Structure of Ustilago maydis Killer Toxin KP6 α-Subunit A ...

8
Structure of Ustilago maydis Killer Toxin KP6 a-Subunit A MULTIMERIC ASSEMBLY WITH A CENTRAL PORE* (Received for publication, March 5, 1999, and in revised form, April 13, 1999) Naiyin Li‡, Mary Erman‡, Walter Pangborn‡, William L. Duax‡, Chung-Mo Park§, Jeremy Bruenn§, and Debashis Ghosh‡i From the Hauptman-Woodward Medical Research Institute, Buffalo, New York 14203, §Department of Biological Sciences, State University of New York at Buffalo, New York 14260, and the Roswell Park Cancer Institute, Buffalo, New York 14263 Ustilago maydis is a fungal pathogen of maize, some strains of which secrete killer toxins. The toxins are encoded by double-stranded RNA viruses in the cell cy- toplasm. The U. maydis killer toxin KP6 contains two polypeptide chains, a and b, having 79 and 81 amino acids, respectively, both of which are necessary for its killer activity. The crystal structure of the a-subunit of KP6 (KP6a) has been determined at 1.80-Å resolution. KP6a forms a single domain structure that has an over- all shape of an ellipsoid with dimensions 40 Å 3 26 Å 3 21 Å and belongs to the a/b-sandwich family. The tertiary structure consists of a four-stranded antiparallel b-sheet, a pair of antiparallel a-helices, a short strand along one edge of the sheet, and a short N-terminal helix. Although the fold is reminiscent of toxins of similar size, the topology of KP6a is distinctly different in that the a/b-sandwich motif has two right-handed bab split crossovers. Monomers of KP6a assemble through crys- tallographic symmetries, forming a hexamer with a cen- tral pore lined by hydrophobic N-terminal helices. The central pore could play an important role in the mech- anism of the killing action of the toxin. Toxins are protein molecules that disrupt cell functions in a number of ways, some by making ion channels in cell mem- branes and others by interacting with membrane channels and/or receptors. Three-dimensional structures of several of these toxins have been determined, including cardiotoxin (1, 2), d-endotoxin (3), hemolysin (4), anthrax toxin (5), colicins (6), and diphtheria toxin (7). The tertiary structures of toxins vary widely from being almost entirely b-sheet, as in cardiotoxin, to nearly all a-helical, such as colicins. Ustilago maydis is a fun- gal pathogen of maize and is one of a number of fungi that secrete cellular killer toxins (8). These killer toxins are encoded by double-stranded RNA viruses in the cell cytoplasm (9). In the absence of an immunity or resistance gene, these toxins are lethal to the organism of origin and sometimes to closely re- lated organisms but not to plant or animal cells. The U. maydis strains P1, P4, and P6 secrete toxins KP1, KP4, and KP6, respectively, all of which are low molecular weight (;100 amino acids) polypeptides (10 –12). The crystal structure of the U. maydis KP4 toxin (a single polypeptide chain of 105 amino acids) has been determined and shown to possess a a/b-sand- wich fold (13) having a central five-stranded anti-parallel b-sheet and two a-helices (14). It appears to kill sensitive cells by blocking Ca 21 channels (14). The SMK 1 secreted by the halo tolerant yeast Pichia farinosa also has a similar a/b-sandwich structure (15). Williopsis mrakii killer toxin, having a four- stranded antiparallel b-sheet structure similar to that of bg crystalline (16, 17), inhibits b-glucagon synthesis. The U. maydis KP6 killer toxin gene has been cloned, se- quenced, and expressed, and the protein was characterized (10). The toxin is unique in that two polypeptide chains, KP6a and KP6b, are necessary for its killer activity (18). The 1234- base pair double-stranded RNA P6M2 codes for a pretoxin of 219 amino acids, which is post-translationally cleaved by two endopeptidases, yielding a- and b-subunits having 79 and 81 amino acids, respectively, as determined by cDNA sequence, N-terminal protein sequence analysis, and by mass spectrosco- py. 2 It has been proposed that the KP6 toxin may act by either interfering with the K 1 channel or binding to its own mem- brane receptor (10, 19), thus depleting cellular K 1 levels and eventually killing the cells. Here we present the crystal struc- ture of the a-subunit of KP6 toxin determined by the isomor- phous heavy atom replacement method and refined at a 1.80-Å resolution. We show that the symmetry-related trimeric or hexameric assembly of the subunit creates a central pore that could have functional implications. EXPERIMENTAL PROCEDURES KP6a was purified to homogeneity as described previously (18, 20). The selenomethionine derivative was prepared by growing cells in a medium supplemented with 20 mg/ml selenomethionine. Native protein crystals were obtained by the hanging drop vapor diffusion technique (21). Hanging drops containing 5 ml of 10 mg/ml protein and 1 ml of 85% ammonium sulfate in 10 mM MES buffer, pH 6.0, were equilibrated against 1 ml of reservoir solution containing 18 –21% ammonium sul- fate in 10 mM MES buffer. Single crystals shaped as hexagonal rods having typical dimensions of 0.15 3 0.30 3 0.30 mm 3 were obtained in about 20 days at room temperature. The crystal of the selenomethi- onine-substituted protein was obtained under the same conditions as the native crystal, but it took longer to grow to a size of 0.1 3 0.2 3 0.2 mm 3 . The heavy atom derivative was prepared by soaking a single native crystal in 200 ml of 0.5 mM PIP (di-m-iodobis(ethylenediamine) diplatinum(II) nitrate) solution in MES buffer, pH 6.0, for 36 days. Three room temperature data sets for native, platinum derivative, and selenomethionine-substituted protein crystals were collected with the in-house RAXIS IIc image plate area detector, receiving x-rays from a Rigaku RU-200 rotating anode generator operated at 50 kV and 90 mA. Graphite-monochromated CuK a radiation was used. Crystals used for * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. The atomic coordinates and structure factors (code 1Kp6) have been deposited in the Protein Data Bank, Brookhaven National Laboratory, Upton, NY. i To whom correspondence should be addressed: Hauptman-Woodward Medical Research Inst., Inc., 73 High St., Buffalo, NY 14203. Tel.: 716- 865-9600 (ext. 316); Fax: 716-852-6086; E-mail: [email protected]. 1 The abbreviations used are: SMK, salt-mediated killer toxin; MES, 2-(N-morpholino)ethanesulfonic acid; PIP, di-m-iodobis(ethylenedia- mine)diplatinum(II) nitrate. 2 J. Bruenn, unpublished results. THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 274, No. 29, Issue of July 16, pp. 20425–20431, 1999 © 1999 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. This paper is available on line at http://www.jbc.org 20425 by guest on April 2, 2018 http://www.jbc.org/ Downloaded from

Transcript of Structure of Ustilago maydis Killer Toxin KP6 α-Subunit A ...

Structure of Ustilago maydis Killer Toxin KP6 a-SubunitA MULTIMERIC ASSEMBLY WITH A CENTRAL PORE*

(Received for publication, March 5, 1999, and in revised form, April 13, 1999)

Naiyin Li‡, Mary Erman‡, Walter Pangborn‡, William L. Duax‡, Chung-Mo Park§,Jeremy Bruenn§, and Debashis Ghosh‡¶i

From the ‡Hauptman-Woodward Medical Research Institute, Buffalo, New York 14203, §Department of BiologicalSciences, State University of New York at Buffalo, New York 14260, and the ¶Roswell Park Cancer Institute,Buffalo, New York 14263

Ustilago maydis is a fungal pathogen of maize, somestrains of which secrete killer toxins. The toxins areencoded by double-stranded RNA viruses in the cell cy-toplasm. The U. maydis killer toxin KP6 contains twopolypeptide chains, a and b, having 79 and 81 aminoacids, respectively, both of which are necessary for itskiller activity. The crystal structure of the a-subunit ofKP6 (KP6a) has been determined at 1.80-Å resolution.KP6a forms a single domain structure that has an over-all shape of an ellipsoid with dimensions 40 Å 3 26 Å 3 21Å and belongs to the a/b-sandwich family. The tertiarystructure consists of a four-stranded antiparallelb-sheet, a pair of antiparallel a-helices, a short strandalong one edge of the sheet, and a short N-terminal helix.Although the fold is reminiscent of toxins of similar size,the topology of KP6a is distinctly different in that thea/b-sandwich motif has two right-handed bab splitcrossovers. Monomers of KP6a assemble through crys-tallographic symmetries, forming a hexamer with a cen-tral pore lined by hydrophobic N-terminal helices. Thecentral pore could play an important role in the mech-anism of the killing action of the toxin.

Toxins are protein molecules that disrupt cell functions in anumber of ways, some by making ion channels in cell mem-branes and others by interacting with membrane channelsand/or receptors. Three-dimensional structures of several ofthese toxins have been determined, including cardiotoxin (1, 2),d-endotoxin (3), hemolysin (4), anthrax toxin (5), colicins (6),and diphtheria toxin (7). The tertiary structures of toxins varywidely from being almost entirely b-sheet, as in cardiotoxin, tonearly all a-helical, such as colicins. Ustilago maydis is a fun-gal pathogen of maize and is one of a number of fungi thatsecrete cellular killer toxins (8). These killer toxins are encodedby double-stranded RNA viruses in the cell cytoplasm (9). Inthe absence of an immunity or resistance gene, these toxins arelethal to the organism of origin and sometimes to closely re-lated organisms but not to plant or animal cells. The U. maydisstrains P1, P4, and P6 secrete toxins KP1, KP4, and KP6,respectively, all of which are low molecular weight (;100amino acids) polypeptides (10–12). The crystal structure of the

U. maydis KP4 toxin (a single polypeptide chain of 105 aminoacids) has been determined and shown to possess a a/b-sand-wich fold (13) having a central five-stranded anti-parallelb-sheet and two a-helices (14). It appears to kill sensitive cellsby blocking Ca21 channels (14). The SMK1 secreted by the halotolerant yeast Pichia farinosa also has a similar a/b-sandwichstructure (15). Williopsis mrakii killer toxin, having a four-stranded antiparallel b-sheet structure similar to that of bgcrystalline (16, 17), inhibits b-glucagon synthesis.

The U. maydis KP6 killer toxin gene has been cloned, se-quenced, and expressed, and the protein was characterized(10). The toxin is unique in that two polypeptide chains, KP6aand KP6b, are necessary for its killer activity (18). The 1234-base pair double-stranded RNA P6M2 codes for a pretoxin of219 amino acids, which is post-translationally cleaved by twoendopeptidases, yielding a- and b-subunits having 79 and 81amino acids, respectively, as determined by cDNA sequence,N-terminal protein sequence analysis, and by mass spectrosco-py.2 It has been proposed that the KP6 toxin may act by eitherinterfering with the K1 channel or binding to its own mem-brane receptor (10, 19), thus depleting cellular K1 levels andeventually killing the cells. Here we present the crystal struc-ture of the a-subunit of KP6 toxin determined by the isomor-phous heavy atom replacement method and refined at a 1.80-Åresolution. We show that the symmetry-related trimeric orhexameric assembly of the subunit creates a central pore thatcould have functional implications.

EXPERIMENTAL PROCEDURES

KP6a was purified to homogeneity as described previously (18, 20).The selenomethionine derivative was prepared by growing cells in amedium supplemented with 20 mg/ml selenomethionine. Native proteincrystals were obtained by the hanging drop vapor diffusion technique(21). Hanging drops containing 5 ml of 10 mg/ml protein and 1 ml of 85%ammonium sulfate in 10 mM MES buffer, pH 6.0, were equilibratedagainst 1 ml of reservoir solution containing 18–21% ammonium sul-fate in 10 mM MES buffer. Single crystals shaped as hexagonal rodshaving typical dimensions of 0.15 3 0.30 3 0.30 mm3 were obtained inabout 20 days at room temperature. The crystal of the selenomethi-onine-substituted protein was obtained under the same conditions asthe native crystal, but it took longer to grow to a size of 0.1 3 0.2 3 0.2mm3. The heavy atom derivative was prepared by soaking a singlenative crystal in 200 ml of 0.5 mM PIP (di-m-iodobis(ethylenediamine)diplatinum(II) nitrate) solution in MES buffer, pH 6.0, for 36 days.Three room temperature data sets for native, platinum derivative, andselenomethionine-substituted protein crystals were collected with thein-house RAXIS IIc image plate area detector, receiving x-rays from aRigaku RU-200 rotating anode generator operated at 50 kV and 90 mA.Graphite-monochromated CuKa radiation was used. Crystals used for

* The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby marked“advertisement” in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.

The atomic coordinates and structure factors (code 1Kp6) have beendeposited in the Protein Data Bank, Brookhaven National Laboratory,Upton, NY.

i To whom correspondence should be addressed: Hauptman-WoodwardMedical Research Inst., Inc., 73 High St., Buffalo, NY 14203. Tel.: 716-865-9600 (ext. 316); Fax: 716-852-6086; E-mail: [email protected].

1 The abbreviations used are: SMK, salt-mediated killer toxin; MES,2-(N-morpholino)ethanesulfonic acid; PIP, di-m-iodobis(ethylenedia-mine)diplatinum(II) nitrate.

2 J. Bruenn, unpublished results.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 274, No. 29, Issue of July 16, pp. 20425–20431, 1999© 1999 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org 20425

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

data collection were mounted in glass capillaries and sealed withmother liquor. One crystal was used for each of the data sets. The imageplate detector was placed at 77.9, 120.0, and 130.3 mm from native, PIPderivative, and selenomethionine protein crystals, respectively, and thelimiting resolutions for the three data sets were 1.8 Å, 2.2 Å, and 2.3 Å,respectively. The space group is P6322, and cell dimensions are a 5 b 548.30 Å, c 5 124.22 Å, with one molecule/asymmetric unit. The specificvolume of the crystal is 2.41 Å3/Da, and the solvent content is about42%. Data were processed using the DENZO program and scaled andmerged using the SCALEPACK program (22, 23). Details of the datastatistics are listed in Table I.

Three platinum and one selenium positions were determined byHASSP (24) and confirmed with the Shake-and-Bake program package(25). Cross-phasing was carried out between the two derivatives tocorrelate platinum and selenium heavy atom positions through a com-mon origin. Heavy atom positions were refined using HEAVYv4 (24) bythe origin-removed Patterson technique. The phases were then calcu-

lated using PHASES (26) combining both isomorphous and anomaloussignals from PIP and the isomorphous signal from the selenomethi-onine derivative. Solvent flattening was carried out before the MIRAS(multiple isomorphous replacement signals from PIP and selenomethi-onine derivatives plus the anomalous signal from the PIP derivative)maps were calculated. Details of phase determination are summarizedin Table II.

MIRAS maps for both hands of the heavy atom positions were cal-culated, and the correct hand was chosen based on the quality of themaps and evidence of right-handed a-helices. Chain-tracing and proteinmodel building were performed manually on a SGI Elan workstationusing CHAIN (27). The location of the selenium position played a keyrole in the tracing of the chain by identifying the only methionine in theb-strand near the C terminus. Identification of the N-terminal a-helixalso helped to complete the model building process. The entire polypep-tide chain of 79 residues was built into this MIRAS electron densitymap prior to refinement. A 2Fo 2 Fc density map of the single methi-

TABLE IIIsomorphous replacement phasing statistics

Derivative Di-m-iodobis(ethylenediamine)diplatinum(II) nitrate Selenomethionine

Phase calculation method SIRa SASb SIRa

Rc deri(F) 0.158 0.065Number of heavy atom sites 3 3 1Rd cullis (%, cent/acent) 0.64/0.78 0.59/0.67No. of reflections phased 1742 1184 1745Phasing powere (3.0 Å, overall/last shell)

Before phase combining 1.59/1.62 1.86/1.60 1.55/1.16After phase combining 1.70/1.61 2.44/1.97 1.52/1.01

Mean figure of meritf (cent/acent)Before phase combining 0.27/0.16 0.32 0.53/0.33After phase combining 0.48/0.26 0.40 0.58/0.37Overall 0.673g

After solvent flatteningNo. of reflections 1792

R factor 0.310Mean FOMh 0.842Correlation coefficient 0.936

a SIR, single isomorphous replacement.b SAS, single anomalous scattering.c Rderi(F) 5 ,u(uFpu 2 uFPHu)u./,0.5x(uFpu 1 uFPHu)u., where uFpu is the protein structure factor amplitude and uFPHu is the heavy atom derivative

structure factor amplitude.d Rcullis 5 uEu/iFpu 2 uFPHi.e Phasing power 5 root mean square (uFHu/E), where uFHu is the heavy-atom structure factor amplitude, and E is the residual lack of closure error.f Figure of merit 5 ,SP(a)eia/SP(a)., in which a is the phase, and P(a) is the phase probability distribution.g For 1793 reflections.h FOM, figure of merit.

TABLE ISummary of data collection for native and derivative crystals

Crystal type

Native Di-m-iodobis(ethylenediamine)adi-platinum(II) nitrate derivative

Seleno-methioninederivative

No. of crystals used 1 1 1Experiment temperature (°C) 20 20 20Wavelength (Å) 1.5418 1.5418 1.5418Detector distance (mm) 77.86 120.02 130.25Space group P6322 P6322 P6322Cell dimensions

a (Å) 48.30 48.43 48.25b (Å) 48.30 48.43 48.25c (Å) 124.22 123.77 124.33

Mosaicity (°) 0.197 0.220 0.220Number of frames 80 24 15Oscillation range/frame (°) 0.75 1.5 3.0Total data collected 109,467 42,112 23,797Unique data 8260 4431 3868Max. resolution (Å) 1.80 2.20 2.30Scale factors 0.88–1.03 1.00–1.21 1.00–1.28Completeness (%) 93.9 91.1 90.3Last 0.05-Å shell completeness 90.0 58.5 56.8F2/s (F2) at last 0.05 Å shell 2.15 5.03 3.47Rb

merge(liner) 0.077 0.062 0.079Rmerge(square) 0.057 0.050 0.062x2c 0.623 0.609 0.444

a The soaking condition for this crystal was 0.5 mM for 36 days.b Rmerge(liner)5S(u(I 2 ^I&)u)/S(I), Rmerge(square)5S((I 2 ^I&)2/S(I2)).c x2 5 S(I 2 ^I&)2/(^s(I)&2 3 N/(N 2 1)).

Structure of Killer Toxin KP6a20426

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

onine-containing b-strand is shown in Fig. 1 along with the independ-ent determination of the selenium position from the cross-differenceFourier with the phases from the PIP derivative alone.

The model was refined with XPLOR (28) using the simulated anneal-ing procedure. The resolution was gradually extended to 1.8 Å. Posi-tional and isotropic parameters of a total of 690 non-hydrogen atomsincluding 77 solvent water oxygens and 1 sulfate ion were refined in thefinal model, using the 6838 reflections (F $ 2sF) in the range 1.8–8.0 Å.Some of these densities modeled as water oxygens could possibly beother ions present in the crystallization medium. The agreement be-tween the final model and the 2Fo 2 Fc density map was excellent forthe main chain and side chains, except for one break at the Cb atom ofresidue Asp37, which is located on the surface of the molecule in ab-turn. The C-terminal Lys79 side chain is also not well defined, prob-ably because of dynamical disorder.

RESULTS AND DISCUSSION

Quality of the Model—The numbers from the final cycle ofrefinement are provided in Table III. The random positionalerror estimated from the Luzzati plot (29) is less than 0.20 Å.The average B values remain low throughout most of the res-idues of the structure. The agreement between side chain den-sities and the amino acid sequence is unambiguous, except in afew cases. Residues Asp37 and His38 are located at a turn whichis exposed to the outer surface; consequently, their average Bvalues are relatively high. The C-terminal Lys79 also has a highB average value probably because of dynamical disorder. Al-though the refinement results shown are based on the unitweight scheme, nonunit weight schemes were also attempted.The atomic coordinates and the electron density maps resultingfrom these two modes of refinement were virtually identical.

Secondary and Tertiary Structures of a KP6a Polypeptide—Amonomer of the KP6 killer toxin a-subunit shown in Fig. 2aforms a single domain structure having an overall shape of anellipsoid of dimensions 40Å 3 26Å 3 21Å. The structure has asingle split bab motif that belongs to the a/b sandwich family(30). It consists of a four- (b1, b3, b4, and b5) stranded antipa-rallel b-sheet, a pair of antiparallel a-helices (a2 and a3) thatlies approximately 20° to the b-strands above one side of the

sheet, a strand b2 along one edge of the sheet, and a shortN-terminal helix a1 on the other side of the b-sheet. The twistof the b-sheet is left-handed, as it usually occurs (13). Amongthe three linkages of the four-stranded antiparallel b-sheet,one belongs to a hairpin connection (b3 to b4), and the other twoare right-handed bab split crossover connections (b1 to b3 andb4 to b5) via a2 and a3, respectively (13, 30). All the secondarystructural elements are connected through six loops. However,only two of them are b turns; one is a b type II turn (13) (Leu39,Ser40, Lys41, and Ser42) connecting the helix a2 with b3, and theother one is a b type I turn (13) (Ser66, Ser67, Leu68, and Asn69)connecting a3 with b5. There are eight cysteines in the struc-ture as indicated in Fig. 2, a and b, all of which form intrachaindisulfide bridges linking these secondary structural elementsinto a compact domain. Helices a2 and a3 are linked to theb-sheet through disulfide bridges: a2 to b4 through Cys35 andCys51 in a right-handed conformation and a3 to b1 throughCys65 and Cys18 in a left-handed conformation. The remainingtwo disulfide bridges are both right-handed, one formed be-tween the N-terminal helix a1 and the loop connecting it to b1

(Cys5 and Cys12) and the other between the two longest anti-parallel strands b1 and b5 (Cys16 and Cys74). The Ca-Ca dis-tances between a pair of cysteines range from 4.05 to 5.86 Å,depending on the conformation of each individual disulfidebridge. Besides disulfide bonds, the hydrogen bonding networkholds the structure together. The hydrogen bonds involvingmain chain atoms among the five antiparallel b-strands areschematically illustrated in Fig. 2c. A total of 44 hydrogenbonds are formed among main chain atoms, ranging from 2.67to 3.14 Å, with an average bond length of 2.93 Å. There are 16hydrogen bonds between main chain and side chain atoms and7 hydrogen bonds among side chain atoms ranging from 2.60 to3.42 Å, with an average value of 2.99 Å. All water molecules arehydrogen-bonded to protein atoms either directly or indirectlythrough another solvent atom. No exposed cluster of hydropho-bic side chains was found in the monomer. Three regions in thismolecule interact through hydrophobic side chains that appearto stabilize the tertiary structure. As shown in Fig. 2b, amongthe three hydrophobic clusters, two located between the twoantiparallel a-helices and the b-sheet are formed by residuesAla20, Leu27, Ala30, and Leu68 at the top of the molecule (op-posite of the N terminus) and by Leu39, Phe53, Leu57, Phe61,and Met72 at the lower middle section of the molecule. Thethird region, located near the N terminus, is comprised ofresidues Ala3, Phe4, and Phe8 forming a hydrophobic pocketwith neighboring symmetry-related molecules. There are four

FIG. 1. The final 2Fo 2 Fc electron density map and the refinedmodel of the methionine-containing b-strand (b5) of the KP6amolecule. Contoured at 1.5s. The position of the independently deter-mined selenium atom is indicated by a large sphere.

TABLE IIIRefinement results of KP6a

Resolution range (Å) 8.0–1.8No. of reflections used (F . 2s(F)) 6838Weighing scheme unit weightPercentage of observed data used 96.

No. of protein-atoms 1 SO422 613

No. of solvent water 77Crystallographic

R (F $ 2sF) 0.164Rfree (F $ 2sF) 0.211

RMS deviation from ideal geometryBond (Å) 0.008Angle (°) 1.309

Average B values (Å2)Whole protein 17.6Backbone 15.1Side chains 24.6Solvent 39.2

Ramachandran plot summary:Residues in most favored regions 91.4%In additional allowed regions 8.6%In disallowed regions 0.0%

Structure of Killer Toxin KP6a 20427

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

histidines in the molecule; although all four are located on thesurface, three of them reside on the helical side of the molecule(His21, His38, and His64), and His50 resides on the b-sheet side,as shown in Fig. 2b. Because the pH of the crystallizationbuffer was 6.0, one of four of these histidines was deprotonated,which made it possible to coordinate to a heavy metal ion. Thesurface of the protein is hydrophilic except for the N-terminalhelix-a1.

The topology of KP6a along with the other two recentlydetermined killer toxins SMK (15) and KP4 (14) is illustratedschematically in Fig. 2d. Although all of them have strikingsimilarity in folding motifs, i.e. a/b sandwich motifs, there aredistinct differences among them; KP6a has two right-handedbab split crossovers, whereas both SMK and KP4 have twoleft-handed crossovers. A survey (30) has shown that there aremore known right-handed a/b sandwich proteins than the left-handed ones. Whether the handedness of connections has anysignificance in the biological function is not clear at this mo-ment, but it is evident that all resulting structures are uniquein their specific functions. Although the KP4 and KP6 killertoxins are both produced by U. maydis, the difference in their

folding connections may indicate that they are not functionallysimilar.

Assembly of KP6a Molecules—As described above, the hydro-phobic region with exposed phenyl rings of Phe4 and Phe8 of themonomer is located in the N-terminal a1-helix. Three moleculesof the KP6a monomer assemble through this region by a crys-tallographic 3-fold rotational symmetry forming a trimer. Twotrimers further associate by a crystallographic 2-fold rotationalsymmetry axis perpendicular to the 3-fold axis forming a hex-amer with the hydrophobic phenol basket in the middle of theassembly, as shown in Figs. 4 and 5. The trimeric association isstabilized by intermolecular salt bridges and hydrogen bonds;the bond lengths range from 2.77 to 3.00 Å with an average of2.88 Å. The salt bridges are formed between Glu15 and Asp45 ofone monomer and Arg47 of another. The Ser78 side chain andmain chain carbonyl at the C terminus of one monomer formhydrogen bonds with side chains of Arg28 and Tyr29 of theneighboring monomer. Intermolecular interactions within atrimer are shown in Fig. 3. The hexamer is stabilized by hy-drophobic interactions between the N termini of the two trim-ers, especially the Gly7-Phe8-Gly9 sequence.

FIG. 2. a, the overall structure is an a/b sandwich composed of a four-stranded antiparallel b-sheet (b1, b3, b4, and b5) with two antiparallela-helices (a2 and a3) on one side of the sheet, a b-strand (b2) along one edge of the sheet, and a short N-terminal a-helix (a1) on another side ofthe b-sheet. Disulfide bridges are drawn in yellow. The secondary structure elements have the following terminal residues a1 (2–8), b1 (14–20),b2 (25–27), a2 (28–33), b3 (42–45), b4 (50–53), a3 (58–67), and b5 (70–77). The figure was drawn with SETOR (33). b, stereo view of a KP6a proteinstructure shown as a Ca trace. Residues in purple are involved in hydrophobic contact within the molecule. Four histidine residues are also shownwith blue labels. Four disulfide bridges are shown in yellow; black labels indicate the locations of the cysteines. c, the main chain hydrogen bondnetwork in the b-sheet. Only the backbone atoms of the polypeptide chain are shown. Hydrogen bonds were drawn with dashed lines. d, acomparison of the topologies of KP6a, SMK, and KP4 toxins. The triangles and the circles represent b-strands and a-helices, respectively. Minorelements were shown in pale colors. The secondary structure assignment for the KP6a toxin was performed by the algorithm of Kabsch and Sander,as implemented in the program PROCHECK (34).

Structure of Killer Toxin KP6a20428

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

The network of intermolecular salt bridges and hydrogenbonds links the monomers to form a solvent-filled funnel-likestructure in which the b-sheet of each molecule spreads on theinner surface and the N-terminal a1-helix sits close to the tip ofthe funnel (Fig. 4, a and b). The funnel is 9–10 Å across at thetop, has a 6-Å diameter opening in the middle and narrows toa 4.2-Å diameter barrel bordered by three symmetry relatedphenyl rings arising from the Gly7-Phe8-Gly9 string of residuesat the N-terminal a1-helix. More than 50 well ordered watermolecules have been located in the funnel. Fig. 4, a and b showsthe KP6a trimer with phenyl rings at the tip of the funnel andwater oxygens inside of the funnel. The assembly of the hex-amer brings the two narrow funnel tips face to face, thusforming an hourglass shaped structure as shown in Fig. 5. Thedistance between two planes of intratrimer salt bridges isabout 34 Å.

The analysis of total accessible surfaces for the monomer, thetrimer, and the hexamer bears out the above description ofmolecular association. The details are listed in Table IV. Thetotal accessible surface area for KP6a monomer is 4783 Å2.Because of the formation of the trimer and the hexamer, totalaccessible surface areas for each monomer decreased by 21.2and 24.2%, respectively. The loss of accessible surface area/monomer because of oligomerization is 1014 Å2 for the trimerand 1157 Å2 for the hexamer. These numbers are in goodagreement with the values of ;800 Å2 for a homodimer and;1000 Å2 for a heterocomplex of a 10-kDa protein calculated byJones and Thorton (31) for 59 different protein-protein com-plexes. Among lost accessible surfaces, the loss of hydrophobicsurface contributed the most, followed by the charged surfaceand the polar surface. As an overall effect of oligomeric associ-ation of KP6a, the percentage of the hydrophilic surface area,consisting of charged and polar residues, increased from 73.0%for the monomer to 75.6 and 78.6% for the trimer and thehexamer, respectively, making them more soluble in an aque-ous medium. By forming hexamer from trimer, the gain or lossof both accessible hydrophilic and hydrophobic surfaces are in

small magnitudes, ;3%, indicating a weak tendency to formthe hexamer from the trimer.

Implications of the Quaternary Structure—It is conceivablethat the crystallographic monomer, the trimer, or the hexamerof KP6a, in association with one or more polypeptides of KP6bforms a molecular assembly that interacts with either the K1

channel or another receptor on the cell membrane or even byitself penetrates the membrane and causes disruption of thecellular ion balance. No conclusive biochemical evidence isavailable yet on any of these possibilities. The KP6a hexamerhas an inner pore opening of 4.2 Å, large enough for monova-lent cations such as K1 to pass through and a length of 34 Å,similar to the thickness of the membrane bilayer. Althoughneither KP4 nor SMK structures exhibit similar oligomericassociation, the distantly related all b-strand g-cardiotoxin (32)is trimeric. However, the central opening of the trimer narrowsto less than 0.5 Å where interfacial atoms are virtually at vander Waals contact distances (32). It was proposed that g-car-diotoxin in its oligomeric state could either interact with amembrane receptor or could insert itself into the membraneforming an ion channel (32). To our knowledge, no other mem-ber of this class of toxins is known to have a central pore likethe one in the KP6a hexamer.

Cardiotoxins are known to act by depolarizing the Ca1 chan-nel. KP4 was shown to kill cells by blocking divalent cationchannels but not Na1 or K1 channels (14). KP4 crystallizes asa monomer and a single polypeptide chain is known to possessthe killer action (14). Structural data suggest that its partici-pation in channel formation is unlikely. The mechanism ofpH-dependent killer action of SMK is still unknown (15). Themonomer containing the a- and the b-peptides forms a dimer inthe crystal (15). Despite the overall similarity of folds of mono-mers of these toxins, there appears to be diversity in the mech-anism of killing action, possibly stemming from diverse oligo-meric states of their functional entities. KP6 is distinctive inthe sense that it kills by depleting the K1, unlike any othertoxin in the family. The architecture of the KP6a hexamer

FIG. 3. The salt bridge network in-volving Asp45, Glu15, and Arg47 in acrystallographic trimer of KP6a. Thebridging distances are shown in Å.

Structure of Killer Toxin KP6a 20429

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

provides a pore structure that has right dimensions to functionas a K1 channel, either outside or within the cell membranebilayer. However, the outer surface of the hexamer as well asthe monomer has polar and charged residues, accounting forthe solubility of the toxin in aqueous medium. In order for theKP6 assembly to enter the membrane bilayer, the outer polarsurface must be concealed. It is conceivable that the b-subunitplays a critical structural role in this aspect because it is knownthat both subunits are necessary for the killer action of thetoxin. Any large rearrangement in the tertiary structure ofKP6a on complexation with the b-subunit is unlikely, because

FIG. 4. Views of a funnel-shaped KP6a trimer with phenylrings at the tip of the funnel and water oxygens inside of thefunnel. a, along the crystallographic 3-fold axis. b, along the directionperpendicular to the 3-fold axis.

FIG. 5. The KP6a hexamer. a, viewed along the crystallographic 3-foldrotation axis. b, viewed along the direction perpendicular to the 3-fold axis.

TABLE IVAccessible surface area (Å2)

Total Charged Polar Hydrophobic

Monomer 4783 1455 (30.4%) 2038 (42.6%) 1289 (27.0%)Trimer 11,306 3060 (27.1%) 5484 (48.5%) 2763 (24.4%)8 vs. monomer (2 21.2%) (2 3.3%) (1 5.9%) (2 2.6%)Hexmer 21,748 6137 (28.2%) 10,957 (50.4%) 4655 (21.4%)8 vs. monomer (2 24.2%) (2 2.2%) (1 7.8%) (2 5.6%)8 vs. trimer (2 3.8%) (1 1.1%) (1 1.9%) (2 3.0%)

Structure of Killer Toxin KP6a20430

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

of its compact nature that is rigidly held by four disulfidebridges. Formation of higher order oligomeric states either bythe KP6a hexamer or by an a-b complex to hide hydrophilicsurfaces from the lipid environment is also a plausible scenario.Further insight into the mechanism of action of KP6 may comefrom the crystal structure of an active complex of a- and b-sub-units of the toxin.

REFERENCES

1. Bhaskaran, R., Huang, C. C., Chang, D. K., and Yu, C. (1994) J. Mol. Biol. 235,1291–1301

2. Rees, B., Samma, J. P., Thierry, J. C., Gilbert, M., Fischer, H., Schweitz, M.,Lazdunski, M., and Moras, D. (1987) Proc. Natl. Acad. Sci. U. S. A. 84,3132–3136

3. Li, J., Carrol, J., and Ellar, D. J (1991) Nature 353, 815–8214. Raghunathan, G., Seetharamulu, P., Brooks, B. R., and Guy, H. R (1990)

Proteins Struct. Funct. Genet. 8, 213–2255. Petosa, C., Collier, R. J., Klimpel, K. R., Leppla, S. H., and Liddington, R. C.

(1997) Nature 385, 833–8386. Parker, M. W., Pattus, F., Tucker, A. D., and Tsernoglou, D. (1989) Nature 337,

93–967. Choe, S., Bennett, M. J., Fujii, G., Curmi, P. M., Kantardjieff, K. A., Collier,

R. J., and Eisenberg, D. (1992) Nature 357, 216–2228. Hankin, L., and Puhalla, J. E (1971) Phytopathology 61, 50–539. Puhalla, J. E. (1968) Genetics 60, 461–474

10. Tao, J., Ginsberg, I., Banerjee, N., Koltin, Y., Held, W., and Bruenn, J. A.(1990) Mol. Cell. Biol. 10, 1373–1381

11. Park, C.-M., Bruenn, J. A., Ganesa, C., Flurkey, W. F., Bozarth, R. G., andKoltin, Y. (1994) Mol. Microbiol. 11, 155–164

12. Park, C.-M., Banerjee, N., Koltin, Y., and Bruenn, J. A. (1996) Mol. Microbiol.20, 957–963

13. Richardson, J. S. (1981) Adv. Protein Chem. 34, 167–339

14. Gu, F., Khimani, A., Rane, S. G., Flurkey, W. H., Bozarth, R. F., and Smith,T. J. (1995) Structure 3, 805–814

15. Kashiwagi, T., Kunishima, N., Suzuki, C., Tsuchiya, F., Nikkuni, S., Arata, Y.,Morikawa, K. (1997) Structure 5, 81–94

16. Najmudin, S., Nalini, V., Driessen, H. P. C., Slingsby, C., Blundell, T. L., Moss,D. S., and Lindley, P. F. (1993) Acta Crystallogr. Sec. D 49, 223–233

17. Antuch, W., Guntert, P., and Wuthrich, K. (1996) Nat. Struct. Biol. 3, 662–66518. Peery, T., Shabat-Brand, T., Steinlauf, R., Koltin, Y., and Bruenn, J. (1987)

Mol. Cell. Biol. 7, 470–47719. Tao, J., Ginsberg, I., Koltin, Y., and Bruenn, J. A. (1993) Mol. Gen. Genet. 238,

234–24020. Steinlauf, R., Peery, T., Koltin, Y., and Bruenn, J. (1988) Exp. Mycol. 12,

264–27421. Hampel, A., Labananskas, M., Conners, P. G., Kirkegard, L., RajBhandary,

U. L., Sigler, P. B., and Bock, T. M. (1968) Science 162, 1384–138722. Otwinowski, Z., and Minor, W. (1994) Methods Enzymol. 276, 307–32623. Minor, W. (1993) XDISPLAYF Program, Purdue University, West Lafayette,

IN24. Terwillinger, T. C. (1995) Heavyv4 Program, University of California at Los

Alamos National Laboratory25. Miller, R., Gallo, S. M., Khalak, H. G., and Weeks, C. M. (1994) J. Appl.

Crystallogr. 27, 613–62126. Furey, W., and Swaminathan, S. (1990) Abstracts of the American Crystallo-

graphic Association Meeting, p. 73, PA33, New Orleans, LA27. Jones, T. A (1978). J. Appl. Crystallogr. 11, 268–27228. Brunger, A. T. (1992) X-PLOR, Version 3.1, User’s Guide, Yale University, New

Haven, CT29. Luzzati, V. (1952) Acta Crystallogr. Sec. D 26, 283–29130. Orengo, C. A., and Thornton, J. M. (1993) Structure 1, 105–12031. Jones, S., and Thornton, J. M. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 13–2032. Bilwes, A., Rees, B., Moras, D., Menez, R., and Menez, A. (1994) J. Mol. Biol.

239, 122–13633. Evans, S. V. (1993) J. Mol. Graphics 11, 134–13834. Laskowski, R. A., MacArthur, M. W., Moss, D. S., and Thorton, J. M. (1993)

J. Appl. Crystallogr. 26, 283–291

Structure of Killer Toxin KP6a 20431

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Bruenn and Debashis GhoshNaiyin Li, Mary Erman, Walter Pangborn, William L. Duax, Chung-Mo Park, Jeremy

ASSEMBLY WITH A CENTRAL PORE-Subunit: A MULTIMERICα Killer Toxin KP6 Ustilago maydisStructure of

doi: 10.1074/jbc.274.29.204251999, 274:20425-20431.J. Biol. Chem. 

  http://www.jbc.org/content/274/29/20425Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/274/29/20425.full.html#ref-list-1

This article cites 30 references, 6 of which can be accessed free at

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from