Document 2g8NVX58jEndMkx2MX5DevRr

I Code: 239-6127 Chlorinated D ibenzofurans in Kanechlors and Rice Oils Used by Patients with Yusho Junya N agayama", Yoshito Masuda:' and Masanori Kuratsune0 Department of Public Health, Faculty of Medicine, Kyushu University, Fukuoka, Japan1'- and Daiichi College of Pharmaceutical Sciences, Fukuoka, Japan" It was reported that the causal agent of the epidemic of Yusho which occurred in the Western Japan in 1968 was polychlorinated biphenyls (PCBs)`\ According to the animal experiments by Vos and his associates marked differences in toxicity were seen among foreign commercial brands of PCBs:r. They reported that the more toxic brands made in France and West Germany contained 20 and 5 ppm of polychlorinated dibenzofurans (PCDF), respectively, while the less toxic one made in the United States no such contaminants" . It seems quite possible that the toxicity of commercial PCBs is greatly affected by their relatively small amount of contaminants such as PCDF" 4 *'. On the other land, it has been well known that `chick edema factor'', the causal agent of an extensive poisoning of chickens which killed more than a million chicks in the United States in .`957, is polychlorinated dibenzo-p-dioxins (PCDD)" " , and that these compounds are extremely toxic and acnegenic to man, too*:,\ All these facts raise a question whether or not PCDF or PCDD are involved in Yusho in any sense. Thus, it is without doubt needed to investigate the possible contamination by these compounds of Kanechlor-400, a commercial brand of PC3s of Japanese make which is known to have caused Yusho epidemic. In view of the fact that very few studies have been made in this .respect, we analyzed various types of Kanechlor and samples of the toxic rice oil used by patients with Yusho for PCDD and PCDF. This paper describes the main results of such analysis. Materials and Methods 1. Materials analyzed 1) Kanechlor-300, -KX). 300. and 600: Kanechlor-300 and -300 were those donated Kanegafuchi Chemical Industry Co., Ltd. in 1968 and Kancchlor-400 and -600 were supplied by the Ministry of Health and Welfare ir. 1972. 2) Rice oil used by patients with Yusho ( "Yusho oil"): 3 samples of rice oil used by 3 independent families with Yusho in 1968. These samples had been kept in glass bottles with glass stoppers at room temperature in our laboratory until 19`d when the analysis was made. 2, Chemicals 1) n-Hexane, ethanol, water, potassium hydroxide, anhydrous sodium sulphate, and silica gel: Purified as described previously111. 2) Carbon tetrachloride and methylen chloride: Reagent grade, distilled with a fractional distillation column. 3) Activated alumina: Alumina for column-chromatography (about 300 mesh), Wako Pure Chemical Industry, Ltd., activated by heating at 15Q*C for 12 hours before every use. 4) PCDF and PCDD: Synthesized by chlorination of dibenzofurans or by conden NEV 0 3 7 0 3 U Page 2 sation of 2. 4. 5-lrichlorophenolli . PCDF consisted mainly of tetra* and pcnta-chlorodihenzofurans. while PCDD mainly of 2. 3. 7, S-tetrachlorodibenzo-p-dioxin. 5) Octachlorodibenzofuran and octachlorodibenzo-p-dioxin: Purchased from Analabs Inc., U.S.A. J . Apparatus 1) Gas chromatograph (ECD-GC): Beckman GC 72-5 with a detecter of electron capture type, using a glass column (4 mm x 2 m) containing chromosorb W AW DMCS (100-120 mesh) coated with 1 % SE-30, Column temperature was 200*C and helium was used as carrier gas, flow rate 30ml/min. 2) Gas chromatograph-mass spectrometer (GC-MS): Type D-100, GC-MS, Japan Electron Optic Laboratory. Column used was similar to the one used for gas chro matography. tonlzation energy was 25eV. 4. Analytical methods The alkaline decomposition method described in the official standard analytical methods for PCBs established by the Ministry of Health and Welfare was applied for the isolation of PCBs fraction from samples of Yusho oil (5-10g), Isolation of PCDF and PCDD from the PCBs fraction mentioned above or from Kanechlor samples (0.10.5 g) was made in the following ways. Each material was placed with 20 ml of n-hexace onto a column ( l x 7.5cm) of activated alumina and then eluted with 120ml of n-hexane containing 20 % carbon tetrachloride, 10 ml of n-hexane, and then with 30 ml of n-hexane containing 20 96 methylenchloride. PCDF and PCDD were expected to be contained in the final eluate. If their separation from PCBs was not adequate, the above chromatographic fractionation was repeated or another similar column chro matography was tried using a shorter column of activated alumina ( l x 2.5cm) and a half volume of each of the eluents mentioned above. The eluate consisting of n-hexane containing 7.096 methylen chloride thus collected was evaporated by a rotary evaporator, then dissolved in n-hexane and subjected to ECD-GC and GC-MS analysis. Determination of PCDF and PCDD was mads by two methods, namely by measur ing the gas chromatographic peak heights and by measuring the gas chromatographic peak area after perchlorinaiion of these compounds. The former consists of comparison of each of the peaks of gas chromatogram condoled as those of PCDF or PCDD by GC-MS with the corresponding one of the synthesized PCDF or PCDD used as standard, assuming that ill the PCDF or PCDD have the same peak height sensitivity regardless of the number of chlorine and the site of substitution by chlorine. The latter consisted of chlorination by the BMC reagent'11 of the evaporation residue of the fraction contain ing PCDF and PCDD to yield octachlorodibenzofuran and octachlorodibenzo-p-dioxin. followed by the gas chromatographic determination of such octachloro-derivatives. For determination of PCBs, the peak height calculation method and the perchlor* nation method, described in the official analytical methods for PCBs set by the Ministry of Health and Welfare, were used. Results The gas chromatograms of PCDF fractions separated from Kanechlor-400 and -300 are shown in Fig. 1, while the gas chromatogram of the corresponding fraction from Yusho oil in Fig. 2. Fig. 3 and 4 show the mass-spectra of tetra- and penta-chlorodibenzofurans isolated from these PCDF fractions, together with those of the standard PCDF. Table l summerizes the results of analysis by ECD-GC and GC-MS, indicating that from dichloro to heptachlorodibenzofurans were contained in the samples analyzed. The Kanechtors designated larger numbers, starting from 300 to 500, contained the I NEV 0 3 7 0 3 1 Page 3 Tri ri j .i h v \ A :i i i A 1 \\ V "1 i T*ftra 1Penta i B Hepta a Q j I' _ He xa j *" ^ ' - I * -- * L 0 10 20 30 40 Retencin time (min) Flij. 1. Gas chromatogram of PCDF fraction A. Kaaechlor-400, B. Kinechlor-JOO. C. Authentic specimen Fig. 2. Retention time (min) Gas chromatogram of PCDF frac tion. A. Rice oil ("Yusho oil';, B. Authentic specimen Tabla 1. Concentration of chlorinated dibenzofuraos in Kaoechlors and "Yusho oil" Samples ! Chlorodibenzofurans Concentration (O P m3 ' DI Tri ; Tetra Penta He*a Hepta a b Kanechlort . I 300 J 400 300 1 600 ; f ^H- *T" , . -r* - -X -r T - ' j. TT if 1 ; Lj IS ; 17 4 IS JiJ i A1 -Yuiho oil" B I . . . - -L c , -r ' -4- i -- " 1 "-y. ^if* I1t -- " . t a: calculated from peak heights, b: calculated by perehlorination method. S ' 4.3 45 J;3 higher chlorinated isomers of dtbenzofuran. The concentrations of FCDF in Kancchlor300, -400, -300 and -600 were 1 18* 4, and 3 ppm by the peak height calculation method and 1.5 17 2.5, and 3 ppm by the perehlorination method, respectively. Thus, NEV 0 3 7 0 3 2 I m Page 4 h f 200 0 300 r%n Fig. 3. GC-MS of tetrachlorodibenzofuran. A. KanechIor-400, B. Rice oil (-Yusho oil**). C. Authentic ipecimen Fig. (, GC-MS of pentachlorodibenzofuran. A. ;Kanechlor-(00. B. Rice oil ("Yusho oil'), C. Authentic specimen Table 3, Concentration of PCBs in hYusho oiP *Yusho oil" Concentration (ppm) ib A 830 3 900 C I.OJO a: calculated from peak heights b: calculated by perchlorination method J 1 370 m 930 two methods gave similar results, demonstrating that Kanechlor-400 contains a consider* ably higher concentration of PCDF than the other types. Three samples of Yusho oil showed similar gas chromatographic patterns, indicat ing that the PCDF contained in the oil consisted mainly of teira* and penta*chlorodi* benzofurans together with a small amount of tri* and hexa-chlorodibenzofurans. Both the peak height calculation method and the perchlorination method gave the same figure, 5 ppmPCDF. Contrary to the above Endings in regard to PCDF, PCDD was not detected in any of the materials examined. As shown a Table 2. the samples of Yusho oil contained 300-1,000 ppm of PCBs. and both calculations gave very close figures. In additional experiments, 10 g of a vegetable oil containing 10 mg of Kanechlor-400 was subjected to a series of analytical procedures, the same as those employed in the above analysts, including the alkaline decomposition process maintained for 6 hours. NEV 0 3 7 0 3 3 Page 5 PCDF found from the oil thus treated was about 0.2 g. just as much as expected from the PCDF content of 10 mg of Kanechlor-400. Similar blank experiments were repeated 3 times, yielding practically the same results. Therefore, any possible transformation of PCBs to PCDF during the analytical procedures could be denied. The recovery ratio of 1pg of PCDF added to 10 g of a vegetable oil was determined to be approximately 9Q?6- Discussion Analysis of a trace amount of PCDF and PCDD is generally known to be fairly difficult. As mentioned, it is not conceivable that the PCDF found in cur analysis were the artificial products due to the analytical procedures employed. In addition, the recovery ratio was also fairly high. Therefore, we can say with confidence that Kanechlors and "Yusho oil" contain PCDF. It seems to be noteworthy that the concentrations of PCDF in Kanechlor-400 and in Yusho oil, as determined by the present analysis, are considerably h:gher than the figures so far reported. Roach et al., found about 1 ppm of PCDF in Kanechlor-4O0,,\ According to Vos. Bowes. Space Science Laboratory. University of California, cculd de monstrate the presence of PCDF in Kanechlor samples and in a sample of "Yusho oil". The figures for the formers have not yet been supplied but the figure for the latter, ap proximately 1 ppm, have been provided us1'*. Thus, the informations so far accumu lated do not very precisely coincide with each other, tut they agree at least in that beth Kanechlors and "Yusho oil" contain PCDF. Since 'Yusho oil" which is known to have been contaminated with Karachi-.r 400 contained 5 ppm of PCDF and 1.000 ppm of PCBs. the ratio of the cor.certr.ainn of PCDF to that of PCBs is 5 : 1.000. much larger than the expected ratio 0 0l< : 1.000 which can be calcurated from our figure concerning PCDF concentration in Karechl'.r400. The reasons for this great discrepancy are not well understood, but it sh u!J ne noted that the sample of Kancch!or-400 we analyzed was as "unused" one. while the Kanechlor-400 present in "Yusho oil" must be "used" one as heat transfer medium. Very few studies have been maJe in which the toxicity of PCDF was directly compared with that of PCBs. Araki observed that PCDF was about i "0 timet tr .ry.-r than Kanechlor-400 in the activity of hepatic enzyme induction" . N.'-h zumi and Masuda. however, could not observe any marked difference in acnegenic.ty lv tween PCBs and PCDF when applied to the skin of the ears of rabbits. The ratio of rhe concentration of PCBs needed to exert a certain degree of icr.egenicity to the corres ponding concentration of PCDF needed to exert the same degree of acncg.nic.ty was shown to be It 0.3:a*. There are several studies, however, in which "Yusho oil" and Kanechlor -too were compared in toxicity. Ikoda reported that "Yusho oil" was twice or more toxic, when fed to broilers, than expected by the concentration of KarechIor-400 o m e n e d in the oil1* ,r-. In feeding experiment with CF s 1 mice, it was shown that both a diet added "Yusho oil" to attain about 110 ppm of PCBs in the diet and a diet containing !OOppm of Kanechlor-400 exerted an approximately similar toxicity. The former diet, however, seemed to affect more the skin and liver than the latter and the authors suggeued a possibility that "Yusho oil" might contain some chlorinated compounds which were more toxic than Kanechlor-4001" . In a feeding experiment with chicken, Gwto *t al. observed that a diet containing 5 # of "Yusho oil" showed similar toxic effects to chase induced by feeding a diet containing 400 ppm of Kacechlor-400. If the concentration of PCBs in the "Yusho oil" used could be assumed to be about 2.000 ppm. the concen tration of PCBs in the diet added "Yusho oil" would be about 100 ppm, considerably NEV 0 3 7 0 3 4 Page 6 lower than that of the diet added KanechIor-400"*. Furthermore, there seems to be a great difference in acute toxicity between PCDF and Kanechlors. LD,, of Kanechlor-400, when fed to mice and rats, was 1.1-1.9 g kg and that of Kanechlor-300 0.6 g kg when fed to rabbits1''. PCDF. on the contrary, killed rabbits due to severe liver necrosis after a single oral administration of as little as about 1 mg/kg, ! . Toxicity of PCBs and possibly of PCDF must greatly vary from isomer to isomer. Rational comparison of toxicities of PCBs and PCDF does not seem to be easy. However, in view of all these past findings described above, PCDF seems to be con siderably more toxic than PCBs. The possible role played by PCDF in Yusho should not be dismissed. Summary In order to answer the question whether or not polychlorinated dibenzofuran* (PCDF) and polychlorinated dibenzo-R-dloxins (PCDD) which are more toxic than PCBs are involved in causation of Yusho. various types of Kanechior, PCBs of Japanese make, and 3 samples of "Yusho oil" used by Yusho families were analyzed for these compounds. Kanechlor-300. -400, -500. and -600 were shown to contain l, 18. 4, and 5 ppm of PCDF, respectively, as calculated by gas chromatographic peak heights and 1.5, 17. 2.5, and 3 ppm of PCDF. as calculated by the perchlorination method. Thus. Kanechlor -400 was most contaminated with PCDF. "Yusho oil" samples also contained about 5 ppm of PCDF, much higher than expected. PCDD, however, was not detected in any materials analyzed. The possible role of PCDF in causation of Yusho was suggested. References 1) Tsukamoto, H. et al.: The chemical studies on detection of toxic compounds in the rice bran oils used by the patients of Yusho. Fukuoka acta Med. 60: *96-512, 1969 (In Japanese;. 2) Vos. J. G. and Koerrun. J. HL: Comparative' toxicologic study with PC3s in chickens with special reference to porphyria, edema formation, liver necrosis and tissue residues. Toxicol. Appl. Pharmacol. 17: 656-663. 1970. J) Vos. J. G., Koeman. J. H., Van der Mass, H. L., Ten Noever de Brauw, ,V|. C. and Ue Vos. R. H.: Identification and toxicological evaluation of chlorinated dibenzofuran and chlor inated naphthalene in two commercial PCBs. Food cosmet. Toxicol. S: 625-6JJ, 1970. 4) Vos, J. G. and Beems, R. 9.: Dermal toxicity studies of technical PCBs and fractions thereof in rabbits. Toxicol. Appl. Pharmacol. 19: 617-633, 1971. 5) Vos. J. G. and Notenboom-Ram. E.: Comparative toxicity study of 2.4.J.2\4\5-hexachloroblphenyl and PCS mixture in rabbits. Toxicol. Appl. Pharmacol. 23: 563-578, 1972. 6) Cantrell. J. S., Webb, N. C. acd Mabis, A. J.: Identification and crystal structure of a hydropericardlum producing factor. Chem. Eng. News. 45: 10, 1967. 7) Firestone, D.. Flick, D. F.. Ress. J. and Higginbotham. G. R .: Distribution of chick edema factor in chick tissues. J. Ass. offle. Anal. Chem. 54: 1293-1298. 1971. 8) Bauer, H.. Schulz, K. H. ur.d Spiegelberg, U.: Berurtiche Vergiftungen bei der Hersteliur.g von Chlorphenol-Verbindungen. Arch. Gewerbepath. Gewerbehvg. 18: 533-555, 1961. 9) Oliver, R. M.: Toxic effects of 2.J. 7.8 tetrachlorodibenzo 1,4 dioxin in laboratory workers. Br. J. Ind. Med. 32: 49-53, 1975. 10) Roach, J. A. G. and Pomerantz, l. H.t The finding of chlorinated dibenzofuran in a Japanese PCB sample. Bull. Environ. Contam. Toxicol. 12: 338-342. 1974. 11) Masuda, Y,, Kagawa. R. and Kuratsune, M. : Polychlorinated biphenyls in Yusho patients and ordinary persons. Fukuoka Acta Med. 65: 17-24, 1974 (In Japanese). 12) Nishlzumi, M,, Masuda, Y. and Kuratsune. M.: Comparison of hyperkeratosis induced by PCBs, PCDF and PCDD application. Fukuoka Acta Med. 66: 600-604, 1965 (In Japanese). NV 0 3 7 0 3 5 Page 7 13) Hutzinger, O., Safe. S. and Zitko. V.: Analysis of chlorinated aromatic hydrocarbons by exhaustive chlorination: Qualitative and structural aspects of the perchloroderivatives of biphenyl, naphthalene, terphenyl, dibenzofuran, dibenzodioxin and DDE. Intern, J. Environ. Anal. Chem. 2: 95-106, 1972. 14) Vos, J. O .: From private communication, 1974. 15) Arakl, Y.: Comparison of enzyme-inducing activities of chlorinated dibe.'.zofuran and chlorinated dibenzodioxin. Fukuoka Acta Med. 65: 61-64, 1974 (In Japanese). 16) National Hygienic Laboratory, Ministry of Health and Welfare: Studies on toxic substances in rice-bran oil, with special reference to organochlorine compounds. (5) Studies on toxicity of noxious substances. Special studies on prevention, diagnosis and treatment of Yusho.: 74-124, 1969 (In Japanese). 17) Ikeda, Y.: Toxicity of PCB. J. Food Hyg. Soc. Japan. 13: J59-J67, 1972 (In Japanese). 18) Livestock Hygiene Experiment Station. Ministry of Agriculture and Forestry: Studies on toxic substances in rice-bran oil, with special reference to organochlorine compounds. (5) Studies on toxicity of noxious substances. Special studies on prevention, diagnosis and treatment of Yusho.: 74-124. 1969 (In Japanese). 19) Qoto, M., Sakaguchi. K. and Ogawa, K.: Hydropericardium assay of rice oil which caused Yusho and of Kanechlor 400 in chickens. Fukuoka Acta Med. 60: 533-531. 1969 (In Japanese). 20) Hofmann, H Th.: Neuere Errahrungen mit hochtoxischen CMorkohler.wasserstoiTen Arch. exp. Pathol. Pharmakol. 232 : 223-230, 1958. NV 3 7 Q 3 6