WO2006114799A1

WO2006114799A1 - Synthetic peptides for the diagnosis and therapy of berillium granulomatous disease

Info

Publication number: WO2006114799A1
Application number: PCT/IT2005/000241
Authority: WO
Inventors: Massimo Amicosante; Cesare Saltini; Floriana Berretta
Original assignee: Universita' Degli Studi Di Roma 'tor Vergata'
Priority date: 2005-04-27
Filing date: 2005-04-27
Publication date: 2006-11-02

Abstract

The present invention regards peptides that have the capacity to block or increase the response to beryllium of beryllium specific T-lymphocytes in subjects suffering from Berylliosis or Chronic Beryllium Disease and Beryllium sensitization and the methods for selection and production of these peptides. The invention encompasses, as well, the use of these peptides as pharmacological compounds and diagnostic agents in the treatment and diagnosis of Berylliosis or Chronic Beryllium Disease and Beryllium sensitization.

Description

SYNTHETIC PEPTIDES FOR THE DIAGNOSIS AND THERAPY OF BERILLIUM GRANULOMATOUS DISEASE

DESCRIPTION Background of the Invention The present invention regards peptides that have the capacity to block or increase the response to beryllium of beryllium specific T- lymphocytes in subjects suffering from Berylliosis or Chronic Beryllium Disease and Beryllium sensitization and the methiods for selection and production of these peptides. The invention encompasses, as well, the use of these peptides as pharmacological compounds and diagnost άc agents in the treatment and diagnosis of Berylliosis or Chronic Beryllium Disease and Beryllium sensitization. State of the Art

It has been shown that the interaction between the immune system and a growing number of toxic environmental agents plays a fundamental role in the alarming increase in the incidence of allergies, autoimmune disorders and chronic granulomatous and inflammatory disorders. An increasing number of these environmental related disorders have been identified as having a con-causal genetic component. The genes coding- for the major histocompatibility complex (MHC) , identified as the "immune response genes" and dominated in man by the human leukocyte antigens (HLA) , play a cen-tral role in the autoimmunity and the susceptibility to hypersensitivity disorders. The understanding of the immunological basis of certain allergies and environmental "toxic agents, such as nickel allergy, cobalt induced large cell pneumonia and beryllium allergy, has led researchers to hypothesize the mechanisms of interaction between H-LA genes and toxic environmental agents. As for other immune disorder, such as asthma, despite the strong evidence for a role of HLA genes in the susceptibility to disease, the mechanisms of interaction between these genes and causal agents of the disease have not yet been completely elucidated. Due to its particular chemical and physiczal properties, Beryllium (Be) is widely used in the aerro- space, military, electronic and telecommunication industry, despite the fact that it has been unequivocal-Iy demonstrated that the inhalation of Be containing dusts is the cause of a hypersensitivity disease that leads to chronic granuloma formation and fibrosis in the lungs with consequent rrespiratory insufficiency. The immunosuppressive therrapy generally indicated has besen shown to be largely inefficacious and lung transplant hias remained the treatment of choice.

Of those subjects who develop Sensitization to Be, approximately 50 % do not develop the disease, rathier only sensitization of "the T - lymphocytes to the metal as demonstrated by their capacity to proliferate in vitzro when stimulated with Be salts. The remaining 50% devel_op a chronic granulomatous disease knoyn as Berylliosis or Chronic Beryllium Disease, which is histopathological_ly indistinguishable from Sarcoidosis, the most common chronic pulmonary granulomatous disease of unknown etiology. Berylliosis , which involves primarily the lungs, is characterized by the accumulation in the lungs of T - lymphocytes of the phenotype "effector memorry" CD4+, CD45R0+, DR+, w-hich proliferate in response to Be as a specific hapten presented in the context of class II MHC.

As in other disorders caused by an exaggerated immune response, it has been hypothesized that there is a genetic susceptibility to developing Berylliosis. A previous study first demonstrated that the susceptibility to Berylliosis is associated with the expression of a variant allele of tine HLA-DP gene, which codes £or glutamic acid instead of lysine in position 69 of the beta chain of HLA-D-P (HLA-DPGlu69) . This amino acid substitution was present in 97% of a group of patiemts while it was present in only 27% of a control group exposed to Be (p<0.00Ol) (Richeldi L et al . Science 1993; 262: 242-244). This observation was independently confirmed in other groups of subjects exposed to Be both by our laboratory and by otherrr researchers (Saltini C et al Eur Respir J 2001; 18: 6~77-84; Maier LA et al. J Immunol. 2003; 171: 6910-8).

The current understanding of the functional role that genetic products coding rfor HLA-DPGlu69 play in the pathogenesis of the direct imirmne reaction against Be are based on the following observations: 1) clones of Be-speciffic CD4⁺ T - lymphocytes obtained from patients with Berylliosis proliferate in response to Be in a dose dependent manner in the context of the class II HLA molecule C Saltini C et al N. Engl. J. Med. 1989; 320: 1103-1109); 2) clones of Be-specific CD4⁺ T - lymphocytes obtained from patients with Berylliosis recognize Be as an antigen only when it is presented in the context of the HLA-DPβGlu69 molecule and not with the other allelic variants of HLA-DP containing Lys69 (Lombardi G et al. J Immunol 2001; 166: 3549-3555)^

3) Be shows an affinity for the HLA-DPGlu69 molecule 40 to 100 times superior to that for the HLA-DP molecule mutated for the codification of a lysine (Lys69) in the same position. This increased affinity is related to the capacity of Be to abrogate the bond that the HLA-DPGlu59, but not HLA-DPLys69, makes with the class II associated invariant-chain derived peptd_de (CLIP) , a peptide which binds with low/moderate affinity to all the class II HLA molecules. (Amicosante M. et al. Human Immunology 2001; 62: 686-693) .

It is not yet known what role the peptides associated with HLA-DP play i_n the binding of Be nor if endogenous peptides can enhance or impair the binding of Be to HLA-DPGlu69 and thereby enhance or impair the immune response to Be.

The class II HLA mol_ecules are receptors for peptides that function mainly by binding peptides produced within the cell in the endoplasmic reticulum or phagocytized out side the cell in phag olysosomes . These peptides are then presented to the superficial surface of the T lymphocytes using a mechanism that is the basis for the identificat ion of foreign antigens in the surveillance against tumors , viruse s and bacterial agents .

The class I I HLA molecules are expressed on a limited number of cell types , known as sntigen presenting cells (APC ) , among which are included macrophages , B- lymphocytes and dendritic cells . They bind peptides derived from proteins that have been processed in the endosomal/lysosomal compartment , inclu- ding phagocytozed antigens . These molecules are highly polymorphic and in humans exist in 3 isotypes , (HLA-DR, HLA-DP e HLA-DQ) , which are expres sed on the cell surface membrane .

The mechani sms which are at the base of the interaction between the class II HILA molecules and peptides , and ttie role of the polymorphic alleles in binding the anti gen peptides , have beeri clarified thanks to the determination of the three dimonsional structure of some allelic variants of the class I Z HLA molecules of the HLA-DR isotope ( Scott CA, et al . Immunity 1998 : 8 : 319-29 ) . In general, the peptide antigens bind to the HLA molecules with certain amino acid resid^~ues or agreotopes , which interact with specific areas on the superficial surface of the HLA molecules known as pockets , while other amino acid residues of the pep -tide or epitopes , which are spatia lly opposite the agret opes in that they are pointed outwards from the molecule, interact with the T cell receptor , forming a three dimensional complex, made up of the HLA molecule , the pept-Lde and the T-cell antigen receptorr , referred to as the "immune synapse" (Reinherz EL, et al . Science 1999 : 286 : 1913-21 ) .

The capacity of the HLA molecule to bind different peptides depends on the form and the el ectrical charge of the pocket, which, are determined by the polymorphic residues of the HL-A molecule, which in turn, dictate the specific repertoire of the peptides bound to each allelic variant of the isotypes of HLA. In tϊiis manner, the allelic variants of HLA determine which antigen peptides will be recognized by the T lymphocytes. While the contribution of polymorphism of the HL.A.-DR and HLA-DQ molecules has been studied in detail, as has been the means of peptide binding to the alleles most frequently occurring in the population, the information on the means of selection of the peptides by the HL-A-DP molecule as well as the inforrmation on the structure of this HLA isotype, are very limited.

Previous studies (Berretta F., et al . Tissue Antigens, 2003, 62: 459-471), have evaluated in detail the structural and functional role in terms of peptide binding in the glutamic acid/lysine (Glu/Lys) polymorphism in position 69 of the D chain of HLA-DP. This polymorphism is involved in the s "usceptibility to Berylliosis and Beryllium sensitizations and has been studied using the molecular modeling system and the competition test on the HLA molecu He produced in recombinant form in the Drosophila m&lanogaster. In these studies, trie CLIP peptide (Clas s II associated Invariant Chain Peptide) was employed ss the reference prototype. This peptide is pressent on the histocompatibility molecule and from here it is displaced when the HLA molecule binds the antigexiic peptides for presentation to T lymphocytes. In addition, the peptide has the unique ability to bind all class II HLA molecules with low/medium affinity and thus can be considered a "universal" ligand .

The detailed structure of the HLA-DP2 molecule, the prototype of HLA-DP molecules having glutamic acid at position 69 of the b chain, was predicted using a system of molecular modeling on the HLA-DP2 molecule and peptide CLIP complex. The molecular model obtai_ned was compared with that of the same molecule mutated so as to have a lysine residue in position 69 of the 3o chain, thereby- rendering it analogous to the natuiral allele HLA- DPBl*0402. The most significant changes that were observed between the two models were in the form and distribution of electrical charge in pocket 4 and to a lesser extent, in the nearby pocket 5 (Figure 2) . In o xder to analyze in detail the role played by the polymorphism Glu/Lys in the binding within these pockets, a library was designed of CLIP peptides mutated in the positions corresponding to and interact-Lng with pocket 4 and pocket 6 in such a way as to cover ail the amino acid variants in these two positions (F. Berretta et al. Tissue Antigens 2003, 62, 459-471) . Th. us, the peptides were tested, in competition with respect to the unaltered CLIP peptide, for affinity of binding- to the HLA-DP2 molecule and to its mutated form HLA-DP2Lys69. In particular, it was observed that pocket 4 demonstrated an elevated affinity for the positively charged residues, the aromatic residues and the polar residues. Pocket 4 of the HLA-DP2Lysβ9 demonstrated instead a reduced selectivity for the positively and negatively charged residues, favoring instead the aromatic residues. Pocket S of both the molecules presented a high, affinity for the aromatic residues. HLA-DP2Lys69 demonstrated a particularly increased affinity for argi nine .

On the basis of this information regarding structure and bonds, it was hypothesized that beryllium binds .Lnside pocket 4 of the HLA-DP molecule "which expresses a glutamic acid in position 69 of the beta chain, the pocket in which positive charges can easily be -accommodated and that beryllium is coordinated, in the tsond with HLA-DP molecule, with the contribution of the electrodonor groups of pocket 4 (which Lncludes the amino acids Glnl3, Glul4, Arg27, Tyr28 e Glu69 of the β chain of the HLA-DP) . In addition, it may be hypothesized that the peptides bound to the HLA molecule may interact with Beryllium and contribute with theirr electrodonor groups in the coordination of Beryllium.

In conclusion, it has been described in the literature that the HLA-DP Glu69 variant is capable of binding beryllium and may be involved the pathogenesis of Berylliosis. On the other hand, separate from this finding, the literature has shown tϊiat specific mutations in the sequence of the CLIP protein provoke an increase or a decrease in the affinity of the bond with the HLA-DP or HLA-DP K69 molecules with respect to the natural CLIP peptide. The state of the knowledge did not permit any specific predictions on the poss ible therapeutic or diagnostic applications of these scientific observations. Summary of the Invention The invention was developed ifrom the hypothesis, validated by the results obtained d_n the present study, that certain peptides can alter the binding capacity of beryllium to the HLA-DP molecules in accordance with models of interaction of HLA-DP/pept ide/beryllium. The invention demonstrates that peptides derived from the CLIP peptide can increase the capacity of the HLA-DP molecule to bind beryllium and to present it more efficiently to beryllium specific T lymphocytes. Alternatively, derivatives from the CLIP peptide, with amino acid residues of high affinity/ for pocket 4 of the HLA-DP Glu69 molecule, can "occupy" the beryllium binding site and prevent Be binding to HLA-DP and presentation to the T lymphocytes.

Object of the present invention are peptides derived from the CLIP peptide with mutations at position 94 and/or 96 of the natural polypeptide sequence of CLIP and having an increased affinity for the HLA variant DP-Glu69 molecule as compared to the natural CLIP peptide, which could be employed in an in vivo o in vitro method of therapy or diagnosis.

These peptides may have a secguence of 13 or more amino acids including the binding frame zone found between positions 91 and 99 of CLIP.

A second object of the present invention is the use of the peptides themselves in th.e treatment of Berylliosis or Chronic Beryllium Disease or as a method for diagnosing Berylliosis and Beryllium, sensitization.

Further objects of the invention are pharmaceutical compositions or diagnostic agents using one or more of the peptides in a pharmacologically acceptable excipient.

Other objectives for the invention are in vitro diagnostic methods for the diagnosis of Beryllium sensitization, methods of selection of peptides derived from the CLIP protein capable of blockiπLg the response to beryllium in patients affected by Beryl liosis or Chronic Beryllium Disease, methods of selection of peptides derived from the CLIP protein capable of increasing the response to beryllium in patients affected by Berylliosis or Chronic Beryllium Disease and BerylLium sensitization and procedures for the preparation of peptides, pharmaceutical compositions and diagnostic agents. The invention offers the following advantages. Peptides capable of increasing the response to beryllium can be utilized in vitro or in vivo as antigens together with beryllium sulfate in order to show the response to beryllium of beryllium specific T lymphocytes. Thus, these peptides would allow for the diagnosis of beryllium sensitivity, even if very weak, in exposed subjects.

In addition, peptides capable of blocking the beryllium specific immune response could be adapted for use as a pharmacological agent in speed-fie immunotherapy of beryllium hypersensitivity and in particular, in Berylliosis.

Brief Figure Descriptions

Figure 1. Analysis of the surface of the binding site in the models of the HLA-DP2 e HLA-DP2K69 molecules. The primary binding pockets that interact with the peptide are indicated by the arrows. The surface distribution of partial charge is demonstrated by a color scale var-ying from red (negative charge) to blue (positive charge) as indicated on the right side of the panels .

Figure 2. Example of competition assay between CLIP and mutated CLIP on the HLA-DP2 molecule. The competition ELISA assay was done on the HLA-DP2 moleciαle, with the CLIP pept-Lde in increasing concentrations (abscissa) . In the presence of a fixed quantity of biotinyl ated CLIP (10 μM) . The competition between the CLIP pep "tide and the mutated CLIP peptide in P4 (P4Q) and mutated CLIP peptide in P6 (P6S) is shown. Also shown are the EC50 values (concentration of peptides necessary to displace 50% of the reference biotinylated peptide from the binding site of the molecule) and the IC50 values (the IC50% value refers to the μM unit of the reference peptide) . The biotinylated peptide signal decreases as the concentration of non biotinylated CLIP increases, on the basis of the different affinities for the peptide. CLIP- P4Q has more affinity than CLIP for HLA-DP2 (lower EC50 and IC50) , while P6S presents a reduced affinity as compared to CLIP (higher EC50 and IC50).

Figur e 3. Competition assay between CLIP and mutated CLIP at P4 or P6 on the so-Luble HLA-DP molecules. The values for IC50 were calculated (shown on the abscissa in each panel) for every mutation (in P4 or P6) in competition with biotinylated CLIP for the HLA- DP2 or the HLA-DP2K69 molecule. The ordinate was designed to show the point of equimolar competition with biotinylated CLIP (1 μM) . Thus, the bars extending to the left of the vertical axis represent peptides that bind to trie molecule indicated with higher .affinity than biotinylated CLIP, while the bars to the right represent peptides that bind with higher affinity to the biotinylated CLIP. The amino acid mutation of the peptide analyzed can be seen near each bar. Panel A: affinity of the mutated peptides in P4 forr the HLA-DP2 molecule; Panel B: affinity of the mutated peptides in P4 fox the HLA-DP2K69 molecule; Panel C: affinity for the mutated peptides in P6 for the HLA-DP2 moILecule; Panel D: affinity of the mutated peptides in P6 fror the HLA-DE>2K69 molecule. Figure 4. Competition assay between CLIP and mutated

CLIP in the biotinylated form and beryllium and the reference non biotinylated peptide of HLA-DP2

(Gluβ ©positive) molecule. The values are shown for- the OD

(on the ordinate in each panel) obtained usincg as a competitor the concentrations of peptide or beryllium shown on the abscissa for each, biotinylated pepti <de at a constant concentration of 10 μM.

Figure 5. Production of IFN-gamma by PBMC of a subject with Chronic Beryll±um Granulomatous Disease stimulated with beryllium in the absence and pres ence of increasing concentrations of CLIP or mutated CLI[P. The ordinate shows the values of IFN-gamma (pg/ml) obtained from culture supernatants collected 5 days after stimulation with Beryllium sulfate (20 μM) in the presence of increasing concentrations of competitor peptide (0.4-250 μM) .

Detailed description of the invention Previous studies have characterized the mechanisms of peptide selection by HLA-DP molecules and to define the role played by residue 69 of the β chain into antigenic peptide selection (Berretta F. et al Tissue Antigens 2003) .

According to the method already described by Beretta et a.1. (see above) the HLA-DP2 molecule (HLA- DPAl* O103/DPB1*0201) has been generated in vitro as the prototype of the Glu69 carrying HLA-DP. At the same time the corresponding engineered HLA-DP2K69, which carries a lysine in position 69 of the β chain has also been generated. Both molecules (H.ILA-DP2 e HLA-DP2K69) have been obtained in soluble form from D. melanogaster Schneider 2 (S2) cells after transfection with pRmHa-3 vectors containing the extra -membrane domains of HLA- DPAl*0103, HLA-DPBl*02012 e HLA-DPBl*02012-K69.

HLA-DP molecules have been purified to h-omogeneity from cell culture supernatants of each cell ILine using affinity chromatography. In the first procedure, a nickel column was used (in order to link histidine residues added to the COOH-terminal end of the HLA-DE³ α and β chains. Thereafter a B7/21 conjugated protein-Av Sepharose was used, as the B7/21 monoclonal antibody is HLA-DP specific and binds only to correctly assembled αβ.dimers. αβ heterodimer: s were hence stabilized by the addition of excess concentration of the tetanus toxoi d 947-967 peptide specific for both HLA-DP molecules. CLIP peptide and derived peptides In order to analyze the bond between HLA-DP and peptides, the CLIP peptide was used as the prototype. This peptide is generated by partial proteolytic digestion of the Ii invariant chain, a molecule that is normally bound to the antigen binding site of MHC class II molecules. The CLIP peptide usually encompasses the region found between residues 82 and 114 of the natd_ve protein (invariant chsin Ii) and its sequence is reported here below:

PKPPKPVSKMRMATPLLMQALPMGALPQGPMQ N (SEQ ED NO: 7) Underlined positions correspond to positions 82 (P), 94 (A) , 96 (P) and 114 (N)

Amino acd-d residues in positions 91 (M) , 94 (A) , 96(P) and 99 <M) interact with pockets Pl, P4, P6 and P9 of the surface of the molecular model of HLA-DE?2 which is illustrated in Figure 1. As it is well known that CLIP binds to class II molecules with medium/low affinity, this peptide d_s widely employed as the standard reference peptide in competition assays. Although the frragment 89- 101 of Ii has been used in competition assays on HLA-DP molecules (AmdLcosante M. 2001) , detailed studies on the binding between HLA-DP molecules and CLIP are not available. The binding frame of the CLIP peptide (aa 89-101) on the HLA-DP2 and the -DP2K69 molecules has been determined by Beretta et al. (eerretta et al . Tissue antigens 2003 - see above) by means of a panel of truncated biotinylated CLIP peptides cover±ng the entire sequence of CLHP (aa86- 114) (Beretta et al _ see above)

Competition ELISA assays of different truncated biotinylated CLIP peptide with HLA-DP2 and HΣ_-JA-DP2K69 have shown that only those peptides having the amino acid sequence 91-99 of Ii, display a significant dose dependent bond to HLA-DP molecules, thereby suggesting that the minimum peptide binding core i.e., the shortest sequence required for binding, is comprised of 9 amino acids. Furthermore, as already reported for other HLA molecules, CLIP amino acid residues located in ^positions Pl, P2 and, even more so, the residues located in. PlO and PIl, determine a manrked increase in affinity towards HLA- DP. This is likely due to the stabilizing role of peptide fragments that extend outside the HLA-DP2 binding site as these interact conserved areas of the HLA molecule. It thus is likely triat they peptide areas are able to increase the stability of the MHC/peptide complete.

Molecular modeis of HLA-DP2 and HLA-DP2K69 tiave been created by means of homology modeling programs. In order to generate the coordinates of the αβ. HLA-DP dimer, the crystallographic structure of HLA-DR4 has been used as a template.

Both models generated for HLA-DP2 and HILA-DP2K69 show the typical HLA class II structure where ttie ccl and βl N-terminal doma±ns are organized in β sheets flanked by two α. helices. This structure represents the peptide- binding site. The cc2 and β2 domains, instead show the typical β. sheet folding of the immunoglubulin dormains.

In order to analyze the interaction between HLA-DP and CLIP, the citrystallographic structure of HLA-DR3 presenting CLIP in -its binding site has been used. As a result, it has been possible to utilize DR3 derived CLIP coordinates for the HLA-DP2 and HLA-DP2K69, as the three molecules present the same binding core for CLIP.

The analysis of the surface of the binding regions of the two molecules (seen from above) after CLIP removal is shown in Figure 1. As expected from homology with the HLA-DR3 and the HLA-DR4 molecules, the binding sites of both molecules show four main contact points (Pl, P 4, Pβ and P9) with the peptide when the peptide is hooked to the binding site. In particular, pocket Pl accommodates the Met residue in position 91 of CLIP, P4 accommodates an Ala residue (Ala94) in position 94 of CLIE³, Pβ accommodates s Pro residue (Pro9β) in position 96 off CLIP and P9 accommodates a Met residue (Met99) in positi_on 99 of CLIP. By analyzing the models in detail, it is possible to observe that HLA-DP2 shows a deeper and more negatively charged P4 pocket as compared to HLA-DP2K69. Furtherr, the CLIP residue "that is accommodated in P4 (Ala94) does not penetrate into the pocket and shows the same conformation in both molecular HLA-DP models. These results suiggest that the diffferences seen in P4 in HLA-DP2 compar^ted to HLA-DP2K69 are structural and not due to the interaction of Ala94 with the HLA-DP molecule. In pocket 4 of the HLA-DP2 molecule there are three negatively charged residues in positions β2β, ββδ and ββ9 and a ifzourth negatively chiarged residue is present in position. β67. The repulsive interaction occurring between these charged residues may account for the wider and deeper conformation of pocket 4. The observation that i n the HLA-DP2K69 model the substitution in position β69 of the GIu residue with a Lys residue having an opposite charge determines a collapse of pocket 4 conformation, charge and size, supports this hypothesis.

Thus, tb_e polymorphism in position β69 determi_nes a structural modification of the HLA-DP molecules, particularly ±n pockets P4 and Pβ.

In order to assess the contribution of these modifications in peptide selection, a panel of mutated CLIP peptides has been designed and synthesized wd_ th mutations in the Ala 94 and Pro 96 residues i.e., those residues specifically/ interacting with P4 and P6. Based upon the sequence SKMRMATPLLQA (SEQ ID N0:l), a series of peptide comprised of 13 amino acids has been synthesized where the peptides carry all possible amino acrid substitutions i.e., 20 substitutions in positions 94 and

96. Peptides are shown in Table 1.

Table 1:

The binding affinity of the HLA-DP2 and HLA-DPGlu69 soluble molecules for each peptide has been measured in the competition assay described in Figurre 2 and the results obtained are shown in Figure 3. Shown in Figure 3 are the IC50 values i.e., the concentration of the competitor peptide necessary to displace from tϊie binding site 50% of the CILIP peptide in biotinylated form, obtained in the competiLtion assay on HLA-DP2 or HLA-DP2K69 for each CLIP peptide mutated in the P4 position (panel A and B) or the Pβ position.

In panel J\, one can see the effect off mutations of the amino acid residue in the P4 position. : HLA-DP2 has greater affinity for CLIP peptides carrying polar or charged residues (GIn, Arg, Lys, Asn) , for non-polar aromatic residues (Trp, Phe, Tyr, His) or S-groups carrying residues (Cys, Met). Instead, non-polar aliphatic residues such as VaI and Leu, lie and Pro residues and the polare residues Thr and Ser are disfavoured. Other residues such as Asp, GXu and GIy, do not display either increased or decreased affinity.

Data shown in panel B indicate that pocket 4 of HLA- DP2K69 has lesser selectivity for amino acid residues than HLA. DP2. Zn particular, one can see hzLgher affinity for non-polar aromatic residues (Phe, Trp, Tyr, His, Pro) and for polar or charged residues, independent of the charge (GIn, A-trg, Asn, Lys, Met, GIu) . Differently from

HLA-DP2, non-polar aliphatic residues are also favored

(Leu, lie) . Otrαer residues (VaI, Ser, Cys, Thr, GIy) are slightly disfavored. Panels C and D show the data obtained with CLIP peptides mutated in position P6.

In this position, HLA-DP2 (panel C) displays increased affinity, with regard to the CLIP P6 residue, for non-polar aromatic residues (Tyr, Trp, Phe, His) and to a lesser extent for positivedly charged residues (Lys, Arg) for the polar residue Gin and for non polar aliphatic residues. Instead, the negatively charged residues (GIu, Asp), polar residues such as Thr and Ser and S-group carrying residues (Met , Cys) as well as the non polar aliphatic residues (Leu_r VaI, Ala, GIy, Asn) are disfavored . The Pβ pocket of HLA-DPK69 <panel D) displays an increased affinity for CLIP peptides that express, in comparison to Pro, non polar aromat ic residues (Phe, Tyr, Trp, His) . However, although the affinity for charged residues is partially independent of the charge itself, as affinity is increased for Arg, Lys and GIu but not for Asp, the positively charged residues seem to be disfavored. Polar residues, such as Asn, GIn, Asp do not show reduced affinity compared to the amino acid Pro. To the contrary, small residues, both polar and non polar aliphatic (Ala, GIy, Leu, Thr, lie, Ser, VaI) and residues carrying S-groups (Cys, Met) appear to be disfavored.

Based upon the data shown above, Clip derived peptides carrying P4 and P6 mutations have been designed which may exprress a widely variable binding affinity for the HLA-DPGlu€>9 positive molecules. Peptides have been synthesized using known methods for peptide synthesis either on solid support or in homo geneous phase, such as the F-MOC or "the B-MOC methodologies which are the most frequently, albeit not exclusively, used.

Table 2 shows the peptides which have been generated. The two letter codes following the "CLIP" acronym indicate the amino acid which substitutes for the native amino acids A and P in positions 94 and 96, respectively, of the CLIP amino acid sequence. Tafcsle 2 :

Table 3 , reproduced below, shows the reference peptide CLIP together wi "th a selection of peptides characterized by a high binding capacity for the HLA-DP molecule carrying GIu 69 triat is shown to be higher than that typical of Be and of a peptide with low binding capacity .

Tcufole 3

Name Sequence Details and use

SEQ I D NO : 1 CLIP SKMRJMATPLLMQA Reference peptide

SEQ ID NO : 2 CLIP-YY SKMRWYTYLLMQA Strong competitor

SEQ I D NO : 3 CLIP-QY SKMRMQTYLLMQA Strong competitor

SEQ I D NO : 4 CLIP-RF SKMRMRTFLLMQA Strong competitor

SEQ ID NO : 5 CLIP-KF SKMRIMKTFLLMQA Strong competitor

SEQ ID NO : 6 CLIP-AA SKMRMATALLMQA Weak competitor

These peptides have been synthesized in biotinylated form and utilized both in competition tests with beryllium using HLA-DP soluble molecules as well as in antigen presentation tests using lymphocytes obtained from the peripheral blood of Berylliosis patients.

The results of the competition assays carried out using HLA-DP soluble molecules with increasing concentrations of beryllium (0.3, 3, 30, 300 μM) , or the non- biotinylated peptide as control, in the presence of a fixed concentration of biotinylated mutated CLIP peptide, are shown in Figure 4. As expected, the following peptides CLIP-YY, CLIP-KF and CLIP-QY cannot be removed from their bond wi^~th the HLA-DP molecule in the presence of beryllium, contrary to CLIP and CLIP-aa which are easily removed. Thus, the binding pocket of HLA- DPGluβ9 can be occupied by high affinity peptides in a stable fashion, making HLA-DPGluβ9 unavailable for beryllium binding.

The same peptides have been thereafter assessed for their ability to block the production of IFN-γ. induced by beryllium in peripheral blood mononuclear cells, obtained from Berylliosis patients and beryllium hypersensitive subjects who were carrying the susceptibility marker HLA- DPGlu69.

Figure 5 shows the results obtained in peripheral blood mononuclear cells (PBMC) of patients with Berylliosis stimulated with Be, in the presence of increasing doses of modified CΣ-.IP proteins. As expected, the CLIP peptides and the CLIP amino acids are not able to significantly reduce the capacity of the PBMC from Berylliosis patients to produce IFN-γ~.To the contrary, the peptide CLIP-YY is able to significantly block the production of .IFN-gamma following stimulation with Be.

Although the interpretatio m of this invention in the context of the scientific theory underlying the mechanism of action of the claimed peptides is beyond the scope of this patent application, it is worthwhile to say that it is generally agreed that the giranulomatous reaction i.e., the formation of the granuloma and the ensuing immunopathology, is mediated by IFN-γ. Thus the in vitro data showing that the CLIP-Y^" Y peptide is capable of blocking the reaction of blood T-lymphocytes to beryllium, demonstrate that the peptides of this invention, all characterized, as the CLIP-YY peptide claimed in this application, by a non-polar aromatic residue in position 94 (P4), Hnave the ability to block this specific granulomatous rea_ction.

In addition, it is worth noticing that the CLIP-RF and CLIP-QY are capable of increasing the response to beryllium in Berylliosis affected subjects. This ability dLs not due to a non-specif d_c ability of the above peptides to increase IFN-γ production. In fact, the CLIP- RF and the CLIP-QY peptides were not capable of increasing IFN-γ production when used to stimulate peripheral blood mononuclear cells from normal, beryllium unexposed subjects both in the presence and in the absence of beryllium.

Finally", it was observed that it is absolutely necessary that a group of electrodonors be present at residue 94 (P4) of the peptide, in "the region which will interact with GIu 69 in the presence of Be. In fact, the use of peptide CLIP-AR where A is not an electrodonors, does not determine an increase in the capacity to produce IFN-γ induced by Be in subjects affected with Berylliosis .

These xesults confirm the ef-ficaciousness of the peptides of the invention having a nonpolar amino acid residue in position 94 such as Y, F, W or H if used medically in the treatment of Chronic Beryllium Disease. Equally so, the results confirm thie efficaciousness of the peptides of the invention having an electrodonors residue, siαch as R, Q, K or N in position 94 as a diagnostic agent in beryllium hypersensitivity. Lastly, drug formulations containing the patent peptides can be employed in the treatment of all forms of beryllium allergy both for systemic and local administrati-on. In order to have e systemic effect the patent peptides can be included in medicinal formulations apt to either oral or parenteral administration. Parenteal drug formulations include solutions, suspensions,, liposome suspensions and emulsions where one or more of the patent peptides are dissolved, resuspended or emulsified in an appropriate fLuid. Diluents apt to perenteral . use are water, saline solutions, sugar solution, ϋnydro-alcoholic solutions, oleic diluents, polyoils like glycerol, ethylene- orr polypropylen-glicol, or any other diluent compatible with the requirement of this administration route with regard to sterility, pH, ionic strength and viscosity. In addition the pharmaceutical compositions of this patent may contain additives such as isotonic agents, sugars, polyalcohols, buffers, chelating agents, antioxidants^ antimicrobials. The patent peptides may also be prepared in solid form, for example as lyophilized material ready for reconstitution with an adequate licjuid diluent or included in formulations such as capsule, granules, tablets or any other formulation for mouth administration. A local lung therapeutic effect may be achieved using formulations apt to th_e intrabronchial administration. To this end, the patent peptides will be formulated in liquid compositions apt to the administration by aerosol or spray - Slow release compositions for oral use may be advantageously employed in the long term treatment of chronic fϊorms of disease. These may include polymers such as polylactate, poly- (meta) -acrylate, polyvinylpyrrolidone, rnethyl-cellulose, carbossimethylcellulose in addition to other known compounds. Slow release compositions Ln the format of subcutaneous implants using as a base compounds like polylactate or other biodegradable polymers could also be used. The patent peptides can be used as diagnostic reagents to reveal either in vivo or in vitro both any clinically apparent disease caused by beryllium and any other form of latent beryllium sensitization. Any composition chosen among those for parenteral use could be used in vivo to enhance the manifestations due to the allergy to beryllium, hence to diagnosis it. The use of the patent peptides in vitro diagnostic tests is preferred however. In these tests, the diagnostic reagent including one or more of the patent peptides will be any aqueous solution compatible with the biological reagents used in the test. These pharmaceutical compositions and the diagnostic reagents of the patent can be prepared in accordance to the current methods well known in pharmaceutics. The diagnostic methods of the patent include steps where cells capable off reacting in a revealable manner against beryllium are exposed to beryllium salts in the presence of one o> r more diagnostic agents of the patent. The response obtained in the presence of the patent peptides will Ioe compared to the values obtained from control subjects. Cellular models that can be employed in the diagnostic test include the measure of T-cell proliferation in response to beryllium or the periphexal blood mononuclear cell response with the production of interferon gamma after similar stimulation. Lastly, the properties characterizing the patent peptides can be exploited to design methods for the selection in vitro of other pep-tides that may be capable either of blocking or of enhancing the response to beryllium of berylliosis patients and beryllium sensitized subjects. Such methods entail experimental steps where the candidate peptide Ls assayed in the presence of tϊie HLA-DPGlu69 molecule to measure its binding affinzLty; where the peptides having binding affinity greater than the natural CLIP peptide are selected; where the selected peptides are tested in competition assays against beryllium for the binding to the HLA-DPGlu69 molecule; where those peptides capable either of blocking or of enhancing the ability of beryllium to form the HLA-DPGlu69/Beiryllium/peptide are selected.

Claims

1. Peptides derived frtrom the CLIP peptide through mutations in positions 94 and/or 96 of the natural polypeptide sequence of CLIP and having an increased affinity for the HLA varzLant DP-Glu69 molecules vszith respect to the natural CLIP peptide for use in a therapeutic or in vivo diagnostic method.

2. Peptides derived from the CLIP peptide according to claim 1, having a sequence of 13 or more amino aoids including the binding frame zone located from position. 91 to 93 of the CLIP peptide.

3. Peptides according to claim 1 or 2 in which the residue Ala in position 94 of the natural CLIP peptide is substituted with a residue selected from Q, R, K, N_r W, F, Y, H, C, M, E or the residue Pro in position 96 of the natural CLIP peptide substituted with a residue selected from Y, W, F, H, K, R, Q, I or both positions 94 and 96 substituted as above indicated.

4. Peptides according to anyone of claims 1 to 3 for use in therapeutic treatment of Chronic Granulomatous

Beryllium Disease or in s diagnostic method for the diagnosis of beryllium hypersensitivity.

5. Peptides accordion "to claim 4 for the use in the therapeutic treatment of Chronic Beryllium Disease in which the residue Ala in position 94 is substituted toy a residue having an aromatic substitution selected from Y, F, W, H.

6. Peptides according" to claim 4 for use in a diagnostic method for the diagnosis of beryllium hypersensitivity in which the residue Ala in position 94 is substituted by an elect rodonor residue selected from R, Q, K, N.

7. Peptides derived fztrom the CLIP peptide in which the residue Ala in posit ion 94 of the - natural CLIP peptide is substituted by a. residue selected from Q_r R, K, N, W, F, Y, H, C, M, E and the residue Pro in position 96 of the natural CLIP peptide is substituted b>:y a residue selected from Y, W, F, H, K, R, Q, IE.

8. Peptides according to claim 7 selected from CLIP- YY, CLIP-YW, CLIP-YR, CLIP-YK, CLIP-YF, CLIP-QY, CLIP-QY, CLIP-QR, CLIP-QK, CLIP-QF, CLIP-WW, CLIP-WY, CLIP-WR, CLIP-WF, CLIP-QR, CLIP-QK, CLIP-QF, CLIP-WW, CLIP-WY, CLIP-WR, CLIP-WF, CLIP-RF, CLIP-RY, CLIP-RW, CLIP-RR, CLIP-RK, CLIP-KF, CLIP-KY, CLIP-KW, CLIP-KR, CLIP-FY, CLIP-FW, CLIP-FR, CLIP-FF, CLIP-HY, CLIP-HW, CLIP-HR, CLIP-HK, CLIP-HF.

9. Peptides, according to anyone of claims 1 to 8, in biotinylated form or otherwise modified or associated with a vehicle or immunomodulators or conjugated with lipids or polyethylene glycol.

10. Methods of preparation of peptides according to anyone of claims 1 to 9 by means of chemica 1 synthesis in homogeneous or heterogeneous phase.

11. Pharmaceutical composition comprrising one or more peptides according to anyone of claims 5 or 8 in a pharmacologically acceptable excipient .

12. Pharmaceutical composition according to claim 10 in liquid form suitable for parenteral, sprrray or aerosol somministration .

13. Diagnostic agent comprising one or more peptides according to anyone of claims 6 cor 8 in a pharmacologically acceptable excipient.

14. Method of preparation for pharmaceutical composition according to claim 10 or 11, comprising the phase in which the peptide or peptides deprived from the natural CLIP protein are formulated in a pharmacologically acceptable excipient.

15. Method of preparation for pharmaceutical composition according to claim 10 or 11, comprising the phase in which the peptide or peptides deprived from the natural CLIP protein are formulated in a pharmacologically acceptable excipient.

16. Diagnostic method in vitro for the diagnosis of Beryllium hypersensitivity comprising the fTollowing steps of : a) putting the lymphocytes ( PBlMC) of diagnosed subjects in contact with beryllium in the presence of a peptide derived from the CLIP pep-tide and able to increase the capacity of the lymphoproL iferative response or of the production of cytokines in subjects with Berylliosis or Beryllium hypersensitivity, b) comparing the response to beryllium in the presence of peptides derived from the CLIP peptide with the response in healthy subjects.

17. Method according to claim 16 in which the lymphocytes are put in contact with beryllium in the presence of a peptide according to claim 6 or 8.

18. Method of selection of peptides derived from the CLIP peptide able to block the response to beryllium in patients affected with Chronic BerylILium Granulomatous Disease comprising the phases of: a) putting the candidate peptide in contact with the HLA-DP Gluβ9 molecule in order to measure the binding capacity; b) selecting those peptides that have an affinity higher to that of the natural CLIE³ peptide; c) putting the selected peptid.es in competition with Be ions for binding with ^"the HLA-DP Gluβ9 molecule; d ) selecting the peptides capab ILe of blocking the capacity of beryllium to form th e Be : peptide : HLA- DPGluβ9 complex .

19 . Method of selection of peptides derived from the CLIP9 peptide able to increase the response to beryllium in patients affected with Chronic Beryllium Granulomatous Disease comprising the phases of : a ) putting the candidate peptide in contact with the HLA-DP Glu69 molecule in order to measure the binding capacity; b ) selecting those peptides tha ^~t have an affinity higher to that of the natural CLIE³ peptide; c) putting the selected peptides in competition with Be ions for binding to the HLA-DP Glu69 molecule; d) selecting the peptides capable of increasing the capacity of bery÷Llium to form the

Be: peptide :HLA-DPGlu69 complex.

20. Use of peptides according to any of the claims 1 to 9 for the preparation of a medication for the treatment of Chronic Beryllium Granulomatous Disease.

21. Use of peptides according to any of the claims 1 to 9 for the preparation of a diagnostic agent for the in vivo diagnosis of Chronic Beryllium Granulomatous Disease .