Fatty acid profiles in adipose tissues and liver differ between horses and ponies

Fatty acids, as key components of cellular membranes and complex lipids, may play a central role in endocrine signalling and the function of adipose tissue and liver. Thus, the lipid fatty acid composition may play a role in health and disease status in the equine. This study aimed to investigate the fatty acid composition of different tissues and liver lipid classes by comparing Warmblood horses and Shetland ponies under defined conditions. We hypothesized that ponies show different lipid patterns than horses in adipose tissue, liver and plasma. Six Warmblood horses and six Shetland ponies were housed and fed under identical conditions. Tissue and blood sampling were performed following a standardized protocol. A one-step lipid extraction, methylation and trans-esterification method with subsequent gas chromatography was used to analyse the total lipid content and fatty acid profile of retroperitoneal, mesocolon and subcutaneous adipose tissue, liver and plasma. In the adipose tissues, saturated fatty acids (SFAs) and n-9 monounsaturated fatty acids (n-9 MUFAs) were most present in ponies and horses. N-6 polyunsaturated fatty acids (n-6 PUFAs), followed by SFAs, were most frequently found in liver tissue and plasma in all animals. Horses, in comparison to ponies, had significantly higher n-6 PUFA levels in all tissues and plasma. In liver tissue, horses had significantly lower hepatic iso-branched-chain fatty acids (iso-BCFAs) than ponies. The hepatic fatty acid composition of selected lipid classes was different between horses and ponies. In the polar PL fraction, horses had low n-9 MUFA and n-3 PUFA contents but higher n-6 PUFA contents than ponies. Furthermore, iso-BCFAs are absent in several hepatic lipid fractions of horses but not ponies. The differences in fatty acid lipid classes between horses and ponies provide key information on the species- and location-specific regulation of FA metabolism, thus affecting health and disease risk.


Introduction
The physiological fundamentals of the lipid metabolism of equids are poorly understood. 44 Several studies have shown that lipid and lipoprotein statuses differ among horse breeds [1][2][3][4][5]. 45 Ponies have higher plasma lipoprotein contents than horses and seem to be more susceptible to inflammation. PUFAs of the n-3 series rather than n-6 PUFAs have commonly been shown to 54 exert molecular actions that result in an improved risk factor profile in relation to metabolic and 55 inflammatory dysregulations [9][10][11]. It is further speculated that the health impact of n-3 PUFAs 56 on whole body homeostasis is mediated by resetting the adipose tissue (AT) function [12]. AT is 57 no longer considered a simple fat storing tissue but rather contributes as an integrative key 58 regulator in energy homeostasis and systemic metabolism [12,13]. AT can influence and 59 communicate with many other tissues, including the brain, heart, vasculature, muscle and liver, 60 on different molecular levels by releasing pro-and anti-inflammatory mediators such as 61 interleukin 1 beta (IL-1β), interleukin 6 (IL-6), tumour necrosis factor alpha (TNF-α) and other 5 62 adipokines [13,14]. In equids, a close link between AT function and health conditions is 63 postulated [15]. Controversial studies about whether visceral fat or subcutaneous fat depots have 64 a greater modulating impact on inflammation exist [16,17]. Thus, it is important to address how 65 tissues vary with respect to FA composition in horses and ponies. 66 The aim of the current study was to compare the FA contents and profiles of different 67 ATs, liver, plasma, and hepatic lipid classes between Shetland ponies and Warmblood horses. 68 Considering the differences in the lipoprotein metabolism of horses and ponies, it was 69 hypothesized that the FA profiles, with a special focus on the n-6 and n-3 PUFA dynamics, were 70 different between equine breeds. Animals and preselection criteria 75 Six Shetland ponies with a mean age (± SD) of 6 ± 3 years and six Warmblood horses 76 with a mean age (± SD) of 10 ± 3 years, all geldings with a median body condition score of (25 th 77 /75 th percentile) 3.7 (2.2/4.4) for ponies and 3.6 (3.1/4.2) for horses on a scale of 1 to 6 [18] and 78 a mean body weight (BW) (± SD) of 118 ± 29 kg (ponies) and 589 ± 58 kg (horses), were 79 included in the study. Animals were individually housed in box stalls and bedded on straw. All 80 animals had turnout onto a sand paddock for at least 5 h a day. During a two-week 81 acclimatization period to the experimental procedure, animals were fed daily 2 kg meadow 6 82 hay/100 kg BW, which was divided into two equal portions, one offered in the morning and one 83 offered in the evening.

84
The animals had free access to water at all times. All animals were assessed for plasma 85 adrenocorticotropic hormone (ACTH) to rule out pituitary pars intermedia dysfunction (PPID).   Thin-layer chromatography 125 The liver (0.1 g) and fat samples (0.5 g) were cut into small pieces and put into 10 ml 126 glass tubes containing a solvent mixture of chloroform and methanol (1:1 v/v). The final dilution 127 was 1:50 for fat and 1:10 for liver tissues (1 g wet material corresponds to 1 ml). Tissue samples 128 were homogenized, and total lipids were extracted for FA analysis.

129
Total lipids of the homogenized liver samples were extracted using a mixture of water,     In addition to SFAs, PUFAs of the n-6 FA family were the most dominant lipid FA in the 215 liver ( Table 2). The hepatic n-3 PUFA content was not different between ponies and horses, but 216 the percentage of the hepatic n-6 PUFA fraction was significantly higher in horses than in 217 ponies. The n-6/n-3 ratios calculated for all tissues and plasma in horses were significantly lower 218 than those in ponies. Except for the total hepatic n-11 MUFA fraction in ponies, MUFA contents 219 were lower in the liver than in AT depots for both horses and ponies. The total n-11 MUFA 220 concentration in the liver of ponies was 5-fold higher than the corresponding MUFA content in 221 horses.

222
Ponies contained a 2-fold higher proportion of n-11 MUFAs in the liver than in the 223 different ATs. Liver iso-BCFA content (10.0%) was 5-fold higher in ponies than in horses.  (Table 2). have inverted n-6/n-3 PUFA ratios and higher percentages of iso-BCFAs in the liver than in the 232 AT depots (Table 2).
17 233 Δ9-desaturase activity indices determined from the 16:1n7/16:0 ratio and 18:1n9/18:0 234 ratio were significantly lower in the liver than in the ATs for both horses and ponies (Fig 1A and   235 B). Increased Δ6-and Δ5-desaturase and elongase indices were found in the liver compared to 236 the AT depots in all animals (Fig 1C-E). In this context, horses had a significantly higher hepatic 237 Δ5-desaturase index than ponies (Fig 1D). Plasma FA lipid composition (Table 1b) and Δ9-, Δ6-and Δ5-desaturase and elongase 250 indices corresponded more closely to the liver profile than to the AT profiles in both horses and 251 ponies (Fig 1A-E).
18 252 In all animals, the majority of hepatic FAs accumulated in the polar PL fraction. Among 253 the neutral FA fractions, TAGs had the highest FA amount, followed by NEFAs and CEs

254
( Table 3). The absolute amounts of FAs in the PL fraction were significantly higher in ponies 255 than in horses (P = 0.004). In ponies, the FA levels in the hepatic NEFA fraction were 2-fold 256 higher (P = 0.03) and that of the CE fraction 3-fold higher (P = 0.004) than those in horses.

257
There were no significant differences in the FA levels of TAGs except for one pony that     (Table 4). Compared to horses, ponies had significant higher n-9 MUFA and n-3 PUFA 20 275 levels but significant lower n-6 PUFA contents in the PL fraction, resulting in a lower n-6/n-3 276 ratio. No significant differences in SFA and iso-BCFA levels were found in the hepatic PL 277 fractions between horses and ponies. Similar to the PL fraction, n-6 PUFA levels and calculated 278 n-6/n-3 ratios in NEFAs, TAGs and CE were significantly higher in horses than in ponies. N-11

279
MUFAs and 280 iso-BCFAs were absent from the hepatic NEFA fraction of horses. In ponies, the hepatic CE 281 fraction had significantly higher n-3 PUFA levels but lower SFA amounts than that in horses.

282
Ponies had the highest iso-BCFA levels and the lowest n-9 PUFA contents in the hepatic CE 283 fraction. These FAs were completely absent from the CE fraction of the horse liver.  the total FA content and FA lipid profile between the distinct AT locations were found in horses 319 and ponies (Table 1).

320
Our results confirmed previous studies that highlighted n-3 PUFAs and, to a lesser extent, from grass species rich in n-3 and n-6 PUFAs into tissues [26,33].

325
In addition to ATs, the liver plays a major role in FA metabolism, which argues that and CEs for both horses and ponies (Table 3). Interestingly, SFA levels were quite similar 331 between ATs and liver, but significant differences were found in MUFA and essential n-6 PUFA 332 levels ( Furthermore, the majority of de novo synthesis of non-essential FAs may occur in the AT depots 335 and not in the liver, as has been recently described for equids [36]. Data from the present study, 336 showing significantly higher 16:1n-7/16:0 and 18:1n9/18:0 ratios in the AT depots than in liver 337 tissue, supports these findings by reflecting a higher AT Δ9-desaturase activity (Fig 1A and B).

355
Species-derived differences in the FA profile of selected ATs and liver were found for 356 n-6 PUFAs. Horses had significantly higher n-6 PUFA contents in the ATs and liver compared to PUFAs in the diet to a n-3/n-6 PUFA ratio of 1:1 was evidenced to prevent excessive n-6 414 eicosanoid action and Toll-like receptor 4 (TLR4)-induced production of pro-inflammatory 415 cytokines via an effective blocking of corresponding signalling pathways by n-3 PUFA action.

416
This approach avoids metabolic disorders and is beneficial for health. Use of n-6 PUFA- neutral TAG fraction (44.1% of total FA) as we did. However, data on the long chain FA were 457 missing, which is partly offset by data from the current study.

458
FA levels and lipid profiles were similar between liver and plasma, suggesting plasma as and plasma, which were both different from those in the ATs (Fig 1A-E).