RibM from Streptomyces davawensis is a riboflavin/roseoflavin transporter and may be useful for the optimization of riboflavin production strains

Abstract

Background: The bacterium Bacillus subtilis, which is not a natural riboflavin overproducer, has been converted into an excellent production strain by classical mutagenesis and metabolic engineering. To our knowledge, the enhancement of riboflavin excretion from the cytoplasm of overproducing cells has not yet been considered as a target for (further) strain improvement. Here we evaluate the flavin transporter RibM from Streptomyces davawensis with respect to improvement of a riboflavin production strain.

Results: The gene ribM from S. davawensis, coding for a putative facilitator of riboflavin uptake, was codon optimized (ribMopt) for expression in B. subtilis. The gene ribMopt was functionally introduced into B. subtilis using the isopropyl-β-thiogalactopyranoside (IPTG)-inducible expression plasmid pHT01: Northern-blot analysis of total RNA from IPTG treated recombinant B. subtilis cells revealed a ribMopt specific transcript. Western blot analysis showed that the his6-tagged heterologous gene product RibM was present in the cytoplasmic membrane. Expression of ribM in Escherichia coli increased [14C]riboflavin uptake, which was not affected by the protonophore carbonyl cyanide m-chlorophenylhydrazone (CCCP). Expression of ribMopt supported growth of a B. subtilis ΔribB::Ermr ΔribU::Kanr double mutant deficient in riboflavin synthesis (ΔribB) and also deficient with respect to riboflavin uptake (ΔribU). Expression of ribMopt increased roseoflavin (a toxic riboflavin analog produced by S. davawensis) sensitivity of a B. subtilis ΔribU::Kanr strain. Riboflavin synthesis by a model riboflavin B. subtilis production strain overproducing RibM was increased significantly depending on the amount of the inducer IPTG.

Conclusions: The energy independent flavin facilitator RibM could in principle catalyze riboflavin export and thus may be useful to increase the riboflavin yield in a riboflavin production process using a recombinant RibM overproducing B. subtilis strain (or any other microorganism).

Figure 1

Characterization of RibM from Streptomyces davawensis as a riboflavin transporter. (A) The gene ribM from S. davawensis was overexpressed in E. coli BL21 using the plasmid pNCO113ribM (grey triangles). An Escherichia coli strain containing the empty vector pNCO113 (black circles) served as a control. E. coli strains were grown in riboflavin-free M9 medium. The cells were collected by centrifugation, suspended in transport buffer (50 mM K₂HPO₄/KH₂PO₄, 50 mM MgCl₂, pH 7.0) and the experiment was started by adding [¹⁴C]riboflavin to a final concentration of 1.6 μM. Samples were filtered, washed with water and the radioactivity was determined by liquid scintillation counting. "1 OD cells" means "the amount of cells that when present in 1 ml will lead to an OD₆₀₀ reading of 1.0". (B) Uptake experiments were performed as described for (A) with 2.2 μM [¹⁴C]riboflavin in the presence of a 10-fold excess of riboflavin (ribM+RF), flavin mononucleotide (ribM+FMN), flavin adenine dinucleotide (ribM+FAD), roseoflavin (ribM+RoF) or carbonyl cyanide m-chlorophenylhydrazone (ribM+CCCP; 130 μM). The uptake activity in the absence of competitors was 4.8 pmol pmol riboflavin × OD cells^-1 min^-1 (ribM). In the presence of CCCP uptake activity was 4.3 pmol riboflavin × OD cells^-1 min^-1. "1 OD cells" means "the amount of cells that when present in 1 ml will lead to an OD₆₀₀ reading of 1.0".

Figure 2

Monitoring expression of the gene ribM from Streptomyces davawensis in recombinant Bacillus subtilis strains containing the plasmid pHT01ribM_opt. (A) Cell samples (normalized to an OD₆₀₀ of 1) were taken 3, 6 and 9 h after treatment with IPTG and total RNA was prepared. Control cells were not treated with the inducer. The level of ribM_opt-transcript was assessed by Northern blot analysis using a ribM_opt-specific digoxigenin labelled DNA probe. The predicted size of the ribM_opt transcript is 0.8 kb (see arrow). Lanes 1, 6 and 10: B. subtilis 168 < pHT01 >; Lanes 2, 7 and 11: B. subtilis 168 < pHT01 > induced with 1 mM IPTG; Lanes 3, 8, and 12: B. subtilis 168 < pHT01ribM_opt >; Lanes 4, 9 and 13: B. subtilis 168 < pHT01ribM_opt > induced with 1 mM IPTG; Lane 5, no sample. (B) After growth in LB and treatment with IPTG B. subtilis < pHT01ribM_opt > cells were collected and cell free extracts were prepared. Cytoplasmic fractions and membrane fractions were subjected to SDS-PAGE, followed by transfer to a nitrocellulose membrane and analysis with anti-penta-his mouse monoclonal antibodies/goat anti-mouse IgG alkaline phosphatase (AP)-coupled secondary antibodies. Lane M, protein marker (in kDa); lane 1, B. subtilis < pHT01 > cytoplasmic fraction; lane 2, B. subtilis 168 < pHT01 > membrane fraction; lane 3, B. subtilis 168 < pHT01ribM_opt > cytoplasmic fraction; lane 4, B. subtilis 168 < pHT01ribM_opt > membrane fraction; (C-terminally his₆-tagged RibM from Streptomyces davawensis, calculated molecular mass: 24.7 kDa).

Figure 3

The overproduction of RibM allowed growth of a ΔribU::Kan^r ΔribB::Erm^r Bacillus subtilis strain. Streaks (top) and drops (bottom, about 50,000 cells) of B. subtilis ΔribU::Kan^r ΔribB::Erm^r cells expressing ribM from plasmid pHT01ribM_opt were applied to LB plates (about 0.5 μM riboflavin) containing the indicated (additional) amount of riboflavin and 100 μM IPTG. Growth was recorded after incubation for 36 h at 37°C. As controls, strains were transformed with the empty vector pHT01. Apparently, only the strains transformed with pHT01ribM_opt could grow on culture media (LB) with low amounts of riboflavin.

Figure 4

Overproduction of RibM enhanced roseoflavin sensitivity of a ΔribU::Kan^r B. subtilis strain. (A) Streaks (top) and drops (bottom, about 50,000 cells) of B. subtilis ΔribU::Kan^R cells expressing ribM from plasmid pHT01ribM_opt were applied to LB plates containing the indicated amounts of the toxic riboflavin analog roseoflavin and 100 μM IPTG. Growth was recorded after incubation for 36 h at 37°C. As controls, strains were transformed with the empty vector pHT01. At 10 μM roseoflavin it was most obvious that the presence of RibM increased roseoflavin sensitivity due to import of the toxic compound. (B) B. subtilis ΔribU::Kan^R expressing ribM from plasmid pHT01ribM_opt was grown in LB broth in the presence of the indicated amounts of roseoflavin and 100 μM IPTG. Growth was recorded at μ = 600 nm. As controls, ΔribU::Kan^R strains containing empty pHT01 were analyzed. At 50 μM roseoflavin, the effect of RibM was most obvious (see bracket). Cells (black triangles) producing RibM imported toxic roseoflavin and consequently grew to an OD₆₀₀ of about 2.5 only. The control strain (open triangles) was less affected by roseoflavin and grew to an OD₆₀₀ of 4.2.

Figure 5

Introduction of the heterologous flavin facilitator RibM from Streptomyces davawensis enhanced riboflavin synthesis in the high-performance riboflavin production strain BSHP. Two BSHP strains containing pHT01ribM_opt (designated as ribM1 and ribM2) and one strain containing pHT01 (control) were grown in shake flasks in the medium used for industrial riboflavin production. For each strain three shake flask experiments were carried out (n = 3). Different amounts of IPTG were used to induce synthesis of RibM. At the end of fermentation (30 h) residual sugar and the optical density at μ = 600 nm were measured. In all shake flasks the carbon source was completely metabolized and the strains had grown to a very similar cell density. For the cultures induced with 100 μM and 1 mM IPTG the t test gave the result that the RibM containing strains produced more riboflavin (mg/L), which was significant at the 5% level.