WO2001035983A1

WO2001035983A1 - Pharmaceutical composition for treatment of diarrhea

Info

Publication number: WO2001035983A1
Application number: PCT/SE2000/002241
Authority: WO
Inventors: Lars G. Burman; Sture Karlsson; Thomas ÅKERLUND
Original assignee: Karolinska Innovations Ab
Priority date: 1999-11-16
Filing date: 2000-11-15
Publication date: 2001-05-25
Also published as: AU1748001A; EP1229926A1; SE9904132D0; CA2402811A1

Abstract

A pharmaceutical composition comprising a colon-specific delivery system and cysteine, cysteine derivatives and/or cysteine-containing compounds, is described. Examples of the active ingredients are cysteine, acetylcysteine, cystine and glutathione. The cysteine, cysteine derivatives and cysteine-containing compounds may be used in a medicament for therapeutic, prophylactic and relapse prophylactic treatment of Clostridium difficile associated diarrhea/disease. A method of therapeutic, prophylactic and relapse prophylactic treatment of Clostridium difficile associa ted diarrhea/disease is also disclosed.

Description

Pharmaceutical composition for treatment of diarrhea.

The present invention relates to a pharmaceutical composition for treatment of diarrhea More precisely, the composition compπses a colon-specific delivery system and an active ingredient for therapeutic, prophylactic and relapse prophylactic treatment of

Clostridium difficile associated diarrhea/disease (CDAD) in both man and animal The active ingredient of the composition is selected from cysteine, cysteine dervatives, cysteine- containmg compounds and mixtures thereof Background Clinical aspects

Clostridium difficile is a spore forming bacteπum and part of the normal large bowel flora of many healthy individuals, and certain wild and domestic animals including horses In 1978 it was recognized as the cause of life threatening large bowel infection m humans, pseudomembraneous colitis (PMC), and resulting in severe diarrhea and occasionally rupture of the bowel Later it was found that C difficile may also cause less severe forms of diarrhea (See review, Wilcox, M H et al)

CDAD is seldom spontaneously occurring, but rather the result of a distmbance of the intestinal flora caused by medication (laxatives, enema, antimicrobial agents) or procedures like surgery or endoscopic examination of the lower bowel. The dominating cause of CDAD is antibiotic therapy (90-95% of the cases), which may lead to suppression of part of the micro flora so that C difficile or its spores gets a growth advantage and can dominate the flora and cause disease The mam virulence factors of C difficile are two similar toxins, named A and B, which attack and destroy the mucous membrane cells of the large bowel CDAD can be treated with either of two antibiotics, metromdazole or vancomycin One problem is that metronidazole-resistant strains of C difficile have started to appear and another, that vancomycin can cause resistance in other potentially pathogenic intestinal bacteπa (l e enterococci) Some strains of vancomycin-resistant enterococci are resistant to all known antibiotics, which has caused world wide concern and lead to international recommendations to avoid the use of vancomycin for CDAD Another problem is that specific antibiotic therapy of CDAD further disrupts the bowel flora Because the spores of C difficile cannot be killed and eliminated by specific antibiotics the therapy results in one or more relapses of the disease m 20% of the patients After an attack of CDAD the immunity against the C difficile toxins is often poor, contπbuting relapses, and otherwise short (only 6 months) making the patient susceptible to later attacks. Also, there is currently no vaccine against CDAD available. In summary, there is a lack of new alternative strategies for prophylaxis and therapy of CDAD. Epidemwlogical and economical aspects

Duπng the last two decades there has been an alarming increase in the incidence of CDAD in the industrialized world and especially among elderly hospitalized patients For example, in the UK the incidence increased 8-fold duπng 1990-1994 and in our nationwide survey in Sweden (population approx. 9 millions) m 1995 over 6 000 cases were reported, up from 500 in 1983. As 6 000 cases corresponds to 8-9 000 episodes, CDAD had become 2-3- fold more common than all domestic cases of bacteπal or parasitic diarrhea taken together in Sweden {Salmonella, Shigella, E coli Yersinia, Campylobacter, amoeba, Giardia) CDAD episodes occur mainly m hospitals (over 75%) leading to prolongation of the hospital stay Therefore, CDAD has become a costly problem in modern health care. It is estimated that of all the money spent on antibiotics in Swedish hospitals an additional 25-50%) is spent on taking care of the resulting cases of CDAD.

The reason for the dramatic increase m CDAD is not known. The most probable explanation is the continuing introduction of new and more powerful antibiotics that disturb the intestinal flora much more than e g penicillin. An aging population, patients with more underlying diseases than before and the ever increasing complexity of diagnostics and therapy may also play a role. Furthermore, due to their diarrhea. CDAD patients spread large numbers of spores, that are naturally resistant also to disinfectants and thus make C difficile very contagious, especially in hospital environments. A new problem is that CDAD may occur after antibiotic therapy also in domestic animals. In horses the amount of fluids lost via the diarrhea is so large that it cannot be replaced, leading to death of the animal. Thus, also for today's highly pπced horses used for breeding, racing and πding, which are frequently given antibiotics, CDAD has become a dangerous and costly problem spreading in stables and veterinary hospitals Toxin production in C difficile The two toxms A and B are similar m structure and are encoded by two genes closely located on the chromosome of C. difficile. The production of the toxms is coordmately regulated, i.e. the same factors apparently control the activity of both genes at the same time. Thus, the toxm A and toxin B genes can be either switched off, or weakly or strongly switched on, depending on the milieu and growth conditions of the organism. To what extent laboratory conditions simulate the natural environment in the antibiotic maltreated large bowel rmcrofiora is not clear.

Effects ofamino acids on toxin production in C difficile

It has previously been demonstrated that a mixture of nine ammo acids (cysteine, glycme, isoleucme, leucme. methionme, proline, threomne, tryptophan and valme,) dramatically suppressed toxm production in C difficile VPI 10463 growing in peptone yeast extract (PY) medium by about 100-fold, whereas a mixture of eight other ammo acids (alamne, arginme, aspartic acid, histidine, lysme, phenylalanme, seπne, and tyrosine) had no effect (Karlsson, S. et al.). Description of the invention

The present invention provides a new alternative strategy for prophylaxis and therapy of CDAD. The invention is based on the novel and surprising finding that only one single ammo acid is needed for down-regulation of the toxm production in C difficile, namely cysteine. Cysteine deπvatives and cysteine-containmg compounds function as well. The exact mode of action of these molecules in the bacteπal cell is currently not known but may be the result of altered bacterial metabolism.

Thus, one aspect of the invention is directed to a pharmaceutical composition compπsing a colon-specific delivery system and at least one active ingredient selected from cysteine, cysteine deπvatives and cysteme-contammg compounds. The colon-specific delivery system compπsed by the pharmaceutical composition of the invention may be any suitable system dehveπng the active ingredient to the colon of the individual human being or animal.

Colon-specific delivery systems are known in the art, see e.g. a review article of Rubinstein A. For example, polysacchaπde preparations are considered to be suitable coatings for colomc drug delivery, see e.g. a review article of Hovgaard L. and Brondsted H., or

Adkin D.A. et al. Specific examples of recently published colomc drug delivery systems are an oral tablet formulation using guar gum as the earner molecule (Prasad Y.V. et al), and chitosan microcores enclosing the drug and entrapped within acrylic microspheres (Lorenzo- Lamosa M.L. et al). Also, oral preparations consisting of capsules coated with an acrylic based resm (Eudragit) for drug delivery to the human colon have been descπbed (Dew M.J. et al).

In a preferred embodiment of the invention the cysteme is selected from free D-, L- and DL- cysteme and the cysteine deπvative is selected from N-substituted cystemes and salts and esters of these cystemes, and the cysteine-containmg compound is selected from di-,

SUBSTITUTE SHEEi (RULE 26) tπ- and polypeptides containing cysteine in the molecule and salts and esters of these compounds The polypeptides preferably contain a high proportion of cysteine residues and are usually peptides with 10 or less ammo acid residues such as 8 or less, and normally 6 or less Both D- and L-cysteine are useful in down-regulatmg the toxin production in C difficile, but L-cysteme is the more potent one

In a presently most preferred embodiment the free cysteine is L-cysteine, the N- substituted cysteine is acetylcysteme, and the di- and tπ- peptides are selected from cystme and glutathione (γ-Glu-Cys-Gly) Another aspect of the invention is directed to the use of at least one of cysteme, cysteine deπvatives and cysteine-containmg compounds for the production of a medicament for therapeutic, prophylactic and relapse prophylactic treatment of Clostridium difficile associated diarrhea/disease

Also in this aspect of the invention, in an embodiment the cysteine is selected from free D-, L- and DL- cysteine and the cysteme deπvative is selected from N-substituted cystemes and salts and esters of these cystemes, and the cysteme-contammg compound is selected from di-, tπ- and polypeptides containing cysteine m the molecule and salts and esters of these compounds In a preferred embodiment the N-substituted cysteine is acetylcysteme, and the di- and tπ-peptides are selected from cystine and glutathione (γ-Glu- Cys-Gly)

Yet another aspect of the invention is directed to a method of therapeutic, prophylactic and relapse prophylactic treatment of Clostridium difficile associated diarrhea'disease compπsing colomc administration of at least one of cysteine. cysteme deπvatives and cysteme-contaimng compounds in an amount sufficient for down-regulatmg toxm production in C difficile in vivo

The colomc administration may be performed by any suitable route, such as by oral administration of colon-specific tablets or capsules containing the active ingredient, or by rectal administration of suppositoπes, enemas or foams containing the active ingredient

The amount of active ingredient will be calculated from research data and will be decided by the manufacturer, and the physician or veteπnary depending on the condition, size and expected effect in the individual to be treated.

Here again, in an embodiment the cysteine is selected from free D-, L- and DL- cysteme and the cysteine deπvative is selected from N-substituted cystemes and salts and esters of these cystemes. and the cysteine-containmg compound is selected from di-, tn- and polypeptides containing cysteine in the molecule and salts and esters of these compounds

In a preferred embodiment of this method aspect of the invention the cysteine is L-cysteine, the N-substituted cysteine is acetylcysteme, and the di- and tπ- peptides are selected from cystme and glutathione (γ-Glu-Cys-Gly).

The invention will now be illustrated with reference to the drawings and descπption of expeπments, but it should be understood that the scope of protection is not limited to the expeπments made. For example, the pharmaceutical composition of the invention may contain other active or inactive ingredients than those specifically mentioned without departing from the spiπt of the invention. Description of the experiments and the drawings

Fig. 1. Toxm yields (U/ml) in anaerobic 48 h 37 °C C difficile cultures in peptone yeast (PY) extract medium and PY supplemented with a mixture of ammo acids Where not otherwise indicated, the concentration of ammo acids added to the media were (g/1), cysteine 0.5, glycme 0.1, isoleucme 0 3, leucme 0 4, methionme 0.2, proline 0 3, valme 0.3, threomne 0.2 and tryptophan 0.1. Toxin A plus B was measured on sonicated culture samples (bacteπa disrupted by ultrasound to release all toxm) using the Ridascreen C. difficile Toxm A/B enzyme lmmunoassay (EIA) kit (r-Biopharm) according to the manufacturer's instructions. A microtiter plate reader (Labsystems Multiscan MCC/340) was used to monitor the absorbance at 450 nm (A^o). An A4 ₀ value of 1.0 was defined to correspond to one unit (U) of toxm. Box plots of (A) six high toxin-producmg reference strains (CCUG 4938, CCUG 8884, CCUG 9004, CCUG 9018, CCUG 19126NPI 10463 and CCUG 20309) and (B and C) 28 recent clinical isolates of C difficile. Boxes enclose 50% of values with lower and upper limits representing 25% and 75% of values, respectively, and median values are indicated. Bars extending from top and bottom of boxes represent values ± 1.5 x box length. All other values are indicated by open circles; values >600 U/ml are given m parenthesis and indicated by arrows (not to scale)

Fig. 2. Toxin yields (U/ml) in anaerobic 37 °C 48 h cultures of (A) C difficile strain VPI 10463 grown in PY or PY supplemented with cysteine, prolme or the seven ammo acids glycme, isoleucme, leucme, methionme, valme, threomne and tryptophan, (B) a high toxin producing clinical isolate of C difficile grown in PY or brain heart infusion (BHI) medium ± proline or cysteine. Ammo acid concentrations are given in the legend to Fig. 1 Each bar represents mean and standard error of three expeπments (* indicates that the toxm yield was < 0.2 U/ml)

Fig. 3. Toxin yield (U/ml, squares) and cell yield (optical density, OD₆₀o, circles) in 48 h C difficile VPI 10463 cultures grown in PY with increasing concentrations of vaπous ammo acids, thioglycolate and cysteme deπvatives Where indicated, the compound supplemented inhibited growth or was not soluble Each value represents mean and standard error of three expeπments

Fig. 4 Effect of cysteme on toxin production in C difficile incubated in fecal suspensions One gram of feces from a healthy human donor was inoculated into 5 ml phosphate buffered salme (PBS) containing 10⁸/ml C difficile VPI 10463 and increasing concentrations of cysteine The cultures were incubated anaerobically at 37°C, and ahquots were withdrawn at different time-points for measurements of toxm levels and bacteπal counts

(not shown) Toxin was measured by EIA (see descπption of Fig 1) and confirmed by a cytotoxin assay using McCoy cells (see below) Fig. 5 Effect of glutathione on toxin production in C difficile incubated in fecal suspensions. For details, see descπption of Fig 4

Fig. 6. Effect of cysteme on toxin production m four C difficile strains each incubated m fecal suspensions from each of three different healthy individuals (1, 2 and 3)

The reference strain VPI 10463 and three recent clinical isolates (HS-2, 8 18 and 5 36) were inoculated after addition +/-10 mM cysteine to the suspensions and toxin was measured after

48 h For details, see Fig 4

Results of experiments

A. In vitro experiments

Effects of amino acids on toxin production As mentioned earlier, a mixture of nine amino acids (cysteine, glycme, isoleucme, leucme, methionme, proline, threomne, tryptophan and valme,) has previously been shown to dramatically suppress toxin production in C. difficile strain VPI 10463 growing in peptone yeast extract (PY) medium, whereas a mixture of eight other ammo acids (alanine. argimne. aspartic acid, histidme, lysine, phenylalanine, seπne, and tyrosme) had no effect (Karlsson. S et al.)

Now it can be revealed that the nine amino acids efficiently suppress toxm production also in five other reference strains (Fig 1A) and in 28 clinical isolates (Fig IB)

For example, for the two highest toxm producing strains the toxm yields were reduced from

13 300 U/ml and 3 710 U/ml to 2 U/ml and 142 U/ml or by 99 999% and 99 6%, respectively These results confirmed that the nine amino acids mentioned above function as general down regulators of toxin production in C. difficile.

Effects of cysteine-containing compounds and proline on toxin production

When tested alone, seven of the nine toxin suppressing amino acids showed a moderate effect whereas proline and particularly cysteine showed strong toxin suppressing activities in C. difficile strain VPI 10463 (Fig. 2A). Cysteine alone efficiently lowered the toxin yields also in the 28 clinical isolates (Fig. 1C), and was active both in PY and brain heart infusion (BHI) medium (Fig. 2B). Moreover, cysteine showed a clear dose-response effect on toxin expression in strain VPI 10463; the highest toxin yield (5000 U/ml) was found at 0.33 mM cysteine and the lowest (80 U/ml) at 33 mM cysteine(Fig. 3A).

The growth (optical density, OD₆oo) of the 48 h cultures were approximately the same regardless of concentrations of e.g. cysteine, showing that the observed reduction of toxin yields were not caused by growth inhibition by cysteine (Fig.3). Control experiments (not shown) in which cysteine (33 mM) was added to sonicated culture samples (bacteria ruptured by ultrasound to liberate all toxin) showed no changes in toxin levels over 72 h, i.e. the toxins were stable over time (e.g. our 48 h growth experiments) and the enzyme immuno assay (EIA) measurements of toxin were not affected by cysteine.

Other molecules containing cysteine residues, such as glutathione (γ-Glu-Cys- Gly), acetylcysteme or cystine, also exhibited a down-regulating and dose-response effect of toxin production in C. difficile strain VPI 10463 (Fig. 3, B, C and D). Also proline showed a dose-response effect on toxin production up to 3.3 mM but not at higher concentrations when added to C. difficile cultures in PY(Fig. 3E). Alanine and serine belonging to the group of amino acids not affecting toxin production and included as negative controls had no effect on toxin production as expected(Fig 3, F and G). No impact of oxidation-reduction potential on toxin production

The oxidation-reduction potential (Eh) of the growth medium has been shown to affect the extracellular toxin levels in C. difficile (see Onderdonk, A. B. et al), and it is thus possible that cysteine exerts its toxin down-regulating activity by lowering the Eh. Increasing cysteine concentrations (0.33 to 33 mM) reduced the Eh in PY medium from -100 to -400 mV (not shown), which correlated with a lowered toxin yield (Fig. 3A). By contrast, the toxin yields in VPI 10463 cultures were largely unaffected by another reducing agent, thioglycolate (Fig. 3H), despite that the Eh was lowered from -200 at 0.33 mM to -300 mV at 10 mM (not shown; levels >10 mM were toxic for growth). The lack of regulation of toxin expression by the Eh was further supported by the fact that also cystine (i.e. the oxidized dimer of cysteine and thus non-reducmg and also found Eh inactive, as expected; not shown) caused a reduction of toxm yields comparable to that of cysteine at 3.3 mM ( c.f. Fig. 3D with 3A; cystine was not soluble at =10 mM). Thus, we concluded that Eh had no or only marginal effect on toxin production in C. difficile Mechanism of action of cysteine and related compounds

The current data indicate that these agents down-regulate toxm production in C difficile by concomitantly alteπng its energy metabolism, but their precise mode of action is not known. It appears that toxm production is switched on as a result of a stress response, and that cysteine prevents or stops this response. B. Ex vivo experiments

For toxin down-regulation to become a realistic approach to prevention and cure of CDAD it is needed for cysteme and other agents to be active also m a milieu representative of the clinical situation, i.e the human fecal flora that has been reduced by antibiotic therapy and thus allowing overgrowth of C difficile (see Background) When C difficile strain VPI 10463 was inoculated into suspensions of feces obtained from healthy donors, diluted 6-fold into phosphate buffered salme (PBS) and incubated anaerobically at 37°C, it started to grow and the toxm levels increased by about 2- fold duπng 48 h. A stopped production and even a significant reduction (by >10-fold) of C difficile toxm was observed when cysteine (Fig. 4) or glutathione (Fig. 5) had been added to the fecal cultures. The reduction below the initial values was probably caused by proteolytic degradation of the toxin by fecal enzymes and now visualized because little or no production of toxin was occurnng.

A similar effect of a moderate level of cvsteme (lOmM) was demonstrated in suspensions of faeces obtained from three different individuals each inoculated with four different strains of C. difficile, strain VPI 10463 and three recent clinical isolates (Fig 6) Suppression of toxm production was most pronounced with the strains that produced most toxin (Fig 6).

A semi-quantitative biological toxm assay based on their ability to destroy cultured human cells (cytotoxin assay) was used to veπfy the results obtained with EIA As expected, we found that the samples from fecal cultures to which only PBS and C difficile had been added (controls) were positive for active toxin, whereas those contaimng PBS, C difficile plus cysteine or glutathione were negative in the cytotoxin assay (data not shown)

Taken together, the results of the ex vivo expeπments confirmed the previous in vitro data. Furthermore, as the concentrations of cysteine and glutathione here found to effectively suppress C. difficile toxin production were the same as in vitro, the ex vivo experiments also showed that these active compounds were not inactivated by fecal components or consumed by the other fecal flora organisms. Summary of the results of the experiments A method to stop toxin production of C. difficile has now been developed. The effect is dose-dependent which enables the calculation of the concentrations of the active ingredient required in the large bowel lumen on order to "pacify" pathogenic (toxin- producing) C. difficile.

The dramatic toxm down-regulating effect of the tested compounds is general. Thus, the effect was demonstrated

A. in vitro (see Expeπments A)

- for all 6 tested reference strains, selected for high toxm production.

- for all 28 tested recent clinical isolates of C. difficile, producing varying amounts of toxin (one even more than the most active reference strains), - in both growth media tested, which like the large bowel milieu lacked oxygen (being so- called anaerobic) and glucose.

B. This effect was also demonstrated ex vivo (see Expeπments B), i.e. for all 4 C. difficile strains tested when actively growing and producing toxin in diluted normal fecal flora specimens obtained from 3 different healthy individuals and kept anaerobically during the expeπment. In all ex vivo expeπments the impact of cysteine and one of its denvatives seen on toxm production in vitro could be reproduced by using the same concentrations of the agents.

The latter result strongly indicates that down-regulation of toxin production in

C. difficile by cystem, cystem-deπvatives and cystein-containing compounds can be obtained also in vivo, when the active agent is delivered to the place of infection, i.e. the large bowel lumen. Thereby, a novel (non-antibiotic) method for prophylaxis, therapy and relapse prophylaxis of CDAD has been developed. References

Adkin D.A., C.J. Kenyon, E.I. Lerner, I. Landau, E. Strauss, D. Caron, A Penhasi, A Rubinstein, and R.I. Wilding. 1997 The use of scmtigraphy to provide "proof of concept" for novel polysacchande preparations designed for colomc drug delivery. Pharmaceutical Research 14(1): 103-7

Dew M.J., P.J. Hughes, M.G. Lee, B.K. Evans, and Rhodes. 1982. An oral preparation to release drugs in the human colon. Bπtish Journal of Clinical Pharmacology 14, 405-408

Hovgaard L. and H. Brondsted H., 1996. Cuπent applications of polysacchaπdes in colon targeting. Cπtical Reviews in Therapeutic Drug Carπer Systems. 13(3-4) 185-223.

Karlsson, S., L. G. Burman. and T. Akerlund. 1999 Suppression of toxm production in Clostridium difficile VPI 10463 by ammo acids. Microbiology 145; 1683-1693

Lorenzo-Lamosa M.L., C. Remunan-Lopez, J.L.Vila-Jato, and M.J. Alonso. 1998. Design of microencapsulated chitosan microspheres for colomc drug delivery Journal of Controlled Release. 52(1-2):109-18.

Onderdonk, A. B.. B. R. Lowe, and J G. Bartlett 1979 Effect of environmental stress on Clostridium difficile toxms levels duπng continuous cultivation Appl. Environ. Microbiol 38; 637-641.

Prasad Y.V., Y.S. Kxishnaiah. and S Satyanarayana. 1998 In vitro evaluation of guar gum as a carπer for colon-specific drug delivery. Journal of Controlled Release. 51(2-3):281-7

Rubinstein A. Approaches and opportunities in colon-specific drug delivery. 1995. Critical Reviews in Therapeutic Drug Carπer Systems.12(2-3): 101 -49.

Wilcox, M..H., Wood, M.J.,Woodford, N., Daly, P.J. 1998. Clostridium difficile infection Report of a Working Group. J Antimicrob Chemother Suppl C, 41 : 1-72.

Claims

1 ?Claims

1. Pharmaceutical composition compπsmg a colon-specific delivery system and at least one active ingredient selected from cysteine, cysteine deπvatives and cysteine - containing compounds

2. Pharmaceutical composition according to claim 1 , wherein the cysteine is selected from free D-, L- and DL- cysteine and the cysteine deπvative is selected from N- substituted cystemes and salts and esters of these cystemes, and the cysteine-containmg compound is selected from di-, tπ- and polypeptides containing cysteine m the molecule and salts and esters of these compounds

3. Pharmaceutical composition according to claim 2, wherein the free cysteine is L-cysteme, the N-substituted cysteine is acetylcysteme, and the di- and tπ- peptides are selected from cystine and glutathione (γ-Glu-Cys-Gly)

4. Use of at least one of cysteine. cysteine deπvatives and cysteine-containmg compounds for the production of a medicament for therapeutic, prophylactic and relapse prophylactic treatment of Clostridium difficile associated diaπhea/disease.

5. Use according to claim 4, wherein the cysteine is selected from free D-, L- and DL- cysteine and the cysteine deπvative is selected from N-substituted cystemes and salts and esters of these cystemes, and the cysteme-contammg compound is selected from di-, tπ- and polypeptides containing cysteine in the molecule and salts and esters of these compounds

6. Use according to claim 5, wherein the N-substituted cysteme is acetylcysteme, and the di- and tπ- peptides are selected from cystine and glutathione (γ-Glu- Cys-Gly)

7. Method of therapeutic, prophylactic and relapse prophylactic treatment of Clostridium difficile associated diarrhea/disease compπsmg colomc administration of at least one of cysteine, cysteme deπvatives and cysteme-contammg compounds in an amount sufficient for down-regulating toxin production in C difficile

8. Method according to claim 7, wherein the cysteine is selected from free D-, L- and DL- cysteine and the cysteine deπvative is selected from N-substituted cystemes and salts and esters of these cystemes, and the cysteme-contammg compound is selected from di-, tπ- and polypeptides containing cysteine in the molecule and salts and esters of these compounds.

9. Method according to claim 8, wherein the cysteine is L-cysteine, the N- substituted cysteine is acetylcysteine, and the di- and tri- peptides are selected from cystine and glutathione (γ-Glu-Cys-Gly).