ABSTRACT
Small RNAs (sRNAs) are crucial regulatory molecules in organisms and are well-known not only for their roles in the control of diverse crucial biological processes but also for their value in regulation rewiring. However, to date, in Gram-positive anaerobic solventogenic clostridia (a group of important industrial bacteria with exceptional substrate and product diversity), sRNAs remain minimally explored, and thus there is a lack of detailed understanding regarding these important molecules and their use as targets for genetic improvement. Here, we performed large-scale phenotypic screens of a transposon-mediated mutant library of Clostridium acetobutylicum, a typical solventogenic clostridial species, and discovered a novel sRNA (sr8384) that functions as a crucial regulator of cell growth. Comparative transcriptomic data combined with genetic and biochemical analyses revealed that sr8384 acts as a pleiotropic regulator and controls multiple targets that are associated with crucial biological processes through direct or indirect interactions. Notably, the in vivo expression level of sr8384 determined the cell growth rate, thereby affecting the solvent titer and productivity. These findings indicate the importance of the sr8384-mediated regulatory network in C. acetobutylicum. Furthermore, a homolog of sr8384 was discovered and proven to be functional in another important Clostridium species, C. beijerinckii, suggesting the potential broad role of this sRNA in clostridia. Our work showcases a previously unknown potent and complex role of sRNAs in clostridia, providing new opportunities for understanding and engineering these anaerobes.
IMPORTANCE The uses of sRNAs as new resources for functional studies and strain modifications are promising strategies in microorganisms. However, these crucial regulatory molecules have hardly been explored in industrially important solventogenic clostridia. Here, we identified sr8384 as a novel determinant sRNA controlling the cell growth of solventogenic Clostridium acetobutylicum. Based on a detailed functional analysis, we further reveal the pleiotropic function of sr8384 and its multiple direct and indirect crucial targets, which represents a valuable source for understanding and optimizing this anaerobe. Of note, manipulation of this sRNA achieves improved cell growth and solvent synthesis. Our findings provide a new perspective for future studies on regulatory sRNAs in clostridia.
INTRODUCTION
Historically, the application of solventogenic clostridia in the large-scale production of the bulk chemicals acetone, n-butanol, and ethanol, a process called ABE fermentation, has demonstrated the value of these anaerobic microorganisms (1, 2). In recent years, in view of the exceptional substrate and product diversity of solventogenic clostridia, the biological production of cost-effective bulk chemicals and biofuels using Clostridium species as chassis has attracted renewed attention (3). To unlock the full potential of solventogenic clostridia in industrial applications, a detailed understanding of metabolic regulation and the discovery of more crucial regulatory elements in these anaerobes are necessary. However, to date, this aspect remains minimally explored, and only a limited number of transcription factors from solventogenic clostridia have been identified and subjected to functional analysis (4); in addition, other types of regulatory molecules and modes (e.g., posttranscriptional and posttranslational modes) remain largely unexplored. The lack of knowledge regarding these aspects will inevitably increase the difficulty in identifying new targets for strain improvement.
Small RNAs (sRNAs), normally dozens to hundreds of nucleotides (nt) in length (5), are crucial regulatory molecules in organisms (6, 7). sRNAs normally exert their regulatory functions by binding to target mRNAs, thereby affecting their transcription and stability as well as the following translation (8). Based on these characteristics, sRNAs have been increasingly regarded as useful tools for metabolic engineering and synthetic biology (9–11), either by exploring native sRNAs or designing synthetic sRNAs (12, 13). Specific to solventogenic clostridia, despite the increasing interests in the function of small RNAs in these anaerobic bacteria, they remain largely unexplored. To date, a few sRNAs have been identified and found to participate in the regulation of acid and alcohol tolerance in Clostridium acetobutylicum (14, 15). A newly reported regulator of SolB in Clostridium acetobutylicum was found to specifically regulate the expression of the genes in the sol locus, leading to a solvent-deficient phenotype after overexpression (16). Gene knockdown of hydA using a constructed sRNA was also performed in Clostridium pasteurianum, achieving higher intracellular levels of reducing cofactors and butanol titer (17). The plasmid-based expression of sRNA molecules also was successfully used for RNA repression in Clostridium cellulolyticum (18). Notably, a comprehensive list of sRNAs in 21 clostridial genomes (including two industrial Clostridium species, C. acetobutylicum and Clostridium beijerinckii) has been computationally predicted, revealing a large number of sRNAs in the genus Clostridium (19). This work, despite not focusing on functional analysis, strongly supports a continued investigation of the important roles of sRNAs in industrial clostridia.
Here, we report the discovery of a novel sRNA (sr8384) in C. acetobutylicum, a representative species of industrial solventogenic clostridia, based on phenotypic screening of a previously established transposon-based random mutant library (20). The sRNA sr8384 was not identified in the previous systematic screening of the intergenic regions of C. acetobutylicum via computational analysis (19). A series of genetic and biochemical analyses were carried out for a detailed functional analysis of sr8384, revealing its pleiotropic regulator role that determines cell growth in C. acetobutylicum. The manipulation of sr8384 could effectively promote cell growth, demonstrating the potential of this sRNA in genetic improvement. Furthermore, we also identified a functional sr8384 homolog in C. beijerinckii, another important Clostridium species that is widely used in the fermentation of lignocellulose hydrolysates, indicating the important functions and broad roles of sr8384-like sRNAs in solventogenic clostridia.
RESULTS
Phenotypic screens reveal a transposon mutant with greatly changed solvent production.In a previous study, we established a mariner-based transposon system in C. acetobutylicum ATCC 824, which generated a mutant library (more than 30,000 mutants) with high randomness (20). As a continuation of that work, we recently used this library to screen for mutants with phenotypic changes in crucial traits, such as growth and solvent synthesis. According to the process shown in Fig. 1A, more than 600 mutants were tested, and we obtained a transposon mutant (Tn mutant) that exhibited greatly impaired solvent formation during fermentation using glucose as the carbon source. This mutant could produce only 6.97 g/liter of total solvents (acetone, butanol, and ethanol) after 96 h of fermentation (Fig. 1B), which is far less than the level (21.04 g/liter) produced by the wild-type (WT) strain, indicating the presence of a transposon insertion at a crucial chromosomal position in the Tn mutant. By sequencing the reverse PCR product of the Tn mutant, we found that this mutant contained a transposon insertion in a 198-bp gene of unknown function (CAC2384) (Fig. 1C).
Identification and characterization of a C. acetobutylicum mutant with significant changes in growth and solvent production. (A) Isolation of transposon mutant with obviously altered ability to form ABE solvents. More than 600 colonies on agar plates were separately picked and inoculated into 5 ml liquid CGM medium for inoculum preparation. The cells reaching exponential growth phase (OD600 of ≈0.8) were transferred into 30 ml P2 medium (5% inoculation amount) for fermentation. The cultures after 96 h of fermentation were used for solvent assays. (B) Comparison of solvent production of Tn mutant and wild-type strain. Data are means ± standard deviations calculated from triplicate independent experiments (***, P < 0.001 by t test). (C) Transposon insertion site (inverted red triangle) on the chromosome of Tn mutant, located between the +81 and +82 sites of the open reading frame. (D) Verification of the intron insertion in the CAC2384 gene by PCR analysis. The 1.5- and 0.5-kb bands represent the PCR-amplified fragment containing the intron-inserted and original CAC2384 genes, respectively. (E) Growth and solvent formation of the Cac-2384m and wild-type strain. 824, the wild-type C. acetobutylicum ATCC 824 strain. Cac-2384m, the mutated C. acetobutylicum ATCC 824 strain with cac2384 disruption. Tn mutant, the mutated C. acetobutylicum ATCC 824 strain with a transposon insertion in the CAC2384 gene. Data are means ± standard deviations calculated from triplicate independent experiments. To determine the statistical significance of the data between the wild-type strain (824) and the mutant strains (Tn mutant and Cac-2384m), one-way analysis of variance (ANOVA) was performed, followed by Dunnett’s test (***, P < 0.001; **, P < 0.01; *, P < 0.05).
Characterization of the Tn mutant reveals a novel small RNA: sr8384.To verify whether the above-mentioned phenotypic changes of the Tn mutant were due to CAC2384 inactivation, we used the group II intron-based gene inactivation method (21) to disrupt CAC2384 in wild-type C. acetobutylicum (Fig. 1D), and the mutant obtained (named Cac-2384m) was used for phenotypic investigation. Additionally, Southern blot analysis was performed to verify that the intron was incorporated only once into the genome of Cac-2384m with no other nonspecific insertions (see Fig. S1 in the supplemental material). As expected, the Cac-2384m strain exhibited greatly impaired growth compared to that of the wild-type strain, resulting in significantly decreased solvent production (acetone, butanol, and ethanol) (Fig. 1E), which was consistent with the phenotypic changes of the Tn mutant strain (Fig. 1E).
However, when genetic complementation was performed by separately introducing four plasmids (the plasmid pP2384-2384, which expressed CAC2384 under the control of its native promoter, P2384; the plasmid pP2384, which harbored only the promoter P2384; the plasmid pPthl-2384, which expressed CAC2384 under the control of a constitutive promoter, Pthl; and the plasmid pPthl, which harbored only the promoter Pthl) back into the Cac-2384m strain (Fig. 2A), a surprising but interesting result was obtained: both the pP2384-2384 and pP2384 plasmids could complement the deficiency of the Cac-2384m mutant in solvent formation, whereas both the pPthl-2384 and pPthl plasmids failed to do so (Fig. 2B). To further confirm this finding, we performed genetic complementation experiments through chromosomal insertion of the target DNA sequence by using ClosTron technology (22, 23). In brief, the above-mentioned four DNA fragments were separately integrated into an intron sequence and then inserted into the chromosome to see if the impaired phenotypes of the Cac-2384m mutant could be restored (Fig. S2A, B, and C). As shown in Fig. S2D, only the P2384-2384 and P2384 modules could recover the solvent-forming ability of the Cac-2384m mutant, which was consistent with the findings shown in Fig. 2B. Taken together, these data strongly suggest that the phenotypic changes in the Cac-2384m mutant can be recovered by the independent expression of the upstream noncoding sequence (P2384) of CAC2384. In other words, there may be some crucial DNA elements in the upstream region of CAC2384, although it is unclear why the insertional disruption of CAC2384 influenced this noncoding region.
Uncovering the functional sequence within the promoter region of CAC2384. (A and B) Genetic complementation of the Cac-2384m mutant indicates an unknown crucial molecule related to the phenotypic changes. The four plasmids (pPthl, pPthl-2384, pP2384, and pP2384-2384) constructed for genetic complementation of Cac-2384m are shown. The native promoter P2384 covered the whole intergenic region (202 bp) between the CAC2384 and CAC2383 genes. Solvent production by the Cac-2384m mutants containing different complementary plasmids and the wild-type strain carrying an empty plasmid (control) were compared. (C) Truncation of the P2384 fragment. (D) Genetic complementation of the Cac-2384m mutant with the above-mentioned truncated fragments. Data are means ± standard deviations calculated from triplicate independent experiments. The statistical significance of the data between the Cac-2384m strain and its derivative strains was assessed by one-way analysis of variance (ANOVA), followed by Dunnett’s test (***, P < 0.001).
A detailed functional analysis of the P2384 sequence was performed next to explore the above-described hypothesis. The whole sequence (202 bp) of P2384 was gradually truncated, yielding 10 truncated fragments, i.e., P2384 minus 10, 20, 30, 40, 50, 60, 100, 120, 140, or 150 bp (Fig. 2C). These DNA fragments were integrated into the expression plasmid and then introduced into the Cac-2384m mutant for genetic complementation analysis. The results showed that all the truncated fragments retained complementation functions except the shortest fragment (with a 150-bp deletion) (Fig. 2D), suggesting that the potential DNA element mentioned above is located within the 62-bp P2384-140 sequence. Given the very low chance that this short 62-bp sequence encodes a functional protein, we reasoned that it may encode an sRNA.
To explore this possibility, the following experiments were performed in sequence: (i) a two-step reverse transcription-PCR (RT-PCR) analysis for determining the transcriptional direction of the P2384-140 sequence and a RACE (5ʹ and 3ʹ rapid amplification of cDNA ends) experiment aiming to determine the actual transcript length of P2384-140; (ii) Northern blotting to verify the role of this transcript (an sRNA or not). As shown in Fig. S3A, in the two-step RT-PCR analysis, theoretically, only the PCR amplification using P-2 as the initial primer will give the desired PCR product (case II). As expected, a 62-bp PCR band was detected from the total RNA of the wild-type C. acetobutylicum when using the primer P-2 to initiate the PCR (Fig. S3B), indicating the native transcriptional direction of the P2384-140 sequence. On this basis, the RACE experiment was carried out (Fig. 3A). The result further revealed a 94-nucleotide (nt) transcript that is located between the CAC2383 and CAC2384 genes and partially overlaps the open reading frame (ORF) of CAC2383 (Fig. 3A). This 94-nt short transcript has a stable and typical secondary structure (Fig. 3B), no Shine-Dalgarno (SD) sequence, and no stop codon. The Northern blotting experiment using a single-stranded oligonucleotide probe targeting this 94-nt transcript was performed next to further confirm the existence of this sRNA. A DNA fragment covering the ORF of CAC2383 and CAC2384 as well as their intergenic region in the chromosome was PCR amplified and then integrated into a replicative plasmid for expression (psRNA) (Fig. 3C), aimed at enriching the in vivo level of this potential sRNA. Encouragingly, a desired, approximately 94-nt hybridization signal was detected for the RNAs (without and with the Terminator 5′-phosphate-dependent nuclease pretreatment) from the C. acetobutylicum strain containing the plasmid psRNA. This result confirms the existence of this sRNA; moreover, it is not a processed RNA from a larger transcript. Simultaneously, no signal was observed from the control strain (pControl) (Fig. 3C), indicating the in vivo low level of this sRNA in the wild-type C. acetobutylicum strain.
Sequence and structure of sr8384. (A) DNA sequence of CAC2383, CAC2384, and their intergenic region. The sr8384 sequence determined by the RACE reactions is shown as an orange box. The Sanger sequencing chromatographs of the 5′ and 3′ RACE are also provided. TSS, transcription start site. The start code of the CAC2384 gene is underlined with blue. (B) The secondary structure of sr8384. (C) Northern blot analysis to identify the predicted sr8384. psRNA, a DNA fragment covering the ORFs of CAC2383 and CAC2384 as well as their intergenic region in the chromosome was PCR amplified and then integrated into a replicative plasmid for expression, aimed at enriching the in vivo level of this potential sRNA. pControl, the control plasmid without the sRNA contained. Two terminators were located at two ends to avoid potential expression running through from other genes on the plasmid. M, marker. TEX, the Terminator 5′-phosphate-dependent nuclease.
In summary, the above-described results suggest the presence of a 94-bp sRNA-coding sequence in the intergenic region between CAC2383 and CAC2384. Notably, this sRNA, named sr8384 here, is not present in the list of sRNAs that were previously identified in Clostridium organisms via computational analysis (19), indicating that sr8384 has some novel genetic features.
sr8384 is crucial for the growth of C. acetobutylicum.Having discovered sr8384, it remained unknown whether this sRNA is a crucial molecule in C. acetobutylicum. Therefore, the sr8384 transcript was disrupted for phenotypic examination using small regulatory RNA-based gene knockdown technology (24, 25). As shown in Fig. 4A, a vector containing a 24-nt target-binding (TB) sequence that targets the middle region of sr8384 was constructed and introduced into the wild-type C. acetobutylicum strain, yielding the mutant strain 824(8384r). The results of quantitative RT-PCR (qRT-PCR) analysis showed that the 824(8384r) strain exhibited a greater than 50% decrease in the sr8384 transcript level compared to that in the 824c strain (the control strain containing the plasmid lacking the 24-nt target-binding sequence) (Fig. 4B), demonstrating effective in vivo knockdown of the sr8384 transcript. Subsequently, in batch fermentation, compared to 824c, 824(8384r) exhibited greatly impaired growth, resulting in significantly decreased production of solvents (Fig. 4C), in which the impact on butanol is especially obvious (10.59 g/liter versus 14.54 g/liter). These data suggest that sr8384 determines cell growth in C. acetobutylicum. In addition, we investigated the effect of three different sugars (glucose, fructose, and arabinose) on the in vivo expression of sr8384; however, no significantly changed sr8384 expression (fold change of <2.0) was observed at any one time point (Fig. 4D), indicating that the expression of this sRNA is not carbon source dependent.
Influence of in vivo sr8384 level on cellular performance. (A) The construct of small regulatory RNA-based gene knockdown. TB, target binding. The 24-nt TB sequence is responsible for targeting against sr8384. (B) Fold change of sr8384 level in C. acetobutylicum after introducing antisense construct. qRT-PCR analysis was performed to measure the changes of in vivo sr8384 levels. The samples of the 824c and 824(8384r) strains for qRT-PCR analysis were taken at 24 h of fermentation. Data are means ± standard deviations from two independent experiments (**, P < 0.01 by t test). (C) Phenotypic effects of repressed sr8384 expression. 824c, the wild-type strain carrying the pIMP1-AS-con plasmid (lacking the 24-nt TB sequence). 824(8384r), the strain carrying the pIMP1-AS-sr8384 plasmid. Data are means ± standard deviations from three independent experiments (***, P < 0.001; **, P < 0.01; *, P < 0.05; all by t test). (D) Comparison of the in vivo sr8384 levels using different sugars (glucose, fructose, and arabinose) as the carbon source. Data are means ± standard deviations from two independent experiments. The statistical significances of the data between fermenting glucose and the other two sugars (fructose and arabinose) were assessed by one-way analysis of variance (ANOVA), followed by Dunnett’s test (***, P < 0.001; **, P < 0.01; *, P < 0.05). (E) Phenotypic effects derived from increased sr8384 levels. A 500-ml working volume was used to perform the fermentation. The red arrows reflect the sampling time points (23 h and 42 h) for microarray assays. Control, the wild-type C. acetobutylicum strain that carries an empty plasmid. CAC-smR, the C. acetobutylicum strain that carries the sr8384-overexpressing plasmid pIMP1-Pthl-sr8384. Data are means ± standard deviations from two independent experiments (**, P < 0.01; *, P < 0.05; both by t test).
Since sr8384 determines cell growth, a derived question is whether increasing the in vivo level of this sRNA could promote the growth of C. acetobutylicum. Therefore, we constructed an expression vector in which the coding sequence of sr8384 was overexpressed under the control of a strong constitutive promoter, Pthl. The plasmid then was introduced into wild-type C. acetobutylicum, yielding the strain CAC-smR. Encouragingly, compared to the control strain, the CAC-smR strain exhibited greatly enhanced growth rate and biomass, leading to the improved production of total solvents (acetone, butanol, and ethanol) (Fig. 4E). Overall, these findings showcase not only the importance of sr8384 in controlling cell growth but also the potential value of this sRNA as a molecular tool in C. acetobutylicum.
Global influence of sr8384 in C. acetobutylicum.Because sr8384 overexpression led to the positive phenotypic changes of C. acetobutylicum (Fig. 4E), we used a comparative transcriptomics approach to search for genes affected by sr8384. The RNA samples for microarray assays were isolated from the sr8384-overexpressing strain CAC-smR and the control strain at two time points, namely, 23 h and 42 h, reflecting acidogenic and solventogenic stages, respectively (Fig. 4E). The results showed that 679 and 380 genes exhibited significantly altered transcriptional levels (fold change of ≥2.0) at 23 h and 42 h (Tables S1 and S2), respectively, of which 172 genes were detected at both time points (Fig. S4A). These differentially expressed genes could be roughly grouped into 16 subsets (Fig. S4B), including some subsets of genes associated with important physiological and metabolic processes. These results indicate a crucial and global influence of sr8384 in C. acetobutylicum. We selected 10 genes that exhibited different degrees of transcriptional repression or activation in the microarray assay after sr8384 overexpression for expression level validation using qRT-PCR. The qRT-PCR results were consistent with the data from the microarray analysis (Fig. S5), indicating that the microarray data were of high quality.
The next task was to experimentally determine whether sr8384 plays a pleiotropic regulator role in C. acetobutylicum. To this end, we first used the online tool IntaRNA (26) to predict putative target sequences based on their potential interaction energy with sr8384. The top 100 sequences (with interaction energies of ≤–18.3083 kcal/mol) within the predicted results were chosen for further investigation. The genes associated with these 100 sequences (located in the promoter or coding region) that exhibited ≥2-fold transcriptional changes (microarray assay data) after sr8384 overexpression were selected, resulting in 26 candidates, and their associated genes were used for further detailed investigations (Fig. 5A). Most of these 26 target sequences spanned both the promoter and coding regions of their corresponding genes. These 26 candidates next were used for RNA hybridization analysis to examine whether these genes interact with sr8384. The results showed that, of the 26 candidates, 15 exhibited distinct binding activity with sr8384 (Fig. 5B), whereas no obvious binding was observed for the remaining candidates (Fig. S6). Among these 15 sequences of direct targets of sr8384, nine were associated with genes with annotated functions (Table S3). The locations of the base-pairing regions of sRNA to these 15 direct targets were also predicted (Fig. S7), showing that these base-pairing regions tend to cluster in the single-stranded region (containing a small hairpin) of sr8384. We next chose three direct targets of sr8384 for a further analysis. Mutations within the sr8384-targeting regions of these three candidates were performed (Fig. 5C), which, as expected, disrupted the in vitro sr8384 binding to these targets (Fig. 5C). Furthermore, the in vivo activity analysis using a reporter gene (lacZ) showed the increased expression of the target genes after these mutations (Fig. 5D). Therefore, these data, along with the microarray analysis results, confirm the pleiotropic regulator role of sr8384 in C. acetobutylicum.
Identification of the direct targets of sr8384 in C. acetobutylicum. (A) The 26 picked genes that are potentially controlled by sr8384 and simultaneously showed over 2-fold expressional changes after sr8384 overexpression. (B) Fifteen identified target sequences that directly bind with sr8384. A concentration of 1 ng/μl of the Cy5-labeled sr8384 (64-nt length) was incubated with 5 ng/μl of the target mRNA (300-nt length) in 20 μl TMN buffer at 37°C for 15 min, aiming to form an RNA-RNA complex for electrophoretic mobility shift assay analysis. These 15 genes, directly regulated by sr8384, are labeled with an asterisk in panel A. (C and D) The in vitro and in vivo data showing that the mutations (mu) within the pairing regions disrupt sr8384 binding to the targets. The predicted base-pairing region of the sRNA (sr8385) to the different targets is highlighted in red. The mutated bases in the lacZ fusion are also shown in red. The reporter gene lacZ was used in the in vivo analysis. The native promoters of the cac0602, cac2617, and cac1485 genes were separately used for expressing lacZ (***, P < 0.001; **, P < 0.01; both by t test).
Manipulation of sr8384 and its homolog leads to improved growth of Clostridium species.To further explore the physiological role of sr8384, we used the native promoter Pthl and its derived artificial weaker promoter, P200-1, for sr8384 overexpression to determine whether this sRNA has a dosage-dependent effect on cell growth (Fig. 6A). The promoter P200-1 has the same sequence as Pthl except for the mutations at the 5′ end of TTG (−35 region), the 3′ end of TATAAT (−10 region), and the intervening spacer between TTG and TATAAT (27). Interestingly, compared to the promoter Pthl, better cell growth was observed with the weaker promoter, P200-1, resulting in higher titers of the solvents (Fig. 6B). These results indicate that the in vivo sr8384 level is associated with the growth of C. acetobutylicum.
Manipulation of sr8384 and its homolog can improve the growth and solvent synthesis of Clostridium. (A) Two promoters with gradually increased activities used for sr8384 overexpression. Data are means ± standard deviations from two independent experiments (***, P < 0.001 by t test). (B) Improved growth and total solvents by sr8384 overexpression using the two promoters shown in panel A. Control, the control strain containing the empty plasmids; thl-sR, the strain with sr8384 overexpression under the control of the promoter Pthl; 200-1-sR, the strain with sr8384 overexpression under the control of the promoter P200-1. Data are means ± standard deviations from three independent experiments. The statistical significance of the phenotypic differences between the control strain and the other two mutants (thl-sR and 200-1-sR) were assessed by one-way analysis of variance (ANOVA), followed by Dunnett’s test (***, P < 0.001; **, P < 0.01; *, P < 0.05). (C) The coding sequence of sr8384 homologs in clostridia. (D) Improved cellular performance of C. beijerinckii by sr8889 overexpression (the sr8384 homolog in C. beijerinckii). Data are means ± standard deviations from three independent experiments (**, P < 0.01; *, P < 0.05; both by t test).
To date, numerous sRNAs have been found to be conserved in several genera (28). Thus, we sought to find sr8384 homologs in other clostridial genome sequences available in NCBI. BLASTN analysis showed that no putative sr8384 homologs were present in any other Clostridium species except two C. acetobutylicum strains (EA2018 and DSM1731) (Fig. 6C). However, when we scanned the genome of C. beijerinckii NCIMB 8052, another major solventogenic Clostridium species, a potential homologous sequence that shares 54.1% identity with the sr8384 sequence was found in the intergenic region between the Cbei1789 and Cbei1788 genes in C. beijerinckii (two homologs in C. beijerinckii corresponding to CAC2383 and CAC2385, respectively, in C. acetobutylicum) (Fig. 6C). Here, this sequence was named sr8889.
We next investigated whether sr8889 also affects the growth of C. beijerinckii. As expected, the overexpression of this sequence indeed resulted in an increased growth rate of C. beijerinckii, leading to improved production of solvent (Fig. 6D). These results showed that sr8889 also plays an important role in C. beijerinckii, although the improvement of growth and solvent production from sr8889 overexpression was not as significant as those caused by sr8384 overexpression in C. acetobutylicum. Such a different effect from overexpressing these two sRNAs may be attributed to the different genomic composition and physiological metabolisms between C. acetobutylicum and C. beijerinckii (4).
DISCUSSION
Discovery and functional analysis of sRNAs have been performed in some representative industrial microorganisms, such as Escherichia coli (29), Saccharomyces cerevisiae (30, 31), and Bacillus subtilis (6, 32). Additionally, native and artificially synthetic sRNAs have been used for improving bacterial physiology and metabolism (12, 13). However, sRNAs, as well as their potential values in metabolic engineering, remain minimally explored in solventogenic Clostridium species. In this study, we discovered the atypical sRNA sr8384 and its utility in the control of growth in C. acetobutylicum, a model organism for clostridia. To the best of our knowledge, such a crucial sRNA determining cell growth in clostridia has not been previously reported.
To date, only a very limited number of sRNAs have been shown to be associated with certain functions in clostridia (14–16). The sRNA sr8384, identified here, is a global rather than specific regulatory molecule in C. acetobutylicum. Given that sr8384 was not predicted or detected in the previous screenings for sRNAs in clostridia based on comparative genomics and transcriptome sequencing (19, 33, 34), this sRNA is likely atypical in sequence or is very poorly expressed in C. acetobutylicum. Moreover, sr8384 exhibited a dose-dependent effect in the regulation of phenotypes of C. acetobutylicum, i.e., negative and positive phenotypic changes were observed upon repression and overexpression, respectively, of this sRNA (Fig. 4C and E), indicating that sr8384 has an application as a target for strain improvement.
The greatly enhanced growth and the resulting improved solvent production upon sr8384 overexpression are two important phenotypic alterations in C. acetobutylicum. Given the global influence of sr8384, these positive effects could be attributed to the direct or indirect regulation of sr8384 on multiple effective targets. This finding is consistent with the previous report that native sRNAs often target multiple mRNAs (35). Such a pleiotropic regulatory role of sr8384 may enable it to be a potential tool for coordinating the expression of multiple target genes.
Broadly, noncoding sRNAs in bacteria consist of cis-acting and trans-acting sRNAs (36). cis-Acting sRNAs are transcribed on the DNA strand complementary to one from which their target mRNAs are transcribed and, thus, antisense, at least partially, to their targets, resulting in expressional repression (37). sr8384 in this study also has the cis-acting sRNA-like characteristics (Fig. 3A); however, no obvious expressional repression was observed for its target gene, CAC2383, after overexpressing sr8384 (shown in the microarray data), which is probably due to their limited base-pairing region (27 bp) (Fig. 3A). We also analyzed the upstream noncoding sequence of the CAC2383 gene using the Riboswitch finder software (http://www.biozentrum.uni-wuerzburg.de/bioinformatik/Riboswitch/) and excluded the possibility that this region contains a riboswitch (a type of cis-acting noncoding RNA). Taking these findings together, sr8384 is unlikely to be a cis-acting sRNA and instead could be a trans-acting sRNA considering its global influence in C. acetobutylicum.
Here, although the comparative transcriptomic analysis for the sr8384-overexpressing strain in combination with the application of the online tool IntaRNA has proven useful for the prediction of sRNA targets in C. acetobutylicum, it cannot be ruled out that some additional targets may have been missed. For example, the expression of some targets might change only when sr8384 is absent, rather than overexpressed, or some targets might not be expressed in the presence of d-glucose due to the CCR (carbon catabolite repression) effect. Therefore, a more comprehensive understanding of the target network of sr8384 in C. acetobutylicum is required. For this purpose, some new techniques, such as MS2 affinity purification and sequencing (MAPS) and cross-linking and immunoprecipitation (CLIP)-based methods, which have been used to characterize sRNA target networks in many typical bacteria (12), can be considered.
Our data here identify sr8384 as a pleiotropic regulator in C. acetobutylicum. With this characteristic, sr8384, to the best of our knowledge, is the first identified sRNA involved in regulating various physiological and metabolic processes in the industrially important Clostridium species. Notably, improved cell growth was achieved via the modulation of the expression of sr8384. Given the discovery of a functional sr8384 homolog in C. beijerinckii, this type of functional sRNA may be widespread in clostridia. In summary, this work provides new insight into the role of sRNAs in clostridia and offers new opportunities for engineering these anaerobes.
MATERIALS AND METHODS
Media and cultivation conditions.Luria-Bertani (LB) medium, supplemented with ampicillin (100 μg/ml) and spectinomycin (50 μg/ml) when needed, was used to cultivate E. coli. C. acetobutylicum was first grown anaerobically (Thermo Forma, Inc., Waltham, MA) in CGM medium (38) for inoculum preparation. Upon reaching the exponential growth phase (optical density at 600 nm [OD600] of 0.8 to 1.0), the cells (5% inoculation amount) were transferred into P2 medium (39) for fermentation. Erythromycin (10 μg/ml) and thiamphenicol (8 μg/ml) were added to the P2 medium when needed. Samples for assays were removed at different time points and then stored at –20°C.
Bacterial strains and plasmid construction.The primers used in this study are listed in Table 1. The strains and plasmids used in this work are listed in Table 2. Top10 cells were used for gene cloning. The plasmids were first methylated by E. coli ER2275 and then electroporated into C. acetobutylicum.
Primers used in this study
Strains and plasmids used in this study
The CAC2384 gene was disrupted using the group II intron-based Targetron system. In brief, a 350-bp DNA fragment was amplified by PCR with the following primers: the EBS universal primer, CAC2384-174,175s-IBS, CAC2384-174,175s-EBS1d, and CAC2384-174,175s-EBS2. Amplification was performed according to the protocol of the Targetron gene knockout system kit (Sigma-Aldrich, St. Louis, MO, USA). After digestion with XhoI and BsrGI, this 350-bp DNA fragment was cloned into the plasmid pWJ1 (40), yielding the plasmid pWJ1-CAC2384.
The pPthl plasmid, derived from pIMP1 (41), was constructed as previously reported (42). The P2384 fragment was amplified by PCR using the primers P2384-for and P2384-rev with the genomic DNA of C. acetobutylicum as the template. After digestion with PstI and BamHI, the P2384 fragment was cloned into the plasmid pPthl, yielding the plasmid pP2384. Similarly, the CAC2384 fragment was amplified by PCR using the primers CAC2384-for and CAC2384-rev. The CAC2384 fragment then was digested with SalI and BamHI and ligated to the plasmids pPthl and pP2384, yielding the plasmids pPthl-2384 and pP2384-2384, respectively.
For the overexpression of sr8384 and sr8889 in C. acetobutylicum and C. beijerinckii, respectively, the sr8384 and sr8889 sequences were amplified by PCR, digested with SalI and BamHI, and then ligated with the plasmid pPthl, which had been digested with the same restriction enzymes, yielding the plasmids pIMP1-Pthl-sr8384 and pIMP1-Pthl-sr8889, respectively.
The plasmid used for small regulatory RNA-based knockdown of sr8384 was constructed according to a previously reported method (24, 25). In brief, the fragment Pthl-AS-sr8384-MicC, which contained a 24-nt target-binding sequence complementary to sr8384 and the MicC sRNA scaffold, was first amplified by PCR using the plasmid pPthl as the template with the primers AS-Pthl-s and AS-sr8384-MicC-a. A fragment containing both the promoter Pptb and HfqEC (Pptb-HfqEC) then was obtained by overlap PCR. Finally, the fragments Pthl-AS-sr8384-MicC and Pptb-HfqEC were assembled by overlap PCR, yielding the large fragment Pthl-AS-sr8384-MicC-Pptb-HfqEC. After digestion with PstI and EcoRI, the fragment Pthl-AS-sr8384-MicC-Pptb-HfqEC was ligated with the plasmid pPthl, which had been digested with the same restriction enzymes, yielding the plasmid pIMP1-AS-sr8384. The method for the construction of the plasmids used for the knockdown of the other target genes in this study was the same as that used for sr8384, changing only the 24-nt target-binding sequence.
Gene disruption and functional complementation using a group II intron (Targetron).Gene disruption in C. acetobutylicum was achieved through chromosomal insertion of a group II intron by using the Targetron plasmid (21). The primers used for retargeting the RNA portion of the intron to target genes are listed in Table 1. In detail, the fragments of Pthl, Pthl-CAC2384, P2384, P2384-CAC2384, and the two separate parts of the initial intron sequence were PCR amplified by using the primer pairs listed in Table 1. The whole functional sequences, which contained both the intron and the insertion sequence, then were acquired by overlap PCR with Intron-for/Intron-rev supplied as primers and inserted into the chromosome by using the Targetron strategy.
Analytical methods.The density of the culture (A600) after cell growth was tested using a spectrophotometer (DU730; Beckman Coulter, Placentia, CA, USA). The concentrations of the solvents (acetone, acetic acid, butyric acid, butanol, and ethanol) were determined using gas chromatography (7890A; Agilent, Wilmington, DE, USA). Isobutyl alcohol and isobutyric acid were used as the internal standards for solvent quantification.
Identification of transposon insertions in the chromosome by reverse transcription-PCR.The transposon mutant library of C. acetobutylicum that was constructed according to a previously reported protocol (20) was used to identify mutants that exhibited greatly changed solvent production. The selected mutant strain was used for reverse transcription-PCR analysis to identify the transposon insertion site in the chromosome as reported previously (20).
Two-step RT-PCR analysis.The two-step RT-PCR analysis was performed according to the previous report (43) to determine the transcriptional direction of the potential small RNA sr8384. In brief, the total RNA of C. acetobutylicum was isolated by TRIzol extraction. A pair of primers that match the sr8384 sequence was separately added into the total RNA to synthesize the first-strand cDNA by using the PrimeScript RT reagent kit (RR047A; TaKaRa). Two primers then were simultaneously used for the second round of PCR amplification using the above-described first-strand cDNA as the template. Finally, the PCR products were separated on 1.5% agarose gels.
Southern blot analysis.Southern blotting was performed using a digoxigenin (DIG) high prime DNA labeling and detection starter kit I (Roche Diagnostics GmbH, Mannheim, Germany), as instructed by the manufacturer. Briefly, 10 μg of genomic DNA was digested with HindIII for about 14 h, separated on a 1.0% agarose gel, and then transferred to a charged nylon membrane. The digoxigenin-labeled DNA probe then was used for Southern hybridization.
Northern blot analysis.Total RNA first was treated with DNase I to remove DNA contamination and then treated with the Terminator 5′-phosphate-dependent exonuclease (TEX) (catalog no. TER51020; Epicentre) that degrades processed RNAs (e.g., rRNA). The TEX-treated RNA was loaded and electrophoretically resolved on a 7% denaturing polyacrylamide gel containing 7 M urea. Nucleic acids then were transferred to an Immobilon-NY+ membrane (INYC00010; Merck KGaA, Darmstadt, Germany) and immobilized by UV-cross-linking. A 94-nt digoxigenin-labeled probe (20 pmol) complementary to the sRNA sequence was synthesized and used to detect the presence of the sRNA. The prehybridization (1 to 2 h) and hybridization (16 h) steps were performed at 37°C using NorthernMax prehybridization and hybridization buffers (AM8677; LifeTech, Thermo Fisher Scientific, Inc., Carlsbad, CA, USA). The membrane was washed twice with 4× SSC (1× SSC is 0.15 M NaCl and 0.015 mM sodium citrate) buffer for 15 min and then washed with 2× SSC buffer (2× SSC, 0.1% SDS) for an additional 15 min. Finally, immunological detection was performed according to the protocol for the DIG northern starter kit (no. 12039672910; Roche, Mannheim, Germany).
5ʹ and 3ʹ RACE analysis.The 5′ and 3′ RACE analyses were performed according to the protocol for the SMARTer RACE 5′/3′ kit (634858 and 634859; TaKaRa Bio USA, Inc.). The whole upstream noncoding region (202 bp) of the CAC2384 gene, together with the ORF of CAC2383 and CAC2384, was PCR amplified from the chromosome of C. acetobutylicum and then integrated into a multicopy plasmid to enrich the transcripts for RACE. For the 3′ RACE analysis, a poly(A) tail was first added to the 3′ end of the RNA template using E. coli poly(A) polymerase (M0276S; New England BioLabs, Beverly, MA, USA).
Microarray analysis.CAC-smR and the control strain were grown in P2 medium (500 ml) with erythromycin supplementation (10 μg/ml). d-Glucose (80 g/liter) was used as the sole carbon source. Samples for microarray analysis were taken at 23 h (acidogenic phase) and 42 h (acid-solvent transition phase). After centrifugation at 4°C for 10 min, the cell pellets were frozen immediately in liquid nitrogen. The cells then were ground into a powder and dissolved in TRIzol reagent (Invitrogen, Carlsbad, CA). Microarray analysis was performed using Agilent custom 60-mer oligonucleotide microarrays (Shanghai Biochip Co., Ltd., Shanghai, China) as described previously (44, 45). Genes that exhibited greater than 2-fold changes in expression in CAC-smR compared to the expression in the control strain were considered differentially expressed.
qRT-PCR.For qRT-PCR analyses, RNA was isolated by TRIzol extraction as described previously (44). RNA was reverse transcribed to cDNA using the PrimeScript RT reagent kit (RR047A; TaKaRa). qRT-PCR then was carried out in a MyiQ2 two-color real-time PCR detection system (Bio-Rad) with the following conditions: 95°C for 2 min, followed by 40 cycles of 95°C for 15 s, 55°C for 20 s, and 72°C for 20 s. The CAC2679 gene (encoding pullulanase) was used as an internal control (44). The primers used for qRT-PCR analysis are listed in Table 1.
Secondary structure analysis and target prediction of sRNA in C. acetobutylicum.The secondary structure of sr8384 was generated using PseudoViewer (46). IntaRNA online software (26) was used to screen for the putative sRNA targets across the genome of C. acetobutylicum (target NCBI RefSeq identifiers NC_003030 and NC_001988), in which the 94-nt whole sequence of sr8384 was entered as the query ncRNA.
In vitro transcription and in vivo β-galactosidase (LacZ) assay.RNAs were synthesized in vitro from PCR-generated DNA fragments using the MEGAscript T7 high-yield transcription kit (AM1334; Thermo Fisher Scientific Baltics UAB, Vilnius, Lithuania) and then purified using the MEGAclear kit (AM1908; Thermo Fisher Scientific Baltics UAB, Vilnius, Lithuania). The primers used for in vitro transcription are listed in Table 1.
The pretreatment of C. acetobutylicum cells for LacZ activity assay was the same as that previously reported (47). The detailed LacZ activity assay was performed according to a protocol reported previously (48).
Analysis of RNA-RNA complex formation.RNA-RNA complex formation analysis was performed as previously reported (32). In brief, the 64-nt sr8384 was synthesized and then labeled with Cy5 at the 5′ end. In the RNA hybridization experiment, the Cy5-labeled sr8384 and target RNA were first resolved in TMN buffer (20 mM Tris-acetate [pH 7.5], 2 mM MgCl2, 100 mM NaCl) and then incubated at 95°C for 2 min. For proper RNA folding, both the Cy5-labeled sr8384 and target mRNA then were incubated on ice for 2 min, followed by 30 min at 37°C, and then 1 ng/μl of the Cy5-labeled sr8384 was incubated with 5 ng/μl of target mRNA in 20 μl TMN buffer (containing 0.1 μg/μl tRNA) (R8508; Sigma-Aldrich Trading, Shanghai, China) at 37°C for 15 min. The RNA-RNA complex formation reaction was stopped by adding stop solution (1× TMN, 50% glycerol, 0.5% bromophenol blue and xylene cyanol). Finally, the mixture was loaded onto a 6% native polyacrylamide gel and resolved by electrophoresis (120 V, 4°C, 1 h). The gel was scanned using an FLA-9000 phosphorimager (FujiFilm, Japan) for visualization. The quantitative image analysis of the gel was performed according to a method reported previously (49).
Small regulatory RNA-based gene knockdown.A synthetic small regulatory RNA-based system, containing a target-binding sequence, MicC sRNA scaffold, and an RNA chaperone, Hfq, has been proven to be effective for controlling gene expression in C. acetobutylicum (24, 25). Here, this technology was used to repress gene expression. In brief, both a 24-nt sequence that binds to the mRNA target and an E. coli MicC scaffold were integrated into a plasmid for expression, yielding a synthetic sRNA. Simultaneously, an Hfq-coding gene was also inserted into this plasmid for expression. The in vivo synthetic sRNA, together with Hfq as a cofactor that facilitates mRNA targeting, is capable of blocking the translation of the target mRNA.
ACKNOWLEDGMENT
This work was supported by the National Natural Science Foundation of China (31670043, 31630003, 31921006, and 31700030).
FOOTNOTES
- Received 19 March 2020.
- Accepted 27 April 2020.
- Accepted manuscript posted online 1 May 2020.
Supplemental material is available online only.
- Copyright © 2020 American Society for Microbiology.
