Team:Freiburg/Project/crrna

From 2013.igem.org

Revision as of 16:36, 4 October 2013 by Bea (Talk | contribs)


CrispR

Targeting

In our uniCAS Toolkit we engineered the CRISPR/Cas9 system for future applications in regulating gene expression. The key components for the binding of the catalytically inactive Cas9 (dCas9) to its targets are two small, non-coding RNAs: the CRISPR-RNA (crRNA) and the tracrRNA. These RNAs that guide our dCas9 to specific DNA sequences have to be cotransfected with the Cas9-plasmids. Therefore we designed an RNA plasmid, termed RNAimer , which contains all required RNAs for efficiently guiding the dCas9 protein to the required DNA - even if multiple DNA sites should be targeted.

As an essential part of our toolkit is the binding of our protein to DNA, we evaluated this in detail: Various DNA target sequences were compared (such as targeting different loci or varying the GC content of the taregt DNAs) to evaluate the best guiding crRNAs. Based on these results, we tested if dCas9 can deal with multiple targets .

In order to facilitate finding potential target sequences, we programmed the crRNA design tool and provide it to te iGEM community. As the binding of the RNA-guided Cas9 requires a protospacer adjacent motif (PAM), the tool determines all possible regions of the desired and pasted DNA sequence. Eventually, you obtain the position of your target sites and the sequences of the oligos that have to be inserted into the RNAimer - ready to order!

RNAimer - targeting dCas9 to its destination

Introduction

As dCas9 requires special RNAs for binding to the DNA, we designed an RNA plasmid containing the tracrRNA, where the crRNA can be introduced easily by digesting with BbsI and inserting two previous annealed oligos. Two of these RNA plasmids (with different crRNAs) can be fused using the iGEM biobrick system. This way it is possible to get two or more crRNAs on one plasmid.

Fig. 1: RNAimer (BBa_K1150034)
Our RNA plasmid contains the tracrRNA and a site where the desired crRNA can be inserted. Both RNAs are driven by different human RNA polymerase III promoters. Assembly of multiple crRNAs can be easily done by digestion with enzymes of prefix and suffix and combining these parts according to iGEM standard assembly.

As it is important that the RNAs are not being marked for protein expression the RNA polymerase III is required for transcription. RNA polymerase III mainly synthesize small non-coding RNAs (e.g. tRNAs or rRNAs) whereas the commonly used polymerase II is responsible for transcription of mainly mRNAs [1,2] . We chose the human U6- and H1-promoter to drive the RNAs as they are exclusively recognized by polymerase III [2] . In order to reduce the amount of plasmids for transfection when intending to target several genes or target sites at once, we wanted to join the required crRNAs on one RNA plasmid. This was easily manageable by using the iGEM standard assembly: We inserted the oligos seperately into BBa_K1150034 by digesting with BbsI and ligation. Afterwards the whole inserts can be combined by using the restriction enzymes of the prefix and suffix (Fig. 6).

Figure 6: Assembly and function of the RNAimer
For each desired crRNA one 'RNA plasmid' (=containing the crRNA after ligation) has to be opened by digestion with BbsI. There the annealed crRNA oligos will be inserted. After that the different RNA inserts can be assembled using the idempotent iGEM cloning strategy. So three different crRNAs will be transcribed.

Results

When compared to two RNA plasmids containing the same crRNAs the RNAimer causes the same SEAP activation in HEK cells transfected with this plasmid (s), CMV:dCas9-VP16, Renilla luciferase and the SEAP reporter plasmid (Fig. 7).
Figure 7: RNAimer in comparison to two RNA plasmids
SEAP activity was divided through luminescence intensity of Renilla luciferase. The bars represent the mean of biological triplicates with standard deviation. All samples were transfected with CMV:dCas9-VP16 in pSB1C3 and with two (left) or one RNA plasmid (right), each time coding for the same crRNAs. T3+4: RNAimer with T3 and T4 crRNA; T3 & T4 two different RNA plasmids with T3 and T4 crRNA.

Discussion

We have shown that our RNA plasmid, the RNAimer, works as expected. The combination of two targets does not seem to affect the efficiency of targeting as no difference is visible when transfecting two separate plasmids or combining both targets in one plasmid. With our toolkit it is possible to induce a gene by transfecting only two plasmids (dCas9 with effector and RNAimer) that can be recombined with no effort in accordance to the experimental setup.
For future application researchers will be able to render gene expression by transfecting even only plasmids containing the crRNA: when dCas9 will be stable integrated into the genome of cells or model organisms, as recently done by Gilbert et al. [4].

References

(1) Dieci G,et al. (2007). The expanding RNA polymerase III transcriptome. Trends Genet. 2007 Dec;23(12):614-22.
(2) Myslinski E, et al. (2001). An unusually compact external promoter for RNA polymerase III transcription of the human H1RNA gene. Nucleic Acids Res. 2001 Jun 15;29(12):2502-9.
(3) Cheng AW, et al. (2013). Multiplexed activation of endogenous genes by CRISPR-on, an RNA-guided transcriptional activator system. Cell Res. 2013 Aug 27. doi: 10.1038/cr.2013.122.
(4) Gilbert LA,et al. (2013). CRISPR-Mediated Modular RNA-Guided Regulation of Transcription in Eukaryotes. Cell. 2013 Jul 18;154 (2):442-51.

Multiple Targeting

Introduction

One of the biggest advantages of the CRISPR/Cas9 system compared to other transcription activators (e.g. Zn fingers, TALEs) is that only one protein is required for targeting several DNA sites: For a new target there has to be just another crRNA. So a RNA plasmid was designed containing the tracrRNA, where the crRNA can be easily introduced.
With this RNA plasmid and another plasmid containing the dCas9-effector fusion protein it is possible to target several DNA sites at once by transfecting only two plasmids. This could mean the simultaneous regulation of different genes or a stricter controlling of one gene by bringing more effector domains to this gene.

But at first we had to evaluate the adequate DNA sequences for targeting, because after our first experiments we recognized that different target loci cause different effects. Thus, we cloned 5 different target sites 26 bp upstream of the CMVmin promoter of a SEAP reporter plasmid. Additionally we designed crRNAs in various distances upstream of this promoter (Tab. 1). This target sites were evaluated by activating SEAP expression with dCas9-VP16.

Table 1: Different target sites

The VEGF and EMX1 target sites were cloned into the SEAP reporter plasmid. T2 till T4 are original sequences of this plasmid. All target loci are upstream of the CMVmin promoter. VEGF and EMX1 are parts of the sequence upstream of the human VEGF or EMX1 gene, respectively.
Name Distance to promoter Sequence GC content [%]
VEGF VZ-573
-26 bp
GTGTGCAGACGGCAGTCACTAGGGGGCGCT
70
VEGF VZ-475
-26 bp
GTGAGTGTGTGCGTGTGGGGTTGAGGGCGT
63
VEGF VZ-8
-26 bp
TTAAAAGTCGGCTGGTAGCGGGGAGGATCG
57
VEGF VZ+434
-26 bp
GACCTGCTTTTGGGGGTGACCGCCGGAGCG
70
EMX1
-26 bp
GGAAGGGCCTGAGTCCGAGCAGAAGAAGAA
57
T2
-148 bp
AAGCATTTATCAGGGTTATTGTCTCATGAG
37
T3
-222 bp
AATGCCGCAAAAAAGGGAATAAGGGCGACA
47
T4
-443 bp
GACCGAGTTGCTCTTGCCCGGCGTCAATAC
60

Results

Evaluation of different target sequences

At first we tested different target sequences at the same loci. For that we had to insert the target sequences into the SEAP reporter plasmid.
HEK cells were transfected with one of these SEAP plasmids and a plasmid containing CAG:dCas9-VP16, the tracrRNA and the appropriate crRNA. The results show different activation efficiencies (Fig. 1): from no activation at all (VEGF VZ -573 and +434) to a 5 fold increase of SEAP activation. With the tested target sequences that have a GC content of 70 % there was no activation, whereas it was possible to induce SEAP expression with the target sequences with a GC content of about 60 % (compare Tab. 1).
Figure 1: Targets with different sequences
The light green bars represent the SEAP activity of HEK cells transfected with CAG:dCas9-VP16 but no crRNA normalized to one. The dark green bars show fold induction of SEAP activity by tansfecting CAG:dCas9-VP16 with crRNA in comparison to the appropriate control without crRNA. All values are means of three biological replicates with standard deviation.
Secondly we checked for different distances upstream of the promoter. Thus, we targeted 4 different loci on the reporter plasmid (EMX1, T2, T3 & T4; see Tab. 1).
HEK cells were transfected with the SEAP plasmid that contains EMX1, CMV:dCas9-VP16 and the RNA plasmid. The results show a very high increase in SEAP expression when EMX1, which is the nearest target site to the promoter, is targeted (Fig. 2). Activation becomes weaker, the bigger the distance between promotor and target sites gets(compare Tab. 1).
Figure 2: Targets with different distances to the promoter
The light green bar represents the SEAP activity of HEK cells transfected with CMV:dCas9-VP16 but no crRNA normalized to one. The dark green bars show fold induction of SEAP activity by tansfecting CAG:dCas9-VP16 with crRNA in comparison to the control without crRNA. All values are means of three biological replicates with standard deviation.

Activation of different genes at once

In order to test the simultaneously activation of several genes we assembled 3 plasmids containing different fluorescent proteins. Every protein is fused to a different signal for intracellular localization. Thus, we were able to distinguish better between the different fluorescent proteins, because there will be no interference of the emitted light.

Fig. 2: Plasmids encoding the fluorescent proteins
Each fluorescent protein is driven by a CMV minimal promoter, that can be switched on by binding of TetR-VP16 to the TetO sequence. Between TetO and CMVmin there is a target site for Cas9 binding, a different on each plasmid. The fluorescent were fused to signal sequences for subcellular localization, so mCherry will be in the nucleus, GFP in the Golgi apparatus and BFP at the membrane. T: terminator.
Fig. 3: Microscopy pictures of fluorescent proteins expressed in HeLa cells
Fluorescence pictures were taken of fixed HeLa cells transfected with Golgi-GFP and mCherry-NLS. Channels of GFP and mCherry were merged. All pictures have the same scale.

HEK cells were transfected with different combinations of these plasmids and dCas9-VP16 or TetR-VP16 (4 fold amount of effector DNA) After 2 days of expression the cells were analysed by flow cytometry. This way the fluorescence intensity of every cell could be determined (Fig. 4).

Fig. 4: Activation of expression of different fluorescent proteins
The fluorescence intensity of each cell was analysed by flow cytometry. The mean fluorescence intensity was calculated with the intensities of the cells which were brighter than untransfected cells. The bars represent the mean with standard deviation of these mean fluorescences of three different cell populations.
blue: only the plasmid containing the fluorescent proteins with minimal promoter were transfected; green: the minimal promoter driven fluorescent proteins were cotransfected with TetR-VP16; yellow: the minimal promoter driven fluorescent proteins were cotransfected with dCas9-VP16.

Unfortunately dCas9-VP16 was not able to increase the intensity of any fluorescent protein at all, even when transfected exclucively, whereas TetR-VP16 strongly activates the expression of fluorescent proteins. This may be due to the higher number of binding sites for TetR (16 in comparison with 1 for dCas9).

Stricter gene regulation by targeting different loci simultaneously

As we wanted to improve the efficiency of our gene regulation toolkit, we tried to target several loci upstream of the reporter gene's promoter at once. Thus, we ordered crRNAs that are complementary to sequences on the SEAP reporter plasmid with different distances to the promoter (Tab. 1).

HEK cells were transfected with the SEAP reporter plasmid, dCas9-VP16 (iGEM standard), one or two RNA plasmids and a plasmid coding for Renilla luciferase for an internal standard (to eliminate variabilities of different cell numbers or expression levels). The total amount of RNA plasmids was always the same, so it can be exclude that an increase of SEAP expression is due to more available crRNA. When combining the targets EMX1 and T2 there could be observed a higher SEAP activation than the sum of the single targets (Fig. 5).

Fig. 5: Effects of different target numbers
SEAP activity was divided through luminescence intensity of Renilla luciferase. The bars represent the mean of biological triplicates with standard deviation. All samples were transfected with CMV:dCas9-VP16 in pSB1C3 and with no (left), one (middle) or two RNA plasmids (right).

Discussion

From the few target sites we evaluated we could draw the conclusion that a target should be chosen like this:

  • The nearer to the promoter the better.
  • Target sites with a low GC content should be preferred.
As we tested our target sequences with dCas9-VP16, stricly speaking, we can only make a statement about the best target sites for gene activation.
Moreover there may be other parameters that influence the effects of dCas9-VP16 on SEAP (e.g. the secondary structure of the crRNA). But for the distance to the promoter Mali et al. [1] had the same result: Only the target site that was nearest to the promoter showed a high increase of activation.

By targeting different loci upstream of a promoter simultaneously the efficiency of transcription activation can be enhanced . While Cheng et al. [2] yielded an up to 8 fold activation by using at least three different targets in comparison to the highest activation of a single target, we were able to achieve a three fold increase by using only two different targets.
About the ability of activating multiple genes we could not make a statement as the activation of our reporter plasmids could not be detected at all with dCas9-VP16. This may be due to the high background expression of CMVmin. Nevertheless Cheng et al. showed that it is possible to target up to three genes at once [2].

References

(1) Mali P,et al. (2013). CAS9 transcriptional activators for target specificity screening and paired nickases for cooperative genome engineering. Nat Biotechnol. 2013 Sep;31(9):833-8.
(2) Cheng AW, et al. (2013). Multiplexed activation of endogenous genes by CRISPR-on, an RNA-guided transcriptional activator system. Cell Res. 2013 Aug 27. doi: 10.1038/cr.2013.122.

crRNA design tool

Enter Target Sequence

Choose Program

Which strand should be targeted?

non coding strand coding strand

This tool helps you to design a crRNA-insert for dCas9 RNA plasmid: "uniCAS RNAimer" (BBa_K1150034).
Using this tool you do not have to do this on your own. Just insert the desired target sequence and you get all different oligo possibilities and their positions. The oligos contain overhangs which fit to this plasmid's BbsI-overhangs and are ready to use.
The two different target possibilities are the coding and non-coding strand, depending on the desired target sequence.

a) For repression of gene transcription by targeting the coding strand the oligos must be designed as follows:

  1. Search at your desired target sequence for a CCN (reverse complement of the PAM sequence) at the coding strand.
  2. Extract the following (3') 30 nucleotides.
  3. Extract the reverse complement.
  4. Add AAAC at the 5' end and GT at the 3' end. This will be your fist oligo.
  5. Take the sequence from step 2 and add TAAAAC at the 5' end. This will be your second oligo.

b) For repression of gene transcription by targeting the non-coding strand the oligos must be designed as follows:

  1. Search at your desired target sequence for a NGG (the PAM sequence) at the coding strand.
  2. Extract the 30 nucleotides before (5') the PAM sequence.
  3. Extract the reverse complement.
  4. Add TAAAAC at the 5' end. This will be your second oligo.
  5. Take the sequence from step 2 and add AAAC at the 5' end and GT at the 3' end. This will be your first oligo.

(the oligos are designed analog to: Cong L, Ran FA, Cox D, Lin S, Barretto R,Habib N, Hsu PD, Wu X, Jiang W, Marraffini LA, Zhang F. Multiplex (2013 Jan 3). Genome Engineering using CRISPR/Cas Systems. Science. DOI: 10.1126/science.1231143 )

Technical Information

view source code