Bacterial Expression, Purification, and Initial Characterization of a Full-length Cas13b Protein from Porphyromonas gingivalis

Abstract

The CRISPR-Cas13b system is a recently identified Class 2, RNA-targeting CRISPR-Cas system. The system has been repurposed to achieve robust mRNA knockdown and precise RNA-editing in mammalian cells. While the CRISPR-Cas13b system has become a powerful tool for nucleic acids manipulation, the mechanisms of the system are still not fully understood. Cas13b endonucleases from different bacterial species show poor overall sequence homologies, suggesting that structural (and probably functional) diversities may exist. It is therefore important to study CRISPR-Cas13b cases from different bacterial species. Here we report the expression, purification, and initial characterization of a Cas13b endonuclease that is associated with the 8^th putative CRISPR locus from Porphyromonas gingivalis genome (Pgi8Cas13b). The full-length Pgi8Cas13b protein (1119 residues) was successfully expressed in E. Coli cells, and purified by affinity and ion-exchange chromatography methods. The purified protein is biologically active, being able to bind its cognate crRNA with high specificity and affinity. Preparation of biologically active Pgi8Cas13b protein provides the basis for further in vitro biochemical and biophysical studies of the Pgi8Cas13b CRISPR system.

1. Introduction

CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) arrays are bacterial or archaeal DNA sequences containing short stretches of variable sequence, termed “spacers”, separated by short invariable repetitive elements, termed “Direct Repeats” (DRs). The spacers are acquired from the DNAs of invading pathogens and therefore store the immunological memory of previous infection. CRISPR loci are typically flanked by sequences that contain a cluster of Cas (CRISPR-associated) genes [1]. The Cas proteins are involved in various stages of CRIPSR-mediated processes. The CRISPR-Cas systems provide adaptive immunity for the bacteria and archaea to combat invading pathogens.

During a CRISPR response, the Cas genes are expressed; the CRISPR loci are transcribed and the transcripts (pre-crRNAs) are processed into crRNAs, which contain DR and individual spacer sequences [2]. The effector complexes responsible for foreign DNA (or RNA) cleavage contain one or more Cas proteins and a crRNA. Some effector complexes additionally contain a trans-activating CRISPR RNA (tracrRNA) [3]. The spacer-derived sequence of a crRNA brings the effector complex to a complementary sequence (called protospacer) in a DNA or RNA target. When the protospacer is adjacent to a PAM (protospacer adjacent motif, in DNAs) or a PFS (protospacer flanking sequence, in RNAs), the target DNA or RNA is cleaved by the effector complex [4].

The CRISPR-Cas systems are divided into two classes. Class 1 systems utilize multiple Cas proteins in the effector complexes, whereas Class 2 systems utilize a single effector endonuclease [5]. Due to the relative simplicity and adaptability of the Class 2 systems, they are more amenable for development as tools for nucleic acids manipulation. During the last 7 years, several Class 2 CRISPR-Cas systems have been identified and characterized, including the well-known Cas9, Cas12a (originally known as Cpf1), Cas12b, Cas12c, Cas13a, and Cas13b [6–9]. These systems target either DNA or RNA for cleavage. Some of these systems (especially Cas9) had been repurposed as powerful tools for nucleic acids manipulation, including genome editing in human cells. Since the CRISPR-Cas9 gene editing tool was first demonstrated in 2012, CRISPR technology has revolutionized biochemical and biomedical research [3, 5, 9–12]. To expand the CRISPR tool kits, efforts are being made to optimize the known CRISPR-Cas systems and to discover new CRISPR-Cas systems.

Cas13b is a more recently identified CRISPR-Cas system [13, 14]. Biochemical and genetic experiments were performed on the Cas13b from Bergeyella zoohelcum (BzCas13b). It was found that the mature crRNA contains 66 nucleotides, with the full 30 nucleotides spacer at the 5′-end followed by the full 36 nucleotides direct repeat [13]. It was also found that purified BzCas13b is capable of cleaving its pre-crRNA transcript to generate the mature crRNAs. Cas13b endonucleases, guided by a crRNA, target single-stranded RNAs for cleavage. Cas13b systems were capable of robust RNA knockdown in mammalian cells. Moreover, a Cas13b system was adapted for precise RNA-editing by fusing a catalytically-inactive Cas13b (dCas13b) to the RNA-editing enzyme ADAR2 (adenosine deaminase acting on RNA). This new system, called REPAIR (RNA editing for programmable A to I replacement) is the first CRISPR-based tool for RNA editing. The tool had been used to correct disease relevant mutations in human cells [14].

While the Cas13b CRISPR-Cas systems have swiftly risen to prominence within a very short period of time, the catalytic mechanisms of the systems are still not well understood. It is interesting to note that Cas13b enzymes from different species may have poor overall sequence homologies (supplementary Table S1). Except the two HEPN (higher eukaryotes and prokaryotes nucleotide-binding) domains at the N- and C-termini (Figure 1), there is no other sequence element that is conserved among all Cas13b proteins from different species. This low sequence homology suggests structural (and probably functional) diversities among family members.

Figure 1.

The Pgi8Cas13b CRISPR system in Porphyromonas gingivalis (ATCC33277). The exact sequences of the crRNAs were determined by next generation sequencing (NGS) of the small RNAs extracted from a host E. coli cell. HEPN: higher eukaryotes and prokaryotes nucleotide-binding domain.

We have been studying CRISPR-Cas13b systems from many species (supplementary table S1). Of particular interest, there are two putative CRISPR-Cas13b systems (associated with the 5^th and the 8^th CRISPR loci) in the Porphyromonas gingivalis genome. Sequences of the Pgi5Cas13b and Pgi8Cas13b proteins share only limited similarity (23% identical and 38% similar), the two systems also differ in the number of CRISPR repeats (13 vs 5), uniformity of the spacer lengths, and presence or absence of an auxiliary protein coding gene within the intervening sequence between the CRISPR locus and Cas13b gene. In-depth studies on the Pgi5 and Pgi8 Cas13b systems should provide valuable insights into the Cas13b family. The availability of Porphyromonas gingivalis chromosomal DNA from the American Type Culture Collection (ATCC) also greatly facilitate the molecular cloning experiments. Here we report the expression, purification, and initial characterization of Pgi8Cas13b endonuclease.

2. Materials and methods

Molecular cloning.

The chromosomal DNA of Porphyromonas gingivalis (ATCC 33277, NC_0107292) was purchased from the American Type Culture Collection (ATCC). The chromosomal DNA contains several CRISPR loci, two of which (5^th and 8^th) encode Cas13b systems. The DNA encoding the Pgi8Cas13b protein was amplified from the chromosomal DNA by PCR using Phusion DNA polymerase, with the forward primer 5’-CAAGGACCGAGCAGCCCCTCAATGACAGAACAAAACGAGAGAC-3’ and reverse primer 5’-ACCACGGGGAACCAACCCTTATGCTTGTATGATTCGTTCTGC-3’. Both of these primers contain 21 nt at the 5’-end as a specific sequence for ligation independent cloning (LIC). The PCR product was purified by 1% agarose gel electrophoresis and processed by T4 DNA polymerase in the presence of 2.5 mM dATP (25°C for 40 min followed by 75°C for 20 min) to generate 5’-overhangs at both ends of the DNA insert. The cloning vector is an in-house-developed LIC vector that contains DNA sequences encoding the Halo tag [15], a His tag, an eight-residue recognition sequence for Human rhinovirus 3C protease [16], and a specific LIC sequence 5’-TCAAGGACCGAGCAGCCCCGGGTTGGTTCCCCGTGGTA- 3’. This LIC sequence contains a SmaI site in the middle. After digestion by SmaI (3 hours at 37°C), the linearized vector was processed by T4 DNA polymerase in the presence of 2.5 mM dTTP (25°C for 40 min followed by 75°C for 20 min) to generate 5’-overhangs at both ends of the linearized vector. The processed PCR insert and vector were mixed together at room temperature and incubated for 20 min. After annealing, the mixture was transformed into Escherichia coli DH5α competent cells. A single colony was picked to grow overnight cultures to amplify the plasmid. Protein expression was carried out using E. coli NiCo21 (DE3) cells (New England Biolabs).

Protein expression and purification.

The Pgi8Cas13b protein, containing 1119 residues, was expressed as a fusion protein with N-terminal Halo and His tags followed by a HRV 3C protease cleavage site (LEVLFQGP). After protease digestion, the target protein contains an artificial sequence GPSSP (from the protease cleavage site and LIC cloning sequence) at the N-terminus. Bacteria containing the Pgi8Cas13b recombinant plasmids were grown in LB medium until the cell density reached an absorbance of 0.8–1.0 measured at a wavelength of 600 nm. Protein expression was induced by adding isopropyl β-d-1-thiogalactopyranoside (IPTG) to the culture to a final concentration of 1 mM. To obtain soluble expression of the Pgi8Cas13b protein, it was necessary to lower the temperature of the cell cultures to 10°C after induction by IPTG. The cell cultures were allowed to grow overnight (16–20 hours) before harvesting. Cell cultures were harvested and the cell pellets were re-suspend in 30 ml buffer (per liter culture) containing 25 mM Tris pH 7.5, and 200 mM NaCl. The cells were lysed by lysozyme and sonication at 4°C. The cell debris was separated by centrifugation at 13,000 rpm for 30 minutes at 4°C. The overexpressed protein was in the supernatant and was purified using cobalt agarose affinity resin from Gold Biotechnology. After the purified fusion protein was eluted from the resin with an elution buffer consisting of 50 mM NaCitrate (pH 7.5), 50 mM NaCl, 200 mM imidazole, His-tagged HRV 3C protease was added to the eluted protein sample. The protein solution (15~25 ml) was dialyzed in 5 L of a buffer containing 50 mM NaCitrate (pH 7.5), 50 mM NaCl, at 4°C overnight. Complete cleavage at the HRV 3C recognition site was achieved after dialysis for 12 h (confirmed by SDS–PAGE). The cleaved Halo and His tags were separated from the target protein by a reverse immobilized metal-affinity chromatography (IMAC) process with cobalt agarose affinity resin. The flow-through from the reverse IMAC column contains the purified target protein (Supplementary Figure S1). The target protein was further purified by SP-Sepharose cation exchange chromatography. After the target protein was loaded into the SP-Sepharose column, the column was washed by 1 L of a buffer containing 100 mM NaCl and 50 mM NaCitratie (pH7.5). The target protein was eluted by an elution buffer containing 50 mM NaCitrate (pH7.5) and a salt gradient from 200 mM to 600 mM NaCl. Fractions were collected with a fraction size of 1.5 ml. The target protein was eluted with NaCl concentration of 300–400 mM (~15 fractions). The fractions with the highest concentration of the target protein were combined and concentrated to a final concentration of ~8 mg/ml in a buffer containing 50 mM NaCitrate (pH 7.5) and 350 mM NaCl. A typical yield of the process is about 5 mg of purified Pgi8 protein from 1 liter LB culture. For SDS-PAGE experiments, 10% Bis-Tris gels with MOPS running buffer were used.

For future structure determination of the Pgi8 cas13b crystal structure by the single- or multi-wavelength anomalous diffraction (SAD or MAD) method, we decided to label the protein with selenomethionine (SeMet). Using SAD and MAD method will allow the structure to be determined independently. To prepare SeMet-labeled protein samples, the bacterial cells were grown in M9 minimal culture until the cell density reached an absorbance of 0.5 measured at a wavelength of 600 nm. The six amino acids leucine, isoleucine, lysine, phenylalanine, threonine and valine were added to the culture to a final concentration of 50–100 mg/L to inhibit the methionine-biosynthesis pathway. The cell cultures were grown for a further 30 minutes at 37°C. The temperature was then lowered to 10°C, selenomethionine was added to the cell cultures to a final concentration of 50 mg/L. Protein expression was induced by IPTG at a final concentration of 1.0 mM. All other steps were identical to the preparation of unlabeled protein samples.

RNA synthesis.

To characterize the interaction between Pgi8Cas13b and its cognate crRNA, the crRNA was prepared by in vitro transcription using T7 RNA polymerase. The DNA template for the in vitro transcription reaction was a PCR product that was amplified from the Porphyromonas gingivalis chromosomal DNA with a forward primer containing a T7 promoter. The Pgi8 crRNA has the sequence of 5’-GCAAGUGGAAUUAAUCUUUUUUCUUCUUUACGUUGGAUCUUCCCUCUAUUCGAAGGGUACACACAAC-3’. The direct repeat (DR) derived sequences in the crRNA is underlined. A 48 nt stem-loop RNA was used as a nonspecific competitor for binding assay. The template for in vitro transcription of the stem-loop RNA were synthetic DNA oligos with a T7 promotor. The stem-loop RNA has the sequence of 5’-GUGGCCUUUAUAAAGGUCAUUCGCAAGAGUGGCCUUUAUAAAGGUCAC-3’. The in vitro transcription reactions were carried out at 37°C for 3 hours. The reactions were stopped by ethanol precipitation of the RNA product. The transcribed crRNAs were separated from abortive transcripts by denaturing 15% PAGE-urea gels. The crRNA band was excised and crRNA was eluted from the gel slice by using an Elutrap device. The concentration of the crRNA was determined by Qubit. For the preparation of Cy5-labeled RNAs, Cy5-UTP was included in the in vitro transcription reaction mixture. All other steps were identical to the preparation of unlabeled RNA samples except that caution was taken to minimize exposure of the sample to light.

Electrophoretic mobility shift assay (EMSA).

EMSA was used to characterize Pgi8Cas13b protein-crRNA interaction. A fixed amount of Cy5-labeled crRNA (at a concentration of 1 nM, comparable to half of the estimated Kd value, measured by Qubit 3.0 fluorimeter) was titrated by increasing amount of the purified Pgi8Cas13b protein (from 0 to 58 nM). The mixtures, in a binding buffer containing 40 mM Tris buffer pH 7.8, 30 mM KCl, 1 mM MgCl₂, 1 mM dithiothreitol (DDT), and 10 μg/mL BSA, were incubated for 30 minutes at room temperature. The mixtures were analyzed by a 6% native polyacrylamide gel (with a 29:1 acrylamide:bisacrylamide ratio), with a running buffer containing 50 mM Tris pH 7.8 and 50 mM glycine. The gels were imaged by using a Li-cor Odyssey scanner at the wavelength of 700 nM. The RNA bands were quantified by ImageJ and the data was analyzed by GraphPad Prism version 7 to calculate the Kd values. To distinguish between specific and nonspecific binding, competition experiments were performed. Unlabeled Pgi8 crRNA was used as a specific competitor. An unlabeled 48 nt stem-loop RNA was used as a nonspecific competitor.

3. Results

The chromosomal DNA of Porphyromonas gingivalis (ATCC 33277, NC_0107292) has 12 putative CRISPR loci [1]. The 5^th and 8^th CRISPR loci potentially harbor Cas13b systems. We have been investigating both systems. In this paper, we report some results from our investigation of the Pgi8Cas13b system. The 8^th CRISPR locus contains 5 DRs and 4 SPs (Figure 1). The flanking sequence of the CRISPR locus contains a large open reading frame (ORF) that encodes a large (1119 residues, MW=132 kDa) basic protein (theoretical pI = 9.08), which is a putative Cas13b endonuclease. The intervening sequence between the Pgi8Cas13b ORF and the CRISPR array has 759 basepairs. A smaller ORF (encoding a putative Csx28 protein with 181 residues) immediately follows the Pgi8Cas13b ORF.

We amplified the Pgi8Cas13b encoding DNA from the chromosomal DNA of Porphyromonas gingivalis and inserted the DNA into the LIC cloning site of a series of in-house developed LIC vector. The LIC cloning site is preceded by DNA sequences that encode a protein tag (NusA, MBP, GST, Halo, SUMO, GB1, Fc, TRX, DnaK, etc), a His tag, and an octapeptide HRV 3C protease recognition motif. We found that the Halo tag LIC vector was the best for obtaining expression of the Pgi8Cas13b protein in a soluble form as a fusion protein. The fusion protein contains N-terminal Halo and His tags, separated from the Pgi8Cas13b protein by the specific HRV 3C protease cutting site. After purification of the fusion protein by Co-affinity resin, the Halo and His tags were cleaved off from the target protein using His-tagged HRV 3C protease. The protein tags together with the His-tagged HRV 3C protease were subsequently removed by a reverse IMAC procedure. The target was further purified by SP-Sepharose cation exchange chromatography (Figure 2). The fractions containing the target protein were combined and could be concentrated to a final concentration of ~8 mg/ml, suitable for crystallization experiments.

Figure 2.

SDS–PAGE of the purified Pgi8Cas13b protein. The lanes are for protein samples from nine different odd consecutive number fractions (starting from fraction #25) eluted from a SP-Sepharose column with increasing NaCl concentration.

To verify that the purified Pgi8Cas13b protein was biologically active, we further investigated whether the purified protein was able to bind to its cognate crRNA. We had used next generation sequencing (NGS) to reveal the exact sequences of the mature crRNAs produced by the Pgi8Cas13b system and found that the majority of Pgi8Cas13b cognate crRNAs had a 36-nt DR-derived sequence flanking by a 30-nt SP-derived sequence at the 5’-end and a 4-nt SP-derived sequence at the 3’-end (details will be published elsewhere). The 36-nt DR-derived sequence of the crRNA has the potential to fold into an overall stem-loop structure with a few mismatches in the stem (Figure 1). We had tried to synthesize a few 70-nt crRNAs by in vitro transcription using T7 RNA polymerase. The crRNA with a 5’-end sequence corresponding to the second spacer (SP2) had the best yield in the transcription reaction. We synthesized and purified this crRNA for the EMSA and crystallization experiments.

As can been seen in the EMSA experiments (Figure 3), the purified Pgi8Cas13b protein binds to its crRNA with specificity and high affinity. The non-specific competitor RNA used in the competition experiment is an unlabeled 48-nt stem-loop RNA with a perfect stem, which cannot disrupt the interaction between Pgi8Cas13b and its cognate crRNA (Figure 3B). The Kd of the Pgi8Cas13b-crRNA complex is 18 nM, indicative of high affinity interaction.

Figure 3.

Electrophoretic mobility shift assays (EMSA) for the characterization of Pgi8Cas13b protein–crRNA interaction. The Pgi8 crRNA is body-labeled with Cy5 for detection at 700 nM wavelength using a Li-Cor Odyssey scanner. A) Cy5 body-labeled Pgi8 crRNA (1 nM) was incubated with increasing amounts of Pgi8 Cas13b or an excess amount of bovine serum albumin (BSA). Samples were loaded onto 6% native polyacrylamide gels with a 29:1 acrylamide:bisacrylamide ratio, using a Tris-glycine buffering system. Positions of bound and free RNA are shown. The binding curve for these data is shown to the left of the gel. The Kd for the binding reaction was calculated to be 18 nM. B) Competitive binding assay using the un-labeled Pgi8 crRNA as the specific competitor, and an un-labeled 48-nt stem-loop RNA (with a perfect 22 bp stem and a GCAA tetra-loop) as the nonspecific competitor.

4. Discussion

We had used E. coli as the bacterial host to express the Pgi8Cas13b protein. Initial trials to express the protein alone was not successful. We then explored to express the protein as a fusion with different protein tags. Although it is a common practice to use protein tags for the expression of recombinant proteins, there is no way to predict which particular protein tag would work for a given protein. Moreover, the commonly used protein tags (such as MBP, NusA, TRX, GST, Halo, SUMO, etc) are available in various vectors from different vendors, making it difficult and time-consuming to carry out a systemic study to test a large number of protein tags. In view of this situation, we had developed a series of LIC cloning vectors which contain the same LIC cloning site but different protein Tags. Parallel cloning of the same DNA insert (encoding the protein of interest) into ten different vectors typically takes only two days. The availability of these vectors had greatly facilitated our studies to systematically screen for the best protein tag for a given protein. We found that the Halo tag worked best for the Pgi8Cas13b protein, while the GST tag worked best for the Pgi5Cas13b protein. Interestingly, although it is necessary to use a fusion tag to achieve soluble expression of the two Cas13b proteins, the proteins alone (after the tags are removed by HRV 3C protease cleavage) actually have great solubility in appropriate buffers.

All known Class 2 CRISPR systems have a large and basic endonuclease. We found that cation exchange chromatography could be used to further purify the basic Cas13b proteins. We expected that the same procedure would work well for other basic Cas endonucleases.

The bacterially expressed Pgi8Cas13b protein is biologically active. By using EMSA, we showed that the purified Pgi8Cas13b protein was able to specifically bind to its cognate crRNA (but not to a stem-loop RNA with perfect stem region) with high affinity. The Kd value we determined for the Pgi8Cas13b-crRNA complex is comparable with that of the previously reported BzCas13b CRISPR system [13].

We had carried out crystallization trials on Pgi8Cas13b protein, alone or in the presence of its cognate crRNA. However, so far we have not been able to obtain crystals that yield good diffraction data for the determination of the structures.

Recently, two crystal structures of BzCas13b and PbuCas13b proteins in complex their cognate crRNAs (only partial sequences were present in the structure) were reported [17, 18]. The two Cas13b protein sequences share 20% identity and 35% homology. Superimposition of the two structures reveal a large pairwise RMSD of 7.8 Å. Substantial structural differences are observed even between corresponding HEPN domains from the two proteins. Most noticeably, the relative orientations of the two HEPN domains in the two structures are quite different. In the PbuCas13b structure, the active site residues of the two HEPN domains are clustered together. In contrast, in the BzCas13b structure, the active site residues of the two HEPN domains are separated by almost 19 Å. None of the structure contains a target RNA. It is possible that binding the target RNA may cause a conformational change that is required for target cleavage. Further structural and biochemical studies are needed to fully elucidate the catalytic mechanisms of Cas13b endonucleases.

5. Conclusion

For in vitro biochemical and biophysical studies of the Pgi8Cas13b CRISPR system, the full-length Pgi8Cas13b was successfully expressed in E. Coli, and purified by affinity and ion-exchange chromatography methods. The purified protein is biologically active, as indicated by its ability to bind its cognate crRNA with high specificity and affinity. The Pgi8Cas13b protein preparation procedure should pave the way for further experiments for the characterization of the Pgi8Cas13b and other related CRISPR systems.

Highlights.

• The full-length Pgi8Cas13b protein was successfully expressed in E. coli

• The protein was purified by affinity and ion exchange chromatography

• The purified protein showed specific and high affinity binding towards its cognate crRNA

• The purified protein is suitable for further biochemical and biophysical experiments

Abbreviations

CRISPR

Clustered Regularly Interspaced Short Palindromic Repeats

HEPN

higher eukaryotes and prokaryotes nucleotide-binding domain

LIC

ligation independent cloning

EMSA

electrophoretic mobility shift assay