Introduction: Road cycling performance is mainly determined by maximal oxygen uptake (VO2max), fractional utilization of VO2max and gross efficiency (GE). However, performance is also determined by anaerobic capacity, especially in short and/or repeated sprints. A high volume of low-intensity endurance training (LIT) improves the ability to perform well in 4-6 h competitions in cyclists (Laursen, 2010, Schumacher and Mueller, 2002) and is shown to upregulate markers of mitochondrial biogenesis and angiogenesis (Skovgaard et al., 2016, Psilander et al., 2010). However, supplementation of training at higher intensities is needed to optimize both aerobic and anaerobic capacities. Training modalities such as repeated 30-s all-out sprints increases markers of mitochondrial biogenesis and angiogenesis (Taylor et al., 2016) but also markers of improved skeletal muscle ion-handling abilities (Iaia et al., 2008). These latter factors potentially explain the improved sprint performance seen after such training (Hostrup and Bangsbo, 2017). Interestingly, the combination of LIT and sprints induces superior increases in markers of mitochondrial biogenesis compared to LIT-only in trained subjects (Skovgaard et al., 2016). In the current study, our main aim was to investigate the effect of adding 30-s sprints to a long-lasting low-intensity ride on muscle signalling, GE, recovery of contractile function in highly trained cyclists. Method: 12 highly trained, male cyclists (VO2max: 73.4 ± 4.0 (Siebenmann et al., 2015) completed a randomized, cross-over design containing a 4-h LIT session (50% of VO2max) with 9 x 30-s sprints included (E&S) or a work-matched 4-h LIT session (E). GE, iEMG from m. Vastus lateralis (VL) and pedal force measurement were recorded in aerobic steady-state periods during exercise. Isokinetic leg extension was performed and muscle samples were collected from VL prior to exercise, as well as 20 min, 3 h and 24 h (isokinetic leg extension only) post exercise. Results: Expression of markers of mitochondrial biogenesis, PGC-1 splice 1, PDK4 and Tfam, increased (P<0.05) or tended to increase (P<0.10) in both conditions. The increase in PGC-1 splice 1 was smaller (P<0.05) after E&S than after E (Figure 1), whereas the increase in PDK4 was greater after E&S than after E. The angiogenic regulators HIF-, VEGFA-stim and thrombospondin increased (P<0.05) to a similar extent in the two conditions. Na+/K+1 mRNA expression was increased (P<0.05) after E only, with the relative change being different (P<0.05) from E&S. ClC1 and NHE1 mRNA expression was decreased (P<0.05) after E&S only, with the relative change being different (P<0.05) from E. GE was reduced (P<0.05) after the first set of sprints in E&S, with the relative change tending to be different from E (P=0.06). GE was reduced to the same extent after 4 hour of cycling in both conditions; 0.9±0.4 percentage points (PP) in E&S and 1.1±0.3 PP in E (Table 1; all P<0.05). Coherent with the decreased GE, the ventilation increased in an invert manner. Neither condition exhibited changes in RER, the angle of peak torque during the pedal stroke, pedalling cadence or iEMG. No correlation was seen between changes in either of these variables and changes in GE. Peak torque in isokinetic leg extension was similar at 24 h post-exercise between conditions.