Structural Basis for the Co-activation of Protein Kinase B byT-cell Leukemia-1 (TCL1) Family Proto-oncoproteins*□S

Received for publication, January 13, 2004, and in revised form, April 30, 2004Published, JBC Papers in Press, May 28, 2004, DOI 10.1074/jbc.M400364200

Daniel Auguin‡, Philippe Barthe‡, Catherine Royer‡, Marc-Henri Stern§, Masayuki Noguchi¶,Stefan T. Arold‡�, and Christian Roumestand‡�

From the ‡Centre de Biochimie Structurale, UMR 5048 CNRS/UM1-UMR 554 INSERM/UM1, Faculte de Pharmacie,BP14491, 15 Avenue Charles Flahault, 34093 Montpellier Cedex 5, France, §Unite INSERM U509, Institut Curie, Sectionde Recherche, 26 rue d’Ulm, F75248 Paris Cedex 5, France, and ¶Division of Cancer Biology, Institute for GeneticMedicine, Hokkaido University, N15 W7, Kita-ku, Sapporo 060-0815, Japan

Chromosomal translocations leading to overexpres-sion of p14TCL1 and its homologue p13MTCP1 are hall-marks of several human T-cell malignancies (1). p14TCL1/p13MTCP1 co-activate protein kinase B (PKB, also namedAkt) by binding to its pleckstrin homology (PH) domain,suggesting that p14TCL1/p13MTCP1 induce T-cell leukemiaby promoting anti-apoptotic signals via PKB (2, 3). Herewe combined fluorescence anisotropy, NMR, and smallangle x-ray-scattering measurements to determine theaffinities, molecular interfaces, and low resolutionstructure of the complex formed between PKB�-PH andp14TCL1/p13MTCP1. We show that p14TCL1/p13MTCP1 targetPKB-PH at a site that has not yet been observed inPH-protein interactions. Located opposite the phospho-lipid binding pocket and distal from known protein-protein interaction sites on PH domains, the binding ofdimeric TCL1 proteins to this site would allow the cross-linking of two PKB molecules at the cellular membranein a preactivated conformation without disrupting cer-tain PH-ligand interactions. Thus this interaction couldserve to strengthen membrane association, promotetrans-phosphorylation, hinder deactivation of PKB, andinvolve PKB in a multi-protein complex, explaining thearray of known effects of TCL1. The binding sites onboth proteins present attractive drug targets againstleukemia caused by TCL1 proteins.

Protein kinase B (PKB)1 is a 60-kDa member of the AGCsuperfamily of serine/threonine kinases composed of a amino-terminal PH domain and linked to a kinase domain by a 30amino acid linker. PKB is frequently called Akt, because it is amammalian homologue of v-Akt, a viral oncogene isolated fromthe Akt8 virus that causes T-cell leukemia in mice (4–7). Ac-tually, the PKB family comprises three members, PKB�,

PKB�, and PKB� (Akt1, Akt2, and Akt3), all of which display,despite some idiosyncrasies, a high level of functional redun-dancy (Fig. 1) (8). Protein kinase B is a central component ofphosphoinositide 3�-kinase signaling pathways and hasemerged as a pivotal regulator of many cellular processes in-cluding apoptosis, proliferation, differentiation, and metabo-lism (3, 9, 10). Deregulation of members of the PKB family hasbeen associated with human pathologies such as cancer anddiabetes (8).

PKB activation in response to growth factors and other ex-tracellular stimuli involves membrane recruitment of PKBtriggered by inositol phospholipid (PtdIns-P) binding of its ami-no-terminal pleckstrin homology (PH) domain. At the mem-brane, PKB is activated by a partially defined process involvinglipid-mediated dimerization and phosphorylation of two criticalresidues, Thr308/Thr309/Thr305 in the kinase activation seg-ment and Ser473/Ser474/Ser472 in the COOH-terminal hydro-phobic motif, on PKB�, PKB�, and PKB�, respectively (6, 11–16). Residue Thr308/Thr309/Thr305 is phosphorylated by the3-phosphoinositide-dependent kinase 1, whereas the mecha-nism responsible for the phosphorylation of Ser473/Ser474/Ser472 has not been resolved (8, 17).

The PH domain of PKB (PKB-PH) has been shown to beessential for mediating the targeting and co-activation of PKBby proteins of the T-cell leukemia-1 (TCL1) family (18, 19).Normally, the cellular expression of TCL1 family genes (TCL1,TCL1b, and MTCP1) is mainly restricted to the lymphoid celllineage and to the early stages of embryogenesis (20) wherethey co-activate PKB, possibly to promote a growth advantageduring development through PKB-stimulated cell survival (3).However, in certain T-cell malignancies, chromosomal translo-cations lead to changes in the expression patterns of TCL1family genes.

Thus the TCL1 oncogene was identified because of charac-teristic chromosomal translocations and inversions at 14q32.1in clonal T-cell proliferations and malignancies (21). Reposi-tioning of T-cell receptor �/� or �-chain control sequences nextto the TCL1 coding region yields deregulated T-cell-specificexpression. The product of the TCL1 gene is a 14-kDa protein(p14TCL1) that has been shown to localize in the cytoplasm andnucleus of expressing cells (22). Crystallographic studies indi-cated that p14TCL1 exhibits a novel �-barrel structure (23). Asimilar structure was found for the 13-kDa product of theMTCP1 gene (p13MTCP1) (23–25). The MTCP1 gene, located inthe Xq28 chromosomal region, was the first gene to be identi-fied in the heterogeneous group of uncommon T-cell leukemiaspresenting a mature phenotype (26). It is involved in the trans-location t(X;14)(q28;q11), recurrently associated with this typeof T-cell proliferations. In addition to structural similarity,

* This work was supported in part by a research grant from the“Association pour la Recherche sur le Cancer,” the “Fondation pour laRecherche Medicale” (to D. A.), and a cancer research investigatoraward and a grant-in-aid from the Ministry of Education, Science andTechnology (Japan) (to M. N.). The costs of publication of this articlewere defrayed in part by the payment of page charges. This article musttherefore be hereby marked “advertisement” in accordance with 18U.S.C. Section 1734 solely to indicate this fact.

□S The on-line version of this article (available at http://www.jbc.org)contains Supplemental Material 1 and 2.

� To whom correspondence may be addressed. E-mail: roume@cbs.cnrs.fr (C. Roumestand) and stef@cbs.cnrs.fr (S. T. A.).

1 The abbreviations used are: PKB, protein kinase B; PH, pleckstrinhomology; PtdIns-P, inositol phospholipid; TCL1, T-cell leukemia-1;p14TCL1, the 14-kDa protein product of the TCL1 gene; p13MTCP1, the13-kDa product of the MTCP1 gene; PDB, Protein Data Bank; R2,relaxation rate; SAXS, small angle x-ray scattering.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 279, No. 34, Issue of August 20, pp. 35890–35902, 2004© 2004 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.org35890

by guest on October 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

p13MTCP1 exhibits high sequence homology (40% identity, 61%similarity) with p14TCL1 and with p14TCL1b (36% identity, 63%similarity), the product of the newly identified TCL1b oncogene(27). Evidence confirming a tumorigenic role for aberrant TCL1and MTCP1 expression has been obtained from transgenicmice (28, 29), suggesting that these proteins are the first iden-tified members of a novel family of proto-oncoproteins.

Functionally, the association of TCL1 proteins with PKB,which is mediated by its PH domain, enhances the phosphoryl-

ation of PKB on Thr308/Thr309/Thr305 and Ser473/Ser474/Ser472,increases PKB-mediated phosphorylation of its substrates gly-cogen synthase kinase-3� and Bc12-antagonist cell of death,and allows PKB to enter the nucleus (18, 19). The underlyingmolecular mechanisms remain controversial. It was suggestedthat TCL1 proteins mimic PtdIns-P binding to PKB-PH (30),facilitate PKB trans-phosphorylation by the formation of het-erotrimeric complexes with PKB (18), or serve as adaptor pro-teins to link PKB with factors containing nuclear localization

FIG. 1. Residues involved in binding as mapped by NMR. Annotated sequence alignment of PKB family PH domains (A) and TCL1 familyproteins (B) is shown. Identical positions are highlighted in blue, and homologous positions are in cyan. Residues with an above-threshold deviationfrom mean in NMR T2-mapping experiments are labeled according to the following: #, caused by p13MTCP1; ^, caused by p14TCL1 (A); #, caused byPKB�-PH (B).

PKB Co-activation by TCL1 Proteins 35891

by guest on October 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

signals (19). Identification of TCL1 and PKB-PH binding sur-faces is an important step in deciphering these complex mech-anisms. A putative site of interaction at the TCL1 surface waspreviously proposed from a mutational analysis of p14TCL1 andmolecular modeling (31, 32). However, the region(s) onPKB-PH that contribute to the interaction are unknown. Ad-vances toward identifying this interaction site were providedrecently by the resolution of the three-dimensional structure ofthe PKB-PH domain (�-and �-isoforms) and the delineation ofthe binding site of the PtdIns-P (33, 34). Here, we used acombination of biophysical techniques (fluorescence anisot-ropy, NMR, and small angle x-ray scattering (SAXS)) to deter-mine the structural basis of the interaction between PKB�-PHand p13MTCP1/p14TCL1. Based on our experimental data, a mo-lecular model of the complex formed between PKB�-PH andp14TCL1 is proposed. Our analysis gives insights into how TCL1family proteins promote their array of cellular effects.

MATERIALS AND METHODS

Protein Preparation—PKB�-PH, p13MTCP1, and p14TCL1 were pro-duced, purified, and, in the case of PKB�-PH and p13MTCP1, 15N-labeledas described previously (23, 35, 36).

NMR Experiments—Chemical shift assignments of PKB�-PH andp13MTCP1 were described previously (24, 35). All of the NMR experi-ments were carried out at 10 °C on a 500-MHz BRUKER AVANCEspectrometer equipped with a 5-mm z-gradient 1H-13C-15N triple-reso-nance cryoprobe. Protein samples were dissolved in 300 �l of buffer (10mM Tris-HCl, 300 mM NaCl, pH 7.4) in Shigemi cells. In all of theexperiments, the 1H carrier was centered on the water resonance and aWATERGATE (37, 38) sequence was incorporated to suppress the sol-vent resonance. All of the NMR spectra were acquired in the phase-sensitive mode with Digital Quadrature Detection in the F2 dimensionand the hypercomplex States-TPPI (Time Proportional Phase Incre-mentation) method in F1 dimension (39) and processed with Gifa (ver-sion 4.22) (40) software utility.

Titration Experiments—For each titration experiment, eleven ali-quots were prepared where the concentration of 15N-labeled PKB�-PH(or 15N-labeled p13MTCP1) was kept constant (60 �M) and the concen-tration of unlabeled p13MTCP1 (or PKB�-PH) and p14TCL1 was increasedfrom 5 to 700 �M and from 2.5 to 200 �M, respectively. [1H,15N]-HSQC(41, 42) spectra were recorded for each sample using a time domain datasize of 64 t1 � 2K t2 complex points and 32 transients/complex t1

increment. The dissociation constants (KD) per residue were measuredfrom the fit of the decreased intensities of all non-overlapping cross-peaks with Equation 1 (43),

y � Imax

�(Imax � Imin) � ([Pt] � [Lt] � KD) � �([Pt] � [Lt] � KD)2 � 4[Pt][Lt]

2[Pt]

(Eq. 1)

where [Pt] and [Lt] are the total concentration of 15N-labeled and unla-beled proteins, respectively.

T2-mapping Experiments—Transverse relaxation rates (R2) weremeasured and analyzed following a standard protocol previously re-ported in details for the backbone dynamics analysis of 15N-p13MTCP1

(24) and 15N-PKB�-PH (34). In this particular case, the single Hahnecho experiment was preferred to the classical CPMG (Carr-Purcell-Meiboom-Gill) experiment for R2 measurements because of its in-creased sensitivity to exchange contributions (44). R2 were measuredfor 15N-p13MTCP1 and 15N-PKB�-PH “free” in solution and comparedwith the ones measured for the same protein in the presence of apredetermined amount from titration experiments of unlabeled partner(15N-p13MTCP1�PKB�-PH (1:1), 15N-PKB�-PH�p13MTCP1 (1:1), and 15N-PKB�-PH�p13MTCP1 (5:1)). In these experiments, the total (labeled plusunlabeled) protein concentration was fixed to 0.5 mM for all of thesamples in order to avoid any parasitic contribution from viscositychanges.

Fluorescence Anisotropy Experiments—Proteins were labeled at pH8.0 at 4 °C for 2 h, conditions under which labeling ratios are far fromunity and that favor unique labeling of the amino terminus. The label-ing ratio for the PKB�-PH proteins was 16%, and that for the p13MTCP1

protein was 10%. Concentrations for the target protein were calculated

for total protein, such that the concentration of fluorophore was in bothcases considerably lower. Binding assays were performed with a Beacon2000 polarization instrument (Panvera Corp., Madison, WI) at 4 °Cusing either Alexa488-labeled PKB�-PH at a total concentration of 32nM or Alexa488-labeled p13MTCP1 at a total concentration of 100 nM. Thebuffer used in the assay was 10 mM Tris-HCl, pH 7.4, and 300 mM NaCl.Anisotropy profiles were obtained in the dilution format. In this ap-proach, data points were taken at equilibrium starting from 240 �l of a0.1 mM solution of p14TCL1 containing 32.5 nM Alexa488-PKB�-PH or0.5 mM solution of PKB�-PH containing 100 nM Alexa488-labeledp13MTCP1. For each subsequent measurement, 60-�l aliquots of solutionwere removed from the initial solution and replaced by 60 �l containing32.5 nM Alexa488-PKB�-PH or 100 nM Alexa488-labeled p13MTCP1 only.Anisotropy values were recorded as a function of time, and the final fiveor six values after stabilization were averaged. Anisotropy data were fitusing a 1:1 association mass action law in a 99% confidence intervalusing a GraphPad Prism Fitter.

SAXS—SAXS data were recorded on beamline D24 at LURE (Orsay,France) at a wavelength of 1.49 Å using a linear 512 channel detectorand sample-detector distances of 1.872 and 0.880 m. For data acquisi-tion, the samples were kept in a quartz capillary of �1.5 mm in diam-eter maintained at a temperature of 10 °C. Protein samples were dia-lyzed into 10 mM Tris-HCl and 300 mM NaCl at pH 7.4 and filteredthrough 0.22-�m pores prior to data recording. Sample concentrationswere adjusted to 0.6 mM. For each detector distance, eight frames of200 s each were averaged for both the protein samples and correspond-ing buffer. The individual frames did not show signs of x-ray damageand were averaged and scaled to transmitted intensity. The buffercontribution was subtracted from the protein-scattering curve. Thedata were merged and processed using the programs Primus and Gnom(45). Dammin and Gasbor were used for ab initio form determination(45). SAXS envelopes determined by these programs showed P2 sym-metry. This symmetry constraint was imposed for subsequent ab initioshape determinations. Ten of the obtained ab initio models were aver-aged to obtain the most likely molecular envelope (using Damaver (45)).The �2 values of atomic models against SAXS data were calculated withCrysol (45) using default parameters.

Molecular Modeling—Manual modeling of the PKB�-PH�p14TCL1

complex was carried out using the program O (46) for visualization ofatomic models and SAXS envelopes. Computational docking was car-ried out with the program FTDock (47). A total of 37,000 possibleorientations of PH relative to p14TCL1 or p13MTCP1 were tested byFTDock for combinations of the x-ray structure of PKB�-PH and NMRstructure of PKB�-PH with x-ray structures of p14TCL1 and p13MTCP1

and an NMR structure of p13MTCP1. This initial set of orientations wasfiltered to suffice the constraint that residues of the interface of bothpartners are �6 Å away from the other molecule. The best-scoredmodels from FTDock were submitted to rigid body and side chainrefinement by the program MultiDock (48).

An alternative approach using the HADDOCK (High AmbiguityDriven DOCKing) (49) program provided us with the means to dock theTCL1 proteins separately onto PKB�-PH. The chemical shift perturba-tion data resulting from NMR-mapping experiments were used as am-biguous interaction restraints to drive the docking process. An ambig-uous interaction restraint is defined as an ambiguous distance amongall of the residues shown to be involved in the interaction. According toestablished criteria of the HADDOCK program, the “active” residuesare those that have been shown to alter the HSQC spectra and also havea high solvent accessibility. The “passive” residues correspond to theresidues that are surface neighbors of the active residues and also havea high solvent accessibility calculated with NACCESS program. There-fore, PKB�-PH residues Lys64, Glu66, and Arg67 in strand �5 andMet100, Arg101, Gln104, Met105, Asn108, and Ser109 in the COOH-terminalhelix are considered to be active residues. Five neighboring PKB�-PHresidues, Leu62, Met63, Pro70, Gly97, and Lys111, are considered to bepassive residues. For the p14TCL1 protein, residues Asp16, Arg17, Trp19,Glu29, Lys30, Gln31, Ile74, Gln77, and Ser87 are considered active resi-dues when the Asp13, His32, Leu58, Pro61, Pro64, Tyr79, and Asp88 arepassive residues. The default HADDOCK parameters were used withthe except that only the 300 initial complex structures were generated,the best 50 solutions in terms of intermolecular energies then wererefined in water, and the 20 final structures were analyzed. All of themolecular images were produced with Aesop2 or PyMOL (50).

2 M. E. M. Noble, unpublished data.

PKB Co-activation by TCL1 Proteins35892

by guest on October 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

RESULTS

Affinities of the PKB-PH-p13MTCP1/p14TCL1 Interactions—The affinity of PKB�-PH for p14TCL1 was determined usingfluorescence anisotropy. In this experiment, the PKB�-PH pro-tein was labeled with an amine-reactive fluorescent dye (Al-exa488) under conditions that favor the selective labeling of theamino terminus. A solution of this labeled protein at a concen-tration of 32 nM in the presence of a high concentration ofunlabeled p14TCL1 was sequentially diluted with a solutioncontaining only the labeled protein and at each point the fluo-rescence anisotropy was measured (see “Materials and Meth-ods”). The total change in the anisotropy was 50 millianisot-ropy units from a value of 185 at very low concentrations ofp14TCL1 into 235 millianisotropy units at saturating p14TCL1.The raw data were fit to a model of simple binding of one dimerof p14TCL1 to PKB�-PH, and the normalized change in anisot-ropy is plotted along with the normalized fit of the raw data(Fig. 2, squares). The fit was satisfactory as evidenced by a lowrelative value of �2 and random residuals and yielded a KD

value of 5.7 �M. We note that, since the total concentration ofthe labeled PKB-PH protein was quite low in order to ensuretrue equilibrium binding, the probability of populating a com-plex containing more than one PKB-PH protein per p14TCL1

dimer was infinitely small.The binding of PKB-PH to p13MTCP1 was also measured by

fluorescence anisotropy using a p13MTCP1 sample labeled in thesame manner by Alexa488 to maximize the total anisotropychange. The labeled p13MTCP1 at a concentration of 100 nM wastitrated by unlabeled PKB-PH (Fig. 2, triangles). The totalchange in anisotropy was from a value of 110 millianisotropyunits at low concentrations of PKB�-PH to a value of 210millianisotropy units at the highest concentration available,which was determined by the limit of solubility of the usedPKB�-PH construct. The data were fit to the same simplebinding model, and the normalized data and fit are plotted alsoin Fig. 2. It is clear from the lack of high concentration plateauthat the affinity of the complex between PKB-PH and p13MTCP1

is significantly lower than that formed by PKB-PH andp14TCL1. The fit of the raw data allowing the undeterminedplateau value to float yields a KD value of 537 �M, �100-fold ofthat found for p14TCL1.

We also carried out NMR titration experiments of PKB-PH

(isoform �; PKB�-PH) by p13MTCP1 or p14TCL1. These meas-urements are not redundant with fluorescence anisotropymeasurements, because they are indicative of the exchangeregime between the free and bound states with regard to theNMR time scale under the conditions of the NMR study. Thisinformation is mandatory for the choice of the NMR methods tobe used for delineating the binding interfaces. Of course, theNMR titrations are expected to yield less accurate KD valuesthan fluorescence anisotropy because of the relatively highconcentration of the labeled target and especially because dif-ferent mechanisms can participate to the line broadening ofresonances belonging to residues located on the binding inter-face (see below), leading to an underestimation of the apparentKD. Nevertheless, only a few residues are expected to belong tothis category, such that only a small bias should be observed inthe average result over all of the residues. Adding increasingamounts of unlabeled p13MTCP1 or p14TCL1 to a solution of15N-labeled PKB�-PH causes a progressive line broadening inthe [1H-15N]HSQC spectrum of PKB�-PH, indicative of an in-termediate-to-slow-exchange process. Under such limiting con-ditions, it has been shown that a dissociation constant can beestimated from progressively disappearing resonances (51–53).With 1H and 15N assignments being available from a previousstudy (35), the intensity decrease can be fitted for most cross-peaks in the HSQC spectra with a simple two-state model (see“Materials and Methods”), allowing the estimation of an “ap-parent” dissociation constant (KD) per residue (Fig. 3). Wetitrated 15N-labeled PKB�-PH with unlabeled p13MTCP1 andp14TCL1 to obtain the binding affinity by averaging the appar-ent KD per residue for all of the residues. Thus the KD values of410 � 140 and 4 � 1.25 �M were established for the complexes15N-PKB�-PH�p13MTCP1 and 15N-PKB�-PH�p14TCL1, respec-tively. A similar KD (380 � 140 �M) was obtained for thecomplex PKB�-PH�15N-p13MTCP1 when 15N-labeled p13MTCP1

was titrated with unlabeled PKB�-PH. The KD values obtainedfor both complexes, PKB�-PH�p14TCL1 and 15N-PKB�-PH�p13MTCP, are in good agreement with those obtained fromthe fluorescence anisotropy profiles.

To distinguish between intermediate or slow-exchange re-gimes, highly accumulated spectra were recorded on proteinsamples under conditions favoring substantially high con-centrations of the PKB�-PH�15N-p13MTCP1 or 15N-PKB�-PH�

p14TCL1 complex as determined from the value of the dissoci-ation constants (PKB�-PH�15N-p13MTCP1 (10:1) and 15N-PKB�-PH�p14TCL1 (1:1), respectively) and compared with referencespectra recorded on “free” species (15N-p13MTCP1 and 15N-PKB�-PH, respectively). In the intermediate chemical ex-change case in addition to extensive broadening, a displace-ment of the resonances should be observed from the resonancevalue in the free state to the resonance value in the boundstate. On the other hand, if the complex dissociation is slow,one should observe one set of resonance for the free protein andone set for the bound protein. During the titration, the free setwill disappear and will be replaced by the bound set. Of course,for similar reasons as previously cited, only resonances belong-ing to the binding interface should be affected. When compar-ing a spectrum recorded on a 15N-PKB�-PH/p14TCL1 (1:1) sam-ple to a reference spectrum (15N-PKB�-PH alone), we cannotidentify any additional resonances or resonance displacements(Supplementary Material 1). This result is strongly indicativeof slow-exchange conditions in the limiting case where theresonances belonging to the bound set are broadened beyonddetection. Such conditions are compatible with the high molec-ular weight expected for the multimeric complex 15N-PKB�-PH�p14TCL1 (56 kDa) (the dimeric protein p14TCL1 is supposedto bind two PH domains). This slow-exchange regime is also

FIG. 2. Fluorescence normalized anisotropy-derived profilesof the interactions between PKB�-PH and p14TCL1 or p13MTCP1.Squares represent the profile obtained upon titration of Alexa488-labeled PKB�-PH by unlabeled p14TCL1, whereas triangles correspondto the titration of Alexa488-labeled by unlabeled PKB�-PH. Linesthrough the points represent the best fit to the data of a simple 1:1binding model. We note that, in both cases, the fits were performed onthe raw data prior to any manipulation and that the raw data and thefits were normalized to facilitate comparison. Buffer conditions were4 °C in 10 mM Tris-HCl, pH 7.4, and 300 mM NaCl.

PKB Co-activation by TCL1 Proteins 35893

by guest on October 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

FIG. 3. Affinities between p13MTCP1 and PKB�-PH as measured by NMR titration experiments. A, evolution of the [1H,15N]HSQCspectra of 15N-labeled sample of p13MTCP1 upon the addition of increasing amounts of unlabeled PKB�-PH. B, left panel, mean-normalized titrationcurve fitted from cross-peak intensity decays of some selected residues representative for the global evolution of the [1H-15N]HSQC spectra ofp13MTCP1. Right panel, “apparent” KD for the complex PKB�-PH�p13MTCP1 as a function of the sequence. The average values are indicated with adashed line on the graphs, and the means � S.D. given in the text have been calculated from all of the residue-specific KD values over the wholesequence. C and D, affinities between p14TCL1 and PKB�-PH as measured by NMR titration experiments. C, evolution of the [1H,15N]HSQC spectraof 15N-labeled sample of PKB�-PH upon the addition of increasing amounts of unlabeled p14TCL1. D, left panel, mean-normalized titration curvefitted from cross-peak intensity decays of some selected residues representative for the global evolution of the [1H-15N]HSQC spectra of p14TCL1.Right panel, apparent KD for the complex PKB�-PH�p14TCL1 as a function of the sequence. The average values are indicated with a dashed lineon the graphs, and the mean � S.D. given in the text have been calculated from all of the residue-specific KD values over the whole sequence.

PKB Co-activation by TCL1 Proteins35894

by guest on October 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

compatible with the dissociation constants measured either byNMR or by fluorescence. When comparing a spectrum recordedon a 15N-p13MTCP1 sample in the presence of PKB�-PH (10:1)

to a reference spectrum (15N-p13MTCP1 alone), we observed anintense line broadening of most resonances. In addition, slightshifts are noticeable for the resonances of residues Val15, Arg22,

FIG. 3—continued

PKB Co-activation by TCL1 Proteins 35895

by guest on October 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Glu24, Gln26, Gln70, and Leu71 (Supplementary Material 1).These residues are likely to be in the binding site as revealed byT2-mapping experiments (see below). Thus this result is infavor of intermediate exchange conditions for the complexPKB�-PH�15N-p13MTCP1.

Binding Interfaces within the Complexes between PKB-PHand p13MTCP1/p14TCL1—Due to the intermediate-to-slowchemical-exchange regime, the widely used chemical shift per-turbation-mapping NMR method cannot be used to map pro-tein interfaces because it relies on fast-exchange conditionsbetween free and bound proteins (54, 55). On the other hand,more recently two different groups have published independ-ently an alternative NMR approach applicable under theseconditions that relies on the analysis of the differential linebroadening of the NMR signals. We have used this so-calledNMR T2-mapping method (56, 57) to determine the imprint ofunlabeled p13MTCP1/p14TCL1 and PKB�-PH, respectively, on15N-labeled PKB�-PH and p13MTCP1.

The theoretical principles of NMR two-site chemical ex-change have been studied extensively (58–63), and the lineshapes during exchange can be simulated with Equation 2 (63).

I(�) � �0

W exp[i( � wE)t � Kt � Rt]1 dt (Eq. 2)

Matrix R contains the transverse relaxation rates, whereasmatrices K and contain the chemical exchange rates andchemical shifts, respectively. Matrix W contains the probabilityof occurrence at each frequency, and E and 1 are the identitymatrix and unity vector, respectively. Considering this equa-tion, when the exchange between the bound and free states isslow or intermediate, two distinct situations can be considered.First, where the chemical shifts in the bound and free statesare the same or very near (�� �� 50 Hz), exchange broadeningdue to chemical exchange is absent or negligible and the linebroadening is very sensitive to the relaxation rate of the boundstate. This is the case for the resonances of residues locatedoutside the binding surfaces. These resonances sense the samechemical environment in the free and in the bound states.Alternatively, where chemical shift differences upon bindingare large (�500 Hz), the line broadening is dominated by theexchange rate in addition to the relaxation rate of the residuein the bound state. Thus NMR signals of residues that formbinding contacts associated with large chemical shift perturba-tions have two sources of exchange broadening as opposed toone for those residues that do not undergo environmentalchanges upon binding. Such exchange broadening will be evi-dent via 15N NMR T2 (R2) relaxation measurements such asthose presented here. Further exchange broadening of contactresidues can also be achieved via secondary processes of con-formational change at the contact site. If this conformationalchange is associated with large chemical shift changes, thiswould promote an additional relaxation pathway contributingefficiently to T2 relaxation. This three-site exchange mecha-nism can be described by modified three-site Bloch equations(64). Note that in case of fast exchange, the line broadening dueto chemical exchange vanishes and a similar increase in T2 isexpected for all of the residues, uniquely from the contributionof the relaxation rate of the bound state.

R2 (1/T2) values were measured on the 15N-labeled protein“free” in solution and compared with the protein in the pres-ence of a predetermined amount of unlabeled partner. Theresults are given as the normalized ratio (R2

comp/R2ref �1) versus

the protein sequence (Fig. 4) where R2ref and R2

comp are hetero-nuclear 15N transversal relaxation rates measured for the la-beled protein free in solution and for a mixture of the twopartners in solution (15N-p13MTCP1�PKB�-PH (1:1), 15N-PKB�-

PH�p13MTCP1 (1:1), and 15N-PKB�-PH�p13MTCP1 (5:1)), respec-tively. Thus a normalized ratio of zero for a particular residueindicates no change in R2 between free and interacting protein,and a normalized ratio of 1.0 indicates an R2 value, whichduring interaction is double that observed in the free state. Formost residues, a similar ratio is observed that corresponds toan expected nearly identical increase of R2 in the complexentirely due to a nearly identical increase of the correlationtime for the corresponding 1H-15N vectors in the complexes. Forsome residues, significant deviations from the average � S.D.are observed, resulting from significant chemical exchange con-tributions to R2 measured in the complex. These residues de-fine finite areas on the surface of each protein, corresponding tothe interaction surfaces.

The residues highlighted by NMR T2 mapping on PKB�-PHupon interaction with p14TCL1 are mainly located on strands �4(residues Val57 and Ala58) and �5 (Lys64, Thr65, Arg67, andArg69) of the �-sandwich, and on the COOH-terminal �-helix(Ser92, Glu98, Met100, Arg101, Ile103, Gln104, Met105, Asn108, andSer109) (Figs. 1 and 5B). The PtdIns-P binding site, formedbasically by the loops VL1 and VL2 of PKB�-PH, appearsunaffected by the association with p14TCL1. No informationcould be obtained for the VL3 loop region because of the lack ofassignments in this particular area (residues shown in black inFig. 5B). This lack of assignment has been attributed to specialdynamic behavior in this loop. Nevertheless it is unlikely thatthis loop interacts with TCL1 proteins because an interaction isexpected to promote dynamic changes leading to additionalpeaks corresponding to VL3 residues in the HSQC spectrum ofthe bound form.

The outlined area does not show any striking features interms of hydrophobicity or charge distribution. The imprint ofp13MTCP1 on PKB�-PH delineates principally the same region,albeit less well defined and including some residues close to theperiphery or outside the area affected by p14TCL1 (residuesThr21, Arg23, Arg25, and Tyr26 on �2 and residues Gly37 andTyr38 on �3) (Figs. 1 and 4B). This can be attributed to a lowersignal-to-noise level of these data, possibly due to the loweraffinity of p13MTCP1 as compared with p14TCL1. Indeed, undersuch conditions, the “hits” due to specific interaction werebarely discernable from those arising from possible nonspecificinteractions.

The residues on p13MTCP1 that are highlighted uponPKB�-PH binding are clustered on one face of the eight-stranded anti-parallel �-barrel (Figs. 1 and 5A). In addition toresidues exhibiting significant deviation of the normalized ra-tio, a small number of resonances were no longer detectable inthe experiment recorded in the presence of PKB�-PH. Thesewere derived from residues His12, Leu61, Ser63, and Met68. Asthese resonances are clearly detectable in the spectrum of thefree p13MTCP1, they must be experiencing a large chemical shiftchange in the presence of the PH domain, thus becoming line-broadened beyond detection. Therefore, it is likely that theseresidues are contact points between the two proteins. The de-lineated area contains a hydrophobic cluster formed by Trp14

on �1, Leu67, Met68, and Leu71 on �5, surrounded by polar andcharged residues (Gln70 on �5, Asn80 on �6, and Arg22 andAsp24 on �2). The PKB�-PH binding site determined onp13MTCP1 by NMR mapping corresponds well to the PH bindingsite on p14TCL1 suggested by mutational analysis (31, 32).From the high structural and sequence homology betweenp13MTCP1 and p14TCL1 and the similarity of the imprint theyproduce on PKB�-PH, we deduce that the molecular complexesPKB�-PH�p13MTCP1 and PKB�-PH�p14TCL1 are structurallycomparable. In addition to the continuous area formed by thesefour �-strands, residues located in the �-helix seem to contrib-

PKB Co-activation by TCL1 Proteins35896

by guest on October 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

ute as well by the interaction with PKB�-PH. This small helixis located in the long flexible loop that connects the two �-me-anders forming the �-barrel structure of p13MTCP1 and facesthe main binding surface.

Low Resolution Structure of the Complex Formed betweenPKB�-PH and p14TCL1—We used SAXS to investigate the lowresolution structure of the molecular complex formed byPKB�-PH and p14TCL1. SAXS data were collected at two dif-ferent sample-detector distances (1.872 and 0.880 m). Thesescattering curves were merged to yield data between 241- and14-Å resolution (Fig. 6A). Using the program Gnom (45), thepair distribution of interatomic distances p(r) was obtainedfrom these data. Data was best fitted when using a maximumparticle diameter of Rmax � 105 Å, yielding a fit of 0.936(classified “excellent” by Gnom) (Fig. 6B). The radius of gyra-tion determined by the pair distribution function was 33.2 �0.094 Å, very close to that obtained by Guinier analysis (32 Å).Based on these data, a molecular envelope was established byaveraging the dummy atom models obtained ab initio from 10individual simulated annealing calculations. Through thisSAXS analysis, the PKB�-PH�p14TCL1 complex appeared as a

P2-symmetric form composed by a rod-shaped form with twoellipsoid extensions (Fig. 6C). Based on the atomic structuresavailable for PKB�-PH (Protein Data Bank (PDB) code 1P6S)and p14TCL1 (PDB code 1JSG) and considering that in solutionPKB�-PH is monomeric and p14TCL1 is dimeric (23, 36, 65), thecenter of the complex was attributed to the p14TCL1 dimer andeach ellipsoid extension was due to one PH domain (Fig. 6C).Indeed, the dimensions of the ellipsoid extensions and therod-shaped central part corresponded well to those of PKB�-PHand dimeric p14TCL1, respectively.

Without any further assumptions or modeling, these struc-tural data revealed that, in solution, p14TCL1 binds PKB�-PHin a 1:1 stoichiometry, forming a dimer of heterodimers wheretwo single PKB�-PH�p14TCL1 complexes are linked via thep14TCL1 dimer interface. In the complex, the PH domains arenot in contact with each other and the PH binding site onp14TCL1 does not overlap with its dimer interface.

Molecular Modeling of the PKB�-PH�p14TCL1 Complex—Mo-lecular docking experiments of PKB�-PH with TCL1 proteins,integrating NMR and SAXS data, were then attempted in orderto clarify the molecular basis of their interaction. For this

FIG. 4. Results of the T2-mappingexperiments. Bar charts of normalizedR2 ratio against the residue number for15N-p13MTCP1 in the presence of PKB�-PH (A) and for 15N-PKB�-PH in the pres-ence of p13MTCP1 (B) and p14TCL1 (C).Residues with above threshold (as definedin the text) shifts are indicated with theirsequence number. In the upper panel,dashed bars correspond to the sequencenumber of 15N-p13MTCP1 residues thathave HN-N correlations that are line-broadened beyond detection upon the ad-dition of PKB�-PH. In the two others pan-els, bold bars indicate PKB�-PH residuesthat are highlighted either by the addi-tion of p13MTCP1 or p14TCL.

PKB Co-activation by TCL1 Proteins 35897

by guest on October 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

reason, we combined our experimental results with manualand computational docking and included a number of assump-tions, as described below, to propose an atomic model of thecomplex formed between p14TCL1 and PKB-PH. However, wenote that the biological and functional conclusions described inthis paper were derived solely on the basis of experimental dataand general structural considerations and do not depend on thedetailed atomic model described below.

Structural studies indicate that the orientation of PtdIns-P-bound PH domains relative to the membrane is well conserved(66, 67). Imposing this PH-membrane orientation, only onepossible PKB�-PH�p14TCL1 arrangement could be obtainedthat satisfied all of the constraints, i.e. to be contained within

the limits of the SAXS envelope (both enantiomers weretested), to display surface complementarity between the li-gands, and to satisfactorily involve the binding sites as deter-mined here by NMR and previously by mutational analysis forp14TCL1 (31, 32). This PKB�-PH�p14TCL1 model was corrobo-rated by computational docking. Under the constraint thatresidues mapped to the binding sites are located within 6 Å ofthe partner molecule, the program FTDock (47) yielded asbest-scored solution a PKB-PH�p14TCL1 arrangement compara-ble to that determined manually (FTDock rpscore � 4.66; nextbest solutions were scored 3.98 and 3.80). After rigid body andside chain refinement by MultiDock (48), this model gave thebest fit to SAXS data compared with the 40 next best-scored

FIG. 5. Interaction of PKB�-PHwith the TCL1 protein family. A, foot-print of PKB�-PH onto 15N-p13MTCP1. Thesolvent-accessible surfaces of residues in-volved in the surface contacts as revealedby the T2-mapping experiments are col-ored in yellow. The residues colored in redare Asp11 and Leu67 that have been alsofound to be involved in the PKB�-PH�p14TCL1 interaction by mutationalanalysis. A ribbon representation of theprotein in the same orientation and withthe same color code is given above. B,footprint of p13MTCP1 and p14TCL1 onto15N-PKB�-PH. The residues colored inyellow, blue, and green are those high-lighted only by the addition of p13MTCP1

or only by the addition of p14TCL1 or bythe addition of either p13MTCP1 orp14TCL1, respectively. A ribbon represen-tation of the protein in the same orienta-tion and with the same color code is givenabove. In both panels, unaffected residuesare colored in white and unassigned resi-dues for PKB�-PH (essentially the VL3loop (34)) are colored black. The two viewspresented in each panel are related by an 180° rotation around the vertical axis.

PKB Co-activation by TCL1 Proteins35898

by guest on October 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

models compiled from different FTDock runs (�2 to raw datawas 3.06 compared with 3.35 and 3.41 for next best solutions(Crysol (45)) (Fig. 7). The buried-accessible surface area of themodeled PKB�-PH�p14TCL1 complex is 1030 Å2 (490 Å2 onPKB-PH and 540 Å2 on p14TCL1), which is within the rangeobserved for protein-protein associations with micromolar af-finities. In the model, PKB�-PH docks onto p14TCL1 mainlyusing an interface constituted by the first �-strand of the sec-ond �-sheet and the COOH-terminal helix. On p14TCL1, thebinding site is centered around Gln77 and involves Asp16 andTrp19, which have been shown to be important for the interac-tion by mutational analysis (32).

This model was further supported by the program HAD-DOCK (49). This program, which carries out computationaldocking under constraints derived from NMR data (see “Mate-rials and Methods”), proposed a structurally similar best solu-tion albeit with a slightly different respective orientation of thebinding surfaces (Supplementary Material 2), giving a lessfavorable �2 score (5.65).

DISCUSSION

Combining fluorescence anisotropy, NMR methods, andSAXS analysis, we have determined the affinities and molecu-lar framework of the complexes formed between TCL1 familyproteins and PKB�-PH. Whereas the affinities measured be-tween PKB�-PH and p14TCL1 using fluorescence anisotropy orNMR are compatible with the slow-exchange regime with re-gard to the NMR time scale, the dissociation constants meas-ured for the PKB�-PH�p13MTCP1 complex are more in favor of afast-exchange regime. Indeed, a rule of thumb is that interac-tions with a KD � 10 �M are slow exchange, whereas interac-tions with low affinity are immediate to fast exchange. How-ever, exceptions can be very dramatic. Slow-exchange

conditions for the interaction of a peptide with 500 �M affinitywas measured for the Hsp70 chaperones (because of a slow kon)(52), and fast-exchange conditions were encountered for thebinding of a phosphate compound to hemoglobin with an affin-ity of �1 nM (because of a multi-site binding mechanism) (68).We believe that the apparent discrepancy between the inter-mediate-to-slow regime and the low affinity observed for thePKB�-PH�p13MTCP1 complex probably arises because the kon issignificantly slower than diffusion-limited. Usually, importantconformational changes for one or more of the two partnersupon binding are invoked to explain such phenomenon. Eventhough such rearrangements cannot completely be discarded inthe present case (see below), it is unlikely that the surfacerecognition mechanism underlying the formation of the PKB�-PH�p13MTCP1 complex would require such important conforma-tional changes. Rather, we believe that, in this particular case,a low kon may have for origin a certain degree of aggregation ofp13MTCP1. Indeed, if the binding surface becomes temporarilymasked by nonspecific aggregation, the kon will be significantlylower than the expected diffusion constant.

It has been shown that NMR is well suited to study suchweakly interacting systems (reviewed in Ref. 55). DifferentNMR techniques have been developed to study these systemsdepending on the exchange regime. Although the chemical shiftperturbation mapping is the most widely used NMR method tomap protein interfaces, it cannot be used in the present casebecause it relies on fast-exchange conditions. On the otherhand, line-broadening analysis has been shown to provide auseful alternative in the case of a slow/intermediate-exchang-ing system. The binding surface of p13MTCP1, as revealed by T2

mapping, is in perfect agreement with previous results ob-tained from single or multiple site-specific mutations (31, 32) of

FIG. 6. SAXS analysis of the PKB�-PH�p14TCL1 complex. A, experimental scattering data. Error bars are indicated by vertical lines. B, pairdistribution function p(r) for the PKB�-PH�p14TCL1 complex. Open circles, experimental curve; solid line, calculated curve for a typical ab initiomodel. C, side and top views of the molecular envelope of the PKB�-PH�p14TCL1 complex obtained by averaging 10 individual ab initio models.Proposed position of PKB�-PH domains (circles) and the p14TCL1 dimer (trapeze) within the SAXS envelope are indicated in the side view.

PKB Co-activation by TCL1 Proteins 35899

by guest on October 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

p14TCL1. It involves mainly residues located in the �1, �2, �5,and �6 strands of the �-barrel structure that form a continuoussurface of �788 Å2, which is highly conserved among TCL1family members. Among the p13MTCP1 residues that differ fromp14TCL1, His12 and Arg22 (respectively Arg17 and Leu27 inp14TCL1) are situated in the center of the binding site and thusare likely to affect the affinity of p13MTCP1 for PKB�-PH. Inaddition, the binding site encompasses residues located in thesmall �-helix located in the long loop that joins the two�-sheets, forming the characteristic barrel structure of theTCL1 proteins. This helix is located on the same face of the�-barrel and is spatially close to the main binding surface asdelineated from NMR experiment. This structural element wasnot proposed as part of the TCL1 interaction site as deduced bysequence homology analysis (32), because its sequence divergessubstantially from one member to another. However, its struc-ture is conserved within the protein family. The differences insequence of this part of the binding surface could also contrib-

ute to the different affinities for PKB-PH measured forp13MTCP1 and p14TCL1. In addition, previous 15N relaxationmeasurements on p13MTCP1 (24) have shown that this helix aswell as the entire loop connecting the two �-meanders formingthe �-barrel structure experiences microsecond-to-millisecondmotions. In the dimeric crystal structure of p14TCL1, contactsexist between the loops of each of the two monomers thatprobably restrict the flexibility of this region when comparedwith p13MTCP1. It is possible that these contacts restrain thehelix position in the dimeric structure of p14TCL1 in a favorableorientation to bind PKB�-PH. This favorable conformationcould be reached only through the conformational changes inthis highly flexible loop in the monomeric structure ofp13MTCP1. Such rearrangements involve microsecond-to-milli-second motions that could explain a significant lower kon forp13MTCP1 and thus could provide an additional explanation forthe higher KD measured for the complex p13MTCP1�PKB�-PH.The binding site of p14TCL1/p13MTCP1 on PKB�-PH comprisesresidues located in the COOH-terminal �-helix as well as res-idues located in the �4 and �5 strands on one face of the�-sandwich. This surface ( 830 Å2) is located opposite thePtdIns-P binding pocket and remote from the G�� binding siteas determined for G protein receptor kinase 2-PH (66, 69),indicating that PKB�-PH is able to engage these three inter-actions simultaneously (Fig. 8). The TCL1 interface onPKB�-PH flanks the region (residues 67–77) necessary forbinding to periplakin (70), a nuclear localization signal-con-taining plakin family protein, suggesting that these bindingevents are also compatible. Finally, TCL1 binding to mem-brane-bound PKB-PH leaves accessible PKB-PH Trp80, a resi-due potentially involved in protein-protein interactions (71).Thus TCL1 proteins could cross-link and possibly stabilize anumber of PKB-PH interactions at the membrane, promotingthe formation of the high molecular weight protein complexes(18) and nuclear relocalization (1, 18).

Our analysis allows us to put forward a model in whichdimeric p14TCL1 cross-links two PKB molecules by binding to asurface region of their PH domains, which has not yet beenobserved in other PH-protein interactions. It should be notedthat NMR experiments can indicate contact points between twoproteins, but that it is not safe to conclude that all such con-tacts are crucial for specificity recognition and/or any subse-quent biological effect. Site-directed mutagenesis will be re-quired to dissect the contributions of individual residues. Themapping of the contact sites offers an important starting pointin such work. Nonetheless, some assumptions can be tenta-tively derived from the present model on how TCL1 familyproteins promote their array of cellular effects.

Contrary to current models (30), the association withp14TCL1 appears compatible with the membrane anchoring ofPKB-PH and even should strengthen significantly its mem-brane attachment by an avidity effect. Indeed, all of the PKB�-PH�p14TCL1 orientations that satisfy the NMR and SAXS con-straints share the fact that the two PtdIns-P binding sites ofboth PH domains point approximately in the same direction,away from the interface with dimeric p14TCL1. Together withstructural data indicating that interaction with TCL1 familyproteins does not affect the PtdIns-P binding site of PKB-PH(this study and Refs. 34 and 71), our analysis suggests that ap14TCL1 dimer is able to associate simultaneously with twomembrane-bound PH domains (Fig. 8). The co-localization ofhigh concentrations of p14TCL1 and activated PKB at the mem-brane has already led French et al. (32) to suggest that themembrane is the site of complex formation. Because constitu-tive membrane anchoring renders PKB oncogenic (8), we spec-ulate that the stimulation of a prolonged membrane association

FIG. 7. Fit of the atomic model of PKB�-PH�p14TCL1 to SAXSdata. A, calculated scattering curve of the atomic model for the PKB�-PH�p14TCL1 complex (solid line) fitted to the experimental scatteringpattern of PKB�-PH�p14TCL1 (open circles). B, atomic model of thePKB�-PH�p14TCL1 complex fitted into the SAXS envelope (gray) in sideand top views. Atomic structures are colored according to the following:blue, PKB�-PH; red, p14TCL1; and green, residues of PKB�-PH impli-cated in binding to PtdIns-P.

PKB Co-activation by TCL1 Proteins35900

by guest on October 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

of PKB contributes to the transforming potential of p14TCL1.TCL1 proteins were suggested to mimic the conformational

change induced in PKB upon PKB-PH binding to PtdIns-P,thus disrupting a postulated PKB conformation where the PHand kinase domains assemble into a catalytically inactive form(30). However, our data show that TCL1 proteins are not directstructural mimics of PtdIns-P because their respective bindingsites are located on the opposite sides of PKB-PH and do notseem to be allosterically linked (this study and Refs. 34 and 71).Although the binding sites of TCL1 and PtdIns-P are not over-lapping, we cannot exclude the possibility that association withTCL1 is incompatible with the postulated “assembled” form ofPKB and hence that TCL1 proteins are able to stabilize PKB ina preactivated “dissembled” conformation. Of note, a TCL1enhancement of membrane attachment would also help to pre-vent PKB from dissociating from the membrane to adopt anassembled inactive conformation.

We have previously reported evidence that p14TCL1 stimu-

lates trans-phosphorylation of PKB molecules on Thr309 invitro (18). This is supported by our current analysis, becausewe show that dimeric p14TCL1 cross-links two PKB moleculesvia their PH domains in a way that promotes proximity be-tween their kinase domains. Indeed, for all of the possiblePKB�-PH�p14TCL1 orientations that bring the binding sites onboth proteins in contact with each other within the limits of theSAXS envelope, the 30 amino acid linker between the PKB PHand kinase domains is long enough to allow the kinase domainsof two p14TCL1-linked PKB molecules to physically contact eachother (Fig. 8). Thus our data support the notion that interactionwith p14TCL1 stimulates trans-phosphorylation of PKB by pro-moting proximity between the two kinase domains. Thr309

phosphorylation enhances the catalytic activity of PKB. There-fore, it is likely that the stimulation of PKB trans-phospho-rylation contributes to the transforming action of TCL1proteins.

In agreement with cellular analysis (31), dimerization of

FIG. 9. Conservation of the binding site within the TCL1 protein family. A, two views of p13MTCP1 showing the sequence homology of thesurface residues between p13MTCP1 and p14TCL1. The same color code has been used as in Fig. 1. The two views are related by a 180° rotation aroundthe vertical axis. B, footprint of PKB�-PH onto 15N-p13MTCP1 (same color code as in Fig. 5). For each panel, a ribbon representation of the proteinin the same orientation and with the same color code is given above.

FIG. 8. Association of p14TCL1 withPKB-PH is nonexclusive with PKB-PHbinding to membrane and G�� andstimulates proximity between twoPKB kinase domains. Side view of themodel proposed for the submembrane com-plex formed by PKB, p14TCL1, and G��.The PtdIns-P-bound PKB-PH domains(green, PDB code 1H10 (33)) were orientedtoward the membrane as suggested bycrystallographic analysis (66, 67). p14TCL1

was docked onto PKB-PH as suggested bycomputational analysis (see “Results”).G�� was orientated with respect to mem-brane and PH domain as determined for Gprotein receptor kinase 2-PH-G�� (PDBcode 1OMW (66)). PKB kinase domain wastaken from Ref. 72. (PDB code 1O6L) andconnected to PKB-PH by a linker modeledin an extended conformation.

PKB Co-activation by TCL1 Proteins 35901

by guest on October 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

TCL1 proteins appears pivotal for their function. Althoughboth p14TCL1 and p13MTCP1 co-activate PKB� and PK� (2, 31),only p14TCL1 appears to be homodimeric in the crystal (23) aswell as in solution (24). Considering that the affinity ofp13MTCP1 for PKB�-PH is significantly weaker than that dis-played by p14TCL1, it is likely that additional bridging mole-cules are necessary for the function of p13MTCP1. When com-paring the homodimerization surface of p14TCL1 with thecorresponding surface on p13MTCP1, the residue homology ap-pears less obvious than for the surface harboring the bindingsite to PKB�-PH (Fig. 9) on the opposite face of the barrel. Thissurface could likely offer potential binding sites for additionalbridging molecules.

Together, the association of TCL1 proteins with PKB-PHseems to stimulate an array of effects on PKB, leading toenhanced enzymatic activity as well as novel and/or extendedprotein-protein interactions. Targeting the herein identifiedPKB-TCL1 interface by molecular compounds could prove use-ful for therapeutic intervention against T-cell leukemiascaused by the overexpression of TCL1 family proteins.

Acknowledgments—We thank P. Vachette, C. Dumas, andL. Ponchon for assistance during SAXS data recording and analysis.

REFERENCES

1. Pekarsky, Y., Hallas, C., and Croce, C. M. (2001) J. Am. Med. Assoc. 286,2308–2314

2. Laine, J., Kunstle, G., Obata, T., and Noguchi, M. (2002) J. Biol. Chem. 277,3743–3751

3. Pekarsky, Y., Hallas, C., and Croce, C. M. (2001) Oncogene 20, 5638–56434. Staal, S. P., Hartley, J. W., and Rowe, W. P. (1977) Proc. Natl. Acad. Sci.

U. S. A. 74, 3065–30675. Bellacosa, A., Franke, T. F., Gonzalez-Portal, M. E., Datta, K., Taguchi, T.,

Gardner, J., Cheng, J. Q., Testa, J. R., and Tsichlis, P. N. (1993) Oncogene8, 745–754

6. Bellacosa, A., Testa, J. R., Staal, S. P., and Tsichlis, P. N. (1991) Science 254,274–277

7. Staal, S. P. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, 5034–50378. Brazil, D. P., and Hemmings, B. A. (2001) Trends Biochem. Sci. 26, 657–6649. Brunet, A., Bonni, A., Zigmond, M. J., Lin, M. Z., Juo, P., Hu, L. S., Anderson,

M. J., Arden, K. C., Blenis, J., and Greenberg, M. E. (1999) Cell 96, 857–86810. Datta, S. R., Brunet, A., and Greenberg, M. E. (1999) Genes Dev. 13,

2905–292711. Andjelkovic, M., Jakubowicz, T., Cron, P., Ming, X. F., Han, J. W., and Hem-

mings, B. A. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 5699–570412. Coffer, P. J., and Woodgett, J. R. (1991) Eur. J. Biochem. 201, 475–48113. Burgering, B. M., and Coffer, P. J. (1995) Nature 376, 599–60214. Kohn, A. D., Takeuchi, F., and Roth, R. A. (1996) J. Biol. Chem. 271,

21920–2192615. Franke, T. F., Kaplan, D. R., Cantley, L. C., and Toker, A. (1997) Science 275,

665–66816. Soskic, V., Gorlach, M., Poznanovic, S., Boehmer, F. D., and Godovac-Zimmer-

mann, J. (1999) Biochemistry 38, 1757–176417. Chan, T. O., Rittenhouse, S. E., and Tsichlis, P. N. (1999) Annu. Rev. Biochem.

68, 965–101418. Laine, J., Kunstle, G., Obata, T., Sha, M., and Noguchi, M. (2000) Mol. Cell 6,

395–40719. Pekarsky, Y., Koval, A., Hallas, C., Bichi, R., Tresini, M., Malstrom, S., Russo,

G., Tsichlis, P., and Croce, C. M. (2000) Proc. Natl. Acad. Sci. U. S. A. 97,3028–3033

20. Narducci, M. G., Fiorenza, M. T., Kang, S. M., Bevilacqua, A., Di Giacomo, M.,Remotti, D., Picchio, M. C., Fidanza, V., Cooper, M. D., Croce, C. M.,Mangia, F., and Russo, G. (2002) Proc. Natl. Acad. Sci. U. S. A. 99,11712–11717

21. Virgilio, L., Narducci, M. G., Isobe, M., Billips, L. G., Cooper, M. D., Croce,C. M., and Russo, G. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 12530–12534

22. Fu, T. B., Virgilio, L., Narducci, M. G., Facchiano, A., Russo, G., and Croce,C. M. (1994) Cancer Res. 54, 6297–6301

23. Hoh, F., Yang, Y. S., Guignard, L., Padilla, A., Stern, M. H., Lhoste, J. M., andvan Tilbeurgh, H. (1998) Structure 6, 147–155

24. Guignard, L., Padilla, A., Mispelter, J., Yang, Y. S., Stern, M. H., Lhoste, J. M.,and Roumestand, C. (2000) J. Biomol. NMR 17, 215–230

25. Fu, Z. Q., Du Bois, G. C., Song, S. P., Kulikovskaya, I., Virgilio, L., Rothstein,J. L., Croce, C. M., Weber, I. T., and Harrison, R. W. (1998) Proc. Natl. Acad.Sci. U. S. A. 95, 3413–3418

26. Stern, M. H., Soulier, J., Rosenzwajg, M., Nakahara, K., Canki-Klain, N.,Aurias, A., Sigaux, F., and Kirsch, I. R. (1993) Oncogene 8, 2475–2483

27. Pekarsky, Y., Hallas, C., Isobe, M., Russo, G., and Croce, C. M. (1999) Proc.

Natl. Acad. Sci. U. S. A. 96, 2949–295128. Gritti, C., Dastot, H., Soulier, J., Janin, A., Daniel, M. T., Madani, A., Grimber,

G., Briand, P., Sigaux, F., and Stern, M. H. (1998) Blood 92, 368–37329. Virgilio, L., Lazzeri, C., Bichi, R., Nibu, K., Narducci, M. G., Russo, G.,

Rothstein, J. L., and Croce, C. M. (1998) Proc. Natl. Acad. Sci. U. S. A. 95,3885–3889

30. Brazil, D. P., Park, J., and Hemmings, B. A. (2002) Cell 111, 293–30331. Kunstle, G., Laine, J., Pierron, G., Kagami Si, S., Nakajima, H., Hoh, F.,

Roumestand, C., Stern, M. H., and Noguchi, M. (2002) Mol. Cell. Biol. 22,1513–1525

32. French, S. W., Shen, R. R., Koh, P. J., Malone, C. S., Mallick, P., and Teitell,M. A. (2002) Biochemistry 41, 6376–6382

33. Thomas, C. C., Deak, M., Alessi, D. R., and van Aalten, D. M. (2002) Curr. Biol.12, 1256–1262

34. Auguin, D., Barthe, P., Auge-Senegas, M. T., Stern, M. H., Noguchi, M., andRoumestand, C. (2004) J. Biomol. NMR 28, 137–155

35. Auguin, D., Barthe, P., Auge-Senegas, M. T., Hoh, F., Noguchi, M., andRoumestand, C. (2003) J. Biomol. NMR 27, 287–288

36. Yang, Y. S., Guignard, L., Padilla, A., Hoh, F., Strub, M. P., Stern, M. H.,Lhoste, J. M., and Roumestand, C. (1998) J. Biomol. NMR 11, 337–354

37. Piotto, M., Saudek, V., and Sklenar, V. (1992) J. Biomol. NMR 2, 661–66538. Sklenar, V. (1995) J. Magn. Reson. A114, 132–13539. Marion, D., Ikura, M., Tschudin, R., and Bax, A. (1989) J. Magn. Reson. 85,

393–39940. Pons, J., Malliavin, T., and Delsuc, M. (1996) J. Biomol. NMR 8, 445–45241. Bax, A., Ikura, M., Kay, L. E., Torchia, D. A., and Tschudin, R. (1990) J. Magn.

Reson. 86, 304–31842. Bodenhausen, G., and Ruben, D. J. (1980) Chem. Phys. Lett. 69, 185–18943. Phizicky, E. M., and Fields, S. (1995) Microbiol. Rev. 59, 94–12344. Wang, L., Pang, Y., Holder, T., Brender, J. R., Kurochkin, A. V., and Zuider-

weg, E. R. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 7684–768945. Svergun, D. I., and Koch, M. H. (2002) Curr. Opin. Struct. Biol. 12, 654–66046. Jones, T. A., Zou, J. Y., Cowan, S. W., and Kjeldgaard, M. (1991) Acta Crys-

tallogr. Sect. A 47, 110–11947. Gabb, H. A., Jackson, R. M., and Sternberg, M. J. (1997) J. Mol. Biol. 272,

106–12048. Jackson, R. M., Gabb, H. A., and Sternberg, M. J. (1998) J. Mol. Biol. 276,

265–28549. Dominguez, C., Boelens, R., and Bonvin, A. M. (2003) J. Am. Chem. Soc. 125,

1731–173750. DeLano, W. L. (2003) PyMOL Reference Manual, DeLano Scientific LLC, San

Carlos, CA51. Zuiderweg, E. R., Hamers, L. F., Rollema, H. S., de Bruin, S. H., and Hilbers,

C. W. (1981) Eur. J. Biochem. 118, 95–10452. van Nuland, N. A., Kroon, G. J., Dijkstra, K., Wolters, G. K., Scheek, R. M., and

Robillard, G. T. (1993) FEBS Lett. 315, 11–1553. McAlister, M. S., Mott, H. R., van der Merwe, P. A., Campbell, I. D., Davis,

S. J., and Driscoll, P. C. (1996) Biochemistry 35, 5982–599154. Markus, M. A., Nakayama, T., Matsudaira, P., and Wagner, G. (1994) Protein

Sci. 3, 70–8155. Zuiderweg, E. R. (2002) Biochemistry 41, 1–756. Matsuo, H., Walters, K. J., Teruya, K., Tanaka, T., Gassner G. T., Lippard,

S. J., Kyogoku, Y., and Wagner, G. (1999) J. Am. Chem. Soc. 121,9903–9904

57. Howard, M. J., Chauhan, H. J., Domingo, G. J., Fuller, C., and Perham, R. N.(2000) J. Mol. Biol. 295, 1023–1037

58. Abragam, A. (1961) Principles of Nuclear Magnetism, Clarendon Press, Oxford59. Rogers, M. T., and Woodbrey, J. C. (1962) J. Phys. Chem. 66, 540–54660. Dwek, R. (1973) Nuclear Magnetic Resonance (N.M.R.) in Biochemistry, Clar-

endon Press, Oxford61. Jardetsky, O., and Roberts, G. (1981) NMR in Molecular Biology, Academic

Press, New York, NY62. Lian, L., and Roberts, G. (1993) NMR of Macromolecules, IRL Press, Oxford63. Cavanagh, J., Fairbrother, W., Palmer, A., III, and Skelton, N. (1996) Protein

NMR Spectroscopy, Academic Press, San Diego, CA64. Sundstrom, J. (1982) Dynamic NMR Spectroscopy (Press, A., ed) Academic

Press, London65. Auguin, D., Gostan, T., Delsuc, M., and Roumestand, C. (2004) C. R. Acad. Sci.

(Paris) 7, 265–27166. Lodowski, D. T., Pitcher, J. A., Capel, W. D., Lefkowitz, R. J., and Tesmer, J. J.

(2003) Science 300, 1256–126267. Rossman, K. L., Worthylake, D. K., Snyder, J. T., Siderovski, D. P., Campbell,

S. L., and Sondek, J. (2002) EMBO J. 21, 1315–132668. Carrington, A., and McLachlan, A. (1967) Introduction to Magnetic Resonance

with Application to Chemistry and Chemical Physics, Harper & Row, NewYork, NY

69. Konishi, H., Kuroda, S., Tanaka, M., Matsuzaki, H., Ono, Y., Kameyama, K.,Haga, T., and Kikkawa, U. (1995) Biochem. Biophys. Res. Commun. 216,526–534

70. van den Heuvel, A. P., de Vries-Smits, A. M., van Weeren, P. C., Dijkers, P. F.,de Bruyn, K. M., Riedl, J. A., and Burgering, B. M. (2002) J. Cell Sci. 115,3957–3966

71. Milburn, C. C., Deak, M., Kelly, S. M., Price, N. C., Alessi, D. R., and VanAalten, D. M. (2003) Biochem. J. 375, 531–538

72. Yang, J., Cron, P., Thompson, V., Good, V. M., Hess, D., Hemmings, B. A., andBarford, D. (2002) Mol. Cell 9, 1227–1240

PKB Co-activation by TCL1 Proteins35902

by guest on October 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Stefan T. Arold and Christian RoumestandDaniel Auguin, Philippe Barthe, Catherine Royer, Marc-Henri Stern, Masayuki Noguchi,

(TCL1) Family Proto-oncoproteinsStructural Basis for the Co-activation of Protein Kinase B by T-cell Leukemia-1

doi: 10.1074/jbc.M400364200 originally published online May 28, 20042004, 279:35890-35902.J. Biol. Chem. 

  10.1074/jbc.M400364200Access the most updated version of this article at doi:

 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/279/34/35890.full.html#ref-list-1

This article cites 64 references, 22 of which can be accessed free at

by guest on October 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from