AU638681B2 - Compound and method for the retardation of collagen cross-linking - Google Patents

Description

CORRECTED I VERSION PCr 33i8 8 43 *3 c, 1B WORLD INTELLECTUAL PROPERTY ORGANIZATION International Bureau INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY (PCT) (51) International Patent Classification 4 (11) International Publication Number: WO 90/06102 A61K 7/40, 7/48, 37/02 Al (43) International Publication Date: 14 June 1990 (14.06.90) (21) International Application Number: PCT/AU89/00422 (74) Agent: F.B. RICE CO.; 28A Montague Street, Balmain, NSW 2041 (AU).

(22) International Filing Date: 28 September 1989 (28.09.89) (81) Designated States: AT (European patent), AU, BE (Euro- Priority data: pean patent), CH (European patent), DE (European pa- PJ 0675 28 September 1988 (28.09.88) AU tent), FR (European patent), GB (European patent), IT (European patent), JP, LU (European patent), NL (European patent), SE (European patent), US.

(71) Applicants (for all designated States except US): COMMONWEALTH SCIENTIFIC AND INDUSTRIAL RESEARCH ORGANISATION Published IAU/AU|; Limestone Avenue, Campbell, 2601 ACT With international search report.

PEPTIDE TECHNOLOGY LIMITED IAU/AUI; 4-10 Inman Road, Dee Why, 2099 NSW (AU).

(72) Inventors; and Inventors/Applicants (for US only): GRIGG, Geoffrey, Walter (AU/AUI; 352 Burns Bay Road, Lane Cove NSW 2066 HANNAN, Gary, Noel IAU/AUJ; 32 Earl Street, Boronia Park NSW 2111 (AU).

(54) Title: COMPOUND AND METHOD FOR THE RETARDATION OF COLLAGEN CROSS-LINKING (57) Abstract The present invention provides a method for reducing or preventing collagen cross-linking in skin and/or damage to skin cell DNA. The method comprises treating the skin with a composition including an antioxidant compound. The antioxidant compound is selected from the group consisting of carnosine, homocarnosine, anserine, 3-methyl-L-histidine, L-alanyl-L-tyrosine, acyl homocarnosine, acetyl carnosine, iodo carnosine, di-iodo carnosine, anserine nitrate, carbenoxylone carnosine, analogues thereof and combinations thereof. The antioxidant compounds of choice are carnosine and homocarnasine with preference for carnosine.

WO 90/06102 (t A61K 7/40 14/1990 under INID Number (71) "Applicant (for all designated States except add "COMMON- WEALTH SCIENTIFIC AND INDUSTRIAL RESEARCH ORGANISATION (AU/AUI; Limestone Avenue, Campbell, ACT 2601 and under INID Numbers (72) "Inventors; and" and "Inventors/Applicants (for US only)", delete "COMMONWEALTH SCIENTIFIC AND IN- DUSTRIAL RESEARCH ORGANISATION IAU/AUI; Limestone Avenue, Campbell, ACT 2601 (RKfrrJ to in P( GwvttL No.0211'42. Stciion II) WO 90/06102 PC/AU89/00422 1 "COMPOUND AND METHOD FOR THE RETARDATION OF COLLAGEN

CROSS-LINKING"

Field of the Invention The present invention relates to a method for reducing or preventing collagen cross-linking in skin and/or damage to skin cell DNA. In particular the present invention relates to the use of specific dipeptides and analogues thereof which have the capability of decreasing or preventing collagen cross-linking either during aging and/or following exposure to UV radiation. The method of the present invention is also applicable for decreasing DNA damage due to UV radiation.

Background of the Invention Oxidative stress and tissue damage by active oxygen species has been suggested as the basis for a number of diseases including cancer and aging (Halliwell and Gutteridge, 1985; Harman, 1987; Saul et al, 1987).

The levels and formation of highly reactive free radicals are known to be enhanced by ultraviolet radiation. Such reactive species can subsequently react with DNA, RNA, proteins and lipids. Natural defence mechanisms to such reactive and potentially damaging species are believed to exist and numerous molecules with anti-oxidant properties have been identified in tissues (eg vitamin E, carnosine and ascorbic acid). However, the physiological role of carnosine and its analogues in vivo remains unclear.

Other roles suggested for carnosine, in addition to being an efficient singlet-oxygen scavenger, include neutralization of lactic acid (Davey, 1960), a copper chelator (Brown, 1981), an activator of myosin ATPase (Parker and Ring 1980) and a regulator of enzymes (Ikeda et al 1980).

During the aging of mammalian skin the quantity of mature collagen cross-links increases, whereas the number WO 90/06102 PCY/AU8900422 2 of immature, reducible cross-links decreases (Ames, 1983). Although the nature of cross-links in other tissues (eg. tendon) may differ, the same general trend exists (Ames, 1983; Rattan et al, 1982). The rates of change in the amounts of cross-links for various species appears to reflect the differences in life span for the different species. For instance, the amount of one mature cross-link HHL (histidinohydroxylysinonorleucine) rises in a linear fashion in human skin up to age 40 years whereas in bovine skin the amount of HHL plateaus at 4 years (Kohen et al, 1988). Cross-links arise from precursor lysine (Vizioli et al, 1983) or hydroxylysine (Boldyrev et al 1987) residues in collagen chains which are oxidatively deaminated. The aldehydes produced then condense with with similar residues to give aldols or with adjacent lysine or hydroxylysine residues to give Schiff-based compounds. The degree of cross-links in collagen also increases on UV irradiation. Thus there is a correlation between the rate of collagen cross-links and the rate of aging of skin and possibly other tissues.

Summary of the Invention The present invention consists in a method for reducing or preventing collagen cross-linking in skin and/or damage to skin cell DJNA comprising treating the skin with a composition comprising a suitable excipient in combination with an active compound, the active compound being selected from the group consisting of carnosine, homocarnosine, anserine, 3-methyl-L-histidine, L-alanyl-L-tyrosine, acyl homocarnosine, acetyl carnosine, iodo carnosine, di-iodo carnosine, anserine nitrate carbenoxylone carnosine, analogs thereof and mixtures of two or more of the foregoing.

In a preferred embodiment of the present the active compound is a naturally occurring found in human tissLe) anti-oxidant compound such as carnosine O h« n Irliin *r PCT/AhRO/n22

WU

9i /uoiu 3 -r 1 7 PYUJUOIUh 3 (?-alanyl-L-histidine).

At-present it is preferred that the active compound is carnosine or homocarnosine on a combination thereof and most preferably carnosine.

In a further preferred embodiment of the present invention the active compound is linked to another molecule, which molecule is such that the composition is improved in regard to skin permeation, skin application and tissue absorption. It is preferred that this other molecule is an amino acid or peptide.

The method of the present invention will generally involve topical application of the composition to the skin, however, the composition may be injected subcutaneously or intramuscularly or may be taken orally.

The concentration of the active compound in the composition will depend upon the route of administration and may be at the direction of a physician, however, it is e:pected that the concentration of the active compound would be in the range of 1 to 100 mg per ml of a skin cream formulation for instance, the preferred range would be from 3 to 20 mg/ml.

It is believed the method of the present invention will reduce aging of the skin by decreasing or preventing collagen cross-linking due to exposure to UV radiation or sunlight. Given that the method is also capable of preventing DNA damage as a result of UV radiation, the method of the present invention has applicability in the prevention of skin cancer.

The composition according to the present invention may include, in addition to the active peptide molecules discussed above, a non-peptide compound which can inhibit or prevent cross-linking of collagen. Such compounds which may be advantageously included in the present composition include bilirubin, carotenoids, mannitol, reduced glutathione, selenium, uric acid, vitamin A, WO 90/06102 PCT/AU8900422 4 vitamin C and Vitamin E.

Detailed Description of the Invention 7n order that the nature of the present invention may be more clearly understood, preferred forms thereof will now be described with reference to the following examples.

Example 1 Mouse Skin Experiments An example of the collagen cross-linking inhibitory activity of carnosine is shown in Table 2 and the effects of UV irradiation on total reducible species in mouse skin in Table 1.

L-carnosine is available from the Sigma Chemical Company or BDH Chemicals Ltd. Poole, England. ChromAr HPLC grade acetonitrile was obtained through Mallinkrodt Australia Pty. Ltd. and all solvents were filtered and degassed before us. KB(H 3 4 was purchased from CEA France, through the Australian Atomic Energy Commission, with a specific activity of 50-70 Ci/mmol.

Dermal fibroblasts were cultured from primary explants of tissue derived from Swiss mice. The cells were maintained throughout in Dulbecco's Modified Eagles Medium (Gibco) and supplemented with 10% foetal bovine serum (Cytosystems Pty. Ltd.) and used between second and third passages. Skin sections were also obtained from neonatal Swiss mice. Sections were trimmed of as much subcutaneous material as possible and rinsed briefly in phosphate-buffered saline before following experimental procedure. All samples were incubated in varying concentrations of L-carnosine in phosphate-buffered saline (PBS) for 1 hour at 37°C prior to UV treatment. Prior to reduction with borohydride, samples were washed extensively in PBS at 40 C to remove as much as possible of contaminating protein, glycoprotein and glycosamino glycans with minimal disruption of the collagen structure.

Both skin sections and open dishes of dermal WO 90/06102 PCAU89/00422 5 fibroblasts were exposed for three hours at 920 uW/cm using a-24 watt germicidal ultraviolet lamp. Samples were kept moist through the UV treatment.

Reduction ith KB(H3) was carried out at room Reduction with KB(H )4 was carried out at room temperature in the same PBS buffer for 1 hour in the ratio of 100:1 wet weight of sample/borohydride. The reaction was stopped by the addition of 4M acetic acid to lower pH to 3.00. Samples were then dialysed against distilled water at 4 C until free of soluble radioactivity.

Samples were hydrolyzed in Sequanal grade 6N hydrochloric acid under nitrogen for 22 hours at 1100. Each sample was then rotary evaporated to dryness twice from distilled water before being taken up in distilled water.

Hydrolysates were neutralized with 0.5M NaOH before HPLC.

HPLC was performed on a Pharmacia HPLC/FPLC system comprising a P3500 HPLC pump, LCC500 programmer, and using a heated column compartment.

Initial separations were performed on a Brownlees cm x 0.46 cm Amino-Spheri 5 column at a temperature of 35°C and an eluent flow rate of 1.0 ml/minute. Column life was considerably lengthened by the use of a Brownlee 4.6 x 30 mm Amino-Spheri 5 pre-column guard cartridge, and a Brownlee 15 x 32 mm Anion Newguard in the solvent line.

Brownlee columns were supplied through Activon Scientific Products Co Pty. Ltd. Australia.

The gradient system comprised two solvents. Solvent A contained 10 mM-potassium phosphate buffer, pH 4.3, and solvent B contained HPLC-grade acetonitrile diluted 50:7 with water. Amino acids could be separated by using a gradient programme similar to that of the prior art, but the programme was modified for separation of reduced collagen components.

Programme 1 was as follows: 1.95% solvent B for min; 2, linear gradient 95% 70% solvent B over 15 min; 3, linear gradient 70% 50% solvent B over 15 min; 4, WO 90/06102 6 PCT/AU89/00422 solvent B for 10 min; 5, linear gradient 50% solvent B over 5 min.

Fractions of volume 0.5 ml were collected directly into 20 ml scintillation vials and 5.0 ml of Amersham PCSII high-efficiency phase combining scintillant was added. Counting was done in an LKB 1215 Rack beta II instrument.

SWO 90/06102 PCf/AU9/00422 7 TOTAL REDUCIBLE SPECIES IN YOUNG MOUSE SKIN TABLE 1 Tissue Skin section Skin (exterior) Skin (exteror) Skin (interior) Skin (interior) Skin cells Fibroblasts Fibroblasts U.V. Treatment Total (3H) CPM 10000 24000 11000 27000 19000 38000 2mg samples were treated with normal light or U.V. light for 2.5 hours then reduced with KB(3H)4 and hydrolyzed plior to HPLC.

TABLE 2 REVERSAL OF UV INDUCED COLLAGEN CROSS-LINKS BY

CARNOSINE

Tissue UV Treatment Skin sections Skin cells Carnosine Total (3H) CPM 15000 23000 14000 14000 20000 Low dose (2mM) 14000 High dose (10mM)10000 Samples were treated as in Table 1 with or without protection by carnosine solution.

WO 90/06102 PCT/AU89/00422 8 Example 2 Dermal Fibroblasts Experiments Primary cultures of mouse dermal fibroblasts (MDF) were isolated from new born mouse skin and were serially passaged in DMEM medium with high glucose plus 10% foetal calf serum for no more than four passages before experimentation. Prior to UV exposure, cell layers were washed with 50 mis of PBS to remove all traces of growth medium. The MDF were then incubated with 25 mls of PBS with or without 10 mM carnosine for 1 hour at 37°C. All the procedures were carried out with minimum exposure to outside light.

At the end of the incubation period the replicate cell populations were exposed to 0,2,4 or 6 minutes of UV A light. Following irradiation the cell monolayers were scraped and transferred to centrifuge tubes containing PBS. After centrifugation the supernatant was removed and the cell pellets resuspended in 30 mls of PBS. All samples were then washed for 3 days at 4 0 C. The PBS changed twice daily.

The samples were then reduced with KB[ H] 4 and acid hydrolysed. The hydrosylates were rotary evaporated, neutralised and lyophilised before HPLC analysis (Smolenski et al, 1983 Biochem 213, 525-532). HPLC fractions of 0.5 mls were collected directly into scintilation vials and 5ml of Amersham PCS II high efficiency phase combining scintillant added. Counting was carried out in a LKB 1215 Rack Beta II. The results of these Experiments are shown in Figures 1-6.

Mouse dermal fibroblasts were scraped and washed in PBS. The isolated cell suspension was then processed for HPLC using the methods described.

The profile shown in Figure 1 is a representation of a typical non ultraviolet irradiated sample (CONTROL).

The fractions in region 40 to 90 consist of isolated 'WO 90/06102 PCMA~B'~U89/00422) 9 reducible collagen crosslinks. The difference in peax height is indicative of the amount of a particular type of crosslinking amino acid complex present in the sample.

The profile shown in Figure 2 depicts the alteration in isolated crosslinks after a 2 minute exposure to UV A light. The major peak at fraction no. 55 has disappeared and there has been a positional shift of the peak at fraction. 83 to 87.

It is speculated that these differences represent modifications of the crosslinking amino acid complexes due to interaction of free radicals generated by the action of UV on water molecules.

As shown in Figure 3 after 4 minutes exposure to UV A ;.ght the profile of reducible crosslinks is showing a large increase in crosslinking amino acid complexes in the region from fraction 55 to As shown in Figure 4, after 6 minutes of UV A exposure there is a marked change in the content of reducible crosslinking amino acid complexes present in fractions 55 to 75. This probably results from modifications of the less complex crosslinks leading to an accumulation in fractions 55 to 75 and possibly the formation of entirely new crosslinks due to the interaction of free radicals.

Figure 5 shows the results obtained when mouse dermal fibroblasts are exposed for 6 minutes to UV A in the presence of 10 mM carnosine. As can be seen there is a marked difference in the profile of isolated crosslinks in comparison to that shown in Figure 4.

Figure 6 shows a comparison of profiles produced when MDF cells were exposed to 2 minutes UV A or to 6 minutes UV A but in the presence of 10 mM carnosine. It is obvious from this overlay of profiles that carnosine has protected the MDF cells from the effects of UV A as seen by the decrease in peaks from fractions 55-75.

WO 90/06102 PcT/AU89/00422 10 Summary When 10 mM carnosine is present in the bathing solution over MDF during exposure to UV A there is a reduction in the formation of altered reducible crosslinks produced. This is exemplified by the appearance of the HPLC profile i.e. the 6 minute UV A exposure 10 mM carnosine profile resembles that of the 2 minute UV A exposure profile.

Example 3 Human Fibroblasts (MRC-5) with UV A Light Experiments The human lung fibroblast cell strain (MRC-5) was used to examine the degree of protection carnosine would give to human collagen when exposed to UV A light.

Fibroblasts are responsible for the maintenance of connective tissue in animals. They actively secrete collagen propeptides into the interstitial spaces, a proportion of which are deposited into the extracellular matrix as connective tissue collagen.

A similar protocol to that used for the MDF cells was employed in these experiments, however, the UV A exposure time was increased to 15 minutes. This was done to increase the collagen crosslinking changes that had been evident with the 6 minute UV A exposure of MDF cells. The results of these experiments are shown in Figures 7-10.

In Figure 7, populations of MRC-5 at similar cell numbers to that of MDF were scraped and processed for HPLC. These cells were not treated with UV A i.e. CONTROL the profile was similar to that produced l. the MDF cells but there were differences. The major collagen crosslinking peaks were in the region from fractions 40-90.

Figure 8 shows the HPLC profile obtained when cells were exposed to 15 minutes UV A light in the presence of 10 mM carnosine. The profile is similar to that of the control (Fig. 7) and very different from the 15 minute UV A exposure in the absence of carnosine WO 90/06102 PCrAU8900422 11 (Fig. 9).

Figure 9 shows the profile produced when MRC-5 cells were exposed to 15 minures UV A light. The region from fractions 40 through 90 are exten'" ,ely increased indicating a putative incorporatiun of newly produced crosslinking complexes into the cellular collagen.

Figure 10 shows an overlay of HPLC profiles isolated from the MRC-5 control and from the MRC-5 after 15 minutes exposure to UV A in the presence of 10 mM carnosine.

Again, there is a great deal of similarity between the two profiles indicating that carnosine is protecting the collagen from free radical attack.

Summary The results from these experiments also supports the premise that carnosine protects collagen against UV A induced crosslinking. This is shown by the remarkable similarity between the MRC-5 control HPLC profile and that of the MRC-5 cells that have been exposed to 15 minutes UV A in the presence of carnosine. Also, the marked changes in the profile obtained from the MRC-5 cells that had UV A exposure but no carnosine present would indicate that carnosine is extremely effective in its protection of collagen against crosslinking. These experiments which had a 150% increase in UV A exposure time still demonstrated a greater than 70% protection with 10 mM carnosine.

Example 4 Rat Tail Tendon Experiments Isometric melting has been widely used to determine a number of age related changes in collagen (Mitchell and Rigby, 1975 BBA, 393, 531-541). Robins and Bailey (1975 Biochem 149, 381-385) have proposed that the density of collagen crosslinks is constant with time but as ageing occurs, the labile reducible aldimine bonds formed from lysine and hydroxylysine, are converted to a thermally WO 90/06102 PCli/AU89/00422 12 stable, non-reducible bond which accounts for the age related collagen changes.

This method of isometric melting was used to examine how UV light may alter the collagen crosslinking. Rat tail tendon, which had been used for a number of other collagen ageing studies (Mitchell and Rigby, 1975 BBA, 393, 531-541; Rigby and Mitchell, 1978, BBA, 532, 65-70 and BBA, 544, 62-68; Rigby et al, 1977, BBRC, 79(2), 400-405 was used in these experiments.

This method has many advantages over other methods as it is a direct measure of age related crosslinking changes. As the present application has relevance to photoageing and in particular collagen crosslinking by UV light this method makes it possible to make quantitative measurements of the extent of UV crosslinking with time.

Further, this method enables the effectiveness of carnosine and other antioxidants in protecting the tendons against UV induced free radicals to be determined.

Materials and Methods The technique of isometric melting was applied to the rat tail tendon in order to determine the effects of photoageing.

Ninety day old Sprague-Dawley rats were used in these experiments.

Isolation of Tendons The tail was removed by excision and transported to the laboratory in ice. the tendons were then carefully dissected out on a saline moistened surgical pad to prevent drying. Each tendon was then measured and cut in half. The top half of the tendon was used for experiments while the bottom was kept at 4 C as a physical control.

UV Irradiation and Isometric Measurement The experimental half of the tendon was incubated at 4°C in the appropriate test solution for 20 hours prior to UV exposure. Following this exposure the tendons were WO 90/06102 PCYIU89/00422 13 washed in PBS and kept at 4°C until analysed.

The-isometric apparatus consisted of a strain gauge Shinkho transducer type UL-100gm connected to an amplifier. After attachment to the strain gauge the sample was immersed in a jacketed pyrex bath containing PBS. During the experiments the bath was heated with a Tamson circulating water heater. To measure the temperature increase a FLUKE thermocouple model 80TK was fixed to a region next to the tendon attachment site.

Measurements of force and temperature were recorded simultaneously with an ICI DP600 dual pen chart recorder.

For analysis the tendon was re-cut to a standard length of 3cm before attachment to the isometric apparatus. The attached tendon was then immersed into the bath of PBS at 20 0 C and a tension of 1 gram applied.

After a relaxation period of 15 minutes the apparatus was turned on and the temperature increase at a rate of 1°C per minute until melting occurred. Final measurements were then obtained from the chart recorder.

To determine the exposure time needed to produce a measurable change in tendon crosslinking a time course experiment was carried out. The results of this experiment are shown in Figure 11. It was found that an exposure of 180 minutes using a light source that consisted of 4 UV A tubes and 2 UV B tubes gave a reproducible result. This time period was used for the following experiments.

Carnosine, homo carnosine and anserine were used to pretreat replicate tendons. These were then exposed to UV AB for 180 minutes The results from these experiments are shown in Figure 12. In these experiments carnosine and homocarnosine at 10 and 100 mM effectively protected the tendons against UV induced crosslinking. Whilst no protective effect was observed with anserine at 10 mM it is believed that anserine may provide a protective effect WO 90/06102 PC/AU9/00422 14 at higher concentrations. (Unfortunately anserine was not tested as higher concentrations due to its lack of availability).

Figure 13 shows the results obtained when a set of other di and tri peptides were examined for their ability to protect tendons against UV induced crosslinking in comparison to carnosine. It was found that none of those tested protected tendons against UV induced crosslinking.

Two of the peptides contained histidine and still were inactive. These results suggest that carnosine is probably acting via its antioxidant property to protect against UV induced collagen crosslinking.

Figure 14 shows comparison of tendon data from Figure 12 with Sm measurements for glutathione at 10 mM Glutathione appears more effective using this method of measurement. Anserine (10A) is not working at this concentration. It is important to note that glutathione would not be available to act as free radical scavenger in vivo at this concentrations.

Summary The results from these experiments using isometric melting techniques prove conclusively that carnosine acts effectively in protecting the rat tail tendon against UV induced collagen crosslinking. Also from the results of the other di and tri peptides tested, it is also obvious that the carnosine effect is specific and occuring because of its antioxidant property.

Claims (13)

1. A method of reducing or preventing collagen crosslinking in the skin of a subject requiring such reduction or prevention comprising treating the skin with a composition comprising a suitable excipient in combination with an active compound, the active compound being selected from the group consisting of carnosine, homocarnosine, anserine, 3-methyl-L-histidine, L-alanyl-L-tyrosine, acyl homocarnosine, acetyl carnosine, iodo carnosine, di-iodo carnosine, anserine nitrate, carbenoxylone carnosine, analogues thereof and combinations thereof.

2. A method as claimed in claim 1 in which the active compound is carnosine or homocarnosine or a combination thereof.

3. A method as claimed in claim 2 in which the active compound is carnosine.

4. A method as claimed in any one of claims 1 to 3 in which the active compound is linked to another molecule, which molecule is such that the composition is improved in regard to skin permeation, skin application and tissue absorption.

5. A method as claimed in claim 4 in which the molecule is an amino acid or peptide. 25

6. A method as claimed in any one of claims 1 to 5 in which the method involves topical application of the composition to the skin.

7. A method as claimed in any one of claims 1 to 6 in which the composition includes a compound selected from 30 the group consisting of bilirubin, carotenoids, mannitol, reduced glutathione, selenium, uric acid, vitamin A, vitamin C, vitamin E and combinations thereof.

8. A method ,f reducing or preventing collagen crosslinking in skin exposed to UV light, said method comprising treating the skin which has been or will be 16 exposed to UV light with a composition comprising a suitable excipient in combination with an active compound, the active compound being selected from the group consisting of carnosine, homocarnosine, anserine, 3-methyl-L-histidine, L-alanyl-L-tyrosine, acyl homocarnosine, acetyl carnosine, iodo carnosine, di-iodo carnosine, anserine nitrate, carbenoxylone carnosine, analogues thereof and combinations thereof.

9. A method as claimed in claim 8 in which the active compound is carnosine or homocarnosine or a combination thereof.

A method as claimed in claim 9 in which the active compound is carnosine.

11. A method as claimed in any one of claims 8 to 10 in which the active compound is linked to another molecule, which molecule is such that the composition is improved in regard to skin permeation, skin application and tissue absorption.

12. A method as claimed in claim 11 in which the molecule is an amino acid or peptide.

13. A method as claimed in any one of claims 8 to 12 in which the method involves topical application of the Scomposition of the skin. S14. A method as claimed in any one of claims 8 to 13 in 25 which the composition includes a compound selected from the group consisting of bilirubin, carotenoids, mannitol, reduced glutathione, selenium, uric acid, vitamin A, vitamin C, vitamin E and combinations thereof. i