Running Speed in Mammals Increases with Muscle n-6 Polyunsaturated Fatty Acid Content Authors: Thomas Ruf, Teresa Valencak, Frieda Tataruch, Walter Arnold Published: 2006, PLOS ONE DOI: 10.1371/journal.pone.0000065 ABSTRACT Polyunsaturated fatty acids (PUFAs) are important dietary components that mammals cannot synthesize de novo . Beneficial effects of PUFAs, in particular of the n-3 class, for certain aspects of animal and human health (e.g., cardiovascular function) are well known. Several observations suggest, however, that PUFAs may also affect the performance of skeletal muscles in vertebrates. For instance, it has been shown that experimentally n-6 PUFA-enriched diets increase the maximum swimming speed in salmon. Also, we recently found that the proportion of PUFAs in the muscle phospholipids of an extremely fast runner, the brown hare ( Lepus europaeus ), are very high compared to other mammals. Therefore, we predicted that locomotor performance, namely running speed, should be associated with differences in muscle fatty acid profiles. To test this hypothesis, we determined phospholipid fatty acid profiles in skeletal muscles of 36 mammalian species ranging from shrews to elephants. We found that there is indeed a general positive, surprisingly strong relation between the n-6 PUFAs content in muscle phospholipids and maximum running speed of mammals. This finding suggests that muscle fatty acid composition directly affects a highly fitness-relevant trait, which may be decisive for the ability of animals to escape from predators or catch prey. INTRODUCTION Comparative studies on locomotor performance and maximum running speed (MRS) in mammals have mostly concentrated on biomechanical models. Factors such as skeletal structure, gravity, bone-length and width, or peak stress on muscles and tendons affect MRS and explain several differences in locomotor performance of mammals of different body size [1] , [2] . Research on muscles specialized on high-frequency contractions, such as the tail shaker muscle of rattlesnakes, has revealed that on the cellular level, muscle cell performance can be modified by both protein isoforms (of troponin, myosin, and ion pumps) and quantitative adaptations of the densities of sarcoplasmatic reticulum (SR) and mitochondrial membranes [3] . To our knowledge, however, it has not been investigated previously whether these membranes also may show systematic qualitative changes concerning their fatty acid composition, namely PUFA content or relative abundance of the n-3 and n-6 PUFA subclasses (which have their first double bond at the 3 rd and 6 th carbon of the fatty acid chain, respectively), that could be associated with MRS across mammals. Several previous studies suggest that PUFAs may indeed affect the performance of skeletal muscles and locomotion in vertebrates. In Atlantic salmon, for instance, dietary fatty acid composition and resulting changes in muscle lipid composition significantly affected maximum swimming speed. These effects were largely due to a positive relation between swimming speed and the most common n-6 fatty acid in muscle lipids, linoleic acid (C18:2n-6) [4] . Also, muscles that undergo very high contraction frequencies, such as the pectoral muscle in hummingbirds, contain very large amounts of docosahexanoic acid (DHA, C22:6n-3), a highly unsaturated PUFA [5] . In humans, training exercise increases muscle PUFA content [6] . Further, we recently found that the proportions of PUFAs in the muscle phospholipids of an extremely fast runner, the brown hare ( Lepus europaeus ), are very high compared to other mammals [7] .Therefore, we hypothesized that muscle phospholipid profiles may be functionally related to locomotor performance, and were interested to see if there is a relation between either total PUFA content, any of the PUFA subclasses, or a particular fatty acid, and locomotor function as measured by running speed. RESULTS MRS increased with body weight in a nonlinear manner ( Fig. 1A ) and levelled off among very large mammals. After adjusting for this body size effect we found that MRS increased significantly with the proportion of total muscle phospholipid PUFAs ( Table 1 ). This association was, however, almost entirely due to a positive relation between MRS and the n-6 PUFA class only, which was the best predictor of MRS in this comparison ( Fig. 1B , Table 1 ). Accordingly, among individual fatty acids, contents of the two most common fatty acids of the n-6 class, Linoleic acid (C18:2n-6) and Arachidonic acid (C20:4n-6) were also positively related to MRS (Linoleic acid: slope = +0.015, t = +3.113, P = 0.0038; Arachidonic acid : slope = +0.012, t = +2.10, P = 0.043). Species with specialized morphological adaptations designed for extremely high running speeds, such as the cheetah ( Acinonyx jubatus ) and blackbuck ( Antilope cervicapra ), are still much faster than predicted from body size and phospholipid n-6 PUFA content alone ( Fig. 1 ), but overall, the relation between muscle membrane composition and MRS was surprisingly strong. When back-transformed to actual velocities, the regression line in Fig. 1B indicates that predicted MRS increased from 13.0 to 49.2 km h −1 as n-6 PUFA content increased from 12.2 to 61.5%. This corresponds to a speed gain of 0.734 km h −1 for each percent increase in the proportion of n-6 PUFAs. (A) MRS as a function of body weight. (B) The relation of MRS to n-6 PUFA content, and to (C) n-3 PUFA content in phospholipids of skeletal muscles. Shown are partial residual plots (i.e. the independent effect of each predictor variable) from general additive mixed models using a smooth term for body weight (estimated df = 2.66, F = 4.77, P = <0.001) and linear terms for n-6 and n-3 PUFA content. To adjust for phylogenetically caused correlations models included the family membership of each species as a random factor. Partial regression coefficients and statistics from 6 GAMM models of muscle phospholipid components from log 10 MRS as a function of log 10 body weight and each fatty acid class (%) as well as the unsaturation index (UI) (n = 36). The taxonomic classification in terms of family membership of each species was entered as a random factor to adjust for phylogenetically caused correlation. The relation between MRS and the proportion of n-3 PUFAs, on the other hand, was slightly negative ( Fig. 1C ), but due to a large scatter in n-3 PUFA abundance, this association was statistically not significant ( Table 1 ). Because n-3 PUFA content dominated species-specific differences in the degree of fatty acid unsaturation, the unsaturation index (UI, i.e. the total number of double bonds per 100 acyl chains), was also unrelated to MRS ( Table 1 ). However, since low levels of n-6 PUFAS were largely replaced by n-3 PUFAs and vice versa (r = −0.71, P<0.001) it is not surprising that the single best predictor of MRS was the difference between n-6 PUFA and n-3 PUFA content (slope+0.0102, t = +4.559, P<0.0001). MRS was not significantly affected by saturated fatty acid (SFA) content, but decreased as the proportion of monounsaturated fatty acids (MUFAs) increased ( Table 1 ). This negative relation between MRS and MUFAs was not independent of the impact of PUFAs however, because MUFA content was negatively correlated with the abundance of both n-6 PUFAs alone (r = −0.549, P<0.001) and, even more strongly, with total PUFA content (r = −0.930, P<0.0001, Fig. 2 ). The saturated fatty acid (SFA) content, on the other hand, varied much less than the other membrane constituents and showed a considerably weaker correlation with PUFA content (r = −0.352, P = 0.035). Skeletal muscle MUFA content was highly determined by PUFA content (% MUFA = 64.18−0.872×% PUFA; t = −14.76, P<0.0001; R2 = 0.865, n = 36.) We found no evidence for the possibility that our finding of a positive relation between MRS and n-6 PUFAs, or the inverse relation to MUFAs, could have been caused by the inclusion of species with a low number of specimens in which fatty acid profiles were determined. A bootstrap analysis based on random combinations of single specimens clearly confirmed the statistical significance of associations between n-6 PUFAs, total PUFAs, and MUFAs ( Protocol S2 ). The above results on the relation between muscle phospholipid fatty acid composition and MRS were further corroborated by phylogenetic GLS regression models (see Material and Methods, using linear terms for both body weight and phospholipid components). Partial regression coefficients from phylogenetic models for the effects of PUFAs, n-6 PUFAs, and MUFAs were statistically significant ( Protocol S3 ) and showed the same direction of effects and similar effect sizes as indicated by our GAMM approach of data analysis. Phylogenetic GLS models indicated, however, that due to a high degree of similarity within clades, body weight had no independent effect on MRS when adjusted for phylogenetic distances between species. Partial regression coefficients for log 10 body weight from 6 models (testing for the effects of different fatty acids and UI) ranged from −0.005 to +0.049 and were not significantly different from zero (0.39
0.10 in all cases). Details on data and sources are given in Protocol S1 . Branch lengths are arbitrary. LIPID EXTRACTION AND ANALYSIS Lipids were extracted from muscle samples (0.5 g each) and lipid classes were separated on silica gel thin layer chromatography plates. All solvents contained butylhydroxytoluol in order to avoid oxidative modification of PUFAs. Phospholipids were extracted into hexane, extracts were transesterified under nitrogen, and analysed by GLC (Perkin Elmer Autosystem XL with autosampler and FID; Norwalk, USA) using a capillary column (HP INNOWax, 30 m×0.25 mm; Hewlett Packard, USA). Fatty acid methyl esters were identified by comparing retention times with those of standards (Sigma-Aldrich, St. Louis, USA) and peaks were integrated using the Turbochrom 4.1 Software (Perkin Elmer, Norwalk, USA). Details about chemical analyses and substances used are given elsewhere [7] . For 27 species, we determined the proportions of the following phospholipid acids: C14:0, C15:0, C16:0, C17:0, C18:0, C16:1n-7, C18:1n-9, C18:2n-6, C18:3n-3, C20:4n-6, C20:5n-3, C22:5n-3, and C22:6n-3. Fatty acid percentages were combined into classes of SFA, MUFA, PUFA, UFA according to their degree of unsaturation and, additionally, PUFAs were attributed to the n-3 and n-6 family. For all animals an unsaturation index was computed [22] , which represents the average number of double bonds per 100 fatty acid molecules (UI = (%MUFA×1)+(%Dienoic×2)+(%Trienoic×3)+(%Tetraenoic×4)+(%Pentaenoic×5)+(%Hexaenoic×6) ). Sample sizes within species ranged from 1 to 244 individuals. We felt justified to include single specimens because fatty acid concentrations showed only little to moderate within-species variation (coefficients of variation for all fatty acid (classes) investigated ranged from 2.3% to 36.67%, overall mean: 15.68%). Nonetheless, we computed additional bootstrap statistics (see below) to test for the possible effects of including species for which no replicates were available. LITERATURE DATA For 9 further species, we used published data on muscle phospholipid composition, in which contents of the same set of fatty acids as determined in our laboratory was given (except for 3 cases in which C14:0, C15:0 or C17:0 had not been determined due to their negligible amount (typically<1%)). We computed mean proportions for each species when more than one source was available. Data on body weight were obtained, if possible, from the actual specimen used for tissue collection (n = 24), or taken from the literature. Also, data on MRS were taken from publications, except for two cases of personal communications of experts on the species in question (see Protocol S1 ), and one case of using own (TR) unpublished observations on Glis glis . STATISTICAL ANALYSES To assess the impact of body weight and phospholipid fatty acid concentration on MRS we used General Additive Mixed Models (GAMM), because this approach allowed us to fit smooth curves (penalized cubic regression splines) to the nonlinear relation of MRS to body weight, and to include possible effects of taxonomic relations. To adjust for phylogenetically caused correlations (e.g., within-clade morphological similarities that may affect MRS), we initially entered order, family, and genus of each species as nested random effects (which corresponds to the nested ANOVA approach for simple linear models). Initial results indicated that inclusion of the factors order and genus decreased the adjusted coefficients of determination and led to over-determined models. Therefore, we included only the taxonomic level family in our final models. Unlike body weight, partial effects of the proportions of fatty acid classes on MRS were best described by simple linear terms ( Fig. 1 ) and were entered as such. GAMMs were computed using the library mgcv [23] within the statistical package R [24] . Because the distribution of MRS was skewed to the right we log transformed MRS. Note that the GAMM models used are an extension of multiple linear regression. Hence, they assess the independent contributions of each independent variable to the prediction of the dependent variable. Thus, using this procedure we statistically eliminated possibly confounding effects of body mass on both MLSP and phospholipid fatty acid composition. We used multiple regression, rather than comparing residuals from individual regressions against body weight because it avoids bias in the estimated coefficients, and computes P-values based on the correct degrees of freedom (with all variables entered into a single model) [25] . To test for the influence of including data with very different levels of uncertainty for each species (i.e., replicates of fatty acid composition ranging from n = 1 to n = 244) we employed a bootstrap procedure. Details on this procedure and its results are given in Protocol S2 . To validate our main results on the relation between phospholipid fatty acids and MRS, using more detailed information on the taxonomic relation between the species investigated [26] , [27] , we computed phylogenetically corrected regressions as outlined in Protocol S3 . SUPPORTING INFORMATION Original data and sources. (0.10 MB DOC) Bootstrap analysis of regression models. (0.02 MB DOC) Phylogenetically corrected regressions. (0.03 MB DOC) REFERENCES [1] Biewener AA 2005 Biomechanical consequences of scaling. J Exp Biol 208 1665 1676 [2] Iriarte-Diaz J 2002 Differential scaling of locomotor performance in small and large terrestrial mammals. J Exp Biol 205 2897 2908 [3] Rome LC Lindtsted SL 1998 The Quest for Speed: Muscles Built for High-Frequency Contractions. News Physiol Sci 13 261 267 [4] McKenzie DJ Higgs DA Dosanjh BS Deacon G Randall DJ 1998 Dietary fatty acid composition influences swimming performance in Atlantic salmon ( Salmo salar ) in seawater. Fish Physiol Biochem 19 111 122 [5] Infante JP Kirwan RC Brenna JT 2001 High levels of docosahexaenoic acid (22:6 n-3)- containing phospholipids in high-frequency contraction muscles of hummingbirds and rattlesnakes. Comp Biochem Physiol B 130 291 298 [6] Helge J Wu B Willer M Daugaard J Storlien L 2001 Training affects muscle phospholipid fatty acid composition in humans J Appl Physiol 90 670 677 [7] Valencak TG Arnold W Tataruch F Ruf T 2003 High content of polyunsaturated fatty acids in muscle phospholipids of a fast runner, the European brown hare ( Lepus europaeus ). J Comp Physiol B 173 695 702 [8] Garland T 1983 The relation between maximal running speed and body mass in terrestrial mammals. J Zool London 199 157 170 [9] Hulbert AJ Rana T Couture P 2002 The acyl composition of mammalian phospholipids: an allometric analysis. Comp Biochem Physiol B 132 515 527 [10] Hulbert AJ 2003 Life, death and membrane bilayers. J Exp Biol 206 2303 2311 [11] Bennett AF 1991 The evolution of activity capacity. J Exp Biol 160 1 23 [12] Weyand PG Bundle MW 2005 Energetics of high-speed running: integrating classical theory and contemporary observations. Am J Physiol Regul Integr Comp Physiol 288 R956 R965 [13] Swanson JE Lokesh BR Kinsella JE 1989 Ca2+-Mg2+ ATPase of mouse cardiac sarcoplasmatic reticulum is affected by membrane n-6 and n-3 polyunsaturated fatty acid content. J Nutr 119 364 372 [14] Ushio H Ohshima T Koizumi C Visuthi V Kiron V 1997 Effect of dietary fatty acids on Ca 2+ -ATPase activity of the sarcoplasmic reticulum of rainbow trout skeletal muscle. Comp Biochem Physiol B118 681 691 [15] Wu BJ Else PL Storlien LH Hulbert AJ 2001 Molecular activity of Na/K-ATPase from different sources is related to the packing of membrane lipids. J Exp Biol 204 4271 4280 [16] Hulbert AJ Else PL 1999 Membranes as possible pacemakers of metabolism. J Theor Biol 199 257 274 [17] Turner N Lee JS Bruce CR Mitchell TW Else PL 2004 Greater effect of diet than exercise training on the fatty acid profile of rat skeletal muscle. J Appl Physiol 96 974 980 [18] Ayre KJ Hulbert AJ 1997 Dietray fatty acid profile affects endurance in rats. Lipids 32 1265 1670 [19] Kriketos AD Pan DA Sutton JR Hoh JFY Baur LA 1995 Relationships between muscle membrane lipids, fiber type and enzyme activities in sedentary and exercised rats. Am J Physiol 269 R1154 R1162 [20] Blackard WG Li J Clore JN Rizzo WB 1997 Phospholipid fatty acid composition in type I and type II rat muscle. Lipids 32 193 198 [21] Nikolaidis MG Petridou A Mougios V 2006 Comparison of the phospholipid and triacylglycerol fatty acid profile of rat serum, skeletal muscle and heart. Physiol Res 55 259 265 [22] Couture P Hulbert AJ 1995 Membrane fatty acid composition of tissues is related to body mass of mammals. J Membr Biology 148 27 39 [23] Wood SN 2004 Stable and efficient multiple smoothing parameter estimation for generalized additive models. J Amer Statist Assoc 99 673 686 [24] R Development Core Team 2006 R: A language and environment for statistical computing (R Foundation for Statistical Computing, Vienna, Austria) ( http://www.R-project.org ) [25] Freckleton RP 2002 On the misuse of residuals in ecology: regression of residuals vs. multiple regression. J Anim Ecol 71 542 545 [26] Arnason U Adegoke JA Bodin K Born EW Esa Y 2002 Mammalian mitogenomic relationships and the root of the eutherian tree. Proc Natl Acad Sci USA 99 8151 8156 [27] Murphy WJ Eizirik E Johnson WE Zhang YP Ryder OA 2001 Molecular phylogenetics and the origins of placental mammals. Nature 409 614 618