In this issue of The Journal of Physiology, Boudreault et al. (2010) present an interesting phenomenon resulting from inactivation or absence of the skeletal muscle surface membrane KATP channel. Muscles with non-functional KATP channels were subjected to two fatigue protocols separated by at least 15 min. During the second fatigue protocol compared to the first, tetanic force was maintained at a higher level, and the concentration of free myoplasmic Ca2+ ([Ca2+]i) and force during the inter-tetanic intervals showed only modest increases. The reason for the improved performance in the second fatigue protocol remains unclear. The KATP channel is abundant in the sarcolemma of striated muscle. It consists of four Kir6.x subunits that form the pore and four SURx ATP binding subunits (for recent review see Flagg et al. 2010). As the name suggests, ATP is the controller of the channel keeping it closed when [ATP] is above 1 mm (Spruce et al. 1985). Other modulators of the KATP channel, e.g. H+, PIP2, and phosphorylation, have also been described but at normal [ATP]i, their effect is modest. Interestingly, electrophysiological, pharmacological and gene analysis data indicate that while the predominant form of the striated muscle KATP channel is composed of Kir6.2 and SUR2A complexes, there can also be significant levels of Kir6.1, SUR2B and SUR1 in some muscles. The Kir6.1 and Kir6.2 pores have widely different conductances (35 and 80 pS, repectively) and this together with the heterogeneity of the SUR subunits underlies the maximal KATP current density seen in different mouse muscles, ranging from 10 pA μm−2 in the slow soleus to 80 pA μm−2 in the fast tibialis anterior (Tricario et al. 2006). The striking behaviour of the channel in opening at low levels of ATP led to its examination as a factor in muscle fatigue. A decline in muscle force or power output (commonly called fatigue) occurs when muscles are repeatedly contracted. Exercise involves a huge increase in ATP consumption and an attractive hypothesis advanced was that as [ATP]i declines, muscle KATP channels would open, partially depolarise the membrane and impair action potential activated release of Ca2+ from the sarcoplasmic reticulum. Thus, there would be a beneficial effect in that muscle energy consumption would be reduced. Despite the fact that [ATP]i rarely falls by more than 25% even during intensive exercise, the concept of ATP compartmentalisation and possibility of far greater declines in [ATP] close to the surface membrane KATP channel allowed this hypothesis to thrive. Pharmacological manipulations of the membrane KATP channel have yielded equivocal results. Opening the channel accelerated the loss of force but so too, in many but not all experiments, did blockers of the membrane KATP channel. When Kir6.2 knockout mice were tested, it was obvious that tetanic force production was markedly less than in wild-type mice. Fatigue developed more rapidly in muscles from Kir6.2 knockout mice than in those from wild-type mice (Cifelli et al. 2007). Boudreault et al. show in the present study that this rapid development of fatigue could be largely prevented by an earlier fatigue run. Thus, the role of KATP channels in fatigue is dubious. If the membrane KATP channel is not important in fatigue, then what is its role? One clue comes from the finding in the present paper of Boudreault et al. that muscles with non-functional membrane KATP channels have a slower recovery of force after the end of the fatiguing exercise. This slower recovery of force after exercise was preceded by a marked elevation in inter-tetanic concentration of internal Ca2+, which entered the muscle cells through verapamil sensitive channels, presumably L-type channels. It is well known that prolonged elevated [Ca2+]i leads to impaired force production and locally activates Ca2+-dependent proteases (Murphy & Lamb, 2009). Thus, the presence of functional membrane KATP channels may assist in limiting the influx of Ca2+ during sustained contractile activity. But for the fortunate individual without a mutation of Kir6.2 or the SUR subunits of the KATP channel, this is not a factor that will limit exercise capacity. Mutations of the subunits in the membrane KATP channel that lead to clinical manifestations are fortunately rare. One of the better known is the developmental delay, epilepsy and neonatal diabetes (DEND) syndrome which is accompanied by muscle flaccidity and motor impairment. Recently, a mouse model carrying a human mutation known to produce the DEND syndrome was developed. When expressed only in the muscles, no muscle or motor deficit was observed. However, when expressed in neurons, clear muscle and motor impairments were found (Clark et al. 2010). These results again reiterate the point that under normal daily activities, non-functional KATP channels in the membrane of skeletal muscle have little consequence but when the isolated muscle is fatigued to the extreme in vitro, some impairment may be seen.