SEARCH

SEARCH BY CITATION

Keywords:

  • cell signalling;
  • Schizosaccharomyces pombe;
  • ras;
  • oncogene;
  • quantitative imaging

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Ras signalling is central to fundamental and diverse cellular processes. In higher eukaryotes ras signalling is highly complex, involving multiple isoforms, regulatory proteins and effectors. As a consequence, the study of ras activity in mammalian systems presents a number of technical challenges. The model organism Schizosaccharomyces pombe has previously proved a key system for the study of human signalling components and provides an ideal model for the study of ras, as it contains just one ras protein (Ras1p), which is non-essential and controls a number of downstream processes. Here we present data demonstrating the quantitative analysis of three distinct Ras1-related signalling outputs, utilizing the three most abundant human ras isoforms, H-Ras, N-Ras and K-Ras4B, in Sz. pombe. Further, we have characterized the localization of these three human ras isoforms in Sz. pombe, utilizing quantitative image analysis techniques. These data indicate that all three human ras isoforms are functional in fission yeast, displaying differing localization patterns which correlate strongly with function in the regulation of pheromone response and cell shape. These data demonstrate that such yeast strains could provide powerful tools for the investigation of ras biology, and potentially in the development of cancer therapies. Copyright © 2013 John Wiley & Sons, Ltd.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Ras proteins are a family of small monomeric GTPases which act as binary switches in signal transduction pathways. In its inactive state, ras is bound to a single molecule of GDP. Upon stimulation by upstream activators, GDP is released to allow binding of the more cellularly abundant GTP and resulting in ras activation, in a process regulated by guanine nucleotide exchange factors (GEFs). Active ras is then able to promote signalling through downstream effectors. Ras may then subsequently be inactivated through hydrolysis of GTP to GDP, in a process regulated by GTPase-activating proteins (GAPs) (Sprang, 1997). Despite the seeming biochemical simplicity of ras activation, ras signalling in higher eukaryotes is highly complex, involving multiple upstream activators and downstream effectors, in addition to the presence of multiple ras isoforms (Malumbres and Barbacid, 2003).

The fission yeast Schizosaccharomyces pombe provides an ideal system for the study of ras signalling, as it contains a single ras protein, Ras1p, which is non-essential and regulates a number of signalling cascades (Fukui et al., 1986; Nielsen et al., 1992). During mitotic growth, Ras1p regulates cell polarity by modulating cytoskeletal elements and maintaining elongated cell morphology through the activation of the Rho GTPase Cdc42p (Garcia et al., 2006). In addition, Ras1p plays a key role in the pheromone-induced mating process. Mating in fission yeast is initiated through the reciprocal exchange of pheromones by haploid cells of opposite mating types. The binding of pheromone to its cognate G protein-coupled receptor (GPCR) causes the Gα subunit Gpa1p to promote nucleotide exchange on Ras1p, although the precise mechanism remains to be determined (Hoffman, 2005). Ras1p–GTP then propagates the signal by stimulation of a mitogen-activated protein kinase (MAPK) cascade, resulting in the transcription of mating genes (Bauman et al., 1998). In addition, Ras1p regulates chemotropic growth towards the source of pheromone (shmoo formation) to allow cells of opposite mating type to meet and fuse (Davey, 1998). Ras1p is differentially regulated by the two GEFs, Ste6p and Efc25p (Hughes et al., 1990; Tratner et al., 1997). It has been suggested that Ste6p and Efc25p act in competition, with Ste6p mediating pheromone-responsive Ras1p activation and Efc25p mediating morphological functions (Papadaki et al., 2002). In addition, previous data have indicated that the localization of Ras1p is key to differential effector activation, with endomembrane-localized Ras1p regulating cell polarity and plasma membrane-localized Ras1p activating pheromone-responsive processes (Chang and Philips, 2006; Onken et al., 2006). Ras1p has an intrinsic GTPase activity to terminate the response, irrespective of cellular location, although this process is significantly enhanced through the action of a single GAP protein Gap1p (Imai et al., 1991).

The development of pheromone-responsive reporter Sz. pombe strains (Didmon et al., 2002) has facilitated the use of Sz. pombe as a model system for studying human signalling components, including GPCRs (Ladds et al., 2003), Gα subunits (Ladds et al., 2007), Gβ subunits (Goddard et al., 2006) and regulator of G protein signalling (RGS) proteins (Hill et al., 2008). As a consequence, heterologous expression of human ras proteins may prove useful in analysing their signalling characteristics. Human ras proteins and Sz. pombe Ras1p share significant sequence homology. The N-terminal region, which is involved in nucleotide binding and effector interaction and contains the guanine nucleotide-binding (G region) and phosphate-binding (P region) domains (Vetter and Wittinghofer, 2001), is highly conserved between human ras proteins and Ras1p. In addition, Ras1p and human ras isoforms all share a CAAX farnesylation motif. Most sequence divergence is observed in the C-terminal region, or hypervariable domain, which is involved in ras localization (Eisenberg and Henis, 2008). Ras1p, in particular, has a considerably larger C-terminus than human ras isoforms, the significance of which remains to be determined. Despite this divergence, the high level of homology in the nucleotide binding and effector interaction domains of these proteins suggests human ras and Ras1p might be functionally interchangeable, a possibility that has been subject to limited exploration (Nadin-Davis et al., 1987).

In this study we present the quantitative analysis of multiple ras signalling outputs for three human ras isoforms, H-Ras, N-Ras and K-Ras4B, in Sz. pombe. H-Ras, N-Ras and K-Ras4B were used in this study as they are the most abundant human ras isoforms, and also the most divergent in their C-terminal localization domains. In addition, we present quantitative imaging data characterizing the localization of these three human ras isoforms in Sz. pombe and demonstrate differing characteristics of ras activity between human ras isoforms and Ras1p.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Strains, reagents and general method

All yeast strains used in this study are detailed in Table 1. Pheromone-responsive strains for analysis of human ras signalling and localization were derived from strain JY544 (mat1-M, ∆mat2/3::LEU2-, leu1-, ade6-M216, ura4-D18, cyr1-D51, sxa2 > lacZ) (Didmon et al., 2002). Ras1p gene disruption was achieved through replacement of the ras1 open reading frame (ORF) with a 1.8 kb ura4 cassette, generating the strain JY1247. Human H-Ras, N-Ras and K-Ras4B ORFs were integrated at the ras1 locus through replacement of the ura4 cassette from JY1247, generating JY1461, JY1462 and JY1463, respectively. The ura4 cassette was removed from JY1247 to create the Δras1 strain JY1279 (mat1-M, ∆mat2/3::LEU2-, leu1-, ade6-M216, ura4-D18, cyr1-D51, sxa2 > lacZ, Δras1). The yeast strains used in this study to analyse mating were derived from strain JY444 (mat1-M, ∆mat2/3::LEU2-, leu1-32, ura4-D18), or where sxa2- were expressing sxa2 from pREP41x, which restores mating to wild-type levels (data not shown). Gene deletion and replacement was confirmed by polymerase chain reaction (PCR) and immunoblotting. All yeast procedures were performed as described in Ladds et al. (1996). Culture media used were yeast extract (routine cell growth) and defined minimal medium (all assays and selective growth) (Davey et al., 1995). DNA manipulations were performed using standard methods. Oligonucleotides were supplied by Invitrogen (Paisley, UK). PCR amplification used FastStart Taq polymerase (Roche Diagnostics, Burgess Hill, UK). All constructs were sequenced prior to use. Human ras ORFs were obtained from the Guthrie cDNA Resource Center (www. cdna.org; University of Missouri at Rolla, USA).

Table 1. Schizosaccharomyces pombe strains
StrainGenotype
  1. The term sxa2 > lacZ is used to indicate a strain in which the lacZ open reading frame is placed under the transcriptional control of the sxa2 promoter.

JY544mat1-M, ∆mat2/3::LEU2-, leu1-, ade6-M216, ura4-D18, cyr1-D51, sxa2 > lacZ (Didmon et al., 2002)
JY1025mat1-P, ∆mat2/3::LEU2-, leu1-32, ura4-D18
JY1279mat1-M, ∆mat2/3::LEU2-, leu1-, ade6-M216, ura4-D18, cyr1-D51, sxa2 > lacZ, Δras1
JY1461mat1-M, ∆mat2/3::LEU2-, leu1-, ade6-M216, ura4-D18, cyr1-D51, sxa2 > lacZ, ras1::h-ras
JY1462mat1-M, ∆mat2/3::LEU2-, leu1-, ade6-M216, ura4-D18, cyr1-D51, sxa2 > lacZ, ras1::n-ras
JY1463mat1-M, ∆mat2/3::LEU2-, leu1-, ade6-M216, ura4-D18, cyr1-D51, sxa2 > lacZ, ras1::k-ras4b
JY1469mat1-M, ∆mat2/3::LEU2-, leu1-, ade6-M216, ura4-D18, cyr1-D51, sxa2-, tea1-mCherry
JY1510mat1-M, ∆mat2/3::LEU2-, leu1-, ade6-M216, ura4-D18, cyr1-D51, sxa2-, tea1-mCherry, ras1::h-ras
JY1411mat1-M, ∆mat2/3::LEU2-, leu1-, ade6-M216, ura4-D18, cyr1-D51, sxa2-, tea1-mCherry, ras1::k-ras4b
JY1569mat1-M, ∆mat2/3::LEU2-, leu1-, ade6-M216, ura4-D18, cyr1-D51, sxa2-, tea1-mCherry, Δras1
JY1570mat1-M, ∆mat2/3::LEU2-, leu1-, ade6-M216, ura4-D18, cyr1-D51, sxa2-, tea1-mCherry, ras1::n-ras

Generation of ras integration constructs

The ras1 ORF, in addition to approximately 500 bp upstream and downstream of the gene, were amplified by PCR, using the sense oligonucleotide JO1752 (CTGAATAGGGAATTCTTCAAAC) and the antisense oligonucleotide JO1753 (AAAAAAAGATCTACGGGA), to allow cloning into EcoRI- and BglII-cut pSP72 (Promega, Southampton, UK), utilizing endogenous EcoRI (JO1752) and BglII (JO1753) restriction sites in the genomic sequence (underlined). The ras1 ORF was removed and an EcoRV site (underlined) introduced between the flaking regions, using inverse PCR (sense oligonucleotide-gatatcCAAGTATTATTGCAGAA and anti-sense oligonucleotide-ATTCACTATTTTATAAAGCACAC). The three human ras isoforms were cloned directly into the new EcoRV site, following PCR amplification with a sense oligonucleotide beginning at the initiating ATG (bold) (H-Ras-ATGACGGAATATAAGC, N-Ras-ATGACTGAGTACAAACTG and K-Ras4B-ATGACTGAATATAAACTTG) and an antisense oligonucleotide beginning at the stop anti-codon (bold) (H-Ras-TCAGGAGAGCACACACTTGC, N-Ras-TTACATCACCACACATGGCAATC and K-Ras4B-TTACATCACCACACATGGCAATC).

Generation of pREP expression constructs

The pREP series of vectors allows expression of genes in Sz. pombe under the control of the thiamine-repressible nmt1 promoter (Maundrell, 1993). All ras isoform ORFs were amplified using a sense oligonucleotide beginning at the initiating ATG (bold) (Ras1p-ATGAGGTCTACCTACTTAAGAGAG, H-Ras-ATGACGGAATATAAGC, N-Ras-ATGACTGAGTACAAACTG and K-Ras4B-ATGACTGAATATAAACTTG) and an antisense oligonucleotide containing a BamHI site (underlined) immediately prior to the endogenous stop anti-codon (bold) (Ras1p-ccggatccCTAACATATAACACAACA, H-Ras-ggggatccTCAGGAGAGCACACACTTGC, N-Ras-ggggatccTTACATCACCACACATGGCAATC and K-Ras4B-ggggatccTTACATCACCACACATGGCAATC). PCR products were digested with BamHI and cloned into a pREP3x vector, modified to contain a unique EcoRV site upstream of the BamHI site. The Ras1pG17V mutant was generated by inverse PCR mutagenesis, using the sense oligonucleotide TGGTGTTGGTAAAAGTGC and antisense oligonucleotide ACATCTCCTACAACTACCA. N-terminal GFP–ras fusions were generated through a two-step cloning strategy. GFP was first amplified using a sense oligonucleotide beginning at the initiating ATG (bold) (ATGAGTAAAGGAGAAGAAC) and an antisense oligonucleotide lacking the endogenous stop anti-codon and containing half an EcoRV site (underlined) (atcTTTGTATAGTTCATCCA). The PCR product was cloned into the modified pREP3x cut with EcoRV (pREP3x-GFP). Ras ORFs were cloned into pREP3x–GFP, using the same strategy as described for pREP3x.

Fluorescence microscopy

Fluorescence microscopy was performed using techniques modified from Ladds et al. (2003). Cells were cultured to a density of 5 × 106 cells/ml in the relevant medium. 5 µl cell culture was transferred to a CoverWellTM imaging chamber (Grace Bio-Labs, OR, USA) containing solid DMM (2% agarose) and imaged using a True Confocal Scanner Leica TCS SP5 microscope (Leica Microsystems, Milton Keynes, UK). Tea1-mCherry was visualized using a Personal DeltaVision (Applied Precision, Issaquah, WA, USA), consisting of an Olympus UPlanSApo × 100, N.A. 1.4, oil-immersion objective and a Photometric CoolSNAP HQ camera (Roper Scientific). Analysis of cell morphology was performed on cells expressing GFP from the nmt1 promoter of pREP3x.

Image analysis

Image analysis was performed using the open source program ImageJ (http://rsb.info.nih.gov/ij/). Cell segmentation was achieved using the BOA plug-in component of the Quantitative Imaging of Membrane Proteins (Quimp) package (http://go.warwick.ac.uk/bretschneider/quimp) (Bosgraaf et al., 2009; Dormann et al., 2002).

Assays for ras signalling in Sz. pombe

β-Galactosidase assays for pheromone-responsive transcription were performed using a method modified from Dohlman et al. (1995) (Didmon et al., 2002; Ladds et al., 2003). Analysis of cell size and cell density was achieved using a Coulter Channelyzer (Beckman Coulter, Luton, UK) (Davey, 1991). Spore formation following mating was determined using differential interference contrast (DIC) microscopy (Egel et al., 1994).

Immunoblotting

Whole-cell protein extracts were prepared from strains expressing Ras1p, H-Ras, N-Ras and K-Ras4B from the endogenous Ras1p promoter, in addition to ΔRas1p cells, using the method described in Ladds and Davey (2000b). Protein samples were resolved using denaturing sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE) and transfered to a polyvinyl difluoride membrane (Bio-Rad, Hemel Hempstead, UK). A primary anti-ras mouse monoclonal antibody (Millipore, Durham, UK) and secondary horseradish peroxidase-conjugated anti-mouse IgG antibody (Promega, Southampton, UK) were used for western blotting. Bound antibody was then detected, using enhanced chemiluminescence reagents (GE Healthcare, Amersham, UK) and visualized using a Syngene G:Box gel documentation system (Syngene, Cambridge, UK).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Quantification of GFP–ras fusion protein localization

All the ras proteins used in this study display a high degree of homology in their effector interaction and nucleotide binding domains, but differ in their C-terminal hypervariable domains (Figure 1A). All isoforms contain a C-terminal CAAX farnesylation motif to promote membrane association. Ras1p and N-Ras each contain a single palmitoyl group to further aid plasma membrane localization. H-Ras, by contrast, contains two palmitoyl groups and K-Ras4B is not palmitoylated but contains a poly-basic domain to allow interaction with lipid head groups at the plasma membrane (Bijlmakers and Marsh, 2003; Eisenberg and Henis, 2008; Smotrys and Linder, 2004) (Figure 1B). In addition, the sequences immediately upstream of these modifications, which have previously been implicated in localization (Hancock, 2003), are highly divergent between the isoforms.

image

Figure 1. The localization of Ras1p and human H-Ras N-Ras and K-Ras4B. (A) Sequence alignment of Sz. pombe Ras1p and human H-Ras, N-Ras, K-Ras4A and K-Ras4B. Sequences were aligned using MutiAlin v. 5.4.1 (Corpet, 1988). Gaps included to optimize the alignment are indicated by -; identical residues are shaded in black and conservative changes in grey. (B) The differing localization domains of these ras proteins (dark grey) containing the 16 carbon palmitoyl and 12 carbon farnesyl groups are indicated in (A). The polybasic domain of K-Ras4B is indicated by a series of lysine (K) residues. (C) Direct in-frame N-terminal GFP–ras fusions were generated in the pREP3x vector, allowing expression from the thiamine-repressible nmt1 promoter. GFP alone was also expressed as a control. Images were obtained using a Leica SP5 scanning confocal microscope equipped with a × 100 oil-immersion objective; scale bar = 10 µm. (D) Image analysis was performed using Quimp image analysis software (Bosgraaf et al., 2009; Dormann et al., 2002), allowing determination of plasma membrane fluorescence. Left hand panel shows a representative cell expressing GFP–ras1p and the right depicts the cell following segmentation; red line, cell perimeter; blue line, membrane–interior partition; determination of pixel intensity is represented visually as a heat map. (E) Percentage plasma membrane fluorescence data presented are mean ± SEM of 10 determinants. Statistical significance compared to Ras1p was determined using one-way ANOVA with Tukey's multiple comparison post-test; *p < 0.05, **p < 0.01, ***p < 0.01

Download figure to PowerPoint

To determine a possible link between ras activity and localization, direct in-frame N-terminal GFP fusions were created with Ras1p, H-Ras, N-Ras and K-Ras4B. GFP–ras fusion proteins were expressed from the thiamine-repressible nmt1 promoter on the pREP3x vector (Maundrell, 1993). Cells were grown in the absence of thiamine, and fluorescence microscopy images were obtained in strains lacking endogenous Ras1p but expressing each individual GFP–ras fusion construct (Figure 1C). Images were analysed using Quimp image analysis software, allowing segmentation of images to isolate individual cells and allow analysis of pixel intensity within the cell (Figure 1D). Relative levels of fluorescence at the periphery compared to the interior of the cell were determined to give a quantitative measure of plasma membrane localization (Bosgraaf et al., 2009; Dormann et al., 2002) (Figure 1E).

GFP–ras1p exhibited predominantly plasma membrane localization (78 ± 3.5%), with a subpopulation of GFP–ras1p present in the interior of the cell (22 ± 3.5%). The internal fluorescence present was not cytosolic and displayed a pattern of localization consistent with endomembrane structures. Human H-Ras, N-Ras and K-Ras4B also displayed localization to both the plasma membrane and endomembrane structures. Quantitative image analysis indicated that K-Ras4B showed the highest level of plasma membrane localization (67 ± 4.6%). H-Ras exhibited an intermediate level of plasma membrane localization (42 ± 8.1%) and N-Ras showed markedly lower levels of plasma membrane localization (11 ± 3.1%), with most internal fluorescence displaying a pattern consistent with localization to endomembranes (Figure 1D).

Quantitative analysis of human ras function in Sz. pombe

Ras1p is responsible for regulating two distinct pathways in Sz. pombe. During mitotic growth Ras1p regulates cell polarization and during the switch from mitotic to meiotic development Ras1p regulates pheromone-responsive changes in both gene expression and cell morphology (Fukui et al., 1986; Garcia et al., 2006; Nielsen et al., 1992). Quantitative assays for all three Ras1p-related processes were used to determine the activity of human ras isoforms in Sz. pombe. Assays were performed in strains expressing H-Ras, N-Ras and K-Ras4B from the endogenous Ras1p promoter. Expression of all three human ras isoforms, in addition to Ras1p, was confirmed by immunoblotting (Figure 2A).

image

Figure 2. Human Ras isoforms promote pheomone-responsive signalling. (A) Expression of ras isoforms was determined in fission yeast cells by immunoblotting, using a monoclonal mouse anti-ras antibody (top panel). Coomassie stains of whole protein are included in the bottom panel as an internal control. JY544 cells modified to express Human H-Ras, N-Ras and K-Ras4B from the endogenous Ras1p promoter, in addition to unmodified JY544 and JY544 lacking Ras1p, were grown in the absence of pheromone and the presence of 100 μm pheromone. Following 16 h incubation, cells were assayed for β-galactosidase activity (B) and median cell volume (C). Data shown are averages of three independent determinants (± SEM). Statistical significance is given between paired columns using one-way ANOVA with Tukey's multiple comparison post-test; **p < 0.01, ***p < 0.01. (D) Following incubation with cells of the opposite mating type, spore formation was detected in cells expressing H-Ras, N-Ras and K-Ras4B from the endogenous ras1 locus, by DIC microscopy. Images were obtained using a Leica SP5 scanning confocal microscope equipped with a × 100 oil-immersion objective; scale bar = 10 µm

Download figure to PowerPoint

Ras1p is key to regulating mating in Sz. pombe cells. Cells which lack Ras1p are sterile and unable to undergo pheromone-responsive transcription and shmoo formation (Nielsen et al., 1992). A strain has previously been developed in which the ORF of the pheromone-responsive gene sxa2, a carboxypeptidase which is expressed rapidly upon pheromone stimulation (Imai and Yamamoto, 1992; Ladds et al., 1996; Ladds and Davey, 2000a), has been replaced by the bacterial β-galactosidase gene lacZ. β-Galactosidase activity, therefore, provides a quantitative assay for pheromone-responsive transcription (Didmon et al., 2002). All analysis was performed in cells lacking adenylate cyclase (cyr1), to allow sexual differentiation during mitotic growth (Maeda et al., 1990). Cells expressing Ras1p from the endogenous ras1 locus exhibit a significant increase in β-galactosidase activity upon incubation with 100 μm pheromone for 16 h (0.7 ± 0.1 to 29.6 ± 1.4 U) (Figure 2B). There was no significant increase in β-galactosidase activity in cells lacking Ras1p, which is consistent with previous published data suggesting an essential role for Ras1p in pheromone-responsive transcription (Nielsen et al., 1992). Cells expressing H-Ras and K-Ras4B displayed significant increases in β-galactosidase activity upon pheromone stimulation (4.9 ± 1.1 and 7.1 ± 0.4 U, respectively), albeit with lower maximal signalling compared to cells expressing Ras1p. Cells expressing N-Ras showed a trend towards increased β-galactosidase activity in the presence of pheromone, displaying a 2.5 ± 1.0 U increase. These data provide the first quantification of human ras function in Sz. pombe (Figure 2B).

Upon stimulation by pheromone, Sz. pombe cells undergo morphological changes in addition to changes in gene expression. Cells display chemotropic growth towards the source of pheromone, elongating from the tip of the cell to form a shmoo (Leupold, 1987). Cell elongation is accompanied by an increase in cell volume, providing a quantitative read-out for shmoo formation (Davey, 1991). The median volumes of cells expressing Ras1p, H-Ras, N-Ras and K-Ras4B, in addition to cells lacking Ras1p, were determined following 16 h growth in the absence and presence (100 μm) of pheromone, to determine whether significant cell elongation could be observed following stimulation (Figure 2C). Cells expressing Ras1p exhibited a significant increase in median cell volume following pheromone stimulation (69 ± 2.7 to 94 ± 1.0 fl). By contrast, cells expressing the human ras isoforms, H-Ras, N-Ras and K-Ras4B, did not display a significant increase in median cell volume following treatment with pheromone, indicating that these cells were unable to undergo shmoo formation. Despite this, all human ras isoforms did support some ascospore formation following incubation with cells of the opposite mating type, indicating a functional mating response (Figure 2D). Quantitative analysis of mating activity was performed in these strains, using heat treatment to enrich for spores (55°C for 10 min in sterile water) followed by assessment of surviving colony forming units (cfu). Significant mating was not seen in the human ras strains using this assay, whereas 32 ± 2% cfu recovery was observed in cells expressing Ras1p.

In addition to regulation of mating, Ras1p is responsible for maintaining elongated cell morphology during mitotic growth (Fukui et al., 1986). Cells lacking Ras1p exhibit abnormal cell shape, becoming shorter and rounder than their wild-type counterparts. Quantitative analysis of cell shape was performed on cells lacking Ras1p, in addition to cells expressing Ras1p, H-Ras, N-Ras and K-Ras4B from the endogenous Ras1p locus, using Quimp software to allow cell segmentation within an image and analysis of cell morphology. To enable detection of cells GFP was expressed from the nmt1 promoter of pREP3x in the absence of thiamine (Maundrell, 1993). A measure of cell elongation was used, ranging between 1 and 0, where a score of 0 represents a perfectly circular cell. Analysis was performed in cells lacking adenylate cyclase (cyr1) as, although they demonstrate a more rounded morphology, the magnitude of change in morphology between cells lacking Ras1p and cells containing Ras1p is comparable to that in a cyr1+ background (data not shown). Cells lacking Ras1p displayed a mean percentage roundness of 0.13 ± 0.02. Expression of Ras1p increased this to 0.28 ± 0.02 and expression of H-Ras, N-Ras and K-Ras4B all led to an increase in cell elongation to above 0.2 (Figure 3A). The increase in elongation observed upon expression of human H-Ras, N-Ras and K-Ras4B, indicating functional activity of human ras isoforms in maintaining elongated cell morphology. The localization of Tea1-mCherry, which displays Ras1p-dependent localization to the cell tip, was also examined to visualize polar cell growth. All three human ras isoforms supported the polar localization of Tea1-mCherry to a similar extent as Ras1p, indicating functional signalling through this pathway (Figure 3B).

image

Figure 3. Human Ras isoforms support polar cell morphology. (A) Cell morphology was determined using Quimp image analysis software (Bosgraaf et al., 2009; Dormann et al., 2002) on populations of cells expressing Ras1p, H-Ras, N-Ras and K-Ras4B, in addition to cells lacking Ras1p. Images were obtained using a Leica SP5 scanning confocal microscope equipped with a × 100 oil-immersion objective; scale bar = 10 µm. Data shown are averages of 10 determinants (± SEM). Statistical significance between strains expressing ras proteins and the strain lacking Ras1p was determined. All statistics were performed using one-way ANOVA with Tukey's multiple comparison post-test; *p < 0.05, **p < 0.01 and ***p < 0.01. (B) The localization of Tea1-mCherry was also determined in cells expressing Ras1p, H-Ras, N-Ras and K-Ras4B, in addition to cells lacking Ras1p, to visualize polar cell growth. Tea1-mCherry was visualized using a Personal DeltaVision; scale bar = 10 µm

Download figure to PowerPoint

Human ras signalling at elevated expression levels

Signalling of all four ras proteins was also investigated through expression from the thiamine-repressible nmt1 promoter on the pREP3x vector. Analysis of pheromone-responsive transcription was performed using an sxa2 > lacZ reporter strain lacking endogenous ras1 (JY1279). Wild-type cells expressing Ras1p exhibited a maximal response to pheromone of 26.8 ± 2.2 U after 16 h stimulation (Figure 4A). Cells expressing H-Ras, N-Ras or K-Ras4B from pREP3x in the absence of thiamine all demonstrated elevated signalling, each producing a maximal response in excess of 35 U and displayed varying levels of pheromone sensitivity (pEC50: Ras1p = 6.37 ± 0.22; H-Ras = 7.25 ± 0.25; N-Ras = 6.60 ± 0.15; K-Ras4B = 6.75 ± 0.18). Wild-type cells, when stimulated with pheromone for 16 h, exhibited a maximal cell volume of 91 ± 1.7 fl and pEC50 of 6.02 ± 0.44 (Figure 4B). Cells expressing human H-Ras, N-Ras and K-Ras4B exhibited maximal cell volumes of 118 ± 2.7, 119 ± 2.7 and 121 ± 1.8 fl, respectively. No significant difference in pheromone sensitivity (pEC50: H-Ras = 6.29 ± 0.11; N-Ras = 6.12 ± 0.11; K-Ras4B = 6.04 ± 0.06) was observed; however, morphological changes were detected at lower pheromone concentrations. Cells lacking Ras1p displayed a slight reduction in cell volume in this assay, where they contain empty pREP3x vector, compared to that seen in Figure 2C. This observation suggests that the change in auxotrophy (leu to leu+) had an influence on morphology in this strain, highlighting the need for control strains with matched auxotrophies.

image

Figure 4. Human Ras activity at elevated expression levels. JY544 cells lacking Ras1p were transformed with pREP3x expression vectors containing Ras1p, H-Ras, N-Ras and K-Ras4B. Cells were then grown in the absence of thiamine and with 100 μm to 10 nm pheromone for 16 h. Assays were then performed for β-galactosidase activity (A) and median cell volume (B). Data shown are averages of three independent determinants (± SEM). (C) Cell morphology was determined using Quimp image analysis software (Bosgraaf et al., 2009; Dormann et al., 2002) on populations of cells expressing Ras1p, H-Ras, N-Ras and K-Ras4B, in addition to cells lacking Ras1p. Images were obtained using a Leica SP5 scanning confocal microscope equipped with a × 100 oil-immersion objective; scale bars = 10 µm. Data shown are averages of 10 determinants (± SEM). Statistical significance between strains expressing Ras and the strain lacking Ras1p was determined using one-way ANOVA with Tukey's multiple comparison post-test; **p < 0.01, ***p < 0.01

Download figure to PowerPoint

Quimp quantitative image analysis software was used to determine the morphology of cells expressing H-Ras, N-Ras or K-Ras4B from pREP3x (Figure 4C). Cells lacking Ras1p exhibited an elongation score of 0.14 ± 0.02. Wild-type cells (0.26 ± 0.02) and N-Ras expressing cells (0.23 ± 0.02) demonstrated significant elongation compared to cells lacking Ras1p. Cells expressing H-Ras (0.19 ± 0.02) and K-Ras4B showed an elongated morphology (0.17 ± 0.01) compared to cells lacking Ras1p, although to a lesser extent. These data indicate that increasing the expression levels of these human ras proteins, unlike in the case of pheromone signalling, did not increase signalling output (cell elongation), and in some cases had a negative influence. As a consequence, these data could indicate that a control of ras protein levels and ras signalling levels is more key to the regulation of cell morphology than pheromone response, an effect that is particularly apparent upon elevated expression of human ras isoforms which differ from Ras1p in their localization and potentially their regulation.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Quantification of human ras function in Sz. pombe

Previous studies have used the fission yeast Sz. pombe as a model system for the heterologous expression of signalling components, including human GPCRs (Ladds et al., 2003), Gα subunits (Ladds et al., 2007), Gβ subunits (Goddard et al., 2006) and RGS proteins (Smith et al., 2009; Hill et al., 2008; Ladds et al., 2005). In this present study, we have demonstrated the quantification of human ras function in Sz. pombe. Expression of human H-Ras, N-Ras and K-Ras4B at endogenous Ras1p levels provided some evidence of signalling in Sz. pombe, with partial rescue of elongated cell morphology being observed. In addition, significant signalling was also noted through human H-Ras and K-Ras4B in promoting pheromone-responsive transcription. These data provide the first quantitative analysis of human ras signalling in fission yeast, building upon previous observations of the partial function of human ras in Sz. pombe (Nadin-Davis et al., 1987). Overexpression of all three human ras isoforms led to high levels of pheromone-responsive transcription and rescued shmoo formation. These assays therefore represent a valuable tool for the accurate analysis of multiple ras signalling events, in addition to allowing the analysis of individual ras isoforms in isolation.

Human ras localization in Sz. pombe

In this study we also demonstrate the use of Sz. pombe as a model for studying the localization of human signalling components. Both H-Ras and N-Ras displayed high levels of endomembrane localization. These observations are in agreement with those in higher eukaryotes, where both H-Ras and N-Ras localize to endomembranes (Bivona et al., 2003; Caloca et al., 2003; Chiu et al., 2002; Perez de Castro et al., 2004; Quatela and Philips, 2006). Ras proteins exhibit a very high level of homology in their N-terminal regions, which are responsible for GTP binding and effector interaction. Despite this homology, human ras isoforms display quite distinct, although overlapping, functional interactions (Shields et al., 2000). By contrast, the C-terminus, which is responsible for localization, lacks significant homology (Eisenberg and Henis, 2008), possibly indicating that the localization of these human ras isoforms is important in dictating functional interactions. In addition, the levels of plasma membrane localization appeared to match well with the levels of maximal signalling H-Ras, N-Ras and K-Ras4B displayed through the pheromone-responsive transcription pathway, with N-Ras displaying the lowest level of pheromone activation and K-Ras4B the highest. These data therefore suggest that plasma membrane localization is important to pheromone-responsive signalling in Sz. pombe.

Human H-Ras, N-Ras and K-Ras4B all displayed quite differing localization patterns in Sz. pombe. In addition, these localization patterns did not just appear to be a function of their C-terminal modifications, as N-Ras and Ras1p displayed highly divergent localizations, despite both containing a single palmitoyl and a single farnesyl modification. Previous studies, however, have suggested the importance of regions upstream of these modifications. Ras1p in particular has a very large C-terminus, which could account for the differences in localization observed between N-Ras and Ras1p (Hancock, 2003).

Recent studies have also suggested that the manipulation of ras localization could provide a valuable tool for the downregulation of oncogenic ras signalling (Dekker, et al., 2010; Dekker and Hedberg, 2010). These observations further illustrate the advantages of a simple system, such as Sz. pombe, in which quantification of ras localization can be linked with the accurate analysis of multiple ras signalling outputs.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This study was supported by a Warwick Postgraduate Research Scholarship (to M.B.), the Warwick Research Development Fund (to M.B. and J.D.) and an Engineering and Physical Sciences Research Council (EPSRC) pre-doctoral traineeship (to W.C. and R.T.). G.L. was supported by the University Hospitals of Coventry and Warwickshire NHS Trust, the Warwick Research Development Fund, the Biotechnology and Biological Sciences Research Council (BBSRC; Grant No. BB/G01227X/1) and Advantage West Midlands (Birmingham Science City). We thank Rachel Forfar for providing the ΔRas1p strains and Byron Stokes for assistance with the production of some of the human Ras constructs.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References