Sequential tyrosine sulfation of CXCR4 by tyrosylprotein sulfotransferases

Abstract

CXC-chemokine receptor 4 (CXCR4) is a G protein-coupled receptor for stromal cell-derived factor-1 (SDF-1/CXCL12). SDF-1-induced CXCR4 signaling is indispensable for embryonic development and crucial for immune cell homing and has been implicated in metastasis of numerous types of cancer. CXCR4 also serves as the major coreceptor for cellular entry of T-cell line-tropic (X4) HIV-1 strains. Tyrosine residues in the N-terminal tail of CXCR4, which are post-translationally sulfated, are implicated in the high-affinity binding of SDF-1 to CXCR4. However, the specific roles of three potential tyrosine sulfation sites are not well understood. We investigated the pattern and sequence of CXCR4 sulfation by using recombinant human tyrosylprotein sulfotransferases TPST-1 and TPST-2 to modify a peptide that corresponds to amino acids 1-38 of the receptor (CXCR4 1-38). We analyzed the reaction products with a combination of reversed-phase HPLC, proteolytic cleavage, and mass spectrometry. We found that CXCR4 1-38 is sulfated efficiently by both TPST enzymes, leading to a final product with three sulfotyrosine residues. Sulfates were added stepwise to the peptide, producing specific intermediates with one or two sulfotyrosines. The pattern of sulfation in these intermediates indicates that with both enzymes Tyr-21 is sulfated first, followed by Tyr-12 or Tyr-7. Using heteronuclear NMR spectroscopy, we demonstrated that the SDF-1 binding affinity of CXCR4 1-38 increases with the number of sulfotyrosines present, which suggests a potential physiological role for sulfation of all three sites in the N-terminus of CXCR4. These results provide a structural basis for understanding the role of post-translational tyrosine sulfation in SDF-1-induced CXCR4 signaling.

FIG. 1. Schematic Representation of Human CXCR4 with Potential Tyrosine Sulfation Sites

The seven transmembrane helices are shown as cylinders. Connecting extracellular and cytosolic loops as well as N- and C-terminal regions are depicted as black lines. Amino acid residues in the CXCR4 N-terminus are represented in single letter code. The potentially sulfated tyrosine residues at positions 7, 12, and 21 are highlighted in red, and the acidic amino acid residues Asp-10, Asp-20, Asp-22, Glu-2, Glu-14, Glu-15, Glu-26, Glu-31, and Glu-32 are shown in green. Cys-28 (dashed circle) forms a putative disulfide bond (dashed line) with Cys-274 in the third extracellular loop of CXCR4.

FIG. 2. TPST Catalyzed In Vitro Sulfation of CXCR4 1-38

(A) Schematic representation of the sequence of peptide CXCR4 1-38. The color coding is the same as in Fig. 1. Ala-28 which replaces Cys-28 of the CXCR4 sequence and the two N-terminal Gly and Ser residues which are not part of the CXCR4 sequence are marked by dashed circles. (B) RP-HPLC analysis of TPST-1 catalyzed CXCR4 1-38 sulfation. Peptide CXCR4 1-38 (50 μM, ≈0.23 mg/ml) was incubated with TPST-1 (40 μg/ml) and in the presence of the sulfotransfer cosubstrate PAPS (400 μM). After 12 h or 168 h at 16°C, 50 μl aliquots were analyzed by RP-HPLC. In negative-control experiments (168 h incubation time) either TPST-1 (No TPST) or PAPS (No PAPS) was omitted. As a standard, unreacted CXCR4 1-38 substrate was analyzed. Peaks were labeled a-e in increasing order of hydrophilicity. To correct for day-to-day variations in elution times caused by slight variations in the eluent compositions, elution times were normalized to the average elution time observed for an unsulfated CXCR4 1-38 standard. (C) RP-HPLC analysis of TPST-2 catalyzed CXCR4 1-38 sulfation. Peptide CXCR4 1-38 (50 μM, ≈0.23 mg/ml) was incubated with TPST-2 (40 μg/ml) as described for TPST-1 in (B). (D) Time course of TPST-1 catalyzed CXCR4 1-38 sulfation. CXCR4 1-38 (50 μM, ≈0.23 mg/ml) was incubated at 16°C with PAPS (400 μM) and TPST-1 (40 μg/ml). At the indicated time points, 50 μl aliquots were analyzed by RP-HPLC and relative amounts for the different peptide species (a - e) were calculated from the peak areas at 220 nm. (E) Time course of TPST-2 catalyzed CXCR4 1-38 sulfation. Analysis was performed as described in (D) with TPST-2 (40 μg/ml) as the sulfotransfer enzyme. (F) Total sulfate incorporation into CXCR4 1-38 by TPST-1 versus TPST-2 was calculated as the sum of the relative amounts of all peptide species in (D) and (E) multiplied by the number of sulfotyrosines present in each peptide species (see Table 1). A value of 100% total sulfation corresponds to a net incorporation of one equivalent of sulfate per equivalent of CXCR4 1-38; the maximum total sulfate incorporation would be 300%, corresponding to complete sulfation of all three sulfation sites present in the peptide.

FIG. 3. Identification of Sulfation Sites in CXCR4 1-38 Sulfation Products

Peak fractions a - e from RP-HPLC were analyzed by (A) negative ion mode MALDI-TOF MS and (B) negative ion mode ESI MS. Peaks are labeled according to the number of identified sulfotyrosines as shown in Table 1; satellite peaks corresponding to sodium adducts and oxidization of Met residues are not labeled. In MALDI-TOF MS (A) the dominant peaks in each spectrum correspond to the [M-1H]^1- and [M-2H]^2- ions. Minor peaks indicate the loss of one or two SO₃ groups during MS analysis. In ESI MS (B) the two dominant peaks in each spectrum correspond to the [M-3H]^3- and [M-4H]^4- ions. (C) Schematic representation of peptide fragments generated by Asp-N cleavage that were used to localize sulfation sites in CXCR4 1-38 sulfation products by MALDI-TOF MS analysis. The color coding is the same as in Fig. 1, and the endoproteinase Asp-N cleavage sites are indicated by red arrows. The peptide fragments are numbered according to the CXCR4 sequence excluding the N-terminal Gly and Ser extension. Fragments 1-19, 1-21, and 10-21 are intermediates of the cleavage reaction, whereas fragments 1-9, 10-19, and 22-38 emerged as the final cleavage products. (D) Reaction scheme for the sequential sulfation of CXCR4 1-38 by TPST-1 and TPST-2. CXCR4 1-38 species that arise from sulfation of non-sulfated CXCR4 1-38 (none) with either TPST-1 or TPST-2 are represented by their sulfotyrosine positions. The corresponding HPLC peak assignments are given in parentheses.

FIG. 4. Effect of CXCR4 1-38 Sulfation on the SDF-1α Binding Affinity

Apparent binding affinities were obtained by ¹⁵N-¹H HSQC spectroscopy titrating ¹⁵N/¹H-labeled SDF-1α with unsulfated (P38), Y21-monosulfated (sY21 P38), and fully sulfated (sY21/12/7 P38) CXCR4 1-38 peptides. (A) Overlaid portions of ¹⁵N-¹H HSQC spectra of SDF-1α showing residue V49 during titration with P38, sY21 P38 and sY21/12/7 P38. SDF-1α to peptide ratios are 1:0 for black peaks, 1:0.5 for gray peaks, 1:1 for light green (P38), light red (sY21 P38) or light blue (sY21/12/7 P38), and 1:2 for green (P38), red (sY21 P38) and blue (sY21/12/7 P38) peaks. Only titration of SDF-1α with sY21/12/7 P38 resulted in saturation of SDF-1α chemical shift perturbations at a SDF-1α to peptide molar ratio of 1:1. (B) Normalized ¹H/¹⁵N chemical shift perturbations for representative residue V49 of SDF-1α are plotted against peptide concentration for P38, sY21 P38 (data from ref. (23)), and sY21/12/7 P38 (this study). For ease of comparison maximum peptide induced chemical shift perturbation for each titration was normalized to a value of 1. (C) Non-linear curve fitting using titration data for SDF-1α residues 23, 25, 29, 31, 40-42, 47-51, 62, and 65-67 yielded apparent K_d values of 4.5 ± 2.2 μM for P38, 1.3 ± 0.5 μM for sY21 P38, and 0.2 ± 0.2 μM for sY21/12/7 P38.