Differential proteomic analysis highlights metabolic strategies associated with balhimycin production in Amycolatopsis balhimycina chemostat cultivations.

Abstract

Proteomics was recently used to reveal enzymes whose expression is associated with the production of the glycopeptide antibiotic balhimycin in Amycolatopsis balhimycina batch cultivations. Combining chemostat fermentation technology, where cells proliferate with constant parameters in a highly reproducible steady-state, and differential proteomics, the relationships between physiological status and metabolic pathways during antibiotic producing and non-producing conditions could be highlighted.
Two minimal defined media, one with low Pi (0.6 mM; LP) and proficient glucose (12 g/l) concentrations and the other one with high Pi (1.8 mM) and limiting (6 g/l; LG) glucose concentrations, were developed to promote and repress antibiotic production, respectively, in A. balhimycina chemostat cultivations. Applying the same dilution rate (0.03 h-1), both LG and LP chemostat cultivations showed a stable steady-state where biomass production yield coefficients, calculated on glucose consumption, were 0.38 ± 0.02 and 0.33 ± 0.02 g/g (biomass dry weight/glucose), respectively. Notably, balhimycin was detected only in LP, where quantitative RT-PCR revealed upregulation of selected bal genes, devoted to balhimycin biosynthesis, and of phoP, phoR, pstS and phoD, known to be associated to Pi limitation stress response. 2D-Differential Gel Electrophoresis (DIGE) and protein identification, performed by mass spectrometry and computer-assisted 2 D reference-map http://www.unipa.it/ampuglia/Abal-proteome-maps matching, demonstrated a differential expression for proteins involved in many metabolic pathways or cellular processes, including central carbon and phosphate metabolism. Interestingly, proteins playing a key role in generation of primary metabolism intermediates and cofactors required for balhimycin biosynthesis were upregulated in LP. Finally, a bioinformatic approach showed PHO box-like regulatory elements in the upstream regions of nine differentially expressed genes, among which two were tested by electrophoresis mobility shift assays (EMSA).
In the two chemostat conditions, used to generate biomass for proteomic analysis, mycelia grew with the same rate and with similar glucose-biomass conversion efficiencies. Global gene expression analysis revealed a differential metabolic adaptation, highlighting strategies for energetic supply and biosynthesis of metabolic intermediates required for biomass production and, in LP, for balhimycin biosynthesis. These data, confirming a relationship between primary metabolism and antibiotic production, could be used to increase antibiotic yield both by rational genetic engineering and fermentation processes improvement.

Full text

Page 1

RESEARCHOpen Access
Differential proteomic analysis highlights
metabolic strategies associated with balhimycin
production in Amycolatopsis balhimycina
chemostat cultivations
Giuseppe Gallo1*, Rosa Alduina1, Giovanni Renzone2, Jette Thykaer3, Linda Bianco2, Anna Eliasson-Lantz3,
Andrea Scaloni2, Anna Maria Puglia1*
Abstract
Background: Proteomics was recently used to reveal enzymes whose expression is associated with the production
of the glycopeptide antibiotic balhimycin in Amycolatopsis balhimycina batch cultivations. Combining chemostat
fermentation technology, where cells proliferate with constant parameters in a highly reproducible steady-state,
and differential proteomics, the relationships between physiological status and metabolic pathways during
antibiotic producing and non-producing conditions could be highlighted.
Results: Two minimal defined media, one with low Pi (0.6 mM; LP) and proficient glucose (12 g/l) concentrations
and the other one with high Pi (1.8 mM) and limiting (6 g/l; LG) glucose concentrations, were developed to
promote and repress antibiotic production, respectively, in A. balhimycina chemostat cultivations. Applying the
same dilution rate (0.03 h-1), both LG and LP chemostat cultivations showed a stable steady-state where biomass
production yield coefficients, calculated on glucose consumption, were 0.38 ± 0.02 and 0.33 ± 0.02 g/g (biomass
dry weight/glucose), respectively. Notably, balhimycin was detected only in LP, where quantitative RT-PCR revealed
upregulation of selected bal genes, devoted to balhimycin biosynthesis, and of phoP, phoR, pstS and phoD, known
to be associated to Pi limitation stress response. 2D-Differential Gel Electrophoresis (DIGE) and protein
identification, performed by mass spectrometry and computer-assisted 2 D reference-map http://www.unipa.it/
ampuglia/Abal-proteome-maps matching, demonstrated a differential expression for proteins involved in many
metabolic pathways or cellular processes, including central carbon and phosphate metabolism. Interestingly,
proteins playing a key role in generation of primary metabolism intermediates and cofactors required for
balhimycin biosynthesis were upregulated in LP. Finally, a bioinformatic approach showed PHO box-like regulatory
elements in the upstream regions of nine differentially expressed genes, among which two were tested by
electrophoresis mobility shift assays (EMSA).
Conclusion: In the two chemostat conditions, used to generate biomass for proteomic analysis, mycelia grew with
the same rate and with similar glucose-biomass conversion efficiencies. Global gene expression analysis revealed a
differential metabolic adaptation, highlighting strategies for energetic supply and biosynthesis of metabolic
intermediates required for biomass production and, in LP, for balhimycin biosynthesis. These data, confirming a
relationship between primary metabolism and antibiotic production, could be used to increase antibiotic yield
both by rational genetic engineering and fermentation processes improvement.
* Correspondence: giuseppe.gallo@unipa.it; ampuglia@unipa.it
1Università di Palermo, Dipartimento di Biologia Cellulare e dello Sviluppo,
Viale delle Scienze, Parco d’Orleans II, 90128 Palermo, Italy
Full list of author information is available at the end of the article
Gallo et al. Microbial Cell Factories 2010, 9:95
http://www.microbialcellfactories.com/content/9/1/95
© 2010 Gallo et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons
Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.

Page 2

Background
Glycopeptide antibiotics have found a successful use as
last resort antibiotics in the treatment of methicillin-
resistant Staphylococcus aureus (MRSA) infections [1].
Leading glycopeptide drugs are vancomycin and teico-
planin, produced by the actinomycetes Amycolatopsis
orientalis and Actinoplanes teichomyceticus, respectively.
The actinomycete Amycolatopsis balhimycina DSM5908,
which produces the vancomycin-like antibiotic balhimy-
cin [2,3], has been investigated as model strain for stu-
dies on glycopeptide biosynthesis since it can be
genetically modified and the sequence of bal cluster,
containing genes devoted to balhimycin biosynthesis, is
available [4-6].
Balhimycin consists of a heptapeptide core made of
the proteinogenic amino acids Leu and Asn and the
nonproteinogenic amino acids 3,5-dihydroxyphenylgly-
cine (DPG), 4-hydroxyphenylglycine (HPG) and b-
hydroxytyrosine (H-Tyr). This heptapeptide is assembled
by a non-ribosomal peptide synthetase (NRPS) and then
extensively modified by the so called tailoring reactions,
such as oxidative cross-linking of the electron-rich aro-
matic side chains, halogenation, glycosylation and
methylation [6].
In Actinomycetes, the biosynthesis of secondary meta-
bolites like antibiotics is generally elicited as develop-
mental program and physiological response to a variety
of environmental stimuli, including high cell density,
nutritional limitation and/or presence of stress-inducing
agents [7-9]. In A. orientalis [10,11] and A. balhimycina
[12] cultivations, performed using minimal defined
media, glucose amount was revealed positively corre-
lated with antibiotic biosynthesis, cell growth rate and
biomass production. On the other hand, inorganic phos-
phate (Pi) limitation is known to negatively affect
growth rate and biomass production but to be beneficial
for the production of glycopeptide antibiotics as shown
for vancomycin [10], A40926 (produced by Nonomuraea
ATCC 39727) [13] and balhimycin [12]. Pi limitation
negatively controls the expression of both primary and
secondary metabolism genes [14-16]. In Streptomycetes,
Pi-controlled regulatory elements, called PHO boxes,
have been reported in the upstream region of Pho regu-
lon genes, which are devoted to Pi-nutritional stress
response [15-19]. The PHO boxes are targets of PhoP, a
transcriptional regulator whose activity is regulated by
the phosphate-sensing membrane protein PhoR. PhoP is
reported to have a dual role acting as either positive or
negative regulator [16-20]. Although balhimycin bio-
synthesis has been extensively studied, the molecular
bases of limitation of nutrients, such as Pi and glucose,
controlling primary metabolism and antibiotic biosynth-
esis have not been investigated yet.
Proteomics was proven useful to identify specific bio-
chemical pathways (or parts thereof) and key enzymes
to be further targeted in genetic manipulations aimed to
maximize the conversion of substrates into useful end-
products in microorganisms [21,22]. In this context, we
have recently used a differential proteomic approach to
demonstrate that balhimycin production in batch cul-
ture is associated with the upregulation of enzymes
involved in the biosynthesis of antibiotic precursors,
thus suggesting that the metabolic apparatus is orien-
tated to sustain balhimycin production [23,24]. Accord-
ingly, an increased precursor availability, in particular
tyrosine, was recently showed to be beneficial for balhi-
mycin production [25].
Up to date, few experimental studies on antibiotic
production have focused attention on continuous cul-
tures, although the advantages of this type of study of
microbial physiology have been recognized for many
years. In fact, this approach provides useful means of
researching the relationships between physiological sta-
tus of an organism and production of antibiotics by
comparing highly reproducible steady-state conditions,
where growth parameters (i.e. biomass production yield,
growth rate, carbon source uptake, O2consumption,
CO2and metabolite production) and medium compo-
nents are constant [10]. In this study, chemostat cul-
tures of A. balhimycina were used to obtain steady-state
conditions for biomass accumulation with the same
growth rate and with or without balhimycin production.
These cultivations were then analyzed by a comparative
proteomic study to elucidate changes in the expression
of genes involved in A. balhimycina primary and sec-
ondary metabolism which are associated with biomass
production and antibiotic synthesis.
Results and Discussion
Chemostat cultivation profile
Preliminary A. balhimycina batch cultivations, per-
formed using a minimal defined medium containing glu-
cose as only carbon source, revealed that to generate 1
g/l of biomass dry weight (BDW) about 3 g/l of glucose
and 0.3 mM of Pi are necessary in a balanced growth.
Thus, two minimal defined media, one with low Pi
(0.6 mM; LP) and proficient glucose (12 g/l) concentra-
tions and the other one with high Pi (1.8 mM) and lim-
iting (6 g/l; LG) glucose concentrations, were developed
to perform A. balhimycina chemostat cultivations.
Appling a dilution rate of 0.03 h-1a highly reproducible
and stable steady-state was achieved in both conditions
(Figure 1: panels A and B), revealing that mycelia grew
with the same rate and with a similar biomass produc-
tion yield, calculated on consumed glucose, i.e. 0.38 ±
0.02 and 0.33 ± 0.02 g/g (BDW/glucose) in LG and LP,
Gallo et al. Microbial Cell Factories 2010, 9:95
http://www.microbialcellfactories.com/content/9/1/95
Page 2 of 14

Page 3

respectively (Figure 1, panel C). In these systems, resi-
dual Pi and glucose concentrations (Figure 1 panels A
and B) were close to zero during steady-state, thus
revealing that glucose and Pi, continuously introduced
by feeding in, are completely up-taken during constant
growth. Despite the similar glucose-biomass conversion
coefficients, CO2production yield, normalized to BDW,
was about 1.4 fold higher in LP than in LG condition
Figure 1 panel D). This result indicates an increased
metabolism throughout TCA cycle in LP. Interestingly,
as demonstrated by HPLC analysis, balhimycin was
revealed only in LP condition, showing a productivity of
0.03 mg/g/h (balhimycin/BDW/time), and the antibiotic
on-set coincided with CO2production yield increment
Figure 1 panel B). Conversely, LG condition resulted in
a complete inhibition of antibiotic production. The bio-
masses, harvested during the steady-state, were then
used to carry out gene expression analysis by quantita-
tive (q)RT-PCR of selected genes and by differential pro-
teomic analysis.
Quantitative RT-PCR of Pho regulon genes
The expression of Pho regulon genes was used as repor-
ter of Pi nutritional stress in chemostat cultures. In
Streptomycetes, the products of these genes, like the
phosphatase PhoD and the high affinity Pi transporter
system PtsSABC, are known to provide extracellular Pi
to fulfil the metabolic needs [15-19] or, like the poly-
phosphate kinase Ppk, to link energetic state of cell and
Pi accumulation [26,27]. A BLAST analysis, performed
using amino acid sequence of S. coelicolor PhoR, PhoP,
PhoD, PstS and Ppk against A. balhimycina ORF data-
base, revealed genes whose products show homology of
65%, 90%, 67%, 59% and 76% with their S. coelicolor
Figure 1 Chemostat cultivation profile of A. balhimycina grown in LG (panel A) and LP (panel B) conditions. The steady-state was
reached after about 3 and 5 residence times (RT), respectively. RT, corresponding to 33.3 h, was calculated as the inverse of DR. Data shown per
each condition are mean values generated by two parallel cultivations. Mean values during steady-state of: biomass dry weight (DW) production
yield, normalized to the glucose consumption (panel C); CO2production yield, reported as partial pressure (%) and normalized to the biomass
DW concentration (panel D); balhimycin concentration (panel E). Vertical bars represent standard deviations.
Gallo et al. Microbial Cell Factories 2010, 9:95
http://www.microbialcellfactories.com/content/9/1/95
Page 3 of 14

Page 4

counterparts, respectively. The identification of these A.
balhimycina proteins was further confirmed by a
BLAST analysis, performed using their amino acid
sequence (Additional file 1 Table1S) against SwissProt
database [28]. In fact, this analysis revealed high degree
of homology (ranging from 85 to 100%) with actinomy-
cete proteins annotated as PhoP, PhoR, PhoD, PstS and
Ppk, respectively (Additional file 1 Table 2S).
As revealed by qRT-PCR, all the tested genes but ppk
were upregulated in LP. In particular, phoR and phoP,
which also showed co-transcription (data not shown),
were induced about 2-fold, while pstS and phoD more
than 20-fold (Figure 2). In agreement with Rodríguez-
García and co-workers (2007) that reported upregulation
of S. coelicolor homologues [16] in phosphate-deficiency,
these data revealed that Pi is limiting under LP
condition.
Surprisingly, ppk expression was downregulated in LP
(Figure 2). Ppk is reported to be involved in polyPi utili-
zation in S. lividans [26] or in polyPi synthesis in Bacil-
lus cereus [29] and Myxococcus xanthus [30]. The
downregulation of ppk in LP condition suggests that in
A. balhimycina the functional role of Ppk is to act
mainly as polyPi kinase for Pi storing. In addition,
accordingly to that it has already been observed in S.
lividans [27], the expression of ppk is negatively related
to antibiotic production in A. balhimycina.
Quantitative RT-PCR of bal genes
The expression of selected bal genes, essential for balhi-
mycin production and chosen as representative of poly-
cistronic transcripts [4], was analysed by qRT-PCR. The
analysed bal genes are involved in DPG (dpgA), HPG
(hmaS) and H-Tyr (bpsD) synthesis, in heptapeptide
backbone assembling (bpsA) and tailoring reactions
(oxyA and bgtfA), as well as in regulatory (bbr) and self-
resistance (vanS) mechanisms. qRT-PCR revealed that
all these genes are upregulated under LP condition (Fig-
ure 3), with the exception of vanS, thus revealing that
the Pi limitation in glucose proficiency positively affects
the expression of balhimycin biosynthetic genes. VanS is
a membrane protein acting with the transcriptional reg-
ulator VanR as a two-component system putatively
devoted to sense extracellular balhimycin. Its downregu-
lation is in agreement with the mechanism proposed by
Walsh et al. [31] where VanR negatively controls the
expression of vanSR bicistronic transcript in the pre-
sence of glycopeptide antibiotics.
Under LG condition, the low expression levels of bal
genes could not explain alone the absence of antibiotic
production. In fact, this finding suggests complex regu-
latory mechanisms involving the synthesis of balhimycin
precursors and the formation of the end-product. The
importance of the availability of precursors in antibiotic
biosynthesis will be elucidated by proteomic results.
Proteomic analysis
Analysis of global protein expression changes between
LP and LG conditions was performed by 2D-Differential
Gel Electrophoresis (2D-DIGE) analysis (Figure 4). This
investigation revealed 72 upregulated and 12 downregu-
lated protein spots in LP condition, showing at least 1.5
fold increased or decreased abundance (calculated as
spot Vol) and at least a probability for null hypothesis
(p) < 0.05 (calculated using the analysis of variance or
ANOVA test). These proteins were identified by MS
analysis or by automatic gel-matching using the A. bal-
himycina protein 2 D reference-maps, available over the
World Wide Web as interactive pages at http://www.
unipa.it/ampuglia/Abal-proteome-maps[23,24] (Addi-
tional file 1 Table 3S, 4S, 5S and 6S). The reliability of
this analysis was firstly demonstrated by the identifica-
tion of groups of proteins showing similar expression
profiles and whose corresponding genes are arranged in
putative operons (Additional file 1 Table 6S), which are
Figure 2 Transcriptional expression profile of selected A.
balhimycina Pho regulon genes.
Figure 3 Transcriptional expression profile of selected A.
balhimycina bal genes.
Gallo et al. Microbial Cell Factories 2010, 9:95
http://www.microbialcellfactories.com/content/9/1/95
Page 4 of 14

Page 5

characterized by a maximum of 60 bp between two
adjacent ORFs. The identified proteins were clustered
into functional categories according to BioCyc http://
biocyc.org/[32], KEGG http://www.genome.jp/kegg/[33]
and ExPASy http://expasy.org/[28] metabolic databases.
The functional clustering revealed that the differentially
expressed proteins are related to many metabolic path-
ways or cellular processes, including glucose and phos-
phate metabolism. On the basis of their function, the
identified proteins could be divided in several groups
which are related to central carbon metabolism (31%),
protein biosynthesis/amino acid metabolism (21%),
redox and energetic balance (10%), nucleotide/fatty acid/
amino sugar metabolism (10%) and balhimycin bio-
synthesis (2%) or are unknown (15%). The role of these
proteins is discussed below in dedicated sections, high-
lighting the mechanisms controlling adaptation to Pi
limitation, glucose catabolism and the metabolic strate-
gies aimed to ensure metabolic intermediates for both
biomass accumulation and antibiotic biosynthesis.
Central carbon catabolism
Most differentially expressed proteins are involved in gly-
colysis, pentose phosphate pathway (PPP), and TCA cycle
(Figure 5; Additional file 1 Table 3S). In particular, glyco-
lytic enzymes glucose-6-phosphate isomerase (PgiA), trio-
sephosphate isomerase (TPI), phosphoglycerate kinase
(Pgk), glyceraldehyde 3-phosphate dehydrogenase
(GAPDH), fructose-bisphosphate aldolase (FDA), PP-
dependent-fructose 6-phosphate 1-phosphotransferase
(P-PFK), enolase (ENO) and pyruvate dehydrogenase
complex members (two dihydrolipoamide dehydro-
genases, LpdA1 and 2, and one dihydrolipoamide acyl-
transferase, SucB) were upregulated in LP condition. The
upregulation of these enzymes suggests an increased
metabolic flux throughout glycolysis. This result is in
agreement with the upregulation of the enzymes transke-
tolase (TrK) and transaldolase (TrA), both belonging to
non-oxidative branch of PPP, since these enzymes use as
substrates the glycolytic intermediates fructose-6P and
glyceraldehyde-3P, respectively. In addition, also the
enzyme F420-dependent glucose-6-phosphate dehydro-
genase (G6PD) was upregulated in LP, thus suggesting an
increased flux throughout the oxidative branch of PPP as
well. The upregulation of glycolytic enzymes is also in
agreement with the upregulation of TCA cycle enzymes
aconitate hydratase (ACO), succinate dehydrogenase fla-
voprotein subunit (SudA), LpdA1 and 2, SucB, succinyl-
CoA ligase alpha and beta subunit (A-and B-SCS) and
malate dehydrogenase (MDH) (Figure 5; Additional file 1
Table 3S). These findings also correlated with the
increased production yield of CO2 in LP condition
(Figure 1D) and also with the upregulation of ThiS and
ThiC, involved in thiamine synthesis and previously asso-
ciated with balhimycin production [23]. Thiamine dipho-
sphate (TDP) is reported to be a co-factor of the
Figure 4 2D-electropherograms, available over the World Wide Web as interactive pages at http://www.unipa.it/ampuglia/Abal-
proteome-maps, of A. balhimycina protein extracts from biomass collected from LP (A) and LG (B) cultivations. Labels indicate protein
up-(A) or downregulated (B) in LP condition (Additional file 1 Table 3S).
Gallo et al. Microbial Cell Factories 2010, 9:95
http://www.microbialcellfactories.com/content/9/1/95
Page 5 of 14

Page 6

Figure 5 Scheme of metabolic pathways where enzymes either upregulated (black boxes) or downregulated (white boxes) in LP are
highlighted. Upregulation of central carbon metabolism enzymes suggests an increased availability of primary metabolite intermediates and
ATP, GTP, NAD(P)H and FADH2for anabolic processes under LP condition. Enzymes related to the synthesis of balhimycin precursors (red
rectangles), such as Leu, Tyr, Asn, DPG and HPG and amino sugars, were upregulated during antibiotic production. Reactions are reported
according to KEGG and BioCy metabolic pathway databases. Enzyme name abbreviations refer to Table 3 S (Additional file 1).
Gallo et al. Microbial Cell Factories 2010, 9:95
http://www.microbialcellfactories.com/content/9/1/95
Page 6 of 14

Page 7

upregulated TrK, pyruvate and alpha-ketoglutarate dehy-
drogenase complexes [34]. Finally, an ABC transport
system ATPase component and an extracellular sugar-
binding protein resulted downregulated in LP. These two
proteins share conserved domains with LivG/F ATPases
and MalE, respectively, which are shown to be required
for the up-take of alternative carbon sources during glu-
cose limitation/starvation [35,36].
The importance of glycolysis [37] and TCA cycle
[10,16,38-40] for energetic balance and antibiotic pro-
duction in Actinomycetes has already been reported.
The upregulation of central carbon metabolism enzymes
leads to an increased availability of NAD(P)H, FADH2,
GTP and ATP and of primary metabolism intermedi-
ates. In fact, pyruvate and acetyl-CoA, from glycolysis,
erytrose-4P, from PPP, and alpha-ketoacids from TCA
cycle are required for anabolic routes such as amino
acid biosynthesis (Figure 5). Interestingly, the upregula-
tion of central carbon metabolism enzymes does not
result in an increased biomass production yield, calcu-
lated on glucose consumption, in LP (Figure 1C). This
result is most probably due to Pi limitation that nega-
tively affects growth. On the contrary, the Pi proficiency
in LG is counterbalanced by the low glucose availability
that results in downregulation of central carbon metabo-
lism enzymes. Thus, the similar biomass production
yield in LP and LG conditions is achieved throughout a
differential expression of central carbon metabolism key
enzymes that provides for metabolic adaptation to ener-
getic imbalance due to Pi limitation in LP condition and
leads to an increased synthesis of primary metabolism
intermediates and cofactors eventually required for bio-
mass production and for balhimycin biosynthesis.
Fatty acid metabolism
Two long-chain-fatty-acid-CoA ligases, namely L-ACS1
and 2, were upregulated and downregulated in LP,
respectively (Figure 5; Additional file 1 Table 3S). Both
are involved in the reversible formation of an acyl-car-
rier protein in the metabolism of fatty acids, with a
mechanism leading to the release of PPi from ATP. In
addition, the enzyme enoyl-ACP reductase (FabI), cata-
lyzing the formation of trans-2,3-dehydroacyl-[acyl-car-
rier protein] and NADH from NAD+and enoyl-acyl-
[acyl-carrier-protein] C4-16 derivatives, was upregulated
in LP. This result is in agreement to Summers et al.,
(1999) [41] that revealed fabI being part of an operon
induced in Pi limited cultures of Sinorhizobium meliloti.
The downregulation of the glycerol kinase (GlpK) cata-
lyzing glycerol phosphorylation, suggests a decreased
synthesis of glycerolipids due to Pi limitation, which,
instead, could be used as phosphorous source, as it has
already been observed in Bacillus subtilis [42].
Oxidative stress and hypothetical proteins
In agreement with the increased expression of TCA
cycle enzymes, some upregulated proteins are involved
in oxidative phosphorylation or counterbalance reactive
oxidative species (ROS) generation and damage. In fact,
upregulation was revealed for NADH:ubiquinone oxi-
doreductase proteins (NuoE and NuoD), 4-hydroxy-3-
methylbut-2-en-1-yl diphosphate synthase (GcpE),
involved in the synthesis of isopentenyl-PP that is a pre-
cursor of CoQ, AhpC, a thiol-specific antioxidant pro-
tein using reducing equivalents derived from either
thioredoxin (TrxR) and glutathione, and TrxR, a small
protein that alters the redox state of target proteins
through the reversible oxidation of its active site (Figure
5; Additional file 1 Table 3S). These results are also in
agreement with the increment of oxidative stress that
has been already observed in E coli during Pi starvation
[43]. In addition, upregulation of ferritin family protein
(FFP) is in agreement with that it has already been
observed in S. coelicolor [16] and in S. meliloti [44] and
with a possible role in free iron detoxification for pro-
tection against ROS [45].
Hypothetical proteins (Additional file 1 Table 3S) that
were previously associated with balhimycin production
[23], such as a forkhead associated domain containing
protein (FDCP) and a MoxR-domain containing ATPase,
were also up-regulated in LP [46].
Protein biosynthesis and nucleic acid metabolism
A number of proteins related to protein biosynthesis or
protein folding and stress-damage response resulted
upregulated in LP (Additional file 1 Table 3S). This is
the case of elongation factor Tu (EF-Tu), trigger factor
(TF), peptidyl-prolyl cis-trans isomerase (PPI), a cold
shock-like protein (CSP-G), chaperone protein DnaK
and Clp protease ATP-binding subunit (ClpB). In con-
trast, 60 kDa chaperonin GroEL and 30 S ribosomal
protein S1 RpsA were downregulated under LP condi-
tions (Additional file 1 Table 3S), as reported also in S.
coelicolor [16], in B. subtilis [42] and in S. meliloti [44]
incubated in Pi limiting condition. This result corre-
lates with the chemostat cultivation conditions used,
where mycelia are in a proliferating state and in LP a
pool of upregulated factors, required for protein bio-
synthesis, may need to counterbalance energetic stress
due to Pi limitation. In fact, the upregulation of factors
necessary for either ribosomal activity or for protein
folding together with the concomitant downregulation
of the ribosomal protein RpsA, the largest ribosomal
component with a pivotal role in ribosome assembly
[47], may reflect the need to optimize the efficiency of
protein biosynthesis under Pi nutritional stress
conditions.
Gallo et al. Microbial Cell Factories 2010, 9:95
http://www.microbialcellfactories.com/content/9/1/95
Page 7 of 14

Page 8

Under Pi limitation, bacteria may reduce their nucleo-
side pool [16] and accordingly inosine 5’-phosphate
dehydrogenase (IMPDH), whose activity is required for
the biosynthesis of xanthosine-5’P, a precursor of both
guanosine and adenosine [48], was downregulated in LP.
Surprisingly, guanosine 5’-phosphate (GMP) synthase,
whose activity is associated with the formation of GMP
and PPi from xantosine-5’P, was instead upregulated LP.
This result might be explained since, as shown in KEGG
pathway database, GMP synthase substrate (xanthosine-
5’P) could derive from biochemical routes involving a
pool of xantine and xantosine-triP or-tetraPi, in a sal-
vage pathway that could be activated in Pi limitation.
Balhimycin production
The amino acids H-Tyr, HPG, DPG, Asn and Leu are
necessary to form the balhimycin heptapeptide back-
bone. Their availability from primary metabolism path-
ways is crucial for antibiotic biosynthesis [25]. Non-
proteinogenic amino acids H-Tyr and HPG are also
derived from metabolism of Tyr, which is also the
amino donor during DPG biosynthesis. Genes encoding
enzymes involved in the last steps of H-Tyr, HPG and
DPG biosynthesis are present in the bal cluster, which
also contain genes coding for enzymes required in the
metabolism of amino sugars, essential for antibiotic
glycosylation.
In agreement with balhimycin production observed
only in LP, primary and secondary metabolism enzymes
involved in the biosynthesis of balhimycin precursors,
such as amino acids and amino sugars resulted upregu-
lated in LP condition. In particular, the metabolic build-
ingblockspyruvate,
4P-erythrose and glucose-6P from central carbon meta-
bolism are required for Leu, DPG, Asn, Tyr, HPG and
amino sugar biosynthesis, respectively (Figure 5) [25].
Thus, bal gene product HmaS-involved in HPG synth-
esis and whose expression was revealed upregulated by
qRT-PCR-, 3-isopropylmalate dehydratase large subunit
(LeuC) and branched-chain amino acid aminotransferase
(IlvE)-catalyzing different reactions in Leu biosynthesis-,
MDH-catalyzing biosynthesis of Asp-precursor oxalace-
tate-, bal gene product Bgtf, phosphoglucosamine
mutase (GlmM) and UTP-glucose-1-phosphate uridylyl-
transferase (GalU)-required in the metabolism of amino
sugars-were upregulated in LP. GlmM is particularly
interesting because it catalyzes the formation of glucosa-
mine-1P, a metabolic intermediate that is converted into
UDP-glucosamine from GalU (Figure 5; Additional file 1
Table 3S). UDP-glucosamine is reported to be likely the
primary metabolism building block required for the bio-
synthesis of modified amino sugars in glycopeptide tei-
coplanin, chloroeremomycin, and balhimycin [49].
Interestingly, a 4-phosphopantetheinyl transferase (PPT)
acetyl-CoA, oxalacetate,
was upregulated in LP. PPTs form a superfamily of
enzymes that transfer prosthetic 4-phosphopantetheine
moiety from CoA to carrier domain of biosynthetic
complexes required for the synthesis of a wide range of
compounds including fatty acid, polyketide (PK) and
nonribosomal peptide (NRP) metabolites. As revealed by
BLAST analysis, A. balhimycina PPT possesses con-
served domains characteristic of Sfp-like phosphopan-
tetheinyl transferases family. Members of this family are
mainly found associated either with NRP synthetase and
PK synthase [50]. Thus, even if an involvement in fatty
acid biosynthesis could not be totally excluded, PPT
might likely be involved in both balhimycin heptapep-
tide assembly line and in the synthesis of DPG, a poly-
ketide derivative, synthesized by dpgABCD products
[51].
Altogether, these data suggest that the destination of
metabolic intermediates in LP and LG cultivations is dif-
ferentially regulated. In particular, the increment of glu-
cose catabolism in LP condition is coupled with the
synthesis of metabolic building blocks related to the
generation of both balhimycin backbone and amino
sugar moieties.
Identification of PHO box regulatory elements in the
upstream of differentially expressed genes
Global gene expression analysis was proven as a good
starting point to identify common regulatory elements,
such as PHO boxes, in the upstream region of differen-
tially expressed genes in S. coelicolor [16]. Similarly,
putative PHO box regulatory elements, upstream the
translational start site of differentially expressed genes,
were searched by sequence homology using published S.
coelicolor PHO box consensus [18]. Thus, PHO box
directed repeats (DRs) in the upstream region of the
genes encoding PstS, Ppk, PhoD, MDH, B-SCS, HmaS,
Gabt1, LpdA2 and P-PFK were revealed by bioinfor-
matics (Additional file 1 Table 7S and 8S). The consen-
sus of A. balhimycina DRs (Figure 6), created using
free-on line available WebLogo software http://weblogo.
Figure 6 Consensus of the direct repeats of 11 nt that forms
the A. balhimycina PHO box. This logo corresponds to a model
that comprises 19-DRs from nine A. balhimycina genes (Additional
file 1 Tables 6S and 7S). The height of each letter is proportional to
the frequency of the base; the height of the letter stack is the
conservation in bits at that position. Error bars are shown at the top
of the stacks.
Gallo et al. Microbial Cell Factories 2010, 9:95
http://www.microbialcellfactories.com/content/9/1/95
Page 8 of 14

Page 9

berkeley.edu/logo.cgi[52], shows high similarity with that
of S. coelicolor revealing a high conservation of the first
seven positions [18].
Experimental validation was carried out by electro-
phoresis mobility shift assay (EMSA) for the putative
PHO boxes in the upstream of pstS and ppk, which
were up-and downregulated in LP respectively, as
revealed by qRT-PCR. In particular, 40 and 38 bp DNA
fragments containing the upstream regions of A. balhi-
mycina pstS (PpstS Amy) and ppk (Pppk Amy), respec-
tively, and A. balhimycina crude extracts from biomass
generated in LP condition were used. This analysis
showed that both PpstS Amy and Pppk Amy specifically
bind to the crude extract (Figure 7). Notably, a 40 bp
DNA fragment containing the PHO box of pstS from S.
coelicolor (PpstS Sco) specifically competed to A. balhi-
mycina crude extract bound, resulting in reduction of
band shift intensity for both PpstS Amy and Pppk Amy.
This result, revealing an interspecific competition
between S. coelicolor and A. balhimycina PHO boxes,
suggests the binding of PhoP to these A. balhimycina
regulatory DNA elements. The presence of a PHO box
in the upstream region of both up-and downregulated
genes is in agreement with PhoP dual role as positive or
negative regulator [17-20]. Even if the expression of ppk
is positively and indirectly controlled by PhoP in S. livi-
dans TK24 [26], it is neither PhoP-or Pi-controlled in S.
coelicolor [16]. Instead, in A. balhimycina the putative
PHO box in the upstream region of ppk could be
explained by a negative control of PhoP over ppk
expression.
Thus, the presence of PHO box regulatory elements in
the upstream of differentially expressed genes suggests a
PhoP-mediated mechanism for the regulation of their
expression. Anyway, further investigations, such as the
construction of a A. balhimycina ΔphoP strains and
EMSAs carried out with a purified A. balhimycina
PhoP, should be performed to better clarify the role of
PHO box-like regulatory elements in controlling the
expression of A. balhimycina genes in Pi limitation.
Conclusions
This is the first report describing a differential proteo-
mic analysis for a glycopeptide producer strain incu-
bated in continuous cultivations. Comparative analyses
of global expression profiles in non-producing and pro-
ducing conditions in batch cultivations may be affected
by growth rate and changes of medium component con-
centrations. In the chemostat conditions which were
set-up in this study to compare A. balhimycina pro-
teomes in producing and non producing conditions,
mycelia grew with the same rate and with similar glu-
cose-biomass conversion coefficients.
Both transcriptional and proteomic analyses high-
lighted that in LP and LG conditions energy balance
and generation of primary metabolism intermediates are
mainly regulated by controlling the expression of central
carbon metabolism enzymes and proteins, such as PstS,
PhoD and an inorganic pyrophosphatase (PPA), whose
activity is required for Pi recovering. The upregulation
in LP of central carbon metabolism genes does not
result in an increased biomass production yield on glu-
cose consumption. Instead, the upregulation of catabolic
and anabolic enzymes, coupled with the upregulation of
bal genes, accounts for the supply of cofactors and pre-
cursors, such as amino acids and/or amino sugars, that
otherwise could eventually become limiting for balhimy-
cin biosynthesis.
Altogether these data correlate with the fact that in
Actinomycetes secondary metabolites are generally
synthesized through multiple intracellular reactions,
which are further affected by cofactor balance and regu-
latory circuits at different levels of cellular metabolism
[53-55]. Thus, altogether these data could be used to re-
design fermentation strategies that are difficult to be
intuitively identified and for approaching the prediction
of new genetic targets from primary metabolism genes
to be engineered for a rational construction of antibiotic
high-yielding producer strains.
Materials and methods
Strain and culture condition
The A. balhimycina DSM5908 strain used in this work
was a gift from Prof. Wohlleben, University of Tubin-
gen, Germany. The strain was stored in 1-ml cryotubes
Figure 7 Gel band shift assays. EMSA was carried out using DNA
fragments upstream A. balhimycina ppk (Pppk Amy) and pstS (PpstS
Amy) and a crude extract from A. balhimycina LP culture. The use of
a molar excess (200 fold) of unlabeled probe as well as of a DNA
fragment upstream S. coelicolor pstS (PpstS Sco), containing a PHO
box, reduced the band shift signals for both probes. No reduction
of band shift signals is visible using a molar excess (200 fold) of an
unspecific competitor (Nonomuraea Pdbv14 [14]).
Gallo et al. Microbial Cell Factories 2010, 9:95
http://www.microbialcellfactories.com/content/9/1/95
Page 9 of 14

Page 10

at -80°C in 15% glycerol and 8 g/l tryptic soy broth
(Difco, Detroit, Mich.), at a biomass dry weight of
approximately 0.2 g/ml. For seed culture preparations, a
250-ml baffled Erlenmeyer flask containing 50 ml of
tryptic soy broth (Difco) was inoculated with one gly-
cerol stock vial and incubated with shaking (150 rpm) at
30°C. After 48 h of incubation, this seed culture was
used for inoculation of the fermentors (5% v/v). The fer-
mentations were performed in 1 l double-jacketed
Applikon fermentors (Applikon, Shiedam, Netherlands)
containing 1 l medium with an agitation rate of 500
rpm and aeration at 1 vvm (air volume/working
volume/min). Dissolved O2tension was monitored with
an O2electrode (Mettler Toledo, Greifensee, Switzer-
land); it did not fall below 50% saturation during any of
the fermentations. The pH was kept at a value of 7.0 by
addition of 1 M NaOH; temperature was maintained at
30°C. The two fermentation media, named LG and LP
for low glucose and phosphate, respectively, contained
the compounds in the concentration reported in Addi-
tional file 1 Table 9S. LP and LG chemostat experiments
were performed in two parallel replicates, respectively,
using a constant dilution rate (DR) of 0.03 h-1. It took
about 3 and 5 residence times (RT, calculated as the
inverse of DR) to achieve the stabilization of growth
parameters (steady-state) in LG and LP cultivations
which was confirmed by the constant off-gas and bio-
mass concentration. The biomass samples used for total
protein and RNA extractions were harvested during
steady-state after about 6.7 and 7.9 RT from LG and LP
cultivations, respectively, and immediately frozen in
liquid N2and then stored at -80°C.
Analysis of biomass dry weight
For biomass concentration, sampling was performed
manually in triplicate from each cultivation. The optical
density was measured on a UV-vis spectrophotometer
UV mini 1240 from Shimadzu at 600 nm. For dry
weight measurements, 3 ml samples were filtered
through a pre-weight 45 mm pore size Sartorius filter
and washed with 0.9% NaCl (w/w). The filters (with
content) were dried 20 min at 150W in a microwave
oven and weighted after cooling.
Analysis of glucose and phosphorus utilization
Glucose concentration in spent medium was measured
by high performance liquid chromatography (HPLC)
(Agilent Technologies, Palo Alto, CA) equipped with an
AMINEX HPX-87 H column (Bio-Rad Laboratories,
Hercules, CA) working at 60 1C. 5 mM H2SO4was
used as mobile phase with a flow rate of 0.6 ml min-1.
Standard UV index detection at 210 nm was used for
quantification.
Phosphorus concentration in spent medium was ana-
lyzed using a spectrophotometric assay kit (Inorganic
phosphorus 80; Abx Diagnostics, Montpellier, France) in
an automatic analyzer (Cobas Miras plus; Roche, Basel,
Switzerland).
Analysis of CO2in exhaust gas
The partial pressure of CO2in the exhaust gas from the
bioreactors was measured using a gas analyzer (Indus-
trial emission monitor type 1311; Brüel & Kjaer,
Denmark).
Analysis of balhimycin
Balhimycin titers in the cultivations were detected and
quantified using liquid chromatography mass spectro-
metry (LCMS) as described elsewhere [25]. LC-DAD-
ESI+-MS data were acquired on an Agilent 1100 HPLC
system (Agilent Technologies, Waldbronn, Germany)
equipped with a diode array detector scanner. The
HPLC system was connected to an Agilent MSD Ion
trap operated in positive electrospray full scan mode
(m/z 100-1500) with 4 scans per second. The amount of
balhimycin was quantified using UV detection at 195
and 280 nm and calculated using a standard curve.
Separation was done on a Phenomenex Gemini C18
110A column (2 × 100 mm2, 3 μm) with a flow rate of
0.3 ml/min at 40°C using water (pH 10.5 with 10 mM
ammonium formate) and methanol solution in the fol-
lowing gradient: t = 0 min, 10% methanol; t = 24 min,
100% methanol; t = 30 min, 10% methanol.
Protein extraction and separation for 2D-Differential Gel
Electrophoresis (2D-DIGE) analysis
Frozen biomass samples, collected from two parallel cul-
tivations for each condition, were sonicated in non redu-
cing conditions using experimental procedures as
previously described [23]. After dialysis against distilled
water, at 4°C, and acetone precipitation at -20°C, proteins
were dissolved in the appropriate 2D-DIGE lysis buffer
(30 mM Tris, 7 M urea, 2 M thiourea, 4% CHAPS).The
four protein extracts (one for each replica per each condi-
tion) were analyzed in two technical replicates. In parti-
cular, two 40 μg protein aliquots from each extraction
were minimally labelled using 320 pmol Cy3 and 320
pmol Cy5 fluorescent dyes (CyDye™DIGE minimal label-
ling kit, GE Healthcare, Uppsala, Sweden), respectively, to
account for florescence bias. Labelling reactions were car-
ried out in the dark on ice for 30 min. and quenched
with 0.2 mM lysine, according to manufacturer instruc-
tions. In addition, four 40 μg protein aliquots from a
standard pool, generated by combining an equal amount
of all the four protein extracts, were minimally labelled
with 320 pmol Cy2 fluorescent dye (CyDye™DIGE, GE
Gallo et al. Microbial Cell Factories 2010, 9:95
http://www.microbialcellfactories.com/content/9/1/95
Page 10 of 14

Page 11

Healthcare) according to manufacturer instructions. To
perform 2D-DIGE analysis, 40 μg of LG and LP labelled
proteins were mixed, combining Cy3 and Cy5 fluorescent
dyes, with the addition of 40 μg of internal standard Cy2-
labelled proteins, thus generating a total of four protein
mixes.
For isoelectrofocusing (IEF), DeStreak rehydration
solution (GE Healthcare) containing 0.5% (v/v) IPG buf-
fer (GE Healthcare) and 1% (w/v) DTE (Sigma) was
added to each mix up to 340 μl (final volume). IEF was
performed as previously described [23] using 4-7 pH
range 18 cm-IPG strips (GE Healthcare) in an Ettan
IPGphor III apparatus (GE Healthcare). The focused
proteins were then separated using 12% sodium dodecyl
polyacrylamide gels (SDS-PAGE) at 10°C in a Hettan
Dalt six (GE Healthcare), with a maximum setting of 40
μA and 110 V per gel.
The four 2D-Gels were scanned with a DIGE imager
(GE Healthcare) to detect cyanin-labeled proteins
according to manufacturer’s instructions. Differential gel
analysis was performed automatically using Image Mas-
ter 2D Platinum 7.0 DIGE software (GE Healthcare),
according to the manufacturer’s instructions. Protein
spots were detected automatically and manually verified.
Individual spot abundance was automatically calculated
from the quadruplicated 2D-Gels as mean spot volume
(Vol, i.e. integration of optical density over spot area)
and normalized to the Cy2-labeled internal pooled stan-
dard. Protein spots showing more than 1.5 fold change
in Vol, with a statistically significant ANOVA value (p <
0.05), were considered differentially abundant and iden-
tified as described in the following section by either MS
analysis or by gel-matching procedures using A. balhi-
mycina protein 2D reference-maps previously obtained
[23,24].
MS analysis and protein identification
Protein spots were excised from the 2D-Gels, alkylated,
digested with trypsin and identified as previously
reported [23]. Peptide mixtures were desalted by μZip-
TipC18 (Millipore, MA) using 50% (v/v) acetonitrile/5%
(v/v) formic acid as eluent before MALDI-TOF-MS
and/or nLC-ESI-LIT-MS/MS analysis.
In the case of MALDI-TOF-MS experiments, peptide
mixtures were loaded on the MALDI target, using the
dried droplet technique and a-cyano-4-hydroxycinnamic
acid as matrix, and analyzed using a Voyager-DE PRO
mass spectrometer (Applied Biosystems, USA) operating
in positive ion reflectron, with an acceleration voltage of
20 kV, a nitrogen laser (337 nm) and a laser repetition
rate of 4 Hz [56]. The final mass spectra, measured over
a mass range of 700-6000 Da and by averaging 400-800
laser shots, were elaborated using the DataExplorer 5.1
software (Applied Biosystems) and manually inspected
to get the corresponding peak lists. Internal mass cali-
bration was performed with peptides deriving from tryp-
sin autoproteolysis.
Tryptic digests were eventually analyzed by nLC-ESI-
LIT-MS/MS using a LTQ XL mass spectrometer (Ther-
moFisher, San Jose, CA) equipped with a Proxeon
nanospray source connected to an Easy-nanoLC (Prox-
eon, Odense, Denmark) [57]. Peptide mixtures were
separated on an Easy C18column (10 × 0.075 mm, 3
μm) (Proxeon). Mobile phases were 0.1% (v/v) aqueous
formic acid (solvent A) and 0.1% (v/v) formic acid in
acetonitrile (solvent B), running at total flow rate of 300
nL/min. Linear gradient was initiated 20 min after sam-
ple loading; solvent B ramped from 5% to 35% over 45
min, from 35% to 60% over 10 min, and from 60% to
95% over 20 min. Spectra were acquired in the range
m/z 400-2000. Acquisition was controlled by a data-
dependent product ion scanning procedure over the
three most abundant ions, enabling dynamic exclusion
(repeat count 2 and exclusion duration 60 s); the mass
isolation window and collision energy were set to m/z 3
and 35%, respectively.
MASCOT search engine version 2.2.06 (Matrix
Science, UK) was used to identify protein spots unam-
biguously from an updated NCBI nonredundant data-
base also containing the A. balhimycina ORF product
database based on A. balhimycina DSM5908 genome
sequencing [23] by using MALDI-TOF-MS data, a mass
tolerance value of 40-80 ppm, trypsin as proteolytic
enzyme, a missed cleavages maximum value of 2 and
Cys carbamidomethylation and Met oxidation as fixed
and variable modification, respectively. Candidates with
a MASCOT score > 83 (corresponding to p < 0.05 for a
significant identification) were further evaluated by the
comparison with their calculated mass and pI values,
using the experimental values obtained from 2-DE.
Raw data files from nLC-ESI-LIT-MS/MS experiments
were searched with SEQUEST (ThermoFisher Scientific,
USA) within the Proteome Discoverer software package
(Thermo Fisher Scientific, San Jose, CA, USA, version
1.0 SP1) against the NCBI nonredundant database
implemented by the preliminary version of the A. balhi-
mycina DSM5908 ORF product database mentioned
above [23]. Database searching was performed by select-
ing Cys carbamidomethylation as a fixed modification
and Met oxidation as variable modification. Searches
were carried out by using a mass tolerance value of 2.0
Da for precursor ion and 0.8 Da for MS/MS fragments,
trypsin as proteolytic enzyme, a missed cleavages maxi-
mum value of 2. Other SEQUEST parameters were kept
as default. Candidates with more than 2 assigned pep-
tides with an individual SEQUEST score versus charge
state > 1.5 for charged state (CS) 1, > 2.0 for CS 2, > 2.4
for CS 3, > 3.3 for CS 4, > 4.2 for CS 5, > 4.5 for CS 6
Gallo et al. Microbial Cell Factories 2010, 9:95
http://www.microbialcellfactories.com/content/9/1/95
Page 11 of 14

Page 12

were considered confidently identified. Definitive peptide
assignment was always associated to manual spectral
visualization and verification.
The sequence of the identified proteins and homology
analysis, performed using on-line available versions of
BLAST search against UniProt Knowledgebase http://
www.expasy.ch/tools/blast, are reported in Additional
file 1 Table 3S and 4S.
Protein identification by gel-matching was automati-
cally performed using A. balhimycina protein 2D refer-
ence-maps [23], available over the World Wide Web as
interactive pages at http://www.unipa.it/ampuglia/Abal-
proteome-maps[24], and Image Master 2D platinum 7.0
software. More than 50 highly reproducible protein
spots were used as landmarks to perform automatic gel-
matching that was then verified by an accurate visual
inspection using 3 D view tool.
Total RNA isolation, RT-PCR and qRT-PCR analysis
Gene expression analysis at the transcriptional level of
selected Pho regulon and bal genes was performed
according to Gallo et al. 2010 [23]. The mycelia were
resuspended in 1 ml P-buffer [58] containing lysozyme
(1 mg/ml) and RNase inhibitor RNaseOUT (Invitrogen)
(40 U/ml), and then incubated for 10 min, at 37°C. The
RNA was extracted by using the RNeasy midi kit (Qia-
gen) according to the manufacturer’s instructions.
DNase I (Roche) treatment was performed at 37°C, for 1
h, and ethanol precipitation in the presence of 0.1 vol of
3 M sodium acetate allowed recovery of the DNase-
treated total RNA. After a washing step with 70% (v/v)
ethanol and air drying, the RNA pellet was resuspended
in water with RNase inhibitor RNaseOUT (200 U/ml). As
control of RNA quality, a RT-PCR with 0.1 μg of total
RNA and primer pairs internal to hrdB was carried out
using the Superscript One-Step RT-PCR kit (Invitrogen)
and the conditions indicated by the supplier. PCRs were
performed on 0.5 μg of RNA samples using 40 cycles
prior to exclude the presence of genomic DNA.
Primer pairs amplifying intragenic regions of the genes
analysed by qRT-PCR are listed in Table 10S (Addi-
tional file 1). The identity of RT-PCR products was con-
firmed by sequencing. The high-capacity cDNA archive
kit (Applied Biosystems) was used to retrotranscribe 2
μg of total RNA, extracted from LP and LG cultures, in
a 100 μl of water (final volume). Gene expression was
analyzed quantitatively by using Applied Biosystems
7300 real-time PCR system (Applied Biosystems). The
expression of hrdB[23] was used as an internal control
to quantify the relative expression of target genes. 2 μl
of cDNAs were mixed with 12.5 μl of SYBR green PCR
master mix (Applied Biosystem) and 10 pmol of each
primer in a final volume of 25 μl. The PCR was per-
formed under the following conditions: 2 min at 50°C
and 10 min at 95°C, followed by 40 cycles of 15 s at
95°C and 1 min at 60°C. A dissociation reaction was
eventually performed using a temperature gradient from
55 to 99°C by increasing 1°C/min. This procedure per-
mitted recording the melting curve of the PCR products
and, consequently, their specificity to be determined. A
negative control (distilled water) was included in all
real-time PCR assays, and each experiment was per-
formed in triplicate or quadruplicate.
Search for putative PHO box sequences
A. balhimycina PHO box sequences were searched by
matching two directs repeats (gttcacccggc and gttcatt-
tacg) of S. coelicolor PHO box in the upstream region of
pstS gene with regions extending 300 bp upstream of
the putative translation start sites of differentially
expressed A. balhimycina genes. To this aim, on-line
available versions of ClustalW http://www.ebi.ac.uk/
Tools/clustalw2/index.html[59] and EMBOSS GUI
matcher http://bips.u-strasbg.fr/EMBOSS/[60] and
BLAST bl2seq http://blast.ncbi.nlm.nih.gov/Blast.cgi[61]
software were used. Only outputs derived from all the
three approaches were accepted (Additional file 1 Table
6S and 7S).
Preparation of labeled DNA fragments
DNA fragments containing the upstream regions of S.
coelicolor pstS (PpstS: tccacaggggttcacccggcgttcatt-
tacgcccttcggc) and the A. balhimycina pstS (PpstS:
gaaaggcttgttcactttgcgttcatctggacaggggaac) and ppk
(Pppk: ggtatcgcgttgagctgttcatctgaccttcaccacgg) were
prepared by incubation of the corresponding oligonu-
cleotides at 90°C for 10 min, followed by slow cooling
to room temperature. The annealed products were
recovered from nondenaturing 20% polyacrylamide gels
by the crush-soak method [62] and labeled with T4
polynucleotide kinase (Invitrogen) according to the sup-
plier’s protocol.
Preparation of A. balhimycina crude extract
A. balhimycina pellet was washed twice with crack buf-
fer (10 mM Tris-HCl, pH 8.0, 0.5 mM EDTA, 0.3 mM
DTT), resuspended in 5 ml of crack buffer, and dis-
rupted by sonication. The cell debris was removed by
centrifugation at 13,000 × g (20 min, 4°C), and the
supernatant was stored at −80°C.
Gel mobility shift assay
The gel mobility shift assay was performed according to
Alduina et al., (2007) [14]. For the binding assay, A. bal-
himycina crude extract was dialyzed against distilled
water over night at 4°C and, then, approximately 200 μg
of proteins were incubated in 20 μl of 12.5 mM Tris-
HCl (pH 7.5), 10% glycerol, 62.5 mM KCl, 0.75 mM
Gallo et al. Microbial Cell Factories 2010, 9:95
http://www.microbialcellfactories.com/content/9/1/95
Page 12 of 14

Page 13

DTT, and 5 mM MgCl2, containing 100 μg of poly(dI-
dC)-poly(dI-dC) ml-1, for 10 min, at 4°C. After 15 min
of incubation with 0.4 ng of32P-labeled DNA, com-
plexes and free DNA were resolved on non-denaturing
5% polyacrylamide gels run in 0.5× Tris-borate-EDTA
buffer at 150 V for approximately 2 h [54] and then
equilibrated in 10% acetic acid, dried, and subjected to
autoradiography. For testing the specificity of binding,
either unlabeled probe or a competitor DNA fragment
containing the upstream region of Nonomuraea sp
ATCC 39727 dbv14 [14] were added before incubation
of the proteins and probe.
Declaration of competing interests
The authors declare that they have no competing interests.
Additional material
Additional file 1: .pdf contains 10 tables reporting: - amino acid
sequence of A. balhimycina Pho regulon gene products (Table 1S);-BLAST
analysis of PHO regulon gene products against SwissProt database (Table
2S);-list of differential expressed proteins with information about their
relative function, relative expression value, either theoretical and
measured values for molecular weight (Mw) and isoelectric point (pI),
protein identification method (Table 3S);-amino acid sequence of the MS-
identified A. balhimycina proteins (Table 4S);-BLAST analysis data,
obtained by using UniProt databank of proteins identified by MS analysis
(Table 5S);-list of A. balhimycina DSM5908 genes arranged in putative
operons (Table 6S);-sequence of upstream regions of selected A.
balhimycina genes, showing PHO box directed repeats (DR) identified by
ClustalW and BLAST bl2seq analysis performed by using S. coelicolor PHO
box DR in the upstream regions of pstS (Table 7S);-EMBOSS-GUI Matcher
analysis of A. balhimycina PHO box DR sequence performed by using S.
coelicolor PHO box DR in the upstream regions of pstS (Table 8S);-
composition of fermentation media (Table 9S);-list of primers used for
qRT-PCR experiments (Table 10S).
Abbreviations
BDW: biomass dry weight. DPG: 3,5-dihydroxyphenylglycine. HPG: 4-
hydroxyphenylglycine. H-Tyr: b-hydroxytyrosine. LG: limiting glucose and
high phosphate. LP: low phosphate and proficient glucose. MS: mass
spectrometry. Pi: inorganic phosphate.
Acknowledgements
The authors acknowledge Wolfgang Wohlleben, Efthimia Stegmann and
Tilmann Weber, for giving us the possibility to access to the Amycolatopsis
balhimycina DSM5908 genome sequence and ORF product databases. The
authors also thank Fabio Sangiorgi for his contribution to the ‘’Amycolatopsis
balhimycina Proteomic Project’’ web pages and Maurizio Noto for technical
assistance.
This study was partly supported by grants from the European Union (LSHB-
CT-2003-503491 and LSHM-CT-2004-005224), MIUR (ex 60%) to AMP and
Rete di Spettrometria di Massa RESMAC (Regione Campania) to AS.
Author details
1Università di Palermo, Dipartimento di Biologia Cellulare e dello Sviluppo,
Viale delle Scienze, Parco d’Orleans II, 90128 Palermo, Italy.2Proteomics &
Mass Spectrometry Laboratory, ISPAAM, National Research Council, 80147
Naples, Italy.3Center for Microbial Biotechnology, Department of Systems
Biology, Technical University of Denmark, Denmark.
Authors’ contributions
GG carried out chemostat cultivations, DIGE analysis, protein identification by
gel-matching, real-time RT-PCR and EMSA experiments, BLAST analyses and
wrote the draft manuscript. RA helped to perform real-time RT-PCR and
EMSA experiments and to wrote the draft manuscript. GR carried out protein
MS-identification. JT designed chemostat experiments, helped to perform
chemostat cultivations and revised the manuscript. LB helped to perform
protein MS-identification. AEL supervised chemostat experiments and revised
the manuscript. AS supervised protein MS-identification and revised the
manuscript. AMP conceived and supervised the study and participated in its
design and coordination and revised the manuscript.
All authors read and approved the final manuscript.
Received: 12 August 2010 Accepted: 26 November 2010
Published: 26 November 2010
References
1.Finch R, Eliopoulos G: Safety and efficacy of glycopeptide antibiotics. J
Antimicrob Chemother 2005, 55(Suppl 2):ii5-13.
2.Nadkarni S, Patel M, Chatterjee S, Vijayakumar E, Desikan K, Blumbach J,
Ganguli B, Limbert M: Balhimycin, a new glycopeptide antibiotic
produced by Amycolatopsis sp. Y-86,21022. Taxonomy, production,
isolation and biological activity. J Antibiot (Tokyo) 1994, 47(3):334-341.
3.Chatterjee D, Sarma P, Jani R, Klesel N, Isert D: Comparative
chemotherapeutic efficacy of balhimycin, desgluco-balhimycin against
experimental MSSA and MRSA infection in mice. Indian J Exp Biol 2000,
38(7):681-686.
4.Pelzer S, Süssmuth R, Heckmann D, Recktenwald J, Huber P, Jung G,
Wohlleben W: Identification and analysis of the balhimycin biosynthetic
gene cluster and its use for manipulating glycopeptide biosynthesis in
Amycolatopsis mediterranei DSM5908. Antimicrob Agents Chemother 1999,
43(7):1565-1573.
5.Shawky R, Puk O, Wietzorrek A, Pelzer S, Takano E, Wohlleben W,
Stegmann E: The border sequence of the balhimycin biosynthesis gene
cluster from Amycolatopsis balhimycina contains bbr, encoding a StrR-
like pathway-specific regulator. J Mol Microbiol Biotechnol 2007, 13(1-
3):76-88.
6. Süssmuth R, Wohlleben W: The biosynthesis of glycopeptide antibiotics-a
model for complex, non-ribosomally synthesized, peptidic secondary
metabolites. Appl Microbiol Biotechnol 2004, 63(4):344-350.
7.Demain A: Induction of microbial secondary metabolism. Int Microbiol
1998, 1(4):259-264.
8. Chakraburtty R, Bibb M: The ppGpp synthetase gene (relA) of
Streptomyces coelicolor A3(2) plays a conditional role in antibiotic
production and morphological differentiation. J Bacteriol 1997,
179(18):5854-5861.
9.Martín J, Liras P: Engineering of regulatory cascades and networks
controlling antibiotic biosynthesis in Streptomyces. Curr Opin Microbiol
2010, 13(3):263-273.
10.McIntyre J, Bull A, Bunch A: Vancomycin production in batch and
continuous culture. Biotechnol Bioeng 1996, 49(4):412-420.
11.Ayar-Kayali H, Tarhan L: Vancomycin antibiotic production and TCA-
glyoxalate pathways depending on the glucose concentration in
Amycolatopsis orientalis. Enzyme Microbiol Technol 2006, 38:727-734.
12.Maiti S, Singh K, Lantz A, Bhushan M, Wangikar P: Substrate uptake,
phosphorus repression, and effect of seed culture on glycopeptide
antibiotic production: process model development and experimental
validation. Biotechnol Bioeng 2010, 105(1):109-120.
13.Gunnarsson N, Bruheim P, Nielsen J: Production of the glycopeptide
antibiotic A40926 by Nonomuraea sp. ATCC 39727: influence of medium
composition in batch fermentation. J Ind Microbiol Biotechnol 2003,
30(3):150-156.
14. Alduina R, Lo Piccolo L, D’Alia D, Ferraro C, Gunnarsson N, Donadio S,
Puglia A: Phosphate-controlled regulator for the biosynthesis of the
dalbavancin precursor A40926. J Bacteriol 2007, 189(22):8120-8129.
15.Sola-Landa A, Moura R, Martín J: The two-component PhoR-PhoP system
controls both primary metabolism and secondary metabolite
biosynthesis in Streptomyces lividans. Proc Natl Acad Sci USA 2003,
100(10):6133-6138.
Gallo et al. Microbial Cell Factories 2010, 9:95
http://www.microbialcellfactories.com/content/9/1/95
Page 13 of 14

Page 14

16.Rodríguez-García A, Barreiro C, Santos-Beneit F, Sola-Landa A, Martín J:
Genome-wide transcriptomic and proteomic analysis of the primary
response to phosphate limitation in Streptomyces coelicolor M145 and in
a ΔphoP mutant. Proteomics 2007, 7(14):2410-2429.
Sola-Landa A, Rodríguez-García A, Franco-Domínguez E, Martín J: Binding
of PhoP to promoters of phosphate-regulated genes in Streptomyces
coelicolor: identification of PHO boxes. Mol Microbiol 2005,
56(5):1373-1385.
Sola-Landa A, Rodríguez-García A, Apel A, Martín J: Target genes and
structure of the direct repeats in the DNA-binding sequences of the
response regulator PhoP in Streptomyces coelicolor. Nucleic Acids Res 2008,
36(4):1358-1368.
Apel A, Sola-Landa A, Rodríguez-García A, Martín J: Phosphate control of
phoA, phoC and phoD gene expression in Streptomyces coelicolor reveals
significant differences in binding of PhoP to their promoter regions.
Microbiology 2007, 153(Pt 10):3527-3537.
Makarewicz O, Dubrac S, Msadek T, Borriss R: Dual role of the PhoP
approximately P response regulator Bacillus amyloliquefaciens FZB45
phytase gene transcription is directed by positive and negative
interactions with the phyC promoter. J Bacteriol 2006, 188(19):6953-6965.
Papini M, Salazar M, Nielsen J: Systems Biology of Industrial
Microorganisms. Adv Biochem Eng Biotechnol 2010, 120:51-99.
Park J, Lee S, Kim T, Kim H: Application of systems biology for bioprocess
development. Trends Biotechnol 2008, 26(8):404-412.
Gallo G, Renzone G, Alduina R, Stegmann E, Weber T, Lantz A, Thykaer J,
Sangiorgi F, Scaloni A, Puglia A: Differential proteomic analysis reveals
novel links between primary metabolism and antibiotic production in
Amycolatopsis balhimycina. Proteomics 2010, 10(7):1336-1358.
Amycolatopsis balhimycina Proteomic Project. [http://www.unipa.it/
ampuglia/Abal-proteome-maps].
Thykaer J, Nielsen J, Wohlleben W, Weber T, Gutknecht M, Lantz A,
Stegmann E: Increased glycopeptide production after overexpression of
shikimate pathway genes being part of the balhimycin biosynthetic
gene cluster. Metab Eng 2010, 12(5):455-6.
Ghorbel S, Smirnov A, Chouayekh H, Sperandio B, Esnault C, Kormanec J,
Virolle M: Regulation of ppk expression and in vivo function of Ppk in
Streptomyces lividans TK24. J Bacteriol 2006, 188(17):6269-6276.
Chouayekh H, Virolle M: The polyphosphate kinase plays a negative role
in the control of antibiotic production in Streptomyces lividans. Mol
Microbiol 2002, 43(4):919-930.
ExPASy Proteomic server. [http://expasy.org/].
Shi X, Rao N, Kornberg A: Inorganic polyphosphate in Bacillus cereus:
motility, biofilm formation, and sporulation. Proc Natl Acad Sci USA 2004,
101(49):17061-17065.
Zhang H, Rao N, Shiba T, Kornberg A: Inorganic polyphosphate in the
social life of Myxococcus xanthus: motility, development, and predation.
Proc Natl Acad Sci USA 2005, 102(38):13416-13420.
Walsh C, Fisher S, Park I, Prahalad M, Wu Z: Bacterial resistance to
vancomycin: five genes and one missing hydrogen bond tell the story.
Chem Biol 1996, 3(1):21-28[http://biocyc.org/], BioCyc Database Collection.
BioCyc Database Collection. [http://biocyc.org/].
KEGG: Kyoto Encyclopedia of Genes and Genomes.[http://www.genome.
jp/kegg/].
Copp J, Neilan B: The phosphopantetheinyl transferase superfamily:
phylogenetic analysis and functional implications in cyanobacteria. Appl
Environ Microbiol 2006, 72(4):2298-2305.
Matsubara K, Ohnishi K, Kiritani K: Nucleotide sequences and
characterization of liv genes encoding components of the high-affinity
branched-chain amino acid transport system in Salmonella typhimurium.
J Biochem 1992, 112(1):93-101.
Bordignon E, Grote M, Schneider E: The maltose ATP-binding cassette
transporter in the 21st century–towards a structural dynamic
perspective on its mode of action. Mol Microbiol 2010, 77(6):1354-1366.
Gunnarsson N, Bruheim P, Nielsen J: Glucose metabolism in the antibiotic
producing actinomycete Nonomuraea sp. ATCC 39727. Biotechnol Bioeng
2004, 88(5):652-663.
Muschko K, Kienzlen G, Fiedler H, Wohlleben W, Schwartz D: Tricarboxylic
acid cycle aconitase activity during the life cycle of Streptomyces
viridochromogenesTü494. Arch Microbiol 2002, 178(6):499-505.
Schwartz D, Kaspar S, Kienzlen G, Muschko K, Wohlleben W: Inactivation of
the tricarboxylic acid cycle aconitase gene from Streptomyces
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
viridochromogenes Tü494 impairs morphological and physiological
differentiation. J Bacteriol 1999, 181(22):7131-7135.
Viollier P, Nguyen K, Minas W, Folcher M, Dale G, Thompson C: Roles of
aconitase in growth, metabolism, and morphological differentiation of
Streptomyces coelicolor. J Bacteriol 2001, 183(10):3193-3203.
Summers M, Denton M, McDermott T: Genes coding for
phosphotransacetylase and acetate kinase in Sinorhizobium meliloti are
in an operon that is inducible by phosphate stress and controlled by
phoB. J Bacteriol 1999, 181(7):2217-2224.
Antelmann H, Scharf C, Hecker M: Phosphate starvation-inducible proteins
of Bacillus subtilis: proteomics and transcriptional analysis. J Bacteriol
2000, 182(16):4478-4490.
Moreau P, Gérard F, Lutz N, Cozzone P: Non-growing Escherichia coli cells
starved for glucose or phosphate use different mechanisms to survive
oxidative stress. Mol Microbiol 2001, 39(4):1048-1060.
Krol E, Becker A: Global transcriptional analysis of the phosphate
starvation response in Sinorhizobium meliloti strains 1021 and 2011. Mol
Genet Genomics 2004, 272(1):1-17.
Carrondo M: Ferritins, iron uptake and storage from the bacterioferritin
viewpoint. EMBO J 2003, 22(9):1959-1968.
Snider J, Houry W: MoxR AAA+ ATPases: a novel family of molecular
chaperones? J Struct Biol 2006, 156(1):200-209.
Sengupta J, Agrawal R, Frank J: Visualization of protein S1 within the 30S
ribosomal subunit and its interaction with messenger RNA. Proc Natl
Acad Sci USA 2001, 98(21):11991-11996.
Gilbert H, Lowe C, Drabble W: Inosine 5’-monophosphate dehydrogenase
of Escherichia coli. Purification by affinity chromatography, subunit
structure and inhibition by guanosine 5’-monophosphate. Biochem J
1979, 183(3):481-494.
Kudo F, Kawabe K, Kuriki H, Eguchi T, Kakinuma K: A new family of
glucose-1-phosphate/glucosamine-1-phosphate nucleotidylyltransferase
in the biosynthetic pathways for antibiotics. J Am Chem Soc 2005,
127:1711-1718.
Copp J, Neilan B: The phosphopantetheinyl transferase superfamily:
phylogenetic analysis and functional implications in cyanobacteria. Appl
Environ Microbiol 2006, 72(4):2298-2305.
Hubbard B, Walsh C: Vancomycin assembly: nature’s way. Angew Chem Int
Ed Engl 2003, 42(7):730-765.
WebLogo. [http://weblogo.berkeley.edu/].
Borodina I, Siebring J, Zhang J, Smith C, van Keulen G, Dijkhuizen L,
Nielsen J: Antibiotic overproduction in Streptomyces coelicolor A3 2
mediated by phosphofructokinase deletion. J Biol Chem 2008,
283(37):25186-25199.
Butler M, Bruheim P, Jovetic S, Marinelli F, Postma P, Bibb M: Engineering
of primary carbon metabolism for improved antibiotic production in
Streptomyces lividans. Appl Environ Microbiol 2002, 68(10):4731-4739.
Avignone Rossa C, White J, Kuiper A, Postma P, Bibb M, Teixeira de
Mattos M: Carbon flux distribution in antibiotic-producing chemostat
cultures of Streptomyces lividans. Metab Eng 2002, 4(2):138-150.
D’Ambrosio C, Arena S, Salzano A, Renzone G, Ledda L, Scaloni A: A
proteomic characterization of water buffalo milk fractions describing
PTM of major species and the identification of minor components
involved in nutrient delivery and defense against pathogens. Proteomics
2008, 8(17):3657-3666.
Scippa G, Rocco M, Ialicicco M, Trupiano D, Viscosi V, Di Michele M,
Arena S, Chiatante D, Scaloni A: The proteome of lentil (Lens culinaris
Medik.) seeds: discriminating between landraces. Electrophoresis 2010,
31(3):497-506.
Kieser T, Bibb MJ, Buttner MJ, Chater KF, Hopwood DA: Practical
Streptomyces genetics. John Innes Foundation, Norwich, United Kingdom;
2000.
ClustalW. [http://www.ebi.ac.uk/Tools/clustalw2/index.html].
EMBOSS GUI. [http://bips.u-strasbg.fr/EMBOSS/].
BLAST. [http://blast.ncbi.nlm.nih.gov/Blast.cgi].
Sambrook J, Fritsch EF, Maniatis T: Molecular cloning: a laboratory manual.
Cold Spring Harbor Laboratory, New York, NY; 1982.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
doi:10.1186/1475-2859-9-95
Cite this article as: Gallo et al.: Differential proteomic analysis highlights
metabolic strategies associated with balhimycin production in
Amycolatopsis balhimycina chemostat cultivations. Microbial Cell Factories
2010 9:95.
Gallo et al. Microbial Cell Factories 2010, 9:95
http://www.microbialcellfactories.com/content/9/1/95
Page 14 of 14