Trinitrophenyl derivatives bind differentlyfrom parent adenine nucleotides toCa2þ-ATPase in the absence of Ca2þChikashi Toyoshima (豊島 近)1, Shin-Ichiro Yonekura, Junko Tsueda, and Shiho Iwasawa

Institute of Molecular and Cellular Biosciences, University of Tokyo, Tokyo 113-0032, Japan

Contributed by Chikashi Toyoshima, November 24, 2010 (sent for review September 29, 2010)

Trinitrophenyl derivatives of adenine nucleotides are widely usedfor probing ATP-binding sites. Here we describe crystal structuresof Ca2þ-ATPase, a representative P-type ATPase, in the absence ofCa2þ with bound ATP, trinitrophenyl-ATP, -ADP, and -AMP at betterthan 2.4-Å resolution, stabilized with thapsigargin, a potent inhi-bitor. These crystal structures show that the binding mode of thetrinitrophenyl derivatives is distinctly different from the parentadenine nucleotides. The adenine binding pocket in the nucleotidebinding domain of Ca2þ-ATPase is now occupied by the trinitrophe-nyl group, and the side chains of two arginines sandwich theadenine ring, accounting for the much higher affinities of thetrinitrophenyl derivatives. Trinitrophenyl nucleotides exhibit apronounced fluorescence in the E2P ground state but not in theother E2 states. Crystal structures of the E2P and E2 ∼ P analoguesof Ca2þ-ATPase with bound trinitrophenyl-AMP show that differ-ent arrangements of the three cytoplasmic domains alter theorientation and water accessibility of the trinitrophenyl group,explaining the origin of “superfluorescence.” Thus, the crystalstructures demonstrate that ATP and its derivatives are highlyadaptable to a wide range of site topologies stabilized by a varietyof interactions.

crystallography ∣ ion pump ∣ nucleotide derivatives

Trinitrophenyl (TNP)-nucleotides (1) are often used for prob-ing the structure of ATP-binding sites and conformational

changes arising from nucleotide binding (2, 3), and for measuringthe affinity of ATP by competition experiments (2, 4). It is apreferred ATP analogue for photochemical crosslinking withazide derivatives (5). These applications utilize the enhancementof fluorescence or absorption of visible light of the TNP groupupon binding to a protein (6). Because of its sensitivity, competi-tion with ATP/ADP has been a valuable means for examiningmutational effects on nucleotide affinity (7). Thus, TNP nucleo-tides have been widely used with F1 (8), myosin (1), and P-typeATPases (2–5, 7, 9, 10), among others.

Nonetheless, whether TNP derivatives are good mimics ofauthentic adenine nucleotides (AxPs) may be questionable. Inseveral proteins TNP nucleotides have much higher affinitiesthan the genuine AxPs. For instance, TNP-ATP is a high affinity(nM) antagonist of P2X receptors, which have IC50 for ATP(or AMPPCP) in the μM range (11). The affinity is at least oneorder of magnitude higher in the E2 states of Ca2þ-ATPase(3, 12) and Naþ, Kþ-ATPase (4), representative P-type ATPases.Furthermore, TNP-AMP binds to Ca2þ-ATPase similarly to oreven more strongly than TNP-ATP (12), in marked contrast toAxPs. Thus, a substantially different binding mode of TNP deri-vatives is suggested. Although more than 20 entries are registeredin the Protein Data Bank (PDB) for Ca2þ-ATPase (reviewed inref. 13), no structure with a bound TNP nucleotide exists. In fact,only three crystal structures have been published with boundTNP nucleotides. They are a bacterial histidine kinase CheA (14),an ABC transporter HlyB (15), and an adenylyl cyclase (16),in which TNP nucleotides bind to the proteins in a somewhat

distorted manner but otherwise similarly to AxPs (or their closeanalogues).

Here we report crystal structures of Ca2þ-ATPase from skele-tal muscle sarcoplasmic reticulum (SERCA1a) with boundTNP-ATP, -ADP, -AMP, as well as ATP and ADP in the absenceof Ca2þ (E2-state) and stabilized with thapsigargin (TG), a verypotent inhibitor (17). Ca2þ-ATPase consists of three cytoplasmicdomains designated A (actuator), N (nucleotide binding), and P(phosphorylation), and 10 transmembrane helices (18). Previouscrystal structures in the presence of Ca2þ have shown that ATP(nonhydrolysable analogue adenosine 5′-(β,γ-methylene) tripho-sphate (AMPPCP) (19, 20) and adenosine 5′-(β,γ-imido)tripho-sphate (21) were used in place of ATP) binds near the hingebetween the N and P domains and cross-links them. The tripho-sphate chain is extended with no Mg2þ bound and the adeninering stacks with Phe487 in the N domain. In the absence of Ca2þ,AMPPCP takes a folded conformation due to bound Mg2þ(22, 23). Here we show that the binding mode of TNP-AxPs tothe ATPase in the E2-state is distinctly different, with the TNPring stacking with Phe487 and the adenine ring sandwichedbetween two arginines.

TNP-nucleotides exhibit greatly enhanced fluorescence(“superfluorescence”) (24) in certain reaction intermediates ofCa2þ-ATPase. Superfluorescence is characteristic of the E2Pground state and not observed with the transition state analogue(25). Hence TNP-AMP bound structures were also determinedfor E2 · BeF3

− (E2P ground state analogue) and E2 · AlF4−

(E2 ∼ P transition state analogue), to explore the origin of thesuperfluorescence. The crystal structures show that the orienta-tion of the TNP ring and the water accessibility vary with thearrangement of the three cytoplasmic domains, thereby explain-ing the different levels of fluorescence in different reaction inter-mediates.

ResultsBinding Mode of ATP and ADP in E2(TG). Crystals of Ca2þ-ATPase inthe E2(TG), E2 · AlF4

−ðTGÞ and E2 · BeF3−ðTGÞ forms were

generated in the presence of TG by microdialysis. Nucleotideswere introduced by placing the dialysis buttons in buffers contain-ing nucleotides. Because the overall structural changes weresmall, molecular replacement based on the atomic model builtfor E2ðTGþ BHQÞ (PDB ID code 2AGV) (26) went smoothly(Table S1).

Author contributions: C.T. designed research; C.T., S.-I.Y., J.T., and S.I. performed research;C.T. and S.-I.Y. analyzed data; and C.T. wrote the paper.

The authors declare no conflict of interest.

Data deposition: Coordinates and structural factors have been deposited in the ProteinData Bank, www.pdb.org (PDB ID codes 3AR4 [E2 · ATPðTGÞ], 3AR3 [E2 · ADPðTGÞ],3AR7 [E2 · TNP − ATPðTGÞ], 3AR6 [E2 · TNP − ADPðTGÞ], 3AR5 [E2 · TNP − AMPðTGÞ], 3AR9[E2 · BeF3

− · TNP-AMPðTGÞ], 3AR8 [E2 · AlF4− · TNP-AMPðTGÞ], and 3AR2 (E1 · AMPPCP).

1To whom correspondence should be addressed. E-mail: ct@iam.u-tokyo.ac.jp.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1017659108/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1017659108 PNAS ∣ February 1, 2011 ∣ vol. 108 ∣ no. 5 ∣ 1833–1838

BIOCH

EMISTR

Y

Dow

nloa

ded

by g

uest

on

Janu

ary

23, 2

021

The crystal structure with bound ATP in the absence of Ca2þstabilized with TG [abbreviated as E2 · ATPðTGÞ] was deter-mined at 2.15-Å resolution (Fig. 1A; Fig. S1 for stereo) and isvirtually identical to that with bound AMPPCP described pre-viously (PDB ID code 2DQS) (22). The adenine ring of ATPis inserted into the binding cavity in the N domain of Ca2þ-AT-Pase, flanked by Phe487 on one side and Leu562 on the other.This binding mode is the same, but the stacking of the adeninering with the Phe487 side chain is not as optimal as in theE1 · AMPPCP crystal structure, now refined to 2.5-Å resolution(Fig. 1B). The orientation is ∼25° offset from parallel, primarilydue to a different orientation of the Phe487 main chain. Thebinding cavity is locally distorted (Fig. 2), due to salt bridgesformed between Arg489 (N domain) and Asp203 (A domain)and between Glu486 (N) and His190 (A).

In the E1 · AMPPCP crystal structure, AMPPCP takes anextended conformation with no Mg2þ bound (Fig. 1B) (19, 20).In E2 · ATPðTGÞ, if Mg2þ is present, the triphosphate chaintakes a folded conformation and coordinates Mg2þ with all threephosphate groups (Fig. 1A). This position of Mg2þ was confirmedby substituting Mg2þ with Mn2+ and calculating an anomalous dif-ference map for Mn2þ and an jFobsj-jFobsj difference Fourier map(i.e., jFobs½E2 · ATPðMn2þÞ�j-jFobs½E2 · ATPðMg2þÞ�j) (Fig. S2).This Mg2þ is also coordinated by three water molecules, twoof which are fixed by hydrogen bonds with Glu439 (Fig. 1A).The carboxyl group of Glu439 itself is more than 3 Å away fromthe Mg2þ. This coordination geometry is different from thatreported by Jensen et al. (23) for the AMPPCP-bound structure(PDB ID codes 2C88 and 2C8K) at 2.8 to 3.1-Å resolution. Theγ-phosphate points away from Asp351, the residue phosphory-lated during the catalytic cycle, due to the position of Mg2þbut is stabilized by a water molecule hydrogen bonded to Gln202in the A domain. Thus, Mg2þ will enhance the binding of ATP toE2(TG) (Kd ¼ 150 μM without Mg2þ and 100 μM with Mg2þ)(23). Nonetheless, the affinity is still much lower than that inthe presence of Ca2þ (3 μM) (27).

The difference maps also located another Mg2þ in the P do-main. It is coordinated by Asp703 and Asp707 (Fig. 1A), insteadof Asp351 as in E1 · AMPPCP and E1 · AlF4

− · ADP. [However,the metal ion in the E1 · AMPPCP crystal structure is most likelyto be Ca2þ rather than Mg2þ (22, 28).] The affinity of this Mg2þappears low because the coordination geometry is far from anideal octahedral one.

The lower affinity for ATP of E2(TG) reflects the lack of directinteractions with residues in the P domain. In the E1 · AMPPCPcrystal structure, AMPPCP crosslinks the N and P domainsby forming many hydrogen bonds (Fig. 1B): They include thosebetween ribose and Arg678 (P), α-phosphate and Arg489 (N),β-phosphate and Asp627 (P)/Arg560 (N), and γ-phosphate andThr353 (P)/Gly626 (P) (19, 20). In E2 · ATPðTGÞ, most of thesesalt bridges are absent (Fig. 1A) because the three cytoplasmicdomains are arranged differently. Instead, the triphosphate chainof ATP is stabilized by Mg2þ-coordination. Consequently, thefolded form, with the γ-phosphate turned away from Asp351,is preferred (Fig. 1A). Then, when the cytoplasmic headpieceopens in transition to E1, ATP binding affinity rises because localdistortion is relieved, in preparation for efficient phosphoryltransfer. If the energy cost of distortion is large, relief of distor-tion (i.e., opening of the headpiece) alone will stabilize the systemand drive the reaction cycle forward.

The crystal structure of the ADP-bound form was determinedat 2.3-Å resolution. Binding geometry of ADP is essentiallythe same as ATP, but the position of the β-phosphate is differentbecause the coordination of Mg2þ is possible only with the α-and β-phosphates. Here the Glu439 side chain is fully extended(Fig. S3).

K515

R560

L562

R489

F487

T441

L445

E442

R678

K492

T353

D703

D707

K684

D351

K515

R560

L562

R489

F487

T441

L445

E442

R678

K492

T353

D703

D707D351

K515

R560

L562

R489

F487T441

L445

E442

R678

K492

T353

D703

D707 D351T625

A

B

C

E439

E439

E439

E680

E680

E680

D627

D627

D627

z y

x

Fig. 1. Binding mode of ATP (or AMPPCP) and TNP-ATP in Ca2þ-ATPase.(A) ATP in the absence of Ca2þ but in the presence of thapsigargin[E2 · ATPðTGÞ], pH 6.1, at 2.15-Å resolution; (B) AMPPCP in the presence ofCa2þ (E1 · AMPPCP) at 2.5-Å resolution; (C) TNP-ATP in the absence ofCa2þ but in the presence of thapsigargin [E2 · TNP-ATP(TG)] at 2.15-Å resolu-tion. The violet net in C represents an jFobsj-jFcalc j electron density map (con-toured at 3σ), before introducing TNP-ATP into the atomic model. The Ndomain appears green and the P domain yellow. Small red spheres representwater molecules; cyan spheres, Ca2þ; green spheres, Mg2þ. Small cyan disksrepresent main chain amide. Broken lines in blue show hydrogen bonds andmetal coordination. A stereo version of this figure is presented as Fig. S1.

1834 ∣ www.pnas.org/cgi/doi/10.1073/pnas.1017659108 Toyoshima et al.

Dow

nloa

ded

by g

uest

on

Janu

ary

23, 2

021

Binding of TNP-ATP in E2(TG). TNP nucleotides cannot bind to theATPase in the same mode as ATP, because the TNP ring wouldcollide with Pro518, Ala517, and Cys561 (Fig. S4). Instead, thePhe487 side chain stacks with the TNP ring in an ideal geometry(Fig. 1C). Moreover, the adenine ring is now sandwiched bythe long side chains of Arg489 (on the N domain) and Arg678(P domain), both in fully extended conformations. One of thethree NO2-groups on the TNP ring forms a hydrogen bond withthe Lys515 side chain. No residues form direct hydrogen bondswith the adenine ring, but the Glu680 carboxyl and the Val679carbonyl fix a water molecule that is hydrogen bonded to N6of the adenine ring (Fig. 1C).

The β-phosphate is stabilized by the Thr353 hydroxyl andGly626 amide, in a way very similar to the γ-phosphate in theE1 · AMPPCP crystal structure (Fig. 1B). Because of this, the tri-phosphate chain is extended and takes a zigzag conformation.The γ-phosphate is free with no Mg2þ bound, even though excessMg2þ was present in the crystallization medium. It comes close toAsp351, even closer than in E1 · AMPPCP, and a water moleculebridges them (small arrow in Fig. 1C).

Thus, TNP-ATP bridges the N and P domains, similar toAMPPCP in the E1 · AMPPCP crystal structure. Likewise Arg560as in E1 · AMPPCP bridges Thr441 (N) and Asp627 (P), furtherstabilizing N and P domain interactions (Fig. 1C). In E2 · ATP,Arg560 bridges only Thr441 and the β-phosphate (Fig. 1A).

Binding of TNP-ADP and TNP-AMP in E2(TG). The binding mode ofTNP-ADP in E2(TG) is essentially the same as that of the tripho-sphate form (Fig. S3), and the β-phosphate occupies the sameposition as the γ-phosphate in E1 · AMPPCP, suggesting thatTNP-ADP might be hydrolyzed even in the absence of Ca2þ.

The binding mode of TNP-AMP in E2(TG) (Fig. 3A) is some-what different from that of TNP-ATP in that the α-phosphategroup of TNP-AMP, stabilized by the Thr353 hydroxyl and theGly626 amide, occupies the position of the β-phosphate of TNP-ATP/ADP. The positions of the TNP group and the adenine ringare slightly shifted so that the whole molecule inclines towardThr353, which forms a hydrogen bond with the α-phosphate. Thisorientation is stabilized by a hydrogen bond between Glu439 anda NO2 group in the TNP ring, in addition to the one between theLys515 and another NO2 group.

Binding of TNP-AMP in E2P and E2 ∼ P Analogues. TNP-AMP hasbeen used for probing structural changes around the nucleotidebinding site of Ca2þ-ATPase. There is particular interest in theorigin of the superfluorescence of TNP-AMP in the E2P groundstate. E2 · BeF3

− is a stable analogue of the E2P ground state,and superfluorescence is observed even in the presence of TG

(25). This is in marked contrast to E2 · AlF4−, a stable transition

state analogue, in which TNP-AMP exhibits little fluorescence(25). Hence we soaked TNP-AMP into E2 · BeF3

−ðTGÞ andE2 · AlF4

−ðTGÞ crystals. These crystals diffracted to at least2.6-Å resolution, which allowed us to visualize tightly boundwater molecules as well as Mg2þ.

The orientation of the TNP ring in E2 · AlF4−ðTGÞ is similar

to that in E2(TG), although Arg678 no longer sandwiches theadenine ring, due to a different position of the P-domain (Fig. 3B).However, the orientation in E2 · BeF3

−ðTGÞ is distinctly differ-ent, being 60° offset from parallel (Fig. 3C and Fig. S5). Thisappears to be an unfavorable orientation, but probably not asmuch as one would think, as the hydrogen bond with Lys515remains and, furthermore, Thr441 and Arg560 form three uniquehydrogen bonds with another NO2 group. Even Arg174 in theA-domain forms a hydrogen bond with the ribose.

The environment of the TNP ring varies substantially and thisexplains the origin of superfluorescence. In E2 · BeF3

−ðTGÞ,there is a tight pocket around the TNP ring, excluding watermolecules (Fig. 4C). TNP-AMP itself is inserted somewhat moredeeply than in E2 · AlF4

−ðTGÞ. Here, Arg174 on the A domain,backed by Ile188 also on the A domain, lines a part of the uppersurface of the binding pocket. In E2 · AlF4

−ðTGÞ and E2(TG)(Fig. 4A and B), there is a relatively large open space on theopposite side of the TNP ring to Phe487, allowing access ofwater molecules (Fig. 2 and 4). Thus, a very hydrophobic envir-onment is realized only in E2 · BeF3

−ðTGÞ, consistent with super-fluorescence in the E2P state.

The orientation of the TNP ring appears to be determined bythe position of Arg174 on the A domain. During the transitionfrom E1P → E2P → E2 · Pi, the A domain rotates approximately110° around an axis approximately perpendicular to the mem-brane (29). As a result, the position of Arg174, located on theA–N interface, changes drastically. In the E2PðE2 · BeF3

−Þ →E2 ∼ PðE2 · AlF4

−Þ transition, the A domain rotates 20° (30)and Arg174 moves approximately 2.3 Å at the Cα atom. This see-mingly small change is enough to alter the hydrogen bondingpartner of Arg174. In the E2 · BeF3

−ðTGÞ crystal structure,Arg174 NH2 forms both a hydrogen bond with the ribose andcation–π interactions with Phe487 and the TNP ring of TNP-AMP (Fig. 3C). When TNP-AMP is absent, the Arg174 side chainoccupies a different position, being fixed by hydrogen bonds withthe Ser186 hydroxyl and the main chain carbonyl (29). InE2 · AlF4

−ðTGÞ, Arg174 forms hydrogen bonds with one ofthe NO2 groups on the TNP ring, fixing the TNP ring in a differ-ent orientation (Fig. 3B).

The position of the phosphate group also varies (Fig. 3). In E2(TG), the phosphate group is stabilized by the Thr353 hydroxyland the Gly626 amide (Fig. 3A). In contrast, in E2 · BeF3

−ðTGÞ,it is stabilized by a hydrogen bond with the Cys561 amide(Fig. 3C), pointing away from Asp351, the phosphorylated resi-due. In none of the crystal structures is Mg2þ bound to TNP-AMP.

DiscussionCrystal Structures of Other Proteins with Bound TNP Nucleotides. ThePDB has only three structures with bound TNP nucleotides:They are CheA with TNP-ATP (PDB ID code 1I5D) (14); HlyB(an ABC transporter) with TNP-ADP (PDB ID code 2PMK)(15); and adenylyl cyclase with TNP-ATP (PDB ID code 2GVD)(16). In all of them, TNP derivatives bind to the proteins withsomewhat distorted but similar modes to the original AxPs.TNP-ATP binds with a >500× higher affinity to CheA, >50×to HlyB, and >1;000× to adenylyl cyclase. TNP-ATP binds toCheA with the TNP ring inserted into a hydrophobic pocketadjacent to the adenine binding site, thereby conferring a higheraffinity (14). In adenylyl cyclase the adenosine moiety is flippedso that the TNP ring is stabilized by several hydrogen bonds(16, 31). The HlyB crystal structure with bound TNP-ADP shows

R489F487

E486

H190

D203R489 F487

E486

H190

D203A

N

A

N

xyz

yz x

Fig. 2. Local distortion of the adenine binding site of Ca2þ-ATPase in theabsence of Ca2þ. Superimposition of the atomic models for E2 · ATPðTGÞ(atom color) and E1 · AMPPCP (green for the N domain, pink for the Adomain; AMPPCP, orange). Broken lines in blue show hydrogen bonds andcoordinations of Mg2þ.

Toyoshima et al. PNAS ∣ February 1, 2011 ∣ vol. 108 ∣ no. 5 ∣ 1835

BIOCH

EMISTR

Y

Dow

nloa

ded

by g

uest

on

Janu

ary

23, 2

021

virtually an identical binding mode (15). Thus, the binding modeof TNP-AxPs described here is distinctly different from those inprevious reports.

Origin of the High Affinity Binding of TNP Nucleotides to Ca2þ-ATPase.TNP modification confers a much higher (>50 times) affinity toadenine nucleotides for Ca2þ-ATPase. Reported Kd for TNP-ATP is 0.1–0.2 μM (3, 12) and that for ATP is around 10 μM[Kd ¼ 10.1 μM (12); 5–9 μM (22); 4 μM (3)] in the absence ofCa2þ. This higher affinity arises primarily because both adenineand TNP moieties are stabilized by hydrophobic interactions withPhe487, Arg489, and Arg678 (Fig. 1C). Stabilization of the phos-phate chain by hydrogen bonds with the Thr353 hydroxyl and theGly626 amide may also contribute. The β-phosphate is stabilizedin TNP-ATP and -ADP, whereas the α-phosphate is stabilized inTNP-AMP (Fig. 3A). It is evident that TNP-ATP would have alower affinity, because the γ-phosphate in TNP-ATP is locatedfairly close to the cluster of aspartates (i.e., Asp351, Asp703, andAsp707) around the phosphorylation site (Fig. 1C). However, it is

obscure why the affinity of TNP-AMP is even higher than that ofTNP-ADP. In fact, Seebregts and McIntosh (5) reported thatTNP-8N3-AMP has a Kd of 0.1 μM and TNP-8N3-ADP∕ATPhave 0.2 μM Kd. Suzuki et al. (12) reported Kds of 7.62 nMand 156 nM for TNP-AMP and -ATP, respectively. Unfortunately,as far as we know, there is no literature that reports on the affinityof all three under the same conditions.

The observation that TNP-AMP binds to E2 · AlF4− and

E2 · BeF3− with high affinity (32) suggests that the primary

origin of the high affinity of TNP-AxPs is hydrophobic stabiliza-tion of the TNP ring. In these two complexes, there is no directstabilization of the adenine ring (Fig. 3 B and C). Yet the surfaceof the binding pocket of the TNP ring is complementary, inparticular, in E2 · BeF3

−ðTGÞ (Fig. 4C).

Explanation of Photolabeling Results.McIntosh reported that TNP-8N3-AMP and -ATP label Lys492 in Ca2þ-ATPase (33). Inesireported that TNP-2N3-AMP is much more efficient than the8N3 derivative in labeling Lys492 (34). For a TNP-8N3-AxP to be

K515

R560

M599

L562

R489

F487

T441

L445E442 K515

R560

M599

L562

R489

F487

T441

L445E442

K515

R560

M599

L562

R489

F487 T441

L445

E442K515

R560

M599

L562

R489

T441

L445

E442

M599

L562

R489

F487

T441

E442

R678 R560

M599

L562

R489

F487

T441

E442

R678

K492 K492

K492 K492

K492 K492

R560

A

B

C

R174 R174

R174 R174

E439 E439

T353 T353

E439 E439

E680E680 G626G626

K515 K515

F487

L445 L445

C561 C561

xyz

xyz

Fig. 3. Binding of TNP-AMP in various intermediates (or analogues) of Ca2þ-ATPase in the absence of Ca2þ. (A) E2(TG), (B) E2 · AlF4−ðTGÞ, an E2 ∼ P transition

state analogue, and (C) E2 · BeF3−ðTGÞ, an E2P ground state analogue. The A, N, and P domains are colored pink, green, and yellow, respectively. Blue broken

lines show hydrogen bonds. Small red spheres represent water molecules; small cyan disks, main chain amide. Carbon atoms in the TNP group are colored gray.

1836 ∣ www.pnas.org/cgi/doi/10.1073/pnas.1017659108 Toyoshima et al.

Dow

nloa

ded

by g

uest

on

Janu

ary

23, 2

021

able to label Lys492, the adenine ring has to flip with respectto the ribose (Fig. 1C), which will cause a considerable stericproblem. In contrast, it is evident that photoaffinity labeling ofLys492 is impossible in the canonical binding mode of AxP, asan azido group attached at the 8 position, and Lys492 cannotcome closer than 6 Å. Thus these reports completely agree withthe crystal structure presented here.

Caviers and colleagues used TNP-8N3-ADP photoaffinitylabeling to identify ATP binding residues in Naþ, Kþ-ATPase(35). With native Naþ, Kþ-ATPase, labeling occurred on theN-domain residue Lys480, which is equivalent to Lys492 inCa2þ-ATPase, indicating that TNP-ADP binds to Naþ, Kþ-AT-Pase similarly to Ca2þ-ATPase. With the FITC-modified enzyme,labeling occurred in a highly conserved region in the P domain(Ala714-Lys728 in Ca2þ-ATPase) just prior to the cytoplasmicend of the M5 helix (35). In the crystal structures, there wasno electron density suggesting a nucleotide in this position,although the concentration of TNP-ADP was as high as 5.6 mM.

Fluorescence of TNP-AMP in E1 · AlF4− and E1 · BeF3

−.Nakamoto andInesi reported that TNP nucleotides fluoresce strongly in E1P(36). Suzuki and colleagues have shown that fluorescence fromTNP-AMP in E1 · AlF4

− is low whereas that in E1 · BeF3− very

high and proposed that the A domain in E1 · BeF3− is located

between those seen in E1 · AlF4− · ADP and E2 · BeF3

− (32).The position of TNP-AMP in either E2(TG) or E2 · BeF3

−ðTGÞwould be incompatible with its binding in E1P, if the struc-ture around the phosphorylation site were the same as in theE1 · AlF4

− · ADP crystal structure (19, 20), because the adeninering would collide with the P domain. The adenine binding site isfairly open to bulk solvent in the crystal structure, and the fluor-escence of TNP-AMP will be correspondingly weak in E1 · AlF4

−.Even though the side chain of Arg174 on the A domain is fairlyfree to move, the A domain should be located within a ratherrestricted range for a hydrogen bond to be formed with the ri-bose. In fact, addition of TNP-AMP to E1 · BeF3

− slowly convertsit into E1 · 2Ca2þ, judging from proteolysis patterns (32). Thus,our crystal structures corroborate the idea that the position of theA domain and the structure of the phosphorylation site in E1Pare different from those in the E1 · AlF4

− · ADP crystal structure(32).

ConclusionThe crystal structures described here demonstrate that AxPs andTNP-AxPs are highly adaptable and can bind to an ATP bindingsite in widely different modes, stabilized by π–π interaction,amine-aromatic sandwiching, and hydrogen bonding. Mg2þ mayalso alter the orientation of the whole nucleotide through itscoordination of the polyphosphate chain and protein residues.These various interactions provide means for adjusting the affi-nity for ATP in different steps of a reaction cycle. The bindingmodes of TNP-AxPs shown here, with the adenine ring stabilizedin distinctly different ways, suggest that authentic AxPs maybind similarly, although with lower affinity. A further corollary ofthis study is that derivatives or inhibitors specific to a particularprotein may be designed by considering the molecular topologyof an area substantially larger than the canonical adenosinebinding site.

MethodsCrystallization. Crystals of Ca2þ-ATPase were prepared from affinity purifiedenzyme by microdialysis. E2(TG) (26), E2 · BeF3

−ðTGÞ (30), and E2 · AlF4−ðTGÞ

(30) crystals were prepared as before. Nucleotides were introduced intocrystals by placing dialysis buttons used for crystallization overnight intothe flash freezing buffer that contained one of the following: 10 mM ATP,10 mM ADP, 10 mM TNP-ATP, 5.6 mM TNP-ADP, or 2.5 mM TNP-AMP.Mg2þ concentration was 1 mM higher than that of nucleotide.

Data Collection and Structure Determination. Crystals were picked with nylonloops in a cold room and flash-frozen in cold nitrogen gas. E2 · BeF3

−ðTGÞcrystals were left to stand for 2.5 min for dehydration after removing excessbuffer (30). Diffraction data were collected from crystals cooled to 100 K atBL41XU of SPring-8 with an ADSC Q315 CCD detector, and processed withDenzo and Scalepack (37). All the E2-nucleotide structures were determinedby molecular replacement using CNS (38) starting from the models publishedpreviously without nucleotides (26, 30). The atomic model of E1 · AMPPCPwas refined with REFMAC including TLS refinement (39). Statistics of thediffraction data and refinement are given in Tables S1 and S2.

ACKNOWLEDGMENTS. We thank Drs. N. Shimizu (Japan Synchroton RadiationResearch Institute) and T. Tsuda (now at Gakushuin University) for datacollection at BL41XU of SPring-8, and Y. Norimatsu (now at Fujitsu) forstructure refinement. We are indebted to Dr. D. B. McIntosh for his helpin improving the manuscript. This work is a part of an ongoing long-termproject (2009B0025) at SPring-8 and was supported in part by a SpeciallyPromoted Project Grant (to C. T.) from the Ministry of Education, Culture,Sports, Science and Technology of Japan.

1. Hiratsuka T, Uchida K (1973) Preparation and properties of 2′(or 3′)-O-(2,4,6-trinitro-phenyl) adenosine 5′-triphosphate, an analog of adenosine triphosphate. BiochimBiophys Acta 320:635–647.

2. Watanabe T, Inesi G (1982) The use of 2′,3′-O-(2,4,6-trinitrophenyl) adenosine5′-triphosphate for studies of nucleotide interaction with sarcoplasmic reticulumvesicles. J Biol Chem 257:11510–11516.

3. Dupont Y, Chapron Y, Pougeois R (1982) Titration of the nucleotide binding sitesof sarcoplasmic reticulum Ca2þ-ATPase with 2′,3′-O-(2,4,6-trinitrophenyl) adenosine5′-triphosphate and 5′-diphosphate. Biochem Biophys Res Commun 106:1272–1279.

4. Moczydlowski EG, Fortes PA (1981) Characterization of 2′,3′-O-(2,4,6-trinitrocyclohexa-dienylidine)adenosine 5′-triphosphate as a fluorescent probe of the ATP site of sodiumand potassium transport adenosine triphosphatase. J Biol Chem 256:2346–2356.

5. Seebregts CJ, McIntosh DB (1989) 2′,3′-O-(2,4,6-trinitrophenyl)-8-azido-adenosinemono-, di-, and triphosphates as photoaffinity probes of the Ca2þ-ATPase of sarcoplas-mic reticulum. J Biol Chem 264:2043–2052.

6. Hiratsuka T (1982) Biological activities and spectroscopic properties of chromophoricand fluorescent analogs of adenine nucleoside and nucleotides, 2′,3′-O-(2,4,6-trinitrocyclohexadienylidene) adenosine derivatives. Biochim Biophys Acta719:509–517.

7. McIntosh DB, Woolley DG, Vilsen B, Andersen JP (1996) Mutagenesis of segment487Phe-Ser-Arg-Asp-Arg-Lys492 of sarcoplasmic reticulum Ca2þ-ATPase produces pumpsdefective in ATP binding. J Biol Chem 271:25778–25789.

8. Grubmeyer C, Penefsky HS (1981) The presence of two hydrolytic sites on beefheart mitochondrial adenosine triphosphatase. J Biol Chem 256:3718–3727.

y

x

A B C

K515

L562

T441

E442

M494

F487 E439

K515

M494

F487

E439

E442

L562

R560

R174K205

R174M494

F487

K515

L562

E442

E439

T441

I188

z

Fig. 4. Surface representation of the binding pocket for TNP-AMP. Views of the binding pocket in E2(TG) (A), E2 · AlF4−ðTGÞ (B), and E2 · BeF3

−ðTGÞ (C).Prepared with PyMol (40). A stereo version of this figure is presented as Fig. S6.

Toyoshima et al. PNAS ∣ February 1, 2011 ∣ vol. 108 ∣ no. 5 ∣ 1837

BIOCH

EMISTR

Y

Dow

nloa

ded

by g

uest

on

Janu

ary

23, 2

021

9. Faller LD (1990) Binding of the fluorescent substrate analogue 2′,3′-O-(2,4,6-trinitro-phenylcyclohexadienylidene)adenosine 5′-triphosphate to the gastric Hþ, Kþ-ATPase:evidence for cofactor-induced conformational changes in the enzyme. Biochemistry29:3179–3186.

10. Moczydlowski EG, Fortes PA (1981) Inhibition of sodium and potassium adenosinetriphosphatase by 2′,3′-O-(2,4,6-trinitrocyclohexadienylidene) adenine nucleotides.J Biol Chem 256:2357–2366.

11. Virginio C, Robertson G, Surprenant A, North RA (1998) Trinitrophenyl-substitutednucleotides are potent antagonists selective for P2X1, P2X3, and heteromeric P2X2/3 receptors. Mol Pharmacol 53:969–973.

12. Suzuki H, Kubota T, Kubo K, Kanazawa T (1990) Existence of a low-affinity ATP-bindingsite in the unphosphorylated Ca2þ-ATPase of sarcoplasmic reticulum vesicles: Evidencefrom binding of 2′,3′-O-(2,4,6-trinitrocyclohexadienylidene)-[3H]AMP and -[3H]ATP.Biochemistry 29:7040–7045.

13. Toyoshima C (2008) Structural aspects of ion pumping by Ca2þ-ATPase of sarcoplasmicreticulum. Arch Biochem Biophys 476:3–11.

14. Bilwes AM, Quezada CM, Croal LR, Crane BR, Simon MI (2001) Nucleotide bindingby the histidine kinase CheA. Nat Struct Biol 8:353–360.

15. Oswald C, Jenewein S, Smits SH, Holland IB, Schmitt L (2008) Water-mediated protein-fluorophore interactions modulate the affinity of an ABC-ATPase/TNP-ADP complex.J Struct Biol 162:85–93.

16. Mou TC, et al. (2006) Broad specificity of mammalian adenylyl cyclase for interactionwith 2′,3′-substituted purine- and pyrimidine nucleotide inhibitors. Mol Pharmacol70:878–886.

17. Sagara Y, Inesi G (1991) Inhibition of the sarcoplasmic reticulum Ca2þ transport ATPaseby thapsigargin at subnanomolar concentrations. J Biol Chem 266:13503–13506.

18. Toyoshima C, Nakasako M, Nomura H, Ogawa H (2000) Crystal structure of the calciumpump of sarcoplasmic reticulum at 2.6 Å resolution. Nature 405:647–655.

19. Sørensen TL, Møller JV, Nissen P (2004) Phosphoryl transfer and calcium ion occlusionin the calcium pump. Science 304:1672–1675.

20. Toyoshima C, Mizutani T (2004) Crystal structure of the calcium pump with a boundATP analogue. Nature 430:529–535.

21. Olesen C, et al. (2007) The structural basis of calcium transport by the calcium pump.Nature 450:1036–1042.

22. Inesi G, Lewis D, Ma H, Prasad A, Toyoshima C (2006) Concerted conformational effectsof Ca2þ and ATP are required for activation of sequential reactions in the Ca2þ ATPase(SERCA) catalytic cycle. Biochemistry 45:13769–13778.

23. Jensen AM, Sorensen TL, Olesen C, Møller JV, Nissen P (2006) Modulatory and catalyticmodes of ATP binding by the calcium pump. EMBO J 25:2305–2314.

24. Berman MC (1986) Absorbance and fluorescence properties of 2′(3′)-O-(2,4,6-trinitro-phenyl)adenosine 5′-triphosphate bound to coupled and uncoupled Ca2þ-ATPase ofskeletal muscle sarcoplasmic reticulum. J Biol Chem 261:16494–16501.

25. Danko S, Yamasaki K, Daiho T, Suzuki H (2004) Distinct natures of beryllium fluoride-bound, aluminum fluoride-bound, and magnesium fluoride-bound stable analoguesof an ADP-insensitive phosphoenzyme intermediate of sarcoplasmic reticulum Ca2þ-ATPase: changes in catalytic and transport sites during phosphoenzyme hydrolysis.J Biol Chem 279:14991–14998.

26. Obara K, et al. (2005) Structural role of countertransport revealed in Ca2þ pump crystalstructure in the absence of Ca2þ. Proc Natl Acad Sci USA 102:14489–14496.

27. Lacapere JJ, Guillain F (1993) The reaction mechanism of Ca2þ-ATPase of sarcoplasmicreticulum. Eur J Biochem 211:117–126.

28. Picard M, et al. (2007) Ca2þ versus Mg2þ coordination at the nucleotide-binding siteof the sarcoplasmic reticulum Ca2þ-ATPase. J Mol Biol 368:1–7.

29. Toyoshima C, Nomura H, Tsuda T (2004) Lumenal gating mechanism revealed incalcium pump crystal structures with phosphate analogues. Nature 432:361–368.

30. Toyoshima C, Norimatsu Y, Iwasawa S, Tsuda T, Ogawa H (2007) How processing ofaspartylphosphate is coupled to lumenal gating of the ion pathway in the calciumpump. Proc Natl Acad Sci USA 104:19831–19836.

31. Mou TC, Masada N, Cooper DN, Sprang SR (2009) Structural basis for inhibition ofmammalian adenylyl cyclase by calcium. Biochemistry 48:3387–3397.

32. Danko S, Daiho T, Yamasaki K, Liu X, Suzuki H (2009) Formation of the stable structuralanalog of ADP-sensitive phosphoenzyme of Ca2þ-ATPase with occluded Ca2þ byberyllium fluoride: Structural changes during phosphorylation and isomerization.J Biol Chem 284:22722–22735.

33. McIntosh DB, Woolley DG, Berman MC (1992) 2′,3′-O-(2,4,6-trinitrophenyl)-8-azido-AMP and -ATP photolabel Lys-492 at the active site of sarcoplasmic reticulumCa2þ-ATPase. J Biol Chem 267:5301–5309.

34. Inesi G, et al. (1992) Long-range intramolecular linked functions in activation andinhibition of SERCA ATPases. Ann NY Acad Sci 671:32–47.

35. Ward DG, Taylor M, Lilley KS, Cavieres JD (2006) TNP-8N3-ADP photoaffinity labelingof two Na, K-ATPase sequences under separate Naþ plus Kþ control. Biochemistry45:3460–3471.

36. Nakamoto RK, Inesi G (1984) Studies of the interactions of 2′,3′-O-(2,4,6-trinitrocyclo-hexyldienylidine)adenosine nucleotides with the sarcoplasmic reticulum (Ca2þþMg2þ)-ATPase active site. J Biol Chem 259:2961–2970.

37. Otwinowski Z, Minor W (1997) Processing of X-ray diffraction data collected inoscillation mode. Methods Enzymol 276:307–325.

38. Brünger AT, et al. (1998) Crystallography & NMR system: A new software suite formacromolecular structure determination. Acta Crystallogr D 54:905–921.

39. Murshudov GN, Vagin AA, Lebedev A, Wilson KS, Dodson EJ (1999) Efficient anisotro-pic refinement of macromolecular structures using FFT. Acta Crystallogr D 55:247–255.

40. DeLano W (2002) The PyMOL Molecular Graphics System (DeLano Scientific, PaloAlto, CA).

1838 ∣ www.pnas.org/cgi/doi/10.1073/pnas.1017659108 Toyoshima et al.

Dow

nloa

ded

by g

uest

on

Janu

ary

23, 2

021