MinK-Related Peptide 1
A β Subunit for the HCN Ion Channel Subunit Family Enhances Expression and Speeds Activation
Abstract—The HCN family of ion channel subunits underlies the currents If in heart and Ih and Iq in the nervous system. In the present study, we demonstrate that minK-related peptide 1 (MiRP1) is a β subunit for the HCN family. As such, it enhances protein and current expression as well as accelerating the kinetics of activation. Because MiRP1 also functions as a β subunit for the cardiac delayed rectifier IKr, these results suggest that this peptide may have the unique role of regulating both the inward and outward channels that underlie cardiac pacemaker activity. The full text of this article is available at http://www.circresaha.org.
The HCN (hyperpolarization-activatedcyclic nucleotide–gated) family of ion channel subunits has been identified as the molecular correlate of the currents If in heart and Ih and Iq in neurons.1 2 3 However, several ion channels are heteromultimers of a large α subunit (like the HCN family members) and smaller β subunits. The cardiac delayed rectifiers IKr4 and IKs5 are examples of this basic principle. Their α subunits derive from the ERG and KCNQ families, respectively, but both also contain β subunits from the KCNE family of single transmembrane–spanning proteins called minK and minK-related peptides (MiRPs). In this study, we report that MiRP1 enhances the expression and speeds the kinetics of activation of the HCN family of channel subunits. From immunoprecipitation experiments, we show that it most probably forms a complex with HCN1. Using RNase protection assays (RPAs), we demonstrate that MiRP1 mRNA is prevalent in the primary cardiac pacemaking region, the sinoatrial (SA) node, and barely detectable in ventricle. Cardiac pacemaker activity is generated by a narrow balance of inward (If) and outward (IKr) currents. Our results demonstrate for the first time the potential importance of a single β subunit in simultaneously regulating both the expression and gating of both inward and outward cardiac pacemaker channels.
Materials and Methods
Heterologous Expression in Xenopus Oocytes
cRNA encoding mouse HCN1 or HCN2, rat MiRP1 with or without an HA tag at the carboxy-terminal, and rat minK were transcribed using the mMessage mMachine kit (Ambion). Xenopus laevis oocytes were isolated, injected with 2 to 5 ng (50 to 100 nL) of cRNA, and maintained in Barth medium at 18°C for 1 to 3 days. For experiments using both HCN1 or HCN2 and MiRP1 or minK, the respective cRNAs were injected in a 1:0.04 to 1 ratio. Electrophysiological studies on oocytes used the 2-microelectrode voltage clamp. The extracellular recording solution (OR2) contained, in mmol/L, NaCl 80, KCl 2, MgCl2 1, and Na-HEPES 5 (pH 7.6). Group data are presented as mean±SEM. Tests of statistical significance for midpoint and slope of activation curves were performed using unpaired Student’s t tests. P<0.05 is considered significant.
RNase Protection Assays
The procedures for the preparation of total RNA from rabbit heart tissues and the performance of the RNase protection assays were essentially identical to those described previously.6 Brain total RNA was obtained commercially from Clontech, and total RNA was isolated from left ventricle, right atrium, and brain using SV Total RNA System (Promega). For each experiment, 2 μg of total RNA was used. A commercially available rat cyclophilin probe was used in each experiment as an internal control over sample loss (Ambion). The undigested rabbit MiRP1 probe is 260 nt, whereas the upper protected fragment is 210 nt and the lower protected fragment is ≈195 nt. The undigested cyclophilin probe is 167 nt, and the protected rabbit cyclophilin band is ≈80 nt. RNA expression was quantified directly from dried RNase protection assay gels using a Storm PhosphorImager (Molecular Dynamics) normalized to the cyclophilin signal in each lane. The MiRP1 signal consisted of two protected fragments in each rabbit tissue where MiRP1 was detected. The presence of two bands is likely the result of the degenerate polymerase chain reaction primers that were designed from mouse and human sequences, used for the cloning of the rabbit RPA probe (GenBank No. AF329636). The combined intensity of both bands was used in the quantification.
All steps were performed in ice. Twenty-five oocytes were washed with Ringer solution (in mmol/L, NaCl 96, CaCl2 1.8, and HEPES 5 [pH 7.4]) and lysed by vortexing with 1 mL lysis buffer 1 (7.5 mmol/L Na2HPO4 [pH 7.4] and 1 mmol/L EDTA) with protease inhibitors (aprotinin, leupeptin, and pepstatin, 5 μg/mL of each, and 1 mmol/L PMSF). The lysate was centrifuged for 5 minutes at 150g to remove yolk proteins and subsequently for 30 minutes at 14 000g. The membrane pellet was washed with lysis buffer 1 and resuspended in 1 mL of lysis buffer 2 (in mmol/L, Tris-HCl 50 [pH 7.5], NaCl 150, EDTA 5, NaF 50, sodium pyrophosphate 50, KH2PO4 100, sodium molybdate 10, and sodium orthovanadate 2 and 1% Triton X-100 and 0.5% NP40) with the same set of protease inhibitors as lysis buffer 1 followed by clarifying by centrifugation at 14 000g for 5 minutes. Protein concentration of the membrane fractions was determined by the Lowry method.
Samples were separated on 10% SDS/PAGE in Tris-glycine buffer (for HCN1) or on 16.5% SDS/PAGE in Tris-tricine buffer7 [for MIRP1(HA)] and electroblotted to Hybond ECL nitrocellulose membranes (Amersham Pharmacia Biotech) or to Immobilon-PSQ membrane (Millipore). Western blots were performed with rabbit anti-HCN1 antibody (Quality Controlled Biochemicals) or rat anti-HA high affinity antibody (Roche Molecular Biochemicals) and with a horseradish peroxidase–coupled secondary antibody (Kirke- gaard Perry Laboratories) or biotin-conjugated affinity-purified secondary antibody (Chemicon International, Inc). Streptavidin-POD conjugate was from Roche Molecular Biochemicals. The immunoreactive bands were visualized using ECL Western blotting detection reagents (Amersham Pharmacia Biotech).
The immunoprecipitation reactions were performed with 750 μg of membrane protein fraction and anti-HCN1 antibody cross-linked to Protein A/G PLUS-Agarose (Santa Cruz Biotechnology, Inc) with dimethyl pimelimidate (Sigma).
We used the Xenopus oocyte as a heterologous expression system and examined the expression of HCN1 and HCN2 individually and coexpressed with either minK (the minimal K+ channel protein, the first identified member of the single transmembrane–spanning protein family) or MiRP1. The results are shown in Figure 1⇓. Both HCN1 (Figure 1A⇓) and HCN2 (Figure 1D⇓) express a small current when injected alone. Coexpression of either HCN1 (Figure 1B⇓) or HCN2 (Figure 1E⇓) with minK results in similar, low levels of current expression. However, a much larger current is observed when either HCN1 (Figure 1C⇓) or HCN2 (Figure 1F⇓) is coexpressed with MiRP1. Injection of MiRP1 by itself did not induce a current nor did injection with 100 nL of H2O (not shown). Our complete set of results for the expression studies of HCN1 and HCN2 with or without minK and MiRP1 is illustrated in Figures 1G⇓ and 1H⇓. The maximal conductance is calculated by dividing the tail current, the result of stepping back to −10 mV from the most negative voltage step, by the driving force (the reversal potential was measured in each oocyte). The results demonstrate an almost 3-fold enhancement of HCN1 conductance when HCN1 is coexpressed with MiRP1 (P<0.01), whereas MiRP1 enhances expression of HCN2 by more than 5-fold (P<0.01). Coexpression of either HCN1 or HCN2 with minK does not significantly alter HCN1 or HCN2 expression (P>0.05). Thus, the enhancement of expression is specific for MiRP1.
We next examined the gating properties of MiRP1 coexpressed with either HCN1 or HCN2. The results are presented in Figure 2⇓. Isochronal activation curves were constructed from tail currents recorded at −10 mV in response to 3-second- (for HCN1) or 8-second- (for HCN2) long hyperpolarizing test pulses. The results demonstrate no significant difference in midpoint but statistically indicate a shallower slope for the activation of HCN channels coexpressed with MiRP1 (Figures 2A⇓ and 2B⇓, see legends for details).
We also examined the kinetics of activation and deactivation (see Figures 2C⇑ through Figure 2F⇑). Raw data are shown for activation of both HCN1 (Figure 2C⇑) and HCN2 (Figure 2D⇑). MiRP1 decreases the time constant of activation. Statistical comparisons were made at each potential (P<0.05). When the same comparisons are made for deactivation (data enclosed within a box in Figures 2E⇑ and 2F⇑), there is no significant difference at any potential (P>0.05) although there is a tendency toward acceleration.
The rectification properties of HCN1 or HCN2 expressed with or without MiRP1 were also studied. Coexpression of either HCN1 or HCN2 with MiRP1 did not alter the linearity of the fully activated current-voltage relationship (not shown).
Previous studies examining the potential role of MiRP1 in generating IKr used Northern blot analysis to demonstrate the presence of MiRP1 mRNA in whole rat heart.4 If MiRP1 also regulates If current expression in vivo, mRNA levels for MiRP1 should be prominent in regions where If currents are large. We used RNase protection assays to quantify the distribution of MiRP1 transcripts in SA node, right atrium, and ventricle of the rabbit heart. The results are provided in Figure 3⇓. MiRP1 transcript levels are highest in the SA node, atrial levels are ≈40% of those in SA node, and ventricular levels are barely detectable (<4% of SA node).
To show that MiRP1 could be a β subunit for the HCN family, we demonstrated that a complex probably exists between members of the HCN family and MiRP1. We chose to pursue this question with HCN1 because we have raised an antibody against this family member.
The HCN1 antibody reveals a single polypeptide with an apparent molecular mass of 145 kDa (possibly glycosylated).8 MiRP1, HA epitope-tagged at the carboxy terminal end, was recognized by anti-HA high-affinity antibody as a 13.5-kDa band. Both proteins were localized in the membrane fraction, and protein expression was enhanced (≈2-fold) when they were coexpressed together (Figures 4A⇓ and 4B⇓).
To test whether a complex of HCN1 and MiRP1(HA) might exist in a heterologous expression system, we performed coimmunoprecipitation experiments using membrane fractions of oocytes injected with HCN1 alone, MiRP1(HA) alone, or both cRNAs. Figure 4C⇑ shows the immunoprecipitation products tested by Western blot analysis. The presence of MiRP1(HA) in the anti-HCN1 immunoprecipitate only for oocytes injected by both HCN1 and MiRP1(HA) cRNAs, together with its absence in oocytes injected with one of these cRNAs, indicates that MiRP1(HA) was pulled down by anti-HCN1 antibody most likely because it was complexed with HCN1.
Our results demonstrate that the proteins are colocalized as a complex in the membrane and enhance each other’s expression. This strongly suggests that MiRP1 is a β subunit for the HCN family of ion channel subunits.
MiRP1 is a member of a family of single transmembrane–spanning proteins that have been demonstrated to alter expression and serve as a β subunit of both KCNQ (minK) and ERG (MiRP1) family members.4 5 In these previous studies, as in our study, the minK family member altered gating and was demonstrated to be a β subunit by coimmunoprecipitation.
In this study, we have shown that minK is largely without effect on the properties of HCN1 and HCN2 channels expressed in Xenopus oocytes. MiRP1, on the other hand, dramatically enhances the current expression of both HCN subunits and hastens the kinetics of current activation. A speeding of deactivation kinetics is seen when MiRP1 associates with HERG to form IKr.4
MiRP1 is expressed in rabbit SA node at significant levels as determined by RPA (see Figure 3⇑), suggesting its potential importance as a β subunit for both IKr and If in SA node. The low level of MiRP1 expression in rabbit ventricle raises questions about its importance in generating IKr in this ventricular tissue and suggests that additional studies in the ventricle of other species would be valuable.
It is well-known that β subunits of Na+ and Ca2+ channels can alter both expression and gating.9 More recently, similar evidence for effects of β subunits on K+ channel expression have also been demonstrated.10 Our results show that MiRP1 and HCN1 probably form a complex in the membrane and so add the nonspecific cyclic nucleotide–gated channels to this growing list of ion channels whose expression or gating is regulated by a β subunit.
Pacemaker activity in the rabbit sinus node is generated by a net inward current of only a few pA.11 This net inward current is attributable to the balance of inward and outward currents more than an order of magnitude larger. Although the biophysical properties of each of the component currents is known, how this fine balance is achieved remains a mystery. Our results imply for the first time that a single β subunit may control the expression of two important pacemaker currents, the outward IKr and the inward If. Thus, MiRP1 could serve as an important regulator of cardiac pacemaker rate.
This work was supported by grants HL28958 and HL20558 from the National Heart, Lung, and Blood Institute and Scientist Development Awards from the American Heart Association to H.Y. and R.W. We gratefully acknowledge the gift of the HCN1 and HCN2 clones from Drs B. Santoro and S. Siegelbaum.
Original received March 16, 2001; resubmission received May 11, 2001; revised resubmission received May 23, 2001; accepted May 25, 2001.
- © 2001 American Heart Association, Inc.
Santoro B, Liu DT, Yao H, Bartsch D, Kandel ER, Siegelbaum SA, Tibbs GR. Identification of a gene encoding a hyperpolarization activated pacemaker channel of brain. Cell. 1998;93:717–729.
Ludwig A, Zong X, Jeglitsch M, Hofmann F, Biel M. A family of hyperpolarization activated mammalian cation channels. Nature. 1998;393:587–591.
Santoro B, Tibbs GR. The HCN gene family: molecular basis of the hyperpolarization-activated pacemaker channels. Ann N Y Acad Sci. 1999;868:741–764.
Abbott GW, Sesti F, Splawski I, Buck ME, Lehmann MH, Timothy KW, Keating MT, Goldstein SA. MiRP1 forms IKr potassium channels with HERG and is associated with cardiac arrhythmia. Cell. 1999;97:175–187.
Sanguinetti MC, Curran ME, Zou A, Shen J, Spector PS, Atkinson DL, Keating MT. Coassembly of KvLQT1 and minK (ISK) proteins to form cardiac ISK potassium channel. Nature. 1996;384:80–83.
Dixon JE, McKinnon D. Quantitative analysis of potassium channel expression in atrial and ventricular muscle of rats. Circ Res. 1994;75:252–260.
Sclägger H, von Jagow G. Tricine-sodium dodecyl sulfate-polyacriglamide gel electrophoresis for separation of proteins in the range from 1 to 100 kDa. Anal Biochem. 1987;166:368–379.
Hansen JE, Lund O, Engelbrecht J, Bohr H, Nielsen JO, Hansen JE. Prediction of O-glycosylation of mammalian proteins: specificity patterns of UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferase. Biochem J. 1995;308:801–813.
Isom LL, DeJongh KS, Catterall WA. Auxiliary subunits of voltage-gated ion channels. Neuron. 1994;12:1183–1194.
Shi G, Nakahira K, Hammond S, Rhodes KJ, Schechter LE, Trimmer JS. β Subunits promote K+ channel surface expression through effects early in biosynthesis. Neuron. 1996;16:843–852.
Vassalle M, Yu H, Cohen IS. Pacemaker channels and cardiac automaticity. In: Zipes D, Jalife J, eds. Cardiac Electrophysiology. From Cell to Bedside. Philadelphia, Pa: W.B. Saunders; 2000:94–103.