Vulnerability of keto bile acids to alkaline hydrolysis

Transcription

1 Vulnerability of keto bile acids to alkaline hydrolysis G. Lepage, A. Fontaine, and C. C. Roy Department of Pediatrics, H6patal Ste-Justine and the University of Montreal, Montreal, Canada Summary Rigorous alkaline hydrolysis of the to primary (cholic and chenodeoxycholic) and of the to preponderant secondary (deoxycholic and lithocholic) bile acids found in bile led to excellent recoveries. Such as not the case ith different keto bile acid standards. Recoveries for a number of standardsereunacceptably lo and a variety of artefactual products ere tentatively identified by gasliquid chromatography. Keto bile acids bearing a keto group on C- ere particularly vulnerable. In vie of these findings, quantitative and qualitative data reported on biological specimens submitted to saponification in ethanol, methanol, or even in ater are of questionable significance. Supplementary key ords primary bile acids. secondary bile acids * gas-liquid chromatography saponification enzymatic hydrolysis Keto bile acids are a major constituent of human feces (-). These products of anaerobic bacterial dehydrogenases () are found in the bile of a number of mammalian species () but have not been reported in man except after cholecystectomy (6, ). Hydrolysis is a necessary step for the analysis of bile acids. Freeing the amino acid residues from the carboxyl end of the molecule can be accomplished by either alkaline or enzymatic hydrolysis. Attention has been dran to the formation of artefactual bile acid metabolites by rigorous alkaline hydrolysis (8, 9). Although the recovery of bile acids after alkaline hydrolysis has been reported to be particularly lo for keto bile acids (8, lo), the extent of structural changes and the particular vulnerability of various keto bile acids have not been studied. Materials and Methods A mixture of bile acid standards consisting of. mg each of lithocholic, nordeoxycholic (NDC), deoxycholic, chenodeoxycholic, cholic, and tri-keto cholanic acid as saponified in open glass tubes ith ml of N NaOH in % ethanol for a period of 9 min in an autoclave, at psi and C. After Abbreviations: Trivial names of bile acids in the text refer to hydroxy-substituted fi-cholanic acids as follos: lithocholic, ; NDC, -nordeoxycholic,,; deoxycholic, a,a; chenodeoxycholic, a,a; cholic, a,a,a. Derivatives ith keto or mixed hydroxyl and keto functions are designated by positions of the functions: tri-keto cholanic acid, dehydrocholic,,,- triketo. GLC, gas-liquid chromatography. acidification to ph (in ice to prevent the formation of ethyl esters'), bile acids ere extracted thrice ith ether and tice ith ethyl acetate.' The samples ere then methylated ith freshly prepared diazomethane () and one drop of acetic acid as added to remove the excess. After adding ater, bile acid methyl esters ere extracted ith ether and ethyl acetate. The dry residue as acetylated (). Five p of the sample dissolved in acetone as analyzed on a Helett-Packard 6A gas chromatograph ith a flame ionization detector. Each sample as chromatographed on to different liquid phase columns. A 6-ft U-shaped glass column as packed ith % QF- on Chromosorb W (HP) - mesh, previously reported () to ensure best resolution of keto forms. The other column as U-shaped and measured ft. It as packed ith % OV- using the same support. Temperature programming as necessary in order to allo better separation and identification of bile acid species ith nearly identical polarities. Reproductibility as checked on QF- from to C as ell as on OV- from to C. Linear programming as used. Peak area measurements ere carried out during GLC using an Autolab 6 automatic digital integrator. Similar experiments ere carried out ith. mg of each of the folloing P-cholanic keto acid derivatives: -monoketo; -keto,a-hydroxy; -keto,a, adihydroxy; a-hydroxy,-keto; a, a-dihydroxy,- keto; Sa-hydroxy, -keto; Sa,a-dihydroxy, -keto;,-diketo;,-diketo; a-hydroxy,,-diketo;,, -triketo. These standards (Steraloids Inc. Wilton, NH) ere estimated to be more than % pure by GLC. Ethanol (%), methanol (%), and ater ere tested as saponification solvents. Each step of the pro- ' Lepage, G., A. Fontaine, and C. C. Roy, unpublished observations. * All organic solvents ere high purity grade. TABLE. Effect of alkaline hydrolysis in ethanol % on a mixture of free bile acid standards Methyl P-Cholanoate Artefactual Products Acetate Recovery of Derivatives" Standard Identificationb % Lithocholic 99. -Nordeoxycholic 9. Deoxycholic. Chenodeoxycholic. Cholic 96.8,,-Triketo -ketodeoxycholic -ketochenodeoxycholic 6,-diketolithocholic 6 a Mixture of. mg of each standard. Tentative identification based on relative retention time as compared to internal standard (NDC) on % QF- and % OV-. Journal of Lipid Research Volume 9, 98 Notes on Methodology

2 % QF on Chromosorb W - mesh m Z u) & & c!i c n U - x U c J 6 minute \ \ J \ L / ; % QF I on Chromosorb W - mesh LJ minuem i Fig.. Representative GLC patterns of free bile acid standards. The bile acids derivatives ere obtained ithout prior saponification (upper chromatogram) and after saponification (loer chromatogram). The,,-triketo derivative as transformed into -ketodeoxycholic, -ketochenodeoxycholic, and,- diketolithocholic by ethanolic alkaline hydrolysis. 6 Journal of Lipid Research Volume 9, 98 Notes on Methodology

3 ~~ TABLE. Effect of alkaline hydrolysis on keto bile acid standards Ethanol % Methanol % Water % Re- % Re- % Recovered of Artefactual Products covered of Artefactual Products covered of Artefactual Products Methyl 8-Cholanoate Original Original Original Acetate Derivative Standard Identificationb % Standard Identification % Standard Identification % keto keto,a hydroxy C keto,a,la dihydroxy, a,a, dihydroxy,,l keto keto, diketo, diketo,. diketo,, triketo a,a,a trihydroxy a,a, a trihydroxy a,a dihydroxy a, hydroxy. a.~ dihydroxy a, keto a, a dihydroxy a,a, dihydroxy, Sa,a,dihydroxy, keto a,a dihydroxy,, keto,, diketo a,a,a 88 trihydroxy 6 a, a dihydroxy, 6,l keto 9 a,a dihydroxy a,la,dihydroxy, 9, keto, keto,, diketo a,a,a trihydroxy N.D.P.d N.D.P N.D.P., N.D.P. N.D.P a,a,a trihydroxy a,a dihydroxy, a,a,l keto ~~ ~ ~~ ~ ~ a. mg of each standard. *Tentative identification based on relative retention time as compared to internal standard (NDC) on % QF- and % OV-., unidentified peak. N.D.P., no detectable peak. 9 cedure as controlled for the presence of solvent contamination by running solvent blanks. The relative retention times of standards and of artefactual products ere measured by comparison of area response ith that of NDC. The possibility that alkaline hydrolysis of the internal standard could contribute to the production of artefactual peaks as ruled out since each GLC run as made ith and ithout NDC ( g). Results and discussion Table summarizes the effect of alkaline hydrolysis on a mixture of free bile acid standards. Rigorous alkalinehydrolysisof the four bileacidsnormally found in bile and of the internal standard (NDC) did not lead to the formation of any artefactual products and recoveries ere excellent. Hoever, this as not the case for the,,-triketo compound. The chromatogram (Fig. ) shos nearly complete disappearance of the,,-triketo-/-cholanic acid that had been added to the bile acid mixture. The artefactual products ere tentatively identified as -ketodeoxycholic, -ketochenodeoxycholic, and,-diketolithocholic acid. Repeated experiments carried out ith each of keto bile acid standards are shon in Table. Ethanolic hydrolysis caused severe transformations. None of the keto bile acids bearing a keto group at C- could be recovered after saponification (Fig. ). The five standards ith keto groups at C- or C- ere Journal of Lipid Research Volume 9, 98 Notes on Methodology

4 Z QF I on Chromosorb W - I mesh cn Z n cn Y c V Ly 8- Ly n minute s X QF I on Chromosorb W I - mesh " 'C r minute S Fig.. These to chromatograms (QF-) illustrate the extensive degradation of -keto-deoxycholic acid after ethanolic saponification. Only % of the original standard as recovered (Table ) (loer chromatogram). 8 Journal of Lipid Research Volume 9, 98 Notes on Methodology

5 more stable; recoveries ranged from to 9%. It should be noted that our recoveries of the -keto, -keto-deoxycholic, -keto-chenodeoxycholic, and, diketo compounds ere loer than those reported previously under similar conditions of ethanolic saponification (). Substitution of ethanol for methanol did not modify the recoveries. The data confirm the previously mentioned (8, ) drastic structural changes that keto bile acids undergo during alkaline hydrolysis. Water hydrolysis appeared to effect feer changes on keto standards; recoveries ere lo for compounds ith to or three keto groups. This could relate to the extraction procedure after hydrolysis since the loss is not made up by artefactual products. Of interest is the fact that the recovery for the,-diketo compound by Grundy, Ahrens, and Miettinen () as also loer than for the monoketo standards, even though a different extraction procedure as used. In vie of our findings, attention is dran to the likelihood that keto bile acids in bile (, ) and in feces (, ), here they are a major constituent, are quantitatively and qualitatively unreliable. In the study by Soloay et al. () the biotransformation of,,-triketo-/-cholanic acid (dehydrocholic acid) in man could also be explained by the effects of alkaline hydrolysis rather than by hepatic metabolism. The same stereospecificity as noted in the present study since only a epimers ere recovered among the artefactual products. Keto bile acids are particularly vulnerable to alkaline hydrolysis hen it is carried out in ethanol or methanol. There is a significant advantage to the use of ater as a saponification medium. Since a prerequisite for enzymatic hydrolysis is the presence of free hydroxyl groups on the steroid skeleton (6, ), it is apparent that enzymatic cleavage of the carbon-nitrogen bond does not provide a solution to the methodological problem underlined in the present reprt.m We thank Doctor A. Kuksis of the University of Toronto for his invaluable help and encouragement. This paper as supported by Grant MT of the Medical Research Council of Canada and by the Canadian Cystic Fibrosis Foundation. Manuscript received January I9 and in revised jorm 8 October 9; accepted Novembpr 9. REFERENCES. Danielson, H., P. Einarsson, K. Hellstrom, S. Lindstedt, and J. Sjovall. 96. On the turnover and excretory products of cholic and chenodeoxycholic acid in man. J, Biol. Chem. 8: Eneroth, P., K. Hellstrom, and J. Sjovall A method for quantitative determination of bile acids in human feces. Acta Chem. Scand. : 9-.. Hill, M. J., and V. C. Aries. 9. Faecal steroid composition and its relationship to cancer of the large boel. J. Pathol. : Drasar, B. S., and M. J. Hill. 9. Human Intestinal Flora. Academic Press, Ne York Hasleood, G. A. D. 96. Comparative Biochemistry. Academic Press, Ne York.. 6. Malagelada, J. R., V. L. W. Go, W. H. J. Summerskill, et al. 9. Bile acid secretion and biliary bile acid composition are altered by cholecystectomy. Amer. J. Dig. Dis. 8: -9.. Hepner, G. W., A. F. Hofmann, J. R. Malagelada, A. P. Szczepanik, and P. D. Klein. 9. Increased bacterial degradation of bile acids in cholecystectomized patients. Gastroenterology. 66: Eneroth, P., and J. Sjovall. 9. Extraction, purification and chromatographic analysis of bile acids in biological materials. In The Bile Acids. P. Nair and D. Kritchevsky, editors. Plenum Press, N.Y Roseleur,. J., and C. M. Van Gent. 96. Alkaline and enzymatic hydrolysis of conjugated bile acids. Clin. Chim. Acta. 66: Norman, A,, and R. H. Palmer. 96. Metabolites of lithocholic--c in human bile and feces. J. Lab. Clin. Med. 6: Evrard, E., and G. Janssen Gas-liquid chromatographic determination of human fecal bile acids. J. Lipid Res. 9: 6-.. Roovers, J., E. Evrard, and H. Vanderhaege An improved method for measuring human blood bile acids. Clin. Chim. Acta. 9: 9-.. Eneroth, P., and J. Sjovall. 9. Extraction, purification and chromatographic analysis of bile acids in biological materials. In The Bile Acids. P. Nair and D. Kritchevsky, editors. Plenum Press, N.Y. 6.. Grundy, S. M., E. H. Ahrens, Jr., and T. A. Miettinen. 96. Quantitative isolation and gas-liquid chromatographic analysis of total fecal bile acids. J. Lipid Res. 6: Soloay, R. D., A. F. Hofmann, P. J. Thomas, J. L. Schoenfield, and P. D. Klein. 9. Triketocholanic (dehydrocholic) acid: hepatic r.letabolism and effect on bile flo and biliary lipid secretion in man. J. Clin. Invest. : Nair, P. P., M. Gordon, and J. Reback. 96. The enzymatic cleavage of the carbon-nitrogen bond in a,a,la-trihydroxy-~-cholan--oxyl glycine. J. Biol. Chem. : -.. Henegouen, G. P., A. Ruben, and K. H. Brandt. 9. Quantitative analysis of bile acids in serum and bile using gas-liquid chromatography. Clin. Chim. Acta. : 9-6. Journal of Lipid Research Volume 9, 98 Notes on Methodology 9