Large-scale Production of Recombinant RNAs on a Circular Scaffold Using a Viroid-derived System in Escherichia coli

Biochemistry

Your institution must subscribe to JoVE's Biochemistry section to access this content.

Fill out the form below to receive a free trial or learn more about access:

 

Summary

Here, we present a protocol to produce large amounts of recombinant RNA in Escherichia coli by co-expressing a chimeric RNA that contains the RNA of interest in a viroid scaffold and a plant tRNA ligase. The main product is a circular molecule that facilitates purification to homogeneity.

Cite this Article

Copy Citation | Download Citations

Cordero, T., Aragonés, V., Daròs, J. A. Large-scale Production of Recombinant RNAs on a Circular Scaffold Using a Viroid-derived System in Escherichia coli. J. Vis. Exp. (141), e58472, doi:10.3791/58472 (2018).

Abstract

With increasing interest in RNA biology and the use of RNA molecules in sophisticated biotechnological applications, the methods to produce large amounts of recombinant RNAs are limited. Here, we describe a protocol to produce large amounts of recombinant RNA in Escherichia coli based on co-expression of a chimeric molecule that contains the RNA of interest in a viroid scaffold and a plant tRNA ligase. Viroids are relatively small, non-coding, highly base-paired circular RNAs that are infectious to higher plants. The host plant tRNA ligase is an enzyme recruited by viroids that belong to the family Avsunviroidae, such as Eggplant latent viroid (ELVd), to mediate RNA circularization during viroid replication. Although ELVd does not replicate in E. coli, an ELVd precursor is efficiently transcribed by the E. coli RNA polymerase and processed by the embedded hammerhead ribozymes in bacterial cells, and the resulting monomers are circularized by the co-expressed tRNA ligase reaching a remarkable concentration. The insertion of an RNA of interest into the ELVd scaffold enables the production of tens of milligrams of the recombinant RNA per liter of bacterial culture in regular laboratory conditions. A main fraction of the RNA product is circular, a feature that facilitates the purification of the recombinant RNA to virtual homogeneity. In this protocol, a complementary DNA (cDNA) corresponding to the RNA of interest is inserted in a particular position of the ELVd cDNA in an expression plasmid that is used, along the plasmid to co-express eggplant tRNA ligase, to transform E. coli. Co-expression of both molecules under the control of strong constitutive promoters leads to production of large amounts of the recombinant RNA. The recombinant RNA can be extracted from the bacterial cells and separated from the bulk of bacterial RNAs taking advantage of its circularity.

Introduction

In contrast to DNA and proteins, protocols for easy, efficient and cost-effective production of large amounts of RNA are not abundant. However, research and industry demand increasing amounts of these biomolecules to investigate their unique biological properties1, or to be employed in sophisticated biotechnological applications, including their use as highly specific aptamers2, therapeutic agents3, or selective pesticides4. In vitro transcription and chemical synthesis are commonly used in research to produce RNA. However, these methods entail important limitations when large amounts of the products are required. The logical alternative is to use the endogenous transcription machinery of living cells, followed by a purification process to separate the RNAs of interest from the cellular companions. Following this strategy, methods have been developed to produce recombinant RNAs in bacterial cells, such as the lab-friendly Escherichia coli5 or the marine purple phototrophic alpha-proteobacterium Rhodovolum sulfidophilum6. Most methods to produce recombinant RNA in bacteria rely on the expression of a native highly stable RNA scaffold, such as a tRNA or an rRNA, in which the RNA of interest is inserted7. This imposes the necessity of releasing the RNA of interest out of the chimeric molecule, if the presence of extra RNA is a problem for the downstream applications8. Another concept in recombinant RNA biotechnology is the production of recombinant ribonucleoprotein complexes that may be the desired product per se or used as a protective strategy to increase the stability of the RNA of interest9,10. Similarly, the production of circular RNAs has also been suggested as a strategy to generate more stable products11.

We have recently developed a new method to produce large amounts of recombinant RNA in E. coli that participates in three of the above concepts: the insertion of the RNA of interest in a highly stable circular RNA scaffold and the co-expression of the recombinant RNA with an interacting protein to likely produce a stable ribonucleoprotein complex that accumulates in remarkable amounts in bacterial cells12. In contrast to previous developments, we used an RNA scaffold completely alien to E. coli, namely a viroid. Viroids are a very particular type of infectious agents of higher plants that are exclusively constituted by a relatively small (246-401 nt) highly base-paired circular RNA13. Interestingly, viroids are non-coding RNAs and, with no help from their own proteins, they are able to complete complex infectious cycles in the infected hosts14. These cycles include the RNA-to-RNA replication in the nuclei or chloroplasts, depending on the viroid family —Pospiviroidae or Avsunviroidae, respectively— movement through the infected plant and evasion of the host defensive response. Viroids must be ranked among the most stable RNAs in nature, as a consequence of being a naked circular RNA and having to survive in the hostile environment of plant infected cells. This property may make viroids particularly suitable as scaffolds to stabilize recombinant RNA in biotechnological approaches. In addition, the new method is based on co-expression of the viroid scaffold with an interacting plant protein. Viroids replicate through a rolling-circle mechanism in which host enzymes are recruited to catalyze the different steps of the process. Notably some viroids, more specifically those that belong to the family Avsunviroidae15, contain ribozymes that are also involved in replication. Depending on the viroid species, transcription of viroid RNAs is mediated by the host RNA polymerase II or the chloroplastic nuclear-encoded RNA polymerase (NEP). Viroid RNA processing seems to be catalyzed by a host type-III RNase, although in viroids with ribozymes oligomeric RNA intermediates self-cleave during replication. Finally, the resulting viroid monomers are circularized, depending on the viroid family, by the host DNA ligase 1 or the chloroplastic isoform of tRNA ligase16,17. This last enzyme is involved in ligation of the monomeric forms of the viroids in the family Avsunviroidae, such as Eggplant latent viroid (ELVd)18.

In the course of a work to analyze the sequence and structural requirements of ELVd that determine recognition by the eggplant (Solanum melongena L.) tRNA ligase, we set up an experimental system based on co-expression of both molecules in E. coli19. We noticed that longer-than-unit ELVd transcripts self-cleave efficiently in E. coli cells through the embedded hammerhead ribozymes and that the resulting viroid monomers with 5'-hydroxyl and 2',3'-phosphodiester termini were efficiently circularized by the co-expressed eggplant tRNA ligase. Even more, the resulting circular viroid RNA reached an unexpected high concentration in E. coli, exceeding those of the endogenous rRNAs12. Absence of replication intermediates indicated lack of ELVd RNA-to-RNA amplification in these bacterial cells. Interestingly, insertion of heterologous RNAs in a particular position of the viroid molecule had a moderate effect on accumulation of the circular viroid-derived RNAs12. These observations made us envision a method to produce large amounts of recombinant RNAs in bacteria. In this method, the cDNAs corresponding to the RNAs of interest are inserted in the ELVd cDNA and the resulting chimeric RNA is expressed in E. coli through a strong constitutive promoter. For the system to work, E. coli must be co-transformed with a plasmid to express the eggplant tRNA ligase. The ELVd-derived longer-than-unit transcript is processed by the embedded hammerhead ribozymes and the resulting monomers with the appropriate termini are recognized and circularized by the co-expressed tRNA ligase. This way, the RNA of interest is inserted into a very stable circular scaffold consisting of the viroid circular molecule. This recombinant chimeric RNA is most probably further stabilized inside the E. coli cells by formation of a ribonucleoprotein complex through interaction with tRNA ligase. Using this method (see protocol below), RNA aptamers, hairpin RNAs and other structured RNAs have been easily produced in amounts of tens of milligrams per liter of E. coli culture in regular laboratory conditions and purified to homogeneity taking advantage of circularity12.

Subscription Required. Please recommend JoVE to your librarian.

Protocol

1. Plasmid Construction

  1. Amplify by PCR (or by reverse transcription PCR if starting from an RNA template) the cDNA corresponding to the RNA of interest using primers with 5’ extension to allow assembly into the expression plasmid. To avoid undesired mutations, use a high-fidelity DNA polymerase.
    1. To insert the cDNA in the expression plasmid by Gibson assembly20, add the following 5’ extensions to the PCR primers: forward, 5’-tctccccctcccaggtactatccccttXXXXXXXXXXXXXXXXXXXXXX-3’; reverse, 5’-ccctcctagggaacacatccttgaXXXXXXXXXXXXXXXXXXXXXX-3’; X represents nucleotides homologous to the terminal ends of the RNA of interest.
    2. Incubate for 30 s at 98 °C, followed by 30 cycles of 10 s at 98 °C, 30 s at 55 °C and 30 s at 72 °C, and a final extension of 10 min at 72 °C.
  2. Digest 100 ng of plasmid pLELVd-BZB with 10 U of the type-IIS restriction enzyme Bpi I for 1 h at 37 °C in a 20 µL reaction in a 0.5 mL tube in buffer G (10 mM Tris-HCl, pH 7.5, 10 mM MgCl2, 50 mM NaCl, 0.1 mg/mL BSA).
    NOTE: All plasmids are available on request to the corresponding author. Bpi I is equivalent to Bbs I.
  3. Separate the PCR and digestion products by electrophoresis in a 1% agarose gel in TAE buffer (40 mM Tris, 20 mM acetic acid, 1 mM EDTA, pH 7.2). Stain the gel for 15 min by shaking in 200-mL of 0.5 µg/mL ethidium bromide. Visualize the DNA using a UV transilluminator and cut the bands corresponding to the amplified cDNA and the Bpi I-digested plasmid (2046 bp) using a scalpel blade.
    NOTE: Bpi I digestion of pLELVd-BZB also releases a 528 bp product corresponding to the LacZ blue-white reporter.
  4. Elute the DNAs from the gel fragments using silica gel columns (gel DNA recovery kit in Table of Materials) and quantify the DNA concentration by spectrophotometric analysis.
  5. Set up a Gibson assembly reaction using the amplified cDNA and the digested plasmid. Use a 3-fold molar excess of insert versus vector20. Incubate for 1 h at 50 °C and purify the reaction products using a silica gel column (DNA clean & concentrator kit in the Table of Materials).
  6. Use the purified products of the Gibson assembly reaction to electroporate competent E. coli DH5α cells. Using 1-mm electroporation cuvettes, apply the following settings: 1,500 V and 5 ms. Incubate for 1 h at 37 °C in super optimized broth with catabolite repression (SOC; 20 g/L tryptone, 5 g/L yeast extract, 0.5 g/L NaCl, 2.5 mM KCl, 10 mM MgCl2, 20 mM glucose, pH 7.0) liquid medium and then spread onto Luria-Bertani (LB; 10 g/L tryptone, 5 g/L yeast extract, 10 g/L NaCl) agar (1.5%) plates containing 50 µg/mL ampicillin.
    1. To screen for colonies corresponding to transformed E. coli clones that likely incorporated the insert, 15 min before plating the electroporated cells, spread 30 µL of 50 mg/mL 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-gal) in dimethylformamide. Incubate plates overnight at 37 °C.
  7. Pick several white colonies and grow overnight at 37 °C in liquid LB media. Purify plasmids using a miniprep kit (see Table of Materials) and analyze their sizes by electrophoresis in a 1% agarose gel in TAE buffer.
  8. Select the most likely recombinant plasmid based on electrophoretic migration compared to the pLELVd-BZB control. Confirm the sequence of the selected plasmid by sequencing using primers 5’-CCTTTTTCAATATTATTGAAGC-3’ and 5’-GATGCTCGTCAGGGGGGCGGAG-3’ that flank the whole expression cassette in pLELVd-BZB.
    NOTE: Remember that this plasmid contains a 528 bp polylinker with the LacZ marker that is replaced by the cDNA of interest. Restriction mapping may help to select the right recombinant plasmids.

2. RNA Expression

  1. Co-electroporate (see conditions in 1.6) the selected E. coli strain (E. coli BL21 or a BL21 derivative) with the pLELVd-BZB-derivative that contains the cDNA corresponding to the RNA of interest and plasmid p15LtRnlSm to co-express the eggplant tRNA ligase. Use the E. coli HT115(DE3), which lacks RNase III21, to express RNAs with long double-stranded regions.
    Note: Both the RNA of interest and the tRNA ligase mRNA are transcribed by the E. coli RNA polymerase. No DE3 lysogen to express T7 RNA polymerase is required in the E. coli strain, although the presence of this lysogen has no deleterious effects.
  2. After 1 h incubation at 37 °C in SOC liquid medium, plate electroporated bacteria in LB solid medium containing 50 µg/mL ampicillin and 34 µg/mL chloramphenicol. Incubate overnight at 37 °C.
  3. Pick a colony and inoculate a 1 L baffled Erlenmeyer flask with 250 mL of liquid Terrific Broth (TB) medium (12 g/L tryptone, 24 g/L yeast extract, 0.4% glycerol, 0.17 M KH2PO4, and 0.72 M K2HPO4), containing 50 µg/mL ampicillin and 34 µg/mL chloramphenicol. Incubate at 37 °C with vigorous shaking at 180 revolutions per minute (rpm). Harvest bacteria between 12 and 16 h after culture inoculation.

3. RNA Extraction and Purification

  1. For analytical purposes, take 2 mL aliquots of the culture at the desired time points and centrifuge at 14,000 x g for 2 min. Discard the supernatant and resuspend the cells in 50 µL of TE buffer (10 mM Tris-HCl, pH 8.0, and 1 mM EDTA) by vortexing.
    1. Add one volume (50 µL) of a 1:1 (v/v) mix of phenol (saturated with water and equilibrated at pH 8.0 with Tris-HCl, pH 8.0) and chloroform. Break the cells by vigorous vortexing and separate the aqueous and organic phases by centrifugation for 5 min at 14,000 x g.
    2. Carefully recover the aqueous phases (upper) that contains total bacterial RNA.
      Note: Preparations can be stored at -20 °C for subsequent analysis.
  2. For preparative purposes, pour culture into a 250 mL centrifuge bottle and spin down cells at 14,000 x g for 10 min. Discard supernatant. Wash the cells by resuspending in 30 mL of water. Transfer to a centrifuge tube and spin down cells again under the same conditions.
  3. Discard the supernatant and resuspend the cells in 10 mL of chromatography buffer (50 mM Tris-HCl, pH 6.5, 150 mM NaCl, 0.2 mM EDTA) by vortexing.
    NOTE: At this time, cells can be frozen at -20 °C to proceed with purification at any other moment.
  4. Using a fresh or thawed bacterial preparation, break cells by adding 1 volume (10 mL) of phenol:chloroform (see step 3.2) and vortexing vigorously.
  5. Centrifuge for 10 min at 12,000 x g, recover the aqueous phase and re-extract with 1 volume (10 mL) of chloroform under the same conditions.
    NOTE: The RNA preparation can be stored at -20 °C at this point.
  6. Further purify total bacterial RNA by anion-exchange chromatography. Filter the RNA preparation through a 45 µm syringe filter and load on a 1 mL diethylethanolamine (DEAE) column.
    1. For chromatography purification, use a liquid chromatography system at a flow rate of 1 mL/min. Before sample loading, equilibrate the column with 10 mL of chromatography buffer (see step 3.3 for composition).
    2. Load the sample and wash the column with 10 mL of chromatography buffer. Elute RNA with 20 mL of elution buffer (50 mM Tris-HCl, pH 6.5, 1 M NaCl, 0.2 mM EDTA) and collect 1 mL aliquots.
      Note: RNA quickly elutes in the initial fractions. The column can be re-used for further purifications. For this, wash the column with 10 mL of water and store at 4 °C in 20% ethanol.
  7. Since a major part of the recombinant RNA accumulates in E. coli in a circular form, this property can be profited for purification to homogeneity12. Separate circular RNAs from the linear counterparts by two-dimensional polyacrylamide gel electrophoresis (2D PAGE) combining non-denaturing and denaturing (8 M urea) conditions12,22.

4. RNA Analysis

  1. Prepare a 5% polyacrylamide gel (37.5:1 acrylamyde:N,N’-methylenebisacrylamide, mass ratio) in TBE buffer (89 mM Tris, 89 mM boric acid, 2 mM EDTA) containing 8 M urea.
  2. Mix 20 µL of RNA preparations with 1 volume (20 µL) of loading buffer (98% formamide, 10 mM Tris-HCl, pH 8.0, 1 mM EDTA, 0.0025% bromophenol blue, and 0.0025% xylene cyanol), incubate for 1.5 min at 95 °C in a heating block, and snap cool on ice.
  3. Load the samples in the polyacrylamide gel and run the electrophoresis at appropriate conditions depending on the gel dimensions (e.g., run 140 x 130 x 2 mm gels for 2.5 h at 200 V). Stain the gel for 15 min in 0.5 µg/mL ethidium bromide, wash with water, and visualize RNA under UV light.

Subscription Required. Please recommend JoVE to your librarian.

Representative Results

To produce recombinant RNA in E. coli using the ELVd-derived system12, the RNA of interest is grafted into an ELVd RNA scaffold. This chimeric RNA is co-expressed along the eggplant tRNA ligase in E. coli. Once processed, cleaved and circularized, the chimeric circular RNA, from which the RNA of interest protrudes, likely forms a ribonucleoprotein complex with the co-expressed eggplant enzyme that reaches remarkable concentration in the bacterial cells (Figure 1). Consequently, the first step to produce recombinant RNA using this system consists of inserting the cDNA corresponding to the RNA of interest into a particular position in the ELVd cDNA, T245-T246 in the ELVd sequence variant AJ536613. Plasmid pLELVd-BZB, which contains a polylinker on this position with two Bpi I recognition sites and a LacZ blue/white marker gene, facilitates this insertion by the very efficient Gibson assembly method20. Selection of transformed E. coli clones containing the desired recombinant plasmids is facilitated by the blue/white screening of LacZ marker. Plasmids in which the double Bpi I digestion was incomplete and could not incorporate the insert likely produce blue colonies in the presence of X-gal. Figure 2 shows the electrophoretic analysis of several recombinant plasmids in which different cDNAs were inserted. These plasmids show different migrations when compared to pLELVd-BZB. Note that pLELVd-BZB contains the LacZ marker (528 bp) that is replaced by the cDNA corresponding to the RNA of interest. So, although this depends on the size of the inserted cDNA, the recombinant plasmids usually migrate faster than the control plasmid (Figure 2).

The pLELVd-BZB-derivatives that contain the cDNAs corresponding to the RNAs of interest are used along p15LtRnlSm to co-electroporate E. coli. Co-transformed bacteria are selected on plates containing ampicillin and chloramphenicol. Then, selected E. coli clones are grown at 37 °C with strong agitation in the rich TB medium. Under this culture conditions, production of the recombinant RNA is maximized12. The presence of the recombinant RNA in the bacteria can be easily monitored by breaking the cells with a mix of phenol and chloroform in the presence of a buffer and analyzing the RNA, which partitions in the aqueous buffer, by denaturing PAGE. Figure 3 show ethidium bromide stained polyacrylamide gels in which total RNAs from different E. coli clones were separated. The pictures show strong bands that correspond to empty ELVd and chimeric ELVd forms in which different RNAs of interest were inserted. Interestingly we find a major fraction of the recombinant RNA as circular form. Using a combination of two PAGEs under denaturing conditions at high and low ionic strength, the circularity of the main fraction of the recombinant RNA can be easily observed (Figure 4).

Depending on the downstream application, the total E. coli RNA preparation that contains the chimeric ELVd-RNA of interest may be the goal of the current protocol. However, when needed, the RNA preparation can be further purified by anion-exchange chromatography. Chromatogram in Figure 5 shows the efficient retention of E. coli RNA on the column at low ionic strength (150 mM NaCl) and subsequent elution at high ionic strength (1 M NaCl). Most of the RNA is collected in fractions 2 and 3. To purify the recombinant RNA to homogeneity, advantage of circularity can be taken. Circular RNAs are delayed with respect to their linear counterparts in denaturing conditions22. These RNAs can be eluted from the gel after ethidium bromide staining. The use of a solubilizable polyacrylamide gel, such as those cross-linked with N,N'-bis(acryloyl)cystamine23, facilitates purification of large amounts of recombinant RNAs (Figure 6).

Figure 1
Figure 1: Schematic representation of the viroid-based system to produce recombinant RNA in E.coli. The cDNA corresponding to the RNA of interest (the 98 nt-long RNA aptamer Spinach in the scheme) is inserted into plasmid pLELVd-BZB. E. coli is co-transformed with the pLELVd-BZB-derivative and plasmid p15LtRnlSm to co-express eggplant tRNA ligase. In E. coli, the chimeric ELVd-Spinach RNA transcript is self-cleaved (red arrowheads) by the viroid hammerhead ribozymes. The resulting monomer is recognized by the co-expressed tRNA ligase and circularized. The recombinant RNA consists of a circular viroid scaffold from which the RNA of interest (in green) protrudes. Accumulation of the recombinant RNA to large amounts in E. coli likely results from stabilization of the ribonucleoprotein complex with tRNA ligase. In addition to the split ELVd cDNA, plasmid pLELVd-BZB contains a pUC replication origin (pUC ori), an ampicillin selection gene (AmpR), the E. coli murine lipoprotein (lpp) promoter and rRNA (rrnC) terminator, and the LacZ marker. Plasmid p15LtRnlSm contains a p15A replication origin (p15A ori), a chloramphenicol resistance gene (CmR), the E. coli lpp promoter and a phage T7 transcription terminator, in addition to the eggplant tRNA ligase cDNA. Please click here to view a larger version of this figure.

Figure 2
Figure 2: Electrophoretic analysis of pLELVd-BZB-derived expression plasmids that contain different cDNAs of interest. Plasmids were separated by electrophoresis in a 1% agarose gel that was stained with ethidium bromide. Lane 1, DNA marker with sizes in kbp of some of the components on the left; lane 2, pLELVd-BZB; lane 3, pLELVd-BZB derivative to express an empty ELVd; lanes 4 to 7, pLELVd-BZB derivatives to express the RNA aptamer Spinach (lane 4), the streptavidin binding aptamer (lane 5), an RNA hairpin with a contiguous 42 bp double-stranded region (lane 6), and the 3' cap-independent translation enhancer (CITE) of a plant virus (lane 7). Please click here to view a larger version of this figure.

Figure 3
Figure 3: Recombinant RNA produced in E. coli using the viroid-based system. Total RNAs from different E. coli clones were extracted by treatment with phenol:chloroform and aliquots separated by denaturing PAGE. Gels stained with ethidium bromide are shown. Recombinant RNAs were produced in E. coli (A) BL21(DE3) and (B) HT115(DE3). (A and B) Lanes 1, RNA marker with sizes in nt on the left; lanes 2, RNAs from E. coli clones to express an empty ELVd (333 nt). (A) Lanes 3 and 4, RNAs from E. coli clones to express a chimeric ELVd form containing the aptamer Spinach (98 nt) and the streptavidin binding aptamer (42 nt), respectively. (B) Lane 3, RNAs from an E. coli clone to express a chimeric ELVd form containing a 42 bp-long double stranded RNA. Red arrows point to the circular recombinant RNAs. Please click here to view a larger version of this figure.

Figure 4
Figure 4: Circularity of the recombinant RNA produced in E. coli using the viroid-based system. Total RNAs from an E. coli clone to express a chimeric ELVd that contains the aptamer Spinach were separated by 2D denaturing PAGE first under high and then under low ionic strength. The gels were stained with ethidium bromide. The blue arrows indicate directions of electrophoretic migration. The red arrow points to the circular ELVd-Spinach that is selectively delayed from the linear counterparts of the same size during the two separations. Please click here to view a larger version of this figure.

Figure 5
Figure 5: Chromatographic purification of a recombinant RNA produced in E. coli using the viroid-based system. Total RNAs from an E. coli clone that produces a chimeric ELVd-3' CITE (55 nt) were loaded on a DEAE chromatography column that was washed in the presence of 150 mM NaCl. RNA was eluted in the presence of 1 M NaCl. The chromatogram shows absorbance at 254 nm (blue line) and conductivity in mS/cm (orange line) versus volume. The different steps of the chromatographic separation (equilibration, injection, wash, elution and column regeneration) are indicated. Collected fractions are also indicated. Please click here to view a larger version of this figure.

Figure 6
Figure 6: Purification to homogeneity of a recombinant RNA produced in E. coli using the viroid-based system. Total RNAs from an E. coli clone that produces a chimeric ELVd-3' CITE were first purified by chromatography using a DEAE-Sepharose column and then separated by 2D PAGE. First gel was run under denaturing conditions (8 M urea, buffer TBE). After staining with ethidium bromide, the gel band that contains the circular form of the recombinant RNA was transferred to the top of a second gel that was run under non-denaturing condition (buffer TAE). The acrylamide of the second gel was cross-linked with N,N'-bis(acryloyl)cystamine, which allowed solubilization of the gel fragment that contained the pure recombinant RNA. Please click here to view a larger version of this figure.

Subscription Required. Please recommend JoVE to your librarian.

Discussion

While researching the ELVd sequence and structure requirements involved in the recognition by the eggplant tRNA ligase, we noticed that co-expression of both molecules in the non-host E. coli led to an unexpected large accumulation of viroid circular forms in bacterial cells19. We understood that the large accumulation of viroid RNA in E. coli most probably was the consequence of co-expressing a highly stable RNA molecule, such as the relatively small (333 nt), highly based-paired, circular viroid, and a protein for which this particular viroid displays high affinity. ELVd RNA must recruit the host tRNA ligase to mediate its circularization during replication in the infected plant17. Although we do not have experimental confirmation, we anticipate that both molecules form a ribonucleoprotein complex in E. coli that further stabilizes the viroid RNA and allows reaching the remarkable concentration of 150 mg per liter of bacterial culture that highly exceeds those of endogenous RNAs, such as rRNAs12. Since ELVd does not replicate in E. coli, we reasoned that insertion of a heterologous RNA was not going to dramatically affect accumulation of the viroid-derived RNA and, in fact, that was the case. This observation is the basis of the method to produce large amounts of recombinant RNA in E. coli that we describe here. The method uses the circular RNA molecule of ELVd as a scaffold on which the RNA of interest is presented. To produce a circular ELVd scaffold in E. coli, we need to express a precursor RNA with the duplicated domain of the viroid hammerhead ribozyme. ELVd ribozymes self-cleave very efficiently in E. coli and the resulting viroid monomers are recognized and circularized by the co-expressed tRNA ligase. Co-expression of the tRNA ligase is a key feature of the system. ELVd RNA is hardly detected in E. coli unless this particular enzyme is co-expressed12.

In our protocol, plasmid pLELVd-BZB allows to insert the cDNA corresponding to the RNA of interest by simple and efficient Gibson assembly between positions U245 and U246 of ELVd. This site corresponds to the terminal loop of a long hairpin present in the most likely ELVd conformation24,25. Insertion in this particular position must promote that the RNA of interest protrudes from the very stable scaffold formed by ELVd and must avoid intramolecular interaction between the RNA of interest and ELVd (Figure 1). We have not explored the effect of inserting the RNA of interest in alternative positions of the ELVd molecule. Plasmid p15LtRnlSm allows co-expression of the eggplant tRNA ligase. RNAs are transcribed in both plasmids under the control of a constitutive strong E. coli promoter, namely murein lipoprotein (lpp). This makes the system constitutive with no need of induction. However, we also build a similar plasmid (p15tRnlSm) to express the tRNA ligase under the control of the phage T7 RNA polymerase in an isopropyl β-D-1-thiogalactopyranoside (IPTG) inducible system. Using this plasmid, accumulation of the recombinant RNA only occurs after induction of tRNA ligase expression by adding IPTG12. In the current system, the RNA of interest is expressed from a high copy number plasmid with pUC replication origin, while the tRNA ligase is expressed from a moderate copy number plasmid with p15A replication origin. Whether altering the copy number of both plasmids involved in the system improves accumulation of the recombinant RNA has not been explored so far. The size range of the RNA of interest admitted by the system has not been systematically investigated either, although small RNAs are logically expected to accumulate to higher concentrations than larger ones.

One of the reasons why production of recombinant RNA in vivo is not easy is most probably due to the intrinsic low half-life of RNA molecules. Our system is not alien to this problem. In fact, at a late E. coli growing phase, accumulation of recombinant RNA starts to decrease to virtually disappear12. This forces one to find the right window to harvest the cells. We observed, though, that this window is large enough to make the system friendly. However, we also observed that the optimum window depends on many factors including the particular E. coli strain, culture medium, growing conditions (agitation, volume of the flask, airing, etc.) and the physiological stage of the bacteria used to start the liquid culture. We recommend starting recombinant RNA production using fresh E. coli colonies from a plate inoculated the previous day and grown at 37 °C. Storing plates in the fridge makes the harvest window more unpredictable. We obtained the most consistent results by starting liquid cultures with bacteria picked from a plate with a toothpick. The use of a liquid pre-culture, particularly if saturated, also drives to variability in the production protocol.

Regarding purification, treatment of bacterial cells with phenol:chloroform is a very effective manner to break the cells and allows for the quantitative recovery of total bacterial RNA in the aqueous phase. Depending on the downstream application of the recombinant RNA, this aqueous phase or the preparation that results from precipitating the RNA from this aqueous phase with an alcohol may be enough. If further purification is needed, we recommend anion-exchange chromatography using a DEAE anion exchange column (Figure 5). This purification step allows efficient removal of DNA and bacterial metabolites that also partitioned in the aqueous phase. If the downstream application of the recombinant RNA requires purification to homogeneity, the viroid-derived system offers a clear advantage when compared to other methods. Since a main fraction of the recombinant RNA is produced as a circular form, advantage of this property can be taken to separate it from the bulk of bacterial RNAs. Circular RNAs experience a delay in electrophoretic migration in denaturing conditions with respect to the linear counterparts of the same size22. This delay is more pronounced when the RNA is electrophoresed under lower ionic strength. In practice, circular RNAs can also been separated by two-dimensional denaturing PAGE under high and low ionic strengths (Figure 4). After electrophoretic separation, the recombinant RNA must be eluted from the gel. The use of a reversible cross-linker of acrylamide, such as N,N'-bis(acryloyl)cystamine23, facilitates recovery, particularly when purifying large amounts of RNA (Figure 6). Finally, if downstream application of the produced RNA demands separation of the RNA of interest from the viroid scaffold, our protocol does not offer any particular advantage to previous methods. The RNA of interest must be excised from the chimeric molecule using some of the previously described strategies, such as the use of ribozymes, DNAzymes or, possibly most efficient, the use of RNase H guided by two DNA oligonucleotides7. Trying to avoid this last cumbersome purification step, we have worked on trying to reduce the size of the viroid scaffold, although with partial success. We were able to produce notable amounts of recombinant RNA using viroid deleted forms (175, 215 and 246 nt) but increasing deletions of the viroid scaffold correlated with decreasing accumulation12.

In sum, here a protocol is described to produce large amounts of recombinant RNA in E. coli that, to our knowledge, improves the yield of previously published methods. The protocol is based of co-expression in E. coli of the RNA of interest inserted into a viroid (ELVd) scaffold and the eggplant tRNA ligase, the enzyme that circularizes this viroid in infected plants. A distinctive feature of our protocol is that recombinant RNA is mainly produced as a circular form, a property that facilitates purification to homogeneity for downstream applications. Using this protocol tens of milligrams of recombinant RNA can be easily produced per liter of E. coli culture in regular laboratory conditions.

Subscription Required. Please recommend JoVE to your librarian.

Disclosures

The authors declare that the technology described in this protocol has been patented (US Patent No. EP14382177.5, PCT/EP2015/060912).

Acknowledgements

This work was supported by grants BIO2017-83184-R and BIO2017-91865-EXP from the Spanish Ministerio de Ciencia, Innovación y Universidades (co-financed FEDER funds).

Materials

Name Company Catalog Number Comments
Phusion High-Fidelity DNA polymerase Thermo Scientific F530S
Bpi I Thermo Scientific ER1011
Agarose Conda 8010 D1 low EEO
Tris PanReac AppliChem A1086,1000
Acetic acid PanReac AppliChem 131008.1214
EDTA Sigma-Aldrich E5134-500G
Ethidium bromide PanReac AppliChem A1152,0025 1%
Zymoclean Gel DNA Recovery Zymo Research D4001
NanoDrop ThermoFisher Scientific ND-3300
NEBuilder HiFi DNA Assembly Master Mix New England BioLabs E2621S
DNA Clean & Concentrator Zymo Research D4003
Eporator Eppendorf 4309000019
Escherichia coli DH5α Invitrogen 18265-017
Tryptone Intron Biotechnology Ba2014
Yeast extract Intron Biotechnology 48045
NaCl PanReac AppliChem 131659.1211
Agar Intron Biotechnology 25999
Ampicillin PanReac AppliChem A0839,0010
X-gal Duchefa X1402.1000
N,N-Dimethylformamide PanReac AppliChem 131785.1611
NucloSpin Plasmid Macherey-Nagel 22740588.250
Escherichia coli BL21(DE3) Novagen 69387-3
Escherichia coli HT115(DE3) Ref. Timmons et al., 2001
Chloramphenicol Duchefa C 0113.0025
Glycerol PanReac AppliChem 122329.1211 87%
KH2PO4 PanReac AppliChem 131509.1210
K2HPO4 PanReac AppliChem 122333.1211
HCl PanReac AppliChem 131020.1211
Phenol Scharlau FE04791000 90%
Chloroform PanReac AppliChem A3691,1000
Filtropur S 0.2 Sarstedt 83.1826.001
HiTrap DEAE Sepharose FF column GE Healthcare Life Sciences 17-5055-01
ÄKTAprime plus liquid chromatography system GE Healthcare Life Sciences 11001313
Acrylamyde PanReac AppliChem A1089,1000 2K
N,N’-methylenebisacrylamide Sigma-Aldrich M7279-100G
Urea PanReac AppliChem 146392.1211
Boric acid PanReac AppliChem A2940,1000
Formamide PanReac AppliChem A0937,2500
Bromophenol blue Sigma-Aldrich B8026-5G
Xylene cyanol Sigma-Aldrich X4126-10G
N,N′-Bis(acryloyl)cystamine Sigma-Aldrich A4929-5G

DOWNLOAD MATERIALS LIST

References

  1. Wang, J., et al. Current Research on Non-Coding Ribonucleic Acid (RNA). Genes. 8, (12), (2017).
  2. Hamada, M. In silico approaches to RNA aptamer design. Biochimie. 145, 8-14 (2018).
  3. Bobbin, M. L., Rossi, J. J. RNA Interference (RNAi)-Based Therapeutics: Delivering on the Promise? Annual Review of Pharmacology and Toxicology. 56, 103-122 (2016).
  4. Airs, P. M., Bartholomay, L. C. RNA Interference for Mosquito and Mosquito-Borne Disease Control. Insects. 8, (1), 4 (2017).
  5. Ponchon, L., Dardel, F. Large scale expression and purification of recombinant RNA in Escherichia coli. Methods. 54, (2), 267-273 (2011).
  6. Kikuchi, Y., Umekage, S. Extracellular nucleic acids of the marine bacterium Rhodovulum sulfidophilum and recombinant RNA production technology using bacteria. FEMS Microbiology Letters. 365, (3), fnx268 (2018).
  7. Ponchon, L., Dardel, F. Recombinant RNA technology: the tRNA scaffold. Nature Methods. 4, (7), 571-576 (2007).
  8. Batey, R. T. Advances in methods for native expression and purification of RNA for structural studies. Current Opinion in Structural Biology. 26, 1-8 (2014).
  9. El Khouri, M., et al. Expression and Purification of RNA-Protein Complexes in Escherichia coli. Methods in Molecular Biology. 1316, 25-31 (2015).
  10. Ponchon, L., et al. Co-expression of RNA-protein complexes in Escherichia coli and applications to RNA biology. Nucleic Acids Research. 41, (15), e150 (2013).
  11. Umekage, S., Kikuchi, Y. In vitro and in vivo production and purification of circular RNA aptamer. Journal of Biotechnology. 139, (4), 265-272 (2009).
  12. Daròs, J. -A., Aragonés, V., Cordero, T. A viroid-derived system to produce large amounts of recombinant RNA in Escherichia coli. Scientific Reports. 8, (1), 1904 (1904).
  13. Daròs, J. A. Viroids: small noncoding infectious RNAs with the remakable ability of autonomous replication. Current Research Topics in Plant Virology. Springer International Publishing Switzerland. 295-322 (2016).
  14. Flores, R., Hernández, C., Martínez de Alba, A. E., Daròs, J. A., Di Serio, F. Viroids and viroid-host interactions. Annual Review of Phytopathology. 43, 117-139 (2005).
  15. Di Serio, F., et al. ICTV Virus Taxonomy Profile: Avsunviroidae. The Journal of General Virology. 99, (5), 611-612 (2018).
  16. Nohales, M. A., Flores, R., Daròs, J. A. Viroid RNA redirects host DNA ligase 1 to act as an RNA ligase. Proceedings of the National Academy of Sciences of the United States of America. 109, (34), 13805-13810 (2012).
  17. Nohales, M. A., Molina-Serrano, D., Flores, R., Daròs, J. A. Involvement of the chloroplastic isoform of tRNA ligase in the replication of viroids belonging to the family Avsunviroidae. Journal of Virology. 86, (15), 8269-8276 (2012).
  18. Daròs, J. A. Eggplant latent viroid: a friendly experimental system in the family Avsunviroidae. Molecular Plant Pathology. 17, (8), 1170-1177 (2016).
  19. Cordero, T., Ortolà, B., Daròs, J. A. Mutational Analysis of Eggplant Latent Viroid RNA Circularization by the Eggplant tRNA Ligase in Escherichia coli. Frontiers in Microbiology. 9, (635), (2018).
  20. Gibson, D. G., et al. Enzymatic assembly of DNA molecules up to several hundred kilobases. Nature Methods. 6, (5), 343-345 (2009).
  21. Timmons, L., Court, D. L., Fire, A. Ingestion of bacterially expressed dsRNAs can produce specific and potent genetic interference in Caenorhabditis elegans. Gene. 263, (1-2), 103-112 (2001).
  22. Schumacher, J., Randles, J. W., Riesner, D. A two-dimensional electrophoretic technique for the detection of circular viroids and virusoids. Analytical Biochemistry. 135, (2), 288-295 (1983).
  23. Hansen, J. N., Pheiffer, B. H., Boehnert, J. A. Chemical and electrophoretic properties of solubilizable disulfide gels. Analytical Biochemistry. 105, (1), 192-201 (1980).
  24. Giguère, T., Adkar-Purushothama, C. R., Bolduc, F., Perreault, J. P. Elucidation of the structures of all members of the Avsunviroidae family. Molecular Plant Pathology. 15, (8), 767-779 (2014).
  25. López-Carrasco, A., et al. The transcription initiation sites of eggplant latent viroid strands map within distinct motifs in their in vivo RNA conformations. RNA Biology. 13, (1), 83-97 (2016).

Comments

0 Comments


    Post a Question / Comment / Request

    You must be signed in to post a comment. Please or create an account.

    Usage Statistics