The influence of sex and gonadectomy on hepatic and brain fatty acid composition, lipogenesis and β-oxidation

Introduction

Fatty acids are important molecules in living organisms, serving as an energy substrate, as the structural components of cell membranes, ligands for transcriptional factors involved in gene expression and as precursors of lipid mediators such as eicosanoids, resolvins and neuroprotectins (Kremmyda et al., 2011; Tvrzicka et al., 2011; Jump et al., 2013). Metabolic diseases, such as diabetes, obesity or metabolic syndrome, as well as several other diseases such as hypertension, coronary heart disease, alcoholism, schizophrenia, depression, Alzheimer's disease, atherosclerosis and cancer, are characterized by disturbances in fatty acid metabolism. The progression of these diseases could be, at least partially, influenced by fatty acid supplementation (Mocellin et al., 2016; Mašek et al., 2014; Ito, 2015; Wani et al., 2015; Endo and Arita, 2016; Pawełczyk et al., 2016). Consequently, fatty acid metabolism is being extensively investigated due to its significant clinical implications (Yashodhara et al., 2009).

De novo lipogenesis and bioconversion of long-chain polyunsaturated fatty acids (LCPUFA) are regulated by desaturation and elongation enzymes, competition for these rate-limiting enzymes between the n3 and n6 lines, partitioning into oxidation, esterification into phospholipids, different metabolites, substrate availability from food, transcriptional factors, hormones and microRNA (Wang et al., 2006; Childs et al., 2010; Tu et al., 2010; Fernández-Hernando et al., 2011; Jump et al., 2013).

Testosterone and oestrogens have an important role in carbohydrate, protein and fat metabolism. Sex hormones influence the fatty acid composition and expression of genes involved in lipid metabolism. Nevertheless, molecular mechanisms have still not been completely elucidated (Kelly and Jones, 2013; Shen and Shi, 2015). Ovariectomy and orchidectomy are still important experimental models for the study of sex hormone deficiency and its influence on fatty acid metabolism (Perez et al., 2009; Alessandri et al., 2011; Kitson et al., 2013). These investigations are becoming even more interesting in the light of the recent dramatic increase in the incidence of metabolic syndrome and diabetes mellitus, and the recently discovered connection between these diseases and sex hormones and their receptors (Chow et al., 2011; Høst et al., 2013; Zhang et al., 2013; Cai et al., 2015).

The aim of this study was to determine, using a rat model, the relationship between sex hormones and tissue fatty acid composition. In addition, we investigated the effects of sex and gonadectomy on the expression of key genes involved in lipogenesis and mitochondrial and peroxisomal β-oxidation. The results of this study will help to explain whether sex should be regarded as a significant variable in scientific studies investigating fatty acid composition, lipogenesis and β-oxidation, and whether the differences are equally significant in both sexes.

Materials and methods

Results

Body weight and uterus weight

As shown in Fig. 1a, female rats had significantly lower body weight compared to the male rats (p < 0.05). Gonadectomy had no impact on the male rats’ body weight, while in females, it caused a significant increase of body weight (p < 0.05). As expected, the uterus index decreased significantly (p < 0.001) in the F OVX group in comparison with the F SHAM group Fig. 1b.

Liver desaturation indices and relative gene expression

Desaturation indices for ∆ desaturation, calculated as the product/precursor ratio and relative expression of genes involved in lipid metabolism, determined by RT-qPCR, are presented in Fig. 2. Desaturation indices for C16:1n7 (C16:1n7/C16:0) were higher in the M ORHX group compared to all the other groups, while the indices for C18:1n9 (C18:1n9/C18:0) were higher in the male rats compared to the female rats. The relative expression of ∆9D increased in male rats after castration. In female rats, castration did not influence ∆9D expression. Desaturation indices for ∆6 desaturation increased after castration in the male rats and decreased in the female rats, while relative expression of the ∆6 desaturase gene increased in both sexes after gonadectomy. Desaturation indices for ∆5 desaturation were not influenced by castration, although they were higher in the F OVX group compared to the male rats. Relative expression of ∆5 desaturase increased in the female rats after castration, while there was no change in the male rats. Relative expression of the FASN gene increased after castration in the female rats, while the relative expression of the ACOX1 and CPT1A genes decreased after castration in the male rats.

Content of C20:4n6 and C22:6n3 in liver and brain phospholipids

We further examined the content of C20:4n6 and C22:6n3 in the liver and brain phospholipids, as important constituents of membranes (Fig. 3). Liver phospholipids showed similar distribution of both fatty acids to that in the total lipids. Intact male rats had lower content of C20:4n6 and C22:6n3 compared to female rats, while castrated males had values between the intact males and females. Neither castration nor sex influenced the brain phospholipid content of C20:4n6 and C22:6n3.

Lipid peroxidation

The intact females had significantly lower lipid peroxidation, measured as TBARS, compared to all other groups (Fig. 4).

Discussion

Ovariectomy was associated with uterine atrophy, which clearly demonstrated a decrease in endogenous oestrogen and successful gonadectomy, as in other studies using ovariectomy as a model for oestrogen deficiency (Alessandri et al., 2011). Increased body mass after castration in female rats is consistent with the influence of oestrogen on the regulation of feed intake and body mass. Following ovariectomy, caloric intake increases, as well as body mass, and then is maintained above the level for the sham control (Chen and Heiman, 2001). The lipogenic effect of ovariectomy was visible in the increased liver index and decreased serum triglyceride content. Serum triglyceride content and liver triglyceride content are inversely related after ovariectomy, probably due to impaired liver VLDL synthesis (Paquette et al., 2007). However, in our trial, we did not observe increased liver triglyceride content in ovariectomized animals. The most probable explanation is that our trial lasted only 4 weeks, in which period only initial lipid accumulation developed.

Saturated fatty acids were strongly influenced by sex, with higher content of C16:0 in males and higher content of C18:0 in females. These results could be explained by the higher elongation rate of C16:0 than C18:0 in female rats and the increased expression of elongase 6 (Elovl6), whose main substrate is C16:0 (Wang et al., 2006; Marks et al., 2013). Nevertheless, other factors, such as uptake by liver from blood, partitioning into triglycerides or phospholipids or into β-oxidation, should be considered (Alessandri et al., 2012). The sex differences observed in C18:0 and C16:0 are not related to the differences in important PUFA examined (C20:4n6 and C22:6n3), because these acids are derived directly from food, or bioconverted from their precursors ingested by food. Although sex differences in C16:0 and C18:0 are probably not as important as the differences in the PUFA content, they must be considered as a variable in trials involving lipid metabolism.

MUFA content was influenced by sex and gonadectomy, but the influence was different for different fatty acids. The most abundant MUFA, C18:1n9, was not influenced by sex or gonadectomy, despite the fact that its direct precursor, C18:0, was significantly higher in the female rats. Therefore, it is evident that the ∆9-desaturation rate for C18:0 was higher in the male rats. Orchidectomy also influenced the ∆9-desaturation rate of C16:0 to C16:1n7, but further decreased elongation of C16:1n7 to C18:1n7, which is the result of elongase 5 and 6. As a consequence, the values for C18:1n7 were similar between the sham and the castrated rats, but higher in males than in females. Despite the complexity of MUFA synthesis and the interplay of different factors involved, these results could be at least partially explained by the significant increase in the expression of ∆9D in the castrated male rats and the absence of change in the female rats. In contrast to our results, Alessandri et al. (2011) found a significant increase in ∆9D expression following ovariectomy. Nevertheless, it should be noted that C18:1n9 concentration in their experimental feed was much higher than in our trial.

Important PUFAs were influenced by sex in our trial, with higher values in the female rats compared to the male rats. An interesting observation was the increase in 20:4n6 and 22:6n3 after castration in the male rats, suggesting the inhibitory effect of testosterone on LC PUFA synthesis. Indeed, several authors have observed the inhibitory effects of testosterone treatment on 22:6n3 (Marra and de Alaniz, 1989; Extier et al., 2009) and on 20:4n6 (Marra and de Alaniz, 1989; Hurtado de Catalfo and de Gómez Dumm, 2005). In contrast, the influence of orchidectomy is still controversial. Some investigations have shown an increase in C20:4n6, similar to our trial (Perez et al., 2009), while others have observed a decrease in LC PUFA (Clejan et al., 1982). Despite increased values for 22:5n3 (a precursor for 22:6n3) in the male rats, 22:6n3 content was higher in the liver tissue of the female rats. That sex-related difference was not visible in brain which was also observed by Kitson et al. (2012). These results suggest increased ‘∆4’ desaturation in females and castrated males. Although ∆4-desaturase does not exist, the 22:6n3/22:5n3 ratio is used as a marker of ∆4-desaturation, which involves elongation of 22:5n3 to 24:5n3, ∆6-desaturation of 24:5n3 to 24:6n3 and finally one cycle of β-oxidation of 24:6n3 to 22:6n3, which takes place in peroxisomes (Sprecher, 2000). The applicability of the ‘∆4’ index is discussed by Alessandri et al. (2012) suggesting that when n3 LC PUFAs are absent from the diet, the ‘∆4’ index is determined by the extent to which 22:5n3 is esterified into phospholipids or bioconverted to its products. As 22:6n3 fatty acid was not detectable in our experimental feed, the ‘∆4’ index could have applicability. Therefore, it could be observed that the differences between males and females in 22:6n3 liver content are related to the desaturation and elongation steps after 22:5n3 and that these steps are influenced by castration in the male rats. The overall influence of castration and sex on the liver fatty acids is shown by the PLS-DA score and loadings plot (Fig. 5). The score plot clearly shows distinct clusters for the intact and castrated male rats and only one cluster for the female rats, which suggests the significant influence of castration only in the male rats. The PLS-DA loadings plot shows the variables that contribute the most to the differences between the groups. Intact males are more related to C18:2n6 and C16:0, castrated males are more related to MUFA and ∆9 desaturation indices, while the female rats are more related to the LC PUFA and ‘∆4′ index. These fatty acids should be considered most in trials with gonadectomy in the experimental design.

The connection between actual fatty acid content, ∆5D and ∆6D is more complex and more difficult to explain. In the male rats, ∆6D expression increased after castration which could be a reason for the increase in PUFA content. It should be noted that most authors have found that testosterone inhibits ∆5D and ∆6D (Marra and de Alaniz, 1989; Cinci et al., 1993; Hurtado de Catalfo and de Gómez Dumm, 2005). Moreover, key enzymes for mitochondrial and peroxisomal β-oxidation significantly decreased after castration in the male rats.

In the female rats, both ∆5D and ∆6D expression increased after castration, but PUFA content was similar in the castrated and intact female rats. In ovariectomized rats, an increase in ∆5D and ∆6D mRNA expression, as well as ∆6D protein content, without any change in liver fatty acid composition, was previously observed by Alessandri et al. (2011) and Kitson et al. (2013). In view of the previously established positive effects of oestrogen supplementation on PUFA content, the increase in ∆5D and ∆6D mRNA expression after ovariectomy could be a compensatory mechanism to maintain normal PUFA values, as well as part of the overall lipogenic effect caused by oestrogen deficiency (D'Eon et al., 2005). In female rats, besides the factors that influence synthesis and partitioning of PUFA, factors that have an impact on the degradation of fatty acids should also be considered. In our trial, the female rats had the lowest rate of lipid peroxidation, as measured by TBARS, which could also be a significant factor contributing to the total PUFA content in female rat tissue.

In conclusion, from our data, it is evident that even under a normal rodent feeding programme, castration and sex have a significant influence on lipogenesis. The influence of sex is observable in the liver, but not in the brain, and it includes two independent groups of fatty acids. The first consists of palmitic and stearic acids, and the second of arachidonic and docosahexaenoic acids. The influence of castration on the liver fatty acid profile is much less evident and primarily involves MUFA in male rats. Castration has significant, but different effects on the sexes, with an increase in de novo lipogenesis in female rats and a decrease in mitochondrial and peroxisomal β-oxidation in male rats. These changes could be at least a partial explanation for the connection between the incidence of different diseases, such as metabolic syndrome and diabetes mellitus, and sex hormone status in humans. In view of these results, both sex and castration should not be neglected in experiments involving fatty acid profile, lipogenesis and β-oxidation in rats.