(Received for publication, January 29, 1997, and in revised form, May 2, 1997)
From the Institute for Enzyme Research, University of
Tokushima, Tokushima 770, Japan and the § Department of
Molecular and Cell Biology, University of California,
Berkeley, California 94720-3200
H-protein of the glycine cleavage system has a lipoic acid prosthetic group. Selenolipoic acid is a lipoic acid analog in which both sulfur atoms are replaced by selenium atoms. Two isoforms of bovine lipoyltransferase that are responsible for the attachment of lipoic acid to H-protein had an affinity for selenolipoyl-AMP and transferred the selenolipoyl moiety to bovine apoH-protein comparable to lipoyl-AMP. Selenolipoylated H-protein was overexpressed in Escherichia coli and purified. Selenolipoylated H-protein was 26% as effective as lipoylated H-protein in the glycine decarboxylation reaction, in which reduction of the diselenide bond of selenolipoylated H-protein is catalyzed by P-protein. The diselenide form of selenolipoylated H-protein was a poor substrate for L-protein, and the rate of reduction was 0.5% of that of lipoylated H-protein. The rate of the overall glycine cleavage reaction with selenolipoylated H-protein was <1% of that with lipoylated H-protein. These results are consistent with the difference in the redox potential between the diselenide and disulfide bonds. In contrast, selenolipoylated H-protein showed three times as high glycine-14CO2 exchange activity as lipoylated H-protein, presumably because the rate of reoxidation of reduced selenolipoylated H-protein is much higher than that of lipoylated H-protein.
The glycine cleavage system is a multienzyme complex that catalyzes the oxidative cleavage of glycine as presented in Equation 1 (1-4),
![]() |
(Eq. 1) |
![]() |
(Eq. 2) |
![]() |
(Eq. 3) |
![]() |
(Eq. 4) |
Lipoic acid attaches to the specific lysine residue of H-protein via an
amide linkage. Attachment of lipoic acid to the protein involves two
consecutive reactions, the activation of lipoic acid to
lipoyl-AMP and the transfer of the lipoyl group to the
apoprotein. In mammals, the two reactions are catalyzed by
separate enzymes in mitochondria: lipoate-activating enzyme catalyzes
the former reaction, and
lipoyl-AMP:N-lysine lipoyltransferase
(lipoyltransferase) catalyzes the latter (7, 8). We have purified two
isoforms of lipoyltransferase, lipoyltransferases I and II, from bovine
liver. They transfer not only the lipoyl group, but also
C6, C8, and C10 acyl groups from
each activated adenylated form to apoH-protein and also catalyze lipoylation of lipoyl domains of the acyltransferase components of the
pyruvate,
-ketoglutarate, and branched chain
-ketoacid dehydrogenase complexes (8, 9).
Selenolipoic acid is a lipoic acid analog in which both of the sulfur
atoms are replaced by selenium atoms. Studies with wild-type Escherichia coli have shown that selenolipoic acid inhibits
growth of the bacteria through formation of selenolipoylated
-ketoacid dehydrogenase complexes that are nonfunctional (10). In
mammals, the biological properties of selenolipoic acid have not been
elucidated so far. In this study, we show that selenolipoic acid can
serve as a substrate for bovine lipoyltransferase. Additionally, we overexpressed selenolipoylated recombinant bovine H-protein in E. coli and characterized the function of selenolipoylated H-protein in the glycine cleavage reaction.
Selenolipoic acid was generously provided by ASTA-Medica (Frankfurt am Main, Germany). [1-14C]Glycine and [2-14C]glycine were purchased from NEN Life Science Products. Porcine dihydrolipoamide dehydrogenase (L-protein) was obtained from Sigma. NAD+ and NADH were from Boehringer Mannheim. Dimedone was from Nacalai Tesque (Kyoto, Japan). Scintilamine-OH was from Wako Pure Chemicals (Osaka, Japan). Lipoyl-AMP and selenolipoyl-AMP were prepared from the mixture of R- and S-forms of lipoic acid and selenolipoic acid, respectively, as described (11), and their concentrations were determined spectrophotometrically with a molar extinction coefficient of 15.4 × 103 at 259 nm (11). Tetrahydrofolate was prepared as described (3). Lipoyltransferases were purified, and their activities were assayed as described previously (8). One unit of lipoyltransferase activity is defined as 1 nmol of H-protein lipoylated per min. E. coli P-protein (12) and T-protein (13) were purified as described previously.
In Vitro Selenolipoylation and Lipoylation of H-proteinThe
selenolipoylation or lipoylation reaction was carried out in a mixture
of 150 µl containing 3 µg of apoH-protein, 40 mM potassium phosphate buffer, pH 7.8, 0.2 mg/ml
BSA,1 10 µM
MnCl2, 30 µM selenolipoyl-AMP or lipoyl-AMP,
and 1.25 × 102 units of lipoyltransferase II. After
incubation at 37 °C for 60 min, 20 µl of the reaction mixture was
removed, and dithiothreitol was added to a final concentration of 0.3 mM. The mixture was heated in a boiling water bath for 1 min and centrifuged. Ten µl of the supernatant was mixed with 5 µl
of 3 × sample buffer (9) and subjected to 20% nondenaturing
polyacrylamide gel electrophoresis and silver-stained. HoloH-protein
migrates faster than apoH-protein since the apoprotein has an
additional positive charge on the unmodified lysine residue (12). The
remaining reaction products were heated without dithiothreitol in a
boiling water bath for 1 min, and H-protein activity was assayed by the
glycine-14CO2 exchange reaction. In the
steady-state kinetic studies, the selenolipoylation and lipoylation
reactions were carried out for 30 min with 1.67 × 10
4 units of either lipoyltransferase I or II as
described previously (8), except that 0.5 mg/ml BSA was added to the
reaction mixture instead of 0.2 mg/ml. Apparent Km
values were determined by varying the concentration of selenolipoyl-AMP
or lipoyl-AMP from 4.16 to 50 µM at an apoH-protein
concentration of 2 µM or by varying the concentration of
apoH-protein from 0.167 to 2 µM at a selenolipoyl-AMP or
lipoyl-AMP concentration of 75 µM. HoloH-protein formed
was determined by the glycine-14CO2 exchange
reaction.
H-protein was expressed in E. coli
BL21(DE3)pLysS carrying the expression plasmid pET-3a containing
cDNA for mature bovine H-protein essentially as described
previously (12) in culture medium containing 40 µM
selenolipoic acid or lipoic acid. Incubation was carried out at
37 °C, and isopropyl--D-thiogalactopyranoside was
added to the medium to a final concentration of 0.4 mM when the A600 of the culture reached 0.8. Cells were
grown further for 3 h and harvested by centrifugation.
Lipoylated and selenolipoylated recombinant H-proteins were purified by the procedures described previously (12) with minor modifications. After DEAE-Sepharose column chromatography, H-protein was further purified by high performance liquid chromatography on a reversed-phase C4 column (4.6 × 250 mm; RP-304, Bio-Rad). The column was developed with an acetonitrile gradient in 0.1% trifluoroacetic acid at room temperature with monitoring at 220 nm. The peak fraction containing selenolipoylated or lipoylated H-protein was pooled, concentrated by lyophilization, and neutralized to pH 7.2 with 1 M Tris solution. The activity of H-protein was determined by the glycine-14CO2 exchange reaction.
Assay of H-proteinThe activity of H-protein was determined
by the following five different ways. 1) The
glycine-14CO2 exchange reaction (sum of the
forward and reverse reactions of Equation 2) was determined with 1 µg
of recombinant H-protein and 7.2 µg of E. coli P-protein
as described previously (8). 2) The glycine decarboxylation reaction
(forward reaction of Equation 2) was carried out in a 0.2-ml reaction
mixture containing 5 mM [1-14C]glycine (2 Ci/mol), 50 mM potassium phosphate buffer, pH 7.0, 1 mM dithiothreitol, 0.1 mM pyridoxal phosphate,
14.4 ng of E. coli P-protein, and 15 µg of recombinant
H-protein in a Warburg-type flask containing 80 µl of Scintilamine-OH
in a center well. The reaction was initiated by the addition of
glycine, carried out for 30 min at 37 °C, and terminated by the
addition of 0.1 ml of 15% trichloroacetic acid from a side arm of the
flask. The mixture was incubated further for 45 min at 37 °C to
facilitate the absorption of 14CO2 into
Scintilamine-OH, and the radioactivity was determined in a liquid
scintillation spectrometer. 3) The overall glycine cleavage (Equations
2-4) was determined in a 50-µl reaction mixture containing 5 mM glycine, 50 mM potassium phosphate buffer,
pH 7.2, 2 mM dithiothreitol, 0.1 mM pyridoxal
phosphate, 0.2 mM tetrahydrofolate, 1 mM
NAD+, 4 pmol of E. coli P-protein, 54 pmol of
recombinant H-protein, 18 pmol of E. coli T-protein, and 2 pmol of porcine L-protein. The ratio of the constituent proteins
employed was that found in the plant system (14). The reaction was
initiated by the addition of glycine, and the rate of NADH production
was measured spectrophotometrically at 340 nm at room temperature. The
reaction without H-protein was employed as a control. 4) The
"C1" formed from glycine was determined in the reaction
mixture for the overall reaction described above with 5 mM
[2-14C]glycine (1 Ci/mol). The reaction was carried out
for 5 min at room temperature and terminated by the addition of 0.1 ml
of 15% trichloroacetic acid and 5 µl of 5 mg/ml BSA. The
C1 formed was isolated as a dimedone adduct, and the
radioactivity was determined as described previously (15). 5) Reduction
of 5,5-dithiobis(2-nitrobenzoic acid) (reverse of Equation 4) was
determined by the method of Ellman (16) in a 0.1-ml reaction mixture
containing 8.89 µg of 5,5
-dithiobis(2-nitrobenzoic acid), 50 mM potassium phosphate buffer, pH 7.4, 10 mM
EDTA, 0.75 mM NADH, 0.5 µg of porcine L-protein, and 2 µg of recombinant H-protein. The reaction was started by the addition
of 20 µl of 5,5
-dithiobis(2-nitrobenzoic acid) solution (1 mg/2.25
ml of 0.5 M Tris-Cl, pH 8.0), and the increase in
absorbance at 412 nm was monitored at room temperature.
The molecular masses of apoH-protein and lipoylated and selenolipoylated H-proteins were analyzed with matrix-assisted laser desorption ionization/time-of-flight mass spectrometry using a PerSeptive Biosystems Voyager RP mass spectrometer. One µl of the sample containing ~70 pmol of each H-protein in 40% acetonitrile containing 0.1% trifluoroacetic acid was mixed with 1 µl of matrix (10 mg of sinapinic acid/ml of 33% acetonitrile containing 0.1% trifluoroacetic acid) on a sample plate. The mass spectrometer was operated with a linear mode at an ion-accelerating voltage of 30 kV. Molecular masses were determined using bovine insulin as an external standard mass (5733.5 + 1 Da).
Other MethodsProtein was determined by the method of Lowry et al. (17) with BSA as a standard. Nondenaturing polyacrylamide gel electrophoresis was carried out as described (12).
To determine whether
lipoyltransferases can recognize selenolipoyl-AMP as a substrate,
in vitro selenolipoylation of bovine apoH-protein was
conducted. As shown in Fig.
1A, lipoyltransferase II
transferred the selenolipoyl moiety from selenolipoyl-AMP to apoH-protein as efficiently as the lipoyl moiety, and selenolipoylated H-protein comigrated with lipoylated H-protein on nondenaturing polyacrylamide gel electrophoresis. The activity of the resultant selenolipoylated H-protein in the mixture of the lipoylation reaction was determined by the glycine-14CO2 exchange
assay. Since the redox potential of the diselenide of selenolipoic acid
(478 mV (10)) is lower than that of the disulfide of lipoic acid
(
325 mV (18)), it was expected that selenolipoylated H-protein would
not be a good substrate for the glycine-14CO2
exchange reaction that accompanies the reductive cleavage of the
diselenide bond by P-protein. Unexpectedly, selenolipoylated H-protein
exhibited about three times higher
glycine-14CO2 exchange activity than lipoylated
H-protein (Fig. 1B). Essentially the same results were
obtained with lipoyltransferase I (data not shown).
The steady-state kinetic studies were carried out in the presence of 0.5 mg/ml BSA as described under "Experimental Procedures." As reported in a previous paper (8), lipoyl-AMP inhibited the reaction at a concentration over 50 µM. The addition of 0.2 mg/ml BSA prevented the inhibition. Similarly, selenolipoyl-AMP inhibited the reaction at a concentration over 50 µM, but the presence of 0.2 mg/ml BSA was not sufficient. In the presence of 0.5 mg/ml BSA, however, the inhibition was not observed, and the reaction proceeded hyperbolically up to at least 100 µM selenolipoyl-AMP (data not shown). It has been reported that lipoic acid binds to BSA with a molar ratio of ~10 mol/mol of BSA (19). As discussed earlier, BSA may interact with lipoyl-AMP to hinder the binding of lipoyl-AMP to a site other than the active site of lipoyltransferase (8). Similarly, BSA may interact with selenolipoyl-AMP and prevent the substrate inhibition, although the affinity of BSA for selenolipoyl-AMP seems to be lower than that for lipoyl-AMP. The apparent Km values for selenolipoyl-AMP and apoH-protein and the Vmax values of selenolipoylation were all similar to those obtained with lipoyl-AMP (Table I). The results indicate that selenolipoyl-AMP is an efficient substrate for lipoyltransferase.
|
To characterize selenolipoylated H-protein in more detail, we overexpressed selenolipoylated bovine H-protein in E. coli and purified it to apparent homogeneity (data not shown). E. coli cells could grow in rich medium containing 40 µM selenolipoic acid as well as in medium containing 40 µM lipoic acid. Purified selenolipoylated H-protein exhibited about three times as high glycine-14CO2 exchange activity as lipoylated H-protein (Table II). The molecular masses of selenolipoylated and lipoylated H-proteins and apoH-protein were 14,143.2 Da (expected value, 14,121 + 1 Da), 14,042.7 Da (expected value, 14,025 + 1 Da), and 13,844.6 Da (expected value, 13,837 + 1 Da), respectively. These results confirm that H-protein purified from cells cultured with selenolipoic acid contains the selenolipoyl moiety instead of the lipoyl moiety. Interestingly, endogenous E. coli H-protein from cells cultured in the presence of selenolipoic acid also incorporates the selenolipoyl moiety since it exhibits ~3-fold higher glycine-14CO2 exchange activity than that from cells cultured with lipoic acid (data not shown). These results indicate that lipoate-protein ligase of E. coli can utilize selenolipoate as a substrate.
|
The specific activity of lipoylated recombinant H-protein in the glycine-14CO2 exchange reaction was comparable to that of native H-protein purified from bovine liver (30.8 nmol/30 min/µg of H-protein (8)). In contrast, in vitro lipoylated H-protein showed a specific activity of about half that of native H-protein (Fig. 1B). These results suggest that lipoyltransferase transfers both R- and S-enantiomers of the lipoyl moiety from lipoyl-AMP to apoH-protein in the in vitro reaction and that only one isomer, probably an R-form, is active in the exchange reaction. The results also suggest that lipoate-protein ligase of E. coli transfers preferentially the active isomer to apoH-protein in vivo and that the resultant lipoylated H-protein is fully active.
To study the function of selenolipoylated H-protein in the glycine cleavage reaction, we examined the various activities of selenolipoylated H-protein. When selenolipoylated H-protein was used, the rate of glycine decarboxylation was 26% of that with lipoylated H-protein (Table II), indicating that the glycine decarboxylation that accompanies the simultaneous cleavage of the diselenide bond by P-protein is more resistant than the cleavage of the disulfide. The overall glycine cleavage reaction was determined with the reconstructed system in which the ratio of the amount of each component protein is that found in the plant system (14) since the stable glycine cleavage complexes of mammalian and E. coli origin have not been isolated. The overall cleavage rate with selenolipoylated H-protein was <1% of that with lipoylated H-protein when determined by the formation of NADH (Table II). In contrast, C1 formation with selenolipoylated H-protein measured under the same reaction condition was 15% of that with lipoylated H-protein (Table II), and selenolipoylated and lipoylated H-proteins turned over 2 and 14 times, respectively, during the reaction for 5 min. The formation of C1 occurred even in the absence of T-protein and L-protein, although the cleavage with lipoylated H-protein absolutely requires four protein components (data not shown). These results suggest that the intermediate attached to the selenolipoyl moiety is cleaved spontaneously to C1 and ammonia concomitant with the formation of the diselenide bond of the selenolipoyl group that will be recycled. The reduction of the diselenide of the selenolipoyl moiety by L-protein was much more resistant than that of the disulfide (0.5% of the reduction of the disulfide (Table II)), suggesting that reoxidation of the diselenol group by L-protein (Equation 4) may be much easier than that of the dihydrolipoyl group and that the glycine synthesis (reverse reaction of Equation 1) may not occur because of the difficulty of reduction of the diselenide bond by L-protein. The above results are consistent with the difference in the redox potential between the diselenide and disulfide. In contrast, the rate of the glycine-14CO2 exchange reaction with selenolipoylated H-protein is higher than that with lipoylated H-protein. The exchange reaction is a sum of the forward and reverse reactions of Equation 2 and accompanies reduction and reoxidation of the diselenide. Since the rate of the forward decarboxylation is 26% of that with lipoylated H-protein, we suppose that the rate of the reverse reaction must be >10-fold faster with selenolipoylated H-protein than with lipoylated H-protein, and as a consequence, the rate of the glycine-14CO2 exchange reaction becomes 3-fold faster. There are many reports about the function of the selenol group of the selenocysteine residue in proteins (20). At physiological pH, selenol is largely ionized and thought to facilitate the catalytic function of the enzymes. In the reverse reaction of the glycine-14CO2 exchange, the remaining selenol group of the selenolipoyl moiety may be ionized and accelerate the rate of the reverse reoxidation reaction.
ConclusionWe have shown, for the first time, the biochemical properties of selenolipoic acid in the reactions catalyzed by lipoyltransferase and the glycine cleavage system. The lipoyltransferases recognized selenolipoyl-AMP and transferred the selenolipoyl group to apoH-protein as efficiently as the lipoyl group. The Km values for lipoyl-AMP and selenolipoyl-AMP were quite similar, indicating that selenolipoic acid is a less effective inhibitor of lipoyltransferase especially when sufficient lipoic acid is present. In the glycine cleavage reaction, as expected from the redox potential, the diselenide bond was resistant to the reduction by P-protein and L-protein. The glycine cleavage system containing selenolipoylated H-protein hardly functions physiologically because of the spontaneous cleavage of the intermediate attached to the selenolipoyl moiety and reoxidation of the selenol group. These results may provide a clue for the investigation of the function of the selenol group in selenoproteins.
We thank Dr. Shiro Futaki (Faculty of Pharmaceutical Sciences, University of Tokushima) for help in the molecular mass determinations and ASTA-Medica for the generous gift of selenolipoic acid.
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
All ASBMB Journals | Molecular and Cellular Proteomics |
Journal of Lipid Research | Biochemistry and Molecular Biology Education |