Team:Newcastle/Results/SHERLOCK
CPLX1 mRNA Detecting via CRISPR SHERLOCK
Summary
Decreased CPLX1 mRNA concentration within blood plasma has been signposted as a specific biomarker for presymptomatic Parkinson's Disease (PD) [1]. The muninn project sought to develop a CRISPR SHERLOCK system to detect different concentrations of artificial CPLX1 mRNA using Cas13a produced in house and crRNA provided by our sponsors - Synthego.
Initial attempts at purification failed leading to attempts to reclone Cas13a into pET28_CD; however, MoClo assembly also failed numerous times. Gibson assembly was used as a final attempt to reclone into PET28_CD.
While site-directed mutagenesis to make Cas13a compatible with Gibson assembly into pET28_CD appeared successful, attempts to produce pET28_CD without a his-tag (pET28_CD_M) were relatively unsuccessful due to low yields. As we were unable to successfully assemble Cas13a into a suitable expression vector, we were unable to obtain Cas13a and subsequently perform mRNA detection assays.
Through information gained via our CRISPR kinetics model, we developed a method for future use to detect CPLX1 mRNA which takes less time and resources than the current standard. Although it was not possible to test our model at this time, here we propose the experimental workflow to continue this work in the future which incorporates model-informed design, purification, and the potential for large-scale system testing aided by liquid-handling automation.
Introduction
Here we propose a proof-of-principle CRISPR SHERLOCK system for the detection of CPLX1 mRNA (Figure 1) - a previously identified biomarker for Parkinson's Disease in Turkish cohorts [1]. CRISPR SHERLOCK - standing for Clustered Regularly Interspaced Short Palindromic Repeats Specific High-sensitivity Enzymatic Reporter un-LOCKing - is a method of detecting specific mRNA molecules by harnessing trans-ssRNA cleavage activity of Cas13a. The guide CRISPR RNA (crRNA) of Cas13a can be designed as the complement of the target mRNA. When the target interacts with the crRNA, the Cas13a becomes enzymatically activated and capable of performing endonuclease activity on both the mRNA bound to the crRNA (cis-cleavage) and neighbouring RNA molecules in the system (trans-cleavage). By adding quenched reporter RNA to an in vitro system, the presence of a target RNA can be detected as the reporter will only become active if it is cleaved by the target-activated trans-activity of Cas13a [2] [3].
Figure 1. Simplified workflow of how CRISPR SHERLOCK detection of CPLX1 mRNA operates.
Through designing a crRNA that is complimentary to a common and constitutive exon of CPLX1, it is possible to detect a number of CPLX1 transcripts. By doing this, the versatility of the system is improved as there is potential to detect different CPLX1 alternative spliceoforms which may vary between individuals. Our system, informed by our model, focuses on 3 concentrations of artificial CPLX1 mRNA; these 3 concentrations - 10 nM, 6.67 nM, and 3.33 nM - reflect the relative change in concentration between person without Parkinson's Disease, a person with early-stage Parkinson's Disease, and a person with late-stage Parkinson's Disease respectively [1]. By measuring difference in fluorescence at these representative CPLX1 mRNA concentrations, it will be possible to demonstrate the potential of the SHERLOCK system to identify those at risk of Parkinson's Disease as indicated by lower fluorescence output compared to a standardised control.
Results
Obtaining Cas13a Plasmid
pC029 - a plasmid encoding Leptotrichia wadei Cas13a, was obtained from Addgene [4]. The plasmid was provided in a Escherichia coli NEB stable preserve and was extracted via Miniprep (Qiagen, De). Plasmid concentration was measured as 35.43 ng.μL-1 and agarose gel electrophoresis (Figure 2) demonstrated pure plasmid was extracted as indicated by a single band at 7,800 bp.
Figure 2. 1% Agarose gel electrophoresis showing the 7,800bp band corresponding to pC029 purified from 2 different E. coli NEB stable colonies via Miniprep (Qiagen, De)
pC029 was transformed into DH5alpha to produce glycerol stocks. As DH5alpha lacks the T7 promoter, it was deemed to be suitable for short-term storage of the plasmid in vivo. Protein expression tests were completed to ensure Cas13a was not being expressed. Sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) analysis (Figure 3) indicated that no Cas13a was being expressed as the predicted 140 kDa band was not present in any sample.
Figure 3. 12% acrylamide SDS-PAGE showing resolve of Marker (ThermoFisher PageRulerTM Unstained Protein Ladder) and Cas13a after 0 hours incubation (t=0), 4 hours uninduced incubation (-ve), and 4 hours incubation after induction by isopropyl β-D-1-thiogalactopyranoside (+ve) in E. coli DH5alpha. Stained using Coomassie Brilliant Blue. Cas13a predicted to be 138.5 kDa.
Preliminary Cas13a Purification in Escherichia coli BL21 (DE3)
Preliminary attempts to purify Cas13a utilised E. coli BL21 (DE3) as the expression strain. Non-commercial competent cells transformed with the plasmid yielded a mix of large and small colonies. As the selected strain typically has few but large colonies, a protein expression test and subcultures were completed. This was completed due to concerns as to whether small colonies were the result of Cas13a toxicity or contamination. Subcultures of large (Figure 4a) and small (Figure 4b) colonies yielded only large or small colonies respectively meaning contamination was likely the explanation.
Figure 4. Observations of A) large colonies and B) small colonies ofE. coli BL21 (DE3) grown on LB agar supplemented with kanamycin after 16 hour incubation at 37˚C
New England Biolabs E. coli BL21 (DE3) were transformed with pC029 and resulted in only large colonies. Protein expression tests were completed due to only the small colony from non-commercial E. coli indicating expression (Figure 5a); however, SDS-PAGE analysis (Figure 5 b) again indicated a lack of strong expression.
Figure 5. 12% acrylamide SDS-PAGE showing resolve of A) Marker (ThermoFisher PageRulerTM Unstained Protein Ladder) and total large and small E. coli BL21 (DE3) colony proteome after 0 hours incubation (t=0), 4 hours uninduced incubation (- IPTG), and 4 hours incubation after induction by isopropyl β-D-1-thiogalactopyranoside (+ IPTG) and B) Marker (ThermoFisher PageRulerTM Unstained Protein Ladder) and total commercial E. coli BL21 (DE3) proteome after 0 hours incubation (t=0), 4 hours uninduced incubation (- IPTG), and 4 hours incubation after induction by isopropyl β-D-1-thiogalactopyranoside (+ IPTG) Stained using Coomassie Brilliant Blue. Cas13a predicted to be 138.5 kDa.
Regardless, a cell pellet was lysed and purified via FPLC using a StrepTrap HP 5 mL column (GE Healthcare, UK). Unicorn 7 trace graph (Figure 6a) indicated that some protein was successfully purified in fractions 4-8. SDS-PAGE analysis of these fractions in addition to the cell pellet, raw lysate, and clarified lysate (Figure 6b) showed the UV peak in fractions 4-7 were not due to Cas13a as no band was present at the predicted size.
Figure 6. A) Elution profile indicating change in protein concentration eluted from a StrepTrap Sepharose HP 5 mL column (GE Healthcare, UK) measured by UV absorbance at 280nm during isocratic change from binding buffer to elution buffer. First peak corresponds to protein eluted in flow through (FT). Second peak corresponds to purified protein in purified fractions (PF). B) 12% SDS-PAGE showing resolve of Marker (ThermoFisher PageRulerTM Unstained Protein Ladder), cell pellet, raw lysate, flow through and purified fractions 4-8 using a StrepTrap Sepharose HP 5 mL column (GE Healthcare, UK). Stained using Coomassie Brilliant Blue. Cas13a predicted to be 138.5 kDa
Cas13a Recloning into Pet28_CD via MoClo
As a result of failed purification, the literature was reconsulted [4] and it was concluded that the Cas13a gene must be recloned into another vector to enable expression and purification. Primers (IDT, USA) complimentary to both the 5’ and 3’ end of the Cas13a gene with BsaI restriction sites (Table 1) were used for Cas13a isolation from pC029 via polymerase chain reaction using Phusion® High-Fidelity DNA Polymerase (NEB, USA).
Table 1. Primer designs for isolation and site-directed mutagenesis of Cas13a to add BsaI C/D restriction sites for MoClo compatibility with pET28_CD| Name | Sequence |
|---|---|
| Cas13a_C Forward | GACGGTCTCTAatgaaggtgaccaaggtggac |
| Cas1_D Reverse | CTGGGTCTCTACCTctccagggccttgtactc |
Gel electrophoresis of gradient PCR (Figure 7a) showed 63oC was most effective for isolating Cas13a. Cas13a_CD was exposed to DpnI to distinguish whether PCR product contained parental DNA. 3,500bp bands were present in both samples during gel electrophoresis (Figure 7b). The band representing sample without DpnI was notably reduced likely due to pipetting error during sample loading. PCR clean-up(Qiagen, De) was completed prior to MoClo assembly.
Figure 7. 0.8% agarose gel electrophoresis showing A) gradient PCR (55 - 65oC) of pC029 to produce Cas13a_CD via site directed mutagenesis and B) DpnI digest of sample produced at 63.0oC.
Our new vector, pET28_CD, was obtained from colonies using Miniprep and concentration of both pET28_CD and Cas13a_CD was quantified using NanoDrop. Cas13a_CD was amplified to obtained more product for future reactions. MoClo assembly was carried out using BsaI-HF® (NEB, USA). A total of 20 fmol DNA was used per assembly. Due to the size of the Cas13a_CD insert different ratios of vector and insert was used, with bias given to the insert. Colony phenotype was used to assess whether assembly was successful as colonies possessing a correctly assembled plasmid were predicted to be white due to a nonfunctional LacZ gene. 1:4 plates yielded 4 blue colonies only (Figure 8a), 1:9 yielded 1 white colony (Figure 8b) and 1:14 yielded no colonies (Figure 8c). Half of the single 1:9 white colony was replated onto plates containing kanamycin, IPTG and X-gal as a method of checking contamination whilst the other half was used for Miniprep.
Figure 8. Observations of E. coli DH5alpha grown on LB plates supplemented with kanamycin, IPTG, and X-gal after transformation using constructs produced by MoClo using A) 1:4, B) 1:9, and C) 1:14 ratios with bias given towards insert. D) Observation of colony subcultured from 1:9 colony on a LB plate supplemented with kanamycin, IPTG, and X-gal. Plates incubated for 16 hours at 37oC.
Replated DH5alpha (Figure 8d) were white and possessed an identical phenotype to the original colony. E. coli Rosetta 2 (DE3) pLysS transformed in with plasmid obtained via Miniprep. Colonies were blue (Figure 9) so and so a further Miniprep was completed to check the plasmid. Gel electrophoresis results (Figure 10) were inconclusive due to issues with ladder separation and a faint band larger than pET28_CD in addition to another, prominent band at approximately 6 kbp in the 1:9 Miniprep sample. Moving forward it was decided that the construct would be redesigned for assembly by Gibson assembly.
Figure 9. Observations of E. coli Rosetta 2 (DE3) pLysS growm on LB plates supplemented with kanamycin, IPTG, and X-gal after transformation using constructs obtained via 1:9 colony Miniprep (Qiagen, De).
Figure 10. 0.8% gel electrophoresis showing Miniprep (Qiagen, De) of E. coli Rosetta 2 (DE3) pLysS transformed with plasmid assembled via Moclo.
Cas13a Recloning into Pet28_CD via Gibson Assembly
As MoClo assembly of Cas13a into PET28_CD was deemed inappropriate; likely due to the large insert size (approximately 3.5 kbp), primers were designed to allow Cas13a to be compatible for Gibson assembly. Established literature [3] and Munich iGEM (2017) [5] both removed the his-tag after FPLC via protease cleavage; however, pET28_CD does not include a cleavage site. Furthermore, Munich iGEM (2017) [5]found that Cas13a is a robust enzyme that can still function efficiently without purification. Because of this, it was deduced that primers must also designed for pET28_CD mutagenesis to remove the his-tag (Table 3) as its inclusion is not necessary and may detrimentally impact the function of Cas13a. In silico PCR using designed primers on the pET28_CD plasmid was successful and removed both the his-tag and LacZa gene.
Table 3. Primer designs for site directed mutagenesis of pET28_CD and Cas13a| Name | Sequence |
|---|---|
| Cas13a_C Forward | CTTTAAGAAGGAGATATACggtctctCATTatgaaggtgaccaaggtggacggcatcagc |
| Cas1_D Reverse | ttgttagcagccggatctcggtctctAGGTCTActccagggccttgtactcgaacatcacttt |
| pET28_D Forward | ACCTagagaccgagatccggctgctaacaaagcccga |
| pET28_C Reverse | AATGagagaccGTATATCTCCTTCTTAAAGTTAAACAAAATTATTTctagaggggaattgttatccg |
Cas13a primers were designed to isolate the LwCas13a gene from pC029 in addition to extending the sequence to include BsaI C/D on the 5' and 3' ends respectively as well as 30bp of homology with the redesigned PET28_CD (Table 3). Again, in silico PCR was completed to ensure the sequence (Cas13a_CD_M) was produced as designed. This was further confirmed by aligning the PCR product to the hypothetical sequence that primer design was based on.
Figure 11. 0.8% gel electrophoresis showing A) gradient PCR (70 - 60oC) of pC029 to produce Cas13a_CD_M via site directed mutagenesis, B) DpnI digest of samples incubated at 60.7oC and 66.1oC, and C) PCR clean-up of samples incubated at 60.7oC and 66.1oC.
0.8% agarose gel electrophoresis of Cas13a PCR products indicated that samples incubated at 60.7 and 66.1oC were subjected to mutagenesis (Figure 11a). Additionally, 0.8% gel electrophoresis of DpnI digests of the samples indicated Cas13a_CD_M samples did not contain parental DNA (Figure 11b) while analysis of PCR cleanup demonstrated samples were pure (Figure 11c). However, gel electrophoresis of pET28_CD mutagenesis (pET28_CD_M) indicated no sample was successfully mutated (Figure 12a). Results show a band at 1,500 bp; however, expected product size was 5.2 kbp. It may be hypothesised that the primers have bound in an unpredicted way but previous in silico PCR analysis and DNA sequence alignment show that the primers are only complimentary to the designed region of pET28_CD. As such, results may be due to contamination and so PCR was repeated.
Figure 12. 0.8% gel electrophoresis showing A) first attempt and B) second attempt of gradient PCR (70 - 60oC) of pET28_CD to produce pET28_CD_M via site directed mutagenesis
Analysis of repeated pET28_CD mutagenesis (Figure 12b) demonstrated 3 samples which appeared to be successfully mutated due to the presence of a band at the expected height. These samples were incubated at 62oC (F), 66.1oC (D), and 69.3oC (B). Nanodrop analysis of PCR clean-up products indicated samples had low concentration of mutated pET28_CD with sample B being the most concentrated at 20.5 ng.μL-1 - as such sample B was used for Gibson assembly.
Figure 13. 0.8% gel electrophoresis showing A) first attempt Gibson assembly using 50 ng total pET28_CD_M DNA and 100 ng total Cas13a_CD_M DNA and B) second attempt Gibson assembly using stock concentration (S) of pET28_CD_M (20.50 ng.μL-1) and Cas13a_CD_M (42.65 ng.μL-1) or dilutions (D) of pET28_CD_M (10.00 ng.μL-1) and Cas13a_CD_M (25.00 ng.μL-1) of gradient PCR (70 - 60oC) of pET28_CD to produce pET28_CD_M via site directed mutagenesis
Initial Gibson assembly (NEBuilder HiFi DNA Assembly) used 50 ng of pET28_CD_M and 100 ng Cas13a_CD_M; however, assembly failed (Figure 13a). As pET28_CD_M concentration was low and the control band was not visible in analysis of the first assembly, additional Gibson assemblies also used undiluted pET28_CD_M and Cas13a_CD_M - 20.50 and 42.65 ng.μL-1 respectively - to approximately the same 1:2 ratio. However, these concentrations also failed to assembly (Figure 13b) further evidenced by the transformation of E. coli Top10 only yielding a two blue colonies (Figure 14).
Figure 14. Observations of E. coli Top10 transformed with Gibson assembly product grown on LB agar supplemented with kanamycin, IPTG, and X-gal after 16 hour incubation at 37 ˚C.
Conclusions
pC029 encoding Cas13a was successfully obtained from E. coli NEB stable via Miniprep and was subsequently transformed into E. coli DH5alpha to produce glycerol stocks for long-term storage. Initially, E. coli BL21 (DE3) was proposed as the expression strain. A commercial variant was used due to contamination of in-house stocks as evidenced by 2 different colony sizes which were subcultured and yielded the same visual phenotype as the original colony on LB agar. While protein expression tests of commercial E. coli BL21 (DE3) did not indicate strong expression, protein purification was completed anyways but did not yield any protein.
Due to the lack of expression and purification, it was decided to reclone Cas13a into a different expression vector; namely pET28_CD. Initial approach aimed to clone using MoClo and so primers were designed for site-directed mutagenesis of pC029 to isolate the Cas13a gene and add C/D BsaI restriction sites. While this was successful, repeated assembly attempts using different ratios were generally unsuccessful. A single white colony was produced using a 1:9 vector:insert ratio, however, E. coli Rosetta 2 (DE3) pLysS transformed with the construct were blue and gel electrophoresis of a Miniprep was ambiguous as to whether the construct was correct.
Final attempts to reclone Cas13a into pET28_CD used Gibson assembly and so primers were designed for site-directed mutagenesis of both Cas13a to make it Gibson compatible, and pET28_CD to remove the his-tag as Munich iGEM (2017) [5] suggested purification is preferential but not essential and a tag that is not removed may detrimentally impact Cas13a function. Mutagenesis of Cas13a appeared to work, although lacking sequence verification; however, mutagenesis of pET28_CD was initially unsuccessful but repeats showed low yield of the DNA at the predicted size. Gibson assembly using a 1:2 ratio was unsuccessful which is likely due to the low concentration of pET28_CD_M or incorrect mutagenesis. The source of the issue is unknown due to lack of sequence verification.
Experimental Continuation
Purifying Cas13a
Due to the timescale of this iGEM project and evidence that purification is not always necessary, Cas13a was to be used in an impure solution. Future work, however, should aim to obtain pure Cas13a to reduce potential confounding effects. However, as a tag may prevent Cas13a from functioning it is more suitable to design a construct with a cleavage site between the Cas13a and tag. Example constructs include Cas13a -> thrombin site-> strep-tag or Cas13a -> SUMO site -> his-tag. A second round of purification, if required, could be achieved by size exclusion or cation exchange chromatography.
CPLX1 mRNA Detection
Through knowledge gained from our CRISPR kinetics model, we designed a different trans-cleavage assay protocol based on established methodology [3]. As the model predicted that maximum GFP concentration would be reached within 30 seconds for all assayed CPLX1 mRNA concentrations, the assays should be ran for 60 seconds. This was to increase the likelihood of maximum GFP being reached in case the model was slightly incorrect.
Furthermore, during model testing it was noted that changing quenched GFP concentration had a negligible impact on maximum active GFP concentration; as such quenched GFP concentration can be reduced from 125nM to 20nM. By doing this, the cost per reaction can be reduced and more assays can be completed.
Finally, the model also predicts a dose-dependant response in that more CPLX1 mRNA correlates to higher GFP concentration. As such, fluorescence of a person without Parkinson's Disease would be predicted to be greater than an individual of with Parkinson's Disease providing sample CPLX1 mRNA concentrations mirror the literature.
Higher Throughput with Automation
As the detection assay would be carried out in a 96-well plate, it is possible to automate the system using liquid handling robots. The use of liquid handling robot provides such as ensuring accurate pipetting and reducing risk of RNase contamination. Accurate pipetting is important as it ensures each system contains equal amounts of Cas13a and crRNA - two factors that the model indicated as being highly important. RNase contamination is a big factor in a SHERLOCK system. The presence of RNase in a sample may impact results in two ways: the first is that the RNase will degrade CPLX1 mRNA in the sample meaning less is available to activate Cas13a, the second is that the RNase may also cleave the quenched GFP to produce active GFP.
Regardless of which occurs, results will be impacted and would likely cause increased false negatives. By using automation, the risk of RNase contamination can be reduced by ensuring only SHERLOCK assays are completed with the robot and that it is regularly cleaned with RNase degrading product. Furthermore, RNase are commonly found on the skin and so reducing human contact with the sample further reduces risk of contamination.
A final benefit of using automation for the SHERLOCK assay is that it allows higher throughput as plates can be filled quicker and more accurately than if completed by a scientist. This also improves the workflow by providing the researcher with more time to prepare the samples. Higher throughput can be capitalised on two ways, either by increasing the number of people screened in a day or by increasing the confidence of the result by completing more replicates. For an indicative Parkinson's Disease test, the latter is more recommended due to the potential impacts of false diagnoses - as discussed in our Human Practices.
Attributions
- Content: Connor Trotter
- Figures: Connor Trotter
- Supervisory help: Jasmine Bird, Dr Jon Marles-Wright, Dr Will Stanley
- Proofreading: Connor Trotter, Jasmine Bird, Bradley Brown
References
- Lahut S, Gispert S, Ömür Ö, Depboylu C, Seidel K, Domínguez-Bautista JA, et al. Blood RNA biomarkers in prodromal PARK4 and rapid eye movement sleep behavior disorder show role of complexin 1 loss for risk of Parkinson’s disease. Dis Model Mech. 2017;10: 619–631. doi:10.1242/DMM.028035
- GBD 2016 Parkinson’s Disease Collaborators. Global, regional, and national burden of Parkinson’s disease, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol. 2018;17: 939–953. doi:10.1016/S1474-4422(18)30295-3
- Gronowski AM. Who or What is SHERLOCK? EJIFCC. 2018;29: 201–204. Available: http://www.ncbi.nlm.nih.gov/pubmed/30479604
- Gootenberg JS, Abudayyeh OO, Lee JW, Essletzbichler P, Dy AJ, Joung J, et al. Nucleic acid detection with CRISPR-Cas13a/C2c2. Science. 2017;356: 438–442. doi:10.1126/science.aam9321
- Abudayyeh OO, Gootenberg JS, Essletzbichler P, Han S, Joung J, Belanto JJ, et al. RNA targeting with CRISPR-Cas13. Nature. 2017;550: 280–284. doi:10.1038/nature24049
- Munich iGEM. 2017 [cited 26 Sep 2019]. Available: https://2017.igem.org/Team:Munich/Cas13a