Sie sind auf Seite 1von 5

Exp Physiol 101.1 (2016) pp 23–27

23

SymposiumSymposium ReportReport

Experimental Physiology

Energy metabolism and the high-altitude environment

Andrew J. Murray

Department of Physiology, Development and Neuroscience, University of Cambridge, Cambridge, UK

New Findings What is the topic of this review? This report describes changes in cardiac and skeletal muscle energy metabolism that occur with exposure to high altitude and considers possible underlying mechanisms. What advances does it highlight? In the human heart, sustained hypoxia at high altitude or shorter-term normobaric hypoxia result in a loss of cardiac energetic reserve. In the hypoxic rat heart, fatty acid oxidation and respiratory capacity fall. In skeletal muscle, prolonged exposure to extreme high altitude results in the loss of mitochondrial density, but even at more moderate high altitude the respiratory capacity may be suppressed. Evidence from cells, genetically modified mice and high-altitude-adapted Tibetans suggests a possible mechanistic role for the hypoxia-inducible factor pathway.

At high altitude the barometric pressure falls, challenging oxygen delivery to the tissues. Thus, whilst hypoxia is not the only physiological stress encountered at high altitude, low arterial P O 2 is a sustained feature, even after allowing adequate time for acclimatization. Cardiac and skeletal muscle energy metabolism is altered in subjects at, or returning from, high altitude. In the heart, energetic reserve falls, as indicated by lower phosphocreatine-to-ATP ratios. The underlying mechanism is unknown, but in the hypoxic rat heart fatty acid oxidation and respiratory capacity are decreased, whilst pyruvate oxidationis also lower after sustained hypoxic exposure. In skeletal muscle, there is not a consensus. With prolonged exposure to extreme high altitude (> 5500 m) a loss of muscle mitochondrial density is seen, but this was not observed in a simulated ascent of Everest in hypobaric chambers. At more moderate high altitude, decreased respiratory capacity may occur without changes in mitochondrial volume density, and fat oxidation may be downregulated, although this is not seen in all studies. The underlying mechanisms, including the possible role of hypoxia-signalling pathways, remain to be resolved, particularly in light of confounding factors in the high-altitude environment. In high-altitude-adapted Tibetan natives, however, there is evidence of natural selection centred around the hypoxia-inducible factor pathway, and metabolic features in this population (e.g. low cardiac phosphocreatine-to-ATP ratios, increased cardiac glucose uptake and lower muscle mitochondrial densities) share similarities with those in acclimatized lowlanders, supporting a possible role for the hypoxia-inducible factor pathway in the metabolic response of cardiac and skeletal muscle energy metabolism to high altitude.

(Received 12 July 2015; accepted after revision 20 August 2015; first published online 28 August 2015) Corresponding author A. J. Murray: Department of Physiology, Development and Neuroscience, University of Cambridge, Downing Street, Cambridge CB2 3EG, UK. Email: ajm267@cam.ac.uk

C

2015 The Authors. Experimental Physiology C 2015 The Physiological Society

DOI: 10.1113/EP085317

24

A. J. Murray

Exp Physiol 101.1 (2016) pp 23–27

Introduction

The defining feature of the high-altitude environment is sustained hypobaric hypoxia. Whilst the atmosphere is 21% oxygen at all altitudes, barometric pressure falls upon ascent, and with it inspired P O 2 , challenging oxygen delivery to the tissues. The majority of the world’s population inhabits regions below 1000 m elevation, but nevertheless, our species has a remarkable capacity for hypoxia tolerance. Most notably, over tens of thousands of years some high-altitude populations have adapted to life in this environment, with genetic signatures revealing natural selection around hypoxia-sensing pathways (Bigham & Lee, 2014). Approximately 140 million people live above 2500 m, but around 40 million others venture into high-altitude regions each year for work or leisure (West et al. 2007). These lowlanders experience a physiological response upon ascent, which has been well documented elsewhere and includes ventilatory, cardiovascular and erythropoietic aspects (West et al. 2007). Even with adequate acclimatization, energy metabolism is altered in human heart and skeletal muscle. This might indicate a failure to compensate fully or may itselfform part of the acclimatization process, but it involves changes in gene expression and thus appears to be a regulated response. The positive aspects of high-altitude acclimatization, most notably decreased susceptibility to acute mountain sickness, also contrast with the less well-understood phenomenon of high-altitude deterioration, which occurs with prolonged exposure to extreme high altitude (> 5500 m) and is characterized by lethargy, fatigue and muscle wasting (Ward, 1954). Hypoxia is not, however, the only stress encountered at altitude (West et al. 2007). Temperature falls with increasing elevation, whilst absolute humidity is extremely low and exposure to solar/ultraviolet radiation high. Visitors frequently experience gastrointestinal upset and appetite loss, which could result from the hypoxia itself, but may be exacerbated by infection, particularly in developing countries. In addition, activity levels are often altered, as oxygen delivery limits exercise capacity and motivation falls; thus, individuals may undergo detraining. In light of these confounding factors, the high-altitude environment is not a perfect setting in which to study the physiological response to hypoxia alone, although other contexts (e.g. confinement to normobaric hypoxia chambers) have limitations too. The study of humans (and other animals) at altitude is nevertheless a valuable research endeavour. First, with increasing numbers of people visiting mountainous regions, an understanding of the integrated response to altitude is vital, particularly the interindividual variation in this response, which is not fully explained by classic descriptions of acclimatization. Second, the study of extreme environment physiology

and a consideration of the limits of human tolerance is a stimulating intellectual challenge in its own right and adds to our understanding of homeostatic regulation. Third, exposure to sustained physiological challenge in extreme environments has been proposed as a possible analogue of human disease (Grocott et al. 2007), and whilst this approach risks overinterpretation of similarities between settings it does allow for the study of preselected and otherwise healthy cohorts in adverse conditions. Finally, the natural experiment of high-altitude adaptation in East Africa, the Tibetan Plateau and the Andes offers an invaluable scenario in which to study human genetics and natural selection (Bigham & Lee, 2014).

Cardiac energy metabolism at high altitude

In subjects returning from Everest Base Camp (5300 m), cardiac phosphocreatine-to-ATP ratios (PCr/ATP; measured using 31 P NMR spectroscopy) decreased by 18% (Holloway et al. 2011b), indicating a loss of energetic reserve. This was accompanied by altered diastolic function and a loss of left ventricular (LV) mass, but all measures returned to normal after 6 months back at sea level. In an accompanying study, shorter term exposure to normobaric hypoxia (20 h, 50–60 mmHg atmospheric P O 2 ) also elicited a loss of cardiac energetics (Holloway et al. 2011 a ). The PCr/ATP fell by 15%, and echocardiographic measurements suggested a mild alteration in diastolic function. Whilst different cellular mechanisms may underlie the observations in these studies, in both instances exposure to a hypoxic stimulus resulted in a loss of cardiac energetic reserve. This is reminiscent of findings in the hearts of highland-dwelling Sherpas, who had lower a cardiac PCr/ATP than lowlanders, which persisted even after 4 weeks spent de-acclimatizing at low altitude (Hochachka et al. 1996). Studies of rats in hypoxic chambers have helped to elucidate the cardiac metabolic response to sustained environmental hypoxia. In rats, exposure to 11% O 2 for 1 week resulted in decreased LV expression of genes regulated by the fatty acid (FA)-activated transcription factor peroxisome proliferator-activated receptor α (PPARα ), including carnitine palmitoyl-transferase 1, pyruvate dehydrogenase kinase 4 and uncoupling protein 3, suggesting downregulation of FA oxidation and increased pyruvate oxidation (Essop et al. 2004). Whilst this was associated with decreased O 2 consumption and ATP synthesis, mitochondrial coupling was preserved (Essop et al. 2004). After 12 weeks, expression of FA oxidation genes decreased in both ventricles, whereas expression of pyruvate dehydrogenase kinase 4 mRNA increased in the LV, suggesting inhibition of pyruvate oxidation with more sustained exposure (Adrogue et al.

2005).

C

2015 The Authors. Experimental Physiology C 2015 The Physiological Society

Exp Physiol 101.1 (2016) pp 23–27

Energy metabolism at high altitude

25

In support of this suggestion, 2 weeks at 11% O 2

decreased respiration rates with FA substrates (palmitoyl- carnitine or palmitoyl-CoA) and pyruvate in both subsarcolemmal and intermyofibrillar subpopulations of mitochondria from rat LV/septum. Curiously,

O 2 consumption with the electron transport chain

complex I substrate, glutamate, was lowered in hypoxic subsarcolemmal but not intermyofibrillar mitochondria, alongside decreased complex I protein (Heather et al. 2012). Work from our laboratory confirmed a loss of complex I-supported respiration, FA oxidation capacity and ATP levels in the LV of hypoxic rats (13% O 2 , 2 weeks), alongside increased oxidative stress. All changes were prevented, however, by dietary supplementation with a moderate dose of nitrate, which improved tissue oxygenation, possibly via improved blood flow and/or tissue metabolic efficiency (Ashmore et al. 2014). In the hypoxic rat heart, therefore, FA oxidation is downregulated, with a loss of respiratory capacity in the LV. Pyruvate oxidation may also be inhibited, suggesting increased glycolysis. Whilst little is known about substrate metabolism in the hypoxic human heart, glucose uptake was higher in the hearts of highland natives than lowlanders (Holden et al. 1995).

Skeletal muscle energy metabolism at high altitude

In comparison with studies of cardiac energy metabolism, more numerous investigations have been carried out into the metabolic response of human skeletal muscle at altitude, perhaps owing to the relative ease of obtaining biopsies. This has not, however, resulted in

a consistent picture, perhaps because of varied ascent

profiles, training status and exposure to confounding features of the high-altitude environment. In a recent review, we attempted to find consensus amongst studies of metabolism in human muscle at altitude, plus humans and rodents in hypoxia chambers (Horscroft & Murray, 2014). Considering only human studies, a picture emerges in which the duration and degree of exposure to hypoxia may be crucial. With prolonged exposure to extreme high altitude, as

experienced on Himalayan mountaineering expeditions, a loss of muscle mitochondrial volume density occurs, with the greatest loss from the subsarcolemmal population of mitochondria (Hoppeler et al. 1990; Levett et al. 2012). It is not clear whether this is caused by hypoxia (or the accompanying oxidative stress) or is instead a detraining response, but notably, the subsarcolemmal population of mitochondria is more susceptible to changes in volume density during training studies at sea level (Desplanches et al. 1993). The response does, however, appear to be regulated, with altered expression of genes including

a 50% loss of mRNA of the mitochondrial biogenesis

factor, peroxisome proliferator-activated receptor gamma

co-activator 1 alpha (PGC1 α ), and corresponding changes in protein levels of metabolic regulators and enzymes (Levett et al. 2012). During Operation Everest II (a gradual decompression to the equivalent of 8840 m in a chamber), however, no such change in mitochondrial volume density was observed (Green et al. 1989). This simulation took place over a slightly shorter time frame than a classic Everest ascent, but with the subjects confined to a chamber a greater loss of mitochondria might have been expected due to detraining. The discrepancy has not been resolved, but whatever the cause the loss of mitochondria at extreme altitude fits neatly alongside Ward’s description of high-altitude deterioration (Ward, 1954). At more moderate high altitude, even with prolonged exposure, no such loss in mitochondrial volume density occurs, although notably, lower muscle mitochondrial densities have been reported in some high-altitude natives (Kayser et al. 1991, 1996). Changes in muscle respiratory function occur at more moderate altitudes, but again this may be dependent on the extent of exposure. Two similar high-resolution respirometry studies by Lundby and co-workers described a loss of respiratory capacity and improved coupling following 28 days at 3454 m (Jacobs et al. 2012), but no changes after 9–11 days at

4559 m (Jacobs et al. 2013). Meanwhile, some studies have

reported the downregulation of FA oxidative enzymes with prolonged exposure to altitudes between 4300 and

8848 m (Roberts et al. 1996; Levett et al. 2012), although

no such differences were found in another study at

5250 m (Mizuno et al. 2008). Despite changes in resting

metabolites, however, muscle PCr recovery half-times following an exercise challenge were remarkably well preserved in subjects returning either from Everest Base Camp or the summit, indicating that muscle capacity for ATP synthesis may in fact be spared (Edwards et al. 2010). The recent discovery of genetic selection on the PPARA gene in Tibetan populations (Simonson et al. 2010) suggests that altered FA metabolism may be a feature of long-term adaptation to high altitude, because the gene encodes the FA-activated transcription factor and regulator of FA oxidation, PPAR α . The functional significance of this polymorphism is currently unresolved, but recent work carried out by our group, in collaboration with Professor Erich Gnaiger and Dr Julian Griffin, has used high-resolution respirometry andmetabolic profiling to investigate FA metabolism in the muscles of lowlanders and Sherpas at altitudes up to 5300 m (Gilbert-Kawai et al.

2015).

Mechanisms, unanswered questions and future directions

What mechanisms might underlie the changes in heart and muscle metabolism at altitude? Hypoxia-signalling

C

2015 The Authors. Experimental Physiology C 2015 The Physiological Society

26

A. J. Murray

Exp Physiol 101.1 (2016) pp 23–27

pathways are anintuitive possibility.Many of themetabolic changes reported in humans at altitude or in the hypoxic rat heart (e.g. pyruvate dehydrogenase inhibition and decreased mitochondrial respiration) have also been observed in hypoxic cells in culture and are associated with stabilization of the hypoxia-inducible factor (HIF) family of transcriptionfactors (reviewed byMurray, 2009). Adding support to a HIF-driven mechanism, oxidative metabolism, and thus exercise capacity, were augmented in the gastrocnemius muscle of mice in which skeletal muscle HIF-1α had been selectively deleted (Mason et al. 2007). The effects of the HIF pathway on muscle metabolism are outlined elsewhere in this issue of Experimental Physiology by Dr Helene Rundqvist (Lindholm et al. 2016). An alternative possibility to low muscle P O 2 per se could be reactive oxygen species (ROS)-mediated effects, because mitochondrial ROS production increases in hypoxia (Guzy & Schumacker, 2006) and is modelled to increase sharply in muscle above altitudes corresponding to those of the highest permanent human settlements (Cano et al. 2014). Reactive oxygen species have sometimes been described as indiscriminate mediators of damage to lipids, protein and DNA, and this may be the case when generated in large quantities, but at more moderate concentrations they play an important signalling role within the cell and can, for instance, bring about stabilization of HIF-1 α (Guzy & Schumacker, 2006). Reactive oxygen species-mediated effects, particularly those that occur via interactions with HIF, are difficult to

separate from hypoxia-mediated responses, although the use of antioxidants in studies of hypoxia can help. Indeed, in contrast to the notion of ROS-mediated suppression of mitochondrial function, antioxidant therapy was found to suppress muscle mitochondrial biogenesis in response to training (Gomez-Cabrera et al. 2008). This might suggest that transient production of ROS during training (possibly as a result of acute hypoxia due to high rates of muscle

O 2 consumption) may elicit training-induced changes, in

agreement with a signalling role. Moreover, as outlined elsewhere in this issue of Experimental Physiology by Prof.

Carsten Lundby (Lundby & Jacobs, 2016), the response to hypoxia may in fact mediate some aspects of endurance training in muscle.

Conclusions

The duration and degree of hypoxic exposure are critical to the metabolic response of skeletal muscle, with the response to acute hypoxia during exercise or relatively

mild hypoxia at lower altitudes differing from that seen with more sustained and severe hypoxia following prolonged stays at higher altitudes. In the human heart,

a loss of energetics is common to both short-term and

long-term exposure, although the underlying mechanisms

are probably distinct. Confoundingfactors such as changes in activity or diet at altitude cannot, however, be ruled out. Future studies combining different hypoxic protocols with controlled activity/inactivity and diet could thus be revealing and have the potential to elucidate the mechanisms underlying the metabolic changes in human heart and muscle at high altitude.

References

Adrogue JV, Sharma S, Ngumbela K, Essop MF & Taegtmeyer

H (2005). Acclimatization to chronic hypobaric hypoxia is

associated with a differential transcriptional profile between the right and left ventricle. Mol Cell Biochem 278, 71–78.

Ashmore T, Fernandez BO, Branco-Price C, West JA, Cowburn AS, Heather LC, Griffin JL, Johnson RS, Feelisch M & Murray AJ (2014). Dietary nitrate increases arginine availability and protects mitochondrial complex I and energetics in the hypoxic rat heart. J Physiol 592,

4715–4731.

Bigham AW & Lee FS (2014). Human high-altitude adaptation:

forward genetics meets the HIF pathway. Genes Dev 28,

2189–2204.

Cano I, Selivanov V, Gomez-Cabrero D, Tegner´ J, Roca J,

Wagner PD & Cascante M (2014). Oxygen pathway modeling estimates high reactive oxygen species production above the highest permanent human habitation. PLoS One 9, e111068. Desplanches D, Hoppeler H, Linossier MT, Denis C, Claassen

H, Dormois D, Lacour JR & Geyssant A (1993). Effects of

training in normoxia and normobaric hypoxia on human muscle ultrastructure. Pflugers Arch 425, 263–267. Edwards LM, Murray AJ, Tyler DJ, Kemp GJ, Holloway CJ, Robbins PA, Neubauer S, Levett D, Montgomery HE, Grocott MP & Clarke K; Caudwell Xtreme Everest Research Group (2010). The effect of high-altitude on human skeletal muscle energetics: P-MRS results from the Caudwell Xtreme Everest expedition. PLoS One 5, e10681. Essop MF, Razeghi P, McLeod C, Young ME, Taegtmeyer H & Sack MN (2004). Hypoxia-induced decrease of UCP3 gene expression in rat heart parallels metabolic gene switching but fails to affect mitochondrial respiratory coupling. Biochem Biophys Res Commun 314, 561–564. Gilbert-Kawai E, Sheperdigian A, Adams T, Mitchell K, Feelisch M, Murray A, Peters M, Gilbert-Kawai G, Montgomery H, Levett D, Kumar R, Mythen M, Grocott M & Martin D (2015). Design and conduct of Xtreme Everest 2: an observational cohort study of Sherpa and lowlander responses to graduated hypobaric hypoxia. F1000Res 4, 90. Gomez-Cabrera MC, Domenech E, Romagnoli M, Arduini A, Borras C, Pallardo FV, Sastre J & Vi na˜ J (2008). Oral administration of vitamin C decreases muscle mitochondrial biogenesis and hampers training-induced adaptations in endurance performance. Am J Clin Nutr 87, 142–149. Green HJ, Sutton JR, Cymerman A, Young PM & Houston CS (1989). Operation Everest II: adaptations in human skeletal muscle. J Appl Physiol 66, 2454–2461. Grocott M, Montgomery H & Vercueil A (2007). High-altitude physiology and pathophysiology: implications and relevance for intensive care medicine. Crit Care 11, 203.

C

2015 The Authors. Experimental Physiology C 2015 The Physiological Society

Exp Physiol 101.1 (2016) pp 23–27

Energy metabolism at high altitude

27

Guzy RD & Schumacker PT (2006). Oxygen sensing by mitochondria at complex III: the paradox of increased reactive oxygen species during hypoxia. Exp Physiol 91,

807–819.

Heather LC, Cole MA, Tan JJ, Ambrose LJ, Pope S, Abd-Jamil AH, Carter EE, Dodd MS, Yeoh KK, Schofield CJ & Clarke K (2012). Metabolic adaptation to chronic hypoxia in cardiac mitochondria. Basic Res Cardiol 107, 268. Hochachka PW, Clark CM, Holden JE, Stanley C, Ugurbil K & Menon RS (1996). 31 P magnetic resonance spectroscopy of the Sherpa heart: a phosphocreatine/adenosine triphosphate signature of metabolic defense against hypobaric hypoxia. Proc Natl Acad Sci USA 93, 1215–1220. Holden JE, Stone CK, Clark CM, Brown WD, Nickles RJ, Stanley C & Hochachka PW (1995). Enhanced cardiac metabolism of plasma glucose in high-altitude natives:

adaptation against chronic hypoxia. J Appl Physiol 79,

222–228.

Holloway C, Cochlin L, Codreanu I, Bloch E, Fatemian M, Szmigielski C, Atherton H, Heather L, Francis J, Neubauer S, Robbins P, Montgomery H & Clarke K (2011 a). Normobaric hypoxia impairs human cardiac energetics. FASEB J 25,

3130–3135.

Holloway CJ, Montgomery HE, Murray AJ, Cochlin LE, Codreanu I, Hopwood N, Johnson AW, Rider OJ, Levett DZ, Tyler DJ, Francis JM, Neubauer S, Grocott MP & Clarke K; Caudwell Xtreme Everest Research Group (2011b). Cardiac response to hypobaric hypoxia: persistent changes in cardiac mass, function, and energy metabolism after a trek to Mt. Everest Base Camp. FASEB J 25, 792–796. Hoppeler H, Howald H & Cerretelli P (1990). Human muscle structure after exposure to extreme altitude. Experientia 46,

1185–1187.

Horscroft JA & Murray AJ (2014). Skeletal muscle energy metabolism in environmental hypoxia: climbing towards consensus. Extrem Physiol Med 3, 19. Jacobs RA, Boushel R, Wright-Paradis C, Calbet JA, Robach P, Gnaiger E & Lundby C (2013). Mitochondrial function in human skeletal muscle following high-altitude exposure. Exp Physiol 98, 245–255. Jacobs RA, Siebenmann C, Hug M, Toigo M, Meinild AK & Lundby C (2012). Twenty-eight days at 3454-m altitude diminishes respiratory capacity but enhances efficiency in human skeletal muscle mitochondria. FASEB J 26,

5192–5200.

Kayser B, Hoppeler H, Claassen H & Cerretelli P (1991). Muscle structure and performance capacity of Himalayan Sherpas. J Appl Physiol 70, 1938–1942. Kayser B, Hoppeler H, Desplanches D, Marconi C, Broers B & Cerretelli P (1996). Muscle ultrastructure and biochemistry of lowland Tibetans. J Appl Physiol 81, 419–425. Levett DZ, Radford EJ, Menassa DA, Graber EF, Morash AJ, Hoppeler H, Clarke K, Martin DS, Ferguson-Smith AC, Montgomery HE, Grocott MP & Murray AJ; Caudwell Xtreme Everest Research Group (2012). Acclimatization of skeletal muscle mitochondria to high-altitude hypoxia during an ascent of Everest. FASEB J 26, 1431–1441.

Lindholm ME & Rundqvist H (2016). Skeletal muscle hypoxia-inducible factor-1 and exercise. Exp Physiol 101,

28–32.

Lundby C & Jacobs RA (2016). Adaptations of skeletal muscle mitochondria to exercise training. Exp Physiol 101,

17–22.

Mason SD, Rundqvist H, Papandreou I, Duh R, McNulty WJ, Howlett RA, Olfert IM, Sundberg CJ, Denko NC, Poellinger

L & Johnson RS (2007). HIF-1α in endurance training:

suppression of oxidative metabolism. Am J Physiol Regul Integr Comp Physiol 293, R2059–R2069. Mizuno M, Savard GK, Areskog NH, Lundby C & Saltin B (2008). Skeletal muscle adaptations to prolonged exposure

to extreme altitude: a role of physical activity? High Alt Med Biol 9, 311–317. Murray AJ (2009). Metabolic adaptation of skeletal muscle to high altitude hypoxia: how new technologies could resolve the controversies. Genome Med 1, 117. Roberts AC, Butterfield GE, Cymerman A, Reeves JT, Wolfel EE

& Brooks GA (1996). Acclimatization to 4,300-m altitude

decreases reliance on fat as a substrate. J Appl Physiol 81,

1762–1771.

Simonson TS, Yang Y, Huff CD, Yun H, Qin G, Witherspoon DJ, Bai Z, Lorenzo FR, Xing J, Jorde LB, Prchal JT & Ge R (2010). Genetic evidence for high-altitude adaptation in Tibet. Science 329, 72–75. Ward M (1954). High altitude deterioration. Proc R Soc Lond B Biol Sci 143, 40–42. West JB, Schoene RB & Milledge JS (2007). High Altitude Medicine and Physiology. Hodder Arnold, London.

Additional information

Competing interests

None declared.

Funding

This report received no direct funding.

Acknowledgements

The author wishes to thank the many colleagues and collaborators with whom he has had the pleasure of carrying out research in this area, but particularly Tom Ashmore, Kieran Clarke, Lindsay Edwards, Martin Feelisch, Edward Gilbert-Kawai, Erich Gnaiger, Julian Griffin, Mike Grocott, Cameron Holloway, Hans Hoppeler, James Horscroft, Randall Johnson, Aleksandra Kotwica, Verena Laner, Denny Levett, Daniel Martin and Hugh Montgomery. The author also thanks the British Heart Foundation, Research Councils UK, BBSRC, Action Medical Research, Oroboros Instruments and theWYNG Foundation for supporting research in his laboratory.

C

2015 The Authors. Experimental Physiology C 2015 The Physiological Society