Supporting Information

Files in this Data Supplement:

Fig. 7. The ¹H,¹⁵N HSQC spectra of the two L27 domain complexes studied in this work. (a) Superposition plot of the HSQC spectra of the L27_Lin2C/L27_Lin7 complexes in the forms of a covalently linked single-chain fusion protein (black) and the two L27 domains purified as two separate chains (red). The cleavable form of the L27_Lin2C/L27_Lin7 complex contains additional eight residues (Leu-Val-Pro-Arg-Gly-Ser-Ser-Gly) necessary for thrombin digestion. The HSQC spectrum of the cleaved L27_Lin2C/L27_Lin7 complex contains a number of additional peaks (red, labeled with asterisks) in the random coil shift region when compared with the HSQC spectrum of the single-chain fusion protein. Amino acid residues originated from the linker regions are lableled. (b) Plot as a function of the residue number of combined ¹H and ¹⁵N chemical shift changes of L27_Lin2C/L27_Lin7 as a result of covalent linking. The combined ¹H and ¹⁵N chemical shift changes are defined as:

D _ppm = [(D d _HN)² + (D d _N × a _N)²]^1/2,

where D d _HN and D d _N represent chemical shift differences of amide proton and nitrogen chemical shifts of L27 domain in covalent linked and unlinked forms. The scaling factor (a_N) used to normalize the ¹H and ¹⁵N chemical shifts is 0.17. The secondary structure of the L27 domains is indicated at the top of the plot. Due to the relatively short linker used in the single-chain fusion protein, we noticed small chemical shift changes induced by covalent linking in the two L27 domains. It is possible that the linker may have some small impact on the conformation of the complex. (c) Superposition plot of the HSQC spectra of the L27_Patj/L27_Pals1N complexes in the forms of single-chain fusion protein (black) and two separate chains (red). The thrombin-cleaved complex has an identical amino acid sequence to the single-chain fusion protein. Therefore, the thrombin cleavage did not generate additional peaks as observed in a, and the two spectra overlap well with each other except for the residues in the linker region. (d) Plot as a function of the residue number of combined ¹H and ¹⁵N chemical shift changes of L27_Patj/L27_Pals1N as a result of covalent linking.

Fig. 8. Surface representation showing the packing interface of the Patj/Pals1 L27 domain complex. In Upper, L27_Patj is in the surface representation; L27_Pals1N is in the worm model. In Lower, L27_Pals1N is in the surface model, and L27_Patj is shown in the worm model. The coloring scheme is the same as in Fig. 1c.

Fig. 9. Interface between the two units of the L27_Patj/L27_Pals1N heterodimers. (a) Selective strips of ¹³C half-filtered NOESY spectrum of ¹³C,¹⁵N-labeled/unlabeled L27_Patj/L27_Pals1N complex mixture showing the contacts between the a C-helices from the two heterodimers. (b and c) Topologies of the central helix bundle of the L27_Patj/L27_Pals1N complex in solution and in the crystal, respectively. The side chains of the residues that display interheterodimer nuclear Overhauser effects (NOEs) shown in a are drawn by using explicit atomic representations. Note that none of these interheterodimer NOEs is compatible with the central helix assembly mode shown in the crystal structure of the L27_Patj/L27_Pals1N complex.

Fig. 10. ¹H, ¹⁵N HSQC spectra of the L27_Patj/L27_Pals1N complex at 1 mM (black) and 10 mM (red).

Fig. 11. Mutational analysis of the interheterodimer interfaces of the two pairs of L27 tetramer complexes. (a) Drawing of the amino acid residues chosen for mutation (Ile-57 of L27_Patj and Ile-169 of L27_Pals1N) in the interface of the L27_Patj/L27_Pals1N complex. Note that these two residues are exclusively involved in the packing of the interheterodimer interface of the complex. (b) Analytical gel filtration chromatography elution profiles of the wild-type (red curve) and the mutant (black curve) forms of single-chain L27_Patj-L27_Pals1N fusion proteins. The wild-type complex was eluted as a dimer of the fused heterodimer. In contrast, the mutant was eluted as a monomer of the fused protein. (c) Chemical cross-linking analysis showing that the wild-type L27_Patj-L27_Pals1N complex could be cross-linked into an »34 kDa band corresponding to the dimer of the heterodimer molecular mass of the L27_Patj/L27_Pals1N complex. In contrast, the mutant L27_Patj-L27_Pals1N remained as a monomer (»15 kDa) even after prolonged incubation with the cross-linking reagent. Both gel filtration and chemical cross-linking data point to the critical role of the aC-helix of each L27 domain in the tetrameric assembly of the L27_Patj/L27_Pals1N complex. (d) Drawing of the amino acid residues (Leu-57 of L27_Lin7 and Val-477 of L27_Lin2C) mutated in the interface of the L27_Lin2C/L27_Lin7 complex. (e and f) Analytical gel filtration (e) and chemical cross-linking (f) analysis showing mutation of Leu-57 of L27_Lin7 and Val-477 of L27_Lin2C disrupted dimer of the heterodimer assembly of the L27_Lin2C/L27_Lin7 complex. In this study, the single-chain L27_Lin2C-L27_Lin7 fusion protein was used for analysis. Analytical gel filtration chromatography was carried out on an AKTA FPLC system by using a Superose 12 10/30 column (Amersham Pharmacia Biotech). Protein samples (»2 mg·ml^-1) were dissolved in 100 mM potassium phosphate buffer containing 1 mM DTT. The column was calibrated with the low molecular mass column calibration kit from Amersham Pharmacia Biotech. Chemical cross-linking was carried out by incubating L27 complexes with Lys-specific cross-linker disuccinimidyl glutarate (DSG). The reaction was carried out in 50 mM Hepes buffer (pH 7.5) with a protein concentration of 0.5 mg/ml at room temperature. The concentration of DSG was adjusted to be 5 M equivalent of total Lys concentration of each protein complex. The cross-linking reactions were quenched by addition of 50 mM Tris to the reaction solution.