Salmonella enterica requires ApbC function for growth on tricarballylate: evidence of functional redundancy between ApbC and IscU.
ABSTRACT Mutants of Salmonella enterica lacking apbC have nutritional and biochemical properties indicative of defects in [Fe-S] cluster metabolism. Here we show that apbC is required for S. enterica to use tricarballylate as a carbon and energy source. Tricarballylate catabolism requires three gene products, TcuA, TcuB, and TcuC. Of relevance to this work is the TcuB protein, which has two [4Fe-4S] clusters required for function, making it a logical target for the apbC effect. TcuB activity was 100-fold lower in an apbC mutant than in the isogenic apbC(+) strain. Genetic data show that derepression of the iscRSUA-hscAB-fdx-orf3 operon or overexpression of iscU from a plasmid compensates for the lack of ApbC during growth on tricarballylate. The studies described herein provide evidence that the scaffold protein IscU has a functional overlap with ApbC and that ApbC function is involved in the synthesis of active TcuB.
- SourceAvailable from: cshlp.orgPCR methods and applications 11/1993; 3(2):85-94.
- [show abstract] [hide abstract]
ABSTRACT: An IscA homologue within the nif regulon of Azotobacter vinelandii, designated (Nif)IscA, was expressed in Escherichia coli and purified to homogeneity. Purified (Nif)IscA was found to be a homodimer of 11-kDa subunits that contained no metal centers or other prosthetic groups in its as-isolated form. Possible roles for (Nif)IscA in Fe-S cluster biosynthesis were assessed by investigating the ability to bind iron and to assemble Fe-S clusters in a NifS-directed process, as monitored by the combination of UV-vis absorption, Mössbauer, resonance Raman, variable-temperature magnetic circular dichroism, and EPR spectroscopies. Although (Nif)IscA was found to bind ferrous ion in a tetrahedral, predominantly cysteinyl-ligated coordination environment, the low-binding affinity argues against a specific role as a metallochaperone for the delivery of ferrous ion to other Fe-S cluster assembly proteins. Rather, a role for (Nif)IscA as an alternate scaffold protein for Fe-S cluster biosynthesis is proposed, based on the NifS-directed assembly of approximately one labile [4Fe-4S](2+) cluster per (Nif)IscA homodimer, via a transient [2Fe-2S](2+) cluster intermediate. The cluster assembly process was monitored temporally using UV-vis absorption and Mössbauer spectroscopy, and the intermediate [2Fe-2S](2+)-containing species was additionally characterized by resonance Raman spectroscopy. The Mössbauer and resonance Raman properties of the [2Fe-2S](2+) center are consistent with complete cysteinyl ligation. The presence of three conserved cysteine residues in all IscA proteins and the observed cluster stoichiometry of approximately one [2Fe-2S](2+) or one [4Fe-4S](2+) per homodimer suggest that both cluster types are subunit bridging. In addition, (Nif)IscA was shown to couple delivery of iron and sulfur by using ferrous ion to reduce sulfane sulfur. The ability of Fe-S scaffold proteins to couple the delivery of these two toxic and reactive Fe-S cluster precursors is likely to be important for minimizing the cellular concentrations of free ferrous and sulfide ions. On the basis of the spectroscopic and analytical results, mechanistic schemes for NifS-directed cluster assembly on (Nif)IscA are proposed. It is proposed that the IscA family of proteins provide alternative scaffolds to the NifU and IscU proteins for mediating nif-specific and general Fe-S cluster assembly.Biochemistry 12/2001; 40(46):14069-80. · 3.38 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Erwinia chrysanthemi causes soft-rot disease in a great variety of plants. In addition to the depolymerizing activity of plant cell wall-degrading enzymes, iron acquisition and resistance to oxidative stress contribute greatly to the virulence of this pathogen. Here, we studied the pin10 locus originally thought to encode new virulence factors. The sequence analysis revealed six open reading frames that were homologous to the Escherichia coli sufA, sufB, sufC, sufD, sufS and sufE genes. Sequence similarity searching predicted that (i) SufA, SufB, SufD, SufS and SufE proteins are involved in iron metabolism and possibly in Fe-S cluster assembly; and (ii) SufC is an ATPase of an ABC transporter. The reverse transcription-polymerase chain reaction procedure showed that the sufABCDSE genes constitute an operon. Expression of a sufB:uidA fusion was found to be induced in iron-deficient growth conditions and to be repressed by the iron-sensing Fur repressor. Each of the six suf genes was inactivated by the insertion of a cassette generating a non-polar mutation. The intracellular iron level in the sufA, sufB, sufC, sufS and sufE mutants was higher than in the wild type, as assessed by increased sensitivity to the iron-activated antibiotic streptonigrin. In addition, inactivation of sufC and sufD led to increased sensitivity to paraquat. Virulence tests showed that sufA and sufC mutants exhibited reduced ability to cause maceration of chicory leaves, whereas a functional sufC gene was necessary for the bacteria to cause systemic invasion of Saintpaulia ionantha. The E. coli sufC homologue was inactivated by reverse genetic. This mutation was found to modify the soxR-dependent induction of soxS gene expression. We discuss the possibility that SufC is a versatile ATPase that can associate either with the other Suf proteins to form a Fe-S cluster-assembling machinery or with membrane proteins encoded elsewhere in the chromosome to form an Fe-S ABC exporter. Overall, these results stress the importance of the connection between iron metabolism and oxidative stress during the early steps of infection by E. chrysanthemi.Molecular Microbiology 03/2001; 39(4):960-72. · 4.96 Impact Factor
JOURNAL OF BACTERIOLOGY, July 2008, p. 4596–4602
Copyright © 2008, American Society for Microbiology. All Rights Reserved.
Vol. 190, No. 13
Salmonella enterica Requires ApbC Function for Growth on
Tricarballylate: Evidence of Functional Redundancy
between ApbC and IscU?
Jeffrey M. Boyd, Jeffrey A. Lewis, Jorge C. Escalante-Semerena, and Diana M. Downs*
Department of Bacteriology, University of Wisconsin, Madison, Madison, Wisconsin 53706
Received 20 February 2008/Accepted 13 April 2008
Mutants of Salmonella enterica lacking apbC have nutritional and biochemical properties indicative of defects
in [Fe-S] cluster metabolism. Here we show that apbC is required for S. enterica to use tricarballylate as a
carbon and energy source. Tricarballylate catabolism requires three gene products, TcuA, TcuB, and TcuC. Of
relevance to this work is the TcuB protein, which has two [4Fe-4S] clusters required for function, making it
a logical target for the apbC effect. TcuB activity was 100-fold lower in an apbC mutant than in the isogenic
apbC?strain. Genetic data show that derepression of the iscRSUA-hscAB-fdx-orf3 operon or overexpression of
iscU from a plasmid compensates for the lack of ApbC during growth on tricarballylate. The studies described
herein provide evidence that the scaffold protein IscU has a functional overlap with ApbC and that ApbC
function is involved in the synthesis of active TcuB.
Three systems for iron-sulfur ([Fe-S]) cluster biosynthesis
have been identified. The first system is encoded by the nif
(nitrogen fixation) operon in Azotobacter vinelandii and is re-
quired for the biosynthesis of nitrogenase (reviewed in refer-
ence 15). The second system is encoded by the iscSUA-hscAB-
fdx-orf3 (iron sulfur cluster) operon of Azotobacter vinelandii.
In Escherichia coli, the iscSUA-hscAB-fdx-orf3 operon encodes
housekeeping [Fe-S] cluster biosynthetic functions (20, 35, 55)
and is regulated by the IscR repressor (49). A third system for
the biosynthesis/repair of the [Fe-S] cluster has been described
for E. coli. In this bacterium, the sufABCDSE (sulfur utilization
factor) operon is induced during times of limited Fe availability
and oxidative stress (20, 28, 35, 37, 44, 54, 63). As in E. coli, the
genome of Salmonella enterica carries both the isc and suf
operons, and cellular viability requires the presence of one of
the two (36).
The [Fe-S] cluster biosynthetic systems mentioned above
have two general functional components. The cysteine desul-
furase enzymes NifS, SufS, and IscS catalyze the removal of
inorganic sulfide from L-cysteine (23, 33, 34, 43, 62), while the
scaffolding proteins NifU, IscU, IscA, and SufA appear to bind
and transfer labile [Fe-S] clusters to apoproteins (12, 27, 41,
56). Additional components can be specific to each system.
It was recently shown that cluster transfer from IscU to
apoferredoxin is stimulated by the addition of HscA, HscB,
and Mg ? ATP (10). These data emphasized a role for ATP-
hydrolyzing proteins, e.g., HscA and SufC, in the process of
cluster maturation (10, 38, 50).
Work with Salmonella enterica serovar Typhimurium LT2
identified several loci outside the above-mentioned operons
that impact [Fe-S] cluster metabolism. These loci include
apbC, apbE, rseC, and yggX, all of which encode proteins with
no characterized biochemical function (3, 4, 18, 45). Initially
isolated as conditional thiamine auxotrophs, strains with le-
sions in these loci displayed phenotypic behavior similar to that
of strains lacking isc operon functions (51–53). The apbC locus
was the most common location of conditional thiamine auxo-
trophs identified in general screens (45). ApbC is a 40-kDa
cytoplasmic protein that contains two conserved C-terminal
cysteine residues separated by two amino acids (CXXC) and a
Walker A box used for ATP binding and hydrolysis (25, 51).
Strains with lesions in apbC were independently isolated as
mutants unable to use tricarballylate as the sole carbon and
energy source for growth (A. R. Horswill and J. C. Escalante-
Semerena, unpublished data). The tricarballylate catabolic
genes tcuRABC (tricarballylate utilization) have been previ-
ously described (31), and a mechanism for this catabolism has
been proposed (29). In this model, TcuC transports tricarbal-
lylate across the inner membrane, where it is oxidized by the
flavoprotein TcuA to cis-aconitate, which can then enter the
Krebs cycle (29). During growth on tricarballylate, the recy-
cling of the reduced flavin adenine dinucleotide of TcuA is
achieved by TcuB, a membrane-bound protein that contains
two [4Fe-4S] clusters and heme (30).
The demonstration that TcuB contains iron-sulfur clusters,
in combination with work on ApbC and homologs, led to the
hypothesis that the [Fe-S] clusters of TcuB were compromised
in an apbC strain, preventing growth on tricarballylate. Con-
sistent with this hypothesis, we show herein that TcuB activity
is 100-fold lower in a strain lacking ApbC. The data further
show that derepression of the isc operon or overexpression of
iscU compensated for the lack of ApbC during growth of an
apbC strain on tricarballylate.
MATERIALS AND METHODS
Bacterial strains, media, and chemicals. All strains used in this study are
derived from S. enterica serovar Typhimurium LT2, and their genotypes are given
in Table 1. MudJ refers to the MudI1734 insertion element (9), and Tn10d(Tc)
refers to the transposition-defective mini-Tn10 described by Way et al. (59).
* Corresponding author. Mailing address: Department of Bacteriol-
ogy, 1550 Linden Drive, 6472 Microbial Sciences Building, University
of Wisconsin, Madison, WI 53706. Phone: (608) 265-4630. Fax: (608)
890-0785. E-mail: email@example.com.
?Published ahead of print on 25 April 2008.
No-carbon essential (NCE) medium of Berkowitz et al. (5) was made with
Milli-Q filtered water and supplemented with 1 mM MgSO4and trace minerals
(1, 11, 58). Glucose and tricarballylate were added to NCE medium at 11 mM
and 20 mM, respectively. Difco nutrient broth (8 g/liter) with NaCl (5 g/liter) or
lysogenic broth (6, 7) was used as rich medium. Difco BiTek agar was added (15
g/liter) for solid medium. When present in the medium, supplements were
provided at the following final concentrations: thiamine, 10 nM or 100 nM;
adenine, 0.4 mM; and nicotinic acid, 20 ?M. When needed, antibiotics were
TABLE 1. Strains and plasmidsa
Strain or plasmidb
ara-9 (wild type)
ara-9 apbC55::Tn10d(tet) rseC::kan
ara-9 apbC::MudJ (kan) apbE42::Tn10d(tet)
ara-9 apbC55:Tn10d(tet) yggX::Gm
ara-9 iscR2::MudJ (kan)
ara-9 apbC55:Tn10d(tet) iscR2::MudJ (kan)
ara-9 apbC55::Tn10d(tet) ?isc5::cat
ara-9 apbC55::Tn10d(tet) sufS::cat
ara-9 iscA1::MudJ (kan)
ara-9 apbC55::Tn10d(tet) iscA1::MudJ (kan)
ara-9 apbC::MudJ (kan)
ara-9 apbC::MudJ (kan) iscR11
ara-9 apbC55::Tn10d(tet) ?iscR11
ara-9 apbC55::Tn10d(tet) yggX::Gm iscR11
ara-9 apbC55::Tn10d(tet) stm2545::Tn10d(cat) iscR6
ara-9 apbC55::Tn10d(tet) stm2545::Tn10d(cat) iscR7
ara-9 apbC55::Tn10d(tet) stm2545::Tn10d(cat) iscR8
ara-9 apbC55::Tn10d(tet) stm2545::Tn10d(cat) iscR9
ara-9 apbC55::Tn10d(tet) stm2545::Tn10d(cat) iscR10
ara-9 staC::Tn10d(tet) stm2545::Tn10d(cat) yggX::Gm
ara-9 apbC::MudJ (kan) staC::Tn10d(tet) stm2545::Tn10d(cat) yggX::Gm
ara-9 apbC::MudJ (kan) stm2545::Tn10d(Cm) yggX::Gm iscR7
ara-9 staC::Tn10d(tet?) stm2545::Tn10d(cat?) yggX::Gm iscR7
J. C. Escalante-Semerena
J. C. Escalante-Semerena
F?ompT gal hsdSB(rB
F?ompT gal hsdSB(rB
?(lacZYA-argF)U169 deoR?80dlac ?(lacZ)M15?
?) dcm lon
?) dcm lon apbC::kan?
?) supE44 thi-1 recA1 gyrA (Nar1r) relA1
J. C. Escalante-Semerena
J. C. Escalante-Semerena
New England Biolabs
iscR (S. enterica)
iscS (S. enterica)
iscA (S. enterica)
iscA-hscB-hscA-fdx-orf3 (S. enterica)
iscU (S. enterica)
tcuABC (S. enterica)
apbC (S. enterica)
iscSUA-hscB-hscA-fdx-orf3 (A. vinelandii)
hscB-hscA-fdx-orf3 (S. enterica)
sufABCDSE (E. coli)
aUnless indicated otherwise in the description, all strains are S. enterica serovar Typhimurium LT2 constructed for this study.
bAll S. enterica strains constructed for this study are derivatives of TR6583 that were transduced to metE?. The TR6583 strain was from the Escalante-Semerena
cGenotypes are given for strains, and inserts (hosts) are given for plasmids. The following alleles have been previously described: apbC55::Tn10d (46), cyaY::Cm (57),
rseC::Kan (3, 53), apbE42::Tn10d(Tc) (4), yggX::Gm (19), iscR2::MudJ (52), ?isc5::Cm (53), and iscA2::MudJ (52). Tc, tetracycline; Cm, chloramphenicol; Gm,
gentamicin; Kan, kanamycin.
dSource also includes Lewis et al. (submitted).
VOL. 190, 2008 apbC IS REQUIRED FOR GROWTH ON TRICARBALLYLATE4597
added to the following concentrations in rich and minimal medium, respectively:
tetracycline, 20 and 10 ?g/ml; kanamycin, 50 and 125 ?g/ml; chloramphenicol, 20
and 4 ?g/ml; ampicillin, 50 and 15 ?g/ml; and gentamicin, 6 and 6 ?g/ml. All
chemicals were purchased from Sigma-Aldrich.
Genetic methods. (i) Mutant isolation. Nine independent cultures of DM10300
(apbC) were grown to full density in nutrient broth medium. One hundred micro-
liters of each culture was spread onto individual minimal tricarballylate thiamine
plates. Colonies spontaneously arose after 48 h of incubation at 37°C. One colony
derived from each culture was saved.
(ii) Isolation of linked insertions. The methods for transduction and the
purification of transductants have been previously described (13, 47, 48).
Transposons [Tn10d(cat)] (14) genetically linked to the suppressor mutations
were isolated by standard genetic techniques (24). In each case, mutant strains
were reconstructed and verified phenotypically prior to characterization. The
relevant insertions were mapped by sequencing using a PCR-based protocol
(University of Wisconsin Biotechnology DNA Sequence Facility) (8, 60).
(iii) Phenotypic analysis. Nutritional requirements were assessed on solid
medium and by the quantification of growth in liquid medium using either 5-ml
cultures in 25-ml shake tubes or 200-?l cultures in a 96-well plate. Protocols for
each have been previously described (3, 4). The starting A650was routinely
between 0.03 and 0.08, with a final A650between 0.5 and 1.1. Each culture had
at least three replicates. Growth on solid medium was scored after replica
printing to the relevant medium and after incubation at 37°C for 48 to 60 h.
Molecular biology. Restriction enzymes and DNA ligase were purchased from
Promega, and Pfu DNA polymerase was purchased from Stratagene. The iscU
and the hscA-orf3 genes were amplified from wild-type S. enterica by using
genomic DNA as the template. The primers were as follows: for iscU, the
forward primer was 5?-CCGAAGCTTATGGCTTACAGCGAAAAAG-3? and
the reverse primer was 5?-CGGGGATCCTTATTTCGCTTCGCGTTTG-3?; for
hscA-orf3, the forward primer was 5?-GGGCAAGCTTTGGATTACTTCACCC
TCTT-3? and the reverse primer was 5?-CCTCGGATCCTTACTCTGCTTCAT
The PCR product of iscU was purified and digested with BamHI and HindIII.
The resulting products were purified and ligated into similarly digested pSU19
(2), creating pIscU. The PCR product containing hscA-orf3 was blunt end ligated
into HincII-digested pSU19, resulting in pHscA-orf3. Plasmids were moved into
the appropriate strains via electroporation (42), and their identities were con-
firmed by restriction digestion and/or sequencing. The plasmids used are given in
Enzyme assays. (i) ?-Galactosidase. ?-Galactosidase assays were performed
according to the method of Winston et al. (61).
(ii) TcuB activity. His6-TcuB was overproduced from plasmid pTCU55 as
previously described (29). Cells from either strain JE6664 [C43(?DE3)] or strain
JE10465 [C43(?DE3) apbC::kan] containing pTCU55 were grown at 18°C on
Terrific broth to an optical density at 600 nm of ?0.4 and then induced with 300
?M isopropyl-?-D-thiogalactopyranoside (IPTG) overnight. His6-TcuB-enriched
extracts were obtained and assayed as previously described (29). Briefly, 200-?l
reaction mixtures contained 2-(N-morpholino)ethanesulfonic acid (MES) (100
mM, pH 6.5, at 30°C), dithiothreitol (1 mM), 1 ?g TcuA, and 20 ?g His6-TcuB-
enriched extract. Reaction mixtures were incubated at 30°C for 10 min following
the addition of tricarballylate (10 mM). Reactions were stopped by the addition
of 40-?l samples to 60 ?l of 166.7 mM H2SO4. Fifty microliters of each sample
was used to quantify cis-aconitate production using a high-performance liquid
chromatography protocol (29).
ApbC, but not Isc or Suf protein, is required for growth on
tricarballylate. An apbC mutant does not grow on tricarbally-
late as a carbon and energy source, although it is proficient for
growth on glucose (Fig. 1) and other carbon sources (e.g.,
succinate and gluconate) (data not shown). The expression of
the tcuABC operon in trans from a nonnative promoter did not
restore growth on tricarballylate, consistent with a posttran-
scriptional effect. A strain carrying a deletion of the iscRSUA-
hscAB-fdx-orf3 operon or a polar insertion in sufS grew well on
tricarballylate (Fig. 1; Table 2). The growth of the ?iscRSUA-
hscAB-fdx-orf3 mutant was reduced compared to that of the
wild type on both tricarballylate and glucose, consistent with a
previously reported global defect (52). The growth of strains
lacking other genes involved in [Fe-S] cluster metabolism was
also assessed on tricarballylate; no defect was found for mu-
tants lacking apbE, rseC, cyaY, or yggX (data not shown). These
growth data suggested a specific role for ApbC function during
FIG. 1. apbC mutants fail to grow on tricarballylate. Strains were
grown at 37°C in NCE medium supplemented with thiamine and nic-
otinic acid and with a sole carbon and energy source. Growth of strains
DM10310 (wild type) (E), DM10300 (apbC) (F), DM10325 (iscSUA-
hscAB-fdx-orf3) (Œ), and DM10667 (iscA-hscAB-fdx-orf3) (?) was
monitored on tricarballylate (A) and glucose (B).
TABLE 2. An apbC mutant is unable to grow on
Strain Relevant genotypeb
Doubling time (h) on
DM10310 Wild type
1.7 ? 0.1
1.9 ? 0.1
4.5 ? 0.2
1.8 ? 0.1
2.1 ? 0.1
2.1 ? 0.1
4.3 ? 0.1
2.2 ? 0.0
2.1 ? 0.0
aDoubling times were calculated using the formula ? ? ln(X/Xo)/T, where ?
is the growth rate, X and Xoare optical density measurements at 650 nm, T is the
time between the absorbance readings X and Xo, and the doubling time (g) was
(ln 2)/? (39). Values are averages of three independent cultures. NG, no growth.
bRelevant genotypes indicate the genes that are defective due to the presence
of a polar mutation or deletion.
cThe defined medium included the indicated carbon source and was supple-
mented with thiamine and nicotinic acid.
dRatio of the doubling time on tricarballylate to the doubling time on glucose.
4598BOYD ET AL.J. BACTERIOL.
TcuB is less active in apbC mutants. To determine if ApbC
function was involved in the synthesis of active TcuB, we mea-
sured the activity of this protein in vitro in a strain lacking
ApbC. E. coli strain C43(?DE3) and an apbC mutant deriva-
tive of that strain were used to overproduce TcuB from plas-
mid pTCU55 (30). The apbC mutant extract produced 100-fold
less cis-aconitate than that of the wild-type strain (220 ? 20
and 25,300 ? 700 pmol cis-aconitate produced after 10 min,
respectively). In contrast, the activities of aconitase, succinate
dehydrogenase, and the non-[Fe-S] cluster protein malate de-
hydrogenase were indistinguishable in the apbC mutant and
wild-type extracts (data not shown).
Conditional growth of an apbC mutant. An apbC strain did
not grow in liquid medium with tricarballylate as a carbon
source but grew on solid tricarballylate medium after ?48 h.
This observation suggested a functionally redundant system
working at low efficiency. Insertions in yggX, iscA, or iscR
eliminated residual growth of the apbC mutants on tricarbal-
lylate medium, while apbC strains with insertions in sufS, apbE,
or rseC retained growth. None of these above-mentioned loci
were required for growth on tricarballylate in an otherwise
Suppressor analysis provides insight into ApbC function.
Nine independent spontaneous mutations that allowed the
growth of strain DM10300 (apbC) on tricarballylate were iden-
tified. Genetic analysis determined that a Tn10d(cat) insertion
in open reading frame STM2545 was ?85% cotransducible by
phage P22 with five of the nine suppressor mutations, placing
the mutations near the isc operon (Fig. 2A). Sequence analysis
determined that each of the five mutations was in iscR, encod-
ing the repressor of the isc operon (49). The suppressor mu-
tations resulted in variant IscR proteins with the following
amino acid changes: L109Q (iscR6), S38F (iscR7), Q94Z (stop)
(iscR8), Y41S (iscR9), and G64A (iscR10). The four substitu-
tions were located in the predicted helix-turn-helix DNA bind-
ing domain of IscR (49) and were expected to generate inactive
proteins and result in the constitutive expression of the isc
Derepression of the isc operon restores the growth of an
apbC mutant on tricarballylate. Three results confirmed that
the suppressor mutations disrupted IscR function and that the
resulting derepression of the isc operon was sufficient to allow
an apbC mutant to grow on tricarballylate. First, the introduc-
tion of an in-frame deletion of iscR (iscR?11) restored the
growth of an apbC mutant on tricarballylate (Fig. 2B). Second,
the expression of the wild-type allele of iscR in trans eliminated
growth on tricarballylate of strain JE10435 (apbC iscR?11) and
strains containing the other iscR alleles (data not shown).
Third, transcription was monitored using a lacZ reporter
under the control of the iscR promoter (piscR-lacZ transcrip-
tional fusion [plasmid pIsc2]) (J. A. Lewis, J. M. Boyd, D. M.
Downs, and J. C. Escalante-Semerena, submitted for publica-
tion). Data given in Table 3 show that the chromosomal iscR
alleles increased the expression of the reporter as efficiently as
the chromosomal deletion of iscR. Together, these results con-
firmed that the IscR variants encoded by mutant iscR alleles
FIG. 2. The overexpression of isc genes allows growth of an apbC
mutant on tricarballylate. Strains were grown at 37°C in NCE medium
supplemented with thiamine and nicotinic acid and with tricarballylate
as a carbon and energy source. (A) A schematic shows the genetic
organization of the S. enterica isc operon. The borders of inserts used
to generate plasmids are diagrammed below the operon. (B) Growth
of strains DM10310 (wild type) (E), JE10435 (iscR11 apbC) (f),
DM10698 (iscR11) (?), DM10474 (apbC iscR7) (Œ), and DM10300
(apbC) (F) on tricarballylate NCE medium. (C) Growth of strains
DM10310 (wild type) with pSU19 (E), DM10300 (apbC) with pSU19
(F), DM10300 (apbC) with pIscU (Œ), and DM10474 (apbC iscR7)
with pSU19 (f) on tricarballylate NCE medium.
TABLE 3. Inactivation of IscR derepresses the isc operona
70 ? 4
70 ? 2
390 ? 72
270 ? 28
430 ? 47
350 ? 31
340 ? 28
290 ? 22
aAll strains contained a plasmid that carried the isc reporter-lacZ fusion as
described in the text. Strains were grown in LB-ampicillin medium to the mid-
exponential phase of cell growth. ?-Galactosidase assays were performed as
described in Materials and Methods.
VOL. 190, 2008 apbC IS REQUIRED FOR GROWTH ON TRICARBALLYLATE 4599
failed to repress iscRSUA-hscAB-fdx-orf3 transcription and
that derepression of the iscRSUA-hscAB-fdx operon was suffi-
cient to restore growth on tricarballylate in an apbC strain.
IscU has functional redundancy with ApbC. The above-
mentioned results strongly suggested that one or more proteins
encoded by the isc operon had a functional overlap with ApbC.
Plasmids encoding one or more isc genes (Fig. 2A) were in-
troduced into strain DM10300 (apbC), and growth on tricar-
ballylate was assessed. Of the plasmids tested, only pIscU af-
fected the growth of the apbC mutant strain on tricarballylate
(Fig. 2C). Several points can be taken from the data in Fig. 2C.
First, pIscU restored the growth of the apbC mutant compared
to the growth of the same strain with the vector-only control.
The uniformly increased lag did not alter the conclusion that
the growth of an apbC mutant on tricarballylate was allowed by
either the derepression of the isc operon or the overexpression
of iscU. Each of the strains retained the pattern of growth upon
reinoculation, indicating that the growth was not due to a
mutant overpopulating the culture. The doubling times for the
strains were 1.5 ? 0.0, 1.7 ? 0.0, and 2.3 ? 0.1 h for the wild
type with pSU19, the apbC iscR7 mutant with pSU19, and the
apbC mutant with pIscU, respectively.
IscU is not sufficient to compensate for ApbC. Plasmid
pIscU failed to restore growth to an apbC mutant that was also
defective in the isc operon (data not shown). This result was
particularly significant for the apbC iscA genetic background,
since the iscA mutation alone did not affect growth on tricar-
ballylate (Table 2). These data suggested that the ability of
IscU overproduction to allow the growth of an apbC mutant on
tricarballylate required iscA and/or at least one gene down-
stream in the operon.
This study was initiated to understand the inability of S.
enterica apbC mutants to grow with tricarballylate as a carbon
and energy source. Previous studies implicating ApbC in [Fe-S]
cluster metabolism (51, 53) and the report that TcuB contained
[Fe-S] clusters (30) generated a hypothesis for the growth
defect. The result that strains lacking apbC displayed 100-fold
less TcuB activity than the wild type supported this hypothesis
and has provided an additional system that can be exploited to
dissect the role of ApbC in Salmonella. The physiological stud-
ies described herein provide data that IscU has a functional
overlap with ApbC and, furthermore, that ApbC has a specific
role in tricarballylate utilization that distinguishes it from the
general [Fe-S] cluster biosynthesis systems encoded by the isc
and suf operons.
Mutants lacking either the complete isc operon or the suf
operon had no growth defect specific to tricarballylate me-
dium. Thus, the lack of growth on tricarballylate was the first
defect described for an apbC mutant that was not shared by
strains lacking the major [Fe-S] cluster biosynthetic system isc.
Recently, ErpA, a protein essential for isoprenoid biosynthesis
in E. coli, was shown to specifically transfer [Fe-S] clusters to
IspG (32). Similarly, Iba57 was shown to be essential for mi-
tochondrial aconitase maturation in yeast (16). The results
with ErpA and Iba57 were parallel to those with ApbC, since
in all cases neither the isc system nor the suf system expressed
at physiological levels could functionally replace these proteins
in isoprenoid biosynthesis, aconitase maturation, or tricarbal-
lylate catabolism, respectively.
Null alleles of iscR restored the growth of the apbC mutant
on tricarballylate and suggested that a gene(s) in the isc operon
had a functional overlap with apbC. This result was confirmed
in a study by Lewis et al. that showed that growth of an apbC
mutant on tricarballylate could be restored by physiological
conditions that increased the expression of the isc operon
(Lewis et al., submitted). When provided in trans, iscU restored
growth on tricarballylate to an apbC mutant. The expression of
iscU in trans failed to restore growth on tricarballylate to an
apbC isc double mutant, showing that IscU required at least
one of the iscA, hscAB, or orf3 genes to restore growth. These
data were reminiscent of studies on the functional redundancy
of the U-type scaffolds in Azotobacter vinelandii. Johnson and
coworkers found that target specificity distinguished the isc
and nif systems when these operons were expressed at chro-
mosomal levels, which was altered if the relevant genes were
overexpressed (21, 22). Transcriptional studies found no effect
of IscR on apbC expression or vice versa (reference 17 and
data not shown).
Eukaryotic ApbC homologues Npb35 and Cfd1 can inde-
pendently bind [Fe-S] clusters and rapidly and efficiently trans-
fer these clusters to the Leu1 apoenzyme (40). Recent data
show that ApbC can similarly transfer an [Fe-S] cluster to Leu1
in vitro (J. M. Boyd, A. J. Pierik, D. J. Aguilar-Netz, R. Lill,
and D. M. Downs, submitted for publication). The data pre-
sented herein suggest that the TcuB protein provides a physi-
ologically relevant system to further explore the biochemical
function of ApbC and to address its specificity in vitro.
This work was supported by competitive grants GM47296 (D.M.D.)
and GM62203 (J.C.E.-S.) and a Kirschstein postdoctoral training grant
(GM079938-02) (J.M.B.) from the National Institutes of Health.
Funds were also provided from a 21st Century Scientists Scholars
Award from the J. M. McDonnell fund to D.M.D.
The content of this study is solely the responsibility of the authors
and does not necessarily represent the official views of the National
Institute of General Medical Sciences or the National Institutes of
1. Balch, W. E., and R. S. Wolfe. 1976. New approach to the cultivation of
methanogenic bacteria: 2-mercaptoethanesulfonic acid (HS-CoM)-depen-
dent growth of Methanobacterium ruminantium in a pressurized atmosphere.
Appl. Environ. Microbiol. 32:781–791.
2. Bartolome, B., Y. Jubete, E. Martinez, and F. de la Cruz. 1991. Construction
and properties of a family of pACYC184-derived cloning vectors compatible
with pBR322 and its derivatives. Gene 102:75–78.
3. Beck, B. J., L. E. Connolly, A. De Las Pen ˜as, and D. M. Downs. 1997.
Evidence that rseC, a gene in the rpoE cluster, has a role in thiamine
synthesis in Salmonella typhimurium. J. Bacteriol. 179:6504–6508.
4. Beck, B. J., and D. M. Downs. 1998. The apbE gene encodes a lipoprotein
involved in thiamine synthesis in Salmonella typhimurium. J. Bacteriol. 180:
5. Berkowitz, D., J. M. Hushon, H. J. Whitfield, Jr., J. Roth, and B. N. Ames.
1968. Procedure for identifying nonsense mutations. J. Bacteriol. 96:215–
6. Bertani, G. 2004. Lysogeny at mid-twentieth century: P1, P2, and other
experimental systems. J. Bacteriol. 186:595–600.
7. Bertani, G. 1951. Studies on lysogenesis. I. The mode of phage liberation by
lysogenic Escherichia coli. J. Bacteriol. 62:293–300.
8. Caetano-Anolles, G. 1993. Amplifying DNA with arbitrary oligonucleotide
primers. PCR Methods Appl. 3:85–94.
9. Castilho, B. A., P. Olfson, and M. J. Casadaban. 1984. Plasmid insertion
mutagenesis and lac gene fusion with mini-Mu bacteriophage transposons. J.
4600BOYD ET AL.J. BACTERIOL.