A short story about a big magic bug

Abstract

Bacillus megaterium, the “big beast”, is a Gram-positive bacterium with a size of 4 x 1.5 µm. During the last years, it became more and more popular in the field of biotechnology for its recombinant protein production capacity. For the purpose of intra- as well as extracellular protein synthesis several vectors were constructed and commercialized (MoBiTec GmbH, Germany). On the basis of two compatible vectors, a T7 RNA polymerase driven protein production system was established. Vectors for chromosomal integration enable the direct manipulation of the genome. The vitamin B12 biosynthesis of B. megaterium served as a model for the systematic development of a production strain using these tools. For this purpose, the overexpression of chromosomal and plasmid encoded genes and operons, the synthesis of anti-sense RNA for gene silencing, the removal of inhibitory regulatory elements in combination with the utilization of strong promoters, directed protein design, and the recombinant production of B12 binding proteins to overcome feedback inhibition were successfully employed. For further system biotechnology based optimization strategies the genome sequence will provide a closer look into genomic capacities of B. megaterium. DNA arrays are available. Proteome, fluxome and metabolome analyses are possible. All data can be integrated by using a novel bioinformatics platform. Finally, the size of the “big beast” B. megaterium invites for cell biology research projects. All these features provide a solid basis for challenging biotechnological approaches.

The “Big Beast”

In 1884 de Bary named the Gram-positive bacterium Bacillus megaterium the “big beast” due to its large size.1 With a cell length of up to 4 µm and a diameter of 1.5 µm, B. megaterium belongs to one of the biggest known bacteria. Compared to the model organism for Gram-negative bacteria, Escherichia coli, it has an up to 100-times higher volume (Fig. 1). Due to the dimensions of the vegetative form and spores, B. megaterium is an ideal model system for morphological research. It is well suited for the investigation of cell-wall biosynthesis, cytoplasmic membrane formation and the sporulation process including spore structure and organization.2 In the 1960s, prior to the development of Bacillus subtilis, B. megaterium was “the” model organism of choice for intensive studies on biochemistry, sporulation and bacteriophages of Gram-positive bacteria.3–6 In 1994, Patricia Vary published the review “Prime-time for Bacillus megaterium.”7 In a recent review from 2007, 13 years later, she repeated this sentence and added “… now it is becoming true.”8

First Steps: Bacillus megaterium and Biotechnology

Due to its application for the production of biotechnological relevant substances, B. megaterium is of general interest for this branch of industry.9 Beside its growth on cheap substrates, the non-pathogenic B. megaterium, in contrast to E. coli, does not produce endotoxins associated with an outer membrane.7 Further, it does not possess any obvious alkaline proteases degrading recombinant gene product. Hence, it was sucessfully used for the production of several α- and β-amylases important for starch modification reactions in the baking industry and of penicillin amidase essential for the synthesis of novel β-lactam antibiotics.10 It is also being used industrially for the intracellular production of HIV antigen by the Abbott Laboratories.11

Sweet Life: Xylose Induced Gene Expression

In 1991, Rygus and Hillen pushed B. megaterium research with the introduction of a strong and xylose-inducible promoter found in the genome of B. megaterium. Under the strict control of the repressor protein XylR, the system was used for the controllable production of intracellular recombinant proteins like β-galactosidase, glucose dehydrogenase, formate dehydrogenase and toxin A.12–15 This promoter provided the basis for the effective and commercialized xylose-inducible vector system for B. megaterium: “…one of the most efficient expression systems described in any organism so far” (MoBiTec, Göttingen, Germany).

Express Yourself: Advanced Protein Production using B. megaterium

A novel series of vectors based on the xylose-inducible system was constructed.16 All vectors belonging to this “1622” series have an identical multiple cloning site. This feature allows for simple subcloning in various of these vectors. Upstream or downstream of the MCS the coding sequences of small affinity tags, His₆- or StrepII-tag, where introduced. The production of GFP using all vector constructs was successfully tested with a maximal GFP recovery of up to 17.9 mg/L in shaking flask cultures.16 The GFP fusion proteins containing an N- or C-terminal His₆- or StrepII-tag17 were easily purified by affinity chromatography. Up to 10.8 mg GFP per liter cell culture were achieved. For upscaling the intracellular production process, GFP was produced under glucose limited high cell density conditions in a 2 liter bioreactor. The induced GFP synthesis, which could be easily pursued via the green fluorescence (Fig. 2), led to 274 mg GFP per liter cell culture. Further, these experiments clearly showed that the xylose-inducible promoter is strictly controlled as no GFP was produced prior induction of gene expression by xylose.16 Another interesting feature was observed: Only 73% of the genetically identical B. megaterium cells showed GFP production while 15% did not (Fig. 2A, B and D).16 This culture heterogeneity was described in a different context for B. subtilis before.18

The Big Picture: Cell Biology with B. megaterium

Within the last years, the in vivo observation of cellular processes (“live cell imaging”) in the bacterial cell became more and more common. Much research was done on the Gram-positive organism B. subtilis especially by Peter Graumann's group and others.20–22 For the transfer of these methodologies to the big bug B. megaterium, the intracellular production of recombinant green fluorescent protein (GFP) fusion proteins was a major prerequisite. Due to the size of B. megaterium, a higher resolution of microscopic pictures is expected.

Inside Out: Recombinant Protein Production and Secretion into the Growth Medium

In contrast to the intracellular production of recombinant proteins, the secretion of proteins into the surrounding growth medium provides an elegant and easy method to directly purify heterologous proteins from the medium after cell removal without cell disruption. Next, a series of secretion vectors based on the xylose-inducible vector system (Fig. 3) including the sequence of the signal peptide of the lipase A from B. megaterium were constructed using a levansucrase from Lactobacillus reuteri as a model protein.23 The production and secretion of up to 4 mg levansucrase per liter culture medium was observed. Analogously, 1 mg of His-tagged and 0.7 mg of Strep-tagged levansucrase were purified directly from the growth medium.23,24 The efficiency of the established secretion system was further demonstrated by secretion of 6,098 U/l of a hydrolase from Thermobifida fusca,25 the recombinant secretion of 40 mg per liter of penicillin amidase from B. megaterium,26 the export of a keratinase with an activity of 166,200 U/L27 as well as the production and secretion of 400 µg per liter of recombinant single chain Fv antibody fragments.28 All vectors of the secretion series are commercialized by the MoBiTec GmbH (Göttingen, Germany).

In 2004, Nahrstedt et al. described the detection of the signal peptidase SipM in the genome of B. megaterium necessary for the removal of the signal peptide of secreted proteins.29 Recombinant protein secretion of (1,3-1,4)-β-glucanase was increased up to 44% when recombinant SipM was co-produced in B. megaterium. Then, one year later, in 2005, these studies were extended on the secretion of a recombinant 200 kDa protein.19 The dextransucrase DsrS from Leuconostoc mesenteroides secreted into the culture medium of B. megaterium showed an activity of up to 28,600 units per liter. By coproduction of SipM and DsrS, encoded on one vector, the extracellular amount of DsrS was increased 3.7-fold.30

Virus Helps: T7 RNA Polymerase Driven Protein Production in B. megaterium

The B. megaterium gene expression system was completed by the viral T7 RNA polymerase-dependent promoter. This demanded a stable two vector system. However, the origin of replication of all members of the constructed intra- as well as the extracellular vectors was derived from pBC16,31 which is replicated by the rolling circle mechanism.32 For the stable presence of two plasmids in one bacterial cell, origins of replication belonging to different compatibility classes are necessary. B. megaterium strain QM B1551 carries seven different plasmids (pBM100–pBM700).33 One of the corresponding replicons, repBM100,34 was chosen for a second in parallel replicating plasmid in B. megaterium.35 While pBC16 derivatives were found to have 50–100 copies per cell, 135 copies of pBM100 are present.33,36 One vector is carrying the gene encoding the T7 RNA-polymerase under control of the xylose-inducible promoter, while the other vector encodes the target protein under control of the T7 RNA dependent promoter. Using GFP as a model protein, almost 50 mg per liter cell culture were produced.35

Pretty in Pink: Vitamin B₁₂ Production with B. megaterium

Next, we decided to directly optimize the vitamin B₁₂-production with the natural producer B. megaterium.37–39 Different individual manipulation steps were followed. Beside the genetic manipulation of the existing vector DNA, further direct manipulations of the genetic information of B. megaterium were demanded.38 The used strategies illustrate the versatility of this bacterium as a biotechnological toolbox. Recombinant overexpression of single genes (cbiX, sirA, fnr and optimized hemA) increased the vitamin B₁₂ concentration up to 13.5-fold. The integration of the strong xylose-inducible promoter upstream of the hemAXCDBL operon and directed enzyme engineering of HemA to HemA_KK by an exchange of two amino acid residues to enhance enzyme stability increased the B₁₂ content up to almost 40-fold (Fig. 4). For this purpose, integration vectors based on the temperature sensitive origin of replication of pE194ts40 were constructed.37,38 The overproduction of the cbiI operon starting from cbiX by chromosomal promoter integration increased the vitamin B₁₂ content almost 20-fold. The biosynthesis of all tetrapyrroles including vitamin B₁₂ and heme uses the same precursor molecule, uroporphyrinogen III. After uroporphyrinogen III, the biosynthetic pathways of vitamin B₁₂ and heme diverge. To repress the flux of this metabolite to the heme biosynthetic path, an antisense RNA strategy to inactivate the hemZ mRNA involved in this competing pathway was established. This was one of the rare examples reported for a successfully employed antisense RNA strategy in bacteria so far. Moreover, overproduction of the recombinant vitamin B₁₂ binding proteins MetH, GlmS and RtpR, respectively, to prevent vitamin B₁₂ dependent feedback inhibition, successfully enhanced the vitamin content almost 5-fold. Here, the twin arginine transport (TAT) system, where folded proteins are secreted, could provide an elegant possibility to export the cofactor vitamin B₁₂ bound to its corresponding vitamin B₁₂ binding domains.41,42

Sometimes More is Less

All directed single manipulations to enhance the cellular vitamin B₁₂ content were successful. The intracellular vitamin B₁₂ amount was increased up to almost 40-fold. Next, the individual strategies were combined. The genetic background of B. megaterium strain HBBm1 containing the xylose-inducible promoter upstream of the hemA_KKXCDBL operon (40-fold increase) was combined with the recombinant production of the antisense mRNA of hemZ (1.2-fold increase) and with the recombinant overproduction of sirA (4.5-fold increase), respectively. By simple multiplication, a 48-fold and a 180-fold increase of B₁₂ production, respectively, was expected. But the experiments proved these assumptions wrong. The vitamin B₁₂ content corresponding to the optical density (µg L⁻¹ OD⁻¹_{578 nm}) showed a maximal enhancement of only 33- and 37.2-fold, respectively. And the volumetric amount was even worse. Only maximal values increased 13.6- and 19-fold, respectively, were detected. This was the same phenomenon observed for cells carrying two vectors overproducing GFP. Especially the production of two proteins mediating antibiotic resistance together with the DNA replication of the plasmids seemed to provide stress to the cells.43 Due to this increased metabolic burden for the B. megaterium cell, the cells grew slower, did not reach higher optical densities and further, were limited in their protein production process.

Looking into the Magic Gene Forest

During the last years, the sequences of more and more bacterial genomes were made available. This amount of data provides the possibility to have a systematic look into the genes present as well as in the regulatory areas between them. Currently, the genome sequences of two biotechnological important B. megaterium strains are being obtained (unpublished data). For an easy access to the genomic data the MegaBac platform was set up, accessible under megabac.tu-bs.de.* Knowledge of the genetic potential of B. megaterium provides the possibility to complete the whole picture of vitamin B₁₂ biosynthesis as well as several other biochemical pathways with impact on the productivity of the host cell. Insights into non-coding regions will give details of regulatory mechanisms. Gene regulating networks can be deduced using the PRODORIC database and the VIRTUAL FOOTPRINT software tool developed in our group (www.prodoric.de).44 Visualization is achieved via ProdoNet.45

The Next Revolution: “Systems Biotechnology”

Now, with help of the genome information, the introduction of modern high-throughput technologies as transcriptomics, proteomics and metabolomics to perform a broad “systems biotechnology” approach is possible. However, in our opinion, “systems biotechnology” is more than just measuring high throughput data: These data have to be combined to establish bioinformatic models for the prediction of cellular processes.

The genome data have already been used to deduce metabolic networks and to model metabolic fluxes. The models have been confronted with the results from metabolome and fluxome analyses.46,47 Consequently now, DNA microarray and proteomic experiments are the experi mental validation for the proposals. Furthermore, integration of regulatory and metabolic networks is desired. The identification of limiting steps in protein production and the deduction of molecular optimization strategies are the major goals in the future.

Figures and Tables

Figure 1 Electron microscope image of Bacillus megaterium (yellow) and Escherichia coli (red) vegetative cells. B. megaterium and E. coli cells were aerobically cultivated separately in LB medium at 37°C. They were grown until reaching the stationary phase. The two cultures were mixed in the ration of 1:1. Aldehyde-fixed bacteria were dehydrated with a graded series of acetone, critical-point-dried with liquid CO₂, and sputter-coated with gold. Samples were examined in a field emission scanning electron microscope (FESEM) Zeiss DSM982 Gemini at an acceleration voltage of 5 kV using the Everhart-Thronley SE-detector and the SE-Inlens-detector in a 50:50 ratio. The image was colored subsequently. Magnification ×15,000. Compared to the E. coli volume of 0.5 µm³ (0.5 × 0.5 × 2), B. megaterium has at least up to 100-times higher volume.

Figure 2 Culture heterogeneity of GFP producing B. megaterium cells (A and B). B. megaterium cells were cultivated on A5 medium19 agarose pad at 37°C and observed using a Zeiss Axiovert 200 M microscope. Pictures were taken using an AxioCam HR under 630x total magnification (63x objective, 10x ocular). (A) fluorescent image of GFP producing cells; (B) bright image of the same cells overlaid with green-colored fluorescent image. (C) A Biostat B2 bioreactor (B. Braun, Melsungen, Germany) with 2 L working volume connected to an exhaust gas analysis unit (Maihak, Hamburg, Germany) was operated and controlled as described previously.16,19 B. megaterium carrying a plasmid coding for GFP-Strep fusion protein was grown in semi-defined minimal medium at 37°C initially in a batch phase with 4 g/L glucose. At the end of the batch phase an exponential feeding profile was started. GFP was visualized by a lamp emitting blue light and a yellow filter using a digital camera. (D) Results of flow cytometric analysis of bioreactor cultivation. Samples taken from bioreactor cultivation of B. megaterium carrying a plasmid coding for GFP before and 4.6 h after induction of the gfp gene expression were stained with propidium iodide (PI) and analyzed in a FACSCalibur (Benton Dickinson, Belgium): Living cells, no GFP: red; living cells, GFP: green; dead cells, no GFP: black; dead cells, GFP: cyan. Percentages of the subpopulation compared to all cells are given.

Figure 3 Series of expression plasmids for the (A) extra- and (B) intracellular production of recombinant proteins by B. megaterium. All expression plasmids shown allow parallel cloning of genes of interest into the identical multiple cloning sites (MCS). SP: signal peptide; Tag: Affinity purification tag; CS: protease cleavage site; Term: Terminator; P_xylA: promoter of xylA; P_T7: T7 RNA polymerase dependent promoter; TEV: tobacco etch virus protease cleavage site; X_a: factor X_a protease cleavage site; SP_lipA: signal peptide of the lipase A; SP_pac: signal peptide of the penicillin G amidase; T_T7: terminator for T7 RNA polymerase.

Figure 4 Shaking flask cultivation of B. megaterium wild type strain and mutant strain HBBm1 (integrated xylose-inducible promoter upstream of the hemA_KKXCDBL operon). Wild type bacterial cells and mutant cells were aerobically grown in LB-medium containing 0.23% xylose. Recombinant gene expression was induced with 0.5% xylose at an optical density of 0.4. The picture was taken with a digital camera 5 hours after induction. Left site: wild type cells, right site: Mutant cells. The strong formation of tetrapyrroles is indicated by the red color.

Acknowledgements

We would like to thank Stefan Leupold for fluorescent images. We gratefully acknowledge financial support granted by the Deutsche Forschungsgemeinschaft, especially “Sonderforschungsbereich 578,” and by the “Fonds der Chemischen Industrie.”

Notes

* The platform is still password protected until the publication of the B. megaterium genome sequence. For accessibility requests contact [email protected].