Advertisement

Neutrophil gelatinase B potentiates interleukin-8 tenfold by aminoterminal processing, whereas it degrades CTAP-III, PF-4, and GRO-α and leaves RANTES and MCP-2 intact

Philippe E. Van den Steen, Paul Proost, Anja Wuyts, Jo Van Damme and Ghislain Opdenakker

Article Figures & Data

Figures

  • Fig. 1.

    Purification and SDS-PAGE analysis of human neutrophil gelatinase B.

    Natural gelatinase B, purified from human neutrophils, occurs in 3 different forms: monomers, disulfide-linked homodimers, and NGAL–gelatinase B complexes.32 Nonreducing SDS-PAGE and Coomassie blue staining before (lane 2) and after (lane 3) removal of the NGAL–gelatinase B complex allows visualization of the positions of the monomers, the dimers, and the NGAL–gelatinase B complexes. After chemical reduction with β-mercaptoethanol, the dimer was dissociated into monomers in the electrophoretically pure preparation (lane 4). Relative molecular masses of the monomer, heterodimer, and homodimer were estimated to be 91, 125 and greater than 200 kd by comparison with the molecular mass markers (lane 1).

  • Fig. 2.

    Processing of the chemokines CTAP-III, MCP-2, PF-4, RANTES, and GRO-α by activated neutrophil gelatinase B.

    Different chemokines were incubated with activated gelatinase B and subsequently analyzed by SDS-PAGE and silver staining. (A) Recombinant human MCP-2 was not affected by treatment with gelatinase B. Natural human CTAP-III and PF-4 were slowly degraded by gelatinase B. Degradation of PF-4 was inhibited by EDTA and o-phenantrolin (PHEN) and was not observed after incubation with stromelysin-1 alone. (B) Recombinant RANTES was not affected by incubation with gelatinase B, whereas natural GRO-α was slowly degraded. Degradation of GRO-α was inhibited by EDTA and PHEN and was not observed after incubation with stromelysin-1. S indicates relative molecular mass standard; 0, no incubation; −, incubation with stromelysin-1 only; +, incubation with activated gelatinase B. Respective inhibitors are indicated at the top of each lane.

  • Fig. 3.

    Degradation of CTAP-III by activated neutrophil gelatinase B.

    CTAP-III was incubated with activated gelatinase B in the presence or absence of the metalloproteinase inhibitors EDTA,o-phenantrolin (PHEN), and TIMP-1. The 3 inhibitors separately inhibited the degradation completely. Stromelysin-1 alone was not able to degrade CTAP-III. S indicates molecular mass standard. The symbols + and − in the upper line indicate the presence or absence of the indicated inhibitors, whereas the symbols 0, −, and + in the line underneath indicate no incubation, incubation with stromelysin-1 alone, and incubation with stromelysin-1–activated gelatinase B, respectively.

  • Fig. 4.

    Determination of the cleavage sites in CTAP-III by gelatinase B.

    After digestion of CTAP-III with gelatinase B, the reaction mixture was subjected to reverse-phase HPLC on a C8 column to separate the cleavage products into different fractions (top). Peptides were identified by aminoterminal sequencing and mass spectrometry analysis (bottom). Sequences in bold letters were determined by tandem MS/MS, and underlined sequences were determined by Edman degradation. Intact disulfide bridges (as determined by MS analysis) are indicated by lines between the cysteine pairs. The experimentally determined molecular mass of the fragments is compared with the calculated theoretical mass data. Positions of the cleavage sites are indicated by asterisks in the sequence of CTAP-III (lowest part of bottom panel). nd, not detectable.

  • Fig. 5.

    Specific conversion of IL-8(1-77) to IL-8(7-77) by activated neutrophil gelatinase B.

    Natural IL-8 occurs as 2 protein variants, IL-8(1-77) and IL-8(6-77), which are separable by SDS-PAGE. Activated gelatinase B processes natural (A) and recombinant (B) human IL-8(1-77) to a shorter form, identified as IL-8(7-77) (Table 1). To illustrate that this conversion was by activated gelatinase B, various inhibitors were tested for their ability to inhibit the chemokine conversion. The metalloproteinase inhibitors EDTA, o-phenantrolin (PHEN), and TIMP-1 and the gelatinase B–inhibiting monoclonal antibody REGA-3G12 inhibited this conversion completely, but no effect was visible with the serine protease inhibitors pefabloc and aprotinin or with the thiolprotease inhibitor E64. Progelatinase B or stromelysin-1 alone was unable to process IL-8. In B, the first 2 lanes (with the controls of the EDTA inhibition) contain twice the amount of IL-8 as the other lanes. S indicates relative molecular mass standard. The + and − symbols in the upper line indicate the presence or absence of the indicated inhibitors, whereas symbols 0, −, +, or P in the lower line indicate no incubation, incubation with stromelysin-1 alone, incubation with activated gelatinase B, or incubation with progelatinase B, respectively.

  • Fig. 6.

    Mass spectrometry analysis of IL-8(1-77) and IL-8(7-77).

    IL-8(1-77) (A) and IL-8(7-77) (B) were desalted and subjected to electrospray mass spectrometry analysis to exclude the possibility of carboxyterminal cleavage by gelatinase B. Unprocessed (m/z) and charge-deconvoluted (m) spectra are shown. Theoretical masses of IL-8(1-77) and IL-8(7-77) are 8922.5 d and 8298.7 d, respectively. In the left panels, the m/z values for the differently charged ions are indicated, as are the number of protons (H+) they carry.

  • Fig. 7.

    Time course of the degradation of IL-8 and CTAP-III by activated gelatinase B.

    Recombinant IL-8 (A) and natural CTAP-III (B) (both at 4 μmol/L) were incubated with activated neutrophil gelatinase B (0.4 μmol/L) at 37°C. A control sample was incubated with activated stromelysin-1 (0.004 μmol/L). Samples were taken at the indicated time intervals and analyzed by SDS-PAGE and silver staining. S indicates relative molecular mass standard; the symbols − and + indicate incubation with stromelysin-1 alone and with activated gelatinase B, respectively.

  • Fig. 8.

    Binding to and intracellular Ca++-mobilizing activity of IL-8(1-77) and IL-8(7-77) in neutrophils.

    (A) Relative binding properties of IL-8(1-77) and IL-8(7-77) to human neutrophils were compared. Neutrophils were incubated with125I–IL-8(6-77) and various concentrations of unlabeled IL-8(1-77) and IL-8(7-77). The decrease in cell-bound radioactivity was used as a measure for competition by the unlabeled IL-8 forms and thus for their affinity to the cells. Mean and SEM (n = 4) are indicated. (B) Intracellular Ca++-mobilizing activities of IL-8(1-77) and IL-8(7-77) were determined by measurement of the fluorescence of the dye fura-2 during stimulation with the IL-8 forms. One representative experiment of 2 is shown. Successive one-third dilutions of both IL-8 forms were analyzed until no increase of [Ca++]i could be measured. The detection limit in [Ca++]i increase was 15 nmol/L. Stimulation with IL-8(1-77) at 0.17 nmol/L and IL-8(7-77) at 0.006 nmol/L resulted in a [Ca++]i increase below the detection limit.

  • Fig. 9.

    Gelatinase B release and chemotactic activities of IL-8(1-77) and IL-8(7-77).

    (A) Purified granulocytes were incubated with various concentrations of IL-8(1-77) and IL-8(7-77) for 30 minutes to induce the release of gelatinase B. Gelatinase B in the culture fluid of stimulated cells was analyzed by substrate zymography, quantified by scanning densitometry, and expressed in scanning units after the subtraction of background levels. Results are indicated as the mean (±SEM) of 3 independent experiments with neutrophil preparations from different donors. (B) The chemotactic activity of IL-8(1-77) and IL-8(7-77) for neutrophils was compared in a modified Boyden chemotaxis chamber. Chemotactic indices are shown as the mean and the standard errors of mean of 4 independent experiments with cell preparations from different donors.

  • Fig. 10.

    Binding of IL-8(1-77) and IL-8(7-77) to CXCR1- and CXCR2-transfected cell lines.

    IL-8(1-77) and IL-8(7-77) were compared for their ability to compete with 125I–IL-8(6-77) for binding (see legend to Figure 8) to CXCR1- and CXCR2-transfected HEK cell lines (A and B, respectively). Mean and SEM (n = 4) are indicated.

  • Fig. 11.

    Intracellular Ca++-mobilizing activity of IL-8(1-77) and IL-8(7-77) in CXCR1- and CXCR2-transfected cell lines.

    Intracellular Ca++-mobilizing capacities of IL-8(1-77) and IL-8(7-77) were compared on cell lines transfected with CXCR1 and CXCR2. In both cell lines, IL-8(7-77) was more active than IL-8(1-77), but this difference was more pronounced in CXCR1- (A) than in CXCR2- (B) transfected cell lines (P < .05). One representative experiment of 3 is shown. Successive one-third dilutions of both IL-8 forms were analyzed until no increase in [Ca++]i could be measured. Detection limits in [Ca++]i increase were 50 and 20 nmol/L for CXCR1 and CXCR2 transfectants, respectively. For each IL-8 variant and CXCR interaction, dosages giving a response below the detection limit of the [Ca++]i increase were included.

Tables

  • Table 1.

    Digestion of IL-8 by gelatinase B and aminoterminal sequence analysis

    Aminoterminal sequencesBefore digestion (%)*After digestion (%)*
    Natural IL-8
     IL-8(-2-77)EGAVLPR …100
     IL-8(1-77)AVLPRSA …350
     IL-8(6-77)SAKELRCQC …5560
     IL-8(7-77)AKELRCQC …040
    Recombinant IL-8
     IL-8(1-77)AVLPRSA …1000
     IL-8(7-77)AKELRCQC …0100
    • *  Relative amounts were determined by aminoterminal sequence analysis.