Mapping the energy of superantigen Staphylococcus enterotoxin C3 recognition of an alpha/beta T cell receptor using alanine scanning mutagenesis

Abstract

Binding of the T cell receptor (TCR) to a bacterial superantigen (SAG) results in stimulation of a large population of T cells and subsequent inflammatory reactions. To define the functional contribution of TCR residues to SAG recognition, binding by 24 single-site alanine substitutions in the TCR Vbeta domain to Staphylococcus aureus enterotoxin (SE) C3 was measured, producing an energy map of the TCR-SAG interaction. The results showed that complementarity determining region 2 (CDR2) of the Vbeta contributed the majority of binding energy, whereas hypervariable region 4 (HV4) and framework region 3 (FR3) contributed a minimal amount of energy. The crystal structure of the Vbeta8.2-SEC3 complex suggests that the CDR2 mutations act by disrupting Vbeta main chain interactions with SEC3, perhaps by affecting the conformation of CDR2. The finding that single Vbeta side chain substitutions had significant effects on binding and that other SEC3-reactive Vbeta are diverse at these same positions indicates that SEC3 binds to other TCRs through compensatory mechanisms. Thus, there appears to be strong selective pressure on SAGs to maintain binding to diverse T cells.

Figure 1

Relative TCR reactivities of SEC3 and SEC3 variant 1A4. Various concentrations of SEC3 wt and SEC3 variant 1A4 were used to inhibit binding of biotinylated scTCR to immobilized mAb KJ16 in an ELISA format. Bound biotinylated scTCR was detected by streptavidin-HRP. The relative reactivity of each SAG was determined as the ratio of SAG to scTCR IC₅₀ by linear regression analysis. Similar results were obtained with immobilized mAb F23.2 (data not shown).

Figure 2

2C scTCR alanine substitutions and protein expression. Alanine mutants that can be expressed and properly refolded are indicated. Proteins listed as having an expression defect possessed minimal or no Vβ8-specific mAb activity by ELISA (see Fig. 3).

Figure 3

Reactivity of scTCR alanine mutants with anti-Vβ8 antibodies. The relative reactivity of the mutant scTCR proteins toward the Vβ8-specific antibodies KJ16 (A) and F23.1 (B) were evaluated in a competition ELISA format. Various concentrations of wt and mutant scTCRs were used to inhibit the binding of biotinylated wt-scTCR to each antibody, followed by detection with streptavidin-HRP. The IC₅₀ ratio of mutant to wt by linear regression analysis was used as a measure of antibody reactivity. The average value for each mutant was used as a normalization factor for properly refolded protein.

Figure 4

Competition ELISA of scTCR alanine mutants with SEC3-1A4. Inhibition of SEC3-1A4 binding by representative 2C scTCR alanine mutants. Wt and mutant scTCRs at various concentrations were used to inhibit the binding of biotinylated SAG SEC3-1A4 to wt-scTCR in an ELISA format. Bound biotinylated 1A4 was detected by streptavidin-HRP. The reactivity of each mutant was determined as the ratio of mutant to wt IC₅₀ by linear regression analysis.

Figure 5

Reactivities of scTCR alanine mutants with SEC3-1A4. Reactivities of mutant proteins were determined as IC₅₀ values relative to wt by linear regression analysis. Positive values (above the wt reference line of 0 reactivity) indicate a decrease in binding, and negative values indicate an increase in binding. Error bars represent three or more independent experiments.

Figure 6

Sensorgrams of 2C TCR alanine mutants binding SEC3 variant 1A4. βK66A (A) and βY48A (B) were injected at the indicated concentrations at a flow rate of 15 μl/min over a surface to which SEC3-1A4 (500 RU) had been immobilized.

Figure 7

Energy map of the 2C scTCR SEC3-1A4 binding site. The face of the TCR Vβ domain that recognizes SEC3 is shown. The color scale indicates the fold reduction in binding for each residue upon alanine substitution (or tryptophan substitution of 52βAla): red shows a >10-fold reduction in binding, purple shows a 1.5–9-fold reduction in binding, yellow shows no effect on binding, and green shows an increase in binding. 106βLeu (yellow) is not shown. 54βSer was unable to be tested, as the alanine mutant lacked Vβ8-specific mAb activity. The 2C TCR surface illustrated is from the 2C TCR crystal structure (reference 35; Protein Data Bank [available at http:\\www.rcsb.org/pdb] accession no. 1TCR).

Figure 8

Binding interface of the Vβ8 domain and SEC3. The binding interface of the TCR Vβ8 (blue) and SEC3 (green) is shown. Residues that were subjected to alanine scanning in the TCR and SEC3 (reference 17) are shown. The color scale indicates the fold reduction in binding for each residue: red shows the greatest reduction in binding, purple shows a moderate effect, yellow shows little effect, and green shows an increase in binding. The SEC3 loop colored red indicates the affinity maturation site of the SEC3-A4 variant. The region of CDR2β colored red (51βGly, 52βAla, 53βGly) contributed the greatest energy to the interaction. A view of the CDR2β loop of the TCR shows the topology of the c′′ β strand in relation to the c′ and d strands. Coordinates are derived from the 14.3.d TCR–SEC3 crystal structure (reference 15; Brookhaven Protein Data Bank accession no. 1JCK).

Figure 9

Reactivities of double and triple scTCR alanine mutants with SEC3-1A4. Fold increases of SEC3-1A4 reactivity were calculated based on the relative IC₅₀ of each mutant compared with wi using linear regression analysis. Error bars represent the SD obtained from three separate ELISA experiments.