- Research article
- Open Access
Neuroanatomical organization of gonadotropin-releasing hormone neurons during the oestrus cycle in the ewe
© Batailler et al; licensee BioMed Central Ltd. 2004
- Received: 11 June 2004
- Accepted: 22 November 2004
- Published: 22 November 2004
During the preovulatory surge of gonadotropin-releasing hormone (GnRH), a very large amount of the peptide is released in the hypothalamo-hypophyseal portal blood for 24-36H00. To study whether this release is linked to a modification of the morphological organization of the GnRH-containing neurons, i.e. morphological plasticity, we conducted experiments in intact ewes at 4 different times of the oestrous cycle (before the expected LH surge, during the LH surge, and on day 8 and day 15 of the subsequent luteal phase). The cycle stage was verified by determination of progesterone and LH concentrations in the peripheral blood samples collected prior to euthanasia.
The distribution of GnRH-containing neurons throughout the preoptic area around the vascular organ of the lamina terminalis was studied following visualisation using immunohistochemistry. No difference was observed in the staining intensity for GnRH between the different groups. Clusters of GnRH-containing neurons (defined as 2 or more neurons being observed in close contact) were more numerous during the late follicular phase (43 ± 7) than during the luteal phase (25 ± 6), and the percentage of clusters was higher during the beginning of the follicular phase than during the luteal phase. There was no difference in the number of labelled neurons in each group.
These results indicate that the morphological organization of the GnRH-containing neurons in ewes is modified during the follicular phase. This transitory re-organization may contribute to the putative synchronization of these neurons during the surge. The molecular signal inducing this plasticity has not yet been identified, but oestradiol might play an important role, since in sheep it is the only signal which initiates the GnRH preovulatory surge.
- Preoptic Area
- GnRH Neuron
- Oestrous Cycle
- GnRH Secretion
- Late Follicular Phase
Gonadotropin-releasing hormone (GnRH), the peptide responsible for the regulation of secretory activity of the pituitary gonadotropes, is found in a diffuse neuronal system situated in the preoptic area and anterior hypothalamic area . The neurons project their terminals to the median eminence where the peptide is released into the hypothalamo-hypophyseal portal vessels. The secretion of GnRH is normally pulsatile, as demonstrated by the measurement of GnRH following the direct sampling of hypothalamo-pituitary portal blood in various species such as sheep [2, 3]. The frequency of this pulsatility is the main characteristic of this secretion and it encodes each part of the sexual cycle in the female . The highest circulating concentration of GnRH is induced during the follicular phase in the female by the sequential action of the two main ovarian steroids (progesterone and oestradiol). These act within the brain to trigger a large and sustained period of GnRH release, the GnRH surge, which stimulates the preovulatory LH surge, and subsequently ovulation.
The pulsatile nature of GnRH secretion, as well as the large amount of peptide released during the preovulatory surge, indicate that the activity of GnRH-containing neurons can be synchronized, and it has been demonstrated that the GnRH neurons of the preoptic area receive synaptic contacts from GnRH neurons [5–7]. Oestradiol, which initiates the preovulatory surge of GnRH, has been shown to be involved in the synaptic plasticity demonstrated in various neuronal populations, including those in the arcuate nucleus [8, 9]. Moreover, in rats, an increase in the expression of the molecular markers of synaptic remodelling is observed at the time of the surge in the preoptic area where most of the GnRH neuron cell bodies are found . In sheep, changes in the expression of the polysialylated form of the neural cell adhesion molecule around the periphery of the GnRH cell bodies, associated with the seasonal changes in GnRH secretion, are indicative of potential plastic changes in this neural system linked with changes in secretory activity . In addition, oestradiol and progesterone have been shown to modify the morphology and staining intensity of GnRH neurons within the ovine hypothalamus , suggesting that ovarian steroids could elicit plastic changes in GnRH neurons during the oestrous cycle. However, very little work has been done regarding neuronal organization at the level of the GnRH cell bodies during the oestrous cycle in the ewe.
The aim of this work was to study the number, distribution and potential contacts present between GnRH-containing cells in intact ewes under natural conditions, focusing particularly on the neuroanatomical organization of these neurons around the preovulatory surge when a high level of oestradiol is released by the growing follicles.
Four of the eight animals in experiment 2 were killed during the LH surge (2 during the ascending and 2 during the descending part). The 4 remaining animals were killed 4–6 hours after the end of the LH surge and were discarded from the study.
The percentage of clusters was higher in group 1 (15.1 ± 1.2) than in groups 2 (11.8 ± 1.0), 3 (8.3 ± 0.4) and 4 (10.7 ± 0.9) (fig 3C).
Mean number of isolated or clustered GnRH neurons counted during and after the preovulatory LH surge. Individual values are indicated for each animal during the first (Exp I) and second experiment (Exp II).
Ascending phase of the surge
Descending phase of the surge
Ewe identification n°
Single GnRH neurons
Clusters of GnRH neurons
Our observations indicate that the organization of GnRH-ir neurons was modified during the oestrous cycle, since more clusters were observed during the follicular phase. However, we did not observe any difference between the ascending and descending phases of LH release in group 2. Therefore, our data support the hypothesis that there are changes in communication between GnRH cell bodies prior to the initiation of the preovulatory GnRH surge.
The increase in the number of clusters was not mirrored by a parallel change in the total number of GnRH cell bodies. However, as the number of clusters was relatively small compared to the GnRH cell population (constituting 10% to 15%), the putative increase in the number of GnRH-ir neurons could have been masked due to the large variations in the number of labelled neurons between animals. An alternative hypothesis is that the number of GnRH neurons remains constant, and that the variations in the number of clusters is linked to variations in the peptide level, which are consistent with the variations in the staining intensity observed in all animals. In this case, the level of GnRH would be higher in clustered neurons than in isolated neurons at the moment of the preovulatory surge, and therefore neuronal communication would be more efficient in clustered neurons. The percentage of clusters increased only in group 1, while the difference between the number of clusters in groups 2, 3 and 4 was not statistically significant. This special morphological organization may constitute an anatomical support for the synchronization of GnRH neuronal activity needed to induce the GnRH preovulatory surge.
Clusters of GnRH-ir cell bodies have been previously described in sheep  (Lehman et al, 1986), rats and monkeys . In rats they represented only 2–7% of the GnRH neuronal population . In our study, the proportion of clusters was similar (10–15%) to that observed in monkeys (3–15%). However, contrary to our observation, the numbers did not change with the different hormonal conditions in the monkey . This might be related to the distribution of GnRH cell bodies which differs between monkeys and sheep , and it is possible that the increase of clusters observed in our experiments was related to a specific GnRH neuronal population involved in the GnRH surge secretion. Indeed, Fos expression at the time of the preovulatory LH surge has been shown to be expressed in a subset of GnRH neurons in rats .
A neuronal re-organization of GnRH-containing neurons has also been previously described in anoestrous ewes where the neurons are ensheathed by glial processes which decrease the number of axo-somatic synaptic contacts . In this latter situation, photoperiod would be the major signal to induce such a modification. Another example of the relationship between synaptic afferents and GnRH secretion has been described in the monkey where the increase of GnRH activity at puberty correlates with the decrease in synaptic afferents on GnRH-containing neurons . This morphological regulation of GnRH-containing neurons, through synaptic contacts or perikaryal apposition, involves glial processes, since in both sheep and monkeys these neurons are ensheathed by numerous glial processes [19, 21].
Steroids, mainly oestradiol, are the most potent regulators in the control of GnRH neuronal activity. Therefore, we may hypothesize that this "plasticity" results from an oestradiol effect, as has been shown in the monkey where oestradiol modulates the number of synaptic contacts on GnRH neurons . In rats and monkeys, numerous studies have demonstrated that oestrogens induce synaptic plasticity in the control of gonadotropin secretion (rat: ; monkeys: ). In our study, the highest number of clusters was found during the preovulatory surge which occurred about 24 hours after the expected oestradiol increase, this delay being consistent with an oestradiol effect.
There was no difference in the number of labelled neurons in the experimental groups, and this observation may indicate that the level of GnRH-immunoreactivity was stable in the neurons of the preoptic area. It has previously been demonstrated in the ewe that GnRH messenger ribonucleic acid expression changes before the onset of the oestradiol-induced luteinizing hormone surge , and that the staining intensity of GnRH-ir perikarya increases after oestradiol treatment, but that the number of labelled neurons does not change [12, 26]. In addition, the variations of LH secretion induced by progesterone withdrawal are not linked to a variation in GnRH mRNA . Unlike the apparent stability of the GnRH level in the perikarya, GnRH-immunoreactivity decreases in the median eminence after ovulation . This information could indicate that variations in GnRH secretion arise from its release from the median eminence terminals without variations in peptides in the perikarya.
In conclusion, we have demonstrated that the morphological organization of the GnRH neurons of the preoptic area is modified during the oestrous cycle, although the overall number of GnRH-ir neurons does not change. During the preovulatory surge, the GnRH neurons are more frequently found in clusters, and the percentage of clusters is significantly higher immediately prior to the preovulatory LH surge. Actual contacts between GnRH-ir neurons cannot be determined using light microscopy, but can only be demonstrated using electron microscopy. Because oestradiol is the most powerful regulator of GnRH activity during the oestrous cycle, we can hypothesize that this plasticity may be induced by steroids. The role of this plasticity remains to be demonstrated, but it could increase interneuronal communication during the preovulatory surge.
All experimental procedures were carried out in accordance with authorisation A37801 of the French Ministry of Agriculture.
Intact adult Ile de France ewes (n = 28) from the laboratory flock, maintained under natural photoperiod were used. Two experiments were performed (one year apart) during the breeding season (between November 15th and December 15th). Animals were fed daily with hay, straw and corn, and water was available ad libitum. All of the ewes had lambed at least once.
Experiment N° 1
Animals (n = 20) were killed at specific time points in the oestrous cycle. In order to synchronize oestrous cycles, animals were treated with an intravaginal progesterone-releasing device (CIDR InterAg, Hamilton, New Zealand) for 14 days. Following removal of the progesterone device, animals were treated (i.m.) with 500 UI of Pregnant Mare Serum Gonadotropin (PMSG), a treatment known to induce the LH surge and ovulation in sheep (for review see ). It has been shown that oestrous behaviour in this breed precedes the LH surge by 7.0 ± 1.6 hours . Therefore, oestrous behaviour was recorded for each ewe every 6 hours, from 18 hours after PMSG administration until the animal was observed to be in oestrus using a ram wearing an apron. The animals were then allocated to four groups:
Group 1: Animals killed 1–2 hours after oestrus, i.e. before the LH surge onset, (n = 5)
Group 2: Animals killed 8–12 hours after oestrus, i.e. during the preovulatory LH surge, (n = 5)
Group 3: Animals killed 8 days after the preovulatory LH surge, (n = 5)
Group 4: Animals killed 15 days after the preovulatory LH surge, (n = 5)
Experiment N° 2
Analysis of LH secretion (see paragraph below) revealed that animals in group 2 in experiment 1 were killed either in the ascending (n = 2) or descending (n = 2) phase of the LH surge. Therefore, in order to increase the N for each sub-group, 8 animals were synchronised as previously described (one year later, during the breeding season) and killed 12 and 20 hours (n = 4) after oestrus.
For both experiments, blood samples were collected by venepuncture every 2 hours from 18 hours after PMSG administration until sacrifice (groups 1 and 2) and daily thereafter (groups 3 and 4).
Plasma samples were assayed in duplicate for LH with 100 μl aliquots of plasma using a previously described RIA method . All samples from one experiment were run in a single assay. Intra-assay coefficient of variation averaged 9% and assay sensitivity was 0.16 ± 0.05 ng/ml (4 assays) of standard 1051-CY-LH (i.e. 0.31 ng/ml of NIH-LH-S1). Progesterone was determined in a single assay after hexane extraction of 100 μl of plasma . The sensitivity was 5 pg/tube and the intra-assay CV 10%.
A solution of heparin (25,000 units) was injected iv 10 min before decapitation. Both carotids were catheterised and the head was perfused with 2 litres of 1% sodium nitrite followed by 4 litres of 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). After perfusion, the brain was quickly removed and a block containing the whole preoptic area was isolated and post-fixed for 2 hours at +4°C in the same fixative. The blocks were immersed in 30% sucrose containing 0.1% sodium azide at +4°C for 5 days.
Serial 40 μm coronal sections were obtained using a freezing microtome and stored at +4°C in PBS containing 0.1% sodium azide. Every fifth section was stained for GnRH immunoreactivity.
Sections were incubated for 2 hours at +4°C in PBS containing 0.3% triton X100 and 1% H2O2 and then preincubated for 30 min at room temperature in blocking solution (PBS, 6% normal sheep serum, 0.3% triton, 0.1% sodium azide). Sections were exposed to GnRH primary antiserum at a dilution of 1/10,000 in PBS containing 0.3% triton, 0.2% HSA, 0.1% sodium azide for 4 days at +4°C. After washing in PBS, sections were incubated for 3 hours at room temperature in the secondary antibody (1/500 with PBS containing 0.2% BSA). After washing in PBS, sections were incubated in rabbit peroxidase-antiperoxidase complex diluted 1/10,000 in PBS containing 0.2% BSA overnight at +4°C. After washing twice in PBS and in Tris buffer (0.05 M pH 7.6), the sections were reacted with 3,3-diaminobenzidine tetrahydrochloride (0.02%) and nickel ammonium sulfate (0.25%) in the same Tris buffer containing 0.0025% H2O2 for 20 min. The reactions were stopped by rinsing the sections with Tris buffer. Sections were mounted onto gelatine-coated slides, dried, counterstained with a solution of neutral red for 5 min and mounted with a cover-slide using Depex.
The characteristics and the specificity of the GnRH antibody were as previously described .
The LH surge was assumed to start when LH concentration exceeded 6 ng/ml of plasma, i.e. about twice the maximum value of a pulse during the follicular phase for this breed and for this LH assay. For hormone values, a global comparison was carried out using a non-parametric ANOVA with exact general score test, and comparison between groups was carried out using the non-parametric exact permutation test for independent samples (Stat Exact Software, Cytel Software Corporation, Cambridge, MA, USA).
The first observation of the sections allowed the immunoreactive area surrounding the OVLT to be determined (fig. 2). We selected two sections where the size of the OVLT was maximum in the third ventricle, and the two rostral and caudal sections were studied. A total of six sections at 160 μm intervals were studied for each ewe. In each section, all immunoreactive neurons for GnRH were counted using a light microscope, and groups of closely-associated neurons were noted. Total number of cells and number of clusters were analyzed by a 2-way (group and section number) ANOVA (SuperANOVA, Abacus concept, California) followed by a student-Newman-keuls post-hoc test for two by two comparisons. The percentage of cells forming clusters was calculated for each animal. It was analyzed by a one-way (group) ANOVA after arcsin transformation followed by a student-Newman-keuls post-hoc test for two by two comparisons. The results are expressed as means ± standard error of the means.
The authors wish to thank Francis Dupont and the shepherds for animal care, and the staff of the RIA laboratory in Nouzilly for performing the LH and progesterone assays.
- Barry J, Hoffman GE, Wray S: LHRH-containing systems. In Handbook of Chemical Neuroanatomy 4, GABA and Neuropeptides in the CNS, Part I. Edited by: Björklund A, Hökfelt T. 1985, Amsterdam; New-York, Oxford: Elsevier, 166-215.Google Scholar
- Clarke IJ, Cummins JT: The temporal relationship between gonadotropin releasing hormone (GnRH) and luteinizing hormone (LH) secretion in ovariectomized ewes. Endocrinology. 1982, 111 (5): 1737-1739.View ArticlePubMedGoogle Scholar
- Caraty A, Locatelli A: Effect of time after castration on secretion of LHRH and LH in the ram. J Reprod Fertil. 1988, 82 (1): 263-269.View ArticlePubMedGoogle Scholar
- Moenter SM, Caraty A, Locatelli A, Karsch FJ: Pattern of gonadotropin-releasing hormone (GnRH) secretion leading up to ovulation in the ewe: existence of a preovulatory GnRH surge. Endocrinology. 1991, 129: 1175-1182.View ArticlePubMedGoogle Scholar
- Witkin JW, Silverman AJ: Synaptology of luteinizing hormone-releasing hormone neurons in rat preoptic area. Peptides. 1985, 6 (2): 263-271. 10.1016/0196-9781(85)90050-6.View ArticlePubMedGoogle Scholar
- Leranth C, Segura LM, Palkovits M, MacLusky NJ, Shanabrough M, Naftolin F: The LH-RH-containing neuronal network in the preoptic area of the rat: demonstration of LH-RH-containing nerve terminals in synaptic contact with LH-RH neurons. Brain Res. 1985, 345 (2): 332-336. 10.1016/0006-8993(85)91011-X.View ArticlePubMedGoogle Scholar
- Thind KK, Goldsmith PC: Infundibular gonadotropin-releasing hormone neurons are inhibited by direct opioid and autoregulatory synapses in juvenile monkeys. Neuroendocrinology. 1988, 47 (3): 203-216.View ArticlePubMedGoogle Scholar
- Garcia-Segura LM, Hernandez P, Olmos G, Tranque PA, Naftolin F: Neuronal membrane remodelling during the oestrus cycle: a freeze-fracture study in the arcuate nucleus of the rat hypothalamus. J Neurocytol. 1988, 17 (3): 377-383.View ArticlePubMedGoogle Scholar
- Olmos G, Naftolin F, Perez J, Tranque PA, Garcia-Segura LM: Synaptic remodeling in the rat arcuate nucleus during the estrous cycle. Neuroscience. 1989, 32 (3): 663-667. 10.1016/0306-4522(89)90288-1.View ArticlePubMedGoogle Scholar
- Kaur G, Heera PK, Srivastava LK: Neuroendocrine plasticity in GnRH release during rat estrous cycle: correlation with molecular markers of synaptic remodeling. Brain Res. 2002, 954 (1): 21-31. 10.1016/S0006-8993(02)03301-2.View ArticlePubMedGoogle Scholar
- Viguié C, Jansen HT, Glass JD, Watanabe M, Billings HJ, Coolen L, Lehman MN, Karsch FJ: Potential for polysialylated form of neural cell adhesion molecule-mediated neuroplasticity within the gonadotropin-releasing hormone neurosecretory system of the ewe. Endocrinology. 2001, 142 (3): 1317-1324. 10.1210/en.142.3.1317.PubMedGoogle Scholar
- Glass JD, Mastran T, Nett TM: Effect of estradiol and progesterone on the gonadotropin-Releasing Hormone (GnRH)-immunoreactive neuronal system of the anestrous ewe. Brain Res. 1986, 381: 336-344. 10.1016/0006-8993(86)90085-5.View ArticlePubMedGoogle Scholar
- Cahill LP, Saumande J, Ravault JP, Blanc M, Thimonier J, Mariana JC, Mauléon P: Hormonal and follicular relationships in ewes of high and low ovulation rates. J Reprod Fertil. 1981, 62 (1): 141-50.View ArticlePubMedGoogle Scholar
- Caldani M, Batailler M, Thiery JC, Dubois MP: LHRH-immunoreactive structures in the sheep brain. Histochemistry. 1988, 89 (2): 129-139. 10.1007/BF00489916.View ArticlePubMedGoogle Scholar
- Lehman MN, Robinson JE, Karsch FJ, Silverman AJ: Immunocytochemical localization of luteinizing hormone-releasing hormone (LHRH) pathways in the sheep brain during anestrus and the mid-luteal phase of the estrous cycle. J Comp Neurol. 1986, 244 (1): 19-35.View ArticlePubMedGoogle Scholar
- Witkin JW, O'Sullivan H, Silverman AJ: Novel associations among gonadotropin-releasing hormone neurons. Endocrinology. 1995, 136: 4323-4330. 10.1210/en.136.10.4323.PubMedGoogle Scholar
- Caldani M, Caraty A, Pelletier J, Thiery JC, Tillet Y: LH pulsatile release and its control. In Reproduction in mammals and man Chap 4. Edited by: Thibault C, Levasseur MC, Hunter RHF. 1993, Paris: Ellipses, 79-96.Google Scholar
- Rubin BS, Lee CE, King JC: A reduced proportion of luteinizing hormone (LH)-releasing hormone neurones express Fos protein during the preovulatory or steroid-induced LH surge in middle-aged rats. Biol Reprod. 1994, 51: 1264-1272.View ArticlePubMedGoogle Scholar
- Lehman MN, Karsch FJ, Robinson JE, Silverman AJ: Ultrastructure and synaptic organization of luteinizing hormone-releasing hormone (LHRH) neurons in the anestrous ewe. J Comp Neurol. 1988, 273 (4): 447-458.View ArticlePubMedGoogle Scholar
- Perera AD, Plant TM: Ultrastructural studies of neuronal correlates of the pubertal reaugmentation of hypothalamic gonadotropin-releasing hormone (GnRH) release in the rhesus monkey (Macaca mulatta). J Comp Neurol. 1997, 385 (1): 71-82. 10.1002/(SICI)1096-9861(19970818)385:1<71::AID-CNE4>3.0.CO;2-9.View ArticlePubMedGoogle Scholar
- Witkin JW, O'Sullivan H, Miller R, Ferin M: GnRH perikarya in medial basal hypothalamus of pubertal female rhesus macaque are ensheathed with glia. J Neuroendocrinol. 1997, 9 (12): 881-885. 10.1046/j.1365-2826.1997.00649.x.View ArticlePubMedGoogle Scholar
- Zsarnovszky A, Horvath TL, Garcia-Segura LM, Horvath B, Naftolin F: Oestrogen-induced in the synaptology of the monkey (Cercopithecus aethiops) arcuate nucleus during gonodotropin feedback. J Neuroendocrino. 2001, 13: 22-28. 10.1046/j.1365-2826.2001.00602.x.View ArticleGoogle Scholar
- Horvath TL, Garcia-Segura LM, Naftolin F: Lack of gonadotropin-positive feedback in the male rat is associated with lack of estrogen-induced synaptic plasticity in the arcuate nucleus. Neuroendocrinology. 1997, 65 (2): 136-140.View ArticlePubMedGoogle Scholar
- Witkin JW, Ferin M, Popilskis SJ, Silverman AJ: Effects of gonadal steroids on the ultrastructure of GnRH neurons in the rhesus monkey: synaptic input and glial apposition. Endocrinology. 1991, 129 (2): 1083-1092.View ArticlePubMedGoogle Scholar
- Harris TG, Robinson JE, Evans NP, Skinner DC, Herbison AE: Gonadotropin-releasing hormone messenger ribonucleic acid expression changes before the onset of the estradiol-induced luteinizing hormone surge in the ewe. Endocrinology. 1998, 139 (1): 57-64. 10.1210/en.139.1.57.PubMedGoogle Scholar
- Moenter SM, Karsch FJ, Lehman MN: Fos expression during the oestradiol-induced gonadotropin releasing hormone (GnRH) surge of the ewe: Induction in GnRH and others neurons. Endocrinology. 1993, 133: 896-903. 10.1210/en.133.2.896.PubMedGoogle Scholar
- Robinson JE, Healey AE, Harris TG, Messent EA, Skinner DC, Taylor JA, Evans NP: The negative feedback action of progesterone on luteinizing hormone release is not associated with changes in GnRH mRNA expression in the ewe. J Neuroendocrinol. 2000, 12: 121-129. 10.1046/j.1365-2826.2000.00426.x.View ArticlePubMedGoogle Scholar
- Polkowska J, Dubois MP, Domanski E: Immunocytochemistry of luteinizing hormone releasing hormone (LHRH) in the sheep hypothalamus during variuos reproductive stages: correlation with the gonadotropic hormones of the pituitary. Cell Tissue Res. 1980, 208 (2): 327-341. 10.1007/BF00234880.View ArticlePubMedGoogle Scholar
- Thimonier J: Control of seasonal reproduction in sheep and goats by light and hormones. J Reprod Fert. 1981, Suppl 30: 33-45.Google Scholar
- Bindon BM, Blanc MR, Pelletier J, Terqui M, Thimonier J: Periovulatory gonadotrophin and ovarian steroid patterns in sheep of breeds with differing fecundity. J Reprod Fertil. 1979, 55 (1): 15-25.View ArticlePubMedGoogle Scholar
- Montgomery GW, Martin GB, Pelletier J: Changes in pulsatile LH secretion after ovariectomy in Ile-de France ewes in two seasons. J Reprod Fertil. 1985, 73: 173-183.View ArticlePubMedGoogle Scholar
- Saumande J, Tamboura D, Chupin D: Changes in milk and plasma concentration of progesterone of cow after treatment to induce super ovulation and their relationship with number of ovulation and embryos collected. Theriogenology. 1985, 23: 719-731. 10.1016/0093-691X(85)90147-5.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.