The Zinc-Responsive Regulon of Neisseria meningitidis Comprises 17 Genes under Control of a Zur Element

Abstract

Zinc is a bivalent cation essential for bacterial growth and metabolism. The human pathogen Neisseria meningitidis expresses a homologue of the Zinc uptake regulator Zur, which has been postulated to repress the putative zinc uptake protein
ZnuD. In this study, we elucidated the transcriptome of meningococci in response to zinc by microarrays and quantitative real-time
PCR (qRT-PCR). We identified 15 genes that were repressed and two genes that were activated upon zinc addition. All transcription
units (genes and operons) harbored a putative Zur binding motif in their promoter regions. A meningococcal Zur binding consensus
motif (Zur box) was deduced in silico, which harbors a conserved central palindrome consisting of hexameric inverted repeats separated by three nucleotides (TGTTATDNHATAACA). In vitro binding of recombinant meningococcal Zur to this Zur box was shown for the first time using electrophoretic mobility shift
assays. Zur binding to DNA depended specifically on the presence of zinc and was sensitive to mutations in the palindromic
sequence. The Zur regulon among genes of unknown function comprised genes involved in zinc uptake, tRNA modification, and
ribosomal assembly. In summary, this is the first study of the transcriptional response to zinc in meningococci.

Full text

The Zinc-Responsive Regulon of Neisseria meningitidis Comprises 17
Genes under Control of a Zur Element
Marie-Christin Pawlik, Kerstin Hubert, Biju Joseph, Heike Claus, Christoph Schoen, and Ulrich Vogel
University of Würzburg, Institute for Hygiene and Microbiology, Würzburg, Germany
Zinc is a bivalent cation essential for bacterial growth and metabolism. The human pathogen Neisseria meningitidis expresses a
homologue of the Zinc uptake regulator Zur, which has been postulated to repress the putative zinc uptake protein ZnuD. In this
study, we elucidated the transcriptome of meningococci in response to zinc by microarrays and quantitative real-time PCR
(qRT-PCR). We identified 15 genes that were repressed and two genes that were activated upon zinc addition. All transcription
units (genes and operons) harbored a putative Zur binding motif in their promoter regions. A meningococcal Zur binding con-
sensus motif (Zur box) was deduced in silico, which harbors a conserved central palindrome consisting of hexameric inverted
repeats separated by three nucleotides (TGTTATDNHATAACA). In vitro binding of recombinant meningococcal Zur to this Zur
box was shown for the first time using electrophoretic mobility shift assays. Zur binding to DNA depended specifically on the
presence of zinc and was sensitive to mutations in the palindromic sequence. The Zur regulon among genes of unknown func-
tion comprised genes involved in zinc uptake, tRNA modification, and ribosomal assembly. In summary, this is the first study of
the transcriptional response to zinc in meningococci.
Trace metals, such as zinc, are essential cofactors of many en-
zymes and DNA-binding proteins (40). On the one hand, bac-
teria have to cope with decreased zinc availability during infection
in the host (25), while on the other, at high concentrations, zinc
can lead to the production of toxic reactive oxygen species (49).
Hence, bacteria tightly control metal homeostasis by metalloregu-
latory proteins, which are specialized metal-sensing transcrip-
tional regulators that change their conformation upon binding of
the metal ions (5). Zinc uptake regulators (Zur) belong to the Fur
protein family of transcriptional regulators that also includes Fur,
Mur, and Nur, which are sensors of iron, manganese, and nickel,
respectively (28). Zinc uptake systems and their regulation by Zur
have been characterized for several bacteria such as Escherichia coli
(45), Bacillus subtilis (13), Listeria monocytogenes (7), and Staph-
ylococcus aureus (31). Furthermore, Zur regulons have been de-
scribed for B. subtilis (14), Mycobacterium tuberculosis (32), Yer-
sinia pestis (29), and Corynebacterium glutamicum (54). Panina et
al. identified the promoter binding motifs for Zur for Gammapro-
teobacteria, the Agrobacteria group, the Rhodococcus group, and
Gram-positive bacteria as well as for the streptococcal zinc repres-
sor AdcR using comparative computational analysis (41).
In this study, gene regulation by zinc exposure was analyzed in
the commensal Gram-negative human pathogen Neisseria menin-
gitidis, causing septicemia and meningitis (50). Several genomes
of N. meningitidis have been sequenced (4,43,60), which allowed
for the bioinformatic prediction of a Zur binding motif for Beta-
proteobacteria in the RegPrecise database (39). Until now, there is
experimental evidence for regulation by meningococcal Zur only
for the TonB-dependent outer membrane receptor, ZnuD, which
is involved in zinc acquisition upon zinc limitation (59). Stork et
al. showed that expression of znuC and znuD is enhanced in a zur
knockout mutant and predicted Zur binding motifs upstream of
both genes based on the E. coli Zur binding sequence (59). Kumar
et al. subsequently demonstrated that N. meningitidis znuD is also
regulated by Fur and allows for heme capture on the cell surface
when expressed in E. coli (26). Yet, even for znuD, binding of Zur
to the promoter of Zur-regulated genes has not been demon-
strated by in vitro experiments.
Here, we characterized the Zur regulon of N. meningitidis. The
transcriptomes of strain MC58 grown under high- and low-zinc
conditions were compared. A total of 15 Zur-repressed and two
Zur-activated genes were found. We established a meningococcal
Zur binding motif (Zur box) based on promoter sequences of
these genes. The direct binding of Zur to proposed Zur boxes was
verified by electrophoretic mobility shift assays (EMSAs) and
shown to be zinc dependent. Our results provide the basis for
further studies characterizing the molecular mechanisms of zinc
adaptation in meningococci.
MATERIALS AND METHODS
Strains and mutants. Serogroup B N. meningitidis strain MC58 (36) was
kindly provided by Richard Moxon (Oxford). The zur gene (nmb1266)of
MC58 was replaced with a kanamycin resistance cassette as follows: a
470-bp DNA fragment upstream of zur was amplified by PCR from strain
MC58 with primers MP16 and MP17, introducing restriction sites of XbaI
and EcoRI, respectively. Likewise, primers MP18 and MP19, introducing
restriction sites of EcoRI and XhoI, respectively, were used to amplify 475
bp downstream of zur. Primer sequences are provided in Table 1. The
upstream and downstream fragments were digested with XbaI/EcoRI and
EcoRI/XhoI, respectively. A kanamycin resistance cassette was obtained
from vector pUC4K (GE Healthcare) by EcoRI digestion. The flanking
upstream and downstream fragments and the kanamycin cassette were
cloned into the XbaI/XhoI-digested expression vector pBluescript II SK
(Stratagene). The resulting plasmid pMP5 was transformed into MC58.
Kanamycin-resistant clones were screened by PCR for replacement of zur
by the kanamycin cassette, and zur deletion was confirmed by Southern
Received 20 June 2012 Accepted 20 September 2012
Published ahead of print 5 October 2012
Address correspondence to Ulrich Vogel, uvogel@hygiene.uni-wuerzburg.de.
Copyright © 2012, American Society for Microbiology. All Rights Reserved.
doi:10.1128/JB.01091-12
6594 jb.asm.org Journal of Bacteriology p. 6594 – 6603 December 2012 Volume 194 Number 23

blotting and sequencing. The resulting zur knockout strain was designated
WUE4812.
Bacterial growth and RNA isolation. Strains MC58 and MC58 ⌬zur
were grown overnight at 37°C and 5% CO
2
on GC agar plates (BD).
Colonies of each overnight culture were resuspended in 20 ml of RPMI
medium (Invitrogen) to an optical density at 600 nm (OD
600
) of 0.2 (2 ⫻
10
8
CFU/ml). The medium was supplemented with 100 ␮M FeCl
3
to
improve bacterial growth. The cultures were incubated at 37°C with shak-
ing at 200 rpm and grown to mid-log phase (OD
600
, 1.0). Both liquid
cultures were then diluted to an OD
600
of 0.2 with RPMI medium plus 100
␮M FeCl
3
. The MC58 culture was consecutively divided into two parts of
20 ml each. To one culture vial, 0.5 ␮M ZnSO
4
was added to establish a
high-zinc condition according to Stork et al. (59). The other culture and
the MC58 ⌬zur culture were further grown without zinc supplementa-
tion. All three 20-ml cultures were grown at 37°C and 200 rpm for 2 h,
resulting in an OD
600
of about 1.0. Bacterial cultures were then mixed with
twice the volume of RNAprotect Bacteria Reagent (Qiagen) to minimize
RNA degradation. Total RNA was isolated using the RNeasy Mini Kit
(Qiagen). Remaining DNA was digested with RNase-free recombinant
DNase I (Invitrogen), and RNA samples were stored at ⫺80°C. RNA
integrity was verified with the Agilent 2100 Bioanalyzer (Agilent Technol-
ogies).
TABLE 1 Oligonucleotides used in this study for PCR, qRT-PCR, and EMSA
a
Assay and target gene Forward Sequence (5=–3=) Reverse Sequence (5=–3=)
Amplicon
length
(bp)
PCR
zur up (nmb1267) MP16 GCGCGCTCTAGAGATGGCGG
AATACATTTTG
MP17 GCGCGCGAATTCTTCTCTGTT
TATGCCGTCTG
470
zur down
(nmb1264-1265)
MP18 GCGCGCGAATTCATAAGCC
TTTCGAAAGGAGC
MP19 GCGCGCCTCGAGCGTT
TTCACCGATAAGGAAC
475
zur (nmb1266) MP183 GCGCGCTCATGAAAACAAATTT
CAAACAGAAAATTA
MP184 GCGCGCAGATCTCTGCTGACAT
TTTTTACAAATC
469
qRT-PCR
nmb0317 MP141 GCAAATCCCTGAAACTCTACCTCTT MP142 TGATGTTGACGCAGTCTTCATG 72
nmb0525 MP143 CGGCGTTCCGAATACCTTT MP144 GCGTAAATCGCGGCATAGAG 66
nmb0546 MP165 CTGCCGCCGGAATCG MP166 GACCAAAGAGCCGACCACTTC 75
nmb0577 MP179 CGGGCGCGGATAACG MP180 AGGCAGACGCTCCGCATA 61
nmb0586 MP145 GCAACTACCAAATGCAGCTCAA MP146 CAGCAGGGACGGCATTAAAT 69
nmb0587 MP147 GGCTTTGGCACGTCCTCTT MP148 GTGTGCCGAGGGCTTGAA 72
nmb0588 MP149 TCGCCCGCGAAAAAATT MP150 CTTGGGCGAGGTAGGGTTCT 66
nmb0817 MP151 AGGATGTCCTGAATGTACTCGAAAT MP152 CTGTCTCTGTGTCCTCCAACCTAA 80
nmb0819 MP153 CCCTCCTCATCCTCGACACA MP154 CAGCACCTGCGGCAGTAA 69
nmb0820 MP155 CCGCCAAAGCCCTAAACA MP156 GGTTGTCGGGATTTGAACGT 72
nmb0942 MP157 CCGCTGTTTTCGCTGGATA MP158 GTGTTGACGTTGCGCTGTTT 71
nmb0964 MP117 CCGTTCCCCGGTTTTGA MP118 CTGCATCGCCTGCTTTTTC 80
nmb0990 MP171 TGCCATTATCGCGCTTGTC MP172 GCCTGCTGCTTCGCAAAT 81
nmb1475 MP159 TCGGACAAAACTTGGAAATCG MP160 TTTGAAGCGTTCGCCTACGT 68
nmb1497 MP161 TGCCCAACATCCAAGAAATGT MP162 CTGGTTTTAAGGCGGTGTGAA 68
nmb2142 MP163 ACGACGGCGGTCATCTTTAC MP164 CGCCGTATGATGCACCATT 68
EMSA
nmb0546 MP191 CTTTCCAAGATGTTATAATATAACA
TATAATCTAT
MP192 AAATATAGATTATATGTTATAT
TATAACATCTTGG
35
nmb0964
Unaltered MP185 TAAAAAATGTAATGTTATATAATAA
CAAACTTTT
MP186 TTTCAAAAGTTTGTTATTATAT
AACATTACATTT
34
TA/CA MP193 TAAAAAATGTAATGTTATATAACGA
TGAACTTTT
MP194 TTTCAAAAGTTCATCGTTATAT
AACATTACATTT
34
CA MP195 TAAAAAATGTAATGTTATATAATAA
TGAACTTTT
MP196 TTTCAAAAGTTCATTATTATAT
AACATTACATTT
34
TA MP197 TAAAAAATGTAATGTTATATAACGA
CAAACTTTT
MP198 TTTCAAAAGTTTGTCGTTATAT
AACATTACATTT
34
C MP199 TAAAAAATGTAATGTTATATAATAA
TAAACTTTT
MP200 TTTCAAAAGTTTATTATTATAT
AACATTACATTT
34
nmb1475
Short MP187 CGATACCTATTTTGTTATAACATAAC
AAAATCTT
MP188 GTTAAAGATTTTGTTATGTTAT
AACAAAATAGGT
34
Long MP189 TCTTCACACGATACCTATTTTGTTATA
ACATAACAAAATCTTTAACCCACA
MP190 CTCGTGTGGGTTAAAGATTTTG
TTATGTTATAACAAAATAGG
TATCGTGTG
51
a
Underlined bases indicate restriction sites.
Zinc-Responsive Regulon of Meningococci
December 2012 Volume 194 Number 23 jb.asm.org 6595

According to Invitrogen’s data sheet, the RPMI medium used here
does not contain any source of zinc. However, we assumed that this RPMI
contains a minimal concentration of zinc because RPMI distributed by
Sigma was shown to comprise at least ⬃1.69 ␮M zinc (59). We therefore
refer to “low” (RPMI) and “high” (RPMI⫹ZnSO
4
) zinc conditions in this
study.
Furthermore, we decided against depletion of zinc, e.g., by an ion
chelator such as N,N,N=,N=-tetrakis-(2-pyridylmethyl)-ethylenediamine
(TPEN), prior to zinc supplementation as conducted in studies of iron
response (18,56) because the probable chelation of ions other than zinc
could have an influence on the general transcriptional response.
cDNA microarray analysis. Microarray analysis was performed using
whole-genome DNA microarrays based on 70-mer oligonucleotides cov-
ering four meningococcal genomes (MC58, Z2491, FAM18, and ␣14) as
described previously (55). In order to characterize the Zur regulon, we
used a common reference design that allows for future extension of the
data set (65). As outlined in Fig. 1, cDNA obtained from strain MC58
grown with or without zinc supplementation (low- and high-zinc condi-
tions) was hybridized separately along with the common reference, i.e.,
cDNA obtained from strain MC58 ⌬zur. MC58 ⌬zur was used as the
common reference under the assumption that Zur mostly represses gene
expression. In the zur knockout strain, mRNA of most of the Zur-regu-
lated genes was therefore expected to be present, which is a prerequisite of
valid comparisons. The transcriptome of MC58 grown without zinc sup-
plementation was consecutively indirectly compared to that of MC58
grown in the presence of high zinc.
Ten micrograms of RNA obtained from each culture was transcribed
into cDNA and labeled with either Cy3-dCTP or Cy5-dCTP (GE Health-
care) using SuperScript II (Invitrogen) and random nonamer primer
(Sigma-Aldrich). Remaining RNA was digested with DNase-free RNase
(Roche). The labeled cDNA was purified using illustra AutoSeq G50 col-
umns (GE Healthcare). For each slide, two differentially labeled cDNA
probes were combined, and probes were then hybridized to a prehybrid-
ized microarray slide. Hybridization was carried out in a Tecan
HS4800TM Pro hybridization station. Three biological replicates (inde-
pendent bacterial cultures and RNA isolations) with at least two technical
replicates (independent slides including dye swap) were used for microar-
ray hybridization. The hybridized slides were scanned with the Genepix
4200 scanner, and the raw data were acquired using Genepix Pro 4.0. The
raw gpr files from the microarray slides were processed with the Limma
package (58) implemented in R (47) to identify significantly differentially
regulated genes. Only genes with a false-discovery rate (FDR) of ⬍0.01
and a B statistic value of ⬎0 were considered for further analyses.
qRT-PCR. Quantitative real-time PCR (qRT-PCR) analysis of the
RNA samples described above was performed as described previously
using the StepOnePlus system and the Power SYBR green Master Mix
(53). nmb1567, encoding a putative membrane-associated peptidyl-prolyl
isomerase, which was not differentially regulated in our microarray, was
applied as endogenous control for data normalization. The oligonucleo-
tides used are given in Table 1. The threshold cycle (⌬⌬C
T
) and relative
quantity (RQ) values of qRT-PCR were obtained using the StepOne soft-
ware v2.0. RQ values represent the fold change of expression of the inves-
tigated gene in MC58 grown under low zinc compared to that in MC58
grown under high-zinc conditions. Microarray results were accepted as
differentially regulated only if comparison of the same RNA by qRT-PCR
analysis yielded RQ values of at least 1.5.
Computational analysis of Zur binding sites. Sequences of genes and
upstream regions of the identified candidate genes of strain MC58 were
retrieved from the NeMeSys database (51). Similarity searches were per-
formed using the BLAST program of the National Center for Biotech-
nology Information (NCBI) (http://blast.ncbi.nlm.nih.gov/Blast.cgi).
Operons were predicted using RNAseq data (B. Joseph and C. Schoen,
unpublished data) and DOOR database (35). Annotation data were re-
trieved from the NeMeSys (51) and Uniprot (62) databases. Upstream
sequences were screened for the presence of a Zur box by alignments using
ClustalW in BioEdit 7.0.9 (20) based on the in silico predicted motif for
Betaproteobacteria as implemented in the RegPrecise database (39) and
the hypothetical motif for znuD (nmb0964), which was recently proposed
(59). Furthermore, upstream regions were searched for conserved motifs
with the GIBBS Motif Sampler (61). The meningococcal Zur binding
motif was visualized using WebLogo 3.2 (6).
Expression and purification of recombinant Zur. The Zur protein
was recombinantly expressed for electrophoretic mobility shift assays. For
gene cloning and protein expression, the QIAexpressionist kit (Qiagen)
was used according to the manufacturer’s instructions. Briefly, the entire
coding region of the zur gene without the start and stop codons (469 bp)
was amplified from MC58 using primers MP183 and MP184 harboring a
BspHI and a BglII restriction site, respectively (Table 1). The PCR product
was cloned between the NcoI and BglII sites of pQE-60 (Qiagen) contain-
ing a C-terminal His tag. E. coli M15(pREP4) was transformed with the
recombinant plasmid, and clones were verified by DNA sequencing. Ex-
pression and purification of Zur were performed under native conditions.
Protein expression was induced by 1 mM IPTG (isopropyl-␤-D-thiogalac-
topyranoside) for 4.5 h at 37°C, and cells were harvested, lysed, and bro-
ken by ultrasonic treatment.
For purification, cell extracts were loaded on a HIS GraviTrap nickel
Sepharose column (GE Healthcare) equilibrated with lysis buffer (50 mM
NaH
2
PO
4
, 0.3 M NaCl, 10 mM imidazole [pH 8.0]). The column was
washed twice with wash buffer (50 mM NaH
2
PO
4
, 0.3 M NaCl, 20 mM
imidazole [pH 8.0]) before elution of the His-tagged Zur protein with
elution buffer (50 mM NaH
2
PO
4
, 0.3 M NaCl, 250 mM imidazole [pH
8.0]). The Zur eluate from the nickel Sepharose column was dialyzed
against Tris-borate (TB) buffer (44.5 mM Tris, 44.5 mM boric acid [pH
8.0]) to remove imidazole and nickel. The purity of Zur was verified by
SDS-PAGE with Coomassie blue staining, and its concentration was de-
termined using the Pierce BCA protein assay kit (ThermoScientific). Pro-
tein sample aliquots were stored at ⫺20°C.
Electrophoretic mobility shift assays to visualize DNA-Zur binding.
EMSAs were performed using the digoxigenin (DIG) Gel Shift kit (Roche)
to determine the ability of Zur to interact with short double-stranded
DNA (dsDNA) fragments containing the Zur box. Oligonucleotides were
purchased from Sigma-Aldrich (Table 1) and resuspended in TEN buffer
(10 mM Tris, 1 mM EDTA, 0.1 M NaCl [pH 8.0]). Complementary oli-
gonucleotides in equal concentrations were annealed to yield dsDNA
fragments by heating the oligonucleotide mixture for 10 min at 95°C and
subsequently slowly cooling it down to room temperature. dsDNA frag-
ments were labeled with digoxigenin (DIG Gel Shift kit; Roche). Binding
reactions were performed in a final volume of 10 ␮l with 0.4 ng (15.5 fmol)
FIG 1 Common reference design of microarray comparison used in this
study. cDNAs obtained from MC58 grown without addition of zinc (low zinc)
and from MC58 grown with addition of zinc for 2 h (high zinc) were hybrid-
ized along with cDNA of the MC58 zur knockout strain grown under low-zinc
conditions as a common reference. MC58 ⌬zur was used as a common refer-
ence under the assumption that total mRNA of this mutant will contain
mRNA of Zur-repressed genes. The zinc-dependent promoter repression by
Zur is visualized by cartoons using the example of znuD.
Pawlik et al.
6596 jb.asm.org Journal of Bacteriology

of the labeled dsDNA fragment and 15 pmol (250 ng) of purified Zur
protein (if not stated differently) in the binding buffer [20 mM Tris-HCl
(pH 8.0), 50 mM KCl, 1 mM dithiothreitol (DTT), 5% glycerol, 0.1 ␮g/␮l
poly(dI-dC), 0.01 ␮g/␮l poly-L-lysine] modified after the method of Shin
et al. (57). If not stated otherwise, the binding buffer was supplemented
with 100 ␮M ZnSO
4
according to the concentration used by Gaballa and
Helmann for B. subtilis (13). We anticipated that zinc is required for
Zur-DNA interaction as already shown for other bacteria (14,29,32,54,
57) and therefore omitted EDTA from the binding buffer as well as from
the running buffer to avoid chelation of zinc. Reaction mixtures were
incubated at room temperature for 20 min. A native 8% polyacrylamide
gel was prerun at 120 V for 30 min, and samples were then applied to the
gel and separated at 120 V for 110 min at 4°C. After electroblotting onto a
nylon membrane in TB buffer, chemiluminescent detection of the bound
dsDNA fragments was carried out.
CaCl
2
, CoCl
2
, CuSO
4
, FeSO
4
, MgCl
2
, MnSO
4
, and NiSO
4
were pur-
chased from Merck or Sigma-Aldrich and added to the binding buffer at a
final concentration of 100 ␮M. The ion chelators EDTA and TPEN were
used at a final concentration of 312.5 ␮M. For binding competition ex-
periments, 125- to 1,000-fold excess of the unlabeled dsDNA probe was
added to the reactions.
Statistics. The Wilcoxon rank sum test with continuity correction was
applied to analyze the impact of mismatches in the Zur box palindromic
sequence on gene expression.
Microarray data accession number. The microarray data discussed in
this publication have been deposited in NCBI’s Gene Expression Omni-
bus database (12) under accession no. GSE38033.
RESULTS
The aim of this study was to provide a data set of genes regulated
by exposure to zinc. Microarray comparison between MC58
grown with and without zinc supplementation (high- and low-
zinc condition, respectively) was employed using a common ref-
erence design (65) that is schematically represented in Fig. 1.
Growth conditions. Growth of MC58 ⌬zur in the chemically
defined medium RPMI supplemented with FeCl
3
was indistin-
guishable from that of the parental wild-type strain (data not
shown). This indicates that deactivation of Zur does not alter
growth in RPMI. Growth conditions for microarray analysis were
validated by qRT-PCR of znuD expression, which previously was
shown to be increased in a zur knockout mutant (59). Strains were
first grown in RPMI medium under low-zinc conditions, followed
by addition of 0.5 ␮M ZnSO
4
to the medium for 1, 2, or 3 h.
After1hofzinc addition, MC58 showed 64.8-fold-reduced
expression levels of znuD by qRT-PCR compared to MC58 ⌬zur.
However, znuD expression in MC58 grown under low-zinc con-
ditions still was 27.3-fold reduced compared to MC58 ⌬zur, prob-
ably due to intracellular zinc remains of the overnight culture on
GC agar. After2hofzinc addition, compared to the zur knockout,
expression of znuD in MC58 was 48.8-fold reduced, whereas
MC58 grown under low-zinc conditions showed similar expres-
sion. After 3 h, znuD expression of MC58 grown at high zinc was
42.5-fold reduced compared to the zur knockout. Based on these
qRT-PCR results, exposure to zinc for 2 h was selected for mi-
croarray analysis.
Transcriptome analysis of zinc-dependent genes. Gene ex-
pression profiles of strain MC58 grown under low-zinc and high-
zinc conditions were compared by a common reference design
using the MC58 zur knockout strain as the common reference
(Fig. 1). Sixteen genes were upregulated, whereas three genes were
downregulated, in MC58 grown at low zinc compared to high
zinc. The genes of this data set were analyzed for operon struc-
tures. To confirm microarray analysis, we directly compared ex-
pression differences between MC58 (low zinc) and MC58 (high
zinc) by qRT-PCR for the first gene of predicted operon structures
and all single genes. All genes except two were confirmed to be
differentially expressed. Therefore, the regulon verified in this
study comprised 15 genes upregulated and only 2 downregulated
in MC58 grown at low zinc compared to high zinc (Table 2). Nine
genes were organized in four transcriptional units: nmb0317 and
nmb0316 encode a 7-cyano-7-deazaguanine reductase and an in-
tegral membrane protein, respectively; nmb0817 and nmb0818
code for hypothetical proteins that belong to the DUF723 family
and may have a role in DNA-binding (62); nmb0942 and nmb0941
encode paralogues of 50S ribosomal proteins; and nmb0588 and
nmb0587, together with one single gene, nmb0586, encode the
components of the putative ABC transporter for zinc, ZnuCBA
(59). The single genes nmb0819 and nmb0820 both encode pro-
teins that contain putative DNA-binding helix-turn-helix motifs.
Highly differential expression of the TonB-dependent outer
membrane receptor znuD (nmb0964), involved in zinc uptake
(59), represents a positive control for our analysis, as znuD was
used for optimization of our microarray conditions as indicated
above. Expression of another putative TonB-dependent receptor
of yet-unknown function, nmb1497, was also increased at low
zinc. Strongly increased at low zinc were expression of nmb0525
(queC), encoding a zinc-binding 7-cyano-7-deazaguanine syn-
thase, and expression of nmb1475, coding for a conserved hypo-
thetical periplasmic protein with similarities to the acetate kinase
AckA of Bacillus spp. Only two genes were repressed at low zinc:
nmb0546 (adhP), which codes for a zinc-containing alcohol dehy-
drogenase, and nmb0577, which shows similarities to the Haemo-
philus influenzae pfkA, encoding a 6-phosphofructokinase.
In silico prediction of promoter organization and the Zur
box. Zur binding to a palindromic sequence upstream of znuD has
been previously postulated (59). The published sequence resem-
bles the in silico-predicted Zur binding motif for Betaproteobacte-
ria in the RegPrecise database (39). We analyzed the promoter
regions of the confirmed zinc-responsive genes for a conserved
Zur box. Homologies to the postulated Zur binding motif were
found in the upstream regions of all regulated transcription units
(genes or operons). This finding suggests that the approach em-
ployed here at least partially elucidated the Zur regulon and that
zinc treatment of the bacteria did not deregulate genes other than
Zur-controlled ones. The deduced meningococcal Zur binding
motif was visualized as the consensus motif and graphically dis-
played by WebLogo 3.2 (Fig. 2A). It is a 23-bp motif with a central
palindromic part comprising hexameric inverse repeats separated
from each other by three nucleotides, which is also found in the
predicted motif for znuD published by Stork et al. (59). The align-
ment of all putative Zur motifs indicated that the strength of gene
regulation by Zur was dependent on the precision and length of
the palindrome (Fig. 2B). The extent of zinc-dependent gene reg-
ulation (as deduced from RQ values) was reduced if at least one
mismatch occurred in the palindrome of the putative Zur box
(Fig. 2C). Gene expression changes as calculated based on the
absolute ⌬⌬C
T
values obtained by qRT-PCR were significantly
higher in genes with a perfect Zur box than in genes with imperfect
Zur boxes having at least one mismatch (Wilcoxon rank sum test
with continuity correction, P⬍0.01). Extension of the length of
the palindromic sequence may compensate for mismatches in the
central palindrome, e.g., in the case of nmb0317 or nmb0588. The
Zinc-Responsive Regulon of Meningococci
December 2012 Volume 194 Number 23 jb.asm.org 6597

longest palindrome was found for nmb1475 with a decameric in-
verse repeat that probably represents the perfect Zur box, because
this gene showed the highest repression of gene expression upon
zinc exposure (RQ ⫽96.4). We detected two putative Zur boxes
upstream of nmb0586. However, this duplication did not affect
the strength of gene expression.
Verification of Zur binding to predicted DNA motifs by elec-
trophoretic mobility shift assay. To demonstrate that Zur binds
to in silico-predicted DNA motifs, we performed EMSAs with a
His-tagged recombinant Zur protein and synthetic dsDNA frag-
ments. A mobility shift of a 34-bp dsDNA fragment comprising
the nmb0964 Zur box was observed upon incubation with the
recombinant protein (Fig. 3A). The shift of the nmb0964 dsDNA
fragment was competed by 125-fold excess of the unlabeled
nmb0964 dsDNA fragment, which confirms binding specificity
(Fig. 3A). A band shift was observed only in the presence of 100
␮M ZnSO
4
, whereas CaCl
2
, CoCl
2
, CuSO
4
, FeSO
4
, MgCl
2
,
MnSO
4
, and NiSO
4
did not support binding of Zur to the dsDNA
fragment, indicating that zinc is needed specifically to mediate
Zur-DNA interaction (Fig. 3B). Addition of EDTA and TPEN,
separately, blocked the zinc effect by chelation, underlining the
importance of zinc (Fig. 3B).
Mutations of the four most conserved nucleotides in the 3=-
inverse repeat of the palindrome (ATAACA) (where the con-
served nucleotides are in boldface) of the nmb0964 dsDNA frag-
ment abrogated or reduced Zur binding (Fig. 3C). However,
interference with shifts was less clear upon mutation of marginal
nucleotides or in the case of a single nucleotide mutation. This
indicates that (i) all four conserved bases are important for Zur
binding but the inner bases (TA) of the palindrome are more
essential than the marginal ones (CA) and (ii) only one mutated
nucleotide is not sufficient to completely abrogate the shift.
We further investigated the binding of Zur to the in silico-
predicted Zur motifs of nmb0546, one of the two genes activated in
response to zinc. A clear band shift that could be inhibited by the
unlabeled nmb0546 dsDNA fragment was seen. For reasons un-
known, competition of the nmb0546 shift required higher concen-
trations of the competing DNA than nmb0964. For nmb1475, the
gene with an extended perfect palindrome in the Zur box, no shift
was observed with a 34-mer dsDNA fragment (data not shown).
Upon extending the surrounding region by 8 and 9 bp on each side
of the extended perfect palindrome, a clear shift became also vis-
ible (Fig. 3D). We can only speculate that Zur binding to the
perfect oversized palindrome may be possible only if the DNA
structure is stabilized by extended flanking DNA sequences.
To furthermore affirm binding specificity, we performed
EMSA with the nmb0964 dsDNA fragment and attempted to com-
pete the shift with unlabeled nmb0964 or nmb1475 dsDNA frag-
ments or mutated versions of the nmb0964 sequence. Indeed, the
unlabeled nmb0964 dsDNA fragment itself and the unlabeled
nmb1475 dsDNA fragment comprising the same palindrome ab-
rogated Zur binding, whereas the nmb0964 dsDNA fragment
comprising mismatches within the 3=-inverse repeat palindrome
(ATAACA) was not able to completely compete the shift (Fig. 3E).
In summary, we detected Zur boxes upstream of all genes that
were verified by qRT-PCR to be differentially expressed in re-
sponse to zinc. Our EMSA results prove that Zur binds to the
predicted binding motifs of three selected genes regulated in re-
sponse to zinc. Therefore, we infer that all transcriptional units
with Zur boxes identified here were indeed regulated by Zur. Zinc
was indispensable for interaction of Zur with the DNA. The pal-
indromic part of the motif was essential for the binding as dem-
onstrated by mutational analysis. In vitro binding of dsDNA frag-
TABLE 2 Differentially expressed genes in N. meninigitidis MC58 observed by comparison of low- to high-zinc conditions
a
Locus Gene Predicted function Predicted localization Size (bp)
Differential gene expression
Fold change
by cDNA
microarray
hybridization
RQ by
qRT-PCR
nmb0546 adhP Alcohol dehydrogenase, propanol preferring Cytoplasmic 1,047 ⫺4.1 ⫺21.7
nmb0577 NosR-related protein Unknown 351 ⫺1.8 ⫺1.8
nmb0588 znuC ABC transporter, ATP-binding protein Cytoplasmic 756 1.4 4.4
nmb0586 znuA Putative ABC transporter substrate-binding
protein
Cytoplasmic
membrane
915 1.5 6.3
nmb0820 Hypothetical protein Unknown 198 1.6 1.9
nmb1497 Putative TonB-dependent receptor Outer membrane 2,766 1.6 1.8
nmb0818 Hypothetical protein Unknown 411 1.9
nmb0817 Hypothetical protein Unknown 384 2.0 3.1
nmb0587 znuB Putative ABC transporter permease protein Cytoplasmic
membrane
876 2.1 3.3
nmb0942 rpmE2 50S ribosomal protein L31 type B Cytoplasmic 276 2.2 41.6
nmb0819 Hypothetical protein Unknown 393 2.3 1.8
nmb0525 queC 7-Cyano-7-deazaguanine synthase Cytoplasmic 660 2.5 3.9
nmb0941 rpmJ 50S ribosomal protein L36 Cytoplasmic 126 2.7
nmb0317 queF NADPH-dependent 7-cyano-7-deazaguanine
reductase
Cytoplasmic 474 3.4 5.8
nmb1475 Conserved hypothetical periplasmic protein Periplasmic 807 3.6 96.4
nmb0964 znuD TonB-dependent receptor Outer membrane 2,277 4.4 51.1
nmb0316 Conserved hypothetical integral membrane protein Cytoplasmic
membrane
687 4.6
a
Gene information retrieved from NeMeSys (51) and Uniprot (62) databases. Genes are ordered by their differential expression in microarray analysis.
Pawlik et al.
6598 jb.asm.org Journal of Bacteriology

ments harboring a Zur box was observed for genes irrespective of
their activation or repression by zinc.
DISCUSSION
In the present study, we analyzed the role of the meningococcal
zinc uptake regulator Zur in transcriptional regulation using a
comparative microarray approach. We defined the Zur box for
meningococci and demonstrated in vitro interaction of recombi-
nantly expressed Zur protein with the predicted Zur boxes up-
stream of three selected target genes.
Previous studies of B. subtilis,Y. pestis,M. tuberculosis, and C.
glutamicum applied direct microarray comparisons of a wild-type
strain with a zur knockout mutant to elucidate the Zur regulon
(14,29,32,54). The advantage of our approach of analyzing the
response to zinc was that pleiotropic effects of a constitutive zur
knockout might be avoided. Indeed, the meningococcal Zur regu-
lon deduced from transcriptome data and consecutive Zur box
analysis is remarkably small. We identified 15 genes downregu-
lated and two genes upregulated at high zinc using microarray
analysis and qRT-PCR. All transcriptional units harbored a Zur
box. The entity of 17 regulated genes is comparable to the number
found in microarray analyses of Zur regulons of other bacteria,
i.e., 18 genes and 32 genes upregulated in a zur deletion mutant of
C. glutamicum (54) and M. tuberculosis (32), respectively. Only in
a comparative microarray analysis of a Y. pestis wild-type strain
and its zur knockout mutant, both grown under zinc-rich condi-
tions, were a much higher number of Zur-regulated genes found
(154 genes). However, as only four genes had a Zur box, regula-
tion of most genes presumably has not been a direct result of Zur
binding to promoters but reflects a general alteration of gene reg-
ulation (29). Following several studies of the iron response (18,
56), we could have applied zinc depletion, achieved by addition of
an ion chelator such as TPEN, to compare with a zinc repletion
condition. However, it cannot be excluded that TPEN chelates
ions other than zinc and interferes with the membrane integrity.
Among the 17 genes regulated by zinc exposure, we detected 2
that are upregulated, i.e., nmb0546 and nmb0577. Gene activation
by Zur has been reported in only two other studies. In a microar-
ray analysis of the B. subtilis Zur regulon, two genes were upregu-
lated in the wild type compared to a zur deletion mutant. How-
ever, Zur motifs were not detected in the genes’ regulatory regions
and it was assumed that their upregulation is probably due to
indirect effects of the altered zinc homeostasis (14). This can be
excluded in this study, as we identified Zur boxes in the upstream
regions of both Zur-activated genes. Only in one other study, of
the phytopathogen Xanthomonas campestris, did Zur act as a di-
rect activator of one gene (22).
Based on the prediction of the Zur binding motif for znuD by
Stork et al. (59), we analyzed the promoter regions of all genes
found to be zinc regulated in our study. We established a binding
motif for meningococcal Zur with the ideal palindromic sequence
TGTTATDNHATAACA, which is identical to the Zur binding
motif proposed for znuD (59) and consistent with the palindrome
in the previously bioinformatically predicted motif for the related
Gammaproteobacteria.
We found an ideal palindrome only within the promoters of
the genes nmb0546,nmb0942,znuD, and nmb1475. These genes
showed the highest differences in expression when low- and high-
zinc conditions were compared. Mismatches in the palindrome
reduced gene expression alteration. This finding was confirmed by
EMSA, also in line with previous investigations of the Zur box of
C. glutamicum (54).
The zinc-activated gene nmb0546 was previously shown to be
Hfq repressed in N. meningitidis (37,42). Hfq is an RNA chaper-
one that stabilizes small regulatory RNAs (sRNAs) and mediates
their binding to their mRNA targets, which leads to subsequent
repression of mRNA translation (37). If nmb0546 and nmb0577
were targets of a yet-unknown sRNA transcribed at zinc depletion
FIG 2 Prediction of the putative meningococcal Zur box. (A) Graphical dis-
play of the Zur box for meningococci based on the consensus sequence of
predicted Zur binding sites for all zinc-responsive genes (generated with
WebLogo 3.2). (B) Nucleotide sequence alignment of putative Zur boxes up-
stream of zinc-regulated genes. Distances of the Zur motif to the start codon of
translation and expression differences determined by qRT-PCR are given for
each gene (RQ; low-zinc to high-zinc conditions). The order of motifs is based
on the RQ values. The two genes upregulated by Zur are in bold gray letters.
The strongly conserved palindrome composed of two hexameric inverted re-
peats separated by three nucleotides is indicated (#) and boxed for all motifs.
Bold nucleotides signify complementary nucleotide pairs between the two
sides of the motifs. Symbols: *, an alternative start for the gene is possible; ‡,
two different motifs were found in the same promoter region. (C) Comparison
of the absolute RQ values of gene expression with the number of mismatches in
the palindrome of each Zur box compared to the znuD (nmb0964) palindrome
sequence (TGTTATDNHATAACA). Gene expression changes were signifi-
cantly higher in genes with a perfect Zur box than in genes with at least one
mismatch in the Zur box (Wilcoxon rank sum test with continuity correction,
P⬍0.01).
Zinc-Responsive Regulon of Meningococci
December 2012 Volume 194 Number 23 jb.asm.org 6599

and downregulated by Zur upon zinc repletion, activation of the
genes would occur indirectly due to Zur-mediated downregula-
tion of the putative sRNA, which otherwise leads to degradation of
both mRNAs. However, RNA sequencing needs to be applied to
analyze the regulation of sRNAs by zinc.
Functions of regulated genes. The previously identified me-
ningococcal genes carrying homologues of the E. coli znuCBA
operon encoding an ABC transporter for high-affinity zinc up-
take, i.e., nmb0588-nmb0587-nmb0586, and nmb0964 (znuD),
coding for a TonB-dependent receptor mediating zinc uptake at
low zinc concentration (59), were shown to be repressed at high
zinc in this study. This finding is reminiscent of E. coli (44), M.
tuberculosis (32), B. subtilis (14), C. glutamicum (54), Y. pestis (29),
and Streptomyces coelicolor (57). We demonstrated by EMSA that
meningococcal Zur binds to the in silico predicted Zur box first
proposed by Stork et al. for znuD (59).
The most strongly regulated gene in our study was nmb1475.It
codes for a conserved hypothetical periplasmic protein that shows
34% similarity to the acetate kinase AckA of Bacillus spp. AckA is
an enzyme involved in the conversion of acetate to acetyl coen-
zyme A (acetyl-CoA) (62), and its E. coli homologue was shown to
bind zinc (24). The protein also harbors a conserved domain that
is similar to the CbiK (COG5266) domain of the periplasmic com-
ponent of an ABC-type Co
2⫹
transport system as identified by
NCBI Blast. Thus, NMB1475 may be involved in the uptake of
zinc and/or other transition metals.
It has been suggested that zinc uptake systems are important
for bacterial survival and virulence upon infection, as the access to
FIG 3 Zur binding to dsDNA fragments in EMSA. Band shifts are indicated with an arrow, unbound dsDNA fragments with an asterisk. (A) Binding of Zur to
the dsDNA fragment comprising the nmb0964 Zur box. Addition of different amounts of Zur from 3.75 pmol to 60 pmol led to a band shift of the 34-mer dsDNA
fragment. The unlabeled nmb0964 dsDNA fragment in 125-fold excess acted as a specific competitor that abrogated the band shift. (B) Effect of divalent ions and
chelators on Zur binding to the nmb0964 dsDNA fragment. Different divalent ions (100 ␮M) were added to the reactions. Only Zn
2⫹
mediated Zur binding to
the nmb0964 dsDNA fragment. Addition of the chelator (312.5 ␮M) EDTA or TPEN reverted the band shift. (C) Mutational analysis of the Zur box. Mutation
of conserved nucleotides in the 3=-inverse repeat of the palindrome of the nmb0964 Zur box inhibited the band shift. The four highly conserved nucleotides are
displayed in bold. Mutated nucleotides are marked in gray. (D) Binding of Zur to dsDNA fragments comprising the Zur boxes of nmb0546 and nmb1475. Zur led
to a band shift of the 34-mer nmb0546 dsDNA fragment and the 51-mer nmb1475 dsDNA fragment covering the respective Zur boxes. Addition of 1,000-fold
excess of the respective unlabeled dsDNA fragment led to competition of the shift. (E) Competition studies with the nmb0964 dsDNA fragment. Incubation with
a 125-fold excess of the unlabeled nmb0964 and nmb1475 dsDNA fragments abrogated Zur binding; 125-fold excess of unlabeled mutated nmb0964 dsDNA
fragments comprising one to four mismatches within the 3=-inverse repeat of the palindrome (panel C) was unable to fully compete the shift.
Pawlik et al.
6600 jb.asm.org Journal of Bacteriology

zinc is limited within the human host (21,59). Of note, expression
of NMB0586 (designated MntC by van Alen et al. [63] or ZnuA by
Stork et al. [59]) and NMB1475 was upregulated in biofilms, and
deletion of nmb0586 reduced biofilm formation (63). Extenuated
biofilm formation has previously also been seen upon deletion of
znuA in gonococci (30) and nmb0587 (znuB)inE. coli (19). A
gonococcal znuA (mntC) mutant in vitro also was more sensitive
to hydrogen peroxide than the wild-type strain and showed re-
duced invasion to primary human cervical epithelial cells (64).
Furthermore, znuA contributes to the in vivo pathogenicity of
Salmonella enterica (3). Moreover, ZnuABC and ZnuD were lately
shown to contribute to bacterial adhesion to epithelial cells in E.
coli (15) and N. meningitidis (26), respectively.
Two meningococcal genes, nmb0317 and nmb0525, that were
repressed at high zinc and harbor an upstream Zur box are possi-
bly involved in queuosine biosynthesis as deduced from protein
similarity analysis. nmb0525 encodes the zinc-binding 7-cyano-7-
deazaguanine synthase QueC, and nmb0317 codes for the
7-cyano-7-deazaguanine reductase QueF (62). To date, queuosine
biosynthesis in bacteria is still not fully understood, but the path-
ways have been studied in E. coli and B. subtilis (16,27). Queuosine
is one of the most complex modified nucleosides that are incor-
porated at the wobble position of a subset of tRNAs (10). Such
modification of tRNAs was shown to improve the efficiency and
correctness of translation (9). Dineshkumar et al. demonstrated a
natural defect in queuosine biosynthesis of E. coli and noticed
reduced fitness under nutrient limitation (8). Lack of queuosine
also attenuated Shigella flexneri virulence (9,10). In meningo-
cocci, queuosine biosynthesis seems to be strongly Zur regulated,
since two enzymes of the biosynthesis are upregulated under zinc-
limiting conditions as they occur in the host (25). It will be of
interest for future studies to address the question of whether en-
hanced queuosine modification of tRNAs might be the result of
zinc depletion and contribute to increased expression of virulence
factors and thus support fitness of N. meningitidis upon infection.
Concordant with what has been observed in E. coli (17,41), Y.
pestis (29), M. tuberculosis (32), S. coelicolor (57), and B. subtilis
(2), the expression of several genes encoding ribosomal proteins
(r-proteins) was also Zur repressed in N. meningitidis. Several bac-
terial genomes code for two paralogous forms of r-proteins (34).
The first form (C⫹) contains a metal-binding Zn
2⫹
ribbon usu-
ally consisting of four conserved cysteines, whereas in the second
form (C⫺) this zinc ribbon is degenerated (52). For the paralo-
gous pair of the L31 r-protein in B. subtilis, RpmE (C⫹) and YtiA
(C⫺), it was shown that YtiA (C⫺) expression is repressed by Zur
and YtiA liberates RpmE (C⫹) from the ribosome under zinc-
deficient conditions (2,29). Panina et al. identified candidate
binding sites for different zinc repressors upstream of genes cod-
ing for C⫺paralogues of the r-proteins L31, L33, L36, and S14 of
a range of different bacterial species and confirmed that upon
existence of a C⫹andaC⫺copy of the r-protein the gene encod-
ing the C⫺copy is regulated by a zinc-dependent repressor (41).
The replacement of zinc-containing r-proteins by non-zinc-con-
taining paralogues upon zinc depletion might liberate zinc for
maintaining zinc homeostasis and may enhance bacterial survival
when facing zinc-restrictive conditions in vivo (25,41). A poten-
tial evolutionary explanation for the duplication of r-proteins
with different zinc contents might be that ribosomal assembly in
any case needs to be maintained under zinc-restrictive conditions.
In the meningococcal genome, the genes for the 50S ribosomal
proteins L31 and L36 are duplicated. Based on sequence similarity
to B. subtilis RpmE (C⫹) and YtiA (C⫺), the paralogous pairs of
N. meningitidis r-proteins L31 and L36 are NMB1956 (C⫹)/
NMB0942 (C⫺) and NMB0164 (C⫹)/NMB0941 (C⫺), respec-
tively (34,51). The nmb0942-nmb0941 (rpmEJ) operon was
shown to be repressed at high zinc in this study, and we identified
a Zur box upstream of nmb0942. Therefore, we assume that also in
meningococci in the absence of zinc the enzyme lacking the zinc
ribbon takes over the function of the C⫹protein.
Comparison to other regulons. In a study by Wu et al., the
manganese-responsive PerR regulon has been described (64).
Gonococcal PerR is a member of the Fur family and shows 96%
protein identity to meningococcal Zur. Wu et al. conducted a
microarray comparison of a Neisseria gonorrhoeae wild-type strain
versus its perR knockout mutant. They found 11 genes upregu-
lated and 1 gene downregulated in the perR mutant. Except for
three genes, all genes do have meningococcal homologues with
93% to 98% protein identity. All of those homologues also have
been identified in our microarray analysis of the zinc-responsive
regulon of Neisseria meningitidis:nmb0586-nmb0588 (ng0170-
ng0168), nmb0941-nmb0942 (ng0931-ng0930), nmb0964 (ng1205),
nmb1475 (ng1049), nmb1497 (ng0952), and nmb0546 (ng1442)(51,
64). This suggests that the meningococcal zinc-dependent Zur
regulation is related to the gonococcal manganese-dependent
PerR regulation.
Several of the Zur-regulated proteins were also regulated dif-
ferentially in a microarray study that compared transcriptional
profiles of N. meningitidis grown with different host iron binding
proteins (i.e., hemoglobin, transferrin, and lactoferrin) as the sole
iron source (23). The genes nmb0941,nmb0942,nmb1475, and
nmb1497 were downregulated, and nmb0546 was upregulated
upon exposure to lactoferrin (compared to hemoglobin or trans-
ferrin) (23). Lactoferrin, which is present in secretions and on
mucosal surfaces of the human host, can be used by meningococci
employing a lactoferrin receptor to catch the iron from lactoferrin
(23,38). Therefore, regulation of these genes upon lactoferrin
exposure may additionally be accomplished by a second regulator
that senses iron. This was shown recently for znuD, whose expres-
sion also was iron induced (26). Besides its upstream Zur binding
site, znuD also harbors a Fur binding site where Fur in vitro binds
to, independently of Zur (26). However, we did not detect any Fur
box upstream of nmb0942-nmb0941,nmb1475,nmb1497, and
nmb0546, nor has znuD been regulated upon lactoferrin exposure
in the study by Jordan and Saunders (23). Hence, we favor a sec-
ond hypothesis to explain the overlap of the zinc and lactoferrin
regulons. As lactoferrin has been reported to also loosely bind zinc
(1), meningococci might use it as an additional source of zinc.
Zinc could then act as a cofactor for Zur, leading to Zur-mediated
regulation of the genes mentioned above as also seen in this study.
The transcriptomic response of N. meningitidis to whole-blood
exposure was recently recorded using microarrays (11). Interest-
ingly, several of the genes deregulated by this approach were also
found to be changed in their expression upon zinc exposure. This
finding suggests that zinc depletion upon transfer of bacteria from
liquid culture to whole blood is responsible for a part of the tran-
scriptional changes observed. Furthermore, this finding also rein-
forces that changes of zinc concentration mimic an important
environmental signal encountered by bacteria during pathogen-
host interaction.
In summary, we elucidated the transcriptional adaptation of N.
Zinc-Responsive Regulon of Meningococci
December 2012 Volume 194 Number 23 jb.asm.org 6601

meningitidis to zinc using strain MC58. The regulon is assumed to
cover at least a significant portion of the Zur regulon, as all tran-
scriptional units (genes/operons) were preceded by promoters
harboring a Zur box. The functionality of representative motifs
was confirmed by EMSA. It will be of interest to investigate the
concerted action of genes derepressed at low-zinc conditions in
vivo, as low zinc is encountered by the bacteria upon infection
(25).
ACKNOWLEDGMENTS
The study was funded by a grant to U.V. provided by the German Federal
Ministry of Education and Research (reference number 0315434) via the
ERA-NET PathoGenoMics program (2nd call).
REFERENCES
1. Ainscough EW, Brodie AM, Plowman JE. 1980. Zinc transport by lac-
toferrin in human milk. Am. J. Clin. Nutr. 33:1314–1315.
2. Akanuma G, Nanamiya H, Natori Y, Nomura N, Kawamura F. 2006.
Liberation of zinc-containing L31 (RpmE) from ribosomes by its paralo-
gous gene product, YtiA, in Bacillus subtilis. J. Bacteriol. 188:2715–2720.
3. Ammendola S, et al. 2007. High-affinity Zn2⫹uptake system ZnuABC is
required for bacterial zinc homeostasis in intracellular environments and
contributes to the virulence of Salmonella enterica. Infect. Immun. 75:
5867–5876.
4. Bentley SD, et al. 2007. Meningococcal genetic variation mechanisms
viewed through comparative analysis of serogroup C strain FAM18. PLoS
Genet. 3:e23. doi:10.1371/journal.pgen.0030023.
5. Chen PR, He C. 2008. Selective recognition of metal ions by metalloregu-
latory proteins. Curr. Opin. Chem. Biol. 12:214–221.
6. Crooks GE, Hon G, Chandonia JM, Brenner SE. 2004. WebLogo: a
sequence logo generator. Genome Res. 14:1188–1190.
7. Dalet K, Gouin E, Cenatiempo Y, Cossart P, Hechard Y. 1999. Char-
acterisation of a new operon encoding a Zur-like protein and an associated
ABC zinc permease in Listeria monocytogenes. FEMS Microbiol. Lett.
174:111–116.
8. Dineshkumar TK, Thanedar S, Subbulakshmi C, Varshney U. 2002. An
unexpected absence of queuosine modification in the tRNAs of an Esch-
erichia coli B strain. Microbiology 148:3779–3787.
9. Durand JM, Bjork GR. 2003. Putrescine or a combination of methionine
and arginine restores virulence gene expression in a tRNA modification-
deficient mutant of Shigella flexneri: a possible role in adaptation of viru-
lence. Mol. Microbiol. 47:519–527.
10. Durand JM, Dagberg B, Uhlin BE, Bjork GR. 2000. Transfer RNA
modification, temperature and DNA superhelicity have a common target
in the regulatory network of the virulence of Shigella flexneri: the expres-
sion of the virF gene. Mol. Microbiol. 35:924–935.
11. Echenique-Rivera H, et al. 2011. Transcriptome analysis of Neisseria
meningitidis in human whole blood and mutagenesis studies identify vir-
ulence factors involved in blood survival. PLoS Pathog. 7:e1002027. doi:
10.1371/journal.ppat.1002027.
12. Edgar R, Domrachev M, Lash AE. 2002. Gene Expression Omnibus:
NCBI gene expression and hybridization array data repository. Nucleic
Acids Res. 30:207–210.
13. Gaballa A, Helmann JD. 1998. Identification of a zinc-specific metallo-
regulatory protein, Zur, controlling zinc transport operons in Bacillus
subtilis. J. Bacteriol. 180:5815–5821.
14. Gaballa A, Wang T, Ye RW, Helmann JD. 2002. Functional analysis of
the Bacillus subtilis Zur regulon. J. Bacteriol. 184:6508– 6514.
15. Gabbianelli R, et al. 2011. Role of ZnuABC and ZinT in Escherichia coli
O157:H7 zinc acquisition and interaction with epithelial cells. BMC Mi-
crobiol. 11:36. doi:10.1186/1471-2180-11-36.
16. Gaur R, Varshney U. 2005. Genetic analysis identifies a function for the
queC (ybaX) gene product at an initial step in the queuosine biosynthetic
pathway in Escherichia coli. J. Bacteriol. 187:6893–6901.
17. Graham AI, et al. 2009. Severe zinc depletion of Escherichia coli: roles for
high affinity zinc binding by ZinT, zinc transport and zinc-independent
proteins. J. Biol. Chem. 284:18377–18389.
18. Grifantini R, et al. 2003. Identification of iron-activated and -repressed
Fur-dependent genes by transcriptome analysis of Neisseria meningitidis
group B. Proc. Natl. Acad. Sci. U. S. A. 100:9542–9547.
19. Gunasekera TS, Herre AH, Crowder MW. 2009. Absence of ZnuABC-
mediated zinc uptake affects virulence-associated phenotypes of uro-
pathogenic Escherichia coli CFT073 under Zn(II)-depleted conditions.
FEMS Microbiol. Lett. 300:36 – 41.
20. Hall TA. 1999. BioEdit: a user-friendly biological sequence alignment
editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp.
41:95–98.
21. Hantke K. 2001. Bacterial zinc transporters and regulators. Biometals
14:239–249.
22. Huang DL, et al. 2008. The Zur of Xanthomonas campestris functions as
a repressor and an activator of putative zinc homeostasis genes via recog-
nizing two distinct sequences within its target promoters. Nucleic Acids
Res. 36:4295–4309.
23. Jordan PW, Saunders NJ. 2009. Host iron binding proteins acting as
niche indicators for Neisseria meningitidis. PLoS One 4:e5198. doi:
10.1371/journal.pone.0005198.
24. Katayama A, Tsujii A, Wada A, Nishino T, Ishihama A. 2002. System-
atic search for zinc-binding proteins in Escherichia coli. Eur. J. Biochem.
269:2403–2413.
25. Kehl-Fie TE, Skaar EP. 2010. Nutritional immunity beyond iron: a role
for manganese and zinc. Curr. Opin. Chem. Biol. 14:218–224.
26. Kumar P, Sannigrahi S, Tzeng YL. 2012. The Neisseria meningitidis
ZnuD zinc receptor contributes to interactions with epithelial cells and
supports heme utilization when expressed in Escherichia coli. Infect. Im-
mun. 80:657–667.
27. Lee BW, Van Lanen SG, Iwata-Reuyl D. 2007. Mechanistic studies of
Bacillus subtilis QueF, the nitrile oxidoreductase involved in queuosine
biosynthesis. Biochemistry 46:12844 –12854.
28. Lee JW, Helmann JD. 2007. Functional specialization within the Fur
family of metalloregulators. Biometals 20:485–499.
29. Li Y, et al. 2009. Characterization of Zur-dependent genes and direct Zur
targets in Yersinia pestis. BMC Microbiol. 9:128. doi:10.1186/1471-2180-
9-128.
30. Lim KH, et al. 2008. Metal binding specificity of the MntABC permease of
Neisseria gonorrhoeae and its influence on bacterial growth and interac-
tion with cervical epithelial cells. Infect. Immun. 76:3569–3576.
31. Lindsay JA, Foster SJ. 2001. zur: a Zn(2⫹)-responsive regulatory element
of Staphylococcus aureus. Microbiology 147:1259–1266.
32. Maciag A, et al. 2007. Global analysis of the Mycobacterium tuberculosis
Zur (FurB) regulon. J. Bacteriol. 189:730–740.
33. Reference deleted.
34. Makarova KS, Ponomarev VA, Koonin EV. 2001. Two C or not two C:
recurrent disruption of Zn-ribbons, gene duplication, lineage-specific
gene loss, and horizontal gene transfer in evolution of bacterial ribosomal
proteins. Genome Biol. 2:research 0033–research0033.14. doi:10.1186/gb-
2001-2-9-research0033.
35. Mao F, Dam P, Chou J, Olman V, Xu Y. 2009. DOOR: a database for
prokaryotic operons. Nucleic Acids Res. 37:D459–D463.
36. McGuinness BT, et al. 1991. Point mutation in meningococcal porA gene
associated with increased endemic disease. Lancet 337:514–517.
37. Mellin JR, et al. 2010. Role of Hfq in iron-dependent and -independent
gene regulation in Neisseria meningitidis. Microbiology 156:2316–2326.
38. Morgenthau A, Livingstone M, Adamiak P, Schryvers AB. 2012. The
role of lactoferrin binding protein B in mediating protection against hu-
man lactoferricin. Biochem. Cell Biol. 90:417–423.
39. Novichkov PS, et al. 2010. RegPrecise: a database of curated genomic
inferences of transcriptional regulatory interactions in prokaryotes. Nu-
cleic Acids Res. 38:D111–D118.
40. O’Halloran TV. 1993. Transition metals in control of gene expression.
Science 261:715–725.
41. Panina EM, Mironov AA, Gelfand MS. 2003. Comparative genomics of
bacterial zinc regulons: enhanced ion transport, pathogenesis, and rear-
rangement of ribosomal proteins. Proc. Natl. Acad. Sci. U. S. A. 100:9912–
9917.
42. Pannekoek Y, et al. 2009. Molecular characterization and identification
of proteins regulated by Hfq in Neisseria meningitidis. FEMS Microbiol.
Lett. 294:216–224.
43. Parkhill J, et al. 2000. Complete DNA sequence of a serogroup A strain of
Neisseria meningitidis Z2491. Nature 404:502–506.
44. Patzer SI, Hantke K. 2000. The zinc-responsive regulator Zur and its
control of the znu gene cluster encoding the ZnuABC zinc uptake system
in Escherichia coli. J. Biol. Chem. 275:24321–24332.
Pawlik et al.
6602 jb.asm.org Journal of Bacteriology

45. Patzer SI, Hantke K. 1998. The ZnuABC high-affinity zinc uptake system
and its regulator Zur in Escherichia coli. Mol. Microbiol. 28:1199–1210.
46. Reference deleted.
47. R Development Core Team. 2008. R: a language and environment for sta-
tistical computing. R Foundation for Statistical Computing, Vienna, Austria.
48. Reference deleted.
49. Reyes-Caballero H, Campanello GC, Giedroc DP. 2011. Metalloregula-
tory proteins: metal selectivity and allosteric switching. Biophys. Chem.
156:103–114.
50. Rosenstein NE, Perkins BA, Stephens DS, Popovic T, Hughes JM. 2001.
Meningococcal disease. N. Engl. J. Med. 344:1378–1388.
51. Rusniok C, et al. 2009. NeMeSys: a biological resource for narrowing the
gap between sequence and function in the human pathogen Neisseria
meningitidis. Genome Biol. 10:R110. doi:10.1186/gb-2009-10-10-r110.
52. Sankaran B, et al. 2009. Zinc-independent folate biosynthesis: genetic,
biochemical, and structural investigations reveal new metal dependence
for GTP cyclohydrolase IB. J. Bacteriol. 191:6936– 6949.
53. Schielke S, et al. 2011. Characterization of FarR as a highly specialized,
growth phase-dependent transcriptional regulator in Neisseria meningi-
tidis. Int. J. Med. Microbiol. 301:325–333.
54. Schröder J, Jochmann N, Rodionov DA, Tauch A. 2010. The Zur
regulon of Corynebacterium glutamicum ATCC 13032. BMC Genomics
11:12. doi:10.1186/1471-2164-11-12.
55. Schwarz R, et al. 2010. Evaluation of one- and two-color gene expression
arrays for microbial comparative genome hybridization analyses in rou-
tine applications. J. Clin. Microbiol. 48:3105–3110.
56. Shaik YB, et al. 2007. Expression of the iron-activated nspA and secY
genes in Neisseria meningitidis group B by Fur-dependent and -indepen-
dent mechanisms. J. Bacteriol. 189:663–669.
57. Shin JH, Oh SY, Kim SJ, Roe JH. 2007. The zinc-responsive regulator
Zur controls a zinc uptake system and some ribosomal proteins in Strep-
tomyces coelicolor A3(2). J. Bacteriol. 189:4070– 4077.
58. Smyth G. 2005. Limma: linear models for microarray data, p 397– 420. In
Gentleman R, Carey V, Dudoit SR, Irizarry R, Huber W (ed), Bioinfor-
matics and computational biology solutions using R and Bioconductor.
Springer, New York, NY.
59. Stork M, et al. 2010. An outer membrane receptor of Neisseria meningi-
tidis involved in zinc acquisition with vaccine potential. PLoS Pathog.
6:e1000969. doi:10.1371/journal.ppat.1000969.
60. Tettelin H, et al. 2000. Complete genome sequence of Neisseria menin-
gitidis serogroup B strain MC58. Science 287:1809–1815.
61. Thompson WA, Newberg LA, Conlan S, McCue LA, Lawrence CE.
2007. The Gibbs centroid sampler. Nucleic Acids Res. 35:W232–W237.
62. UniProt Consortium. 2012. Reorganizing the protein space at the Uni-
versal Protein Resource (UniProt). Nucleic Acids Res. 40:D71–D75.
63. van Alen T, et al. 2010. Comparative proteomic analysis of biofilm and
planktonic cells of Neisseria meningitidis. Proteomics 10:4512–4521.
64. Wu HJ, et al. 2006. PerR controls Mn-dependent resistance to oxidative
stress in Neisseria gonorrhoeae. Mol. Microbiol. 60:401–416.
65. Yang YH, Speed T. 2002. Design issues for cDNA microarray experi-
ments. Nat. Rev. Genet. 3:579–588.
Zinc-Responsive Regulon of Meningococci
December 2012 Volume 194 Number 23 jb.asm.org 6603