Phenotypic Mutation 'mayday' (pdf version)
Mutation Type large deletion
Base Change
Gene Kcnj8
Gene Name potassium inwardly-rectifying channel, subfamily J, member 8
Synonym(s) Kir6.1, sltr, gnite, slmbr
Chromosomal Location 142,564,939-142,571,356 bp (-)
MGI Phenotype Mice homozygous for a targeted null mutation exhibit sudden cardiac death due to dysregulation of the vascular tonus in the coronary arteries, and exhibit a phenotype resembling Prinzmetal (or variant) angina in humans.
Accession Number

NCBI RefSeq: NM_008428; MGI: 1100508

Mapped Yes 
Amino Acid Change
Institutional SourceBeutler Lab
Ref Sequences
Ensembl: ENSMUSP00000032374 (fasta)
Gene Model not available
SMART Domains

Pfam:IRK 37 380 3.5e-155 PFAM
Phenotypic Category cardiovascular system, homeostasis/metabolism, immune system, lethality-postnatal, MCMV susceptibility
Penetrance 100% 
Alleles Listed at MGI

All alleles(7) : Targeted, knock-out(2) Spontaneous(1) Chemically induced(4

Lab Alleles
AlleleSourceChrCoordTypePredicted EffectPPH Score
IGL00337:Kcnj8 APN 6 142570235 missense probably damaging 1.00
IGL02303:Kcnj8 APN 6 142570111 missense probably benign 0.01
IGL03026:Kcnj8 APN 6 142566473 missense probably damaging 0.99
goodnight UTSW 6 large deletion
slumber UTSW 6 large deletion
solitaire UTSW 6 large deletion
sos UTSW 6 142565927 missense probably damaging 1.00
R0278:Kcnj8 UTSW 6 142570348 missense probably benign 0.12
R0927:Kcnj8 UTSW 6 142565901 missense possibly damaging 0.82
R1680:Kcnj8 UTSW 6 142570189 nonsense probably null
R1864:Kcnj8 UTSW 6 142570240 missense probably damaging 1.00
R1865:Kcnj8 UTSW 6 142570240 missense probably damaging 1.00
R2087:Kcnj8 UTSW 6 142565696 missense probably benign 0.02
R4900:Kcnj8 UTSW 6 142566495 missense probably damaging 1.00
R5863:Kcnj8 UTSW 6 142565688 missense not run
X0018:Kcnj8 UTSW 6 142565914 missense probably benign 0.17
X0020:Kcnj8 UTSW 6 142565914 missense probably benign 0.17
X0026:Kcnj8 UTSW 6 142565914 missense probably benign 0.17
X0027:Kcnj8 UTSW 6 142565914 missense probably benign 0.17
X0061:Kcnj8 UTSW 6 142570120 missense probably damaging 1.00
X0065:Kcnj8 UTSW 6 142565914 missense probably benign 0.17
Mode of Inheritance Autosomal Recessive
Local Stock Live Mice, Embryos, Sperm, gDNA
MMRRC Submission 029579-UCD
Last Updated 05/13/2016 3:09 PM by Anne Murray
Record Created unknown
Record Posted 01/18/2008
Phenotypic Description

The mayday phenotype was identified in a genetic screen for susceptibility to normally sublethal inocula (5 x 104 PFU) of mouse cytomegalovirus (MCMV) (MCMV Susceptibility and Resistance Screen) (1).  Whereas wild type mice appear robust and healthy 1.5 to 3 days post inoculation, homozygous mayday mice die suddenly during this time period, when viral titres in the spleen are equivalent to those in wild type mice and no prior evidence of sickness is visible (Figure 1A, B).  Mayday homozygotes also die when infected with normally sublethal doses of Listeria monocytogenes, but not when infected with vesicular stomatitis virus. Mayday homozygotes exhibit hypersensitivity to Toll-like receptor ligands: mayday mice die in response to intraperitoneal injection of low, normally survivable doses of lipopolysaccharide (LPS), poly I:C or CpG DNA, ligands for TLR4, TLR3, and TLR9, respectively. It should be noted that poly I:C is also a ligand for the intracellular nucleic acid sensing protein MDA5.  However, cytokine production by homozygous mayday macrophages in response to either MCMV or TLR ligands is essentially normal (Figure 1C, D).  Finally, injection with 1.5 ml thioglycolate for the purpose of harvesting peritoneal macrophages often causes death of mayday homozygotes.  Thus, it appeared that the systemic innate host response, rather than inability to restrict the growth of MCMV infection, caused exaggerated susceptibility in mayday mice.

Compound homozygosity for the mayday mutation and Myd88Poc, TrifLps2, Stat1Dom, or TnfPanR1 did not suppress the lethal effect of LPS; neither compound homozygosity for mayday and null alleles of Tnfrsf1a, Tnf, Ifnb1, and Ifng, which encode the p55 TNF receptor, TNF-α, IFN-β, and IFN-γ, respectively.  TLR3 signaling requires signaling through TRIF, but homozygosity for the TrifLps2 mutation failed to suppress poly I:C-induced lethality in mayday mice, suggesting the sufficiency of MDA5 signaling in causing a lethal outcome. However, the CpG DNA-induced lethality of mayday mice was fully suppressed by homozygosity for Myd88poc.
Reciprocal bone marrow transplantation between Ly5.1+ and mayday mice was performed to determine whether MCMV susceptibility in mayday mice was mediated by cells of hematopoietic origin.  Of 5 mayday mice that achieved stable engraftment, 2 died within 36 hours of MCMV infection (5 x 104 PFU).  Zero out of 13 Ly5.1+ controls died under the same conditions, indicating that susceptibility in mayday mice is mediated by extrahematopoietic cell types.
The transcriptome of whole heart from wild type and mayday homozygotes was analyzed before and after injection of 0.5 μg LPS. Among 45,100 transcripts surveyed by microarray analysis, only Kcnj8 itself (the gene found mutated in mayday mice) and Fxdy6, a regulator of the Na+/K+ ATPase pump, were found to be differentially expressed in wild type versus mayday mice. Thirty minutes after LPS administration, the expression of more than 600 (mostly non-overlapping) genes was modulated in both wild type and mayday homozygotes. Three hours after LPS administration, almost complete resolution of transcriptional change was observed in wild type heart (only nine modulated transcripts were identified), but more than 3,000 transcripts were modulated in homozygous mayday hearts, many of them related to stress, injury and/or apoptosis. These included genes involved in protection against brain ischemia (Bdnf, Crk and the two metallothionein genes Mte1 and Mt2a), in vasoconstriction (Ptk2, which encodes FAK; and Ednra, encoding an endothelin receptor), in angiogenesis (Timp2 and the two neuropilin genes Nrp1 and Nrp2), in coagulation (genes encoding coagulation factor F3, Fibulin 1, the Fcγ receptor 2a, the adhesion molecule ICAM1, and the platelet-derived molecules PDGFβ and SELP) and in immune responses (Il4r, Fas and the prostaglandin-endoperoxide synthase gene Ptgs1). Notably, the voltage-dependent Ca2+ channel encoded by Cacna1c was also strongly modulated in the heart.
Nature of Mutation
The mayday mutation consists of a deletion of exon 1 and most of exon 2 from Kcnj8, due to a compound inversion within the locus (Figure 2). All of exon 1 and 74 nucleotides from exon 2 encode the 5’ UTR. The remaining 374 nucleotides of exon 2 encode the N-terminus of the Kir6.2 protein, from the start methionine to amino acid 125.
The mayday mutation arose spontaneously, and not as a result of ENU-induced mutagenesis. Three other mutant lines (goodnight, slumber, and solitaire) exhibited the same phenotype as mayday mice (2).  Failure of the mayday mutation to complement these other mutations, and sequence analysis of Kcnj8 in each strain, indicated that all four phenotypes were caused by the same mutation. Another Kcnj8 allele, designated sos, was identified in a screen designed to detect mutations that sensitize mice to immunological or other stresses. The LPS Hypersensitivity Screen tests for death induced by injection of a normally sublethal dose of LPS (5-10 μg).
Protein Prediction
Figure 3. Membrane topology and domain structure of a Kir6.1 monomer. Kir proteins contain two transmembrane helices designated M1 (outer helix) and M2 (inner helix), a pore loop, and N- and C-terminal cytoplasmic domains. The pore loop contains the selectivity filter (SF), the narrowest part of the conduction pathway. The mayday mutation consists of a deletion of exon 1 and most of exon 2 from Kcnj8, affecting the N terminal cytosolic region, outer helix M1, and the N terminal extracellular portion of the pore loop of Kir6.1. This image is interactive. Click on the image to view other mutations found in Kcnj8 (red). Click on the mutations for more specific information.   

Figure 4. Side view of the KirBac1.1 structure. Two monomers are illustrated side by side, but the depiction is not the native conformation of an intact channel; the bundle crossing cannot be seen. The M1 helix (purple), M2 helix (pink), and pore loop (gray) are transmembrane spanning. The KirBac structure has an additional helix not present in other Kir proteins called the ‘side helix” (dark cyan). The C-terminal cytoplasmic domains are shown in cyan. The K+ channel signature sequence, which forms the selectivity filter, is black. UCSF Chimera model based on PDB 1P7B, Pegan et al. Nat Neurosci. 8, 279-287 (2005).This image is interactive. Click on the 3D structure to view it rotate.
The defining properties of inwardly rectifying K+ (Kir) channels are their selectivity for conduction of K+ ions, and their ability to allow a greater influx than efflux of ions when the electrochemical gradient favors the outward flow of K+ (i.e. upon depolarization).  Kir channels stabilize the resting membrane potential and thereby regulate membrane excitability to control heart rate, vascular tone, insulin release, and salt flow across epithelia (3). There are seven subfamilies of Kir channels (Kir1-7), which are distinguished by their strength of rectification and their responses to cellular signals. Kir6 channels, also known as ATP-sensitive K+ channels (KATP), are only weakly rectifying and are regulated by the cellular ATP:ADP ratio (ATP is inhibitory, ADP is stimulatory) (4). There are two Kir6 isoforms, Kir6.1 and Kir6.2, which differ in their tissue expression patterns. Kir6.1 is composed of 424 amino acids and is 71% identical to Kir6.2 (4;5).
Kir channels are tetramers, and likewise four Kir6 proteins form the pore of the KATP channel (6;7). However, KATP channels also contain four obligate sulfonylurea receptor (SUR) proteins, each of which associates with one Kir6 protein (5;7;8). SUR proteins are required for proper membrane localization of the channel and modulate channel activity in response to cellular ATP levels (see Background) (9). Study of the crystal structures of KirBac1.1 (10), and the cytoplasmic domains of Kir3.1 (11) and Kir2.1 (12) has provided insight into the structure of the entire Kir protein family, which shares a common organization. Kir proteins contain two transmembrane (TM) helices designated M1 (outer helix) and M2 (inner helix), a pore (P) loop (also called H5 region), and N- and C-terminal cytoplasmic domains (Figure3)(Figure 4; PDB ID 1U4F) (3).
The transmembrane M2 helices of four Kir proteins line the pore of the channel, forming an “inverted teepee” in the closed channel conformation. The M2 helices cross over each other within the transmembrane space close to the intracellular surface of the membrane (the “bundle crossing”). Some evidence suggests that the bundle crossing has a gating function, being both narrow enough to restrict the passage of hydrated K+ ions (13-15), and having hydrophobic residues unfavorable for K+ ions (10).  The cryo-electron microscopic structure of KirBac3.1 demonstrates that during channel opening the M2 helices kink at a central glycine residue to widen the channel (16). Pore-facing residues in the M2 helix can bind to Mg2+ ions, which physically block K+ efflux at membrane potentials that are more positive than the resting potential (17-19). The occlusion of the transmembrane pore by Mg2+, as well as by polyamines (20), forms the basis of inward rectification of Kir channels. 
Figure 5. A, Crystal structure of the N- and C-terminal cytoplasmic domains of mouse Kir2.1. The domains (N-terminus, residues 41-64; C-terminus, residues 189-248) are fused to stabilize the protein for crystalization. B, View from outside toward the inside of the cell of the homotetramer formed by four Kir2.1 N- and C-terminal domains. The central space forms the pore of the channel. UCSF Chimera model based on PDB 1U4F, Kuo et al. Science 300, 1922-1926 (2003). This image is interactive. Click on each 3D structure to view it rotate.
The pore loop, consisting of the sequence between the M1 and M2 helices and located at the external face of the channel, contains the pore helix and the K+ channel signature sequence (TXGYG or TXGFG in Kir6 channels), present in all K+ channels. The main-chain carbonyl oxygens of the K+ channel signature sequence form the selectivity filter, the narrowest part of the conduction pathway in the open channel (21). Point mutations in the K+ channel signature sequence abolish K+ selectivity (22); the residues surrounding the signature sequence also affect K+ selectivity. In addition to mediating ion selectivity, the selectivity filter may also function in channel gating. Mutations in the pore helix, the sequence between the selectivity filter and the M2 helix, or the K+ channel signature sequence can affect gating rather than K+ selectivity (23-25).
The N- and C-terminal domains of Kir proteins constitute about 67% of the total Kir protein sequence and form the large intracellular cytoplasmic pore. Crystal structures of Kir3.1 and Kir2.1 reveal three β sheets and two α helices as the core structural elements of the Kir cytoplasmic domain (Figure 5; PDB ID 1P7B) (11;12). The β sheets, containing many polar and charged residues, surround a large pore ~30 Å in length and 7-15 Å in diameter. An α helix at the end of the C-terminal domain projects downward into the cytoplasm. The transmembrane and cytoplasmic pores together form a channel almost 60 Å in length, through which a K+ ion must diffuse to reach the cytoplasm.  The activity of Kir channels is modulated by a variety of cytoplasmic factors, which bind to residues of the cytoplasmic pore and affect channel gating. Kir6 channels are inhibited by intracellular ATP (with or without bound Mg2+), which binds to residues in both the N- and C-terminal domains, particularly R50 in the N-terminus and K185 in the C-terminus (26-32). Each of the four Kir proteins in a KATP channel has one ATP binding site, and ATP binding to one site is sufficient to close the channel (33). The phosphatidylinositol phospholipid PI(4,5)P2 enhances the currents of Kir channels, including the Kir6 channels (34;35). PI(4,5)P2 antagonizes ATP inhibition of KATP channels by competing with ATP for binding to the C-terminus of Kir6 proteins (36).
Figure 6. Probable organization of the complete KATP channel. The central pore is formed by four Kir6 proteins (purple) and surrounded by four SUR proteins (pink).
Kir6 proteins require partnership with one of two SUR proteins for proper channel function. The complete KATP channel is thought be organized with four Kir6 proteins forming a central pore, surrounded by four peripheral SUR proteins (Figure 6) (39;41). SUR1 and SUR2 (including SUR2A and SUR2B splice forms) are “atypical” ATP-binding cassette (ABC) proteins, in that they do not directly transport molecules across membranes but still contain two highly conserved ABC domains (also called nucleotide-binding fold, NBF1 and NBF2) and two transmembrane domains (TMD1 and TMD2) (see Figure 7) (9). SUR proteins contain an additional five-helix transmembrane domain called TMD0, which is connected to TMD1 by a linker, L0. Kir6 strongly associates with TMD0 in coimmunoprecipitation experiments, and TMD0 and L0 regions can regulate KATP channel opening (37;38). Protein trafficking-based assays and single particle cryo-electron microscopy of the Kir6.2/SUR1 KATP channel support the idea that SUR TMD0 and L0 interact, respectively, with the outer transmembrane helix (M1) and N-terminus of Kir6.2 (39;40). The activating effect of ADP (in the form of Mg2+-bound ADP, MgADP) on KATP channels is a result of MgADP binding to the ABC domains of SUR proteins, which is thought to induce a conformational change in SUR that results in opening of the channel pore (see Background).
The mayday mutation deletes exon 1 and most of exon 2 from Kcnj8, which encode the 5’ UTR and the N-terminus of Kir6.1, including the N-terminal cytoplasmic domain, the outer transmembrane helix (M1), and the N terminal extracellular portion of the pore loop. The large deletion is believed to result in non-functional Kir6.1-containing KATP channels, possibly due to protein degradation or misfolding.
Northern blot analysis demonstrates that Kcnj8 mRNA is expressed ubiquitously (4). High levels of expression are detected in the heart, ovary, and adrenal gland; moderate levels of expression in skeletal muscle, lung, brain, stomach, colon, testis, thyroid, and pancreatic islets; and low levels of expression in kidney, liver, small intestine, and pituitary gland (4). In the heart, strong Kcnj8 mRNA expression is observed by in situ hybridization in cardiomyocytes, and in vascular smooth muscle cells of coronary arteries (42).  Antibody staining confirmed these findings, demonstrating Kir6.1 expression in ventricular myocytes, endothelial cells, and coronary smooth muscle cells (from arteries but not veins), with the highest expression levels in the vasculature (43-45). SUR2B is expressed in coronary arterial smooth muscle but not in veins, and is thought to form KATP channels with Kir6.1 in these blood vessels (43;46;47). Cultured human pulmonary arterial smooth muscle cells have been shown to express transcripts encoding Kir6.1, but not Kir6.2, and SUR2B by RT-PCR (48). Kir6.1 is localized at the cell membrane (43). Its localization at the mitochondrial inner membrane remains controversial (45;49-51).
Figure 7. Topography of SUR and Kir6 proteins. (A) The interactions between Kir6 and SUR in the closed channel state are shown. The SUR TMD0 and LO regions interact with the M1 helix and N terminus of Kir6. ATP inhibits Kir6 by binding to residues in the cytoplasmic domains, formed by the N- and C-termini. ATP binding also brings the two ABC domains of SUR together into a dimer. (B) Hydrolysis of ATP and subsequent binding of MgADP to ABC2 drives a conformational change in SUR that is communicated to the Kir6 channel pore via the TMD0/LO region, relieving Kir6 inhibition by ATP and opening the channel.
SUR regulation of KATP channels
Kir6 proteins assemble into hetero-octameric complexes with either SUR1 or SUR2 to form KATP channels. The binding of SUR to Kir6 proteins serves at least two functions. First, the interaction is required for the translocation of properly assembled KATP channels from the ER to the plasma membrane. Exposure of an ER retention/retrieval signal, an RKR motif present in each protein (in the C-terminus of Kir6 proteins and adjacent to the first ABC domain of SUR1), prevents monomers and partial complexes from being expressed on the cell surface (52). Formation of the Kir6/SUR octamer results in a masking effect in which the RKR motif from each subunit is buried, allowing the complex to reach the cell surface.  The last 13 amino acids of SUR1 also contain an anterograde signal that promotes channel exit from the ER (53). Interactions between SUR1 and Kir6.2 in the ER have also been shown to protect Kir6.2 from degradation (54).
In addition to the effects on KATP channel assembly and trafficking, the binding of SUR proteins to Kir6 proteins directly regulates channel activity. Thus, different combinations of Kir6.1 or Kir6.2 with SUR1 or SUR2 yield channels with different properties in various cell types (55).  Studies of the bacterial ABC proteins Rad50, a DNA repair enzyme, and MJ0796, a bacterial transporter, demonstrated that two ABC domains form ‘head-to-tail’ dimers upon ATP binding. Two ATP molecules are sandwiched between the domains, with the nucleotide binding sites (NBS) formed at the dimer interface by elements from each ABC domain (56;57). Subsequent ATP hydrolysis and binding of ADP is proposed to drive a conformational change that mediates ABC protein function (e.g. transmembrane transport, DNA release, etc.) (58;59). Although the crystal structures of SUR proteins have not been reported, mutational analysis and coimmunoprecipitation experiments with SUR2 suggest that its ABC domains also dimerize upon ATP binding (60;61).  ATP hydrolysis followed by binding of MgADP to the second ABC domain, is thought to induce a conformational change in SUR that is communicated to the Kir6 channel pore via the SUR TMD0/L0 region, relieving Kir6 inhibition by ATP and opening the channel (Figure 7) (41).
Physiological function of Kir6 KATP channels
KATP channels sense changes in intracellular adenine concentration, being inhibited by intracellular ATP and activated by MgADP, and thereby couple the metabolic state of a cell to its membrane potential [reviewed in (62;63)]. KATP channels are found in many tissues, including heart, pancreatic β-cells, brain, skeletal and smooth muscle, and kidney, where they perform tissue-specific functions. One of the most extensively studied KATP channels is the Kir6.2/SUR1 channel found in pancreatic β-cells, which regulates glucose-induced insulin secretion (64). Increased glucose metabolism leads to an increase in ATP concentration that closes KATP channels, thus depolarizing the cell membrane and activating Ca2+ influx through voltage-dependent Ca2+ channels. The rise in intracellular Ca2+ triggers exocytosis of insulin-containing granules from β-cells. KATP channels regulate the tonus of vascular smooth muscles to affect blood pressure. They also play important roles in different tissue responses to metabolic stress. For example, KATP channel activity protects neurons from damage during ischemic stress (62).  In the heart, activation of Kir6.2/SUR2A KATP channels is important for vasodilation of the coronary artery during ischemia, and is thought to minimize cardiac damage in subsequent, more severe episodes of ischemia (65;66).
Kir6.1 is ubiquitously expressed, but its highest levels are found in arterial smooth muscle cells of the heart, where it forms KATP channels with SUR2B to regulate vascular smooth muscle tone. In these cells, native KATP channels are characteristically inhibited by ATP and activated by nucleotide diphosphates (NDP), are closed by sulphonylureas such as glibenclamide and tolbutamide, and are opened by K+ channel openers such as pinacidil and cromakalim (67-69). KATP channels ultimately regulate smooth muscle tone by affecting the concentration of free calcium within muscle cells. Opening of K+ channels results in membrane hyperpolarization due to efflux of K+, an effect that results in the closure of voltage-dependent Ca2+ channels, reduced Ca2+ entry into the cell, and vasodilation (70). Conversely, inhibition of K+ channels leads to membrane depolarization and vasoconstriction (71)
Endogenous vasodilators and vasoconstrictors affect KATP channel activity and shed light on the cellular pathways that modulate channel activity. Adenosine, calcitonin gene-related peptide (CGRP), and β-adrenergic agonists are vasodilators that stimulate KATP channel current that is glibenclamide-sensitive in vascular smooth muscle cells. Their effect on KATP channels depends on activation of adenylyl cyclase, an increase in intracellular cAMP, and activation of protein kinase A (PKA) (72-74). Other KATP-dependent vasodilators, such as atrial natriuretic factor, act through guanylyl cyclase, which leads to increased cGMP levels and activation of KATP channels (75). Vasoconstrictors, such as angiotensin, endothelin, vasopressin, serotonin and noradrenaline, inhibit vascular smooth muscle KATP channels via PKC activation. These molecules bind to G-protein coupled receptors, leading to generation of diacylglycerol and inositol 1,4,5-triphosphate (IP3) by phospholipase C, which activates PKC (74;76;77). KATP channel phosphorylation sites are under investigation, but the mechanisms by which PKA- and PKC-mediated phosphorylation affect KATP channel activity remain unknown.
Electrophysiological and expression studies have shown that Kir6.1/SUR2B constitute one of the native coronary vascular smooth muscle KATP channels (48;78). However, Kir6.1/SUR2B channels reconstituted in HEK-293 cells display a very different ATP sensitivity than native KATP channels from coronary smooth muscle (48;78). This may be due to cell type differences affecting channel properties. Alternatively, heteromultimerization between Kir6.1 and Kir6.2, also expressed in vascular smooth muscle, may produce KATP channels with distinct properties from homomeric channels (79). It has been suggested that coronary vascular smooth muscle cells express multiple types of KATP channel, including the Kir6.2/SUR2B channel (80;81), as electrophysiological recordings of native channels demonstrate heterogeneity in channel conductances and ATP/NDP sensitivity (69).
However, the phenotype of mice with a targeted deletion of Kcnj8 supports a predominant role for Kir6.1 in the regulation of vascular tone in coronary arteries (42). These mice are prone to sudden, premature death at 5 to 6 weeks of age while showing no prior signs of ill health.  Electrocardiogram (ECG) analysis revealed spontaneous ST segment elevation followed by atrioventricular block in these mice, indicating that death is due to myocardial ischemia.  Although Kir6.1 is present in cardiomyocytes, no electrophysiological abnormalities were found in the plasma membrane of Kir6.1 null cardiomyocytes.  Specifically, pinacidil induced similar glibenclamide-sensitive outward currents in both wild type and Kir6.1 null cardiomyocytes. On the other hand, KATP channel function is defective in aortic vascular smooth muscle of Kir6.1 null mice, where pinacidil fails to stimulate vasodilation, as indicated by a lack of reduction in arterial blood pressure and tension in aortic rings from Kir6.1 null mice.  Furthermore, whereas pinacidil evokes significant K+ currents that are blocked by glibenclamide in aortic smooth muscle cells from wild type mice, no significant K+ currents are induced by pinacidil in Kir6.1-deficient aortic smooth muscle cells. In contrast, Kir6.2 null mice display normal KATP channel currents in aortic smooth muscle, but lack KATP channel currents in cardiomyocytes (82). These findings strongly suggest that Kir6.1 predominantly or solely constitutes KATP channels in arterial vascular smooth muscle, and that Kir6.2 constitutes KATP channels in cardiomyocytes. These data also argue against the existence of Kir6.1-Kir6.2 heteromultimers.
Kir6.1 null mice have a greatly reduced ability to survive the innate immune response following endotoxin challenge with LPS (22/22 Kcnj8-/- deaths, 6/22 WT deaths following 15 μg/g LPS) (83). LPS administration induced severe and persistent ST segment change on ECG, indicating myocardial ischemic injury. In addition, Kcnj8-/- mice did not increase cardiac performance in response to LPS. Dissection revealed disrupted architecture of Kcnj8-/- hearts, consistent with ischemic damage, induced by LPS.  Further study revealed that Kir6.1 null cardiac vessels failed to vasodilate in response to LPS-induced TNFα, which was similarly produced in wild type and Kir6.1 null animals. Direct administration of TNFα, or the endogenous vasodilator adenosine, failed to induce vasodilation in Kcnj8-/- hearts. Thus, death following LPS injection was attributed to an impaired vasodilatory response to endotoxemia, mediated in part by TNFα, and resulting in deterioration of cardiac activity, ischemic myocardial damage, and contractile dysfunction.
Targeted deletion of SUR2 also results in sudden death, elevated blood pressure, and coronary artery vasospasm, similar to that observed in Kir6.1-deficient mice (84).  Whether LPS challenge of SUR2 mutants causes death is unknown.
In contrast to the only known role of Kir6.1 in vascular smooth muscle, studies of Kir6.2 knockout mice reveal that Kir6.2 mediates many physiological processes in cells throughout the body. Kir6.2-containing KATP channels regulate glucose- and sulphonylurea-induced insulin secretion from pancreatic β-cells; regulate glucagon secretion by hypothalamic neurons during hypoglycemia; oppose hypoxia-induced seizure; protect skeletal muscles against fatigue; protect cardiac function against adrenergic stress; and contribute to ischemic preconditioning in the heart [discussed in (62)].
Putative Mechanism
During infection, the innate immune response is triggered upon sensing of microbial molecules, leading to localized inflammation. The term sepsis describes this inflammatory response at a systemic level, and is known to involve the induction of multiple cytokines including interferon-γ (IFN-γ), tumor necrosis factor (TNF), and interleukin-12p70 (IL-12p70).  The coordinated activities of tissues and cells throughout the body contribute to an organism’s ability to survive sepsis (85). In particular, cytokine-induced cardiovascular hypercontractility and systemic vasodilation are induced during the septic state (86;87). By permitting the efflux of K+ ions, activation of KATP channels in the heart and vascular smooth muscle favors hyperpolarization, reduces Ca2+ entry, and promotes vasodilation, an important part of the body’s adaptive response to cytokines induced by LPS or MCMV.
Study of Kcnj8-/- mice and mayday mice has revealed a role for Kir6.1 in the innate immune response, but one which resides outside of the immune system itself. Although Kcnj8-/- hearts display an impaired vasodilatory response to TNFα, homozygous mayday mice survive and show no sickness when injected with 10 μg of TNF (1/25 the LD50 in wild type mice), suggesting that other cytokines triggered by LPS signaling mediate lethality. However, compound homozygosity for mayday and the ENU-induced mutations Myd88Poc, TrifLps2, Stat1Dom, and TnfPanR1 did not suppress the lethal effect of LPS; neither did double mutants of mayday with null alleles of Tnfrsf1a, Tnf, Ifnb1, and Ifng, which encode the p55 TNF receptor, TNF-α, IFN-β, and IFN-γ, respectively.  These results indicate that neither TLR, TNF, type I IFN, or IFN-γ signaling individually result in the death of mayday mice.  Interestingly, death is fully suppressed in Myd88Poc/ Poc Kcnj8mydy/mydy mice in response to CpG DNA injection.  Sensitization of mayday mice to CpG DNA thus depends completely upon the integrity of MyD88 signaling. Together with results from microarray analyses demonstrating modulation of hundreds to thousands of genes upon LPS injection, these studies strongly suggest that impaired vasodilation of coronary arteries following LPS-induced sepsis leads to ischemic myocardial damage and rapid deterioration of cardiac function in mice deficient in Kir6.1.  However, the nature of the cytokine signal normally opposed by KATP, without which severe vasoconstriction and ischemic damage occur, remains mysterious.
Kir6.1 null mice have been proposed as a model of Prinzmetal angina in humans (42). This rare and severe form of angina, first described in 1959, occurs while at rest and is associated with elevation of ST segments on ECG during the attack. Prinzmetal angina can lead to myocardial infarction, severe atrioventricular block, ventricular tachycardia, and sudden death if coronary vasospasm is prolonged (88;89). Similarly, Kir6.1 null mice display spontaneous coronary vasospasms leading to lethal vasoconstriction. Kir6.1-deficient mice also exhibit coronary artery spasms in response to ergot alkaloids (42).  So far, no genetic linkage between human Kcjn8 and Prinzmetal angina has been established, nor has any connection between microbial infection and Prinzmetal angina been documented (90).
The Drosophila homologue of SUR2, dSUR2, is expressed in the heart.  Knockdown of dSUR2 expression by RNAi throughout the body or specifically in the heart, but not in the gut, sensitizes flies to infection with the RNA virus Flock house virus (FHV) (1).  Viral load was found to be increased in FHV-infected dSUR2-knockdown flies relative to wild type flies. Together with results from mice, these data support a critical, evolutionarily conserved role for Kir6.1-containing KATP channels in the heart, where it maintains homeostasis during the innate immune response.
Primers Primers cannot be located by automatic search.
Mayday genotyping is performed by PCR. Fragments of different lengths are amplified from wild type versus mayday alleles due to the rearrangement of the Kcnj8 locus (Figure 8).
mayday_3: 5’- ggacacctcgcttgctaaaa -3’
PCR program
1) 94°C             2:00
2) 94°C             0:30
3) 55°C             0:30
4) 72°C             0:45
5) repeat steps (2-4) 34X
6) 68°C             7:00
7) 4°C              ∞
PCR products (Figure 9)
mayday allele- Primers 1 and 3 yield a 602-bp product.
wild type allele- Primers 1 and 2 yield a 418-bp product.

Primers 1 + 2 amplify a 418 bp PCR product from NM_008428.4 Mus musculus potassium inwardly-rectifying channel, subfamily J, member 8 (Kcnj8), mRNA:


195               ggaaagg agccacaggt tcaggcaggt gcataggcgg gctatggtga
241 aaggaagatg ttggccagga agagcatcat cccggaggag tatgtgctgg cgcgcatcgc
301 agcggagaac ctgcgcaaac cgcgcatccg cgaccgtctc cccaaagccc gcttcatcgc
361 caagagcgga gcctgcaacc tggcacacaa gaacatccga gagcaaggtc gcttcctgca
421 ggacatcttc accaccttgg tagacctgaa gtggcgtcac acgctggtca tcttcaccat
481 gtccttcctc tgcagctggc tgctcttcgc tatcatgtgg tggctggtgg ccttcgccca
541 cggggacatc tatgcttaca tggagaaagg caccatggag aagagtggcc tggagtccgc
601 tgtctgtgtg ac


Primers 1 + 3 amplify a 602 bp PCR product from EF562532.1 Mus musculus ATP-sensitive inwardly rectifying potassium channel subfamily J member 8 (Kcnj8) pseudogene, complete sequence :


1032             ggacacctc gcttgctaaa aaaatctcaa gtgaaaaaaa aaaagccttc
1081 aatagagcag ggctgagggc ctggtatcag ctgaacaata atgagaaaaa taacataaaa
1141 taacatagtt tctattttcg gactcacttt ggaaagtcat ctttgcacaa aggagaaagt
1201 gagccaccct aaagtaagag caaccaaggg atgttggaaa tactggtgga cagcctctga
1261 gaggaaggtg cttccgaatc tgatgatttg cctgacacca aagctgcctg acaacaaatt
1321 ggcaatgtaa acagttggaa ttgtgcatgt ttaaagttct ctaggaagag aatgagcttt
1381 aaatatgcta aggagcgaga tttaaaagct ggcgctgttg agagcaaggt cgcttcctgc
1441 aggacatctt caccaccttg gtagacctga agtggcgtca cacgctggtc atcttcacca
1501 tgtccttcct ctgcagctgg ctgctcttcg ctatcatgtg gtggctggtg gccttcgccc
1561 acggggacat ctatgcttac atggagaaag gcaccatgga gaagagtggc ctggagtccg
1621 ctgtctgtgt gac


Science Writers Eva Marie Y. Moresco
Illustrators Diantha La Vine
AuthorsBen Croker, Karine Crozat, Bruce Beutler
Edit History
01/07/2011 9:20 AM (current)
10/11/2010 1:35 PM
10/11/2010 1:31 PM
10/08/2010 3:00 PM
08/17/2010 4:19 PM
08/17/2010 4:04 PM
08/17/2010 2:21 PM
08/17/2010 2:09 PM
08/16/2010 1:08 PM
08/16/2010 1:07 PM
08/16/2010 1:06 PM
08/16/2010 1:04 PM
07/22/2010 10:35 AM
05/13/2010 1:27 PM
05/13/2010 1:27 PM
05/13/2010 12:24 PM
05/13/2010 12:23 PM
05/13/2010 12:22 PM
05/13/2010 12:21 PM
05/13/2010 12:19 PM
03/22/2010 2:52 PM
02/03/2010 5:15 PM