Ferritin from the obligate anaerobe Porphyromonas gingivalis: purification, gene cloning and mutant studies

Dinath B. Ratnayake1, Sun Nyunt Wai2, Yixin Shi1, Kazunobu Amako2, Hiroaki Nakayama1 and Koji Nakayama1

Department of Microbiology, Faculty of Dentistry1 and Department of Bacteriology, Faculty of Medicine2, Kyushu University, Fukuoka 812-8582, Japan

Author for correspondence: Koji Nakayama. Tel: +81 92 642 6332. Fax: +81 92 642 6263. e-mail: knak{at}dent.kyushu-u.ac.jp


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Porphyromonas gingivalis is an obligate anaerobe that utilizes haem, transferrin and haemoglobin efficiently as sources of iron for growth, and has the ability to store haem on its cell surface, resulting in black pigmentation of colonies on blood agar plates. However, little is known about intracellular iron storage in this organism. Ferritin is one of the intracellular iron-storage proteins and may also contribute to the protection of organisms against oxidative stresses generated by intracellular free iron. A ferritin-like protein was purified from P. gingivalis and the encoding gene (ftn) was cloned from chromosomal DNA using information on its amino-terminal amino acid sequence. Comparison of the amino acid sequence deduced from the nucleotide sequence of ftn with those of known ferritins and bacterioferritins identified the protein as a ferritin and positioned it between proteins from the Proteobacteria and Thermotogales. The P. gingivalis ferritin was found to contain non-haem iron, thus confirming its identity. Construction and characterization of a P. gingivalis ferritin-deficient mutant revealed that the ferritin was particularly important for the bacterium to survive under iron-depleted conditions (both haemin and transferrin starvation), indicating that intracellular iron is stored in ferritin regardless of the iron source and that the iron stored in ferritin is utilized under iron-restricted conditions. However, the ferritin appeared not to contribute to protection against oxidative stresses caused by peroxides and atmospheric oxygen.

Keywords: Porphyromonas gingivalis, ferritin, iron storage, oxidative stress

Abbreviations: Em, erythromycin; TMBZ, 3,3',5,5'-tetramethylbenzidine

The GenBank accession number for the sequence reported in this paper is AB016086.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Iron is an essential growth requirement for almost all living organisms including pathogenic bacteria. There is an abundance of iron in the human body: however, the availability of iron to infecting bacteria is strictly limited. Almost all iron in the human body is sequestered by iron-binding proteins, and the availability of such proteins varies with place and time. Therefore, it is critically important for infecting bacteria to have the ability to store iron intracellularly while in an iron-rich environment and utilize the stored iron under iron-depleted conditions. Moreover, in the presence of oxygen, intracellular free iron poses the problem of cytotoxicity because it reacts with reduced forms of oxygen, leading to oxidative damage of nucleic acids, lipids and proteins (Halliwell & Gutteridge, 1992 ; Miller & Britigan, 1997 ).

Micro-organisms have developed two types of iron-storing proteins: ferritins and bacterioferritins (Andrews, 1998 ). The former contain iron, whereas the latter contain haem. Amongst prokaryotes, ferritins have been isolated from Bacteroides fragilis (Rocha et al., 1992 ), Escherichia coli (Hudson et al., 1993 ), Helicobacter pylori (Frazier et al., 1993 ) and Campylobacter jejuni (Wai et al., 1995 , 1996 ). In addition, ferritin-encoding genes have been found in the genomes of Haemophilus influenzae, Methanobacterium thermoautotrophicum, Clostridium acetobutylicum, Thermotoga maritima, Archaeoglobus fulgidus, Mycobacterium tuberculosis and Vibrio cholerae (Andrews, 1998 ). However, the complete genome sequences of Bacillus subtilis, Methanococcus jannaschii, Mycoplasma pneumoniae and Synechocystis PCC6803 do not contain a ferritin-encoding gene. This indicates that the iron-storage system varies amongst bacterial species. Evidence to demonstrate the contribution of ferritin to protection against oxidative stress has also been provided by analysis of a ferritin-deficient mutant of Cam. jejuni (Wai et al., 1996 ). In E. coli, however, a mutation in the ferritin-encoding gene (ftnA) confers no sensitivity to oxidative stress on the cell (Abdul-Tehrani et al., 1999 ).

Porphyromonas gingivalis is a Gram-negative anaerobic bacterium belonging to the division Cytophagales (Hugenholtz et al., 1998 ; Woese, 1987 ). This bacterium is frequently isolated from periodontal pockets of patients with advanced adult periodontitis and is recognized as a major pathogen in the disease. P. gingivalis is a poor utilizer of free iron but can utilize haem, haemoglobin and transferrin efficiently as iron sources, and these are available in periodontal regions. It has the ability to bind haem, haemoglobin (Nakayama et al., 1998 ) and transferrin (Tazaki et al., 1995 ), and a haemoglobin-receptor protein on the cell surface has been biochemically and genetically characterized (Nakayama et al., 1998 ). One characteristic and interesting property of P. gingivalis is that the organism produces black-pigmented colonies when grown on haemolysed blood agar plates. The black pigment identified as the µ-oxo dimeric form of iron–porphyrin IX is accumulated on the cell surface (Smalley et al., 1998 ), and may serve as a source of iron for utilization under iron-depleted conditions, although intracellular iron storage has not been characterized in this organism.

The aim of this study was to clarify the mechanism for intracellular iron storage in P. gingivalis. After isolating ferritin from P. gingivalis cells, the encoding gene (ftn) was cloned and sequenced. We also constructed and characterized an ftn-disrupted mutant of this organism.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and plasmids.
P. gingivalis ATCC 33277, E. coli DH5{alpha} and E. coli BL21(DE3) were used in this study. Chromosomal DNA of P. gingivalis ATCC 33277 was used as the source of DNA for cloning the ferritin-encoding gene from P. gingivalis. Plasmids pUC19 and pET11a were used as vectors for gene cloning and for the overproduction of ferritin, respectively. Preparation of the competent cells was carried out according to Sambrook et al. (1989) . Plasmids pKD381, pKD382, pKD383 and pKD384 were constructed in this study, and are described below.

Media and growth conditions.
Unless otherwise specified, P. gingivalis was grown in an anaerobic atmosphere (10% CO2, 10% H2, 80% N2) at 37 °C. Enriched BHI broth [containing 37 g brain–heart infusion (Difco), 5 g yeast extract (Difco), 1 g cysteine, 5 mg haemin and 1 mg vitamin K1 per l], enriched TS agar [containing 40 g Trypto-Soya agar (Nissui), 5 g brain–heart infusion, 1 g cysteine, 5 mg haemin and 1 mg vitamin K1 per l] and blood agar prepared by adding 5% haemolysed defibrinated sheep blood to enriched TS agar were used for P. gingivalis. E. coli was grown at 37 °C in L broth or on L agar (L broth solidified with 1·5% agar). Erythromycin (Em; 10 µg ml-1), tetracycline (0·5 µg ml-1) and ampicillin (50 µg ml-1) were added as required for selection and maintenance of the strains.

Chemicals.
The proteinase inhibitors N{alpha}-p-tosyl-L-lysine chloromethyl ketone and leupeptin were obtained from Sigma and Peptide Institute (Osaka), respectively.

Purification of ferritin.
P. gingivalis ATCC 33277 was grown in 1 l enriched BHI broth for 48 h. The cells were harvested by centrifugation at 15000 g for 20 min at 4 °C. The pellet was resuspended in 30 ml buffer A (10 mM Tris buffer, pH 7·5), and N{alpha}-p-tosyl-L-lysine chloromethyl ketone and leupeptin were added to final concentrations of 0·1 mM and 1 mM, respectively. The cells were broken by sonic vibration (25 W; 30 pulses min-1; 1 s pulse length) in a Branson sonicator at 1 min intervals for 10 min on ice as described by Wai et al. (1995) . The sonicate was shaken for 15 min at 37 °C and centrifuged to remove the unbroken cells. The supernatant was saved, to which CsCl was added to a final concentration of 40% (w/v), and centrifuged at 80000 g for 24 h at 4 °C. The solution was carefully separated into fractions, and each fraction was dialysed overnight against buffer A by changing the buffer at 6 h intervals. The fractions were then examined by electron microscopy (see below). The fractions containing ferritin-like particles were concentrated by ultrafiltration with a Microcon YM-10 (Millipore). For further purification, the concentrated ferritin sample was subjected to gel filtration on a Sephadex FPLC column (FPLC column TM 200; Pharmacia). The system was equilibrated with buffer A, proteins were eluted at a flow rate of 0·5 ml min-1 and 2 ml fractions were collected. Elution of the proteins was monitored by measuring absorbance at 280 nm. The fractions were also examined by electron microscopy and SDS-PAGE to examine for the presence of ferritin.

Electron microscopy.
Samples were negatively stained with a 0·5% uranyl acetate solution and examined in a JEM 2000EX electron microscope (JEOL) at 100 kV.

Protein analysis.
SDS-PAGE was performed under reducing conditions on 12·5% gels essentially according to Laemmli (1970) , using a low-molecular-mass electrophoresis calibration kit (Pharmacia) as reference protein markers. The gels were stained with 0·1% Coomassie blue R-250 or the Sil-Best staining kit (Nacalai), according to the manufacturer’s instructions. For immunoblot analysis, proteins on SDS-PAGE gels were electrophoretically transferred to nitrocellulose membranes according to the method of Towbin et al. (1979) . The membranes were immunostained with a 2000-fold dilution of an anti-ferritin antiserum. The antiserum was prepared from a rabbit immunized with ferritin purified from cells of an ftn-overexpressing E. coli strain. The reacting proteins were detected using the ECL Western blotting system (Amersham).

DNA purification, colony hybridization and Southern blot analysis.
Chromosomal DNA was isolated from P. gingivalis cells by the guanidine isothiocyanate method (Lippke et al., 1987 ) with the Isoquick DNA extraction kit (MicroProbe). Plasmid DNA was purified from E. coli cells using the Wizard DNA purification system (Promega). Colony hybridization and Southern blot hybridization were performed essentially according to Wallace & Miyada (1987) , and Sambrook et al. (1989) , respectively.

Plasmid construction.
A 5·5 kb BamHI–HindIII fragment containing the ferritin gene (ftn) was isolated from the chromosome of P. gingivalis ATCC 33277 and introduced into the BamHI/HindIII sites of pUC19 to construct plasmid pKD381. This plasmid was digested with HincII and HindIII, the resulting 4·4 kb linear plasmid DNA was subjected to filling-in with Klenow and was self-ligated, resulting in pKD382. The ermF ermAM DNA block (end-filled EcoRI–BamHI fragment) of pVA2198 (Fletcher et al., 1995 ) was inserted into the filled-in XhoI site within the ftn gene of pKD382, to generate pKD383. The ferritin-encoding DNA region of pKD381 was amplified by PCR (upper primer, 5'-CCATATGAAAATAAGCGAAAACGTA-3'; lower primer, 5'-AGGATCCCTTCATTTGTCCATTCGG-3'). The PCR product was then subcloned into the pET11a vector using NdeI and BamHI sites, which had been introduced at both ends of the ftn gene by oligonucleotide primers with base changes necessary to produce the restriction sites, resulting in pKD384.

Electrotransformation.
An overnight culture of P. gingivalis ATCC 33277 in enriched BHI broth (1 ml) was mixed with 20 ml enriched BHI broth and grown anaerobically for 5 h at 37 °C. The cells were harvested by centrifugation at 4000 g and 4 °C for 15 min, gently washed with 10 ml 0·3 M sucrose, and resuspended in 2 ml of the same solution. Ten microlitres of plasmid solution (10 µg ml-1) was mixed with 400 µl cell suspension and the mixture placed in a cuvette for electroporation (Pulser cuvette with a 0·2 cm electrode gap; Bio-Rad). Electroporation was carried out at a voltage of 2·0 kV at a time constant of 6·1 ms with a Gene pulser (Bio-Rad). The cells were immediately transferred to 4 ml prewarmed enriched BHI broth and incubated anaerobically overnight at 37 °C. The culture (0·2 ml aliquots) was then spread on enriched TS plates containing Em and incubated anaerobically for 7 d at 37 °C. Em-resistant colonies were isolated from these plates.

Determination of amino-terminal amino acid sequence and nucleotide sequence.
To determine the amino-terminal amino acid sequence of ferritin, the protein on the polyacrylamide gel was electrophoretically transferred onto a PVDF membrane and stained with Coomassie brilliant blue R-250. The stained protein band was cut out and analysed in an automatic protein sequencer (Applied Biosystems model 476A; Perkin Elmer). DNA sequencing was achieved using plasmid templates and a dideoxy sequencing kit (AutoRead sequencing kit; Pharmacia) with ALF DNA sequencer II (Pharmacia). The sequence data were analysed with the GeneWorks software program (IntelliGenetics).

Non-denaturing PAGE and staining methods for non-haem iron and haem.
Vertical non-denaturing PAGE (5% acrylamide at 5 V cm-1) was performed according to Ausubel et al. (1995) . The gels were stained for iron as described by Chung (1985) with a solution that contained 0·75 mM Ferene S (Sigma) and 15 mM thioglycolic acid in 2% (v/v) acetic acid. Haem staining was achieved using hydrogen peroxide and 3,3',5,5'-tetramethylbenzidine (TMBZ; Dojindo) according to Thomas et al. (1976) .

Iron starvation.
To determine the ability to grow under iron-starvation conditions, cells of test strains were first grown in the presence of haemin or transferrin and then deprived of the iron source. In each case, the initial inocula were prepared by growing the strains in haemin-containing enriched BHI broth overnight. In the case of haemin deprivation, the inocula were diluted 10-fold with haemin-free enriched BHI broth and incubated. Every 24 h, the OD540 of the cultures was measured, and 10-fold dilution of the cultures with haemin-free enriched BHI broth was repeated. In the case of transferrin deprivation, the inocula were diluted 10-fold with enriched BHI broth containing transferrin (5 µM) and incubated for 24 h. Ten-fold dilution followed by 24 h incubation was repeated twice in the same medium. Finally, the cultures were diluted 10-fold into transferrin-free enriched BHI broth and subsequent dilutions were the same as those for haemin starvation.

Assays for oxidative-stress sensitivity.
The sensitivity of strains to hydrogen peroxide and cumene hydroperoxide was determined by diluting 10-fold an overnight culture of the test strain in enriched BHI broth with the same medium containing various concentrations of either chemical and measuring the OD540 using a spectrophotometer (150-20; Hitachi) before and after an overnight anaerobic incubation. Sensitivity to the chemicals was also determined by agar-diffusion assays. Test strains were grown in enriched BHI broth for 18 h and a 500 µl portion from each culture was spread on an enriched TS plate, which was then dried at 32 °C for 45 min. Hydrogen peroxide (30%; Wako) was diluted twofold in distilled water, and cumene hydroperoxide (~80%; Sigma) was diluted 10-fold with DMSO. Filter paper discs (5 mm diameter; Whatman) were blotted with 5 µl portions of these solutions and the discs were placed at the centres of the plates. The plates were anaerobically incubated at 37 °C for 48 h, and the diameters of growth-inhibition zones around the discs were measured. To determine aerotolerance of strains, overnight cultures in enriched BHI broth were diluted twofold with the same medium and exposed to atmospheric oxygen by shaking at 125 cycles per min. Samples were taken at intervals and plated on enriched TS plates, which were incubated anaerobically for 7 d.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Partial purification of ferritin-like particles from P. gingivalis
Ferritin-like particles were partially purified from extracts of P. gingivalis ATCC 33277 cells by CsCl equilibrium centrifugation and gel filtration. Each fraction of the gel filtration was subjected to electron microscopy to examine for the presence of ferritin-like particles (Fig. 1a). All of the fractions in which the ferritin-like particles, exhibiting a spherical structure with an inner electron-dense core, were observed contained a protein with a molecular mass of 18 kDa, which was the expected molecular mass of ferritin (Frazier et al., 1993 ; Hudson et al., 1993 ; Wai et al., 1995 ). Determination of the amino-terminal amino acid sequence of the protein revealed the sequence MKISENVTKAINDQIKAEM, which is 58% identical to the amino-terminal sequence of Bact. fragilis ferritin (Rocha et al., 1992 ).



View larger version (79K):
[in this window]
[in a new window]
 
Fig. 1. Electron micrographs of P. gingivalis ferritin. The ferritin particles were purified from cell extracts of P. gingivalis (a) and E. coli overexpressing the P. gingivalis ftn gene (b), and are negatively stained with uranyl acetate. Bars, 50 nm.

 
Molecular cloning and sequencing of a gene encoding the 18 kDa protein of P. gingivalis
P. gingivalis ATCC 33277 chromosomal DNA was digested with various restriction enzymes and subjected to Southern blot hybridization with a degenerate DNA probe based on the amino-terminal amino acid sequence of the 18 kDa protein. A BamHI–HindIII fragment of ~5·5 kb was found to hybridize with this probe. BamHI–HindIII fragments of ~5·5 kb were purified from the gel and ligated to pUC19 double-digested with BamHI/HindIII. A plasmid, pKD381, which contained the 5·5 kb BamHI–HindIII fragment identified by hybridization with the probe, was obtained by colony hybridization.

The 1·3 kb BamHI–BglI region of the P. gingivalis chromosomal DNA from pKD381 was then sequenced (Fig. 2). One ORF encoding a protein with a calculated molecular mass of 18507 Da and the same predicted amino-terminal amino acid sequence as the 18 kDa protein was revealed, indicating that it encoded the expected 18 kDa protein. The deduced amino acid sequence of the protein shared similarities with those of ferritins from E. coli, Hel. pylori and Cam. jejuni (Fig. 3). Several conserved amino acid residues of the eukaryotic ferritin H chain, which participate in iron chelation, were also found at the corresponding positions (Fig. 3). Since this similarity strongly suggested that the 18 kDa protein was P. gingivalis ferritin, we tentatively designated the ORF as the ftn gene of P. gingivalis. Comparison of the predicted amino acid sequence with those of ferritins and bacterioferritins, including putative ones from genomic sequences, revealed that the ftn gene product belonged to the ferritin group rather than the bacterioferritin group and showed higher similarities to proteins from T. maritima and members of the Archaea than with those from members of the Proteobacteria (Fig. 4).



View larger version (10K):
[in this window]
[in a new window]
 
Fig. 2. Restriction map of P. gingivalis ATCC 33277 chromosomal DNA in the vicinity of the ftn gene. Top: the 5·5 kb BamHI–HindIII chromosomal DNA region of pKD381. Bottom: the 1·3 kb sequenced region showing the insertion of the Em cartridge (pKD383).

 


View larger version (51K):
[in this window]
[in a new window]
 
Fig. 3. Alignment of the amino acid sequence of P. gingivalis ferritin with those of the ferritins previously isolated and determined in other prokaryotes (Wai et al., 1996 ; Frazier et al., 1993 ; Izuhara et al., 1991 ; Rocha et al., 1992 ). Amino acid residues which correspond to the residues of the eukaryotic ferritin H participating in iron chelation are indicated by asterisks.

 


View larger version (13K):
[in this window]
[in a new window]
 
Fig. 4. Phylogenetic relationships of ferritins and bacterioferritins. The dendrogram was constructed using the UPGMA clustering method (Nei, 1987 ) with GeneWorks software. The database accession numbers or websites for the sequences of ferritin- and bacterioferritin-encoding genes are as follows. Ferritins: Ftn (A. fulgidus) and Ftn (T. maritima), http://www.ncbi.nlm.nih.gov/cgi-bin/blast/nph-tigrbl; Ftn (Met. thermoautotrophicum), http://www.cric.com/htdocs/sequences; Ftn 1/Ftn 2 (Hae. influenzae), P43707/P43708; Ftn/YecI (E. coli), P23887/D90832; Cft (Cam. jejuni), D64082; Pfr (Hel. pylori), P52093. Bacterioferritins: Bfr (E. coli), P11056; BfrA/BfrB (Neisseria gonorrhoeae), http://dna1.chem.uoknor.edu; Bfr (Brucella melitensis), P49944; Bfr (Rhodobacter capsulatus), Q59738; Bfr (Pseudomonas putida), U66717. See http://www.ncbi.nlm.nih.gov/BLAST/unfinishedgenome.html for sequences without accession numbers.

 
To confirm that the ftn gene product was ferritin, we constructed an ftn-overexpressing E. coli strain using the T7 RNA polymerase expression system. The E. coli strain thus obtained produced large amounts of ferritin-like particles showing the same structure as the ferritin purified from P. gingivalis cells (Fig. 1b).

Presence of non-haem iron within the recombinant ferritin particle
Ferene S stains iron in ferritin, transferrin and lactoferrin, but hardly stains that in haem (Chung, 1985 ). We used this to determine whether P. gingivalis ferritin contains non-haem iron or haem. The cell extract of the ftn-overexpressing E. coli, transferrin, BSA and haemin were electrophoresed through non-denaturing gels, and stained with Ferene S and TMBZ (Fig. 5). Ferene S stained both transferrin and the ftn gene product, but not BSA and haemin. As expected, TMBZ stained only haemin. None of the other proteins was stained with TMBZ. These results showed that P. gingivalis ferritin contained non-haem iron within the shell, indicating that the ferritin has the ability to store iron.



View larger version (54K):
[in this window]
[in a new window]
 
Fig. 5. Non-denaturing PAGE profiles. Gels were stained with Coomassie brilliant blue R250 (a), Ferene S (b) and TMBZ (c). Lanes: 1, human transferrin; 2, cell extracts of P. gingivalis ftn-overexpressing E. coli; 3, haemin; 4, BSA.

 
Construction of an ftn mutant of P. gingivalis
To gain insight into the biological significance of ferritin in P. gingivalis, a ferritin-deficient mutant was constructed. SmaI-linearized pKD383 DNA, which contained the ftn gene disrupted by the Em cartridge (see Methods), was introduced into P. gingivalis ATCC 33277 cells by electroporation. A number of Em-resistant colonies were obtained and one such transformant (KDP139) was chosen for further characterization. To confirm the proper construction of the transformant, chromosomal DNA of the transformant and the wild-type parent was subjected to Southern blot analysis after double digestion with BamHI and HindIII (Fig. 6a). The expected hybridizing fragments of 1·8 kb for the mutant and 5·5 kb for the wild-type parent in the digests were obtained, indicating that the disrupted ftn gene had been substituted for the wild-type ftn gene in the Em-resistant transformant. To confirm the absence of ferritin in the cells of the ftn mutant, immunoblot analysis using anti-ferritin antiserum was performed (Fig. 6b). The ftn mutant showed no immunoreactive protein of 18 kDa. We also subjected cell extracts from the ftn mutant to fractionation by the procedure described above. Electron microscopy of gel filtration fractions that were expected to contain ferritin particles if it was present revealed none in the ftn mutant (data not shown). These results clearly showed that the ftn gene was responsible for the production of ferritin in P. gingivalis and that the ftn mutant was ferritin deficient.



View larger version (54K):
[in this window]
[in a new window]
 
Fig. 6. Proof of authenticity of the P. gingivalis ftn mutant KDP139. (a) Southern blot analysis of the chromosomal DNA. The chromosomal DNAs of the wild-type ATCC 33277 (lane 1) and KDP139 (lane 2) were digested with BamHI/HindIII. The resulting DNA fragments were subjected to agarose gel electrophoresis followed by blotting. Hybridization was performed using the 0·4 kb BamHI–XhoI fragment of pKD381 (Fig. 2) as a DNA probe. (b) Immunoblot analysis. After the cell extracts of ATCC 33277 (lane 1) and KDP139 (lane 2) were electrophoresed through an SDS-polyacrylamide gel, proteins were transferred to a nitrocellulose membrane and immunoreacted with antiserum against P. gingivalis ferritin.

 
Growth of the ftn mutant with haemin or transferrin as the iron source
To determine the effect of different iron sources on the growth of the ftn mutant, the mutant and the wild-type were grown in enriched BHI containing haemin or transferrin. There was no significant difference in growth between the two strains with either iron source. In transferrin-containing medium, however, cell lysis occurred earlier in the ftn mutant than in the wild-type parent (data not shown).

Effect of iron depletion on the growth of the ftn mutant
To determine the contribution of ferritin to intracellular iron storage, the mutant and the wild-type strains were iron-starved after growth in enriched BHI broth containing haemin or transferrin as the iron source (Fig. 7). Iron starvation resulted in gradual growth depression, which was more marked in transferrin-grown cells than in cultures of haemin-grown cells and occurred earlier in the mutant than in the wild-type.



View larger version (12K):
[in this window]
[in a new window]
 
Fig. 7. Iron starvation of the P. gingivalis strains following growth in enriched BHI broth containing transferrin or haemin as the iron source. {bullet}, Wild-type ATCC 33277 deprived of transferrin; {circ}, ftn mutant KDP139 deprived of transferrin; {blacktriangleup}, ATCC 33277 deprived of haemin; {triangleup}, KDP139 deprived of haemin. Data are given as means and standard error values of triplicate experiments.

 
Sensitivity of the ftn mutant to oxidative stresses
To compare the sensitivity to peroxides between the mutant and the wild-type, we used two different methods: measurement of MIC in liquid medium and agar-diffusion assays. Neither method revealed any difference in sensitivity against hydrogen peroxide and cumene hydroperoxide between the two strains, indicating that ferritin does not play a direct role in protecting the cell against these peroxides (data not shown). Furthermore, it was shown that the ftn mutant was tolerant of atmospheric oxygen to the same degree as the wild-type parent (data not shown). These results clearly indicated that ferritin of P. gingivalis had no direct role in protecting the cells from oxidative-stress damage.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In this study we purified a ferritin from P. gingivalis, cloned the ferritin-encoding gene (ftn) from the P. gingivalis chromosome and determined its nucleotide sequence. Comparison of the deduced amino acid sequence of this ferritin with those of other organisms placed it closer to those of A. fulgidus, Mec. thermoautotrophicum and T. maritima than to those of the Proteobacteria represented by E. coli, Hae. influenzae, Cam. jejuni and Hel. pylori, which was consistent with the notion that P. gingivalis belongs to the division Cytophagales and not to the Proteobacteria. Eukaryotic ferritin H chains have several conserved amino acid residues participating in iron chelation, i.e., Glu-27, Tyr-34, Glu-61, Glu-62, His-65, Glu-107 and Gln-141 (Andrews et al., 1991 ). All of these critical residues forming the ferroxidase centre are found in the corresponding sites of P. gingivalis ferritin. In addition, P. gingivalis ferritin had the same structure under electron microscopy as ferritins purified from other organisms. Analysis by staining indicated that P. gingivalis ferritin contained non-haem iron. Taken together, these findings indicate that the P. gingivalis ftn gene encodes an authentic ferritin and that the P. gingivalis ferritin may serve as an iron-storage protein in the same way as ferritins of other organisms.

Using the ftn mutant, we found that ferritin was important for P. gingivalis to survive under iron-restricted conditions. Thus, the mutant ceased growth under iron starvation more rapidly than the wild-type parent strain, which was consistent with previous studies using ferritin-deficient mutants of other micro-organisms (Wai et al., 1996 ; Abdul-Tehrani et al., 1999 ). The increased sensitivity of the ftn mutant to iron-source starvation was observed in both transferrin- and haemin-grown cells, although the growth decline started much earlier after transferrin depletion than after haemin depletion. That the ftn mutant and the wild-type grew equally well up to the fourth transfer under haemin-depleted conditions may be explained as follows. Cells of P. gingivalis store haem on the cell surface; this surface stock of haem will serve as the preferential iron source after haemin deprivation, resulting in the same growth rate in the wild-type and mutant strain. Although P. gingivalis may possess several iron uptake/storage systems which have developed for different iron sources such as haemin, haemoglobin or transferrin, it appears that intracellular iron is stored as ferritin regardless of iron source and the stored iron is utilized under iron-restricted conditions.

Free iron has the ability to form a highly reactive oxidant, the hydroxyl radical, through the Fenton reaction, which can cause damage to cellular components including DNA and membranes (Abdul-Tehrani, 1999 ; McCormick et al., 1998 ; Keyer et al., 1995 ; O’Connell et al., 1985 ). As a result of this detrimental effect of iron, regulated uptake and storage is important for the organism (Andrews, 1998 ; Wai et al., 1996 ; Guerinot, 1994 ). Ferritins have the ability to store iron in a nontoxic, bio-available form within a shell, which appears to protect the cell from the harmful effects of iron while making this essential element available when needed (Guerinot, 1994 ; Theil, 1987 ). In addition, Andrews (1998) mentioned that ferritin of Bact. fragilis (Rocha et al., 1992 ), which is taxonomically close to P. gingivalis, may be insignificant in iron storage but play a part in iron detoxification, especially during transient exposure to oxygen. However, the data presented here show that the P. gingivalis ferritin has no direct role in protecting the cells from oxidative damage. This is consistent with the results obtained by Abdul-Tehrani et al. (1999) , who showed that the E. coli ferritin plays no significant role in protecting the cells from oxidative damage in the wild-type genetic background. However, as shown by Wai et al. (1996) , ferritin does play an important role in protecting the organism from oxidative damage in Cam. jejuni. Since each organism possesses its own repertoire of anti-oxidant mechanisms, it is plausible that an organism may possess an alternative protein to substitute for ferritin or use some other protective mechanism(s) which compensates for the lack of ferritin. In this context, it is of note that in E. coli lack of ferritin causes hypersensitivity to various peroxides in the fur genetic background (Abdul-Tehrani et al., 1999 ).

P. gingivalis contains superoxide dismutase, which is essential for tolerance to atmospheric oxygen (Nakayama, 1994 ). Recent studies using E. coli have demonstrated that superoxide toxicity is mainly due to its role in accelerating the Fenton reaction (Keyer & Imlay, 1996 ; McCormick et al., 1998 ). Thus, superoxide attacks iron–sulfur proteins, leading to the release of iron that can participate in the Fenton reaction (McCord & Day, 1978 ). However, Lynch & Kuramitsu (1999) have recently found that P. gingivalis superoxide dismutase seems not to provide protection against hydrogen peroxide. On the other hand, Amano et al. (1988) reported that P. gingivalis does not have catalase, peroxidase, iodine peroxidase or NADH peroxidase activity. Which protein(s) is responsible for protection against peroxide species in P. gingivalis? One candidate is the Dps protein, which has structural and functional similarity to ferritins and bacterioferritins (Bozzi et al., 1997 ; Wolf et al., 1999 ) and abilities to protect DNA against oxidative agents (Almiron et al., 1992 ; Martinez & Kolter, 1997 ). We have found that P. gingivalis possesses a Dps-like protein (data not shown). Another candidate is rubrerythrin, which is a member of the ferritin/bacterioferritin/rubrerythrin superfamily and has thus far been found only in ‘anaerobic’ bacteria (Andrews, 1998 ; Lehmann et al., 1996 ; Bonomi et al., 1996 ; Coulter et al., 1999 ). In fact, a gene encoding rubrerythrin has been identified in the partial genome sequence of P. gingivalis (http://www.tigr.org).

In conclusion, we have shown here that ferritin is important for survival of P. gingivalis under iron-deprived conditions, but that it is not protective against oxidative stresses, at least in the wild-type background. Animal-infection experiments with the ftn mutant will reveal whether intracellular iron storage by ferritin contributes to in vivo survival and virulence of P. gingivalis.


   ACKNOWLEDGEMENTS
 
We thank A. Takade and K. Sakai for technical assistance in electron microscopy and general assistance, respectively. Preliminary P. gingivalis genome sequence data were obtained from the website of The Institute for Genomic Research (http://www.tigr.org). This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture, Japan.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Abdul-Tehrani, H., Hudson, A. J., Chang, Y., Timms, A. R., Hawkins, C., Williams, J. M., Harrison, P. M., Guest, J. R. & Andrews, S. C. (1999). Ferritin mutants of Escherichia coli are iron deficient and growth impaired, and fur mutants are iron deficient.J Bacteriol 181, 1415-1428.[Abstract/Free Full Text]

Almiron, M., Link, A. J., Furlong, D. & Kolter, R. (1992). A novel DNA-binding protein with regulatory and protective roles in starved Escherichia coli.Genes Dev 6, 2646-2654.[Abstract]

Amano, A., Tamagawa, H., Takagaki, M., Murakami, Y., Shizukuishi, S. & Tsunemitsu, A. (1988). Relationship between enzyme activities involved in oxygen metabolism and oxygen tolerance in black-pigmented Bacteroides.J Dent Res 67, 1196-1199.[Abstract]

Andrews, S. C. (1998). Iron storage in bacteria.Adv Microb Physiol 40, 281-351.[Medline]

Andrews, S. C., Smith, J. M. A., Yewdall, S. J., Guest, J. R. & Harrison, P. M. (1991). Bacterioferritins and ferritins are distantly related in evolution.FEBS Lett 293, 164-168.[Medline]

Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A. & Struhl, K. (editors) (1995). Short Protocols in Molecular Biology, 3rd edn. Chichester: Wiley.

Bonomi, F., Kurtz, D. M.Jr & Cui, X. (1996). Ferroxidase activity of recombinant Desulfovibrio vulgaris rubrerythrin.J Biol Inorg Chem 1, 67-72.

Bozzi, M., Mignogna, G., Stefanini, S., Barra, D., Longhi, C., Valenti, P. & Chiancone, E. (1997). A novel non-heme iron-binding ferritin related to the DNA-binding proteins of the Dps family in Listeria innocua.J Biol Chem 272, 3259-3265.[Abstract/Free Full Text]

Chung, M. C. (1985). A specific iron stain for iron-binding proteins in polyacrylamide gels: application to transferrin and lactoferrin.Anal Biochem 148, 498-502.[Medline]

Coulter, E. D., Shenvi, N. V. & Kurtz, D. M.Jr (1999). NADH peroxidase activity of rubrerythrin.Biochem Biophys Res Commun 255, 317-323.[Medline]

Fletcher, H. M., Schenkein, H. A., Morgan, R. M., Bailey, K. A., Berry, C. R. & Macrina, F. L. (1995). Virulence of Porphyromonas gingivalis W83 mutant defective in prtH gene.Infect Immun 63, 1521-1528.[Abstract]

Frazier, B. A., Pfeifer, J. D., Russell, D. G., Falk, P., Olsen, A. N., Hammar, M., Westblom, T. U. & Normark, S. J. (1993). Paracrystalline inclusions of a novel ferritin containing nonheme iron, produced by the human gastric pathogen Helicobacter pylori: evidence for a third class of ferritins.J Bacteriol 175, 966-972.[Abstract]

Guerinot, M. L. (1994). Microbial iron transport.Annu Rev Microbiol 48, 743-772.[Medline]

Halliwell, B. & Gutteridge, M. C. (1992). Biologically relevent metal iron-dependent hydroxyl radical generation: an update.FEBS Lett 307, 108-112.[Medline]

Hudson, A. J., Andrews, S. C., Hawkins, J. M., Williams, J. M., Izuhara, M., Meldrum, F. C., Mann, S., Harrison, P. M. & Guest, J. R. (1993). Overproduction, purification and characterization of Escherichia coli ferritin.Eur J Biochem 218, 985-995.[Abstract]

Hugenholtz, P., Goebel, B. M. & Pace, N. R. (1998). Impact of culture-independent studies on the emerging phylogenetic view of bacterial diversity.J Bacteriol 180, 4765-4774.[Free Full Text]

Izuhara, M., Takanume, K. & Takata, R. (1991). Cloning and sequencing of an Escherichia coli K12 gene which encodes a polypeptide having similarity to the human ferritin H subunit.Mol Gen Genet 225, 510-513.[Medline]

Keyer, K. & Imlay, J. A. (1996). Superoxide accelerates DNA damage by elevating free-iron levels.Proc Natl Acad Sci USA 193, 13635-13640.

Keyer, K., Gort, A. S. & Imlay, J. A. (1995). Superoxide and the production of oxidative DNA damage.J Bacteriol 177, 6782-6790.[Abstract]

Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4.Nature 227, 680-685.[Medline]

Lehmann, Y., Meile, L. & Teuber, M. (1996). Rubrerythrin from Clostridium perfringens: cloning of the gene, purification of the protein, and characterization of its superoxide dismutase function.J Bacteriol 178, 7152-7158.[Abstract]

Lippke, J. A., Strzempko, M. N., Raia, F. F., Simon, S. L. & French, C. K. (1987). Isolation of intact high-molecular-weight DNA by using guanidine isothiocyanate.Appl Environ Microbiol 53, 2588-2589.[Medline]

Lynch, M. C. & Kuramitsu, H. (1999). Role of superoxide dismutase activity in the physiology of Porphyromonas gingivalis.Infect Immun 67, 3367-3375.[Abstract/Free Full Text]

McCord, J. M. & Day, E. D.Jr (1978). Superoxide-dependent production of hydroxyl radical catalyzed by iron–EDTA complex.FEBS Lett 86, 139-142.[Medline]

McCormick, M. L., Buettner, G. R. & Britigan, B. E. (1998). Endogenous superoxide dismutase levels regulate iron-dependent hydroxyl radical formation in Escherichia coli exposed to hydrogen peroxide.J Bacteriol 180, 622-625.[Abstract/Free Full Text]

Martinez, A. & Kolter, R. (1997). Protection of DNA during oxidative stress by the nonspecfic DNA-binding protein Dps.J Bacteriol 179, 5188-5194.[Abstract]

Miller, R. A. & Britigan, B. E. (1997). Role of oxidants in microbial pathophysiology.Clin Microbiol Rev 10, 1-18.[Abstract]

Nakayama, K. (1994). Rapid viability loss on exposure to air in a superoxide dismutase-deficient mutant of Porphyromonas gingivalis.J Bacteriol 176, 1939-1943.[Abstract]

Nakayama, K., Ratnayake, D. B., Tsukuba, T., Kadowaki, T., Yamamoto, K. & Fujimura, S. (1998). Haemoglobin receptor protein is intragenically encoded by the cysteine proteinase-encoding genes and the haemagglutinin-encoding gene of Porphyromonas gingivalis.Mol Microbiol 27, 51-61.[Medline]

Nei, M. (1987). Molecular Evolutionary Genetics. New York: Columbia University Press.

O’Connell, M. J., Ward, R. J., Baum, H. & Peter, T. J. (1985). The role of iron in ferritin and haemosiderin-mediated lipid peroxidation in liposomes.Biochem J 229, 135-139.[Medline]

Rocha, E. R., Andrews, S. C., Keen, J. N. & Brock, J. H. (1992). Isolation of ferritin from Bacteroides fragilis.FEMS Microbiol Lett 95, 207-212.

Sambrook, J., Fritsch, E. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Smalley, J. W., Silver, J., Marsh, P. J. & Birss, A. J. (1998). The periodontopathogen Porphyromonas gingivalis binds iron protoporphyrin IX in the µ-oxo dimeric form: an oxidative buffer and a possible pathogenic mechanism.Biochem J 331, 681-685.[Medline]

Tazaki, K., Inoshita, E., Amano, A., Hanioka, T., Tamagawa, H. & Shizukuishi, S. (1995). Interaction of Porphyromonas gingivalis with transferrin.FEMS Microbiol Lett 131, 161-166.[Medline]

Theil, E. C. (1987). Ferritin: structure, gene regulation and cellular function in animals, plants and microorganisms.Annu Rev Biochem 56, 289-315.[Medline]

Thomas, P. E., Ryan, D. & Levin, W. (1976). An improved staining procedure for detection of the peroxidase activity of cytochrome p-450 on sodium dodecyl sulfate polyacrylamide gels.Anal Biochem 75, 168-176.[Medline]

Towbin, H., Staehelin, T. & Gordon, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications.Proc Natl Acad Sci USA 76, 4350-4354.[Abstract]

Wai, S. N., Takata, T., Takade, A., Hamasaki, N. & Amako, K. (1995). Purification and characterization of ferritin from Campylobacter jejuni.Arch Microbiol 164, 1-6.[Medline]

Wai, S. N., Nakayama, K., Umene, K., Moriya, T. & Amako, K. (1996). Construction of a ferritin-deficient mutant of Campylobacter jejuni: contribution of ferritin to iron storage and protection against oxidative stress.Mol Microbiol 20, 1127-1134.[Medline]

Wallace, R. B. & Miyada, C. G. (1987). Oligonucleotide probes for the screening of recombinant DNA libraries.Methods Enzymol 152, 432-442.[Medline]

Woese, C. R. (1987). Bacterial evolution.Microbiol Rev 51, 221-271.

Wolf, S. G., Frenkiel, D., Arad, T., Finkel, S., Kolter, R. & Minsky, A. (1999). DNA protection by stress-induced biocrystallization.Nature 400, 83-85.[Medline]

Received 26 October 1999; revised 28 January 1999; accepted 7 February 2000.