Obesity-related alterations in adipose tissue fatty acid composition have attracted scientific interest recently, as 18:0 content has been specifically shown to be altered in obesity.4
The aim of the present study was to examine the fatty acid content of SC and OM adipose tissues in women with or without abdominal obesity. We tested the hypothesis that adipose tissue 18:0 content would be reduced in expanded fat compartments with large adipocytes and that this would be observed with concomitant increases of the desaturation index and SCD1 expression. We found that fatty acid content of each adipose tissue compartment was relatively similar to that of the diet. However, 18:0 content was reduced in OM fat of women with visceral obesity and large OM adipocytes, while overall obesity was related to increased SC desaturation, as well as SCD1 mRNA and protein levels. Adipose tissue 18:0 content has previously been studied in relation to obesity, but we are the first to address this question with respect to fat distribution.
Previous studies relating adipose tissue fatty acid composition and dietary fatty acids are equivocal. General consensus is that adipose tissue fatty acids reflect dietary composition.5
Accordingly, we found that the relative content of most fatty acids was similar in the diet and adipose tissue. However, when examining the correlations between specific dietary fatty acids and their content in adipose tissue, results were less consistent. Only dietary 14:0 and 20:1 were significantly correlated with their counterparts in a given compartment. We cannot exclude a type 2 error related to our sample size. The trend for a positive association that we found between dietary intake of 18:2 and its content in adipose tissue is consistent with previous data,5
showing a correlation between dietary polyunsaturated fatty acids and their adipose tissue concentration. On the other hand, we did not find correlations between trans fatty acid consumption and adipose tissue content. This may be due to the new legislation banning trans fat in Canadian foods, which could have lead to an overestimation of trans fat intake with our food frequency questionnaire. Two studies by Garaulet et al.10, 17
examined adipose tissue and dietary fatty acid content. In the first study, 18:1n-9, 16:0 and 18:2n-6 were the most abundant adipose tissue fatty acids, matching the results of our study. The same group also found that SFAs were more abundant and MUFAs less abundant in visceral adipose tissue. We observed opposite results for SFAs and a trend for MUFAs. In the second study,17
a higher 16:1 and 18:1 content in visceral and SC adipose tissue, respectively, was reported in overweight subjects compared with obese and morbidly obese patients. We did not reproduce these findings. Discrepancies could be related to population differences such as the degree of obesity, sex or dietary patterns.
We examined two subgroups of 10 women with either small or large OM adipocytes that were matched for age, total BFM and SC adipocyte size. No difference was observed in the dietary profile of these subgroups. Adipose tissue fatty acid composition was also similar, except for 18:0 content, elongation and desaturation index, which were, respectively, lower and higher in women with large OM adipocytes. Results obtained with desaturation (18:1/18:0) and elongation (18:0/16:0) reflected those obtained with 18:0 alone, as 18:1, 16:0 and 16:1 contents were not related to any measure performed. Studies performed in human adipocytes suggest that elongation and desaturation processes are coordinately regulated to produce oleate.18, 31
Moreover, the abundance of SFAs is harmful to normal cell function32
and higher desaturation rates, and SCD1 expression was shown to protect against such lipotoxicity.33
This could possibly be the role of fatty acid desaturation and elongation in adipose tissue of women with abdominal obesity and large OM adipocytes. The absence of appreciable variation of 18:1 content may come from the fact that oleate is by far the most abundant fatty acid present in adipose tissue (). Recent studies in mice and humans indicate that the content of 16:0 and 16:1 is associated with insulin resistance.34, 35, 36
Subjects in our cohorts were healthy and did not suffer from pronounced metabolic disorders, including diabetes. However, the fact that adiposity is related to the substrate rather than the product of the SCD1 reaction needs to be further investigated.
The 18:0 content of SC fat has been related to total adiposity.4
When examining the measures of SC or overall obesity, such as BMI, fat mass, SC adipose tissue area and SC adipocyte size, we found an inverse correlation between these measures and 18:0 content in the SC depot. Therefore, variables representing the SC depot are correlated to the size of that same compartment. This is not in contradiction with our findings in the matched design, where SC differences were not apparent. When examining OM 18:0 content in a matched design so that differences in visceral adipose tissue are emphasized, lower OM 18:0 content is observed in women with visceral obesity and large OM adipocytes. However, when focusing on the SC 18:0 content in relation to overall obesity, a correlation is also observed between BFM and 18:0 content in SC adipose tissue.
SCD1 is the enzyme that introduces a single double bound between carbons 9 and 10 of the 18:0 fatty acid, converting it to oleate.37
In our study, we found positive correlations between SC desaturation (18:1/18:0), an indirect measure of SCD1 activity, and BMI as well as SC adipose tissue area. When focusing on the visceral depot, we found a positive correlation between the desaturation index and visceral adipose tissue area. In the SC depot, a positive correlation between the levels of SCD1 mRNA/protein expression and total BFM () was observed. A similar correlation was also found with BMI (data not shown). These results are consistent with a recent study from Garcia-Serrano et al.38
showing that the variable mainly associated with SCD1 protein level in SC adipose tissue was BMI.
The activation of ERK1/2 is an early event of adipocyte differentiation.21, 39
Consistent with the study of Laviola et al.
we observed higher ERK1/2 expression and phosphorylation in OM vs SC adipose tissue of non-obese women. In obese patients, the ratio OM/SC was diminished, which suggests that the capacity for adipocyte differentiation may be impaired in OM adipose tissue of obese women. Accordingly, a depot-specific difference occurs in obese subjects showing lower SCD1
gene expression in OM vs SC adipose tissue. Our results are consistent with a recent study showing higher SCD1 expression in SC compared with OM fat in obese women. SCD1 expression was also associated with DGAT2 expression, the rate-limiting enzyme in TG synthesis.41
SCD1 expression is mainly regulated by SREBP-1c at the transcriptional level in response to insulin by a PI3-kinase-dependent signaling pathway.42, 43
A similar depot-specific difference in non-diabetic obese subjects was already observed for SREBP-1c.44
We and others also observed a depot-specific difference regarding the level of IR () as well as the phosphorylation state of insulin sensitive pathways, such as ERK1/2 and PI3-kinase ().40, 45
Indeed, in obese subjects, insulin was shown to have a more pronounced effect in activating its associated signaling pathways, such as PI3-kinase in the SC adipose tissue.45
Additional analyses in our cohort revealed that SCD1 mRNA and protein levels are highly and significantly associated in the SC depot (r
<0.01), whereas this correlation was absent in the OM depot. This result indirectly suggests the predominance of transcriptional regulation of SCD1 in the SC depot. In the OM depot, an alternative post-transcriptional mechanism might take place. To date, only polyunsaturated fatty acids have been shown to impair SCD1 mRNA stability in adipocytes.46
This may account for the depot-specific difference observed in our study.
Taken together, and in agreement with other studies,15, 16, 47
our observations suggest that enhanced fat storage in the insulin-sensitive SC depot is associated with increased SCD1 transcription and activity. We speculate that this may be associated with the insulin sensitivity levels in our patients, who have relatively minor metabolic alterations.
Limitations of the study should be taken into consideration. The cross-sectional nature of the design prevents us to establish cause-and-effect relationships. The use of a food frequency questionnaire can also be seen as a limitation. However, as the fatty acid intake was generally well represented in adipose tissue, we suggest that the questionnaire was representative, at least for lipid composition. Further studies are needed with a larger pool of subjects as well as male participants.
In conclusion, our study demonstrates that in women, the presence of large adipocytes and increased adipose mass in a given fat compartment is related to reduced 18:0 content and increased fatty acid desaturation. This is also associated with an altered fat depot distribution in SCD1 and IR expression, as well as ERK1/2 expression and activation. These alterations are independent of dietary fat intake and suggest that SCD1-mediated desaturation in SC adipose tissue may exert a protective effect from metabolic disorders.