IL-16 is a proinflammatory cytokine implicated in the pathogenesis of asthma and other conditions characterized by recruitment of CD4+ T cells to sites of disease. It is postulated that CD4 is an IL-16 receptor, although other receptors or coreceptors may exist. Among several known functions, IL-16 is a chemoattractant factor for CD4+ T cells and it inhibits MLR. We previously reported that an oligopeptide corresponding to the 16 C-terminal residues of human IL-16 inhibits chemoattractant activity. To identify functional domains with greater precision, shorter oligonucleotides containing native or mutated C-terminal IL-16 sequences were tested for IL-16 inhibition. Within the 16 C-terminal residues, the minimal peptide RRKS (corresponding to Arg106 to Ser109) was shown to mediate inhibition of IL-16 chemoattractant activity. Inhibition was lost when either arginine was substituted with alanine. Point mutations in IL-16 revealed that Arg107 is critical for chemoattractant activity, but MLR inhibition was unaffected by mutation of Arg107 or even deletion of the C-terminal tail through Arg106. Deletion of 12 or 22 N-terminal residues of IL-16 had no impact on chemoattractant activity, but MLR inhibition was reduced. Deletion of 16 C-terminal plus 12 N-terminal residues abolished both chemoattractant and MLR-inhibitory activity of IL-16. These data indicate that receptor interactions with IL-16 that activate T cell migration are not identical with those required for MLR inhibition, and suggest that both N-terminal and C-terminal domains in IL-16 participate in receptor binding or activation.
Interleukin-16 is a proinflammatory lymphokine with chemoattractant activity for resting CD4+ T lymphocytes. A growing body of published data indicates that IL-16 activates signal transduction and stimulates a variety of biological activities in addition to chemotaxis in CD4+ target cells, including monocytes, eosinophils, and pro-B cells. Among these activities are inhibition of retroviral replication (1, 2, 3), up-regulation of IL-2R and synergy with IL-2 for CD4+ T cell proliferation (4), induction of RAG-1 and RAG-2 expression in CD4+ pro-B cells (5), and transient inhibition of MLR (6). Investigation of certain human diseases and experimental murine models indicates that IL-16 participates in inflammatory conditions characterized by tissue recruitment of CD4+ T lymphocytes and other CD4+ cell types.
The predicted amino acid sequence of IL-16 contains a central PDZ module, and structural studies confirm that IL-16 assumes a core PDZ-like conformation with flexible N-terminal and C-terminal tails of 17 and 14 residues, respectively (7). We previously reported that a synthetic oligopeptide corresponding to the 16 C-terminal amino acids of human IL-16 (Arg106 to Ser121) inhibits the chemoattractant activity of natural and recombinant human or murine IL-16 (8). This peptide partially displaced binding of the anti-CD4 mAb OKT4, suggesting that the peptide functions as a competitive inhibitor for receptor binding, and that the C-terminal tail of IL-16 is a potential binding and signal-inducing domain. Using a series of smaller native sequence and substituted peptides for IL-16 inhibition assays, we now demonstrate that C-terminal oligopeptides as short as four residues can inhibit IL-16 chemoattractant activity. Experiments with mutated rIL-16 constructs reveal that both N-terminal and C-terminal domains are involved in IL-16 functions, and that receptor interactions stimulating T cell motility differ from those required for MLR inhibition.
Materials and Methods
Peptides and Abs
Synthetic oligopeptides corresponding to native or altered C-terminal IL-16 sequences were purchased from Research Genetics (Atlanta, GA). Monoclonal and polyclonal anti-IL-16 Ab against human rIL-16 were produced in our laboratory.
Human PBMC were isolated as described (9, 10, 11) from the blood of healthy volunteers by density centrifugation on Ficoll-Hypaque (Pharmacia, Piscataway, NJ). The mononuclear cell layer was washed with medium 199 (M.A. Bioproducts, Walkersville, MD) supplemented with 0.4% BSA, 25 mM HEPES buffer, 100 U/ml penicillin, and 100 μg/ml streptomycin (M199-HPS). Samples were enriched for T lymphocytes by nylon wool adherence (12). The nonadherent cells were >95% CD3+ by flow cytometry.
Human rIL-16 corresponding to the 121 C-terminal amino biologically active cytokine cleaved from natural pro-IL-16 was produced in Escherichia coli as a polyhistidine fusion protein using the expression vector pET-30 LIC (Novagen, Madison, WI). Following lysis of transformed bacteria, the protein was purified by metal chelation chromatography, and the N-terminal polyhistidine tag was removed by cleavage with enterokinase. The native IL-16 expression vector (pET-30/IL-16121) was used as a template for PCR mutagenesis to create four rIL-16 mutant constructs with progressive 4-aa deletions at the C terminus (C-4 to C-16), as well as deletions of 12 or 22 N-terminal residues. Two double deletion constructs lacking the first 12 or 22 N-terminal residues as well as the last 16 C-terminal residues of IL-16 were also produced. Point mutations in C-terminal residues of rIL-16 were generated by site-directed mutagenesis using the Stratagene Quick Change Kit (Stratagene, La Jolla, CA), according to the manufacturer’s specifications. The point mutations included alanine substitution for Arg106, Arg107, and Arg106 plus Arg107.
Western blot analysis
Native and mutated rIL-16 proteins were subjected to electrophoresis through a 15% SDS-polyacrylamide gel, then electrophoretically transferred to nitrocellulose membranes. The membranes were probed with either polyclonal rabbit anti-rIL-16 or a murine anti-rIL-16 mAb designated clone 17.1. Secondary HRP-conjugated anti-Igs were used at a concentration of 1:5000, and the signal was visualized by chemoluminescence (Pierce, Rockford, IL).
Cell migration was measured using a modified Boyden chemotaxis chamber, as described (9, 10, 11). Cells were suspended (5 × 106 cells/ml) in M199-HPS and loaded into the upper wells, separated by an 8-μm-pore-size nitrocellulose membrane from lower wells. The lower wells were loaded with control buffer or experimental chemoattractant stimuli, with or without inhibitory peptides. After a 4-h incubation at 37°C, the membranes were removed and stained with hematoxylin, dehydrated by sequential washes in ethanol and propanol, then washed in xylene to clarify the filter for cell counting by light microscopy. Cell migration was quantified by counting the number of cells in the filter that had moved beyond a depth of 50 μm in five separate fields in duplicate wells for all conditions. Cell counts were compared with unstimulated control cell migration that was normalized to 100%. Results are expressed as mean percentage of control migration, and the data were analyzed for statistical significance (p < 0.05) by Student’s t test.
Mixed lymphocyte reaction
Stimulator cells for mixed lymphocyte reactions were prepared by incubating PBMC (106/ml) with 25 μg/ml mitomycin C for 30 min. The cells were then washed four times with RPMI 1640 medium supplemented with 25 mM HEPES buffer, 100 U/ml penicillin, and 100 μg/ml streptomycin (RPMI 1640-HPS), then resuspended in RPMI 1640-HPS supplemented with 10% FBS (complete medium) at 106 cells/ml. Responder cells were prepared from an unrelated donor, suspended in complete medium at 106 cells/ml, and preincubated (1 h, 37°C) with control buffer, or with rIL-16 or mutated rIL-16 constructs (10−9–10−11 M). Stimulator cells were then added (1:1) and the cell mixtures were transferred in quadruplicate to 96-well round-bottom plates. Cell cultures were pulsed with [3H]thymidine on day 5, harvested with a Titertek cell harvester, and counted in a Becton Dickinson (Franklin Lakes, NJ) scintillation counter on day 6. Results are expressed as mean percentage of cpm above background ± SEM. Data were analyzed for statistical significance (p < 0.05) by Student’s t test.
Inhibition of IL-16-stimulated T lymphocyte motility by C-terminal oligopeptides
We previously reported that an oligopeptide corresponding to the 16 C-terminal residues of IL-16 inhibited the chemoattractant activity of human and murine IL-16 (8). A series of smaller oligopeptides derived from that sequence were prepared to further define the critical residues within this region of IL-16 (Fig. 1). Normal human T lymphocytes were stimulated with rIL-16, and motility was measured using a modified Boyden chemotaxis chamber assay. Cells were stimulated with IL-16 (10−9–10−11 M) in the presence or absence of two 8-mer peptides corresponding to amino acids Arg106 to Lys113, and Glu114 to Ser121 of IL-16. As shown in Fig. 2 A, only the Arg106 to Lys113 peptide inhibited IL-16 in this assay. The 6-mer RRKSLQ also inhibited IL-16-stimulated T cell migration, but a scrambled peptide containing the same residues in a randomly chosen sequence demonstrated no inhibitory activity. To further define the residues mediating inhibition, the eight-residue sequence from Arg106 to Lys113 was divided into RRKS and LQSK. Only RRKS inhibited IL-16 chemoattractant activity. These data demonstrate that IL-16 chemoattractant activity can be effectively inhibited by a four-residue peptide and indicate that the inhibitory activity of the 16-mer Arg106 to Ser121 peptide is entirely contributed by the first four residues RRKS.
The contribution of individual residues within RRKS was analyzed by alanine scanning (Fig. 2,B). Substitution of either Arg106 or Arg107 was associated with loss of inhibitory activity against IL-16-induced chemotaxis. In contrast, the peptides RRAS and RRKA inhibited IL-16 as effectively as the native RRKS. These data demonstrate that Arg106 and Arg107 are required for peptide inhibition and suggest that these corresponding residues in IL-16 may be functionally important. Inhibition by RRKS of chemotaxis in response to CD4 stimulation is specific for IL-16. Cell migration in response to HIV-1 gp120 (13), or divalent anti-CD4 mAb Leu3a was not blocked by this peptide (Fig. 3).
Chemoattractant activity of IL-16 deletion mutants
The ability of RRKS-containing peptides to block IL-16-stimulated T lymphocyte migration suggests that the corresponding region in IL-16 is functionally important. To test this hypothesis, IL-16 constructs were created with progressive deletions of four C-terminal amino acids from C-4 through C-16 (Fig. 4). The C-12 construct terminates at Ser108, retaining the RRKS motif. The C-16 construct terminates at Ile105, deleting RRKS and succeeding downstream residues. Chemotaxis assays demonstrate that C-12 is as active as native rIL-16, while the C-16 deletion completely eliminates chemoattractant activity (Fig. 5,A). In similar experiments, C-4 and C-8 deletion constructs demonstrated chemoattractant activity comparable with native rIL-16 (data not shown). The C-terminal deletion results are consistent with the peptide studies, suggesting that residues within the RRKS motif are required for IL-16-stimulated chemoattractant activity. Two additional constructs with deletion of 12 or 22 N-terminal amino acids were also tested to identify any contribution of N-terminal structures to chemotactic signaling. Both the N-12 and N-22 deletion mutants demonstrated chemoattractant activity comparable with native IL-16 (Fig. 5 A). Thus, receptor interactions activating cell motility appear to involve only the C-terminal domain of IL-16.
Contribution of specific C-terminal residues to IL-16 chemoattractant activity
To determine the contribution of individual residues within the RRKS motif to chemoattractant signaling, and to test the activity of IL-16 mutants with minimal structural alterations, a series of point mutations using alanine substitution were generated (Fig. 4). Replacement of Arg107 alone, or Arg106 plus Arg107, completely abrogated chemoattractant activity of the recombinant protein (Fig. 5 B). In contrast, substitution of Arg106 alone retained full activity. These data indicate a critical role for Arg107 in IL-16-stimulated CD4+ T cell migration. The identical pattern of motile responses was observed using a different IL-16-responsive cell type, human peripheral blood monocytes (data not shown).
Aggregation of native IL-16 monomers is required for chemoattractant activity, presumably mediated by receptor cross-linking (8, 10). A mutation could inactivate IL-16 if it disrupted multimer formation. However, aggregation was not disrupted in the point or deletion mutants used in our studies, because all of these constructs formed multimers similar to native IL-16, as assessed by HPLC (data not shown). These observations suggest that mutation of Arg107 directly interferes with CD4 binding or activation by IL-16. Structural constraints in the interaction of the full-length folded rIL-16 protein with its receptor may account for the precise involvement of Arg107, whereas both synthetic peptides ARKS and RAKS can function as IL-16 inhibitors.
Western analysis of IL-16 mutant proteins
Using rIL-16 as an immunogen, we previously generated and characterized rabbit polyclonal anti-IL-16 Ab, as well as a murine monoclonal anti-IL-16 (clone 17.1). This mAb was isolated by screening hybridoma supernatants for neutralization of IL-16 chemoattractant activity. Western blot analysis was performed with native rIL-16 and the C-terminal deletion mutants (Fig. 6), using either the polyclonal Ab or the mAb for detection. As expected, the polyclonal Ab recognized native rIL-16 and all of the deletion mutants. The mAb 17.1 detected native rIL-16 and the deletion mutants lacking 4, 8, or 12 C-terminal residues, as well as the N-terminal deletion mutants (data not shown). However, mAb 17.1 failed to bind to C-16. The epitope for the neutralizing anti-IL-16 mAb 17.1 therefore maps to the identical domain shown to be required for IL-16 chemoattractant activity by peptide inhibition and mutation experiments.
Inhibition of the mixed lymphocyte reaction
To determine whether other biological activities of IL-16 are mediated by the C-terminal domain, we tested the capacity of native and mutated rIL-16 constructs to inhibit the one-way MLR. Responder T lymphocytes were pretreated with rIL-16 or control buffer, then cultured with mitomycin C-treated stimulator PBMC from an unrelated donor. Pretreatment with 10−8 M native rIL-16 reduced thymidine incorporation on day 6 by nearly 70%, compared with untreated cells. Surprisingly, the C-terminal point mutations that lose chemoattractant activity retain full capacity to inhibit the MLR (Fig. 7). We next tested the activity of IL-16 deletion mutants in the MLR inhibition assay. The C-16 deletion was nearly as active as native rIL-16, with a ∼1 log shift of the dose response (Fig. 8). Deletion of 12 or 22 N-terminal residues resulted in a similar pattern as C-16; MLR inhibition was reduced, but not eliminated. In contrast, constructs that combine the C-16 deletion with N-12 or N-22 lose all capacity to inhibit the MLR. These data demonstrate significant differences in the structural requirements of IL-16 for MLR inhibition as compared with the induction of T cell motility. The results also suggest that both N-terminal and C-terminal domains of IL-16 are involved in receptor binding and activation.
Previous investigations revealed that an oligopeptide comprising the C-terminal IL-16 amino acids Arg106 to Ser121 specifically inhibits IL-16 chemoattractant activity (8). Based on that finding, we initiated the current studies to more precisely identify the functionally important C-terminal domain. A series of small peptides were tested for their ability to antagonize IL-16 chemoattractant activity. Within the 16-aa domain, only peptides containing or limited to the RRKS motif possessed inhibitory activity. Alanine substitution of either arginine in RRKS abrogated the inhibitory activity. The importance of this site was further supported by experiments testing the chemoattractant activity of rIL-16 mutants with serial C-terminal deletions or point mutations of these bases to alanine. Deletion of Arg107 through Ser121, or replacement of Arg107 alone, completely eliminated the chemoattractant activity of rIL-16. Furthermore, the epitope of an anti-IL-16 mAb that neutralizes chemoattractant activity was mapped to the identical location. All of these mutated rIL-16 constructs retained the capacity for noncovalent multimer formation, which was previously shown to be a requirement for bioactivity. We conclude that Arg107 in native IL-16 is essential for the induction of CD4+ cell motility by binding or activation of the receptor. In this regard, Arg107 is conserved in the IL-16 sequence of the mouse and in six primates species (14). In the mouse, Lys108 is replaced by threonine, while Arg106 is replaced by lysine in two primate species studied. All of these IL-16 homologues induce human T cell motility.
Contrary to our expectation, the IL-16 C-terminal point mutations that lost chemoattractant activity fully retained a different biological activity of IL-16, inhibition of the MLR. This suggests that another domain of IL-16 could participate in receptor binding or signal induction. Deletion of 12 or 22 N-terminal residues had no effect on chemoattractant activity, but shifted the dose response for MLR inhibition by ∼1 log. Deletion of 16 C-terminal residues had a similar effect of reducing, but not eliminating, MLR inhibition. Both chemoattractant activity and MLR inhibition were completely lost with combined deletion of 12 or 22 N-terminal residues plus 16 C-terminal residues. These data suggest that sequences within both the N-terminal and C-terminal tails of IL-16 interact with CD4. This is consistent with recent studies from our laboratory that indicate two discrete touch points on CD4 are involved in IL-16 binding and activation of T cell motility (27).
An NMR structure of human IL-16 reported by Muhlhahn et al. (7) confirmed that the folded core is a PDZ-like domain characterized by an atypically small GLGF cleft that is additionally blocked by a tryptophan side chain in its center. The N-terminal and C-terminal tails extending from the hydrophobic core structure are flexible. PDZ domains bind to C-terminal residues of target proteins, including transmembrane receptors or ion channels, and may also participate in heterodimeric binding with PDZ domains of other proteins (15). Binding of IL-16 to peptides representing common PDZ-binding sequences could not be demonstrated, so it remains uncertain whether the PDZ structure of IL-16 mediates binding interactions similar to PDZ domains of other proteins. The functional significance of this domain in IL-16, either in the intracellular or extracellular environment, is presently unknown. In their report, Muhlhahn et al. also constructed two truncated forms of IL-16 extending from Ser15 to Glu114, and from Ala19 to Ser121, which retained chemoattractant activity. Our data presented in this study are consistent with that finding. These results do not exclude an extracellular function for the GLGF cleft of IL-16, but indicate that point mutations far removed from that site, and unlikely to perturb the core PDZ structure, are sufficient to destroy its bioactivity.
The induction of cell migration is known to require signal transduction initiated by receptor cross-linking. In this regard, several anti-CD4 Ab possess CD4+ T cell chemoattractant activity, whereas monomeric F(ab′)2 fragments do not stimulate motility (F(ab′)2 fragments of OKT4 inhibit IL-16-stimulated motility). In contrast, inhibition of the MLR might be caused by steric hindrance alone. Physical studies indicate that CD4 forms homodimers in solution, based on interactions between structures in the fourth Ig-like domain (D4 (16)). It is postulated that CD4 dimerization on the cell surface, in the context of MHC class II binding, results in the costimulatory signaling of CD4 with the TCR. In this regard, Satoh et al. (17) reported that synthetic oligopeptides based on D4 sequences could inhibit the MLR, presumably by monomeric binding to the corresponding site on CD4, thereby preventing homodimer formation.
The region of CD4 that interacts with IL-16 has recently been demonstrated to lie within the D4 domain, overlapping structures involved in CD4 dimer formation (27). Thus, IL-16 binding to this site on CD4 would be expected to inhibit the MLR in a similar manner as the D4-sequence oligopeptides. Because N-terminal deletions of IL-16 affected MLR inhibition but not chemotaxis, it appears that this domain might interfere with CD4 dimerization by a steric mechanism. We predict that this interaction is weak, and requires additional binding interactions of the C-terminal tail of IL-16 with CD4 for full stability. This model provides an explanation for the reduction in MLR inhibition with C-terminal deletion and the complete loss of MLR inhibition with deletion of both IL-16 tails.
Abundant evidence links IL-16 expression in the bronchial epithelium to the induction and maintenance of airways inflammation in asthma (18, 19, 20, 21, 22). Other inflammatory conditions in which IL-16 is implicated in the recruitment of CD4+ leukocytes include rheumatoid arthritis, systemic lupus erythematosus, inflammatory bowel disease, and multiple sclerosis (23, 24, 25, 26). Identifying functional domains on IL-16 and developing potent antagonists will help to establish the role that IL-16 plays in normal and pathological immune responses. This knowledge might also provide a rational basis for the design of therapeutic antagonists useful for diseases in which overexpression of IL-16 may contribute to pathogenesis.
We thank Gregg Fine for technical assistance, and Sue Kim and Dr. Yujun Zhang for valuable advice on the design and implementation of these studies.
This work was supported by National Institutes of Health Grant HL32802. W.W.C. is a Career Investigator of the American Lung Association.