Previous Article | Next Article ![]()
Journal of Bacteriology, July 2002, p. 3931-3940, Vol. 184, No. 14
0021-9193/02/$04.00+0 DOI: 10.1128/JB.184.14.3931-3940.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
and C. Peter Wolk*
MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, Michigan 48824
Received 26 December 2001/ Accepted 24 April 2002
|
|
|---|
|
|
|---|
Regulatory DNA-binding proteins that are known to affect heterocyst formation in Anabaena sp. include a histone-like protein, HU (27); transcription factor cyclic AMP receptor protein-related proteins NtcA (16, 45) and DevH (21); and PatB, a ferredoxin-like protein with a helix-turn-helix motif (33). However, RNA polymerase sigma factors SigA, which is expressed in both the presence and absence of combined nitrogen, and SigB and SigC, which are expressed transiently under nitrogen-limiting conditions, and five additional sigma factors (SigD, SigE, SigF, SigG, and SigH) are not involved specifically in the regulation of heterocyst differentiation (6, 7; I. Khudyakov, X. Wu, and J. W. Golden, Abstr. 10th Int. Symp. Phototroph. Prokar., p. 193, 2000.). HU is degraded during heterocyst differentiation, and a new, similar but different protein is synthesized in the heterocyst (37). Mutation of hanA, which encodes HU, results in a highly pleiotropic phenotype that includes slow growth, altered pigmentation, resistance to phage A-4(L), cellular fragility, and the inability of heterocysts to differentiate (27). NtcA binds upstream from xisA, which encodes a site-specific recombinase, but also upstream from glnA, rbcL, nifH, and hetC (10, 22, 36) and is required for heterocyst differentiation in Anabaena sp. (16, 36, 46). Despite this accumulation of known factors, the regulatory mechanisms that underlie the progression of differentiation in cyanobacteria remain largely unknown (48).
Earlier, hepB, hepK, and neighboring genes hepC and hepA were shown to be required for the normal formation of a principal structural element of heterocysts, the outer, polysaccharide layer of their envelope (48, 50). A DNA sequence upstream from hepC greatly influences induction of hepA upon nitrogen stepdown (50). We have identified four proteins that bind to a 150-bp sequence, X12, upstream of hepC and have shown that two of these proteins play an important role in the differentiation of heterocysts and in their ability to fix N2 aerobically.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Cyanobacterial strains and plasmids
|
DNA affinity chromatography. PCR-amplified X12 (Table 2) oligomerized by self-ligation with T4 DNA ligase was coupled to CNBr-activated Sepharose (28), and the resulting sequence-specific DNA affinity resin was used to purify DNA-binding proteins as described previously (25). One milliliter of the DNA affinity resin thus generated was equilibrated in an Econo-Column (Bio-Rad, Hercules, Calif.) with 20 ml of buffer Z (25 mM K+-HEPES [pH 7.8], 12.5 mM MgCl2, 1 mM DTT, 20% glycerol) containing 0.1 M KCl. Proteins partially purified by heparin column chromatography and extensively desalted and then diluted with 4 volumes of buffer Z containing 0.1 M KCl were allowed to bind for 1 h at 4°C with sonicated herring sperm DNA (20- to 50-fold excess by mass relative to the amount of X12 bound to the column) and then overnight with the DNA affinity resin. The protein-DNA mixture, 7 to 10 ml for one column, was allowed to pass through the Econo-Columns under gravity at ca. 15 ml/h. The resin in each column was then washed four or five times with 2 ml of buffer Z containing 0.1 M KCl. Protein bound to the column was eluted by two successive washings with 1.2 ml of buffer Z containing 1.0 M KCl. The eluted proteins were concentrated by precipitation with trichloracetic acid, the acid was removed with acetone, and the proteins were redissolved in buffer Z containing 0.05 M KCl and subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The gel was stained with silver (23) or with Coomassie brilliant blue (41).
|
View this table: [in a new window] |
TABLE 2. Primer sequences used to generate DNA fragments by PCR
|
Overexpression and purification of proteins Abp1, Abp2, Abp3, and Abp4. Genes abp1, abp2, abp3, and abp4 were amplified by PCR (for primers, see Table 2). All PCR products were verified by sequencing. Corresponding products of PCR were cloned without a His tag (43, 49) between the NdeI (CATATG) and BamHI sites (GGATCC) of plasmid pET-21b (Novagen, Inc., Madison, Wis.). The resulting plasmids were denoted, respectively, pRL2489, pRL2490, pRL2491, and pRL2492. To overproduce Abp1, Abp2, Abp3, and Abp4, E. coli strain BL21(DE3) transformed with each of these plasmids was grown in 250 ml of Luria-Bertani medium supplemented with 100 µg of ampicillin ml-1 at 37°C to an optical density at 600 nm of 0.5 to 0.6, 0.5 mM isopropyl-ß-D-thiogalactopyranoside was added, and incubation was continued for 4 h at 37°C for Abp1, Abp2, and Apb3 and overnight at 22°C for Abp4. Following Frías et al. (17), these four overexpressed proteins were partially purified as follows. E. coli suspended in 50 mM Na phosphate-100 mM NaCl (pH 7.0) was broken with a French pressure cell. The supernatant solution from centrifugation at 4°C for 20 min at 10,000 rpm and then at 4°C for 20 min at 32,500 rpm was fractionated with ammonium sulfate. The protein fraction precipitating between 30 and 50% ammonium sulfate was used for mobility shift assays.
Gel mobility shift assays for DNA-protein complexes. Mobility shift assays (18, 19) were performed by using the DIG gel shift kit (Roche) according to the manufacturer's procedures. DNA fragments A1, B2, Y, X1, X12, and X2 used in these experiments were prepared by PCR with the primers shown in Table 2; double-stranded control oligonucleotidev6, is a component of the DIG gel shift kit. The DNA fragments were 3'-end-labeled by using DIG-11-ddUTP and terminal transferase. Labeled DNA probes (0.4 to 0.8 ng) were incubated with Anabaena sp. protein extracts in binding buffer (25 mM HEPES [pH 7.9], 5 mM MgCl2, 25 mM NaCl, 0.5 mM DTT, 5% glycerol, 0.25 µg of bovine serum albumin µl-1) containing 0.5 µg of poly(dI-dC) · poly(dI-dC) in a final volume of 20 µl. For gel shift competition, protein was allowed to bind with 100-fold excess by weight (100- to 194-fold molar excess), unless otherwise specified, of unlabeled competitor DNA for 15 min at room temperature before addition of the probe. Incubation with the labeled probe lasted for 20 min. The mixture was then loaded on an 8% polyacrylamide gel (30:1 acrylamide-bisacrylamide in Tris-glycine buffer containing 50 mM Tris, 380 mM glycine, and 2 mM EDTA [pH 8.5]) (45) which had been prerun for 1 h at 4°C in Tris-glycine buffer at 25 mA. Electrophoresis was performed at 4°C in Tris-glycine buffer for 2 to 2.5 h at 25 mA. The gel was then electroblotted onto Hybond-N+ nylon membranes (Amersham) at 200 mA in Tris-glycine buffer at 4°C for 1 h, the membrane was baked at 80°C for 2 h, and chemiluminescence was detected according to the instructions of Roche.
Insertional mutagenesis. Copies of the four abp genes, truncated at both ends (the truncation denoted with a prime sign), were prepared by PCR with the primers shown in Table 2 for abp1', abp2', abp3', and abp4' and were treated with T4 DNA polymerase to blunt their ends. The resulting PCR copies were cloned between the SmaI sites of the polylinker of pRL498 (12) and then transferred by conjugation into wild-type cells of Anabaena sp. strain PCC 7120. Insertional mutants were selected on AA plus nitrate agar in the presence of 25 or 50 µg of neomycin ml-1.
Luciferase assays. Luciferase activity of cyanobacterial cell suspensions was measured with a model TD-20/20 luminometer (Turner Designs, Sunnyvale, Calif.) (13) and was normalized to the concentration of chlorophyll a in the sample, which was measured in methanolic extracts (35).
RNA isolation and Northern blot analysis.
Total RNA was isolated from wild-type and mutant cells of Anabaena sp. strain PCC 7120 by using TRIzol reagent (Life Technologies Division, Invitrogen Corp., Carlsbad, Calif.) according to the manufacturer's instructions. RNA samples were denatured and run on a 1.3% denaturing formaldehyde gel (30 µg/lane) in MOPS buffer (40 mM morpholinepropanesulfonic acid [pH 7.0], 10 mM Na acetate, 1 mM EDTA) and blotted onto Hybond-N+ nylon membranes (Amersham) according to standard procedures (41). Concentrations of RNA were measured spectrophotometrically at 260 nm before loading and were validated by the intensity of the ethidium bromide fluorescence from the rRNA bands observed under UV light before and after running the gels. RNA was fixed to the nylon membrane by baking at 80°C for 2 h. The blots were hybridized with
-32P-labeled DNA probes prepared by random primer labeling and purified on Micro Bio-Spin P-30 Tris chromatography columns (Bio-Rad). The hybridization solution contained 7% SDS, 50% formamide, 5x SSC (1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate), 2% blocking reagent (Roche), 50 mM Na phosphate buffer (pH 7.0), and 0.1% sodium N-lauroylsarcosinate (Sigma Chemical Co., St. Louis, Mo.). Hybridization was performed overnight at 42°C. Blots were washed twice for 5 min with 2x SSC-0.1% SDS at room temperature, twice for 10 min with 1x SSC-0.1% SDS at 57°C, and once for 10 min with 0.2x SSC-0.1% SDS at 57°C. Bands were quantified with a Molecular Dynamics (Amersham) PhosphorImager and Image Quant software. The radioactivity was plotted as counts per minute per square millimeter after background correction.
Probes for Northern blot analysis. DNA probes for hepA, hepC, and rnpB were prepared by PCR with the corresponding primers shown in Table 2.
Thin-layer chromatography of lipid extracts. Aliquots of methanol-chloroform (1:2, vol/vol) extracts of wild-type and mutant filaments, containing equal amounts of chlorophyll, were chromatographed on thin-layer chromatography plates precoated with silica gel (EM Reagents, E. Merck, Darmstadt, Germany) with a solvent system of chloroform-methanol-acetic acid-water (170:30:20:7.4, vol/vol) (38).
|
|
|---|
|
View larger version (17K): [in a new window] |
FIG. 1. Schematic representation of the region from bp 3452256 (on the left) to bp 3456061 (on the right) of the chromosome of Anabaena sp. strain PCC 7120 showing genes hepC and hepA and the origins of the DNA fragments used in this study. Numbers represent base pairs relative to the first ATG codon of hepC (for fragments X1, X2, X11, and X12) and relative to the transcriptional start site of hepA (for A1 and B2). DNA fragment X12 was used for protein purification.
|
![]() View larger version (148K): [in a new window] |
FIG. 2. Mobility shift assay. Protein extract partially purified from Anabaena sp. strain PCC 7120 by precipitation between 50%- and 70%-saturated (NH4)2SO4 binds specifically to sequence X12. Labeled X12 was used as the probe with the following additions. Lanes: 1, no added protein; 2, protein extract; 3, protein extract plus 100-fold nanogram excess of unlabeled X12; 4, protein extract plus 400-fold nanogram excess of unlabeled X12; 5, protein extract plus 100-fold nanogram excess of unlabeled B2; 6, protein extract plus 100-fold nanogram excess of unlabeled X1; 7, protein extract plus 100-fold nanogram excess of unlabeled A1; 8, protein extract plus 100-fold nanogram excess of unlabeled X2. The shifted bands are near the top of the photographic image, which lies beneath the loading wells of the gel.
|
![]() View larger version (36K): [in a new window] |
FIG. 3. Comparison of the sequences of fragments X12, B2, and Y, aligned with Multalin (http://www.toulouse.inra.fr). Proteins Abp1, Abp2, Abp3, and Abp4 bind specifically to each of these fragments.
|
![]() View larger version (44K): [in a new window] |
FIG. 4. SDS-PAGE analysis, on a 10% polyacrylamide gel, of proteins eluted from specific-DNA affinity chromatography. The gel was stained with silver. Lane 1, protein purified by using the oligomerized X12 fragment of DNA coupled to CNBr-activated Sepharose; lane 2, molecular masses of protein standards in kilodaltons.
|
|
View this table: [in a new window] |
TABLE 3. Characteristics of polypeptides obtained from the Harvard University Microchemistry Facility
|
|
View this table: [in a new window] |
TABLE 4. Characteristics of four Anabaena sp. proteins
|
Single recombination of pRL1069a with Anabaena sp., producing derivative SR1069a, placed Vibrio fischeri luxAB under the control of the hepA promoter while retaining the ability of the strain to fix N2 aerobically. Northern blot hybridization experiments showed that expression of hepC and hepA is induced after 10 h of nitrogen deprivation of cells of wild-type Anabaena sp. (Fig. 5A and B) or of strain SR1069a (Fig. 6A) but not of SR1069a derivatives CPB8226 (mutated in abp2) and CPB8563 (mutated in abp3) (Fig. 5A and B and 6B and C). Increasing activity of luciferase was observed in cells of strain SR1069a but not in the abp2 and abp3 mutant derivatives of strain SR1069a (Fig. 6C). It is clear from growth experiments (data not shown) that genes abp2 and abp3 are unimportant during growth on nitrate but required for growth on N2. Surprisingly, electron microscopy has shown that abp2 and abp3 mutants ABP2 and ABP3, despite their weak expression of hepA and hepC, form an envelope polysaccharide layer but neither a laminated envelope layer of glycolipids (Fig. 7) nor heterocyst envelope glycolipids, as shown by thin-layer chromatography of lipid extracts (Fig. 8). Evidently, even very weak expression of hepA and hepC suffices for the production of heterocyst envelope polysaccharide. In the abp1 and abp4 mutants ABP1 and ABP4, which grow with or without nitrate in the medium, hepA and hepC are expressed but are induced only after 24 h of nitrogen deprivation, much later than they are in the wild-type strain (data not shown). Although not essential for heightened expression of hepC and hepA, Abp1 and Abp4 evidently play a minor role in the regulation of transcription of those two genes.
![]() View larger version (23K): [in a new window] |
FIG. 5. Expression of hepC (A) and hepA (B) derived by quantitation of Northern blots of mutants CPB8226 and CPB8563 in Fig. 6 and of wild-type Anabaena sp. and luciferase activity (ATP photometer units per microgram of chlorophyll a) of the hepA::luxAB transcriptional fusions in strain SR1069a and in its derivatives, CPB8226 and CPB8563 (C).
|
![]() View larger version (56K): [in a new window] |
FIG. 6. Transcripts of hepC and hepA in derivatives SR1069a, CPB8226, and CPB8563 of Anabaena sp. strain PCC 7120. Northern hybridization with the hepC and hepA probes (see Materials and Methods) shows induction of expression of corresponding genes in 10 h in response to nitrogen deprivation of cells of derivative SR1069a of Anabaena sp. (the same result was obtained for cells of wild-type Anabaena sp. [see Fig. 6A and B]). No induction of expression was seen in the abp2 (CPB8226) and abp3 (CPB8563) derivatives of strain SR1069a. An rnpB (44) probe was used as a loading control. Lanes: 1, cells grown with nitrate, unwashed; 2, cells washed; 3 through 8, washed cells deprived of fixed nitrogen for 3, 6, 10, 24, 30, and 48 h, respectively.
|
![]() View larger version (114K): [in a new window] |
FIG. 7. abp2 and abp3 mutants form a heterocyst envelope polysaccharide layer but no glycolipid layer. (B [a magnification of the boxed area in panel A]) In a heterocyst of wild-type Anabaena sp., the laminated layer of glycolipids (47) (the laminations vertically arrayed in panel B) is enveloped by a layer of polysaccharide. In contrast, the only envelope layer seen in a heterocyst of an abp3 mutant (D [a magnification of the boxed area in panel C]) or an abp2 mutant (F [a magnification of the boxed area in panel E] and G) is the polysaccharide layer. H, heterocyst; V, vegetative cell; GL, glycolipid layer; PS, polysaccharide layer.
|
![]() View larger version (120K): [in a new window] |
FIG. 8. Thin-layer chromatography of lipid extracts shows that heterocyst envelope glycolipids are absent from abp2 and abp3 mutants. A lipid extract of wild-type Anabaena sp. (lanes 1 and 2) and its derivative strains ABP2 (lanes 3 and 4), ABP3 (lanes 5 and 6), SR1069a (lanes 11 and 12), CPB8226 (lanes 9 and 10), and CPB8563 (lanes 7 and 8) were chromatographed. The cyanobacteria had been grown in the presence (odd-numbered lanes) or in the absence (even-numbered lanes) of nitrate in the medium. The arrow marks heterocyst-specific glycolipids.
|
![]() View larger version (46K): [in a new window] |
FIG. 9. Mobility shift assay with Abp proteins that were overexpressed in E. coli, showing specificity for sequence X12. Labeled X12 was used as the probe, with the following additions: (A) Lanes: 1, no added protein; 2, Abp2 protein extract; 3, Abp2 protein extract plus 100-fold nanogram excess of unlabeled X12; 4, Abp2 protein extract plus 100-fold nanogram excess of unlabeled v6 (see Materials and Methods); 5, Abp3 protein extract; 6, Abp3 protein extract plus 100-fold nanogram excess of unlabeled X12; 7, Abp3 protein extract plus 100-fold nanogram excess of unlabeled v6. (B) Lanes: 1, Abp1 protein extract; 2, Abp1 protein extract plus 100-fold nanogram excess of unlabeled X12; 3, Abp1 protein extract plus 100-fold nanogram excess of unlabeled v6; 4, Abp4 protein; 5, Abp4 protein plus 100-fold nanogram excess of unlabeled X12; 6, Abp4 protein plus 100-fold nanogram excess of unlabeled v6; 7, no added protein. The arrows point to labeled DNA-protein complexes whose formation was reduced after addition of unlabeled DNA fragment X12.
|
Abp2 shares similarity (E = 7 x 10-10) with the mitochondrial processing peptidase-like protein, Mpp, of Bradyrhizobium japonicum, and it was recently shown that a B. subtilis homolog, MlpA, of mitochondrial processing peptidases may regulate the transcription and/or secretion of a group of proteins (4). The authors discuss the possibility that MlpA binds upstream from and represses one of the regulated sequences but consider this possibility unlikely because MlpA lacks known DNA-binding motifs. From our results, this possibility may warrant further consideration. Alternatively, the authors suggest, MlpA may modulate the activity of a transcriptional regulator. It is tempting to speculate that Abp1 and Abp2, which are encoded by neighboring genes, bind DNA, and resemble proteases, act together.
More than 300 different species of eubacteria and archeabacteria possess a regularly arranged structure, named the S (surface)-layer, which covers the cell envelope. S-layers are paracrystalline monolayered assemblages of proteins, some of which are glycosylated or phosphorylated (42). Several S-layer proteins and some other cell wall proteins contain one or more copies of a domain of about 50 to 60 residues, known as an S-layer homology domain, which may anchor the protein to the peptidoglycan (32, 34). Alternatively, O-polysaccharide side chains of the lipopolysaccharide component of the cell walls of gram-negative bacteria may serve as receptors for the attachment of the S-layer (5). Abp3, which contains three S-layer homology domains (amino acids 360 through 402, 420 through 462, and 482 through 526) (ScanProsite website), showed similarity (BLAST score, 105; E = 10-21) to an S-layer-associated multidomain endoglucanase of an anaerobic bacterium, Thermoanaerobacterium polysaccharolyticum (Table 4). Unlike Abp1, Abp2, and Abp4, whose indicated similarities extend throughout the proteins, only about one-third of Abp3 showed similarity to less than 20% of that endoglucanase. Interestingly, the S-layer-like protein, SlpM, of Thermus thermophilus HB8 is one of three proteins that control the expression of slpA, the gene that encodes the protein that comprises the crystalline surface layer (15). Insertional inactivation of slpM demonstrated its in vivo function in the control of slpA transcription: SlpM acts as an activator. Notably, SlpA itself acts as a specific repressor of its own translation by binding directly to the 5'-untranslated region of its own transcript.
Abp4 is most similar (BLAST score, 340; E = 2 x 10-92) to a hemolysin secretion-like protein from Synechocystis sp. strain PCC 6803 (Table 4) and to related ATP-binding-cassette transporters (E = 8 x 10-22 and 6 x 10-20). Abp4 is encoded by the central ORF of three closely spaced ORFs (http://www.kazusa.or.ip/cyano/Anabaena) that may constitute an operon. The predicted products of the other two ORFs are similar, respectively, to a Synechocystis sp. protein that resembles a hemolysin secretion protein (BLAST score, 1,021; E < 10-180) and to the protein predicted by ORF slr0655 of Synechocystis sp. (BLAST score, 196; E = 10-49).
Protein Abp4 contains three leucine (bold L) zipper motifs: LEFVRSRLPQEATFLLRSRAAL, corresponding to amino acids 167 through 188, LPQEATFLLRSRAALVAENELL, corresponding to amino acids 174 through 195, and LTQTTVKLQDTQSSLAIQQRIL, corresponding to amino acids 241 through 262. Leucine zipper motifs are present in many proteins with diverse functions and are characteristic of some eukaryotic DNA-binding, gene regulatory proteins (8, 31). The leucine side chains facilitate dimerization of proteins (39) and appear to be involved in protein-protein interactions.
In conclusion, we have identified four proteins that bind sequence specifically to DNA of Anabaena sp. strain PCC 7120, even though none has a known DNA-binding domain and all bear extensive similarity to proteins that have other known roles. Mutants lacking two of these proteins show developmental defects and greatly reduce expression of hepC and hepA, suggesting that these proteins indeed function as transcriptional regulators. Our data add new examples to other examples that have been reported (2, 11, 30, 40) of proteins that lack a known DNA-binding domain but nonetheless bind sequence specifically to DNA and regulate transcription.
Permanent address: N. Vavilov Institute of General Genetics, Moscow 119991, Russia. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»