Involvement of a putative intercellular signal-recognizing G protein-coupled receptor in the engulfment of Salmonella by the protozoan Tetrahymena.

In an effort to investigate the molecular basis of protozoa engulfment-mediated hypervirulence of Salmonella in cattle, we evaluated protozoan G protein-coupled receptors (GPCRs) as transducers of Salmonella engulfment by the model protozoan Tetrahymena. Our laboratory previously demonstrated that non-pathogenic protozoa (including Tetrahymena) engulf Salmonella and then exacerbate its virulence in cattle, but the mechanistic details of the phenomenon are not fully understood. GPCRs were investigated since these receptors facilitate phagocytosis of particulates by Tetrahymena, and a GPCR apparently modulates bacterial engulfment for the pathogenic protozoan Entamoeba histolytica. A database search identified three putative Tetrahymena GPCRs, based on sequence homologies and predicted transmembrane domains, that were the focus of this study. Salmonella engulfment by Tetrahymena was assessed in the presence of suramin, a non-specific GPCR inhibitor. Salmonella engulfment was also assessed in Tetrahymena in which expression of putative GPCRs was knocked-down using RNAi. A candidate GPCR was then expressed in a heterologous yeast expression system for further characterization. Our results revealed that Tetrahymena were less efficient at engulfing Salmonella in the presence of suramin. Engulfment was reduced concordantly with a reduction in the density of protozoa. RNAi-based studies revealed that knock-down of one the Tetrahymena GPCRs caused diminished engulfment of Salmonella. Tetrahymena lysates activated this receptor in the heterologous expression system. These data demonstrate that the Tetrahymena receptor is a putative GPCR that facilitates bacterial engulfment by Tetrahymena. Activation of the putative GPCR seemed to be related to protozoan cell density, suggesting that its cognate ligand is an intercellular signaling molecule.

Introduction

Protozoa engulf Salmonella and, while the bacteria reside inside the protozoa, host cell invasion capabilities are hyperactivated and the bacteria are hypervirulent after exiting the protozoa. This protozoa-mediated hypervirulence occurs only for Salmonella strains bearing the multiresistance integron designated as SGI1, and its in vivo relevance has only been associated with protozoa that engulf Salmonella in the bovine rumen (Rasmussen et al., 2005; Carlson et al., 2007; Xiong et al., 2010; Brewer et al., 2011). Free-living protozoa, like Acanthamoeba and Tetrahymena, are models for studying this phenomenon (Carlson et al., 2007; Xiong et al., 2010; Brewer et al., 2011).

Protozoal determinants of engulfment are unclear but Renaud et al. (1995) determined that G protein-coupled receptors (GPCRs) are apparently involved in Tetrahymena phagocytosis of particulates. Furthermore, Picazarri et al. (2005) demonstrated that bacterial engulfment is governed by a GPCR in the pathogenic protozoan Entamoeba histolytica (Picazarri et al., 2005). Additionally, Lampert et al. (2011) presented evidence for GPCR-encoding genes in the Tetrahymena database (Lampert et al., 2011).

The aim of this study was to assess the role of three putative GPCRs in the engulfment of Salmonella by Tetrahymena. These three putative GPCRs were identified in a GenBank database search. We used RNAi to preliminarily determine if any of these putative Tetrahymena GPCRs are involved in Salmonella engulfment. A heterologous yeast expression system was used to partially characterize one of these candidate Tetrahymena GPCRs. Using these approaches we were able to identify a putative Tetrahymena GPCR involved in the engulfment of Salmonella.

Materials and Methods

Tetrahymena culture

Tetrahymena thermophila was obtained from ATCC and axenically grown in the recommended ATCC medium (5.0 g/L proteose peptone, 5.0 g/L tryptone, and 0.2 g/L K2HPO4) at 25°C. Media was replaced every four days and cells were diluted 1:14 in fresh media.

Tetrahymena engulfment of Salmonella in the presence of suramin

Approximately 105 to 109 CFUs/mL of bacteria were added to approximately 101 to 105 Tetrahymena/mL, maintaining a multiplicity of infection equal to 104. The Salmonella-Tetrahymena mixture was then gently rolled for 16 hrs at 37°C in a sealed 5 mL glass tube, in the presence of various concentrations of suramin (0-200μM)- a non-specific inhibitor of GPCRs (Freissmuth et al., 1999). At the end of the 16-hour incubation period, extracellular Salmonella were killed using 300 μg/mL florfenicol [Schering-Plough; (Rasmussen et al., 2005)].

Protozoa were then lysed for 60 sec at 4,800 rpm using 2.5 mm glass beads and a mini-beadbeater (Biospec Products). The lysate was centrifuged at 10,000 x g for 2 min, then resuspended in 350 μL Lennox L broth (Difco) that was plated on Salmonella-selective XLD agar plates (Fisher Scientific) and then incubated at 37°C overnight. Characteristic black colonies were then counted the following day. Percent engulfment was calculated as such: 100x (number of Salmonella recovered from protozoa/number of Salmonella added to Tetrahymena).

RNAi experiments

A preliminary GenBank database search identified three putative GPCR genes in Tetrahymena (XM_001009792.2, XM_001027519.2, and XM_001010055.2). siRNA was designed, using the Invitrogen webportal (), to silence expression of these three genes in Tetrahymena. The sequences GAGATTACTACTAATAGCCTCTCTT, GCTGATTCATTTAATAGCCTTGCTT, and TGGCTCAGTGTAAGTGACTTAATAT were deemed to be appropriate siRNA targets for the three putative GPCRs, respectively. A random sequence (CTGACGACAGTTGCATAAAGC) was used as a control siRNA. Semi-quantitative RT-PCR (Carlson et al., 2007) was used to confirm the knock-down of the genes encoding the putative GPCRs (Table 1).

Table 1

Semi-quantitative RT-PCR of GPCR targets in Tetrahymena transformed with siRNA. Tetrahymena were transformed with siRNA and then DNA-free RNA was isolated and subjected to an RT-PCR assay that semi-quantitates transcript levels based on the number of PCR cycles required for amplicon visualization using agarose gel electrophoresis (Carlson et al., 2007).

GenBank accession numbers corresponding to the source of the siRNA used in Tetrahymena	Number of PCR cycles required for visualization of amplicons specific to the following GenBank sequences in Tetrahymena transformed with siRNA
	XM_ 00100979.2	XM_ 001027519.2	XM_ 001010055.2
XM_00100979.2	50**	10	15
XM_001027519.2	15	>50**	10
XM_001010055.2	10	15	50**
random RNAi	15	10	15
mock RNAi	10	10	10

on-target effects on transcription.

Tetrahymena thermophila were grown in 5 mL of media to reach confluency (3 × 104 cells/mL), and were then harvested by centrifugation at 3,000 x g for 10 min. The pellet was washed twice in 15 mL of deionized water and resuspended in 200 μL of deionized water. These cells were then electroporated (0.2 cm electrode gap, 10 μF, and 5 milliseconds) with 5 nM siRNA.

At 24 hours after electroporation with siRNA, Tetrahymena cells were centrifuged at 4,000 x g for 5 min. Tetrahymena were then spectrophotometrically enumerated and resuspended in 1 mL of Luria-Bertani broth containing SGI1-bearing Salmonella at a multiplicity of infection equal to 104. After 1 hr of co-incubation, non-engulfed bacteria were then killed with florfenicol (300 μg/mL) and protozoa were lysed with bead-beating. Protozoal lysates were recovered and plated on selective agar (XLD) for the enumeration of Salmonella engulfed by Tetrahymena, as described herein.

Manipulation of the gene encoding the Tetrahymena GPCR

RNAi studies identified one of the putative GPCRs [XM_001009792.2] as a potential determinant of Salmonella engulfment.

In order to deorphanize this receptor in the yeast heterologous expression system, its gene was cloned into a yeast expression vector. Since Tetrahymena genes have read-through stop codons encoding glutamine residues (Adachi and Cavalcanti, 2009), the gene sequence encoding this GPCR was synthesized (GeneScript) whereby the 13 read-through stops codons (TAA or TAG) were exchanged for glutamine codons (CAA or CAG). Codons were also optimized for expression in yeast.

Construction and cloning of the yeast expression vector containing the Tetrahymena GPCR gene

The synthetic Tetrahymena GPCR gene was PCR-amplified with forward (5’GCCATACCATGGACCAATCATTTGGAAATCAA3’) and reverse (5’GCCATAGGATCCTCAAGTTAGATTTATTTCACGTGAAT3’) primers, which included filler sequences (underlined) and the restriction sites NcoI and BamHI (italicized) incorporated into the 5’ and 3’ ends of the amplicon, respectively. Purified amplicons and the linearized yeast expression vector Cp4258, which bears a leucine prototrophic marker (Kimber et al., 2009; Wang et al., 2006), were co-digested with NcoI and BamHI restriction endonucleases (New England BioLabs).

The digested vector and amplicons were ligated with T4 DNA ligase (New England BioLabs) and the resulting plasmids were transformed into Escherichia coli K12 and individual clones were selected in ampicillin (resistance encoded by the Cp4258 vector) and grown aerobically in LB broth overnight at 37°C. Plasmid DNA was purified using the HiSpeed Plasmid Mini Kit (Qiagen) and inserts were verified using PCR and then sequenced to confirm gene orientation and fidelity.

Transformation of yeast with the Cp4258/GPCR expression vector

Saccharomyces cerevisiae strain CY 19043 (J. Broach, Princeton University, USA) was used as the yeast recipient since these cells are leucine/histidine auxotrophs and exhibit a histidine prototrophic phenotype upon GPCR activation even if the receptor is exogenous (Wang et al., 2006; Kimber et al., 2009).

Non-transformed CY 19043 yeast were grown in YPD media supplemented with all essential amino acids. Cells at mid-log phase (OD600 equal to 0.3 to 0.5) were co-incubated with 1μg of Cp4258/GPCR construct, or Cp4258 bearing a non-Tetrahymena GPCR gene [an unpublished putative GPCR sequence from Cryptosporidium parvum (C. parvum)], in the presence of 200 μg salmon sperm DNA (Invitrogen) and 0.1M LiAc (Sigma-Aldrich).

Yeast cells were then incubated at 30°C and then heat shocked at 42°C for 15 minutes. Cells were plated on leucine-deficient media [1x YNB (Difco), 1x yeast synthetic dropout medium supplement without leucine (Sigma), 10mM ammonium sulfate (Sigma), and 50% glucose] to select for transformation of Cp4258/GPCR. Transformants were verified by isolating plasmids (Promega Wizard Prep kit) and PCR-based detection of the Tetrahymena GPCR gene insert and its proper orientation.

Partial deorphanization of the putative Tetrahymena GPCR using the yeast heterologous expression system

A volume of 3 mL of leucine-deficient media was inoculated with yeast expressing the Tetrahymena GPCR, or the control, and grown at 30°C to an OD600 equal to one. Cells were washed three times with leucine/histidine deficient medium [1x YNB (Difco), 1x yeast synthetic drop out medium supplement lacking leucine and histidine (Sigma), 10mM ammonium sulfate, 50% glucose, 50mM 4-morpholinepropanesulfonic acid, pH 6.8], and then resuspended in 1 mL leucine/histidine-deficient media to a density of 15-20 cells/µL. Approximately 3,000 cells were added to each well of 96-well plates containing the same medium along with 50 µL of Tetrahymena culture media, Tetrahymena lysates (15-20 cells/µL lysed by bead-beating), or Tetrahymena lysates plus suramin (200μM). Cells were grown at 30°C for approximately 24 hours after which growth was measured spectrophotometrically at OD600. Yeast cells transformed with the C. parvum GPCR-encoding vector were used as a control for each treatment.

Statistical analysis

Statistical comparisons were made using ANOVA with Scheffe’s F test for multiple comparisons. Comparisons were made across protozoal densities, suramin treatments, siRNA transformations, and yeast treatments. GraphPad Prism 5.0 was the software used, with p<0.05 indicating statistical significance.

Hydropathy analysis of the putative Tetrahymena GPCR

Since GPCRs are comprised on seven transmembrane domains (Kobilka et al., 1987), putative transmembrane domains were assessed using Kyte-Doolittle plots (Kyte and Doolittle, 1982) from a web-based platform (). Transmembrane signatures were ascribed to peptide regions in which hydropathy scores rose from negative values to values greater than one. The codon-optimized deduced version of the protein, where glutamine codons replace read-through amber (TAG) or ochre (TAA) stop codons, was used for the hydropathy analysis.

Results

Suramin-mediated inhibition of the protozoa density-dependent engulfment of Salmonella by Tetrahymena

In order to assess the possible role of a GPCR in the engulfment of Salmonella by Tetrahymena, we incubated Salmonella-Tetrahymena co-cultures with the non-specific GPCR inhibitor suramin. As shown in Fig. 1, bacterial engulfment increased with the density of Tetrahymena. As shown in Fig. 2, this engulfment was significantly hampered by suramin in a concentration-dependent manner at all Tetrahymena densities.

Fig. 1

Salmonella engulfment at various densities of Tetrahymena. For each density of protozoa, Salmonella and Tetrahymena were co-cultured using a multiplicity of infection equal to 10,000 (i.e., bacteria:protozoa ratios of 109:105, 108:104, 107:103, 106:102, and 105:101). Bacterial engulfment (average number of bacteria engulfed per cell) was determined after lysing protozoa. Each data point represents the mean ± sem for three independent experiments each performed in triplicate. *p<0.05 versus the rest of the data.

Fig. 2

Salmonella engulfment by Tetrahymena exposed to suramin. For each concentration of suramin, Salmonella and Tetrahymena were co-cultured using a multiplicity of infection equal to 10,000 (i.e., bacteria:protozoa ratios of 109:105, 108:104, 107:103, 106:102, and 105:101). Bacterial engulfment (average number of bacteria engulfed per cell) was determined after lysing protozoa. Each data point presents the mean ± sem for the five different Salmonella:Tetrahymena ratios. *p<0.05 versus the results from suramin-free assays.

RNAi-based screening of a GPCR involved in bacterial engulfment by Tetrahymena

Our preliminary database query identified three Tetrahymena genes encoding putative GPCRs (XM_001009792.2, XM_001027519.2, and XM_001010055.2).

In order to assess the association of these GPCRs with bacterial engulfment, we knocked-down receptor expression with siRNA and then evaluated bacterial engulfment in Tetrahymena.

As shown in Fig. 3, bacterial engulfment was significantly hampered in Tetrahymena electroporated with one of the siRNAs corresponding to the gene with GenBank accession number XM001009792.2. Expression of this receptor was diminished as verified by semi-quantitative RT-PCR, and no off-target effects were noted in the transformants (Table 1).

Fig. 3

Salmonella engulfment by Tetrahymena electroporated with RNAi corresponding to putative GPCR genes. A preliminary database search identified three putative GPCR-encoding genes (GenBank accession numbers provided in the x-axis) in Tetrahymena and siRNA was designed based on these sequences. Bacterial engulfment (average number of bacteria engulfed per cell) was determined after electroporation with siRNAs. Random RNAi and mock transformants served as controls. Data presented are the mean ± sem for three independent experiments each performed in triplicate. *p<0.05 versus the rest of the data.

Partial deorphanization of the putative Tetrahymena GPCR using Tetrahymena lysates and suramin

Results presented in Fig. 3 identified a putative GPCR involved in Salmonella engulfment by Tetrahymena. In order to deorphanize this receptor, we expressed its codon-optimized cDNA in a yeast heterologous expression system that exploits a histidine prototrophic phenotype upon GPCR activation by a cognate ligand (Kimber et al., 2009; Wang et al., 2006). We hypothesized that a Tetrahymena surface molecule activates the putative GPCR on neighboring Tetrahymena cells, thus we used Tetrahymena lysates as “ligands”.

We also assessed the activation of the putative GPCR in the presence of a suramin, a non-specific GPCR inhibitor (Freissmuth et al., 1999).

Treatment-control yeast were grown in leucine-deficient media in the absence of lysates or suramin. Transformation-control yeast were transformed with a vector encoding a non-Tetrahymena GPCR whose gene was cloned from C. parvum. As shown in Fig. 4, Tetrahymena lysates elicited significant increases in yeast growth in the transformants expressing the Tetrahymena GPCR. This growth was blocked by the addition of suramin. Tetrahymena lysates had no effect on yeast expressing the non-Tetrahymena GPCR cloned from C. parvum. However, suramin was able to reduce the basal activity associated with this non-Tetrahymena GPCR.

Fig. 4

Tetrahymena lysate-mediated activation of histidine prototrophism in yeast expressing the Tetrahymena GPCR. Yeast transformants were exposed to the lysates and yeast growth was measured spectrophotometrically at OD600. To determine background growth, yeast transformants were grown in media lacking leucine and histidine (untreated control). As an additional control, yeast cells were transformed with a vector encoding an unrelated GPCR (from C. parvum) and then grown in the same conditions. Growth is quantitated as compared to growth observed in untreated controls. Data presented are the mean ± sem for three independent experiments each performed in triplicate. *p<0.05 versus untreated controls.

Secondary structure analysis of the putative Tetrahymena GPCR

To assess the presence of motifs suggestive of a GPCR, a Kyte-Doolittle hydropathy plot (Kyte and Doolittle, 1982) was created for the putative Tetrahymena GPCR. Fig. 5 reveals that the putative Tetrahymena GPCR has the characteristic seven transmembrane domains.

Fig. 5

Kyte-Doolittle hydropathy plot of the putative Tetrahymena GPCR. Putative transmembrane domains were assessed using Kyte-Doolittle plots (Kyte and Doolittle, 1982) from a web-based platform (). Transmembrane signatures were ascribed to peptide regions in which hydropathy scores rose from negative values to values greater than one. Putative transmembrane regions are indicated numerically from one to seven.

Discussion

The objective of this study was to identify protozoal determinants involved in the engulfment of Salmonella, a phenomenon that hyperactivates the virulence of the bacteria (Rasmussen et al., 2005; Carlson et al., 2007; Xiong et al., 2010; Brewer et al., 2011). Protozoa enhance the virulence of SGI1-bearing Salmonella by hyperactivating the expression of the pro-invasion gene designated as hilA, and this hyperexpression leads to enhanced virulence when the bacteria escape from the protozoa. The hyperactivation of hilA expression is dependent on the SGI1-specific gene designated as SO13 (Carlson et al., 2007).

This phenomenon specifically occurs when SGI1-bearing Salmonella are engulfed by and then are liberated from free-living protozoa, such as Acanthamoeba and Tetrahymena (Carlson et al., 2007), or from host-associated protozoa (Rasmussen et al., 2005; Carlson et al., 2007; Xiong et al., 2010; Brewer et al., 2011). The protozoa/SGI1-bearing Salmonella hypervirulence relationship appears to be relevant to the ingestion of water contaminated with free-living protozoa harboring SGI1-bearing Salmonella (Xiong et al., 2010), and cattle that naturally harbor large numbers of protozoa in their rumen (Rasmussen et al., 2005; Carlson et al., 2007; Xiong et al., 2010; Brewer et al., 2011).

GPCRs were the focus of the present study since these receptors are apparently involved in Tetrahymena phagocytosis of particulates (Renaud et al., 1995), and another group presented evidence for GPCR-encoding genes in the Tetrahymena database (Lampert et al., 2011). Additionally, a GPCR was implicated in the engulfment of bacteria by the pathogenic protozoan Entamoeba histolytica (Picazarri et al., 2005). In the present study we provide evidence that a putative GPCR governs Salmonella engulfment by Tetrahymena. It also appears that this engulfment process is dependent upon the density of protozoa, suggesting intercellular communication by Tetrahymena.

It appears that a specific seven transmembrane-spanning protein mediates the engulfment of Salmonella. This protein is putatively a GPCR given its ability to stimulate histidine prototrophism in a yeast heterologous expression in which G proteins govern de novo synthesis of histidine following ligand occupancy of a GPCR. Histidine prototrophism was observed in the presence of Tetrahymena lysates while suramin was able to block the histidine prototrophism in the presence of Tetrahymena lysates, further indicating that the protein is a GPCR. Suramin has some unknown effects, but we found that this molecule inhibited the Tetrahymena GPCR in two disparate systems.

Additionally, suramin reduced the basal activity of a non-Tetrahymena GPCR in the yeast system. Because of the unknown effects of suramin, however, we are unable to completely rule out GPCR-independent mechanisms contributing to its effects on Tetrahymena and the yeast.

Since Tetrahymena lysates activated the putative GPCR, an unknown Tetrahymena factor(s) appears to be an autocrine-like ligand for this receptor. This factor may be membrane-associated, whereby the receptor could be constitutively activated due to continual ligand occupancy when Tetrahymena are confluent. Alternatively, this may be a secreted factor. Although Tetrahymena has been established as an important model protozoan, there is little information on GPCRs in this organism except for a recent study (Lampert et al., 2011). Interestingly, there has been debate regarding the existence of Tetrahymena GPCRs (Renaud et al., 1991).

In conclusion, we determined that a putative Tetrahymena GPCR is responsible for the engulfment of Salmonella by the protozoan. We also partially characterized this unique receptor that appears to be activated by intercellular ligands whose threshold appears to be dependent upon the density of protozoa.