Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Metabolic Engineering of Cofactor F420 Production in Mycobacterium smegmatis

  • Ghader Bashiri ,

    Affiliations Structural Biology Laboratory, School of Biological Sciences, The University of Auckland, Auckland, New Zealand, Maurice Wilkins Centre for Molecular Biodiscovery, School of Biological Sciences, The University of Auckland, Auckland, New Zealand

  • Aisyah M. Rehan,

    Affiliations Structural Biology Laboratory, School of Biological Sciences, The University of Auckland, Auckland, New Zealand, Maurice Wilkins Centre for Molecular Biodiscovery, School of Biological Sciences, The University of Auckland, Auckland, New Zealand

  • David R. Greenwood,

    Affiliations Maurice Wilkins Centre for Molecular Biodiscovery, School of Biological Sciences, The University of Auckland, Auckland, New Zealand, Centre for Genomics and Proteomics, School of Biological Sciences, The University of Auckland, Auckland, New Zealand

  • James M. J. Dickson,

    Affiliations Structural Biology Laboratory, School of Biological Sciences, The University of Auckland, Auckland, New Zealand, Maurice Wilkins Centre for Molecular Biodiscovery, School of Biological Sciences, The University of Auckland, Auckland, New Zealand

  • Edward N. Baker

    Affiliations Structural Biology Laboratory, School of Biological Sciences, The University of Auckland, Auckland, New Zealand, Maurice Wilkins Centre for Molecular Biodiscovery, School of Biological Sciences, The University of Auckland, Auckland, New Zealand

Metabolic Engineering of Cofactor F420 Production in Mycobacterium smegmatis

  • Ghader Bashiri, 
  • Aisyah M. Rehan, 
  • David R. Greenwood, 
  • James M. J. Dickson, 
  • Edward N. Baker


Cofactor F420 is a unique electron carrier in a number of microorganisms including Archaea and Mycobacteria. It has been shown that F420 has a direct and important role in archaeal energy metabolism whereas the role of F420 in mycobacterial metabolism has only begun to be uncovered in the last few years. It has been suggested that cofactor F420 has a role in the pathogenesis of M. tuberculosis, the causative agent of tuberculosis. In the absence of a commercial source for F420, M. smegmatis has previously been used to provide this cofactor for studies of the F420-dependent proteins from mycobacterial species. Three proteins have been shown to be involved in the F420 biosynthesis in Mycobacteria and three other proteins have been demonstrated to be involved in F420 metabolism. Here we report the over-expression of all of these proteins in M. smegmatis and testing of their importance for F420 production. The results indicate that co–expression of the F420 biosynthetic proteins can give rise to a much higher F420 production level. This was achieved by designing and preparing a new T7 promoter–based co-expression shuttle vector. A combination of co–expression of the F420 biosynthetic proteins and fine-tuning of the culture media has enabled us to achieve F420 production levels of up to 10 times higher compared with the wild type M. smegmatis strain. The high levels of the F420 produced in this study provide a suitable source of this cofactor for studies of F420-dependent proteins from other microorganisms and for possible biotechnological applications.


The cofactor F420 was first identified chemically in methanogenic Archaea in 1972 [1], although a compound with similar characteristics was previously described in Mycobacteria in the early 1960s [2], [3]. Since its discovery, F420 and its precursor FO (so called 5-deazaflavins) have been found in a variety of (micro)organisms, including Archaea, bacteria and eukaryotic species (Table 1). F420 is named on the basis of its intense absorbance/fluorescence at 420 nm (emission 480 nm), which is redox dependent and is lost upon reduction of the cofactor. It also has unique chemical and biological characteristics; the isoalloxazine chromophore of F420 is structurally very similar to that of the flavins (FMN and FAD), although it is functionally similar to NAD(P)+ (Figure 1). Functionally, F420 is a two–electron carrier involved in hydride transfer reactions. The redox potential of F420H2/F420+2e (−360 mV) is lower than those of the classical hydrogen carriers NAD(P)H/NAD(P)+2e (−320 mV) and FADH2/FAD+2e (−219 mV) [4], [5].

Figure 1. The molecular structures of F420, the flavins and NADP+.

(A) Schematic representation of F420 showing different parts of the molecule, whereas (B) and (C) show the molecular structures of the flavins and NADP+, respectively. The atoms involved in oxidoreductive reactions are numbered in all structures.

Table 1. Deazaflavin–dependent reactions in different (micro)organisms.

A key biosynthetic precursor of F420 is FO (7,8-didemethyl-8-hydroxy-5-deazariboflavin), comprising an isoalloxazine ring and ribitol moieties. Formation of F420 follows a series of biochemical reactions and is completed by the addition of a phospholactate group, and finally a poly–glutamate tail in which L–glutamate residues are linked together via γ–glutamyl bonds (Figure 2) [6], [7]. The length of the poly–glutamate tail constitutes the main difference between the F420 cofactors from different microorganisms, the number of residues varying from 2–9. There are suggestions, however, that the type of α– or γ–glutamyl linkage in the terminal glutamate residue could also be different in some Archaeal species [6], [8], [9], [10].

Figure 2. The proposed biosynthetic pathway for cofactor F420.

The FbiA, B and C are Mycobacterial proteins whereas the CofA, B, C, D, E, G and H are Archaeal proteins involved in the biosynthetic reactions. The pathway to formation of the activated phospholactate moiety (LPPG) is yet to be experimentally established in Mycobacteria. In some Archaeal species an α–linked terminal glutamate residue caps the γ–linked poly–glutamate tail, the addition of which is catalyzed by CofF.

F420 is not commercially available and researchers working on F420–dependent proteins have to prepare it as required. With the discovery of new F420–dependent enzymes and increasing interest in F420–dependent reactions, especially in the case of the pathogen Mycobacterium tuberculosis (Mtb), a resource with high yields of F420 production is required. F420 has been previously purified from various microorganisms, including Archaea (Methanobacterium, Methanococcus and Methanosarcina species) and Actinomycetes (Actinomadura, Actinoplanes, Streptomyces, Rhodococcus, Nocardia and Mycobacteria species), with differing yields [11]. F420 purification in all cases, however, essentially follows the same principle; precipitation of cellular proteins using heat or an organic solvent, followed by separation of F420 from remaining cellular components based on its acidic nature [4]. In order to purify F420, a number of different chromatographic steps have been used, including ion exchange, adsorption, HPLC and gel filtration chromatography [6], [9], [11]. Isabelle et al. have reported thorough analyses of F420–producing microorganisms, and based on “ease of growth, fewer hazards, and lower costs” concluded that M. smegmatis is the best source for F420 production, providing there is no requirement for a particular number of glutamate residues in the F420 poly–glutamate tail [11].

Our initial F420 purification trials indicated that M. smegmatis transformed to over–express the M. tuberculosis protein FGD1 (F420–dependent glucose-6-phosphate dehydrogenase 1) could produce higher levels of F420 compared with the wild type strains. This observation prompted us to thoroughly investigate the effects on F420 production of over–expression of other proteins known to be involved in F420 biosynthesis and metabolism in Mycobacteria. These include three proteins in the F420 biosynthetic pathway, viz. FbiA (Rv3261) [12], FbiB (Rv3262) [12] and FbiC (Rv1173) [13] and three other proteins which are shown to be involved in F420 metabolism: FGD1 (Rv0407) [9], [14], Ddn (Rv3547) [15] and Rv0132c (author's unpublished data).

Here we describe the development of vectors to co–express Mtb proteins in M. smegmatis. We further show that by co–expressing enzymes associated with F420 production and manipulating growth conditions, greatly increased levels of F420 can be obtained. With the growing recognition that F420 plays a crucial role in Mycobacteria and other organisms, this readily available source of the cofactor will be useful for testing its physiological and biochemical roles, and for possible applications in biotechnology.

Materials and Methods

Preparation of New Mycobacterial Vectors

The pYUB1049 vector (5795 bp) is a product of ligation between the vectors pMS134 and pET28b–cmaA2 [16], resulting in a vector with a cloned gene between NdeI and BamHI restriction sites. The pYUB1049 vector was subjected to restriction digestion using NcoI (single site) and BlpI (two sites) restriction sites, in order to obtain a linear vector without the multiple cloning site. The plasmid was first digested to completion with NcoI (Roche Applied Science) and dephosphorylated using calf intestinal alkaline phosphatase (New England Biolabs) followed by ethanol precipitation. The NcoI–cut linear pYUB1049 vector was subjected to a partial digestion with BlpI (BpuI102I isoschizomer, Fermentas) for 20 minutes and the reactions stopped using 5 µL 0.5 M EDTA. The digested vector was run on a 0.5% agarose gel and a DNA fragment corresponding to 4705 bp was excised and gel–purified.

The pET28b and pETDuet–1 vectors (Novagen) were double–digested using NcoI and BlpI enzymes. The resulting multiple cloning site fragments, 216 and 382 bp respectively, were ligated separately into the NcoI/BlpI fragment of the pYUB1049 vector using T4 DNA ligase (Roche Applied Science). Ligation mixtures were electroporated into E. coli TOP10 cells and the positive colonies were selected on low salt LB agar plates (tryptone 10 g/L, yeast extract 5 g/L, NaCl 5 g/L and agar 15 g/L, pH 8.0) containing 50 µg/mL hygromycin B. Positive clones were verified using restriction digestion and sequencing. The resulting vectors were designated as pYUB28b (4921 bp) and pYUBDuet (5087 bp), respectively.

PCR Amplification and Cloning

The open reading frames (ORFs) encoding Rv3261 (FbiA), Rv3262 (FbiB), Rv1173 (FbiC), Rv0407 (FGD1), Rv3547 (Ddn) and Rv0132c were amplified from M. tuberculosis H37Rv genomic DNA using Pwo, Pfx or PrimeStar polymerases with the primers outlined in Table 2. All constructs were cloned with either N– or C–terminal His6–tags. The amplified products for the FGD1 [17] and Rv0132c constructs were cloned using restriction/ligation cloning into the pYUB1049/pYUB28b vectors. The constructs were transformed into E. coli Top10 cells and plated on low salt LB agar medium supplemented with 50 µg/mL hygromycin B to select for colonies harbouring the plasmid. Positive clones were verified using colony PCR, restriction digestion and sequencing.

Table 2. Oligonucleotide primers used in the amplification of the protein coding sequences in this study.

All other ORFs were cloned using the Gateway® cloning system into the pDESTsmg vector [18]. The Gateway® cloning system uses a nested PCR method involving two rounds of amplification in which the second round uses the product of the first round as template. Gene–specific primers are used in the first round PCR to amplify the gene of interest and generic primers are used in the second round amplification to incorporate the required recombination sites for subsequent cloning. The PCR products were cloned by recombination into pDONR221 (Invitrogen) using BP Clonase™ (Invitrogen), to generate the entry clones. The constructs were transformed into E. coli Top10 cells and plated on LB agar medium containing 50 µg/mL of kanamycin. Positive clones were verified using BsrGI digestion and sequencing. These positive entry clones were recombined in vitro with pDESTsmg, in an LR reaction using LR Clonase™ (Invitrogen), to generate a M. smegmatis expression construct. Following transformation of recombinant pDESTsmg plasmids, positive clones were selected on low salt LB agar plates supplemented with 50 µg/mL hygromycin B and were verified using BsrGI digestion.

The pYUBDuet vector was used to clone the F420 biosynthetic ORFs (FbiAB and FbiC) together using restriction/ligation cloning. Both FbiC and FbiAB ORFs were amplified using PfuUltra Fusion HS DNA polymerase (Stratagene) using the primers outlined in Table 2. FbiC was first cloned using NcoI/HindIII restriction sites and the FbiAB operon was subsequently cloned using NdeI/EcoRV restriction sites.

Expression in M. smegmatis

All expression constructs were electroporated individually into the M. smegmatis strain mc24517. Preparation of electrocompetent cells and electroporation procedures were performed following published protocols [19]. Briefly, M. smegmatis mc24517 cells were grown at 37°C in 7H9/ADC/Tween80 or LB/Tween80 containing 50 µg/mL kanamycin until an OD600 ∼0.7. Cells were harvested and washed three times in 10% ice–cold glycerol and finally resuspended in 10% ice–cold glycerol. Single aliquots of the resulting competent cells (40 µL) were transformed with 1 µL of DNA and a further 260 µL of 10% glycerol in 0.2 cm cuvettes. Electroporation was performed using a Bio–Rad Gene Pulser set to the following parameters: R = 1000 Ω, Q = 25 µF and V = 2.5 kV. Cells were immediately harvested with 1 mL 7H9/ADC/Tween80 (Difco™ and BBL™ Middlebrook) or LB/Tween80 and incubated for 3 h at 37°C with shaking. Positive transformants were selected by plating on 7H10/ADC (Difco™ and BBL™ Middlebrook) or LBT agar plates containing 50 µg/mL each of kanamycin and hygromycin B.

Protein expression was performed either in autoinduction [20], LB or 7H9/ADC media supplemented with 0.05% Tween80 and 50 µg/mL each of kanamycin and hygromycin B. A single transformed colony was selected from a 7H10/ADC plate and used to inoculate a starter culture in MDG media (25 mM Na2HPO4, 25 mM KH2PO4, 50 mM NH4Cl, 5 mM Na2SO4, 2 mM MgSO4, 0.5% D-glucose, 0.25% L-aspartate, 0.2× metal mix) [20]. The starter culture was grown for 48–72 h at 37°C and was freshly used at a dilution of 1∶100 to inoculate expression cultures of ZYM–5052 autoinduction (1% tryptone, 0.5% yeast extract, 25 mM Na2HPO4, 25 mM KH2PO4, 50 mM NH4Cl, 5 mM Na2SO4, 2 mM MgSO4, 0.5% glycerol, 0.05% glucose, 0.2% alpha–lactose, 1× metal mix), LB or 7H9/ADC. The expression cultures were grown for 4 days at 37°C for maximal expression [17]. LB, MDG and 7H9/ADC cultures were induced using IPTG at a final concentration of 0.1 or 1 mM.

Western Blot Analyses

M. smegmatis cells expressing different constructs were lysed twice using a cell disruptor (Constant Systems Ltd.) and centrifuged at 16,000×g to pellet non–lysed cells and other insoluble material. Protein samples were separated on a 15% SDS–PAGE gel and transferred to polyvinylidene difluoride (PVDF) membranes using a wet transfer protocol (200mA, 3 hours) [21]. His–tagged recombinant proteins were detected using a mouse monoclonal anti–His antibody and horseradish peroxidase–conjugated anti–mouse antibody (GE Healthcare). The Luminol (ECL plus kit, GE Healthcare) chemiluminescence was detected using an LAS4000 imaging system (Fujifilm).

FO and F420 Characterization

M. smegmatis cells expressing different M. tuberculosis proteins were grown in identical conditions to late log phase or stationary phase. In all expression cultures the ZYP–5052 autoinduction media was used for F420 production experiments and the media to flask volume ratio was kept constant at 20%. In order to optimize the media for F420 production, the ZY component of ZYM–5052 media was replaced by commonly used media bases including 2× ZY, YT (0.8% tryptone, 0.5% yeast extract and 42.77 mM NaCl), TB (1.2% tryptone, 2.4% yeast extract and 0.4% glycerol), SOB (2% tryptone, 0.5% yeast extract, 8.56 mM NaCl, 2.5 mM KCl and 10 mM MgCl2) and SOC (SOB with 20 mM glucose). Iron and sulphur supplements (ferric ammonium citrate, ferric citrate and ferrous sulphate all at 0.1 mg/mL and L–cysteine at 1 mM) were also added to the expression media as a possible requirement for the FbiC enzyme. L–glutamate and manganese chloride (1 mM final concentration) were also added to the expression media to evaluate their necessity for FbiB–mediated F420 production [22].

To ascertain the optimum growth period for F420 production, eight identical cultures of M. smegmatis cells expressing the recombinant FbiABC construct were set up. Each culture had a wild type M. smegmatis culture as a control. At 24 h intervals, one culture each of control and recombinant FbiABC–expressing M. smegmatis cells were harvested and processed to monitor the F420 production level. The procedure was carried out for eight days and the F420 production ratio for each day was calculated by dividing the F420 fluorescence from FbiABC–expressing cells by fluorescence of the wild type control.

M. smegmatis cells were centrifuged for 15 min at 16000×g and the resulting media were used for FO characterization. The cell pellets were washed with 25 mM sodium phosphate buffer, pH 7.0 and were subsequently resuspended in 1 mL of the same buffer per 100 mg of cells (wet weight). The cell suspensions were autoclaved at 121°C for 15 min to break the cells open and were then centrifuged for 15 min at 16000×g. Fluorescence of the media and the extract were monitored using excitation wavelength of 420 nm (405±10 nm filter) and emission wavelength of 480 nm (485±15 nm filter). All fluorescence experiments were performed using an EnVision Multilabel plate reader (Perkin Elmer) in a 96–well plate format and were carried out in triplicate.

The autoclaved cell extracts were further purified using a HiTrap QFF ion exchange column (GE Healthcare) to separate the intracellular FO from the F420. The extract was run on the column pre–equilibrated with 25 mM sodium phosphate buffer, pH 7.0 and was subsequently washed with five column volumes of buffer. Two yellow fractions were eluted at 200 and 500 mM NaCl, respectively. The purified fractions were used for mass spectrometry analysis, together with the media from the previous step. The media (1 mL) was treated with an equal volume of cold acetone to precipitate the protein and the solution was then evaporated down to <0.5 mL to drive off the acetone. A mix of water and 5% aqueous methanol with 0.1% formic acid was added to bring the final concentration of methanol to less than 1% (total volume 4 mL). All samples were then applied to a pre–equilibrated Alltech Maxi–Clean 300 mg large pore 100Å C–18 SPE cartridge and washed with 4 mL 5% methanol containing 0.1% formic acid followed by 4 mL 10% methanol. Compounds were eluted with 4 mL 80% methanol containing 5 mM ammonium bicarbonate pH 8.5. Eluates were evaporated under nitrogen and redissolved in 80% methanol and 20 mM ammonium acetate ready for mass spectrometry. Samples were infused at 3 µL/min under negative electrospray conditions into an LTQ–FT mass spectrometer (Thermo Scientific). The ion intensity data were obtained using a source voltage of 2.5 kV and capillary temperature of 225°C. Ions were examined in both the ion trap and ion cyclotron resonance cells, the latter to obtain high resolution (100,000 at m/z 400) accurate mass data. This was necessary to confirm the atomic composition of the ions and help deconvolute the contribution of metal ion adducts (Na+/K+) to the levels of individual poly–glutamate species. Up to four sodium ions were adducted to produce some double charged negative ions.


New Mycobacterial Expression Vectors

The pYUB1049 vector does not provide an intact multiple cloning site and does not support C–terminal His–tag expression. In order to overcome these obstacles, the pYUB1049 vector was subjected to a restriction digestion using NcoI and BlpI enzymes and a linear fragment lacking the multiple cloning site was obtained. The resulting fragment was used as a backbone that could be ligated to the intact multiple cloning site from the pET28b or pETDuet–1 vectors to produce the pYUB28b and pYUBDuet vectors, respectively. Figure 3 provides a schematic representation of the vectors with the list of unique restriction sites that can be used for cloning.

Figure 3. A schematic representation of the vectors designed in this study.

The unique restriction sites in the multiple cloning sites of each vector are indicated.

PCR Amplification and Cloning

Six different ORFs which are believed to be involved in F420 biosynthesis (FbiA, FbiB and FbiC) or F420 metabolism (FGD1, Ddn and Rv0132c) were amplified and cloned for expression in M. smegmatis as His–tagged proteins. Assuming that FbiA and FbiB ORFs are transcribed as a single operon, we investigated the possibility of cloning and co–expression of the whole F420 biosynthetic pathway (FbiAB and FbiC) in order to boost F420 production yield. The pYUBDuet co–expression vector was designed, prepared and subsequently used to clone FbiC and FbiAB ORFs, making it possible to express three different proteins from a single vector. All three proteins were expressed in their native form without His–tags.

Expression of Proteins in M. smegmatis

The six F420 biosynthetic or metabolic ORFs cloned into pYUB1049/pYUB28b/pDESTsmg vectors were expressed in M. smegmatis as individual proteins. Each of these proteins were cloned with either N– or C–terminal His–tags, making it possible to detect the protein expression using monoclonal anti–His antibodies. The western blotting experiments indicated that all proteins were expressed in M. smegmatis cells, as shown by appearance of correct–sized bands for the appropriate proteins (data not shown).

The expression of proteins from the pYUBDuet vector could not be detected using western blotting, as they did not contain any tags; however, their successful expression could be inferred from FO and F420 production as discussed later.

Cofactor F420 Production

Individual M. smegmatis cultures harbouring six different constructs (FbiA, FbiB, FbiC, FGD1, Ddn and Rv0132c) were grown in order to find out the over–expression effect of these targets on F420 production. Three different media were initially used to express the proteins; LBT with IPTG induction, MDG with no or low induction using IPTG, and ZYM–5052 autoinduction media. Based on growth rate and cell mass, ZYM–5052 media was selected as the best media and was used to continue F420 production experiments. The fluorescence signals of the expression media and the cell extracts were monitored at 420 nm, enabling the detection of both FO and F420. It has been reported that FO comprises 1–7% of the total intracellular deazaflavin in Mycobacteria [8]; we used fluorescence at 420 nm to evaluate the F420 contents of the cellular extracts without taking into account the small portion of the fluorescence signal coming from FO.

The experimental results indicate that FGD1 over–expression increases F420 production by almost two–fold compared to the wild type strain (Figure 4, A). Cells expressing other Mtb proteins did not show a significant increase in F420 yield, however. Cells expressing the FbiC construct (pDEST–FbiC) showed a strong blue–green colour in the media. This is presumably due to the presence of fluorescent FO in the media which diffuses out of the cells as FO does not have any charge on the molecule to cause retention inside the cell (Figure 4, A) [23]. Mass spectrometry confirmed that FO was indeed responsible for the distinct fluorescence of the media (m/z 362.09870 [M-H]; C16H16N3O7 requires 362.09882). This observation could be explained by over–expression of the FbiC protein leading to higher FO synthesis. Because the cells could not convert the over–produced FO to F420, the excess was presumably lost from the cells, either by diffusion or by active export.

Figure 4. FO/F420 production by M. smegmatis cells expressing different recombinant proteins.

(A) comparative FO/F420 production between wild type strain and six different recombinant strains expressing proteins involved in F420 biosynthesis or metabolism. (B) FO/F420 production using co–expression vector compared to the wild type strain. (C) F420 production ratio by M. smegmatis cells expressing FbiABC construct over wild type strains for eight days. (D) Effect of iron/sulphur and L–glutamate/manganese additives on the F420 production (FO is shown as black and F420 as grey). In panels (A), (B) and (D) the error bars are derived from experiments carried out in triplicate.

This observation provided the motivation for us to co–express the FbiAB operon together with FbiC, hoping that over–expressed FbiA and FbiB proteins would be able to convert the synthesised FO into F420 inside the cells. The pYUBDuet vector was used to clone FbiABC ORFs together; FbiC was first cloned, resulting in the pYUBDuet–FbiC construct, after which FbiAB was introduced to obtain pYUBDuet–FbiABC. Both these constructs were used to investigate the effect on FO/F420 production (Figure 4, B). Cells expressing FbiC alone (pYUBDuet–FbiC) consistently showed more than 10–fold higher FO levels in the expression media compared to the wild type strains. It is an interesting observation that FO production by the pYUBDuet–FbiC construct is much higher (>50%) than by the pDESTsmg–FbiC construct, with the former expressing FbiC as the native protein whereas the latter has an N–terminal His–tag. In contrast, F420 production from pYUBDuet–FbiC was not significantly elevated compared to wild type. By expressing the FbiAB operon together with FbiC (pYUBDuet–FbiABC), however, F420 production was consistently more than five times higher inside the cells (Figure 4, B). These results clearly indicate that the cells express functional recombinant proteins resulting in much higher intracellular F420 levels.

M. smegmatis cells expressing the pYUBDuet–FbiABC construct were then used to find out the optimum time period for F420 production. The F420 production was monitored for eight days using ZYM–5052 media and the F420 production ratio was calculated and plotted versus the day of culture. The results indicated that the F420 levels were the highest on day four of the culture, after which the levels gradually decreased. Based on this result, the best time to harvest the cells for F420 purification is 4–5 days after setting up the expression culture (Figure 4, C). Subsequently, a set of experiments was performed to find out the best media formulation to grow the cells for F420 production using an autoinduction protocol. ZY produced the highest F420 yield among ZY, YT, TB, SOB and SOC media. Bioinformatic analysis has indicated that FbiC is a protein with possible Fe–S clusters. In addition the reaction catalyzed by an archaeal homologue of FbiB requires L–glutamate and manganese chloride [22]. The expression media were therefore also supplemented with iron/sulphur and L–glutamate/manganese additives. The results indicated that supplementation of the expression media with either of these additives does indeed increase the F420 production yield (Figure 4, D). Surprisingly, cultures with an L–glutamate/manganese supplement did not have extra FO in the media, implying that the cells could convert all the produced FO into F420 inside the cells (Figure 4, D). It seems that the limiting factor in producing F420 from over–produced FO was the supply of the required L–glutamate/manganese.

The FO/F420 produced by the cells expressing the FbiABC construct was purified and analysed using mass spectrometry. The results show two predominant fractions; a 200 mM NaCl fraction mainly composed of FO and a 500 mM NaCl fraction of exclusively F420 with more than 95% being F420-6 and F420-7 species (Figure 5). This result is in line with the previously published results of F420 extracted from the wild type M. smegmatis cells having the major species of F420-5 to F420-7 [9], [11], implying that the over–expression of the FbiABC construct does not change the F420 production profile.

Figure 5. The F420 production profile from M. smegmatis cells over–expressing the FbiABC construct.

The F420–6 and F420–7 species constitute the main species as deduced using mass spectrometry.


The cofactor F420 has an important role in the metabolism of Archaea and has been the subject of numerous studies over the years since its identification. It is now clear that this importance applies also to Mycobacteria, for which there is growing evidence that F420 plays a key role in defence against oxidative and nitrosative stress [5], [24]. Consistent with this, the number of identified F420–dependent enzymes from Mycobacteria is growing, with nine new examples recently described [25]. A recent partial phylogenetic profiling study has proposed that there are at least 28 separate F420–dependent enzymes in M. tuberculosis, suggesting that F420 has a pivotal role in redox reactions of this pathogenic mycobacterium [26]. Few of these enzymes have been characterised, however, and research into their functions, and the role of F420, are handicapped by the fact that there is no commercial source for this cofactor, which can only be obtained in relatively low yield from the wild type M. smegmatis strain. A major aim of this study was to increase the F420 production yield in M. smegmatis by cloning and expression of the genes involved in F420 production and metabolism.

Mycobacterial Expression Vectors

The pYUB1049 plasmid is a T7 promoter–based vector for which expression can be induced by IPTG or autoinduction. This vector has previously been used as a shuttle vector for cloning of Mycobacterial genes into E. coli and subsequent expression of proteins in M. smegmatis [17], [27]. The pYUB1049 vector has been also converted to a Gateway® cloning system compatible vector, pDESTsmg [18]. In this study, two different vectors were designed and prepared from the parental pYUB1049 vector; the pYUB28b vector is used for restriction/ligation cloning of single genes with the capability of expressing N– and C–terminal His–tags, whereas the pYUBDuet vector is a co–expression vector for simultaneous expression of two genes in a Mycobacterial host. Our experimental results demonstrate the application of T7–promoter based co–expression vectors in M. smegmatis that could also be useful in other contexts. Although there have been previous reports of co–expression systems for Mycobacteria [28], [29], [30], [31], [32], [33], to the best of our knowledge, this is the first T7–promoter based co–expression vector for a Mycobacterial host. The pYUB28b and pYUBDuet vectors, together with the pDESTsmg vector which has been previously developed in the authors' lab for the Gateway® cloning system, represent a repertoire of T7 promoter–based vectors which can be routinely used for expression of a wide range of ORFs in a Mycobacterial host.

F420 Production

FbiC is annotated as FO synthase [34], catalysing the transfer of the hydroxybenzyl group from 4-hydroxyphenylpyruvate (a tyrosine precursor) to 5-amino-6-ribitylamino-2,4(1H,3H)-pyrimidinedione (an intermediate in flavin biosynthesis) to form FO (Figure 2) [23]. FO is the first intermediate with a complete deazaflavin chromophore in the F420 biosynthesis pathway [13], [23] providing the rationale for believing that this reaction might be the rate limiting step in F420 biosynthesis. FbiA and FbiB are believed to be involved in production of F420 from the precursor FO molecule; FbiA in generating F420–0 from FO and FbiB in adding glutamate residues to F420–0 to produce F420 with a poly–glutamate tail of variable length (Figure 2) [12]. In all Mycobacterial species with the genome sequences completed to date (21 in total, as of October 2010) FbiA is located immediately upstream of FbiB ( A detailed analysis indicates that the start site for the FbiB ORF overlaps with the last four base pairs of the FbiA ORF, though in a different reading frame, implying that they might be transcribed as a single operon for expression. In fact, it has been shown in M. bovis that these two ORFs are transcribed together as a single mRNA species [12]. This genetic arrangement made it possible to co–express the FbiAB operon and FbiC gene together, using the pYUBDuet vector, in order to increase F420 yield. Based on our results, the optimum condition to produce F420 by M. smegmatis cells expressing recombinant FbiABC is a culture with autoinduction media using ZY base over 4–5 days supplemented with iron, sulphur, L–glutamate and manganese. Using these optimal conditions, the F420 production yield was up to 10–times higher compared with the wild type strains.

The main limiting factor in F420 production, based on our results on over–expression of the three enzymes FbiA, FbiB and FbiC from the F420 biosynthetic pathway, appears to be the availability of the FbiB reaction substrate/cofactor. It does not seem that the FbiC reaction is the limiting step of the pathway even when the media are not supplemented with L–glutamate/manganese; excess FO was always present in high quantities in the media, indicating that the over–expressed FbiA and FbiB proteins are still not capable of converting all FO to F420. An alternative possibility is that FbiA and FbiB need other accessory protein(s) in order to perform the conversion more efficiently; in fact another ORF in M. smegmatis (MSMEG_2392) has been shown, by transposon mutagenesis studies, to be involved in F420 biosynthesis from FO [35]. Biochemical studies need to be performed using purified enzymes in order to study the kinetics in detail and determine the rate limiting step of the pathway.

Our previous crystal structures of M. tuberculosis FGD1 [9], together with other F420–containing crystal structures from different Archaeal species [36], [37], [38], [39], have indicated that the F420 poly–glutamate tail is not required for reaction catalysis; the poly–glutamate tail is extended into the solvent and it seems that this is a conserved feature of the enzymes that use F420 in oxidoreduction reactions. We propose that, therefore, the high yields of F420 from M. smegmatis strains expressing the recombinant FbiABC proteins, regardless of the number of glutamate residues in the poly–glutamate tail, identify this as a valuable source of F420 that might be used with enzymes purified from other microorganisms. Furthermore, the high yield of FO/F420 also opens a door for possible biotechnological applications.


The nucleotide sequences for pYUB28b and pYUBDuet vectors have been deposited in the National Centre for Biotechnology Information (NCBI) under GenBank HQ247814 and HQ247815 accession numbers, respectively. The vectors are available upon request.


The pYUB1049 vector and M. smegmatis mc24517 were kindly provided by Professor W. R. Jacobs, Albert Einstein College of Medicine.

Author Contributions

Conceived and designed the experiments: GB. Performed the experiments: GB AMR DRG. Analyzed the data: GB DRG. Contributed reagents/materials/analysis tools: JMJD. Wrote the paper: GB JMJD ENB. Acquisition of funding/supervision: ENB.


  1. 1. Cheeseman P, Toms-Wood A, Wolfe RS (1972) Isolation and Properties of a Fluorescent Compound, Factor 420, from Methanobacterium Strain M.o.H. J Bacteriol 112: 527–531.
  2. 2. Cousins FB (1960) The prosthetic group of a chromoprotin from mycobacteria. Biochim Biophys Acta 40: 532–534.
  3. 3. Sutton WB (1964) Properties of a new TPN-like electron transport component from Mycobacterium phlei. Biochem Biophys Res Commun 15: 414–419.
  4. 4. DiMarco AA, Bobik TA, Wolfe RS (1990) Unusual coenzymes of methanogenesis. Annu Rev Biochem 59: 355–394.
  5. 5. Purwantini E, Mukhopadhyay B (2009) Conversion of NO2 to NO by reduced coenzyme F420 protects mycobacteria from nitrosative damage. Proc Natl Acad Sci U S A 106: 6333–6338.
  6. 6. Eirich LD, Vogels GD, Wolfe RS (1978) Proposed structure for coenzyme F420 from Methanobacterium. Biochemistry 17: 4583–4593.
  7. 7. Eirich LD, Vogels GD, Wolfe RS (1979) Distribution of coenzyme F420 and properties of its hydrolytic fragments. J Bacteriol 40: 20–27.
  8. 8. Bair TB, Isabelle DW, Daniels L (2001) Structures of coenzyme F(420) in Mycobacterium species. Arch Microbiol 176: 37–43.
  9. 9. Bashiri G, Squire CJ, Moreland NJ, Baker EN (2008) Crystal structures of F420-dependent glucose-6-phosphate dehydrogenase FGD1 involved in the activation of the anti-tuberculosis drug candidate PA-824 reveal the basis of coenzyme and substrate binding. J Biol Chem 283: 17531–17541.
  10. 10. Graupner M, White RH (2003) Methanococcus jannaschii Coenzyme F420 Analogs Contain a Terminal α-Linked Glutamate. J Bacteriol 185: 662–4665.
  11. 11. Isabelle D, Simpson DR, Daniels L (2002) Large-scale production of coenzyme F420-5,6 by using Mycobacterium smegmatis. Appl Environ Microbiol 68: 5750–5755.
  12. 12. Choi KP, Bair TB, Bae YM, Daniels L (2001) Use of transposon Tn5367 mutagenesis and a nitroimidazopyran-based selection system to demonstrate a requirement for fbiA and fbiB in coenzyme F(420) biosynthesis by Mycobacterium bovis BCG. J Bacteriol 183: 7058–7066.
  13. 13. Choi K-P, Kendrick N, Daniels L (2002) Demonstration that fbiC is required by Mycobacterium bovis BCG for coenzyme F420 and FO biosynthesis. J Bacteriol 184: 2420–2428.
  14. 14. Purwantini E, Daniels L (1996) Purification of a novel coenzyme F420-dependent glucose-6-phosphate dehydrogenase from Mycobacterium smegmatis. J Bacteriol 178: 2861–2866.
  15. 15. Singh R, Manjunatha U, Boshoff HI, Ha YH, Niyomrattanakit P, et al. (2008) PA-824 kills nonreplicating Mycobacterium tuberculosis by intracellular NO release. Science 322: 1392–1395.
  16. 16. Huang CC, Smith CV, Glickman MS, Jacobs WR Jr, Sacchettini JC (2002) Crystal structures of mycolic acid cyclopropane synthases from Mycobacterium tuberculosis. J Biol Chem 277: 11559–11569.
  17. 17. Bashiri G, Squire CJ, Baker EN, Moreland NJ (2007) Expression, purification and crystallization of native and selenomethionine labeled Mycobacterium tuberculosis FGD1 (Rv0407) using a Mycobacterium smegmatis expression system. Protein Expr Purif 54: 38–44.
  18. 18. Goldstone RM, Moreland NJ, Bashiri G, Baker EN, Shaun Lott J (2008) A new Gateway vector and expression protocol for fast and efficient recombinant protein expression in Mycobacterium smegmatis. Protein Expr Purif 57: 81–87.
  19. 19. Cirillo JD, Weisbrod TR, William R, Jacobs J (1993) Efficient electrotransformation of Mycobacterium smegmatis. Richmond, California: Bio-Rad Laboratories.
  20. 20. Studier FW (2005) Protein production by auto-induction in high-density shaking cultures. Protein Expr Purif 41: 207–234.
  21. 21. 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 U S A 76: 4350–4354.
  22. 22. Nocek B, Evdokimova E, Proudfoot M, Kudritska M, Grochowski LL, et al. (2007) Structure of an amide bond forming F(420):gamma-glutamyl ligase from Archaeoglobus fulgidus – a member of a new family of non-ribosomal peptide synthases. J Mol Biol 372: 456–469.
  23. 23. Graham DE, Xu H, White RH (2003) Identification of the 7,8-didemethyl-8-hydroxy-5-deazariboflavin synthase required for coenzyme F(420) biosynthesis. Arch Microbiol 180: 455–464.
  24. 24. Darwin KH, Ehrt S, Gutierrez-Ramos J-C, Weich N, Nathan CF (2003) The proteasome of Mycobacterium tuberculosis is required for resistance to nitric oxide. Science 302: 1963–1966.
  25. 25. Taylor MC, Jackson CJ, Tattersall DB, French N, Peat TS, et al. (2010) Identification and characterization of two families of F420H2-dependent reductases from Mycobacteria that catalyze aflatoxin degradation. Mol Microbiol 78: 561–575.
  26. 26. Selengut JD, Haft DH (2010) Unexpected Abundance of Coenzyme F420-dependent enzymes in the Genomes of Mycobacterium tuberculosis and other Actinobacteria. J Bacteriol 192: 5788–5798.
  27. 27. Robson J, McKenzie JL, Cursons R, Cook GM, Arcus VL (2009) The vapBC operon from Mycobacterium smegmatis is an autoregulated toxin-antitoxin module that controls growth via inhibition of translation. J Mol Biol 390: 353–367.
  28. 28. Chang Y, Mead D, Dhodda V, Brumm P, Fox BG (2009) One-plasmid tunable coexpression for mycobacterial protein-protein interaction studies. Protein Sci 18: 2316–2325.
  29. 29. George KM, Yuan Y, Sherman DR, Barry CE 3rd (1995) The biosynthesis of cyclopropanated mycolic acids in Mycobacterium tuberculosis. Identification and functional analysis of CMAS-2. J Biol Chem 270: 27292–27298.
  30. 30. Harth G, Maslesa-Galic S, Horwitz MA (2004) A two-plasmid system for stable, selective-pressure-independent expression of multiple extracellular proteins in mycobacteria. Microbiology 150: 2143–2151.
  31. 31. Kaps I, Ehrt S, Seeber S, Schnappinger D, Martin C, et al. (2001) Energy transfer between fluorescent proteins using a co-expression system in Mycobacterium smegmatis. Gene 278: 115–124.
  32. 32. Luo Y, Chen X, Szilvasi A, O'Donnell MA (2000) Co-expression of interleukin-2 and green fluorescent protein reporter in mycobacteria: in vivo application for monitoring antimycobacterial immunity. Mol Immunol 37: 527–536.
  33. 33. Slayden RA, Lee RE, Barry CE 3rd (2000) Isoniazid affects multiple components of the type II fatty acid synthase system of Mycobacterium tuberculosis. Mol Microbiol 38: 514–525.
  34. 34. Cole ST, Brosch R, Parkhill J, Garnier T, Churcher C, et al. (1998) Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature 393: 537–544.
  35. 35. Guerra-Lopez D, Daniels L, Rawat M (2007) Mycobacterium smegmatis mc2 155 fbiC and MSMEG_2392 are involved in triphenylmethane dye decolorization and coenzyme F420 biosynthesis. Microbiology 153: 2724–2732.
  36. 36. Aufhammer SW, Warkentin E, Berk H, Shima S, Thauer RK, et al. (2004) Coenzyme binding in F420-dependent secondary alcohol dehydrogenase, a member of the bacterial luciferase family. Structure 12: 361–370.
  37. 37. Aufhammer SW, Warkentin E, Ermler U, Hagemeier CH, Thauer RK, et al. (2005) Crystal structure of methylenetetrahydromethanopterin reductase (Mer) in complex with coenzyme F420: Architecture of the F420/FMN binding site of enzymes within the nonprolyl cis-peptide containing bacterial luciferase family. Protein Sci 14: 1840–1849.
  38. 38. Ceh K, Demmer U, Warkentin E, Moll J, Thauer RK, et al. (2009) Structural basis of the hydride transfer mechanism in F(420)-dependent methylenetetrahydromethanopterin dehydrogenase. Biochemistry 48: 10098–10105.
  39. 39. Warkentin E, Mamat B, Sordel-Klippert M, Wicke M, Thauer RK, et al. (2001) Structures of F420H2:NADP+ oxidoreductase with and without its substrates bound. EMBO J 20: 6561–6569.
  40. 40. Jacobson FS, Daniels L, Fox JA, Walsh CT, Orme-Johnson WH (1982) Purification and properties of an 8-hydroxy-5-deazaflavin-reducing hydrogenase from Methanobacterium thermoautotrophicum. J Biol Chem 257: 3385–3388.
  41. 41. Deppenmeier U, Blaut M, Mahlmann A, Gottschalk G (1990) Membrane-bound F420H2-dependent heterodisulfide reductase in methanogenic bacterium strain Göl and Methanolobus tindarius. FEBS Lett 261: 199–203.
  42. 42. Seedorf H, Dreisbach A, Hedderich R, Shima S, Thauer RK (2004) F420H2 oxidase (FprA) from Methanobrevibacter arboriphilus, a coenzyme F420-dependent enzyme involved in O2 detoxification. Arch Microbiol 182: 126–137.
  43. 43. Tzeng SF, Wolfe RS, Bryant MP (1975) Factor 420-dependent pyridine nucleotide-linked hydrogenase system of Methanobacterium ruminantium. J Bacteriol 121: 184–191.
  44. 44. Deppenmeier U, Blaut M, Mahlmann A, Gottschalk G (1990) Reduced coenzyme F420: heterodisulfide oxidoreductase, a proton- translocating redox system in methanogenic bacteria. Proc Natl Acad Sci U S A 87: 9449–9453.
  45. 45. Kunow J, Linder D, Stetter KO, Thauer RK (1994) F420H2: quinone oxidoreductase from Archaeoglobus fulgidus. Characterization of a membrane-bound multisubunit complex containing FAD and iron-sulfur clusters. Eur J Biochem 223: 503–511.
  46. 46. Tzeng SF, Bryant MP, Wolfe RS (1975) Factor 420-dependent pyridine nucleotide-linked formate metabolism of Methanobacterium ruminantium. J Bacteriol 121: 192–196.
  47. 47. Zeikus JG, Fuchs G, Kenealy W, Thauer RK (1977) Oxidoreductases involved in cell carbon synthesis of Methanobacterium thermoautotrophicum. J Bacteriol 132: 604–613.
  48. 48. Fuchs G, Stupperich E (1982) Autotrophic CO2 fixation pathway in Methanobacterium thermoautotrophicum. Zentralbl Bakteriol Hyg Abt 1 Orig C 3: 277–288.
  49. 49. Hartzell PL, Zvilius G, Escalante-Semerena JC, Donnelly MI (1985) Coenzyme F420 dependence of the methylenetetrahydromethanopterin dehydrogenase of Methanobacterium thermoautotrophicum. Biochem Biophys Res Commun 133: 884–890.
  50. 50. Ma K, Thauer RK (1990) Purification and properties of N5, N10-methylene-tetrahydromethanopterin reductase from Methanobacterium thermoautotrophicum (strain Marburg). Eur J Biochem 191: 187–193.
  51. 51. Widdel F, Wolfe RS (1989) Expression of secondary alcohol dehydrogenase in methanogenic bacteria and purification of the F420-specific enzyme from Methanogenium thermophilum strain TCI. Arch Microbiol 152: 322–328.
  52. 52. Johnson EF, Mukhopadhyay B (2005) A new type of sulfite reductase, a novel coenzyme F420-dependent enzyme, from the methanarchaeon Methanocaldococcus jannaschii. J Biol Chem 280: 38776–38786.
  53. 53. Vermeij P, Detmers FJ, Broers FJ, Keltjens JT, Van der Drift C (1994) Purification and characterization of coenzyme F390 synthetase from Methanobacterium thermoautotrophicum (strain delta H). Eur J Biochem 226: 185–191.
  54. 54. Vermeij P, Vinke E, Keltjens JT, Van der Drift C (1995) Purification and properties of coenzyme F390 hydrolase from Methanobacterium thermoautotrophicum (strain Marburg). Eur J Biochem 234: 592–597.
  55. 55. Coats JH, Li GP, Kuo MS, Yurek DA (1989) Discovery, production, and biological assay of an unusual flavenoid cofactor involved in lincomycin biosynthesis. J Antibiot (Tokyo) 42: 472–474.
  56. 56. McCormick JRD, Morton GO (1982) Identity of cosynthetic factor 1 of Streptomyces aureofaciens and fragment FO from coenzyme F420 of Methanobacterium sp. J Am Chem Soc 104: 4014–4015.
  57. 57. Rhodes PM, Winskill N, Friend EJ, Warren M (1981) Biochemical and genetic comparison of Streptomyces rimosus mutants impaired in oxytetracycline biosynthesis. J Gen Microbiol 124: 329–338.
  58. 58. Ebert S, Rieger PG, Knackmuss HJ (1999) Function of coenzyme F420 in aerobic catabolism of 2,4, 6-trinitrophenol and 2,4-dinitrophenol by Nocardioides simplex FJ2-1A. J Bacteriol 181: 2669–2674.
  59. 59. Eker AP, Hessels JKC, Velde Jvd (1988) Photoreactivating enzyme from the green alga Scenedesmus acutus. Evidence for the presence of two different flavin chromophores. Biochemistry 27: 1758–1765.
  60. 60. Eker AP, Kooiman P, Hessels JK, Yasui A (1990) DNA photoreactivating enzyme from the cyanobacterium Anacystis nidulans. J Biol Chem 265: 8009–8015.
  61. 61. Glas AF, Maul MJ, Cryle M, Barends TR, Schneider S, et al. (2009) The archaeal cofactor F0 is a light-harvesting antenna chromophore in eukaryotes. Proc Natl Acad Sci U S A 106: 11540–11545.
  62. 62. Kiener A, Gall R, Rechsteiner T, Leisinger T (1985) Photoreactivation in Methanobacterium thermoautotrophicum. Arch Microbiol 143: 147–150.