Abstract

Glutaric aciduria type 1 (GA1) is an autosomal recessive neurometabolic disorder caused by mutations in the glutaryl-CoA dehydrogenase gene (GCDH), leading to an accumulation and high excretion of glutaric acid and 3-hydroxyglutaric acid. Considerable variation in severity of the clinical phenotype is observed with no correlation to the genotype. We report here for the first time on expression studies of four missense mutations c.412A > G (p.Arg138Gly), c.787A > G (p.Met263Val), c.1204C > T (p.Arg402Trp) and c.1240G > A (p.Glu414Lys) identified in GA1 patients in mammalian cells. Biochemical analyses revealed that all mutants were enzymatically inactive with the exception of p.Met263Val which showed 10% activity of the expressed wild-type enzyme. Western blot and pulse-chase analyses demonstrated that the amount of expressed p.Arg402Trp protein was significantly reduced compared with cells expressing wild-type protein which was due to rapid intramitochondrial degradation. Upon cross-linkage the formation of homotetrameric GCDH was strongly impaired in p.Met263Val and p.Arg402Trp mutants. In addition, GCDH appears to interact with distinct heterologous polypeptides to form novel 97, 130 and 200 kDa GCDH complexes. Molecular modeling of mutant GCDH suggests that Met263 at the surface of the GCDH protein might be part of the contact interface to interacting proteins. These results indicate that reduced intramitochondrial stability as well as the impaired formation of homo- and heteromeric GCDH complexes can underlie GA1.

INTRODUCTION

Glutaric aciduria type 1 (GA1, MIM# 231670) is caused by the inherited deficiency of the mitochondrial matrix protein glutaryl-CoA dehydrogenase (GCDH, MIM# 608801, E.C. 1.3.99.7), which catalyzes the oxidative decarboxylation of glutaryl-CoA, an intermediate in the degradation of the amino acids lysine and tryptophan. The enzymatic reaction produces crotonyl-CoA and CO2 and requires flavin adenine dinucleotide (FAD) as coenzyme (1). Defective GCDH leads to formation and accumulation of the metabolites glutaric acid and 3-hydroxyglutaric acid in tissues and body fluids of affected patients. During catabolic states, GA1 patients are prone to the development of encephalopathic crises, which are characterized by destruction of striatal neurons and a subsequent irreversible dystonic-dyskinetic movement disorder. The risk for the development of encephalopathic crises is highest during a time window of vulnerability between 6 and 36 months of age and can be reduced by the administration of a lysine-restricted diet, the supplementation of carnitine and an intensive emergency management. Newborn screening techniques enabled the early identification of affected patients, and allow the implementation of therapy prior to the development of encephalopathic crises. The clinical course of GA1 shows a high degree of variability, and some patients remain asymptomatic (2–5).

The GCDH gene is localized on human chromosome 19p13.2, spans ∼7 kb and is composed of 11 exons and 10 introns (6). The encoded protein comprises 438 amino acids. After import into mitochondria 44 N-terminal amino acid residues are cleaved off (7). The active GCDH enzyme represents a homotetramer (2). More than 150 disease-causing mutations in the GCDH gene have been described so far in GA1 patients (8–11). However, certain mutations show predominance in specific populations, e. g. with a high incidence of the p.Ala421Val allele or the intronic mutation IVS1 + 5G > T in the old-order Amish in PA, USA and the Island Lake Ojibway Indians in Canada, respectively (6,12). The most common Caucasian mutation identified is p.Arg402Trp (8,10). The majority of GCDH mutations are associated with an enzymatic activity in patient fibroblasts of 0–5% of controls (10,13). For others such as p.Val400Met and p.Arg227Pro occurring in compound heterozygosity or the homozygous mutation p.Met263Val, GCDH activities as high as 30% of controls have been reported (10,13–15). About 15 of these mutations identified have been characterized for enzymatic activity and capability to oligomerize after expression in Escherichia coli (E. coli) (6,7,16,17).

In this report, we have investigated the effects of the mutations p.Arg138Gly, p.Met263Val, p.Arg402Trp and p.Glu414Lys on biochemical parameters of GCDH in mammalian cells. The mutations were introduced into human GCDH cDNA, expressed in cells, and tested for synthesis, cellular stability and capability of mutant GCDH to oligomerize. The data were completed by three-dimensional (3D) modeling analyses. The results show that loss or reduction of enzymatic activity can be explained by mutations in amino acids involved in the catalytic reaction, reduced stability in mitochondria, and by impaired homo- and heterologous protein interaction.

RESULTS

Expression of wild-type and mutant GCDH in hamster cells

To investigate the biological significance, mutations were introduced into the wild-type GCDH cDNA and transiently expressed in baby hamster kidney (BHK) cells. Non-transfected BHK cells were used as controls. The expression of the wild-type enzyme resulted in a 60-fold increase in GCDH activity above levels in non-transfected cells (Table 1, 363 ± 12 versus 6.1 µmol/h × g protein). The endogenous GCDH activity in BHK cells was not affected by transfection with vector alone (data not shown). With the exception of the p.Met263Val mutant which had 10% of wild-type GCDH activity, all other mutants, p.Arg138Gly, p.Arg402Trp and p.Glu414Lys, were enzymatically inactive (Table 1).

Table 1.

GCDH enzyme activity of wild-type- and mutant GCDH-transfected BHK cells

Transfected GCDH variant GCDH activity (% of wild-type) 
Non-transfected 1.7 
Wild-type 100a 
p.Arg138Gly 1.7 
p.Met263Val 10.4 
p.Arg402Trp 1.6 
p.Glu414Lys 1.7 
Transfected GCDH variant GCDH activity (% of wild-type) 
Non-transfected 1.7 
Wild-type 100a 
p.Arg138Gly 1.7 
p.Met263Val 10.4 
p.Arg402Trp 1.6 
p.Glu414Lys 1.7 

aGCDH activity of cell lysates of wild-type GCDH-transfected BHK cells was 363 ± 12 µmol/h × mg protein.

Western blot analysis showed that both wild-type and mutant GCDH were expressed in transfected cells as a single immunoreactive 43 kDa band (Fig. 1). In non-transfected cells, background levels of immunoreactive polypeptide bands (5 ± 8% of wild-type GCDH) were detectable. The steady state concentrations of p.Arg138Gly, p.Met263Val and especially p.Arg402Trp were significantly lower (81 ± 13, 68 ± 16 and 14 ± 13% of wild-type, respectively; P < 0.05 in five independent experiments) than those in cells expressing wild-type or mutant p.Glu414Lys GCDH. Similar amounts of co-expressed green fluorescent protein in BHK cells indicate that differences in transfection efficiencies are rather unlikely to account for the different steady state expression levels of wild-type and mutant GCDH protein (data not shown).

Figure 1.

Expression of wild-type and mutant GCDH protein. BHK cells were transfected with wild-type or mutant GCDH cDNAs, protein extracts were prepared and aliquots containing equal amounts of protein were analyzed by SDS–PAGE (10% acrylamide) and GCDH western blotting. Results were compared with non-transfected cells. The immunoreactive polypeptides were quantified by densitometric scanning. The signal intensity of wild-type GCDH was set to 100%. Data were normalized to content of MnSOD, a mitochondrial marker protein. Of note, the signal intensity of the p.Arg402Trp protein was only 12% of the wild-type control, indicating either reduced synthesis or increased degradation. The figure shows one representative experiment and its densitometric evaluation, average values of five independent experiments are given in the text.

Figure 1.

Expression of wild-type and mutant GCDH protein. BHK cells were transfected with wild-type or mutant GCDH cDNAs, protein extracts were prepared and aliquots containing equal amounts of protein were analyzed by SDS–PAGE (10% acrylamide) and GCDH western blotting. Results were compared with non-transfected cells. The immunoreactive polypeptides were quantified by densitometric scanning. The signal intensity of wild-type GCDH was set to 100%. Data were normalized to content of MnSOD, a mitochondrial marker protein. Of note, the signal intensity of the p.Arg402Trp protein was only 12% of the wild-type control, indicating either reduced synthesis or increased degradation. The figure shows one representative experiment and its densitometric evaluation, average values of five independent experiments are given in the text.

Stability of mutant GCDH protein

To evaluate whether the reduced amount of mutant GCDH protein, particularly the p.Arg402Trp mutant, was caused by decreased synthesis or by increased degradation, wild-type and mutant-transfected BHK cells were metabolically labeled with (35S)-methionine for 2 h and either directly harvested or chased for 24 h, followed by immunoprecipitation of labeled GCDH protein, sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE) and fluorography. After 2 h of metabolic labeling, in lysates of wild-type and mutant GCDH expressing cells two polypeptides of 48 and 43 kDa were immunoprecipitated, representing the precursor and mature GCDH protein, respectively (Fig. 2). The intensities of these precursor and mutant GCDH forms were comparable indicating similar rates of synthesis. After 24 h of chase, GCDH was detected in wild-type, p.Arg138Gly, p.Met263Val and p.Glu414Lys cell extracts as a mature 43 kDa form. The amounts of (35S)-labeled mutant p.Arg138Gly, p.Met263Val, and p.Glu414Lys proteins were 81, 55 and 108%, respectively, of wild-type protein. Mutant p.Arg402Trp protein was no longer detectable in cellular extracts suggesting an increased degradation. In order to determine the half-life of the mutant GCDH Arg402Trp more precisely, metabolically labeled cells were subjected to chase periods of 2, 4, 8 and 24 h, followed by immunoprecipitation (Fig. 3). A rapid disappearance of the mutant p.Arg402Trp polypeptide was observed allowing the estimation of a half-life of 4 h, whereas the wild-type protein showed an approximate half-life span as high as 34.5 h.

Figure 2.

Pulse-chase labeling of wild-type and mutant GCDH. BHK cells expressing wild-type or the indicated mutant GCDH were labeled for 2 h with (35S)-methionine (75 µCi/ml). Afterwards, cells were either harvested or chased for 24 h in the presence of unlabeled methionine. GCDH was immunoprecipitated from cell extracts and analyzed by SDS–PAGE and fluorography. The positions of the precursor and mature GCDH forms are indicated. GCDH-specific signal intensities were quantified by densitometric scanning, the wild-type signal without chase was set to 100%. Note absence of p.Arg402Trp protein after 24 h chase indicating increased degradation of the protein.

Figure 2.

Pulse-chase labeling of wild-type and mutant GCDH. BHK cells expressing wild-type or the indicated mutant GCDH were labeled for 2 h with (35S)-methionine (75 µCi/ml). Afterwards, cells were either harvested or chased for 24 h in the presence of unlabeled methionine. GCDH was immunoprecipitated from cell extracts and analyzed by SDS–PAGE and fluorography. The positions of the precursor and mature GCDH forms are indicated. GCDH-specific signal intensities were quantified by densitometric scanning, the wild-type signal without chase was set to 100%. Note absence of p.Arg402Trp protein after 24 h chase indicating increased degradation of the protein.

Figure 3.

Determination of half-life of the mutant p.Arg402Trp protein. Cells overexpressing wild-type and p.Arg402Trp mutant GCDH were pulse-labeled with (35S)-methionine and chased for the indicated time periods as described in the legend of Figure 2. The intensities of GCDH polypeptides visualized by fluorography were evaluated by densitometry and related to the signal intensities directly after the pulse.

Figure 3.

Determination of half-life of the mutant p.Arg402Trp protein. Cells overexpressing wild-type and p.Arg402Trp mutant GCDH were pulse-labeled with (35S)-methionine and chased for the indicated time periods as described in the legend of Figure 2. The intensities of GCDH polypeptides visualized by fluorography were evaluated by densitometry and related to the signal intensities directly after the pulse.

Intracellular localization of mutant GCDH Arg402Trp

To examine whether the degradation of mutant p.Arg402Trp protein occurs before or after import of the protein into mitochondria, its intracellular localization was compared with wild-type GCDH by double immunofluorescence microscopy using co-staining for myc-tagged GCDH and the mitochondrial marker manganese-dependent superoxide dismutase (MnSOD) (Fig. 4). Wild-type GCDH was found to be completely co-localized with MnSOD, indicating a localization of wild-type GCDH within mitochondria. The p.Arg402Trp mutant showed partial co-localization with MnSOD indicating that mutant p.Arg402Trp protein is imported into mitochondria followed by rapid intramitochondrial degradation. The capability of the mutant protein to be imported into mitochondria is in agreement with the finding of proteolytic processing of the mutant precursor GCDH to the mature form (Figs 2 and 3).

Figure 4.

Intracellular localization of wild-type and p.Arg402Trp GCDH. At 36 h after transfection, BHK cells were stained for GCDH-myc (green) and the mitochondrial marker MnSOD (red). Yellow indicates a co-localization of GCDH with the marker.

Figure 4.

Intracellular localization of wild-type and p.Arg402Trp GCDH. At 36 h after transfection, BHK cells were stained for GCDH-myc (green) and the mitochondrial marker MnSOD (red). Yellow indicates a co-localization of GCDH with the marker.

Oligomerization of wild-type and mutant GCDH

Human GCDH functions as a homotetramer (2). The capability of mutant GCDH to form oligomeric structures was examined by cross-linkage experiments. Extracts of BHK cells overexpressing wild-type or mutant GCDH were incubated in the presence and absence of the non-cleavable cross-linker bis(sulfosuccinimidyl)suberate (BS3). Subsequent immunoblot and densitometric analyses showed that after cross-linkage 36 ± 3 and 19 ± 6% of wild-type GCDH was observed as a 86 and 172 kDa dimer and tetramer, respectively (Fig. 5A and C). Additionally, GCDH immunopositive cross-link products of 97, 130 and 200 kDa were detectable (Fig. 5A). Both mutants p.Arg138Gly and p.Glu414Lys showed a cross-linker-induced oligomeric pattern similar to the wild-type GCDH forms. The mutant p.Glu414Lys, however, almost lost its ability to form the 97 kDa polypeptide complex. In cells expressing the p.Met263Val mutant the ability to form tetrameric GCDH forms was significantly reduced to 5 ± 1% of total (Fig. 5A and C). Due to the low expression and high degradation rate of mutant p.Arg402Trp (Figs 1 and 3) cross-linkage products could be visible only after longer exposure of the western blot (Fig. 5B). The data showed that the capability of the mutant p.Arg402Trp to form oligomeric complexes was reduced or lost.

Figure 5.

Oligomerization of wild-type and mutant GCDH. (A) Extracts of transfected wild-type and mutant GCDH-expressing BHK cells were incubated with (plus) and without (minus) 200 µm of the chemical cross-linker BS3. The samples were solubilized, separated by SDS–PAGE followed by western blot analysis using anti-GCDH antibodies. The positions of molecular mass markers and mono-, di- and tetrameric GCDH are indicated. The identities of the heteromeric GCDH-immunoreactive complexes (arrowheads) are unknown. (B) Section of western blot (A), lanes of untransfected Control and p.Arg402Trp protein, with 5-fold longer exposure time compared with (A). (C) Densitometric quantification of GCDH signal intensity of mono-, di- and tetramers, respectively, of three independent experiments. Significance was determined by one-way analysis of variance followed by Scheffé’s test. *P < 0.05 versus the respective wild-type mono- or oligomer.

Figure 5.

Oligomerization of wild-type and mutant GCDH. (A) Extracts of transfected wild-type and mutant GCDH-expressing BHK cells were incubated with (plus) and without (minus) 200 µm of the chemical cross-linker BS3. The samples were solubilized, separated by SDS–PAGE followed by western blot analysis using anti-GCDH antibodies. The positions of molecular mass markers and mono-, di- and tetrameric GCDH are indicated. The identities of the heteromeric GCDH-immunoreactive complexes (arrowheads) are unknown. (B) Section of western blot (A), lanes of untransfected Control and p.Arg402Trp protein, with 5-fold longer exposure time compared with (A). (C) Densitometric quantification of GCDH signal intensity of mono-, di- and tetramers, respectively, of three independent experiments. Significance was determined by one-way analysis of variance followed by Scheffé’s test. *P < 0.05 versus the respective wild-type mono- or oligomer.

Comparative modeling of GCDH mutants

In order to understand the role of the investigated mutations on a structural level, the individual mutations were generated and mutated on the 3D structure of human GCDH (Fig. 6A, 18). The initial protein structure was taken from the X-ray crystal structure of GCDH lacking the first 44 N-terminal amino acids in complex with 4-nitrobutyryl-CoA (NB-CoA; MMDB# 29361, PDB# 1SIQ; 18). The GCDH monomer itself has an overall ellipsoid shape and consists of three parts, an α-helical amino-terminus (residues 45–167) oriented side by side with the α-helical part formed by the C-terminus (residues 282–438), both parts forming the interaction surface to the right of the ligands (with respect to Fig. 6A). The middle region of GCDH (residues 168–281) forms a domain composed of two stacked β-sheets, interacting with the ligands from the left (with respect to Fig. 6A). Interestingly, functional GCDH is organized as a tetramer as it is in the crystal, although there is only one molecule in the asymmetric unit (18).

Figure 6.

Localization, arrangement and effects of the four investigated point mutations on GCDH 3D structure. The model is depicted in yellow. The carbons of FAD and NB-CoA are shown in ball and stick mode and colored in cyan and magenta, respectively, nitrogen in blue and oxygen in red. (A) Surface representation of GCDH with the sites bearing point mutations shown in blue. Note that the mutation p.Arg138Gly (=R138G) is located within the active site pocket and thus not directly visible (indicated by the dashed arrow). (B) Representation of modeled active site mutants p.Arg138Gly and p.Glu414Lys (=E414K). Shown in ball and stick mode are in the upper panel wild-type residue (carbons in yellow) and in the lower panel the respective point mutation in red. (C) The p.Arg402Trp (=R402W) mutation changes the side chain arrangement from pointing toward the surface of the protein (left panel) to an inward orientation (right panel). (D) The mutation p.Met263Val (=M263V) is located on the surface of the molecule (wild-type left panel and mutation right panel). (E) Surface potential representation of wild-type GCDH (left) and M263V mutant protein (right). Top: complete surface; bottom: magnification images of the regions marked by rectangles. Areas with altered surface potential are highlighted by dashed circles. (F) The mutation M263V is located in the vicinity of the dimerization region of GCDH. Left panel: surface representation with the mutation marked in blue; right panel: cartoon mode with the mutated side chain in ball and stick mode in blue and the dimerizing molecule in orange.

Figure 6.

Localization, arrangement and effects of the four investigated point mutations on GCDH 3D structure. The model is depicted in yellow. The carbons of FAD and NB-CoA are shown in ball and stick mode and colored in cyan and magenta, respectively, nitrogen in blue and oxygen in red. (A) Surface representation of GCDH with the sites bearing point mutations shown in blue. Note that the mutation p.Arg138Gly (=R138G) is located within the active site pocket and thus not directly visible (indicated by the dashed arrow). (B) Representation of modeled active site mutants p.Arg138Gly and p.Glu414Lys (=E414K). Shown in ball and stick mode are in the upper panel wild-type residue (carbons in yellow) and in the lower panel the respective point mutation in red. (C) The p.Arg402Trp (=R402W) mutation changes the side chain arrangement from pointing toward the surface of the protein (left panel) to an inward orientation (right panel). (D) The mutation p.Met263Val (=M263V) is located on the surface of the molecule (wild-type left panel and mutation right panel). (E) Surface potential representation of wild-type GCDH (left) and M263V mutant protein (right). Top: complete surface; bottom: magnification images of the regions marked by rectangles. Areas with altered surface potential are highlighted by dashed circles. (F) The mutation M263V is located in the vicinity of the dimerization region of GCDH. Left panel: surface representation with the mutation marked in blue; right panel: cartoon mode with the mutated side chain in ball and stick mode in blue and the dimerizing molecule in orange.

Models of the GCDH mutants were obtained by exchanging the residues at the respective positions Arg138, Met263, Arg402 and Glu414 of the full-length GCDH precursor molecule and optimized by energy minimization. No substantial changes were observed in the backbone of mutated molecules compared with wild-type GCDH when the individual models were superimposed. Of the mutated residues, two mutated residues, Arg138 and Glu414, are located at the active site pocket whereas the two other mutated amino acids, namely Met263 and Arg402, are localized on the surface of the left and right domain of GCDH, respectively (Fig. 6A).

Comparison of the wild-type and of the mutations led to the following observations. The Arg138 and Glu414 residues mutated in the active site mutants, p.Arg138Gly and p.Glu414Lys, are both located at the bottom of the active site. Arg138 is located at the end of the pocket whereas Glu414 is oriented toward the pocket entry. Both mutants are directly interfering with the binding of the ligand, NB-CoA, an alternate substrate, but isosteric and isoelectric to glutaryl-CoA. The FAD coordinates the orientation of NB-CoA from the top.

Arg138 interacts directly with nitroyl group of NB-CoA and stabilizes its position (18). The Arg138Gly substitution has a major impact on the proposed interactions of Arg138 with, and stabilization of, the γ-carboxylate of glutaryl-CoA as well as the neutralization of the negative charge on C4 in the transient crotonyl-CoA anion intermediate and the decarboxylation transition state (18). Due to the lack of any side chain p.Arg138Gly is unable to perform any of the aforementioned interactions (Fig. 6B top).

Glu414 is directly involved in the positioning of the NB-CoA between itself and the isoalloxazine ring of the FAD by interacting via the amide nitrogen with the carbonyl ester oxygen of the NB-CoA, whereas the C2–C3 bond of NB-CoA is coordinated by the carboxylate group of Glu. The mutation Glu414Lys affects the catalytic base of GCDH that abstracts protons, and impairs the orientation of the substrate. Substitution of Glu to Lys changes the charge of the side chain from a negative to positive which in turn affects the local charge distribution. The Glu414 interaction is partially lost and has the potential to destabilize the other ionic interactions required for proper positioning of the NB-CoA (Fig. 6B bottom).

The remaining two mutations investigated are localized distant from the active site, suggesting a different way of influencing the activity of GCDH. Position Arg402 is localized at the end of a helix that is in the vicinity of the helix which harbors the active site Arg138 and the short loop region preceding this helix (Fig. 6C). The side chain of Arg402 is oriented toward helix and the loop, and its hydrogen bonding with the neighboring amino acids, namely Asp134, Val133 and Arg132, might play an important role in stabilization of the enzyme by positioning the two helices. The Arg402Trp changes the charged elongated side chain into a bulky and hydrophobic one (Fig. 6C), thereby replacing the hydrogen bonds by a sole hydrophobic contact with Asp134 only (data not shown). This strongly suggests that the mutation of Arg to Trp affects the spatial arrangement and the dynamics of this region, which might as well result in a decrease of stability, rendering the molecules prone to degradation.

Met263 is localized in a beta sheet on the surface of GCDH (Fig. 6D). The mutation Met263Val results in a change from an elongated unpolar side chain to a shorter and bulky unpolar side chain with only minor changes of the surface potential from neutral to slightly more positive (Fig. 6E). The interaction pattern of p.Met263Val forms two more hydrophobic contacts than wild-type GCDH (data not shown). Moreover, Met263 is located in a region that is not directly involved in the dimerization of the subunits of GCDH (Fig. 6F), nor with a significant influence on the active site arrangement, strongly arguing against any effect on stability and/or activity.

DISCUSSION

We investigated the molecular pathology underlying GCDH dysfunction in GA1. More than 150 mutations in the GCDH gene are known and cause considerable variations in clinical symptoms ranging from an asymptomatic course to severe disabling dystonia (10). We selected four GCDH missense mutations which affect the catalytic center of the enzyme, i. e. p.Arg138Gly and p.Glu414Lys (1,18), or showed the highest prevalence in Caucasian patients (p.Arg402Trp, 2). The homozygous mutation p.Met263Val was found in a patient with a severe dystonic-dyskinetic movement disorder although 30% of control GCDH activity was detected in this patient’s fibroblasts (15). All amino acid residues investigated in this study are highly conserved in human, mouse, rat, cow and frog GCDH proteins.

We analyzed for the first time the impact of these mutations in the GCDH gene on enzymatic activity, stability, localization, and their capability to affect oligomerization of GCDH monomers in mammalian cells. The introduction of these mutations into human GCDH cDNA, followed by overexpression in BHK cells, showed that almost all mutations resulted in the complete loss of enzymatic activity with the exception of p.Met263Val exhibiting 10% of wild-type GCDH expressing cells. The discrepancy in the relative GCDH activity in overexpressing BHK cells and patient fibroblasts (10 versus 30% of activity in control cells) cannot be explained by differences in the steady state concentration of the mutant GCDH which was only reduced by 20–30% in the p.Met263Val expressing BHK cells. It is more likely that overexpression of mutant p.Met263Val GCDH impairs other parameters such as folding or the rate and efficiency of transport to mitochondria, which contribute to cellular GCDH activity. Whereas both p.Arg138Gly and p.Glu414Lys exhibited no enzymatic activity in BHK cells, only p.Arg138Gly lost enzymatic activity when expressed in E. coli (1,8). GCDH modeling analysis confirmed that the two active site residues Arg138 and Glu414 are both required for proper arrangement of the ligand. In the investigated GCDH mutants this function is abrogated, the ligand not held in place properly, which results in the observed loss of activity. In contrast, mutant GCDH Arg402Trp or Glu414Lys expressed in E. coli showed residual activities of 3–4.5% of controls (6,8), whereas respective homozygous mutations in patient cells showed no enzymatic activity at all (13) demonstrating the advantage of mammalian expression systems in comparison with prokaryotic expression analyses.

The relatively low steady state concentration of the mutant GCDH Arg402Trp in comparison with wild-type GCDH may be due to a decreased rate of synthesis or an increased degradation. Pulse-chase experiments indicated that an increase in degradation is responsible for the low level of mutant GCDH Arg402Trp resulting in a calculated half-life of the protein of ∼4 h in comparison with 34.5 h of wild-type GCDH. Arg402 is located close to the surface of the GCDH polypeptide and the mutation changes an elongated charged side chain pointing into a bulky hydrophobic side chain. The altered arrangement of the local area might result in a destabilization and faster degradation of the molecule. In addition, the biosynthetic studies demonstrated that all mutants including p.Arg402Trp are transported normally to mitochondria. GCDH is synthesized at ribosomes as a 48 kDa precursor protein which is proteolytically processed upon import into the mitochondrial matrix to a 43 kDa mature form (7). The mitochondrial localization signal of GCDH is part of the 44 amino acid comprising propeptide which is cleaved after translocation by mitochondrial processing peptidases (19). Both the 48 and 43 kDa form were observed after 2 h of labeling with (35S)-methionine and were completely converted to the 43 kDa form during the 24 h chase period. Double immunofluorescence microscopy provided additional evidence that the mutant GCDH Arg402Trp co-localizes with the mitochondrial marker protein MnSOD, suggesting that the mutant is degraded intramitochondrially by yet unknown proteases.

GCDH functions as a homotetramer exhibiting a tetrahedral symmetry in which 30% of each monomer’s surface is buried by the other three monomers (2,18). By analyzing GCDH mutants expressed in E. coli by size exclusion chromatography, it has been suggested that mutations in the carboxy-terminal region of GCDH (amino acid residues 282–438) resulted in a reduced capability to form tetrameric complexes leading to inactive monomers and/or dimers (6,16). In this study we used chemical cross-linkage to demonstrate the capability of wild-type as well as mutant GCDH to form dimeric and tetrameric complexes of 86 and 172 kDa, respectively. The relative ratios of tetrameric, dimeric and monomeric GCDH are 0.4:0.8:1 in wild-type, 0.2:0.8:1 in the mutant GCDH Arg138Gly, and 0.5:1:1 in the mutant GCDH Glu414Lys. The mutant GCDH Met263Val and Arg402Trp, however, have almost or completely lost their capability to form tetrameric complexes. It appears that in addition to the reduced capability to form tetrameric complexes, other factors contribute to the loss of enzymatic activity. Of note, the used method for determination of GCDH activity uses an artificial electron acceptor which might not consider mutations affecting the binding of GCDH to the electron transferring flavoprotein (ETF). ETF is responsible for transferring electrons from GCDH to the membrane-bound respiratory chain (20). Most notable are the novel GCDH protein complexes of 97, 130 and 200 kDa detected by cross-linkage. These data demonstrate that monomeric GCDH can interact with proteins with apparent molecular masses of 54, 87 and 157 kDa (after subtraction of 43 kDa corresponding to GCDH), or they represent multimeric protein complexes. It is possible that at least one of these GCDH protein complexes contains the 58 kDa dimeric ETF. Further studies are required to identify the interacting proteins and their biological significance for GCDH structure and function. The lack and the weak signal intensities of the 200 and 130 kDa cross-linkage products, respectively, in cells expressing mutant GCDH Met263Val suggests that Met263 might be involved in this novel GCDH–protein interactions. Met263 is located on the surface of the protein and is neither involved in any structural function nor in dimerization. The loss of GCDH activity of mutant p.Met263Val suggests that the Met263-mediated formation of 130 and/or 200 kDa protein complexes might function in regulation of enzymatic activity.

Taken together, in this study four GA1-causing missense mutations in the GCDH gene have been expressed in mammalian cells. The data indicate that all GCDH mutants are synthesized to a similar extent as the wild-type enzyme. All mutants are transported into mitochondria but they differ in enzymatic activities and stability. The failure of p.Met263Val and p.Arg402Trp to form tetrameric complexes appears to be associated with increased degradation rates. Our study for the first time provides evidence that GCDH may also interact with other proteins to form high molecular mass complexes. The identification of the interacting proteins and their role in regulation of GCDH enzymatic activity, stability, and their potential relevance for the clinical variability in GA1 remains to be investigated.

MATERIALS AND METHODS

Materials

(35S)-Methionine and the prestained Rainbow molecular mass marker were purchased from Amersham Biosciences Europe (Freiburg, Germany). The following reagents were obtained commercially as indicated: restriction enzymes and T4 DNA ligase were from New England Biolabs (Schwalbach, Germany), Pfu Turbo polymerase and QuikChange™ site-directed mutagenesis kit were from Stratagene (Amsterdam, The Netherlands), Dulbecco’s minimal essential medium (DMEM), fetal calf serum (FCS), penicillin, streptomycin, trypsin/EDTA and OPTIMEM from Gibco (Eggenstein, Germany), Lipofectamine 2000 and the expression vector pcDNA6.2/V5/GW/TOPO from Invitrogen (Karlsruhe, Germany). Plasmid-Mini and Midi Kit and gel extraction kit were from Qiagen (Hilden, Germany). Oligonucleotide primers used for cloning and sequencing were synthesized by MWG Biotech (Munich, Germany). Protein A-agarose and protease inhibitor cocktail were purchased from Sigma (Munich, Germany). Enhanced chemiluminescence (ECL) reagents and the cross-linker BS3 were purchase from Pierce (Rockford, IL, USA).

DNA constructs

Expression constructs coding for full-length GCDH with a C-terminal myc-tag were generated by PCR using the human GCDH cDNA (German Resource Center for Genome Research (IRAUp969D038D6, RZPD Berlin, Germany), as template and the following primers: GCDH-myc-for: 5′-CACCATGGCCCTGAGAGGCGTCTCC-3′ and GCDH-myc-rev: 5′-TCACAGATCCTCTTCTGAGATGAGTTTTTGTTCCTTGCTGGCCGTGAA-3′. Mutations were introduced into the human GCDH cDNA (p.Arg138Gly, p.Met263Val, p.Arg402Trp, p.Glu414Lys) using the QuikChange™ site-directed mutagenesis kit (Stratagene) and the following primers (mutated nucleotides are underlined) Arg138Gly-myc-for: 5′-GGTGGACAGTGGCTACGGGTCGGCGATGAGTGTCC-3′ and Arg138Gly-myc-rev: 5′-GGACACTCATCGCCGACCCGTAGCCACTGTCCACC-3′; Met263Val-myc-for: 5′-GGGCCTCAGCCACAGGCGTGATCATCATGGACGG-3′ and Met263Val-myc-rev: 5′-CCGTCCATGATGATCACGCCTGTGGCTGAGGCCC-3′; Arg402Trp-myc-for: 5′-CGAGTATCACGTGATCTGGCACGCCATGAACCTGGAGG-3′ and Arg402Trp-myc-rev: 5′-CCTCCAGGTTCATGGCGTGCCAGATCACGTGATACTCG-3′; Glu414Lys-myc-for: 5′-GGCCGTGAACACCTACAAAGGTACACATGACATTCACGCCC-3′ and Glu414Lys-myc-rev: 5′-GGGCGTGAATGTCATGTGTACCTTTGTAGGTGTTCACGGCC-3′. Wild-type and mutated GCDH cDNAs were cloned into expression vector pcDNA6.2/V5/GW/TOPO according to the manufacturer’s instructions. All constructs were sequenced using the Big Dye terminator kit (Perkin Elmer Life Sciences, Waltham, MA, USA) on an Applied Biosystems model 377 automated DNA Sequencer (in the University Medical Center Service Center, Hamburg).

Cell culture and transfection

BHK cells were cultured in DMEM containing 10% FCS and antibiotics in a humidified atmosphere containing 5% CO2 at 37°C. Cells were seeded on 60 mm dishes at a density of 1 × 106 cells/dish and on 35 mm dishes at a density of 0.5 × 106 cells/dish, respectively. After 24 h of cultivation the cells were transiently transfected with wild-type and mutant GCDH cDNAs in pcDNA6.2/V5/GW/TOPO or with vector alone using Lipofectamine 2000. At 6 h after transfection the medium was replaced with DMEM containing 10% FCS and penicillin/streptomycin.

Antibodies

Rabbit anti-human GCDH antibody was kindly provided by Dr. Michael Woontner and Professor Stephen I. Goodman (University of Colorado Health Sciences Center, Denver). The monoclonal anti-myc antibody was purchased from Sigma and rabbit anti-MnSOD was from Upstate (Biomol, Hamburg, Germany). Peroxidase-conjugated goat anti-rabbit IgG was from Dianova (Hamburg, Germany). Sheep anti-mouse IgG coupled to fluorescein isothiocyanate (FITC) and anti-rabbit IgG-Cy3 were from Sigma.

Cell fractionation and western blotting

At 48 h after transfection cells were washed with ice cold 10 mm phosphate-buffered saline pH 7.4 (PBS) and 10 mm Tris–HCl pH 7.4 containing 0.25 m sucrose (buffer A) and 2 mm EDTA. The cells were harvested in buffer A and centrifugated for 5 min at 500g at 4°C. The cell pellet was homogenized with a 20G syringe in 0.2 ml buffer A and centrifugated for 5 min at 1000g at 4°C. The postnuclear supernatant was centrifugated for 15 min at 10 000g at 4°C and the mitochondrial-enriched pellet was resuspended in buffer A containing protease inhibitor cocktail. Protein concentration was calculated with the Bio-Rad protein assay (Munich, Germany). Solubilized cell extracts (100 µg) were separated on SDS–PAGE (10% acrylamide), blotted onto Protran® nitrocellulose membranes (Whatman, Dassel, Germany) as described previously (21). After blocking, the membranes were incubated with rabbit anti-GCDH antibody (1:10 000) or rabbit anti-MnSOD (1:1000) and visualized with peroxidase-conjugated anti-rabbit IgG (1:10 000) and ECL.

Metabolic labeling and immunoprecipitation

At 24 h after transfection, cells were metabolically labeled with (35S)-methionine (75 µCi/ml) for 2 h as described previously (22). After removing the labeling medium, cells were washed twice with PBS and either harvested or chased in DMEM containing 0.1% bovine serum albumin (BSA) and 0.25 mg/ml methionine. Cells were lysed in lysis buffer (0.4% Triton X-100, 0.2% sodium deoxycholate, 0.2% sodium dodecylsulfate, 1% BSA in PBS). After successive removal of DNA by means of 50 U benzonase and 0.03% protamine sulphate, extracts were preabsorbed with rabbit IgG (1:500) and protein A agarose for 60 min at 4°C. After precipitation of the agarose beads (500g for 20 min at 4°C), supernatants were mixed with rabbit anti-human GCDH antibody (1:500) and incubated for 12 h at 4°C. The immunocomplexes were precipitated by means of protein A agarose, washed, and processed for SDS–PAGE under reducing conditions (10% acrylamide) and fluorography.

Immunofluorescence microscopy

For double immunofluorescence microscopy 2.5 × 105 BHK cells were grown on glass coverslips in a six-well plate for 6 h and transfected with 2.5 µg pcDNA 6.2-GCDH-myc and pcDNA6.2-GCDH-Arg402Trp-myc. The medium was changed 12 h after transfection and the cells were grown for additional 24 h. Cells were fixed with 4% paraformaldehyde for 30 min, washed with PBS, permeabilized with 1% Triton X-100 for 10 min and blocked with PBS containing 2% FCS (PBS–FCS). Cells were incubated using monoclonal mouse anti-myc antibody (1:2000) and rabbit anti-MnSOD antibody (1:200) in PBS–FCS for 90 min at room temperature. Incubation with secondary antibodies was performed at room temperature for 60 min using anti-mouse FITC (1:400) and anti-rabbit Cy3 (1:400) in PBS–FCS. Coverslips were mounted in Mowiol (Merck, Darmstadt, Germany) and viewed with a Leica DM IRE2 digital scanning confocal microscope with TCS NT software (Leica Microscopy Scientific Instruments Group, Wetzlar, Germany).

Determination of GCDH activity

GCDH activity of non-transfected and BHK cells expressing wild-type and mutant GCDH was determined as described previously (23).

Cross-linking of GCDH

At 24 h after transfection, BHK cells grown on 60 mm dishes were harvested in PBS and centrifugated at 500g for 10 min at 4°C. Cells were lysed in 100 µl binding buffer (100 mm HEPES pH 7.6, 120 mm NaCl, 1.8 mm MgSO4, 5 mm KCl, 8 mm glucose, 0.25% saponine, 1% Triton X-100). After centrifugation at 20 000g for 10 min at 4°C, equal aliquots of the supernatant were incubated in the absence and presence of 200 µm BS3 for 30 min at 4°C. The reaction was stopped by addition of 10 µl of 1 M Tris pH 7.4, followed by SDS–PAGE and western blotting.

In silico modeling of effects of GCDH mutations on 3D protein structure

Models of the GCDH mutants were obtained by exchanging the residues at the respective positions: Arg138 to Gly, Met263 to Val, Arg402 to Trp, and Glu414 to Lys according to the sequence numbering of the full-length GCDH precursor molecule, and the energy minimized models were generated subsequently by SYBYL (TRIPOS, INC). Resulting figures were generated using PYMOL (The PYMOL Molecular Graphics System 2002, DeLano Scientific, San Carlos, USA). Interactions between single amino acid residues, hydrophobic contacts and hydrogen bonds were calculated and displayed using LIGPLOT™ v.4.4.2 software (24).

Data analysis and software

Densitometric quantification of chemiluminiscence signals was performed by using AIDA Image Analyzer software version 4.14.025 (Raytest, Straubenhardt, Germany). Significance was determined by one-way analysis of variance followed by Scheffé’s test using SPSS 15.0 software (SPSS, Chicago, IL, USA) and accepted at P < 0.05.

FUNDING

This work was supported by Deutsche Forschungsgemeinschaft (DFG grant MU1778/2-1 to B.K. and C.M.).

ACKNOWLEDGEMENTS

We are grateful to Dr. Michael Woontner and Professor Stephen I. Goodman (University of Colorado Health Sciences Center, Denver) for providing the rabbit anti-human GCDH antibody.

Conflict of Interest statement. All authors decline any financial conflict of interest that might be construed to influence the results or interpretation of data.

REFERENCES

1
Dwyer
T.M.
Rao
K.S.
Westover
J.B.
Kim
J.-J.P.
Frerman
F.E.
The function of Arg-94 in the oxidation and decarboxylation of glutaryl-CoA by human glutaryl-CoA dehydrogenase
J. Biol. Chem.
 , 
2001
, vol. 
276
 (pg. 
133
-
138
)
2
Goodman
S.I.
Frerman
F.E.
Scriver
C.R.
Beaudet
A.L.
Sly
W.S.
Valle
D.
Childs
B.
Kinzler
K.W.
Vogelstein
B.
Organic acidemias due to defects in lysine oxidation: 2-ketoadipic acidemia and glutaric acidemia
The Metabolic and Molecular Bases of Inherited Disease
 , 
2001
8th edn
New York, USA
McGraw-Hill Inc.
(pg. 
2195
-
2204
)
3
Kölker
S.
Garbade
S.F.
Greenberg
C.R.
Leonard
J.V.
Saudubray
J.-M.
Ribes
A.
Kalkanoglu
S.
Lund
A.M.
Merinero
B.
Wajner
M.
, et al.  . 
Natural history, outcome, and treatment efficacy in children and adults with glutaryl-CoA dehydrogenase deficiency
Pediatr. Res.
 , 
2006
, vol. 
59
 (pg. 
840
-
846
)
4
Mühlhausen
C.
Hoffmann
G.F.
Strauss
K.A.
Kölker
S.
Okun
J.G.
Greenberg
C.R.
Naughten
E.R.
Ullrich
K.
Maintenance treatment of glutaryl-CoA-dehydrogenase deficiency
J. Inherit. Metab. Dis.
 , 
2004
, vol. 
27
 (pg. 
885
-
892
)
5
Lindner
M.
Kölker
S.
Schulze
A.
Christensen
E.
Greenberg
C.R.
Hoffmann
G.F.
Neonatal screening for glutaryl-CoA dehydrogenase deficiency
J. Inherit. Metab. Dis.
 , 
2004
, vol. 
27
 (pg. 
851
-
859
)
6
Biery
B.J.
Stein
D.E.
Morton
D.H.
Goodman
S.I.
Gene structure and mutations of glutaryl-coenzyme A dehydrogenase: impaired association of enzyme subunits that is due to an A421V substitution causes glutaric acidemia type I in the Amish
Am. J. Hum. Genet.
 , 
1996
, vol. 
59
 (pg. 
1006
-
1011
)
7
Goodman
S.I.
Kratz
L.E.
DiGiulio
K.A.
Biery
B.J.
Goodman
K.E.
Isaya
G.
Frerman
F.E.
Cloning of glutaryl-CoA dehydrogenase cDNA, and expression of wild type and mutant enzymes in Escherichia coli.
Hum. Mol. Genet.
 , 
1995
, vol. 
4
 (pg. 
1493
-
1498
)
8
Goodman
S.I.
Stein
D.E.
Schlesinger
S.
Christensen
E.
Schwartz
M.
Greenberg
C.R.
Elpeleg
O.N.
Glutaryl-CoA dehydrogenase mutations in glutaric acidemia (type I): review and report of thirty novel mutations
Hum. Mutat.
 , 
1998
, vol. 
12
 (pg. 
141
-
144
)
9
Schwartz
M.
Christensen
E.
Superti-Furga
A.
Brandt
N.J.
The human glutaryl-CoA dehydrogenase gene: report of intronic sequences and of 13 novel mutations causing glutaric aciduria type I
Hum. Genet.
 , 
1998
, vol. 
102
 (pg. 
452
-
458
)
10
Busquets
C.
Merinero
B.
Christensen
E.
Gelpi
J.L.
Campistol
J.
Pineda
M.
Fernandez-Alvarez
E.
Prats
J.M.
Sans
A.
Arteaga
R.
, et al.  . 
Glutaryl-CoA dehydrogenase deficiency in Spain: evidence of two groups of patients, genetically, and biochemically distinct
Pediatr. Res.
 , 
2000
, vol. 
48
 (pg. 
315
-
322
)
11
Zschocke
J.
Quak
E.
Guldberg
P.
Hoffmann
G.F.
Mutation analysis in glutaric aciduria type I
J. Med. Genet.
 , 
2000
, vol. 
37
 (pg. 
177
-
181
)
12
Greenberg
C.R.
Reimer
D.
Singal
R.
Triggs-Raine
B.
Chudley
A.E.
Dilling
L.A.
Philipps
S.
Haworth
J.C.
Seargeant
L.E.
Goodman
S.I.
A G-to-T transversion at the +5 position of intron 1 in the glutaryl CoA dehydrogenase gene is associated with the Island Lake variant of glutaric acidemia type I
Hum. Mol. Genet.
 , 
1995
, vol. 
4
 (pg. 
493
-
495
)
13
Christensen
E.
Ribes
A.
Merinero
B.
Zschocke
J.
Correlation of genotype and phenotype in glutaryl-CoA dehydrogenase deficiency
J. Inherit. Metab. Dis.
 , 
2004
, vol. 
27
 (pg. 
861
-
868
)
14
Pineda
M.
Ribes
A.
Busquets
C.
Vilaseca
M.A.
Aracil
A.
Christensen
E.
Glutaric aciduria type I with high residual glutaryl-CoA dehydrogenase activity
Dev. Med. Child Neurol.
 , 
1998
, vol. 
40
 (pg. 
840
-
842
)
15
Mühlhausen
C.
Christensen
E.
Schwartz
M.
Muschol
N.
Ullrich
K.
Lukacs
Z.
Severe phenotype despite high residual glutaryl-CoA dehydrogenase activity: a novel mutation in a Turkish patient with glutaric aciduria type 1
J. Inherit. Metab. Dis.
 , 
2003
, vol. 
27
 (pg. 
713
-
714
)
16
Westover
J.B.
Goodman
S.I.
Frerman
F.E.
Pathogenic mutations in the carboxyl-terminal domain of glutaryl-CoA dehydrogenase: effects on catalytic activity and the stability of the tetramer
Mol. Genet. Metab.
 , 
2003
, vol. 
79
 (pg. 
245
-
256
)
17
Rao
K.S.
Fu
Z.
Albro
M.
Narayanan
B.
Baddam
S.
Lee
H.-J.K.
Kim
J.-J.P.
Frerman
F.E.
The effect of a Glu370Asp mutation in glutaryl-CoA dehydrogenase on proton transfer to the dienolate intermediate
Biochemistry
 , 
2007
, vol. 
46
 (pg. 
14468
-
14477
)
18
Fu
Z.
Wang
M.
Paschke
R.
Rao
K.S.
Frerman
F.E.
Kim
J.P.
Crystal structures of human glutaryl-CoA dehydrogenase with and without an alternate substrate: structural bases of dehydrogenation and decarboxylation reactions
Biochemistry
 , 
2004
, vol. 
43
 (pg. 
9674
-
9684
)
19
Koppen
M.
Langer
T.
Protein degradation within mitochondria: versatile activities of AAA proteases and other peptidases
Crit. Rev. Biochem. Mol. Biol.
 , 
2007
, vol. 
42
 (pg. 
221
-
242
)
20
Toogood
H.S.
Leys
D.
Scrutton
N.S.
Dynamics driving function - new insights from electron transferring flavoproteins and partner complexes
FEBS J.
 , 
2007
, vol. 
274
 (pg. 
5481
-
5504
)
21
Stellmer
F.
Keyser
B.
Burckhardt
B.C.
Koepsell
H.
Streichert
T.
Glatzel
M.
Jabs
S.
Thiem
J.
Herdering
W.
Koeller
D.M.
, et al.  . 
3-Hydroxyglutaric acid is transported via the sodium-dependent dicarboxylate transporter NaDC3
J. Mol. Med.
 , 
2007
, vol. 
85
 (pg. 
763
-
770
)
22
Muschol
N.
Matzner
U.
Tiede
S.
Gieselmann
V.
Ullrich
K.
Braulke
T.
Secretion of phosphomannosyl-deficient arylsulphatase A and cathepsin D from isolated human macrophages
Biochem. J.
 , 
2002
, vol. 
368
 (pg. 
845
-
853
)
23
Christensen
E.
Improved assay of glutaryl-CoA dehydrogenase in cultured cells and liver: application to glutaric aciduria type I
Clin. Chim. Acta
 , 
1983
, vol. 
129
 (pg. 
91
-
97
)
24
Wallace
A.C.
Laskowski
R.A.
Thornton
J.M.
LIGPLOT: a program to generate schematic diagrams of protein–ligand interactions
Protein Eng.
 , 
1995
, vol. 
8
 (pg. 
127
-
134
)

Author notes

The authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors.