4. Hirota, S., Miyamoto, M., Kasugai, T., Kitamura, Y., and Mo-rimura, Y. (1990) Arch. Pathol. Lab. Med. 114(4), 429–431.

5. Laemmli, U. K. (1970) Nature (London) 227, 680–685.6. Harlow, E., and Lane, D. (1988) Antibodies: A Laboratory Man-

ual, pp. 493–649. Cold Spring Harbor Laboratory, Cold SpringHarbor, NY.

7. Hancock, K., and Tsang, V. C. W. (1983) Anal. Biochem. 133,157–162.

8. Burnette, W. N. (1981) Anal. Biochem. 122, 157–162.9. Rohrinnger, R., and Holden, D. W. (1985) Anal. Biochem. 144,

118–127.10. Jacobson, G. (1994) in Protein Blotting: A Practical Approach

(Dunbar, B. S., Ed.), pp. 53–70. IRL Press, New York.

Determination of Heme Oxygenase Activity inMurine Macrophages for Studying OxidativeStress Inhibitors

Victor Turcanu,1 Mounir Dhouib,and Philippe PoindronDepartement d’Immunologie, Immunopharmacologie etPathologie, Faculte de Pharmacie, Universite Louis Pasteur,BP 24, F-67401 Illkirch, France

Received June 9, 1998

Heme oxygenase (HO)2 (EC1.14.99.3) is the microso-mal enzyme which degrades heme, producing biliver-din IXa which is further reduced to bilirubin by biliv-erdin reductase (1). In mammalian cells, HO inductionis a universal response to oxidative stress (2) and HOmRNA measurement was established as a sensitivemarker of oxidative stress. Since its induction is inhib-ited by antioxidants, HO induction can also be used tomeasure a free radical-scavenging activity (3).

HO activity can be quantified spectrophotometricallymeasuring bilirubin production by tissular or cellularmicrosomal fractions (4, 5). Such methods are adequatefor tissue analysis but they are relatively inconvenientfor in vitro work as they require relatively large quan-tities of cells. A more sensitive assay using gas chro-matography was established to quantify the carbonmonoxide production due to the HO activity of cellularcytosolic fractions (6), while another recent techniqueis based on the detection of [14C]bilirubin formation bycellular microsomal fractions, using [14C]heme as a

substrate (7). Although both methods are sensitive,they require certain skills and access to equipmentthat is not standard in many laboratories. This and thenecessity of making kinetic assays in order to investi-gate the effects of different oxidative stress inhibitorsupon HO induction made us develop an assay based onthe quantitation of the bilirubin excreted into the cul-ture medium. As HO is strongly induced in macro-phages after erythrophagocytosis or exposure to heminwhich also induces a stress-like effect (5), we chose touse macrophage cultures as an in vitro cell system toestablish this new HO assay. The addition of antioxi-dants to the system should decrease HO inductionaccording to their free radical scavenging effect, so thissimple in vitro cellular system should allow the screen-ing of antioxydant agents.

Materials and Methods. Benzene was purchasedfrom Fluka (Buchs, Germany), and BaCl2, 2H2O fromProlabo (Paris, France). Protoporphyrin IX zinc (II)was from Aldrich (St. Quentin Fallavier, France) whilebilirubin, dimethyl sulfoxide (DMSO), hemin (Hm),and Trizma base were purchased from Sigma (St.Quentin Fallavier, France). Porphyrin solutions wereprepared as follows: the porphyrin and Trizma base (ina 1:10 molar ratio) were dissolved in a minimal volumeof 1 M NaOH, then neutralized with 1 M HCl; finally,deionized water was added to make a 10 mM finalconcentration. This stock solution was aliquoted andkept frozen at 220°C. Its addition to the cell culturedid not significantly influence the pH of the culturemedium.

BALB/c bone marrow-derived murine macrophages(BMMf) were obtained according to the method estab-lished by Munder et al. (8) by the culture of bonemarrow cells in hydrophobic Teflon bags (a kind gift ofDr. Manuel Modolell from the Max-Planck Institut furImmunbiologie from Freiburg-im-Breisgau, Germany).After 12 days of culture, BMMf were collected anddistributed in 6-well cell culture plates (purchasedfrom Nunc, Roskilde, Denmark) at a concentration of2 3 106 cells/well/2 ml culture medium. The culturemedium was RPMI 1640 (from Gibco, Life Technolo-gies, Eragny, France) containing 10% fetal calf serum(FCS, decomplemented 30 min at 56°C), 50 mM 2-mer-captoethanol, 100 U/ml penicillin, and 100 mg/ml strep-tomycin).

HO activity was measured as follows: after plating,BMMf were left overnight to adhere, and then theporphyrins were added. All further manipulationswere carried out in dim light and the plates wereprotected against light by wrapping in aluminium pa-per. The plates were incubated for different periods oftime, then 0.5 ml of each culture supernatant wascollected and 250 mg BaCl2, 2H2O/probe was added.After vortexing (10–15 s), 0.75 ml benzene was added;

1 To whom correspondence should be addressed at Universite LouisPasteur, Faculte de Pharmacie, Departement d’Immunopharmacologie,74, route du Rhin, BP 24, F-67401 Illkirch, France. Fax: 133 388 66 01 90.

2 Abbreviations used: HO, heme oxygenase; DMSO, dimethyl sul-foxide; Hm, hemin; BMMf, bone marrow-derived murine macro-phages; FCS, fetal calf serum.

251NOTES & TIPS

ANALYTICAL BIOCHEMISTRY 263, 251–253 (1998)ARTICLE NO. AB9828060003-2697/98 $25.00Copyright © 1998 by Academic PressAll rights of reproduction in any form reserved.

then, tubes were vortexed vigorously (50–60 s) leadingto the formation of a relatively stable milky-whiteemulsion. After centrifugation (13,000g, 30 min, with-out cooling), the upper benzene layer was collected andthe absorbance at 450 nm with reference wavelength at600 nm was measured using a Shimadzu UV-160Aspectrophotometer. In a separate tube, 0.5 ml of freshculture medium was processed in the same way andthe benzene layer was collected and used as a blank.The quantity of bilirubin produced was calculated us-ing a molar extinction coefficient of bilirubin dissolvedin benzene e450 5 27.3 mM21 cm21.

Results and discussion. Bilirubin can be completelyextracted from cell culture medium with benzene in asingle step. Bilirubin was dissolved in DMSO (stocksolution, 10 mM), diluted in RPMI 1640 1 10% FCS atconcentrations close to those observed in activatedmacrophage cultures (5 mM), and then extracted withbenzene (Fig. 1). Complete extraction needed supra-saturating concentrations of BaCl2 and was optimal, inthis range of concentrations, for a ratio of 1:3 for cul-ture supernatant:benzene. The absorbance at 450 nmof the bilirubin dissolved in benzene is proportional tothe concentration of bilirubin for the range of concen-trations susceptible to be found in cell culture super-natants. We also found that bilirubin concentration isrelatively stable in the culture medium and that it isnot taken up or further catabolized by BMMf. Exoge-nous bilirubin was added in a BMMf culture; samplesfrom the culture supernatant were collected immedi-ately after addition and after 24 and 48 h of incubationat 37°C in an atmosphere containing 5% CO2. Bilirubinwas extracted in benzene and its apparent decreasewas calculated, showing that after 24 h of incubation,87% of the initial quantity could be recovered.

Finally, we tested the HO assay in a BMMf culture.We observed a dose-dependent increase of bilirubinconcentration in the culture medium related to thedose of hemin added to the BMMf (Fig. 2). We alsomeasured the bilirubin content of the BMMf and ob-served that practically all the produced bilirubin wasexcreted. We also observed that a HO inhibitor, zinc(II) protoporphyrin IX, inhibits the production of bili-rubin (Fig. 3) by BMMf.

The assay of the HO activity that we describe herecould be useful to measure the efficiency of free radicalscavenging agents and could also be used for kineticassays. The main advantage of this assay resides in thesmall number of cells which are necessary for everyassay (allowing complex experiments and the use ofslow-proliferating cells). This advantage results fromthe increase in sensitivity due to the benzene extrac-

FIG. 1. Bilirubin extraction with benzene in the presence of supra-saturating concentrations of BaCl2 (column 1, 5 mmol/liter bilirubindissolved in culture medium; column 2, medium alone).

FIG. 2. Bilirubin production by BMMf incubated for 24 h in thepresence of different concentrations of hemin.

FIG. 3. Inhibition of bilirubin production by BMMf cultured in thepresence of hemin (100 mM) by ZPP (for a 24-h incubation time).

252 NOTES & TIPS

tion and to the higher amount of detectable bilirubindue to the longer incubation time.

Other authors extracted the HO-dependently pro-duced bilirubin in chloroform before spectrophotome-try (9) to increase sensitivity, but it seems that some-times chloroform extraction does not succeed in aquantitative extraction of bilirubin (7, 10). We chose tomodify a method used previously to assay nonconju-gated bilirubin in the serum (11). That method usednonsaturating barium chloride concentrations and sev-eral successive extractions of bilirubin in benzene. Weinduced the formation of an emulsion by adding muchlarger quantities of barium chloride, which allowed theuse of a single-step extraction.

The linearity of the calibration curve in a regionwhich largely encompasses the interval of cell cultureexcreted bilirubin values allows the concentration ofthe produced bilirubin to be calculated according to theabsorption at 450 nm. In the case of cells having lowHO activity or for shorter incubation periods, the sen-sitivity of this assay can be further increased by usingmore cells, by concentrating the supernatant beforeextraction or by decreasing the ratio between benzeneand the culture supernatant.

In the case of very high level induction of HO, theactivity of biliverdin reductase might theoretically be-come limiting for the generation of bilirubin. In vivo,such a situation is not found in rodents or humanswhich do not excrete biliverdin even when they mustdegrade high amounts of heme (as in hemolytic ane-mias). In the case of the BMMf culture that we use, ifsome biliverdin were excreted, it would be measured as

well given its absorbtion at 450 nm, which is close tothat of bilirubin.

Therefore, this new method of HO activity determi-nation in cell cultures represents a good alternative toother assays due to its simplicity and sensitivity, andthe possibility of using it for kinetic experiments.

Acknowledgments. We are indebted to Robert Deyes from theUniversity Louis Pasteur, Strasbourg, France, for critically readingthe manuscript and making helpful suggestions.

REFERENCES

1. Tait, G. H. (1978) in Handbook of Experimental Pharmacology(De Matteis, F., and Norman Aldridge, W., Eds.), Vol. 44,pp. 1–24. Springer-Verlag, Berlin.

2. Applegate, L. A., Luscher, P., and Tyrrell, R. M. (1991) CancerRes. 51, 974–978.

3. Tyrrell, R. M., and Basu-Modak, S. (1994) Methods Enzymol.234, 224–235.

4. Yoshida, T., and Kikuchi, G. (1978) J. Biol. Chem. 253, 4224–4229.

5. Gemsa, D., Woo, C. H., Fudenberg, H. H., and Schmid, R. (1974)J. Clin. Invest. 53, 647.

6. Vreman, H. J., and Stevenson, D. K. (1988) Anal. Biochem. 168,31.

7. Sierra, E. E., and Nutter, L. M. (1992) Anal. Biochem. 200, 27.8. Munder, P. G., Modolell, M., and Wallach, D. F. H. (1971) FEBS

Lett. 15, 191.9. Balla, G., Jacob, H. S., Balla, J., Rosenberg, M., Nath, K., Apple,

F., Eaton, J. W., and Vercelotti, G. M. (1992) J. Biol. Chem. 267,18148.

10. Kutty, R. K., and Maines, M. D. (1982) J. Biol. Chem. 257, 9944.11. Rodier, J., and Mallein, R. (1968) Manuel de Biochimie Pratique.

Maloine, Paris.

253NOTES & TIPS