Gene Expression Toolbox for B. subtilis

Sarah Guiziou, from the Jerome Bonnet lab at the University of Montpellier, has graciously donated a large set of plasmids and strains comprising a toolkit allowing tunable gene expression in Bacillus subtilis. The amyE integration vectors in the set contain various arrangements of natural promoters, optimized RBS sequences, and protein degradation tags. By fusing the constructs to sfGFP reporters, Guiziou et al. achieved a range of expression corresponding to an average number of GFP molecules per cell varying from 15 to 270000, a span of more than five orders of magnitude. (Some of the higher expression levels result in B. subtilis constructs that look distinctly bright yellow-green under ordinary room lighting!) A complete listing of the plasmids and B. subtilis strains in the set are beyond the scope of this news item, but I encourage you to read the paper. Supplementary data file 1, an Excel spreadsheet detailing expression levels, is especially helpful. We thank Guizhiou and colleagues for these valuable tools.

A Safe Simulant for Bacillus anthracis

Safe simulants that closely mimic the select agent Bacillus anthracis are needed both for laboratory and field studies. B. anthracis belongs to the Bacillus cereus group (BCG) of species. Members of the BCG are nearly identical is cell and spore morphology due to highly similar genome sequence and content. They differ primarily in toxins and virulence factors, many of which are encoded by megaplasmids that differ among isolates. B. thuringiensis (Bt) likewise belongs to the BCG. Some naturally insecticidal Bt strains have been safely used in agriculture for over half a century. Non-insecticidal derivatives of standard Bt strains would seem to make ideal simulants for B. anthracis. For this reason, Alistair Bishop and colleagues at the Defence Science and Technology Laboratory, Salisbury, Wiltshire, UK, have developed plasmid-cured derivatives of Bt kurstaki strain HD1. One of them, HD-1 Cry-, has been demonstrated to be particularly useful in studies of spore aerisolization, dispersal, and decontamination. We thank Dr. Bishop for depositing B. thuringiensis HD-1 Cry- in the BGSC. It is available as BGSC 4D24.

ordering issues

Some have reported recent problems with the online ordering system. When we receive an order, we always acknowledge it within 24 hours. If you do not hear back from us within that time period, please contact me ( and let me know of the problem.


New! Protease-free Bacillus subtilis host

Bacillus subtilis is widely used as a production platform for synthesizing enzymes, pharmaceuticals, and fine chemicals (1). Unfortunately, B. subtilis 168 secretes no fewer than seven proteases during vegetative growth and stationary phase. Strains in which multiple protease genes have been inactivated have proved to be superior to wild type strains in production of foreign proteins (2, 3). I have now constructed a seven-protease deletion strain that is free from antibiotic resistance genes or integrated plasmids. This strain, B. subtilis KO7, was generated from the commonly used laboratory host, PY79, by sequentially knocking out the coding sequences. At each step, I transformed the strain with the appropriate BKE cassette for knocking out one of the loci, removing the erythromycin resistance gene with the Cre-producing plasmid pDR244, and finally heat-curing the plasmid. All seven knockouts in KO7 were confirmed by sequencing to be marker-free, in-frame deletions with a 150-bp scar replacing the coding sequence. KO7 is prototrophic and grows rapidly in standard minimal media for B. subtilis. I am placing strain KO7 in the public domain and disclaim all downstream rights to any process or product that you develop with it. It will be available from the BGSC as accession number 1A1133. Standard user fees apply. Later this summer I hope to introduce further elaborations of KO7, such as restriction-negative and asporogenous variants. Feel free to put in requests for particular features!

BGSC Accession: 1A1133

Original Code: Bacillus subtilis KO7

Reference: Zeigler DR, unpublished

Genotype: ΔnprE ΔaprE Δepr Δmpr ΔnprB Δvpr Δbpr

Description: Free of secreted proteases; marker-free deletions in PY79 genetic background; prototrophic

B. subtilis essential gene knockdown library

The BGSC is excited to announce the availability of a new collection of Bacillus subtilis 168 mutants designed to explore the functions of 289 essential genes in this organism. The paper describing the construction of this library and its initial characterization will be released online today (26 May 2016) and will appear in the June 2 edition of Cell. The paper is a collaboration among labs at the University of California, San Francisco, Stanford University, University of California, Berkeley, and McMaster University, Hamilton, Ontario. The co-first authors are Jason M. Peters of UCSF and Alexandre Colavin and Handuo Shi of Stanford.

The library uses a CRISPR interference (CRISPRi) strategy to created a tunable “knockdown” of individual essential genes. Every strain in the library has a Streptococcus pyogenes dcas9 gene integrated into the B. subtilis lacA locus, where it has been placed under control of the xylose-inducible Pxyl promoter. Each strain also has a single-guide RNA (sgRNA) targeting a specific essential gene. The sgRNA coding sequence is integrated into B. subitlis amyE, where it has been placed under the control of the strongly constitutive Pveg promoter. The dCas9 protein lacks nuclease activity. But when dCas9 is present, the sgRNA enables it to bind to the 5’ end of the target gene, where it effectively blocks transcription via steric hindrance. Basal level expression of dcas9 in the absence of xylose knocks down expression of the targeted essential gene about 3-fold. This reduction creates subtle phenotypes, such as increased sensitivity to specific antibiotics and chemical inhibitors, but allows for essentially normal growth under standard laboratory conditions. Full induction of dcas9 with xylose (1%) reduces expression of the essential gene ~150-fold, with drastic consequences for cell morphology and viability. Varying the concentration of xylose between 0.001% and 0.1% allows tunable expression of the essential gene. Peters et al. have not only reported the construction of the library, but have demonstrated its power for analyzing essential genes. They used chemical genomics, for example, to reveal the essential gene network of B. subtilis, revealing interesting connections between seemingly unrelated processes.

These strains provide an invaluable tool for a systematic study of essential genes in a bacterial model system. We thank Jason Peters, Carol Gross, and the entire consortium for donating the library to the BGSC, and we look forward to supplying strains from it to scientists from the B. subtilis research community and beyond. For a complete list of the genes targeted in the library, please see the Peters et al. publication. Summaries of what has been learned previously about most of these genes can be accessed at SubtiWiki. It will take a little while for us to update the BGSC online database to include these strains, but they are available for immediate distribution. But their naming convention is simple. The numeric portion of the gene’s locus tag is appended to the prefix “BEC” to produce the strain name. Hence the knockdown strain for the essential gene ligA, which encodes DNA ligase and carries the locus tag BSU06620, is BCE06620. The full genotype of this strain is lacA::Pxyl-dcas9 amyE::Pveg-sgRNA(ligA) trpC2, and it carries resistance markers for erythromycin and chloramphenicol. Users may request these strains by giving us the targeted gene name or locus tag. Standard user fees apply.

pLIKE vector sequences

Over the past two years, there has been considerable interest in the pLIKE vectors from the Thorsten Mascher lab at Ludwig-Maximilians-University (LMU) Munich. These expression vectors feature a bacitracin-inducible promoter, available both in a replicative plasmid (our ECE255) and an amyE-integration vector (ECE256). For the convenience of our users, we have uploaded maps and DNA sequences of these plasmids onto our website. You can download them here. Thanks to Prof. Mascher for providing this document.

Low-Noise Fluorescent Protein Vectors

Some of the most useful tools for Bacillus geneticists are integrative vectors designed to fuse a promoter of interest to a coding sequence specifying a fluorescent protein. Strains bearing these vectors are among the most widely requested items in the BGSC collection, and improved versions offering some advantage over first generation vectors have regularly appeared on this page. Such is the case today, as we feature the pXFP_Star reporter system, developed and kindly donated by Stephanie Trauth and Ilka B. Bischofs in the Bischofs lab at the University of Heidelberg. These vectors, which come in green, cyan, and yellow “flavors,” feature extremely low noise. A B. subtilis strain with an empty pXFP_Star vector integrated into its chromosome has no greater background fluorescence than the host strain alone. This significant improvement was achieved by including a strong transcription terminator just upstream of the promoter-cloning site. The pXFP_Star vectors will be useful for studying all kinds of promoters, but should be especially advantageous for analyzing those that are weakly expressed. As a bonus, they also are designed to allow ligation independent cloning (LIC). The vectors are available in the following BGSC strains. Note that the NCBI sequence files may not be released for a few weeks, but in the meantime are available from the BGSC upon request.

ECE295 | pGFP_Star | KJ411636

ECE296 | pYFP_Star | KJ411637

ECE297 | pCFP_Star | KJ411638

Easily transformable mutant of B subtilis NCIB3610

The well-studied Gram-positive model organism, Bacillus subtilis 168, is a highly domesticated descendent of type strain of the species, B. subtilis Marburg. Forensic sequencing evidence (Zeigler 2008) suggests that the wild type strain was passaged multiple times in the laboratory, acquiring a few key mutations that eliminated swarming and motility behaviors, allowing for more uniform, regular colonies to grow on agar (Aguilar 2007). This domesticated Marburg was further adapted to life in the laboratory by selecting for variants that showed improved growth on glucose-ammonium minimal medium (Burkholder, Giles 1947). Finally, this variant was subjected to X-ray mutagenesis, resulting in a tryptophan-requiring mutant, strain 168 (Burkholder, Giles 1947). When Spizizen discovered that it could develop natural genetic competence during early stationary phase (Spizizen 1958), strain 168 quickly emerged as the organism of choice for genetic studies and eventually for biotechnology applications.

Curiously, the wild type parent for strain 168, B. subtilis NCIB 3610 (BGSC accession 3A1), is only weakly competent for transformation--about 1000-fold less than its domesticated offspring. This deficiency is unfortunate, because in recent years much attention has been focused on the wild type strain. Strain NCIB 3610 is capable of many social behaviors that require a sophisticated degree of mulicelluarity, including biofilm formation, swarming motility, and cell signaling (Kearns 2010; Shank, Kolter 2011; Romero 2013; Vlamakis 2013). A highly competent derivative of NCIB 3610 that nevertheless retain multicellarity would be a boon to research aimed at elucidating these behaviors. Such a strain has now been described. During its domestication or mutagenesis, strain 168 lost an 84-kb endogenous plasmid that is native to NCIB 3610 (see GenBank: KF365913.1 (Konkol 2013). This plasmid encodes a 30-amino acid peptide, ComI, that is largely responsible for the inhibition of competence development in strain NCIB 3610. Strain DK1042 is identical with NCIB except for a single point mutation, comI(Q12L), that inactivates the inhibitor and increases competence 100-fold (Konkol 2013).

The Dan Kearns laboratory at Indiana University, who constructed this mutant, has now kindly donated it to the BGSC. We have accessioned it as Bacillus subtilis 3A38. We anticipate that BGSC 3A38 will be of great use to those who wish to dissect the genetics of mulicelluarity in B. subtilis. We thank the Kearns lab for making it available!

Standardized, reusable B. subtilis tools

The BGSC is pleased to announce the acquisition of a toolbox of standardized, reusable building blocks for Bacillus subtilis genetics. The Thorsten Mascher lab at Ludwig-Maximilians-Universität München, Germany, has constructed several vector tools following the BioBrick standard. Full details are available in the cited reference, including sequence files in the online supporting material. For convenience, I am including a brief description of these tools.

First, the Bacillus BioBrick Box includes five integration vectors. Each vector contains flanking homology regions for integration into the B. subtilis chromosome; a resistance cassette for selection of integrants; and a multiple cloning site that contains EcoRI, NotI, and XbaI upstream from an rfp cassette, and SpeI, NotI, and PstI downstream from the cassette. The rfp cassette turns the E. coli host bright red unless the cassette is removed by cloning within the multiple cloning site, making it simple to screen for recombinants. Except for pBS2E, all can be linearized by ScaI digestion prior to transformation into B. subtilis. Vector pBS2E can be linearized with BsaI or PciI digestion prior to transformation. The five vectors include:

pBS1C (BGSC accession ECE257), an empty vector that integrates into amyE with chloramphenicol selection;

pBS2E (ECE258), an empty vector that integrates into lacA with erythromycin selection;

pBS4S (ECE259), an empty vector that integrates into thrC with spectinomycin selection;

pBS1ClacZ (ECE260), for integration of lacZ reporter into amyE with chloramphenicol selection;

pBS3Clux (ECE261), for integration of luxABCDE (luciferase) reporter into sacA with chloramphenicol selection.

Second, the Box includes six promoters. Each has been cloned into the EcoRI and SpeI sites of the plasmid backbone pSB1C3 ( These plasmids are chloramphenicol-resistant in E. coli and do not replicate in B. subtilis or other Gram-positives.

Four of the promoters are constitutive in B. subtilis:

Pveg (ECE262), very strong constitutive promoter;

PliaG (ECE263), constitutive promoter;

PlepA (ECE264), strong constitutive promoter;

J23101 (ECE266), very weak constitutive promoter (although strong in E. coli)

Two of the promoters are inducible in B. subtilis:

PliaI (ECE267), bacitracin-inducible promoter;

PxylA (ECE268), xylose-inducible promoter

Finally, the Box contains five commonly used epitope tags. Each epitope tag fragment has likewise been cloned into the EcoRI and SpeI sites of pSB1C3.

His10 (ECE269), for tagging proteins with 10xHis;

FLAG (ECE270), for tagging proteins with FLAG;

StrepII (ECE271), for tagging proteins with Streptactin;

HA (ECE272), for tagging proteins with HA;

cMyc (ECE273), for tagging proteins with cMyc

These tools should greatly facilitate a wide variety of genetic experiments for B. subtilis. We are grateful to the Mascher lab for sharing them with us, with special thanks to Jara Radeck for her patient work in preparing the plasmids for the BGSC and answering questions about them.

Benchmarked GFP variants

The laboratory of Jan-Willem Veening at the University of Groningen have kindly provided a collection of Green Flourescent Protein (GFP) vectors that have been benchmarked for performance in Bacillus subtilis, Streptococcus pneumoniae, and Lactococcus lactis. Seven GFP variants were selected for analysis (see below).

The seven gfp variant genes were inserted into each of three plasmid constructs. For B. subtilis, each gene was placed under the control of the IPTG-inducible Phyperspank promoter in the amyE integration vector pDR111. For S. pneumoniae, each gene was placed under the control of the Zn(++)-inducible promoter PZn in a bgaA integration vector. For L. lactis, each gene was placed under the control of a constitutive promoter in the pseudo 10 integration vector pSEUDO.

The seven benchmarked GFP variants are as follows (with more information available in Table 2 and in the main text of the publication):

GFPmut1; optimized for original A. victoria; 35-fold brighter than wild GFP

GFP+; optimized for E. coli; 130-fold brighter than wild GFP

GFP+(htrA); optimized for E. coli; improved translation efficiency for S. pneumoniae

GFP(Sp); optimized for S. pneumoniae; functions as monomer

sfGFP(Bs); optimized for B. subtilis; superfolder

sfGFP(Sp); optimized for S. pneumoniae; superfolder

sfGFP(iGEM); optimized for B. subtilis/E. coli compromise; superfolder

Interestingly, Overkamp et al. found that the variant producing the strongest signal in B. subtilis---whether in planktonic cells or biofilms---was GFP(Sp), which had been optimized with S. pneumoniae codon usage. Conversely, the variant producing the strongest signal in S. pneumoniae and L. lactis was sfGFP(Bs), which had been optimized for B. subtilis. The entire collection of 21 plasmids is available from the BGSC in our strains ECE275 through ECE295A. Probably the most heavily used plasmids will be pDR111_GFP(Sp), found in our E. coli host ECE278; pKB01_sfGFP(Bs), found in our ECE286; and pSEUDO::Pusp45-sfGFP(Bs), found in our ECE293. They provide the strongest signals in B. subtilis, S. pneumoniae, and L. lactis, respectively. But the entire collection may find a use in future optimization experiments for related species and for other Gram-positive organisms. We thank the Veening lab for their generosity!

pDR244 for Markerless Deletions in the BKE series

Last year we announced the availability of a knockout library for Bacillus subtilis 168. This BKE series includes over 4000 strains. In each of them a single gene has been deleted and replaced with an erythromycin resistance cassette. We are also pleased to offer a vector, pDR244, that makes it a simple matter to loop out the BKE cassette, leaving behind only a small scar consisting of bar code and universal primer sequences. The plasmid contains a functional cre gene that catalyzes a site-specific recombination between the lox sites that flank the BKE cassette. Because it is based on the pE194-ts replicon, pDR244 is quickly cured from its B. subtilis host by growing it at 42°C. Using a BKE strain together with pDR224 makes it possible to generate a markerless deletion in only a few days with a minimum amount of hands-on effort.

A map of pDR244 and a summary of the steps required to generate a markerless deletion can be found here. Plasmid pDR244 is available in the BGSC E. coli strain ECE274.

B. subtilis 168 gene knockout library available!

The BGSC is excited to announce the availability of the newly constructed gene knockout library for Bacillus subtilis 168. As many of you know, Byoung-Mo Koo and colleagues in the Carol Gross laboratory at the University of California, San Francisco assisted by members of David Rudner’s laboratory at Harvard Medical School have been constructing this and a similar library for the past couple of years. The work has been performed under a Grand Opportunity grant awarded to a consortium led by David Rudner. Dr. Koo, along with all those involved in the project, have generously consented to deposit one of these libraries in the BGSC collection prior to publication.

The library currently comprises nearly 4000 strains, each containing an erythromycin-resistance cassette inserted into one of the non-essential genes in the B. subtilis 168 genome. (The library will be augmented with a few additional strains in the near future.) Very soon we will upload data concerning these strains in our online searchable database. The strains are already available for distribution, however. The strains follow a simple naming scheme. Each begins with “BKE” (Bacillus Knockout Erythromycin), prefixed to the locus number of the knocked-out gene. Suppose, for example, one wished to obtain a spo0A::erm knockout. The spo0A gene has the locus_tag “BSU24220”. The library strain with this knockout (along with trpC2, of course) is called “BKE24220.” If you wish to request this strain, you could either request strain BKE24220 or ask for our spo0A knockout from the BKE library. Please note that the BGSC does not have the capacity to distribute the entire collection or any large subsets of it. But we are ready and able to send individual strains or smaller sets in keeping with our normal distribution policy.

We thank Byoung-Mo Koo, along with the Gross and Rudner labs for making this library available to the Bacillus subtilis research community!