Dehydroepiandrosterone (DHEA) and its sulfate, DHEA-S, are the most abundant circulating steroids in man and the precursors for most steroid hormones (Orentreich et al. Reference Orentreich, Brind, Rizer and Vogelman1984). Serum concentrations of DHEA and DHEA-S are age dependent; in man, they rapidly increase at puberty, reach their peak levels between 20 and 30 years of age, and then decrease gradually (Yamaji & Ibayashi, Reference Yamaji and Ibayashi1969; Orentreich et al. Reference Orentreich, Brind, Rizer and Vogelman1984; Vermeulen, Reference Vermeulen1995; Macario et al. Reference Macario, Mazza, Ramunni, Oleandri, Savio, Grotto, Rosetto, Procopio, Gauna and Ghigo1999). This evolution, coincident with the incipient loss of physical performance, has led these hormones to be known as ‘the hormones of youth’ (Nawata et al. Reference Nawata, Yanase, Goto, Okabe and Ashida2002).
Far from being just biochemical intermediates, these steroids per se have been reported to have positive effects in the prevention and treatment of certain pathologies, especially the age-related ones, such as cancer (Schwartz et al. Reference Schwartz, Whitcomb, Nyce, Lewbart and Pashko1988; Ratko et al. Reference Ratko, Detrisac, Mehta, Kelloff and Moon1991; Kawai et al. Reference Kawai, Yahata, Nishida, Nagai and Mizushima1995), CVD (Ebeling & Koivisto, Reference Ebeling and Koivisto1994), cognitive deterioration (Yanase et al. Reference Yanase, Kukahori, Taniguchi, Nishi, Sakai, Takayanagi, Heji and Nawata1996), insulin resistance and obesity (Williams et al. Reference Williams, Boyden, Pamenter, Lohman and Going1993).
In man, the action of DHEA on obesity is not generally agreed. Some studies report no relationship between plasma levels of these steroids and body weight and fat (Azziz et al. Reference Azziz, Zacur, Parker, Bradley and Boots1991; Phillips, Reference Phillips1993; Barret-Connor & Ferrara, Reference Barret-Connor and Ferrara1996; Macario et al. Reference Macario, Mazza, Ramunni, Oleandri, Savio, Grotto, Rosetto, Procopio, Gauna and Ghigo1999), while others find a negative correlation between serum DHEA-S or DHEA and obesity (De Pergola et al. Reference De Pergola, Giagulli, Garruti, Cospite, Giorgino, Cignarelli and Giorgino1991; Tchernof et al. Reference Tchernof, Despres, Belanger, Dupont, Prud'homme, Morjani, Lupien and Labrie1995). Regarding its pharmacological use, some authors doubt that exogenous DHEA has any effect on weight loss in obese human subjects (Clore, Reference Clore1995). In contrast, others have suggested a role for DHEA-S treatment in fat-mass loss (Nestler et al. Reference Nestler, Barlascini, Clore and Blackard1988). Furthermore, DHEA-S plasma levels show a negative correlation with visceral fat distribution in women (Garaulet et al. Reference Garaulet, Pérez-Llamas, Fuente, Zamora and Tebar2000) and its administration seems to improve glucose tolerance (Haffner & Valdez, Reference Haffner and Valdez1994; Richards et al. Reference Richards, Porter and Svec2000) and to reduce serum cholesterol and TAG (Macario et al. Reference Macario, Mazza, Ramunni, Oleandri, Savio, Grotto, Rosetto, Procopio, Gauna and Ghigo1999), so ameliorating these features of the metabolic syndrome.
In rodents, DHEA has been reported to decrease dietary fat and energy intakes as well as body weight and fat content (Taniguchi et al. Reference Taniguchi, Yanase, Haji, Ishibashi, Takayanagi and Nawata1995; Richards et al. Reference Richards, Porter and Svec1999; Pham et al. Reference Pham, Porter, Svec, Eiswirth and Svec2000; Abadie et al. Reference Abadie, Malcom, Porter and Svec2001; Kajita et al. Reference Kajita, Ishizuka, Muna, Miura, Ishiwaza, Kanoh, Kawai, Natsume and Yashuda2003; Ryu et al. Reference Ryu, Kim, Kim, Song, Park, Lee, Kim and Lee2003). However, the mechanisms of action of this hormone on body composition are not yet fully understood, although a role in food intake regulation has been suggested (Shepherd & Clearly, Reference Shepherd and Clearly1984; Abadie et al. Reference Abadie, Wright, Correa, Browne, Porter and Svec1993; Svec et al. Reference Svec, Hilton, Wright, Browne and Porter1995; Wright et al. Reference Wright, Svec and Porter1995; Svec & Porter, Reference Svec and Porter1996; Pham et al. Reference Pham, Porter, Svec, Eiswirth and Svec2000), or even in the utilisation or storage of ingested energy (Clearly et al. Reference Clearly, Billheimer, Finan, Sartin and Schwartz1984; Mohan et al. Reference Mohan, Ihnen, Levin and Clearly1990). In addition, there are studies reporting direct effects of DHEA on muscle, suggesting another mechanism for the hormone action on body composition (Tsuji et al. Reference Tsuji, Furutama, Tagami and Ohsawa1999; Abadie et al. Reference Abadie, Malcom, Porter and Svec2001; Aragno et al. Reference Aragno, Mastrocola, Catalano, Brignardello, Danni and Boccuzzi2004; Campbell et al. Reference Campbell, Caperuto, Hirata, Araujo, Velloso, Saad and Carvalho2004). However, no specific studies have been found in the literature that focused on the influence of DHEA or DHEA-S on the digestibility of the different macronutrients, i.e. protein, carbohydrates or fat.
In the present study, the main objective was to analyse whether the effects of DHEA on body weight and composition of aged, fat-fed rats are exerted at a digestive level, i.e. dietary protein and fat digestibility or metabolic use. A second objective was to study the specific effects of DHEA in skeletal muscle composition, particularly in the gastrocnemius muscle.
Materials and methods
Animals and housing conditions
Twenty-one female Sprague–Dawley rats were provided by our University's animal care facilities, and kept in a temperature-controlled room (24 ± 2°C) in a 12 h light–dark schedule with lights on at 08.00 hours. Rats were bred with a high-energy diet, with 40 % of energy in the form of fat, from 7 weeks of age.
When rats were 72 weeks old and had an average body weight of 345 ± 6 g, they were housed in individual metabolism cages with free access to water and food. Dietary intake was recorded every 2 d, weighing dispensed, remaining and spilled food. Body weight was monitored weekly. From these measurements, the feed efficiency ratio (FER) was calculated as follows:
Dietary and hormonal treatments
The semi-purified high-fat diet (Portillo et al. 2001) is described in Table 1. This diet was freshly prepared once per week and stored at 5°C to avoid rancidity.
Table 1 Composition of the experimental high-fat diet*
* Palm oil supplied by Croexa (Barcelona, Spain); casein supplied by Hero (Murcia, Spain); sucrose supplied by a local market; cellulose (Avicel) supplied by FMC Corp. (Madrid, Spain); choline and methionine supplied by J. Escuder (Barcelona, Spain).
† Mineral and vitamin mixes were formulated according to the AIN-92 dietary guidelines for laboratory rodents' care (Reeves et al. Reference Reeves, Nielsen and Fahey1993) and supplied by Tegasa (Barcelona, Spain) and Sigma (St Louis, MO, USA).
After a 3-week adaptation to the metabolism cages, when the rats were 75 weeks old and had an average body weight of 354 ± 7 g, they were randomly assigned to one of two experimental groups: the control group (n 10) and the DHEA group (n 11). The control group kept on being fed the high-fat diet, without any change, while the DHEA group received the high-fat diet supplemented with DHEA at 0·5 % (w/w) (Roig Farma, S.A., Terrasa, Barcelona, Spain; 99·5 % purity). This hormonal treatment lasted for 13 weeks.
Digestibility of dietary protein and fat
The digestibility assay provides information on dietary use: the analysis of faecal N and fat is needed to determine the apparent digestibility coefficients for dietary protein and fat, while urine N is an index of the metabolic utilisation of body protein.
The assays were conducted on fourteen out of the twenty-one animals (seven from the control group and seven from the DHEA group) and consisted of the collection of 24 h urinary and faecal excretions during 1 week. In faeces, N and fat contents were determined to further estimate the intestinal digestibility coefficients of dietary protein and fat. Urinary N and net protein utilisation (NPU) were assessed to obtain information about total protein catabolism. All analyses were performed according to the official methods of the Association of Official Analytical Chemists International (Reference Cunniff1997).
The N content of urine was measured by the Kjeldahl method and expressed as mg N/100 g body weight. The N content of faeces was determined following the same procedure, and then protein was calculated by multiplying by the conversion factor 6·25. The fat content of faeces was assessed by diethyl ether extraction in a Soxhlet apparatus (Foss, Hillerød, Denmark), with a previous digestion with hydrochloric acid. Protein and fat faecal excretion were determined and apparent digestibility coefficients (ADC) were calculated as follows:
where N i is the nutrient intake (g) and N f is the nutrient content of faeces (g). In addition, NPU was calculated as the excreted N:digested N ratio, as follows:
where N u is the nutrient content of urine.
Before the beginning of the hormonal treatment, an assay of digestibility was performed, in order to confirm the homogeneity of the population (A0). To study the short- and long-term effects of DHEA treatment on protein and fat digestibility, two more assays were carried out following the same procedure, but at different times. The first one (A1) took place 2 weeks after the beginning of the hormonal treatment, and the second (A2) just at its end, after 13 weeks.
Assessment of body and muscular composition
At the end of the 13-week experimental period and after an overnight fast, all animals were anaesthetised with diethyl ether and killed by cardiac puncture, at the beginning of the light phase. Blood samples were collected and centrifuged to obtained plasma for DHEA-S concentration determination.
Peri-ovarian, mesenteric and subcutaneous fat depots were dissected, weighed, frozen in liquid N2 and stored at − 80°C. Isolated adipocytes were obtained by digestion of adipose tissue with collagenase A and filtration through nylon mesh, following the method of Rodbell (Reference Rodbell1964) with minor modifications by Langin et al. (Reference Langin, Portillo, Saulnier-Blache and Lafontan1991). Fat cell size was measured by optic microscopy, with the aid of a computerised image analysis system (MIP 4·5 Microm Image Processing Software; Consulting Image Digital, S.L., Barcelona, Spain) and the mean diameter was calculated by measuring 200 cells. Adipocyte number was estimated in each depot considering average cell weight and depot weight.
Hindlimb gastrocnemius muscles were dissected, weighed, frozen in liquid N2 and stored at − 20°C, in order to analyse the effect of DHEA administration on skeletal muscle. To determine whether the actions of DHEA on body weight were tissue specific, the relative gastrocnemius size was calculated as a percentage of total body weight.
Carcasses were homogenised by mincing in a grinder for the analysis of total body fat and protein. Muscle and carcass fat contents were determined in the Sohxlet apparatus, and protein contents were measured by N determination by the Kjeldahl method and multiplying by 6·25, as described earlier. The sample size for muscle analysis was 0·5 g for N quantification and 1·5 g for fat quantification. Body fat was calculated considering carcass fat and dissected adipose depots, and both body fat and protein were expressed as percentages of total body weight. In parallel, gastrocnemius muscle composition was expressed as percentages of total gastrocnemius weight.
Statistical analysis
All results are presented as mean values with their standard errors. Statistical analysis was performed using SPSS 12·0 (SPSS Inc., Chicago, IL, USA). Student's t test was used to compare energy intake, FER, circulating DHEA-S values, body weight, fat depots and body and muscle compositions between DHEA and control groups and the two-way ANOVA test (assay × DHEA treatment) was carried out for comparisons of the digestibility results. In all cases, significance was assessed at the P < 0·05 level.
Results
Body weight and fat and energy intake
In order to know if orally administered DHEA had been absorbed and incorporated into the bloodstream, DHEA-S plasma concentrations were measured, proving the oral treatment to be effective and showing that DHEA-treated rats had significantly higher DHEA-S concentrations than control rats (829·6 (sem 93·3) and 71·8 (sem 26·0) ng/ml, respectively; P < 0·0001).
Table 2 shows the changes in average body weight and body fat from DHEA and control groups after the hormonal treatment. Weekly body-weight changes in treated and non-treated groups are shown in Fig. 1(A). It can be seen that, although the initial weights were similar in the two experimental groups, final body weight was significantly lower in rats treated with DHEA. In addition, this decrease in body weight started in the first week after the beginning of the treatment and reached statistical significance as soon as in the third week. Body and carcass fat percentages were also reduced following DHEA administration and the reduction affected all fat depots studied. The significant changes found in fat cell size and number with DHEA treatment were depot specific: fat cell size was decreased in visceral (peri-ovarian and mesenteric) adipose tissue in DHEA rats, while in subcutaneous adipose tissue there was a reduction in adipocyte number (Table 2).
Table 2 Changes in body weight, body fat and cellularity of three different fat depots in the two experimental groups (Mean values with their standard errors)
DHEA, dehydroepiandrosterone.
Mean value was significantly different from that of the control group: *P < 0·05, **P < 0·01, ***P < 0·001.
Fig. 1 Body weight (A) and weekly food intake (B) in control (–♦–) and dehydroepiandrosterone-treated (–□–) rats throughout the experimental period. Values are means, with standard errors represented by vertical bars. *Mean value was significantly different from that of the control group (P < 0·05).
Regarding energy intake, it was smaller in the group that was fed DHEA compared with the control group (Table 3 and Fig. 1(B)). In order to know if the reduction in body weight was due to a diminished energy intake, the FER was calculated. Data show that DHEA-treated rats were less efficient in transforming the nutrients fed into their own biomass (Table 3).
Table 3 Body-weight change, cumulative intake and feed efficiency ratio (FER) in 13 weeks of the dehydroepiandrosterone (DHEA) treatment period (Mean values with their standard errors)
Mean value was significantly different from that of the control group: *P < 0·05, ***P < 0·001.
† FER = (weight change (g)/diet fed (g)).
Digestibility assays
To determine whether the diminished feeding efficiency had a digestive origin, three digestibility assays were performed at different times: before the beginning of the treatment (A0); after 2 weeks of treatment or short-term treatment (A1) and after 13 weeks of treatment or long-term treatment (A2) (Table 4). Data show that after 2 weeks of hormonal treatment there was a significant reduction in protein digestibility in the treated group (Table 4). Similar results were observed for urinary N excretion, which was significantly lower in the DHEA-treated rats in the A1 assay (short term). In agreement with these data, the NPU was higher in DHEA-treated rats, although differences did not reach statistical signification (Table 5). In the long-term assay (A2), we found a similar trend both in protein digestibility and N excretion, although without statistical significance (Table 5). With regard to fat digestibility, no significant differences were found between the DHEA-treated and non-treated rats, neither in the short- nor in the long-term assays (Table 4).
Table 4 Protein and fat apparent digestibility coefficients in the three digestibility assays (Mean values with their standard errors)
DHEA, dehydroepiandrosterone; A0, assay before the beginning of the treatment; A1, assay after 2 weeks of treatment; A2, assay after 13 weeks of treatment.
* Mean value was significantly different from that of the control group (P < 0·05).
Table 5 Urine nitrogen excretion and net protein utilisation (NPU) in the digestibility assays (Mean values with their standard errors)
DHEA, dehydroepiandrosterone; A0, assay before the beginning of the treatment; A1, assay after 2 weeks of treatment; A2, assay after 13 weeks of treatment.
** Mean value was significantly different from that of the control group (P < 0·01).
Body protein and muscular composition
DHEA treatment exerted a significant and positive effect on total body protein. Indeed, body protein percentage was significantly higher in DHEA-treated rats than in the control ones, although no significant differences were found in the other protein parameters studied, such as muscle protein content and percentage (Table 6).
Table 6 Body protein and muscle composition (Mean values with their standard errors)
DHEA, dehydroepiandrosterone.
Mean value was significantly different from that of the control group: *P < 0·05, **P < 0·01.
† Gastrocnemius weight expressed as a percentage of total body weight.
Regarding relative gastrocnemius weight percentage, significant differences were found between DHEA-treated and non-treated rats, the percentage being higher in the treated group. However, the gastrocnemius weight itself was significantly smaller. The lower weight of DHEA-rats' muscles could be due to the significant reduction of fat content of the muscles from the DHEA rats compared with control ones (Table 6).
Discussion
The involvement of exogenous DHEA and DHEA-S as anti-obesity agents in rodents seems to be generally accepted, although there is still some divergence about the effects on body weight and body-fat loss.
The present study confirms the effectiveness of a DHEA treatment in reducing body weight and the proportion of body fat in aged, high-fat-fed rats. These results are in accordance with other studies which show the anti-obesity properties of this hormone (Mohan et al. Reference Mohan, Ihnen, Levin and Clearly1990; Tagliaferro et al. Reference Tagliaferro, Ronan, Payne, Meeker and Tse1995; Lea-Currie et al. Reference Lea-Currie, Wen and McIntosh1997a, Reference Lea-Currie, Wu and McIntoshb). However, other studies found no effect on body weight (Lea-Currie et al. Reference Lea-Currie, Wen and McIntosh1997a, Reference Lea-Currie, Wu and McIntoshb; Aragno et al. Reference Aragno, Mastrocola, Catalano, Brignardello, Danni and Boccuzzi2004). These differences could be a consequence of the length of the treatment. In the present study, the slimming effect of DHEA was immediate; it was noticeable after just 1 week of hormone administration.
We observed that the effect of DHEA on adipose tissue was depot specific. All adipose regions studied were smaller in DHEA-treated than in control rats, but this reduction was mediated by diminished adipocyte size in mesenteric and peri-ovarian fat depots, while in the subcutaneous adipose tissue it was due to a 2·4-fold drop in fat cell number.
The present results also show a significant effect of DHEA on energy intake. Treated rats reduced their energy intake at the beginning of the experiment and, although it was steadily increased throughout the study, it remained lower than the intake of the control group and total energy consumption was significantly less than that of the control group. This behaviour has been previously observed (Abadie et al. Reference Abadie, Malcom, Porter and Svec2001; Ryu et al. Reference Ryu, Kim, Kim, Song, Park, Lee, Kim and Lee2003), although other authors report no alteration in food intake due to DHEA administration (Hansen et al. Reference Hansen, Han, Nolte, Chen and Holloszy1997). Again, as was postulated for body-fat reduction, the different impact of this hormone on intake could be influenced by the treatment period; in the short term the reduction is rather evident, but normal intake is recovered in the long term (Porter & Svec, Reference Porter and Svec1995). In the present study, however, this was not the case, since our experimental group ate less than the control one during the 13 weeks of DHEA administration.
In the revised literature, there is disagreement on whether the decreased energy intake could be the reason for the reduction in body weight and fat or if there are other factors influencing this effect. To elucidate this question, we calculated the FER. The present results show that FER was higher in control than in treated rats, indicating the DHEA-treated rats were less efficient in transforming the nutrients fed into their own biomass. In this sense, the study by Ryu et al. (Reference Ryu, Kim, Kim, Song, Park, Lee, Kim and Lee2003) showed that rats given DHEA lost more weight than their pair-fed, non-treated counterparts, even when the same energy intake was consumed. These data suggest that the weight loss observed in the DHEA-treated rats was not due exclusively to lower food intake, but to other processes that were being altered by DHEA.
The mechanisms by which DHEA acts on body composition still remain to be clarified. Authors have suggested different targets for the anti-obesity properties of DHEA, such as alteration of pre-adipocyte proliferation and differentiation (Lea-Currie et al. Reference Lea-Currie, Wen and McIntosh1998), increase of thermogenesis in brown and white adipose tissues (Ryu et al. Reference Ryu, Kim, Kim, Song, Park, Lee, Kim and Lee2003), or changes in the central regulation of food intake (Tagliaferro et al. Reference Tagliaferro, Davis, Truchon and Van Hamont1986; Wright et al. Reference Wright, Svec and Porter1995; Svec & Porter, Reference Svec and Porter1997; Gillen et al. Reference Gillen, Porter and Svec1999). However, it has not been reported yet whether DHEA affects the digestive process.
For that reason, for the first time, we analysed the possible effect of DHEA administration on the digestion and/or absorption of dietary protein and fat. The present results showed a significant reduction in protein digestibility in the short-term treatment with DHEA. No significant effect was found on fat digestibility neither in the short- nor in the long-term assay. These data as a whole could indicate that the anti-obesity properties of DHEA could be related to a decreased digestibility of dietary protein, but not to a specific action on the dietary fat digestibility.
Further studies are needed in order to find out how DHEA treatment affects protein digestibility, whether it interferes with the digestive process, interacting for instance with receptors in protease-secreting pancreatic cells or altering the function of peptidases in the luminal cell membranes. Perhaps DHEA acts at the absorptive level, interfering with amino acid and oligopeptide transporters (Martínez de Victoria et al. Reference Martínez de Victoria, Mañas, Yago and Gil2005).
Another mechanism for the anti-obesity effects of DHEA and DHEA-S could be mediated by altered utilisation of ingested macronutrients (Clearly et al. Reference Clearly, Billheimer, Finan, Sartin and Schwartz1984; Mohan et al. Reference Mohan, Ihnen, Levin and Clearly1990). We determined N urinary excretion and NPU, so as to estimate the metabolic degradation of proteins. DHEA supplementation was accompanied by a reduction in the urinary excretion of N in the short term. The DHEA-related decrease of urinary N excretion, together with the preserved NPU in spite of the lower food intake and body-mass loss, suggest a protective effect of DHEA on body protein. The fact that the reduction in dietary protein digestibility was followed by a decrease in N excretion could be a consequence of a possible compensatory effect of DHEA on protein balance.
To fully understand the impact of the previous results on body protein, we analysed total body protein and muscle composition and found that the percentage of body protein was significantly greater in DHEA-treated rats than in controls. Because of its accessibility, the gastrocnemius muscle has been previously studied to analyse the specific effects of DHEA on skeletal muscle (Hansen et al. Reference Hansen, Han, Nolte, Chen and Holloszy1997; Aragno et al. Reference Aragno, Mastrocola, Catalano, Brignardello, Danni and Boccuzzi2004; Campbell et al. Reference Campbell, Caperuto, Hirata, Araujo, Velloso, Saad and Carvalho2004). In the present study, a larger gastrocnemius size relative to total body mass was found in DHEA-treated rats compared with control ones. However, regarding gastrocnemius muscle itself, it was smaller. To search for a possible explanation for these results, the gastrocnemius muscle composition was analysed and data showed that the fat content of the muscle was reduced up to a third, while protein content was not altered. This suggests that the lower muscle weight in the DHEA group was due to the fat loss provoked by the hormonal treatment specifically in skeletal muscle tissue. These findings indicate that DHEA acts directly on skeletal muscle. In fact, Tsuji et al. (Reference Tsuji, Furutama, Tagami and Ohsawa1999) found two specific receptor sites for DHEA-S in skeletal muscle, and Liu & Dillon (Reference Liu and Dillon2002) described a G-protein-linked membrane receptor for DHEA. Also, DHEA has been described to exert metabolic effects on skeletal muscle, such as stimulation of glucose uptake (Campbell et al. Reference Campbell, Caperuto, Hirata, Araujo, Velloso, Saad and Carvalho2004), changes in fatty acid profile (Abadie et al. Reference Abadie, Malcom, Porter and Svec2001) and improvement of muscular function (Aragno et al. Reference Aragno, Mastrocola, Catalano, Brignardello, Danni and Boccuzzi2004).
In conclusion, the present results confirm that DHEA administration in aged rats fed a high-fat diet significantly reduces energy intake, body weight and body fat, with selective changes in fat cell size and number depending on the fat depot. We demonstrated for the first time that DHEA exerts a specific action at a digestive level. In the short term, DHEA treatment is followed by a reduction in protein digestibility compensated by a decrease in urine N excretion, indicating changes in protein digestibility and in catabolism. As a consequence, both body and muscle compositions were affected, showing an important reduction in fat content and preservation of protein content. It can be therefore suggested that the anti-obesity and anti-ageing properties of DHEA could be related to a reduction in protein digestibility and a protective effect on body protein, with a selective mass loss from body fat, and that DHEA's properties vary depending on the treatment length.
Acknowledgements
The authors acknowledge the companies Hero S.L. (Murcia, Spain), Norte-CRSA (Barcelona, Spain) and Tegasa S.L. (Barcelona, Spain), and the Spanish Ministry of Education and Culture, for their economical and technical support. Both F. P. De H and D. C have equally contributed to the elaboration of the present study.
