Ship2-Sam and EphA2-Sam were expressed as previously reported . A synthetic genes construct containing residues from 1 to 80 of human Arap3 (UniprotKB/TrEMBL code: Q8WWN8), encompassing the Sam domain (residues from 4 to 68), was purchased from Celtek (Nashville, TN). Genes were cloned into the PET15b plasmid and transformed using BL21-Gold (DE3) competent cells (Stratagene).
The PET15b plasmid carrying genes for the Arap3-Sam triple mutant (H37D, R77D, R80D) was purchased from Celtek (Nashville, TN). These protein constructs have each an N-terminal 6His-tag (See Additional File 4).
Unlabeled proteins were expressed at 37°C in LB medium. Protein over-expression was induced at OD600 = 0.6 for 4 hours with isopropyl β-D-thiogalactopyranoside (IPTG) at 1 mM concentration. Expression of 15N/13C double labeled and 15N labeled proteins was carried out in M9 minimal medium containing 2 g/l of 13C-Glucose and/or 0.5 g/l of 15NH4Cl. 10% fractional 13C labeling for stereo-specific assignments of Leu-CH3δ1,2/Val-CH3γ1,2 methyl groups  was obtained by adding 3.6 g of 12C-glucose (natural abundance) and 0.4 g of 13C-glucose to the M9 medium.
After dissolving the pellet in the following buffer: 50 mM Tris (pH = 8), 500 mM NaCl, 5 mM imidazole, cell were harvested by sonication. The protein was purified on a nickel column (His-trap TM FF, 5 ml, Amersham) by using an AKTA prime plus FPLC system; the elution buffer consisted of 50 mM Tris (pH = 8), 500 mM NaCl, 200 mM imidazole. To avoid non-specific interactions with the Shiptide during ITC experiments, the His-tag tail of Arap3-Sam was cut away from the protein by incubating it overnight at 4°C with thrombin. The thrombin was then removed with a benzamidine column (FF (HS), 1 ml, Amersham).
The protein concentration was estimated by a nanodrop ND-1000 spectrometer.
Resonance assignments of Arap3-Sam
Experiments for resonance assignments were recorded at 25°C on a Bruker Avance 600 MHz or a 700 MHz Bruker AvanceIII spectrometers both equipped with TCI cryoprobes. NMR samples consisted of 15N or 15N/13C labeled Sam domains (1 mM) in phosphate buffer saline (PBS, 11.9 mM phosphate, 137 mM NaCl, 2.7 mM KCl) (Fisher Scientific) at pH = 7.7 with 0.3% NaN3. Sample volumes of 500 μl (95% H2O/5% D2O) were used. The Bruker software Topspin version 2.0 was implemented to process NMR spectra; NEASY http://www.nmr.ch/ was used to analyze the data.
Backbone assignments were obtained by analyzing triple resonance experiments (HNCA, HNCACB, HNCO) . The (H)CC(CO)NH spectrum was used to assign carbon side chains. Comparison of 3D 15N resolved-[1H, 1H] NOESY (100 ms mixing time) and 3D 15N resolved-[1H, 1H] TOCSY (70 ms mixing time) together with analysis of HCCH-TOCSY experiments allowed to assign proton side chains.
HN, NH and Cα backbone atoms of the Ship2-Sam/Arap3-Sam complex were identified in HNCA experiments acquired on samples containing either 15N/13C double labeled Ship2-Sam at a concentration of 1 mM and unlabeled Arap3-Sam at a concentration of 1 mM or double labeled Arap3-Sam (1 mM concentration) and unlabeled Ship2-Sam (1 mM concentration). Stereo-specific assignments for Leu-CH3δ 1,2 and Val-CH3γ 1,2 methyl groups of Arap3-Sam were obtained by recording a 2D [1H, 13C]-HSQC experiment of a fractionally 13C labeled Arap3-Sam sample at a concentration of 1 mM .
Experiments for measuring backbone 15N nuclear spin relaxation parameters, were recorded at 25°C on a 600 MHz Bruker Avance DRX spectrometer equipped with a TXI probe. Longitudinal relaxation rates (R1) and transverse relaxation rates (R2) were obtained for 15N-labeled samples of Arap3-Sam at the concentrations of 150 μM and 1.4 mM and for a sample of the Sam-Sam complex containing 15N-labeled Ship2-Sam (1 mM) and unlabeled Arap3-Sam (2 mM). R1 and R2 relaxation data were collected and analyzed as we have previously reported for the Ship2-Sam and EphA2-Sam interaction . Briefly, five relaxation delays (0.01, 0.1, 0.3, 0.6, 1.0 s) were used for R1 measurements; and seven relaxation delays were implemented for R2 data sets (i.e., 0.01, 0.03, 0.05, 0.07, 0.11, 0.15, 0.19 s). The rotational correlation time was estimated with the software tmest (Palmer A. G. III, Columbia University) by using the average R2/R1 values.
Solution structure of Arap3-Sam
Structure calculations were performed with the program CYANA version 2.1 . A 3D-15N resolved [1H, 1H] NOESY-HSQC spectrum  (100 ms mixing time) together with a 3D 13C resolved [1H, 1H] NOESY-HSQC spectrum (150 ms mixing time) and 2D [1H, 1H] NOESY  (100 ms mixing time), for the aliphatic to aromatic region, that was recorded after dissolving the lyophilized protein sample in 99% D2O, were used to obtain distance constraints. Structure calculations were initiated from 100 random conformers; the 20 structures with the lowest CYANA target functions were analyzed with the programs MOLMOL  and PROCHECK-NMR . Colour figures were produced with MOLMOL . MOLCAD , as implemented in Sybyl, was used to generate surface representations. NMR structures have been deposited in the Protein Data Bank under accession code 2KG5.
NMR Binding studies
Chemical shifts perturbation studies by means of 2D [1H, 15N]-HSQC were carried out to study the protein-protein interaction [15, 16]. First, 15N-labeled Ship2-Sam (200 μM concentration) was titrated with increasing amounts of unlabeled Arap3-Sam (50, 74, 130, 186, 233 μM). Then, a 15N labeled Arap3-Sam sample (200 μM concentration) was titrated with unlabeled Ship2-Sam (concentrations: 50, 100, 200 μM). Details on the NMR chemical shift perturbation studies with the Arap3-Sam triple mutant are reported in the Additional File 2.
2D spectra were compared with the program Sparky (T. D. Goddard and D. G. Kneller, SPARKY 3, University of California, San Francisco).
Isothermal titration calorimetry
ITC measurements were carried out at 25°C with a VP-ITC apparatus (Microcal, USA). The Shiptide (Ac-EGLVHNGWDDLEFLSDITEEDL-NH2) , was purchased by the Protein/DNA Facility of the Medical College of Wisconsin. A solution of peptide (1 mM concentration) was titrated into a solution of Arap3-Sam at a concentration of 75 μM. For these studies, the peptide was dissolved in 1 × PBS at pH = 7.7 and Arap3-Sam was extensively dialyzed in the same buffer.
ITC experiments were repeated twice to evaluate the reproducibility of the data. Data were fit to a standard one binding site model using Origin as supplied by Microcal.
A Beckman ProteomeLab™ Optima XL-I analytical ultracentrifuge was used to carry out sedimentation equilibrium analysis. Three runs were performed by using samples with protein concentrations of 1 mg/ml, 0.33 mg/ml, and 0.11 mg/ml, respectively. Data were collected at the angular velocity of 30,000 rpm and at 20°C. Data were analyzed with the software HeteroAnalysis (James L. Cole; http://www.biotech.uconn.edu/auf/).
The program Haddock 1.3  was used to generate a model of the Ship2-Sam/Arap3-Sam complex. The NMR structures number one of both Ship2-Sam (pdb code: 2K4P, ) and Arap3-Sam (pdb code: 2KG5) were implemented for these studies. Chemical shift perturbation data were exploited to produce ambiguous interaction restraints (AIR). Residues H47, N48, W50, D51, D52, E54, F55, S57, D58, I59, T60, E61, E62, E66, Q70 of Ship2-Sam were set as active. For Arap3-Sam, residues H37, S70, T72, G73, K76, R77, R80, Q83 were considered active, whereas H61, E62, E63, K65, Q66 were set as passive. Residues for the AIR restraints were chosen among the ones with the greatest chemical shift perturbation because they either show high solvent exposure (> 30% as evaluated with MOLMOL ) or because they could provide interactions at the interface as shown in experimental structures of Sam-Sam complexes. The limit for the AIR restraints was kept to the default value of 2 Å. During the rigid body energy minimization stage, 2000 structures were calculated; in the second iteration a semi-flexible simulated annealing of the best 200 solutions was performed, finally a refinement in water was carried out. Segments 48-66 and 70-80 of Ship2-Sam and Arap3-Sam respectively, were set as semi-flexible and movements of their side-chains were permitted during the semi-rigid body docking protocol. Besides, residues of the C-terminal tail of Arap3-Sam (88-100) were set completely flexible during the whole docking calculation.