Identification and Characterization of 5' Untranslated Regions (5'UTRs) in Zymomonas mobilis as Regulatory Biological Parts

Abstract

Regulatory RNA regions within a transcript, particularly in the 5' untranslated region (5'UTR), have been shown in a variety of organisms to control the expression levels of these mRNAs in response to various metabolites or environmental conditions. Considering the unique tolerance of Zymomonas mobilis to ethanol and the growing interest in engineering microbial strains with enhanced tolerance to industrial inhibitors, we searched natural cis-regulatory regions in this microorganism using transcriptomic data and bioinformatics analysis. Potential regulatory 5'UTRs were identified and filtered based on length, gene function, relative gene counts, and conservation in other organisms. An in vivo fluorescence-based screening system was developed to confirm the responsiveness of 36 5'UTR candidates to ethanol, acetate, and xylose stresses. UTR_ZMO0347 (5'UTR of gene ZMO0347 encoding the RNA binding protein Hfq) was found to down-regulate downstream gene expression under ethanol stress. Genomic deletion of UTR_ZMO0347 led to a general decrease of hfq expression at the transcript level and increased sensitivity for observed changes in Hfq expression at the protein level. The role of UTR_ZMO0347 and other 5'UTRs gives us insight into the regulatory network of Z. mobilis in response to stress and unlocks new strategies for engineering robust industrial strains as well as for harvesting novel responsive regulatory biological parts for controllable gene expression platforms in this organism.

Keywords: 5′ UTR; 5′ rapid amplification of cDNA ends (RACE); Zymomonas mobilis; regulatory RNA; stress response.

Figure 1

Pipeline for the selection of 5′UTR candidates. Initial 5′UTR candidates were selected from transcriptomic data and then filtered for comparable expression level to adjacent mRNA, with length more than 35 bp. 5′UTR candidates that met these criteria were analyzed for homology with known RNA regulators by Rfam and for general conservation in other organisms by BLAST. From this analysis, 101 candidates were selected for experimental confirmation.

Figure 2

RT-PCR analysis to experimentally confirm 5′UTR candidates. (A) To confirm the 5′UTRs detected in transcriptomic data are actually transcribed with their downstream mRNAs, two sets of primers were designed. Primer Set A amplifies from the middle of the predicted UTR into the adjacent mRNA coding region. Primer Set B, as a control, amplifies the coding region only, indicating the amplification level of the transcript. (B) Examples of positive (confirmed) and negative (undetected) 5′UTR candidates are shown. RNA was reverse-transcribed and then subjected to PCR with Primer Set A and Primer Set B for each 5′UTR candidate. As a negative control, RNA was left without reverse transcription to indicate background levels of any residual DNA that could be left undigested from DNaseI treatment. Samples were then visualized on agarose gels, as shown in the examples. (C) For each 5′UTR confirmed by RT-PCR, 5′RACE was performed to determine the transcription start site by attaching an adapter of known sequence to the 5′ end and sequencing from the mRNA coding region upstream toward the adapter.

Figure 3

Establishment of a green fluorescent protein (GFP)-based high-throughput reporter gene screening system to characterize regulatory 5′UTR regions. Control-GFP plasmid was developed in which GFP is located under P_tet promoter control to demonstrate expression of the GFP protein in Z. mobilis (A). The theophylline synthetic riboswitch (sequence shown) was cloned in front of GFP to regulate its expression and show that a well-characterized regulatory 5′UTR could be used to see measurable differences in GFP expression upon the addition of the ligand it senses (theophylline in this case) (B). Upon induction of P_tet with tetracycline (black), the Control-GFP strain shows a fluorescence shift by flow cytometry compared to the un-induced sample (light gray) (C). The strain containing plasmid theophylline RNA stability element (RSE)-GFP was induced with 2.5 mM theophylline in DMSO (black) and compared by flow cytometry to a sample to which only DMSO (light gray) was added. In both cases, P_tet was induced with tetracycline (D). To allow screening of 5′UTRs identified in this study, the plasmid was designed with BsmBI sites flanking the PheS cassette in front of GFP. The PheS cassette was replaced with each UTR+90 bps of initial mRNA sequence by Golden Gate cloning (E).

Figure 4

Identification of 5′UTRs that regulate GFP expression under ethanol stress. UTR_ZMO0347 and UTR_ZMO1142 showed fluorescence shifts under different levels of ethanol compared to control-GFP. White gray line: RMG (normal media). Light gray line: 1% ethanol supplemented media. Dark gray line: 3% ethanol supplemented media. Black line: 5% ethanol supplemented media. Median fluorescence level of each condition is shown in a table and the fluorescence ratios relative to the RMG control are shown in the bar graphs. GFP protein levels were detected by western blotting using anti-GFP antibody. These experiments were done in triplicate and the fluorescence ratio bar graph error bars represent standard deviation from the mean.

Figure 5

Native regulation of Hfq in ethanol stress. Transcript levels of hfq were quantified by qPCR with (5%) and without (0%) ethanol supplemented to the media (A). Bars represent averages of biological triplicates calculated by the comparative delta-delta threshold cycle (ΔΔC_T) method and error is standard deviation. The Hfq protein levels were quantified by western blots of chromosomally FLAG-tagged Hfq strains with the UTR (B) and without (C). After growth in a range of ethanol concentrations, 20 ng of total protein were loaded into an SDS-PAGE gel, then transferred to a membrane, and blotted with the anti-FLAG antibody. Band intensity was quantified and is shown as averages of biological triplicates with error as standard deviation.

Figure 6

Summary of observed Hfq regulation in ethanol stress. In native conditions with the UTR of Hfq intact (ZMO0347), the transcript and protein levels increase in 5% ethanol stress. When the UTR is deleted from the genome, the protein expression increases upon a more mild ethanol stress (1%), indicating the sensitively of Hfq expression without the UTR.