Scavenging effect of MIF-

Scherer TA, Lauredo IT, Abraham WM: Scavenging effect of MIF-A3; Scavenging of Reactive Oxygen Species by a Glycolipid Fraction of Mycobacterium Avium Serovar 2. The Internet Journal of Pulmonary Medicine 1997; Vol1N1: http://www.ispub.com/journals/IJPM/Vol1N1/scav.htm

Previous experiments indicated that MIF-A3, a glycolipid extracted from Mycobacterium avium serovar 2 (Mycobacterium paratuberculosis 18), inhibits the killing of Candida albicans by activated bovine peripheral blood derived macrophages and murine thioglycollate elicited peritoneal macrophages in vitro. Subsequent in vitro data from our laboratory indicated that this reduction in killing may be related to the ability of MIF-A3 to scavenge reactive oxygen species (ROS).

In this study we examined this hypothesis further by determining if the reduction in ROS (around 50%) as measured by H2O2 accumulation previously observed in the presence of MIF-A3 in adherent alveolar macrophages affected candidacidal activity. For the present studies we used two in vitro Candida albicans killing assays. First we incubated Candida albicans with H2O2 (4mM) alone or together with different concentrations of catalase. At high levels of catalase, the fungicidal effect of H2O2 was completely blocked. However, reducing the amount of catalase from 6.8 U/ml to 3.4 U/ml resulted in a loss of scavenging activity and this was associated with a 50%increase in killing. In the second assay we replaced catalase with MIF-A3 (400 mg/ml). At this concentration MIF-A3 was shown to reduce AM induced H2O2 accumulation by 45%. In the absence of MIF-A3 H2O2 reduced Candida albicans to less than 10³ CFU/ml. However, in the presence of MIF-A3 the CFU/ml of Candida increased 7.5-fold. These findings a) provide direct evidence for the H2O2 scavenging ability of MIF-A3 and b) a possible mechanism for reduced killing of Candida albicans seen in various in vitro systems in the presence of MIF-A3.

Key words: Mycobacteria, host defense, reactive oxygen species, killing of Candida albicans.

MIF-A3, a glycolipid from Mycobacterium avium serovar 2 (Mycobacterium paratuberculosis 18) is a water-soluble peptidoglycolipid containing trehalose as the only sugar moiety ^1,2. This compound has been reported to inhibit the candidacidal activity of activated peripheral blood derived bovine macrophages and thioglycollate elicited murine peritoneal macrophages ².

Since it has been shown, that killing of Candida albicans by human alveolar macrophages (AM) is correlated with the amount of superoxide anion released during phagocytosis ³, we postulated, that MIF-A3 may have exerted its inhibitory effects in the aforementioned studies by acting as a scavenger of reactive oxygen species (ROS). Preliminary experiments from our laboratory support this conclusion as MIF-A3 was observed to reduce measurable H2O2 produced in cell free systems and by adherent alveolar macrophages ⁴.

In the present study we extended our previous observations by determining if the addition of MIF-A3 at a dose that was seen to inhibit H2O2 accumulation is able to prevent H2O2 induced killing of Candida albicans.

Hydrogen peroxide (30% solution), penicillin 10'000 U/ml and streptomycin 10 mg/ml and catalase were obtained from Sigma (St. Louis, MO). Sterile Dulbecco's PBS (calcium and magnesium free) (D-PBS) was obtained from Gibco Laboratories (Grand Island, NY), tryptic soy broth from DIFCO Laboratories (Detroit,MI) and sterile pyrogen free distilled water from Baxter Healthcare Corporation (Deerfield, IL). Tryptic soy blood agar plates (TSA II 5% SB) were obtained from Becton Dickinson Microbiology Systems (Cockeysville, MD).

MIF-A3 was kindly provided by M.E. Hines II. It was dissolved in D-PBS at a concentration of 1 mg/ml and stored at -80° C.

The exact concentration of the H2O2 solution (30%) before use was determined by its absorption at 230 nm using an extinction coefficient of 81/M/cm ⁵. On the day of the experiment from this solution a 1/5 dilution was made in sterile pyrogen free distilled water and another 1/5 dilution in PBS to achieve a final concentration of 400mM.

Immediately before use catalase (3400U/mg protein) was dissolved in D-PBS to a concentration of 1360U/ml. From this solution serial 1:1 dilutions were made achieving final concentrations of 680U/ml, 340U/ml, 170U/ml, 85U/ml and 42.5U/ml.

A 24 hour culture of Candida albicans (ATCC 60193), grown in 5 ml tryptic soy broth, was centrifuged at 2500 x g at 4° C and after discarding the supernatant resuspended in 5 ml of D-PBS. Aliquots of 10 ml of this solution were placed into different polypropylene tubes together with 10 ml of penicillin (10'000U/ml) and streptomycin (10 mg/ml). Thereafter 10 ml of the 400mM H2O2 solution and 10 ml of the serial dilutions of the catalase were added to some tubes. Finally the volume of all tubes was adjusted to 1 ml by adding D-PBS. Our tubes contained therefore Candida albicans without any additional agents, with 4mM H2O2 alone or with different concentrations of catalase (13.6 U/ml, 6.8 U/ml, 3.4 U/ml, 1.7 U/ml, 0.85 U/ml and 0.425 U/ml) or catalase (13.6 U/ml) alone. Preliminary experiments showed that 4mM H2O2 was the lowest concentration able to almost completely kill Candida albicans in these experiments. The tubes were incubated in a shaking water bath at 37° C for three hours. Thereafter, from each reaction tube 1/100 and 1/1000 dilutions were made and 100 ml of each dilution spread on blood agar plates. The plates were cultured at 37° C for 48 hours. Killing of Candida albicans was assessed by counting colony forming units (CFU/ml) on the blood agar plates and expressed as CFU/ml x 10⁴.

In the second assay catalase was replaced by MIF-A3 (400 mg/ml). We used a final volume of 0.5 ml in the reaction tubes. Otherwise the experiments were done as described above.

Data were analyzed using Kruskal Wallis ANOVA and t-test where appropriate. If the null hypothesis was rejected, then pairwise comparisons were made with Duncan's Multiple Range Test. Significance was accepted when p < 0.05 using a two-tailed test.

Table 1 shows the effects of catalase on H2O2 killing of Candida albicans. Without H2O2 the CFU/ml was 11.34 +/- 2.03 x 10⁴. Addition of H2O2 suppressed the number of CFU almost completely (0.375 +/- 0.144 x 10⁴ CFU/ml). Catalase in concentrations higher than 3.4 U/ml protected Candida albicans from the fungicidal effects of H2O2; reducing the amount of catalase to 3.4 U/ml resulted in killing of about 50% of Candida.

Having established that changes in the level of H2O2 scavenging can result in increased killing of Candida albicans we used another assay to determine if MIF-A3 had similar scavenging properties.

Table 1: CFU of Candida albicans in the presence and absence of H2O2 and catalase (Back to text)

Added agents to incubation solution	Candida albicans (CFU/ml x 104)
None	11.34 +/- 2.03
H2O2	0.375 +/- 0.144 *
catalase (13.6 U/ml)	12.28 +/- 1.73
H2O2 + catalase (13.6 U/ml)	14.5 +/- 2.76
H2O2 + catalase (6.8 U/ml)	12 +/- 0.74
H2O2+ catalase (3.4 U/ml)	5.93 +/- 1.18 *
H2O2+ catalase (1.7 U/ml)	1.1 +/- 0.25 *
H2O2+ catalase (0.85 U/ml)	0.6 +/- 0.12 *
H2O2+ catalase (0.42 U/ml)	0.58 +/- 0.38 *

To show that MIF-A3 itself can protect against H2O2 induced killing we incubated Candida albicans with H2O2 alone and together with MIF-A3 (Table 2). With H2O2 alone <10³ CFU/ml of Candida were present. This number increased at least 7-fold to 7.5 ( 0.8 x 10³ CFU/ml in the presence of MIF-A3. These findings support the idea that the scavenging ability of MIF-A3 is able to protect against H2O2 induced killing of Candida albicans.

Added agents to incubation solution	Candida albicans(CFU/ml)
None	22 x 10⁴
H2O2	< 10³ *
H2O2 +MIF-A3	7.5 +/- 0.8 x 10³ *
MIF-A3	28 x 10⁴

In this study we provide direct evidence for the protective effect of MIF-A3 on H2O2 induced killing of Candida albicans. These results provide a possible mechanism for the reported inhibition of the candidacidal activity of activated bovine and murine macrophages by MIF-A3 ². These results also extend our previous data in which we found that MIF-A3 acted as an ROS scavenger ⁴. In these previous experiments we showed that in an in vitro system, using X-XO as a source of ROS, MIF-A3 was able to decrease the accumulation of superoxide and H2O2. Further experiments with adherent, zymosan stimulated sheep alveolar macrophages showed, that MIF-A3 was able to reduce accumulation of H2O2 in a concentration dependent manner. Compared to baseline, MIF-A3 at concentrations of 400, 200 and 100 mg/ml reduced H2O2 accumulation by 45%, 34% and 17%, respectively. To rule out, that the decrease in ROS was caused by interfering with the ROS generating enzymes, we directly incubated MIF-A3 with H2O2 and showed that MIF-A3 caused a concentration dependent decrease in measurable H2O2. Collectively these findings indicated that MIF-A3 had the ability to scavenge H2O2.

Although these results showed that MIF-A3 was able to scavenge ROS, we had not demonstrated whether or not this reduction in H2O2 accumulation had biological relevance. To do this we used in vitro assays to assess the killing of Candida in the presence and absence of MIF-A3 and catalase. Our results show that MIF-A3 at a dose previously shown to provide a 45% reduction in H2O2 accumulation was effective in preventing candidacidal activity of H2O2.

We can not directly compare the scavenging activity of MIF-A3 in this in vitro assay to that previously found in the zymosan stimulated adherent macrophages. The concentration of H2O2 used in the present studies was approximately 250 times higher than the concentrations produced by zymosan stimulated macrophages. We used this concentration of H2O2 because our preliminary studies showed this concentration to almost completely kill Candida. However, despite this high concentration of H2O2, MIF-A3 was able to protect Candida from the H2O2 induced killing. This finding supports our hypothesis and furthermore demonstrates that the scavenging effect of MIF-A3 has biological relevance.

Previous studies have shown that small changes in ROS generation can affect host defense mechanisms. Vecchiarelli et al ^3,6 have shown that the candidacidal activity of polymorphonuclear leukocytes and alveolar macrophages correlates with the amount of superoxide released. A gradual increase in the killing of Candida albicans was caused by gradual increases in superoxide production. Nathan et al ⁷ correlated maximal and minimal killing of Trypanosoma cruzi by murine peritoneal macrophages with a 30% change in H2O2 production (9 nMol vs. 6 nM H2O2 production per minute). This decrease is in the range we attained with MIF-A3 at concentrations of 200 and 400 mg/ml in adherent and zymosan stimulated macrophages.

A glycolipid extracted from Mycobacterium leprae, phenolic glycolipid I (PGLI), at concentrations of 100 to 300 mg / 2 x 106 cells, suppressed the release of superoxide by monocytes from healthy human volunteers from 50% to 70% ⁸. Monocytes from healthy persons, pretreated with 100 mg PGL I / 2 x 106 cells, experienced a reduction in superoxide secretion around 40 % ⁹. Thus, PGL I, a glycolipid considered to have relevant adverse effects on the host, affects ROS production to an extent that is comparable to the effects of MIF-A3 in adherent alveolar macrophages.

In summary, we have shown that MIF-A3, a glycolipid extracted from Mycobacterium paratuberculosis 18, is able to scavenge ROS, and that this effect inhibits the fungicidal action of H2O2. This scavenging effect may be responsible for the inhibition of the candidacidal activity previously shown in activated bovine and murine macrophages. These findings also suggest that breakdown products from mycobacteria can act in a protective fashion against ROS-induced killing and thus, act to increase the infectivity of the organism.

MIF-A3: macropage inhibiting factor A3; ROS: reactive oxygen species; CFU: colony forming units; D-PBS: sterile Dulbecco's phosphate buffered saline; H2O2: hydrogen peroxide.

1. Hines, M.E., II, Long A.R., Snider T.G.III, Barker S.A.. Lysis and fractionation of Mycobacterium paratuberculosis and Escherichia coli by matrix solid phase dispersion. Anal. Biochem. 1991; 195:197-206. (Back to text)

2. Hines, M.E., II, Jaynes J.M., Barker S.A., Newton J.C., Enright F.M., Snider T.G. III. Isolation and partial characterisation of glycolipid fractions from Mycobacterium avium serovar 2 (Mycobacterium paratuberculosis 18) that inhibit activated macrophages. Infect. Immun.1993; 61:1-7. (Back to text)

3. Vecchiarelli, A., Dottorini M., Cociani C., Pietrella D., Todisco T., Bistoni F. Mechanism of intracellular candidacidal activity mediated by calcium ionophore in human alveolar macrophages. Am. J. Resp. Cell Mol. Biol. 1993; 9:19-25. (Back to text)

4. Hines M.E. II, Scherer T., Silwany O., Lauredo I.T., Wanner A., Stein-Streilein J., Altman N.H., Abraham W.M. A glycolipid compound derived from Mycobacterium avium serovar 2 that inhibits the candidacidal activity of macrophages scavenges reactive oxygen species, but has no effect on nitric oxide production. Proceedings of the Fourth International Colloquium on Paratuberculosis. 1995: 240-249. (Back to text)

5. Pick, E., Keisari Y.. A simple colorimetric method for the measurement of hydrogen peroxide produced by cells in culture. J. Immunol. Methods 1980; 38:161-170. (Back to text)

6. Vecchiarelli, A., Dottorini M., Beccari T., Cociani C., Todisco T., Bistoni F. Inhibition of candidacidal activity of polymorphonuclear cells by alveolar macrophage-derived factor from lung cancer patients. Am. Rev. Resp. Dis. 1993; 147:414-419. (Back to text)

7. Nathan, C., Nogueira N., Juangbhanich C., Ellis J., Cohn Z.. Activation of macrophages in vivo and in vitro. J. Exp. Med. 1979; 149:1056-1068. (Back to text)

8. Vachula, M., Holzer T.J., Andersen B.R.. Suppression of monocyte oxidative response by phenolic glycolipid I of Mycobacterium leprae. J. Immunology 1989; 142:1696-1701. (Back to text)

9. Vachula, M., Worobec S., Andersen B.R.. A comparison of monocyte oxidative responses in leprosy patients and healthy subjects as influenced by mycobacterial lipid pretreatment. Int. J. Leprosy 1990; 58:534-539. (Back to text)