini G, Fried G, Ostlund E, Hagenas L: Plasma leptin in infants: relations to birth weight and weight loss. Pediatrics. 1998, 101: 429-432. 10.1542/peds.101.3.429.View ArticlePubMedGoogle Scholar
- Matsuda J, Yokota I, Iida M, Murakami T, Yamada M, Saijo T, Naito E, Ito M, Shima K, Kuroday Y: Dynamic changes in serum leptin concentrations during the fetal and neonatal periods. Pediatr Res. 1999, 45: 71-75. 10.1203/00006450-199901000-00012.View ArticlePubMedGoogle Scholar
- Kratzsch J, Schubring C, Stitzel B, Bottner A, Berthold A, Thiery J, Kiess W: Inverse changes in the serum levels of the soluble leptin receptor and leptin in neonates: relations to anthropometric data. J Clin Endocrinol Metab. 2005, 90: 2212-2217. 10.1210/jc.2004-1454.View ArticlePubMedGoogle Scholar
- Jaquet D, Leger J, Levy-Marchal C, Oury JF, Czernichow P: Ontogeny of leptin in human fetuses and newborns: effect of intrauterine growth retardation on serum leptin concentrations. J Clin Endocrinol Metab. 1998, 83: 1243-1246. 10.1210/jc.83.4.1243.View ArticlePubMedGoogle Scholar
- Collinson A, Moore S, O'Connell M, Charalambos C, Prentice A: Developmental changes in leptin as a measure of energy status in human infants in a natural ecologic setting. Am J Clin Nutr. 2005, 81: 488-494.PubMedGoogle Scholar
- Huang L, Wang Z, Li C: Modulation of circulating leptin levels by soluble receptor. J Biol Chem. 2001, 276: 6343-6349. 10.1074/jbc.M009795200.View ArticlePubMedGoogle Scholar
- Oken E, Kleinman KP, Rich-Edwards J, Gillman MW: A nearly continuous measure of birth weight for gestational age using a United States national reference. BMC Pediatr. 2003, 3: 6-10.1186/1471-2431-3-6.View ArticlePubMedPubMed CentralGoogle Scholar
- Gordon CC, Chumlea WC, Roche AF: Stature, recumbent length, and weight. Anthropometic Standardization Reference Manual. Edited by: Lohman TG, Roche AF, Martorell R. 1988, Champaign, IL: Human Kinetics Books, 3-8.Google Scholar
- Callaway CW, Chumlea WC, Bouchard C: Circumference. Anthropometic Standardization Reference Manual. Edited by: Lohman TG, Roche AF, Martorell R. 1988, Champaign, IL: Human Kinetics Books, 39-54.Google Scholar
- Hammami M, Koo WWK, Hockman EM: Technical considerations for fan beam dual energy x-ray absorptiometry body composition measurements in pediatric studies. JPEN. 2004, 28: 328-333.View ArticleGoogle Scholar
- Koo WWK, Hammami M, Hockman EM: Use of fan beam dual energy X-ray absorptiometry to measure body composition of piglets. J Nutr. 2002, 132: 1380-1383.PubMedGoogle Scholar
- Chauhan S, Koo WWK, Hammami M, Hockman EM: Fan beam dual energy X-ray absorptiometry body composition measurements in piglets. J Amer Coll Nutr. 2003, 22: 408-414.View ArticleGoogle Scholar
- Koo WWK, Hockman EM, Hammami M: Dual energy X ray absorptiometry measurements in small subjects: conditions affecting clinical measurements. J Amer Coll Nutr. 2004, 23: 212-219.View ArticleGoogle Scholar
- Gluer CC, Blake G, Lu Y, Blunt BA, Jergas M, Genant HK: Accurate assessment of precision errors: how to measure the reproducibility of bone densitometry techniques. Osteoporos Int. 1995, 5: 262-270. 10.1007/BF01774016.View ArticlePubMedGoogle Scholar
- Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, Flegal KM, Mei Z, Wei R, Curtin LR, Roche AF, Johnson CL: 2000 CDC growth charts for the United States: methods and development. Vital Health Stat 11. 2002, 246: 1-190.PubMedGoogle Scholar
- Koo WWK, Bush AJ, Walters J, Carlson SE: Postnatal development of bone mineral status during infancy. J Amer Coll Nutr. 1998, 17: 65-70.View ArticleGoogle Scholar
- Koo WWK, Walters JC, Hockman EM: Body composition in human infants at birth and postnatally. J Nutr. 2000, 130: 2188-2194.PubMedGoogle Scholar
- Hammami M, Koo WWK, Hockman EM: Body composition of neonates from fan beam dual energy X-ray absorptiometry measurement. JPEN. 2003, 27: 423-426.View ArticleGoogle Scholar
- Hassink SG, de Lancey E, Sheslow DV, Smith-Kirwin SM, O'Connor DM, Considine RV, Opentanova I, Dostal K, Spear ML, Leef K, Ash M, Spitzer AR, Funanage VL: Placental leptin: an important new growth factor in intrauterine and neonatal development?. Pediatrics. 1997, 100: E1-10.1542/peds.100.1.e1.View ArticlePubMedGoogle Scholar
- Van Vliet G, Polar M: Thyroid disorders in infancy. Pediatric Endocrinology. Edited by: Lifshitz F. 2007, New York, NY: Informa Healthcare USA, Inc, 391-404.Google Scholar
- Harigaya A, Onigata K, Nako Y, Nagashima K, Morikawa A: Role of serum leptin in the regulation of weight gain in early infancy. Biol Neonate. 1999, 75: 234-238. 10.1159/000014100.View ArticlePubMedGoogle Scholar
- Toprak D, Gokalp AS, Hatun S, Zengin E, Arisov AE, Yumuk Z: Serum leptin levels of premature and full-term newborns in early infancy: metabolic catch-up of premature babies. Turk J Pediatr. 2004, 46: 232-238.PubMedGoogle Scholar
- Wang J, Liu R, Hawkins M, Barzilai N, Rossetti L: A nutrient-sensing pathway regulates leptin gene expression in muscle and fat. Nature. 1998, 393: 684-688. 10.1038/31474.View ArticlePubMedGoogle Scholar
- Kishida Y, Hirao M, Tamai N, Nampei A, Fujimoto T, Nakase T, Shimizu N, Yoshikawa H, Myoui A: Leptin regulates chondrocyte differentiation and matrix maturation during endochondral ossification. Bone. 2005, 37: 607-621. 10.1016/j.bone.2005.05.009.View ArticlePubMedGoogle Scholar
- Reseland JE, Syversen U, Bakke I, Ovigstad G, Eide LG, Hiertner O, Gordeladze JO, Drevon CA: Leptin is expressed in and secreted from primary cultures of human osteoblasts and promotes bone mineralization. J Bone Miner Res. 2001, 16: 1426-1433. 10.1359/jbmr.2001.16.8.1426.View ArticlePubMedGoogle Scholar
- Kratzsch J, Lammert A, Bottner A, Seidel B, Mueller G, Thiery J, Hebebrand J, Kiess W: Circulating soluble leptin receptor and free leptin index during childhood, puberty, and adolescence. J Clin Endocrinol Metab. 2002, 87: 4587-4594. 10.1210/jc.2002-020001.View ArticlePubMedGoogle Scholar
- Ruhl CE, Everhart JE: Leptin concentrations in the United States: relations with demographic and anthropometric measures. Am J Clin Nutr. 2001, 74: 277-278.Google Scholar
- Vatten LJ, Nilsen ST, Odegard RA, Romundstad PR, Austgulen R: Insulin-like growth factor I and leptin in umbilical cord plasma and infant birth size at term. Pediatrics. 2002, 109: 1131-1135. 10.1542/peds.109.6.1131.View ArticlePubMedGoogle Scholar
- Lammert A, Kiess W, Bottner A, Glasow A, Kratzsch J: Soluble leptin receptor represents the main leptin binding activity in human blood. Biochem Biophys Res Commun. 2001, 283: 982-988. 10.1006/bbrc.2001.4885.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.