Chassis Selection
In paratransgenesis, the chassis of choice needs to be very well thought, as the efficiency of the strategy is directly related to the survival rate of bacteria and the production rate of the effector molecule.
The ideal for paratransgenesis is to use symbiotic microorganisms of vectors as chassis, as these organisms are more likely to stay longer within the vector. However, Lutzomyia longipalpis (visceral leishmaniasis vector) does not have any symbionts, so this search is extended to commensal organisms.
With this in mind, we did extensive bibliographical research on papers that identified sandfly microbiota and papers that indicated possible model bacteria for paratransgenesis application.
As a result, we obtained three candidates: Enterobacter cloacae dissolvens, Pantoea agglomerans and Bacillus subtilis. The advantages and disadvantages of these chassis are as follows:
E. cloacae dissolvens
Advantages
Most abundant bacteria in the intestine of some insects, including L. longipalpis
Was used in other paratransgenesis researches
Disadvantages
In a study, P. papatasi was not attracted to food containing E. cloacae. This may have been caused by the odor of the bacteria.
Some other strains are pathogenic.
hard to get a strain
Poorly characterized
Pantoea agglomerans
Advantages
Used by the USP-Brazil 2017 team (BioTROJAN) and other researches with paratransgenesis
Similar to E. coli
Naturally found in L. longipalpis
We could get the strain easily
Disadvantages
Gram-negative - Less resistant to our AMP
Poorly characterized
Does not sporulate
Bacillus subtilis
Advantages
GRAS (Generally recognized as safe)
We would get the strain easily
Was already used paratransgenesis researches
Gram-positive - More resistant to our AMP
Well Characterized Genetic Parts
Great protein producer
Sporulation may be advantageous for field application.
Disadvantages
It is not naturally found in L. longipalpis, but studies show that it can survive in the midgut.
Not similar to E. coli
B. subtilis we choose you!
Comparing the advantages and disadvantages of our candidates, B. subtilis was an easy choice to make. It's amenable to genetic modifications, relatively easy to work, and its sporulation trait could protect our circuit in unstable environments.
Protection Module
In the implementation section, we discussed the contribution of sugar baits to the application of our project. These structures can be distributed in hotspots and carry out vector control normally through toxin ingestion[1]. Here, sugar baits are implemented differently, where we only use a sugar solution containing a concentration of our chassis without any toxin. Nonetheless, it's necessary to consider that these sugar baits are subjected to climatic variations and are not amenable to maintaining stable conditions for microorganisms' cultivation. This variation can affect the chassis viability and the entire system's efficiency, as it needs to be viable for leishmanicidal production when within the vector.
However, our chassis can differentiate between the vegetative to the sporulated form. This sporulation mechanism provides resistance to stresses generated in the environment (i.e., nutrient-limitation, extreme conditions, etc.), allowing them to withstand the different stresses events[2]. To optimize the chassis delivery to the vector population and to take advantage of the differentiation capacity of B. subtilis, we introduced the Protection Module.
Design
The Protection Module idealization came from a meeting with our advisor Tiago Lubiana. We discussed the instability of conditions in sugar baits and the costs to maintain the cultivation of B. subtilis. Then, we saw an advantage in using this chassis' differentiation mechanism, where we could produce the sporulated form to fill the sugar baits and engineer a module to induce germination upon entry into the vector.
The iGEM SZU 2017 Team described the PsspB-gerA part BBa_K2232020 to induce germination in B. subtilis, and we used it as the basis for this Module. The main component is the gerA receptor, an endogenous Bacillus subtilis receptor located on spores coat, which recognizes nutrients in the medium and triggers germination[3]. L-alanine is the main nutrient that triggers differentiation from binding to gerA and probably an abundant amino acid in the midgut of the vector (or any other organism). The PsspB promoter overexpresses this receptor during the forespore stage formation and with that, we can produce the chassis in its sporulated form and store it in sugar baits, waiting for the sand fly's digestion.
Indeed, L-alanine is the main nutrient that triggers differentiation from binding to gerA and abundant amino acid in the midgut of the vector (or any other organism). It is well known that the spores of B. subtilis can support enzymes and gastric fluids in humans, remaining viable and germinating in the intestine[4][5]. This suggests that an analogous process occurs in the sandflies, as this circuit will guarantee these sporulation-germination cycles.
Figure 1: Germination Measurement Device
We designed a Germination Measurement Device BBa_K4075555 to assess sporulation and the germination process, which has potential for Paratransgenesis assays in sandflies or mosquitos with Bacillus subtilis. The system is composed of the PsspB and gerA promoter with a GFP reporter expressed downstream and a mCherry reporter expressed by the constitutive promoter Pveg. This promoter has activity only in the vegetative phase, that is, when the spores underwent differentiation within the vector.
Detection Module
Overview
Synthetic biology as a platform for developing paratransgenesis technologies can promote an efficient application in the control of vector-borne diseases and, especially, guarantee a safe application. Safety is the main challenge of paratransgenesis, as releasing engineered organisms into the environment can be harmful. Our first biocontainment against this event is a killswitch mechanism (see Biocontainment Module), and the second one is this module that identifies the vector's midgut.
The Detection Module was designed to be specific to Lutzomyia longipalpis, the vector target. Although it is challenging to find biomarkers in this insect, through the literature, we have identified putative conserved miRNAs present in the midgut (referred here as miRNA-Lulo). Then, we built a library of de-novo-designed Toehold RNA switches specifically for miRNA-Lulo using our (slow but efficient) Toehold Switch Generator, the Carpincho Tool.
How we design it?
TOEHOLD-BASED SWITCH INSPIRATION
The Detection Module is the strategy found for the chassis to recognize when it is in the midgut of L. longipalpis. Initially, this was not the primary objective as the project focused on the parasite as a target. The first efforts were related to searching for biomarkers in leishmania during its promastigote stage, carrying the system trigger in the parasite presence. However, in a meeting with our advisor Prof. Marcelo Ramalho-Ortigao, we concluded that the ability to detect the presence of the parasite might not generate a suitable time response to effectively kill leishmania, a crucial factor for the project's success.
Therefore, the parasite detection strategy concentrates on the search for midgut biomarkers of sandflies and the leishmanicidal molecule's production from the chassis's entry into the insect. The search for protein receptors, metabolites, salivary proteins, and digestion conditions required the exploration of the complete physiological mechanisms of this vector. Still, these were insufficient to find a potential candidate since the molecules characterized in L. longipalpis could be found in any other organism. In fact, the lack of information regarding sandflies was confirmed by Prof. Marcelo Ramalho-Ortigao and Dr. Daniel Solomon, both specialists in sandflies.
In addition, this change could generate other drawbacks in the system, as production in high quantities could affect chassis growth and harm the insect's fitness. These problems were solved from the design of proAMPs, this mechanism is described in the Elimination Module section.
In a meeting with Dr. Daniel Solomon, an issue about the unspecificity of sugar baits was raised (many insects consume glucose and could be off-targets, such as Apis mellifera ), which reinforced our need to find a specific strategy for detection. Thus, the search for alternative detection methods resulted in the option described by the iGEM USP 2017 team, a Toehold RNA Switch system as a potential tool to be applied in paratransgenesis.
We sought miRNAs in the sandfly to act as activators of the Toehold, which led our team to a bibliographic review. The best miRNAs target candidates were those secreted in the places of interest for detection, such as the digestive tract and the midgut. Based on Oliveira (2016), which performs a comparative analysis of miRNAs in different species of Diptera, including L. longipalpis, we made a table comparing miRNAs of different places and levels of expression described in the sandfly.
| miRNA | Localization | Ref |
|---|---|---|
| aga-miR-1175 | Aedes aegypti: midgut[3] Anopheles gambiae: midgut and saliva[2] | VectorBase |
| aga-miR-1174 | A aegypti: midgut[3] A. gambiae: midgut and[2] | VectorBase |
| aga-miR-281 | A. gambiae: saliva[3] Lutzomyia longipalpis[4] Aedes albopictus: midgut[2] | VectorBase miRbase |
| aga-miR-12 | A. gambiae: midgut[3] A. aegypti: midgut[2] | VectorBase |
| aga-miR-283 | A. albopictus: midgut[2] A.: saliva and midgut[2] L. longipalpis: Highly expressed[6] | VectorBase miRbase |
| aae-miR-1890 | Ae. albopictus: midgu[2] An. gambiae: midgut[3] | VectorBase |
| Bantam | D. melanogaster, A. albopictus, A. gambiae[2][3] and L. longipalpis[4] A. gambiae: midgut y saliva[3] | VectorBase |
| aga-mir-317 | L. longipalpis: Highly expressed[4] A. albopictus: head[2] A. gambiae: saliva and head[2][3] | VectorBase miRbase |
Among the best candidates, we find mir-283 with a high level of expression and present in the midgut and saliva. Other potential miRNAs were presented, but they still need to be further described.
The Toeholds design for the recognition of miRNA represents a good alternative for the detection of Leishmania infantum within the sandfly; however, several factors need to be evaluated. One of them is the ability of our Bacillus subtilis chassis to internalize the extracellular RNA to the intracellular medium to be recognized and activate the Riboswitch. This was mentioned by the Stanford team, in which a mechanism for the import of DNA is cited, it is suggested that B. subtilis has a natural competence to internalize extracellular DNA sequences from the environment.
However, in a meeting with Professor Danielle Pedrolli, we perceived that the importation of RNA can follow different mechanisms. There are studies on the interactions between the microbiome and the host through miRNA, acting as molecular regulators and gene expression, the host exerts signaling mechanisms through the release of miRNA in extracellular vesicles or those associated with high-density lipoproteins or argonaut protein, which are taken up by microorganisms in the environment[8][9]. These interactions require different mechanisms of internalization of extracellular RNA within Bacillus subtilis that involve transcytosis processes, such as clathrin-dependent endocytosis, caveolin or other independent mechanisms.[10][11]. More tests are needed to verify that the competition of Bacillus subtilis allows the internalization of sufficient quantities of miRNAs for the detection and activation of our system.
Design
For miRNA detection, we planned to use the Toehold RNA Switch riboregulator. These regulators are an RNA sequence with a secondary hairpin structure that bears a strong ribosome binding site (RBS) and start codon. This conformation prevents ribosome binding and begins translation of the coding sequence. The activation occurs when the trigger RNA (or DNA) binds to the switch sequence and induces the secondary structure removal, making the RBS and start codon available to ribosome binding and translation. Furthermore, it presents positive results for the recognition of DNA or RNA molecules as biomarkers[6][7].
This switch is versatile and robust for synthetic biology applications since it offers a wide dynamic range, programmability, and orthogonality1. When exploring its applications and implementations, we observed that the iGEM Standford 2020 team developed a Toehold adapted to Bacillus subtilis as well as the iGEM Ulaval 2017 and many other teams developed software for the design of Toeholds according to their needs. As discussed in Uptaking RNA From Vector Midgut, we aimed to use the Toehold Switch to detect miRNAs acquired from the extracellular space. These inspired us to design our own Toehold to be applied in Bacillus subtilis through activation in the presence of miRNAs secreted in the midgut of Lutzomyia longipalpis.
Toehold Design
We decided to build a toehold library for detecting L. longipalpis in B. subtilis through our Carpincho Toehold generator. This tool is based on the de-novo-designed Toehold Switch Type II described by Green and coworkers[12](Figure 2). The Toehold is composed of a half-trigger domain, containing a "GGG" sequence linked to 15bp complementary trigger sequence; the lower stem, the other complementary sequence to the trigger that is paired with an arbitrary complementary sequence for hairpin formation; start codon, translation start sites when toehold is activated; the upper stem, a sequence composed of weak links that stabilize the hairpin structure; and the loop, comprising a spacer sequence and RBS. After that, a linker sequence joint the Toehold Switch to the coding region, this prevents interference from downstream secondary structure.
Figure 2: proAMP and AMP representation.
Carpincho Tool
Although there is already software available for toehold design, the low robustness makes it challenging to design specific features. Therefore, we built the Carpincho Toehold Designer Tool based on NUPACK[13] for thermodynamic calculations. In the Modeling section, we describe how we design each variation of Toeholds applied to our project[1].
[1] Green, A. et al. (2014). Toehold Switches: De-Novo-Designed Regulators of Gene Expression. Cell, 159(4), 925–939. Article
[2] Feng, X., Zhou, S., Wang, J., & Hu, W. (2018). microRNA profiles and functions in mosquitoes. PLoS Neglected Tropical Diseases, 12(5). Article
[3] Arca, B., et al. (2019). MicroRNAs from saliva of anopheline mosquitoes mimic human endogenous miRNAs and may contribute to vector-host-pathogen interactions. Scientific Reports 2019 9:1, 9(1), 1–16. Article
[4] Oliveira, K. P. (2016). Analise comparativa da evoluçao de miRNAs utilizando insetos como modelo. Link
[5] Angenent-Mari, N., et al. (2020). A deep learning approach to programmable RNA switches. Nature Communications 2020 11:1, 11(1), 1–12. Article
[6] Pardee, K., et al. (2016). Rapid, Low-Cost Detection of Zika Virus Using Programmable Biomolecular Components. Cell, 165(5), 1255–1266. Article
[7] Takahashi, M., et al. (2018). A low-cost paper-based synthetic biology platform for analyzing gut microbiota and host biomarkers. Nature Communications 2018 9:1, 9(1), 1–12. Article
[8] Williams, M., et al. (2017). MicroRNAs-Based Inter-Domain Communication between the Host and Members of the Gut Microbiome. Frontiers in Microbiology, 0(SEP), 1896. Article
[9] Rath, H. et al. (2020). Management of Osmoprotectant Uptake Hierarchy in Bacillus subtilis via a SigB-Dependent Antisense RNA. Frontiers in Microbiology, 0, 622. Article
[10] Ul Haq, I., Muller, P., & Brantl, S. (2020). Intermolecular Communication in Bacillus subtilis: RNA-RNA, RNA-Protein and Small Protein-Protein Interactions. Frontiers in Molecular Biosciences, 7. Article
[11] Rubio, A. et al. (2020). Transcytosis of Bacillus subtilis extracellular vesicles through an in vitro intestinal epithelial cell model. Scientific Reports 2020 10:1, 10(1), 1–12. Article
[12] Green, A.A. et al. (2020). Complex cellular logic computation using ribocomputing devices. Nature, 548, p. 117-121. Article
[13] J. N. Zadeh, C. D. Steenberg, J. S. Bois, B. R. Wolfe, M. B. Pierce, A. R. Khan, R. M. Dirks, N. A. Pierce. NUPACK: analysis and design of nucleic acid systems. J Comput Chem, 32:170–173, 2011.Article
Elimination Module
For the attack! (but safely)
Leishmania parasite infects the vector and starts to reproduce, waiting for the following passage to another host. Interruption of this cycle by the Biopank system can reduce or even end the parasite transmission. However, even though many leishmanicidal molecules are available, an elimination mechanism should consider the strategies for time and toxicity required to produce enough effector to kill the parasite since these features can be harmful to the chassis or even the vector.
The Elimination Module design holds an innovative strategy utilizing target vector physiological events to tune the production on our chassis and have an efficient response. The output of proAMP production inactivated forms of AMPs, and their activation by trypsin post blood meal in the vector's midgut enables a rapid and modulated response against the parasite. This Module is our attack mechanism after the chassis gets inside the vector and triggers the Detection Module, aiming to interrupt the transmission cycle of the parasite.
How we design it?
We aimed to kill the parasite with a low burden to our chassis and host vector. As Biotrojan taught us, "killing the pathogen may seem like the no-brainer choice, but it isn't always as simple as it sounds". And we tightly agree with it! Our primary candidates were single-chain antibodies (scFv) and antimicrobial peptides (AMPs).
The scFv might show high specificity, contributing to prevent adverse effects on the bacterium or host vector. It was required in different approaches, such as blocking parasite attachment in the midgut or acting directly against the parasite, which has low lethality or requires high amounts, in other words, low effectiveness [4]. Meanwhile, AMPs already were effectively applied in different paratransgenic approaches [1] [2], and the diversity of molecules offers a lot of possibilities to target Leishmania spp. [3]. Although the scFv is suitable for bacterial production, some secondary structures hamper efficient production. Otherwise, AMP production does not require a long synthesis pathway or engineering a metabolic pathway, and its short amino acids sequences can be produced efficiently in bacteria chassis.
Even though the AMPs have many exciting features, this molecule can only discriminate between host and target membranes by charge and composition. This non-specificity factor can affect our chassis growth, induce vector death, or risk non-target hosts. However, our Detection device increases the specificity, which reduces the risk to non-target hosts. Regarding the chassis, AMPs can be produced in an inactive form and be activated when cleavaged by a specific molecule, enabling high production amounts without significantly affecting the chassis and rapid activation by enzymatic digestion [4].
Therefore, we decided to select AMPs! To evaluate our design, we decided to test four AMPs (DRS-N1, DRS-H3, DRS-S1, and CAM-W). Each candidate demonstrated in vitro leishmanicidal activity and some convenient features for our paratransgenic design.
Candidates
Dermaseptins
Dermaseptins (DRSs) are peptides produced by Hylid frogs, a group of a-helical shaped polycationic and short peptides (21-34 residues) containing a highly preserved tryptophan residue on N-terminal 3rd position, with hydrophobic residues and the polar cationic residues clusters in opposite sides [5]. It shows strong conservation of precursor pro-domains, containing a signal peptide and acidic pro-peptide that inactivate the peptide until the target action region [6]. Regarding safety concerns, these peptides show effectiveness in vitro against many pathogens (i.e. bacterias, parasites, viruses, etc) and several human cancer types. The negatively charged membranes of some pathogens increase the interaction with DRSs cationic peptides, which induce membrane destabilization and cell lysis. Although negatively charged membranes are present in normal erythrocytes, there is not great interaction with DRSs. However, in vivo and clinical DRSs experiments remains to be investigated to understand the safety aspects [5].
Dermaseptin-N1 (DRS-N1) peptide is the shortest dermaseptin ever described, and this short cationic peptide contains 20 amino acids and was isolated from the skin secretion of Phyllomedusa nordestina. DRS-N1 was in vitro evaluated positively against the promastigote form of L. infantum in previous studies. Also, the peptide resulted in very low toxicity for mammalian cells [8].
To better understand the DRS-N1, we decided to meet Dr. Brand (responsible for DRS-N1 characterization), he pointed out to us that DRS-N1 isn't a potent peptide against leishmania as great as other options. However, besides the medium-high level to tackle leishmania, DRS-H10 shows a low level of toxicity against bacteria, an essential feature for optimum bacterial expression. Finally, this dermaseptin might be activated after the natural acidic propeptide cleavage described previously in similar dermaseptin peptides [7].
DShypo01
DRS-H3, before called DShypo01, was isolated from the skin secretion of Phyllomedusa hypochondrialis [7]. The Leishmania amazonensis bioassays revealed that DRS-H3 is an efficient anti-L. amazonensis promastigotes agent (IC50 = 15uM) when compared with glucantime (a drug used for the treatment of leishmaniasis). Our advisor Profesor Marcelo Ramalho-Ortigão highlighted that L. amazonensis has more resistance than L. infantum. Therefore, if this peptide shows great activity against the first one probably has even more effectivity against the second one. The natural DRS-H3 acidic propeptide was identified as well, which we selected for DRS-N1 experiments
Moreover, DRS-H3 acts more effectively against gram-negative (MIC = 6.6 uM) than gram-positive (MIC = 26.5 uM), which might contribute to our chassis (B. subtilis, gram-positive) production. In addition, DRS-H3 at concentrations up to 53 uM demonstrate no activity against white and red blood cells when incubated with blood samples. More analysis against mammalian cells is needed to evaluate their harmfulness to humans.
DRS-S1 was isolated from the skin secretion of Phyllomedusa sauvagei. This peptide is well-known for its lowest hemolytic on human dendritic cells, but it showed significant cytotoxic activity (IC50 = 3 uM) [9]. Also, it was effectively evaluated against L. major (MIC = 12 uM or IC50=2.4). We selected this peptide because of its leishmanicidal strength. Otherwise, more caution should be taken with cytotoxicity tests.
The CAM-W is a cecropin A-melittin hybrid peptide comprising the cationic N-terminal sequence of cecropin A and hydrophobic N-terminal sequence of melittin, which was modified through four-tryptophan-substitution (KWKLWKKIEKWGQGIGAVLKWLTTWL-NH2) from Cecropin A-melittin (CA(1-8)M(1-18), KWKLFKKIEKVGQGIGAVLKVLTTGL). Also, this peptide was already efficiently expressed in B. subtilis WB700 [13], which can make it easier to fine-tune for our design.
Many antimicrobial peptides (AMPs) are synthesized, delivered or stored in an inactive form, known as AMPs precursors (proAMPs), that require proteolytic cleavage to become active (i.e. cathelicidins, defensins, etc). Therefore, gene expression is not the primary regulatory mechanism, and the abundance of appropriate proteases to activate it became a potential strategy to regulate it.
These precursor forms often have a tripartite structure as peptide signal, acidic pro-peptide (or pro-domain) and the antimicrobial itself [15] [16]. The acidic pro-peptide has charge densities that neutralize the positive charge of AMPs. The electrostatic neutralization turns the AMPs less toxic in an inactive form, which molecularly is still unclear [15].
This mechanism has been exploited for biotechnological applications as efficiently peptide recombinant production, reducing the toxicity to the host bacteria, or the pro-drug strategy, whereas can activate the AMP upon the presence of a specific pathogen [15].
Júnior et al. (2018) designed a synthetic deca acidic model pro-peptide based on glutamic acids (Glu x 10) to modulate the conformational change of crotalicidin suggest that attaching this part leads to complete inhibition of antimicrobial activity. Still, it will be necessary to probe the inactivation mechanism through various antimicrobial peptides. In addition, the Dermaseptin pro-peptides are highly conserved, which allows the application of this pro-peptide in many Dermaseptin [7]. Therefore, we selected pro-peptides to act as an inactivator for leishmanicidal candidates such as CAM-W , DRS-N1, DRS-S1 e DRS-H3.
Figure 3: Elimination Module constitutively expresses proAMPs upon activation of the Detection Module. The digestion of AMPs causes the separation of propeptide and AMP through cleavage by trypsin, enzymes that increase their activity after the vector's blood meal.
The Ellimination Module was the first part of the system designed this season. We had several meetings with our advisor, specialist in paratransgenesis, Ph.D Marcelo Ortigao. In these meetings, Prof. Marcelo suggested that AMP can be produced in its inactivated form (proAMP) and activated at a specific moment of interest. The chassis could have a killing response to the parasite without harming itself or even sandfly.
Following the suggestion of Prof. Marcelo, who described how the change of digestive enzymes occurs in the midgut post blood meal, we used the increase of trypsin to activate the proAMP[11]. In our system, the proAMP becomes activated (AMP) seconds after the blood meal, ensuring that if there is a parasite in the blood, it does not develop in the vector after the bite of an infected animal/human (Figure 1). Thus, the mechanism quickly reached toxic quantities, considering that proAMP, the inactivated form, will have a faster enzymatic activation than instantaneous transcription/translation activation; and modular, as we can use the regulation of digestive enzymes to define the moment of activation.
The Elimination Module is linked to the Detection Module constitutively expressed by the promoter Pveg. The chassis recognizes that it is inside the vector's midgut with the Detection Module with Toehold RNA switch and starts the proAMP translation (Click here for more information). This inactivated form of AMP enables its production without harming chassis growth and promotes sufficient levels to kill the parasite efficiently and quickly after trypsin cleavage.
The proAMP consists of three components: the acidic propeptide, a sequence of peptides used to neutralize the AMP charge and reduce toxicity; a trypsin cleavage site, a sequence located between acidic propeptide and AMP; and the AMP itself, a sequence of peptides release after trypsin digestion (Figure 2). All proAMPs contain an upstream signal peptide that guides the secretion pathway. These assembled parts form our main component in the elimination mechanism as a fast and modular response against the parasite.
Figure 4: proAMP and AMP representation.
In a meeting with AMP expert Dr. Guilherme Brand, we found that limited digestion is a common reaction for obtaining AMPs in nature and in fact, it would have a similar behavior in the midgut of the vector with trypsin digestion. This is because the propeptide aggregates the stability of AMP, making enzymatic digestion difficult. Therefore, we decided to evaluate this mechanism in our laboratory experiments, check our Results section.
[1] Arora, A. K., Pesko, K. N., Quintero-Hernández, V., Possani, L. D., Miller, T. A., & Durvasula, R. v. (2018). A paratransgenic strategy to block transmission of Xylella fastidiosa from the glassy-winged sharpshooter Homalodisca vitripennis. BMC Biotechnology 2018 18:1, 18(1), 1–10.
[2] Fang, W., Vega-RodrÃguez, J., Ghosh, A. K., Jacobs-Lorena, M., Kang, A., & Leger, R. J. st. (2011). Development of Transgenic Fungi That Kill Human Malaria Parasites in Mosquitoes. Science, 331(6020), 1074–1077.
[3] Robles-Loaiza, A. A., Pinos-Tamayo, E. A., Mendes, B., Teixeira, C., Alves, C., Gomes, P., & Almeida, J. R. (2021). Peptides to Tackle Leishmaniasis: Current Status and Future Directions. International Journal of Molecular Sciences 2021, Vol. 22, Page 4400, 22(9), 4400.
[4] Wijerathna, T., Gunathunga, S., & Gunathilaka, N. (2020). Recent developments and future directions in the paratransgenesis based control of Leishmania transmission. Biological Control, 145, 104260.
[5] Bartels J. et al. Dermaseptins, Multifunctional Antimicrobial Peptides: A Review of Their Pharmacology, Effectivity, Mechanism of Action, and Possible Future Directions.
[6] Nicolas, P., & el Amri, C. (2009). The dermaseptin superfamily: A gene-based combinatorial library of antimicrobial peptides. Biochimica et Biophysica Acta (BBA) - Biomembranes, 1788(8), 1537–1550.
[7] Brand, G. et al. (2006). Novel dermaseptins from Phyllomedusa hypochondrialis (Amphibia). Biochemical and Biophysical Research Communications, 347(3), 739–746.
[8] Brand, G. et al. (2013). The Skin Secretion of the Amphibian Phyllomedusa nordestina: A Source of Antimicrobial and Antiprotozoal Peptides. Molecules 2013, Vol. 18, Pages 7058-7070, 18(6), 7058–7070.
[9] Pérez-Cordero et al. (2011). Leishmanicidal activity of synthetic antimicrobial peptides in an infection model with human dendritic cells. Peptides, 32(4), 683–690.
[10] Dermaseptin-S1 precursor - Phyllomedusa sauvagei (Sauvage's leaf frog). (2021).
[11] Telleria, E. et al. (2007). Constitutive and blood meal-induced trypsin genes in Lutzomyia longipalpis. Archives of Insect Biochemistry and Physiology, 66(2), 53–63.
[12] Ji, S. et al. (2014). Cecropin A–melittin mutant with improved proteolytic stability and enhanced antimicrobial activity against bacteria and fungi associated with gastroenteritis in vitro. Biochemical and Biophysical Research Communications, 451(4), 650–655.
[13] Ji, S. et al. (2017). Efficient biosynthesis of a Cecropin A-melittin mutant in Bacillus subtilis WB700. Scientific Reports 2017 7:1, 7(1), 1–10.
[14] DÃaz-Achirica, P. et al. (1998). The plasma membrane of Leishmania donovani promastigotes is the main target for CA(1–8)M(1–18), a synthetic cecropin A–melittin hybrid peptide. Biochemical Journal, 330(1), 453–460.
[15] Júnior, N. G. O. et al. (2018). An acidic model pro-peptide affects the secondary structure, membrane interactions and antimicrobial activity of a crotalicidin fragment. Scientific Reports 2018 8:1, 8(1), 1–11.
[16] Mahlapuu, M. et al. (2016). Antimicrobial Peptides: An Emerging Category of Therapeutic Agents. Frontiers in Cellular and Infection Microbiology, 0(DEC), 194.
Biocontainment Module
Overview
Programming a system to act only within the vector (i.e. mosquitos, sandflies, etc) is the main challenge to control undesirable GMOs spreading in paratransgenesis strategies even though it has few strategies to solve it. Kill switches (KS) are programmed death circuits, which prevent engineered bacteria from escaping into the environment. This Module can enhance the biocontainment of paratransgenesis strategies. Therefore, based on UNESP’s 2018 team idea, our team designed a photoactivatable CRISPR-Cpf1 to prevent our chassis from remaining functional when getting out of the midgut.
- Why light as a condition signal?
L. longipalpis midgut has only common nature compounds already identified that hampers a specific live or death condition. For instance, N-acetylglucosamine (GalNac) and Fe2+ are potential compounds to target if it weren't for their low specificity since they are present in several places in nature7 (Evangelista et al., 2002). Our advisor, PhD Marcelo Ramalho-Ortigao confirmed these findings due to the meager amount of biochemistry and physiological studies of sandfly vectors compared to other disease vectors. However, environmental conditions may be the choice to control bacterial survival. After all, we don't want our bacteria anywhere beyond in the midgut of the vector. Thus, based on UNESP 2018 team biocontainment, the light signal was selected as the condition to activate the killing system. As we can keep our chassis out of light while in sugar baits and the vector's midgut has a low incidence of light, the choice of using light as a signal activation is promising in allowing our chassis to live only in the midgut.
- Why Magnets as light receptors?
The Magnets domains are known as pMag (positive) and nMag (negative), VVD engineered variants that were optimized to selectively heterodimerization each other when exposed to blue light and fine-tuned switch-off kinetics to differed by up to four orders of magnitude within the range of seconds to hours[2][10]. Furthermore, this one-way regulation does not have a wavelength that generates a reversible switch compared to devices already applied in B. subtilis [8], which can be interfered with by natural light (i.e. solar radiation).
- Why CRISPR-Cpf1 is the killer mechanism?
The killer mechanism should be fast and effective in order to avoid bacterial survival and undesirable synthetic genes transfer. During the design of our Killswitch Module, we had partnered with the UANL team to compile different cell lysis mechanisms in B. subtilis. This was a way to gather and compare potential strategies for this chassis.
We selected CRISPR as a killer mechanism due to its great potential for generating multiple deletions, low toxicity and robust design that may prevent low death effectiveness and synthetic genes spreading. Nonetheless, the well-known CRISPR-Cas9 has many drawbacks, such as high molecular weight, low deletion efficiency and complexity of sgRNA build for multiplex genome editing.
The nuclease Cpf1, a type V CRISPR system, improves these aspects with the capability to process pre-crRNA to mature crRNA (or sgRNAs) without trancrRNA sequences, making circuit build easy for multiple editing targets. Also, it shows reduced molecular weight and toxicity. For B. subtilis, Cpf1 is more efficient for multiple editing and has low toxicity than Cas9, as shown previously3,9.
All in one, CRISPR-Cpf1 is our key component for this Module.
Design
CRISPR-based photo-switches have been used in optogenetic control, which offer photo-sensitive devices that are activated with post-translational modifications. CRISPR nuclease is produced into two fragments (C-terminal and N-terminal) connected to a paired photoswitchable dimerization domain. Upon exposure to light, the parts heterodimerize, connect the N-terminal and C-terminal fragments and make it functional. After that, a single-guide RNA (sgRNA), composed of crRNA and trancrRNA, guides the nuclease functional to bind and cleave a target DNA sequence. This system was already tested with split Cas9 and other variants with photoswitches such as Cryptochrome 2 (CRY2) and CIB1, a blue light-dependent protein and its binding partner from Arabidopsis thaliana, or Vivids (VVDs), the smallest photoreceptors already tested[1].
Based on the optogenetic tool from Nihongaki et al. (2019), we designed a photoactivatable KS Module containing split CRISPR-Cpf1 fragments ( BBa_K4075100 and BBa_K4075101). These fragments are connected to photoswitches Magnets (known as pMag and nMag), VVDs variants engineered to heterodimerize through electrostatic interactions upon blue light exposure. The split Cpf1 fragments linked with magnets are constitutively expressed and thus their post-translational mechanism may make the activation process faster.
After the C-Cpf1-pMags and N-Cpf1-nMag heterodimerization, the functional Cpf1 will be guided by shorter crRNA to specific gene targets. One significant versatility of Cpf1 is its capability to process a single transcript crRNA array into multiple crRNAs. That is, we can design a cassette for multiple targets. Even though this light-activated split-Cpf1-Mag has been applied in mammalian cells, Cpf1 has shown high efficiency for multiple targets deletions in B. subtilis [3][4]. Finally, regarding our KS Module goals, we designed a multiplex crRNA to target several essential genes (and protein reporter for future tests) to trigger cell lysis.
sgRNA Design
The sgRNAs were designed to cleave essential genes and eliminate synthetic DNA. We selected candidate targets from the DEG15 database[11], a database for several essential genes. We set divIB (BSU15240), which plays an important role in the sporulation process; dnaA (BSU00010), necessary for chromosomal replication; and the reporter mCherry, for future experimental tests purposes. Then we used the web tool CHOPCHOP[5][6] to design the guide RNAs. The complementary target candidates were evaluated considering the off-targets, self-complementarity and efficiency, respectively. Cpf1 recognizes a T-rich protospacer adjacent motif (PAM) sequence, which can improve the efficiency in B. subtilis[9]. Thus, we set PAM sequence = TTTN.
| Target | PAM and Target sequence |
|---|---|
| divIB BSU15240 | (TTTG)CTTGTCCAAACTCCAAAACTCAGT (TTTG)CAGGTACACAATGATCAGCACCA |
| dnaA BSU00010 | (TTTC)CCTAAGCCGACGCCCCCATAGATA (TTTC)GGCATAAAGTCCTCAACATCTTGA |
| dnaA BSU00010 | (TTTA)ACATATGCTTTGCTGCCATACATA (TTTC)AGTGCACCATCTTCCGGATACAT |
Future Challenges
In our project, GMOs releasing was was a concern that was studied from the beginning and highlighted in meetings with experts. Prof. Salomon, a tropical diseases specialist in Argentina and Prof. Margareth Capurro, a transgenic mosquitoes specialist in Brazil, warned us of the risks of implementing paratransgenesis. Horizontal gene transfer and recombinant expression hazardous in off-target organisms are some problems that can occur (and a considerable barrier for regulatory institutions!), even more with B. subtilis spores release risks.
Finally, we highlight some future challenges to improve our paratransgenesis biocontainment:
Spore release control: Prevention of re-sporulation after germination in the vector's midgut;
Physical biocontainment: Propose a sugar baits architecture regarding safety aspects;
Kill Switch experiment: Evaluate the efficiency of the designed killswitch in Wet Lab;
[1] Nihongaki, Y. et al. (2015) "Photoactivable CRISPR-Cas9 for optogenetic genome editing". Nature Biotechnology, v. 33, p. 755-760.
[2] Nihongaki, Y. et al. (2019) "A split CRISPR-Cpf1 platform for inducible genome editing and gene activation." Nature Chemical Biology (15): 882-888.
[3] Hao, W. et al. (2020) "Design and Construction of Portable CRISPR-Cpf1-Mediated Genome Editing in Bacillus subtilis 168 Oriented Toward Multiple Utilities. Frontiers in Bioengineering and Biotechnology, 8:524676.
[4] Wu, Y. et al. (2020) "CAMERS - B: CRISPR/Cpf1 assisted multiple-genes editing and regulation system for Bacillus subtilis". Biotechnology Bioengineering, 117(6), p. 1817-1825.
[5] Labun, K. et al. (2021) "CRISPR Genome Editing Made Easy Through the CHOPCHOP website". Current Protocols, 1(4), e46.
[6] Labun, K. et al. (2019) "CHOPCHOP v3: expanding the CRISPR web toolbox beyond genome editing". Nucleic Acids Research, 47(W1), p. W171-174.
[7] Evangelista, L. G. et al. (2002) "Histochemical Localization of N-Acetyl-Galactosamine in the Midgut of L. longipalpis." Journal of Medical Entomology, 39(3), p. 432-439.
[8] Castillo-Hair, S. M. et al. (2019) "Optogenetic control of Bacillus subtilis gene expression". Nature Communications, 10(3099).
[9] Wu, Y. et al. (2020) "CAMERS - B: CRISPR/Cpf1 assisted multiple-genes editing and regulation system for Bacillus subtilis". Biotechnology Bioengineering, 117(6), p. 1817-1825.
[10] Kawano, F. et al. (2015) "Engineered pairs of distinct photoswitches for optogenetic control of cellular proteins". Nature Communications, 6(6256).
[11] Luo, H. et al. (2020) "DEG 15, an update of the Database of Essential Genes that includes built-in analysis tools." Nucleic Acids Research, 49(D1), p. D677-D686.