Team:Groningen/Description
Blue Bio Energy
Description
The world is changing and so are the energy needs of humanity. Fossil resources are being depleted and it is clear that the transition to clean sustainable energy has to be made. We believe synthetic biology can be an important catalyst in this process. Specifically, we engineer a Bacillus subtilis biofilm to function as a cation exchange membrane. Such a membrane can be used in Reverse Electrodialysis (RED), a technique to generate energy where salt and fresh water mix, for example where rivers flow into the sea. We call this application of the bacterial biofilm Blue Bio Energy. To make the Bacillus subtilis suitable for RED, we first make it more robust by overexpressing the biofilm genes tasA and bslA while preventing reversion to the motile state by knocking out the abrB regulator and overexpressing slrR, another biofilm regulator. Simultaneously, the amino acid polymer gamma-polyglutamic acid was modelled extensively to see if it could be used to make the biofilm ion selective. This was found to be the case. Experiments showed that Bacillus subtilis 3610 comI biofilms are slightly ion selective and, surprisingly, that ion selectivity is improved by knocking out abrB as well as by overexpression of bslA and slrR.
Growing biofilms
The word biofilm is used to denote a pleidae of different extracellular structures created by bacteria. The willingness of bacteria to stick together on a surface using secreted extracellular materials is usually considered the defining property of a biofilm, though it is hard to define a clear boundary between closely packed colonies and a low density biofilm. Biofilms are usually grouped according to the type of surface they inhabit, depicted in Table 1.
Surface
Typical biofilm
Solid-liquid
Thin and uniform
Solid-air
Thicker and wrinkly, heterogeneous
Liquid-air
Floating pellicles
The effect of the surface on the biofilm phenotype
Since (by definition) liquid-air pellicles cannot be grown on a solid strength-providing carrier material, they are uninteresting for our project. The solid-air type is the most well-known and can be grown on a simple agar plate. This is convenient, but our setup is nothing like an agar plate, and it is not clear that a biofilm grown on a solid-air interface can cope with the water flow in our setup. The solid-liquid type has the advantage of living at the right kind of surface, but is difficult to grow. In our project we tried growing both the biofilms on the solid-air and on the solid-liquid interface. We found that the solid-air type biofilms were strongest and therefore most suitable for our application.
Different states of Bacillus subtilis
The Bacillus subtilis that is used in our project is able to form a biofilm naturally. However not every Bacillus subtilis with the same genotype is in the biofilm state. One Bacillus subtilis strain knows different behaviours or states. For example, a bacterium can swim around freely, which is called the planktonic state. But B. subtilis can also form spores, or work together in a biofilm community. This biofilm state aids in the survival of the bacteria under harsh conditions.
A biofilm originates from one or more bacteria that adhere to a surface. By growing in aggregates and producing an extracellular matrix of polymeric substances, a complex community of microbiological life is established. In the biofilm every bacterium has its own role in maintaining the community. Although these bacteria have the same DNA, they express different genes. This phenotypic heterogeneity is the result of certain genetic pathways being activated. When a biofilm matures, some cells start to express different genes and divert into a planktonic state. In this planktonic state the bacterium becomes motile and starts to disperse into the environment. From this planktonic state a new biofilm can be formed by only a few bacteria. The bacteria once again start to change their gene expression and start to express matrix genes to form a biofilm (Vlamakis et al., 2013).
Engineering the biofilm state
Our project requires a stable, robust biofilm. To accomplish this we need to fix the cells in the biofilm (matrix) producing phenotype. The phenotype of Bacillus subtilis is determined by the activity of different genetic pathways. A network of regulators, activators and repressors regulate the expression of genes resulting in a certain phenotype. The expression of biofilm genes is also under control of this complex network. We decided to adopt both a top-down approach (where we modify the genetic pathways controlling expression of biofilm genes and repression of motility genes), and a bottom-up approach (where we overexpress biofilm genes directly). Our biofilm must not only be stable, it must also be ion-selective. Specifically, we wanted to engineer the biofilm to function as a cation exchange membrane. However, selectivity can only be achieved in a reasonably robust biofilm.
Bottom up approach for robustness
The most obvious way to increase the robustness of the biofilm is the direct overexpression of biofilm genes. This bottom-up approach aims to have continuous expression of biofilm genes as a result, maintaining the bacteria in a biofilm state even if they would otherwise revert to a motile state. Gene candidates are the tasA gene, which results in the formation of TasA amyloid fibres and provides structural integrity to the biofilm (Romero et al, 2010), and the bslA gene, which encodes hydrophobic proteins which form a layer on the surface of a biofilm. This hydrophobic layer of BslA is a major contributor to surface repellency of the biofilm (Kobayashi et al, 2012).
Structural fibres (TasA)
TasA is the main protein component of the Bacillus subtilis biofilm. It forms amyloid fibres which provide structural integrity to the biofilm by binding together the cells in the biofilm (Romero et al, 2009). It is also known that biofilms lacking the TasA amyloid fibres are very fragile (Romero et al, 2010). TasA is part of the tapA-sipW-tasA operon (Kearns et al, 2005). In this operon the sipW gene is responsible for the secretion of the TasA and TapA proteins (Branda et al, 2006). TapA is required for the polymerisation of TasA proteins into amyloid fibres. However TapA is only a minor component of the TasA amyloid fibres (Romero et al, 2011;2014). Since TasA forms the bulk of the amyloid fibres we decided to overexpress the tasA gene specifically. The genes for the TasA fibres were isolated from the B. subtilis 168 genome and were submitted to the parts registry as the K1597002 biobrick. However, early overexpression of tasA might be toxic for the cell because tasA is a prion and also has high metabolic burden on the cell, so a constitutive promoter was not considered an option. To bypass the toxicity of tasA, the production of TasA proteins should be postponed to a later stage in the life of the cell. This is why the choice for an inducible promoter was made. Specifically a salt inducible promoter as seen in figure 1 a & b, which is explained in more detail further on.
Hydrophobic coating (BslA)
BslA is another prominent biofilm protein. It forms a hydrophobic coating on the outer layer of the biofilm. Bacillus subtilis mutants lacking BslA lose the ability to repel water (Kobayashi et al, 2012). Since our biofilm should be robust under water, a continuous expression of this hydrophobic coating gene bslA seems desirable. The bslA hydrophobicity gene was also isolated from the Bacillus subtilis 168 strain and submitted to the parts registry as biobrick K1597003. Just like tasA, early expression of this gene in the cell life could be toxic. Therefore, a salt promoter was added to regulate gene expression in the same manner as for tasA as seen in figure 1 c & d.
Salt promoter
The expression of the tasA and bslA genes should be under control of an inducible promoter since early expression of those genes could result in toxicity for the cells. So to control the expression a salt inducible promoter K1597000 is created, which is already present in the B. Subtilis 168 genome and is is called PproH. The promoter was isolated and engineered in front of our tasA fibre genes and bslA hydrophobicity genes. This gives us the ability to control the expression of the biofilm genes tasA and bslA by placing the B. subtilis in a salty environment or not. Since our Bacillus subtilis will come in contact with seawater, this provides an elegant way to regulate the expression of the TasA amyloid fibres and BslA hydrophobic coating.
a) The construct with salt inducible promoter PproH and fibre forming tasA. b) The construct with salt inducible promoter PproH and fibre forming tasA. c ) The construct with salt inducible promoter PproH and hydrophobic coating forming bslA. d) The construct with salt inducible promoter PproH and hydrophobic coating forming bslA.
Combining structural fibres (tasA) and hydrophobic coating (bslA)
After the creation of biobrick K1597002 for the TasA amyloid fibres and the biobrick K1597003 for the bslA hydrophobic coating, we placed the biobricks behind each other. In front of this we placed our PproH salt inducible promoter called K1597000. This insert was placed in the BBa_K823023 integration vector from Munich 2012, which uses the amyE locus and replicates in E. coli. This construct was transformed in the Bacillus subtilis strain we selected, which was the Bacillus subtilis 3610 ::comI strain. See figure 2 a & b
a) The construct with salt inducible promoter PproH, fibre forming tasA and hydrophobic coating forming bslA. b) The construct with salt inducible promoter PproH, fibre forming tasA and hydrophobic coating forming bslA. c) The construct desgine with salt inducible promoter PproH and negative charge forming PGA. d) The construct design with salt inducible promoter PproH and negative charge forming PGA.
Top down approach for robustness
Apart from the bottom up approach, where individual biofilm genes were overexpressed, a top down method was also done. In this method the genetic pathways responsible for the behavioural decision making of Bacillus subtilis get engineered in such a way that biofilm behaviour is constantly on. The reason the biofilm formation is tightly regulated, is because the production of a biofilm requires high investment energetically. Phosphorylation of the transcription factor Spo0A is central to biofilm formation (Branda et al, 2001). The Spo0A transcription factor itself controls two parallel pathways which are responsible for expression of biofilm operons. It also has other functions like repressing cell motility which are beneficial for biofilm formation. However, direct engineering of the Spo0A transcription factor would be difficult, since it controls genetic pathways in a very subtle manner. Very specific amounts of phosphorylated Spo0A have to be present to regulate biofilm formation in a beneficial way (Fujita et al, 2005). The two pathways Spo0A regulates are repressors of biofilm formation. The abrB gene is controlled by Spo0A and directly represses biofilm gene expression. The second pathway is the sinR pathway, which represses biofilm gene expression (Vlamakis et al, 2013). Our goal was to disrupt the activity of those two biofilm repressing pathways.
Pathway 1 (abrB)
abrB works by binding directly to the promoter regions of operons responsible for biofilm formation (Hamom et al, 2004). This results in the repression of biofilm gene expression as seen in figure 3a. Under natural circumstances derepression of biofilm operons gets initiated by the alleviation of the repressive effect of abrB by abbA. However instead of making an anti-repressor abbA, an abrB knockout could also be made to have similar effects. In literature it was also shown that a knock-out of the abrB gene resulted in a 3.3 fold increase in volume of biofilms formed (Bridier et al, 2011). So our plan was to knock out the abrB gene to prevent it from repressing the expression of biofilm operons. Our approach was done by making a knock-out with an antibiotic resistance in the abrB sequence to disrupt it. Since repression of biofilm operons by abrB is now completely gone, this would result in constant biofilm gene transcription through this pathway as seen in figure 3b. However there were two pathways repressing biofilm genes (Vlamakis et al, 2013), so the second pathway also should be engineered.
Pathway 2 (sinR)
The second pathway also works by repression of the biofilm operons, specifically the gene sinR acts as the repressor of biofilm operon transcription. It operates in a similar way as the abrB gene. It would be obvious to make a knock-out of this gene too, to inhibit the repression of the biofilm operons in the same way as the abrB. However the presence of sinR is needed to aggregate with another protein called slrR to perform other regulatory functions. In our case the inhibition of cell motility genes is regulated by this aggregation. So knocking out the sinR would also result in the impossibility to repress cell motility, which is required for good biofilm formation. So instead of making a sinR knock-out, the sinR antirepressor slrR was overexpressed constitutively. This was done by making the biobrick BBa_K1597004, which consists of the slrR gene with strong promoter BBa_K090504 and a strong RBS BBa_K780001 in front of it as seen in figure 3c.
a) Biofilm operons eps, bslA and tapA-sipW-tasA get repressed by abrB and Spo0A.
b) Biofilm operons eps, bslA and tapA-sipW-tasA get derepressed, by abrB repressor knock out.
c) Biofilm operons eps, bslA and tapA-sipW-tasA get derepressed as sinR is repressed by slrR.
d) Biofilm operons eps, bslA and tapA-sipW-tasA get derepressed by abrB knock out and by repressing sinR with slrR.
thrC locus and our integration vector
Since the tasA and bslA genes were already in the amyE locus, another site for our biobrick with the slrR gene had to be found. abrB doesn’t need an integration vector because it is a knockout of a gene which is in the genome of the B. subtilis.We selected the thrC locus for the overexpression of the slrR gene. The BBa_K823023 integration vector from Munich 2012 was modified by replacing their amyE locus with a thrC locus from pdg1664. Also the chloramphenicol resistance gene was replaced by an erythromycin resistance gene which also came from pdg1664. The integration vector we created is called BBa_K1597001. Via this integration vector we were able to put the slrR biobrick BBa_K1597004 in the thrC locus.
Combining pathway 1 and 2
So now the repression of the biofilm operons of pathway 1 has been deactivated by knocking out the abrB gene. Also the repression of biofilm operons through pathway 2 has been deactivated by overexpressing the repressor of the biofilm operon repressing pathway 2 as seen in figure 3d.
Combining top down approach and bottom up approach
So via the bottom up approach specific biofilm genes were overexpressed by a salt inducible promoter. This bottom up construct was placed in the amyE locus via the K823023 backbone from Munich 2012. This was combined with the top down approach where genetic pathways resulting in biofilm behaviour were derepressed. The top down approach construct was placed in the thrC locus via the K1597001 backbone we created. See figure 4b. This combination resulted in B.subtilis specifically forming a biofilm and the biofilm formed proved to be robust. However to generate energy, the biofilm had to be more ion selective.
Selectivity
So after a robust biofilm was engineered via the top down and bottom up approach, secreted proteins should be engineered to make the biofilm layer ion selective. Specifically cation selective in our case. In literature we found the biofilm matrix could function as a molecular sieve, sequestering away cations, anions and apolar materials from the water (Flemming, H.-C. & Leis, in Encyclopedia of Environmental Microbiology (ed. Bitton, G.) 2958–2967 (Wiley, New York, 2002).). It is shown that cations from the water can form ionic bonds with negative groups in the biofilm. So we decided to engineer the secretion of negative molecules into the biofilm. We selected the highly negative poly-γ-glutamic acid (PGA), which is already present in the Bacillus subtilis natto strain. To prove that PGA can be cation selective in a layer, its interactions in water with positive and negative ions were modelled with Molecular Dynamics. The pgsBCA genes from were designed to be behind salt inducible promoter K1597000 we created. Since the amyE locus already had a construct in it, this would fit in the thrC with the slrR by using our integration vector. The pgsBCA genes, which form PGA, are secreted into the biofilm like in figure 2 c & d.
Total construct plan
To combine the genes which enhance stability and robustness ( tasA, bslA, ΔabrB, slrR) with the selectivity enhancing operon pgsBCA the following construct was designed as seen in figure 4c.
Constitutive overexpression of slrR and inducible overexpression of pgsBCA in the thrC locus by using the K1597001 backbone.
Induced overexpression of bslA and tasA in the amyE locus by using the K823023 backbone.
Knocking out the abrB in the Bacillus subtilis genome.
a) Biofilm operons eps, bslA and tapA-sipW-tasA get derepressed by abrB knock outand also by repressing sinR with slrR from the thrC locus.
b) Biofilm operons eps, bslA and tapA-sipW-tasA get derepressed by knocking out abrB and also by repressing sinR with slrR from the thrC locus.
c)As in Figure 4b, where the complete robust construct is created. But with addition of the negative charge creating pgsBCA.