Article

• The EMBO Journal (2006) 25, 2315 - 2325
• doi:10.1038/sj.emboj.7601145

Published online: 18 May 2006

• Subject Categories:

  • Membranes and Transport | Structural Biology

Crystal structure of CHP2 complexed with NHE1-cytosolic region and an implication for pH regulation

Youssef Ben Ammar^1,a, Soichi Takeda^2,3,a, Takashi Hisamitsu¹, Hidezo Mori² and Shigeo Wakabayashi¹

1. Department of Molecular Physiology, National Cardiovascular Center Research Institute, Suita, Osaka, Japan
2. Department of Cardiac Physiology, National Cardiovascular Center Research Institute, Suita, Osaka, Japan
3. Laboratory of Structural Biochemistry, RIKEN Harima Institute at SPring-8, Kouto, Mikazuki-cho, Sayo, Hyogo, Japan

Correspondence to:

Shigeo Wakabayashi, Department of Molecular Physiology, National Cardiovascular Center Research Institute, Fujishirodai 5-7-1, Suita, Osaka 565-8565, Japan. Tel.: +81 6 6833 5012; Fax: +81 6 6835 5314; E-mail: wak@ri.ncvc.go.jp

^aThese authors contributed equally to this work

Received 5 December 2005; Accepted 24 April 2006

• Keywords:

  • calcineurin homologous protein,
  • crystal structure,
  • Na⁺/H⁺ exchanger,
  • pH regulation

Introduction

Maintenance of intracellular pH (pH_i), Na⁺ concentration, and cell volume is crucial for all living cells to survive and to ensure a variety of cellular functions, such as cell metabolic processes, muscle contraction, secretion, and higher-order brain activity. The Na⁺/H⁺ exchanger (NHE) is an important transporter regulating such ionic homeostasis and catalyzing the electroneutral countertransport of Na⁺ and H⁺ through the plasma membrane and other intracellular organellar membranes in various animal species (Wakabayashi et al, 1997; Counillon and Pouyssegur, 2000; Putney et al, 2002; Orlowski and Grinstein, 2004; Zachos et al, 2005). Of the nine different NHE isoforms identified to date (NHE1–NHE9), the ubiquitous isoform NHE1 has been studied most extensively. NHE1 is known to be activated rapidly in response to various extracellular stimuli, such as hormones, growth factors, and mechanical stressors (Wakabayashi et al, 1997; Orlowski and Grinstein, 2004). Such activation of NHE1 is often linked to various diseases. For example, elevated NHE1 activity is thought to be a risk factor causing heart failure and hypertrophy, as evidenced by the remarkable protective effects of specific inhibitors (Karmazyn, 2001; Engelhardt et al, 2002). Regulation of NHE1 is thought to occur through interaction of multiple signaling molecules with the carboxyl (C)-terminal cytoplasmic domain of NHE1 and subsequent conformational change of the amino (N)-terminal transmembrane domain responsible for catalyzing NHE (see Figure 1A for membrane topology). Importantly, this regulation of NHE1 is attributable to a change in the affinity for intracellular H⁺. A previous biochemical experiment using membrane vesicles suggested that the exchangers possess a cytoplasmic 'H⁺-modifier' or 'pH-sensor' site(s), distinct from the H⁺-transport site (Aronson et al, 1982; Wakabayashi et al, 2003a). The exchangers were thus considered to be activated through conformational changes caused by protonation of this regulatory site. Our recent study, using cells expressing NHE isoforms (NHE1–3), further supported this concept by measuring the reverse reaction of exchange (Na⁺ efflux) (Wakabayashi et al, 2003a). However, another recent study reported an allosteric model with no additional H⁺-modifier site to explain the sigmoidal cytosolic H⁺ dependence (Lacroix et al, 2004).

Figure 1.

Sequence alignments of CHP and its binding domain in NHE isoforms. (A) Membrane topology model of NHE1. NHE1 consists of the N-terminal transporter domain with 12 transmembrane domains (TM) and a long C-terminal regulatory domain. Six extracellular (EL) and five intracellular loops (IL) are mapped. CHP binds to the juxtamembrane cytoplasmic domain. Arg440 was identified as an important residue for NHE1 regulation. (B) Amino-acid sequences for human NHE1–NHE5 (Genbank Accession Numbers: NM_003047, NM_003048, NM_004174, NM_177084, and NM_004594, respectively) are aligned together with sequence of CNB-binding region in CNA (Q08209). Hydrophobic residues involved in interaction with CHP or CNB are highlighted in green. Residues in NHE1 forming hydrogen bonds with CHP2 are shown in red. (C) Amino-acid sequences of CHP1/2 and CNB (Q99653, Q9D869, and P63098). Secondary structural elements are based on the structure of CHP2/NHE1-peptide complex. Loops of four EF-hands are shown by green boxes, but EF1 and EF2 are ancestral sites that do not coordinate Ca²⁺. Hydrophobic residues involved in interaction with NHE1 or CNA are highlighted in blue and red, respectively. Central CHP-unique region is boxed. Residues in CHP2 forming hydrogen bonds with NHE1 are shown in red.

View full figure (358 KB)

Of the signaling molecules that interact with the exchanger, calcineurin B (CNB) homologous protein (CHP) is particularly important. CHP was initially identified as a protein (p22) involved in vesicular transport (Barroso et al, 1996) and that interacts with NHE (Lin and Barber, 1996). CHP is a Ca²⁺-binding protein with EF-hand motifs and is myristoylated at the N-terminus (Gly2) (Barroso et al, 1996; Lin and Barber, 1996). To date, three CHP isoforms with different tissue expression patterns have been identified in mammalian tissues. While CHP1 is expressed ubiquitously in virtually all tissues, the expression of CHP2 is restricted to cancer cells (Pang et al, 2002) and the small intestine (Inoue et al, 2003), and that of CHP3 (also called tescalcin) is restricted to the heart, brain, stomach, and testis (Mailander et al, 2001; Perera et al, 2001; Gutierrez-Ford et al, 2003). Previously, we reported that CHP (at least CHP1 and CHP2) is an essential cofactor supporting the physiological activity of the plasmalemmal NHE by interacting with the juxtamembrane cytoplasmic domain of NHE1 to NHE3 (Pang et al, 2001, 2002). CHP binding-defective mutant exchangers (NHE1-3) exhibited low exchange activity (Pang et al, 2001) (5–10% of the wild-type level) and greatly shifted the pH_i-dependence to the acidic side (Pang et al, 2004), suggesting that the juxtamembrane region with bound CHP would therefore function as a key structure maintaining the physiologically active conformation of NHE1. Recently, the crystal structure of CHP1 was solved as a single polypeptide without the target molecule (Naoe et al, 2005). However, this structure did not provide information about the interaction(s) with NHE.

To obtain insights into the molecular mechanisms of the interaction(s) with NHE and regulation of NHE by CHP, we determined the crystal structure of CHP (CHP2 isoform) complexed with its binding region in NHE1. Herein, we report that the -helical CHP-binding region of NHE1 is inserted into the hydrophobic cleft formed by residues in the N- and C-lobes of CHP2 through strong hydrophobic interactions as well as hydrogen bond formation. Structure-based mutational analysis revealed that CHP may activate the exchanger by associating its hydrophobic cleft and also regulate the pH-sensing of NHE1 via its central CHP-unique region connecting the two lobes of CHP.

Overall structure of CHP2 in complex with the cytoplasmic binding region of NHE1

We crystallized the His₆-tagged human CHP2 (1–196) complexed with its binding domain (aa 503–545) in NHE1 (referred to as CHP2/NHE1-peptide) in the presence of yttrium, initially solved the structure by MAD phasing, and refined the structure model to 2.7 Å (Table I). Figure 2 shows the overall structure of the complex CHP2/NHE1-peptide. The structural model contains 181 amino acids from CHP2 (Ile11–Glu99 and Lys105–Lys196), and 25 amino acids from NHE1 fragment (Arg516–His540). The polypeptide chain of CHP2 with dimensions of 55 47 28 Å is folded into two globular domains (N- and C-terminal lobes) composed of 12 -helices (1–12) and four short -strands (1–4). Two Y³⁺ ions were found to associate with EF3 and EF4 in the C-lobe (Figure 2), consistent with biochemical data showing the high-affinity binding of Y³⁺ (1 nM) to CHP2 (Supplementary Figure 1). The overall structure of CHP2 is close to the Ca²⁺-bound form of NHE-free CHP1, which was solved recently (Naoe et al, 2005; Supplementary Figure 2). Indeed, the distances between metal ions and coordinating oxygen atoms are almost the same for Y³⁺ in CHP2 (2.380.2 Å) and Ca²⁺ in CHP1 (2.400.1 Å). In addition, the Y³⁺-bound form of CHP2 preserved the ability to interact with the full-length NHE1 (Supplementary Figure 1). Therefore, we consider our solved structure compatible with the Ca²⁺-bound form of CHP2. The overall structure of CHP2 is also similar in folding topology to the structures of other EF-hand-containing proteins, including CNB (Kissinger et al, 1995), K⁺-channel interacting protein (KChIP1) (Zhou et al, 2004; Supplementary Figure 2), AtCBL2 (Nagae et al, 2003), NCS-1 (Bourne et al, 2001), neurocalcin (Vijay-Kumar and Kumar, 1999), and CIB (Gentry et al, 2005). A common structural feature of these proteins is that they all have four EF-hands in pairs such that EF1 and EF2 form the N-lobe and EF3 and EF4 form the C-lobe. However, some local differences were observed between CHP2 and CNB or between CHP2 and KChIP1. For example, the orientation of EF-hand -helices is different among these molecules because of the swiveling motion between the N- and C-lobes of each molecule, although each pair of EF hands is relatively fixed via hydrogen bonds. This may explain the incomplete superposition between overall structures of CHP2 and CNB (root mean square deviation, r.m.s.d. 1.7 Å), in contrast to the relatively good superposition between the N-lobes or between the C-lobes themselves (r.m.s.d. 1.3 or 1.0 Å, respectively). Although CHP2 and CNB have about 36% identity in their amino-acid sequences (Figure 1C), CHP2 coordinates two Ca²⁺ ions in EF3 and EF4, unlike CNB capable of coordinating four Ca²⁺ ions in its four EF-hands (Kissinger et al, 1995). Lack of Ca²⁺-coordination in EF1 and EF2 would be derived from significant deviation from the canonical EF-hand sequence and resulting atypical structure, as described previously in detail (Naoe et al, 2005).

Figure 2.

Stereo view of the CHP2/NHE1-peptide complex showing the overall structure. N- and C-lobes of CHP2 are colored red and blue, respectively. The NHE1 peptide is colored green. Pink spheres represent the two yttrium ions coordinated by EF3 and EF4.

View full figure (141 KB)

Table 1: Data collection, phasing and refinement statistics

View full table

The most remarkable difference between CHP2 and other calcium binding proteins is that the N- and C-lobes of CHP2 are linked by a long flexible region, which protrudes by a length of about 14 Å for the determined region (Figure 3A). This domain, referred to as the CHP-unique region, was not found in other Ca²⁺-binding proteins. Five residues, Thr100 to Lys104, were not assigned in the structure due to the absence of their electron density map. Particularly, the region from aa 93–108 did not make any contact with other regions of CHP2. In addition, we found that the side chain oxygen atoms of Asp95 in this region coordinate Y³⁺ in the neighboring molecule in the crystal and thus participate in stabilization of this region. This observation explains why CHP2/NHE1-peptide was crystallized only in the presence of Y³⁺. These findings suggest that the CHP-unique region is very flexible in aqueous solution. This is consistent with the finding that this region was not seen in the structure of NHE-free CHP1 (Naoe et al, 2005), which was solved in different crystal packing.

Figure 3.

Target specificity of hydrophobic cleft. (A–C) Surface features of CHP2, CNB, and KChP1 are presented together with -helices of target peptides, respectively. In CHP2, front (upper) and side (lower) views are shown. The N- and C-lobes are colored light blue and light green, while the CHP-unique region is colored red. (D, E) Surface view of NHE1 (left) and CNA (right) peptides, respectively. The upper panels indicate the side facing outside the cleft and the lower panels represent the side facing the cleft. Hydrophobic residues are colored yellow.

View full figure (238 KB)

Interaction between CHP2 and its binding domain in NHE1

Structure determination of the complex revealed the orientation of CHP2 toward the cytoplasmic region of NHE1. The N- and C-terminal halves of the cytoplasmic -helix of NHE1 are inserted into the hydrophobic cleft in such a way that they face the C- and N-lobes of CHP2, respectively (Figure 3A). This cleft constitutes the region that traps the cytoplasmic -helix of NHE1 and maintains the stability of the complex by means of hydrophobic interactions. The contact area between CHP2- and NHE1 peptide is 1466 Ų, which accounts for 13.0 and 59.8% of the total surface areas, respectively, implying that a large area is used for interaction between the two molecules. The NHE1 peptide (Arg516–His540) solved in this structure contains different hydrophobic residues, all of which face toward the hydrophobic cleft of CHP2 (Figure 3D). Ile518, Ile522, Phe526, Leu527, and Leu530 are accommodated into the cavity formed by hydrophobic residues provided mainly from the C-lobe of CHP2, whereas Leu531, Ile534, and Ile537 are inserted into the cavity formed mainly by the N-lobe of CHP2 (Figure 4A and B). All hydrophobic side chains of the NHE1 peptide, with the exception of Ile518, make van der Waals contacts with side chain atoms from many hydrophobic residues of CHP2. For example, side chain atoms of Ile534 make van der Waals contacts with those of Leu57, Ile66, Phe70, Val86, Leu87 Phe90, Tyr118, and Ile192 of CHP2 (Figure 4C), while side chain atoms of Leu527 make interactions with those of Ala119, Leu181, Val186, Met184, and Met190 (Figure 4A). In addition to hydrophobic interactions, hydrogen bonds strengthen the association between CHP2 and NHE1 peptide. In the N-lobe of CHP2, the guanidinium group of Arg30 makes hydrogen bonds with the main chain carbonyl oxygen of Gly539, while the guanidinium group of Arg34 makes hydrogen bonds with the main chain carbonyl oxygens of Asp536 and Ile537 (Figure 4B). These hydrogen bonds are very important for the C-terminal end of the NHE1 helix to fix to the edge of the hydrophobic cleft of CHP2. The conserved Gly539 residue of NHE1 would allow the following cytoplasmic stretch to leave the cleft by inducing clear bending of the NHE1 peptide (Figures 3A and 4B). Several other hydrogen bonds are also formed. The hydrogen bond between the imidazole group (N) of His523 in NHE1 and the side chain oxygen of Tyr123 in CHP2 is particularly interesting because it is the only polar interaction in the hydrophobic surface of the cleft (Supplementary Figure 3). These polar interactions are important to determine the specificity and orientation of CHP2 toward the cytoplasmic -helix of NHE1.

Figure 4.

Closeup view showing the interaction between CHP2 and NHE1-peptide. (A) The NHE1-peptide backbone is shown in magenta, while hydrophobic side chains are shown in yellow. The hydrophobic pocket surrounding Leu527 is marked in red. His523 of NHE1 and Tyr123 of CHP2 are marked in blue. (B) Closeup view showing interaction between the N-lobe of CHP2 and the C-terminus of NHE1-peptide. (C) Ile534 is accommodated into the hydrophobic pocket produced by residues mainly in the N-lobe of CHP2.

View full figure (209 KB)

Although CNB and KChIP1 have similar hydrophobic clefts (Figure 3B and C), their size and shape are clearly different from those of CHP2. Interestingly, target peptides show sequence features that just fit to their clefts. While NHE1 peptide shows a uniform distribution of hydrophobic residues all facing toward the cleft of CHP2 (Figure 3D), CNA contains more hydrophobic residues that face toward both the CNB crevice and the outside (Figure 3E). In addition, CNA peptide contains four aromatic residues (Phe346, Phe350, Trp352, and Phe356) in the N-terminus, which are accommodated in corresponding hydrophobic cavities in the C-lobe of CNB, whereas NHE1-peptide contains only one aromatic residue (Phe526) located in the central part of CHP2. These observations suggest that highly selective fitting of each peptide to its corresponding crevices may be caused by differences in the distribution of hydrophobic residues of the target peptides and in the shape of the cleft formed by -helices of EF-hands.

Effect of mutations on interaction between NHE1 and CHP2 in cells

The present structure indicated that the correct CHP-binding domain of NHE1 consists of the region from aa 516–540, extended by 10 residues toward the C-terminus of the region (aa 510–530), as predicted previously (Pang et al, 2001). This prompted us to mutate residues in this extended region of NHE1 interacting with the N-lobe of CHP2. We mutated the Ile534 and Ile537 residues of NHE1 into charged residues (Lys or Asp) and stably expressed these mutants in exchanger-deficient PS120 cells. These mutants were expressed in the plasma membrane (as indicated by the existence of surface-expressed mature NHE1 in Supplementary Figure 4). We examined the interactions of these mutants with CHP2 by assessing whether stably transfected green fluorescent protein (GFP)-tagged CHP2 was localized to the plasma membrane. As shown in Figure 5A, most of the CHP2 was localized to the plasma membrane of cells expressing the wild-type NHE1 but not in nontransfected PS120 cells (not shown; see Pang et al, 2001), indicating a strong interaction. Of note, in most cells expressing the wild-type NHE1 almost no GFP-fluorescence was detected in the intracellular space (Figure 5A), despite the accumulation of immature NHE1 in the intracellular membranes under forced expression (data not shown), suggesting that CHP2 interacts predominantly with mature NHE1 in the plasma membrane. This was further confirmed by the finding that almost no immature NHE1 co-immunoprecipitated with CHP2 (Supplementary Figures 1 and 4). In contrast to the wild-type NHE1, the plasma membrane localization of CHP2 was almost completely abolished when transfected into cells expressing the NHE1 mutants, I534K, I534D and I537K (Figure 5, and also D for summarized data), indicating that these mutations drastically reduced the affinity for interaction with CHP2.

Figure 5.

Effects of mutations on the interaction between NHE1 and CHP2 in cells. (A) Low magnification confocal images of cells coexpressing GFP-tagged CHP2 and the wild-type (left) or I537K (right) mutant exchangers. Inset shows the intensity profile of GFP fluorescence along the dotted line in a marked cell. In most cells expressing the wild-type NHE1 but not I537K, strong fluorescent signals were detected at the cell edge. (B) Subcellular localization of CHP2 expressed in cells. GFP-tagged CHP2 was coexpressed in cells stably expressing the wild-type or mutant NHE1 variants and GFP-fluorescence was observed by confocal microscopy. (C) Subcellular localization of mutant CHP2. GFP-tagged CHP2 mutants were expressed in cells stably expressing the wild-type NHE1. Eleven residues from Glu94 to Lys104 of CHP2 were deleted in 94–104, while the 10 N-terminal residues from Met1 to Val10 were deleted in N10. For one control experiment, GFP-tagged human CNB was expressed in NHE1-transfectants. (D) Summary data for membrane localization of GFP-tagged CHP2. Intensity profile analysis was performed on confocal images as shown in (A). The number of cells with strong fluorescence signal at the cell edge (at least three times more than the average of fluorescence in the internal cell region) was counted. Data are expressed as the means.d. from 6–8 images (total cell number analyzed, 99–341).

View full figure (216 KB)

Furthermore, as expected, the plasma membrane localization of CHP2 was also markedly reduced in cells expressing the wild-type NHE1 when Leu87 or Tyr118 of CHP2 was changed to Lys (Figure 5C and D), indicating that these mutants are not able to replace completely endogenous CHP1 bound to NHE1. As the expression of GFP-tagged CHP2 was approximately 10-fold higher than that of endogenous CHP1 (data not shown), these results suggest that these mutations would reduce the affinity for interaction with NHE1 at least one-order of magnitude. Similar analysis was performed with mutation of the Arg residues (Arg30 and Arg34) in the N-terminal -helix of CHP2. The plasma membrane localization of CHP2 was abolished when these two residues were simultaneously changed to Glu (Figure 5C and D), while it was still partially preserved when either Arg residue alone was changed to Glu. Similar results were also obtained for the NHE1/CHP2 interaction in a co-immunoprecipitation experiment (Supplementary Figure 4). Thus, the hydrophobic and polar interactions involving these residues would be important in determining the strong and highly specific association between NHE1 and CHP2, Furthermore, in addition to the C-lobe, the N-lobe of CHP2 plays a crucial role in the tight interaction with NHE1, suggesting that our crystal structure reflects the interaction between native NHE1 and CHP2 molecules expressed in cells. On the other hand, structure determination indicated the existence of a unique protruding region in CHP2. In contrast to mutations of residues involving formation of the cleft, a mutant with deletion of this CHP-unique region (94–104) was mostly localized to the plasma membrane of cells expressing wild-type NHE1 (Figure 5C), indicating that deletion did not disrupt the interaction of CHP2/NHE1. Hence, as also predicted from structure, the CHP-unique region was not involved in the interaction between CHP2/NHE1. The 10 N-terminal residues of CHP2 were also found not to be involved in the interaction, as indicated by the plasma membrane localization of N10 (Figure 5C).

Functional consequence of mutations

We measured the NHE activity in cells expressing mutant exchangers. Figure 6A demonstrates the pH_i-dependence of EIPA-sensitive ²²Na⁺-uptake activity. While the wild-type NHE1 exhibited high ²²Na⁺-uptake activity, with pK 6.5 for pH_i (Hill coefficient, 1.7), the activity of mutant exchangers, I534D, I534K and Ile537, which lack CHP-binding was very low (see Supplementary Table I for kinetic parameters). The inhibitory effect of the mutations was characterized by changes in two parameters: (i) a drastic reduction of maximal ²²Na⁺-uptake activity (V_max) at acidic pH_i, and (ii) a large acidic shift of the pH_i-dependence of uptake (>0.6 pH units) (see Figure 6A, inset for normalized activity). We then compared the pH_i-dependence of ²²Na⁺-uptake in cells co-expressing NHE1 with GFP-tagged CHP2 or its deletion mutant 94–104, which preserves interaction with CHP. Interestingly, expression of 94–104 significantly shifted the pH_i-dependence of uptake towards acidic pH_i (0.2 pH unit) without any change in V_max, while it preserved the cooperative pH_i dependence, with a Hill coefficient 1.5 (Figure 6B, see also inset for the sigmoidal internal H⁺ dependence), suggesting that the CHP-unique region is involved in regulation of pH-sensing of NHE1. In general, the sigmoidal H⁺-dependence has been interpreted as exhibiting involvement of at least two H⁺-binding sites: H⁺-transport and H⁺-regulatory sites in NHE (Wakabayashi et al, 1997). To examine the effect of deletion of a CHP-specific region on the H⁺-regulatory site, we measured the reverse reaction of NHE1, that is, ²²Na⁺ efflux coupled to H⁺-influx. If we assume that NHE1 catalyzes a counter-transport reaction only involving the transport site, intracellular acidification should result in inhibition of ²²Na⁺ efflux due to cytosolic Na⁺/H⁺ competition. In contrast to this thermodynamically expected inhibition, modest acidification from 7.5 to 7.2 dramatically stimulated the rate of EIPA-inhibitable ²²Na⁺ efflux in cells co-expressing the wild-type NHE1 and CHP2 (Figure 6D), suggesting that protonation of H⁺-regulatory sites activated NHE1 (Figure 6C), as reported in detail, previously (Aronson et al, 1982; Wakabayashi et al, 2003a). Consistent with the acidic shift of pH_i-dependence of ²²Na⁺ uptake, we observed that deletion of a CHP-unique region significantly reduced the rate of ²²Na⁺ efflux (Figure 6D). EIPA-sensitive fraction of ²²Na⁺ efflux at pH_i 7.2 during the initial 3 min was reduced to 30% upon expression of 94–104, suggesting that the affinity of the H⁺-regulatory site for H⁺ decreased upon deletion of this region.

Figure 6.

Effect of mutations on the exchange activity. (A), pH_i-dependence of EIPA-sensitive ²²Na⁺-uptake in cells expressing wild-type NHE1 () or CHP binding-defective mutants, I534D (), I534K () and I537K (). pH_i was clamped at various values with K⁺/nigericin. Data were fitted to Hill equations with the kinetic parameters shown in Supplementary Table I and plotted after normalization by the maximal activity at pH_i=5.4 (inset). (B) pH_i-dependence of EIPA-sensitive ²²Na⁺-uptake in cells co-expressing wild-type NHE1 and GFP-tagged CHP2 () or deletion mutant 94–104 of CHP2 (). Data were fitted to Hill equations with the kinetic parameters shown in Supplementary Table I. The ²²Na⁺-uptake activity was also plotted against intracellular H⁺ concentration up to 1 M (inset). (C) Schematic drawing of ²²Na⁺-efflux protocol. The efflux experiment was done in ²²Na⁺-loaded, pH_i-clamped cells at extracellular pH 7.4. At lower pH_i²²Na⁺ efflux would be accelerated by H⁺ binding to the regulatory site, while at high pH_i it would be inhibited by H⁺-release from the regulatory site. (D) Time courses of ²²Na⁺ efflux. Cells were loaded with ²²Na⁺ and at the same time pH_i-clamped at pH_i 7.5 or 7.2. After removal of the radioactive solution, cells were exposed to the nonradioactive solutions with or without 0.1 mM EIPA. Data are expressed as the means.d. of three determinations. Error bars are sometimes smaller than symbol sizes.

View full figure (50 KB)

In the present study, we solved the first crystal structure of the CHP2/NHE1-peptide complex at 2.7 Å. This structure provides detailed information regarding the interaction between NHE1 and CHP: (i) the interaction with NHE1 occurs in a hydrophobic cleft encompassing the N- and C-lobes of CHP2, (ii) this interaction occurs by extensive Van der Waals contact between hydrophobic residues, as well as by some specific hydrogen bonds, (iii) the interacting region of NHE1 forms an -helix which bends at a conserved Gly539 and (iv) two domains of CHP2 are connected by a large flexible CHP-unique region. These features strengthened our previous functional study (Pang et al, 2001, 2002, 2004) from structural aspects and at the same time provided new information to be addressed regarding the function of NHE1.

A unique feature of interaction between CHP and NHE

Structure comparison among Ca²⁺-binding proteins provided valuable information regarding the target specificity of CHP1/2 proteins. Although the crystal structure of CNB/CNA complex (Kissinger et al, 1995) revealed a similar mode of interaction through its hydrophobic groove, comparison between the hydrophobic clefts of CHP2 and CNB showed several different structural features (Figure 3A and B). In addition, there is a marked difference in the distribution of hydrophobic residues between NHE1- and CNA peptides (Figure 3D and E). Such local structural differences between CHP2 and CNB, together with differences between target peptide sequences, may explain the selectivity of CHP1/2 toward the NHE members and of CNB toward CNA, and led us to predict that NHE1 may not be a target for CNB. In fact, we observed that GFP-tagged CNB was not localized to the plasma membrane when co-expressed with NHE1 (Figure 5C). On the other hand, since the C-lobe of KChIP1 is occupied by the 10 helix of itself (Figure 3C), KChIP1 associates with its target K⁺-channel molecule only through the N-lobe (Zhou et al, 2004). Crystal structures have been also solved for several other Ca²⁺-binding proteins belonging to the NCS family: recoverin (Flaherty et al, 1993), NCS-1 (Bourne et al, 2001), neurocalcin (Vijay-Kumar and Kumar, 1999), AtCBL2 (Nagae et al, 2003), and CIB (Gentry et al, 2005). Although all these proteins have hydrophobic crevices that would interact with the target molecules, the sizes and shapes of these crevices are clearly different. In contrast to calmodulin, which is able to interact with a variety of proteins with broad specificity, target molecules for CNB and NCS family proteins appear to be limited. Structural differences in the hydrophobic cleft together with highly specific polar interactions would enable the diverse array of Ca²⁺ binding proteins belonging to these families to associate selectively and tightly with the target molecules.

Of note, the N-lobe of CHP2 associates tightly with the C-lobe within a large area (Figure 3A). The contact area between the two domains is 1066 Ų, accounting for 16.6 and 15.2% of the total surface areas of the N- and C-lobes, respectively. Such interaction between the two lobes would result in the formation of a rigid cleft structure independent of the target peptide. In fact, the structure of the hydrophobic cleft of CHP2 is very similar to that of target-free CHP1 (Naoe et al, 2005). In addition, the N-lobe of CHP2 does not contain the methionine residue that was suggested to produce plasticity allowing fine-tuning toward various target molecules in calmodulin (Osawa et al, 1998). Therefore, NHE1-peptide would enter the preformed cleft, rather than the interaction being caused by a conformational change in the cleft. This target-independent rigid structure would allow the CHP-unique region to exert specific physiological functions, beyond a role as a connecting linker. This is in contrast to calmodulin the flexible liker region between the two domains of which is known to act as a hinge when it recognizes various target molecules (Zhang et al, 1995).

CHP1/2 is myristoylated and has two Ca²⁺ ions. However, these properties are different from those of other CNB and NCS family proteins. Mutation of CHP1 that prevents myristoylation has no apparent effect on plasma membrane expression or the exchange activity of NHE1. Therefore, myristoylation of CHP1/2 would not be required for the plasma membrane translocation of NHE1 or for the exchange activity (Pang et al, 2001, 2004), in contrast to KChIP1 (An et al, 2000). In addition, unlike recoverin, which undergoes the Ca²⁺-myristoyl switch (Ames et al, 1997), the myristoyl moiety of CHP1/2 would always be exposed outside the core structure upon interaction with NHE1 and probably embedded in the plasma membrane. Thus, the physiological function of myristoylation is still unknown. On the other hand, unlike many other Ca²⁺-binding proteins, only EF3 and EF4 of CHP1/2 are able to coordinate Ca²⁺ ions. Removal of Ca²⁺ with a Ca²⁺-chelator significantly reduced the interaction between NHE1 and CHP2 (Supplementary Figure 1) and a double mutation of EF3 and EF4 of CHP1 abolished the interaction with NHE1 (Pang et al, 2004). Furthermore, the Ca²⁺ affinity of CHP1 increases markedly (40-fold) upon interaction with NHE1 (Pang et al, 2004). These findings suggest that Ca²⁺-binding sites and the hydrophobic cleft are structurally coupled, that is, the Ca²⁺-bound conformation of EF-hands is required for the formation of the appropriate structure of the hydrophobic cleft for the interaction with NHE1. By analogy with calmodulin (Zhang et al, 1995), we consider that Ca²⁺ binding opens the hydrophobic pockets of EF-hands so that the residues of the C-lobe can fit into the hydrophobic cleft. However, such Ca²⁺-induced conformational changes of CHP1/2 bound to NHE1 would not occur physiologically in cells, since the Ca²⁺ affinity for such a complex would be extremely high, that is, K_d for Ca²⁺ (23 nM) is much less than the physiological cytosolic Ca²⁺ concentration (0.1–1 M) (Supplementary Figure 1 and see also Pang et al, 2004). Hence, we suggest that Ca²⁺ bound to EF-hands plays a structural role in stabilizing the hydrophobic cleft of CHP1/2. In the case of NHE1, Ca²⁺-induced activation would take place on another Ca²⁺-binding protein CaM, which binds to the middle of the cytoplasmic domain of NHE1 (Bertrand et al, 1994; Wakabayashi et al, 1994).

Implications for roles of CHP in NHE function

We demonstrated that Ile534 and Ile537 in NHE1 are critical residues for interaction with CHP2 (Figure 5). Mutations of these residues resulted in a disruption of NHE1 function, as characterized by a drastic reduction of the maximal exchange activity and a large acidic shift of the pH_i-dependence of exchange (Figure 6). This is consistent with our previous findings obtained by simultaneous mutation of four hydrophobic residues (Phe526, Leu527, Leu530 and Leu531) of NHE1, substituted with Gln or Arg (Pang et al, 2001, 2004). On the other hand, deletion of the CHP-unique region (94–104) significantly reduced the physiological exchange activity by inducing an acidic shift of the pH_i-dependence of ²²Na⁺-uptake (Figure 6B) and inhibited the rate of ²²Na⁺ efflux (Figure 6D), while preserving the maximal activity at acidic pH_i (Figure 6B), suggesting that this region functions as an important segment controlling pH_i by interacting with NHE1. Based on these findings, we propose that CHP1/2 plays two important roles in the function of NHE1. First, CHP functions as an obligatory subunit, which activates almost non-functional NHE1 by tightly associating with its cytoplasmic domain via the hydrophobic cleft. This activation is characterized by increases in both V_max and the H⁺-affinity. Secondly, CHP would participate in the delicate pH_i regulation of NHE1 through the CHP-unique region, which is involved in modulation of the H⁺-affinity, but not V_max.

We predict that CHP may stabilize the structure of the juxtamembrane domain by inducing a stable -helix, thereby preserving NHE1 in the functional conformation. Indeed, circular dichroism measurements indicated that the CHP-free, CHP-binding peptide has no secondary structure in aqueous solution (data not shown), although we do not exclude the possibility that this region is folded within the entire structure. On the other hand, the dominant-negative effect of 94–104 raises the interesting possibility that it may regulate the pH_i-sensing by interacting with some regions of NHE1. We observed that NHE1 and CHP2 crosslinked with each other through IL5 of NHE1 and the CHP-unique region (Supplementary Figure 5), suggesting that these regions in the two proteins are located in accessible positions beneath the membranes. Since, similar to 94–104 of CHP2, mutation of Arg440 in IL5 results in a large acidic shift of the pH_i-dependence of ²²Na⁺-uptake and efflux (Wakabayashi et al, 2003a, 2003b), we consider that IL5 would be one of targets for the CHP-unique region. Although we do not have direct evidence, IL4 may be another target region because mutation of Arg327 in IL4 has also been reported to lead to a similar functional defect of NHE1(Lacroix et al, 2004). Activation of ²²Na⁺ efflux by cytosolic acidification predicted the existence of an additional H⁺-regulatory site(s) (Figure 6C and D; Aronson et al, 1982; Wakabayashi et al, 2003a). This phenomenon does not appear to be fully explained by the recent allosteric model, in which only H⁺-transport sites are assumed to participate in the cooperative H⁺-activation of NHE1 (Lacroix et al, 2004). We speculate that multiple charged residues in loops of NHE1, as well as in CHP1/2, may serve as 'pH-sensor' sites, which accept protons, although further investigation will be required to support this concept. The recently reported crystal structure of the bacterial Na⁺/H⁺ antiporter, NhaA, predicted that modification of charged residues provided by cytoplasmic loops may activate NhaA by exposing the cation transport site via reorientation of transmembrane helices (Hunte et al, 2005).

Finally, it should be noted that the relatively long juxtamembrane domain (aa 503–595) of NHE1 is essential for the physiological exchange activity (Ikeda et al, 1997; Wakabayashi et al, 1997), although we focused on the CHP-binding domain in the present study. Similar to the CHP-binding region, it is likely that the region covering aa 503–595 forms a stable structure by interacting with several accessory factors. Indeed, this region appears to be structurally unstable when expressed in Escherichia coli or in mammalian cells as a single polypeptide, as evidenced by its rapid degradation (unpublished observation), despite the fact that the whole cytoplasmic domain can be easily expressed in these cells. In addition, a previous study (Aharonovitz et al, 2000) indicated that positively charged clusters (aa 509–516 and aa 552–560 in human NHE1) interact with polyphosphoinosides (PIP₂). These regions were also reported to interact with ezrin, although interaction with ezrin is independent of the exchange activity (Denker et al, 2000; Baumgartner et al, 2004). Furthermore, we recently reported that a region (aa 560–580) of NHE1 may be involved in dimeric interactions (Hisamitsu et al, 2004). Such homotypic interactions may provide another way for stabilization of the structure, as deletion of aa 560–580 resulted in a marked reduction of exchange activity by inducing the acidic shift of pH_i-dependence (Hisamitsu et al, 2004). These findings raise the possibility that multiple homotypic and heterotypic interactions may induce stabilization of the juxtamembrane region and thereby form a substructure regulating the exchange activity. Clearly, CHP is an obligatory member in such regulatory machinery of NHE1.

In summary, we first determined the crystal structure of CHP2 complexed with its binding region in NHE1 and clarified the mechanism by which CHP tightly and specifically associates with plasma membrane members of the NHE family. Structure-based mutagenesis revealed the physiological importance of CHP in pH_i regulation by NHE1. Most Ca²⁺-binding proteins have long been recognized to function as Ca²⁺-sensors or Ca²⁺-buffers. In contrast to this general understanding, CHP does not appear to be a Ca²⁺-sensor, but rather acts as a critical regulator of pH-sensing activity in the exchangers, presumably mediated by the CHP-unique region. The results of the present study cast new light on a unique function of CHP that was not reported in other canonical EF-hand Ca²⁺-binding proteins.

Protein expression and purification

The complex of human full-length CHP2 (aa 1–196) with its binding region (aa 503–545) in NHE1 was expressed and purified essentially as described previously (Pang et al, 2004; Ben Ammar et al, 2005).

Crystallization and data collection

Crystals of the complex CHP2/NHE1-peptide were grown using the sitting-drop vapor diffusion technique at 20°C. To obtain the best crystal quality, 1.5 l of the protein complex solution containing 20 mM yttrium chloride was mixed with 1 l of the reservoir solution containing 200 mM ammonium acetate, 100 mM Bis–Tris, pH 5.5, and 25% (w/v) PEG 3350 as described previously (Ben Ammar et al, 2005). Crystals were not obtained without adding yttrium chloride to the crystallization solution, suggesting that yttrium ion(s) were involved in crystal packing. Therefore, taking advantage of the yttrium ion(s) in the crystal, we collected multiple wavelength data sets around the absorption edge of the yttrium atom at SPring-8 beamline BL44B2. A high-resolution single-wavelength data set (=1 Å) was collected at beamline BL41XU at SPring-8. The statistics of the data collection are summarized in Table I.

Structural determination and refinement

The structure of the CHP2/NHE1-peptide complex was determined by MAD phasing. The program SOLVE (Terwilliger and Berendzen, 1999) was used to determine two yttrium positions and the initial phases with a figure of merit of 0.43 at 3.1 Å resolution. The program RESOLVE (Terwilliger, 2000) improved the phases and produced the initial polyalanine model, which covered 65% of the residues of the CHP2/NHE1–peptide complex. The model was built with TURBO-FRODO and refined against the high resolution data set to 2.7 Å using CNS (Brunger et al, 1998), and subsequent rounds of model building and refinement produced the final structural model. The residues in the final model lie in 87.2% in the most favorable and 12.8% in the additionally allowed regions of the Ramachandran plot. Refinement statistics are summarized in Table I. Figures 2, 3 and 4 were generated using PyMOL graphing software (http://www.pymol.org).

Construction of mammalian expression plasmids

The plasmid carrying the cDNA encoding NHE1 containing unique restriction sites cloned into the mammalian expression vector pECE was described previously (Wakabayashi et al, 1992). Construction of plasmids for various NHE1 or CHP2 mutants tagged with GFP or hemagglutinin (HA) epitope YPYDVPDYAS was carried out by a PCR-based strategy as described previously (Wakabayashi et al, 2000). PCR fragments were digested and cloned into the appropriate restriction sites of vectors pECE or pGEFP-N1 (Clontech, Palo Alto, CA) for NHE1 or CHP2, respectively.

Cell culture and plasmid transfection

The exchanger-deficient cell line PS120 (Pouyssegur et al, 1984) and corresponding transfectants were maintained in Dulbecco's modified Eagle's medium containing 25 mM NaHCO₃ and supplemented with 5% (v/v) fetal calf serum. All cDNA constructs were transfected into PS120 cells with Lipofectamine 2000 (Invitrogen), and stable clones for NHE1 and its mutant constructs were selected by repetitive H⁺-killing selection procedures as described previously (Wakabayashi et al, 1992). GFP-tagged CHP2 variants were stably expressed in cells expressing various mutant NHE1 proteins and fluorescence images were taken under a confocal fluorescence microscope (BioRad).

Measurement of ²²Na⁺-uptake and ²²Na⁺-efflux activities

²²Na⁺-uptake activity was measured by the K⁺/nigericin pH_i clamp method as described previously (Ikeda et al, 1997). Data were simulated by fitting the values to the sigmoidal dose–response equation, rate of EIPA-sensitive (pK, pH_i giving half maximal ²²Na⁺ uptake; n, Hill coefficient), using the simulation program included in Graphpad Prizm (Microsoft Corp., Redmond, WA). ²²Na⁺ efflux was measured as described previously (Wakabayashi et al, 2003a). Briefly, serum-depleted cells in 24-well dishes were loaded with ²²Na⁺ by preincubating them for 30 min at 37°C in chloride/KCl medium containing 1 mM ²²NaCl (37 kBq/ml) and at the same time pH_i-clamped at 7.5 or 7.2 in presence of 5 M nigericin. The radioactive solution was removed and ²²Na⁺ efflux was initiated by adding the nonradioactive medium. At the times indicated in figures, cells were rapidly washed four times with ice-cold PBS and the remaining ²²Na-radioactivity in the cells was counted.

Coordinates

Coordinates and structure factor amplitudes have been deposited in the Protein Data Bank with accession code 2BEC.

We thank the staff at beamline BL44B2 and BL41XU, SPring-8, for support with data collection, Dr Sugawara and Dr Miyano (RIKEN Harima Institute) for fruitful discussion, Dr Tomoe Y Nakamura for critical reading of manuscript, and Dr Tianxiang Pang for the initial participation in this study. This work was supported by Grant nano-001 for Research on Advanced Medical Technology from the Ministry of Health, Labor, and Welfare of Japan and Grant-in-Aid for Priority Areas 13142210 for Scientific Research from the Ministry of Education, Science, and Culture of Japan. YBA is a Japan Society for the Promotion of Science (JSPS) Postdoctoral Fellow.

Aharonovitz O, Zaun HC, Balla T, York JD, Orlowski J, Grinstein S (2000) Intracellular pH regulation by Na(+)/H(+) exchange requires phosphatidylinositol 4,5-bisphosphate. J Cell Biol 150: 213–224 | Article | PubMed | ISI | ChemPort |

Ames JB, Ishima R, Tanaka T, Gordon JI, Stryer L, Ikura M (1997) Molecular mechanics of calcium-myristoyl switches. Nature 389: 198–202 | Article | PubMed | ISI | ChemPort |

An WF, Bowlby MR, Betty M, Cao J, Ling HP, Mendoza G, Hinson JW, Mattsson KI, Strassle BW, Trimmer JS, Rhodes KJ (2000) Modulation of A-type potassium channels by a family of calcium sensors. Nature 403: 553–556 | Article | PubMed | ISI | ChemPort |

Aronson PS, Nee J, Suhm MA (1982) Modifier role of internal H+ in activating the Na+-H+ exchanger in renal microvillus membrane vesicles. Nature 299: 161–163 | Article | PubMed | ISI | ChemPort |

Barroso MR, Bernd KK, DeWitt ND, Chang A, Mills K, Sztul ES (1996) A novel Ca2+-binding protein, p22, is required for constitutive membrane traffic. J Biol Chem 271: 10183–10187 | PubMed | ISI | ChemPort |

Baumgartner M, Patel H, Barber DL (2004) Na(+)/H(+) exchanger NHE1 as plasma membrane scaffold in the assembly of signaling complexes. Am J Physiol Cell Physiol 287: C844–C850 | Article | PubMed | ISI | ChemPort |

Ben Ammar Y, Takeda S, Sugawara M, Miyano M, Mori H, Wakabayashi S (2005) Crystallization and preliminary crystallographic analysis of the human calcineurin homologous protein CHP2 bound to the cytoplasmic region of the Na+/H+ exchanger NHE1. Acta Crystallograph Sect F 61: 956–958 | ISI |

Bertrand B, Wakabayashi S, Ikeda T, Pouyssegur J, Shigekawa M (1994) The Na+/H+ exchanger isoform 1 (NHE1) is a novel member of the calmodulin-binding proteins. Identification and characterization of calmodulin-binding sites. J Biol Chem 269: 13703–13709 | PubMed | ISI | ChemPort |

Bourne Y, Dannenberg J, Pollmann V, Marchot P, Pongs O (2001) Immunocytochemical localization and crystal structure of human frequenin (neuronal calcium sensor 1). J Biol Chem 276: 11949–11955 | Article | PubMed | ISI | ChemPort |

Brunger AT, Adams PD, Clore GM, DeLano WL, Gros P, Grosse-Kunstleve RW, Jiang JS, Kuszewski J, Nilges M, Pannu NS, Read RJ, Rice LM, Simonson T, Warren GL (1998) Crystallography & NMR system: a new software suite for macromolecular structure determination. Acta Crystallogr D 54 (Part 5): 905–921 | Article

Counillon L, Pouyssegur J (2000) The expanding family of eucaryotic Na(+)/H(+) exchangers. J Biol Chem 275: 1–4 | Article | PubMed | ISI | ChemPort |

Denker SP, Huang DC, Orlowski J, Furthmayr H, Barber DL (2000) Direct binding of the Na–H exchanger NHE1 to ERM proteins regulates the cortical cytoskeleton and cell shape independently of H(+) translocation. Mol Cell 6: 1425–1436 | Article | PubMed | ISI | ChemPort |

Engelhardt S, Hein L, Keller U, Klambt K, Lohse MJ (2002) Inhibition of Na(+)-H(+) exchange prevents hypertrophy, fibrosis, and heart failure in beta(1)-adrenergic receptor transgenic mice. Circ Res 90: 814–819 | Article | PubMed | ISI | ChemPort |

Flaherty KM, Zozulya S, Stryer L, McKay DB (1993) Three-dimensional structure of recoverin, a calcium sensor in vision. Cell 75: 709–716 | Article | PubMed | ISI | ChemPort |

Gentry HR, Singer AU, Betts L, Yang C, Ferrara JD, Sondek J, Parise LV (2005) Structural and biochemical characterization of CIB1 delineates a new family of EF-hand-containing proteins. J Biol Chem 280: 8407–8415 | PubMed | ISI | ChemPort |

Gutierrez-Ford C, Levay K, Gomes AV, Perera EM, Som T, Kim YM, Benovic JL, Berkovitz GD, Slepak VZ (2003) Characterization of tescalcin, a novel EF-hand protein with a single Ca2+-binding site: metal-binding properties, localization in tissues and cells, and effect on calcineurin. Biochemistry 42: 14553–14565 | Article | PubMed | ISI | ChemPort |

Hisamitsu T, Pang T, Shigekawa M, Wakabayashi S (2004) Dimeric interaction between the cytoplasmic domains of the Na+/H+ exchanger NHE1 revealed by symmetrical intermolecular cross-linking and selective co-immunoprecipitation. Biochemistry 43: 11135–11143 | Article | PubMed | ISI | ChemPort |

Hunte C, Screpanti E, Venturi M, Rimon A, Padan E, Michel H (2005) Structure of a Na+/H+ antiporter and insights into mechanism of action and regulation by pH. Nature 435: 1197–1202 | Article | PubMed | ISI | ChemPort |

Ikeda T, Schmitt B, Pouyssegur J, Wakabayashi S, Shigekawa M (1997) Identification of cytoplasmic subdomains that control pH-sensing of the Na+/H+ exchanger (NHE1): pH-maintenance, ATP-sensitive, and flexible loop domains. J Biochem (Tokyo) 121: 295–303 | PubMed | ChemPort |

Inoue H, Nakamura Y, Nagita M, Takai T, Masuda M, Nakamura N, Kanazawa H (2003) Calcineurin homologous protein isoform 2 (CHP2), Na+/H+ exchangers-binding protein, is expressed in intestinal epithelium. Biol Pharm Bull 26: 148–155 | Article | PubMed | ISI | ChemPort |

Karmazyn M (2001) Role of sodium-hydrogen exchange in cardiac hypertrophy and heart failure: a novel and promising therapeutic target. Basic Res Cardiol 96: 325–328 | Article | PubMed | ISI | ChemPort |

Kissinger CR, Parge HE, Knighton DR, Lewis CT, Pelletier LA, Tempczyk A, Kalish VJ, Tucker KD, Showalter RE, Moomaw EW, Gastinel LN, Habuka N, Chen X, Maldonado F, Barker JE, Bacquet R, Villafranca JE (1995) Crystal structures of human calcineurin and the human FKBP12-FK506-calcineurin complex. Nature 378: 641–644 | Article | PubMed | ISI | ChemPort |

Lacroix J, Poet M, Maehrel C, Counillon L (2004) A mechanism for the activation of the Na/H exchanger NHE-1 by cytoplasmic acidification and mitogens. EMBO Rep 5: 91–96 | Article | PubMed | ISI | ChemPort |

Lin X, Barber DL (1996) A calcineurin homologous protein inhibits GTPase-stimulated Na-H exchange. Proc Natl Acad Sci USA 93: 12631–12636 | Article | PubMed | ChemPort |

Mailander J, Muller-Esterl W, Dedio J (2001) Human homolog of mouse tescalcin associates with Na(+)/H(+) exchanger type-1. FEBS Lett 507: 331–335 | PubMed | ISI | ChemPort |

Nagae M, Nozawa A, Koizumi N, Sano H, Hashimoto H, Sato M, Shimizu T (2003) The crystal structure of the novel calcium-binding protein AtCBL2 from Arabidopsis thaliana. J Biol Chem 278: 42240–42246 | Article | PubMed | ISI | ChemPort |

Naoe Y, Arita K, Hashimoto H, Kanazawa H, Sato M, Shimizu T (2005) Structural characterization of calcineurin B homologous protein 1. J Biol Chem 280: 32372–32378 | Article | PubMed | ISI | ChemPort |

Orlowski J, Grinstein S (2004) Diversity of the mammalian sodium/proton exchanger SLC9 gene family. Pflugers Arch 447: 549–565 | Article | PubMed | ISI | ChemPort |

Osawa M, Swindells MB, Tanikawa J, Tanaka T, Mase T, Furuya T, Ikura M (1998) Solution structure of calmodulin-W-7 complex: the basis of diversity in molecular recognition. J Mol Biol 276: 165–176 | Article | PubMed | ISI | ChemPort |

Pang T, Hisamitsu T, Mori H, Shigekawa M, Wakabayashi S (2004) Role of calcineurin B homologous protein in pH regulation by the Na+/H+ exchanger 1: tightly bound Ca2+ ions as important structural elements. Biochemistry 43: 3628–3636 | Article | PubMed | ISI | ChemPort |

Pang T, Su X, Wakabayashi S, Shigekawa M (2001) Calcineurin homologous protein as an essential cofactor for Na+/H+ exchangers. J Biol Chem 276: 17367–17372 | Article | PubMed | ISI | ChemPort |

Pang T, Wakabayashi S, Shigekawa M (2002) Expression of calcineurin B homologous protein 2 protects serum deprivation-induced cell death by serum-independent activation of Na+/H+ exchanger. J Biol Chem 277: 43771–43777 | Article | PubMed | ISI | ChemPort |

Perera EM, Martin H, Seeherunvong T, Kos L, Hughes IA, Hawkins JR, Berkovitz GD (2001) Tescalcin, a novel gene encoding a putative EF-hand Ca(2+)-binding protein, Col9a3, and renin are expressed in the mouse testis during the early stages of gonadal differentiation. Endocrinology 142: 455–463 | Article | PubMed | ISI | ChemPort |

Pouyssegur J, Sardet C, Franchi A, L'Allemain G, Paris S (1984) A specific mutation abolishing Na+/H+ antiport activity in hamster fibroblasts precludes growth at neutral and acidic pH. Proc Natl Acad Sci USA 81: 4833–4837 | PubMed | ISI | ChemPort |

Putney LK, Denker SP, Barber DL (2002) The changing face of the Na+/H+ exchanger, NHE1: structure, regulation, and cellular actions. Annu Rev Pharmacol Toxicol 42: 527–552 | Article | PubMed | ISI | ChemPort |

Terwilliger TC (2000) Maximum-likelihood density modification. Acta Crystallogr D 56 (Part 8): 965–972 | Article

Terwilliger TC, Berendzen J (1999) Automated MAD and MIR structure solution. Acta Crystallogr D 55 (Part 4): 849–861 | Article

Vijay-Kumar S, Kumar VD (1999) Crystal structure of recombinant bovine neurocalcin. Nat Struct Biol 6: 80–88 | Article | PubMed | ISI | ChemPort |

Wakabayashi S, Bertrand B, Ikeda T, Pouyssegur J, Shigekawa M (1994) Mutation of calmodulin-binding site renders the Na+/H+ exchanger (NHE1) highly H(+)-sensitive and Ca2+ regulation-defective. J Biol Chem 269: 13710–13715 | PubMed | ISI | ChemPort |

Wakabayashi S, Fafournoux P, Sardet C, Pouyssegur J (1992) The Na+/H+ antiporter cytoplasmic domain mediates growth factor signals and controls 'H(+)-sensing'. Proc Natl Acad Sci USA 89: 2424–2428 | PubMed | ChemPort |

Wakabayashi S, Hisamitsu T, Pang T, Shigekawa M (2003a) Kinetic dissection of two distinct proton binding sites in Na+/H+ exchangers by measurement of reverse mode reaction. J Biol Chem 278: 43580–43585 | Article | PubMed | ISI | ChemPort |

Wakabayashi S, Hisamitsu T, Pang T, Shigekawa M (2003b) Mutations of Arg440 and Gly455/Gly456 oppositely change pH sensing of Na+/H+ exchanger 1. J Biol Chem 278: 11828–11835 | Article | PubMed | ISI | ChemPort |

Wakabayashi S, Pang T, Su X, Shigekawa M (2000) A novel topology model of the human Na(+)/H(+) exchanger isoform 1. J Biol Chem 275: 7942–7949 | Article | PubMed | ISI | ChemPort |

Wakabayashi S, Shigekawa M, Pouyssegur J (1997) Molecular physiology of vertebrate Na+/H+ exchangers. Physiol Rev 77: 51–74 | PubMed | ISI | ChemPort |

Zachos NC, Tse M, Donowitz M (2005) Molecular physiology of intestinal Na+/H+ exchange. Annu Rev Physiol 67: 411–443 | Article | PubMed | ISI | ChemPort |

Zhang M, Tanaka T, Ikura M (1995) Calcium-induced conformational transition revealed by the solution structure of apo calmodulin. Nat Struct Biol 2: 758–767 | Article | PubMed | ISI | ChemPort |

Zhou W, Qian Y, Kunjilwar K, Pfaffinger PJ, Choe S (2004) Structural insights into the functional interaction of KChIP1 with Shal-type K(+) channels. Neuron 41: 573–586 | Article | PubMed | ISI | ChemPort |

Main navigation

• Search EMBO Journal

  Search

• Advanced search

• The EMBO Journal

  • Home
  • Aims and scope
  • Current issue
  • Advance online publication
  • Web focuses
  • Archive:
  • Browse by issue
  • Browse by subject
  • Browse by category
  • Free online sample issue
  • Press releases

• Authors and Referees

  • Editorial process
  • Guide for authors
  • Submit an article
  • Guide for referees
  • Editors and Editorial Board
  • Contact editorial office

• Customer Service

  • Subscribe
  • Order sample copy
  • Purchase articles
  • Reprints and permissions
  • Contact NPG
  • Advertising