Background of the invention

The present invention relates to a family of potential drug compounds binding to cellular receptors which are capable of recognizing carbohydrates - preferably mannose and fructose and their derivates, such as mannose-6-phos- phate (M6P) or fructose-1-phosphate (F1P). The receptors in question are either able to recognize these carbo¬ hydrate moieties alone or in association with other oligo- and/or polysaccharide structures which might or might not be coupled to proteins, peptides or other non-peptidic compounds. An example of such a receptor is the mannose-6- phosphate (M6P) receptor.

Introduction

Mannose binding proteins are known to participate in many important physiological as well as pathophysiological mechanisms including inflammatory processes.

The lysosomal proteolytic enzymes are transported between the subcellular compartments in the cell by M6P binding receptors (Kornfeld et al., Ann. Rev. Cell Biol. 5_, 483- 525 (1989)). Such enzymes are known to play important roles in inflammatory processes, and it has been shown that animal models of allergic encephalomyelitis can be inhibited by various phosphosugars, particularly M6P. This sugar specificity suggests that inhibition may be due to depletion of lysosomal enzymes on the cell-surface of the lymphocytes. These enzymes are essential for the passage of lymphocytes across the vascular endothelium and the entry of lymphocytes into the central nervous system

parenchyma (Willenborg et al., FASEB. J. 3, 1968-1971 (1989)). Also it has been shown that administration of M6P to rats suffering from adjuvant arthritis inhibits joint inflammation, possibly by a similar mechanism (Willenborg et al., Immun. Cell Biol. 70, 369-377 (1992)).

There is also evidence suggesting that high-mannose con¬ taining oligosaccharides play a role in the regulation of the immune response (Muchmore et al., J. Leukoc. Biol. 48, 457-464 (1990)). The T helper cells exert important central regulatory functions in the immune system and play as such an important role in many diseases involving the immune system, such as rheumatoid arthritis, diabetes mellitus, transplant rejection etc.

T cells recognize protein antigens presented on the sur¬ face of the so-called antigen presenting cells. They do not recognize, however, native antigen per se. Rather, they appear to recognize a complex ligand consisting of two components, a "processed" (fragmented) protein antigen and a Major Histocompatibility Complex class II molecule; see Werdelin et al., Immunol. Rev. 106, 181 (1988).

The antigen fragments are produced intracellularly in an acidic compartment, most probably the lysosome itself; see McCoy et al., Immunol. Rev. 106, 129-147 (1988); Mouritsen et al., J. Immunol. 148, 1438 (1992); Peters et al., Nature 349, 669 (1991). Thus, if the supply of lysosomal enzymes from the Golgi complex to the lysosomes somehow is blocked, the processing of protein antigens will be inhibited leading to a decrease in the amount of presented antigen fragments on the surface of the antigen presenting cell. This may eventually lead to an inhibition of the T cell response towards the antigens.

A blockade of the transport mediated by the M6P receptor using the compounds according to the present invention can lead to such an immunosuppressive effect. There are other examples of immunosuppressive drugs acting on the antigen presenting cells. Oxychlorokin is known to raise the intracellular acidic lysosomal pH leading to inhibition of the lysosomal proteolytic enzymes which have optimal activity at low pH values (Ziegler et al., Proc. Natl. Acad. Sci. USA 7!J, 175-182 (1982)). Oxychlorokin is also known to be a potent anti-malaria drug, so the compounds according to the present invention may also be effective against this disease.

The powerful immunosuppressive drug deoxyspergualin is also believed to mediate its effect via the antigen presenting cells. Deoxyspergualin has been shown to bind itself to the constitutively expressed heat shock protein HSc70 (Nadler et al.. Science 258, 484-486 (1992)). Heat shock proteins are believed to participate actively in the regulation of antigen processing.

M6P acts on two different receptors: a cation dependent (CD) low molecular weight receptor and a cation indepen¬ dent (CI) high molecular weight receptor. The CI-M6P receptor is identical to the insulin-like growth factor II receptor (IGF-II receptor) which is considered to be important in the regulation of cell growth. This receptor is also believed to be an important receptor in the patho- genesis of breast cancer. Proteins which contain M6P modified polysaccharides are able to inhibit the inter¬ action of IGF-II with the CI-M6P receptor (Vignon et al., Breast Cancer Res. Treat. 22, 47-57 (1992)). For that reason it is likely that compounds which mimic such poly¬ saccharides would affect the binding of IGF-II and as such could affect breast cancer cells.

The technical field

The M6P receptor as well as other carbohydrate binding receptors are known to bind most avidly to complex oligo- saccharides whereas the receptor affinity to e.g. the M6P molecule alone is at least 1000 fold lower (Vignon et al., supra). The M6P receptor only recognizes the two terminal M6P molecules whereas the rest of the complex oligosaccha- ride molecule functions as a non-specific, although important, scaffolding part or back-bone which serves to orientate the M6P molecules properly for recognition by the receptor. The exact chemical structure of the oligo- saccharide, which binds optimally, is generally unknown and would in any case be very complicated to synthesize chemically compared to the present compounds.

The present invention relates to the surprising discovery that the said oligosaccharide back-bone structure or scaf¬ folding part can be substituted with a different molecular entity, e.g. a peptide template. These compounds bind avidly to e.g. the M6P receptor and will be able to exert the anti-inflammatory and immunosuppressive effects described above.

More specifically, the invention concerns a family of carbohydrate-containing compounds which contain mannose and/or fructose or derivatives thereof, and which bind to carbohydrate binding receptors such as the mannose-6- phosphate (M6P) receptor. The compounds of the invention are characterized by the general formula

R ¹ R ² R ³

I I I

A ^X A ² -(A ³ -)_.A ⁴ -(A ⁵ -) _m A ⁶ A ⁷ (I)

wherein

R 1, R2 and R3 independently are selected from

a) the group of natural L- or D-monosaccharides: Glc, Man, Gal, Fuc, Rha, GlcNAc, GalNAc, Fru and Neu5Ac or oligosaccharides thereof or phosphorylated or sulfated mono- or oligosaccharides thereof, or

b) chemically modified mono- or oligosaccharides as described in (a), which are derived by atom or group substitution,

A 1 and A7 independently are selected from the group con¬ sisting of -H, -OH, - H ₂ , D- or L-amino acids, peptides, glycopeptides, peptidomimetics and oligonucleotides,

A 2, A4 and A6 independently are selected from the group of

D- or L-hydroxy amino acids, e.g, Ser, Thr, Hyl, Hyp, Tyr or D- or L-carboxamido amino acids, e.g. Asn and Gin, and

A 3 and A5 independently are selected as a variety from the group of genetically encoded or non-encoded amino acids in their D- or L-form or peptidomimetics or nucleotides, and wherein m and n independently are integers between 1 and 15,

and wherein A 5, A6 and R3 optionally are omitted,

and wherein any residue in the linear sequence A 1-A7 may be covalently linked to form a cyclic compound.

The compounds of the invention may contain mannose- 6- phosphate (M6P ) or fructose- 1 -phosphate ( F1P) or derivatives thereof according to the formulae

wherein X is CR ₂ , S, 0 or NR, where R is H, alkyl or aryl, and wherein Y and Z independently are H, SR', OR', NR.R^ alkyl or aryl, where R', R ₁ and R ₂ independently are H, alkyl or aryl, and is 0, S or NH, and that they bind to the CI-M6P/IGF receptor and/or the CD-M6P receptor as defined above.

The present compounds can be used for the treatment of inflammatory diseases or to suppress the immune response. Further, they can be used for the treatment of cancer, such as breast cancer, for the treatment of malaria and other diseases which require a normal lysosomal function, for the treatment of diseases which require a normal IGF- II receptor function and for the treatment of diabetes mellitus.

The following examples further illustrate the present invention without limiting the scope thereof.

Glycopeptides containing two mannose disaccharides phos- phorylated at the terminal 6' hydroxyl groups and linked (l,2) or α(l,6) have been synthesized. These glyco¬ peptides also contain a fluorescence probe, which allows easy monitoring in biological assays.

EXAMPLE 1

Synthesis of M6P modified glycopeptides which are ligands for the M6P receptor

The most convenient method for the synthesis of the glyco¬ peptides containing 6'-O-phosphorylated α(l,2) and/or o(l,6) linked disaccharides is to utilize phosphorylated, glycosylated threonine (or serine) building blocks for use in multiple peptide synthesis, e.g. MCPS (Meldal et al.. Int. J. Pept. Protein Res. 41., 250-260 (1993)). M6P (Man- 6-P) is the non-reducing carbohydrate unit in both the o(l,2) and α(l,6) linked disaccharides, and it has been found most convenient to synthesize a protected, phospho- rylated monosaccharide unit, which could be incorporated into both disaccharides by glycoside synthesis. Therefore phenyl-2,3,4-tri-O-benzoyl-l-thio-α-D-mannopyranoside was phosphorylated using bis(2,2,2-trichloroethyl)phospho- chloridate and pyridine. The resulting compound was con- verted into the corresponding glycosyl bromide by addition of bromine and used in silver triflate promoted glycosyla- tions of phenyl-2,3,4-tri-O-benzoyl-l-thio-α-D-manno- pyranoside and 1,3,4,6-tetra-O-acetyl-β-D-mannopyranose, respectively. The resulting disaccharides were converted into glycosyl bromides and used in silver triflate promoted glycosylations of Fmoσ-Thr-O-Pfp to give the two phosphorylated building blocks in high yields.

To find optimal cleavage conditions for the 2,2,2-tri- chloroethyl groups the peptide Ac-Thr(Bz.-.-Man-6-0- P(0)(OCH ₂ CCl ₃ ) ₂ -o(l,6)-Bz ₃ -Man)NH ₂ was synthesized. Various conditions were studied, but only deprotection in pyridine containing 10% acetic acid using Zn and AgC0~ gave quantitative deprotection. The phosphate moiety has to be deprotected before the hydroxyl groups to avoid cyclic phosphate formation and migration.

The building blocks were used in solid phase peptide synthesis, using the Fmoc strategy. Thirteen glycopeptides were synthesized by multiple column peptide synthesis (MCPS) (Meldal et al., supra) on PEGA resin (Meldal et al.. Tetrahedron Lett. 33, 3077-3080 (1992)) derivatized with 4-[o[1-(9H-fluoren-9-yl)-methoxyformamido]-2,4-di- methoxybenzyl]-phenoxyacetic acid. Fluoren-9-y1-methoxy- carbonyl amino acid pentafluorophenyl esters (Fmoc-amino acid-OPfp, 3 eqv. ) with tert-butyl (Boc and Bu ) based side chain protection and 3-hydroxy-3,4-dihydro-4-oxo- 1,2,3-benzotriazine (Dhbt-OH, 1 eqv.) as catalyst were used. Fmoc-Lys was derivatized at the e-amino group by reaction with Boc-anthranilic acid-ODhbt ester (1 eqv.) and converted into the Pfp ester with dicyclohexyl carbo- diimide (1 eqv. each). The Fmoc-groups were removed with 20% piperidine in dimethylformamide. The peptides were cleaved off the resin with 95% trifluoracetic acid and purified by preparative HPLC (overall 50% yield). The glycan part was deprotected as described above and the glycopeptides were repurified (overall 50-60% yield). The phosphate moieties were deprotected as described above, the acyl groups were removed with hydrazine in chloro- form:methanol 1:4 and the glycopeptides were purified by gel filtration and HPLC. All the peptides contain the fluorescence probe anthranilic acid linked to the N ^e of Lys. The CI-M6P receptor is known to contain two binding sites for M6P. For that reason peptides containing two mannose disaccharides were also synthesized.

The following peptides have been synthesized and characte¬ rized:

Ac-Thr-Lys-Thr-NH,-, Ac-Thr-Gly-Lys-Gly-Thr-NH, (MKC 6.43) (MKC 6.34) ^Δ

** **

Ac-Thr-Lys-Thr-NH, Ac-Thr-Gly-Lys-Gly-Thr-NH, (MKC 6.25) ^Δ (MKC 6.32)

** **

Ac-Thr-Lys-Thr-NH, Ac-Thr-Gly-Lys-Gly-Thr-NH, (MKC 6.27) (MKC 6.33)

** ** ** **

Ac-Thr-Lys-Thr-NH, Ac-Thr-Gly-Lys-Gly-Thr-NH, (MKC 6.26) (MKC 6.31)

** **

Ac-Thr-Lys-Gly-Thr-NH, Ac-Thr-Lys-Thr-NH, (MKC 6.29) ^Δ (MKC 6.08) ⁴

** Ac-Thr-Lys-Gly-Thr-NH, Ac-Thr-Lys-Thr-NH,

(MKC 6.30) (MKC 6.09) ⁴

** **

Ac-Thr-Lys-Gly-Thr-NH, (MKC 6.30) ^Δ

*: P-6-D-Man-α(l,6)-D-Man- **: P-6-D-Man-α(l,2)-D-Man-

EXAMPLE 2

The binding of M6P modified peptides to purified bovine CI-M6P receptor

96-well microtiter plates were coated overnight at 4 °C with the M6P containing polysaccharide, phosphoraannan (2 μg/ml, 100 μl/well in 0,1 M carbonate buffer, pH 9.6). The wells were subsequently blocked for two hours with 1% bovine serum albumin in the same buffer and washed at least 3 times with PBS, pH 7.4, containing 0.1% Triton X- 100 and 0,5 M NaCl (washing buffer).

M6P containing compounds as well as glucose-6-phosphate were added to the microtiter wells in 5-fold dilutions in a 50 mM imidazole buffer, pH 6.5, and co-incubated for two hours with 80 αg/ml purified CI-M6P receptor. After a thorough washing procedure, as described above, followed a one hour incubation with a rabbit anti-CI-M6P receptor antiserum diluted 1:1000 in washing buffer. After repeated washing the wells were incubated for one hour with a 1:1000 dilution of horse radish peroxidase (HRP) labelled polyclonal porcine anti-rabbit serum. The detection of bound HRP was performed using the colorigenic substrate o- phenylene diamine and H ₂ 0 ₂ and the absorbance was read at 490 nm.

The results are shown in Fig. 1. A strong specific inhibi¬ tion by the compound MKC 6.26 of the interaction between the CI-M6P receptor and the solid phase bound M6P is observed. As expected, free M6P and F1P also inhibit this interaction although to a much lesser extent.

EXAMPLE 3

The use of M6P modified glycopeptides for suppression of antigen processing in antigen presenting cells

The following experiment illustrates the effect of M6P or F1P containing glycopeptides on antigen processing: Anti¬ gen presenting cells (APC) of e.g. the monocyte/macrophage cell lineage eHi are incubated with various amounts of glycopeptides which bind to the M6P receptor with high affinity. Also a suitable protein antigen, e.g. ovalbumin (OVA), is added to the APC. After various periods of incubation OVA the OVA specific T cell hybridoma DO.1.11 which recognizes the OVA peptide 323-339 is added to the APC.

The cell/peptide mixture is cultured using conventional cell culture techniques for at least 24 h in a humidified atmosphere containing 5% CO,. The cell supernatant is harvested and the content of 11-2 id determined as a measure of specific T cell recognition of processed OVA.

This assay shows that the presence of said glycopeptides will inhibit said specific T cell response due to inhibi- tion of the antigen processing of OVA.

EXAMPLE 4

The use of M6P modified glycopeptides for inhibition of inflammatory responses

The following experiment illustrates the effect of M6P or F1P containing glycopeptides on inflammation: The glyco¬ peptides described in Example 1 are administered to rats suffering from adoptively transferred adjuvant arthritis or from allergic encephalomyelitis. The peptides are

administered intraperitoneally.

Histological examination of joint tissue from glycopeptide treated rats reveals reduced inflammatory infiltrates into the synovium and surrounding tissue compared to non- treated rats. Similar beneficial effect can be observed when examining brain tissue from rats suffering from allergic encephalomyelitis which have been treated with said glycopeptides.

EXAMPLE 5

The use of M6P modified glycopeptides for inhibition of the growth of tumor cells

The growth of Balb/c 3T3 tumor cells which contain IGF-II receptors is stimulated with recombinant IGF-II. This can bbee ddeetteecctteedd bbyy mmeeaassuurriinngg tthe incorporation of H-labelled thymidine into these cells.

Addition of the glycopeptides described in Example 1 will inhibit this stimulation of 3T3 cells by IGF-II.