Biomol NMR Assign DOI 10.1007/s12104-014-9583-x

ARTICLE

Resonance assignment of the ligand-free cyclic nucleotide-binding domain from the murine ion channel HCN2 Claudia Bo¨rger • Sven Schu¨nke • Justin Lecher Matthias Stoldt • Friederike Winkhaus • U. Benjamin Kaupp • Dieter Willbold

•

Received: 31 July 2014 / Accepted: 26 September 2014 Ó Springer Science+Business Media Dordrecht 2014

Abstract Hyperpolarization activated and cyclic nucleotide-gated (HCN) ion channels as well as cyclic nucleotidegated (CNG) ion channels are essential for the regulation of cardiac cells, neuronal excitability, and signaling in sensory cells. Both classes are composed of four subunits. Each subunit comprises a transmembrane region, intracellular N- and C-termini, and a C-terminal cyclic nucleotidebinding domain (CNBD). Binding of cyclic nucleotides to the CNBD promotes opening of both CNG and HCN channels. In case of CNG channels, binding of cyclic nucleotides to the CNBD is sufficient to open the channel. In contrast, HCN channels open upon membrane hyperpolarization and their activity is modulated by binding of cyclic nucleotides shifting the activation potential to more positive values. Although several high-resolution structures of CNBDs from HCN and CNG channels are available, the gating mechanism for murine HCN2 channel, which leads to the opening of the channel pore, is still poorly understood. As part of a structural investigation, here, we report C. Bo¨rger  S. Schu¨nke  J. Lecher  M. Stoldt  D. Willbold (&) Institute of Complex Systems, Structural Biochemistry (ICS-6), Forschungszentrum Ju¨lich, 52425 Ju¨lich, Germany e-mail: [email protected] C. Bo¨rger  J. Lecher  M. Stoldt  D. Willbold Institut fu¨r Physikalische Biologie, Heinrich-Heine-Universita¨t, 40225 Du¨sseldorf, Germany F. Winkhaus  U. B. Kaupp Department of Molecular Sensory Systems, Center of Advanced European Studies and Research, 53175 Bonn, Germany Present Address: F. Winkhaus Pharma Technical Development Penzberg, Roche Diagnostics GmbH, Nonnenwald 2, 82377 Penzberg, Germany

the complete backbone and side chain resonance assignments of the murine HCN2 CNBD with part of the C-linker. Keywords HCN channel  cAMP  Apo state  Heteronuclear NMR

Biological context Cyclic nucleotide-sensitive ion channels play important physiological roles in the excitation of neurons, the regulation of the heart rate, and signal transduction in sensory cells. They can be subdivided into two classes: hyperpolarization activated and cyclic nucleotide-gated (HCN) ion channels, and cyclic nucleotide-gated (CNG) ion channels (for reviews see Cukkemane et al. 2011; Kaupp and Seifert 2001, 2002). HCN and CNG channels are composed of four subunits; each subunit consists of six a-helical transmembrane segments (S1–S6) with intracellular N- and C-termini. The C-terminal part contains a cyclic nucleotide-binding domain (CNBD) that is connected via the C-linker to the transmembrane segment S6. Despite sharing a similar topology the activity of both channels is regulated differently. CNG channels are mainly voltage independent and direct binding of cyclic nucleotides to the CNBD is required to open the channel pore. In contrast, HCN channels open upon membrane hyperpolarization. Cyclic nucleotide binding modulates HCN channel activity by shifting the activation potential to more positive values. In the ligand bound state structural investigation using crystallization studies of HCN2 CNBD constructs including the C-linker revealed a fourfold symmetric tetramer formation, which is also called gating ring (Zagotta et al. 2003). The gating ring seems to play a crucial role for the

123

C. Bo¨rger et al.

modulation by cyclic nucleotides. Intersubunit interactions in the gating ring are almost entirely formed by the C-linker which consists of six a-helices. The C-linker show an ‘‘elbow-on-shoulder’’ formation, in which the first helix–turn–helix motif (‘‘elbow’’) of one subunit rests on the adjacent two helices (‘‘shoulder’’) of the neighboring subunit (Craven et al. 2008; Taraska et al. 2009; Zagotta et al. 2003). The protein fold of the HCN2 CNBD is very similar to CNBD structures of the HCN1 and HCN4 channel (Lolicato et al. 2011; Taraska et al. 2009; Xu et al. 2010; Zagotta et al. 2003) and other cyclic nucleotideregulated proteins like the exchange protein directly activated by cAMP (EPAC), the protein kinases A and G, and the bacterial catabolite activator protein (CAP) (Eron et al. 1971; Kim et al. 2005; Rehmann et al. 2003; Vaandrager and Jonge 1996). The CNBD is composed of a central eight-stranded antiparallel b-roll with an internal phosphate binding cassette (PBC) helix topped by a bundle of three a-helices. Even though high-resolution CNBD structures of both CNG and HCN channels are available, the molecular mechanisms relaying ligand binding to channel opening remained unknown until recently. Akimoto et al. (2014) showed that in HCN4 steric clashes causing inactivation of the CNBD are eliminated upon binding of cAMP. In case of HCN2, the CNBD crystal structures in the ligand-free and -bound state are virtually identical. The last short helix in the C-linker is not formed and the C-terminal helix of the CNBD is shortened by five residues in the apo state (Taraska et al. 2009; Zagotta et al. 2003). In contrast to HCN2, structural information about the prokaryotic CNBD of the Mesorhizobium loti K1 channel in solution showed fundamental rearrangements upon binding of cyclic nucleotides (Clayton et al. 2004; Mari et al. 2011; Schu¨nke et al. 2009, 2011). In this study our intention is to analyze the structure and dynamics of the murine HCN2 CNBD using NMR spectroscopy in solution to gain further insights into the ligand dependent regulation of cyclic nucleotide-modulated HCN2 channels. As a first step, we describe the almost complete backbone and side chain assignments of the murine HCN2 CNBD with part of the C-linker.

Methods and experiments Protein expression and purification For recombinant protein expression of the murine HCN2 CNBD with part of the C-linker the DNA sequence coding for amino acids between residues asparagine 507 and histidine 645 was cloned into the vector pGEX-6P-2. E. coli (BL21 (DE3) cod ? RP) were transformed with the

123

plasmid to express HCN2N507 as GST fusion protein. Cells were grown in M9-minimal media at 37 °C ([15N]-ammonium chloride and [13C]-glucose were used for uniform labeling). Protein expression was induced with 1 mM IPTG after the optical density at 600 nm reached 0.8 and cells were further grown at 20 °C overnight. E. coli cells were harvested by centrifugation and the cell pellet was washed with 19 PBS. Lysis was done via cell disruption (Constant System Limited) in 19 PBS supplemented with DNaseI, MgCl2, and Protease Inhibitor Tablets (Roche). Cell debris was removed from the soluble protein by centrifugation at 50,0009g for 1 h. Subsequently, the supernatant was applied on glutathionesepharose (4B, GE Healthcare). Bound fusion protein was incubated with GST-tagged PreScission protease (GE Healthcare) at 4 °C for 4 h. Elution fractions containing HCN2N507 were pooled and further purified using a size exclusion chromatography (Superdex 75 high load, 16/60 prep grade, GE Healthcare) and cation exchange chromatography (Resource S, GE Healthcare). HCN2N507 containing fractions were merged and dialyzed against citrate buffer (25 mM sodium citrate pH 5.6, 50 mM sodium chloride, 0.5 mM EDTA, and 0.5 mM dithiothreitol). Protein concentrations were determined by absorbance at 280 nm and calculated using the theoretical extinction coefficient (8,940 M-1 cm-1).

NMR experiments In all NMR experiments a uniformly [13C, 15N] isotopically enriched sample of 0.45 mM HCN2N507 in dialysis buffer containing 7 % (v/v) 2H2O was used. All experiments were performed at 298 K on a Varian VNMRS instrument at a proton frequency of 900 MHz, provided with a 5 mm cryogenic Z-axis pulse field gradient (PFG) triple resonance probe. Resonance assignment was achieved using the following 2D and 3D spectra: (1H–15N)-HSQC, ct-(1H–13C)-HSQC, HNCA, HNCACB, HNHA, HNCO, H(C)CH-COSY, (1H–1H–15N)NOESY–HSQC (120 ms mixing time), (1H–13C–1H)-HSQC– NOESY (100 ms mixing time), ((1H)–13C)-HMQC–NOESY(1H–15N)-HSQC (120 ms mixing time). For the assignment of aromatic side chains a 2D ct-(1H–13C)-HSQC and 3D (1H–13C–1H)-HSQC–NOESY (100 ms mixing time) was used. 15N and 13C nuclei decoupling during proton acquisition has been performed by application of GARP and adiabatically by WURST sequences, respectively. Direct referencing for 1H and 13C chemical shifts was done using DSS. 15N chemical shifts were referenced indirectly using the absolute frequency ratio (Wishart et al. 1995). For NMR data processing NMRPipe 8.1 was used (Delaglio et al. 1995). Subsequent analysis was performed with CcpNmr analysis 2.4.0 (Vranken et al. 2005).

Resonance assignment of the ligand-free CNBD

T531 G581 G505

T599

G589 S514 T627 ε2 Q558

G568 G548

G637

F580 N610

G577 G541

T587

N640

H559

δ2

N520 S575

A593

N507

δ2

T549

110 N612

δ2

F518

N569

δ2

I545

Q5 3

N520

L586

ε2

δ2

δ2

Y579 L517 E617 δ2 N524 V537 R623 T592 E616 V512 S641 D576 F625 Y618 K510 N524 M620 N569 K553 F556 A513 K532 V561 N610 M554 D609 V562 C508 E626 M621 V526 N640 L511 R632 V614 L585 L533 Q539 R590 D634 Y600 Y543 L615 R622 N507 + R591 E613 A521 V628 D631 V595 K552 F611 F538 L574 I557 A624 I642 R602 L504 K639 D542 D522 I550 V608 M572 S607 H645 L603 A528 F525 R588 L644 K534 K573 E571 K567 A519 R546 L643 V564 A629 E547 D598 M515 F535 I544

S605

I630 M529

+

L606

I636

K638

L633 R509

G560 S506 N612 T566

T527

115

Y555 S578 L530

15

G551

S563

δ( N) ppm

Y604

125 +R

η

A597 C601

10.0

R596

L565

130

Q558

9.0

1

δ( H) ppm

Fig. 1 2D (1H–15N)-HSQC spectrum of [U-15N, 13C] isotopically enriched HCN2N507 (protein concentration: 450 lM, pH 5.6, 7 % (v/v) 2 H2O, T = 298 K) recorded at 900 MHz. Backbone resonance assignments are presented by one-letter amino acid code and sequence number. Resonances, which could not be assigned, are

Assignment and data deposition Analysis of triple resonance NMR spectra led to an almost complete backbone assignment of 94 % (Fig. 1). 15N chemical shifts of five proline residues and their preceding 13 0 C were not assigned. Assignment of aliphatic and aromatic side chains was achieved to 94 % (1H–15N and 15N of arginine and lysine residues, OH, SH, side chain 13C0 , 13 f C , and quaternary 13C were excluded). Several unusual chemical shifts (I544, Y555, Q558, E582, V595, R596, and S605) have been assigned, which were checked carefully. In addition, a resonance signal, belonging to an arginine ggroup, was detected. Furthermore, the first four residues of the protein that result from cloning and fusion protein cleavage seem to adopt two different conformations, since two distinct resonance sets for these amino acids were assigned. Analysis of the secondary structure was achieved based on backbone and Cb chemical shifts using TALOS-N

8.0

7.0

marked with ‘plus symbol’. Furthermore, resonances belonging to a second conformation of the N-terminal residues are indicated by ‘asterisk’. Side chain amide groups of asparagine and glutamine residues are connected by horizontal lines

Secondary Structure Propensity

K570

E536

1.00

0.75

0.50

0.25

0.00 510

530

550

570

590

610

630

Residue Number

Fig. 2 Secondary structure prediction using TALOS-N. The height of the bars indicates the secondary structure probability (white a-helix, black b-strand)

(Shen and Bax 2013). The analysis leads to a prediction of seven a-helical and seven b-stranded regions (Fig. 2). The seven a-helices comprise residues C508–A513, P516–F518,

123

C. Bo¨rger et al.

P523–T531, I583–L586, V608–E617, P619–T628, and I630–L633. Seven b-stranded regions were predicted for K534–F538, Y543–R546, K553–T566, E571–L574, Y579–F580, S594–A597, and Y600–S607. All predicted secondary structure segments are in good agreement with the previously reported crystal structure of the apo-state HCN2 CNBD (Taraska et al. 2009). Backbone and side chain resonance assignments were deposited in the BMRB with the accession number 25111. Acknowledgments We gratefully acknowledge support (and training) from the International NRW Research School BioStruct, granted by the Ministry of Innovation, Science and Research of the State North Rhine-Westphalia, the Heinrich-Heine-University of Du¨sseldorf, and the Entrepreneur Foundation at the Heinrich-Heine-University of Du¨sseldorf.

References Akimoto M, Zhang Z, Boulton S, Selvaratnam R, VanSchouwen B, Gloyd M, Accili EA, Lange OF, Melacini G (2014) A mechanism for the auto-inhibition of hyperpolarization-activated cyclic nucleotide-gated (HCN) channel opening and its relief by cAMP. J Biol Chem. doi:10.1074/jbc.M114.572164 Clayton GM, Silverman WR, Heginbotham L, Morais-Cabral JH (2004) Structural basis of ligand activation in a cyclic nucleotide regulated potassium channel. Cell 119:615–627. doi:10.1016/j. cell.2004.10.030 Craven KB, Olivier NB, Zagotta WN (2008) C-terminal movement during gating in cyclic nucleotide-modulated channels. J Biol Chem 283:11. doi:10.1074/jbc.M710463200 Cukkemane A, Seifert R, Kaupp UB (2011) Cooperative and uncooperative cyclic-nucleotide-gated ion channels. Trends Biochem Sci 36(1):55–64. doi:10.1016/j.tibs.2010.07.004 Delaglio F, Grzesiek S, Vuister G, Zhu G, Pfeifer J, Bax A (1995) NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J Biomol NMR 6:277–293. doi:10.1007/ bf00197809 Eron L, Arditti R, Zubay G, Connaway S, Beckwith JR (1971) An adenosine 30 :50 -cyclic monophosphate-binding protein that acts on the transcription process. Proc Natl Acad Sci USA 68:4 Kaupp UB, Seifert R (2001) Molecular diversity of pacemaker ion channels. Annu Rev Physiol 63:235–257 Kaupp UB, Seifert R (2002) Cyclic nucleotide-gated ion channels. Physiol Rev 82(3):769–824 Kim C, Xuong N-H, Taylor SS (2005) Crystal structure of a complex between the catalytic and regulatory (RIa) subunits of PKA. Science 307:690–696. doi:10.1126/science.1104607

123

Lolicato M, Nardini M, Gazzarrini S, Mo¨ller S, Bertinetti D, Herberg FW, Bolognesi M, Martin H, Fasolini M, Bertrand JA, Arrigoni C, Thiel G, Moroni A (2011) Tetramerization dynamics of C-terminal domain underlies isoform-specific cAMP gating in hyperpolarization-activated cyclic nucleotide-gated channels. J Biol Chem 286:44811–44820. doi:10.1074/jbc.M111.297606 Mari SA, Pessoa J, Altieri S, Hensen U, Thomas L, Morais-Cabral JH, Mu¨ller DJ (2011) Gating of the MlotiK1 potassium channel involves large rearrangements of the cyclic nucleotide-binding domains. Proc Natl Acad Sci USA 108:20802–20807. doi:10. 1073/pnas.1111149108 Rehmann H, Prakash B, Wolf E, Rueppel A, de Rooij J, Bos JL, Wittinghofer A (2003) Structure and regulation of the cAMPbinding domains of Epac2. Nat Struct Biol 10(1):26–32. doi:10. 1038/nsb878 Schu¨nke S, Stoldt M, Novak K, Kaupp UB, Willbold D (2009) Solution structure of the Mesorhizobium loti K1 channel cyclic nucleotide-binding domain in complex with cAMP. EMBO Rep 10:729–735. doi:10.1038/embor.2009.68 Schu¨nke S, Stoldt M, Lecher J, Kaupp UB, Willbold D (2011) Structural insights into conformational changes of a cyclic nucleotide-binding domain in solution from Mesorhizobium loti K1 channel. Proc Natl Acad Sci USA 108:6121–6126. doi:10. 1073/pnas.1015890108 Shen Y, Bax A (2013) Protein backbone and sidechain torsion angles predicted from NMR chemical shifts using artificial neural networks. J Biomol NMR 56(3):227–241. doi:10.1007/s10858013-9741-y Taraska JW, Puljung MC, Olivier NB, Flynn GE, Zagotta WN (2009) Mapping the structure and conformational movements of proteins with transition metal ion FRET. Nat Methods 6:532–537. doi:10.1038/nmeth.1341 Vaandrager A, Jonge H (1996) Signalling by cGMP-dependent protein kinases. Mol Cell Biochem 157:23–30. doi:10.1007/ bf00227877 Vranken WF, Boucher W, Stevens TJ, Fogh RH, Pajon A, Llinas M, Ulrich EL, Markley JL, Ionides J, Laue ED (2005) The CCPN data model for NMR spectroscopy: development of a software pipeline. Proteins 59:687–696. doi:10.1002/prot.20449 Wishart D, Bigam C, Yao J, Abildgaard F, Dyson HJ, Oldfield E, Markley J, Sykes B (1995) 1H, 13C and 15N chemical shift referencing in biomolecular NMR. J Biomol NMR 6:135–140. doi:10.1007/bf00211777 Xu X, Vysotskaya ZV, Liu Q, Zhou L (2010) Structural basis for the cAMP-dependent gating in the human HCN4 channel. J Biol Chem 285:37082–37091. doi:10.1074/jbc.M110.152033 Zagotta WN, Olivier NB, Black KD, Young EC, Olson R, Gouaux E (2003) Structural basis for modulation and agonist specificity of HCN pacemaker channels. Nature 425:200–205. doi:10.1038/ nature01922