Team:Imperial College/Proof Of Concept

Software validation

Aim

·  ·  ·

As we have explained in other parts of our Wiki the key thrust of our iGEM project was the development of our software – SOAP-Lab. The purpose of this software is to increase the accessibility of automation to wet lab scientists for both the Design and Build stages of D-B-T-L synthetic biology projects by generating Opentrons scripts for one of three different DNA assembly standards. Therefore, a key aim of our wet lab work was to demonstrate that SOAP-Lab works as intended and to demonstrate its utility. Through our wetlab experiments we aimed to achieve PROOF OF CONCEPT by assembling GFP and RFP in E.coli, using BioBricks, BASIC and MoClo. Our goal was to generate single and multiple gene fluorescent assemblies.

Design

·  ·  ·

1.

As previously mentioned a key goal of our wet lab work was to validate that SOAP-Lab works as intended. This involves demonstrating that the Design stage of SOAP-Lab allows for rapid design in a specific standard while allowing users to make use of the key attributes of that standard. Therefore, we first designed the plasmid constructs we wanted to build for each method using SOAPlab (click to learn more):

2.

Then, we carried out dry runs of the assemblies to make sure the OT-2 robots were functioning as expected.

The image depicts a dry run (without any liquid). NOTE: We had to set up the lab at OpenCell from scratch. At this point in time the laboratory did not contain any reagents.

3.

We then performed the assemblies of the previously designed constructs for each assembly method, both manually and using the OT-2 robots.

Learn more about each assembly!

BioBricks

BASIC

MoClo

4.

Finally, we plated, inoculated and mini-prepped the assembly products and evaluated their fluorescence under ultraviolet light for GFP and blue light for RFP.

CHECK OUT OUR RESULTS TO FIND OUT MORE!

Experimental design summary

Figure showing the experimental design summary. Plasmids were designed according to the assembly method requirements. All will constitutive promoters, a ribosomal binding site (RBS), green fluorescence protein (GFP) and/or red fluorescence protein (RFP) coding sequences and terminators. These are then assembled and their fluorescence analysed.

BioBricks

For BioBricks construct design, it was important to demonstrate the software allowed to design plasmids with multiple parts. In order to facilitate the design stage for the user, our softwares allows to search for desired parts directly from the iGEM registry.

Considering the key purpose of this assembly was the validate the software rather than build anything novel, the obvious choice was to build plasmids containing gene fluorescence. This would speed up screening and minimise troubleshooting. In this case, due to delayed lab access we only built single gene constructs to meet iGEM deadlines whilst still delivering proof of concept. We therefore used the following parts to build the following assemblies:

Illustration of the BioBrick constructs built. Only two parts can be assembled simultaneously using 3A assembly. Therefore two assembly rounds were required. The receiving backbone always needs a different antibiotic resistance than the assembled parts, illustrated by the difference in backbone colour. (Brown = Ampicillin, Pink = Chloramphenicol)

The BioBricks parts used in the validation of automated BioBricks assembly were kindly donated by Will Shaw from Ellis Lab (Imperial College London). The BioBricks parts were sourced from the the 2016 and 2018 DNA Distribution Kits from the iGEM Foundation.

Name	Role	Description
BBa_B0015	Terminator	Double terminator consisting of BBa_B0010 and BBa_B0012
BBa_B0030	RBS	Strong RBS
BBa_E1010	CDS	RFP
BBa_I746916	CDS	GFP
BBa_J23108	Promoter	Anderson Family Constitutive Promoter
BBa_J23119	Promoter	Anderson family Constutive Promoter
ColE1_AmpR	Backbone	Destination plasmid vector with ampicillin resistance
Col1E1_CamR	Backbone	Destination plasmid vector with chloramphenicol resistance

Automated Assembly

The automated assembly was carried out on the Opentrons 2 (OT2) with scripts generated from SOAP Lab. The automated assembly protocol was developed from the manual NEB Digestion and Ligation Protocol for BioBrick assembly kit. A total of 4 level 1 constructs (promoter + RBS, CDS + terminator) were assembled with repeats, followed by 4 fully assembled transcriptional units from the level 1 constructs. All assembled constructs were then then transformed into chemically competent DH5α E. coli. A summary of the constructs and parts used is shown.

Level 1 Assembly

Construct	Upstream	Downstream	Plasmid
pro_rbs_1_Var_BBa_J23119	BBa_J23119	BBa_B0030	ColE1_AmpR
cds_ter_2_Var_BBa_I746916	BBa_I746916	BBa_B0015	ColE1_AmpR
pro_rbs_1_Var_BBa_J23108	BBa_J23108	BBa_B0030	ColE1_AmpR
cds_ter_1_Var_BBa_I746916	BBa_I746916	BBa_B0015	ColE1_AmpR
cds_ter_2_Var_BBa_E1010	BBa_E1010	BBa_B0015	ColE1_AmpR
pro_rbs_2_Var_BBa_J23119	BBa_J23119	BBa_B0030	ColE1_AmpR
pro_rbs_2_Var_BBa_J23108	BBa_J23108	BBa_B0030	ColE1_AmpR
cds_ter_1_Var_BBa_E1010	BBa_E1010	BBa_B0015	ColE1_AmpR

Level 2 Assembly

Construct	Upstream	Downstream	Plasmid
assembly_2_Var_pro_rbs_2_Var_BBa_J23108_cds_ter_2_Var_BBa_E1010	pro_rbs_2_Var_BBa_J23108	cds_ter_2_Var_BBa_E1010	ColE1_CamR
assembly_2_Var_pro_rbs_2_Var_BBa_J23119_cds_ter_2_Var_BBa_I746916	pro_rbs_2_Var_BBa_J23119	cds_ter_2_Var_BBa_I746916	ColE1_CamR
assembly_1_Var_pro_rbs_1_Var_BBa_J23108_cds_ter_1_Var_BBa_E1010	pro_rbs_1_Var_BBa_J23108	cds_ter_2_Var_BBa_E1010	ColE1_CamR
assembly_1_Var_pro_rbs_1_Var_BBa_J23119_cds_ter_1_Var_BBa_E1010	pro_rbs_1_Var_BBa_J23119	cds_ter_2_Var_BBa_E1010	ColE1_CamR
assembly_2_Var_pro_rbs_2_Var_BBa_J23119_cds_ter_2_Var_BBa_E1010	pro_rbs_2_Var_BBa_J23119	cds_ter_2_Var_BBa_E1010	ColE1_CamR
assembly_2_Var_pro_rbs_2_Var_BBa_J23108_cds_ter_2_Var_BBa_I746916	pro_rbs_2_Var_BBa_J23108	cds_ter_2_Var_BBa_I746916	ColE1_CamR
assembly_1_Var_pro_rbs_1_Var_BBa_J23119_cds_ter_1_Var_BBa_I746916	pro_rbs_1_Var_BBa_J23119	cds_ter_2_Var_BBa_I746916	ColE1_CamR
assembly_1_Var_pro_rbs_1_Var_BBa_J23108_cds_ter_1_Var_BBa_I746916	pro_rbs_1_Var_BBa_J23108	cds_ter_2_Var_BBa_I746916	ColE1_CamR

Results

All Level 1 construct plates contained recombinant bacterial colonies displaying ampicillin resistance. Due to time constraints, it was not possible to conduct a resistance digest or sequencing analysis to verify the identity of the assembled constructs. As the transformations of the Level 2 assembly constructs were unsuccesful, it was not possible to verify the efficiency of assembly. It was not possible to conduct a repeat of the Level 2 assembly due to time constraints.

Conclusion

The simplicity of the BioBricks assembly standard makes it highly amenable to automation. Future iGEM teams will benefit from parallelizing the assembly and validation of large construct libraries with the aid of a single Opentrons, accelerating the DBTL workflow in the lab and enabling the construction of complex assemblies with BioBricks parts. The open source nature of Opentrons also fosters collaboration between different teams, and pushes the iGEM community towards greater consistency and standardization of workflows.

 Go back to validation

BASIC

For BASIC design it was important to be able to demonstrate that our software could design plasmids with both an operon type configuration and a multiple transcriptional unit configuration. We needed to demonstrate that the software would be allow users to utilise both functional and neutral linkers. Finally we needed to demonstrate that from these designs the build would be implemented in a single tier assembly.

As the key purpose of these assemblies was the validate the software rather than build anything novel, the obvious choice was to build multi gene fluorescent protein plasmids. This would speed up screening and minimise troubleshooting. We therefore used the following parts to build the following assemblies:

Illustration of the two BASIC constructs built

The DNA parts used in the validation of the automated BASIC assembly workflow was provided by the Baldwin Lab (Imperial College London).

Name	Role	Description
BASIC_mCherry_ORF.1	CDS	RFP
BASIC_sfGFP_ORF.1	CDS	GFP
BASIC_L3S2P21_J23108_RiboJ.1	Terminator, Promoter, Insulator	Multi-part construct consisting of terminator, promoter, and RiboJ insulator
BASIC_L3S2P21_J23101_RiboJ.1	Terminator, Promoter, Insulator	Multi-part construct consisting of terminator, promoter, and RiboJ insulator
BASIC_SEVA_36_CmR-p15A.1	Backbone	Destination plasmid vector
LMP_Prefix	Linker	Methylated linker prefix containing integrated prefix
LMS_Prefix	Linker	Methylated linker prefix containing integrated prefix
LMP_Suffix	Linker	Methylated linker prefix containing integrated prefix
LMS_Suffix	Linker	Methylated linker prefix containing integrated prefix
L1RBS3_Suffix	Linker	RBS linker
L1RBS3_Prefix	Linker	RBS linker

Manual Assembly

The manual assembly of BASIC was carried out to verify the correct assembly of the BASIC DNA parts. One construct was assembled and transformed into chemically competent DH5α E. coli (NEB 5-alpha Competent E. coli (High Efficiency)).

Assembly Construct	Linker 1	Part 1	Linker 2	Part 2	Linker 3	Part 3
construct1	LMS	BASIC_SEVA_36_CmR_p15A1	LMP	BASIC_L3S2P21_J23101_RiboJ1	L1RBS3	BASIC_sfGFP_ORF1

Results

The red flourescence reported reflect the prescence of the RFP dropout casette present on the BASIC SEVA 36 plasmid, and indicates the lack of integration of parts into the backbone. Several colonies did not dexpress any fluorescence, likely indicating a misassembly.

Automated Assembly

The automated assembly was carried out on the Opentrons 2 (OT2) with scripts generated from SOAP Lab. The automated assembly protocol was developed from DNABot. A total of 4 constructs were assembled with 2 different promoters (J23101, J23108) and 2 different fluorescent proteins (sfGFP, mCherry). The assembled constructs were then transformed into chemically competent DH5α E. coli. A repeat was also performed for the automated assembly of the same BASIC constructs. 

Assembly Construct	Linker 1	Part 1	Linker 2	Part 2	Linker 3	Part 3
A1	LMS	BASIC_SEVA_36_CmR_p15A1	LMP	BASIC_L3S2P21_J23101_RiboJ1	L1RBS3	BASIC_mCherry_ORF1
A2	LMS	BASIC_SEVA_36_CmR_p15A1	LMP	BASIC_L3S2P21_J23101_RiboJ1	L1RBS3	BASIC_sfGFP_ORF1
A3	LMS	BASIC_SEVA_36_CmR_p15A1	LMP	BASIC_L3S2P21_J23108_RiboJ1	L1RBS3	BASIC_sfGFP_ORF1
A4	LMS	BASIC_SEVA_36_CmR_p15A1	LMP	BASIC_L3S2P21_J23108_RiboJ1	L1RBS3	BASIC_mCherry_ORF1

Results

All plates contained colonies expressing red fluorescence. A colony expressing green fluorescence was observed in plate "B1" which contains the BASIC_SEVA_36_J23101_GFP construct (identical to the construct assembled by hand), indicating the possible presence of a correct assembly. Sequencing analysis is required to verify that the construct was properly assembled.

Conclusion

GoldenGate is an advantageous assembly method given its facile one-pot synthesis protocol. However, the creation of large libraries of diverse constructs becomes a challenge with increasing numbers of parts and variants given that each pot would contain a different set of parts. Automating complex assembly workflows is a viable solution that reduces human errors in the most complex step of the workflow - the distribution of parts amongst the assembly reactions.

 Go back to validation

Moclo

For MoClo construct design, it was important to demonstrate the software allowed to design plasmids with multiple parts. Furthermore, we also wanted to show that multiple level Golden Gate assembly is achievable.

Motivation

As the key purpose of this assemblies was the validate the software rather than build anything novel, the obvious choice was to build gene fluorescent protein plasmids. This would speed up screening and minimise troubleshooting. For this parts of the project we assembled YFP rather than GFP. This was decided after Alberto Scarampi at the University of Cambridge kindly offered to send us the required parts for this part of the experiment. We believe this does not impede the main goal of the experiment, which is validating the software.

Due to lab and time constraints imposed by Covid-19, we decided gene expression data of the three different methods would not be compared. Moreover, gene expression is also heavily influenced by the environment. The nature of the assembly methods makes it impossible to have the exact same construct sequences, questioning the value of a comparison.

We used the following parts to build the following assemblies:

Illustration of the constructs built. We chose to try out different Anderson promoters in each construct, equivalent to (BBa_J23119, BBa_J23100, BBa_J23102, BBa_J23106 and BBa_J23114) in BioBricks. Each promoter contains a ribosomal binding site (RBS) embedded. Moreover, we also tried two different backbones and terminators.

The DNA parts were graciously donated by Alberto Scarampi del Cairo from Howe Lab (University of Cambridge). All parts provided have been made to be MoClo compatible.

DNA Parts

Name	Role	Description	Length (bp)
pICH47732	Backbone	Destination Plasmid Vector	4372
pICH47742	Backbone	Destination Plasmid Vector	4372
BBa_J23119	Promoter	Anderson Family Constitutive Promoter	91
Bba_J23102	Promoter	Anderson Family Constitutive Promoter	91
BBa_J23100	Promoter	Anderson Family Constitutive Promoter	91
BBa_J23114	Promoter	Anderson Family Constitutive Promoter	91
BBa_J23106	Promoter	Anderson Family Constitutive Promoter	91
pC0_009	CDS	YFP	720
pC0_069	Terminator	E. coli arcA Terminator	50
pC0_062	Terminator	V. fischeri luxICDABEG Operon Terminator	50

Manual Assembly

The purpose of the manual assembly was to verify by hand the correct assembly of the DNA parts provided. Two constructs were assembled and transformed into chemically competent DH5α E. coli (NEB 5-alpha Competent E. coli (High Efficiency)).

Assembly Constructs

Name	Backbone	Promoter	CDS	Terminator
pICL0.1	pICH47732	Bba_J23119	pC0_009	pC0_062
pICL0.6	pICH47742	Bba_J23119	pC0_009	pC0_069

Results

Given the lack of XGal at the point in time, we were unable to perform blue/white screening to verify that the recombinant bacteria contained assemblies with inserts. We were also unable to observe any fluorescence in any colonies.

Automated Assembly

The automated assembly was carried out on the Opentrons 2 (OT2) with scripts generated from SOAP Lab. The automated assembly protocol was developed from DAMP Lab OT2 Modular Cloning (MoClo) and Transformation in E. coli Workflow. A total of ten constructs were assembled using the automated protocol using the various DNA parts as shown. The assembled constructs were then then transformed into chemically competent DH5α E. coli (NEB 5-alpha Competent E. coli (High Efficiency)).

Assembly Constructs

Name	Backbone	Promoter	CDS	Terminator
construct1	pICH47742	Bba_J23119	pC0_009	pC0_069
construct2	pICH47732	Bba_J23102	pC0_009	pC0_062
construct3	pICH47732	Bba_J23100	pC0_009	pC0_062
construct4	pICH47732	Bba_J23114	pC0_009	pC0_062
construct5	pICH47732	Bba_J23106	pC0_009	pC0_062
construct6	pICH47742	Bba_J23100	pC0_009	pC0_069
construct7	pICH47742	Bba_J23106	pC0_009	pC0_069
construct8	pICH47742	Bba_J23114	pC0_009	pC0_069
construct9	pICH47732	Bba_J23119	pC0_009	pC0_062
construct10	pICH47742	Bba_J23102	pC0_009	pC0_069

Results

Due to insufficient quantities of pC0_009 (YFP), some assembly reaction mixes did not receive or received insufficient quantities of the YFP part. The affected constructs are: construct1, construct5, construct6, construct9, and construct10. As observed from the plates (refer to Notebook), the affected constructs had a low yield of recombinant bacterial colonies. Plates that contained the unaffected constructs appeared to have a high yield of recombinant bacteria.

As the assemblies carried out by hand are amongst the affected constructs in the automated assembly (construct1 and construct9, equivalent to pICL0.6 and pICL0.1 respectively), more repeats will need to be performed to compare the assembly efficiencies between the two workflows.

No fluorescence could be observed in all recombinant bacterial colonies containing successful inserts. A restriction digest could be carried out to verify if these colonies contained the correct number of inserts, and sequencing analysis could be performed to verify the correct assembly of the parts.

Conclusion

GoldenGate is an advantageous assembly method given its facile one-pot synthesis protocol. However, the creation of large libraries of diverse constructs becomes a challenge with increasing numbers of parts and variants given that each pot would contain a different set of parts. Automating complex assembly workflows is a viable solution that reduces human errors in the most complex step of the workflow - the distribution of parts amongst the assembly reactions. 

 Go back to validation 

Conclusion

·   ·   ·

We have successfully validated our pipeline on a simple and standard BASIC construct. We created our design, sourced our parts, and generated a script to run on the Opentrons to assemble our construct of choice. This proves that our software is able to generate the correct liquid handler programs just from the designs without any coding necessary. We were also able to exchange labware as needed if one of our modules was not performing well thanks to the modularity of our hardware specification. Our small-scale proof of concept evidences the utility of our pipeline for more ambitious builds with lots of combinatorial variations, such as with our Tryptophan optimisation, as well as showcasing the ability of automation to match the accuracy of a manual assembly protocol. As our pipeline is design-based, scaling to more ambitious projects is simply a matter of specifying a more detailed SBOL file.

BASIC Assembly

Biopart Assembly Standard for Idempotent Cloning (BASIC) cloning relies on a physical DNA standard where the biopart is flanked by the integrated prefix and suffix sequences which contain BsaI recognition sites.

Cloning then relies on orthogonal linkers which are complementary to the overhangs resulting from BsaI digestion. These linkers can be designed to be functional parts themselves for example RBSs or fusion protein linkers.

Figure illustrating the BASIC assembly method

• Key references: 
• 1. Storch, M., Casini, A., Mackrow, B., Fleming, T., Trewhitt, H., Ellis, T. & Baldwin, G.S. 2015, "BASIC: A New Biopart Assembly Standard for Idempotent Cloning Provides Accurate, Single-Tier DNA Assembly for Synthetic Biology", ACS Synthetic Biology, vol. 4, no. 7, pp. 781-787.

 Close

MoClo Assembly

MoClo assembly is a cloning method that permits to directionally and simultaneously assembly multiple DNA fragments together. This method relies on Type II restriction enzymes and T4 ligase.

Commonly BsaI, BsmBI, and BbsI are used, as they cut out outside their recognition sites, creating non-palindromic overhangs. Many overhang sequences are possible so multiple fragments can be assembled together.

The final product does not have a restriction site, making the reaction irreversible.

Illustration summarising the MoClo assembly method

• Key references:
• 1.Moore, S.J., Lai, H., Kelwick, R.J.R., Chee, S.M., Bell, D.J., Polizzi, K.M. & Freemont, P.S. 2016, "EcoFlex: A Multifunctional MoClo Kit for E. coli Synthetic Biology", ACS Synthetic Biology, vol. 5, no. 10, pp. 1059-1069.

 Close

 Biobricks Assembly

Each BioBrick Part is flanked by restriction sites. EcoRI and Xbal on the left (prefix), and SpeI and Pstl on the right (suffix).

The 3 antibiotic (3A) assembly is a method that allows to assemble two BioBrick parts into a linearised backbone as well as selection of correct assembly through antibiotics. Effective antibiotic selection eliminates the downstream need for gel purification and colony PCR.

3A assembly relies on the restriction sites for assembly. After a digestion step, the desired parts are ligated together. The new assembled part will have the same prefix and suffix and contain a scar at the site of cutting/re-ligation.

Image shows a summary of a 3A BioBricks assembly. Modified from the iGEM webpage. 

• Key references:
• 1. iGEM 2017a, , Help:Assembly/3A Assembly. Available: http://parts.igem.org/Help:Assembly/3A_Assembly.
• 2. iGEM 2017b, , Help:Protocols/3A Assembly
  . Available: http://parts.igem.org/Help:Protocols/3A_Assembly.
• 3. Knight, T. 2003, Idempotent Vector Design for Standard Assembly of Biobricks.

 Close