fungal metabolic model for 3-methylcrotonyl-coa ... · pdf filedehydrogenase complex lead to...

Fungal metabolic model for 3-methylcrotonyl-CoA carboxylase

deficiency

José M. Rodríguez1, Pedro Ruíz-Sala2, Magdalena Ugarte2 and Miguel Á. Peñalva1,3

1 Centro de Investigaciones Biológicas CSIC, Ramiro de Maeztu 9, Madrid 28040,

Spain

2 Centro de Biología Molecular CSIC-UAM, Universidad Autónoma de Madrid, Canto

Blanco, Madrid 28034, Spain

3 Corresponding author

Prof. Miguel Á. Peñalva

Departamento de Microbiología Molecular

Centro de Investigaciones Biológicas CSIC

Ramiro de Maeztu, 9

Madrid 28040, Spain

Phone: + 3491 8373112, ext. 4358

Fax: +3491 5360432

e-mail: [email protected]

RUNNING TITLE : A model for 3-methylcrotonyl-CoA carboxylase deficiency

1

Copyright 2003 by The American Society for Biochemistry and Molecular Biology, Inc.

JBC Papers in Press. Published on November 11, 2003 as Manuscript M310055200 by guest on A

pril 19, 2018http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

SUMMARY The filamentous fungus Aspergillus nidulans is able to use Leu as sole carbon source

through a metabolic pathway leading to acetyl-CoA and acetoacetate that is homologous

to that used by humans. mccA and mccB, the genes encoding the subunits of 3-

methylcrotonyl-CoA carboxylase, are clustered with ivdA encoding isovaleryl-CoA

dehydrogenase, a third gene of the Leu catabolic pathway, on the left arm of

chromosome III. Their transcription is induced by Leu and other hydrophobic amino

acids and repressed by glucose. Phenotypically indistinguishable ∆mccA, ∆mccB and

∆mccA ∆mccB mutations prevent growth on Leu but not on lactose or other amino

acids, formally demonstrating in vivo the specific involvement of 3-methylcrotonyl-

CoA carboxylase in Leu catabolism. Growth of mcc mutants on lactose plus Leu is

impaired, indicating that Leu metabolite(s) accumulation resulting from the metabolic

block is toxic. Human patients carrying loss-of-function mutations in the genes

encoding the subunits of 3-methylcrotonyl-CoA carboxylase suffer from

methylcrotonylglycinuria. GC/MS analysis of culture supernatants revealed that fungal

∆mcc strains accumulate 3-hydroxyisovaleric acid, one of the diagnostic compounds in

the urine of these patients, illustrating the remarkably similar consequences of

equivalent genetic errors of metabolism in fungi and humans. We use our fungal

model(s) for methylcrotonylglycinuria to show accumulation of 3-hydroxyisovalerate

on transfer of 3-methylcrotonyl-CoA carboxylase-deficient strains to the isoprenoid

precursors acetate, 3-hydroxy-3-methylglutarate or mevalonate. This represents the first

reported genetic evidence for the existence of a metabolic link involving 3-

methylcrotonyl-CoA carboxylase between isoprenoid biosynthesis and Leu catabolism,

providing additional support to the mevalonate shunt proposed by Edmond and Popjáck.

2

by guest on April 19, 2018

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

INTRODUCTION

In humans, Leu is a ketogenic amino acid that is converted to acetyl-CoA through a six-

step pathway (1)(see Fig. 1). The first step in this pathway is catalyzed by the branched

chain aminoacid (BCAA)1 transaminase which appears to act on all three BCAAs (Leu,

Ile, Val) to yield the corresponding branched chain ketoacids (BCKAs). All three

ketoacids are substrates of the BCKA dehydrogenase complex which catalyzes their

oxidative decarboxylation to the respective –1 acyl-CoA derivatives (step II, Fig. 1,

isovaleryl-CoA in the case of Leu catabolism). Isovaleryl-CoA is oxidized to acetyl-

CoA and acetoacetate through four additional steps, specific for Leu catabolism (III-VI,

Fig. 1) (2).

Individuals deficient for any of the six enzymes of this pathway have a clinical

phenotype (see Fig. 1). For example, inborn errors of metabolism resulting from loss-of-

function mutations in any of the structural genes encoding subunits of the BCKA

dehydrogenase complex lead to maple syrup urine disease, a severe, not infrequent

panethnic recessive disorder (2). However, in spite of the clinical relevance of this

pathway, the molecular basis of two enzyme deficiencies in Leu catabolism had not

been established when this work was started. These were the deficiencies of 3-

methylcrotonyl-CoA carboxylase (MCC; E.C.6.4.1.4.), resulting in 3-

methylcrotonylglycinuria (OMIM 210200), and 3-methylglutaconyl-CoA hydratase

(E.C.4.2.1.18), resulting in type I 3-methylglutaconic aciduria (OMIM 250950). While

the characterization of the 3-methylglutaconyl-CoA hydratase-mediated step will be the

subject of another paper, this paper describes the genetic and biochemical

characterization of the MCC-mediated step in A. nidulans. The derived amino acid

sequences of the fungal MCC subunits were instrumental for our identification of their

3


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

human homologues and subsequent characterization of the molecular basis of the

disease (3) which was simultaneously reported by Baumgartner et al. (4).

The mitochondrial enzyme MCC is, with propionyl-CoA carboxylase (PCC),

pyruvate carboxylase (PC) and acetyl-CoA carboxylase (ACC), one of the four biotin-

dependent carboxylases in humans. All four enzymes share a biotin carrier domain

carrying the covalently bound prosthetic group, a biotin carboxylating domain and a

carboxyltransferase domain catalyzing carboxyl group transfer from carboxybiotin to

the specific substrate. In the current model (5), the biotin carrier domain acts as a

swinging arm oscillating between the biotin carboxylase and the carboxyltransferase

domains of the enzyme. As PCC , MCC is a heterodimeric enzyme consisting of α and

β subunits in an (αβ)6 configuration (6), whose corresponding genes were first

characterized in plants (7, 8). MCCα contains the biotin-carboxylating and biotin carrier

domains and MCCβ contains the carboxyltransferase domain (3, 4, 7-9).

The metabolically versatile fungus A. nidulans is able to use most amino acids as

sole carbon source through catabolic pathways homologous to those used by human

hepatocytes (10, 11; this work). High level expression of genes for an amino acid

catabolic pathway usually requires their induction by the corresponding pathway

substrate and their transcription is low or undetectable in its absence, which facilitates

the cloning of the corresponding cDNAs using subtractive procedures (10-13) or cDNA

libraries from cells cultured under inducing conditions (this work). Amino acid

catabolic genes are often clustered (10-13; this work) which, with the small average

intron size (less than 100 bp), facilitates the molecular genetic characterization of the

pathways. The fungus has a relatively compact, haploid genome of only 31 Mb whose

complete sequence is available (http://www-genome.wi.mit.edu). Haploidy, the absence

of significant gene redundancy and the relative ease with which A. nidulans can be

4


http://ww

w.jbc.org/

Dow

nloaded from

http://www-genome.wi.mit.edu/

http://www.jbc.org/

manipulated by reverse genetics facilitate gene inactivation by homologous

recombination/gene replacement, using a relatively rapid procedure. Mutations can be

recombined using the sexual cycle, which is completed within two weeks. Lastly, the

fungus can be cultured in a chemically defined medium containing inorganic salts and

sole nitrogen and carbon sources which can be changed as appropriate, permitting

normal growth of mutants deficient in amino acid catabolic pathways by using

alternative carbon sources, such as carbohydrates. These advantages over human cell

cultures are extensively exploited in this work.

In addition to its role in Leu catabolism, MCC has been proposed to be involved

in a shunt pathway by which mevalonate-derived isopentenyl pyrophosphate is

converted to mitochondrial 3HMG-CoA (Fig. 1). This pathway is of notable interest

because it might divert isoprene units from cholesterol biosynthesis (14). However,

evidence supporting a physiological role for this pathway is solely based on tracer

studies using radioisotope-labeled mevalonate (for a review, see (14)). Here we describe

the molecular and biochemical characterization of A. nidulans genes encoding subunits

of MCC and the construction of fungal models for MCC deficiency. Strains carrying

null mutations in the genes encoding MCC subunits were used to demonstrate

unambiguously the in vivo involvement of this enzyme in Leu catabolism. Biochemical

characterization of these strains illustrates further the similar metabolic consequences

caused by homologous amino acid degradation enzyme deficiencies in fungi and

humans. Finally, we exploit these models and the ease with which the composition of

the A. nidulans growth medium can be manipulated à la carte to provide genetic

evidence on the presence of a metabolic link between isoprenoid biosynthesis and Leu

catabolism involving MCC .

5


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

EXPERIMENTAL PROCEDURES

Fungal strains, media and growth condition. A. nidulans strains carried markers

in standard use (15). Complete and minimal medium were as in (16). Lactose was used

at 0.05% (w/v) as sole carbon source and amino acids at 30 mM, unless otherwise

indicated.

RNA isolation. 7x106 conidiospores/ml were inoculated on minimal medium

containing 0.3 % (w/v) glucose as carbon source and 5 mM ammonium tartrate as

nitrogen source. Primary cultures were incubated at 37ºC. When the glucose

concentration fell below 0.01%, mycelia were transferred to 5 mM urea minimal

medium without any carbon source, and incubated for a further 1 h before adding an

appropriate carbon source. These secondary cultures were incubated for a further 3 h

before collecting mycelia for RNA extraction. Mycelia were frozen in liquid nitrogen,

ground to fine powder, resuspended in 6 M guanidine hydrochloride, 0.1 mM DTT and

20 mM sodium acetate, pH 5.0 and incubated for 10 min at 0ºC. This suspension was

spun at 10,000 rpm and 4ºC for 10 min in a SS-34 rotor and nucleic acids in the

supernatant were recovered by ethanol precipitation. Nucleic acids were resuspended in

20 mM EDTA pH 8.0 and the RNA was precipitated with 3 volumes of 4 M sodium

acetate for 1 h at –20ºC and recovered after centrifugation in a refrigerated

microcentrifuge. RNA was resuspended in RNase-free water, re-precipitated with

ethanol and stored in water at –80ºC.

Probes for genomic library screens and Northern blot analysis. The screens of

the genomic library were carried out using PCR-amplified DNA fragments obtained

from genomic DNA (mccA) or from A. nidulans ESTs n01c10 (mccB)

(http://bioinfo.okstate.edu/pipeonline/). In Northern blots, the EST n01c10 insert was

also used as mccB-specific probe, the mccA probe was a PCR fragment containing the

6


http://ww

w.jbc.org/

Dow

nloaded from

http://bioinfo.okstate.edu/pipeonline/

http://www.jbc.org/

complete mccA coding sequence and the ivdA-specific probe included its coding region

between codon 69 and the stop codon.

Genomic and cDNA libraries. The A. nidulans genomic library was a standard

λEMBL4 library constructed from a Glasgow wild-type strain. mRNA used to construct

the cDNA library was obtained from mycelia cultured in the presence of 30 mM Leu as

sole carbon source, which strongly induces transcription of Leu catabolism genes. After

oligo-dT selection of messenger RNA, double stranded cDNA was synthesized using

the SMARTTM technique (Clontech), ligated into SfiI-digested λTriplEx2 phagemid

arms and packed in vitro using Stratagene Gigapack II Plus extracts. The primary

library contained 1.5 x 106 recombinant clones, with an average insert size of 1.39 kbps.

Transformation. To construct ∆mccA and ∆mccB mutants, strains MAD003 biA1

methG1 argB2 and MAD559 yA2 pabaA1 argB2 pyrG89 riboB2 were used as recipient

strains in transformation experiments (17), respectively. Southern blot analysis with

mccA- and argB-specific probes or mccB- and pyr-4-specific probes was used to

confirm the expected deletion events in ∆mccA and ∆mccB strains, respectively.

Enzyme assays. These were carried out essentially as described (18), by

determining the amount of [14C]-HCO3- incorporated by protein extracts into acid

soluble material in a 3-methylcrotonyl-CoA- and ATP-dependent manner. The reaction

mixture contained, in a final volume of 0.2 ml, 0.1M Tris-HCl pH 8.0, 0.1 M KCl, 6

mM MgCl2, 1.5 mM DTT, 1 mM ATP and 6 mM [14C]-HCO3- (16 µCi/µmol).

Reactions were started by addition of 300 µM 3-methylcrotonyl-CoA. After a 7 min

incubation at 30ºC, they were stopped by addition of 0.1 ml of HCl 6N. Samples were

microcentrifuged for 15 min at 14,000 rpm and aliquots of the supernatant were spotted

on Whatmann 3MM paper, dried and counted in a liquid scintillation counter.

7


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

GC/MS analysis of culture supernatants. Fungal mycelia were pre-cultured as

described for RNA isolation and transferred to minimal medium with one of the

following carbon sources: 0.05% w/v lactose; 3 mM Leu; 10 mM Ile; 100 mM sodium

acetate; 30 mM 3-hydroxy-3-methylglutaric acid or 30 mM DL-mevalonate. Secondary

cultures were incubated for a further 21 h. Culture supernatants were collected after

removing mycelia by filtration. After addition of 0.1 mM (final concentration)

phenylacetate (which was used as an internal control for extraction of organic acids),

culture supernatants were acidified to pH <2 with 6N HCl in the presence of saturating

NaCl and extracted with isobutylacetate. The organic phase was dried under nitrogen

and derivatized with bis(trimethylsilyl)trifluoroacetamide. TMS derivatives were

analyzed in a 5%-phenylmethylsilicone column (30 m x 0.25 mm; 0.25 µM film

thickness), using a HP 5890 Series II gas chromatographer and a HP 5971 mass

spectrometer, with the following temperature program: after 5 min at 70ºC, temperature

was raised to 300ºC (4.5 ºC/min) and kept at this limit for a further 5 min. Identification

of peaks was carried out with the ChemStation software of Hewlett Packard.

8


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

RESULTS

Previous work showed that Leu is able to support A. nidulans vegetative growth and

conidiation when used as sole carbon source, although both growth strength and

conidiation density were notably reduced on a range of Leu concentrations as compared

to more favorable carbon sources. Supplementation of the medium with pyridoxine or

thiamine (the co-factors of BCAA transaminase and ΒCΚΑ dehydrogenase,

respectively) did not improve either to any significant extent, indicating that cofactor

biosynthesis is not limiting. Thus, unless otherwise indicated (see GC/MS analysis

below), we routinely used Leu at 30 mM for growth tests and induction of the synthesis

of Leu catabolic enzymes.

Molecular characterization of fungal genes encoding subunits of 3-methylcrotonyl-CoA

carboxylase

Fungal candidate cDNAs for MCCA and MCCB were identified by in silico

screening of the University of Oklahoma A. nidulans EST database

(http://www.genome.ou.edu/fungal.html) using the deduced Arabidopsis thaliana α-

and β−MCC and human α− and β−PCC amino acid sequences in TBLASTN searches

(3). PCR-derived DNA probes, based on the nt sequences of candidate ESTs, were

obtained and used to screen A. nidulans genomic (in λEMBL4) and cDNA (from

inducing growth conditions, see below) libraries. The genomic screen revealed, in

addition to phage clones hybridizing only to ΜCCα− or MCCβ−specific probes, a third

class of clones cross-hybridizing to both, which strongly suggested that A. nidulans

candidate genes encoding MCCα and MCCβ, that we denoted mccA and mccB,

respectively, are physically linked. Restriction enzyme mapping and nt sequencing of

9


http://ww

w.jbc.org/

Dow

nloaded from

http://www.genome.ou.edu/fungal.html

http://www.jbc.org/

an ∼15 kb A. nidulans genomic DNA region reconstructed with overlapping

bacteriophage clone inserts in this third class revealed that this is indeed the case. mccA

and mccB ORFs are divergent, with their predicted initiation Met codons separated by

3692 bp.

A cDNA library made from mRNA isolated from cells grown in the presence of

Leu as sole carbon source (a condition inducing transcription of Leu catabolic genes,

see below) was screened to isolate full length mccA and mccB cDNAs, which were used

to determine intron/exon borders after comparison of genomic and cDNA sequences.

The 2688 bp mccA gene (GenBank accession number AY387592) is split by three

introns ranging from 44 to 309 bps, whereas The 1843 bp mccB gene (GenBank

accession number AY387593) is split by a single 80 bp intron.

The mccA cDNA encodes a 763 residue MccA polypeptide that shows 40%

amino acid sequence identity to Arabidopsis and human ΜCCα. Triple (fungal, plant,

human) MCCα alignment (not shown) showed that patches of amino acid sequence

conservation predominate in the NH2-terminal two-thirds of the protein and in the ∼65

residue C-terminal domain containing the biotinylation motif, while the sequence

separating these regions (residues 480 through 640 in A. nidulans MccA) does not

contribute to sequence conservation. Near the C-terminus of MccA there is a 668-

SMKM-671 tetrapeptide biotin binding motif found in biotin-dependent carboxylases

(19), where the ε-amino group of Lys670 is predicted to bind the cofactor covalently

through an amide bond. In addition, A. nidulans MccA contains the sequence 197-

GGGGKGMR-204, which is completely conserved amongst Aspergillus, Arabidopsis

and human MCCα polypeptides and strongly conserved among biotin-dependent

carboxylases. The structure of bacterial acetyl-CoA carboxylase revealed that this

peptide makes a major contribution to ATP binding (20).

10


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

The mccB cDNA encodes a 567 residue MccB polypeptide showing 59% and

51% amino acid sequence identity to human and Arabidopsis ΜCCβ, respectively.

MccB contains a putative acyl-CoA binding motif (21, 22) within residues 354 and 392

which is strongly conserved among MCCβ subunits and less so among biotin-dependent

carboxylases, possibly reflecting the differing substrate specificities of these enzymes.

In agreement with substrate specificities residing in the β chains of heterodimeric

biotin-dependent carboxylases, MccB and PCCβ are markedly more dissimilar in

amino acid sequence than the MccA/PCCα pair (data not shown).

Transcriptional regulation of mccA and mccB

Northern analysis of mycelia pre-cultured under glucose-limiting conditions and

transferred to media with different carbon sources revealed that mccA and mccB show a

similar transcription pattern, in agreement with these genes encoding subunits of a

heterodimeric enzyme (Fig. 2). Basal transcript levels, comparable to those observed in

the absence of carbon source, were detected in the presence of acetate (a gluconeogenic

C source), but also in the presence of Thr, Glu, Pro or Trp (Fig. 2, lanes 1-4 and 9). In

marked constrast, BCAAs (Leu, Ile, Val), Phe and Met resulted in strong, similar

induction of mccA and mccB transcription (Fig. 2, lanes 5-8 and 12). Transcription was

repressed by glucose to nearly undetectable levels, even in the presence of an inducing

amino acid (Fig. 2, lanes 10 and 11). These findings agree with the predicted role of

mccA and mccB in the Leu degradation pathway.

In A. nidulans, pathway-specific transcription factors of the zinc binuclear

cluster family (23, 24) control expression of structural genes for catabolic pathways.

These factors are typically positive-acting regulatory proteins whose action absolutely

requires induction by the pathway substrate. Therefore, the above results suggest that

11


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

one such pathway-specific transcriptional activator, whose action requires induction by

Leu or certain other hydrophobic amino acids such as Met and Phe, is required for full

expression of mccA and mccB. The negative effect of glucose on transcript levels

suggests that transcription is carbon catabolite repressible. The regulatory proteins

mediating induction and repression of genes in the Leu catabolic pathway have not been

investigated further.

A three gene cluster mediating two sequential steps in the Leu degradation pathway

mccA and mccB are located within contig 80 (nt positions 69,585 to 71,752) on

the left arm of chromosome III (http://www-

genome.wi.mit.edu/annotation/fungi/aspergillus) (Fig. 3). Fig. 3 also depicts the relative

position of sC encoding ATP sulphurylase, which is the closest (at ~ 250 kb) gene to

mccA that has been functionally characterized (25). mccA and mccB are flanked by

genes which almost certainly are not involved in Leu catabolism. Centromere distal

from mccA (see Fig. 3) is arbD, a gene encoding a 290 residue protein showing high

amino acid sequence identity to Candida albicans D-arabitol dehydrogenase.

Centromere proximal from mccB and separated from it by an 822 bp intergenic region is

a gene predicted to encode a chitosanase, denoted csnA (Fig. 3). Similar csnA transcript

levels were found on glucose, Leu or a mixture of glucose and Leu (Fig. 3), as expected

for a gene that is not involved in Leu catabolism. In contrast, Northern analysis using a

probe containing the region between mccA and mccB revealed the presence of a third

gene whose transcription is Leu-inducible and glucose repressible, strongly suggesting

that this region contains another gene of the Leu catabolic pathway, that was denoted

ivdA (GenBank accession number AY387594). Sequence analysis of ivdA cDNA clones

revealed that the gene is split by three introns and that its transcript encodes a 430

12


http://ww

w.jbc.org/

Dow

nloaded from

http://www-genome.wi.mit.edu/annotation/fungi/aspergillus

http://www-genome.wi.mit.edu/annotation/fungi/aspergillus

http://www.jbc.org/

residue protein showing high amino acid sequence identity to isovaleryl-CoA

dehydrogenase (IVD) enzymes (55% and 63% to human and Pseudomonas aeruginosa

IVD, respectively). IVD (E.C. 1.3.99.10) catalyzes conversion of isovaleryl-CoA to 3-

methylcrotonyl-CoA with formation of FADH2 (Fig. 1). The divergently transcribed

mccB and ivdA genes are separated by an intergenic region of 385 bp (Fig. 3). We

conclude that this region of chromosome III contains a three gene cluster encoding two

sequentially acting enzymes of the Leu degradation pathway.

Construction and phenotypic characterization of ∆mccA and ∆mccB alleles

Null ∆mccA and ∆mccB alleles were constructed by transformation, using a gene

replacement procedure. The mccA transcription start and polyadenylation sites are

located at –31 (relative to the initiation codon) and +89 (relative to the stop codon),

respectively. To construct a null ∆mccA allele, a 5423 bp DNA fragment in which

sequences from –14 (relative to the ATG) to +3 (relative to the stop codon) had been

substituted by a 3133 bp DNA fragment containing the argB+ gene was used to

transform an argB2 arginine-requiring strain (Fig. 4a). Prototrophic transformants were

purified on minus-arginine minimal medium. Of these, ∼15% were unable to grow on

leucine as sole carbon source. All ten transformants unable to grow on leucine that were

analyzed by Southern blotting showed the hybridization pattern expected for the double

crossover event shown in Fig. 4a and were therefore carrying the ∆mccA allele. Two,

chosen for further analysis, were phenotypically indistinguishable in that they showed

wild type growth on lactose, Ile, Phe, Pro, Thr or Val but their growth was severely

reduced on Leu minimal medium as compared to the wild-type (data not shown and

Ref. (3)). Northern blot analysis of mRNA isolated from cells grown under leucine-

inducing conditions revealed that, in agreement with the complete deletion of the gene,

13


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

the ∆mccA mutation indeed results in undetectable levels of the mccA transcript, while

levels of the ivdA transcript remain essentially unaffected (Fig. 5). As expected from

Northern analysis, enzyme assays showed that the ∆mccA mutation nearly abolishes

MCC activity (Table I).

An apparently full length mccB cDNA starting 39 bp upstream from the

initiation codon that is polyadenylated 57 bp downstream from the stop codon was

characterized. To construct a null ∆mccB allele, a 5624 bps DNA fragment in which

sequences from –21 (relative to the ATG) to +62 (relative to the stop codon) had been

substituted by a ~3200 bp DNA fragment containing the Neurospora crassa pyr-4 gene

was used to transform a pyrG89 pyrimidine-requiring strain (Fig. 4b) (pyr-4 is the N.

crassa homologue of A. nidulans pyrG). Prototrophic transformants were purified on

minus-pyrimidine minimal medium and analyzed by Southern blotting and growth tests

as described above. As with ∆mccA strains, two transformants showing the expected

hybridization pattern for a mccB deletion event were phenotypically indistinguishable,

being unable to grow on Leu as sole carbon source, although they grew as the wild type

on lactose, Ile, Phe, Thr and Val (Fig. 4c). ∆mccB results in undetectable mccB

transcript levels and, like ∆mccA, it does not affect ivdA transcription (Fig. 5) and

abolishes MCC activity (Table I).

A double ∆mccA ∆mccB double mutant was constructed after crossing the

corresponding single mutants. As expected, while single mutants abolished transcription

only of the corresponding deleted gene, the double ∆mccA ∆mccB mutation abolished

transcription of both, but had no major effect on ivdA (Fig. 5). As single ∆mccA or

∆mccB mutants (see below), ∆mccA ∆mccB strains showed residual growth on Leu,

similar to that observed on agar plates lacking an added carbon source. In enzyme

assays (Table I), the double mutant showed the same nearly null levels of activity than

14


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

either of the single mutants. Therefore, ∆mccA, ∆mccB or ∆mccA ∆mccB strains are

phenotypically identical.

The inability of ∆mccA, ∆mccB or ∆mccA ∆mccB strains to use Leu as sole

carbon source demonstrates the in vivo involvement of MCC in Leu catabolism.

Fungal model for 3-methylcrotonyl-CoA carboxylase deficiency

Patients suffering from isolated MCC deficiency excrete in their urine markedly

elevated amounts of 3-hydroxy-isovaleric acid (3-HIVA) and 3-methylcrotonylglycine

(3-MCGly), which are diagnostic biochemical markers of the disease (Fig. 6a and b) (1,

26). A. nidulans can be grown on synthetic medium containing Leu as sole carbon

source. If an amino acid degradation pathway is blocked by a loss-of-function, structural

gene mutation, metabolites accumulating as a result of the enzyme deficiency exit the

cell and appear in the culture supernatant which, analogously to human urine, can be

analyzed by GC/MS (11, 13). In contrast to the wild type, ∆mccA ∆mccB cells

accumulate 3-HIVA in a leucine-dependent manner, as expected form the specific

involvement of MCC in Leu catabolism. Fig. 6e-h shows that the GC/MS profile of

supernatants from wild type and ∆mccA ∆mccB cultures is virtually indistinguishable

when cells were incubated in synthetic medium containing lactose or isoleucine as sole

carbon source. In contrast, a marked accumulation of 3-HIVA was detected when the

double mutant but not the wild type was cultured in medium containing leucine as sole

carbon source (Fig. 6c-d). This marked accumulation of 3-HIVA was also detected

using ∆mccA or ∆mccB single mutants (data not shown). Of note, 3-MCGly was

detected in neither.

The specific accumulation of 3-HIVA upon culturing MCC-deficient fungal

cells in Leu medium illustrates the remarkable similarity of the metabolic consequences

15


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

of equivalent enzyme deficiencies in human and fungi. As in humans, 3-HIVA almost

certainly results from hydration of 3-methylcrotonyl-CoA by crotonase (enoyl-CoA

hydratase) followed by spontaneous deacylation (1). Although A. nidulans enzymes

with enoyl-CoA hydratase activity have not yet been functionally characterized, the A.

nidulans genome contains an almost certain candidate gene to encode such protein

(J.M.R. and M.A.P., unpublished). In contrast, the human plasma Gly-conjugating

enzyme converting organic acids to their corresponding glycine amides would appear to

be absent from A. nidulans.

2-Hydroxy-3-methylvaleric acid (2H3MV) resulting from hydration of α-keto-β-

methylvaleric acid, the product of Ile transamination, accumulates in the culture

supernatants of both the wild-type and the MCC-deficient mutant grown in minimal

medium with 10 mM Ile as sole carbon source. This indicates that, as in humans (27),

BCKA dehydrogenase activity is limiting under these conditions. BCKA dehydrogenase

catalyzes the oxidative decarboxylation of all three BCAA transamination products. In

agreement, a marked elevation of 2-hydroxyisocaproic acid resulting from hydration of

α-ketoisocaproic acid, the Leu transamination product (Fig. 1), was seen when the Leu

concentration in fungal cultures was raised to 10 mM or above (data not shown). It is

worth noting that also in humans, α-ketoisocaproic acid decarboxylation is the rate

limiting step in Leu catabolism.

While MCC-deficient strains grow normally on lactose, ∆mccA or ∆mccB

mutations impair growth on lactose plus leucine to a significant extent, indicating that

3-HIVA and/or other metabolites accumulating in the deficient strains are toxic (Fig. 7).

As A. nidulans is aerobe, such toxicity could reflect an impairment of mitochondrial

function by these metabolites.

16


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

The mevalonate shunt

The biosynthetic cholesterol pathway proceeds via mevalonate formed by 3HMG-CoA

reductase. Metabolic tracing studies showed that not all mevalonate is converted to

sterols and strongly supported the existence of a link between the cholesterol

biosynthetic pathway and Leu catabolism (see (14, 28-30) and references therein).

Edmond and Popják (29) proposed that mevalonate is converted to acetyl-CoA via the

enzymes of the Leu catabolic pathway after its conversion to 3-methylcrotonyl-CoA by

the shunt pathway shown in Fig. 1. Another possible link is at the level of non-cyclic

isoprenoids such as geranoyl-CoA, which can be catabolized by Pseudomonas

citronellolis and plants by a series of reactions also leading to 3-methylcrotonyl-CoA

(31, 32) (Fig. 1).

Microbial or plant mutants, or human patients specifically deficient in this

pathway have not yet been reported nor have animal models been devised. In view of

the apparent similarity between fungal and human leucine degradation pathways, we

used our fungal model deficient in MCC to obtain genetic evidence for the existence of

this link, using the GC/MS procedure described above. While the metabolic profile of

the wild type culture supernatant was markedly similar to that of a ∆mccA ∆mccB strain

when transferred to medium with 0.05% (w/v) lactose as the sole carbon source (Fig.

8a-b), the mutant specifically accumulated 3-HIVA on transfer to acetate medium (Fig.

8c-d)(acetate is converted to acetyl-CoA by acetyl-CoA synthetase (33)). This finding

agrees with the existence of the above mentioned link, as among compounds directly

derived from acetyl-CoA only isoprenoid biosynthesis intermediates can be converted

to a β-methyl branched compound such as 3-methylcrotonyl-CoA. Marked

accumulation of 3-HIVA in the mutant but not in the wild type supernatant was also

observed on transfer to the mevalonate precursor 3-hydroxy-3-methylglutaric acid (Fig.

17


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

8e-f). Moreover, in spite of the apparently inefficient utilization of mevalonate by A.

nidulans, the MCC-deficient mutant but not the wild type clearly accumulated 3-HIVA

on transfer to mevalonate (a commercial preparation which is a mixture of D and L

isomers with mevalolactone (Fig.8g-h)). These results strongly support the existence of

a mevalonate shunt in A. nidulans

DISCUSSION

Our work has pioneered the use of A. nidulans knock-out strains as fungal models

for human inborn errors of metabolism ((11-13); this work). Of particular note, we

found that enzymes for A. nidulans amino acid catabolism are ∼50% identical in amino

acid sequence to their human homologues, a finding which has been crucial for the

identification of as yet uncharacterised human amino acid catabolism genes by in silico

searches of human databases (see Ref. (34) for a review). This approach was

successfully exploited in the characterisation of the human genes for alkaptonuria (12,

13 , 35 , 36), maleylacetoacetate isomerase deficiency (13) and 3-methylcrotonyl-CoA

carboxylase deficiency (3). Alkaptonuria, Garrod’s prototypic inborn error of

metabolism and the first human disease to be recognized as a Mendelian trait (37, 38), is

a notable example of a gene that was first characterised using our fungal model, despite

the facts that the human gene had previously been genetically mapped (39) and a mouse

model was available (40).

We report here the genetic and biochemical characterization of fungal models

for 3-methylcrotonyl-CoA carboxylase deficiency. Fungal cell models for inborn errors

of Leu metabolism are particularly advantageous over human cell models for two

reasons. First, Leu metabolism involves extensive interplay of metabolites between

muscle and liver as a result of non-uniform distribution of BCAA transaminase (which

18


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

is high in muscle and low in liver) and BCKA dehydrogenase (which is low in muscle

and high in liver) (2), complicating the use of a single human cell type. Second, Leu,

which is catabolized to acetyl-CoA and acetoacetate, is a ketogenic amino acid in

humans/mammals. In contrast, A. nidulans can convert acetyl-CoA into carbohydrates

through the glyoxylate bypass of the Krebs cycle (41-44). Therefore the fungus can use

Leu as sole carbon source and simple growth tests can be used to determine whether

mutations impair the catabolic pathway (see Fig. 7). Results reported here together with

data in Ref. (3) show that null mutations in either of the genes encoding MCC subunits

prevent growth on Leu, but have no effect on utilization other amino acids, formally

demonstrating that MCC is specifically involved in Leu catabolism in vivo.

As expression of the metabolic defect in Aspergillus mutants is dependent upon

the presence of the pathway substrate, cells can be pre-cultured using carbohydrates and

challenged by the pathway substrate simply by transferring mycelia to fresh medium.

Metabolites accumulating as a result of the metabolic block leak into the medium and

can be analysed by HPLC or GC/MS. These studies revealed that fungal amino acid

catabolism models faithfully resemble most aspects of the biochemical phenotype in

human patients. For example, the fungal alkaptonuric model accumulated the ochronotic

pigment (12) and the fungal model for type 1 tyrosinemia accumulated succinylacetone,

the diagnostic compound of the disease in the urine of human patients (11).

When cultured in the presence of Leu, MCC-deficient strains accumulate 3-

HIVA in the culture supernatant, as determined by GC/MS. 3-HIVA is one of the

diagnostic compounds in the urine of MCC-deficient patients, strongly supporting the

validity of the metabolic model. In the presence of lactose as sole carbon source,

addition of Leu impairs growth of MCC-deficient strains to a significant extent, strongly

indicating that one or more of the metabolites accumulating in the mutants is/are toxic

19


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

to the cells. One possibility is that high levels of 3-HIVA accumulating in the

mitochondria affect its normal functioning. A. nidulans is aerobe and therefore its

growth rate is highly dependent on normal mitochondrial function

Accumulation of 3-HIVA in the culture supernatant of the MCC deficient strains

is indicative of metabolite flow through the Leu degradation pathway. Growth of these

strains on 0.05% lactose results in a minor yet detectable accumulation of 3-HIVA,

possibly reflecting that such growth conditions result in carbon source limitation that

triggers catabolism of amino acids resulting from protein turnover. The levels of 3-

HIVA detected upon transfer of a MCC deficient strain to Ile were similar to those on

lactose, strongly indicating that Ile catabolism does not produce shunt metabolites that

can be converted into 3-methylcrotonyl-CoA.

In humans, as much as 16.5% of mevalonate is diverted away from sterol

biosynthesis (45). Both the mevalonate shunt proposed by Edmond and Popják (29) and

the geranoyl-CoA catabolic proposed in bacteria (31) and plants (32) involve the three

final steps of Leu catabolism in the conversion of isopentenyl pyrophosphate and

geranoyl-CoA, respectively, into acetoacetate and acetyl-CoA. Therefore, if either or

both were operative in Aspergillus, MCC-deficient strains should predictably

accumulate 3-HIVA on exposure to isoprenoid precursors. 3HMG-CoA is the

immediate precursor of mevalonate, the building block of isoprenoids. The high levels

of 3-HIVA found in culture supernatants of the double mcc null mutant upon transfer to

a medium containing 3HMG strongly indicate that a fraction of the 3HMG pool is

converted into metabolites of the lower Leu catabolic pathway through isopentenyl

pyrophosphate or linear isoprenoids. The finding that 3-HIVA was also detected in the

mutant culture supernatant upon transfer to mevalonate strongly supports this

interpretation .

20


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

A link between sterol biosynthesis and Leu catabolism is also suggested by

certain biochemical findings detected in patients with Smith-Lemli-Opitz syndrome

(SLOS, OMIM 270400) (46) who, as a result of a deficiency of a cholesterol

biosynthetic enzyme, are likely to accumulate high levels of cholesterol precursors.

Patients with the lowest cholesterol levels show abnormally increased plasma levels of

3-methylglutaconic acid, as would be expected if forced flow through the

mevalonate/isoprenoid shunt strains 3-methylglutaconate hydratase, the enzyme

downstream of MCC (47). Our work provides the first reported genetic evidence for the

mevalonate/isoprenoid shunt, whose precise enzymatic basis remains to be determined.

MCC is a member of the biotin-dependent carboxylase family of enzymes. In

Escherichia coli ACC, the biotin carboxylating, biotin carrier and carboxyltransferase

domains reside in three different polypeptide chains, in contrast to fungal and

mammalian ACCs, where all three activities reside in a single polypeptide. MCC and

PCC display an intermediate organization. They are α/β heteromeric complexes where

α subunits contain an N-terminal biotin carboxylating domain and a smaller, C-terminal

biotin carboxyl carrier domain separated by a relatively long non-conserved arm. The

carboxyltransferase domain is located in the β subunits where substrate specificity

resides. The high level of identity between the catalytic domains of PCCα and ΜCCα

would suggest the possibility that PCCα could at least partially substitute for MCCα in

vivo. We show that ∆mccA, ∆mccB and ∆mccA ∆mccB strains are phenotypically

indistinguishable in their inability to grow on Leu and their nearly undetectable in vitro

MCC activity. These findings rule out the above possibility and suggest that the

interaction between α modules and their corresponding β subunits is specific. The

structures of the biotin-carboxylating (5, 20) and biotin carrier domains (48) of E. coli

ACC, the carboxyltransferase domain of yeast ACC (49) and the Propionibacterium

21


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

shermanii transcarboxylase (50) have been determined. However, the structural

determinants of the specific α/β interaction in PCC and MCC and the basis for substrate

discrimination by their β subunits remains to be described.

ACKNOWLEDGEMENTS Thanks are due to the DGICYT and the FIS for their financial support through grants

2FD97-1292 and G03/054, respectively, to E. Reoyo for technical assistance, To E.

Cerdá-Olmedo for discussion, to H.N. Arst for his critical reading of the manuscript and

to members of the Aspergillus Sequencing Project at the Whitehead Institute/MIT

Center for Genome Research (http://www-genome.wi.mit.edu) for the A. nidulans

genomic sequence

22


http://ww

w.jbc.org/

Dow

nloaded from

http://www-genome.wi.mit.edu/

http://www.jbc.org/

Reference List

1. Sweetman L. and Williams J.C. (1994) Branched chain organic acidurias. In

Scriver, C. R., Beaudet, A. L., Sly, W., and Valle, D., editors. The Metabolic

Basis of Inherited Disease, MacGraw Hill, New York

2. Chuang DT and Shih VE (1994) Disorders of branched chain amino acid and

ketoacid metabolism. In Scriver, C. R., Beaudet, A. L., Sly, W., and Valle, D.,

editors. The Metabolic Basis of Inherited Disease, MacGraw Hill, New York

3. Gallardo, M. E., Desviat, L. R., Rodríguez, J. M., Esparza-Gordillo, J., Pérez-

Cerdá, C., Pérez, B., Rodríguez-Pombo, P., Criado, O., Sanz, R., Morton, D. H.,

Gibson, K. M., Le, T. P., Ribes, A., Rodríguez de Córdoba, S., Ugarte, M., and

Peñalva, M. A. (2001) Am.J. Hum.Genet. 68, 334-346

4. Baumgartner, M. R., Almashanu, S., Suormala, T., Obie, C., Cole, R. N.,

Packman, S., Baumgartner, E. R., and Valle, D. (2001) J.Clin.Invest. 107, 495-

504

5. Waldrop, G. L., Rayment, I., and Holden, H. M. (1994) Biochemistry 33, 10249-

10256

6. Lau, E. P., Cochran, B. C., and Fall, R. R. (1980) Arch.Biochem.Biophys. 205,

352-359

7. McKean, A. L., Ke, J., Song, J., Che, P., Achenbach, S., Nikolau, B. J., and

Wurtele, E. S. (2000) J.Biol.Chem. 275, 5582-5590

23


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

8. Song, J., Wurtele, E. S., and Nikolau, B. J. (1994) Proc.Natl.Acad.Sci.U.S.A. 91,

5779-5783

9. Holzinger, A., Roschinger, W., Lagler, F., Mayerhofer, P. U., Lichtner, P.,

Kattenfeld, T., Thuy, L. P., Nyhan, W. L., Koch, H. G., Muntau, A. C., and

Roscher, A. A. (2001) Hum.Mol.Genet. 10, 1299-1306

10. Hull, E. P., Green, P. M., Arst, H. N., Jr., and Scazzocchio, C. (1989)

Mol.Microbiol. 3, 553-559

11. Fernández-Cañón, J. M. and Peñalva, M. A. (1995) Proc.Natl.Acad.Sci.USA 92,

9132-9136

12. Fernández-Cañón, J. M. and Peñalva, M. A. (1995) J.Biol.Chem. 270, 21199-

21205

13. Fernández-Cañón, J. M. and Peñalva, M. A. (1998) J.Biol.Chem. 273, 329-337

14. Landau, B. R. and Brunengraber, H. (1985) Methods Enzymol. 110, 100-114

15. Clutterbuck, A. J. (1993) Aspergillus nidulans. In O'Brien, S. J., editor. Genetic

maps. Locus maps of complex genomes, Cold Spring Harbor Laboratory Press,

Cold Spring Harbor

16. Cove, D. J. (1966) Biochim.Biophys.Acta 113, 51-56

17. Tilburn, J., Scazzocchio, C., Taylor, G. G., Zabicky-Zissman, J. H., Lockington,

R. A., and Davies, R. W. (1983) Gene 26, 205-211

18. Suormala, T., Wick, H., Bonjour, J. P., and Baumgartner, E. R. (1985)

Clin.Chim.Acta 145, 151-162

24


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

19. León del Río, A. and Gravel, R. A. (1994) J.Biol.Chem. 269, 22964-22968

20. Thoden, J. B., Blanchard, C. Z., Holden, H. M., and Waldrop, G. L. (2000)

J.Biol.Chem. 275, 16183-16190

21. López-Casillas, F., Bai, D. H., Luo, X. C., Kong, I. S., Hermodson, M. A., and

Kim, K. H. (1988) Proc.Natl.Acad.Sci.U.S.A 85, 5784-5788

22. Ha, J., Daniel, S., Kong, I. S., Park, C. K., Tae, H. J., and Kim, K. H. (1994)

Eur.J.Biochem. 219, 297-306

23. Todd, R. B. and Andrianopoulos, A. (1997) Fungal.Genet.Biol. 21, 388-405

24. Davis, M. A. and Hynes, M. J. (1989) Trends.Genet. 5, 14-19

25. Borges-Walmsley, M. I., Turner, G., Bailey, A. M., Brown, J., Lehmbeck, J., and

Clausen, I. G. (1995) Mol.Gen.Genet. 247, 423-429

26. Gibson, K. M., Bennett, M. J., Naylor, E. W., and Morton, D. H. (1998) J.Pediatr.

132, 519-523

27. Matthews, D. E., Bier, D. M., Rennie, M. J., Edwards, R. H., Halliday, D.,

Millward, D. J., and Clugston, G. A. (1981) Science 214, 1129-1131

28. Fogelman, A. M., Edmond, J., and Popják, G. (1975) J.Biol.Chem. 250, 1771-

1775

29. Edmond, J. and Popják, G. (1974) J.Biol.Chem. 249, 66-71

30. Weinstock, S. B., Kopito, R. R., Endemann, G., Tomera, J. F., Marinier, E.,

Murray, D. M., and Brunengraber, H. (1984) J.Biol.Chem. 259, 8939-8944

25


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

31. Seubert, W. and Remberger, U. (1963) Biochemische Zeitschrift 338, 245-264

32. Guan, X., Diez, T., Prasad, T. K., Nikolau, B. J., and Wurtele, E. S. (1999)

Arch.Biochem.Biophys. 362, 12-21

33. Connerton, I. F., Fincham, J. R. S., Sandeman, R. A., and Hynes, M. J. (1990)

Mol.Microbiol. 4, 451-460

34. Peñalva, M. A. (2001) Fungal.Genet.Biol. 34, 1-10

35. Fernández-Cañón, J. M., Granadino, B., Beltán-Valero de Bernabé , D., Renedo,

M., Fernández-Ruiz, E., Peñalva, M. A., and Rodríguez de Córdoba, S. (1996)

Nat.Genet. 14, 19-24

36. Griffiths, A. J. F., Miller, J. H., Suzuki, D. T., Lewontin, R. C., and Gelbart, W.

M. (2003) Introduction to Genetic Analysis, 7th Ed., W. H. Freeman & Co., New

York

37. Garrod, A. E. (1902) Lancet 2, 1616-1620

38. Garrod, A. E. (1908) Lancet 2, 73-79

39. Pollak, M. R., Chou, Y. W., Cerda, J. J., Steinmann, B., La Du, B. N., Seidman, J.

G., and Seidman, C. E. (1993) Nat.Genet. 5, 201-204

40. Montagutelli, X., Lalouette, A., Coudé, M., Kamoun, P., Forest, M., and Guénet,

J. L. (1994) Genomics 19, 9-11

41. Kornberg, H. L. (1966) Biochem.J. 99, 1-11

42. Armitt, S., McCullough, W., and Roberts, C. F. (1976) J.Gen.Microbiol. 92, 263-

282

26


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

43. Armitt, S., Roberts, C. F., and Kornberg, H. L. (1970) FEBS Lett. 7, 231-234

44. Armitt, S., Roberts, C. F., and Kornberg, H. L. (1971) FEBS Lett. 12, 276-278

45. Feingold, K. R., Wiley, M. H., Searle, G. L., Machida, B. K., and Siperstein, M.

D. (1980) J.Clin.Invest. 66, 361-366

46. Pappu, A. S., Steiner, R. D., Connor, S. L., Flavell, D. P., Lin, D. S., Hatcher, L.,

Illingworth, D. R., and Connor, W. E. (2002) J.Lipid Res. 43, 1661-1669

47. Kelley, R. I. and Kratz, L. (1995) Pediatr.Res. 37, 671-674

48. Athappilly, F. K. and Hendrickson, W. A. (1995) Structure 3, 1407-1419

49. Zhang, H., Yang, Z., Shen, Y., and Tong, L. (2003) Science 299, 2064-2067

50. Hall, P. R., Wang, Y. F., Rivera-Hainaj, R. E., Zheng, X., Pustai-Carey, M.,

Carey, P. R., and Yee, V. C. (2003) EMBO J. 22, 2334-2347

27


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

FOOTNOTES The abbreviations used are:

ACC: acetyl-CoA carboxylase

bp(s): base pair(s)

BCAA: branched chain amino acid

BCKA: branched chain ketoacid

EST: expressed sequence tag

GC: gas chromathography

3-HIVA: 3-hydroxyisovaleric acid

3HMG: 3-hydroxy-3-methylglutaric acid

2H3MV: 2-hydroxy-3-methylvaleric acid

3-MCGly: 3-methylcrotonylglycine

IVD: isovaleryl-CoA dehydrogenase

MCC: 3-methylcrotonyl-CoA carboxylase

MS: mass spectrometry

nt: nucleotide

PCC: propionyl-CoA carboxylase

PC: pyruvate carboxylase

SLOS: Smith_Lemli-Opitz syndrome

28


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

FIGURE LEGENDS

FIG. 1. The leucine degradation pathway and the mevalonate shunt Enzymes for

Leu catabolism in Aspergillus and humans, with other possible pathways involving 3-

methylcrotonyl-CoA carboxylase. Enzyme deficiencies in Leu catabolism cause the

following diseases: I, hypervalinemia/ hyperleucine-isoleucinemia; II, Maple syrup

urine disease; III, isovaleric acidemia; IV, isolated 3-methylcrotonyl-CoA carboxylase

deficiency, methylcrotonylglycinuria; V, methylglutaconic aciduria, type 1; VI, 3-

hydroxy-3-methylglutaric aciduria. The isoprenoid biosynthetic pathway is schematized

on the right, with indication of proposed intermediates of the mevalonate shunt (see

text). Geranoyl-CoA degradation could be another possible link between isoprenoid

metabolism and the Leu degradation pathway.

FIG. 2. Northern analysis of mccA and mccB Mycelia were grown in minimal

medium with 0.3% (w/v) glucose as sole carbon source for 16 h at 37ºC and transferred

to media with the indicated carbon sources, where amino acids were used at 30 mM,

glucose (indicated as Gluc) at 1% (w/v) and acetate (indicated as AcH) at 100 mM. –C

indicates medium with no carbon source added. After a 3 h additional incubation of

these secondary cultures, mycelia were harvested and used to isolate RNA, that was

analyzed by Northern blotting, using mccA- or mccB-specific probes, as indicated. Actin

was used as a loading control.

FIG. 3. A three gene cluster mediating two sequential steps in the Leu degradation

pathway A diagram of chromosome III, with the position of the centromere (CEN) and

the most telomeric genetic markers on either arm, is shown, indicating the approximate

position of contig 1.80 including the three gene cluster that contains the mccA, ivdA and

29


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

mccB genes (hatched). Panels show Northern analyses of mRNA from repressed,

induced and induced-repressed conditions (as in Fig. 2) using the indicated probes.

Equal loading was confirmed by hybridization with an actin-specific probe, as in Fig. 2.

FIG. 4. Constructing null mccA and mccB alleles (A) Strategy for deleting mccA . (B)

Strategy for deleting mccB. For (A) and (B), EcoRI sites are indicated by E. The relative

position of genes within the cluster should be consulted in Fig. 3. Detailed coordinates

are given in the text. (C) Growth phenotype of a ∆mccB strain on the indicated carbon

sources, compared to the wild type. Valine is a very poor carbon source. The phenotype

of the ∆mccA mutant has been published elsewhere (3), and is identical to that of the

∆mccB mutant shown here.

FIG. 5. Northern analysis of mutant mcc strains Northern blot analysis of RNA

isolated from strains with the indicated relevant genotype, using probes specific for the

genes indicated on the right of each panel. Mycelia were cultured under Leu-inducing

conditions as in Fig. 2. Approximately equal amounts of total RNA were loaded on each

track, as determined by ethidium bromide staining of rRNAs (not shown).

FIG. 6. GC/MS analysis of culture supernatants of MCC-deficient strains Panels

(A) and (B) show GC/MS analyses of urinary organic acids from a normal individual or

from a patient suffering from isolated 3-methylcrotonyl-CoA carboxylase deficiency. In

panels (C-H) , A. nidulans wild-type or mutant ∆mccA ∆mccB strains were pre-

cultured as for Northern analysis and transferred to fresh media with the indicated

carbon sources, where they were incubated for a further 21 h before analyzing organic

acids extracted from culture supernatants by GC/MS. Lactose was used at 0.05% (w/v),

30


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Leu at 3 mM and Ile at 10 mM (Ile is a poorer carbon source than Leu). Phenylacetic

acid (PhAc) was added exogenously to culture supernatants and served as an internal

control for the extraction procedure. Ethylvanilline (EV) was used as control for

derivatization. 2HIC is 2-hydroxyisocaproic acid; 3-HIVA is 3-hydroxyisovaleric acid

(the marker of the MCC deficiency); MC is methylcrotonic acid; 2H3MV is 2-hydroxy-

3-methylvaleric acid; 3-MCGly is 3-methylcrotonylglycine.

FIG. 7. Leucine toxicity to MCC-deficient strains Wild-type and mutant strains were

inoculated onto minimal medium plates containing the indicated carbon sources and

incubated at 37ºC for 5 days.

FIG. 8. Mevalonate shunting requires 3-methylcrotonyl-CoA carboxylase A.

nidulans wild-type or mutant ∆mccA ∆mccB strains were pre-cultured as in Fig. 6

before analyzing organic acids extracted from culture supernatants by GC/MS. Carbon

compounds were used as follows: lactose at 0.05% (w/v), acetate at 100 mM, 3-

hydroxy-3-methylglutarate and mevalonate at 30 mM. Phenylacetic acid (PhAc) was

added exogenously to culture supernatants and served as an internal control for the

extraction procedure. Ethylvanilline (EV) was used as control for derivatization. Their

peaks are not visible in the mevalonate conditions, as for panels (G) and (H) we injected

1/40 of the material used for panels (A-F), due to the fact that mevalonate appears to be

inefficiently taken up by A. nidulans which results in column saturation. 3H3MV is 3-

hydroxy-3-methylvaleric acid, almost certainly proceeding from mevalonate. 3HMG is

3-hydroxy-3-methylglutaric acid. 3-HIVA is 3-hydroxyisovaleric acid. 2-HIVA is 2-

hydroxyisovaleric acid. 5 and 6 are succinic and fumaric acids, respectively. MVA is

DL-mevalonic acid and MVL is mevalolactone.

31


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

TABLE I

Absence of MCC activity in mccA or mccB extracts

Protein extract Specific activityb % control Human fibroblasts 1.2 ± 0.3 Mouse liver 13.8 ± 1.2 A. nidulans wild typea 25.03 ± 2.9 100 A. nidulans ∆mccA 0.15 ± 0.03 0.59 A. nidulans ∆mccB 0.12 ± 0.02 0.48 A. nidulans ∆mccA ∆mccB 0.16 ± 0.04 0.63

a A. nidulans strains were grown under inducing conditions b Specific activities are in nmoles of [14C]-HCO3

-/min x mg protein

32


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

mevalonate

isopentenylpyrophosphate

3,3-dimethylallylpyrophosphate

3-methylcrotonaldehyde

3-methylcrotonicacid

3-methylglutaconyl-CoA

3-hydroxy-3-methylglutaryl-CoA

L-leucine

α-ketoisocaproic acid

isovaleryl-CoA

3-methylcrotonyl-CoA

acetyl-CoA+

acetoacetate

‘Mevalonateshunt’

3,3-dimethylallylalcohol

Geranoyl-CoA

I

II

III

IV

V

VI

Branched chain amino acid

transaminase

Branched chain ketoacid

dehydrogenase

Isovaleryl-CoAdehydrogenase

3-methylcrotonyl-CoAcarboxylase

3-methylglutaconyl-CoAhydratase

3-hydroxy-3-methylglutaryl-CoAlyase

isoprenoids

3-hydroxy-3-methylglutaryl-CoA

acetyl-CoA

Rodríguez et al., Figure 1


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Thr

Glu

Pro

Trp

Phe

IleVal

AcH

Met

Glu

cG

luc+

Leu

-CLeu

Thr

Glu

Pro

Trp

Phe

IleVal

AcH

Met

Glu

cG

luc+

Leu

-CLeu

1 2 3 4 5 6 7 8 9 10 11 12 13

1 2 3 4 5 6 7 8 9 10 11 12 13

mccA

actin

mccB

actin



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

galE areACENChromosome III

sC

Contig 1.80/ 335,440 bpsC

mccA ivdA csnAarbD mccB

EcoRI

Glu

Leu

Glu

+Leu

Glu

Leu

Glu

+Leu

Glu

Leu

Glu

+Leu

Glu

Leu

Glu

+Leu

mccA

mccB csnA

probes

2 kb

mccA

ivdA

Rodíguez et al., Figure 3

by guest on April 19, 2018 http://www.jbc.org/ Downloaded from

http://www.jbc.org/

w/o carbon

lactose

Ile

Leu

Phe

Thr

Val

∆mccB wild-typeargB+

argB+

E E E E

mccA

ivdA

E E E E

ivdA

E

argB+

argB+argB+

E E E E

mccA

ivdA

E E E E

ivdA

E

pyr-4

ivdA

csnA

E E E

E

pyr-4

E E

mccB

ivdA

csnA

pyr-4

ivdA

csnA

E E E

E

pyr-4

E E

pyr-4

E E

mccB

ivdA

csnA

A

B

CarbD

arbD

Rodríguez et al. , Figure 4


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

wild

type

∆mcc

A

∆mcc

A ∆m

ccB

∆mcc

B

mccA

mccB

ivdA

wild

type

∆mcc

A

∆mcc

A ∆m

ccB

∆mcc

B

mccA

mccB

ivdA



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

14.00 18.00 22.00 26.00 30.00 34.000

2e+06

4e+06

6e+06

Time-->

Abundance

citrate

normalA

14.00 18.00 22.00 26.00 30.00 34.000

2e+06

4e+06

6e+06

Time-->

Abundance

citrate

3-HIVA 3-MCGly patientB

14.00 18.00 22.00 26.00 30.00 34.000

5e+06

1e+07

1.5e+07

2e+07

Time-->

Abundance

EV

PhAc

2HIC

wild typeleucineC

14.00 18.00 22.00 26.00 30.00 34.000

5e+06

1e+07

1.5e+07

2e+07

2.5e+07

Time-->

Abundance

12.00

2HICMC

3-HIVA

PhAcEV

∆mccA ∆mccBleucine D

14.00 18.00 22.00 26.00 30.00 34.000

1.5e+06

3e+06

4.5e+06

Time-->

Abundance

PhAcEV2H3MV

fumarate

wild typeisoleucineE F

14.00 18.00 22.00 26.00 30.00 34.000

1.5e+06

3e+06

4.5e+06

6e+06

Time-->

Abundance

PhAcEV

2H3MV

fumarate3-HIVA

∆mccA ∆mccBisoleucine

Abundance

14.00 18.00 22.00 26.00 30.00 34.000

1.5e+06

3e+06

4.5e+06

6e+06

Time-->

PhAcEV

3-HIVA

∆mccA ∆mccBlactose

14.00 18.00 22.00 26.00 30.00 34.000

1.5e+06

3e+06

4.5e+06

Time-->

Abundance EVPhAc wild typelactose

G H



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

∆mccA∆mccB∆mccA ∆mccBwt

Lactose

Leucine

Lactose &Leucine



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

14.00 18.00 22.00 26.00 30.00 34.000

1.5e+06

3e+06

4.5e+06

Time-->

Abundance PhAc EV wild typelactoseA Abundance

14.00 18.00 22.00 26.00 30.00 34.000

1.5e+06

3e+06

4.5e+06

6e+06

Time-->

EV

3-HIVA

PhAc ∆mccA ∆mccBlactoseB

14.00 18.00 22.00 26.00 30.00 34.000

8e+06

1.6e+07

2.2e+07

Time-->

Abundance

2-HIVA

2H3MV

PhAc

EV

wild typeacetateC

14.00 18.00 22.00 26.00 30.00 34.000

8e+06

1.6e+07

2.2e+07

Time-->

Abundance 3-HIVA

EV

M

PhAc2-HIVA

∆mccA ∆mccBacetateD

14.00 18.00 22.00 26.00 30.00 34.000

1e+07

2e+07

3e+07

Time-->

Abundance

3HMG

M

3-HIVAPhAc ∆mccA ∆mccB

3HMG

F

14.00 18.00 22.00 26.00 30.00 34.000

1e+07

2e+07

3e+07

Time-->

Abundance

PhAc

3HMG wild type3HMGE

14.00 18.00 22.00 26.00 30.00 34.000

60000

120000

180000

240000

300000

Time-->

Abundance MVL MVA

3H3MV

wild typemevalonateG

14.00 18.00 22.00 26.00 30.00 34.000

60000

120000

180000

240000

300000

Time-->

Abundance MVL MVA

3H3MV

3-HIVA

∆mccA ∆mccBmevalonate

H



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Jose M. Rodríguez, Pedro Ruiz-Sala, Magdalena Ugarte and Miguel A. PeñalvaFungal metabolic model for 3-methylcrotonyl-CoA carboxylase deficiency

published online November 11, 2003J. Biol. Chem.

10.1074/jbc.M310055200Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M310055200

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;M310055200v1&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2003/11/11/jbc.M310055200.citation

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=early/2003/11/11/jbc&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2003/11/11/jbc.M310055200.citation

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/

fungal metabolic model for 3-methylcrotonyl-coa ... · pdf filedehydrogenase complex lead to...

Documents