Both acute block of Sh activity (DTx) and loss of function of Sh expression significantly reduced IKfast ( Figure 3B; WT 40.5 ± 1.9 versus WT + DTx 29.3 ± 2.7 versus Sh[14] 26.1 ± 1.7 pA/pF; p ≤ 0.01 and p ≤ 0.01, respectively). Moreover, the IKfast recorded in dMNs under both conditions JAK phosphorylation (WT + DTx 29.3 ± 2.7 and Sh[14] 26.1 ± 1.7 pA/pF) was indistinguishable from that of vMNs in WT (26.1 ± 2.3 pA/pF, DTx p = 0.38, Sh p = 1), which is in full agreement with our model. To further support the notion that the difference in IKfast that exists between dMNs and vMNs is due, at least in part, to expression of Sh in dMNs, we recorded IKfast in vMNs under the same conditions. As expected,

neither the presence of DTx, nor loss of Sh, had any marked effect on IKfast in vMNs (p = 0.51 and 0.23, respectively; Figure 3B). To further verify the differential expression of Sh in dMNs versus vMNs we assessed transcription of Sh in these two cell types by in situ hybridization. We designed probes that specifically recognize the Sh pre-mRNA. These intron probes label the unspliced Sh transcript at the site of transcription within the nucleus, but not the fully mature message in the cytoplasm. We detected Sh transcription in dMNs, labeled with Eve antibody ( Figure 3C, black arrowheads), but not in vMNs, labeled by expression of

GFP (Lim3 > nlsGFP; Figure 3D, white arrowheads). Taken together, both electrophysiology and in situ hybridization are consistent Autophagy animal study with dMNs expressing Sh while the vMNs do not. Next, we tested whether Islet is sufficient to repress Sh-mediated K+ currents in cells where Sh, but not islet, is normally expressed. We used two different preparations for these experiments. First, we ectopically expressed islet in dMNs. Driving a UAS-islet transgene with GAL4RN2-0 significantly reduced IKfast (34.4 ± 2.6 versus 41.2 ± 1.9 pA/pF, experimental versus controls which consisted of WT and heterozygous GAL4 driver line, p

≤ 0.05; Figure 4A). These recordings were carried out in the presence of external Cd2+ to eliminate Ca2+-dependent K+ currents. The observed reduction in IKfast in dMNs could, however, be due to a reduction in either Sh- or Shal-mediated K+ currents. To distinguish between these two possibilities, we tested for DTx sensitivity, which is observed in WT dMNs and is an indicator for the presence of Sh currents. Casein kinase 1 DTx sensitivity was lost when islet was ectopically expressed in dMNs ( Figure 4A). In addition, when we expressed ectopic islet in dMNs in a Sh−/− background, there was no further reduction in IKfast compared to ectopic islet expression in a WT background ( Figure 4A). We conclude from this that ectopic expression of islet in dMNs is sufficient to downregulate Sh-mediated IKfast. The second preparation we used takes advantage of the fact that IKfast in body wall muscle is solely due to Sh and Slowpoke (the latter of which can be easily blocked [Singh and Wu, 1990]).