Targeting 16S ribosomal DNA for stable recombinant gene expression

Subscriber access provided by UNIV OF SOUTHERN INDIANA

Article

Targeting 16S ribosomal DNA for stable recombinant gene expression in Pseudomonas Maike Otto, Benedikt Wynands, Thomas Drepper, Karl-Erich Jaeger, Stephan Thies, Anita Loeschcke, Lars M. Blank, and Nick Wierckx ACS Synth. Biol., Just Accepted Manuscript • DOI: 10.1021/acssynbio.9b00195 • Publication Date (Web): 12 Jul 2019 Downloaded from pubs.acs.org on July 18, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

Targeting 16S ribosomal DNA for stable recombinant gene expression in Pseudomonas Maike Otto1,2, Benedikt Wynands1,2, Thomas Drepper3, Karl-Erich Jaeger2,3,4, Stephan Thies3,4, Anita Loeschcke3,4, Lars M. Blank1,4 and Nick Wierckx1,2*

1 Institute

of Applied Microbiology, RWTH Aachen University, 52074 Aachen, Germany

2 Institute

of Bio- and Geosciences (IBG-1: Biotechnology), Forschungszentrum Jülich GmbH, 52425

Jülich, Germany 3 Institute

of Molecular Enzyme Technology, Heinrich Heine University Düsseldorf,

Forschungszentrum Jülich GmbH, 52425 Jülich, Germany 4 Bioeconomy

Science Center (BioSC), Forschungszentrum Jülich GmbH, 52425 Jülich, Germany

*: corresponding author, Nick Wierckx, Institute of Bio- and Geosciences, IBG-1: Biotechnology, Forschungszentrum Jülich GmbH, 52425 Jülich, Germany, e-mail: [email protected], phone: +49 2461 61 85247

1 ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 37

Abstract Ribosomal RNA (rRNA) operons have recently been identified as promising site for chromosomal integration of genetic elements in Pseudomonas putida, a bacterium that has gained considerable popularity as microbial cell factory. We have developed a tool for targeted integration of recombinant genes into the rRNA operons of various Pseudomonas strains, where the native context of the rRNA clusters enables effective transcription of heterologous genes. However, a sufficient translation of foreign messenger RNA (mRNA) transcriptionally fused to rRNA required optimization of RNA secondary structures, which was achieved utilizing synthetic ribozymes and a bicistronic design. The generated tool further enabled the characterization of the six rRNA promoter units of P. putida S12 under different growth conditions. The presence of multiple, almost identical rRNA operons in Pseudomonas also allowed the integration of multiple copies of heterologous genetic elements. The integration of two expression cassettes and the resulting disruption of rRNA units only moderately affects growth rates and the constructs were highly stable over more than 160 generations.

Keywords Synthetic biology, ribosomal RNA, ribozymes, bicistronic design, recombinant gene expression, Pseudomonas


Page 3 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Bacteria of the genus Pseudomonas are receiving increasing attention as microbial cell factories for various biotechnological applications. Pseudomonads are endowed with an extraordinarily versatile metabolism that allows the assimilation of many different substrates1,2 and feature a distinct tolerance towards organic solvents3–5 and oxidative stress6, allowing growth under harsh environmental conditions. It is thus no surprise that many different industrially relevant compounds, including aromatics7,8,9, rhamnolipids10, terpenes11, and prodiginines12, have recently been synthesized with Pseudomonas. Their efficient engineering is only possible due to a rapidly extending molecular toolbox, enabling genetic manipulations such as deletion or disruption of genomic regions13–16 or the expression of genetic elements from standardized vector platforms17. Vector systems allow straightforward implementation of heterologous gene expression, but are associated with a number of disadvantages, including plasmidinduced growth defects18, and a lack of reproducibility as a result of clonal variability19,20 due to plasmid instability21, loss22 or varying copy numbers23. These drawbacks can be avoided by integrating an expression cassette into the chromosome of a host organism24. Prominent examples for genomic integration are the utilization of recombinases25 or transposons26 as delivery system. A widely used tool in the Pseudomonas community is Tn7-based transposition27, where heterologous cassettes including a promoter unit can be directed into the attTn7-site, a specific neutral site downstream of the glmS gene28 for non-interfering heterologous gene expression. As this site-specific transposon is only functional for the respective chromosomal site, further “landing pads” for genetic elements remain to be identified. This issue has recently been addressed with the establishment of a bacteriophage BxB1-integrase system in P. putida KT2440, specifically targeting the attB site29. This site however is not natively present in Pseudomonas and prior integration of the Ptac-integrase/attB cassette as target must be performed to allow integration. Non-biased transposition (e.g. Tn5-based) delivers an alternative method, which offers the opportunity to integrate gene cassettes at additional genomic sites30. The expressional control of a genetic element delivered via non-biased transposition can either occur through regulated promoter units included in the delivered cargo (e.g. LacIQ/Ptrc31,32) or by a native chromosomal expression unit33. Nevertheless, this method requires a laborious screening effort to obtain sufficient expression levels and to determine the respective chromosomal insertion site. 3 ACS Paragon Plus Environment


Page 4 of 37

A recent study demonstrated the genomic integration of the 21 kb prodigiosin biosynthesis cluster (pig) from Serratia marcescens into P. putida12 via the Tn5-based TREX-system34, relying on recombinant expression from a native promoter. Sequencing analysis of the most efficient prodigiosin producers revealed that in these clones, transposition invariably occurred within genes encoding for ribosomal RNA (rRNA)35. To maintain the high abundance of rRNA that is required to enable translation rates appropriate to the cell’s growth condition36, most microorganisms, including Pseudomonads, have multiple copies of rRNA operons (rrn) transcribed from highly active promoter units37,38. The promoter organization of rRNA operons can vary between different species39,40 and the transcriptional activity is significantly influenced by upstream cis-acting UP elements and trans-acting transcription factors41, altogether enabling an exceptionally high transcriptional activity. While in trans cloning of the rrn promoters into vectors can deliver alternative heterologous expression vectors for Pseudomonas, the above-mentioned drawbacks of plasmid systems endorse in cis targeting of the native locus, relying on multiple rrn copies to compensate for any negative effects of the insertion. Furthermore, rRNA is highly stable and only degraded under specific conditions42. The integration of heterologous genes into a stable rRNA context might also positively influence the stability and accessibility of the target gene mRNA, possibly leading to improved translation. Ribosomal operons are thus a promising target location for directed integration, to “hijack” and make use of the high native promoter activity for heterologous gene expression. In this work, we have developed a tool for targeted integration of recombinant genes into rrn operons of Pseudomonas species, based on the pEMG-vector system for genome editing13. Via homologous recombination, the coding sequence for a monomeric superfolder green fluorescent protein (MsfGFP)43 was introduced into the coding region of the 16S subunit. As revealed by low expression of the reporter gene, translation of mRNA in an rRNA fusion background yet requires optimization of RNA secondary structures. This was achieved by introducing minimal hammerhead ribozymes44 that resulted in autocatalytic cleavage of flanking rRNA regions from the mRNA, thereby removing rRNA secondary structures. In addition, the replacement of the classical ribosomal binding site (RBS) by a translational coupler (bicistronic design, BCD45) further enhanced translation rates. The herewith developed tool also enabled the characterization of the six rrn promoters in the strain P. putida S12, unraveling differential 4 ACS Paragon Plus Environment

Page 5 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


expression from the various clusters. The presence of multiple, almost identical rRNA operons further allows the integration of multiple genetic elements at different rrn operons, with only slight impact on growth rates despite the disruption of one functional rrn unit. Sequential batch cultivations of a strain bearing two rrn insertions over more than 160 generations had no impact on the integration stability, confirming the high suitability of rRNA operons for genomic integration and expression of heterologous genes.



Page 6 of 37

Results and Discussion Development of a tool for directed integration at 16S rDNA sites Tool development for heterologous gene expression in Pseudomonas putida is crucial to improve its potential as biotechnological workhorse. While plasmid-based systems remain an important instrument for straightforward recombinant gene expression, genomic integration represents an alternative with numerous advantages. These include an enhanced expression stability compared to plasmid systems and reduced growth defects due to the use of selection markers18,28. We therefore aimed for targeted gene integration into the chromosomal sites encoding ribosomal RNA in Pseudomonas. The rrn operons in P. putida S12 are transcribed from two promoters and consist of three rRNA (16S, 23S, 5S) and two tRNA genes (Figure 1A)38,46. The operons are transcribed as a single molecule of RNA which is subsequently processed by different RNases into the mature ribosomal components47. The 16S gene has the closest proximity to the promoter region and an integration within this sequence as opposed to the 23S gene has recently been shown to deliver higher expression for heterologous genes transcribed from rrn promoters35. Hence, the tool was designed for the integration of a recombinant genetic element between bases 652 and 653 of the 1574 bp 16S structural gene. Targeted integration at the 16S sites was achieved by employing vectors based on the pEMG plasmid system used as gene editing platform in Gram-negative bacteria13. The msfgfp43 reporter gene was cloned between two 400 bp sequences homologous to a section of the 16S rDNA of Pseudomonas, amplified from strain P. putida S12 (Figure 1B) in a pEMG backbone to generate the integration vector (Figure 1C). Two multiple cloning sites (MCS) flanking msfgfp were included in the resulting vector, which was designated as pMO-msfgfp, to facilitate the exchange of genes of interest. The pMO-msfgfp vectors contain one I-SceI site as target for the I-SceI nuclease to generate a double strand break, leading to curing of the pMO backbone after single crossover by homologous recombination. The original pEMG construct contains two I-SceI sites, thereby increasing the efficiency of recombination in this step. The presence of only one I-SceI site in the pMO vector however had no impact in the applicability of the tool, as recombination and selection of integrants was possible with a moderate screening effort. After successful integration, msfgfp is under control of the constitutively active rRNA promoter region (Figure 1D). 6 ACS Paragon Plus Environment

Page 7 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


As highly conserved regions within the 16S gene were amplified for tool construction, it enables the integration at all existing 16S sites in P. putida S12 and in different strains and species of Pseudomonas (Figure 1B). In this study, chromosomal integration of the pMO vector series was achieved in P. putida S12, P. putida KT2440 and P. taiwanensis VLB120. Although the frequency of homologous recombination of genetic elements with the chromosome is known to increase with an increasing length (and sequence similarity) of the homologous regions48, those were deliberately designed with moderate length. Since the sequence identity of the 16S region within the Pseudomonas genus lies between 97100 %, this design enables an integration also in a larger number of Pseudomonas representatives. Pseudomonads have multiple rrn operons at varying chromosomal loci. A schematic representation of the chromosomal loci of all rrn operons in P. putida S12, P. putida KT2440 and P. taiwanensis VLB120 is given in Figure 1E. While both P. putida KT2440 and P. taiwanensis VLB120 inherit seven rrn sites as a result of one tandem operon (rrnA/rrnB), P. putida S12 holds six rrn operons at different chromosomal locations. To exclude the possibility of a DNA sequence assembly error, the missing tandem rrn operon in P. putida S12 was confirmed by PCR (data not shown). Integration of the pMO backbone occurs randomly in one of these loci. A PCR-based identification of the insertion site was therefore developed for the six (P. putida S12) or seven (P. putida KT2440, P. taiwanensis VLB120) individual loci with specific primer combinations that extend into the flanking regions of the rrn operons (Supplement Figure S1) to enable site mapping to the specific Pseudomonas rrn operons (Figure 1E).



Page 8 of 37

Figure 1. Integration principle of recombinant genes at 16S rDNA sites. (A) General organization of rrn operons in Pseudomonas putida S12. The 16S, 23S and 5S rRNA subunits and two tRNAs are clustered and transcribed from two promoters. (B) Alignment of the different copies of 16S rDNA of P. putida S12 (NCBI reference sequence CP009974), P. putida KT2440 (NCBI reference sequence NC_002947.4) and P. taiwanensis VLB120 (NCBI reference sequence CP003961.1). Blue areas represent 100% sequence identity, the yellow bars indicate single nucleotide differences compared to the reference sequence (S12 rrnA). Dotted lines indicate the two 400 bp regions chosen as homologous recombination flanks. (C) Map of the pMO-msfgfp integration vector. Annotations: RBS, ribosomal binding site; MCS, multiple cloning site; KmR, kanamycin resistance marker; traJ and oriT elements for conjugative mobilization; ori R6K, π protein-dependent origin of replication; I-SceI, recognition site for the I-SceI homing endonuclease; 16S-TS1/2, 16S target site 1/2; DSB, DNA double strand break. An abbreviated workflow is depicted below the vector, leading to (D) the final, chromosomally integrated construct, where the gene of interest is integrated between bases 652 and 653 of the 16S gene and transcribed from the native rrn promoter unit. Genetic elements are not drawn to scale. (E) Linear representation of the chromosomes of the three Pseudomonads tested in this study, starting from the origin of replication (oriC) and relative positions of the rrn operons therein. The published genome sequence of P. putida S12 was rotated and inverted to match the relative positions of the operons. The colored boxes next to the operons indicate matching flanking regions of 2000 bp (including the respective promoter sequences) with at least 77% DNA sequence similarity. Operon size and distances are not drawn to scale.


Page 9 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Improving mRNA translation for enhanced protein expression In an initially designed construct, the msfgfp gene including a standard upstream ribosomal binding site (RBS, sequence ggagga)49 was integrated between bases 652 and 653 of the 16S gene into each rrn operon (A-F) of P. putida S12 (Figure 2A). Low MsfGFP fluorescence was detected for all clones, independent of the integration site (Supporting information, Figure S2), shown for rrnD in figure 2B (rrnD::msfgfp). RT-qPCR analysis revealed sufficient amounts of msfgfp mRNA, indicating successful transcription of heterologous genes from rrn promoters (Figure 2B). As controls, the same msfgfp was expressed from calibrated promoters with increasing expression strengths, P14a (weak), P14d (medium) and P14g (strong) integrated via Tn7-transposition at the attTn7-site28. Here, increasing expression strengths of the synthetic promoters led to increasing amounts of transcripts which were effectively translated into MsfGFP as reflected by the respective fluorescence values. In contrast, expression of msfgfp with a standard RBS from the rrnD (rrnD::msfgfp) promoter led to transcript levels which were only slightly lower compared to promoter P14g. At the same time, much weaker MsfGFP fluorescence in the range of promoter P14a was observed, indicating insufficient translation of the msfgfp mRNA with this setup. The msfgfp is flanked by 16S rRNA fragments (Figure 2A) and transcription of this construct leads to an rRNA-mRNA hybrid. As rRNA is not translated and forms strong secondary structures50, we hypothesized that interactions of the rRNA with the msfgfp can inhibit translation by blocking the RBS and impeding proper ribosome recruitment. We therefore added different synthetic elements to the integration vector to alter RNA secondary structures and subsequently analyzed expression from the same promoter (operon rrnD) to maintain equal transcription rates. Ribozymes described by Lou et al.44 were originally designed to insulate 5’- untranslated regions of mRNA to achieve quantitatively identical output of heterologous expression systems. They consist of an sTRSV-ribozyme51 (satellite RNA of tobacco ringspot virus) with autocatalytic cleavage activity at a defined residue and a 23-nt hairpin52 for enhanced accessibility of the RBS. The ribozymes riboJ and ltsvJ were cloned upstream of the RBS and/or downstream of msfgfp, respectively, to remove the 16S flanks post-transcriptionally (Figure 2A, rrn::riboJmsfgfp and rrn::riboJmsfgfpltsvJ). PCR analysis using cDNA of the different constructs confirmed cleavage activity of riboJ and ltsvJ (Figure 2C). RiboJ cleavage of the upstream 16S-flank in the 9 ACS Paragon Plus Environment


Page 10 of 37

rrn::riboJmsfgfp construct led to a 17-fold enhanced MsfGFP fluorescence compared to the initial integration setup (Figure 2B). We attribute this primarily to the likely reduction of rRNA-mRNA interactions and steric hindrance of ribosome docking, thereby enhancing translation efficiency by the removal of the upstream 16s region, further aided by the hairpin formed by the ribozyme. In addition, a 58% increase of msfgfp mRNA was detected by RT-qPCR. As expression was analyzed from the same rrn promoter and the transcriptional output can be assumed to remain identical due to the long distance between promoter and construct, this effect can most likely be explained by an improvement of mRNA stability due to riboJ cleavage. The insertion of heterologous genes in the center of a 16S gene will interfere with correct folding of the 16S rRNA subunit after transcription, alter structures that otherwise protect the rRNA against cellular RNases, and inhibit correct ribosome assembly. These factors trigger rRNA degradation42,53 and thus might also have an impact on the stability of the msfgfp transcripts. RiboJ cleavage and the resulting lack of an upstream 16S rRNA flank might trigger these quality control mechanisms less intensively, leading to slower degradation. Furthermore, enhanced translation rates lead to a higher mRNA ribosome load, which further protects from RNase activity54,55. The additional insertion of ltsvJ downstream of the msfgfp gene in the clone P. putida S12 rrnD::riboJmsfgfpltsvJ leads to cleavage of the downstream 16S flank, leaving an unprotected 3’- end of the mRNA (Figure 2A). Strains containing the ltsvJ element showed low MsfGFP fluorescence and decreased transcript levels (Figure 2B), as free 3’ends are quickly recognized by exoribonucleases, leading to a rapid decay56. The strong secondary structures of the downstream 16S rRNA thus appear to protect the msfgfp mRNA from degradation, resulting in stable transcripts. Further translation improvement was achieved by replacing the standard RBS by a synthetic bicistronic design (BCD)45. In this study, the BCD2 element was utilized, a translation initiation sequence consisting of two Shine-Dalgarno sequences and a 16-amino-acid leader peptide transcribed from the first RBS, overlapping by one base with the start codon of a gene of interest (GOI). This element aids in reducing GOI-based variation in expression45, and can also enhance GFP expression compared to a single RBS28. The replacement of the standard RBS by BCD2 in the rrn integration constructs led to a 23-fold enhanced MsfGFP signal (Figure 2B, rrnD::BCDmsfgfp). The insertion of this 77-bp spacer between the upstream 16S flank and the msfgfp apparently improves the accessibility of the RNA to 10 ACS Paragon Plus Environment

Page 11 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


initiate translation, indicating that steric hindrance is a predominant mode of translation inhibition in the original construct. The combination of BCD2 with cleavage of the upstream 16S region by riboJ in the rrnD::riboJ-BCDmsfgfp setup led to a 73% enhanced fluorescence signal compared to BCD2 alone and a 40-fold expression improvement compared to the initial integration constructs with a standard RBS (Figure 2B). This is a significant improvement over the single modifications with riboJ or BCD2, but the effect is not cumulative, indicating that they both mostly impact a similar phenomenon of ribosome access. In both expression variants, increased transcript levels compared to the rrn::msfgfp integration variants were detected, which can be linked to the secondary structure alterations leading to improved stability of the mRNA as discussed above. In both cases, where riboJ is added (rrnD::BCDmsfgfp vs. rrnD::riboJ-BCDmsfgfp), the transcript increases by the same factor (1.8). Promoter activity of the BCD2 element itself can be ruled out28. The same expression trend for riboJ cleavage and BCD2 translation improvement was observed in the strain P. putida KT2240 (Figure S4, supporting information). In P. taiwanensis VLB120, riboJmsfgfp integration at site rrnF was performed as proof-of-concept and resulted in GFP fluorescence at a similar level to the equivalent P. putida strains (Figure S4). While absolute fluorescence values vary depending on the construct, the trend of fluorescence development is similar for all constructs, as exemplarily shown for P. putida S12 rrnE::riboJmsfgfp in the supplementary material (Figure S3). Fluorescence develops parallel to the OD600, as expression of rrn operons is coupled to growth. Only a slight increase in fluorescence is visible shortly after onset of the stationary phase, likely due to MsfGFP maturation.



Figure 2. Translation of mRNA in an rRNA fusion background requires secondary structure alterations to enhance protein expression. (A) The insertion of different synthetic elements and combinations thereof were constructed to obtain differently structured RNA transcripts generated from the rrn sites. Genetic elements and corresponding RNA molecules are not drawn to scale. Annotations: RBS, ribosomal binding site; riboJ, synthetic ribozyme; ltsvJ, synthetic ribozyme; BCD, bicistronic design. (B) Transcript levels, determined by RT-qPCR, and MsfGFP fluorescence from different constructs integrated into the rrnD site of P. putida S12 in MSM containing 20 mM of glucose. The normalized expression of MsfGFP is represented by the slope of the linear increase of fluorescence and OD600 during exponential growth. As controls, P. putida S12 strains expressing MsfGFP from calibrated synthetic promoters with increasing strength, chromosomally integrated at the attTn7-site, were analyzed. Error bars indicate the standard error of the mean (n = 3). (C) Map of primer sites used for confirmation of RNA ribozyme cleavage and agarose gel electrophoresis of corresponding PCR products from reactions on cDNA of the indicated strains carrying the integration constructs in rrnD with different primer combinations. A PCR product for primer combinations 153/156 and 155/154 indicates intact RNA between 16S flanks and msfgfp mRNA. A missing product confirms cleavage between rRNA and mRNA.

However, the translational efficiency of the constructs remains subject to further improvement. Compared to the calibrated promoter P14g, the construct rrnD::riboJ-BCDmsfgfp showed 80% higher transcript levels in P. putida S12, but only a 20% higher MsfGFP fluorescence (Figure 2B). The full translational potential might be recovered by introducing additional sequence motifs into the constructs. While the 16S downstream flank stabilizes the mRNA compared to free 3’-end, it might to some extent impair translation through RNA interactions. This hypothesis is further supported by the results shown 12 ACS Paragon Plus Environment

Page 12 of 37

Page 13 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


in Figure S5 (Supporting information). Here, a replacement of the transcriptional terminator in the attTn7-integration cassette of the pBG14x vector28 with a section of the 16S region led to reduced MsfGFP expression. An insertion of a transcriptional terminator57,58 into the rrn integration vector could eliminate these effects and still sufficiently stabilize the 3’-end of the transcript by the formation of less “bulky” loop structures compared to the 16S flank. Nevertheless, it will also lead to transcriptional termination of the downstream 23S, 5S and tRNA sequences and the influence of this effect on the expression of the whole rrn operon remains to be investigated. In a further attempt to boost the expression strength, the synthetic promoter P14g28 was introduced upstream of the riboJ-BCDmsfgfp cargo (Supplement Figure S6). Integration of this construct at site rrnD however led to slightly decreased expression compared to

riboJ-BCDmsfgfp,

in the range of msfgfp

expression from P14g integrated at site attTn7. In dual promoter systems, the expression strength of a promoter can be impaired by the transcribing complex approaching from the rrn promoter unit further upstream59. In this case, the transcriptional output might result from impaired transcription from both promoter units. Nevertheless, the expression strengths of the native rrn promoter units alone deliver transcriptional outputs comparable to established synthetic expression systems28,45. In conclusion, the integration tool enables efficient expression of heterologous genes in rrn operons, and it delivers a library of varying expression strengths for heterologous gene expression by modulating translation efficiency of mRNA transcripts.



Page 14 of 37

Expression analysis of different rrn promoters The transcriptional output from the different rrn promoters was subsequently analyzed using the rrn::riboJmsfgfp expression cassette in P. putida S12. Integration into each rrn site occurred in a random manner and single integrations into each of the six operons were selected by PCR screening as described before. Growth and MsfGFP fluorescence of the strains were analyzed during BioLector® cultivations in mineral salts medium (MSM) using either succinate, glucose or fructose as sole carbon source to analyze the effect of the different carbon sources on activation of the respective rrn operons (Figure 4). The integration of our heterologous gene construct into an rrn operon leads to a disruption of one 16S coding sequence, likely resulting in a loss of functional 16S subunits stemming from this operon. As observed for other organisms such as Escherichia coli60 and Corynebacterium glutamicum61, the loss of one rrn unit can be compensated by the remaining intact units to maintain stable growth rates. Similar effects were observed for the P. putida rrn integration clones, as an insertion and disruption of 16S genes had little or no impact on growth rates compared to the wild-type, regardless of the carbon source. On succinate, insertion at operon rrnE led to a significant 20% decrease in growth rate, while the remaining strains displayed rates comparable to the wild-type, indicating the importance of this operon during the fast growth on succinate. On glucose, an integration into site rrnB and rrnE led to 10% decreased growth rates, while on fructose, an integration at site rrnA, rrnB, rrnC and rrnF led to 1020% decreased rates. Overall, cells remained vital and displayed stable growth behavior, despite the disruption of one out of six 16S subunits. The remaining intact operons can apparently compensate negative effects of the integration, underlining the suitability of rrn operons for chromosomal integration of heterologous genetic elements. In line with increasing growth rates due to varying metabolism of the different carbon sources62,63, an increase of expression from each rrn promoter was observed. The amount of ribosomes within a cell determines the translation rate of proteins, it thus correlates directly with the growth rate under a given condition36. The rate of ribosome synthesis is adjusted via the regulation of gene activity64, leading to an up- or downregulation of rDNA transcription at different growth rates. At the same time, the promoter units of the different rrn sites exhibited varying expression strengths with identical trends on all carbon sources. Sites rrnA, rrnB, rrnC, and rrnF display comparable 14 ACS Paragon Plus Environment

Page 15 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fluorescence, while in comparison sites rrnD and rrnE show about 150% and 30% higher expression levels, respectively. Sequence alignment of the upstream region of the six 16S coding regions of P. putida S12 (Supporting information, Figure S7) indicates variations in the promoter regions of the rrn operons which correlate with these trends. While the -10/-35 region of the first promoter (closest proximity to the 16S coding region) and the sequence downstream of this promoter are largely identical for all operons, the region further upstream shows clear sequence differences for rrnD and rrnE compared to the remaining operons. Most striking are (i) sequence aberrations in the second -10/-35 box, (ii) different distances between the -10/-35 boxes of the two promoters, with 19 additional base pairs at both promoter regions of site rrnE and rrnD compared to the other four operons, and (iii) multiple sequence differences upstream of the second promoter. Regulation of the rrn operons involves, besides recognition of the -10/-35 boxes by the sigma factor, further activators and repressors binding surrounding regions of the promoter. As observed in E. coli, UP elements, which are AT-rich regions located upstream of the second -35 box, are recognized by the α subunit of the RNA polymerase (RNAP) and regulate expression strengths. In addition, the binding of a trans-acting transcription factor FIS (factor for inversion stimulation) to a region upstream of the UP element attribute to the transcriptional output of ribosomal promoters41. This gives a clear indication for differential regulation and the resulting enhanced strength of the operons rrnE and rrnD, as especially the regions upstream of the first promoter differ from the sequences of the other four rrn sites. Furthermore, the effect of cellular stress on the expression of the different operons was tested by the addition of 0.8 M urea in mineral salts medium containing 20 mM of glucose (Figure 4). Due to its chaotropic effects, urea impairs protein stability and reduces enzyme activity65, accompanied by osmotic stress66. This is reflected by the reduced growth rates of both the rrn integration strains and the wildtype. Growth rates remain comparable for all strains, further underlining the strains capability to compensate for the loss of one intact rrn unit, even under conditions with enhanced stress. However, while growth rates in the presence of urea are reduced by around 45%, expression of MsfGFP remains in the range of expression (rrnA, rrnC, rrnF) or reduced by only around 20% (rrnB, rrnD, rrnE). This indicates that in this case the induction of stress response mechanisms includes an increased transcriptional activity of the rrn units, possibly to increase the cellular protein content as a 15 ACS Paragon Plus Environment


Page 16 of 37

compensation for reduced enzyme activity, or to facilitate the expression of stress response genes. While temperature38 or nutrient limitation67 have shown to upregulate specific, single rrn operons, the upregulation of one specific operon in P. putida S12 as a result of urea stress was not observed.

Figure 3. Low impact of rrn integration on growth rates and differential msfgfp expression from the different rrn promoters of P. putida S12. The construct riboJmfsgfp was integrated into the different rrn sites in P. putida S12. Growth and fluorescence were monitored in BioLector® experiments with MSM with a five-fold increased buffer capacity supplemented with different carbon sources (30 mM succinate, 20 mM glucose, 20 mM fructose) and a combination of 0.8 M urea and glucose (20 mM). The normalized expression of MsfGFP is represented by the slope of the linear increase of the fluorescence and OD600 during exponential growth. Error bars indicate the standard error of the mean (n = 3). Growth rate values that differ significantly from the corresponding values of the wild-type are indicated with * for p

Targeting 16S ribosomal DNA for stable recombinant gene expression

Recommend Documents