Progress Indicator Animation

Summary
Our Troygenics are a versitile yet specific tool for the transformation of eukaryotic cells and their potential applications are wide spread. For our proof of concept we decided to focus on the application to target and specifically attack eukaryotic pathogens, especially pathogenic fungi, by the introduction of a Cell Death inducing System (CeDIS). The CeDIS is based on a Cas13a, which induces colleteral cleavag after recognition of a specific guide RNA. We designed and constructed a guide RNA array fot both of our model organisms Saccharomyces cerevisiae and the filamentous fungi Aspergillus niger. After the cloning process we transformed S.cerevisae with three different Cas13a variants Cas13a Leptotrichia wadei (Lwa), Leptotrichia buccalis (Lbu) and Leptotrichia shahii (Lsh). We performed growth experiments for all three and were able to verify the activity of both Cas13a Lbu and Cas13a within yeast. In summary, we delivered a proof of concept, that CeDIS has the potential to be used as an efficient method to induce the death of a targeted cell. Both Cas13a Lbu and Cas13a Lsh show a high potential to be used in this context and are viable options for the implementation of our CeDIS.

	Plasmid design and construction
	In virto Cas Analysis
	CeDIS proof of concept
	Fungicide simulation

Plasmid design and construction

Design of the Basic Insert

As the CeDIS plasmid is a rather complex construct, which requires many cloning steps, we first designed a “basic Insert”. This basic insert is the frame for the guide RNA array as well as the Cas13a, which will be added later on. It represents a cascade of the needed promoter for the translation of the Cas System and the corresponding terminator, followed by the consecutive promoter for the gRNAs and the corresponding terminator. Furthermore, a spacer has been added between the first terminator and second promoter, to ensure the correct structure of the terminator, and avoid any interactions between them.

As a strong constitutive promoter for the expression of Cas13a in S. cerevisiae, we considered the use for the final application of the Troygenics the pTDH3 which is the strongest native promoter under almost all conditions. Additionally, it displays a high stability throughout the cultivation (Xiong et al., 2018). It has been demonstrated that the use of the pTDH3 promoter combined with the TPS1 Terminator, leads to the highest expression rates, based on that we decided to use this combination for the highest possible expression of Cas13a in S.cerevisiae in future applications (Yamanishi et al., 2011).

For the assays to prove the functionality of the CeDIS, however we used the GALL promoter, a shortened version of the GAL1 promoter. pGAL1 is the promoter controlling the gene encoding galactokinase in S. cerevisiae (Johnston & Davis, 1984). It is induced by addition of galactose or its analogue Isopropyl-β-D-thiogalactopyranosid (IPTG) and almost completely repressed by glucose (Peng et al., 2018). pGALL is missing one of the three upstream activating sequences (UAS) and is therefore fully inactivated in the presence of glucose (Mumberg et al., 1994) which makes it the perfect candidate for testing our CeDIS.

Previously, Cas13a has only been used in fission yeast (Jing et al., 2018) but never in S. cerevisiae. Hence, there are no records of promoters and terminators used for the expression of single guide RNAs in S. cerevisiae. In fission yeast however, the same promoter/terminator combination has been used for Cas13a guide RNAs, that has been used for Cas9 guide RNAs before (Jacobs et al., 2014). Therefore, we decided to use the SNR52 promoter and the SUP4 terminator for the Cas13a guide RNAs, which are commonly used for Cas9 guide RNAs in S. cerevisiae (Bao et al., 2015; DiCarlo et al., 2013; Horwitz et al., 2015; Stovicek et al., 2015).

Plasmid map of pRS304 containing the intact TRP1 gene, which serves as a selection marker when using a TRP auxotrophic yeast strain

In our initial design, we included inverted terminal repeats as flanking sequences right before the first promoter and behind the second terminator, to achieve an integration into the genome, as well as an overlap to pSB1C3, to enable the integration into the backbone via Gibson assembly.

However, based on the advice of Lara Petersen we eventually decided to use a Tryptophan auxotrophic yeast strain and a TRYP gen as a selective marker and integration mechanism. This leads to a higher transformation efficiency and makes the selection process of yeast significantly easier.

Therefore, we decided to integrate and test our basic insert in the yeast transformation vector pRS304 instead of pSB3C1, as we would be able to start testing right away once the Cas and gRNA were integrated into the basic insert. Additionally, we decided to put effort into cloning the TRYP auxotrophic marker from pRS304 into pSB1C3 in order to make the iGEM Backbone feasible for the transformation of yeast.

Cas13a

For our CeDIS we intended to use a system based on CRISPR/Cas13a – a protein that is commonly used to cause collateral cleavage of RNA in prokaryotic cells (Abudayyeh et al., 2016). Our project however, set out to harness the power of Cas13a as an application in eukaryotic cells.

Cas13a consists of a recognition site including crRNA, which binds to complementary ssRNA within the cell. Upon binding, the nuclease of the Cas13a complex is activated. This results in collateral RNA cleavage, ultimately leading to cell death (Liu et al., 2017).

Within the organisms it has been found in so far, for example Leptotrichia wadei (Lwa), Leptotrichia buccalis (Lbu) and Leptotrichia shahii (Lsh), it has been assumed to serve as an immune system inducing cell death upon infection with RNA viruses, or, more probable, upon expression of dormant DNA viruses (Koonin & Zhang, 2017). It is known to be able to differentiate very similar RNA strains, in some systems even single nucleotide polymorphisms (Gootenberg et al., 2017). Therefore, Cas13a can be considered to be highly specific and can be used to differentiate between species. The Cas13a versions we used during our project were kindly provided to us by iGEM Munich. We decided to test Cas13a Lwa (BBa_K2323000), Cas13a Lbu (K2926001) and Cas13a Lsh (K2926000).

Design of guide RNA arrays

In order to ensure that our CeDIS is only activated within the target organism, we constructed a target specific guide RNA array. As the first step, we had to find appropriate genes to target. We used several databases to find essential genes of our model organisms, which we then used to construct the appropriate guide RNAs.

S. cerevisiae

S. cerevisiae is used as a model organism, as it is one of the best studied eukaryotic organisms and information about its genome and mechanisms are readily accessible. For the gRNA design the goal is to not only target essential but also yeast specific genes, to achieve the maximal effectiveness of our CeDIS. Using the Saccharomyces Genome database we set up a list of all the essential S. cerevisiae genes, using the YeastMine application.
YeastMine enables the generation of a filtered list of genes. In order to get a list of only essential genes, the template “Phenotype→Genes” was selected. This template allows us to retrieve genes that are annotated to a specified phenotype. In this case, the phenotype “inviable” was selected, which generates a list of all genes in which a null-mutation leads to an inviable cell. That way, we received a list of all essential yeast genes.

Additional to the fact that the genes are absolutely necessary for the organism to survive, our project is also aiming to create a very specific and selective CeDIS. To ensure that the CeDIS is only activated within the target cell we had to construct unique guide RNAs for the targeted organism. To find genes which are essential but also unique to yeasts, a blast search of all the essential genes using the NCBI-Database (February 2018) was performed. The results of the search were annotated using a python script kindly provided to us by our advisor Dr. Boas Pucker. We selected the seven genes, which generated not only the least amount of hits but also only hits which were found in closely related yeast-species.
The targeted genes for S. cerevisiae are: ATP16, BRR6, FAP7, LCP5, RQN1, SLA2 and SWP1.

Furthermore, we used a short python script to generate every possible guide RNA with a length of 21 nucleotides for each of the target genes. Afterwards, another blast search was performed for each guide RNA with a wordsize of 7, to see if it would specifically bind to the yeast gene and cannot be found in any other organism. In the initial blast search S. cerevisiae was excluded as it was only interesting to see in which other organisms the guide RNA would bind. As the CeDIS consists of a highly specific Cas13a System, a guide RNA was deemed appropriate if it showed a maximum similarity of 85% during the blast search, as the Cas is not activated at that level of variation (Gootenberg et al., 2017).

Mechanism of construction of target specific gRNAs.
First all possible gRNAs with a length of are constructed via a python script. Then an extensive blast search is performed. if the maximum similarity of a guide RNA is 85% or lower it is a suitable candidate and can be used.

Following this approach, we constructed seven suitable and specific guide RNAs for S. cerevisiae. The sequence of each guide RNA is as following:

Sequence of the chosen guide RNAs for *S. cerevisiae* for each gene, as well as the highest BLAST hit in other organisms
Gene	gRNA sequence	highest Blast hit
ATP16	ATTTCGTCGCTAAGCGTTCAT	Accession:CP01924.1 Coverage: 85%
BRR6	GAGCCCAAGTTTAATTGCGGA	Accession:CP033737.1 Coverage: 85%
FAP7	GGGTGTCGTCAAACAGGATGC	Accession: CP032429.1 Coverage: 85%
LCP5	CTGCTGAGAAAGTATCGCTGC	Accession: LL596195.1 Coverage: 85%
RQN1	AGAGGGTTTGACGTAGGGACT	Accession: LR53582.3 Coverage: 85%
SLA2	ATTCTGGCGGTTCGTCGTACA	Accession: CP033824.1 Coverage: 85%
SWP1	GAATAGGAATCGATGAAACTG	Accession: CP020857.1 Coverage: 85%

Aspergillus niger

Since we also aim to establish our Troygenics as an effective transformation for a variety of complex eukaryotic cells and with regards to CeDIS decided to focus on pathogenic fungi as a potential target. It was important for us to test our system on more complex fungi. Based on the advice of Prof. Dr. Mark Varrelmann we decided to use Aspergillus niger as a second model organism, as the filamentous growth of A.niger resembles the growth of plant pathogens more closely.

It would be good to expand your proof of concept from yeast to a filamentous fungi, which resembles the growth of pathogenic fungi more closely. Aspergillus niger would be a suitable candidate for that purpose.

Prof. Dr. Mark Varrelmann
Institut of sugarbeet research at Universität Göttingen

Even though, A. niger is a commonly used organism and rather well studied than other filamentous fungi, the abundance of information regarding its genome was rather sparse compared to S. cerevisiae. This circumstance allowed us to test our method of gRNA construction with only few known essential genes, which makes the task even more challenging. We constructed the guide RNAs in the same manner as previously described for S. cerevisiea , with the difference, that we used the Aspergillus Genome Database to determine the known essential genes of our model fungus. The following genes are essential for A. niger and have been chosen as target genes: gmtA, hemF, rmsA, secA and hemH.
We then proceeded with the construction of all potential guide RNAs for each essential gene and then narrowing down the options to six gRNAs via an extensive BLAST search.

Through this process we determined the following gRNAs for A.niger. Due to the unavailability of information regarding essential genes we constructed two guide RNAs for the gene HemF, as this gene showed more variation to other closely related species and was more viable for the construction of unique gRNAs as other essential genes.

Sequence of the chosen guide RNAs for *A. niger* for each targeted gene, as well as the highest BLAST hit in other organisms
Gene	gRNA sequence	highest Blast hit
gmtA	GTCACTTCTTCGACCACTTAC	Accession:LM002073.1 Coverage: 85%
hemF 1	TACTCCTCGGAACAGAAGCCT	Accession:LR584072.1 Coverage: 85%
hemF 2	TTCTCATCTCGGACCTGTCAC	Accession: Xm_018616956.1 Coverage: 85%
rmsA	CTACCAGCTGGGTGTTGCCCG	Accession: CP037423.1 Coverage: 85%
secA	TGGCGCGGTTCATCCAGAATG	Accession: CP031068.1 Coverage: 85%
hemH	CGAATTCCCCGGGTGTCAAGC	Accession: LR597559.1 Coverage: 85%

Construction gRNA array

In a gRNA array the gRNA is flanked by two double repeats, which are used as spacers in the processing of the gRNAs and assist in the binding of the Cas13a to the target RNA. Furthermore, they have a highly repetitive sequence, which leads to problems when attempting to synthesize the array, as there are too many repetitions within the array.

Therefore, we decided to use a Golden Gate Cloning approach to construct the array. In order to do so, we designed the gRNAs and double repeats as oligo fragments with unique overlaps to ensure the correct order. These have been ordered as a forward and reverse primer and then annealed and a phosphorylation was performed. In that way we were able to generate all fragments needed and set off to fuse them together via Golden Gate. We chose an approach which added one gRNA fragment at a time. Furthermore, we split it up onto two backbones, which allowed us to clone two guide RNAs at the same time and speed up our process.

Concept of the Golden Gate stratey used to clone the gRNA array.

Following this approach, we successfully cloned the guide gRNAs 1-3 (K2926008) and 4-7 (K2926010) for S. cerevisiae, with the needed double repeat spacers into the pSB3C1 backbone and confirmed the correct assemble via Sanger sequencing. Additionally, we have sucessfully assembled the full gRNA array containing both gRNA 1-3 and 4-7 in one backbone (K2926009) via Gibson Assembly.

Assembled guide RNAs for S. cerevisiea. It shows the first two steps of creating an array with four (1) and three (2) guide RNAs, as well as the fully assembled gRNA array containig all seven (3).

For A.niger we also successfully cloned gRNAs 1-3 within pSB3C1 (K2926011), however, due to the limited time available during the iGEM competition we decided to focus on the establishing of CeDIS within yeast.

Assembled guide RNAs for *A. niger* . It shows the first three guide RNAs assembled in pSB1C3.

Plasmid assembly

After the design and construction of the different parts the next step was to construct the full plasmid. As a first step, we assembled the entirety of CeDIS in pSB1C3. As we designed the basic insert with an overlap to pSB1C3 we performed a Gibson assembly to add it to the plasmid. Afterwards the three versions of Cas13a were added via another Gibson assembly.

As Cas13a is not active without the detection of a guide RNA, the previously described gRNA assay was added between the consecutive promoter and the corresponding terminator. After this the CeDIS was fully constructed within pSB1C3. As previously mentioned, we decided to embed our CeDIS into the common yeast transformation vector pRS304, as it contains a tryptophan gene, which can easily be used as a selection marker, when using a TRP auxotrophic strain. We cloned the fully assembled basic insert into pRS304 with one simple Gibson assembly. After the confirmation of the correct sequence via Sanger sequencing our CeDIS was ready for testing.

Besides, the CeDIS application we also would to establish our Troygenics as a novel transformation method for a variety of organisms. As a proof of concept for the functionality of the basic insert and to establish the Troygenics as a transformation method with an application in the laboratory, the Lab application plasmid was constructed.

The lab application consists of the basic insert containing a sfGFP right behind the inducible promoter. It does not contain the second promoter or terminator. It allows us an easy detection of the expression via fluorescence.

In vitro Cas13a analysis

In order to ensure the functionality of the Cas13a variants which iGEM Munich kindly provided us, we decided to perform an in vitro analysis of the Cas13a activity. Therefore, we cloned all three Cas systems, Cas13a Lwa, Lbu and Lsh, into pTXB1 and used E.coli ER2566 for expression. The protein was purified using the IMPACT-Kit from NEB. The Bradford assay and a subsequent SDS Page showed that we were able to purify Cas13a Lsh with a molecular weight of 166.2 kDa and a yield of 42.7 µg out of 1.39 g cell mass, Cas13a Lbu with a molecular weight of 138.468 kDa and a yield of 1.09 mg out 1.50 g of purified protein, Cas13a Lwa with a molecular weight of 143.7 kDa and a yield of only 3.1 ng out of 2.11g cultivate.

SDS-PAGEof purified Cas13a proteins Lsh, Lbu and Lwa. Marked bands indicate the further used proteins bands for analysis by MALDI-ToF MS/MS.

To further analyze the expressed Cas proteins and compare them to the expected protein sequence, the marked bands were excised from the SDS-PAGE, washed, digested with trypsine and analyzed in a MALDI-ToF MS/MS. The generated mass spectra and mass lists were evaluated using the software BioTools.

Mass spectrum of the proteins Cas13a Lbu (1) and Cas13a Lsh (2) after tryptic digestcompared to the theoretical mass spectrum. Excised bands from the SDS-PAGE of Cas13a Lbu and Cas13a Lsh were washed, digested over night with trypsine and co-cristallyzed with a α-Cyano-4-hydoxycinnamic acid-matrix on a MALDI target. The mass spectrum was recorded in a MALDI-ToF MS from Bruker Daltronics and data was evaluated using the software BioTools.

Using this technique, we successfully confirmed that Cas13a Lbu and Lsh were expressed and purified from the expression strain. For the Cas13a activity in-vitro assay we designed single guide RNAs (gRNAs) targeting a RFP gene. The following gRNA was ordered via RNA synthesis from IDT.

CAAAGCTTACGTTAAACACCCGATTTAGACTACCCCAAAAACGAAGGGGACTAAAAC

The target RNA was isolated from an overnight culture of E. coli DH5α with pSB1K3_RFP, purified with the RNA isolation kit from ZYMO Research. The activity of the Cas protein was determined using our Cas13a activity assay protocol based on the RNAse Alert-Kit by Thermo Fischer and evaluated via fluorescent measurement with a plate reader. We tested the Cas with the gRNA and the target RNA as well as the Cas and gRNA without the target RNA, to account for any offsite effects. Due to the low yield we did not conduct the experiment with Lwa. Lbu has been used in a concentration of 2.3 µM, while the concentration of Lsh was 0.08 µM.

*In vitro* activity of Cas13a Lsh and Cas13a Lbu using the Cas13a activity array based on the RNAse Alert kit (Thermo Fischer). Development of the fluorescence intensity based on the activity of Cas13a Lbu and Lsh over a period of 12 h. For each Cas variant the activity was measured with all essential parts for the activation of Cas13a including the Cas13a protein, gRNA and the target RNA (for Lsh shown in dark purple and Lbu in dark blue) To show any possible offsite events, the activity was also measured without providing of the target RNA (for Lsh pictured in red and Lbu in light blue).

We validated the functionality of both Cas13a Lsh and Lbu. Lsh as well as Lbu show an activity. However, Lsh showed a higher activity than Lbu. Furthermore, the negative controls without any target RNA also showed a slight increase of fluorescence intensity. Thereby, the Lbu negative control showed a higher activity than the Lsh control. The activity without the target RNA present can indicate offsite activity but it can also be influenced by airborne RNAse. As both of the received parts are functional, we performed growth experiments with the complete CeDIS system in pRS304 in S. cerevisea INVSc1. Additionally, we also transformed S. cerevisiae with pRS304 Cas13a Lwa to perform growth experiments and test its functionality.

CeDIS proof of concept

In order to test the functionality of our CeDIS system, we conducted growth experiments with INVSc1, S. cerevisiae yeast strain containing the system. The strain carries a tryptophan autotrophy. All experiments for the proof of concept were performed with the CeDIS encoded on pRS304. The yeast was grown in liquid SD media without tryptophan, to ensure the selective growth of transformed yeast.

Our initial tests were conducted with yeast, where the cultures had been grown on YPD media. Once an OD of 0.8 had been reached the cells were washed and a medium change was performed. Afterwards the cells were induced by the usage of an YP medium containing galactose as a carbon source. In this initial test Cas13a Lwa and Cas13a Lbu in pRS304 were tested.

Growth experiment of *S.cerevisea INVSc1* transformed with pRS304_Lbu to assess the functionality of the CeDIS. All yeasts were cultivated were cultivated on glucose containing SD medium in 250 mL shaker flasks at a temperature of 30°C and 180 rpm. The results were normalized to the starting value of each culture to allow for a better comparison between them. As a control the wild type S. cerevisea was grown on glucose (red) containing media, and equal treated to the galactose induced cells (dark purple). S. cerevisea transformed with pRS304_Lbu was grown on the inhibitor glucose (dark blue), to show the effects in an undinduced state. Furthermore a galactose induced culture containing pRS304_Lbu (light blue) was measured.

Growth experiment of *S.cerevisea INVSc1* transformed with pRS304_Lwa to assess the functionality of the CeDIS. All yeasts were cultivated were cultivated on glucose containing SD medium in 250 mL shaker flasks at a temperature of 30°C and 180 rpm. The results were normalized to the starting value of each culture to allow for a better comparison between them. As a control the wild type S. cerevisea was grown on glucose (red) containing media, and equal treated to the galactose induced cells (dark purple). S. cerevisea transformed with pRS304_Lbu was grown on the inhibitor glucose (dark blue), to show the effects in an undinduced state. Furthermore a galactose induced culture containing pRS304_Lwa (light blue) was measured.

During the growth experiments there was no significant difference between the growth of S. cerevisea with and without the Cas13a protein. For both the growth on galactose however is significantly decreased than it is on glucose. After the growth experiment the presence of both variants of Cas13a was verified by colony PCR.

In order to figure out why there was no difference in growth with and without the presence of the Cas13a protein, we tested the Lab application in both glucose containing medium and galactose containing medium, as well as simulated a change of medium similar as performed in the growth experiments. As the Lab application plasmid contains a fluorescent marker, the activation of the GALL promoter was easily detectable via a plate reader.

*S. cerevisiae* transformed with the Lab application was cultivated in SD media containing raffinose (Raff), glucose (Glu) or galactose (Gal) at 30 °C and 180 rpm. Samples marked with ind. were induced with 2 % galactose at an OD of 0.4. The measurement took place 2 h after induction. Three samples were taken of each culture and the fluorescence intensity was measured using the TECAN plate reader infinite M200 (λEx=449 nm, λEm=525 nm, gain calculated from 2.5 µM fluorescein). The results were normalized to the max. fluorescence intensity of cells continuously grown on raffinose.

The results indicate, that there is a slight fluorescence in the presence of glucose (in Fig. 14 shown in red). Especially interesting is the sample with the medium exchange from glucose containing medium to galactose containing medium, as there is no significant increase in fluorescence intensity detectable (Fig. 14 shown in dark purple). The promoter activity is almost completely inhibited in the presence of glucose and does not show any activation 2 h after induction. In galactose however, a clear fluorescence intensity, proving the functionality and its proper induction of the promoter in yeast (Fig. 14 shown in light blue). This is due to the fact, that the GALL promoter is strongly inhibited by glucose and the activation by galactose proceeds slow (Hovlanda et al., 1989). To avoid the slow activation, we tested the use of raffinose in the media for the growth of the cultures and added galactose when higher cell densities were reached to induce the GALL promoter. In comparison to the induction after growth in glucose medium, with the growth on raffinose there is a strong increase of fluorescence intensity when induced afterwards (Fig. 14 shown in dark blue). It is shown, that the cultivate in raffinose instead of glucose is beneficial for the fast and efficient induction of the promoter

According to these results, the cultures were grown in an SD medium with raffinose as sole carbon source and without tryptophan. Even though the inhibition of the GALL promoter with raffinose is weaker than with glucose, the activation by galactose is much faster, which is desirable for our purpose. Furthermore, based on the advice of experts the activation will occur at an OD600 of 0.4.

Growth experiment of S. cerevisea INVSc1 transformed with pRS304_Lbu to assess the functionality of the CeDIS. All yeasts were cultivated on 2% raffinose containing SD medium in 250mL shaker flasks at a temperature of 3°C and 180 rpm. As a control, the wild type S. cerevisea was cultivated in raffinose (red) containing mediumand treated equal to the 2% galactose induced cells (dark purple). S. cerevisea transformed with pRS304_Lbu was grown on selective raffinose containing SD medium (dark blue). Furthermore a galactose induced cells containing pRS304_Lbu (light blue) were cultivated.

Growth experiment of S.cerevisea INVSc1 transformed with pRS304_Lwa to assess the functionality of the CeDIS All yeasts were cultivated on 2% raffinose containing SD medium in 250mL shaker flasks at a temperature of 30 degress celcius at 180 rpm. As a control the wild type S.cerevisea was grown on raffinose(red), as well as treated as if it would be induced with 2% galactose (dark purple). S. cerevisea transformed with pRS304_Lwa was grown on selective raffinose SD medium (dark blue). Furthermore an galactose induced culture containing pRS304_Lwa (light blue) was measured

While yeast containing Cas13a Lwa shows a slight decrease in growth, there is no indication, that the CeDIS is fully activated or functional within the cell. It has previously been reported, that Cas13a Lwa does not produce any unspecific cleavage events in eukaryotes (Cox et al., 2017). Therefore, no unspecific cleavage events occur, making Lwa unfeasible for our system (Wolter & Puchta, 2018). This function is useful when it comes to downregulation of a gene, however it is not feasible for the purpose of a complete knockout or cell death induction. However, for other Cas13a variants collateral cleavage has been described (Abudayyeh et al., 2016).
The results of the cultivation with Cas13a Lbu however shows that the cells reach a premature stationary phase. This indicates, that the growth of the yeast is decreased and indicates collateral cleavage events of the Cas13a. Over the duration of 10 hours there is no significant increase in the OD600 of the induced cells carrying the CeDIS. However, the yeast containing the Cas13a show a significant decrease of growth even with the uninduced cells. The control WT S. cerevisiae also shows a difference in growth on raffinose and galactose. The variation of the OD600 for the Cas carrying yeast can be assigned to the variation in carbon source. While these results show that the CeDIS containing Cas13a is active and effective within yeast, it also indicates that there is a background activity when raffinose is used as an inhibitor of the GALL promoter. Even in the uninduced state, Cas is expressed a on a low level , which leads to a decrease in growth, even if it is less effective than in the induced state.
This can potentially be explained by the structure of raffinose. Raffinose is a trisaccharaide consisting of galactose, glucose and fructose.

Structure of raffinose containing in this order galactose, glucose and fructose

While the galactose, which activates the promoter, is readily accessible due to its position, the inhibitor glucose however is bound by the other two saccharides, galactose and fructose, and might not be fully available for the inhibition of the promoter. This could explain why the difference between induced and uninduced growth is smaller than expected. Furthermore, after transformation the yeast cells have to produce the amino acid tryptophan on their own, as they are grown on a selective medium. This also increases the stress on the cells and could also contribute to the decrease in growth.

According to the previously obtained results we altered our experimental set up. The yeasts are further cultivated on an SD-medium containing raffinose as a sole carbon source, however 4 % (w/v) of galactose added to induce the cells, while the uninduced cultures received 4 % (w/v) of glucose. Thereby we inhibit the GALL promoter in our control samples which will lead to the cells recovering from the previous damage they received by the production of the CeDIS.
The tests have been conducted for both Cas13a Lwa and Cas13a Lsh.

Growth experiment of S. cerevisea INVSc1 transformed with pRS304_Lsh to prove the functionality of the CeDIS.
All yeasts were cultivated on 2% raffinose containing SD medium in 250mL shaker flasks at 30°C and180 rpm. As a control the wild type S. cerevisea was grown on raffinose and at the time of induction was inhibited with 4% glucose (red), as well as treated equal to the cells induced with 2 % galactose (dark purple). S. cerevisea transformed with pRS304_Lsh was cultivated in SD medium with raffinose as sole carbon source and at of the time of induction inhibited by addition of 4% glucose (dark blue). Furthermore, a galactose induced culture containing pRS304_Lsh (light blue) was cultivated.

In accordance with the previous experiments,previous experiments, Cas13a Lwa effects a significant reduction in growth, however it does not lead to a premature stationary phase. This confirms the downregulation of RNA rather than an induction of cell death. Even though, this is useful for many applications but not suitable for the design of our CeDIS. Cas13a Lsh however clearly indicates an induction of cell death, when induced with galactose, as a stationary phase with only slight fluctuations after a cultivation time of 10 h. Furthermore, the relative OD600 reached is three times smaller than the OD600 reached by the control group of WT yeast grown on raffinose with the addition of galactose. The inhibited culture containing Lsh shows a strong reduction of growth rate compared to the control WT cultivation and reaches only 50% of the maximum OD of the control but opposed to the induced sample has a constant growth. This indicates, that the culture was already struggling before the inhibition due to the background activity, but an active inhibition does relieve some of the stress on the culture. It reaches its plateau after 22h which corresponds well with the control grown on both galactose and glucose.

In summary, we showed that CeDIS has the potential to be used as an efficient method to induce the death of a targeted cell. Both Cas13a Lbu and Cas13a Lsh show a high potential to be used in this context and are viable options for the implementation of our CeDIS. However, Lsh showed a higher activity and less off target activation during the in vitro analysis and could be more suitable for the CeDIS.

Fungicide Simulation

To support our argument about the importance of specificity and the risks of resistance formation in conventional systems, we conducted a series of experiments which comprises the effects of chemical fungicides to our model organisms S. cerevisiae and A. niger. The experiment was inspired from this theory and was conducted following this protocol. We came up with the following results:

S. cerevisiae

The growth experiment of S. cerevisiae during sic hours of cultivation in yeast Peptone Dextrose (YPD) medium with different fungicide concentrations resulted in the following growth rate:

Growth rate of *S. cerevisiae* with Proline (concentrations 10x – 1:1000) with a negative control (-) which contained no fungicide in the medium.

From both diagrams, it could be concluded that even the least amount of fungicide (1:1000) would already interfere with the growth speed of S. cerevisiae compared to the negative control (-). Meanwhile, increased amounts of the fungicide (1:100 and 1:10) would slow down the growth respectively, but not enough to stop it ultimately. It is also interesting that starting from the recommended amount (1x), no growth was detectable, suggesting that the recommended amount and higher have indeed lethal effects to the growth of S. cerevisiae.
Apart from the similar tendencies that both fungicides exhibited to this organism, it is still recognizable that Folicur might be more effective than Proline, given that cultivation with Proline during 6 hours ended with a higher OD than the cultivation with Folicur (see fig. 6).
After a week of incubation at 30 °C, the YPD agar plates were also analyzed.

YPD agar plates containing plated 100 μl *S. cerevisiae* overnight culture with both fungicides at different concentrations (10x – 1:1000) and one plate without fungicide at all as negative control (-) on the right side.

From the visually identifiable growth on agar plates, it can also be concluded that the recommended amount (1x) and higher affected the growth of S. cerevisiae in a lethal way and that Proline might really be less effective than Folicur. At 1:10, both YPD Proline and YPD Folicur plates still exhibit growth, although not visible enough to be caught on camera. At decreasing rate of fungicide (1:100 and 1:1000), faint growth spurs are also visible on the YPD Folicur plates. Meanwhile, the YPD plates with Proline showed an essentially clearer growth spurs at the same fungicide amount, which furthermore supports our statement about Proline’s relatively lower activity compared to Folicur.
In accordance to the results, we started the resistance test from the concentration of 1:10 and went up further to 1:8, 1:4, 1:2, and finally 1x. After 5 days of measurement every 12 hours and inoculating into a medium with increased fungicide concentration every 24 hours, we came up with the following result:

The resulted quantitative resistance formation against Folicur from *S. cerevisiae*. The plotted OD values for each day and each concentration are based on the ODs from the start and the end of each cultivation.

The resulted quantitative resistance formation against Proline from *S. cerevisiae*. The plotted OD values for each day and each concentration are based on the ODs from the start and the end of each cultivation.

In both diagrams, the formation of quantitative resistance against both fungicides are made visible and comparable. Despite decreasing growth rate with the increasing concentration, the “evolved” S. cerevisiae cells are proved to be able to withstand the lethal concentration (1x) of both fungicides at the end of the experiment. Additionally, Proline still exhibited a more rapid growth rate at the higher concentrations, which once again could correlate to Proline’s less efficiency relative to Folicur.

Aspergillus niger

Compared to S. cerevisiae, the whole growth experiment with A. niger was more complicated due to the organism’s characteristics and slow growth rate, which interfered with the OD measurement.

Despite the circumstances, we were able to detect the growth through measurement of dry biomass.

Measured dry biomass of *A. niger* during 12 hours of cultivation with Folicur and without as negative control (-). The start weights are standardized to 0.1 mg.

Measured dry biomass of A. niger during 12 hours of cultivation with Proline and without as negative control (-). The start weights are standardized to 0.1 mg.

Like S. cerevisiae, A. niger also showed no signs of growth starting from the 1x concentration. Furthermore, the cells also exhibited a higher growth rate with Proline than with Folicur. These results finally supported the claims from the S. cerevisiae results. After some days, the growth rate on the agar plates were also evaluated.

CM agar plates containing plated 100 μL A. niger overnight culture (cultivated for about three nights) with both fungicides at different concentrations (10x – 1:1000) and without fungicide at all as negative control (-) on the left side.

From the growth of A. niger on the agar plates, growth spurs are also identifiable until the fungicide concentration of 1:10 before disappearing completely on the agar plates with 1x concentration and above. It is also noticeable that the Proline plate showed a more advanced growth than the Folicur plate. However, we observed an unusual tendency at the very faint growth spurs on the plates with 1:100 and 1:1000 fungicide concentration. One of the possible explanations for this inconsistency is the uneven dispersion of the A. niger overnight culture, which might have contributed to the lack of the cells plated on the respective agar plates.
During these experiments, we also stumbled upon an interesting result which was based on the contamination on our first batch of A. niger agar plates. Before starting over, we took time to observe the contaminants briefly and came to the assumption that the contamination might originate from colonies of unidentifiable bacteria due to the rapid growth (incubation time less than 2 days).

Interestingly, this contamination also stopped at the concentrations of 1x and 10x, which may suggest that the fungicide does not only stop at disrupting fungal growth activities, but also the growth of other organisms that happen to be around, including bacterial contaminants.
From these series of experiments, we could conclude that the recommended usage dose from Bayer AG is appropriate and sufficient to eliminate pathogens in the field. Therefore, it should not be necessary to increase this concentration and it might even be reasonable to determine the lowest effective concentration before applying the fungicide to avoid using higher concentrations unnecessarily.
On the other side, Folicur has been showing signs of superior ability to eliminate pathogens compared to Proline throughout the experiment. This superiority might be due to higher concentration aimed by the safety data sheet (Proline: 0.8 L/ha in 200 – 400 L/ha water. Folicur: 1.5 L/ha in 200 – 400 L/ha water). Furthermore, Folicur is described with more hazards than Proline according to their safety data sheets and therefore proved itself to be more toxic than Proline, which could explain the increased effectivity in eliminating pathogens.
In this case, it might be essential to plan a fungicide treatment on crops thoroughly. Apart from the severity of the infection, the effect of the fungicide usage on the surrounding environment should also be included into consideration.
Overall, it is also worth mentioning that the whole experiment took place within laboratory scales (agar plates and cultivations in 30 mL medium) with relatively simple model organisms, which obviously contributed to the rapid progresses and the possibilities to obtain results in a relatively short time. On the scale of an agriculture field, all these progresses would take a significantly longer time, also considering that fungal pathogens in the field are morphologically and physiologically more complex than S. cerevisiae or A. niger. Nevertheless, this experiment could serve as another perspective to the utilization of fungicide and the danger of quantitative resistance.

References

Abudayyeh, O. O., Gootenberg, J. S., Konermann, S., Joung, J., Slaymaker, I. M., Cox, D. B. T., Zhang, F. (2016). C2c2 is a single-component programmable RNA-guided RNA-targeting CRISPR effector. Science, 353(6299), aaf5573. Link

Bao, Z., Xiao, H., Liang, J., Zhang, L., Xiong, X., Sun, N., Zhao, H. (2015). Homology-integrated CRISPR-Cas (HI-CRISPR) system for one-step multigene disruption in Saccharomyces cerevisiae. ACS Synthetic Biology, 4(5), 585–594. Link

Cox, D. B. T., Gootenberg, J. S., Abudayyeh, O. O., Franklin, B., Kellner, M. J., Joung, J., & Zhang, F. (2017). Rna editing with CRISPR-Cas13. Science (New York, N.Y.), 358(6366), 1019–1027. Link

DiCarlo, J. E., Norville, J. E., Mali, P., Rios, X., Aach, J., & Church, G. M. (2013). Genome engineering in Saccharomyces cerevisiae using CRISPR-Cas systems. Nucleic Acids Research, 41(7), 4336–4343. Link

Gootenberg, J. S., Abudayyeh, O. O., Lee, J. W., Essletzbichler, P., Dy, A. J., Joung, J., Zhang, F. (2017). Nucleic acid detection with CRISPR-Cas13a/C2c2. Science (New York, N.Y.), 356(6336), 438–442. Link

Horwitz, A. A., Walter, J. M., Schubert, M. G., Kung, S. H., Hawkins, K., Platt, D. M., Newman, J. D. (2015). Efficient Multiplexed Integration of Synergistic Alleles and Metabolic Pathways in Yeasts via CRISPR-Cas. Cell Systems, 1(1), 88–96. Link

Hovlanda, P., Flick, J., Johnston M., & Sclafani R. A. (1989). Galactose as a gratuitous inducer of GAL gene expression in yeasts growing on glucose. Gene, 83(1), 57–64. Link

Jacobs, J. Z., Ciccaglione, K. M., Tournier, V., & Zaratiegui, M. (2014). Implementation of the CRISPR-Cas9 system in fission yeast. Nature Communications, 5, 5344. Link

Jing, X., Xie, B., Chen, L., Zhang, N., Jiang, Y., Qin, H., Li, X. [Xuan] (2018). Implementation of the CRISPR-Cas13a system in fission yeast and its repurposing for precise RNA editing. Nucleic Acids Research, 46(15), e90. Link

Johnston, M., & Davis, R. W. (1984). Sequences that regulate the divergent GAL1-GAL10 promoter in Saccharomyces cerevisiae. Molecular and Cellular Biology, 4(8), 1440–1448. Link

Koonin, E. V., & Zhang, F. (2017). Coupling immunity and programmed cell suicide in prokaryotes: Life-or-death choices. BioEssays : News and Reviews in Molecular, Cellular and Developmental Biology, 39(1), 1–9. Link

Liu, L., Li, X. Xueyan, Ma, J., Li, Z., You, L., Wang, J., Wang, Y. (2017). The Molecular Architecture for RNA-Guided RNA Cleavage by Cas13a. Cell, 170(4), 714-726.e10. Link

Mumberg, D., Müller, R., & Funk, M. (1994). Regulatable promoters of Saccharomyces cerevisiae: Comparison of transcriptional activity and their use for heterologous expression. Nucleic Acids Research, 22(25), 5767–5768. Link

Peng, B., Wood, R. J., Nielsen, L. K., & Vickers, C. E. (2018). An Expanded Heterologous GAL Promoter Collection for Diauxie-Inducible Expression in Saccharomyces cerevisiae. ACS Synthetic Biology, 7(2), 748–751. Link

Stovicek, V., Borodina, I., & Forster, J. (2015). CRISPR–Cas system enables fast and simple genome editing of industrial Saccharomyces cerevisiae strains. Metabolic Engineering Communications, 2, 13–22. Link

Wolter, F., & Puchta, H. (2018). The CRISPR/Cas revolution reaches the RNA world: Cas13, a new Swiss Army knife for plant biologists. The Plant Journal, 94(5), 767–775. Link

Xiong, L., Zeng, Y., Tang, R.‑Q., Alper, H. S., Bai, F.‑W., & Zhao, X.‑Q. (2018). Condition-specific promoter activities in Saccharomyces cerevisiae. Microbial Cell Factories, 17(1), 58. Link

Yamanishi, M., Katahira, S., & Matsuyama, T. (2011). Tps1 terminator increases mRNA and protein yield in a Saccharomyces cerevisiae expression system. Bioscience, Biotechnology, and Biochemistry, 75(11), 2234–2236. Link

Team:Bielefeld-CeBiTec/CeDIS

Plasmid design and construction

Design of the Basic Insert

Cas13a

Design of guide RNA arrays

S. cerevisiae

Aspergillus niger

Construction gRNA array

Plasmid assembly

In vitro Cas13a analysis

CeDIS proof of concept

Fungicide Simulation

S. cerevisiae

Aspergillus niger