The Signal Sequence of the Abundant ... - ACS Publications

Jun 22, 2016 - The Signal Sequence of the Abundant Extracellular Metalloprotease. PPEP‑1 Can Be Used to Secrete Synthetic Reporter Proteins in...
0 downloads 0 Views 1MB Size
Letter pubs.acs.org/synthbio

The Signal Sequence of the Abundant Extracellular Metalloprotease PPEP‑1 Can Be Used to Secrete Synthetic Reporter Proteins in Clostridium dif f icile Ana M. Oliveira Paiva, Annemieke H. Friggen, Shabnam Hossein-Javaheri, and Wiep Klaas Smits* Department of Medical Microbiology, Section Experimental Microbiology, Leiden University Medical Center, 2333 ZA Leiden, The Netherlands S Supporting Information *

ABSTRACT: Clostridium dif f icile is an opportunistic pathogen and the main cause of antibiotic-associated diarrhea. Adherence of C. dif ficile to host cells is modulated by proteins present on the bacterial cell surface or secreted into the environment. Cleavage of collagen-binding proteins is mediated by the zinc metalloprotease PPEP-1, which was identified as one of the most abundant secreted proteins of C. dif f icile. Here, we exploit the PPEP-1 signal sequence to produce novel secreted enzymes. We have constructed two functional secreted reporters, AmyEopt and sLucopt for gene expression analysis in C. dif f icile. AmyEopt extracellular activity results in starch degradation and can be exploited to demonstrate promoter activity in liquid or plate-based assays. sLucopt activity could reliably be detected in culture supernatant when produced from an inducible or native promoter. The secreted reporters can be easily assessed under aerobic conditions, without the need of complex sample processing. KEYWORDS: amylase, PPEP-1, secretion, luciferase

C

Gene expression analysis of C. dif f icile can be challenging. Many common reporters used to study gene expression in bacteria are not suitable to use in C. dif f icile studies, as they require oxygen for maturation or are produced at insufficient levels due to different codon usage. A number of genetic tools are available to study gene expression in C. dif f icile, including βglucuronidase (gusA), alkaline phosphatase (phoZ) and various fluorescent proteins.9−14 However, no secreted reporters have been described for C. dif f icile to date. In the present study, we show that the signal sequence of PPEP-1 can be fused to synthetic constructs to yield secreted proteins in C. dif f icile. We exemplify this strategy by generating two novel secreted reporters, AmyEopt and sLucopt, that allow screening of gene expression activity on plates as well as in liquid medium, without the need for complex processing of samples. Our lab strain of C. dif ficile, 630Δerm15,16 is not capable of breaking down starch, suggesting that no functional amylase is produced under laboratory conditions (Figure 1A). Bacillus subtilis does produce a functional α-amylase, encoded by the amyE gene, and this feature has been exploited to ascertain double crossover integration of DNA into this locus.17 We reasoned that it might be possible to use the properties of this enzyme to engineer a synthetic amylase that could function

lostridium dif f icile is a Gram-positive anaerobic bacterium and is the leading cause of antibiotic associated diarrhea in the healthcare environment. Symptoms of Clostridium dif f icile infection (CDI) can range from mild diarrhea to pseudomembranous colitis and even death.1,2 The incidence and severity of CDI have increased worldwide in the past decades due to the appearance of epidemic strains. Recently, an increase of CDI cases in the community has been noted.2 Consequently, the interest in the physiology of the bacterium has increased. The ability of C. dif f icile to adhere to intestinal epithelial cells plays a crucial role in the development of the disease. Adherence is modulated by proteins present on the cell surface or secreted into the environment, such as the S-layer proteins that cover the C. dif f icile cell surface3 or the components of the flagella that confer motility to the cells.4 The secreted toxins TcdA and TcdB compromise the intestinal barrier by disrupting the actin cytoskeleton of the epithelial cells, leading to morphological alterations and eventually cell death.2,5 Recently, the metalloprotease PPEP-1 (CD2830; EC 3.4.24.89) has been identified among the most highly secreted proteins in both the laboratory strain 630Δerm as well as the epidemic strain R20291 (a representative of the PCR ribotype 027, BI, NAP01).6 PPEP-1 has been suggested to regulate the switch between adhesion and motility phases through the cleavage of Pro-Pro peptide bonds in the collagen binding protein CD2381 and other proteins.6−8 © XXXX American Chemical Society

Received: April 5, 2016

A

DOI: 10.1021/acssynbio.6b00104 ACS Synth. Biol. XXXX, XXX, XXX−XXX

Letter

ACS Synthetic Biology

Figure 1. Visualization of secreted amylase activity. (A) Staining with 0.2% iodine solution of BHI agar plates supplemented with 0,1% soluble starch. Growth of the C. dif f icile strains 630Δerm (no plasmid), WKS1588 (Ptet‑gusA), WKS1594 (Ptet-amyEopt) and SJ113 (Pveg-amyEopt) growth after 24 h in the presence or absence of 500 ng/mL ATc. Clear halos (yellow) indicate starch degradation by the α-amylase. (B) Quantification of amylase activity of the strain WKS1594 (Ptet-amyEopt) at 0, 60, or 180 min after induction with 200 ng/mL ATc. (C) Quantification of amylase activity of the strain SJ113 (Pveg-amyEopt) assayed at 0 and 360 min after inoculation. (D) Quantification of the relative halo size (halo size of Ptet-amyEopt/halo size of PvegamyEopt) on BHI agar plates supplemented with 0,1% soluble starch in the presence of increasing amounts of ATc (0, 20, 50, 100, 200, and 500 ng/ mL). Error bars represents the ± standard deviation of triplicate samples. Using a Student’s t test for statistical analysis, the following combinations were found to be significantly different from each other (p < 0.05): 0 vs 50, 100, 200 and 500; 20 vs 200 and 500; 50 vs 500 ng/mL ATc.

as a plate-based reporter for promoter activity in C. dif f icile. We therefore fused a codon optimized AmyE from B. subtilis to the signal sequence of PPEP-1, resulting in the secreted reporter AmyEopt. To ensure high level induction the previously published anhydrotetracyclin (ATc) inducible promoter was used.9 We found that the C. dif f icile strain expressing AmyEopt was able to hydrolyze the starch, as visible by halo formation after staining starch containing plates with an iodine solution (Figure 1A). As expected, the halo was larger when the strain was induced with 500 ng/mL ATc compared to the noninduced condition. It has previously been noted that the Ptet promoter is leaky and prolonged incubation in combination (in our case 24 h) with the efficient enzymatic activity of the synthetic amylase could explain the observed breakdown of starch under noninducing conditions (Figure 1A). Nevertheless, halo formation was specific for the amylase containing plasmid, as it did not occur with cells containing the control plasmid pRPF1859, harboring Ptet-gusA (Figure 1A). Amylase-producing colonies could easily be distinguished from amylase-negative colonies; in a 5:1 mixed culture of wild type and Ptet-amyE harboring C. dif f icile, 21% of the colonies were found to hydrolyze the starch when plated (Supplemental Figure S3). Several methods exist to quantify amylase activity in liquid samples. We used a colorimetric assay (see Methods) to determine amylase activity in liquid cultures. A C. dif f icile strain capable of expressing AmyEopt was induced with 200 ng/mL ATc in BHI and supernatants collected for quantification of the amylase activity (Figure 1B). At the time of induction (0 min) no amylase activity was detected, suggesting the leakiness of the

Ptet promoter is less pronounced in the time frame of this assay. At 60 min amylase activity was detected (43,27 ± 7,5 nmol/ min/mL) with a further 2-fold increase at 180 min (95,38 ± 8,4 nmol/min/mL). In liquid medium the expression of AmyEopt does not allow the growth on starch as sole carbon source (data not shown), likely due to an inability to further break down the disaccharides that result from the degradation of starch.18 We conclude that the signal sequence of PPEP-1 is able to drive the secretion of the functional AmyEopt and the predicted extracellular activity is correlated with the starch degradation observed in the solid medium and amylase activity in liquid cultures. Therefore, AmyEopt can be used as a reporter for gene activity in C. dif f icile. Next, we wanted to demonstrate whether the observed effect could be extended beyond the inducible promoter Ptet. Different organisms demonstrate different preferences for promoter sequences. As a result, it is frequently necessary to optimize promoter sequences for inducible gene expression systems, or screen a library of promoter fragments to determine desired characteristics. A plate based screening method could facilitate these promoter-trap experiments. We placed the well-characterized B. subtilis promoter Pveg19,20 upstream of the amyEopt sequence to determine whether it can be used to drive gene expression in C. dif f icile. The incubation of the C. dif f icile strain carrying the Pveg-amyEopt construct produced large halos on starch plates (Figure 1A), which appeared comparable to the strain carrying the inducible PtetamyEopt construct. However, when amylase activity of the C. dif ficile strain carrying the Pveg-amyEopt was quantified after 6 B

DOI: 10.1021/acssynbio.6b00104 ACS Synth. Biol. XXXX, XXX, XXX−XXX

Letter

ACS Synthetic Biology h of inoculation (Figure 1C), values were approximately 2-fold lower (52,8 nmol/min/mL) than those measured for the strain with the induced Ptet-amyEopt (Figure 1B). The results above suggest that a plate-based readout of amylase activity has its limitations. Pveg is believed to be a constitutively expressed promoter.19,20 We reasoned that the use of a Pveg-amyEopt control might allow for a semiquantitative measure of amylase activity that can overcome some of the inherent variability of a plate based assay. We determined the relative halo size at varying amounts of ATc (0, 20, 50, 100, 200, and 500 ng/mL) after 24 h of incubation. An increase in halo size was evident with increasing amounts of ATc (Figure 1D), although they only reached statistical significance when the highest and lowest concentrations of inducer were compared (Figure 1D). Together, these data show that Pveg from B. subtilis is a functional and highly expressed promoter in C. dif f icile and that AmyEopt can be used to verify the presence and relative strength of promoter sequences in plate-based or liquid assays. To extend the use of the signal sequence of PPEP-1 for the secretion of synthetic proteins beyond the amylase, we generated a novel secreted luciferase reporter. Luciferasebased bioluminescence assays are widely used for gene expression studies and promoter activity. However, due to requirement for oxidation in the reaction with the substrate the application to anaerobic organisms is limited. We anticipated that efficient secretion of the luciferase would allow assaying of culture supernatant in an aerobic environment without removing the original culture from the anaerobic environment, which could cause stress that affects gene expression levels. We fused the PPEP-1 signal sequence to a codon optimized luciferase based on NanoLuc,21 yielding the sLucopt reporter. We analyzed Ptet dependent expression of sLucopt by assaying luciferase activity in diluted culture supernatants at different concentrations of ATc (0, 20, 50, 100, 200, and 500 ng/mL) at 180 min after induction (Figure 2A). In the absence of inducer, a low background signal was detected (830 ± 30 RLU/OD). Induction with ATc increased the signal up to ∼500-fold (452713 ± 6022 RLU/OD, P < 0.000001) in extracellular luciferase activity, in a dose dependent manner. No significant increase of luciferase activity was observed when cells were induced with >100 ng/mL of ATc, suggesting maximal expression from this promoter. The luciferase activity largely mirrored the detection of sLucopt in culture supernatant by immunoblotting using antiNanoLuc antibodies (kindly provided by Promega) (Figure 2B). Much greater sample volumes were required for the immunoblot detection, suggesting less sensitivity than the luciferase assay. The signal decay in our hands was identical to that described for NanoLuc activity21 and after storing cell-free supernatants for 1 month at −20 °C identical signals were obtained (data not shown). Thus, samples can be harvested throughout a growth experiment and assayed for luciferase activity in a microtiter plate after all samples have been collected. As a result of the stability of sLucopt, continuous expression of the reporter is expected to result in increased signals over time. To determine if this was the case, we assayed the extracellular luciferase activity of C. dif f icile strains expressing sLucopt and gusA under the control of Ptet, induced with 200 ng/mL ATc over time (Figure 2C). No signal was detected for the Ptet-gusA containing strain. Upon induction of sLucopt, luciferase activity increased significantly between 60 (529815 ± 771 RLU/OD)

Figure 2. Luciferase reporter assays with the inducible Ptet promoter. (A) Luciferase activity of 1:100 diluted culture supernatant of C. dif ficile strain harboring Ptet-sLucopt induced with different amounts of anhydrotetracyclin (ATc) at 180 min. (B) Western blot analysis of the same samples with anti-NanoLuc antibody. (C) Luciferase activity of C. dif f icile strains harboring Ptet-sLucopt, assayed at 0, 60, and 180 min after induction. Ptet-gusA and noninduced samples were used as controls and for these luciferase activity was measured at 180 min. (D) Luciferase activity of the C. dif f icile strain harboring Ptet-sLucopt. Extracellular (line) and intracellular (broken line) luciferase activity following induction at 5, 15, 30, 60, 90, 120, 150, and 180 min. (E) Western blot analysis of the culture supernatant of the same samples. Error bars represents the ± standard deviation of triplicate samples. Student’s t test was used for statistical analysis, * indicates a P value of