Changing of Val47 to Asp47 or to Lys47 enhances the immunomodulatory activity of the human Interleukin-l peptide 47–55 Μ BRUHASPΑΤΗΥ* and S K KAR Centre for Biotechnology, Jawaharlal Nehru University, New Delhi 110 067, India MS received 13 May 1996; revised 10 October 1996 Abstract. The 47–55 domain of the maturehumanInterleukin- was predicted to be exposed by our computational analysis and confirmed to be so by comparing with X-ray crystallographic as well as nuclear magnetic resonance (NMR) spectroscopic data. Four peptides representing fully or part of this domain with sequences 47–55, 41–61, 45–61and 50–66 were synthesized and tested for their ability to modulate in vivo, the humoral immune response of Balb/c mice to Shigella dysenteriae 116 kDa antigen(s). The smallest immunomodulatory peptide amongst them was found to be the nonapeptide 47–55. To ascertain the structurefunction relationships of this 47-55 peptide, various mutant peptides were synthesized and tested for IL-1ß like activity in vivo. Change of Val47 to Asp47 or to Lys47 enhanced its immunomodulatory activity significantly while the change of Gly49 to Asp49 or Glu50 to Ile50 or Asp54 to Ile54 had no such effect. The peptides 47–55 and its mutants were first tested for their ability to elicit inflammatory response like PGE2 synthesis by a sensitive radioimmunoassay. The peptides which did not have any inflammatory activity were then tested for their ability to stimulate antigen primed T-cells in vitro in the presence of sub-optimal concentration of the antigen. Keywords. Interleukin-l peptide 47–55; immunomodulation; Shigella dysenteriae 116 kDa antigen; exposed domain; mutant peptides.
1. Introduction Interleukin-lß(IL-1ß) is known to act as an immunomodulatory protein by activating antigen primed TH cells to divide and secrete Interleukin-2 (IL-2) with concommitant expression of IL-2 receptors (Staruch and Wood 1983; Kaye et al 1984; Lowenthal et al 1986). Besides being an immunomodulator for TH cells it is also known to be involved in various other interactions such as differentiation of B-cells, activation of natural killer (NK) cells and induction of prostaglandin E2 (PGE2) synthesis in human fibroblasts (Falkoff et al 1983; Pike and Nossal 1985; Dinarello 1991). Therefore IL-1ß is known as one of the most Pleiotropic lymphokines (Dinarello 1991). For a globular protein like IL-1ß it is likely that its biological activities are mediated through the exposed domains which are available for interactions with the external milieu. Therefore Antoni et al (1986) used computational methods to predict them. The peptide 47–55 which represents one of the domains (previously referred as 163–171) predicted by them to be exposed when tested showed the ability to activate T-cells and did not stimulate inflammatory responses in vitro. Thus it was proposed that this nonapeptide 47–55 may represent one of the domains of human IL-1ß responsible for * Corresponding author (Fax, 91-11-6165886; Email, [email protected]).
Μ Bruhaspathy and S Κ Kar
its immunostimulatory activity. In another study by Nencioni et al (1987) it was shown that the same nonapeptide 47–55 could act as an immunoadjuvant for Τ dependent antigens like sheep red blood cells (SRBC) as well as Τ independent antigens like Streptococcus pneumoniae type III polysaccharide. When Rao and Nayak (1990) synthesized a composite peptide consisting of the sequence 12-32 derived from hepatitis Β virus surface antigen S1 and the human IL-1 sequence 47–55 and tested it in mice it was observed that the immunogenicity of the hepatitis Β virus epitope was significantly enhanced. Manivel and Rao (1991) also showed that the inclusion of the same nonapeptide (47–55) with a hepatitis Β vaccine preparation induced higher antibody response against hepatitis antigens in mice. We have analysed the human IL-1β sequence by a computational approach that predicts the exposed domains of a protein by taking into consideration the possible occurrence of β-turns in it. Then we have compared our predictions with the X-ray crystallographic and NMR spectroscopic data available for IL-1β (Priestle et al 1989; Driscoll et al 1990; Clore et al). By this approach we have identified three sequences of human IL-1ß to be exposed on the surface of the molecule. Since the domain 47–55 predicted by us to be exposed was already shown to be immunomodulatory and we have focused our attention on this peptide. With a view to find out the minimum structure necessary for immunomodulatory activity we synthesized four peptides with sequences 47–55, 41–61, 45–61 and 50–66 of human IL-1ß and tested them in vivo in Balb/c mice using S. dysenteriae antigens. We found that the peptide representing the sequence 47–55 is the smallest molecule capable of immunomodulation in vivo. To understand the relationship of structure with immunomodulatory function of the peptide 47–55 we synthesized mutant peptides representing alterations of Val47 to Asp47 (mutant-1), Val47 to Lys47 (mutant-2), Gly49 to Asp49 (mutant-3), Glu50 to Ile50 (mutant-4) and Asp54 to Ile 54 (mutant-5) and tested them in vivo for any IL-ß like activity. We find that the replacement of a hydrophobic amino acid Val47 with either a negatively charged or a positively charged and hydrophillic amino acid Asp or Lys respectively enhances the immunomodulatory activity of the peptide while replacement of Gly49 with Asp or Glu50 with Ile or Asp 54 with Ile did not show any enhancement. 2. Materials and methods 2.1 Animals Female Balb/c mice (6-8 weeks old) obtained from National Institute of Nutrition, Hyderabad, were used in our experiments. 2.2 Peptide synthesis and purification The peptides were synthesized by Merrifield's solid phase peptide synthesis method (Atherton and Sheppard 1989) using Fmoc chemistry on an applied biosystems automatic peptide synthesizer model 43 1A by using HMP resin or on a Biolynx 4175 peptide synthesizer from Pharmacia using Ultrosyn A resin. The peptides were cleaved from the resin with 95 % aqueous trifluoro acetic acid. Crude peptides were purified by preparative HPLC using a C18 reverse phase column on a Waters HPLC system and characterized by amino acid analysis.
Immunomodulatory activity ofthehuman IL- 1ß
2.3 Antigen S. dysenteriae, (serotype 1–7), a human isolate was grown in peptone-water medium (Bacto-Peptone 1%, NaCl 0·5%, pH 7·3) at 37°C with shaking at 200 rpm. Cells were harvested at O·D650 = 0·5 and washed with phosphate buffered saline (PBS) pH 7·3 containing proteolytic enzyme inhibitors such as phenyl methyl sulphonyl fluoride (PMSF)-10(µg/ml, L-l-tosylamide-2-phenyl-ethyl chloro methyl ketone (TPCK)-100 µg/ml and N-tosyl-lysine chloromethyl ketone (TLCK)-50 µg/ml. After suspending the organisms at 1011 cells/ml in washing buffer they were sonicated at 4° using a microtip probe for 40s at 40 % duty cycle on Heat Systems Ultrasonics Inc., W-385 model. The protein profile of the sonicate was checked by 12% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) (Laemmli 1970). 2.4 Immunization protocol Groups of 5 mice each were immunized with 5 × 107 S. dysenteriae whole cell antigens per mouse emulsified in IFA on day 0 and on day 21 by subcutaneous route. Mice belonging to different groups were given either 5 ng of IL-1ß per mouse (obtained from National Institute for Biological Standards and Control, Hertfordshire, England) five times more than the optimum dose decided earlier (Nencioni et al 1987) or 50 ng of IL-lra (obtained from NIBSC, Hertfordshire, England) or 500 nmol ofone ofthe peptides 47–55 or 41–61 or 45–61 or 50–66 or irrevalent peptide (reverse sequence of 47–55) per mouse along with the S. dysenteriae whole cell antigens on day 0 and on day 21. The optimum dose of the peptides was decided on the basis of previous published observation (Manivel and Rao 1991). As a control group, five mice were immunized with S. dysenteriae whole cell antigens but using CFA as an adjuvant on day 0 and IFA on day 21. Sera were collected from individual mice in each group on day 28 and stored at –70°C till further use. In the second set of experiment mice belonging to different groups were given either 500 nmol of one of the peptides 47–55 or its point mutants or irrelevant peptide per mouse along with the same antigen emulsified in IFA and processed as above. 2.5 PGE2 assay The Induction of PGE2 secretion in HeLa cells by human recombinant IL-1β or its peptides was measured by radioimmunoassay (RIA) (Simon et al 1993). HeLa cells obtained from National Institute of Immunology, New Delhi were cultured in RPMI1640 medium supplemented with 10% FBS, 50 µg/ml gentamycin sulphate and 2 mM glutamine. The cells were plated on a 24-well plate at 9 × 104 cells/well and grown overnight prior to assay. The growth medium was removed, the cells rinsed with medium and 200 µl RPMI1640 containing 0·1%FBS and the respective concentrations of IL-1ß (25IU/ml) or its peptides (100 µg/ml) was added to the cells. After 6 h of incubation, 100µl of medium was removed from each well and assayed for PGE2 by RIA using a commercial kit (Amersham). 2.6 Enzyme linked immunosorbent assay The 116kDa antigen obtained from 107 S.dysenteriaewhole cell sonicate suspended in 0·05Μ carbonate/bicarbonate buffer, pH 9·5 was coated on to each well of a 96 well
M Bruhaspathy and S Κ Kar
Maxisorp microtitre plate (Nunc, Denmark) and by using different antisera, enzyme linked immunosorbent assay (ELISA) was done (Engvall and Perlmann 1972). 2,2'Azinobis [3-ethylbenzothianzoline sulphonic acid (ABTS)] was used as the chromogen and the absorbance was read at 405 nm. 2.7 Antigen induced T-cell proliferative responses in vitro Balb/c mice were primed with S. dysenteriae whole cell antigens (100 µg/mouse) emulsified in CFA through subcutaneous route 8-10 days prior to their sacrifice for in vitro stimulation of T-cells. The inguinal lymph nodes of the mice were removed under asceptic conditions and their single cell suspension was made in RPMI1640 medium containing 10% FBS, 100 µg/ml penicillin, 100 µl/ml streptomycin sulphate, 20 mM HEPES and 2 mM L-glutamine, pH 7·4 at 5 × 106 cells/ml. The cell suspension (100 µl) corresponding to 5 × 105 cells were dispensed into each well of a 96 well microtitre plate. The optimal and suboptimal concentration of the Shigella whole cell antigens was decided on the basis of a dose response curve. The cells were stimulated with submitogenic concentration of the S. dysenteriae whole cell antigens (1 µg/ml) in a volume of 50 µl. The human recombinant IL-1ß, 47–55 and its mutant peptides under test were added to the wells in triplicate over a range of dilutions (for IL-1ß 1-100 IU/ml and for peptides 1–100 µg/ml). The cells were cultured for 72 h at 37° in humidified atmosphere of 5% CO2. The cells in each well were pulsed with 1 µCi of [3H ] thymidine for 24 h, the cells were harvested and the radioactivity incorporated was measured using a ß-scintillation counter. 2.8 Statistical analysis The significance of the differences between two groups was determined by their P-values calculated by Students 't' test. Two values are considered as significantly different from each other only when Ρ < 0.05. 3. Results 3.1 Determination of the smallest peptide representing the domain 47-55 capable of inducingimmun omodulation in vivo The IL-1ß molecule and the peptides representing exposed domains 47–55 of human IL-1ß namely 47–55, 41–61, 45–61 and 50–66 were tested for their ability to stimulate antibody response against S. dysenteriae 116 kDa antigens in vivo in Balb/c mice. The whole IL-1ß molecule stimulated maximum antibody response followed by significantly lower but almost similar type of immunomodulatory responses by the peptides as measured by ELISA (figure 1A). As shown in figure 1B and 1C the antibody response to S. dysenteriae antigen is mostly of IgG1 isotype. Immunoblot analysis using total antigen also gave similar type of results except showing that the peptide 41–61 is the most immunomodulatory amongst the four peptides tested on the basis of number of bands visualized (data not shown).
Immunomodulatory activity of the human IL-1ß
Figure 1. End point titres of antigen specific total lgG(A) and lgG1 (B) and lgG2a (C) isotype response of Balb/c mice to S. dysenteriae 116 kDa antigen in the presence of various peptides encompassing the domain 47–55 as measured by ELISA. The values represent the average end point titres of the sera from 5 different mice, (a), Ag in IFA after two immunizations; (b), Ag + IL-1ß in IFA after two immunizations; (c), Ag+47–55 peptide in IFA after two immunizations; (d), Ag + 41–61 peptide in IFA after two immunizations; (e), Ag+ 45–61 peptide in IFA after two immunisations; (f), Ag + 50–66 peptide in IFA after two immunizations; (g), Ag + Irrelevant peptide in IFA after two immunizations.
3.2 Immunomodulation by IL-Iβ and mutant 47–55 peptides in vivo Quantitation of antigen specific Ig response in Balb/c mice to S. dysenteriae 116 kDa antigens induced by IL-1ß the 47–55 peptide and its mutants is given in figure 2. The mutant peptide-1 (Val 4 7 → Asp 47 ), peptide-2 (Val 4 7 → Lys 4 7 ) and peptide-5 (Asp54 → Ile 54) showed significantly higher level of antigen specific IgG response in comparison to when no peptide was used (figure 2A-e, f and i). Similarly the end point
Μ Bruhaspathy and S Κ Kar
Figure 2. Total antigen specific lgG (A) and end poin titres of lgG1 (B), lgG2a (C) and lgG3 (D) isotype response to S. dysenteriae 116 kDa antigen in the presence of lL-1ß lL-lra, 47–55 peptide and its point mutants as measured by ELISA. The values represent the average end point titres of the sera from 5 different mice. The P-values between native 47-55 and Mu-1 or Mu-2 or Mu-3 or Mu-4 or irrelevant peptides <0.05 in (A), (B) and (C) but not (D). (a), Ag in IFA after two immunizations; (b), Ag in CFA after two immunizations; (c), Ag + IL-1ß in IFA after two immunizations; (d), Ag + IL- 1ra in IFA after two immunizations; (e), Ag + Native 47–55 in IFA after two immunizations; (f), Ag + Mutant-1 in IFA after two immunizations; (g), Ag + Mutant-2 in IFA after two immunizations; (h), Ag + Mutant-3 in IFA after two immunisations; (i), Ag + Mutant-4 in IFA after two immunizations; (j), Ag + Mutant-5 in IFA after two immunizations; (k), Ag + Irr peptide in IFA after two immunizations.
titres of antigen specific IgGl, IgG2a and lgG3 under these conditions are shown in figure 2B, C and D. Again here the mutant peptide-1 and peptide-2 showed significant modulation of antigen specific IgG1 isotype response (figure 2B-e and f). The P-values when compared with the native 47–55 and Mu-1 or Mu-2 or Mu-3 or Mu-4 or irrelevant peptide with respect to antigen specific IgG, IgGl, and IgG2a were significant (P < 0·05).
Immunomodulatory activity of the human IL-1ß
Figure 3. Quantitation of prostaglandin E2 secreted by 9 × 104 HeLa cells in the presence of IL-1ß or the native peptide 47–55 or its mutants as determined by RIA. Values are the mean of duplicate determinations after deducting the control (a), IL-1ß (25 IU/ml); (b), Native 47–55 peptide (100 µg/ml); (c), Mu-1(100 µg/ml) (d), Mu-2 (100µg/ml); (e) Mu-3 (100 µg/ml) (f), Mu-4 (100µg/ml) (g), Mu-5 (100 µg/ml); (h), Irrpeptide (100 µg/ml).
3.3 Inflammatory activity of the peptide 47–55 and its mutants in vitro The ability of IL-1ß and the peptides representing the domains 47–55 and its point mutants to stimulate PGE2 secretion by HeLa cells as measured by a sensitive RIA is shown in figure3. While the whole IL-1ß caused inflammation by enhanced PGE2 secretion from the HeLa cells the peptide 47–55 is not having any inflammatory property. Like the native 47–55 peptide the mutant 1 and 2 caused no stimulation of HeLa cells to secrete PGE2 in vitro indicating that they are non inflammatory in nature. Though the same was observed with the mutant-5, the other mutants (mutants 3 and 4) showed slight inflammatory activity as illustrated in figure 3. 3.4 Proliferation of sub-optimal antigen primed lymph node cells from Balb/c mice by IL-1β, the peptide 47–55 and its mutants in vitro To see if IL-1ß the peptide 47-55 and its mutants can induce proliferation of sub-optimal antigen primed lymph node cells from Balb/c mice different concentrations S. dysenteriae antigens was used. A bell shaped response curve is obtained when increasing concentration of the antigen was used to stimulate the lymph node cells derived from antigen primed mice (figure 4A). From the above stimulation data a suboptimal concentration of the antigen is selected. As shown in figure 4B there is an increase in the proliferation of the primed lymph node cells when they are stimulated in
M Bruhaspathy and S Κ Kar
Figure 4. (A) Determination of optimal concentration of S. dysenteriae whole cell antigens required for stimulation of primed lymphocytes of Balb/c mice. (B) Stimulation of lymphocytes from Balb/c mice by IL-1β and the native peptide 47–55 and two of its mutants in the presence of sub-optimal concentration of S.dysenteriae whole cell antigens. The P-values between native (c) and Mu-5 (e) or irrelevant peptide (f) ( P <0·01) (a), Sub-optimal antigen alone; (b), Sub-optimal antigen +IL-1β (c), Sub-optimal antigen +47-55 peptide; (d), Sub-optimal antigen + mutant-1 peptide; (e), Sub-optimal antigen + mutant-5 peptide; (f), Sub-optimal antigen + Irr. peptide.
the presence of sub-optimal concentration of the antigen and IL-1ß as measured by [3H] thymidine incorporation assay. Of the noninflammatory 47–55 point mutant peptides the mutant-1 induced the maximum proliferation of the lymph node cells when compared to the native 47–55 peptide and mutant-5.
4. Discussion The domain 47–55 of mature human IL-1ß molecule (previously referred as 163–171) has been predicted to be hydrophilic in nature and several groups have shown the same peptide to be immunostimulatory but not inflammatory (Antoni et al 1986; Nencioni et al 1987; Rao and Nayak 1990). We have predicted the exposed domains of human IL-1ß by using a computational method which besides considering factors like local hydrophilicity, acrophilicity, accessibility, flexibility also takes into consideration the occurrence of turns in the sequence. In our opinion this approach would give better prediction of exposed regions of a protein as backbone groups forming a ß-turn would tend to protrude the neighbouring domains of the molecule to the surface (Pauletti et al 1985). The region 47–55 of IL-1ß predicted to be exposed by us was confirmed to be so after analysis of X-ray crystallographic and 3D and 4D NMR spectroscopic data taken from Protein Data bank. As the region 47–55 is shown to have the ability to activate T-cells and unlike the whole IL-1ß molecule it does not induce any inflammatory responses as shown by us and others (Antoni et al 1986), we have studied this peptide in greater detail. In the native structure of the whole IL-1ß The domain 47–55 is made up of a type II ß-turn
Immunomodulatory activity of the human IL-1ß
Figure 5. Amino acid sequences of the peptides encompassing the domain 47–55 used for minimum domain analysis.
spanning the sequence 52–55 residues which is encompassed by a rigid ß-strand spanning the sequence 56–62 on the C-terminus and an irregular ß-strand spanning the sequence 41–51 onthe N-terminus. The presence of a slightly flexible ß-bulge spanning the sequence 46–49 is responsible for the irregular structure of ß-strand on the upstream of the ß-turn. The hydrophobic nature of the residues forming these two ß-strands is likely to stabilize the -turn and thus may be contributing to the functions attributed to the domain 47–55. In order to see the effect of these two ß-strands on the function of this domain and to find out the minimum structure that is required for the immunomodulatory activity of this exposed domain so that we can then make mutant peptides to study structure-function relationships we synthesized four different peptides representing the sequences 47–55 (ß-bulge-ß-turn), 41–61 (ßstrand-ß-turn-ßstrand), 45–61 (ß-bulge-ß-turn-ß-strand) and 50–66 (ß-turn-ß-strand) (figure 5) (Clore et al 1991). In our study we have used a complex antigen mixture i.e., S. dysenteriae whole cell antigens and we have examined in vivo response of animals to these antigens emulsified in a weak adjuvant like IFA alone and IFA in the presence of either IL-1ß whole molecule or peptides 47–55 or 41–61 or 45–61 or 50–66. We have used five times more IL-1ß than what has been reported mainly because our route of immunization is different and we have used recombinant IL-1ß supplied by NIB SC, Hertfordshire, UK. Further; it has been shown that eventhough the immunomodulatory effect saturates at a lower dose, use of higher dose does not have any adverse effect. We found that IL-1ß modulates antigen specific IgG1 isotype response much more than any other isotype (M Bruhaspathy, unpublished observation). Therefore we have measured antigen specific IgG1 and IgG2a response in the presence of IL-1ß and various peptides as an indicator of IL-1ß like immunomodulation. From the observations as shown in figure 1 based on the in vivo assay for immunomodulation, it is clear that the activity is not increased significantly by including the -strands on either side of the peptide 47–55 as represented by the peptides 41–61, 45–61 or 50–66. This is surprising given the observations that Ile56 and aromatic ring of Phe46 are critical in the maintenance of the binding cleft on IL-1ß Therefore we found that the peptide 47–55 is the smallest peptide tested which had immunomodulatory activity. With a view to analyse the immunomodulating functions of the domain 47–55 peptide with respect to its structure, we designed and synthesized peptides with point mutations at different positions. Even though the peptide 41–61 is a better immunomodulator we selected the shortest peptide i.e., the peptide 47–55 for further study only because number of mutant peptides to be synthesized in this case will be smaller. Careful examination of the 47–55 sequence of IL-1ß from mouse, human,
Μ Bruhaspathy and S Κ Kar
Figure 6. Amino acid sequences of rat, mouse, rabbit, sheep, bovine and human IL-1ß after their alignment.
Figure 7. Amino acid sequences of mutant peptides of 47–55.
rabbit, bovine and sheep revealed that the four contiguous N-terminal amino acids namely Val, Gin, Gly and Glu (position 47 to 50) are conserved in all of them (figure 6). Therefore these four amino acids are likely to be playing some important role in the function of the molecule. Out of these four amino acids Val is the only hydrophobic amino acid and therefore we thought replacement of Val with an amino acid which is hydrophilic and not very different in size but negatively charged Asp (mutant-1) or positively charged Lys (mutant-2) may afect the biological activity of the peptide. Further, our computational analysis had predicted a ß-turn in the peptide 47–55 spanning Ser52-Asn-Asp-Lys55 and therefore we felt replacement of Asp at position 54 (i + 2 position of turn) with an amino acid which is a ß-turn breaker like Ile (Chou and Fasman 1979) may affect the structure of the peptide and therefore may be its immunomodulatory activity. This is our rationale for synthesis of mutant-1, mutant-2 (Val 47 → Lys 47 ) and mutant-5. Mutant-3(Gly 49 → Asp 49 )andmutant-4(Glu 50 → Ile 50 ) were synthesized to see the effect of change of amino acids at other positions (figure 7).
Immunomodulatory activity of the human IL-1ß
We have examined in vivo response of animals to S. dysenteriae antigens emulsified in a weak adjuvant like IFA alone and IFA in the presence of either IL-1ß whole molecule or IL-1 receptor antagonist which binds to the IL-1 receptor but does not induce any signal as a control for IL-1ß or 47–55 peptide and its point mutants or an irrevalent peptide. We have used IF A because it being a weak adjuvant immunostimulatory response, if any, brought about by IL-1ß or any of the peptides can be measured. CFA being a strong adjuvant itself there will be no scope for any further immunostimulation by the IL-1ß or peptides. We have deliberately selected IFA as an adjuvant only because the antigen and immunomodulators (IL-1ß and various peptides) trapped inside an oil in water emulsion will be released slowly and have a prolonged effect. This cannot be achieved by using A1(OH)3 that effectively. We have used ELISA and Western blotting to analyse the antibody response of mice against the Shigella antigens. We found that S. dysenteriae whole cell lysate contained a 116 kDa antigen which is very immunogenic. Since this antigen could be pelleted easily from the whole cell lysate by centrifugation and it could be coated on to the ELISA plate more quantitatively and antibody titration curves were much more reproducible in comparision to the whole cell lysate it was used for measuring immune response of mice by ELISA. We find that while the peptide 47–55 at 500nmol concentration is moderately immunostimulatory, the mutant-1 and mutant-2 at that concentration are much better immunomodulators (figure 2A). Unlike the whole IL-1ß which is active at pmol concentration native 47–55 peptide and the mutants 1 and 2 were needed at 500 nmol concentration to show significant immunomodulatory effect. This is probably because the conformation which is actually involved in immunomodulation is maintained in the whole IL-1ß molecule because of structural constrains. But the peptides being small are highly flexible and therefore the biologically active conformation is achieved by a very small proportion of the molecule at any given time. It is also likely that a very small proportion of the peptides assume a particular conformation after interacting with a cell surface receptor by induced fit mechanism. We find that the replacement of a hydrophobic amino acid valine47 (Val) with negatively charged and hydrophilic amino acid aspartic acid (mutant-1) or positive ly charged and hydrophilic lysine (mutant-2) enhances the immunomodulatory activity of the peptide while replacement of glycine49 with aspartic acid (mutant-3) or glutamic50 acid with isoleucine (mutant-4) did not show any enhancement. As mentioned earlier IL-1ß modulated antigen specific IgGl response more strongly than any other isotype. We believe that IL-1ß or its peptides can stimulate antigen primed TH2 cells more specifically. Thus more amount of IL-4 will be produced. This would then induce B-cells to synthesize and secrete greater amount of IgG1 isotype antibodies. Therefore we are observing significantly higher titre of lgG1 response in the presence of IL-1ß and its peptides as represented in figure 2. In order to see the involvement of the residues at various positions of the sequence 47–55, we tested another function attributed the whole IL-1ß molecule, namely, the ability of these mutants to cause any inflammation as measured by quantitation of PGE2 secretion. Unlike their immunomodulatory property, the mutants-1,2 and 5 did not stimulate HeLa cells to secrete PGE2 indicating that the residues at these positions are not involved in the inflammatory reaction. In contrast, mutants-3 and 4 have gained the inflammatory property not associated with the 47–55 peptide but with the whole IL-1ß molecule though not to a significant extent (figure 3). All these observations indicate that retention of Gly49 and Glu50 are important for the non-inflammatory nature of the domain 47–55.
Μ Bruhaspathy and S Κ Kar
Having seen the immunomodulatory properties of the 47–55 and its mutant peptides behaving like IL-1ß in vivo we wanted to see whether these peptides exhibit any other IL-1ß like activity in vitro. As lymph node cells derived from antigen primed mice respond to IL-1ß in the presence of sub-optimal dose of the antigen which can be measured by [3H ] thymidine incorporation, we used this system to assay the ability of the peptide 47–55 and its mutants to cause T-cell proliferation in vitro. The optimal and suboptimal doses of the S. dysenteriae antigens were chosen from the dose response curve (figure 4 A). We have focussed on the ability of only those peptides corresponding to the domain 47–55 without any inflammatory response. We used S. dysenteriae antigens to prime the Balb/c mice and stimulated their lymph node cells in vitro using the same antigen. Though S. dysenteriae antigens contain a mixture of T-dependent and T-independent antigens it was used to compare the in vivo and in vitro responses using the same antigen. While IL-1ß elicited maximum T-cell proliferation, the mutant-1 among the non inflammatory mutant peptides of 47–55 could stimulate a higher response (figure 4B). The irrelevant peptide used by us gave non-specific stimulation of suboptimal antigen primed T-cells. But the same peptide did not show any non-specific stimulation when defined antigens like gelonin or thyroglobulin were used (M Bruhaspathy, unpublished data). We believe that the non-specific stimulation is due to the nature of S. dysenteriae antigens used to prime the T-cells. Thus far we have shown that co-injection of IL-1ß can modulate antibody response to S. dysenteriae 116 kDa antigen and the peptide representing the domain 47–55 can also do the same in a dose dependent manner without showing any inflammatory response. Some of the mutants of the 47–55 peptide are able to modulate response much more effectively than the native peptide. The change of Val47 to Asp47 in the mutant-1 and Val47 to Lys47 in mutant-2 would have made the peptides more hydrophillic. But it is not clear how they bring about immunomodulation. We have also studied circular dichroism spectra of the peptides (data not shown) to see if they have any secondary structure. The CD spectra of 47–55 peptide and the mutants recorded in water and in 90% trifluoroethanol did not show any propensity for the formation of α-helix or ß-sheets. Thus these peptides are probably not having any secondary structure. Therefore it remains to be seen how the peptide 47–55 and its mutants are able to show IL-1ß like activity without binding to its receptor.
References Antoni G, Presentini R, Perin F, Tagliabue A, Ghiara P, Censini S, Volpini G, Villa L and Boraschi D 1986 A short synthetic peptide fragment of human Interieukin-1 with immunostimulatory but not inflammatory activity; J. Immunol, 137 3201-3204 Atherton Ε and Sheppard R C 1989 Solid Phase Peptide Synthesis: A practical approach (Oxford: IRL Press) Chou Ρ Υ and Fasman G D 1979 Prediction of beta turns; Biophys. J. 26 367-373 Clore G M, Wingfield Ρ Τ and Gronenborn Α Μ 1991 High-Resolution Three-Dimensional Structure of IL-1 in solution by Three- and Four-Dimensional Nuclear Magnetic Resonance Spectroscopy; Biochemistry 30 2315-2323 Dinarello C A 1991 Interleukin-1 and Interleukin-1 antagonism; Blood 77 1627-1652 Driscoll Ρ C, Gronenborn A M, Wingfield Ρ Τ and Clore GM 1990 Determination of the secondary structure and molecular topology of Interleukin-1 by use of two and three dimensional heteronuclear 15 N-1H NMR spectroscopy; Biochemistry 29 4668-4682 Engvall Ε and Perlmann Ρ 1972 Enzyme linked Immunosorbent assay (ELISA), III. Quantitation of specific antibodies by Enzyme labelled antiimmunoglobulin in antigen coated tubes; J. Immunol. 109 129-135
Immunomodulatory activity of the human IL-1ß
Falkoff R J Μ, Muraguchi A, Hong J X, Butler J L, Dinarello C A and Fauci A S 1983 The effects of Interleukin-1 on human Β cell activation and proliferation; J. Immunol. 131 801-805 Kaye J, Gillis S, Mizel S B, Shevac Ε Μ, Malek Τ R, Dinarello C A, Lachman L Β and Janeway C A Jr 1984 Growth of a cloned helper Τ cell line induced by a monoclonal antibody specific for the antigen receptor: Interleukin-1 is required for expression of receptor for Interleukin-2; J. Immunol. 133 1339-1345 Laemmli U Κ 1970 Cleavage of structural proteins during the assembly of head of bacteriophage T4; Natural (London) 227 680-685 Lowenthal J W, Cerottini JC and MacDonald HR 1986 Interleukin-1 dependent induction of both Interleukin-2 secretion and Interleukin-2 receptor expression by thymoma cells; J. Immunol. 137 1226-1231 Manivel V and Rao Κ V S 1991 Interleukin-1 derived synthetic peptide as an added coadjuvant in vaccine formulation; Vaccine 9 395-397 Nencioni L, Villa L, Tagliabue A, Antoni G, Presentini R, Perin F, Silvestri S and Boraschi D 1987 In-vivo immunostimulatory activity of the 163-171 peptide of human IL-1 ~; J. Immunol. 139 800-804 Pauletti D, Simmonds R, Dressman G R and Kennedy R C 1985 Application of a modified computer algorithm in determining potential antigenic determinants associated with the AIDS virus glycoproteins; Anal. Biochem. 151 540-546 Pike Β L and NossalG J V 1985 Interleukin-1 can act as a Β cell growth and differentiation factor; Proc. Natl. Acad. Sci. USA 82 8153-8157 Priestle J P, Schar Η Ρ and Grutter Μ G 1989 Crystallographic refinement of IL-1ß at 2A resolution; Proc. Natl. Acad. Sci. USA 86 9667-9671 Rao Κ V S and Nayak A R 1990 Enhanced immunogenicity of a sequence derived from hepatitis Β virus surface antigen in a composite peptide that includes the immunostimulatory region from human Interleukin-1; Proc. Natl. Acad. Sci. USA 87 5519-5522 Simon Ρ L, Kumar V, Lillquist J S, Bhatnagar P, Einstein R, Lee J, Porter T, Green D, Sathe G and Young Ρ R 1993 Mapping of neutralizing epitopes and the receptor binding site of human Interleukin-1ß; J. Biol. Chem. 268 9771-9779 Staruch Μ J and Wood D D 1983 The adjuvanticity of Interleukin-1 in vivo; J. Immunol. 130 2191-2194