首页   按字顺浏览 期刊浏览 卷期浏览 Gas-liquid chromatographic determination of alpha-, beta-, gamma- and delta-BHC levels ...
Gas-liquid chromatographic determination of alpha-, beta-, gamma- and delta-BHC levels in human blood, depot fat and various organs with the use of 2,2-dimethylpropane-1,3-diol succinate as the stationary liquid phase

 

作者: G. Czeglédi-Jankó,  

 

期刊: Analyst  (RSC Available online 1973)
卷期: Volume 98, issue 1173  

页码: 863-872

 

ISSN:0003-2654

 

年代: 1973

 

DOI:10.1039/AN9739800863

 

出版商: RSC

 

数据来源: RSC

 

摘要:

Analyst, December, 1973, Vol. 98, $9. 863-872 863 Gas - Liquid Chromatographic Determination of Alpha-, Beta-, Gamma- and Delta-BHC Levels in Human Blood, Depot Fat and Various Organs with the Use of 2,2=Dimethylpropane- 1 ,3-diol Succinate as the Stationary Liquid Phase BY G. CZEGLEDI- JANKG (Institute for Chemistry and Food Analysis, H-1022 Hermann Ottd u. 15, Budapest, Hungary) The presence of various BHC isomers in the human organism has received relatively little attention, and studies were often restricted to only one BHC isomer. A previously described one-step extraction and clean-up procedure before gas - liquid chromatographic determination of organochlorine pesticide residues in human blood has now been applied to various organs and depot fat. The identification of the BHC isomers was performed by gas - liquid chromatography with several stationary liquid phases.2,Z-Dimethylpropane- 1,3-diol succinate was found t o be the most satisfactory stationary liquid phase for the present purpose, as it gave distinctly separated peaks and characteristic relative retention times. OF the investigations that have been carried out on the accumulation of organochlorine pesticide residues in the human organism, only a few have dealt with the presence of BHC isomers and most papers have been concerned with DDT and its metabolites. For instance, the presence of gamma-BHC in the depot fat of the French population was first demonstrated by Hayes, Dale and LeBreton,l but they treated the results with some reservation. Their findings were later confirmed by other studies, but nevertheless the presence of the various BHC isomers in human biological material has not received great attention.Many workers do not mention the presence of BHC isomers in the human organism at all, and Table I, which is far from complete, demonstrates that some authors who have investigated this problem restricted their study to only one BHC isomer. TABLE I REFERENCES TO BHC ISOMERS IN THE HUMAN ORGANISM Authors Reference BHC isomer mentioned Specimen Hayes, Dale and LeBreton . . .. 1 Dale, Curley and Cueto . . .. 4 Dale, Curley and Hayes . . .. 5 DBnes and TarjAn . . . . .. 6 Engst, Knoll and Nickel . . .. 7 Abbott, Goulding and Tatton . . 8 deVlieger et al. . . .. .. 9 Curley, Copeland and Kimbourgh . . 10 Milby, Samuels and Ottoboni .. 11 Dacre . . . . .. . . . . 12 Kadis, Breitkreitz and Jonasson . . 13 Samuels and Milby . . .. . . 14 Radomski et aE. . . .. . . 15 Egan, Goulding, Roburn and Tatton Radomski and Fiserova-Bergerova . . 2 3 Gamma Beta + “total BHC” Gamma Beta Beta, gamma Alpha, beta, gamma Gamma Beta + “total BHC” Alpha, beta, gamma Gamma Gamma Alpha, beta, gamma Alpha Gamma Alpha, beta, gamma Adipose tissue Adipose tissue, milk Adipose tissue, organs Blood Blood Foetal liver (one case) Adipose tissue Adipose tissue Adipose tissue, organs, blood Adipose tissue Blood Adipose tissue, organs, blood Adipose tissue, organs Blood Blood The intention of the present study was to investigate the determination of the four important BHC isomers in human biological material by means of an appropriate gas - liquid chromatographic method.@ SAC and the author.864 MATERIALS- Brain, liver, kidney and gonad and abdominal depot fat samples taken from ten healthy persons killed in accidents and dissected at the Institute of Forensic Medicine of the Semmel- weis Medical School, Budapest, were investigated. The specimens were selected randomly from subjects of both sexes (aged 32 to 66 years) and stored in a refrigerator between dissection and processing. Whole blood samples were taken from ten healthy donors and investigated without delay. CZEGLI~DI-JANK~ : GAS - LIQUID CHROMATOGRAPHIC DETERMINATION [Analyst, Vol. 98 EXPERIMENTAL METHODS- In a previous paper,16 a one-step extraction and clean-up procedure for the gas - liquid chromatographic determination of organochlorine pesticide residues in human blood samples was described.This method was modified so as to be suitable for the determination of alpha-, beta-, gamma- and delta-BHC isomers in human blood, adipose tissue (depot fat) and various organ samples by altering the composition of the material used for the clean-up (sodium sulphate and Florid were not used) and by using several different stationary liquid phases. Extraction and clean-up of samples-Organ samples containing about 0.01 to 0-1 g of fat were ground with an equal amount of sand and then dehydrated by lyophilisation in an Edwards High Vacuum Ltd. centrifugal freeze dryer, Model 30 PI-512, with a secondary drying unit, Model 30 SI-332, or equivalent equipment. The blood samples (2 to 10 ml) were lyophilised without pre-treatment.One-step extraction and clean-up was carried out in the apparatus described previously.16 In the present study, only 1.8 ml of fuming sulphuric acid (10 per cent. of SO,) mixed with 2 g of Hy-Flo Super Cel was used in the clean-up and 25 to 30 ml of light petroleum (boiling range 35 to 40 “C) or n-pentane (gas-chromatographic purity) was used as the extraction solvent. Adipose tissue samples were also lyophilised and subsequently extracted with light petroleum. The extracted pure depot fat (0-1 g) was placed in the apparatus above the clean-up material and processed as described above. With extracted depot fat, solvent extrac- tion for 30 minutes was sufficient. Most of the light petroleum or n-pentane in the flask was evaporated by means of a rotary evaporator, and the remainder was transferred into a conical tube and 1 ml of n-hexane was added.After the evaporation of the residual light petroleum (n-pentane), the 1 ml of n-hexane solution (diluted, if necessary) proved to be suitable for gas - liquid chromatographic analysis. Determination of the fat content of organ samples-The organochlorine pesticide level can be expressed in parts per million of wet organ tissue or of extractable fat, as recommended by Kadis, Breitkreitz and Jonasson,13 and the fat content of the organs therefore had to be determined. For this purpose, organ samples were ground with sand, lyophilised and extracted with light petroleum. Hence in this study the “fat content” represents the lipids extracted with light petroleum from the lyophilised organs (see Table VI).Gas - liquid chromatographic analysis of puyi$ied extracts-A Packard gas chro- matograph, Model 7934, was used. The general operating conditions were as follows- detector: electron capture, tritium foil, 200 mC, applied potential 50 V; electrometer range : 1 x A; column: Pyrex glass, 4 mm i.d., on-column injection; solid support : Chromosorb W HMDS, 60 to 80 mesh; carrier gas: nitrogen, high purity; temperature: inlet, 235 “C; detector, 200 “C; outlet, 235 “C; chart speed: 30 in h-l. Operating conditions that were varied for the different columns are given in Table 11. The extraction time was 3 hours. IDENTIFICATION AND RECOVERY OF BHC ISOMERS IN HUMAN BIOLOGICAL MATERIAL- The BHC isomers were identified by comparing the retention times on each of the four columns with the retention times of the reference substances. Figs.1 to 4 show the chromato- grams obtained from a human liver extract (female, 46 years old) on the four columns, together with the peaks and retention times of pure alpha-, beta-, gamma- and delta-BHC isomers and of Pp’-DDE as the internal standard. Table I11 gives the retention times of the BHC isomers relative to those of gamma-BHC and to pP’-DDE. pp’-DDE was chosen because it was present in all of the samples examined owing to environmental pollution.December, 19731 OF BHC IN HUMAN BLOOD, DEPOT FAT AND VARIOUS ORGANS TABLE I1 OPERATING CONDITIONS FOR GAS - LIQUID CHROMATOGRAPHY 865 Column r A \ Variable . . .. .. .... I I1 I11 IV Stationary liquid phase . . .. * . QF-1 Concentration of liquid phase, per cent. Column length/cm . . .. .. 180 Inlet pressure/p.s.i. . . . . .. 10 (silicone resin) 5 Carrier gas flow-ratelm1 min-1 . . .. 50 Column temperature/”C . . .. .. 170 OV- 1 7 (silicone resin) 5 60 50 6 185 XF-1112 (silicone nitrile resin) 120 60 13 180 2.5 2,2-Dimethyl- propane-1,3- diol succinate 2 120 70 12-5 185 For the identification of the BHC isomers thin-layer chromatography, according to Kov~cs,~’ was used as a complementary method. The extracts of each organ were mixed together and, after evaporation of the solvent in a rotary evaporator, a suitable amount of the residue was placed on a thin layer of alumina G. The developed chromatograms were compared with those for a series of standard BHC isomers, for which the RF values of the standards were: alpha-BHC 0.28, gamma-BHC 0.21, beta-BHC 0.04 and delta-BHC 0.02.F R.eference substances K, I X I 4 L Fig. 1. Gas -liquid chromatogram of an acid-cleaned extract from 2 g of lyophilised human liver. Volume of extract, 1 ml; volume injected, 10 pl; liquid phase, QF-1 silicone resin. Peaks: A, alpha-BHC (0.004 ng) ; B, gamma-BHC (0.02 ng) ; C, beta-BHC (1.1 ng) : D, delta-BHC (0.005 to 0.01 ng) ; E, pfi’-DDE; F, pp’-DDD; G, pp’-DDT; and X, unidentified. Reference substances: H, alpha-BHC (0.05 ng) ; I, gamma- BHC (0.05 ng) ; J, beta-BHC (0.5 ng) ; K, delta-BHC (0.1 ng) ; and L, pp’-DDE (internal standard) The reliability of the one-step extraction and clean-up of the lyophilised samples and extraction time have been discussed in a previous paper,16 and it was shown that the recovery of added gamma-BHC was highly unsatisfactory owing to its evaporation during lyophilisation.Therefore, it was necessary to investigate whether or not the BHC isomers, transferred through metabolic pathways to the organs, would be lost during lyophilisation. (Dale, Miles and Gained8 used compounds labelled with carbon-14 and found that DDT incorporated by metabolic pathways is bound more strongly to biological material than is added DDT).X ,Injection point X Fig. 2. Gas - liquid chromatogram of an acid-cleaned extract from 2 g of lyophilised human liver. Volume of extract, 1 ml; Peaks: A, alpha-BHC; B, gamma-BHC ((3.02 ng); C, beta-BHC (1.0 ng); Reference substances: H, alpha-BHC volume injected, 10 p1; liquid phase, OV-17 silicone resin. D, delta-BHC (0.005 to 0.01 ng) ; E, pp'-DDE; F, pp'-DDD; G, pp'-DDT; and X, unidentified.(0-025 ng) ; I, gamma-BHC (0.025 ng) ; J , delta-BHC (0.05 ng) ; and K, pp'-DDE (internal standard) Z 's 0 Z 3A n ‘Injection point F I II 1 Fig. 3. Gas - liquid chromatogram of an acid-cleaned extract from 2 g of lyophilised human liver. Volume of extract, 1 ml; volume injected, Peaks: A, alpha-BHC; B, gamma-BHC (about 0.02 ng) ; C, beta-BHC (1.2 ng) ; D, delta-BHC Reference substances: G, alpha-BHC (0.025 ng) ; H, gamma-BHC 10 pl; liquid phase, XF-1112 silicone nitrile resin. (about 0.015 ng) ; E, pp’-DDE; F, pp’-DDT plus fip’-DDD; and X, unidentified. (0.025 ng) ; I, beta-BHC (0.5 ng) ; J, delta-BHC (0.05 ng) ; and K, fip’-DDE (internal standard)868 CZEGLkDI-JANK6 : GAS - LIQUID CHROMATOGRAPHIC DETERMINATION [AflaZySi!, VOl.98 X -Injection point Reference substances I F Fig. 4. Gas - liquid chromatogram of an acid-cleaned extract from 2 g of lyophilised human liver. Volume of extract, 1 ml; volume injected, 5 pl; liquid phase, 2,2-dimethyl- propane-1,3-diol succinate. Peaks: A, alpha-BHC (0.002 ng) ; B, gamma-BHC (0.01 ng) ; C , pP’-DDE; D, delta-BHC (0.005 ng) ; E, beta-BHC (0.52 ng) ; and F, pp‘-DDT plus pp’-DDD. Reference substances: G, alpha-BHC (0.005 ng); H, gamma-BHC (0.005 ng); I, pp’-DDE (internal standard) ; J, delta-BHC (0.01 ng) ; and K, beta-BHC (0.05 ng) In this study, such evidence could be obtained only by applying an indirect method of recovery, i.e., the method described by Stanley and LeFavoure.l9 The samples were digested with a 1 + 1 mixture of acetic acid and 70 per cent.perchloric acid. The final clean-up step of this method was modified by using our circulation apparatus with the clean-up material described above. Amounts of 0.01 p.p.m. each of alpha-, gamma- and delta-BHC and 0.1 p.p.m. of beta-BHC were added to whole blood, homogenised liver and homogenised adipose tissue samples, which TABLE I11 RELATIVE GAS - LIQUID CHROMATOGRAPHIC RETENTION TIMES OF BHC ISOMERS AND P$’-DDE Relative retention times Stationary liquid phase* Stationary liquid phaset A A r 3 \ Compound OF-1 OV-17 XF-1112 DPDSt OF-1 01-17 XF-1112 DPDSZ alpha-BHC 0.28 0.18 0.26 0.30 0.79 0.75 0.79 0.68 beta-BHC 0.42 0.29 0.33 1.64 1-16 1.20 2.17 3-72 delta-BHC 0.47 0.34 0.79 1.34 1.32 1.44 2.44 3.11 pp’-DDE 1.00 1.00 1.00 1.00 2.78 4.20 3.10 2.27 gamma-BHC 0.36 0.24 0.7 1 0.44 1.00 1.00 1.00 1.00 * Relative to pp’-DDE = 1.00.t Relative to gamma-BHC = 1-00. # 2,2-Dimethylpropane-1,3-diol succinate.December, 19731 869 were then exarnined by the method of Stanley and LeFa~0ure.l~ Samples without added isomers were also examined. Our results for recovery were expressed as the means of results from three samples, as recommended by Kadis, Jonasson and Breitkreitz2* Table IV shows that the recovery of added BHC isomers was satisfactory. OF BHC IN HUMAN BLOOD, DEPOT FAT AND VARIOUS ORGANS TABLE IV RECOVERY OF BHC ISOMERS FOLLOWING THE ADDITION OF 0.01 p.p.m.OF ALPHA-, GAMMA- AND DELTA-BHC AND 0.1 p.p.m. OF BETA-BHC TO WHOLE BLOOD, HOMO- GENISED LIVER AND HOMOGENISED ADIPOSE TISSUE, USING THE MODIFIED METHOD OF STANLEY AND LEFAVOURE19 Results are mean values & standard deviations, from three samples. Results obtained by gas - liquid chromatography with 2,2-dimethyIpropane-l,3-dio~ succinate as the stationary liquid phase Recovery of pesticide added, per cent. A I \ Sample alpha-BHC beta-BHC gamma-BHC delta-BHC Whole blood . . . . 96.4 f 2-8 98.0 f 0.8 98.7 f 1.2 94.2 f 2.4 Liver . . .. . . 92.6 f 7.6 95.6 f 1.2 101.2 & 0.2 97.2 f 1.0 Adipose tissue* . . 88.4 f 7.2 105.2 f 0.8 89-8 & 3.2 90.0 f 3.6 * The results in this trial are referred to the total adipose tissue. Subsequently, samples were examined repeatedly by means of our lyophilisation and one-step extraction - clean-up method and by the method of Stanley and LeFavoure. The results of our lyophilisation method and those obtained by the perchloric acid digestion (the latter confirmed by the recovery of added BHC isomers) are in good agreement, with the usual accuracy achieved in gas - liquid chromatographic analyses (see Table V).RESULTS Quantitative determinations were made by comparing the gas - liquid chromatographic peak areas of the BHC isomers with those of the reference substances, and the results are summarised in Table VI for 2,2-dimethylpropane-1,3-diol succinate as the stationary liquid phase. As the amounts of BHC isomers in the samples differ considerably, it cannot be assumed that all of the BHC isomers are in the linear range of detection in the solution injected into the chromatographic column.The results therefore do not refer to a single reference solution, but to a series of standards of various concentrations, which were prepared so as_ to cover the range of the sample peaks. The results in Table VI also show the agreement between the results obtained following the two methods of preparation. Owing to the large number of investigations that would be necessary, it was not possible to obtain all of the corresponding results following lyophilisa- tion and perchloric acid digestion, but the range and mean values of the results verify the agreement between the two methods. DISCUSSION In applying the modified one-step method to demonstrate the presence of BHC isomers in human biological material, the following difficulties had to be overcome.Firstly, the original procedurel6 proved suitable for the recovery of added DDT, its metabolites and dieldrin, but not for that of added BHC. We therefore had to find an indirect method by which we could prove reliably that although added gamma-BHC would evaporate during lyophilisation, gamma-BHC and other BHC isomers incorporated through metabolic pathways would be retained. Lyophilisation, subsequent one-step extraction and clean-up in the circulation apparatus have the advantages that the lyophilised samples can be stored without refrigeration and can be easily and cleanly processed. The volume of solvent needed for extraction is not more than 25 to 30 ml. The second problem was to find the optimum gas - liquid chromatographic operating conditions that would result in good separations and make the identification of BHC isomers reliable.The chromatographic columns in general use for determining organochlorine pesticide residues from biological materials have relatively short retention times for BHC isomers.870 CZEGLBDI- JANKd GAS - LIQUID CHROMATOGRAPHIC DETERMINATION [Analyst, VOl. 98 On OV-17 as the stationary liquid phase, the determination of alpha-BHC is very uncertain owing to interference by an unknown component extracted from acid-cleaned biological material. In the present study, four stationary liquid phases were used in the identification of BHC isomers. It was found that for the separation of BHC isomers, the use of 2,2-dimethylpropane- 1,3-diol succinate as the stationary liquid phase has the particular advantage that it gives distinctly separated peaks and characteristic relative retention times, especially for beta- and delta-BHC.Dale, Curley and Cueto4 and McClure21 have already reported that diethylene- glycol succinate has the same advantage, but it has not been widely used. Its main dis- advantage is its volatility and its tendency to contaminate the electron-capture detector. The use of 2,2-dimethylpropane-1,3-diol succinate as the stationary liquid phase enables temperatures above 200 “C to be used without the risk of rapid “end-blooding.” As shown in Figs. 1 to 4, acid-stable DDT, its metabolites and unidentified peaks, which were probably due to co-extracts or artifacts, also appeared in the gas - liquid chromatograms.In this paper, however, only the gas - liquid chromatographic analytical problems with BHC isomers are considered. TABLE V REPEATED INVESTIGATIONS OF ALPHA-, BETA-, GAMMA- AND DELTA-BHC CONTENTS OF LIVER, ADIPOSE TISSUE AND WHOLE BLOOD SAMPLES I N ORDER TO COMPARE THE LYOPHILISATION AND PERCHLORIC ACID DIGESTION METHODS Gas - liquid chromatographic analysis with 2,2-dimethylpropane-1,3-diol succinate as the stationary liquid phase Sample Method alpha-BHC beta-BHC gamma-BHC delta-BHC Liver* Lyophilisation 0.0008 0.320 0.0052 0.0003 0.0009 0.340 0-0068 0.0003 0*0010 0.342 0.0060 O*OOO 25 0.00 12 0,334 0.0064 0*0003 0.0012 0.340 0.0058 0.0003 0~0010 0.330 0.0062 0.0003 0~0010 0.338 0.0068 0*0003 0.0260 7.16 0.124 0.0062 0,0265 7.84 0.142 0.0062 0.0300 7.30 0.134 0.0080 0.0284 7.42 0.120 0.0060 0.132 0.0042 0.0280 7.24 0.0320 7.46 0.1 10 0.0068 0.0250 7.32 0.118 0*0080 0.0264 7.38 0.138 0.0056 0.012 24.6 2.44 0.026 0.016 24.0 2.00 0.020 0-015 24.2 2.16 0.022 0.015 24.9 2-28 0.032 0.010 24.8 2.00 0.024 0.020 25.0 2.22 0.026 0.018 24.8 2-24 0.020 0.014 24-0 2.22 0.024 Perchloric acid digestion 0.00 10 0.328 0-0056 0~0002 Adipose Lyophilisation tissue? Perchloric acid digestion Whole Lyophilisation blood $ Perchloric acid digestion * Pesticide content expressed in p.p.m.t Pesticide content expressed in p.p.m., referred t o the total adipose tissue. 1 Pesticide content expressed in p g 1-l. We also used thin-layer chromatography as a complementary method. The thin-layer chromatograms alone, especially those for compounds with low RF values, such as beta- and delta-BHC, would not provide conclusive evidence of identification. However, proof of identity was provided by the four gas - liquid chromatographic columns, and corresponding evidence was obtained when spots with the same RF values as those of the reference substances, and with corresponding areas as expected, also appeared on the thin-layer chromatograms.The results of the representative runs given in Table VI do not include the results of preliminary experiments and ancillary investigations. Nevertheless, it is worth noting thatDecember, 19731 87 1 these other results fell within the range of the tabulated results for the representative runs, except for the blood used to compare the lyophilisation and perchloric acid digestion methods of preparation, which had a lower alpha-BHC content than the lowest alpha-BHC level recorded for the ten samples in the representative series.OF BHC I N HUMAN BLOOD, DEPOT FAT AND VARIOUS ORGANS TABLE VI PESTICIDE CONTENT (p,p.m.) IN VARIOUS ORGANS AND PURE DEPOT FAT OF HEALTHY PERSONS KILLED IN ACCIDENTS AND IN WHOLE BLOOD (pg I-') OF HEALTHY DONORS Mean values and ranges are given for ten determinations, except where stated otherwise. Gas - liquid chromatography with 2,2-dimethylpropane-2,3-diol succinate as the stationary liquid phase Mean fat content, Specimen per cent. Method Brain, frontal 3.68 Lyophili- sation Liver Perchloric acid digestion 4.67 Lyophili- sation Cerebellum 5.62 Kidney Gonads 6-24 1.64 Abdominal - depot fat Whole - blood? Perchloric acid digestion Lyophili- sation Perchloric acid digestion Lyophili- sation Perchloric acid digestion Lyophili- sation Perchloric acid digestion Lyophili- sation Lyophili- sation Value Mean Range Mean Range Mean Range Mean Range Mean Range Mean Range Mean Range Mean Range Mean* Range* Mean* Range* Mean Range Mean Range alpha-BHC 0~0010 0-0001 to 0.0034 o*ooo 73 0*0001 to 0.0029 0.001 39 0~0002 to 0.0033 0.001 30 0.0002 to 0.0029 0.002 22 0~0002 to 0.0057 0.002 27 0.0002 to 0.0060 0.001 85 0.0003 to 0.0043 0.001 67 0-0003 to 0-0042 0.000 78 0.000 03 to 0.0021 0.000 77 0.000 06 to 0.0018 0.041 0.009 to 0.086 0.334 0.06 to 0.80 beta-BHC 0.0512 0.0323 to 0.1060 0.051 86 0.0296 to 0.1114 0.3842 0.0620 to 0.7721 0.3308 0.0543 to 0.6412 0-106 10 0,0161 to 0.2985 0.068 60 0.0169 to 0.2870 0.297 73 0.0337 to 1.040 0.2784 0.0345 to 0.8563 0.172 37 0.0783 to 0.3910 0.166 08 0,0756 to 0.3801 9.39 3.56 to 16.3 25.9 16.0 to 36.0 gamma-BHC 0,005 79 0.0023 to 0-0234 0.006 71 0.0027 to 0.0248 0.005 32 0.0009 to 0-0223 0.0055 0.0006 to 0.0286 0.005 80 0.0008 to 0.0268 0,005 75 0.0007 to 0.0280 0.006 49 0.0017 to 0.0224 0.006 39 0.0017 to 0.0192 0.001 99 0*0009 to 0.0048 0.001 89 0.0006 to 0.0048 0-135 0.020 to 0-348 1.628 0.149 to delta-BHC 0.000 22 0.000 05 to 0.0005 0~000 21 0.000 05 to 0.0005 0-000 47 0~000 02 to 0.0015 0.000 40 0.000 02 to 0.0014 O*OOO 38 O*OOO 03 to 0*0020 0.000 50 0.000 03 to 0.0024 0.0006 0-000 02 to 0.0033 0.000 64 0.000 02 to 0.0036 0-000 144 0~000 02 to 0.0004 0.000 15 o*ooo 02 to 0*0004 0.008 0.001 to 0.034 0.104 0.0 (not 2.88 detectable) to 0.63 * Five determinations.t Results expressed in pg ml-l. In this study, the gas - liquid chromatographic analytical problems of the BHC isomers in biological materials were considered from the point of view of their satisfactory separation. However, we did not deal with the problem of the separation of other organochlorine com- pounds from the BHC isomers; these compounds have recently become important in the pollution of the environment, e.g., hexachlorobenzene, chlorine-containing plasticisers and polychlorobiphenyls. These compounds can appear as interferents in both thin-layer and gas - liquid chromatography. At present in Hungary, the presence of these compounds in the environment and in the general population does not cause any problems, as confirmed by other investigations. Nevertheless, if the presence of these compounds cannot be neglected872 CZEGLI~DI- J A N K ~ in the gas - liquid chromatographic separation of BHC isomers, one should remove such interfering materials, such as in pesticide residue analysis in the presence of polychlorobi- phenyls, as described by Reynolds,22 or in the investigation of the identity of hexachloro- benzene according to Zemann.23 The scope of the present study does not permit a statistical evaluation to be made of the results in terms of the sex and age of subjects. Also, it was not our aim to discuss the toxicological implications of the BHC isomers, to study the possibility of post mortem changes in BHC, mentioned by French and Jefferies,Z4 or to consider the problem of large amounts of beta-BHC. Our aim was simply to develop a method by which we could separate the BHC isomers from human biological materials with good reproducibility.The author expresses his thanks to Dr. Zsuzsanna Balla for making available the organ samples, Mr. A. Ho116 and Miss Zsuzsanna Pksztor for technical assistance and Mr. K. Krajcziir for lyophilisation of samples. This work was carried out in the State Institute of Hygiene, Budapest. 1. 2. 3. 4. 5. 6. 7 8 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22 23 24 REFERENCES Hayes, W. J., jun., Dale, W. E., and LeBreton, R., Nature, Lond., 1963, 199, 1189. Egan, H., Goulding, R., Roburn, J., and Tatton, J. O’G., BY. Med. J., 1965, 2, 66. Radomski, J . L., and Fiserova-Bergerova, V., Ind. Med. Surg., 1965, 34, 934. Dale, W. E., Curley, A., and Cueto, C., jun., Life Sci., 1966, 5, 47. Dale, W. E., Curley, A., and Hayes, W. J., jun., Ind. Med. Surg., 1967, 36, 275. DCnes, A., and TarjAn, R., Magy. Tudom. Akud. V . Osztdly Kozl., 1967, 18, 379. Engst, R., Knoll, R., and Nickel, B., Pharmazie, 1967, 22, 654. Abbott, D. C., Goulding, R., and Tatton, J. O’G., BY. Med. J., 1968, 3, 146. deVlieger, K. M., Robinson, J., Baldwin, M. K., Crabtree, A. N., and Dijk, M. C., Archs Envir. Curley, A., Copeland, M. F., and Kimbourgh, D. R., Ibid., 1968, 19, 628. Milby, T. H., Samuels, A. J., and Ottoboni, F., J . Occup. Med., 1968, 10, 584. Dacre, J. C., Proc. Univ. Otago Med. Sch., 1969, 47, 74. Kadis, V. W., Breitkreitz, W. E., and Jonasson, 0. J., Can. J . Publ. Hlth, 1970, 61, 413. Samuels, A. J., and Milby, T. H., J . Occup. Med., 1971, 13, 147. Radomski, J. L., Astolfi, E., Deichmann, W. B., and Rey, A. A., Toxic Appl. Pharmuc., 1971, CzeglCdi- Jank6, G., and Cieleszky, V., Analyst, 1968, 93, 445. Kovacs, M. F., J . Ass. 08. Analyt. Chem., 1950, 33, 130. Dale, W. E., Miles, J. W., and Gaines, T. B., Ibid., 1970, 53, 1287. Stanley, R. L., and LeFavoure, H. T., Ibid., 1965, 48, 666. Kadis, V. W., Jonasson, 0. J., and Breitkreitz, W. E., Can. J . Publ. Hlth, 1968, 39, 357. McClure, V. E., J . Chromut., 1972, 70, 168. Reynolds, L. M., Residue Rev., 1971, 34, 27. Zemann, A., Naturwissenschaften, 1971, 58, 276. French, H. C., and Jefferies, D. J., Nature, Lond., 1968, 219, 164. Hlth, 1968, 17, 759. 20, 186. Received July 29th, 1971 Amended June 18th, 1973 Accepted July 19th, 1973

 

点击下载:  PDF (795KB)



返 回