Task-5 needs family members to leave its "silence"

Abstract

The invention relates to a method for the identification, resp. characterisation of compounds which are especially effective and / or have no, resp. less, negative side-effects compared to the state of the art when used for treating different illnesses. The method for the identification of a compound being suitable to be used as active pharmaceutical ingredient comprises the following steps (i) to (iii): (i) Expression of a TASK-5 containing heteromeric channel complex in vertebrate cells or insect cells, the TASK-5 containing heteromeric channel complex being a heteromeric channel complex of a vertebrate species or a heteromeric channel complex of an insect species or a heteromeric channel complex of a bacteria species, so that cells are acquired having the TASK-5 containing heteromeric channel complex located in the outer cell membrane. (ii) Performing at least one measurement of the activity of the TASK-5 containing heteromeric channel complex being located within the outer cell membrane of the cells acquired according to step i) in the presence of the compound and in the absence of the compound, thus receiving values of the activity of the TASK-5 containing heteromeric channel complex, measured in presence of the compound and values of the activity of the TASK-5 containing heteromeric channel complex measured in absence of the compound. (iii) Comparing the values acquired according to step ii) in presence of the compound with the values acquired according to step ii) in absence of the compound, wherein the compound is identified as active pharmaceutical ingredient if the values measured in presence of the compound differ mathematically significantly from the values measured in absence of the compound.

Description

[0001] Patent Application

[0002] TASK-5 needs family members to leave its "silence"

[0003] The invention relates to a method for the identification, resp. characterisation of compounds which are especially effective and / or have no, resp. less, negative side-effects compared to the state of the art when used for treating different illnesses.

[0004] Description

[0005] Field of the invention

[0006] TASK-5 (KCNK15) channels belong to the acid-sensitive subfamily of two-pore domain potassium (K2P) channels including TASK-1 and TASK-3. Since TASK-5 was described as non-functional in heterologous expression systems, the role of this K2P channel remained elusive, even twenty years after its first description. For naming of the members of the acid-sensitive subfamily of two-pore domain potassium (K2P) channels generally two different variants (with or without dash) are used synonymously herein; for example, TASK3 is synonymous to TASK-3, TASK5 is synonymous to TASK-5 etc.

[0007] Background of the technology

[0008] The TASK-5 channel, encoded by KCNK15, is a member of the acid-sensitive subfamily of K2P channels including TASK-1 and TASK-3. K2P channels have emerged as major pharmacological drug targets for a wide range of human diseases, i.e. ranging from depression, migraine, atrial fibrillation, sleep apnea to pulmonary hypertension. The TASK-5 channel is on the other hand, one of the few K2P channels for which almost no data is available, as it was considered, by four independent studies describing the initial cloning of this channel in 2001 , to be a non-functional or an intracellular channel. However, its distinct expression pattern in humans in adrenal gland, pancreas, liver, kidney, lung, ovary, testis and heart does not support the idea of TASK-5 being an intracellular ion channel, as those would rather show a homogenous or wide spread expression pattern. Another peculiar observation was a common heterozygous polymorphism in the selectivity filter of KCNK15 (TASK-5G95E) which would destroy, in parts of our population, at least for one allele the channel function of an already non-functional channel. As in two of the initial reports for the cloning of TASK-5 co-expression experiments with TASK-1 or TASK-3 failed to indicate a heteromerization, it was concluded, that TASK-5 does not form heteromers within the TASK family. Thus, over twenty years, it was concluded, that TASK-5 may require a still unidentified accessory protein to form functional channels in the plasma membrane or it may form a channel in an intracellular organelle.

[0009] Content of the invention

[0010] Here we describe various approaches to functionally express TASK-5 channels, with the result that homomeric TASK-5 remained a “silent” K2P channel. Here we found that TASK-5 is engaged in the formation of heteromeric channel complexes with TASK-1 and TASK-3. While TASK-5 negatively modulates the surface expression of TASK channels, the heteromeric TASK-5 channel complexes which are present at the plasma membrane, are characterized by alterations in singlechannel conductance, Gq-coupled receptor-mediated channel inhibition and pharmacology. The here identified unique pharmacology of TASK-1 / TASK-5 heterodimers, which is in addition modulated by the TASK-5G95Epolymorphism, should be considered in the development of future drugs targeting TASK-1 channels in diseases like atrial fibrillation, sleep apnea and pulmonary arterial hypertension. TASKS resists to form homomeric channels at the plasma membrane

[0011] Until today, it was not possible to record TASK-5 channels in different heterologous expression systems. Also in our hands, currents of TASK-5 injected oocytes, were undistinguishable from those of non-injected oocytes (Figure 1 B,C). Previously described attempts to gain functional TASK-5 channels included for instance experiments probing chimeric channels with TASK-3, the removal of a putative N- terminal di-arginine (2RR3) endoplasmic reticulum retention sequence, the exchange of the C-termini of TASK3 and TASK-5 or recordings at different physiological extracellular pH values between 6.0 and 8.4. In the current study we applied a brought approach to record currents of TASK-5 channels.

[0012] First, we examined whether more alkaline extracellular pH values may result in channel function, similar as observed for TALK channel family members like TASK- 4. However, also very alkaline solutions with a pH of up to 10.5 did not lead to a functional expression (Figure 1 D). Recently, we described an inner gate in TASK-1 channels, the so-called “X-gate” structure, which stabilizes the closed state of the channel and provides the molecular correlate for drug trapping and the very low open probability of TASK-1. The X-gate structure is stabilized by an interaction of the late M4 segment with regions in the early M1 and distal M2 segment forming the so called “latch” region of the X-gate. As the amino acid sequence of the X-gate region is highly similar within the TASK family, it is very likely, that TASK-3 and TASK-5 may also contain a similar X-gate like structure. Therefore, we studied mutations in the latch region, that caused in TASK-1 a strong destabilization of the inner gate and a gain-of-function, in TASK-5, as the channel might be non- conductive due to a very stable X-gate. However, the corresponding mutations localized in the early N-terminus or the distal M2 segment, TASK-5R7Dand TASKS’*1310(Figure 1A, Figure 7), did not lead to functional TASK-5 expression (Figure 1 E)

[0013] After studying chimeric channels of TASK-5 with TASK-3, Karschin et al. proposed that the first extracellular linker and the M2 region contribute to the nonfunctionality of TASK-5. Strikingly, this region involves the extracellular M1 -P1 linker which is now known as the “cap structure” that is essential for the dimeric assembly of K2P channels. Since a cysteine residue, at the tip of the cap structure, albeit not conserved in all K2P channels, was discussed to be involved in K2P channel dimerization, we introduced a cysteine residue at the corresponding amino acid 53 in TASK-5. However, also theTASK-5G53Cmutant did not yield any currents (Figure 1 F). We previously found that the inner leaflets of the cap domains can interact with each other in order to stabilize the cap structure and identified several amino acid residues at the tip of the cap structure of TASK-1 (52YNLS55), as essential for the functional expression. This52YNLS55sequence also contains an N-glycosylation site and glycosylation at this site was postulated to be important for TASK-1 and TASK- 3 channel stability. To make the TASK-5 cap structure more TASK-1 -like and compensate for a putative weakness in the self-assembly or stability of the channel, we mutated the tip of the cap structure of TASK-5 from52FGFS55to52YNLS55(Figure 1A), but again, also this TASK-5 channel construct was not functional (Figure 1G).

[0014] Most of the approaches to functionally express TASK-5 were performed with storage solutions that contain the antibiotic gentamycin and / or the phosphodiesterase inhibitor theophylline. As the antibiotic or high cAMP levels might have either blocked TASK-5 or interfered with its functional expression, we removed those drugs from the storage solution, however without any success to record TASK-5 mediated currents (Figure 1 H).

[0015] Studying the surface expression of TASK-5 containing an extracellular HA epitope (TASK-5HA’Ex) (position of the tag see Figure 1A) in Xenopus laevis oocytes, using an ELISA based luminometric assay, the channel was not detected at the plasma membrane, whereas TASK-1HA’Ex, serving as a positive control, was robustly expressed at the cell surface (Figure 11). In fluorescence imaging experiments TASK-5 was localized in intracellular compartments, presumably, given the typical web-like fluorescence pattern, primarily in the ER (Figure 8). The idea that TASK-5 channels get stuck in the ER was, as mentioned above, previously probed by mutating the putative N-terminal ER retention signal2RR3. Here, we reinvestigated the role of2RR3and additional putative di-arginine signals that were predicted in the M2-M3 linker at position137RR138, as well as in the cytosolic C- terminus at position278RR27. However, mutating these putative ER retention signals did not gain any functional TASK-5 expression (Figure 1 J).

[0016] Binding of 14-3-3 to the distal C-terminus of TASK-1 and TASK-3 promotes the functional expression of these two channels at the plasma membrane. As TASK-5 also contains a putative C-terminal binding motif for 14-3-3 binding, we coexpressed 14-3-3s or 14-3-3^ with TASK-5 in the hope to measure TASK-5 channels at the plasma membrane. However, also after co-expression with 14-3-3 we could not measure any TASK-5 currents (Figure 1 K).

[0017] TASKS forms heterodimers with TASK-1 and TASKS changing channel composition at the plasma membrane

[0018] TASK-5 is co-expressed with TASK-1 and TASK-3 in the human heart, pulmonary VSMCs, neurons that regulate respiration, as well as the nasopharynx (human protein atlas). VSMC’s are vascular smooth muscle cells. These co-expression patterns in tissue relevant for the development of TASK-1 modulators against atrial fibrillation (AFib), sleep apnea or pulmonary arterial hypertension (PAH), tempted us to re-investigate whether TASK-5 channels heteromerize with other TASK channel family members, despite that heteromerization was previously not observed. The expression modulator is denoting herein a chemical compound being able to activate or to inhibit a TASK channel. A modulator may modulate a single homomeric TASK channel, e.g. a TASK-1 channel or a heteromeric TASK channel, e.g. a heteromeric TASK-1 / TASK-5 channel. Other diseases for which the development of inhibitors and activators of heteromeric TASK-1 / 3 channels containing TASK-5 could be relevant include primary hyperaldosteronism (PHA), diabetes, obesity, central sleep apnea, hypertension, tinnitus and various forms of malignant cancers. The PHA indication relates to the fact that TASK channels are strongly expressed in the adrenal cortex and that TASK-1 knockout mice exhibit a primary hyperaldosteronism. Obesity is an indication because TASK channels are expressed in brown adipose tissue (BAT), where they serve as marker proteins for BAT, potentially playing a role in oxidative dyscoupling. Diabetes is a putative indication for drugs targeting heteromeric TASK channel complexes containing TASK-5, because TASK-1 channels are associated with diabetes and TASK-1 knockout mice show an altered insulin secretion. TASK-1 channels have been identified as therapeutic targets for central sleep apnea (for instance gain-of function mutations in TASK-1 channels lead to developmental delay and sleep apnea (DDSA)) and TASK channels are under investigation as drug targets against obstructive sleep apnea. TASK channel block has been reported to affect the blood pressure and unspecific ion channel blockers (local anesthetics) which also block TASK channels are effective in models against tinnitus. It is to be noted that - with respect to the subject matter of the current invention - the term blocker refers to all types of compounds being able to reduce the activity of a TASK channel, independent from the special mechanism by which the reduction of the activity is achieved. Thus, a blocker according to the invention may either reduce the activity of a TASK channel by binding to a TASK protein within the pore of the TASK channel or by binding to a TASK protein outside but close to the pore of the TASK channel or by binding to a TASK protein apart from the pore of the TASK channel. In this invention, blocker is used mechanistically as a synonym for inhibitor. This invention does not make a mechanistic distinction between blockers and inhibitors. According to the definition of a blocker, the phrase blocking is to be understood as comprising all types of reduction of the activity of a TASK channel, independent of the specific mechanism of reduction of the activity. Similarly, all types of activation mechanisms are covered in this invention. The role of TASK-5 in these tissues is still unclear, yet heteromeric channels are promising targets again hypertension or tinnitus. TASK channels are dysregulated in a variety of cancer types and are being discussed as promising new drug targets. We have also shown that TASK-3 inhibitors modulate cell division. The role of TASK-5 containing heteromers has not been studied in most cancers. For all the diseases discussed above, the development of TASK-5 containing heteromer-specific modulators (inhibitors and / or activators) would be of great potential therapeutic significance. In addition, TASK-5 is not sufficiently studied in tissue like the ganglia cells of the retina, fibroblast of the heart and immune cells, where other TASK channels play a role and therefore TASK-5 containing heteromeric channels could also be targets against other diseases like retinopathies, different cardiac arrhythmias or autoimmune disorders. The different modulators, i.e. compounds being identified with the method revealed herein as being able to modulate the activity of heteromers of TASK-proteins (e.g. TASK-1 + TASK-5, TASK-3 + TASK-5, TASK-1 + TASK-3), are advantageously be used as active pharmaceutical ingredients (API - active pharmaceutical ingredient). Due to their specifity in binding and modulating the TASK proteins they exert tremendous effects upon the physiological activity of the TASK proteins and show less unwanted side effects upon usage as APIs. Heteromerization within the family of acid-sensitive K2P channels, of TASK-1 and TASK-3, was already described by us and others. Here we tested whether TASK-5 may, in contrast to the current expectations, form heterodimers with its subfamily members TASK-1 and / or TASK-3. First, we coexpressed TASK-1 (Figure 2A) or TASK-3 (Figure 2C) with increasing amounts of TASK-5. In both cases, TASK-5 decreased current amplitudes in a concentrationdependent manner (Figure 2A-D). Analyzing the effect of TASK-5 on the surface expression of extracellularly HA-tagged TASK-1HA’Exor TASK-3HA’Exwe found, that the current reduction was paralleled by a reduced surface expression of both channels in the presence of TASK-5 (Figure 2E,F). As a control, TASK-3 coexpression with TASK-1HA-Exled to an increased surface expression of TASK-1 , while TASK-1 co-expression with TASK-3HA’Exdid not change the robust TASK-3 signal at the plasma membrane (Figure 2E), as previously described. Next we probed, whether TASK-1 or TASK3 may assist TASK-5 to reach the cell surface. To this end we co-expressed an extracellularly HA-tagged TASK-5 channel construct (TASK-5HA’Ex) with non-tagged TASK1 (Figure 2G) or TASK-3 (Figure 2H) subunits. Strikingly, co-expression with TASK-1 as well as TASK-3 induced a localization of TASK-5HA-Exat the plasma membrane with highly significant fluorescent signals (Figure 2G,H). Moreover, this surface expression of TASK-5 by co-expression with either of the two TASK channels was further increased in a concentration-dependent manner (Figure 2G,H).

[0019] Co-expression with selectivity filter dead mutants in which the GYG sequence is changed to EYG is an approach commonly used to probe for heteromerization of K2P potassium channels. As our data strongly indicate a heteromerization of TASK- 5 with TASK-1 and TASK-3, we also co-expressed TASK-1 or TASK-3 with TASK- 5G95E. As a control, we co-expressed wild-type TASK-3 channels or the dominantnegative TASK-3G95Emutant with TASK-1 or TASK-3 (Figure 2I,J). Here, as previously described, TASK-3 co-expression with TASK-1 led to a drastic increased channel amplitude compared to TASK-1 injected alone (Figure 2I) and coexpression of a TASK-3G95Emutant with TASK-1 or TASK-3 led to strong dominantnegative effects with a concentration dependent current reduction (Figure 2I, J), confirming previous data and the validity of our experiments. In the same experimental setting, TASK-5G95Edid not conduct currents on its own (Figure 2K,L), while it reduced the current amplitude of TASK-1 and TASK-3 in a dominant- negative manner and significantly stronger, than by the co-expression with wild-type TASK-5 (Figure 2K,L). Summarizing, our data strongly promote the hypothesis, that TASK-5 forms heterodimers with TASK-1 and TASK-3, changing channel composition at the cell surface.

[0020] Inside-out single-channel patch-clamp recordings support the hypotheses of a formation of heteromeric TASKS channel complexes

[0021] To further support the hypothesis that TASK-5 forms heterodimers with other TASK channel family members, we performed inside-out single-channel patch-clamp recordings of Xenopus oocytes to search for single-channel events characterized by heteromeric TASK-3 / TASK-5 channels. To this end, we expressed TASK-3 alone or together TASK-5. TASK-3 expressed alone yielded characteristic single-channel events with an average amplitude of about 8.2 pA at -100 mV, whereas coexpression of TASK3 with TASK-5 led to a novel additional conductance with a single-channel amplitude of about 5.5 pA at -100 mV, presumably reflecting TASK- 3 / TASK-5 heteromers (Figure 3A-B). Figure 3A illustrates representative recordings from patches, containing either (i) TASK-3 alone, (ii) heterodimeric TASK-3 / TASK-5 channels or (iii) TASK-3 together with heteromeric TASK-3 / TASK- 5 channels. After injection of TASK-3 cRNA alone, a TASK-3-like channel was observed in 15 out of 15 patches. The co-expression of TASK-3 and TASK-5 yielded TASK-3-like events in 18 / 27 patches, while in 9 / 27 patches a heterodimeric TASK- 3 / TASK-5-like channel were observed (Figure 3C). In five out of these nine patches containing a TASK-3 / TASK-5-like channels, an additional TASK-3-like channel was present in the patch, while in four of these patches only the heteromeric channel was detected (Figure 3C). The single-channel conductance (SCC) of the heteromer-like channel was about 93 pS whereas the SCC of TASK3 was, similar as previously described, about 113 pS (Figure 3D). While the open times of the heteromer-like channels was with To of 1.98 ms highly similar to that of TASK-3 channels (T0= 1.92 ms) (Figure 3E), there were strong differences in the closed times for the heteromer-like channels (Figure 3F,G). Here, the short closed time (TCI ) were prolonged from 55.1 ms for TASK-3 to 171.3 ms for the heteromeric channels and the long closed times (TC2) were prolonged from 340.6 ms for TASK- 3 to 982.0 ms for the TASK-3 / TASK-5-like channels, respectively. Consistent with this prolonged closed time, the NPo of heteromeric TASK-3 / TASK-5-like channels was with about 0.01 % very low and about one third of that of TASK-3 alone.

[0022] Taken together, the patch-clamp experiments revealing an additional conductance with specific single-channel properties further strongly support the idea that there is a formation of heteromeric TASK-3 / TASK-5 channel complexes. The reduction in the single-channel conductance, the increased closed times together with the reduced surface expression, contributes to the reduced channel amplitudes of TASK-5 heterodimers with TASK channels recorded in voltage-clamp experiments.

[0023] TASKS alters the gating of TASK heterodimers

[0024] The results described above reveal an altered gating of the heteromeric channel complexes with TASK-5. Thus, we probed next for TASK-5-dependent changes in the extracellular pH gating of TASK-1 and TASK-3. However, co-expression of TASK-5, did not induce significant changes in the pHso values of TASK-1 (Figure 4A) or TASK-3 (Figure 4B). As for TASK-1 and TASK-3 a role in thermosensation was proposed, we also analyzed temperature-dependence of TASK-5 containing heterodimers. However, TASK-5 did not influence the thermosensitivity (Q10 values) of both, TASK-1 and TASK-3 channels (Supplemental Figure S3). However, the single-channel recordings of TASK-3 / TASK-5 heteromeric channels revealed prolonged closed times, indicating a stabilization of the closed state, which we postulated for TASK-1 to occur at the so called X-gate. As described above R7 and R131 are essential for the stabilization of the inner gate of TASK-1 and mutation of these residues led to a strong current increase in TASK-1 currents, but not in homomeric non-functional TASK-5 channels (Figure 1 E). Therefore, we probed whether these non-functional TASK-5 mutants exert effects on the putative X-gate in heteromeric channel complexes with TASK-1 or TASK-3. To this end, we coexpressed TASK-5R7Dor TASK-5R131 Dwith TASK-1 (Figure 4C,E,G) or TASK-3 (Figure 4D,F,H). In the case of TASK-1 , both mutants led to a significant increase in current amplitudes compared to the co-expression of wild-type TASK-5 (Figure 4C,E), while the surface expression was not significantly altered (Figure 4G). These data demonstrate that TASK-5 is involved in heteromeric complexes with TASK-1 and that these heteromeric channels must have an inner X-gate-like structure that can be de-stabilized by mutants, similar as in TASK-1 channels. Also for heteromeric channels with TASK-3, TASK-5R7Dled to a current increase, while here however, TASK-5R131 Ddid not alter the gating (Figure 4D,E,H). Thus, TASK-5 is involved in heteromerization with TASK-1 and TASK-3, while the X-gate-like structure of heteromeric TASK-5 channels may not be completely similar between TASK-1 and TASK-3.

[0025] As our data shows that TASK-5 channel subunits are involved in the formation of an inner X-gate-like structure in heteromeric channel complexes with TASK-1 and TASK-3, and since Gq-coupled inhibition of homomeric TASK-1 and TASK-3 channels mechanistically occurs at the inner gate, we next probed for altered Gq- coupling of TASK-5 heteromers. Here, we co-expressed TASK-1 or TASK-3 together with TASK-5 and the ai -adrenoreceptor and compared the receptor- mediated channel inhibition via application of the ai -adrenoreceptor methoxamine (10 pM) (Figure 4I-J). TASK-1 current amplitudes were strongly diminished due to Gq-coupled receptor activation (Figure 4I), while this reduction was less efficient and somewhat transient for TASK-3 channels, similar as previously described (Figure 4J). Interestingly, co-expression of TASK-5 did not alter the Gq-receptor mediated inhibition of TASK-3 channels (Figure 4J), while the receptor-mediated inhibition of heteromeric TASK-1 / TASK-5 channels was strongly reduced. The reduced Gq-coupling, specific to heteromeric TASK-1 / TASK-5 channel complexes, might be of physiological relevance in tissue were these both channels are coexpressed. Similar as for the X-gate mutants, heteromeric TASK-1 and TASK-3 channel complexes containing TASK-5 subunits behaved differentially, indicating a structural variability at the inner gate, leading to the differences in the functional Gq- coupling that we have observed here.

[0026] TASK-5 / TASK-1 heterodimers show specific pharmacological properties

[0027] TASK-1 blockers are promising drugs for the treatment of AFib, obstructive and central sleep apnea (OSA and CSA) while TASK-1 activators might be highly beneficial for the treatment of both heritable and non-heritable forms of PAH. Thus it is of highest interest, whether the pharmacology of TASK-1 / TASK-5 heterodimers is different compared to that of TASK-1. We tested different TASK1 blockers, including BAY100493, BAY2586116, A293 and A1899 and Doxapram (Figure 5B, Supplemental Figure S4). For BAY100493, BAY2586116 and A293, an unexpected and very strong increase in sensitivity was observed for the TASK- 1 / TASK-5 heteromeric channels with ICso values changing from 19.7 nM to 3.0 nM, 27.6 to 1.0 nM and 85.7 to 6.6 nM, respectively (Figure 5A,B). In contrast, A1899 sensitivity was not different to TASK-1 (20.0 nM versus 27.4 nM) (Figure 5A,B). Thus, TASK-1 / TASK-5 heterodimers have a unique pharmacology, which is different to that of TASK-1 , with an almost up to 30-fold increase in affinity in the case of BAY2586116. Highly potent TASK-1 blockers are trapped in the central cavity, whereas A1899 which is binding also to residues below the X-gate is not trapped. Thus, the increased affinity for selected TASK-1 blockers might be explained by a more pronounced trapping in heteromeric channels, which is in line with altered gating at the X-gate (increased closed times, reduced Gq-coupling) of heteromeric channels.

[0028] Systematic studies of drug-binding sites in TASK-1 revealed that the L122, located in the M2 segment, is as an essential residue for the binding of highly potent TASK-1 blockers. This leucine residue is highly conserved in the TASK subfamily. Therefore, we wondered, whether the TASK-5L122Awhich does not conduct currents in homomeric channels (not illustrated) may change the drug affinity in heteromeric channels with TASK-1. For BAY100493 and A293 a strongly reduced drugsensitivity was observed and ICso shifted about 45.8- and 28-fold, respectively (Figure 5C,D), whereas in the case of BAY2586116 and A293 almost no changes in drug-sensitivity were observed (Figure 5C,D). The fact that the co-expression of a silent channel which harbors a pore mutation alters the pharmacology of drugs that bind to the central cavity of TASK-1 , again confirms the formation of TASK-5 containing heteromeric channel complexes.

[0029] Ruthenium Red (RR) is a commonly used valuable tool, to discriminate currents in native tissue that are mediated by TASK-3 from those conducted by TASK-1 , since only TASK-3 is RR sensitive. Czirjak et al. reported that the cationic dye inhibited homomeric TASK-3 channels, whereas TASK-1 homomers and TASK- 1 / TASK-3 heteromers were not affected, since two E70 residues are necessary for the inhibition by this compound and only TASK-3 contains a glutamate at this position (K70 in TASK-1 ). In our experiments TASK-3 / TASK-5 heteromers were equally sensitive to RR (5 pM) as TASK3 (Figure 5E,F), whereas TASK-1 and TASK-1 / TASK-5 heteromers showed no RR sensitivity (Figure 5E,F). The fact that TASK-5 channels do not introduce a RR-sensitivity into heteromeric TASK-1 channel complexes or remove RR-sensitivity of heteromeric complexes with TASK- 3 is in perfect agreement with the fact that TASK-5 contains a glutamate at position 70 and that in the dimeric channel two E70 residues are required to render the channel RR sensitive.

[0030] K2P channels are activated at the level of the selectivity filter by so called negatively charged activators (NCAs). However, it was not reported whether TASK- 1 channels, despite that these channels have an additional inner gate, respond to NCAs. TASK-1 activating compounds, are however, promising novel drugs for the treatment of heritable and non-heritable forms of PAH. We tested activation of heteromeric channels by different NCAs and found that NCAs also activate TASK- 1 K2P channels, however less efficiently (Figure 5G,H). Strikingly, PD118057 caused a potent activation of heteromeric TASK-1 / TASK-5 channel complexes, while there was only a minor activation of homomeric TASK-1 channels (Figure 5G). Also ML67-33 showed stronger activation of heteromeric TASK-1 / TASK-5 channels (Figure 5H), while other NCAs, had similar efficiencies to activate heteromeric channels containing TASK-5, as NS3623 or BL-1249, or the drugs caused less activation, as in the case of NS11021 (Figure 5H). In conclusion, TASK-5 containing heterodimers show a unique pharmacology that should be considered in future studies regarding basic research in native tissue and in the research and development of novel drugs targeting TASK channels.

[0031] About half of the general population carries a TASK-5G95Epolymorphism in one allele (46%) and about 31 % are homozygous for this variant (GnomAD Exac database). This variant corresponds to the non-conducting selectivity filter dead TASK-5G95Emutant (GYG sequence is changed to EYG). As described above this variant acts dominant-negative and the current amplitudes of TASK-1 co-expressed with only TASK-5G95E, reflecting the homozygous state, further suppressed the currents compared to the co-expression of the two wild-type channels. Here the question arises, if homozygous polymorphism carriers may have altered drugsensitivities as TASK-5G95Econtaining TASK-1 / TASK-5 heteromers which are nonconducting might not be able to confer higher drug-sensitivity to TASK-1 as wildtype TASK-5 does. To this end, we heterologously expressed TASK-1 (Figure 6A, upper panel), TASK-1 plus TASK-5 (Figure 6A, lower panel) or TASK-1 plus TASK-5G95E(Figure 6A, lower panel) in oocytes and applied the activator PD118057 (Figure 6B). While currents of co-expressed TASK-1 / TASK-5 were activated by about 45%, currents of co-expressed TASK-1 / TASK-5G95Ewere only activated by 31 % (Figure 6B,C). Similarly, the TASK blocker A293 showed significantly reduced effects on TASK-5G95Econtaining heteromers compared to wild-type heteromeric TASK-1 / TASK-5 channels (Figure 6D,E). As a control, for A1899, a drug that did not show an increased affinity for heteromeric TASK-1 ZTASK- 5 channels, co-expression of TASK-1 with TASK-5G95Edid not alter drug-affinity (Figure 6F,G). In conclusion, the enhanced drug-sensitivity of heteromeric TASK-5 channels might be altered for some drugs in native tissue, especially for homozygous polymorphism carriers which might be considered in future clinical trials or pharmacogenomics studies.

[0032] Ever since the cloning of TASK-5 two decades ago, it was thought that TASK-5 is non-functional and that it may need to associate with another protein to form a functional channel or that it only acts as a channel in intracellular compartments. This perception mainly emerged, as heteromerization of silent TASK-5 channels with TASK-1 or TASK-3 was not further considered, after two of the three initial cloning papers postulated that TASK-5 is not engaged in heteromerization. Therefore all attempts in the past to functionally express homomeric TASK-5 channels had to fail if the channel is only active in heteromeric complexes. Our study however, convincingly shows that TASK-5 is involved in heteromeric channel complexes at the plasma membrane, not only since the selectivity filter mutant TASK-5G95Esuppressed TASK-1 and TASK-3 currents in a dominant-negative manner (similar as TASK-1 currents are suppressed by TASK-3G95E), but also since the co-expression with the “silent” TASK5 channel altered the single channel properties, the Gq-receptor coupling and the pharmacology of TASK channel family members. In addition, co-expressing mutants of the “silent” TASK-5 also altered the gating (TASK-5R7D) and the pharmacology (TASK-5L122A) of TASK channels. All these changes listed above must occur at plasma membrane and strikingly our surface expression assays revealed that TASK-5 reaches the plasma membrane when co-expressed with TASK-1 or TASK-3 channels. Clearly TASK-5 alone is not expressed at the plasma membrane and TASK-1 and TASK-3 assist TASK5 channels to reach the surface membrane. However, the exact mechanism why homomeric TASK-5 channels are retained in intracellular compartments, presumably the ER, still remains elusive, as we and others could not identify a functional retention signal in TASK-5. For sure, it needs the heteromerization with another TASK family member to get the “ticket” to the plasma membrane. Without being bound to a certain theory it may be that one or more of the following effects are at work:

[0033] (i) the heteromeric TASK-5 containing complexes may be more stable than the homomeric TASK-5 complexes and therefore can reach the outer membrane just simply due to their greater stability;

[0034] (ii) it may be that only the heteromeric TASK-complexes do associate with trafficking-relevant proteins, thus being guided to outer membranes, while TASK-5 homomeric complexes cannot associate with trafficking-relevant proteins and therefore cannot be guided to outer membranes.

[0035] Strikingly, TASK-1 and TASK-3 harbor either di-acidic forward transport signals and / or allow 14-3-3 binding that increase surface expression, while such a di-acidic motif is not present in TASK-5 and 14-3-3 did not rescue surface expression of homomeric TASK-5 channels.

[0036] TASK-1 channel blockers are promising drugs targeting AFib, OSA and CSA, while there is a high medical need to develop TASK-1 activators for the treatment of PAH. Thus, given the reported co-expression of TASK-5 with TASK-1 and / or TASK-3 in the corresponding disease-relevant human tissue, the altered pharmacology of TASK- 1 / TASK-5 heterodimers that we have observed for blockers as well as activators, might be highly relevant for the future drug development. Strikingly, we found for two drugs that were in clinical trials against OSA (BAY2586116) or AFib (A293=AVE1231 ) strong changes in affinity. For the activators tested, we even found evidence for a drug that preferentially acts on heteromeric TASK-1 / TASK-5 channels (PD118057), which might be desirable for the treatment of PAH, given the fact that TASK-5 is in pulmonary VSMCs one of the most strongly expressed potassium channels together with TASK-1 . On the other hand designing TASK-1 blockers that do not block heteromeric TASK-1 / TASK-5 channels might be beneficial for the treatment of AFib, as these might avoid a PAH as putative side effect, previously discussed for TASK-1 blockers. Overall, the unique pharmacology of TASK-1 / TASK-5 heterodimers should be considered in future therapy approaches targeting TASK-1 channels. The experimental results provided herein, are revealed as examples only, and are in no way to be understood as being confining the scope of the invention.

[0037] Besides altering the Gq-coupled receptor inhibition of TASK-1 , TASK-5 primarily modulates the currents of TASK-1 and TASK-3, by a reducing the surface expression and gating of the heteromeric channel complexes. As mentioned above, about half of the general population carries a G95E polymorphism in one allele (46%) and about 31 % are homozygous for this variant. The resulting amino acid exchange renders the selectivity filter dead. Our data suggest that this polymorphism works synergistically in the suppression of TASK1 and TASK-3 currents by TASK-5. While wild-type TASK-5 already alters the trafficking and gating of TASK-1 and TASK-3, the TASK-5G95Evariant will cause an additional abrogation of the gating at the selectivity filter itself. To which extent this polymorphism contributes to a TASK-1 and TASK-3 suppression in vivo remains unclear currently, as it will depend on the transcription levels of TASK-1 , TASK-3 and TASK-5 in different native tissue. In addition, this polymorphism might in some tissue of homozygous TASK-5G95E'carriers’ result in a reduced number of conductive TASK- 1 / TASK-5 heteromeric channels at the plasma membrane, meaning a functional loss of those heteromeric channels with the hallmark of an enhanced drugsensitivity. Therefore, TASK-1 inhibition by some blockers might, depending on the targeted tissue and disease, result in responding or non-responding patients, similar as observed in the KOALA clinical OSA study utilizing the TASK-1 blocker BAY2586116 or as frequently observed in human medicine in general. Thus, this polymorphism should be considered in future clinical and / or pharmacogenomics studies.

[0038] However, the involvement of K2P channels in autoimmune diseases and cancer should not be underestimated. Similar to other K+channels, K2PS have recently emerged as novel targets in cancer disease. This can be clearly seen in the example of TASK-3, here a pro-oncogenic potential was demonstrated in cell lines and in various tumor entities. Although TASK channels are widely studied in cancer, the role of TASK-5 in cancer remained elusive, as the channel was thought to be nonfunctional and / or not involved in heteromerization with other TASK channel family members. Strikingly, TASK-5 is dysregulated for instance in pancreatic cancer or hepatocellular carcinoma. Thus, the information that TASK-5 is engaged in heteromerization and can be characterized by its unique single-channel conductance and pharmacological properties is of major relevance for this research field, also providing an access to functionally study the role of TASK-5 in cancer.

[0039] Since TASK-5 channels alone (as homomers) cannot be functionally characterized, it is also not possible to functionally investigate whether, for example, a genetic defect in the KCNK15 gene (encoding TASK-5 channels) causes the disease. In this case, for the diagnostics of a disease (putative candidate diseases discussed above), it would be necessary to functionally examine the consequences of mutant (genetic variant) TASK-5 in heteromeric complexes with TASK-1 and / or TASK-3 channels. Also genetic defects in TASK-1 or TASK-3 might be only dysfunctional and disease-causing in heteromeric channel complexes with TASK-5. For any type of genetic disease in TASK-1 , TASK-3 or TASK-5 that involves TASK- 5 containing heterodimeric channel dysfunction, studying the heterodimeric channel complex by co-expression the channels (as described below) is relevant for the identification of the disease causing mechanism and subsequently for testing of putative drugs that are able to rescue channel dysfunction, presenting a prerequisite to heal, prevent or stop the progression of the disease. One example would be for instance that in addition to TASK-1 also TASK-5 and TASK-3 channel gene mutations might cause pulmonary arterial hypertension (PAH), and that one would need to functionally examine the TASK-5-cointaining heteromers of these channels to diagnose the basis of the disease and derive a therapy by identifying TASK-5 containing heteromer-specific activators or blockers with our assay.

[0040] In addition, our assay of co-expression TASK-5 with other TASK channels can be used to probe for and / or develop drugs, that avoid a side effect or to develop a tissue specific therapy. For instance, TASK-1 blockers might be beneficial against atrial fibrillation and the assay can be used to develop blockers that do not affect heteromeric TASK-5 / TASK-1 channel complexes and thus preventing the side effect of a pulmonary hypertension. Conversely, cardiac side effects could be prevented by developing blockers (i.e. inhibiting modulators) of heteromeric TASK- 1 / TASK-5 channels to treat PAH without affecting the homodimeric TASK-1 channels in the heart. Thus, with our assay it is possible to identify specific compounds, preventing side effects and gaining tissue-specificity for a medication.

[0041] The assay can be used to probe whether a specific drug therapy with a TASK channel modulator (blocker or activator) will be successful in a patient, as this might depend on the KCNK15 polymorphism leading to the TASK-5 G95E variant channel subunits. For personalized medicine it is important to know whether a patient carries the G95E polymorphism or not, in order to determine whether a therapy with a TASK channel modulator could work or not. Thus depending on the targeted disease, activators or blockers can be tested on the TASK-5-containing TASK channel complex with or without the G95E polymorphism, simply by co-expression of the different channels with wild-type or the polymorphic variant. In the future novel KCNK15 variants may be identified and can be tested in a comparable way by the co-expression experiments of our assay.

[0042] The assay to develop specific modulators (activators or blockers) of TASK-5- containing channel complexes with TASK-1 and TASK-3 can be performed in any suitable cell system, at any suitable temperature and with any method of gene transfer (i.e. transient or stable co-expression), as well as with any suitable method to detect changes in the pharmacology of the TASK-5-containing TASK channel complexes.

[0043] In detail: Instead of using Xenopus oocytes or any other sort of oocytes derived from i.e. Xenopus laevis, Xenopus tropicalis or any Axolotl species like Ambystoma mexicanum, the assay can be performed in any type of isolated cells, tissue or immortalized cell lines (i.e. mammalian cell line like CHO, HEK293 or insect cell line like Sf9 or yeast cell lines) without leaving the scope of the invention. The assay can be performed at any suitable temperature dependent on the choice of the expression system. It is well known to any person of ordinary skill in the art which temperatures are suitable for each expression system, so that it is possible to use all generally known parameters according to the state of the art without leaving the scope of the invention. For example: If oocytes derived from frog, toad or Axolotl species are used, the preferred range of temperature for performing the expression of the heteromeric TASK channel complex is 10°C-25°C, more preferred is the range of 14°C-23°C, most preferred is the range of 16°C-21 °C. If mammalian cells are used, the preferred range of temperature for performing the expression of the heteromeric TASK channel complex is 20°C-50°C, more preferred is the range of 35°C-42°C, most preferred is the range of 36°C-38°C. If insect cells are used, the preferred range of temperature for performing the expression of the heteromeric TASK channel complex is 20°C-50°C, more preferred is the range of 25°C-45°C, most preferred is the range of 30°C-35°C. Whenever a range of parameter values is given herein throughout the document, all single values within the range are also implicitly included. The genes can be co-expressed with any other method than cRNA-injection into oocytes, including for example stable and transient cotransfection, transduction or injection of genetic material in cells or tissue. In addition, any suitable method can be chosen to detect changes in the pharmacology of the TASK-5-containing TASK channel complexes, ranging from electrophysiological techniques like TEVC, standard Patch Clamp or any different electrophysiological technique, including automated Patch Clamp technique and methodological derivatives without leaving the scope of the invention. Any method can be utilized to directly (electrophysiological) or indirectly measure currents or membrane potential (voltage-sensitive dyes, i.e. using DiBac derivatives) to detect changes in the pharmacology by heteromerization of TASK-1 or TASK-3 with TASK- 5 without leaving the scope of the invention. The co-expression assay can be also used in combination with Rubidium Efflux experiments to indirectly determine changes in pharmacology of TASK channels by the presence of TASK-5 without leaving the scope of the invention. As the co-expression of TASK-5 with TASK-1 or TASK-3 is suitable for the detection of altered pharmacology, this assay can be also adapted and performed in silico by docking experiments and or virtual screenings to identify drugs with the desired effect or lack of effect on TASK-5-containing heteromers without leaving the scope of the invention.

[0044] Summarizing, we found that TASK-5 is engaged in the formation of heteromeric channel complexes with TASK-1 and TASK-3. In particular the altered Gq-coupled receptor-mediated channel inhibition might be of physiological relevance in some tissue in which TASK-1 and TASK-5 are co-expressed. Moreover, the unique pharmacology of TASK-1 / TASK-5 heterodimers is highly relevant for the research and development of novel drugs targeting atrial fibrillation, obstructive or central sleep apnea and pulmonary arterial hypertension.

[0045] Ion channels have evolved as promising drug targets against insects, bacteria, viruses or other infectious organisms like plasmodium falciparum, causing the malaria disease. Thus, TASK-5 containing heteromeric channel complexes present in these species (orthologous) are promising targets for the development of insecticides or drugs against many infectious diseases.

[0046] As is explained above (i. e. the involvement of TASK-channels within numerous tissues) it is obvious that the scope of the invention generally comprises the identification of API’s against any illness being related to a tissue within which TASK-5 channels are expressed.

[0047] The TASK-5 containing heteromeric channel complex according to the invention is composed of subunits. Examples of subunits of the TASK-5 containing heteromeric channel complex are protein sequences being orthologous to SEQ ID 1 and / or being orthologous to SEQ ID 2 and / or being orthologous to SEQ ID 3.

[0048] Further examples of subunits of the TASK-5 containing heteromeric channel complex are also proteins, comprising proteins having at least 10 % sequence identity to SEQ ID 1 , at least 10 % sequence identity to SEQ ID 2, at least 10 % sequence identity to SEQ ID 3.

[0049] Detailed embodiments of the invention

[0050] The embodiments prevented herein are to be understood as being examples of embodiments of the invention and are not to be understood as being confining the scope of the invention. Any variants or alternatives of - for example - expressing, measuring, handling / comparing of measured values, evaluating of measured values etc. which are known to persons of ordinary skill in the art can be applied without leaving the scope of the invention.

[0051] Materials and methods

[0052] Cloning and site directed mutagenesis

[0053] Human (h)TASK-5 (AF294350.1 ), hTASK-1 and hTASK-3 cDNAs were subcloned into the oocyte expression vector pSGEM, 14-3-3s and 14-3-3^ in pGEM-HE and a1 receptor in pBluescript. Mutations were introduced with the QuikChange Site- Directed Mutagenesis Kit (Agilent) following manufacturer’s instructions and confirmed by Sanger sequencing (Seqlab).

[0054] Isolation ofXenopus laevis oocytes, cRNA synthesis and injection

[0055] Oocytes were obtained from anesthetized Xenopus laevis frogs and incubated in OR2 solution containing in mM: 82.5 NaCI, 2 KCI, 1 MgCl2, 5 HEPES (pH 7.5) substituted with 1.5 mg / ml collagenase (Nordmark) to remove residual connective tissue. Subsequently, oocytes were stored in ND96 solution containing in mM: NaCI 96, KCI 2, CaCl2 1 .8, MgCl2 1 , HEPES 5; pH 7.4 with NaOH, supplemented with Na- pyruvate (275 mg / l), theophylline (90 mg / l) and gentamicin (50 mg / l) at 18°C.

[0056] TASK-1 , TASK-3, TASK-5 and 14-3-3 cDNAs were linearized with Nhel and cRNA was synthesized using the HiScribe T7 ARCA mRNA Kit (New England Biolabs). cDNA of the ai receptor was linearized with BamHI and cRNA synthesis was done with the mMESSAGE mMACHINE™ T3 Kit (ambion). Quality was tested using agarose gel electrophoresis and cRNAs were quantified by a spectrophotometer (NanoDrop, Thermo Fisher Scientific). Stage IV and V oocytes were each injected with 50 nl of cRNA.

[0057] Two-electrode voltage clamp (TEVC) recordings

[0058] All two-electrode voltage clamp recordings were performed at room temperature (20-22 °C) with an Axon Axoclamp 900A Microelectrode Amplifier (Molecular Devices) and a Digidata 1440 Series (Axon Instruments) as an analog / digital converter or with a TurboTEC 10CD (npi) amplifier and a Digidata 1200 Series (Axon Instruments). Micropipettes were made from borosilicate glass capillaries (GB 150TF-8P, Science Products) and pulled with a DMZ-Universal Puller (Zeitz). Recording pipettes had a resistance of 0.5-1 .5 MQ when filled with 3 M KCI solution. ND96 (pH 7.5) was used as recording solution. Block / activation was analyzed with a voltage step protocol from a holding potential of -80 mV. A first test pulse to 0 mV of 1 s duration was followed by a repolarizing step to -80 mV for 1 s directly followed by another 1 s test pulse to +40 mV. The sweep time interval was 10 s. Current amplitudes were analyzed at +40 mV after applying a ramp protocol. From a holding potential of -80 mV voltage was ramped from -120 mV to +45 mV within 3.5 s. Data were acquired with Clampex 10 (Molecular Devices) and analyzed with Clampfit 10 (Molecular Devices) and Origin 2016 (OriginLab Corp.).

[0059] Temperature sensitivity recordings were performed by using the TC-10 / 20 temperature controller (npi, Tamm, Germany). Usage of the HPT-2A heated perfusion tube (ALA Scientific Instruments, New York, United States) and the TS- 200 miniature thermistor probe (npi, Tamm, Germany) in the recording chamber allowed for the precise heating of the bath solution to the desired temperature value. Temperature sensitivity was recorded using a ramp protocol, starting from a holding potential of -80 mV, down to -120 mV within 0.1 s and followed by a 3.5 s long ramp to +45 mV. The sweep time interval was 10 s. Temperature dependent current changes were continuously recorded in a range between 15 °C and 35 °C and analyzed at the end of the voltage ramp (+45 mV). Differences in temperature sensitivity between homomers and heteromers were quantified and compared by calculating the temperature coefficient (Q10 value) for different temperature ranges:

[0060] Inside-out single-channel patch-clamp recordings

[0061] Single-channel patch-clamp recordings in the inside-out configuration of manually devitell inized Xenopus leavis oocytes were performed at room temperature 24-48 h after cRNA injection (0.1 ng / pl TASK-3 and / or 2.5 ng / pl TASK-5 per oocyte). Patchpipettes were pulled from borosilicate glass capillaries GB 150TF-8P (Science Products) using a DMZ-Universal Puller (Zeitz) and had resistances of 4-6 MQ when filled with bath solution containing (in mM): KCI 140, HEPES 5, EGTA 1 at pH 7.4 (adjusted with KOH / HCI). Single-channel currents where amplified with an Axopatch 200B amplifier (Axon Instruments) and recorded with pCIampI O software (Axon instruments) at a sampling rate of 15 kHz and the analog filter frequency set to 5 kHz using a Digidata 1550B A / D converter (Axon instruments). Single-channel analysis was performed and the data was subsequently filtered with a 3 dB 8-pole

[0062] Bessel filter at 2 kHz for illustrations using ClampfitIO (Axon Instruments).

[0063] Quantification of surface expression

[0064] An extracellular hemagglutinin (HA) tag followed and preceding by a PGG sequence was introduced in human TASK-1 , TASK-3 and TASK-5 at amino acid position 214. Surface expression of HA tagged channel constructs was analyzed in Xenopus laevis oocytes 48 h after cRNA injection. To block unspecific binding of antibodies, oocytes were incubated in ND96 solution supplemented with 1 % (w / v) bovine serum albumin (BSA) at 4°C for 30 min. Subsequently, oocytes were incubated for 1 h at 4°C with rat anti-HA antibodies (clone 3F10, Roche), washed at 4°C with 1 % BSA / ND96 for 30 min and incubated for 30 min at 4°C with peroxidase conjugated secondary anti-rat antibodies (Dianova). After washing for 1 h at 4°C in 1 % BSA / ND96 and for 15 min in ND96 solution, chemiluminescence of single oocytes was measured as relative light units (RLUs) using SuperSignal Elisa Femto solution (Pierce) and a luminometer (Promega). Non-injected oocytes served as a control.

[0065] Confocal microscopy

[0066] For confocal microscopy, HEK293T cells were transfected in 6 cm dishes with 50 ng pEGFP-C1 / hTASK5 and 950 ng pcDNA3. On the next day, cells were seeded on polylysin-coated coverslips. Cells were imaged 2 days post transfection using the 488 nm line of an argon laser on a Leica SP5 confocal microscopy equipped with a 63x / 1.4 oil immersion lens.

[0067] Drugs

[0068] Drugs were resolved in DMSO or H2O (ruthenium red) and added to ND96 recording solution directly before recordings. Final DMSO content did not exceed 0.1 %. The IC50 was determined from Hill plots using five concentrations for each construct.

[0069] Statistical analyses

[0070] All values are expressed as means ± S.E. For all oocyte experiments N > 3 different batches were used. Error bars represent S.E.M. values. Significance was assessed using an unpaired two-tailed Student's t test. Asterisks indicate significance: *, p < 0.05; **, p < 0.01 ; ***, p < 0.001. According to the invention, the identification of a compound as being a pharmaceutical active compound comprises determining whether the activity of the heteromeric channel complex (being a K2P channel) in presence of the compound to be tested is mathematically significantly different from the activity of the heteromeric channel complex in absence of the compound to be tested. Any person of ordinary skill in the art well knows about how to determine mathematical significance, the above mentioned Student’s t test being just an example.

[0071] Description of the drawings

[0072] Figure 1: TASK-5 resists to form homomeric channels at the plasma membrane. (A) Cartoon of a TASK5 subunit, illustrating the location of amino acid mutations, potential trafficking signals and the localization of the extracellular HA- epitope that was introduced. (B) Representative current traces of TASK-5 and TASK-3 recorded in Xenopus laevis oocytes after injection of 25 ng TASK-5 or 50 pg TASK3. A voltage-step protocol from a holding potential of -80 mV to +40 mV was applied, ni: non-injected oocytes. (C) Mean current amplitudes analyzed at +40 mV. (D) Expression of TASK-5 in oocytes and voltage-clamp recordings with different extracellular pH (pH 7.5 or 10.5). (E) Current amplitudes after injection of 25 ng TASK-5 or the X-gate mutants TASK-5R7Dor TASK-5R131 D, analyzed at +40 mV. (F) Current amplitudes of mutants in the cap structure, TASK-5G53Cor (G) TASK-5FGF’YNL, analyzed at +40 mV. (H) Current amplitudes of TASK-5 recorded after storage in solution containing gentamycin and theophylline or after removing these drugs from the bath solution, analyzed at +40 mV. (I) Analysis of the surface expression of TASK-5, after injection of 5, 12.5 or 20 ng TASK-5HA’ExcRNA per oocyte. RLUs: relative light units. TASK-1HA’Ex(5 ng / oocyte) served as a positive control. (J) Mutation of cytosolic retention signals in TASK-5: TASK-5R2A / R3A(N- terminus), TASK-5R137A / R138A(M2-M3 linker) or TASK-5R278A / R279A(C-terminus). 25 ng cRNA of each mutant was injected per oocyte and currents were analyzed at +40 mV. (K) Current amplitudes analyzed at +40 mV after co-expression of TASK-5 with 14-3-3s or 14-3-3^. Numbers of experiments are provided within the bar graphs.

[0073] Data are presented as mean ± s.e.m..

[0074] Figure 2: TASK-5 forms heterodimers with TASK-1 and TASK-3, changing channel composition at the plasma membrane. (A) TASK-1 (2.5 ng) was expressed alone or co-expressed with increasing amounts of TASK5 (2.5, 5, 12.5 ng) in oocytes. Currents were analyzed at +40 mV and normalized to TASK-1. (B) Representative current traces of TASK-1 (2.5 ng, blue) and TASK-1 co-expressed with TASK-5 (2.5 ng+5 ng, red). The currents were measured applying 200 ms voltage-steps ranging from -70 to +70 mV, with +10 mV increments, from a holding potential of -80 mV. (C) TASK-3 (0.05 ng) was expressed alone or co-expressed with increasing amounts of TASK-5 (0.05 ng, 0.25 ng or 1.25 ng). Currents were analyzed at +40 mV and normalized to TASK-3. (D) Representative current traces of TASK-3 (0.05 ng, green) and TASK-3 co-expressed with TASK-5 (0.05 ng + 1 .25 ng, light red). The currents were measured with the same protocols as described in B. (E) Analysis of TASK1HA’Exsurface expression in an ELISA-based chemiluminescence assay after injection of T1HA-Exalone (2.5 ng / oocyte) or together with either 5 ng TASK-5 or 1 .25 ng TASK-3, ni: non-injected oocytes; RLUs: relative light units. (F) Analysis of TASK-3HAExsurface expression after injection of TASK- 3HA’Exalone (2.5 ng / oocyte) or together with either 5 ng TASK-5 or 5 ng TASK-1. (G) Analysis of TASK-5HA’Exsurface expression after injection of TASK-5HAExalone (5 ng / oocyte) or together with 2.5 ng or 5 ng TASK-1 or (H) together with 1 .25, 2.5 or 5 ng TASK-3. (I) Relative current amplitudes of TASK-1 expressed alone (2.5 ng) or co-expressed with either 12.5 ng TASK-5 or TASK-5G95E, or after expression of 12.5 ng TASK5G95Ealone. Currents were analyzed at +40 mV and normalized to that of TASK1 . (J) Expression of TASK-3 (0.05 ng) alone or after co-expression with either 1.25 ng TASK-5 or TASK-5G95E, or after expression of 1.25 ng TASK-5G95Ealone. Currents were analyzed at +40 mV and normalized to that of TASK-3. (K) Relative current amplitudes of TASK-1 (2.5 ng) alone or after co-expression with either 2.5 ng TASK-3 or TASK3G95E, or after expression of 12.5 ng TASK-3G95Ealone. Currents were analyzed at +40 mV and normalized to that of TASK-1. (L) Relative current amplitudes of TASK-3 (0.05 ng) alone or after co-expression with 0.05 ng or 1 .25 ng TASK-3G95E. Currents were analyzed at +40 mV and normalized to that of TASK-3. Numbers of experiments are given within the bar graphs. Data are presented as mean ± s.e.m.. Significance was probed using an unpaired Student's T-test, **, p<0.01 ; ***, p<0.001.

[0075] Figure 3: Inside-out single-channel patch-clamp recordings support the hypotheses of a formation of heteromeric TASK-5 channel complexes. (A) Representative current traces of patches containing either TASK-3 (T3), TASK-3 plus TASK3 / TASK-5 (T3+T3 / T5) or heteromeric TASK-3 / TASK-5 (T3 / T5), recorded at -100 mV. (B) Single-channel amplitudes of TASK-3 or TASK-3 / TASK-5 heteromers. The numbers of patches are provided within the respective bar. (C) Statistical overview illustrating the number of patches containing TASK-3 alone (T3), TASK-3 / TASK5 heteromers alone (T3 / T5) or TASK-3 and TASK-3 / TASK-5 heterodimers (T3+T3 / T5), after injection of TASK-3 cRNA (left bar) or co-injection of TASK-3 with TASK-5 (right bar). Analysis of (D) the single-channel conductances (n = 4 - 6) and the (E) open-times To of TASK-3 and TASK-3 / TASK-5 heterodimers. (F) Analysis of the short TCI and long closed-times TC2 of TASK3 or (G) TASK3 / TASK-5 heterodimers. N = 3 - 5. Data are presented as mean ± s.e.m..

[0076] Figure 4: TASK-5 alters the gating of TASK heterodimers. (A) TASK-1 was expressed in Xenopus oocytes alone or co-expressed with TASK-5. Extracellular pH was changed by perfusion with ND96 recording solution with the respective pH between 5.5 and 10. Data were analyzed at +40 mV and normalized to currents at pH 10. (B) TASK-3 was expressed in oocytes alone or co-expressed with TASK-5. Extracellular pH was changed in the range of pH 5.5 to 9.0. Data were analyzed at +40 mV and normalized to currents at pH 9.0. (C, O’, C”) Representative current traces of TASK-1 or (D, D’, D”) TASK-3 after co-expression with wild-type TASK-5 or putative X-gate mutants in TASK-5 (TASK5R7Dor TASK-5R131 D), corresponding to previously identified mutations in the latch region of the TASK-1 X-gate. (E) Current amplitudes of C and D, analyzed at +40 mV, were normalized to wild-type heteromeric TASK-1 +TASK5 or (F) wild-type heteromeric TASK-3+TASK-5 current amplitudes. Surface expression of (G) TASK-1HA-Exor (H) TASK-3HA’Ex, after coexpression with either wild-type TASK-5 or the TASK-5 mutants, homologues to X- gate mutants in TASK-1 , analyzed in an ELISA-based chemiluminescence assay. RLUs: relative light units, ni, non-injected oocytes. (I-J) Gq-coupled receptor- mediated inhibition of homomeric and TASK-5 heteromeric channel complexes of (I) TASK-1 channels or (J) TASK-3 channels, ai receptors were co-expressed with TASK-1 alone (I, squares) (n = 16) or the complex of TASK-1 and TASK-5 (I, circles) (n = 18) or TASK-3 was co-expressed (J, squares) (n = 14) with the ai receptor alone or together with TASK-5 (J, circles) (n = 14). Currents were recorded by applying a voltage-step to +40 mV from a holding potential of -80 mV, in the absence or presence of the receptor agonist methoxamine (10 pM). Numbers of experiments are provided within the bar graphs. Data are presented as mean ± s.e.m.. Significance was probed using an unpaired Student's T-test, *, p<0.05; ***, p<0.001 .

[0077] Figure 5: TASK-1 / TASK-5 heterodimers show specific pharmacological properties. (A) Dose-response curves of BAY100493, BAY256116, A293 or A1899 on TASK-1 channels expressed alone (squares) or after co-expression of TASK-1 with TASK-5 (circles), recorded in Xenopus oocytes. The ICso (given in the graph) were determined from Hill plots using at least five concentrations with four to nine recordings at each concentration. (B) Fold-change in ICso of different compounds showing drug-specific increases in affinity for heteromeric TASK-1 / TASK-5 channel complexes. (C) Dose-response curves of BAY100493, BAY256116, A293 or A1899 on heteromeric TASK-1 / TASK-5 channels (circles) (also illustrated in A) or heteromeric channel complexes of TASK-1 with TASK-5L122A(squares). The ICso (given in the graph) were determined from Hill plots using at least five concentrations with four to nine recordings at each concentration. (D) Fold-change in ICso by the putatively pore facing TASK-5L122Amutant, compared to heterodimers containing wild-type TASK-5, results in strongly reduced affinities specific for different drugs. (E) Representative current traces illustrating the block of TASK-3 (upper graph) or heteromeric TASK-3 / TASK-5 channels by 5 pM ruthenium red (RR), recorded by applying a voltage-step to +40 mV from a holding potential of -80 mV. (F) Percentage of block of TASK-3, TASK3 / TASK-5 (left) or TASK-1 , TASK-1 / TASK-5 currents (right) by 5 pM RR, analyzed at +40 mV. (G) Representative current traces illustrating the activation of TASK-1 (upper graph) or heteromeric TASK-1 / TASK-5 (red) by 20 pM PD118057, recorded by applying a voltage-step to +40 mV from a holding potential of -80 mV. Percentage of activation of TASK1 or heteromeric TASK1 / TASK5 currents (right) by 20 pM PD118057 analyzed at +40 mV. (H) Percentage of activation of TASK1 or TASK-1 / TASK-5 currents by ML67-33, NS3623, BL-1249 or NS11021 analyzed at +40 mV. Data are presented as mean ± s.e.m.. Significance was probed using an unpaired Student's T-test, *, p<0.05, **, p<0.01 ; ***, p<0.001.

[0078] Figure 6: The TASK-5G95Epolymorphism alters drug-sensitivity to TASK-1 channel blockers. (A) Representative currents traces recorded after cRNA injection of TASK-1 (upper panel), TASK-1 and TASK-5 (dashed line, lower panel) or TASK-1 and TASK-5G95E(continuous line, lower panel). Currents were recorded by applying a voltage ramp from -120 mV to +45 mV within 3.5 s. (B) Representative current traces illustrating TASK-1 (continuous line left graph), heteromeric TASK- 3 / TASK-5 channels (continuous line center panel) or heteromeric TASK-1 ZTASK- 5G95Echannels (continuous line right panel) under control conditions and after application by 20 pM PD118057 (dashed line). Currents were recorded by applying a voltage-step to +40 mV from a holding potential of -80 mV. (C) Analyses of the percentage of activation by 20 pM PD118057, analyzed at +40 mV. (D) Representative current traces illustrating the block of TASK-1 (continuous line left panel), heteromeric TASK-3 / TASK-5 channels (continuous line center panel) or heteromeric TASK-1 / TASK-5G95Echannels (continuous line right panel) by 100 nM A293 (dashed line), recorded by applying a voltage-step to +40 mV from a holding potential of -80 mV. (E) Percentage of block by 100 nM A293, analyzed at +40 mV. (F) Representative current traces illustrating the block by 40 nM A1899 (dashed line). (E) Percentage of block by 40 nM A1899, analyzed at +40 mV. (C,E,G) Numbers of experiments are provided within the bar graphs and data are presented as mean ± s.e.m.. Significance was probed using an unpaired Student's T-test, *, p<0.05; ***, p<0.001 *, p<0.05. ***, p<0.001

[0079] Figure 7 (resp. Supplemental Figure 1): Amino acid sequence alignment of human TASK-1 (SEQ ID 1) and TASK-5 (SEQ ID 2) and cartoon of a TASK-5 subunit. (A) Sequence alignment of human TASK-1 (NM_002246.3) and TASK-5 (AF294350.1 ).2RR3: putative di-arginine signal; R7 or R131 , putative latch residues of the X-gate; G53: site at which the G53C dimerization mutation was introduced and glycosylation site in TASK-1 but not TASK-5;52FGFS55and52YNLS55: difference in sequence and sequence to mutate to a more TASK-1 like cap structure;137RR138,278RR279: putative di-arginine signals;326RWKSI330: putative 14-3-3 binding site; L214: position at which HA-tag was introduced in TASK-1 , TASK-3 or TASK-5. (B) Schematic drawing of a TASK-5 subunit and the localization of modifications introduced. (C) Depiction of aminoacid sequence of TASK-3 (SEQ ID 3).

[0080] It is well known to any person of ordinary skill in the art that the protein sequences of orthologous TASK proteins can vary very broadly between biological species. It may therefore well be that the degree of sequence identity between orthologous TASK proteins of different biological species is less than 70%, less than 60%, less than 50%, less that 40%, less than 30%, less than 20% or even less than 10%. But, due to the fact that TASK channels are existing in numerous completely different organisms which are relevant with respect to illnesses, it is obvious that the scope of the invention comprises all types of TASK channels of organisms which can cause illnesses, for ion channels have evolved as promising drug targets against insects, bacteria, viruses or other infectious organisms like plasmodium falciparum, causing the malaria disease. Thus, TASK-5 containing heteromeric channel complexes present in these species (orthologous) are promising targets for the development of insecticides or drugs against many infectious diseases. For example, application of orthologic protein sequences of TASK proteins of bacteria allow for the screening of antibiotic API’s, application of orthologic protein sequences of TASK proteins of fungi allow for the screening of antifungal API’s, application of orthologic protein sequences of TASK proteins of insects allow for the screening of insecticides etc.

[0081] Figure 8 (resp. Supplemental Figure 2): Intracellular localization of heterologously expressed homomeric TASK-5 channels. Exemplary confocal fluorescence images of TASK-5 pEGFP transfected HEK293 cells, 48 h after transfection. All three confocal fluorescence images are of same scale so that the scaling bar within the first image is valid for the others also. Figure 9 (resp. Supplemental Figure 3): Temperature-sensitivity of TASK-5- containing TASK heterodimers. (A) Representative temperature-sensitivity measurements of TASK-1 and heteromeric TASK-1 / TASK-5 channels or (B) TASK- 3 and heteromeric TASK-3 / TASK-5 channels recorded using a voltage-ramp protocol, starting from a holding potential of -80 mV, quickly ramping to -120 mV within 0.1 s, followed by a slow voltage ramp of 3.5 s to +45 mV, analyzing temperatures between 15 °C and 35 °C. Q10 values Currents were analyzed at the end of the voltage ramp (+45 mV) and plotted for (C) homomeric TASK-1 or heteromeric TASK-1 / TASK-5 and (D) TASK-3 or heteromeric TASK-3 / TASK-5. Numbers of experiments are given within the bar graphs. Data are presented as mean ± s.e.m..

[0082] Figure 10 (resp. Supplemental Figure 4): Doxapram-sensitivity of TASK-5 containing heterodimers with TASK-1. (A) Block of TASK-1 or heteromeric TASK-1 / TASK-5 channels by 500 nM Doxapram, analyzed at +40 mV. (B) Block of TASK-3 or heteromeric TASK-3 / TASK-5 channels by 50 pM Doxapram, analyzed at +40 mV. Numbers of experiments are given within the bar graphs. Data are presented as mean ± s.e.m..

[0083]

Claims

Claims1. Method for identification of a compound being suitable to be used as active pharmaceutical ingredient, characterized in that the method comprises the following steps i) expression of a TASK-5 containing heteromeric channel complex in vertebrate cells or insect cells, the TASK-5 containing heteromeric channel complex being- a heteromeric channel complex of a vertebrate species or- a heteromeric channel complex of an insect species or- a heteromeric channel complex of a bacteria species, so that cells are acquired having the TASK-5 containing heteromeric channel complex located in the outer cell membrane; ii) performing at least one measurement of the activity of the TASK-5 containing heteromeric channel complex being located within the outer cell membrane of the cells acquired according to step i) in the presence of the compound and in the absence of the compound, thus receiving values of the activity of the TASK-5 containing heteromeric channel complex, measured in presence of the compound and values of the activity of the TASK-5 containing heteromeric channel complex measured in absence of the compound; iii) comparing the values acquired according to step ii) in presence of the compound with the values acquired according to step ii) in absence of the compound, wherein the compound is identified as active pharmaceutical ingredient if the values measured in presence of the compound differ mathematically significantly from the values measured in absence of the compound.

2. Method according to claim 1 , characterized in that the subunits of the TASK-5 containing heteromeric channel complex are selected from protein sequences being orthologous to SEQ ID 1 and / or being orthologous to SEQ ID 2 and / or orthologous to SEQ ID 3.

3. Method according to claim 1 , characterized in that the subunits of the TASK-5 containing heteromeric channel complex are being selected from the list of proteins comprising proteins having at least 10 % sequence identity to SEQ ID 1, at least 10 % sequence identity to SEQ ID 2, at least 10 % sequence identity to SEQ ID 3.

4. Method according to any one of claims 1 to 3, characterized in that the at least one measurement of the activity of the TASK-5 containing heteromeric channel complex according to step ii) is selected from the list of measuring methods comprising standard patch clamp methods, manual patch clamp methods, automatized patch clamp methods, two-electrode voltage clamp (TEVC) measurements, cell based rubidium efflux assays, virtual screenings, voltage-sensitive dye-based assays.

5. Method according to any one of claims 1 to 4, characterized in that the TASK-5 containing heteromeric channel complex is a heterodimer channel complex comprising- a protein having at least 70 % sequence identity to SEQ ID 2 and a protein having at least 70 % sequence identity to SEQ ID 1; or- a protein having at least 70 % sequence identity to SEQ ID 2 and a protein having at least 70 % sequence identity to SEQ ID 3.

6. Method according to any one of claims 1 to 5, characterized in that the expression of the TASK-5 containing heteromeric channel complex is performed- in vertebrate cells or vertebrate tissue, for example in cells of tissue sections or in isolated cells; or- in immortalized cells, for example in HEK293 cells or CHO cells or- in insect cells, for example in Sf9 cells or- in oocytes from Xenopus laevis or Xenopus tropicalis or Axolotl.

7. Usage of the method according to any one of claims 1 to 6 for the identification of an active pharmaceutical ingredient, the active pharmaceutical ingredient binding to TASK-5 containing heteromeric channel complexes.

8. Usage of the method according to any one of claims 1 to 6 for the identification of an active pharmaceutical ingredient, the active pharmaceutical ingredient being applicable for use in the treatment or prevention of an illness, selected from the list of illnesses comprising atrial fibrillation, obstructive sleep apnea, central sleep apnea, hypertension, pulmonary arterial hypertension, malignant cancer, primary hyperaldosteronism (PHA), diabetes, obesity, tinnitus, retinopathies, cardiac arrhythmias, autoimmune disorders, malaria, bacterial infections.

9. Usage of the method according to any one of claims 1 to 6 for the identification of an active pharmaceutical ingredient being applicable for use in the treatment or prevention of an illness of a certain patient.

10. Usage of the method according to any one of claims 1 to 6 as assay to develop specific modulators (activators or inhibitors) of TASK-5 containing heteromeric channel complexes.

11. An apparatus for carrying out the method according to any one of claims 1 to 6, characterized in that the apparatus comprises- an expression unit being suitable to express a TASK-5 containing heteromeric channel complex;- a measurement unit being suitable to measure the activity of the TASK-5 containing heteromeric channel complex in presence and / orin absence of a compound;- a comparing unit being suitable to determine whether the activity of the TASK-5 containing heteromeric channel complex in presence of the compound is mathematically significantly different from the activity of the TASK-5 containing heteromeric channel complex in absence of the compound;- a display unit being suitable to display if the activity of theTASK-5 containing heteromeric channel complex in presence of the compound is mathematically significantly different from the activity of the TASK-5 containing heteromeric channel complex in absence of the compound.

Images