HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 18, 15293-15302, May 3, 2002

This Article Abstract Full Text (PDF) Supplemental Data All Versions of this Article: 277/18/15293    most recent M111478200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Pathania, R. Articles by Dikshit, K. L. Search for Related Content PubMed PubMed Citation Articles by Pathania, R. Articles by Dikshit, K. L. Social Bookmarking                      What's this?

Mycobacterium tuberculosis Hemoglobin HbO Associates with Membranes and Stimulates Cellular Respiration of Recombinant Escherichia coli^*,

Ranjana Pathania, Naveen K. Navani, Govindan Rajamohan, and Kanak L. Dikshit^§

From the Institute of Microbial Technology, Sector 39 A, Chandigarh 160036, India

Received for publication, December 2, 2001, and in revised form, January 15, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The truncated hemoglobins HbN and HbO of Mycobacterium tuberculosis H37Rv share little sequence similarity and display structural differences in their EF-loop regions, suggesting distinct function(s) for these hemoglobins. HbO of M. tuberculosis was expressed in Escherichia coli and Mycobacterium smegmatis as a 14.5-kDa homodimeric heme protein exhibiting nearly 50-fold (P₅₀ ~0.51) lower oxygen affinity than HbN. 40-50% of HbO remained associated with the cell membranes and significantly enhanced its respiration in comparison with the membrane fractions of control cells or cells overproducing HbN. Oxygen uptake of HbO-associated membranes was decreased by washing and restored by adding HbO. Additionally, membrane vesicles prepared from terminal oxidase-deficient (cyo, cyd) mutants of E. coli did not exhibit significant enhancement in oxygen uptake in the presence of HbO, suggesting its interaction(s) with the electron transport chain. Expression of HbO in Mycobacterium bovis bacillus Calmette-Guérin, an experimental model of M. tuberculosis, was observed (0.2-0.5% of total cellular proteins) throughout its aerobic growth. These results provided evidence for the involvement of HbO with the component of aerobic electron transport chain, suggesting that its function may be related to the facilitation of oxygen transfer during aerobic metabolism of M. tuberculosis. Membrane association properties of HbO may thus play a crucial role in sequestering oxygen and facilitating its availability to internalized M. tuberculosis (an obligate aerobe) under the hypoxic conditions of its intracellular habitat.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

With about one-third of the world population infected and causing nearly three million deaths annually (1-3), Mycobacterium tuberculosis arguably is one of the world's most successful human pathogens. Efforts to combat mycobacterial disease are hampered by our inadequate understanding of the underlying cellular metabolism and pathogenicity. Pathogenic mycobacteria such as M. tuberculosis and Mycobacterium leprae can survive within the host cell because of their ability to withstand the hostile cellular environment. According to conventional wisdom, M. tuberculosis is an obligate aerobe (4), but during its intracellular regime it encounters severe hypoxia and nitrosative stress. Within macrophages and avascular calcified granulomas, where the organism primarily resides during the long latent period of infection, oxygen tension may be even lower (5, 6). M. tuberculosis appears to have not only the remarkable capability of adapting its metabolism to environmental changes but also has the ability to successfully compete with the lungs for oxygen to sustain its aerobic metabolism within the oxygen-deficient environment of its intracellular habitat. The glbN and glbO genes, encoding hemoglobin-like proteins (HbN and HbO, respectively) have been detected in the genome sequence of M. tuberculosis (7). Recently, the glbN gene of M. tuberculosis has been overexpressed in Escherichia coli and has been shown to encode a functional homodimeric hemoglobin displaying very high oxygen binding affinity and cooperativity that makes it unsuitable as an oxygen carrier (8). Resonance Raman spectra of HbN revealed that the structural features of the distal heme pocket of HbN are distinct from those of other globins and carries unstrained heme iron-proximal coordination ideally suited for oxygen and nitric oxide interactions (9). It has been speculated that HbN may be involved in oxygen-sustained detoxification of nitric oxide, providing a defense mechanism to the bacillus against macrophage-generated reactive nitrogen species. At present we do not have any information on the other mycobacterial hemoglobin, HbO. The presence of two different Hb encoding genes in the genome sequence of M. tuberculosis indicates that these two hemoglobins may have distinct functions in the oxygen metabolism of the tubercle bacillus.

The mycobacterial hemoglobins HbN and HbO belong to the superfamily of truncated hemoglobins consisting of small heme proteins of 110-130 amino acid residues per chain that carry substantial deletions within their N or C termini along with the replacement of crucial heme-binding F-helix with an extended polypeptide loop (10, 11). The aligned amino acid sequences predict that the distal heme pocket of trHbs¹ hosts an almost invariant tyrosine at the B10 position and a phenylalanine or tyrosine at the CD1 position. The E7 position in the distal pocket of the trHbs is variable, whereas in other vertebrate and non-vertebrate hemoglobins it is conserved (usually a histidine or glutamine), suggesting that other structural factors may modulate the stabilization and overall ligand binding affinity of these trHbs. Thus far, small trHbs from the ciliated protozoa (12, 13), unicellular alga (14), cyanobacteria (15), and M. tuberculosis (8, 9, 11) have been studied. High resolution crystal structures of trHbs from Chlamydomonas eugametos, Paramecium caudatum, and M. tuberculosis HbN (10, 11) have revealed an alternative folding pattern comprising a two-over-two sandwich of -helices in these hemoglobins instead of the three-over-three -helical sandwich of the classical globin fold. One of the most striking structural features of trHbs, emerged from the studies on HbN, is the presence of a continuous tunnel/cavity connecting the heme distal pocket to the protein surface at two distinct sites (11), which may facilitate the diffusion of ligands such as oxygen and nitric oxide. An analysis of the currently available microbial genome sequence data suggests that truncated hemoglobins may be distributed widely in several pathogenic and nonpathogenic microbes. The functional roles of trHbs are virtually unknown at present and may be various. For example, in the cyanobacterium, Nostoc commune, the hemoglobin is localized along the cytosolic face of the cell membrane and is expressed under low oxygen and presumably help in the electron transfer (16). In the unicellular alga, C. eugametos, trHb is induced in response to activated photosynthesis (14), whereas M. tuberculosis HbN synthesis is induced during the stationary phase (8). The presence of trHbs in a wide variety of single-celled microorganisms indicates that these unique oxygen-binding heme proteins may play a pivotal role in the cellular metabolism.

In the present work, we have cloned and overexpressed the glbO gene from M. tuberculosis in E. coli and Mycobacterium smegmatis in an attempt to study the characteristics and potential function of the mycobacterial hemoglobin, HbO. We present evidence demonstrating that HbO interacts with the respiratory membranes of E. coli and M. smegmatis and participates in oxygen uptake and transfer during aerobic growth. These observations have vital implications in the context of the initial stages of intracellular growth and survival of the obligate aerobe M. tuberculosis in the oxygen-deficient environment of its host. Our results strongly suggest that the presence of HbO in M. tuberculosis provides an efficient way of sequestering and competing for oxygen during different stages of intracellular growth to sustain aerobic metabolism.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Bacterial Strains, Plasmids, and Culture Conditions-- E. coli JM109 and E. coli BL21DE3 strains were used routinely for cloning and expression of recombinant genes. E. coli strains ECL 936 (cyo, cyd⁺) and ECL 937 (cyo⁺ cyd) were kindly provided by Dr. E. C. C. Lin (17). Cultures of E. coli strains were grown in Luria-Bertani (LB) medium at 37 °C at 180 rpm. M. bovis bacillus Calmette-Guérin (BCG) ATCC 35734 and M. smegmatis mc²155 (18) were grown in Middlebrook 7H9 broth or 7H10 agar (Difco) supplemented with ADC (10% bovine serum albumin fraction V, dextrose, and sodium chloride), 0.2% glycerol, and 0.05% Tween 80 or in Sauton's minimal medium. When required, kanamycin sulfate (Sigma) and hygromycin B (Roche Molecular Biochemicals) were added at 25 and 50 µg/ml, respectively, for M. smegmatis or 50 and 200 µg/ml, respectively, for E. coli. Plasmids BlueScript KS⁺ (Stratagene), pET9b (Promega), and p19Kpro (19) were utilized for cloning and expression of recombinant genes. Restriction and modifying enzymes were obtained from New England Biolabs Inc. (Beverly, MA). ATP estimation kit (Sigma), Middlebrook 7 H9, Middlebrook 7H10, ADC enrichment, and OADC enrichment (Difco Laboratories, Detroit, MI) were available commercially. The oligonucleotides were custom synthesized from PMK International.

Cloning and Expression of M. tuberculosis glbN and glbO Genes in E. coli and M. smegmatis-- The forward primers for the glbN (5'-GCTGTTCCATTGGGACTACTGTCACGCTT-3') and glbO (5'-GATCATATGCCGAAGGTCTTTCTACGA-3') genes and the reverse primers for the glbN (5'-CGGGA TCCTCAGACTGGTGCCGTGGT-3') and glbO (5'-CCATCAAAACGGGGA GTTGACCAGCGA-3') genes were designed on the basis of the published M. tuberculosis genome sequence (7). An NdeI site was incorporated at the 5' end of the forward primers of the two genes. Two genes, glbO and glbN, were retrieved from the genomic DNA of M. tuberculosis H37Rv (kindly provided by Dr. M. Dube, PGIMER, Chandigarh, India) and M. bovis BCG following PCR amplification and sequenced completely to check the authenticity of the cloned genes. The glbO and glbN genes from M. bovis were found to be identical to the M. tuberculosis glbO and glbN genes, respectively. For the overexpression of hemoglobin-encoding genes in E. coli, the NdeI-BamHI-digested fragments of glbN and glbO genes were further subcloned on the expression vector pET9b and subsequently transformed into E. coli BL21DE3. These plasmid constructs overexpressing HbO and HbN were designated pET-glbO and pET-glbN, respectively. For the overexpression of mycobacterial hemoglobin in E. coli, a single colony carrying the respective plasmid construct was inoculated in LB medium supplemented with kanamycin (30 µg/ml) and allowed to grow at 37 °C until A₆₀₀ reached ~0.45. The cultures were then induced with 0.4 mM isopropyl--D-thiogalactopyranoside and further incubated for another 8 h. The expression of cloned protein was monitored after running the cell lysate of recombinant strains on 15% SDS-PAGE followed by Coomassie Brilliant Blue staining.

For the expression of hemoglobin genes in M. smegmatis, glbN and glbO genes were amplified separately through PCR using gene-specific primers having a BamHI restriction site on the forward primer and a PstI restriction site on the reverse primer, respectively. The PCR products were sequenced and cloned on a mycobacterial expression vector, p19Kpro (19), under the constitutive promoter of a 19-kDa antigen gene of mycobacterium and electrotransformed separately into M. smegmatis cells. The expression of the two hemoglobin proteins was confirmed separately after running the cell lysate on a 15% SDS-PAGE.

Isolation, Purification, and Characterization of HbO-- For protein purification, cell culture of E. coli BL21DE3 overexpressing HbO was harvested by centrifugation at 6000 rpm for 10 min at 4 °C and resuspended in 10 mM Tris-Cl (pH 8.0) having 10 mM dithiothreitol, 1 mM EDTA, 45 µg/ml phenylmethylsulfonyl fluoride, 500 µg/ml RNase A, and 100 units/ml DNase I. Cells were lysed after passing through French pressure cell and then subjected to ultracentrifugation (45,000 rpm, 4 °C, 2 h). The clear reddish brown supernatant thus obtained was loaded on an ion exchange column (DEAE-Sepharose CL-6B, Amersham Biosciences), pre-equilibrated with 10 mM Tris-Cl (pH 8.0), and eluted using 0.12 M NaCl. This step resulted in about 80% pure preparation of HbO exhibiting distinct reddish brown color. This fraction was further purified by gel-filtration chromatography. Briefly, partially purified recombinant HbO protein preparation was loaded onto a Superdex-75 column (Amersham, 30 × 10 cm) and eluted in 0.15 M NaCl in10 mM Tris-Cl (pH 8.0) at a flow rate of 0.3 ml/min. The protein and hemoglobin elution profiles were monitored at 280 and 410 nm, respectively.

Spectral and Oxygen Binding Studies-- Absorption spectra of whole cells or protein preparation were recorded using a Shimadzu or Cary 210 spectrophotometer. CO difference spectra of whole cells carrying HbO were obtained after bubbling CO for 1 min in the cell suspension (20 mg/ml wet weight) of recombinant cells. The oxygen dissociation curve of HbO was obtained by the tonometer method (20).

Isolation of Respiratory Membranes-- E. coli cells grown in LB medium at 37 °C for 10 h were harvested after centrifugation at 5000 rpm for 15 min at 4 °C and washed twice with 10 mM Tris-Cl (pH 7.2). Spheroplasts were then prepared by incubating 10 g of cell paste in 30 ml of 50 mM Tris-Cl (pH 7.2) containing 250 mM sucrose, 5 mM EDTA, and 8 mg of lysozyme for 30 min at 25 °C. After incubation, the spheroplasts were harvested by centrifugation at 17,000 × g for 30 min at 4 °C and resuspended in 5 ml of 10 mM Tris-Cl (pH 7.2). 100 units of DNase I and 1 mg of RNase A were added to the suspension, which was incubated at 25 °C for 3 h. This suspension was then sonicated for 5 min to disrupt the spheroplasts and centrifuged at 48,000 × g for 1 h at 4 °C. The supernatant cytosol and the pelleted membranes were separated. The membranes were washed once with 10 mM Tris-Cl (pH 7.2), and oxygen uptake was determined before and after washing of the membranes with 10 mM Tris-Cl (pH 7.2).

Measurement of Specific Oxygen Consumption Rate-- The specific oxygen consumption rate was measured with a Yellow Springs Instruments model 55 oxygen monitor in air-saturated 0.1 M potassium phosphate buffer (pH 7.2) at 25 °C. 1 ml of cell culture (the total number of cells/ml was simultaneously determined by plating on LB) was concentrated by centrifugation at 12,000 × g for 10 min and washed twice with 0.1 M potassium phosphate buffer (pH 7.2). The resulting pellet was added quantitatively to 4 ml of air-saturated buffer. The change in oxygen concentration of the buffer containing cells was recorded with respect to time. Succinate was used as an external substrate where required.

Estimation of Cellular ATP Level-- ATP estimation of whole cells was done using the Sigma ATP estimation kit according to the manufacturer's instructions. Briefly, 1 ml of cell suspension equivalent to 1 × 10⁸ cells from the late log phase cultures was pelleted in a microcentrifuge and washed twice with 2 mM potassium phosphate buffer (pH 7.2). The washed pellet was then suspended in 500 µl of the same buffer and sonicated to disrupt the cell. The cell debris was removed after centrifugation at 14,000 K, and the clear supernatant was treated with 10% trichloroacetic acid to separate proteins from the cell lysate and subsequently used for the ATP estimation as described elsewhere (21).

Preparation of HbO Antiserum and Western Blot Analysis-- Polyclonal antisera against HbO and HbN were raised individually by immunizing rabbits using standard procedures (22). Briefly, HbO or HbN was purified from the cell extracts of recombinant E. coli and run on a 15% preparatory SDS-PAGE. The 14.5-kDa protein band corresponding to trHb (HbO or HbN) was eluted from the gel, emulsified in incomplete Freund's adjuvant, and injected subcutaneously into the rabbit at multiple sites. Boosters were given at 21-day intervals. After the third booster dose, blood was collected, and the serum was prepared. The polyclonal antibodies were saturated with the crude cell extract of E. coli to block the nonspecific sites, and the titer of antiglobin (HbN and HbO) was determined by enzyme-linked immunosorbent assay.

For Western blotting, purified proteins or cell extracts (500 ng, 3 µg of protein/slot) were resolved on SDS-PAGE and transferred onto a nitrocellulose membrane (0.45 µm) in a mini-transblot apparatus (Bio-Rad). Immobilized proteins were probed with primary (anti-HbO or anti-HbN) and secondary (horseradish peroxidase-conjugated anti-rabbit IgG) antibodies and developed using diaminobenzidine and H₂O₂.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Structural Features of HbO Encoded by the glbO gene of M. tuberculosis H37 Rv-- A number of putative trHbs have been identified within the available finished and unfinished genome sequences of various organisms including many severely pathogenic microorganisms (Ref. 23, Fig. 1), e.g. M. tuberculosis, M. leprae, Mycobacterium avium, Bacillus anthracis, Corynebacterium diphtheriae, etc. A structure-based sequence alignment of different mycobacterial trHbs (HbN and HbO) along with other microbial trHbs is given in Fig. 1. A comparative analysis of the main structural features of trHbs with various mycobacterial hemoglobins indicated clear structural conservation of the main protein regions in HbO, which is crucial for the stabilization of the trHb fold. Among these are three conserved Gly motifs, an extended pre-F-loop region, and a conserved phenylalanine-tyrosine at B9 and B10, phenylalanine at E14, and histidine at the F8 position. However, the conserved CD1 and ligand-binding E7 positions are occupied by variable amino acid residues in HbO-type trHbs, suggesting that these positions are not very crucial to the structural and functional integrity of these hemoglobins. M. tuberculosis HbO also appears to lack a 10-amino acid residue-long highly polar pre-A region that ties the N terminus with the core of the protein and modulates overall packing of M. tuberculosis HbN (11). Sequence identity between various mycobacterial HbN and HbO ranges only between 13 and 18%, whereas sequence similarity between various HbO-type trHbs is much higher and ranges between 30 and 85%. From the sequence comparison of various trHbs some interesting differences between HbN and HbO can easily be visualized (Fig. 1). The majority of the bacterial trHbs share sequence similarity with HbO rather than HbN of M. tuberculosis, which is more similar to protozoan, algal, and cyanobacterial hemoglobins (28 to 47%). A comparison of the sequence composition of trHbs indicates that HbO and related truncated Hbs exhibit a 15-16-residue-long EF-loop region carrying a highly charged and polar sequence motif, GHP(R/M)LRNRH, that is more or less conserved within this group. In contrast, HbN and other related protozoan and cyanobacterial trHbs have a shorter pre-EF-loop (10-11 residues) and do not exhibit any sequence conservation within their F-loop region. Because the EF-loop and F-helix significantly affect the structural orientation of His F8 imidazole and the oxygen binding properties of HbN of M. tuberculosis (11), elongation of the EF-loop region and unique sequence composition of the F-helix region of HbO and related truncated Hbs suggest that the functional properties of HbN and HbO-type truncated Hbs may be different. Most of the HbO-type trHbs carry a single cystine residue at the G10 position. This position is surface-exposed in the trHb fold (11) and may be crucial for the maintenance of the subunit contact within the HbO homodimer.

View larger version (95K): [in this window] [in a new window]   Fig. 1.   Sequence alignment of mycobacterial hemoglobins HbN and HbO along with various truncated hemoglobins. Sequence comparison was done using Clustal W (38). The globin fold topological positions are shown above the aligned sequences. Important residues with respect to coordination of the heme and the ligand binding residue properties are marked. Residues that are conserved in the trHb family are highlighted in black. Conserved residues of HbO related trHbs are shown in the gray-shaded boxes with bold letters. Sequence data for various hemoglobins were obtained from the web sites of NCBI and TIGR (the respective accession numbers of various trHbs are provided in the Supplemental Materials).

Cloning and Expression of the glbO Gene of M. tuberculosis H37RV in E. coli and M. smegmatis-- The glbO gene from M. tuberculosis H37Rv was expressed in E. coli under T7 promoter that resulted in the accumulation of 15-20% of heme protein inside the cell imparting a reddish brown tinge to the recombinant E. coli cells. SDS-PAGE analysis of these cells confirmed the presence of a 14.5-kDa protein corresponding to the expected size of HbO (Fig. 2A). The absolute absorption spectra of the cell lysate of the HbO-carrying cells indicated the presence of a Soret peak at 413 and  and  peaks at 544 and 581 nm (Fig. 3A), which is very similar to oxyhemoglobin and myoglobins, suggesting that the predominant form of HbO remains oxygenated in vivo. CO difference spectra of HbO-carrying cells exhibited a sharp peak at 419 nm (Fig. 3B) as has been observed previously in the case of other bacterial hemoglobins (8, 24-26). The glbO gene was expressed in M. smegmatis under the constitutive promoter of 19-kDa antigen of mycobacterium. The SDS-PAGE profile (Fig. 2B) and its densitometric analysis indicated that HbO constitutes 2-3% of total cellular proteins. Spectral analysis of these cells further confirmed the presence of HbO in M. smegmatis.

View larger version (41K): [in this window] [in a new window]   Fig. 2.   Overexpression of HbO in E. coli and M. smegmatis. The total cellular protein content (10-15 µg) of recombinant cells expressing HbO was resolved on 15% SDS-PAGE as described under "Experimental Procedures," and protein bands were visualized after Coomassie Blue staining. A, expression of HbO in E. coli. Lane 1, E. coli BL21DE3 carrying pET-glbO plasmid construct; lane 2, E. coli BL21DE3 carrying pET9b alone; M, molecular mass marker. The numbers denote the positions of the molecular mass markers. B, expression of HbO in M. smegmatis. Lane 1, M. smegmatis cells carrying p19Kpro-glbO plasmid construct; lane 2, M. smegmatis cells carrying p19Kpro alone; M, molecular size markers. The numbers denote the size of the molecular size markers.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Spectral characteristics of HbO. A, absolute absorption spectrum of purified HbO. The inset indicates expansion of spectra. B, CO difference spectra of recombinant E. coli expressing HbO. The recombinant E. coli cells were collected by centrifugation and washed with sodium phosphate buffer (pH 7.5). The cell pellet thus obtained was resuspended in the same buffer at a final concentration of 25 mg/ml. The CO difference spectra were recorded after bubbling CO into the sample cuvette for 2 min. C, oxygen equilibrium curves of HbO at 10 and 20 °C measured in 50 mM Tris-Cl, 1 mM dithiothreitol, and 50 mM KCl.

Growth and Specific Oxygen Consumption Rates of Recombinant E. coli and M. smegmatis Carrying Mycobacterial HbO-- To determine whether the functional expression of HbO had any metabolic effect(s) on its host, we compared the growth and oxygen consumption properties of recombinant E. coli and M. smegmatis carrying HbO with the control cells carrying similar plasmid without the glbO gene. In E. coli leaky expression of the glbO gene occurred even without isopropyl--D-thiogalactopyranoside, which constituted nearly 2-3% of total cellular protein more or less similar to the level of HbO expressed in M. smegmatis under the constitutive promoter of 19-kDa antigen using the mycobacterial expression plasmid p19Kpro. HbO carrying E. coli and M. smegmatis cells outgrew control cells and resulted in a higher cell mass of aerobically growing cells (Fig. 4, A and B). The difference in growth rate of these strains became more obvious during late log and stationary phase. In all cases the correlation between optical density (OD) and the wet and dry cell mass was linear as determined by intermittent cell mass weight determination, thus justifying the use of OD as a measure of growth. The presence of HbO in recombinant E. coli and M. smegmatis resulted in a significantly higher increase in oxygen uptake as compared with their respective control cells (Table I). In contrast, E. coli cells, carrying more or less similar levels of HbN, exhibited only a marginal increase in oxygen consumption. The oxygen uptake rate of HbO-carrying E. coli cells was nearly 2-fold higher in comparison with the control or HbN-carrying cells. Similarly, a 1.5-fold enhancement in oxygen uptake was observed when HbO was expressed in M. smegmatis (Table I).

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Growth profile of E. coli and M. smegmatis carrying M. tuberculosis HbO. A, E. coli cells, carrying pET9b and pET-glbO, were grown in 250-ml baffled flasks carrying 100 ml of LB supplemented with 30 µg/ml kanamycin and incubated at 37 °C at 200 rpm on a gyratory shaker and A600 was monitored at different time intervals. B, growth characteristics of M. smegmatis, carrying p19Kpro and p19Kpro-glbO. The cells were allowed to grow in Middlebrook 7H9 broth at 37 °C at 200 rpm.

Isolation Purification and Characterization of HbO-- The HbO protein was purified to near homogeneity (>95% purity) from the cell extract of recombinant E. coli using two chromatographic steps involving ion exchange and gel filtration. The purified HbO migrated with an apparent molecular mass of 14.5 kDa on SDS-PAGE, whereas gel-filtration analysis of HbO indicated its molecular mass at around 29 ± 1 kDa, suggesting that the protein was present primarily as a homodimer.

The UV-visible spectrum of the recombinant HbO (Fig. 3, A and B) exhibited the typical characteristic of an oxygen-bound heme protein, with maxima at 412 (Soret), 580 (-band), and 542 nm (-band). After deoxygenation of HbO with sodium dithionite, the Soret peak shifted to 422 nm, and the  and  bands converged to a broad peak at 554 nm. This behavior is consistent with the formation of the deoxygenated forms and is similar to the effect seen with deoxyhemoglobin. The addition of carbon monoxide to the reduced protein resulted in the Soret peak at 419 nm, indicating the presence of the ferrous form of HbO, capable of reacting with CO. The addition of NAD(P)H to ferric HbO under aerobic conditions gave spectral characteristics very similar to the oxyform of myoglobin (27) and Vitreoscilla hemoglobin (28), exhibiting a Soret peak at 419 nm and - and -bands at 545 and 580 nm, respectively. The calculated oxygen concentration at 50% saturation (P₅₀ = ~0.51; Fig. 3C) of HbO is around 50-fold higher than HbN (0.01). Thus, the oxygen affinity of HbO is significantly lower than that of HbN. This difference suggests that the cellular function of HbO differs from that of HbN.

HbO Appears as a Peripheral Membrane-associated Heme Protein in E. coli-- During its purification from recombinant E. coli, a substantial fraction of HbO was found associated with the membrane preparations of cells that constituted nearly 40-50% of the total HbO present. A significant amount of the protein remained associated with the membrane fractions even after several washings (Fig. 5B, I and II) and could only be removed in the presence of the chaotropic agent, 0.2 M potassium thiocyanate (KSCN), that interrupts hydrophobic interactions. For comparison, when E. coli cells, overexpressing HbN, were checked under similar conditions, more than 90% HbN was found in the clear lysate and the membrane fractions did not show any significant level of associated HbN (Fig. 5A, I and II). When the pattern of hydrophobic stretches in HbO were checked for the presence of solvent accessible surface area and compared with that of HbN, hydrophobicity profile of HbO was found significantly higher than the HbN. HbO contains two distinctly hydrophobic patches (35-55 and 75-95 aa residues) that in HbN are mainly hydrophilic and may be crucial for determining the membrane association properties of HbO.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Membrane association properties of M. tuberculosis hemoglobin HbO. Recombinant E. coli cells expressing HbN (A) and HbO (B) were grown in LB for 6 to 8 h. The membrane fraction and the total soluble proteins were isolated as mentioned under "Experimental Procedures." Fractions from equal number of cells were examined by Coomassie Blue staining (I) and Western blotting (II) after electrophoresis on SDS-PAGE (15%). A, Lanes 1 and 4, total soluble proteins (10 µg); lane 3, total membranes (10 µg); lane 2, total membrane proteins after two washes with buffer. B, Lanes 1 and 6, total soluble proteins (15 µg); lane 2, total membrane proteins (20 µg); lane 3, total membrane proteins after two washes with the buffer; lane 4, total membrane proteins after one wash with KSCN; lane 5, total proteins that appeared in the supernatant after one wash with KSCN. M, denotes molecular mass marker.

Membrane Vesicles of E. coli Carrying HbO Exhibit Enhanced Oxygen Consumption Activity-- Because the observations above indicate a close association of HbO with the E. coli respiratory membranes, unlike HbN, which appears predominantly in the soluble fraction, a logical next step was to compare the oxygen uptake properties of cell membranes prepared from the recombinant strains carrying these mycobacterial hemoglobins. The respiration of membrane fraction prepared from HbO-carrying cells were about twice that of the membrane preparation of the control cells (Fig. 6A, lanes 1 and 6). The oxygen uptake of both control and HbO-carrying membranes was stimulated in the presence of succinate and decreased in the presence of cyanide (data not shown). However, oxygen uptake of HbO-carrying cells decreased after successive washings of the membranes. The increase in oxygen consumption was observed again when washed membranes were sonicated (Fig. 6A), presumably because of the release of HbO trapped in the membrane vesicles. In contrast, no significant change in oxygen uptake was observed from the membrane fractions prepared from the control or E. coli cells overexpressing HbN.

View larger version (46K): [in this window] [in a new window]   Fig. 6.   A, oxygen uptake characteristics of cellular membranes of E. coli carrying HbO and HbN. The preparation and washing of membrane vesicles are described under "Experimental Procedures." The oxygen uptake rate was measured in the presence of 25 mM succinate. The protein concentration of the membrane was 15 mg/ml. Column 1, membranes of E. coli expressing HbO and HbN, respectively; column 2, membrane fraction of E. coli expressing HbO and HbN after three washes; column 3, after seven washes; column 4, membranes were sonicated after seven washes and washed once again (column 5) after sonication; column 6, membranes of control E. coli cells. B, oxygen uptake of the isolated membrane vesicles of control E. coli cells after addition of purified HbO and HbN. Column 1, membrane fraction (0.15 mg/ml) of E. coli BL21DE3; column 2, membrane fraction after addition of HbO- and HbN-carrying cytosol (0.30 mg/ml), respectively; column 3, membrane fraction plus HbO- or HbN-carrying cytosol plus 3 mM KCN; column 4, cytosol alone exhibiting negligible respiration in the absence of membranes; column 5, membrane fraction plus purified HbO/HbN (15 µM); column 6, cell membrane plus purified HbO/HbN (15 µM) plus 3 mM KCN.

The respiration lost by washing of the membrane fraction prepared from HbO-carrying E. coli cells was regained by adding HbO-carrying cytosol or purified HbO (data not shown). These results prompted us to further test the interaction of HbO with the cellular membranes of E. coli not expressing any hemoglobin. Purified HbO or cytosol prepared from the HbO-carrying cells was added to the membrane preparation of control E. coli, and its oxygen uptake characteristics were studied. As control experiments, membranes were omitted in each case to rule out the possibility that succinate or any component of the cytosol reduces HbO. The addition of purified HbO or cytosol containing HbO enhanced the oxygen uptake rate of the membrane fraction by 4- and 3-fold, respectively. In contrast, the addition of HbN resulted in only a marginal change in the consumption of oxygen (Fig. 6B). These results provide further evidence in support of the participation of HbO in oxygen/electron-transfer process. Because total energy status in terms of ATP level is a direct indication of the metabolic condition of the cells, the energetic consequences of HbO expression on E. coli cells was further checked to explain the response of HbO on the physiology and metabolic activity of cells. The intracellular ATP concentration of late log phase cells of E. coli carrying HbO increased 1.5- and 2-fold relative to HbN-carrying and control cells, respectively (Fig. 7).

View larger version (9K): [in this window] [in a new window]   Fig. 7.   Cellular ATP pool of recombinant E. coli carrying HbN and HbO.

Oxygen Uptake of Membrane Vesicles of Terminal Oxidase-deficient Mutants of E. coli in the Presence of HbO-- Because the initial results on the interaction of HbO with cell membranes suggested that the action of HbO in vivo may be linked to cell respiration, the effect of HbO on the kinetics of oxygen consumption by respiratory membranes of E. coli mutants lacking one of the terminal oxidases was monitored. The first investigation was done on how HbO would affect the oxygen uptake rate of cells if the respiratory membranes carry only one terminal oxidase. The oxygen uptake rate of the cellular membranes of cyo and cyd mutants of E. coli was measured after adding a purified preparation of HbO or HbO-carrying cell lysate. The oxygen uptake rate of respiratory membranes of the wild type and the cytochrome d-deficient mutant (cyo⁺ cyd) of E. coli increased nearly 2- and 1.5-fold, respectively, in the presence of HbO (Table II). In contrast, membrane preparation of the cytochrome o-deficient (cyo cyd⁺) mutant of E. coli exhibited only a marginal change in oxygen consumption in the presence of HbO as compared with the control membrane fraction. These results provided evidence for the interactions of HbO with the terminal oxidases, presumably cytochrome o.

Expression of HbO during Growth Cycle of M. bovis-- Because our physiological studies on recombinant E. coli and M. smegmatis suggested a correlation of HbO with oxygen utilization and electron transport, it was further investigated as to whether HbO is essentially required for the aerobic growth of the Mycobacterium or is needed only under certain growth and physiological conditions. The pattern of HbO biosynthesis in M. bovis (ATCC 35734) was examined; M. bovis is used widely as a model system for M. tuberculosis and carries the identical glbO gene (deposited in GenBank^TM under accession No. AF213450). Polyclonal antibodies raised against HbO were utilized to detect the presence of HbO under different physiological conditions of M. bovis. In aerobic culture, HbO was detected during all growth phases of M. bovis and constituted 0.2-0.5% of total cellular proteins (Fig. 8). Western blot and densitometric analysis of cellular proteins, taken from different growth periods, indicated a monotonic increase in HbO expression reaching a plateau maximum at 100 h. These results indicated the requirement of HbO for the life cycle of M. bovis and presumably of M. tuberculosis.

View larger version (13K): [in this window] [in a new window]   Fig. 8.   Expression profile of HbO in M. bovis. HbO is expressed throughout the life cycle of M. bovis. M. bovis BCG cells were grown at 37 °C at 200 rpm. Aliquots of cell culture were taken out at the indicated times for the A600 determination, and Western blot analysis of total cellular proteins was performed using HbO-specific polyclonal antibodies. Each lane contains 5 µg of total protein.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

All aerobic cells require the transfer of oxygen from the environment in order to respire. Eukaryotic globins, well known oxygen-transporting molecules, usually mediate the process of oxygen transfer. Their extraordinary ability to do so is the result of two central properties. First, these heme proteins are localized so as to connect regions of oxygen availability with those in which oxygen is utilized. Second, each globin has optimized its oxygen binding affinity so as to be most effective in its local environment. Detection of hemoglobins or hemoglobin-like proteins in virtually all kingdoms of life (29, 30), including vertebrates, invertebrates, plants, and various microbes, suggests a widespread requirement for this protein in cellular metabolism. Small hemoglobins exhibiting unusual folding characteristics due to truncation and reorientation in their overall globin fold have recently been identified in several pathogenic and nonpathogenic unicellular organisms (23), suggesting the vital requirement of these heme proteins in cellular metabolism. The human pathogen M. tuberculosis synthesizes two truncated hemoglobins, HbN and HbO, of unknown functions. Structural analysis and comparison of the biochemical properties of HbN and HbO suggested that these two mycobacterial hemoglobins differ in many ways and may have distinct cellular functions.

Sequence comparison of various HbN and HbO type trHbs suggests that, despite having structural conservation of the main protein regions thought to be crucial for the stabilization of the trHb fold, there are major differences within the pre-EF and F-loop regions of HbN and HbO. The F-loop region of HbO appears to be more elongated and carries a highly charged sequence motif, HP(R/M)LRNRH, which appears more or less conserved in other HbO-related trHbs. The functional relevance of this sequence motif is currently unknown. The crystal structure of mycobacterial HbN (11) has indicated that the pre-EF-loop region makes specific interactions with the porphyrin group and plays a key role in modulating the conformation of proximal heme pocket. Therefore, these differences between HbO and HbN may result in differences in their functional properties. There are several surface-exposed positively charged residues within the F-loop region of HbO, and it is quite possible that this region of HbO interacts with other cellular components also and may play a specific role. Our results show that M. tuberculosis HbO has lower oxygen affinity than HbN. Structural studies on M. tuberculosis HbN (11) have indicated that the distal network of H-bonds between Tyr at B10, Leu at E7, and Gln at E11 position contribute to the stabilization of ligated oxygen, resulting in high oxygen affinity of this protein. Structures of other HbN-type trHbs from C. eugametos and P. caudatum also show that hydrogen bonding between Tyr at B10, Gln at E7, and Leu at E11 contribute to the effective structuring of the distal sites of these trHbs. Although, M. tuberculosis HbO carries a tyrosine residue at the B10 position, its E7 and E11 positions are occupied by alanine and leucine residues, respectively, and may be less effective in generating a close network of hydrogen bonding within the distal oxygen binding pocket, resulting in lower oxygen affinity of HbO.

To explore the possible mechanism(s) of HbO function, we checked its physiological response in E. coli, which does not carry any truncated hemoglobin. Spectral studies conducted on recombinant E. coli and M. smegmatis indicated that HbO remains in the oxygenated form and enhances the oxygen utilization properties of these heterologous hosts. HbO is more hydrophobic and exhibits lower oxygen affinity than HbN. Expression of the hydrophobic and presumably membrane-associated HbO in E. coli results in a much higher increase in oxygen uptake compared with HbN, which remains mainly cytoplasmic. The close proximity of oxygenated HbO with cellular membranes could facilitate oxygen delivery to the oxygen binding sites of the terminal oxidases, which remain oriented toward the cytoplasm (31). In fact human hemoglobin has been shown to possess the ability to associate reversibly with the erythrocyte membranes (32). The P₅₀ value of HbO is reasonably close to human hemoglobin and myoglobin. Its presence in the close proximity of cell membranes indicates that it is well tailored for the need of cell respiration. If the K_d is too large, then the relative contribution of the globin-mediated oxygen flux becomes insignificant. On the other hand, if the K_d is too small, as in case of HbN, then even at limiting oxygen concentrations, the protein remains largely saturated and is less useful for the cell. It is quite possible that HbO results in cytoplasmic gradients of free and bound oxygen, and these could play a significant physiological role. M. tuberculosis is an obligate aerobe. One of the major challenges encountered by this bacterium is to sustain its aerobic growth and metabolism within the intracellular environment where oxygen availability is extremely low. The close proximity of HbO with the respiratory membranes and its moderate oxygen-binding capability, very similar to human hemoglobin, may help it to compete effectively with the lungs for oxygen and to be unloaded easily at the site of membranes, to facilitate cellular respiration. Truncated hemoglobins from Nostoc commune (16) and bacterial hemoglobin (VHb) from Vitreoscilla have been shown closely associated with cellular membranes (33). It has been proposed that these proteins may be involved in electron transport and oxygen transfer.

Respiratory membranes of lysed recombinant E. coli and M. smegmatis retained significant amounts of HbO, indicating that the role of HbO may be linked to respiration. This presumption is also supported by the fact that recombinant E. coli and M. smegmatis, carrying HbO, exhibit higher respiratory activities; the addition of purified HbO to the membrane vesicles of control E. coli cells results in enhanced oxygen uptake. A candidate mechanism of HbO action may be that it interacts specifically with one or more components of the aerobic respiratory chain. A preliminary investigation into this possibility was carried out using E. coli strains in which one of the terminal oxidases, e.g. cytochrome o or cytochrome d, was nonfunctional. These results suggest that HbO action may be linked with one of the component of terminal oxidases, specifically cytochrome o. It is quite possible that the presence of HbO in the close proximity of respiratory membranes increases effective oxygen concentration inside the cell and facilitates oxygen transfer to the oxygen-binding sites of the terminal oxidases, which are oriented toward the cytoplasm (34), thus resulting in an increase in the number of protons extruded by the respiratory chain per oxygen molecule reduced. Subsequent entry of these extra protons into the cell via the ATPase increases the ATP production rate. The higher level of ATP pool within the HbO-carrying recombinant E. coli as compared with the control cells strongly supports this presumption.

Our analysis of the temporal expression pattern of HbO in M. bovis indicated the presence of HbO during all growth phases, as opposed to HbN, which appears specifically at the stationary phase of M. bovis (8). It is not known at present whether the cellular level of HbO changes during the intracellular regime of mycobacterium. Results presented in this work on the structural and biochemical features and the expression of mycobacterial HbO strongly support the role of HbO as a facilitator of oxygen to the terminal respiratory apparatus to increase the energetic status of the cells. However, other possible functions for this hemoglobin cannot be excluded and will require further exploration. Experimental data on any bacterial HbO-type trHb is not available at present. Truncated hemoglobin (GLB3), homologous to HbO, has been reported recently from Arabidopsis thaliana (35). GLB3 expression occurs in various parts of the plant and was found reduced under hypoxic condition, suggesting its role in aerobic metabolism and oxygen transfer within the plant cells. M. bovis also produces small amount of HbO throughout its growth cycle and may have a similar function(s).

Why should M. tuberculosis have hemoglobin for its aerobic growth? The following observations need consideration in order to speculate on the cellular function(s) of HbO. Although most mycobacterial species are obligate aerobe, their cell wall is highly complex, thick, and impermeable and thus likely to hinder the easy diffusion of oxygen. For pathogenic mycobacteria such as M. tuberculosis, M. bovis, and M. avium, etc. the presence of an oxygen-binding protein adjacent to their cellular membrane may prove beneficial in sustaining their aerobic metabolism and electron transfer during the intracellular regime, where these pathogenic bacteria encounter severe oxygen paucity. HbO binds oxygen reversibly and has oxygen affinity, which may allow it to release the bound oxygen readily and to generate sufficient local oxygen flux to enable M. tuberculosis to respire, survive, and adapt efficiently under low oxygen. Alternatively, HbO may provide, shortly after infection, a new electron pathway for the bacterium in the unfavorable environment of its host. Bacterial hemoglobin, VHb, has been suggested to function as an alternative terminal oxidase in the absence of conventional cytochrome oxidase in an E. coli mutant (36). One of the major contributing factors to the success of M. tuberculosis as a human pathogen is its ability to persist in the dormant state within the human host for decades, with subsequent reactivation later in life. Perhaps the presence of more than one Hb-like protein with different oxygen binding characteristics may enable the mycobacterium to cope with varying metabolic demands during different stages in its pathogenicity. HbO, being one of the components of most of the pathogenic bacteria, appears to play a pivotal role during the survival and adaptation of mycobacterium inside the host and may constitute an interesting target in understanding the physiology, adaptation, and pathogenicity of M. tuberculosis.

	ACKNOWLEDGEMENTS

We thank Prof. D. A. Webster and Sandhya Ahuja (Illinois Institute of Technology, Chicago) for providing help with spectral measurements. We are also grateful to Prof. Stefan Hoyle for making available the expression plasmid p19Kpro. Technical assistance provided by S. Muthukrishnan is gratefully acknowledged.

	FOOTNOTES

^* This work was supported by the Department of Science and Technology and the Council of Scientific and Industrial Research, India.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The on-line version of this article (available at http://www.jbc.org) contains supplementary Table 1, which has the NCBI and TIGR accession numbers of various truncated hemoglobins included in Fig. 1, and Table 2, which shows the relative percent similarity among various truncated hemoglobins.

Present address: National Bureau of Animal Genetic Resources, Karnal 132001, India.

^§ To whom correspondence should be addressed. Tel.: 91-172-695215; Fax: 91-172-690632/690585; E-mail: kanak@imtech.res.in.

Published, JBC Papers in Press, January 16, 2002, DOI 10.1074/jbc.M111478200

	ABBREVIATIONS

The abbreviations used are: trHb, truncated hemoglobin; BCG, bacillus Calmette-Guérin.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles: