Enzymatic Determination of Bile Acids. The NADP-specific 7a-Hydroxysteroid Dehydrogenase from P. testosteroni (ATCC 11996)
Scand J Gastroenterol Downloaded from informahealthcare.com by Chulalongkorn University on 01/12/15 For personal use only.
BJORN A. SKALHEGG & OLAV FAUSA Dept. of Biochemistry, Nyegaard & Co N S , and Medical Dept. A, National Hospital of Norway, Rikshospitalet, Oslo, Norway
Skilhegg, B. A. & Fausa, 0. Enzymatic determination of bile acids. The NADP-specific 7a-hydroxysteroid dehydrogenase from Pseudomonas testosteroni (ATCC 11996) Scand. J . Gastroent. 1977,12,433-439 By use of the specific substrates, 7a-hydroxy-5t!3-cholanic acid and 3a, 7a, 12a-trihydroxy-3,12-diacetyl-5~-cholanic acid methyl ester, crude extracts of P. testosteroni grown on a steroid-containing medium have been shown to exhibit 7a-hydroxysteroid: NADP-oxidoreductase activity. The enzyme is highly specific for NADP. Both free and conjugated 7a-hydroxy bile acids can act as substrates, but those of low polarity (few hydroxyl groups) seem to be preferred, judging from initial reaction velocity studies. Optimal conditions appears to be at pH 8.5-9.5 and at 25 OC. Free SH-groups are essential for maximum catalytic activity, since the enzyme is inhibited by SHreacting substances such as p-chloromercuribenzoate and monoiodoacetic acid. Also the ketone-trapping agents hydrazine hydrate and semicarbazide act as inhibitors. Upon dilution, the storage stability is severely reduced, but this effect may be counteracted by the addition of glycerol at concentration of 20% or more. By gel filtration experiments on Sephadex G-100, the molecular weight was estimated to about 80,000. Key-words: Bile acids; enzyme assay; 7a-hydroxysteroid dehydrogenase; 3ahydroxysteroid dehydrogenase Bjorn A. Skdlhegg, Dept. of Biochemistry, Nyegaard & Co AIS, Nycovn. 2, Oslo 4, Norway
The enzymatic quantitation of 3a-hydroxy bile acids by the use of the NAD-linked 3a-hydroxysteroid dehydrogenase (E.C.1.1.1.50) (18) from Pseudomonas testosteroni (ATCC 11996) (10, 19) is well documented as a simple, specific and sensitive method (1, 4-6, 8, ll, 13-16). It may also be used in combination with thin-layer chromatography (TLC)* for the quantitation of the individual bile acids
present in various biological samples such as bile and duodenal aspirates (2, 4).Unfortunately, the dihydroxy bile acids, chenodeoxycholic acid, and deoxycholic acid and their corresponding taurine and glycine conjugates cannot be properly separated by any known solvent system on TLC and must therefore be determined together enzymatically or by relatively insensitive colorimetric methods. Indi-
* Abbreviations used in this paper: C=cholic acid, DC =deoxycholic acid, CDC= chenodeoxycholic acid, LiC =lithocholic acid, TC =taurocholic acid, TDC= taurodeoxycholic acid, TCDC =taurochenodeoxycholic acid, TLiC =taurolithocholic acid,
GC =glycocholic acid, GDC =glycodeoxycholic acid, GCDC =glycochenodeoxycholic acid, GLiC =glycolithocholic acid. 3a-hydroxysteroid dehydrogenase =3a-HSD. 7a-hydroxysteroid dehydrogenase=7a-HSD.
4 - Scand. J. Gastroent.
Scand J Gastroenterol Downloaded from informahealthcare.com by Chulalongkorn University on 01/12/15 For personal use only.
434
Bjorn A . Skdlhegg & Olav Fausa
vidual quantitation may certainly be obtained by using gas liquid chromatography (GLC), but this method is not well suited to clinical, routine studies, which have become more relevant during the last few years. One reason has been the promising results in the dissolution of gallstones by oral administration of chenodeoxycholic acid (3). The combined use of the fa-hydroxysteroid dehydrogenase (3aHSD) with a 7a-hydroxysteroid dehydrogenase (7a-HSD) isolated from E.coli has recently been introduced (7, 9). Both enzymes use NAD as cofactor. This approach has therefore been made possible through the fact that the extracts from P.testosteroni seem to contain little or no 7a-HSD while the content of 3a-HSD in the E.coli extract is negligible. The present paper reports evidence of the presence of a new enzyme, a 7a-hydroxysteroid: NADP-oxidoreductase in crude extracts from P.testosteroni grown on a steroid (testosterone) containing medium. The enzyme cannot utilize NAD as cofactor, thus permitting the determination of both 3a- and 7a-hydroxy bile acids by a combined, sequential use of the two cofactors NAD and NADP.
MATERIALS AND METHODS Reagents
The following four bile acids were used as main substrates:
chased from Steraloids. If not otherwise stated, the substrates were dissolved in methanol and used without any further purifications. All other chemicals were of the highest grade commercially available. The water content of the standards was determined by Karl Fisher titration (12). Enzyme preparations
I Acetone powder (1 g) from P.testosteroni grown in our own laboratory (17) was extracted with 9 ml of 0.03M phosphate buffer pH 7.2, which also contained EDTA (1O-j M), dithiothreitol M), and nbutanol (2%). After stirring for 6 h at +4 OC, the suspension was centrifuged at 20,000 x g for 1 h. The clear, yellowish supernatant containing about 19 mg of protein/ml was diluted to give a final concentration of 5 mg/ml. I1 The same procedure as described above was performed with 1 g of dried cells of P.testosteroni purchased from Sigma Chem. Corp. Final concentration of protein was adjusted to 5 mg/ml. I11 Partially purified 3a-HSD from Sigma was dissolved in the same buffer as given above, so that the final concentration of protein was 5 mg/ml. IV To the enzyme kit ‘Sterognost3a’ from Nyegaard & Co. A/S, Oslo, Norway, was added distilled water instead of the recommended 0.1M hydrazine hydrate solution in order to make this 3a-HSD preparation as comparable to I, 11, and 111 as possible.
1) Chenodeoxycholic acid from Canada Packers Ltd. (Toronto, Canada). Purity (from GLC and enzymatic determination) - 98.8%. Water content 0.8%. 2) Cholic acid from Sigma Chem. Corp. (St. Louis, USA). Purity from GLC - 95.5%. Water content The assay system 0.92%. If not otherwise stated, the enzyme activities 3) 7a-hydroxy-5~-cholanicacid (7a-CA). were determined in a 3-ml assay system conKindly synthesized by Prof. Eyssen and Dr. sisting of: 200 pmol sodium pyrophosphate, Parmentier, Leuven, Belgium. Purity: 98 %. 0.5 pmol of NADP (or NAD), and 35 pM of substrate in methanol. The reaction was Water content 1%. 4) 3a, 7a, 12~-trihydroxy-3,12-diacetyl-5/?-cho-started by adding 0.024.1 ml of the enzyme lanic acid methyl ester (3,12-DA) from solution. For the solution IV, however, to 2.9 Steraloids (Pawling, N.Y., USA). Purity ml of the Sterognost-3a solution was added 0.1 approx. 95%. Other bile acids were pur- ml of the substrate. Readings of the absorbance
Enzymatic Determination of Bile Acids
435
Table I. The activities of preparations I, 11,111. and IV towards 7a-hydroxy-5P-cholanicacid (7a-CA) and 3a, 7a, 12a-trihydroxy-3,12-diacetyl-5~-cholanic acid methyl ester (3,12-DA) as substrates. Final concentrations of both substrates in the assay mixture were 35 pM. The initial velocities were measured at 25 "C and pH=9.0 with both NAD and NADP as cofactors (0.5 pM) Initial velocity (units d n - ' ml- l) with NAD
Preparation I
Scand J Gastroenterol Downloaded from informahealthcare.com by Chulalongkorn University on 01/12/15 For personal use only.
11
111 IV Substrate
with NADP
-
-
-
-
-
-
7a-CA
3,12-DA
7a-CA
3,12-DA
-
at 340 nm against the time were made after 1 - 2 - 3 - 4 - 5 - 10 - 30 and 60 minutes at room temperature. From these values the initial velocity was calculated and expressed as units min -1, where 1 Unit is the amount of enzyme that will cause a AE340nmof 0.001 per minute, corresponding to 0.48 nmol of NADPH (NADH) formed per minute. Isolation and identification of reaction products were performed by using thin-layer chromatography on 20 x 20 cm glass plates covered with a 0.5 mm gel layer (Silicagel, Merck Darmstadt, Germany). As solvent system chloroform:ethanol (9:1, v/v) was used. After drying in the air, the plates were sprayed with a 1% anisaldehyd solution in HzSOa:CH3COOH (150, v/v) and then developed at 120 OC for 10 minutes. A description of the preparations and operations of the Sephadex G-100 columns, the determination of protein concentrations and zymography have been given in detail elsewhere (17).
.
RESULTS The four enzyme preparations (I-IV) were tested for 7a-HSD activity using either 7a-CA or 3,12-DA as substrates. Both NAD and NADP were used as cofactors. The results are presented in Table I. All extracts were also tested with LiC as substrate and NADP as cofactor. In preparations I and I1 the maximum optical density at 340 nm after 60 minutes of incubation was 0.015 or less. The formation of the NADPH therefore seems to be due to the
15000 7000
750 400
presence of a NADP-specific 7a-hydroxysteroid dehydrogenase. In order to support this result further, the assay mixture was investigated by TLC in order to detect the eventual reaction product 7-ketolithocholic acid when CDC was the substrate. Two incubating mixtures A1 and A2 contained: 0.2 ml of CDC (7.5 ,umol/ ml of methanol), 0.2 ml of NADP-solution (5 ymol/ml), 1.0 ml of crude bacterial extract (I), and 2.6 ml of 0.1M sodium pyrophosphate buffer pH=9.0. After incubation of A1 for 1 rnin and A2 for 60 rnin at room temperature, 1.0 ml of 1N HCl was added to both of them, followed by 8 ml of ethy1actetate:heptane (l:l, v/v). After shaking, the upper layers were transferred to two flasks. The extraction procedures were repeated, and the combined extracts from A1 and those from A2 were separately evaporated to dryness and redissolved in 0.25 ml of methanol. A solution containing 0.1 ml of 7ketolithocholic acid (7.5 ,umol/ml of methanol) +4.0 ml of buffer was treated in a similar way (solution C). As standards were also run 7-ketolithocholic acid in methanol (solution D) and CDC (solution E). The results from the TLC runs and colour developments are shown in Fig. 1, indicating that little or no 7-ketolithocholic acid is formed from CDC after 1 min of incubation, but significant amounts of the former (the reaction product) are found after 60 rnin of incubation. Substrate specificity The substrate specificities of the enzyme were measured by using different bile acids and
Bjorn A . Skdlhegg & Olav Fausa
436
Al: Complete system incubated 1 rnin A2: 11 ,, 60 mins (8
C : Contrd with 7-krtditocholic a t i d D : Standard 7-ketolitdcholic acid E : Standard CM:
group, both being present on the same steroid molecule as for instance in the CDC and C. The initial velocities were also determined by using 7a-CA and C substrates, the results compared with those obtained with CDC, can be expressed as follows, setting the initial velocity found with 7a-CA as 100%:
Scand J Gastroenterol Downloaded from informahealthcare.com by Chulalongkorn University on 01/12/15 For personal use only.
VVa-CA
:VCDC :VC = 100 : 10 : 1
It therefore seems as if the 7a-HSD enzyme prefers substrates having low polarity (few 0 . . e e 0 hydroxyl groups). In order to assure the measurements of all kinds of 7a-hydroxy bile acids E D A l A 2 C D E substrates, the incubation time was always set Fig. 1. Thin-layer chromatography (TLC) of an to 60 min. extract of the assay mixture consisting of bacterial In order to control the possible influence of extract, CDC, NADP, and buffer. The mixture other steroid dehydrogenase, the crude bacwas extracted with ethy1acetate:heptane (l:l,v/v). After repeated extraction, the combined extracts terial extract was incubated with NADP and were evaporated to dryness, redissolved in meth- the following substrates: Androsterone (3a), anol and applied for TLC using ch1oroforrn:etha- Epiandrosterone (38), Testosterone (178), and no1 (9:1, v/v) as the solvent system. Afterwards of the spots were developed by spraying the plates Ursodeoxycholic acid (78). The presence with 1% anisaldehyd in H2S04:CHjCOOH (150, methanol and ethanol were also checked. In v/v), followed by heating them at 120 OC for 10 all cases, even after prolonged incubation time minutes. (90 min), no formation of NADPH could be detected as determined by increased absorbance their corresponding taurine and glycine con- at 340 nm. jugates. For comparison, all those substrates also having a 3a-hydroxyl group were tested Properties of the enzyme It was found that both hydrazine hydrate with the enzyme kit Sterognost-3a as a control. The results are given in Table 11. It is seen that and semicarbazide inhibited the enzyme to the two enzyme preparations gave comparable some extent, especially when the final concenresults although one system determines the 7a- trations were higher than 0.01M. The optimal hydroxyl group and the other the 3a-hydroxyl condition for the oxidation of the 7a-hydroxyl Table 11. The substrate specificity of the 7a-HSD. The different substrates, dissolved in methanol, were added to the assay mixture to give a final concentration of 35 pM. Readings were made after 60 min of incubation at 25 "C. The theoretical values given are all corrected for the water content of the different compounds as determined by titration after K. Fisher (17). The values obtained with Sterognost-3~: (the 3a-hydroxy bile acids) are also given for comparison Bile acid
EAs4,,after 60 min with
crude 7a-HSD/NADP
AEsdoafter 60 min with
Sterognost-3a
Theoretical values
CDC TCDC GCDC C TC GC 7a-CA 3,12-DA
0.200 0.146 0.140 0.186 0.129 0.161 0.200 0.171
0.196 0.136 0.144 0.209 0.135 0.165 0.005 0.006
0.207 0.145 0.156 0.206 0.139 0.167 0.206 0.180
Scand J Gastroenterol Downloaded from informahealthcare.com by Chulalongkorn University on 01/12/15 For personal use only.
Enzymatic Determination of Bile Acids
437
t (+4'c)
Fig. 2. The effect of storage upon the stability of undiluted (-x-) and 10-fold diluted ( - 0 - ) solutions of crude 'la-hydroxysteroid dehydrogenase. The concentrations of protein were 10.3 mg/ml and 1.03 mg/ml, respectively. The substrate was always CDC present in the assay mixture at a final concentration of 35 pM. The protecting effect of diluting the enzyme preparation in a buffer containing 20% of glycerol is also shown (- o -).
I
1 2 3 4 5 6
group was found to be at pH=8.5-9.5 and the temperature about 25 "C. The reaction velocity was unchanged when the temperature was varied from 20-35 OC.
1 2 3 4 5 6 Storage time (days)
tered. It therefore seems as if free SH-groups are necessary for the catalytic activity of the 7a-HSD. Molecular size
Stability
Several samples of the crude, bacterial extract were tested for activity either undiluted or diluted 10-fold with the extraction buffer. The results after 6 days of storage at +20 OC and at + 4 OC of such samples are shown in Fig. 2, A and B. The results clearly indicate that the enzyme activity is drastically decreased when diluted solutions are stored. When including 20% of glycerol in the enzyme solutions, this effect is effectively counteracted, as indicated from the results also illustrated in Fig. 2. Similar protecting effects could also be seen when the glycerol was exchanged with sucrose and glucose at the same concentration. Inactivation
Both p-chloromercuribenzoate (pCMB) at a M and monoiodofinal concentration of acetic acid (MIA) at a final concentration of 10-3 M inhibited the 7a-HSD activity about 80% (CDC as substrate). Some of the activity could be restored upon the addition of cysteine to the assay system at a final concentration of M. At concentrations of cysteine higher than this, a small, inhibitory effect was regis-
Crude preparations of the 7a-HSD was applied on Sephadex G-100 gel in a thin-layer gel chromatography (TLG) apparatus as described in detail elsewhere (17). Prints of the protein distribution were made by covering the gel layer with a filter paper (Whatman 3MM) for about '/z minute. The wet paper was removed and stained for proteins. A duplicate was developed by zymography (17) using CDC as substrate. The results indicate that the molecular weight of the enzyme is about 80,000. Special care must be taken during such experiments, since gel filtrations always leads to sample dilution with a concomitant loss of activity. The inclusion of 20% of glycerol into all solutions and buffers to be used is therefore recommended. As a final experiment, the possibility of using the 7a-HSD enzyme preparation for detecting the amount of CDC present in a mixture of various amounts of CDC and D C was evaluated. Six mixtures containing unknown concentrations of either CDC, D C or both were prepared. The total concentration of bile acids present was initially determined by use of Sterognost-3a. Samples of each mixture were then added to an assay mixture containing 7a-
438
Bjorn A . Skdlhegg & Olav Fausa
Table 111. Identification of CDC at various concentrations in mixtures with different amounts of DC. Determinations were performed with 7a-HSD and NADP as cofactors at pH=9.0 and 25 "C. Incubation time: 60 min. The concentration of DC was calculated from the difference between values obtained with the Sterognost3a (total bile acids) and that obtained with the 7a-HSD system True amounts, (pg/ml) Solution
CDC
DC
17.5 8.8 26.2 24.5
17.5 26.2 8.8 10.5 17.5
33.6 33.9 34.5
0
0
I
Scand J Gastroenterol Downloaded from informahealthcare.com by Chulalongkorn University on 01/12/15 For personal use only.
Found with Sterognost-3a
I1 I11 IV V VI
0 0
HSD and NADP. The values obtained after 30 minutes of incubation were taken as a measure of the CDC concentration in the mixture. The concentration of DC was then calculated by subtracting the '7a-HSD value' from that obtained with the Sterognost3a. The results are shown in Table 111.
DISCUSSION By use of the specific substrate 7a-hydroxy-5Pcholanic acid (7a-CA), crude extracts of P.testosteroni grown on a steroid-containing medium have been shown to exhibit 7a-hydroxysteroid dehydrogenase activity, provided that the cofactor is NADP. Exchanging NADP with NAD completely abolished the activity found. This fact makes it possible to use the enzyme for quantitations of 7a-hydroxy bile acids even in the presence of 3a-hydroxy bile acids, since the 3a-HSD known to be present in such extracts (10, 18, 19) is strictly NAD-specific. Measurements of the initial reaction rates by using substrates of different polarity such as C, CDC and 7a-CA indicate that there is a preference for substrates of low polarity (few hydroxyl groups). Furthermore, when the enzyme is used for determinations of bile acids in different biological fluids, a sufficiently long incubation time must be used. In general, it is considered to be 60 min at the optimal conditions, here found to be pH=8.5-9.5 and a temperature of 25 OC. The present results also indicate that the SH-
33.4
15.9
CDC found with the 7a-HSD. pg/ml (% recovery)
@C/ml) calculated ( % =oven9
16.5 (94.3) 9.3 (105.6) 26.9 (102.4) 22.3 (91.0)
17.1 (97.7) 24.6 (93.9) 7.5 (86.3) 11.1 (105.7) 15.9 (100)
0 0
0
reacting substances should be avoided. These results suggest the inclusion of EDTA in the assay mixture in order to chelate heavy metal ions. When the enzyme preparations are being diluted, this should always be performed with solvents containing 20% (or more) of glycerol. Otherwise the storage stability is severely reduced. Experiments are now being performed in order to evaluate the possible practical significance of the 7a-HSD.
REFERENCES l.Beeke, R., De Werdt, G,A., Pares, I. & Barbier, P. Clin. Chim. Acta 1976, 71, 27-29 2.Bruusgaard, A. Clin. Chirn. Acta 1970, 28, 495-504 3. Danziger, R. G., Hofmann, A. F., Schoenfield, L. J. & Thistle, J. L. New Engl. J . Med. 1972, 286,1-8 4. Fausa, 0. & SkBlhegg, B. A. Scand. I . Gastroent. 1974,9,249-254 5.Fausa, 0. Scand. 1. Gastroent. 1975, 10, 747752 6.Fausa, 0. Scand. J. Gastroent. 1976, 11, 229232 7.Haslewood, G. A. D., Murphy, G. M. & Richardson, J. M. Clin. Sci. 1973, 44, 95-98 8. Iwata, T.& Yamasaki, K.1. Biochem. (Tokyo) 1964,56,424-431 9. MacDonald, I. A., Williams, C.N. & Mahony, D. E. J. Lipid. Res. 1975, 16, 244-246 lO.Marcus, P. I. & Talalay, P. J . Biol. Chem. 1956,218,661671
Enzymatic Determination of Bile Acids
Scand J Gastroenterol Downloaded from informahealthcare.com by Chulalongkorn University on 01/12/15 For personal use only.
11. Mashige, F., Imai, K. & Osuga, T. Clin.Chim. Acta 1976,70,79-86 12. Mitchell, I. Analyt. Chem. 1951, 23, 1069 13.Murphy, G. M., Billing, B. & Baron, D. M. 1. Clin.Path. 1970, 23, 594-598 14. Palmer, R. H., Meth. Enzymol. 1969, 15, 280288 KSchwarz, H. P., Bergmann, K. V. & Paumgartner, G. Clin. Chim. Acta 1974, 50, 197-206 Received 10 March 1977 Accepted 2 April 1977
439
16. Sheltaway, M. J. & Losowsky, M. J. Clin. Chim. Acta 1975,64,127-133 17.SkPlhegg, B. A. Europ. J. Biochem. 1974, 46, 117-125 18.Talalay, P.,Dobson, M. M. & Tapley, D. F. Nature (Lond.) 1952, 170, 620-621 19.Talalay, P. & Marcus, P. I. J. Biol. Chem. 1956,218,675-691