Oleoyl‐Estrone

Oleoyl‐estrone (OE) is a powerful slimming agent that is also present in plasma and adipose tissue, where it is synthesized. It acts through the formation of a derivative W. OE effects (and W levels) are proportional to the dose. OE reduces food intake but maintains energy expenditure (thermogenesis). The energy gap is fulfilled with adipose tissue fat, sparing body protein and maintaining glycemia (and glycogen) with lower insulin and leptin levels. OE (in fact W) acts through specific receptors, different from those of estrogen. OE increases cholesterol catabolism, reducing hypercholesterolemia in obese rats. The main metabolic effect on adipose tissue is lowering of lipid synthesis, maintaining unchanged the intracellular lipolytic processes; the imbalance favors the progressive loss of fat, which is largely used by the muscle. OE administration induces additive effects with other antiobesity agents, such as β3‐adrenergic agonists, forcing a massive loss of lipid. Corticosteroids markedly limit OE action by altering the liver control of lipogenesis. OE also inhibits the action of 17β‐hydroxysteroid dehydrogenase, decreasing the synthesis of β‐estradiol and testosterone. Discontinuous treatment allows for maximal efficacy both in rats and humans. OE has the advantage that the loss of fat is maintained and does not require additional dietary limitations.


INTRODUCTION
Oleoyl-estrone (OE) is both a natural hormone (or hormone derivative) and a powerful drug inducing the loss of body fat with permanent effects and practically no metabolic alterations except the massive fat loss. This ideal panorama, however, has its limitations, such as the marked interference of glucocorticoids and the ominous and direct relationship with estrogens. The latter point has severely limited research on this peculiar compound. Nevertheless, there is a considerable amount of information on OE. The marked lack of workable antiobesity drugs 1 is combined with the focussing of most pharmacologically oriented efforts to the development of new long-term use of drugs that help maintain a lower energy intake or support a moderate weight loss for incremental improvement of health. OE is somewhat different: is devised to solve the problem of obesity, not to decrease its effects, it is a lipophilic substance, a steroidal hormone nonetheless, instead of being either a peptidic hormone, cytokine modulator, or, essentially, a factor interfering with the appetite control circuits as are most of the drugs studied, including those already abandoned. These two ''peculiarities'' have significantly thwarted the development of parallel research on obesity based on sexrelated steroidal hormones, a trend that nevertheless at present is gaining momentum in metabolic syndrome research.
There are only three previous reviews on OE, the first is one decade old but was fairly complete, 2 the second is essentially a sketchy explanation of OE, 3 and the third is so far the most complete and independent description of drug with potential for the treatment of obesity. 4 The present analysis intends to encompass everything that has been published on OE so far, since the considerable amount of papers published on the subject from many points of view allows us, finally, to obtain a fairly wide panorama of OE, its potential and drawbacks.

OLEOYL-ESTRONE CHEMISTRY
Oleoyl-3-estrone (OE) is the ester of oleic acid (cis-D 9-10 octadecenoic) and estrone (Fig. 1). It has a waxy consistence and high hydrophobicity. It is insoluble in water, but soluble in dimethyl-sulfoxide and most organic solvents and vegetable oils. It is soluble in pure ethanol and methanol, but small portions of water rapidly decrease its solubility. 5 The metabolic effects are closely related to its particular structure, since modification of either the fatty acid or the steroid moiety results in the loss of biological activity. 6 OE chemical synthesis is relatively simple; it is formed by the reaction of oleoyl-chloride with estrone in an organic medium containing an organic base (i.e. pyridine) to take away the protons and facilitate the coupling. 7 The yield, even at ultramicroescale conditions, exceeds 60-70%. OE purification from estrone and remaining oleoyl-chloride is slightly more difficult, but high degrees of purity up to 98% can be easily achieved if the purity of the initial reagents is also high. Impure oleic acid (i.e. containing the trans isomer, elaidic acid, other fatty acids or methyl-esters) results in a softer product that keeps most of these impurities difficult to eliminate.
Tritium-or 14 C-labelled estrone can be synthesized in small amounts following the same procedure, and a simple thin-layer chromatography is sufficient to separate labelled OE from labelled estrone (498% label purity). 5,7 Tritium labelling in the estrone moiety allows for most tracer experiments because of its high specific activity and cheapness. When using biological materials, the main inconvenient is 3 H quenching, which make any quantitative study of the fate of the estrone moiety very difficult. The use of 14 C-estrone is even more limited, because of its high cost, scarce commercial supply, and, especially, its low specific radioactivity, which precludes its use as a tracer in most specific binding and hormone metabolism and distribution analyses. Synthesis of limited amounts of deuterium-and 14 C-labelled OE are carried out following the same procedures outlined above for tritium. 8 Oral administration of tritium-or deuterium-labelled OE results in the loss of part of the label, interchanged with the medium under the acidic conditions of the stomach. 9 Saponification of labelled OE produces a similar problem, because of the high pH that results in the displacement of estrone tritium/deuterium in significant proportions to form tritiated or deuterated water. 8 Thus, drastic pH exposures alter the specific activity of labelled OE and may result in artifactual measurements if the labelled water formed is not taken into account or removed.
Oleoyl-estrone can be hydrolyzed by a number of intestinal and cellular aryl-esterases. 10 Cholesterol-ester esterases break up OE with efficiency comparable to that of their breakup of oleoyl-cholesterol; nevertheless, OE hydrolysis is not dependent on taurocholate as is the hydrolysis of cholesterol-esters. 10

OLEOYL-ESTRONE ANALYSIS: LOW REAL CIRCULATING LEVELS
The initial mass-spectrometric analyses of OE were focussed to find out whether OE was indeed a natural compound, since the first approximation to its existence and eventual function was based on pure deduction and experimental evidence of effects, not being the result of the discovery of the molecule in a biological medium. The analyses showed that a compound with the mass of OE was present in lipoprotein and adipose tissue lipid extracts. 7 The coincidence of the fraction with internal OE standards and the presence of estrone resulted in the positive identification of OE as a natural compound; however, the results were only qualitative.
The systematic analysis of OE in biological materials were initially done by first extracting the compound, then hydrolyzing the ester bond and then measuring the estrone evolved through radioimmunoassay. 11 The extraction step should prevent carrying estrone sulfate, a major metabolite of OE in the rat. 12 Later, the dried extraction residue is saponified with potassium hydroxide in methanol-water, and estrone is extracted with ethyl ether or ethyl-acetate. 11 The organic extract is washed with water, dried, and then the residue is used for estrone radioimmunoassay, using tritiated estrone as standard. 11 The radioimmunoassay poses a number of potentially crippling problems. The first is the difficulty in obtaining adequate antibodies, cross-reactivity with estradiol and dehydroepiandrosterone, as well as the tolerance to impurities carried over from the biological medium extraction and the second extraction after saponification. In spite of ample variability in the measurements in the plasma of humans 13,14 and rats [15][16][17][18][19] due to the problems of the extraction-saponificationextraction-RIA method, the results were directly comparable only within the same experiment. The term ''acyl-estrone'' was used instead of OE because the method used could not determine the nature of the acyl moiety of the ester.
A ''macro'' method was developed for the analysis of OE in foods, 20 which was later refined for its application to dairy products, 21 both based on the above described principles, but most of the methodological development efforts were devoted to the measurement of OE changes in the plasma under varying physiological situations, such as starvation, 16,17 genetic obesity, 19 and exposure to high-fat 18 or cafeteria 19 diets.
A number of labelled OE studies have shown that ''acyl-estrone'' (i.e. the 14 C, 3 H, or 2 H-labelled estrone nucleus in a form other than estrone, estradiol, or their oxidized forms or sulpfates) is essentially carried in plasma by lipoproteins, mainly in the HDL fraction; 22 but it is also present in VLDL and LDL, as well as in the remaining plasma fraction. 22 When given orally, most of the absorbed non-hydrolyzed OE is carried by the portal vein with HDLs or, mainly, through the lymph as larger (chylomicra, VLDL) lipoproteins; the liver hydrolyzes part of the OE carried in larger lipoproteins to estrone or its catabolites, but leaves HDLacyl-estrone unaffected. 23 4. THE CIRCULATING OLEOYL-ESTRONE DERIVATIVE W, STRUCTURE, AND RELATIONSHIP TO LIPOPROTEINS Later, to simplify the cumbersome and partly unreliable OE analysis method available, we tried to develop a quantitative method for OE using a mass spectrometry approach and deuterated internal standards. Practically no ions were obtained for the whole molecule using electrospray or chemical ionization methods, thus, the direct approach was unviable for the analysis from biological samples. The MS/MS approach and the search for estrone fragments was better, and gave fair results for standards, but failed to show the expected OE levels in biological samples. [24][25][26] To increase ionization, several approaches were tested, but derivatization of OE on C17 with a Girard mixture gave the best results. 27 Most other procedures resulted in the early hydrolysis of OE and induced some of the problems indicated above for the RIA and the impossibility to undoubtedly identify OE at the levels ''showed'' by our RIA-based analyses of acyl-estrone.
HPLC-MS analyses of plasma samples resulted in an unexpected surprise that was systematically repeated with better sophistication (HPLC-MS/MS, chemical ionization analysis, derivatization, etc.): the actual levels of OE in plasma were in the range of 5-20 nM even after loading the rats with a large oral dose of OE. 26 The results obtained using the ''old'' RIA method gave yields in the range of 100 nM under comparable basal conditions, with several-fold increases after a load of OE. 12,15,28 This difference could not be attributed to manipulation or analysis errors, since in both cases internal and external standards (tritiated, deuterated, and just unlabelled OE) were used, and the data derived from labelled OE were corrected for the actual specific activity (i.e. discounted/evaporated labelled water). The number of tries and the diversity of approaches could not explain the differences either. The influence of dehydroepiandrosterone sulfate was analyzed and found irrelevant, and the presence of estrone sulfate 12 could justify-stretching the results-perhaps 10-20% of the ''old'' method yields. Free estrone was found to be in similar levels using both approaches 11,24 in the 0.1-2 nM range depending on the OE load (heavy loads resulted in much higher values, but never in the range of those of acyl-estrone). 12,15 The levels of ''real OE'' in humans were in the o5-50 nM range. 25 Thus, the HPLC-MS/MS values obtained for true OE levels in rat plasma (even after an oral load of OE) 24,26 were only a tiny fraction of the results obtained using the hydrolysis-RIA methods, and in no way justified the high levels of labelled estrone found in plasma after an oral gavage of 3 H-or 2 H-labelled OE. The analysis of samples with tritium or 14 C-labelled estrone proved that the estrone nucleus label did not correspond to the elution time of OE in the HPLC-MS/MS setting. 29 The data unequivocally proved, then, that the active form of OE transported in the plasma does not correspond to the chemical structure of OE.
The possibility of changes in the overall molecular weight, such as the hydroxylation of the estrone moiety in positions 2 or 6, or the conversion of the C17 keto group to alcohol, were also investigated, but the expected molecular weights did not appear in the scans. 29 In addition, it has been found that most of the body estrone is excreted through the urine or feces as sulfate and glucuronate conjugates, without chemical alteration of the estrone moiety. 30 In fact estrone is a very tough molecule that withstands adverse chemical and biological conditions and is used as an index of environmental contamination. 31 Thus, most (if not all) estrone nuclei are maintained unaffected throughout metabolic change up to excretion.

OE INTESTINAL UPTAKE, HYDROLYSIS, AND THE FORMATION OF THE OE DERIVATIVE W
The active form of OE carried by plasma has a different chemical structure than OE, as indicated above. This OE ''derivative'' (we call it simply ''W'') contains the estrone nucleus and has a higher solubility in water than OE, probably because of the substitution of the oleoyl moiety for a more hydrophilic compound, perhaps a peptide, susceptible of saponification freeing a RIA-recognizable estrone derivative, but not free estrone (HPLC-MS/MS). 29 After an oral load of labelled OE, there is a rapid surge of label in plasma, peaking at a few hours after gavage. 12 This is in agreement with OE being largely taken up in the upper intestine and, especially, the stomach, that in the rat retains label for more than 12 hr. 10 The intestines, essentially jejunum and ileum, contain esterases able to hydrolyze OE. In addition, pancreatic secretions contain OE-hydrolyzing cholesterol-ester esterase. The liver also contains powerful aryl-esterases that break down OE. 10 The combined hydrolyzing power of these enzymes is difficult to associate with most of the OE being absorbed or processed in a way different from free estrone, i.e. without hydrolysis. Probably this is a consequence of the process of synthesis of W, which apparently does not proceed from free estrone, but is easily formed from OE. This substitution of the oleoyl moiety by another more hydrophilic attachment is very probably carried out in the upper digestive tract wall itself, since the product is then found associated in the lymph-carried lipoprotein fraction. 23 This hypothesis is reinforced by the observation that oleoyl-DHEA, i.e. the substitution of estrone by another 17-keto steroid results in the practically quantitative hydrolysis of the compound, with a fast kinetics of plasma levels of DHEA, 32 a significant difference with the case for estrone.
It must be taken into account that the large array of esterases may contribute to the complete hydrolysis of food-borne acyl-estrone (i.e. from dairy products), rendering sizeable amounts of free estrone, which is taken up and disposed of by the liver in the adults. 23 In sucklings, however, this estrone may play a significant role as signal for energy efficiency, favoring the deposition of fat and protein and decreasing energy waste, 33 hence the large presence of acyl-estrone in milk. 20,21 Under physiological conditions we assume that OE is produced by adipose tissue and stored there, 34,35 being used for the formation of W and its release into the bloodstream.

IS THERE A ''TRANSPORTING PROTEIN''?
These two sets of data should take into account the particularly extreme insolubility of OE in water and the transport of acyl-estrone in plasma loosely bound to proteins (mainly lipoproteins). 22 Analysis of OE label distribution after oral dosing suggests that a large part of the label can be recovered in the organic phase of a biphasic extraction system, such as the classical trichloromethane/methanol Folch method. 36 In our early studies we assumed that this was the consequence of OE being pried out of the proteins which we assumed would remain in the methanol-water phase or precipitated. The possibility of W being attached in some way to a peptide or protein was thus investigated.
It is unclear whether W itself is present in plasma in free form or whether it is attached to a larger transporting protein. The ability to cross-molecular sieves suggests that W runs alone, but can attach to larger lipoproteins because of its innate lipophilia. Treatment of plasma with protein denaturing agents did not yield either free estrone nor changed the solvent distribution capabilities of W. 29 However, the absorption of nonhydrolyzed OE (i.e. not free estrone) from the gut requires the intervention of lipoproteins, since most of the label is carried away from the intestine not via portal vein but via lymph lipoproteins. 22,23 W either free or attached to lipoproteins is not susceptible to esterases as OE does; in cultured cells, exposure to microprecipitated OE (from OE solution in adequate organic solvents) results in its rapid hydrolysis. 34 Injection of liposomes loaded with OE in vivo result in the appearance of massive amounts of estrone in plasma 15,28,37 (a problem circumvented by oral administration), consequence of the massive hydrolysis of the real OE in tissues. 28 In addition, as indicated above, cholesterol-ester esterase also easily hydrolyzes OE. 10 In plasma, however, W is maintained at levels up to 100 nM range and free estrone is kept at less than one hundredth of that, 26 suggesting that, also in vivo, it is well ''protected'' from ubiquitous esterases. These estimations of W levels were arrived at by measuring the amount of 14 C and 3 H label present in plasma after a gavage of labelled OE in the estrone moiety after discounting the levels of all other normal plasma components containing the estrone nucleus (RIA and HPLC-MS/MS). 29

WHITE ADIPOSE TISSUE AS PHYSIOLOGICAL SOURCE OF OLEOYL-ESTRONE
OE is present 7 and synthesized in white adipose tissue. When 3T3 L1 cells are exposed to labelled-free estrone, a significant part of the label is recovered in the harvested cells as a more lipophilic fraction than free estrone (estimated through thin-layer chromatography) coincident with OE. 35 This synthesis is modulated by the presence of several hormones (insulin and glucocorticoids) in the medium and by leptin. 34,35 After the intravenous administration of liposomes loaded with labelled OE, as well as after the oral administration of either OE or free estrone, there is a considerable accumulation of estrone label in several tissues, including white adipose tissue, which is not in the form of free estrone but as ''acylestrone''. 38,39 Overloading of animals with free estrone, however, do not result in its massive accumulation, since the estrone balance is well maintained 40 : the amount of total estrone (assumedly, most of it as tissue OE) in the body is rather constant irrespective of the dosing of either OE or free estrone; 39,40 this indicates that the capacity of the organism to get rid of excess estrone is very powerful, but also suggests that the total amount of OE in the body is maintained at a closely controlled level by an unknown mechanism. 39 In animals that contain excess body fat, such as the genetically obese Zucker fa/fa rats, the body content of OE is lower than that expected from their large fat reserves. 39 Using different rat stocks it has been found that ''acyl-estrone'' and the body total estrone are directly related to the mass of body fat reserves (i.e. body lipids). 39 Zucker obese rats do not follow this relationship; their acylestrone (W's?) levels (RIA) are in the same range as their lean controls. 39 In humans, the situation is similar; in lean and overweight individuals there is a direct relationship between plasma acyl-estrone (RIA) and body fat. 13 In the obese and morbidly obese, however, this relationship is lost, with circulating levels lower than expected from body fat. 14

OE, W, AND THE PONDEROSTAT
The relationship between body fat and ''acyl-estrone'' (a highly plausible surrogate for W) levels 13,14,39 strongly supports its role as ponderostat signal, i.e. a signal from all adipose tissue masses in the body that contrive to build-up certain levels of a signal that is detected by the brain to adjust the whole body mass of fat reserves. 41 This indicates that adipose tissue in addition to synthesizing OE, is able to convert it into W, as indicated by the direct relationship between its circulating levels and body adiposity. 13,14 The hypothalamus monitors changes in the fat reserves, increasing appetite, favoring the incorporation of available energy into fat, and increasing its storage, and slowing down the lipolytic processes when the body reserves are detected to be in low levels. 42 Similarly, a detected excess of energy reserves elicits a decrease in food intake, increased thermogenesis, and predominance of lipolysis upon lipogenesis until the levels of the ponderostat signal return to the preset levels. When these levels increase sharply (as is the case when we administer OE as a drug), the situation is interpreted as a gross excess of stored fat energy, eliciting the physiological response of selectively giving rid of it, as observed under OE administration. 43 The experimental results obtained with OE treatment fit smugly with the ponderostat hypothesis. 7,43 OE induces a rapid loss of appetite, but energy expenditure is maintained. 43 Adipose tissue rapidly mobilizes the lipid energy it contains, 44 which is used mainly by the muscle and other peripheral tissues, 45 with no other metabolic alteration and without compromising the energy homoeostasis. 19,43 In addition to this purported role as ponderostat signal, OE treatment has been found to decrease the ponderostat reference setting. 46,47 The precise adjustment of the ponderostat is a critical factor for survival. It is closely related to the so-called ''thrifty genes'' that appear especially concentrated in human groups that have endured and survived long stretches of food scarcity. 48 Groups possessing thrifty genes can rapidly adapt to low food, and nevertheless strive. 49 These special set of genes, however, has a penance: the tight energy control that is the basis for survival becomes a liability when there is an excess of energy available. The high incidence of the metabolic syndrome-associated diseases (obesity, type 2 diabetes, hyperlipidemias, hypertension, cardiovascular diseases, gout, etc.) in our modern societies can be traced to the genetic presence of thrifty genes, 50 and the deep effects of the environment (availability of food, excess of fats, stress, sedentarism, etc.). [51][52][53] But also to the prenatal programming of ''thriftiness'', to be developed in an environment which is known only by the fetus from the energy availability signals crossing into the maternal womb. 54 Hypernutrition of the pregnant woman may convey to her fetus developing brain the notion that ''outside'' there is plentiful availability of food; fasting (for lack of food or medical/aesthetical reasons), on the other side may be interpreted by the fetus as a signal to prepare for an scarce-energy environment, thus fully developing its genetic potentiality for thriftiness. 55 There are many probabilities that (if the environmental conditions are adequate) a fetus raised in utero under limited maternal food availability may develop into an obese adult. There are a number of studies proving so. 56,57 The role of thrifty genes, in conjunction with perinatal exposure to environmental agents, are probably key factors explaining the present explosive growth of the incidence and severity of obesity (and of the metabolic syndrome) in the whole World. Exposure of neonatal rats to OE results in different body buildup and slower weight gain, 58 another factor suggesting the modulation of the ponderostat setting by OE.
The ponderostat is set at different body mass levels (i.e. the size of fat reserves) during the life-span. But at any given moment, the ponderostat setting of each individual is fiercely defended against external ''aggression'': starvation and the consequent decrease in energy availability is countered by decreased energy expenditure 59 and the controlled use of lipid reserves (as well as of body protein). 60 Refeeding results first in the recovery of the energy lost and only then energy expenditure is re-established. 61 This helps explain the generalized lack of success of hypocaloric dieting for the treatment of obesity: it is an uphill fight against one's own homoeostasis. 62 A number of diseases (mainly psychiatric) result in rapid changes in body weight, 63 and a few drugs produce iatrogenic obesity. 64 It may be speculated that these agents may also affect the ponderostat setting it at a higher body fat level. However, even the obese maintain their body weight (at a setting higher than that objectively optimal): they often encounter severe difficulties to decrease, but also to increase their body weight. 65 In this aspect, OE is peculiar between the agents used for the treatment of obesity in its down-regulation of the ponderostat setting. 46 OE not only decreases food intake and maintains energy expenditure, but also facilitates the mobilization of adipose tissue fats and their oxidation elsewhere; Once the fat has been shed off and the drug is discontinued, there is no rebound effect, 46,47,66 the weight lost remains lost and no accumulation of fat substitutes the oxidized fat. OE resets the ponderostat setting, opening a possible avenue for the cure, i.e. not only the treatment, of obesity.

OLEOYL-ESTRONE TREATMENT AND ENERGY BALANCE, EFFECTS ON FOOD INTAKE
Appetite is a main signal of the brain center for the control of energy homoeostasis to modulate the intake of (food) energy. The drive to eat (apart from conscious or other cortical influences) proceeds from two sets of signals: (a) emptiness of the gut 67 and (b) low levels of circulating energy metabolites, 68 which are in part complemented by cyclic energy and alert cycles in itself controlled by the hypothalamus-pituitary-adrenals (HPA) axis. 69 The gut and metabolic signals are mainly detected by the hypothalamus, which reacts through the control of neuropeptide Y (NPY) and other orexigenic peptides that elicit food intake, 70 and the counter-regulation of this appetite by melanocortins 71 results mainly in anorexigenic signals that decrease food intake. In both cases, efferent signals are mainly sent through nerves, but also using peptidergic paths.
In addition to the drive to eat, i.e. appetite, there are signals that decrease or stop feeding once initiated, inducing satiation. OE has been found not to alter the levels or hypothalamic NPY. 72 In addition, OE does not eliminate the sensation of appetite, but induces satiety very shortly after food is being eaten. 18 OE is not, thus, an appetite suppressant, but largely a satiation signal promoter. OE thus does not prevent food intake but markedly diminish it, safeguarding at least a minimum external energy supply. The effects of OE on appetite are more complex than simply eliciting satiation; intracerebroventricular injection of OE dissolved in dimethylsulfoxide provokes a loss of appetite (and maintenance of energy expenditure) in rats, 73 which demonstrates a ''central'' effect of OE on the brain. On the other side, OE has been found to dramatically inhibit, in the stomach, the expression of the main orexigenic gut peptide, ghrelin; 74 this peptide is directly involved in the maintenance of feeding, and its absence decreases food intake. OE also affects negatively a number of anorexigenic peptides along the intestinal channel, 74 but the effects are less marked than those of ghrelin. Since OE is largely taken up by the stomach, 10 the fast satiating effects of OE are probably conveyed through modulation of ghrelin synthesis and release.
Probably, the principal path for OE treatment decreasing food intake may be conceptually more simple, but metabolically complex: the maintenance of glycemia. 44,75 Plasma/ blood glucose is one of the main and strongest signals to awoke or damp appetite. Since glucose is the main inter-organ energy staple, and its levels are well maintained by the insulin system, a marked decrease in the gut-derived glucose supply, not adjusted by liver glucose output, results in hypoglycemia, which induces food intake. On the other side, hyperglycemia (assuming a normal insulin function) blocks appetite and elicits satiety. 76 OE increases insulin sensitivity, 77,78 but decreases peripheral glucose utilization 79 (at the expense of massive utilization of fat from internal stores), which leaves a well-maintained supply of glucose in plasma (normal levels, normalized insulin sensitivity), 44 full glycogen stores in liver and low peripheral demand by muscle (it uses mainly fats) or adipose tissue (in negative energy balance mode). Since ''nobody'' uses glucose massively, there are no signals forcing the body to look for more glucose (food) to replenish the gut, hence the decreased food intake. The combination of central effects, satiety signalling, and maintained glycemia result in a powerful combined drive to eat less. This is in part compounded by glucocorticoids, whose secretion is elicited by OE 80 that help maintain circulating glucose at the expense of other substrates, but nevertheless reinforcing the glucose homoeostasis and the negative effects on food intake induced by normoglycemia.
Treatment with OE results in the maintenance of overall energy expenditure in spite of the severe energy drainage experienced; 7,81,82 this can only be explained through central signals from the hypothalamus, as observed in rats directly injected into the brain ventricles. 73 In rats, adaptive thermogenesis represents a significant part of energy expenditure, and the factor more easily subjected to regulation. Under starvation or limited access to food, thermogenesis decreases, trying to maintain the energy balance grossly unbalanced by decreased energy intake. 83 In rats treated with OE, the expression of uncoupling protein-1 in brown adipose tissue, the main site for rat thermogenesis, is largely unaffected by OE. 84,85 Curiously, in brown adipose tissue of rats under limited food intake, the energy stress posed upon the metabolic machinery is less marked than in the tissue of OE-treated rats, since the latter has to sustain the severe energy drainage of uncoupling protein operation.

EFFECTS OF OLEOYL-ESTRONE ADMINISTRATION
The liver is a key organ for the mechanism of action of OE. It has a powerful OE-esterase capability 10 and may dispose of free estrone through additional hydroxylation 86 or conjugation with sulfate or glucuronate. 87 OE administration affects only slightly the liver energy metabolism, but the differences are more marked when compared with the effects of a limited food availability similar (pair-fed) to that encountered by OE-treated animals. 75 In the pair-fed rats, lipid used as substrate is complemented with the exhaustion of glycogen stores and the inhibition of lipogenesis. In OE-treated rats, however, the overall availability of glucose allows for the maintenance of full glycogen reserves 15,78,88 and the maintenance (at a lower setting) of lipogenesis from glucose, including the activation of the pentosephosphate pathway (and malic enzyme) as sources for NADPH. 75 In addition, lipoprotein synthesis is increased because the flow of fatty acids (from adipose tissue) to the liver is increased and there is no accumulation of fat in the liver; in addition, OE induces a marked increase in cholesterol turnover, 89,90 largely a consequence of reduced circulating levels of cholesteroyl-esters. 90 Cholesterol disposal toward the synthesis of bile acids is increased, 89 which agrees with the early and sustained decrease of cholesterolemia induced by OE treatment. 44 The effects on cholesterolemia appear rapidly after OE administration, 90 and are more marked in obese rats. 45 In muscle, glucose uptake is decreased in spite of maintained insulin sensitivity and normal expression of glucose transporters (GLUT4); 91 the small rise in circulating nonesterified fatty acids 44 facilitates their uptake, but do not induce insulin resistance. 79 The main source of energy for muscle is, however, an increased lipoprotein lipase 45 activity, which facilitates the sustenance of muscle essentially from circulating lipids, sparing glucose.
The conspicuous presence of full glycogen reserves in tissues, 15,88 together with a lack of stimulation of lipolysis 92 or thermogenesis 7,43,85 under OE treatment points to a diminished catecholamine stimulation, which is in part due to decreased expression of b-adrenergic receptors in adipose tissues, 93 but also to a marked decrease in the expression of the catecholaminergic pathway enzymes in the adrenal glands, 94 which prevents the logical catecholamine surge induced by starvation or limited feeding. 95 In the white adipose tissue, OE acts in slightly different ways at different sites. 92 The most marked effects can be seen on the mesenteric WAT, which venous output is carried to the liver via the porta vein. Other locations, such as the retroperitoneal or perigonadal (periovaric, in females and epididymal, in males) masses probably serve a function more related to medium and long-term storage, whereas the mesenteric has a direct relationship with intestinal function and liver signalling, and shows more rapid and extensive changes than the other. 92 Cell size is smaller in the mesenteric mass, but subcutaneous adipose tissue shows a higher degree of metabolic implication than the deep storage retroperitoneal and perigonadal sites, but less intense than that of mesenteric. 96 Production of cytokines, such as leptin, significantly varies between different sites, suggesting different role, regulation, and also different presence of nonadipocyte cells. 97 OE deeply inhibits the expression of leptin, 15 but needs a functional leptin receptor to do so, since it does not affect the expression of the LEP gene in the LEPR ob mutant obese Zucker fa/fa rats. 98 OE also inhibits the expression of antagonistic cytokines such as adiponectin, resistin, TNFa, and interleukins in adipose tissue. 99,100 Long-term effects of OE on adipose tissue include a decrease in cell size, 92,101 but also a decrease in cell numbers, mediated by OE-induced apoptosis. 102 OE reduces the amount of total WAT RNA. 92,99 The cells are limited to just physiological maintenance, with strongly inhibited lipogenesis and triacylglycerol synthesis 92 with limited uptake and utilization of glucose. There is no special activation of catecholamine signalling, nor increased lipase activities, just inhibition of the synthesis. 92 Lipoprotein lipase is strongly inhibited; 45,103 thus, the closure of the main energy gates for adipose tissue: glucose (low transport, decreased metabolism, and normal circulating levels with normal insulin), lipids (low or maintained lipoprotein levels in plasma and inhibition of lipoprotein lipase), and the hampering of the synthesis of fat 92,104 allows for a constant drainage of the products of an unchecked lipolysis and the consequent shrinking of the lipid droplets. The additional process of apoptosis helps release of huge amounts of energy progressively into the bloodstream for the liver to reshape as lipoproteins to feed off the rest of the body.
One of the most important peculiarities of OE-induced wasting of body energy is the preservation of protein. 7,43,44 It does not seem to exist a direct effect of OE on nitrogen metabolism. 18 In the liver, the loss of protein is limited (if any), and there is no damage to its structures (normal values for transaminases even at large OE doses). 44,46 The lack of negative nitrogen balance in spite of massive losses of energy can only be explained by a conjunction of factors: (a) glucose levels are maintained throughout, indicating a well-controlled energy homoeostasis system; (b) insulin activity is maintained at lower circulating levels, contrarily to obese insulin-resistant animals, where it is increased; (c) no extra needs for energy are apparent, especially, no need for 3C skeletons as substrate for gluconeogenesis (there is enough glucose available in spite of decreased food intake); (d) there is a decreased (albeit sustained) intake of energy, containing protein; (e) the overwhelming availability of lipidderived energy protects the diet and body protein from their utilization as energy substrates. Protein turnover, and amino acid oxidation, continue to proceed at a normal pace, which results in the protein pool remaining largely unchanged, and nitrogen balance unaffected. 19,44 Organ shrinking and apoptosis may generate minor adjustments, but, overall, protein homoeostasis is also maintained.
Again, the key to this maintenance probably lies on the stabilization of glycemia. Amino acid oxidation under starvation proceeds mainly to provide 3C precursors for gluconeogenesis, necessary to maintain glycemia. Most hypocaloric diets also result in negative nitrogen balances, since hypocaloric diets supplemented with protein become ''relatively'' hyperproteic and, consequently, the body machinery uses these amino acids as energy substrates instead of saving them to cover the obligatory nitrogen losses. In the case of OE this is not a problem because the diet ingested has an unchanged, ''normal'', proportion of protein, and there is no need for additional amino acid oxidation because glycemia is maintained.
In adipose tissues, apoptosis is enhanced by OE, but to a lesser extent than needed to justify the net loss of cells in a given white adipose tissue site. 102 It can be partly explained by the depressed cell turnover (lowered cell proliferation signals), maintained or partially increased apoptotic signals and more than probable specific loss of non-adipocyte cells.
Inflammation markers, such as cytokine levels, insulin resistance, and cell damage, are diminished under OE treatment, 100 in spite of the severe draining of adipose tissue reserves. There are no specific data overall suggesting the differential specific loss of macrophages or other nonstorage cells in adipose tissue, but the physiological consequences of OE treatment suggest such an effect. A line or particularly difficult analysis, for which we only have indirect data, is the fate of intramuscular adipose tissue, since it is difficult to discern between muscle lipid droplet depots and muscle-interspersed adipose cells. Analysis of the expression of cytokine genes in muscle and the careful quantization of adipose tissue masses by weight in rats compared with total rat lipid content hints at muscle (which stands for about 45% of a lean rat mass, compared with adipose tissue in the range of 5%, but total lipids justifying close to 9%) containing a sizeable part of lipid stores, probably in the form of small WAT masses in muscle, between other tissues, around vessels, etc. The interspersed WAT may be an important source of energy for muscle, especially red fibers, but may be also part of the problem of muscle-specific insulin resistance when insulin function is altered. 105 The data obtained so far with OE-treated rats suggest that non-adipose tissue fat stores tend to be used at a higher rate than adipose tissue, which may help exhaust excess lipid in muscle and thus markedly improving insulin resistance.

OLEOYL-ESTRONE AND INSULIN RESISTANCE
Treatment with OE markedly improves the hyperglycemic condition of prediabetic Zucker obese rats. 98 This is achieved with a parallel decrease in circulating insulin, thus reducing insulin resistance 79 and enhancing the insulin response to glucose. 106 OE can only partially improve the situation in streptozotocin-diabetic rats, 107 but is highly effective on other diabetic models, such as the Goto-Kakizaki 107 or the beta-diabetic rats. 108 In normal rats, OE decreases insulin levels, but maintains glycemia at normal and stable levels, increasing insulin sensitivity. 15,104 This effect, not paralleled by an increase in glucose uptake by peripheral tissues, 79 is a consequence of decreased expression of IRS-1 and GLUT4. 100 The elevated lipid turnover and utilization by muscle limits glucose uptake, which amounts in the practice to insulin resistance, since the circulating glucose is not taken by the tissues at a normal rate. However, this situation is characterized by decreased insulin levels (increased insulin sensitivity). 15,79 The relative increase in circulating non-esterified fatty acids may help limit glucose uptake by affecting the deployment of GLUT4 (a characteristic of classical insulin resistance), favoring instead the use of lipids as energy source. This seems to be the ''original'' reason for insulin resistance, but in most other scenarios this causes (or is a consequence of) lowered insulin sensitivity. This is not the case with OE, which favors exhaustion of lipid in adipose tissue, preservation of glucose in muscle, and lipogenesis from glucose in liver 75 but not in adipose tissue. 92 To further compound the situation, OE decreases the expression and levels of leptin and adiponectin, and the expression of resistin (and to a lower extent of TNFa and visfatin) in WAT. 100 Leptin and resistin have been associated with insulin resistance or decreased insulin sensitivity, 109,110 but visfatin is an insulin-like adipocytokine 111 and adiponectin is considered one of the key indicators of insulin resistance, 112 its increased levels announcing improvement of these conditions. 113 But OE inhibits the expression of both groups of cytokines. 100 In addition, when OE is given together with a thiazolidinedione, such as rosiglitazone, there is no additional improvement of glycemia or insulin sensitivity, both drugs acting through markedly different paths. 104 Their effects on adipose tissue metabolism are markedly different, their handling of lipids being practically opposite, since OE inhibits PPARa and PPARg, 104 the latter being the target of rosiglitazone. 114 In any case, the final result of both diverging agents is coincident: maintenance of glycemia and improvement of insulin resistance.
OE reduces insulin resistance, hyperlipidemia, and body fat in rats with defective leptinergic pathways (fa/fa, with impaired leptin receptor gene, thus insensitive to leptin), 45,78 but also on mice lacking either a functional leptin receptor (db/db) or a viable leptin gene (ob/ob). 115 This suggests that OE does not act centrally nor peripherally through leptinergic pathways. In addition, OE blocks the expression of the leptin gene in wild type animals and lowers their circulating leptin levels. 15

OLEOYL-ESTRONE AND THE GLUCOCORTICOIDS
Evidently, in the case of obesity and insulin resistance, glucocorticoids are a key factor. OE elicits increases in plasma ACTH and corticosterone in rats, an effect in part due to increased hypothalamic CRH. 80 This process is interfered by the lack of leptin receptors, and thus Zucker fa/fa rats are not as responsive as the wild type, 80 in spite of considerable strain variability in the responses to OE. 116 However, the plasma surge of corticosterone and ACTH cannot be directly correlated to changes in CRH, which points to additional or complementary mechanisms affecting corticosterone secretion. Treatment of adrenalectomized rats with OE results in an accelerated wasting of body fat reserves, which in this case also include losses of protein and a profound depression of appetite. 117 OE decreases the circulating levels of the serpin corticosteroid-binding globulin CBG, which increases the free plasma corticosteroid fraction. 118 In addition, OE treatment decreases the expression of CBG in liver and adipose tissue. 118,119 CBG modulates the availability of corticosteroids to adipose tissue, 120 largely because of a barrier-like distribution of the globulin on the cell surface. 121 Thus, the decreased CBG availability induced by OE may result in a relative increase of the accessibility of corticosteroids to adipose tissue. OE also increases the liver expression of 11b-hydroxisteroid dehydrogenase 1, 122 which locally increases the efficiency of glucocorticoids by favoring the predominance of the more active corticosterone over dihydro-corticosterone. 123 Inactivation of corticosteroids by the liver is also enhanced by OE treatment by increasing the expression of 5-reductase. 94 On the other hand, glucocorticoids strongly inhibit the lipid mobilizing effects of OE. 124 Combination of glucocorticoids and OE results in the marked enhancing of liver lipogenesis, 125 with the loss of the inhibition of lipogenesis in adipose tissue elicited by OE and due to glucocorticoid action. 125 The consequence is the loss of the capacity to effectively mobilize lipids, which may even induce the net accumulation of fat at higher concentrations. 124 The effects on appetite are also mainly lost, because of the direct (and indirect, through inhibition of CRH) effects of glucocorticoids on appetite, 126 but also on glycemia and, especially on insulin resistance. 126 The combination of OE and glucocorticoids plays havoc on the regulation of metabolism and insulin action, exacerbating insulin resistance and promoting lipid synthesis. 124 Anecdotic evidence in humans suggests that stress or other glucocorticoid sources may effectively block any slimming effects elicited by OE. Counterregulation of OE by glucocorticoids is probably the main problem that OE treatment faces. OE effects on energy metabolism are substantially opposed to those elicited by glucocorticoids: OE maintains glucose, diminishing liver glucose output and increasing the sensitivity to insulin, and glucocorticoids increase insulin resistance, increase glycemia, and the liver glucose output (Table I); OE decreases plasma lipids and inhibits lipogenesis, thus tipping the scales in favor of lipolysis, 92,125 and glucocorticoids favor lipid deposition 127 and in direct cell effects activate lipolysis, 128 raising the circulating levels of lipids; OE does not affect proteins, sparing body protein from catabolism, 19,44,81 and glucocorticoids stimulate proteolysis and the metabolic utilization  129 which results in a negative nitrogen balance. Glucocorticoids are related to inflammation processes and the increase of non-adipocyte adipose tissue cells; 130 OE probably decreases the number, function, and proportion of these cells. Finally, OE raises the levels of glucocorticoids 80 and these block its metabolic effects, eliciting OE resistance. 124 In any case, it is unclear whether glucocorticoids modulate the levels of OE, apart from modifying its rate of synthesis in cultured cells. 35 It may be speculated whether the continuous alteration of glucocorticoid secretion induced by sustained stress may result in a deranged OE ponderostat system in a similar way it deeply alters the glucose homoeostatic system by inducing insulin resistance.

OLEOYL-ESTRONE, ESTROGENS, AND ANDROGENS
The relationship of OE to estrogens seems obvious, since OE hydrolysis releases an estrogen, estrone. However, estrone is only a mild estrogen when compared with the main physiological one, 17b-estradiol. Estrone has powerful energy storage effects as described above, but its easy interconversion with estradiol by means of 17b-hydroxysteroid dehydrogenase may result in increased levels of estradiol following the massive hydrolysis of OE. Intravenous administration of OE produced significant estrogenic effects, which were not observed when OE was given orally to rats. 28 The limited human data available suggest that there is indeed an increase in estradiol after OE administration, but of limited extent. 131 Data from rats showed that estrone-sulfate is the main plasma catabolite of OE, 12 and is probably the main form in which the estrogen nucleus is excreted. 132 The increase in plasma estradiol elicited by OE is less marked than expected from the rise in estrone, and can be explained by an inhibition of the 17b-hydroxysteroid dehydrogenase activity, 133 which prevents the production of too much active estrogen. A secondary consequence of this inhibition is the marked decrease in active androgen activity. The interconversion of androstenedione to testosterone by the same OE-inhibited 17b-hydroxysteroid dehydrogenase results in a lowered production of testosterone, 133 which may affect the androgenic function under heavy doses extended for a long time. 131 The eventual combination of slight estrogenic and more marked hypoandrogenic effects of OE is one of the main drawbacks of its eventual pharmacological use in humans. A number of studies has shown that at comparable (allometrically corrected) doses, OE effects are more marked in male than in female rats. 78 This was initially taken as a hint that OE may act through the estrogenic pathway, since females are supposed to be more used to synthesize and dispose off estrogen than males. Recent studies on the metabolic effects of estrogen do not support this assumption, 134 but the fact remains that males are more sensitive to OE. Probably the main reason for this difference may lie on the different size and distribution of fat reserves, and, probably, on the ''basal'' effects of estrogen on the insulinergic path.
OE shares with estrogen insulin-sensitizing and cardiovascular risk factor protection abilities, but the marked estrogenic effects of estradiol and the high mobilizing effects of OE are not shared. In any case, the possibility that OE (or W) may act just as ''carrier'' or estrone (to yield later estradiol) to target cells/organs was tested; this avenue could not be considered true because the substitution of OE by estrone or estradiol gavages results in completely different effects in the case of estrone, 40 or acute estrogenization without loss of weight in the case of estradiol, 135 even at much lower concentrations than those used for OE. The limited appearance of estradiol in plasma after OE administration, largely unrelated to the dose of OE 12 and the trend to convert estrogen into OE observed in cultured cells (not the reverse), 29 suggest that OE (or W) acts through a pathway that does not require its previous conversion to estradiol. Combined treatment of rats with tamoxifen showed that the relationship between OE and estrogen is not simple, either, since tamoxifen in some ways mimicked the effects of OE, while blocking the estrogenic effects of OE metabolites. 136 In fact tamoxifen showed a few ''W-like'' properties 136 that did not appear when it was substituted by a more pure estrogen receptor antagonist, fulvestrant. 137 Fulvestrant actually inhibited a few of the actions of OE, but not all. The data available so far suggest that it is possible that in some way the estrogen receptor may be involved in how OE exerts its metabolic effects, 137,138 but not all and not directly, since the overall consequences of estrogen receptor stimulation are widely different from the effects of OE administration, and inhibition of these receptors does not completely block either the OE-elicited metabolic changes. OE did not mimic the effects of estrogens in brain, which indicates that OE does not act through estrogen-mediated pathways. 82 Under physiological conditions, the relationship of OE to estrogen (and androgen) is probably complex. Androgen and estrogen are progressively being considered main metabolic regulation players, clearly beyond their better-known sex-related roles. Protection of protein buildup by androgens 139 and prevention of cholesterol and other lipid alterations by estrogen 140 are widely recognized. Aromatase is a critical enzyme converting androgen into estrogen; this enzyme is present mainly in the gonads, adrenal cortex, brain, breast, adipose tissue, skin, and placenta. 141,142 The brain actively uses all steroid hormones as signals 143 and thus also possesses the ability to synthesize estrogens. 144 However, the reason why breast and adipose tissue can synthesize estrogen, and the function of these hormones is not as readily apparent. In breast its function could be related to the production of acyl-estrone, for export into the milk (as discussed above). But in adipose tissue the ability to synthesize estrone 145 (and then esterify it into OE) is probably part of the production of OE (or W) as a bloodcarried ponderostat signal.
The synthesis of the conjugated linoleic acid ester of estrone (CLE) 146 allowed us to test a molecule analogous to OE in which the fatty acid itself has a specific apoptosis-producing effect by itself when delivered specifically into the cells. 147 CLE reunites in the same molecule the overall OE effect (i.e. its expected conversion into W) because of its chemical structure, but its hydrolysis under physiological conditions may result in insulin modulating 148 and apoptosis eliciting action of conjugated linoleic acid. 149 The results obtained with CLE compared with OE showed that on a molar basis, CLE was not as powerful as OE in mobilizing fat, but CLE was more effective than OE on the control of insulin resistance. 146

THE W RECEPTOR
OE does not bind the estrogen receptor: it is unable to displace estradiol in vitro. 28 The known limited effects of estrogen lowering body weight (essentially lipid, too) 150 could not be considered responsible of the slimming effects of OE. However, we know now that OE is not the actual agent responsible for the metabolic effects elicited by the oral (or injected) administration of OE, since OE is not present in significant amounts in the plasma, being substituted by what we call OE-derivative, W. 29 We do not know the chemical nature of W, and thus we have not been able to carry on in vitro binding tests. However, we have tested the binding of W to nuclear receptors in an indirect way. Using liver cells from rats loaded with tritium-labelled OE we obtained a cytosol extract which contained tritium label in the form of W free or bound to a specific receptor. 29 When the cytosol of liver cells from rats having received a gavage of 3 H-OE was incubated with nuclei from untreated rats, part of the cytosol OE-derived label (W) was found bound to the nuclei; the presence of cold estrone or estradiol did not affect this binding. 29 In consequence, estrogens did not displace W binding to nuclear sites, which agrees with our postulate that OE acts through specific receptors that identify the active form of OE, W, circulating in plasma and present in liver cell cytosol. Addition of labelled plasma to a cold cytosol-nuclei system also resulted in the binding of label to nuclei, unaffected by estradiol or estrone. Adipose tissue nuclei show a similar behavior, but the small yields of this tissue and the frank metabolic response of liver decided us to continue using this organ for most of the receptor-binding studies.
It may be assumed that the W receptor shares some of the particularities of other steroid receptors: cytoplasmatic receptors that eventually hybridize with other (or the same) loaded receptors that are brought into the nucleus and affect the translation of DNA segments. It can be speculated that the estrogen-like interactions manifested by partly modifying some effects of tamoxifen and fulvestrant on OE activity 136,137 may be due to possible hibridization of the W receptor with others, including the estrogen's, thus achieving a much wider range of effects and helping minimize unwanted secondary effects.

OLEOYL-ESTRONE AS ANTIOBESITY DRUG. RODENT MODELS
The first OE experiments were done on rats using the intravenous continuous infusion of a liposome suspension (Intralipid, 30% lipid) with OE administered using Alzet osmotic minipumps. 15,151 The rationale was that OE would be hydrolyzed along the intestinal path, yielding unwelcome estrone. After finding out that the oral administration resulted in lower hydrolysis, and, consequently, less estrone (and estrogen) interference, 28 the intravenous approach was forfeited. It was soon found that the fat-elimination effects of OE were dosedependent, the optimal range being established for rats in the 1-20 nmol/g per day, 44 and establishing the 10 nmol/gd dose as standard for most experiments, since it yielded significant losses of fat and metabolic changes amounting to 10-15% of body energy in just 10 days. 18,44 A study involving leptin deficient (ob/ob) and leptin receptor deficient (db/db) mice showed that OE was able to make them lose weight and body fat 115 as occurred with rats deficient in the leptin receptor (fa/fa) 19 and wild type rats. 44 However, the wild type mice did not respond to OE, only high doses resulted in a minimal loss of body weight, most doses (including those that make ob/ob and db/db mice lose weight) in fact elicited an increase in body weight similar to that experienced by rats receiving low doses of OE (the estrone effect). 7 Treatment of mice with higher doses of OE, however, elicited in them similar results to those observed in rats. 152 OE also induced a marked loss of weight in an inbred strain of rats with high resistance to insulin (beta-rats). 108 The type of vehicle for oral OE administration was investigated in order to optimize the efficiency of intact (i.e. pharmacologically effective) absorption of OE. The best results were obtained using natural oils of C16-C20 chains, typically sunflower oil, which contains a large proportion of polyunsaturated and monounsaturated fatty acids. The overall effects of sunflower oil bested those of olive oil (essentially triolein) and fish oil (containing mostly polyunsaturated fatty acids); short chain fats, saturated fats, and OE in liposomes gave poor results altogether. 153 The standard volume of 0.2 mL was established because of its limited energy content (7 kJ), and adequate volume for handling and gavage.

SAFETY ISSUES AND DRUG SYNERGIES
The possible dangers posed by estrogenicity were studied using rats in the OE1tamoxifen study described above. 136 The advantages of discontinuous vs. prolonged OE administration were also studied using the rat model, finding that discontinuous treatment has the clear advantage of minimizing possible negative hormonal effects, and there is no recovery of the weight loss during the interspersed ''rest and recovery'' periods. 47,66 Following the same rationale that in the end resulted in the withdrawal of fenfluramine/ dexfenfluramine, its combination with another drug, phenthermine, enhanced the antiobesity effects of fenfluramine, 154 the possible synergies/dangers of combination of OE with known drugs used for the treatment of obesity were studied. 96 First, a model of overweight male rats was established, using adult males treated for three months with a cafeteria diet followed by a stabilizing period of 2 weeks after which they were used in the experiments. 96,155 These animals had a proportion of body fat in the range of 15-18%, comparable to that of adult overweight male humans.
The combined effects of food intake limitation and OE were investigated using two levels of food restriction comparable to the low-calorie diets and very low-calorie diets used for the treatment of obesity in humans. No additional benefit was obtained by further restricting food intake in terms of lipid loss or metabolic homoeostasis. 155 The differences between OE-treated and partially starved rats were considerable when looking at the handling of glucose and amino acids, but the consequences on lipid mobilization were more similar between both groups. In any case, the data support the possible use of OE as slimming drug without the additional concourse of hypocaloric diets, a considerable difference with all the other drugs used for the treatment of obesity. OE effects on body fat are fully apparent and demonstrable in animals being fed a hypercaloric (hyperlipidic) diet: energy intake decreases independently of the energy density of the food. 18,81 The combination of OE and a serotonin-reuptake drug such as dexfenfluramine or sibutramine resulted in a synergistic effect, especially in the combination of OE with sibutramine, which induced more marked losses of fat that either drug alone. 96 This effect was not observed with phentermine, probably because this drug by itself induced very little change on body fat. 96 Combination of OE and the lipase inhibitor orlistat produced results very difficult to interpret because of the complexity of the experimental setup. In any case, the combination of both drugs did not result in any advantage either metabolic or inducing the loss of fat, especially because orlistat itself induced little effect on rats, 156 since their normal diet is poor in lipid. The use of a hyperlipidic diet did not elicit any additional benefit either. The preparation of rat chow balls amassed with water containing orlistat or not induced a type of taste-aversion on OE-treated rats that maintained a group of animals voluntarily eating nothing for almost 10 days. Surprisingly, most metabolic parameters, including glycemia, were maintained, but the loss of weight and energy was massive. 157 The combination of OE and a specific b 3 -adrenergic receptor agonist (CL316,243) resulted in a powerful synergistic effect of mobilization of body fat: the adrenergic agonist increased energy expenditure 158 and accelerated lipolysis, while OE decreased food intake and enhanced lipid mobilization. In only 10 days, a group of rats lost more than one quarter of all their body energy without any apparent change in their energy homoeostasis markers. 158 The short-term loss of effect that characterizes the use of b 3 -adrenergic receptor agonists usually prevents the prolongation of its thermogenic effects beyond a few days, but combination with OE, which reduces food intake and eliminates the markers for emergency wasting, may help maintain its effects for a longer time, including a marked potentiation of thermogenesis via increased expression of WAT UCP1. 100 Rimonabant was also investigated in combination with OE. No significant combined effects were observed, in part because the effects of rimonabant were limited to the very first part of the study. 159 The combination of OE and a glitazone, rosiglitazone, 104 has been commented above in relation to insulin resistance. With respect to body weight (fat), rosiglitazone treatment resulted in the accumulation of fat, which was effectively removed when combined with OE; however, OE alone induced a greater loss of energy that the combination of both drugs, since rosiglitazone stimulated lipogenesis in WAT, inhibited by OE. 104 This series of experiments helped to know what to expect from the point of view of safety in the combination of OE and other available drugs. However, the most important finding may be to prove that OE does not act through the same paths of serotonin-reuptake activators (dexfenfluramine and sibutramine), thermogenic enhancers such as phentermine or CL316,243, cannabinoid CB1 inhibitors such as rimonabant, or PPAR agonists such as rosiglitazone. The mechanism of action of OE is, thus, different from these, allowing for possible additive effects that become synergistic in the cases of sibutramine and, especially the b 3 -adrenergic receptor agonist CL316,243. 100,158 The possible influence of OE on growth and development was investigated as a consequence of the finding of large amounts of acyl-estrone in the milk 21 and the ''estrone'' effect inducing growth and fat deposition. Treatment of pregnant rats with OE resulted in them losing more weight than controls, but the weight and biochemical parameters of the pups were unaffected at birth. 160 The OE effects during pregnancy, both on the dam and its foetuses, were relatively milder than those obtained when the treatment was applied to dams and their pups just after birth. 58,161 This was a direct consequence of differential expression of genes controlling the handling of lipids. 162 In any case we were unable to demonstrate the crossing of OE-derived label through the placenta, only free estrone label (and in small amounts) was found in fetal tissues. 161 Treatment of lactating dams with OE also affected their fat reserves, with little initial effect on the pups. 161 However, the posterior development of these pups showed that exposure to OE during lactation reduced their rates of growth, 58,160,161 affecting differently male and female rats and had permanent effects on their ability to store fat and grow. 163 The results were compatible with the theory of perinatal setting of the ponderostat.

INTERVENTION STUDIES IN HUMANS
Shortly after finding that oral OE does not produce significant estrogenic effects, the only published OE study on a human was started. 47 To prevent the appearance of problems due to alteration of steroid hormone control and secretion, a standard treatment cycle was established, consisting of 3-weeks of OE administration followed by a 2-3-month ''resting and recovery'' period. The doses initially tested were in the 1-1.5 nmol/gd, i.e. in the range of 100 mg/day, allowing for allometric differences in metabolic rate between standard rats and humans (BMR being proportional to BW 0.75 ). The results were substantial: the loss of weight was not immediate, peaking just at the beginning of the recovery period. In 2½ years of ''treatments'' there was a net loss of 42 kg (initial BW 172 kg), with improvement in metabolic indicators and well being, and no apparent estrogenization.
After Manhattan Pharmaceuticals took control of the development of OE as a drug, the necessary safety studies were carried out in animals and OE was found to be safe for initial tests with humans. Then, a phase-I study was carried out, showing that OE was safe; the data showed a hint of slimming effects at the highest doses. Then, a phase-II was devised with the emphasis on ''safety'' rather than on effects. However, a ''safe'' level for indefinite treatment with OE was established at a very low dose compared with those used previously with rats and humans for limited-time studies.
In phase II, two approaches were used, the discontinuous procedure of two 2-week treatment stretches with two rest and recovery periods. The doses used 5, 10, or 20 mg/day were too low to obtain a significant loss of weight (a maximal loss of 0.8 kg was expected by extrapolation of the previously published human data). 47 A thorough analysis of metabolic and hormonal parameters was included, showing no significant negative effects at the end of the study. 131 The second approach included a single group of morbidly obese patients treated continuously for 1 month with 30 mg/day OE followed by 30 days rest. In this case, there was a 2-month mean net loss of 2.671.5 kg compared with a loss of 1.170.9 kg for the control (placebo) group, i.e. about 1.6 kg, 131 in range with the expected 1.3 kg weight loss for dose and duration calculated from the published data. 47 The difference with placebo was not significant and the development of the product was abandoned because of insufficient Other drugs refer to: serotonin and noradrenaline reuptake inhibitors, adrenergic agonists, cannabinoid 1 receptor inhibitors, thiazolidinediones, metabolic blockers, lipase inhibitors, and CCK agonists.

OLEOYL-ESTRONE K 19
Medicinal Research Reviews DOI 10.1002/med slimming effects, in spite of not observing any change in safety or metabolic components at the end of the study. 131 The normalization of estradiol and testosterone levels at the end of the rest period was in agreement with the benefits of the discontinuous treatment pattern.
The changes in sexual hormones, however, became a crippling problem in the eventual development of OE for the treatment of obesity, since the Drug Control Agencies are prone to look for drugs with minimal (if any) secondary effects, putting the emphasis on safety more than on efficacy, the rationale being the need for lifelong treatments. This may be the case for most drugs: in the market, in most Companies' pipelines or withdrawn, but it is not applicable to OE, since treatment with OE was not expected to be indefinite but of a limited duration in time.

PERSPECTIVES
Most studies on OE metabolic effects and the whole development of OE as a potential drug for the treatment of obesity were developed assuming that it was OE itself the physiological agent (ponderostat signal) responsible for all the changes in energy utilization and appetite observed. The discovery that it was not OE but W the compound circulating in plasma and binding the cell nuclear receptors occurred after the development of OE as antiobesity drug candidate was abandoned. Only part of the oral OE seems to be converted to W, the rest being largely converted to estrone derivatives (estrone sulfate) and excreted. Under physiological conditions we can also expect that W would also produce estrone in its catabolism; however, there is a question of scale in the quantities of estrone involved to obtain a similar metabolic effect. It is highly probable that the full identification of W may provide us with a good candidate for the further development of steroidal antiobesity drugs of which OE has been the first (failed) candidate. We continue working on this line because the clear advantages of OE over any of the so far known antiobesity drugs (Table II) are considerable, and its main (albeit significant) drawbacks on the realm of sexual hormones may be dramatically minimized by the direct use of its effective metabolic form, W.

ACKNOWLEDGMENTS
This study was supported by grant SAF2009-11739 of the Plan Nacional de Investigacio´n en Biomedicina of the Government of Spain. The contribution od CIBER Fisiopatologı´a de la Obesidad y Nutricio´n of the Spanish Institute of Health Carlos III is gratefully acknowledged. Disclaimer: None of the Authors has a current interest in discovery or development drug companies.
Biology (1985)(1986)(1987)(1988)(1989)(1990)(1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003), and later, of Nutrition and Food Science (2001)(2002)(2003) at the University of Barcelona. His main research interests have been the study of amino acid metabolism during pregnancy, lactation, and development, the effects of ethanol during development, the effects of exercise on substrate handling by obese rats, and, lately, the study of the metabolic syndrome, mainly focussed on the finding of new pharmacological approaches for the treatment of obesity, with the decades-long study of oleoyl-estrone as antiobesity drug candidate. He has obtained one patent and has published about 200 papers in peer-reviewed journals. He has directed 10 PhD theses and has been the PI of six competitive research projects. He was the co-founder of a biotechnological company, OED SL in 2001, a spin-off of the University of Barcelona, dissolved in 2010.  Barcelona (1994Barcelona ( -2001. His main research interests have been the study of intestinal absorption of nutrients, the effects of exercise on substrate handling by obese rats, and, essentially, the study of the metabolic syndrome, mainly focussed on the finding of new pharmacological approaches for the treatment of obesity, with the decades-long study of oleoyl-estrone as antiobesity drug candidate. He has obtained one patent and up to date has published 140 papers in peer-reviewed journals. He has directed seven PhD theses and has been the PI of three competitive research projects. He was the co-founder of a biotechnological company, OED SL in 2001, a spin-off of the University of Barcelona, dissolved in 2010. Maria`Alemany (1946 Barcelona, Catalonia, Spain) is currently a full Professor of Nutrition and Food Science at the University of Barcelona (from 2001). He obtained his PhD in Sciences (Biology) in the same University in 1972, and did a post-doc stay at the Washington Univeristy (1973)(1974) in St. Louis, MO. He was an Assistant Professor of General Physiology at the University of Barcelona (1974Barcelona ( -1977, Associate Professor (and later full Professor) of Biochemistry at the University of Balearic Islands (1977)(1978)(1979)(1980)(1981), and Professor of Biochemistry and Molecular Biology at the University of Barcelona (Faculty of Chemistry at Tarragona and later Biology) . His main research interests have been the study of glycogen metabolism, amino acid metabolism in mammals and birds during development, the effects of exercise on substrate handling by obese rats, and, lately, the study of the metabolic syndrome, mainly focussed on the finding of new pharmacological approaches for the treatment of obesity, with the decades-long study of oleoyl-estrone as antiobesity drug candidate. He has written 7 books, (including a novel), obtained 7 patents, and published about 300 papers in peer-reviewed journals. He has directed 17 PhD theses and has been the PI of 18 contracts and competitive research projects. He was the co-founder of a biotechnological company, OED SL in 2001, a spin-off of the University of Barcelona, dissolved in 2010. He has obtained five relevant awards, including the Wasserman Award (2001), from the European association for the Study of Obesity.