JOURNAL OF BACTERIOLOGY, Nov. 2010, p. 5923–5933
0021-9193/10/$12.00 doi:10.1128/JB.00602-10
Copyright © 2010, American Society for Microbiology. All Rights Reserved.
Vol. 192, No. 22
Functional Characterization of a Cyanobacterial OmpR/PhoB Class
Transcription Factor Binding Site Controlling Light Color Responses䌤
Ryan P. Bezy and David M. Kehoe*
Department of Biology, 1001 East Third Street, Indiana University, Bloomington, Indiana 47405
Received 26 May 2010/Accepted 26 July 2010
Complementary chromatic acclimation (CCA) allows many cyanobacteria to change the composition of their
light-harvesting antennae for maximal absorption of different wavelengths of light. In the freshwater species
Fremyella diplosiphon, this process is controlled by the ratio of red to green light and allows the differential
regulation of two subsets of genes in the genome. This response to ambient light color is controlled in part by
a two-component system that includes a phytochrome class photoreceptor and a response regulator with an
OmpR/PhoB class DNA binding domain called RcaC. During growth in red light, RcaC is able to simultaneously activate expression of red light-induced genes and repress expression of green light-induced genes
through binding to the L box promoter element. Here we investigate how the L box functions as both an
activator and a repressor under the same physiological conditions by analyzing the effects of changing the
position, orientation, and sequence of the L box. We demonstrate that changes in the local sequences surrounding
the L box affect the strength of its activity and that the activating and repressing functions of the L box are
orientation dependent. Also, the spacing between the L box and the transcription start site is critical for it to work
as an activator, while its repressing role during light regulation requires additional upstream and downstream DNA
sequence elements. The latter result suggests that the repressing function of RcaC requires it to operate in
association with multiple additional DNA binding proteins, at least one of which is functioning as an activator.
as PBS structural proteins called linkers, are highly abundant
in RL and virtually undetectable in GL (10, 12, 34). Conversely, RNA levels from the cpeBA operon, which encodes the
apo-PE subunits, and the cpeCDESTR operon (hereafter
called the cpeC operon), whose first three genes encode the PE
linkers, are high in GL and low in RL (16–18, 36). Many
additional genes are CCA regulated, including pcyA, which
encodes the biosynthetic enzyme necessary to produce the
chromophore used for RL absorption by PC and is expressed
five times more highly in RL than in GL in F. diplosiphon (1).
There are also many genes that are either weakly or not CCA
regulated, including a number that encode components of the
PBS that are present in both RL and GL, such as the operon
encoding PC1, cpcB1A1 (herein called the cpc1 operon) (11,
37, 43).
CCA is regulated by two distinct signal transduction pathways (30). The best characterized of these is the Rca system,
which is a complex two-component system. It consists of a
sensor called RcaE that contains a chromophore binding domain and is a member of the phytochrome superfamily (28,
45), a single-domain response regulator named RcaF that appears to act after RcaE (27), and a multidomain response
regulator called RcaC that is a member of the OmpR/PhoB
family of transcription factors and apparently acts after RcaF
(7, 27). The Rca system controls both RL- and GL-upregulated
genes by the binding of RcaC to a 7-bp direct repeat DNA
sequence, separated by 4 bp, called the L box (5⬘-TTGCACA
N4TTGCACA-3⬘) (32). Each of the promoters of the genes
that are upregulated in RL, i.e., pcyA and the cpc2 operon,
contains an L box, with the 3⬘ end located at positions ⫺22 and
⫺24, respectively, relative to the CCA-regulated transcription
start site, and these function as activating elements in RL (1)
(Fig. 1A). There is also an L box upstream of the cpeC operon,
Cyanobacteria possess the ability to modify their photosynthetic light-harvesting antennae, or phycobilisomes (PBS), in
response to changes in a wide range of abiotic cues, including
both light intensity and color (20–22). Color responsiveness is
called chromatic acclimation. This process is widespread in
marine and freshwater environments and exists in multiple
forms, the best characterized of which is called complementary
chromatic acclimation (CCA; historically referred to as “adaptation”) (15, 30, 38, 44). The model organism for the study of
CCA and its regulation is the freshwater filamentous cyanobacterium Fremyella diplosiphon UTEX 481/Tolypothrix sp. PCC
7601. During CCA, PBS production in this organism is influenced by the ratio of green light (GL) to red light (RL) in the
environment. Growth in RL leads to the accumulation of two
different forms of an RL-absorbing chromoprotein called phycocyanin (PC) in the outer regions of the PBS. These two
forms are called PC1 and PC2. Growth in GL results in the
accumulation of PC1 and another chromoprotein, called phycoerythrin (PE), which maximally absorbs GL, in the outer
portions of the PBS. Thus, cyanobacteria that are capable of
CCA can maximize photon capture for photosynthesis by tailoring the color absorption profiles of their PBS to the spectral
distribution of ambient light in the green and red regions of the
spectrum.
CCA-controlled changes in PC2 and PE abundance have so
far been found to occur entirely at the level of transcription.
Transcripts from the cpcB2A2H2I2D2 operon (hereafter called
the cpc2 operon), which encodes the PC2 apoproteins as well
* Corresponding author. Mailing address: Department of Biology,
Indiana University, 1001 E. Third Street, Bloomington, IN 47405.
Phone: (812) 856-4715. Fax: (812) 855-6705. E-mail: dkehoe@indiana
.edu.
䌤
Published ahead of print on 10 September 2010.
5923
5924
BEZY AND KEHOE
FIG. 1. Locations and sequences of L boxes in the genomes of
Fremyella diplosiphon and two other cyanobacteria. (A) Placement
and orientation of known L boxes (closed triangles) within the F.
diplosiphon, Synechococcus sp. PCC 7335, and Pseudanabaena sp.
PCC 7409 genomes. The orientation of each triangle denotes the
directionality of the L box in front of each open reading frame,
indicated with arrows. (B) Alignment of L box sequences upstream
of pcyA and the cpc2 and pebAB operons in Synechococcus sp. PCC
7335 (Syn), pcyA and the cpc2 and cpeC operons in F. diplosiphon
(Fd), and the cpc2 operon in Pseudanabaena sp. PCC 7409 (Psa). L
box repeats are underlined. Shaded letters indicate shared sequence
identity with the consensus L box repeat (see the text). Nonitalicized numbers are for known transcription start sites in F. diplosiphon and Pseudanabaena sp. PCC 7409, and italicized numbers are
for presumptive translation start sites in Synechococcus sp. PCC
7335. The dotted line demarks the sequence that makes up the L
box repeats and flanking sequence, as defined in the text. (C) Analysis of similarity of L box repeats and flanking sequences within F.
diplosiphon (F. dip.), between F. diplosiphon and Synechococcus sp.
PCC 7335 (7335), and within Synechococcus sp. PCC 7335.
J. BACTERIOL.
but it is in the opposite orientation from that of the pcyA and
cpc2 L boxes, and its 3⬘ end is located at position ⫺71 (1). Thus
far, no other GL-upregulated operons have been found to be
flanked by an L box, consistent with the fact that their regulation by the Rca system appears to be through the cpeC operon,
since the final gene in that operon encodes an activator called
CpeR that is required for the expression of cpeBA and other
GL-upregulated genes (2, 8, 40). The cpeC L box was recently
shown to function as a negative regulator of transcription in
RL (32). Thus, in RL the L box acts as both an activating and
a repressing element, and these activities have been correlated
with the orientation and location of this element relative to the
transcription start sites of the genes it controls.
In addition to the ability of L boxes associated with different
genes to either activate or repress transcription, each specific L
box imparts a different relative degree of induction or repression of light-regulated gene expression. The cpc2 L box mediates a ⱖ20-fold increase in expression in RL compared to the
GL level (6), the pcyA L box imparts 5-fold induction in RL
(1), and although the differences in location and orientation
make comparisons to the other two L boxes difficult, the cpeC
L box provides 2- to 3-fold repression of expression in RL (32).
Interestingly, the apparent binding affinity of RcaC for each of
these L boxes mirrors these differences in activity, since it is
strongest for the cpc2 operon and weakest for the cpeC operon
(32). These differences in activity and binding may be due to
variations in the sequences within the two repeats or flanking
them: the repeats are identical in the cpc2 and pcyA L boxes
but contain a 1-bp change in the cpeC L box, while the sequences between and flanking the pcyA L box repeats are quite
diverged from those found around both the cpc2 and cpeC
repeats, which are very highly conserved (1).
In addition to the issue of how local sequence variations
affect L box activity levels, there is also the question of how
these elements function as both activating and repressing elements under the same physiological conditions. The most likely
possibility is that the differences are controlled by the position
and orientation of these elements relative to the transcription
start site. It is also very possible that the activities of some or
all of these L boxes are modified by other DNA sequence
elements and their binding proteins, although at least for the
cpc2 L box, the sequences within the region approximately 250
bp upstream of the L box do not further boost activity in RL
(6). However, there is strong evidence that the cpc2 and pcyA
promoters are very weakly active in the absence of the L box or
the Rca system, while the cpeC promoter is quite strong (1, 32,
33), and although the reason(s) for these differences in promoter strength is not yet understood, it is clear that the
strength of these promoters in the absence of Rca system
influence predisposes them to control via repression or activation.
The study of RcaC and L box function is important because
OmpR/PhoB class transcription factors are abundant in cyanobacteria (32), but unlike the case for other bacteria (31), we
have little knowledge about the mechanisms through which
they operate. Here we provide a functional analysis of the
effects of changing the L box sequence, position, and orientation on the ability of this element to both activate and repress
transcription via the Rca regulatory pathway.
VOL. 192, 2010
CYANOBACTERIAL L BOX CHARACTERIZATION
5925
TABLE 1. Oligonucleotides used in this study
Primer
Construct
PCc3⬘B
PCc5⬘P
PC1Gjxn
400cpeC
Sequence (5⬘ to 3⬘)
GCGGGATCCGCTTGGGAGACTACTTTTGC (BamHI site underlined)
GCGCTGCAGGAATTCTGTATTGATGTCTTCTAC (PstI site underlined)
GCGGGATCCGCTTGGGAGACTACTTTTGC (BamHI site underlined)
GCGGCATGCAATACTGCCAACATCGGTGGT (SphI site underlined)
Primers for cpeC promoter truncations
200cpeC
pRB2
GCGGCATGCTGGGGATGAGGAGGATGAGG (SphI site underlined)
150cpeC
pRB3
GCGGCATGCAATAACCATTACCCATTCCCCATTC (SphI site underlined)
110cpeC
pRB4
GCGGCATGCTCCCATGCCCATTACAAATAGTT (SphI site underlined)
52cpeC
pRB5
GCGGCATGCAGATTTACAAATCTTGATGTACA (SphI site underlined)
Primers for cpeC substitutions (⫺150
130DRF
pRB1⌬A
130DRR
pRB1⌬A
InvRepF
pRB1⌬B
InvRrepR
pRB1⌬B
NearlbF
pRB1⌬C
NearlbR
pRB1⌬C
130lnv-Dbl
pRB1⌬AB
130DRR
pRB1⌬AB
to ⫺90 region)
ACTAGTCTATCCGGATAGCTCACTAAGCAAGTTCTCATGCCCATTACAAATAGTTTGTG
GCTTAGTGAGCTATCCGGATAGACTAGTGGTTATTTCCCCTCATCCCCCTC
ATTACCCAGATCCAACTGAAGCATACAAATAGTTTGTGCAAATTGAGTGCAA
GGAATGGGTAATGGTTATTTCCCCT
CCCTAGCGCTATACTCTGACATTGTGCAAATTGAGTGCAAAATTCT
CACAATGTCAGAGTATAGCGCTAGGGAAATTGGGTAATGGGCAATGG
GGGAAATAACCATTACCCATTCCCCATTCCCCATTACCCAGATCCAACTGAAGCATA
GCTTAGTGAGCTATCCGGATAGACTAGTGGTTATTTCCCCTCATCCCCCTC
Primers for cpc1 L box variation
PCc-LBrand1
pPCc1
PCc-LBrand2
pPCc1
PCc-cpc2lb1
pPCc2
PCc-cpc2lb2
pPCc2
PCc-cpc2-5lb1
pPCc3
PCc-cpc2-5lb2
pPCc3
PCc-pcyalb1
pPCc4
PCc-pcyalb2
pPCc4
PCc-pcya5lb1
pPCc5
PCc-pcya5lb2
pPCc5
PCc-cpeC-LB1
pPCc7
PCc-cpeC-LB2
pPCc7
PCc-33cpeC1
pPCc10
PCc-33cpeC2
pPCc10
PCc-cpcRlb1
pPCc8
PCc-cpcRlb2
pPCc8
PCc-cpcMlb1
pPCc9
PCc-cpcMlb2
pPCc9
PCc-pcy⫹2-1
pPCc11
PCc-pcy⫹2-2
pPCc11
cpc1Lbox2
pPCc12
cpc1Lbox1
pPCc12
cpc1Lmut1
pPCc13
cpc1Lmut2
pPCc13
TCGTTGTAGACTATTTGTACGTAATTTAAGTTAGTCATACTAAGAGGCTGAATTGCTT
TAAATTACGTACAAATAGTCTACAACGAACGATATAGTATAAACAAGTAATGGAAAGCTTC
GTTGGATTGTGCAAATTTTGTGCAATGTCAAGTTAGTCATAGTAAGAGGCTGAAT
TTGCACAAAATTTGCACAATCCAACGATATAGTATAAACAAGTAATGGAAAG
TCGTTTGTGCAAATTTTGTGCAAATTTAAGTTAGTCATACTAAGAGGCTGAATTGCTT
TAAATTTGCACAAAATTTGCACAAACGAACGATATAGTATAAACAAGTAATGGAAAGCTTC
TAGCCTGTGCAATATCTGTGCAAATTGTCAAGTTAGTCATACTAAGAGGCTGA
TTGCACAGATATTGCACACCAACGATATAGTATAAACAAGTAATGGAAAG
TAGCCTGTGCAATATCTGTGCAATTTGCCAAGTTAGTCATACTAAGAGGCTGAATTGC
GCAAATTGCACAGATATTGCACAGGCTAGATATAGTATAAACAAGTAATGGAAAGCTTCAC
TAGTTTGTGCAAATTGAGTGCAAAAAAATTCCCTCTTAGTATGACTAACTTGACAATTCGT
GAATTTTGCACTCAATTTGCACAAACTACCTTTATTAACAAAATTTTTAATGCTTAACAC
TAGTTTGTGCAAATTGAGTGCAAAATTCAGTTAGTCATACTAAGAGGCTGAAT
GAATTTTGCACTCAATTTGCACAAACTAACGATATAGTATAAACAAGTAATGGAA
TAAATTTGCACAAAATTTGCACAAACGAAGTTAGTCATACTAAGAGGCTGAATTGCTT
TAGTTTGTGCAAATTGAGTGCAAAATTCCTTTATTAACAAAATTTTTAATGCTTAACAC
TCGTTTGTGCAAATTTAGTGCAAATTTAAGTTAGTCATACTAAGAGGCTGAATTGCTT
TAAATTTGCACTAAATTTGCACAAACGAACGATATAGTATAAACAAGTAATGGAAAGCTTC
TAGCCTGTGCAATATCTGTGCAATTTGCAGTTAGTCATACTAAGAGGCTGAATTGCTT
GCAAATTGCACAGATATTGCACAGGCTAACGATATAGTATAAACAAGTAATGGAAAGCTTC
TTGCACTCAATTTGCACAAACTATTTGTAATGGGGGCATGGGAAATT
CAAATAGTTTGTGCAAATTGAGTGCAATTCAGCCTCTTAAAGTATGACTAACTTGACAAT
CAAATAGTTACCGTCAATTGCCTACTATTCAGCCTCTTAGTATGACTAACTTGACAAT
TAGTAGGCAATTGACGGTAACTATTTGTAATGGGCATGGGAAATT
cpc2 primers
PCiPR
PCiPF
GCGCTGCAGCTGTTTTTTGAACCTGAAATTGATT (Pstl site underlined)
GCGGCATGCCATAAATACATTACAAAATCTGCTA (Sphl site underlined)
pPC2-C
pPC2-C
MATERIALS AND METHODS
Strains, growth conditions, and transformation. The shortened filament mutant strain SF33 (9) of Fremyella diplosiphon UTEX 481 (also known as Tolypothrix sp. PCC 7601) was the wild type, and cultures were grown as previously
described (40), with or without 10 ␮g of kanamycin (Kan) per ml. Cells were
grown in light intensities of approximately 15 ␮mol photons m⫺2 s⫺1 and transformed by electroporation as described previously (29).
Plasmid construction. All position numbers used are relative to the transcription start site unless otherwise noted. All PCR-amplified DNA regions and
ligation junctions were checked by sequencing to ensure that no mutations had
occurred during cloning.
Constructs with the cpc1 promoter harboring various versions of the L box
were created by separate PCR amplifications of the upstream and downstream
regions. An outside primer, PCc5⬘P, was paired with one of several inside
primers to amplify the upstream portion. A second outside primer, PCc3⬘B, was
paired with one of several other inside primers to amplify the downstream
portion of the promoter and the 5⬘ leader of cpc1. The inside primers introduced
mutated sequences into the cpc1 promoter that created different versions of the
L box and provided a region of sequence overlap between the upstream and
downstream amplified fragments that allowed annealing and a final PCR amplification using the outside primers from the first two amplifications. This created
a series of cpc1 promoters from positions ⫺412 to ⫹286 with altered L box
sequences, as described below. The upstream and downstream outside primers
also added PstI and BamHI sites, respectively, which were used to cut the final
PCR amplification product for insertion into similarly cut pPCc412G (6). Primer
sequences and the plasmids they were used to create are listed in Table 1.
pPC2-C was made by PCR amplifying the cpc2 promoter from positions ⫺285
to ⫹20, using the PCiPR and PCiPL primers. The product was digested with
SphI and PstI and inserted into similarly cut p400cpeC (32). The name of
plasmid p400cpeC was changed to pRB1 for simplicity.
To create a series of 5⬘ cpeC promoter truncations (Table 1), the cpeC promoter and cpc1 5⬘ leader were PCR amplified, using pRB1 as a template, a series
of upstream primers corresponding to sequences at truncation endpoints, and
the downstream primer PC1Gjxn. The upstream primers contained an SphI site,
5926
BEZY AND KEHOE
and PC1Gjxn contained a BamHI site. Amplification products were cut with
SphI and BamHI and inserted into similarly cut pRB1. Mutations in the cpeC
promoter between positions ⫺150 and ⫺90 (Table 1) were created via two-part
PCR amplification, using pRB1 as a template. The upstream region was amplified using the p400cpeC primer paired with a series of different inside primers,
and the downstream region was amplified with PC1Gjxn and a series of different
inside primers. All inside primers were designed to generate a random sequence
to replace portions of the region comprising positions ⫺150 to ⫺90 (see Results)
and to provide the overlapping sequence needed to allow annealing of the two
initial amplification products in the third PCR amplification, which was conducted using the primers 400cpeC and PC1Gjxn, containing SphI and BamHI
sites, respectively. The final product was cut with SphI and BamHI and inserted
into similarly cut pRB1 to produce the constructs listed in Table 1.
To create the two plasmids containing fusions between the cpeC upstream and
cpc1 promoter and 5⬘ leader regions (with and without the L box) (Table 1), the
region of the cpeC operon from positions ⫺421 to ⫺71 was PCR amplified either
with primers 400cpeC and cpc1Lbox1 or with primers 400cpeC and cpc1Lmut1.
A second PCR amplified the cpc1 promoter and 5⬘ leader regions from positions
⫺70 to ⫹286, using either primers PCc3⬘B and cpc1Lbox2 or primers PCc3⬘B
and cpc1Lmut2. The cpc1Lbox1 and cpc1Lbox2 primers and the cpc1Lmut1 and
cpc1Lmut2 primers were designed to introduce complementary sequences into
both products that were used to anneal the products from the first two amplifications, which were then reamplified using primers 400cpeC and PCc3⬘B. The
product was cut with SphI and BamHI and inserted into similarly cut pRB1 to
create the plasmids.
Bioinformatics. Nucleotide sequences for the CCA-capable cyanobacteria
Pseudanabaena sp. PCC 7409 (http://www.crbip.pasteur.fr/fiches/fichecata.jsp?crbip
⫽PCC%207409) and Synechococcus sp. PCC 7335 (http://www.crbip.pasteur.fr
/fiches/fichecata.jsp?crbip⫽PCC%207335) are available through NCBI accession
number M99427 and at https://research.venterinstitute.org/moore/SingleOrganism.do
?speciesTag⫽S7335&pageAttr⫽pageMain, respectively.
GUS assays. Beta-glucuronidase (GUS) assays were modified from previous
protocols (6, 24). F. diplosiphon transformants harboring the respective promoter-gusA plasmids were grown in 50 ml liquid BG-11 with 10 ␮g/ml Kan to an A750
of ⬃0.7, and then 0.35 ml of each culture was centrifuged at room temperature
for 3 min at 4,500 ⫻ g. Pellets were resuspended in 1 ml GUS assay buffer (50
mM NaPO4, pH 7.0, 1 mM EDTA) supplemented with 12 ␮g/ml chloramphenicol and centrifuged as before. Pellets were then resuspended in 1 ml of GUS
assay buffer supplemented with 20 ␮l of 0.1% SDS and 40 ␮l of chloroform, and
samples were vortexed for 10 s. Three replicates were conducted for each transformant, and each construct was assayed using a minimum of four independently
transformed lines. In each assay, 20 ␮l of cell lysate was mixed with 180 ␮l of
GUS assay buffer containing 1.25 mM ␣-p-nitrophenyl ␤-D-glucuronide (PNPG)
and incubated at room temperature. Absorption at 405 nm was recorded every
2 min for 30 min with a Molecular Devices SpectraMax 190 spectrophotometer
(Molecular Dynamics, Sunnyvale, CA). The total protein concentration of each
sample was determined using a Pierce BCA protein assay reagent kit per the
manufacturer’s instructions. GUS activity was quantified as nmol of product per
mg of protein per min.
RESULTS
L box contexts are comparable in promoters from several
CCA-capable cyanobacteria. In F. diplosiphon, the L boxes
upstream of the pcyA gene and the cpc2 and cpeC operons
differ in orientation and position, and these differences correspond to the expression of these genes in either RL or GL (1)
(Fig. 1A). The L boxes within the pcyA and cpc2 promoters,
which are both upregulated in RL, are in the same orientation
and location relative to the transcription start site. We designate this the “forward” orientation of the L box relative to the
direction of transcription of the affected gene. The L box
upstream of the GL-upregulated cpeC operon is in the opposite direction, which we refer to as the “reverse” orientation.
The pcyA and cpc2 L boxes act as activating elements in RL,
while the cpeC L box acts as a repressing element in RL (1, 32).
The locations, orientations, and flanking sequences of these
three L boxes are all different, even though the core repeats
J. BACTERIOL.
within them are completely conserved, with the exception of a
single base pair within the cpeC operon.
Thus far, the only cyanobacterium that is capable of CCA
whose completely sequenced genome is available is Synechococcus sp. PCC 7335. Its genome contains the genes that encode the three components of the Rca system, RcaE, RcaF,
and RcaC, as well as L boxes immediately 5⬘ of the pcyA gene
and the cpc2 and pebAB operons (Fig. 1A). In addition, the L
boxes upstream of pcyA and the cpc2 operon in Synechococcus
sp. PCC 7335 are in the forward orientation, while the pebAB
operon, which has been shown to be CCA regulated and more
highly expressed in GL in F. diplosiphon (2), is flanked by an L
box in the reverse orientation. Thus, it is possible that Synechococcus sp. PCC 7335 uses the Rca system and L boxes to
control CCA in a manner quite similar to that of F. diplosiphon. Pseudanabaena sp. PCC 7409 also undergoes CCA and
contains an L box upstream of the cpc2 operon, in the forward
orientation and the same position relative to transcription as
that in F. diplosiphon (13, 44). Collectively, these data suggest
that although some features of the Rca regulatory pathway
differ in these species, they share basic features of this system,
including the correlation between the orientation of the L
boxes and their apparent use in activating and repressing gene
expression in RL.
The L box repeats and flanking sequences for F. diplosiphon,
Synechococcus sp. PCC 7335, and Pseudanabaena sp. PCC
7409 are shown in Fig. 1B, and a comparative analysis of the
first two is provided in Fig. 1C. There is an extremely high level
of sequence conservation within the six known F. diplosiphon
repeats (41 of 42 bp identical) and somewhat less, but still
strong, conservation within the Synechococcus sp. PCC 7335
repeats (37 of 42 bp identical) and between the repeats of the
three species (36 of 42 bp identical between F. diplosiphon and
Synechococcus sp. PCC 7335 and 13 of 14 bp identical between
F. diplosiphon and Pseudanabaena sp. PCC 7409). It is also
noteworthy that the differences between the repeats occur only
at the outer nucleotide of the 5⬘ or 3⬘ end of the repeats, never
in the internal regions. The sequences flanking the repeats
(defined as the 5 bp on each outer side of the repeats) and the
4 bp between the repeats are much more diverged. Only the F.
diplosiphon cpc2 and cpeC flanking sequences (10 of 14 bp
identical), the cpc2 flanking sequences from F. diplosiphon and
Synechococcus sp. PCC 7335 (6 of 14 bp identical), and the
cpc2 flanking sequences from F. diplosiphon and Pseudanabaena sp. PCC 7409 (7 of 14 bp identical) showed aboverandom similarity. Overall, the variations in both the L box
repeat and flanking sequences appear likely to affect the activity of this cis-acting element.
Variation in sequences flanking the L box repeats controls
promoter activity. We examined the relative RL activation
capabilities of the cpc2, pcyA, and cpeC L boxes and the influence of the DNA sequences flanking the direct repeats of each
of these by placing them into the non-CCA-regulated cpc1
promoter and 5⬘ leader region. Each was joined as a translational fusion to the gusA reporter gene (Fig. 2A). Every L box
region included 5 bp of flanking DNA on each side, since this
was the region protected by RcaC in previous footprinting
studies (32), and was positioned so that the center of the 4-bp
intervening sequence between the two repeats was located at
position ⫺32/⫺33 relative to the cpc1 transcription start site
VOL. 192, 2010
FIG. 2. Relative strengths of the F. diplosiphon L boxes from pcyA and
the cpc2 and cpeC operons. (A) Diagrams representing the constructs used
for panel C. The cpc1 promoter and 5⬘ leader regions were used to drive
expression of the gusA reporter gene. The L box sequences from pcyA, the
cpc2 and cpeC operons, and the cpc2 operon with a single base pair mutation
(asterisk) (pPCc9) to match the sequence within the cpeC L box repeat or a
random sequence was used to replace the equivalently located sequence
within the cpc1 promoter at the same position and orientation as the L box in
the cpc2 operon. The cpc2 promoter from positions ⫺285 to ⫹14, fused to
the cpc1 5⬘ leader from positions ⫹20 to ⫹286, was also used to drive gusA
expression (pPC2-C). (B) Alignment of cpc1 promoter sequences, either wild
type, with a random sequence (pPCc1) substituted, or with the various L
boxes substituted. The direct repeat sequences of the L boxes are underlined,
while the entire region that was mutated is shown in bold. The asterisks
denote the single nucleotide difference in the direct repeat of the cpeC L box.
(C) Relative mean rates of GUS activity per mg of protein for cell lysates of
F. diplosiphon transformed with the indicated plasmids and grown in RL or
GL. The 100% value was 428.6 nmol of product per mg protein per min and
was derived from cells transformed with pPCc412G and grown in GL. At
least four independently transformed lines were tested for each plasmid and
light condition. The RL-to-GL ratio of mean GUS activities is provided in
parentheses above each construct. Error bars show standard errors of the
means.
CYANOBACTERIAL L BOX CHARACTERIZATION
5927
(11) (Fig. 2B). The base construct, pPCc412G, consisting of
only the cpc1 promoter and 5⬘ leader region, was slightly more
active in GL than in RL (Fig. 2C) (6). A negative-control
construct (pPCc1) containing 28 bp of random sequence in the
same position as constructs containing the L boxes showed
reduced activity that was also higher in GL than in RL (Fig.
2C). The cpc2 L box (pPCc3) conferred the greatest RL induction and overall activity of the three, with 8-fold higher RL
activity than that of the negative control. In fact, the addition
of this L box to the cpc1 promoter increased its activity to a
level comparable to that measured for the cpc2 promoter and
cpc1 5⬘ leader sequence translationally fused to gusA (Fig. 2C,
pPC2-C), suggesting that this L box is the major, and likely
only, component controlling the RL-mediated transcriptional
activation of cpc2. This is supported by previous results from a
cpc2 deletion analysis (6). The pcyA L box (pPCc11) provided
significantly less RL induction than the cpc2 L box (only 4-fold
above the negative-control level), although the activities of
these two promoters in GL were very similar. Since the 7-bp
core repeats of the pcyA and cpc2 L boxes are identical (Fig.
2B), the sequences flanking the repeats must be the major
determinants of the capacity for RL transcriptional activation.
The cpeC L box (pPCc10), which is a repressing element in its
native promoter (32), functioned as an activator when its orientation was reversed and it was placed in the same promoter
location as the cpc2 L box (Fig. 2C). However, it was less
effective than the cpc2 and pcyA L boxes at transcriptional
activation in RL, increasing levels to only 3 times that of
pPCc1. Compared to the other two L boxes, there is a 1-bp
difference within the direct repeats of the cpeC L box (Fig. 2B)
that affects its capacity to activate transcription in RL, since
changing only this base pair within the context of a cpc2 L box
and flanking sequence (pPCc9) decreased RL transcriptional
induction by approximately 50% (Fig. 2C). Therefore, the relative weakness of the cpeC L box must be due primarily to this
single base pair change, although the sequences flanking the
direct repeats must also contribute, since a promoter with the
1-bp mutant version of the cpc2 L box was still 1.4 times more
active in RL than the promoter containing the cpeC L box
element (Fig. 2C).
L box flanking sequences have a major effect on RL-mediated transcriptional activation. We analyzed the influence of
the 5-bp regions upstream and downstream of the cpc2 and
pcyA L boxes to further assess their importance in RL transcriptional activation. Two versions of each of these were created, one containing the normal flanking sequences and one in
which these regions had been modified (Fig. 3A). These were
placed into the cpc1 promoter and 5⬘ leader region, which was
translationally joined to gusA (Fig. 3B). The cpc2 L box was
positioned as in Fig. 2B, while the pcyA L box was placed 2 bp
closer to the transcription start site to match its location in its
native promoter (1). For the cpc2 L box, the loss of the upstream and downstream flanking sequences led to a decrease in
transcriptional activity of approximately 35% in RL and 16%
in GL (Fig. 3C). This shows that the cpc2 L box flanking
sequences contribute significantly to the effectiveness of L
boxes as transcriptional activation elements, while the pcyA
flanking sequences apparently contribute little more to the
effectiveness of that L box than the direct repeat sequences
alone.
5928
BEZY AND KEHOE
FIG. 3. Effect of L box flanking sequences on the strength of their
activity. (A) Diagrams representing the constructs used for panel C.
The cpc1 promoter and 5⬘ leader regions were fused to the gusA
reporter with L box sequences from the cpc2 operon with (pPCc3) or
without (pPCc2) the flanking 5 bp centered at position ⫺33/⫺32 replacing the cpc1 sequence. The pcyA L box was substituted for the cpc1
sequence with (pPCc5) or without (pPCc4) the flanking 5 bp centered
at position ⫺31/⫺30 relative to transcription, as in the pcyA promoter.
(B) Sequence alignment of the L box repeats (underlined) and substituted sequences (bold) in the cpc1 promoter. (C) Relative mean rates
of GUS activity per mg of protein for cell lysates of F. diplosiphon
transformed with the indicated plasmids and grown in RL or GL. The
100% value was 741.2 nmol of product per mg of protein per min and
was derived from cells transformed with pPCc3 and grown in RL. At
least four independently transformed lines were tested for each plasmid and light condition. The RL-to-GL ratio of mean GUS activities
is provided in parentheses above each construct. Error bars show
standard errors of the means.
Capacity for RL activation depends on L box orientation. A
central question concerning the function of L boxes is whether
their orientation is critical for their activity, particularly for
those relatively close to the transcription start site. We addressed this by reversing the orientation of the cpc2 L box
within the cpc1 promoter while maintaining its position relative
to the transcription start site (Fig. 4A and B). This reorientation involved both the direct repeat sequence and the flanking
sequence, as shown in Fig. 2B. The results clearly show that the
orientation of the L boxes that activate transcription in RL is
critical for their function (Fig. 4C). An L box placed in its
normal orientation imparted RL responsiveness (pPCc3),
J. BACTERIOL.
FIG. 4. Effect of orientation of an activating L box on its function.
(A) Diagrams representing the constructs used for panel C. The cpc1
promoter and 5⬘ leader regions were used to drive expression of the
gusA reporter gene, with the L box from the cpc2 operon substituted
for the cpc1 sequence centered at position ⫺33/⫺32, in the forward or
reverse orientation. (B) Sequence alignment of the cpc1 promoter
containing the cpc2 L box in either the wild-type (pPCc3) or reverse
(pPCc8) orientation. All numbering is relative to the transcription start
site for the cpc1 operon. L box repeats are underlined, and substituted
sequences are shown in bold. (C) Relative mean rates of GUS activity
per mg of protein for cell lysates of F. diplosiphon transformed with the
indicated plasmids and grown in RL or GL. The 100% value was 741.2
nmol of product per mg of protein per min and was derived from cells
transformed with pPCc3 and grown in RL. At least four independently
transformed lines were tested for each plasmid and light condition.
The RL-to-GL ratio of mean GUS activities is provided in parentheses
above each construct. Error bars show standard errors of the means.
while reversing the orientation of the L box actually led to the
repression of transcriptional activity in RL (pPCc8). This was
presumably the result of the continued binding of RcaC to the
L box in RL, but in a manner that interfered with DNAdependent RNA polymerase binding and/or transcription initiation. These results suggest that very specific contacts must
be made between RcaC and the basal transcription machinery
to achieve enhanced transcriptional activity in RL.
cpeC L box transcriptional repression in RL is context dependent. The ability of the cpeC L box to act as a negative
element during growth in RL raised the question of whether
this L box, if placed in the same location within a heterologous
promoter, could repress its expression in an RL-dependent
fashion or if it required the surrounding cpeC sequences to
function. We inserted the cpeC L box and 5 bp of flanking
sequence on either side into the cpc1 upstream region, promoter, and 5⬘ leader at the identical location and orientation as
in cpeC and translationally fused this construct to gusA (Fig.
5A). Cells expressing this construct (pPCc7) showed no signif-
VOL. 192, 2010
FIG. 5. Capacity of the cpeC L box to repress expression of a
heterologous promoter in RL. (A) Diagrams representing the constructs used for panel B. The cpc1 upstream, promoter, and 5⬘ leader
regions were used to drive expression of gusA without (pPCc412G) or
with (pPCc7) the cpeC L box and 5 bp of flanking sequence on each
side substituted for the cpc1 sequence and placed in the same position
and orientation as those within the cpeC promoter, centered at position ⫺79/⫺80 relative to the transcription start site. (B) Relative mean
rates of GUS activity per mg of protein for cell lysates of F. diplosiphon
transformed with the indicated plasmids and grown in RL or GL. The
100% value was 428.6 nmol of product per mg of protein per min and
was derived from cells transformed with pPCc412G and grown in GL.
At least four independently transformed lines were tested for each
plasmid and light condition. The GL-to-RL ratio of mean GUS activities is provided in parentheses above each construct. Error bars show
standard errors of the means.
icant decrease in GUS activity in RL compared to the control
containing the cpc1 upstream region, promoter, and 5⬘ leader
region only (pPCc412G) (Fig. 5B). These data suggest that the
RL-dependent repressing activity of the cpeC L box is context
dependent. It is possible that RcaC binding to the cpeC L box
in RL disrupts the action of other DNA elements that function
as activators within the upstream region and that similar activating elements are not in equivalent regions of cpc1.
An element(s) that activates high-level expression of cpeC is
located upstream of the L box. To determine whether the cpeC
promoter and upstream sequences contained any activating
DNA elements, we constructed a series of 5⬘ deletions of this
region. The base construct, pRB1, contained the cpeC promoter from positions ⫺421 to ⫹38 fused to the 5⬘ leader
region of cpc1 (from positions ⫹10 to ⫹286, encoding the first
11 amino acids), which had been joined translationally to gusA
(Fig. 6A). This construct was expressed 2.5 times more in GL
than in RL (Fig. 6B), and this difference was due to the Rca
system operating through the cpeC L box (32).
Analysis of constructs in which regions upstream of cpeC
had been removed demonstrated that the sequence down to
position ⫺150 could be changed without affecting GL expression but that removal of the sequence from positions ⫺150 to
CYANOBACTERIAL L BOX CHARACTERIZATION
5929
FIG. 6. 5⬘-Deletion analysis of the region upstream of cpeC.
(A) Diagrammatic representation of the constructs used for panel B.
The cpeC upstream and promoter regions and cpc1 5⬘ leader were used
to drive gusA reporter gene expression. Negative numbers on the left
refer to deletion endpoint nucleotides for cpeC. (B) Relative mean
rates of GUS activity per mg of protein for cell lysates of F. diplosiphon
transformed with the indicated plasmids and grown in RL or GL. The
100% value was 285.6 nmol of product per mg of protein per min and
was derived from cells transformed with pRB2 and grown in GL. At
least four independently transformed lines were tested for each plasmid and light condition. The GL-to-RL ratio of mean GUS activities
is provided in parentheses above each construct. Error bars show
standard errors of the means.
⫺110 resulted in the loss of high-level expression in GL and
the loss of CCA regulation (Fig. 6B). Further loss of the cpeC
sequence down to position ⫺52, which eliminated the L box,
resulted in slightly less overall expression than that with the
⫺110 truncation, with a continued loss of the CCA response.
Taken together, these results suggest that there is one or more
positively acting element in the sequence between positions
⫺150 and ⫺110 upstream of the cpeC start site and that this
element(s) is required for the normal CCA response of cpeC.
The cpeC activating element(s) between positions ⴚ150 and
ⴚ110 is redundant or extended. Because the DNA sequence
from positions ⫺150 to ⫺110 of cpeC increases transcriptional
activity, particularly in GL (Fig. 6B), it was further examined
for an activating element(s). This region contains a direct repeat, shown in Fig. 7A and B. Four internal substitution mutants were created, using pRB1. These replaced the sequence
from positions ⫺142 to ⫺118 (pRB1⌬A), which eliminated
direct repeat A (5⬘-CCATTACCCATTN2CCATTCCCCATT
A-3⬘); the sequence from positions ⫺114 to ⫺100 (pRB1⌬B),
which eliminated region B, immediately downstream of direct
repeat A; the sequence from positions ⫺107 to ⫺90
5930
BEZY AND KEHOE
J. BACTERIOL.
FIG. 7. Internal substitution analysis of the region immediately upstream of the cpeC L box to examine activating elements. (A) Diagram
representing the constructs used for panel C. The cpeC upstream, promoter, and 5⬘ leader from positions ⫺421 to ⫹38 were fused to the cpc1 5⬘
leader sequence from positions ⫹10 to ⫹286. A direct repeat centered at position ⫺130 (A), a 15-bp region centered at position ⫺107 (B), and
a region upstream of the L box (C) were identified as possible positively acting elements or regions that operate upstream of the L box and were
eliminated by replacement with a random sequence. (B) Sequence alignments of the various constructs used for panel C. The cpeC upstream region
is shown from the L box (boxed region) at position ⫺71 up to position ⫺150. The bold underlined sequences were mutated to a random sequence
in order to remove possible cis-acting regions A, B, and C, which are denoted at the top. Arrows denote the direction of the A direct repeat.
(C) Relative mean rates of GUS activity per mg of protein for cell lysates of F. diplosiphon transformed with the indicated plasmids and grown
in RL or GL. The mean value was 263.4 nmol of product per mg of protein per min and was derived from cells transformed with pRB1 and grown
in GL. At least four independently transformed lines were tested for each plasmid and light condition. The GL-to-RL ratio of mean GUS activities
is provided in parentheses above each construct. Error bars show standard errors of the means.
(pRB1⌬C), which eliminated region C, immediately upstream
of the L box; and the sequence from positions ⫺142 to ⫺100
(pRB1⌬AB), which combined the substituted regions of
pRB1⌬A and pRB1⌬B and eliminated the A and B regions
(Fig. 7A and B).
The GL expression level did not decrease for the pRB1⌬A
and pRB1⌬B mutants, as would be expected if a sufficient
amount of a single essential activating element had been eliminated in either of these regions (Fig. 7C). The GL expression
level was somewhat elevated in the combination mutant
pRB1⌬AB, which suggests that there are no elements between
positions ⫺142 and ⫺100 that work redundantly to transcriptionally activate cpeC. For construct pRB1⌬C, the activity increased significantly in both GL and RL compared to that
measured for pRB1.
cpeC L box function requires cpeC sequences downstream of
the L box. The results in Fig. 6 demonstrated that the L box
which represses cpeC expression in RL (32) cannot function
in the absence of sequences upstream of position ⫺110,
which are required for high levels of cpeC transcriptional
activation. It is possible that cpeC sequences downstream of
the L box are also required for a sufficiently high level of
cpeC expression in GL to allow CCA activity. To test this,
the region of cpc1 from positions ⫺412 to ⫺71 from
pPCc412G was replaced with either the native cpeC sequence from positions ⫺421 to ⫺71, which contains the L
box (pPCc12), or the equivalent region with a mutated L box
sequence (pPCc13) (Fig. 8A). Neither of these constructs
was expressed in a significantly different fashion in either
RL or GL from that for the base construct pPCc412G, even
though they were all highly expressed (Fig. 8B). This indicates that the CCA regulation of cpeC transcriptional activity through the repressing action of the L box cannot occur
without specific DNA sequences downstream of the L box,
specifically between positions ⫺70 and ⫹30, since only this
region was different between pPCc12 and pRB1 (Fig. 6),
which was CCA regulated.
DISCUSSION
The functional studies presented here demonstrate that the
sequence context, position, and orientation of L boxes are all
important for their function as CCA activating and repressing
elements during growth in RL, a hypothesis that emerged from
VOL. 192, 2010
FIG. 8. Effectiveness of the cpeC L box in the absence of downstream cpeC sequence. (A) Diagrams representing the constructs used
for panel B. The cpc1 upstream, promoter, and 5⬘ leader regions
(pPCc412G), the cpeC upstream region from positions ⫺400 to ⫺71
(with the L box) fused to the cpc1 promoter and 5⬘ leader sequence
from positions ⫺70 to ⫹286 (pPCc12), or the cpeC upstream region
from positions ⫺400 to ⫺71 and containing random sequence substituted for the L box fused to the cpc1 promoter and 5⬘ leader sequence
from positions ⫺70 to ⫹286 (pPCc13) were used in translational fusions to drive the expression of gusA. (B) Relative mean rates of GUS
activity per mg of protein for cell lysates of F. diplosiphon transformed
with the indicated plasmids and grown in RL or GL. The mean value
was 428.6 nmol of product per mg of protein per min and was derived
from cells transformed with pPCc412G and grown in GL. At least four
independently transformed lines were tested for each plasmid and light
condition. The GL-to-RL ratio of mean GUS activities is provided in
parentheses above each construct. Error bars show standard errors of
the means.
the comparison of L box properties in three cyanobacterial
species capable of CCA (32) (Fig. 1). In F. diplosiphon, the L
box repeats are highly conserved (either 13 and 14 bp or 14 and
14 bp), while the flanking sequences are more variable (from 5
and 14 bp to 10 and 14 bp). The data presented here demonstrate that the flanking sequences are major determinants of
the element’s activity. A clear example of this is provided in
Fig. 2, where the relative strengths of the L boxes with their
flanking sequences are compared in identical promoter backgrounds. The cpc2 L box conferred twice the level of RL
activity as the pcyA L box (Fig. 2C), and this difference was due
only to the differences in flanking sequences (Fig. 2A and B).
These results are supported by the apparent binding affinity
differences of RcaC for the cpc2 and pcyA L boxes (32) and the
relative activities of the cpc2 and pcyA promoters in RL (data
not shown), although the latter data could be influenced by
other aspects of promoter architecture. However, although the
flanking sequences of the cpc2 and cpeC L boxes are much
more similar (10 and 14 bp) than those of the cpc2 and pcyA L
boxes (5 and 14 bp), our results demonstrate that the cpeC L
box is actually weaker in RL than the pcyA L box (Fig. 2C).
CYANOBACTERIAL L BOX CHARACTERIZATION
5931
This might be partly because the 4 bp that differ between the
cpc2 and cpeC L box flanking sequences are critical for highaffinity RcaC binding. In addition, the single nucleotide
change at the end of one of the cpeC repeats (compared to
the cpc2 repeats) (Fig. 2B) also influences the strength of
RcaC binding, as demonstrated by the 2-fold lower activity
of the cpc2 L box containing this change than that of the
wild-type L box (Fig. 2C).
The L box repeat sequences are also quite highly related
between species. Despite the fact that neither Synechococcus
sp. PCC 7335 nor Pseudanabaena sp. PCC 7409 is closely
related to F. diplosiphon within the cyanobacteria (23), the
repeats present in the cpc2 promoters of these three species
are identical in at least 12/14 positions (Fig. 1D). Overall, the
L box repeats appear to be more highly conserved between
promoters than the repeats found within the binding sites of
other members of this response regulator class, such as OmpR
and PhoB (5, 19, 31), although it remains possible that many
other, less highly conserved RcaC binding sites exist but have
not yet been identified. This L box conservation may be evidence of recent lateral gene transfer of the Rca system between cyanobacterial species or of a strong fitness advantage
that is conferred by the level of interaction that exists between
RcaC and the consensus repeat sequences within the L box.
We note that in the experiments presented here, the 10-fold
difference in RNA abundance that is normally measured for
the cpeC operon during CCA (2, 4, 40) was reduced to a 2- to
3-fold response for pRB1, which consists of the cpeC sequence
from positions ⫹38 to ⫺421 and the cpc1 5⬘ leader region from
positions ⫹10 to ⫹286 (Fig. 6). This discrepancy between GUS
activity differences and cpeC RNA accumulation differences
has been observed previously (32) and is the result of the
absence of influence of the second CCA regulatory pathway,
called the Cgi system (30), which appears to function posttranscriptionally through the 5⬘ leader of the cpeC operon (our
unpublished data).
For each of the L boxes identified thus far (Fig. 1C), the fact
that the repeats are either identical or differ in only one of the
seven positions may provide a clue to how RcaC interacts with
L boxes. PhoB and OmpR, the best-characterized members of
this group of response regulators, operate through different
mechanisms. The dimerization and subsequent DNA binding
enhancement of PhoB appear to be driven by its phosphorylation, while OmpR appears to initially bind to DNA as a
monomer and then to undergo enhanced phosphorylation,
dimerization, and binding of a second monomer (3, 35, 39).
The sequence similarity between the half-sites within the DNA
target sequence is quite high for PhoB (5), while for OmpR the
greater differences in the sequences of the half-sites result in
different binding affinities and allow the stepwise addition of
OmpR monomers to the DNA (39). Therefore, the high degree of sequence conservation within the repeat sequences of
the L boxes, together with the fact that RcaC appears to be a
dimer in its unbound, active state (L. Li and D. M. Kehoe,
unpublished data), suggests that the mechanism underlying
RcaC-L box interactions may be more similar to that controlling PhoB than to that controlling OmpR.
The activating function of the L box is clearly orientation
specific (Fig. 4C). This differs from the apparent situation for
RpaB, another cyanobacterial OmpR/PhoB class factor which
5932
BEZY AND KEHOE
binds direct repeat HLR1 elements within the promoters of
light-intensity-regulated genes and, depending on the gene,
can act as either an activator or a repressor under low-light
conditions (14, 25, 26, 41, 42). Although the dual function of
RpaB via HLR1 elements is similar to the action of RcaC in
RL through the L box, there are indications that the mechanisms through which they act are quite different. The most
notable of these is that unlike L boxes, both positively and
negatively acting HLR1 elements can be present near or just
upstream of the transcription start sites of the genes they
regulate. More importantly, although not yet tested directly,
HLR1 elements also appear to be capable of functioning in
either orientation, regardless of whether they are acting positively or negatively. For CCA regulation, the inability of an
oppositely oriented L box centered at position ⫺33/⫺32 to
activate transcription is most likely due to the loss of specific
interactions between RcaC and the basal transcription machinery that enhance transcription rates or to steric interference,
since this is a large response regulator (ca. 73 kDa) that contains at least four domains (7, 27).
The repressing function of the L box is strongly context
dependent. It failed to repress cpc1 expression in RL, despite
its placement at a location equivalent to where it is found in the
cpeC promoter (Fig. 5B). It also required sequences both upstream (Fig. 6B) and downstream (Fig. 8B) of the cpeC L box.
In particular, our results suggest that the upstream activating
elements required to attain sufficient expression for the cpeC L
box to exert its repressing function, which appear to reside
between positions ⫺88 and ⫺150, may be redundant, since the
stepwise replacement of virtually all of the sequences within
this region failed to eliminate promoter activity (Fig. 7C), even
though deletion of this region to position ⫺110 led to its loss
(Fig. 6B). Although there was some decrease in the CCA
response for the pRB1⌬C construct, this may have been the
result of having decreased RcaC binding to the L box, since
almost all of the flanking sequence upstream of the L box was
replaced in this construct (Fig. 7). Overall, our results show
that both the expression of cpeC and its regulation by CCA
involve multiple cis-acting elements that must be positioned
correctly relative to each other in order to function properly.
In summary, the data presented here advance our understanding of the mechanisms of action underlying light regulation and OmpR/PhoB class transcription factors in cyanobacteria. This study has further characterized the features of the L
box, which is the binding site of RcaC, the best-studied cyanobacterial OmpR/PhoB family member. Our results demonstrate that the L box location, orientation, and both core repeat
and flanking sequences all contribute in very specific ways to
the strength of this element and its ability to function as an
activator or repressor.
ACKNOWLEDGMENTS
This work was supported by the National Science Foundation (grant
MCB-0519433 to D.M.K.).
We thank Matt Careskey, Peter Gracheck, and Noyan Olcum for
their assistance with the work described here. We also thank the other
members of the Kehoe laboratory for thoughtful discussions and comments on this work.
REFERENCES
1. Alvey, R. M., R. P. Bezy, N. Frankenberg-Dinkel, and D. M. Kehoe. 2007. A
light regulated OmpR-class promoter element coordinates light harvesting
J. BACTERIOL.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
protein and chromophore biosynthetic enzyme gene expression. Mol. Microbiol. 64:319–332.
Alvey, R. M., J. A. Karty, E. Roos, J. P. Reilly, and D. M. Kehoe. 2003.
Lesions in phycoerythrin chromophore biosynthesis in Fremyella diplosiphon
reveal coordinated light regulation of apoprotein and pigment biosynthetic
enzyme gene expression. Plant Cell 15:2448–2463.
Bachhawat, P., G. V. T. Swapna, G. T. Montelione, and A. M. Stock. 2005.
Mechanism of activation for transcription factor PhoB suggested by different
modes of dimerization in the inactive and active states. Structure 13:1353–
1363.
Balabas, B. E., B. L. Montgomery, L. E. Ong, and D. M. Kehoe. 2003. CotB
is essential for complete activation of green light-induced genes during
complementary chromatic adaptation in Fremyella diplosiphon. Mol. Microbiol. 50:781–793.
Blanco, A. G., M. Sola, F. X. Gomis-Ruth, and M. Coll. 2002. Tandem DNA
recognition by PhoB, a two-component signal transduction transcriptional
activator. Structure 10:701–713.
Casey, E. S., and A. Grossman. 1994. In vivo and in vitro characterization of
the light-regulated cpcB2A2 promoter of Fremyella diplosiphon. J. Bacteriol.
176:6362–6374.
Chiang, G. G., M. R. Schaefer, and A. R. Grossman. 1992. Complementation
of a red-light-indifferent cyanobacterial mutant. Proc. Natl. Acad. Sci.
U. S. A. 89:9415–9419.
Cobley, J. G., A. C. Clark, S. Weerasurya, F. A. Queseda, J. Y. Xiao, N.
Bandrapali, I. D’Silva, M. Thounaojam, J. F. Oda, T. Sumiyoshi, and M. H.
Chu. 2002. CpeR is an activator required for expression of the phycoerythrin
operon (cpeBA) in the cyanobacterium Fremyella diplosiphon and is encoded
in the phycoerythrin linker-polypeptide operon (cpeCDESTR). Mol. Microbiol. 44:1517–1531.
Cobley, J. G., E. Zerweck, R. Reyes, A. Mody, J. R. Seludo-Unson, H. Jaeger,
S. Weerasuriya, and S. Navankasattusas. 1993. Construction of shuttle plasmids which can be efficiently mobilized from Escherichia coli into the chromatically adapting cyanobacterium, Fremyella diplosiphon. Plasmid 30:90–
105.
Conley, P. B., P. G. Lemaux, and A. R. Grossman. 1985. Cyanobacterial
light-harvesting complex subunits encoded in two red light-induced transcripts. Science 230:550–553.
Conley, P. B., P. G. Lemaux, and A. R. Grossman. 1988. Molecular characterization and evolution of sequences encoding light-harvesting components
in the chromatically adapting cyanobacterium Fremyella diplosiphon. J. Mol.
Biol. 199:447–465.
Conley, P. B., P. G. Lemaux, T. L. Lomax, and A. R. Grossman. 1986. Genes
encoding major light-harvesting polypeptides are clustered on the genome of
the cyanobacterium Fremyella diplosiphon. Proc. Natl. Acad. Sci. U. S. A.
83:3924–3928.
Dubbs, J. M., and D. A. Bryant. 1993. Organization and transcription of the
genes encoding two differentially expressed phycocyanins in the cyanobacterium Pseudoanabaena sp. PCC 7409. Photosynth. Res. 36:169–183.
Eriksson, J., G. F. Salih, H. Ghebramedhin, and C. Jansson. 2000. Deletion
mutagenesis of the 5⬘ psbA2 region in Synechocystis 6803: identification of a
putative cis element involved in photoregulation. Mol. Cell. Biol. Res. Commun. 3:292–298.
Everroad, C., C. Six, F. Partensky, J. C. Thomas, J. Holtzendorff, and A. M.
Wood. 2006. Biochemical bases of type IV chromatic adaptation in marine
Synechococcus spp. J. Bacteriol. 188:3345–3356.
Federspiel, N. A., and A. R. Grossman. 1990. Characterization of the lightregulated operon encoding the phycoerythrin-associated linker proteins
from the cyanobacterium Fremyella diplosiphon. J. Bacteriol. 172:4072–4081.
Federspiel, N. A., and L. Scott. 1992. Characterization of a light-regulated
gene encoding a new phycoerythrin-associated linker protein from the cyanobacterium Fremyella diplosiphon. J. Bacteriol. 179:5994–5998.
Glauser, M., D. A. Bryant, G. Frank, E. Wehrli, S. S. Rusconi, W. Sidler, and
H. Zuber. 1992. Phycobilisome structure in the cyanobacteria Mastigocladus
laminosus and Anabaena sp. PCC 7120. Eur. J. Biochem. 205:907–915.
Goh, E. B., D. F. Siino, and M. M. Igo. 2004. The Escherichia coli tppB (ydgR)
gene represents a new class of OmpR-regulated genes. J. Bacteriol. 186:
4019–4024.
Grossman, A. R., D. Bhaya, K. E. Apt, and D. M. Kehoe. 1995. Lightharvesting complexes in oxygenic photosynthesis: diversity, control, and evolution. Annu. Rev. Genet. 29:231–288.
Grossman, A. R., D. Bhaya, and Q. He. 2001. Tracking the light environment
by cyanobacteria and the dynamic nature of light harvesting. J. Biol. Chem.
276:11449–11452.
Grossman, A. R., L. G. van Waasbergen, and D. M. Kehoe. 2003. Environmental regulation of phycobilisome biosynthesis. Adv. Photosynth. Respir.
13:471–493.
Herdman, M., R. W. Castenholz, and R. Rippka. 2001. Bergey’s manual of
systematic bacteriology, 2nd ed., vol. 1. The archaea and the deeply branching and phototrophic bacteria. Springer, New York, NY.
Jefferson, R. A., S. M. Burgess, and D. Hirsh. 1986. Beta-glucuronidase from
Escherichia coli as a gene fusion marker. Proc. Natl. Acad. Sci. U. S. A.
83:8447–8451.
VOL. 192, 2010
25. Kappell, A. D., D. Bhaya, and L. G. van Waasbergen. 2006. Negative control
of the high light-inducible hliA gene and implications for the activities of the
NblS sensor kinase in the cyanobacterium Synechococcus elongatus strain
PCC 7942. Arch. Microbiol. 186:403–413.
26. Kappell, A. D., and L. G. van Waasbergen. 2007. The response regulator
RpaB binds the high light regulatory 1 sequence upstream of the high-lightinducible hliB gene from the cyanobacterium Synechocystis PCC 6803. Arch.
Microbiol. 187:337–342.
27. Kehoe, D. M., and A. R. Grossman. 1997. New classes of mutants in complementary chromatic adaptation provide evidence for a novel four-step
phosphorelay system. J. Bacteriol. 179:3914–3921.
28. Kehoe, D. M., and A. R. Grossman. 1996. Similarity of a chromatic adaptation sensor to phytochrome and ethylene receptors. Science 273:1409–1412.
29. Kehoe, D. M., and A. R. Grossman. 1998. Use of molecular genetics to
investigate complementary chromatic adaptation: advances in transformation and complementation. Methods Enzymol. 297:279–290.
30. Kehoe, D. M., and A. Gutu. 2006. Responding to color: the regulation of
complementary chromatic adaptation. Annu. Rev. Plant Biol. 57:127–150.
31. Kenney, L. J. 2002. Structure/function relationships in OmpR and other
winged-helix transcription factors. Curr. Opin. Microbiol. 5:135–141.
32. Li, L., R. M. Alvey, R. P. Bezy, and D. M. Kehoe. 2008. Inverse transcriptional activities during complementary chromatic adaptation are controlled
by binding of the response regulator RcaC to red and green light responsive
promoters. Mol. Microbiol. 68:286–297.
33. Li, L., and D. M. Kehoe. 2005. In vivo analysis of the roles of conserved
aspartate and histidine residues within a complex response regulator. Mol.
Microbiol. 55:1538–1552.
34. Lomax, T. L., P. B. Conley, J. Schilling, and A. R. Grossman. 1987. Isolation
and characterization of light-regulated phycobilisome linker polypeptide
genes and their transcription as a polycistronic mRNA. J. Bacteriol. 169:
2675–2684.
35. Mack, T. R., R. Gao, and A. M. Stock. 2009. Probing the roles of the two
different dimers mediated by the receiver domain of the response regulator
PhoB. J. Mol. Biol. 389:349–364.
CYANOBACTERIAL L BOX CHARACTERIZATION
5933
36. Mazel, D., G. Guglielmi, J. Houmard, W. Sidler, D. A. Bryant, and N.
Tandeau de Marsac. 1986. Green light induces transcription of the phycoerythrin operon in the cyanobacterium Calothrix 7601. Nucleic Acids Res.
14:8279–8290.
37. Mazel, D., J. Houmard, and N. Tandeau de Marsac. 1988. A multigene
family in Calothrix sp. PCC 7601 encodes phycocyanin, the major component
of the cyanobacterial light-harvesting antenna. Mol. Gen. Genet. 211:296–
304.
38. Palenik, B. 2001. Chromatic adaptation in marine Synechococcus strains.
Appl. Environ. Microbiol. 67:991–994.
39. Rhee, J. E., W. Y. Sheng, L. K. Morgan, R. Nolet, X. B. Liao, and L. J.
Kenney. 2008. Amino acids important for DNA recognition by the response
regulator OmpR. J. Biol. Chem. 283:8664–8677.
40. Seib, L. O., and D. M. Kehoe. 2002. A turquoise mutant genetically separates
expression of genes encoding phycoerythrin and its associated linker peptides. J. Bacteriol. 184:962–970.
41. Seino, Y., T. Takahashi, and Y. Hihara. 2009. The response regulator RpaB
binds to the upstream element of photosystem I genes to work for positive
regulation under low-light conditions in Synechocystis sp. strain PCC 6803. J.
Bacteriol. 191:1581–1586.
42. Seki, A., M. Hanaoka, Y. Akimoto, S. Masuda, H. Iwasaki, and K. Tanaka.
2007. Induction of a group 2 sigma factor, RPOD3, by high light and the
underlying mechanism in Synechococcus elongatus PCC 7942. J. Biol. Chem.
282:36887–36894.
43. Stowe-Evans, E. L., J. Ford, and D. M. Kehoe. 2004. Genomic DNA microarray analysis: identification of new genes regulated by light color in the
cyanobacterium Fremyella diplosiphon. J. Bacteriol. 186:4338–4349.
44. Tandeau de Marsac, N. 1977. Occurrence and nature of chromatic adaptation in cyanobacteria. J. Bacteriol. 130:82–91.
45. Terauchi, K., B. L. Montgomery, A. R. Grossman, J. C. Lagarias, and D. M.
Kehoe. 2004. RcaE is a complementary chromatic adaptation photoreceptor
required for green and red light responsiveness. Mol. Microbiol. 51:567–577.