| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
INSERM, U-549 (V.T., M.T.B.P., C.F., R.D., P.Z., J.E.), and Clinique des Maladies Mentales et de lEncéphale (C.F., R.D.), 75014 Paris, France; and Services dEndocrinologie et de Psychiatrie (M.K., D.F., C.B., C.M., F.L., B.E.), Hôpital de Bellevue, 42100 Saint Etienne, France
Address all correspondence and requests for reprints to: Dr. Jacques Epelbaum, INSERM, U-549, IFR Broca-Sainte Anne, 2 ter rue dAlésia, 75014 Paris, France. E-mail: epelbaum{at}broca.inserm.fr.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
The neuroendocrine mechanisms regulating GH secretion are strongly dependent on nutritional state. In humans, fasting stimulates GH release, and AN patients show elevated basal GH in presence of low IGF-I levels (9). However, the mechanisms subserving these changes remain unclear. No data concerning the somatotropic axis in constitutional thinness have been reported.
Ghrelin, an endogenous ligand for the GH secretagogue receptor (GHS-R) (10), has recently been identified in rat stomach as a 27- or 28-amino acid peptide in which the serine 3 residue is N-octanoylated (11, 12). In addition to its GH-releasing properties, exogenous ghrelin injections exert potent orexigenic effects in rodents (13, 14, 15, 16) and humans (17, 18). Both GH-releasing and orexigenic effects are likely to be mediated through intrahypothalamic GHS-R (19). In humans (20) and rats (21), endogenous plasma ghrelin levels are directly related to food intake episodes, and the stomach is the major source of circulating ghrelin (12). Moreover, preliminary data in a single sampling experiment suggested that plasma ghrelin levels are markedly elevated in AN patients and are negatively correlated with BMIs (22).
In the present work we assessed plasma ghrelin levels in AN patients before and after renutrition and in constitutionally thin subjects without feeding behavior disturbance and compared them with levels in control subjects. The relationships between plasma ghrelin levels and other neuroendocrine and nutritional status parameters such as GH, leptin, T3, and cortisol were also investigated.
| Subjects and Methods |
|---|
|
|
|---|
Subjects
Four groups of women (1536 yr old) participated in the study. Nine women were diagnosed with AN. All patients met the criteria of the Diagnostic and Statistical Manual of Mental Disorders (22A ). They took no medication or estroprogestative treatment and did not present bingeing or vomiting behavior. Secondary amenorrhea was a constant feature for at least 6 months. The same patients were followed during free refeeding. Blood sampling was performed at the initiation of this study and after a minimum weight gain of 10% according to initial weight, without preestablished delay (time range of the refeeding period, 215 months). Seven constitutionally thin subjects presented with BMI similar to the AN group before renutrition. Due to the rarity of these subjects, blood sampling was performed at different times in the menstrual cycle (2 in the early phase, 1 during the ovulatory peak, and 4 in the late phase of the cycle). These subjects had no body image disturbance or psychological disorders and were not amenorrheic. The control subjects were 10 age-matched cycling healthy women of normal weight without estroprogestative treatment. Blood sampling in these subjects was performed in the early phase of the cycle. None of the women included in the study was taking any medication.
Experimental protocols
Body composition was evaluated by electrical impedance biometry with an ANALICOR 2 (Eugedia, Chambly, France). The patients rested in the supine position for at least 15 min. Two electrode needles were placed on clean and degreased skin at the limb ends. Two frequencies (50 and 100 kHz) were used at a current of 400 µA. Fat mass was recorded as a percentage of body weight as a relative value. For statistical analysis, relative values were computed using normal theoretical parameters.
Peripheral hormones (free T3, IGF-I, and 17ß-estradiol) were measured at 0800 h after an overnight fast. Hormone measurements were performed in controls in the early phase of the menstrual cycle and in constitutionally thin subjects in a later phase.
For evaluation of circadian variations in ghrelin, leptin, cortisol, and GH, blood samples were collected on EDTA at 0800, 1200, 1600, 2000, 2400, and 0400 h. Meals were taken immediately after the 0800 and 1200 h samplings and at 1900 h, before the 2000 h sampling.
For the dynamics of GHRH-induced GH release, an indwelling catheter was placed in an antecubital vein of the forearm, and blood samples were collected 20 and 0 min before and 15, 30, 60, 90, and 120 min after GHRH (Ferring Pharmaceuticals Ltd., Gentilly, France; 100 µg, iv) injection. Blood samples were immediately centrifuged after collection, and plasma was stored at -80 C before assays.
Assay methods
Plasma GH levels were measured by immunoradiometric assay (Immunotech, Beckman Coulter, Inc., Marseilles France; intra- and interassay coefficients of variations, 2% and 14%, respectively; detection limit, 0.1 mU/liter; manufacturers reference level, <5 mU/liter). Plasma IGF-I levels were measured by immunoradiometric assay (Immunotech; intra- and interassay coefficients of variations, 9% and 16%, respectively; detection limit, 12 µg/liter; manufacturers reference range, 107310 µg/liter). Plasma ghrelin levels were determined by RIA (Phoenix Pharmaceuticals, Inc., Belmont, CA; intra- and interassay coefficients of variations, <10%; detection limit, 53 ng/liter). Plasma leptin levels were measured by RIA [Nichols Institute Diagnostics, San Juan Capistrano, CA; intra- and interassay coefficients of variations, <5%; detection limit, 0.5 µg/liter; manufacturers reference range for normal BMI (1825), 3.711.1 µg/liter]. Plasma cortisol levels were measured by RIA (Immunotech; intra- and interassay coefficients of variations, 6% and 8%, respectively; detection limit, 13 nmol/liter; normal range at 0800 h, 263724 nmol/liter). Free T3 was measured by RIA (Beckman; intra- and interassay coefficients of variations, 6% and 5%, respectively; detection limit, 0.5 pmol/liter; manufacturers reference range, 2.55.8 pmol/liter). 17ß-Estradiol levels were determined by RIA (DiaSorin, Inc., Antony, France; intra- and interassay coefficients of variations, 6% and 7%, respectively; detection limit, 5 ng/liter).
Statistical analysis
Data are the mean ± SEM, and statistical analyses were performed by ANOVA and paired t test using StatView 4.5 software (Abacus Concepts, Inc., Palo Alto, CA).
| Results |
|---|
|
|
|---|
Clinical and hormonal parameters are summarized in Table 1
. In patients with AN, body weight and BMI were strongly reduced compared with healthy controls in the same age range. This denutrition concerned mainly body fat mass, which represented only 10.7% of the total body mass. Morning fasting IGF-I, 17ß-estradiol, and free T3 levels were decreased. After renutrition, body weight, BMI, and percentage of fat mass only partially recovered, whereas peripheral hormone levels returned to control values.
|
Plasma ghrelin-like immunoreactivity levels in controls, AN patients, AN patients after renutrition, and constitutionally thin subjects
The mean morning fasting plasma ghrelin level was 208 ± 21 ng/liter in young control women (n = 9) and 270 ± 45 ng/liter in constitutionally thin subjects (n = 5). It reached 491 ± 68 ng/liter (n = 8) in AN patients and returned to 269 ± 24 ng/liter (n = 8) in AN patients after renutrition. Individual values are illustrated in Fig. 1
. The mean decrease in ghrelin levels in AN patients after renutrition reached 39 ± 8%.
|
In control subjects plasma ghrelin levels, sampled every 4 h, displayed only small variations across the nychthemere (Fig. 2
). In contrast, a marked circadian variation in ghrelin was observed in AN patients, with higher plasma levels at night (by paired t test: P < 0.05, 2000 vs. 0400 h). These patients exhibited increased plasma ghrelin levels compared with controls (P < 0.001), and levels returned to control values after renutrition and partial weight recovery. Constitutionally thin subjects displayed less regular and intermediate ghrelin levels, significantly different from levels in controls (P < 0.001), AN patients (P < 0.001), and AN patients after renutrition (P < 0.05).
|
Plasma GH levels were markedly elevated in AN patients compared with controls (mean 24-h concentrations, 23.5 ± 14.1 vs. 4.7 ± 1.3 µg/liter; P < 0.001). GH levels were not significantly different from controls in constitutionally thin subjects (7.6 ± 1.7 µg/liter) or AN patients after renutrition (5.8 ± 1.2 µg/liter).
Plasma cortisol levels were significantly increased in AN patients compared with controls, AN patients after renutrition, and constitutionally thin subjects (P < 0.001). However, the circadian rhythmicity of plasma cortisol was maintained in AN patients, with higher levels at 0800 h and a decline until 2400 h (by paired t test: P < 0.0001).
Correlations between circulating ghrelin and BMI, GH, leptin, cortisol, and T3
These correlations are illustrated in Fig. 3
. Controls, constitutionally thin subjects, and AN patients before renutrition (group A) were compared with controls, constitutionally thin subjects, and AN patients after renutrition (group B). In group A the mean plasma ghrelin levels were negatively correlated with BMI (r = -0.596; P < 0.001), leptin (r = -0.410; P < 0.05), and free T3 (r = -0.610; P < 0.001). After partial weight recovery (group B), the negative correlation between ghrelin and leptin was no longer significant, whereas ghrelin and BMI (r = -0.395; P < 0.05) and ghrelin and free T3 (r = -0.482; P < 0.05) were still correlated. Correlations were not significant between ghrelin and cortisol or between ghrelin and GH in both groups.
|
The amplitude of GHRH-induced GH release was 4-fold higher in AN patients than in controls (P < 0.01; Fig. 4
). In all groups, GHRH-induced GH release reached a maximum 3060 min after the injection and returned to basal values after 120 min, except for the AN group (Fig. 4A
, left panel).
|
Mean ghrelin plasma levels were positively correlated to the GH area under the curve (AUC) after GHRH stimulation for controls, constitutionally thin subjects, and AN patients (r = 0.602; P < 0.01; Fig. 4B
, left panel), but not for AN patients after partial weight recovery (r = 0.277; P = NS; Fig. 4B
, right panel). Correlations were not significant between GH and IGF-I AUCs in both groups.
| Discussion |
|---|
|
|
|---|
AN is a severe eating disorder characterized by decreased food intake, important weight loss (BMI, <17.5), and reduced body fat. It is associated with multiple endocrine changes that have been well described: hypercortisolism (23, 24), hypothyroidism (25), alteration of the GH-IGF-I axis with GH hypersecretion coupled with low IGF-I (2, 9), hypoleptinemia (26, 27), and amenorrhea (28). In contrast, constitutionally thin women, whose BMI is equivalent to AN patients, do not exhibit abnormal feeding behavior and caloric deficit, as indicated by the normal level of T3, a well established nutritional factor influenced by caloric restriction (29, 30) or overnutrition (31). Moreover, most endocrine parameters (17ß-estradiol, cortisol, GH, and IGF-I) are not different from controls, which make them distinct from AN patients.
In both conditions of psychogen and hereditary thinness, morning fasting ghrelin concentrations increased compared with those in healthy age-matched women and remained higher across the nychtemere. The 2-fold increase observed in patients with AN and the return to control values after renutrition were consistent with the findings of previous studies (22, 32). The circadian rhythmicity of plasma ghrelin levels previously observed in normal subjects (20) was observed in constitutionally thin women and seemed to be maintained in AN patients, who exhibited maximal ghrelin levels at night.
In AN, leptin levels are considerably decreased along the 24-h sampling period and return to control values after renutrition, as previously shown (26, 27, 33, 34, 35). This hormone is also decreased in constitutionally thin subjects who displayed intermediate values between AN patients and control subjects. Leptin concentrations are directly related to body fat mass (36, 37). Despite an equivalent BMI between AN patients and constitutionally thin subjects, body composition in this later group is equivalent to control subjects. In contrast, in AN patients, fat mass is strongly diminished. The net body mass in constitutionally thin subjects is consequently higher than that in AN women, in keeping with their intermediate plasma leptin concentrations.
Intermediate ghrelin levels in constitutional thinness indicate that circulating ghrelin is also dependent on the fat content, as suggested by the negative correlation between ghrelin and BMI found herein and previously (22, 32). However, despite the close correlation between BMI and circulating leptin or ghrelin, body adiposity is not the only determinant of leptin and ghrelin levels. In a recent study of AN, bulimia nervosa, and binge-eating disorders, Monteleone et al. (38) proposed that factors other than body weight may also play a role in the determination of leptin levels (38). Leptin levels have been significantly correlated with eating behavior scores in patients with distinct BMI and feeding behaviors such as AN and bulimia nervosa (39).
Circulating ghrelin levels are also influenced by changes in energy intake. During short-term fasting, an increase in circulating ghrelin is observed independently of changes in body fat content (15). Furthermore, in the present study ghrelin levels returned to control values in AN subjects after renutrition despite an only partial body weight recovery. These data together with the negative correlation between ghrelin and T3 in individuals exhibiting different nutritional status (i.e. control, AN, and constitutional thinness) as well as in populations with identical nutrition (i.e. control, AN after renutrition, and constitutional thinness) suggest that ghrelin is a good nutritional indicator.
A considerable amount of evidence indicates that ghrelin and leptin exert opposite metabolic actions. Ghrelin elicits orexigenic and adipogenic effects in both rats (13, 14, 15, 40) and humans (17, 41), whereas leptin is anorexigenic and promotes adipolysis (42, 43). These antagonistic effects are likely to be relayed via modulation of neuropeptide Y/agouti-related protein hypothalamic neurons bearing ghrelin and leptin receptors (44, 45). Indeed, although ghrelin activates neuropeptide Y/agouti-related protein neurons (16, 40), leptin inhibits them (46, 47). Furthermore, circulating levels of both hormones are regulated in an opposite manner. In conditions of negative energy balance, such as fasting or AN, leptin is negatively regulated (review in Ref. 48). In contrast, plasma ghrelin levels are reduced and plasma leptin levels are increased in conditions of positive energy balance, such as overfeeding (49) or obesity (50).
If ghrelin behaves as an orexigenic factor, the increase in endogenous ghrelin levels in AN could be considered an adaptive mechanism, promoting energy intake and increasing body fat stores in response to a deficit in energy balance. Normalization of circulating ghrelin after partial weight recovery is consistent with this hypothesis. In this study ghrelin levels were assayed every 4 h to avoid intensive blood sampling in the patients. Thus, we could not observe the dramatic changes in circulating ghrelin before (increase) and after (decrease) each meal in normal subjects (20). Such pre- and postprandial ghrelin levels may be of great relevance in the control of feeding behavior. However, in AN patients, increased plasma ghrelin levels are not associated with increased food intake. Desensitization of GHS-Rs implicated in the control of food intake could explain in part this paradoxical response. In a first study the GH response to hexarelin, a powerful GHS, was lower in AN than in controls, suggesting a desensitization of the GHS-R implicated in the GH response (51), but a second study did not confirm these modifications (52). Thus, further experiments are needed to draw conclusions about ghrelin resistance in AN in terms of both feeding and GH control. The two simplest (and not mutually exclusive) hypotheses that can fit with a desensitization of ghrelin receptors involved in feeding behavior in AN, despite the fact that these patients exhibit increased basal GH secretion and GH responses to GHRH, are 1) that ghrelin receptors involved in these two functions are structurally different; and 2) that they are differentially located in the brain and at the periphery. In both cases ghrelin receptors might be differently affected by agonist-induced desensitization. At any rate, therapeutic treatment with ghrelin seems inappropriate in such a psychological disorder.
GH secretion is influenced by modifications of nutritional status, as evidenced by GH deficit in obesity (53) or GH hypersecretion during fasting (54). AN has been associated with high GH levels (8). Most studies investigating the GH response to GHRH in these patients described high GH release after GHRH stimulation (55, 56, 57). One study (51) did not show differences between AN and controls, and another study showed time-dependent responses (58). In the present study we also observed an increase in basal and stimulated GH release associated with low IGF-I levels in AN women. Circulating IGF-I levels are nutritionally regulated (59) and decrease acutely with caloric deprivation (60). In AN it has been proposed that high GH levels reflect the altered negative feedback via IGF-I (9), because increased circulating IGF-I after renutrition (2) or injection of recombinant human IGF-I (61) are associated with normalization of GH levels. However, IGF-I is not modified in constitutionally thin women and returns to control values in AN patients after renutrition. Thus, IGF-I modifications cannot explain the enhanced GH response to GHRH in these two groups of subjects compared with controls. Thus, another factor has to be involved in the enhancement of GH secretion. As ghrelin can inhibit somatostatin release centrally (21), it may be postulated that high ghrelin levels induce a low somatostatinergic tone that would be responsible for the enhanced GHRH-induced GH response (62, 63).
Plasma cortisol levels were elevated in AN and declined after body weight gain, as previously reported (1, 23). However, cortisol circadian rhythmicity was not altered in AN, with a maximal level at 0800 h and a progressive decrease until midnight in all groups of subjects (23, 34). In contrast with other nychthemeral hormones, cortisol was not modified in constitutional thinness despite an increase in ghrelin levels. These observations suggest that hypercortisolism in AN is related to characteristics other than nutritional status. Experimental hypophagia caused by restraint stress has similarities with AN: both are associated with an increased release of corticotropin-releasing factor (64, 65). Thus, differences between AN and constitutional thinness may reflect differences in emotional status between these two groups (66).
In summary, as previously reported in the rat, endogenous ghrelin in humans is not strictly correlated with basal GH secretion, although it appears to be involved in GHRH-induced secretion. Ghrelin and leptin, two peripheral hormones that signal changes in the energy balance to the central nervous system, are reciprocally modified in AN as well as in constitutionally thin subjects. This later group, being distinct from AN subjects by both feeding behavior and neuroendocrine parameters, deserves further characterization. Finally, ghrelin is not only dependent on body fat mass, but is also directly influenced by feeding behavior and/or energy intake. This is further pointed out by the closer relationship between T3 and ghrelin than between T3 and leptin, suggesting that ghrelin may be a better nutritional indicator.
| Acknowledgments |
|---|
| Footnotes |
|---|
Abbreviations: AN, Anorexia nervosa; AUC, area under the curve; BMI, body mass index; GHS, GH secretagogue; GHS-R, GH secretagogue receptor.
Received April 25, 2002.
Accepted September 19, 2002.
| References |
|---|
|
|
|---|
2-Adrenoceptor sensitivity in anorexia nervosa: GH response to clonidine or GHRH stimulation. Biol Psychiatry 25:256264[CrossRef][Medline]
This article has been cited by other articles:
![]() |
J. H. LIU and A. H. BILL Stress-Associated or Functional Hypothalamic Amenorrhea in the Adolescent Ann. N.Y. Acad. Sci., June 1, 2008; 1135(1): 179 - 184. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Misra, R. Prabhakaran, K. K. Miller, M. A. Goldstein, D. Mickley, L. Clauss, P. Lockhart, J. Cord, D. B. Herzog, D. K. Katzman, et al. Prognostic Indicators of Changes in Bone Density Measures in Adolescent Girls with Anorexia Nervosa-II J. Clin. Endocrinol. Metab., April 1, 2008; 93(4): 1292 - 1297. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Galusca, M. Zouch, N. Germain, C. Bossu, D. Frere, F. Lang, M.-H. Lafage-Proust, T. Thomas, L. Vico, and B. Estour Constitutional Thinness: Unusual Human Phenotype of Low Bone Quality J. Clin. Endocrinol. Metab., January 1, 2008; 93(1): 110 - 117. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Germain, B. Galusca, C. W Le Roux, C. Bossu, M. A Ghatei, F. Lang, S. R Bloom, and B. Estour Constitutional thinness and lean anorexia nervosa display opposite concentrations of peptide YY, glucagon-like peptide 1, ghrelin, and leptin Am. J. Clinical Nutrition, April 1, 2007; 85(4): 967 - 971. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Bossu, B. Galusca, S. Normand, N. Germain, P. Collet, D. Frere, F. Lang, M. Laville, and B. Estour Energy expenditure adjusted for body composition differentiates constitutional thinness from both normal subjects and anorexia nervosa Am J Physiol Endocrinol Metab, January 1, 2007; 292(1): E132 - E137. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Stevanovic, V. Milosevic, D. Nesic, V. Ajdzanovic, V. Starcevic, and W. B. Severs A BRIEF COMMUNICATION: Central Effects of Ghrelin on Serum Growth Hormone and Morphology of Pituitary Somatotropes in Rats. Experimental Biology and Medicine, November 1, 2006; 231(10): 1610 - 1615. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Brauner, C. Trivin, M. Zerah, J.-C. Souberbielle, F. Doz, C. Kalifa, and C. Sainte-Rose Diencephalic Syndrome due to Hypothalamic Tumor: A Model of the Relationship between Weight and Puberty Onset J. Clin. Endocrinol. Metab., July 1, 2006; 91(7): 2467 - 2473. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Miljic, S. Pekic, M. Djurovic, M. Doknic, N. Milic, F. F. Casanueva, M. Ghatei, and V. Popovic Ghrelin Has Partial or No Effect on Appetite, Growth Hormone, Prolactin, and Cortisol Release in Patients with Anorexia Nervosa J. Clin. Endocrinol. Metab., April 1, 2006; 91(4): 1491 - 1495. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Misra, K. K. Miller, V. Stewart, E. Hunter, K. Kuo, D. B. Herzog, and A. Klibanski Ghrelin and Bone Metabolism in Adolescent Girls with Anorexia Nervosa and Healthy Adolescents J. Clin. Endocrinol. Metab., September 1, 2005; 90(9): 5082 - 5087. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Misra, K. K. Miller, K. Kuo, K. Griffin, V. Stewart, E. Hunter, D. B. Herzog, and A. Klibanski Secretory dynamics of leptin in adolescent girls with anorexia nervosa and healthy adolescents Am J Physiol Endocrinol Metab, September 1, 2005; 289(3): E373 - E381. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Misra, K. K. Miller, K. Kuo, K. Griffin, V. Stewart, E. Hunter, D. B. Herzog, and A. Klibanski Secretory dynamics of ghrelin in adolescent girls with anorexia nervosa and healthy adolescents Am J Physiol Endocrinol Metab, August 1, 2005; 289(2): E347 - E356. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Erdmann, F. Lippl, S. Wagenpfeil, and V. Schusdziarra Differential Association of Basal and Postprandial Plasma Ghrelin With Leptin, Insulin, and Type 2 Diabetes Diabetes, May 1, 2005; 54(5): 1371 - 1378. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. E. Foster-Schubert, A. McTiernan, R. S. Frayo, R. S. Schwartz, K. B. Rajan, Y. Yasui, S. S. Tworoger, and D. E. Cummings Human Plasma Ghrelin Levels Increase during a One-Year Exercise Program J. Clin. Endocrinol. Metab., February 1, 2005; 90(2): 820 - 825. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Hotta, R. Ohwada, H. Katakami, T. Shibasaki, N. Hizuka, and K. Takano Plasma Levels of Intact and Degraded Ghrelin and Their Responses to Glucose Infusion in Anorexia Nervosa J. Clin. Endocrinol. Metab., November 1, 2004; 89(11): 5707 - 5712. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. R. Vulliemoz, E. Xiao, L. Xia-Zhang, M. Germond, J. Rivier, and M. Ferin Decrease in Luteinizing Hormone Pulse Frequency during a Five-Hour Peripheral Ghrelin Infusion in the Ovariectomized Rhesus Monkey J. Clin. Endocrinol. Metab., November 1, 2004; 89(11): 5718 - 5723. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Grinspoon, K. K. Miller, D. B. Herzog, K. A. Grieco, and A. Klibanski Effects of Estrogen and Recombinant Human Insulin-Like Growth Factor-I on Ghrelin Secretion in Severe Undernutrition J. Clin. Endocrinol. Metab., August 1, 2004; 89(8): 3988 - 3993. [Abstract] [Full Text] [PDF] |
||||
![]() |
< |