Cleavage of transaldolase by granzyme B causes the loss of enzymatic activity with retention of antigenicity for multiple sclerosis patients

Abstract

Multiple sclerosis (MS) is an autoimmune demyelinating disease of the CNS resulting from a progressive loss of oligodendrocytes. Transaldolase (TAL) is expressed at selectively high levels in oligodendrocytes of the brain, and postmortem sections show concurrent loss of myelin basic protein and TAL from sites of demyelination. Infiltrating CD8(+) CTLs are thought to play a key role in oligodendrocyte cell death. Cleavage by granzyme B (GrB) is predictive for autoantigenicity of self-proteins, thereby further implicating CTL-induced death in the initiation and propagation of autoimmunity. The precursor frequency and CTL activity of HLA-A2-restricted TAL 168-176-specific CD8(+) T cells is increased in MS patients. In this paper, we show that TAL, but not myelin basic protein, is specifically cleaved by human GrB. The recognition site of GrB that resulted in the cleavage of a dominant TAL fragment was mapped to a VVAD motif at aa residue 27 by N-terminal sequencing and confirmed by site-directed mutagenesis. The major C-terminal GrB cleavage product, residues 28-337, had no enzymatic activity but retained the antigenicity of full-length TAL, effectively stimulating the proliferation and CTL activity of PBMCs and of CD8(+) T cell lines from patients with MS. Sera of MS patients exhibited similar binding affinity to wild-type and GrB-cleaved TAL. Because GrB mediates the killing of target cells and cleavage by GrB is predictive of autoantigen status of self proteins, GrB-cleaved TAL-specific T cell-mediated cytotoxicity may contribute to the progressive destruction of oligodendrocytes in patients with MS.

FIGURE 1

Cleavage of TAL by GrB. A, SDS-PAGE analysis of TAL and MBP exposed to GrB. TAL (38 kDa)and MBP (17.5 kDa) were incubated for 1 h at 37°C with 150 nM purified GrB. Reaction products were loaded into 12% Tris-glycine polyacrylamide gel, separated by SDS-PAGE, and stained with Coomassie brilliant blue. B, Western blot analysis of GrB-digested TAL electroblotted onto nitrocellulose and detected with anti-TAL Ab 169 (22). C, Western blot analysis of rTAL (rTAL, 100 ng/lane) and native TAL in lysates of MO3.13 cells (40 μg/lane) following incubation without (−) or with GrB (+) for 1 h at 37°C. The 70-kDa components of U1 small nuclear ribonucleoprotein and vinculin were detected as positive and negative controls, respectively, for monitoring GrB cleavage. D, Western blot analysis of native TAL in lysates of MO3.13 cells (40 μg/lane) following incubation without (−) or with (+) GrB for 1 h at 37°C. The 70-kDa components of U1 small nuclear ribonucleoprotein and actin were detected as positive and negative controls, respectively, for monitoring GrB cleavage. The TAL/actin ratio was determined with automated densitometry (48) and normalized to lysates incubated without GrB (1.0). Results reflect four independent digestions of MO3.13 lysates with GrB.

FIGURE 2

Mapping of GrB cleavage motifs in human TAL. A, Locations of GrB recognition motifs are underlined, and sites of cleavage are indicated by vertical arrows. CTL epitope residues 168–176 are italicized. B, Digestion of TAL-WT and mutated TAL carrying an aspartic-to-alanine substitution at aa position 27 (TAL-D27A). A total of 100 ng of recombinant protein was incubated without (−) or with (+) 150 nM GrB for 1 h at 37°C and detected by anti-TAL Ab 169, using Western blot.

FIGURE 3

Western blot analysis of TAL-WT (recombinant clone 1425) and the 310-aa C-terminal GrB cleavage product (TAL-GrB, recombinant clone 8427) and TAL protein expression in Hmy A3.1, Hmy A2.1, MO3.13, MO3.13/A2, and EBV-transformed B cells of MS patient JAB (JAB/EBV). A total of 100 ng of TAL-WT or TAL-GrB recombinant protein and 40 ug of whole-cell lysates was analyzed per lane and developed by simultaneous incubation with Abs to TAL and human β-actin, as earlier described (15).

FIGURE 4

Proliferative responses to TAL-WT (clone 1425) and TAL-GrB (clone 8427) by PBMCs from HLA-A2⁺ MS patient JAB and HLA-A2⁺ control donor BN. A total of 10⁵ cells per well were stimulated for 72 h with 5 μg/ml TAL-WT, TAL-GrB, negative control Ag BSA (left y-axis) or polyclonal stimulator Con A (right y-axis). Data show mean ± SD of SIs of six parallel cultures. *p < 0.0001 reflects comparison of TAL-WT–stimulated or TAL-GrB–stimulated cells with BSA-stimulated cells of MS patient JAB. **p = 0.021; ***p = 0.010.

FIGURE 5

Cytotoxic activity of TAL-WT–stimulated and TAL-GrB–stimulated PBMCs from HLA-A2⁺ MS patients (JAB, PET, RYD, SOL), HLA-A2⁻ MS patients (PCA, DET, GIR), and HLA-A2⁺ HCs (RUS, NIL) against MO3.13/A2 (HLA-A2–transfected) and control MO3.13, as well as Hmy-A2.1 (HLA-A2–transfected) and control Hmy-A3.1 (HLA-A3–transfected) target cells. Target cells were pulse labeled with or without TALpep. A total of 10⁶/ml PBMCs were stimulated with and without 5 μg/ml TAL-WT, TAL-GrB, or BSA for 7 d and added to target cells at 10:1 effector/target ratio. Values significantly exceeding control levels are indicated by brackets. *p < 0.005; **p < 0.0005; ***p < 0.0001.

FIGURE 6

Cytotoxicity by TAL-specific TCLs from MS patients was assessed against MO3.13/A2 (HLA-A2–transfected) and control MO3.13, as well as Hmy-A2.1 (HLA-A2–transfected) and control Hmy-A3.1 (HLA-A3–transfected) target cells. JTE and KTE TCLs were generated from HLA-A*0201–positive MS patients and maintained by weekly stimulation with autologous EBV-transformed B cells pulsed with 5 μg/ml rTAL (49). Values significantly exceeding control levels are indicated by brackets. *p < 0.005; **p < 0.0005; ***p < 0.0001.

FIGURE 7

Testing of binding affinity to TAL-WT and TAL-GrB by sera of 12 MS patients and 6 HCs, using ELISA. Binding affinity of MS sera exceeded the binding affinity of control sera to both TAL-WT (p = 2.7 × 10¹³) and TAL-GrB (p = 9.4 × 10¹¹), using a two-tailed t test.