LAUSR

Dear Editor, Voltage-gated potassium channels (KV) play vital roles in electrically excitable and non-excitable cells. They usually open with membrane depolarization and allow the flow of K ions. Ion flow through these channels is curtailed by time-dependent entry into non-conducting inactivated states. Inactivation allows channels to close even in the face of continued stimulation and can occur rapidly (in ms, N-type) or slowly (in s, C-type). This tight governance of ion flow by inactivation is essential for the timing and control functions of KV channels. KV1.3 was the first ion channel discovered in immune cells three decades ago and exhibited only C-type inactivation. During antigen presentation, the channel clusters at the immunological synapse and promotes Ca signaling. Effector memory T cells up-regulate KV1.3 during activation. Many toxins from scorpions, sea anemones, and parasitic worms block KV1.3 by binding to an external vestibule at the outer entrance to the channel’s pore. Protein engineering of these peptide toxins has resulted in selective KV1.3 inhibitors that preferentially suppress proliferation, cytokine secretion, and in vivo migration of effector memory T cells. One inhibitor, dalazatide, advanced to human trials where it ameliorated symptoms in patients with plaque psoriasis. These physiological and therapeutic importance of KV1.3 motivated us to elucidate its molecular structure. To produce a stable homogeneous sample for structural study, we removed residues 1–52 of human KV1.3; these residues are absent in mouse and rat KV1.3. The channel’s voltage-dependence of activation, use-dependent inactivation, and sensitivity to the KV1.3-specific inhibitor ShKEWSS at low picomolar concentrations (Fig. 1a) matched that of KV1.3 in T cells . Since the auxiliary subunit KVβ2 was reported to promote expression of hKV1.3 , we expressed hKV1.3 with KVβ2.1, purified the complex, and finally obtained an overall 3.2 Å resolution map (Supplementary Fig. S1). The hKV1.3–KVβ2.1 complex assembled as a tetramer with a four-fold symmetry (Fig. 1b). Each subunit of KV1.3 contained a transmembrane domain (TMD) and a cytoplasmic T1 domain, which was a docking platform for the auxiliary β subunit. The TMD consisted of a voltage sensor domain (VSD, helices S0–S4), which responded to changes in membrane potential, and a poreforming domain (helices S5–S6). The KV1.3 T1 domain, helices S5–S6, and KVβ2.1 were at higher resolution (~2.5–3.5 Å), which allowed for accurate model building. The local resolution for the VSD was between 4 and 5 Å. As a result, most side chains were invisible in these regions. We built the VSD model based on its strong main chain density and the corresponding region in crystal structure of KV1.2-2.1 chimera (PDB ID 2R9R) . The final model includes KVβ2.1 with a NADP + molecule (Supplementary Fig. S2a), which is a cofactor for KVβ2 subunit , the T1 domain, and the TMD. Human KV1.3–KVβ2.1 complex exhibits overall dimensions of ~140 Å × 100Å × 100 Å, and the length and width of the TMD are ~55 and ~80 Å, respectively (Fig. 1b). Unsurprisingly, the overall architecture of hKV1.3–KVβ2.1 is remarkably similar to that of rat KV1.2-2.1 chimera-KVβ2 10 (Supplementary Fig. S2b). By aligning the TMDs of KV1.3 and the chimera, we observed a small shift of the KV1.3 T1 domain and KVβ2.1 (Supplementary Fig. S2c), which was also mentioned in the cryo-EM structure of the chimera in nanodiscs comparing to its crystal structure.