Eukaryotic Cell
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Other Versions of this Article:
EC.00204-06v1
6/2/245    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Strmecki, L.
Right arrow Articles by Pears, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Strmecki, L.
Right arrow Articles by Pears, C.
Eukaryotic Cell, February 2007, p. 245-252, Vol. 6, No. 2
1535-9778/07/$08.00+0     doi:10.1128/EC.00204-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Proteomic and Microarray Analyses of the Dictyostelium Zak1-GSK-3 Signaling Pathway Reveal a Role in Early Development{triangledown}

Lana Strmecki,1 Gareth Bloomfield,2 Tsuyoshi Araki,3 Emma Dalton,4 Jason Skelton,5 Christina Schilde,3,{dagger} Adrian Harwood,4 Jeffrey G. Williams,3 Al Ivens,5 and Catherine Pears1*

Biochemistry Department, Oxford University, South Parks Rd., Oxford OX1 3QU, United Kingdom,1 MRC Laboratory of Molecular Biology, Hills Rd., Cambridge CB2 2QH, United Kingdom,2 School of Life Sciences, Wellcome Trust Biocentre, University of Dundee, Dundee DD1 5EJ, United Kingdom,3 Cardiff School of Biosciences, Cardiff University, Museum Ave., Cardiff CF10 3US, United Kingdom,4 The Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1SA, United Kingdom5

Received 28 June 2006/ Accepted 26 October 2006


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
GskA, the Dictyostelium GSK-3 orthologue, is modified and activated by the dual-specificity tyrosine kinase Zak1, and the two kinases form part of a signaling pathway that responds to extracellular cyclic AMP. We identify potential cellular effectors for the two kinases by analyzing the corresponding null mutants. There are proteins and mRNAs that are altered in abundance in only one or the other of the two mutants, indicating that each kinase has some unique functions. However, proteomic and microarray analyses identified a number of proteins and genes, respectively, that are similarly misregulated in both mutant strains. The positive correlation between the array data and the proteomic data is consistent with the Zak1-GskA signaling pathway's functioning by directly or indirectly regulating gene expression. The discoidin 1 genes are positively regulated by the pathway, while the abundance of the H5 protein is negatively regulated. Two of the targets, H5 and discoidin 1, are well-characterized markers for early development, indicating that the Zak1-GskA pathway plays a role in development earlier than previously observed.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
GSK-3 is a multifunctional serine/threonine protein kinase that regulates a large number of key eukaryotic cellular processes, including intermediate metabolism, cytoskeleton maintenance, and development (12). A significant question is how GSK-3 can be active in a broad range of signal pathways yet retain pathway specificity. For example, in the same cell, insulin can control glycogen synthesis and Wnt can control ß-catenin protein levels, both in a GSK-3-dependent manner but without crossover between the two pathways (10). This specificity appears to arise through a number of mechanisms. Phosphorylation by GSK-3 generally requires prior phosphorylation of a substrate by another kinase at a priming site four residues toward the C terminus from the phosphoacceptor site. As a consequence, in most instances, GSK-3 operates in combination with other kinases. Substrate specificity can be determined by the identity and activation profile of the priming kinase, such as in the case of CRMP-2 and CRMP-4 (8). Both are substrates for GSK-3 but can be phosphorylated by GSK-3 at different times, as phosphorylation is dependent on different priming kinases which are not coordinately regulated. In addition, GSK-3 and its priming kinases may be brought into close contact with its substrate through the action of scaffold proteins, such as Axin, a protein which brings together the kinases CK1 and GSK-3 with their substrate ß-catenin. Wnt signaling can disrupt this protein complex to block ß-catenin phosphorylation in a manner that is distinct from the regulation via serine phosphorylation at the N terminus of GSK-3 seen in response to signals such as insulin (24).

A third, but less understood, regulatory mechanism is tyrosine phosphorylation at residue 216 (in GSK-3ß). This phosphorylation event is not required for kinase activity but may influence substrate interaction with the active site (9). Although in most cases, phospho-Tyr216 does not change during GSK-3 regulation, there are a number of situations where it does regulate GSK-3 function (4, 17, 29). A well-characterized example of phosphotyrosine regulation of GSK-3 during the multicellular development of Dictyostelium has been observed (22).

Dictyostelium amoebae exist in a unicellular state while feeding on bacteria, but starvation triggers the formation of a multicellular organism. The aggregation of single cells into a mound is coordinated by the pulsatile release of cyclic AMP (cAMP). cAMP acts as a chemoattractant and as a regulator of the developmental gene expression program, acting through high-affinity cAMP receptors cAR1 and cAR3. In the mound, extracellular [cAMP] rises to millimolar levels, stimulating the low-affinity cAMP receptors cAR2 and cAR4. The process eventually leads to the generation of a terminal structure consisting of a ball of spores supported on a stalk composed of vacuolated cells (34).

Deletion of the gene encoding GskA, the Dictyostelium homologue of GSK-3, leads to ectopic expression of a marker of stalk cell differentiation, ecmB, suggesting that GskA forms part of the repressive signaling pathway that prevents premature stalk cell differentiation (18, 30). Both ecmB expression and stalk cell formation are repressed by extracellular cAMP, whereas the formation of the spore precursor cells (prespore cells) requires cAMP and the loss of GskA or cAR3 can disrupt these effects during the multicellular stages of development (18, 27, 30). A dual-specificity kinase, Zak1 (22), acts downstream of cAR3 and directly phosphorylates GskA on Tyr214 (the equivalent of Tyr216 in the mammalian GSK-3ß). The activation profile of Zak1 coincides with that of GskA, and the peak of GskA activation normally seen around the mound stage of development is lost in a zak1– strain. The balance between Zak1 kinase activity and an unidentified phosphatase activity has been proposed to cause the differential activation of GskA in different cell types, thus regulating the expression of cell type-specific markers (21).

Despite the importance of GskA, only one substrate for this kinase has been identified. The transcription factor STATa is phosphorylated by GskA, and this event leads to its enhanced nuclear export (15). In order to establish the prevalence of the Zak1-GskA regulatory mode, we set out to identify other cellular effectors of the two kinases by searching for changes in the proteomes and the transcriptomes of cells bearing null mutations in the genes encoding Zak1 or GskA. We identify a number of features that are coordinately altered in the two mutant strains, providing potential targets in a unified Zak1-GskA pathway. Some targets identified are genes or proteins whose expression is regulated during the early stages of development, suggesting that the Zak1-GskA pathway has a role at a much earlier stage than previously thought.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell culture and development. Axenic Dictyostelium Ax2 cells were grown at 22°C in HL5 medium (40). The Ax2/gskA-null and statA-null cells have been described previously (25, 30). For development, exponentially growing cells were resuspended in KK2 (16.5 mM KH2PO4, 3.8 mM K2HPO4) at 2 x 107 cells/ml and shaken at 120 rpm and 22°C for 5 h, being pulsed with 5 nM cAMP every 5 min.

Construction of the zak1 null strain in an Ax2 background. The gene encoding Zak1 was disrupted by an in vitro transposition technique (1) using an artificial transposon carrying a blasticidin S resistance cassette for selection in Dictyostelium. The integration position of the transposon was determined by sequencing, and the transposon disrupts the coding sequence corresponding to the second, or DI,kinase domain of Zak1 after amino acid 574. The original zak1– strain also contains a disruption corresponding to this second kinase domain, predicted to be a tyrosine kinase. Transcription could not be detected by real-time (RT)-PCR, but we cannot rule out the possibility that a truncated Zak1 containing a functional N-terminal serine/threonine kinase domain is expressed. Four independent clones showed identical phenotypes, so one was chosen for further analysis and called zak1–Ax2 to distinguish it from the original zak1– strain. When developed on nutrient-free filters, most aggregates arrested at the tipped-mound stage. A few aggregates developed further into slugs, and a small number of multitipped structures were apparent after 48 h of development, while the development of control strains into fruiting bodies was complete after 24 h. The observed aberrant development is consistent with the previously described zak1– phenotype, produced in a different genetic background (22).

Western blotting and immunoprecipitation. Whole-cell extracts were prepared by lysing 1.5 x 107 cells in 500 µl of ice-cold buffer (150 mM NaCl, 50 mM Tris-HCl, 2 mM EDTA, 1% Triton X-100, 1 mM Na3VO4, 50 mM NaF) and cleared by centrifugation at 12,000 x g for 10 min. Thirty micrograms was run on a NuPAGE bis-Tris 4 to 12% gel (Invitrogen). Following transfer, membranes were incubated with 4G10 anti-pTyr antibody (Upstate Biotechnology) and visualized by chemiluminescence.

Nuclear cell extracts were prepared by filtering 3 x 107 cells in 2 ml of ice-cold modified buffer [10 mM Tris at pH 8, 2 mM EDTA, 5 mM Na3VO4, 0.2 mM 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride, 1 mM NaF, Mg-free protease inhibitor tablet (Roche)] through Nuclepore Track-Etch membranes (Whatman) (method adapted from reference 20). Nuclei were collected by centrifugation at 3,000 x g for 2 min and lysed in 750 µl of nuclear lysis buffer [300 mM NaCl, 2.5 mM MgCl2, 2.5 mM Mg(OAc)2, 25 mM Tris (pH 7.4), 5 mM HEPES (pH 7.9), 0.05 mM EDTA, 5 mM Na3VO4, 0.2 mM 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride, 1 mM NaF, Mg-free protease inhibitor tablet (Roche)].

Two-dimensional gel electrophoresis. Aliquots containing 100 µg of soluble whole-cell extracts were acetone precipitated and resuspended in 125 µl of sample buffer {5 M urea, 2 M thiourea, 4% (wt/vol) 3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate (CHAPS), 4% (wt/vol) NSDB-256 (dimethylbenzylammonium propane sulfonate), 1% (wt/vol) TBP (tributylphosphine), 1% (wt/vol) dithiothreitol, 10 mM benzamidine, 1 mM sodium orthovanadate, 1 mM sodium fluoride, and a trace of bromophenol blue}. The samples were loaded onto nonlinear immobilized pH gradient strips (pH range, 3 to 11; Amersham), and isoelectric focusing was performed on a MULTIPHOR II apparatus (Amersham) as recommended by the manufacturer. Equilibration of isoelectric focusing strips was performed for 10 min in equilibration buffer (4 M urea, 2 M thiourea, 2% [wt/vol] dithiothreitol, 2% [vol/vol] sodium dodecyl sulfate [SDS], 0.05 M Tris [pH 6.8], 30% [vol/vol] glycerol, and a trace of bromophenol blue). The second dimension was performed by standard gel electrophoresis on NuPAGE bis-Tris 4 to 12% ZOOM gels (Invitrogen). The gels were subsequently stained using colloidal blue (Invitrogen). Spots were analyzed using ImageMaster 2-DGE Platinum software (Amersham).

Protein identification. Protein mass fingerprint data were obtained by matrix-assisted laser desorption ionization-tandem time of flight (tandem mass spectrometry [MS-MS]) analysis performed at the University of Dundee "Fingerprints" Proteomics Facility using an Applied Biosystems (AB) 4700 proteomic analyzer. Excised protein spots from two-dimensional electrophoresis analysis were prepared and in-gel digested by trypsin (Roche; modified sequencing grade) as previously described (38). One-tenth of each digest was then applied to a 192-well matrix-assisted laser desorption ionization sample plate (AB), allowed to air dry, and then supplemented with 0.5 µl of a 5-mg/ml solution of {alpha}-cyano-4-hydroxy-trans-cinnamic acid matrix (Sigma) plus 10 mM ammonium dihydrogen phosphate in 50% (vol/vol) acetonitrile in 0.1% (vol/vol) trifluoroacetic acid, mixed and allowed to air dry prior to analysis. The mass spectrometer was internally calibrated using the AB 4700 proteomic analyzer calibration mix. Using the 4000 series Explorer software (AB), MS spectral data were acquired from the samples and an MS-MS list was automatically generated for further analysis based on the top five most intense ions present (trypsin and major keratin ions were excluded). The MS and MS-MS spectral data obtained were exported from the 4700 proteomic analyzer by using the global proteome server Explorer software (AB). The data were then submitted to a local Mascot search engine for comparison against entries in the NCBInr and Dictyostelium databases for identification. Methionine oxidation and cysteine carbamidomethylation modifications were allowed for with a peptide mass tolerance of 50 ppm and one missed cleavage.

Northern transfer analyses. Total RNA was extracted from approximately 107 cells by using a TRIzol RNA extraction kit (Sigma) according to the manufacturer's protocol. Samples (10 µg) of total RNA were separated on a 1% formaldehyde-containing gel, blotted, and probed by using standard methods (19).

QRT-PCR. Total RNA was treated with 2 U/µg DNase I (Promega), which was subsequently deactivated by heating at 75°C for 10 min. Quantitative RT-PCR (QRT-PCR) was performed in two steps using an ABsolute MAX QRT-PCR kit (ABgene) according to the manufacturer's protocol. Extensive Western blotting and Northern analysis had revealed that cadA expression levels were not altered in the zak1–Ax2 and gskA– cells, which was confirmed by the analysis of microarray data. Therefore, levels of cadA were used as an internal control.

Genome-wide expression profiling. Total RNA was extracted from three independent samples of control, zak1–Ax2, and gskA– cells which had been developed in shaken suspensions for 5 h with nanomolar pulses of cAMP. Each sample was primed with oligo(dT) and separately labeled with Cy3 and Cy5 by using Superscript III reverse transcriptase (Invitrogen). Each set of labeled mutant cDNA was paired with the control cDNA labeled with the complementary fluorophore, and the mixture was hybridized to a DNA microarray. Two pairs of replicate samples of the two mutants were also compared directly. Hybridization and data analysis with three biological replicates were carried out as previously described (7). Arrays were scanned using an Axon Instruments GenePix 4000B scanner, and fluorescence was quantified using the GenePix 3.0 software. Subsequent data processing steps were carried out using the limma package, part of the Bioconductor project, using the R statistical environment (14, 32, 33). Background fluorescence was subtracted using the method of Kooperberg et al. (23), and data were then normalized using the print-tip loess algorithm. Differences in the levels of gene expression, and their significance, among strains were assessed using linear models and empirical Bayes methods. After adjusting to correct for multiple comparisons by using the method of Benjamini and Hochberg (3), and after ranking, genes with a P value of less than 0.05 were provisionally accepted as having altered expression in the mutant cell lines.

Microarray data accession numbers. The microarray data have been deposited in Array Express with the following accession numbers: array, A-SGRP-3; experiment, E-SGRP-4.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Tyrosine phosphorylation events induced by cAMP in gskA– and zak1–Ax2 strains. We first disrupted the zak1 gene in Ax2 cells in order to allow comparison with an existing gskA– strain that was generated in the Ax2 background (30). The resulting zak1-null mutants in the Ax2 background (zak1–Ax2 cells) have a phenotype similar to that reported for a different (Ax4) genetic background (22) (see Materials and Methods). The Zak1 and GskA kinases are regulated by millimolar concentrations of extracellular cAMP at the mound stage of development, and consistent with this pattern, the two mutants develop relatively normally until mound formation (ca. 8 to 10 h) (data not shown). We therefore chose to study cells at a nominally earlier stage than the mound stage: by starving cells in shaking suspensions for 5 h. In a further attempt to ensure that comparisons were made when the strains were at similar developmental stages, we chose to drive the developmental program by pulsing with exogenous nanomolar cAMP.

We first analyzed the known components of the signaling system. After 5 h of pulsing with low-level cAMP, cells were exposed to 5 mM cAMP for 1 min, conditions expected to induce tyrosine phosphorylation of STATa (2). For the control strain and both the mutant strains, there was an increase after 1 min in the phosphotyrosine content of a band migrating at the position predicted for STATa (Fig. 1A). The use of an antibody specific for the tyrosine-phosphorylated form of STATa (data not shown) and the loss of this band in the STATa-null (dstA–) cells (Fig. 1B) confirmed the identity of the band as STATa. Thus, all three strains were competent to respond to cAMP by activating STATa.


Figure 1
View larger version (71K):
[in this window]
[in a new window]

  		FIG. 1. Tyrosine phosphorylation in response to extracellular cAMP. (A) Exponentially growing parental control, zak1–Ax2, and gskA– cells were harvested, washed twice, and resuspended in KK2 at 2 x 107 cells/ml. Cells were developed for 5 h in shaken suspensions at 120 rpm and 22°C and pulsed with 5 nM cAMP every 5 min. Cells were then lysed in LDS sample buffer (Invitrogen) following a 1-min incubation with or without the addition of 5 mM cAMP. Western blot analysis was carried out using anti-phosphotyrosine antisera 4G10. The band believed to correspond to tyrosine-phosphorylated STATa is marked. (B) Phosphotyrosine proteins induced by cAMP in control and STATa-null (dstA–) cells were analyzed as described above. (C) Following the development of control, zak1–Ax2, and gskA– cells as described above, phosphotyrosine-containing proteins were immunoprecipitated from nuclear extracts by using 4G10 monoclonal antibody. The immunoprecipitated proteins were then resolved by SDS-polyacrylamide gel electrophoresis and subjected to Western blotting using 4G10. The band believed to correspond to tyrosine-phosphorylated Zak1 is marked.

 
A tyrosine-phosphorylated band of approximately 52 kDa was missing from gskA– cells, and this is the molecular mass expected for GskA. There was no apparent increase in the level of tyrosine phosphorylation of this protein following treatment with 5 mM cAMP in control cells, presumably because the pulsing-shaking conditions led to the activation of tyrosine kinases that modify GskA. Zak1 appears not to be the only kinase responsible for the tyrosine phosphorylation of the band we hypothesize to be GskA because it was tyrosine phosphorylated in the zak1–Ax2 null mutant. Interestingly, a close homologue of Zak1 is present in the genome (16). This other kinase, known as DPYK4, could be responsible for the phosphorylation of GskA during these earlier stages of development.

In Western analysis of zak1–Ax2 cells, there was no obvious loss of a tyrosine-phosphorylated protein with the molecular mass predicted for Zak1. However, this reflects the low abundance of Zak1 because, upon enrichment with phosphotyrosine-containing protein by immunoprecipitation from nuclear extracts with an anti-phosphotyrosine antibody, a difference between the control and zak1–Ax2 cells was apparent. One phosphotyrosine-containing band was missing in the zak1–Ax2 cells (Fig. 1C), and the molecular mass of this band was consistent with its being a tyrosine-phosphorylated form of Zak1. The total level of tyrosine phosphorylation of this nuclear band was also unchanged upon treatment of control and gskA– cells with 5 mM cAMP.

Changes in the proteome induced by loss of Zak1 and GskA function. In order to identify transcriptionally and nontranscriptionally regulated, high- to medium-abundance targets in the Zak1 and GskA signaling pathway, we first performed proteomic analysis. We compared the patterns of features detectable on two-dimensional gels bearing whole-cell extracts from control, zak1–Ax2, and gskA– cells. Samples were harvested following 5 h of development in shaken suspensions with exposure to pulses of cAMP. Representative pairs of two-dimensional gels are shown in Fig. 2A. The majority of features show equal staining intensities in all three strains, and a number of these constant spots were used for normalization. Several features were reproducibly altered in one or both of the mutant strains (Fig. 2 and 3 and Table 1),and their identities were determined by mass spectrometry.


Figure 2
View larger version (48K):
[in this window]
[in a new window]

  		FIG. 2. Two-dimensional gel analysis of zak1–Ax2 and gskA– cells, developed for 5 h. (A) Whole-cell extracts of control, gskA–, and zak1–Ax2 cells, developed in shaken suspensions for 5 h with 5 nM pulses of cAMP, were resolved by two-dimensional gel electrophoresis followed by staining with colloidal Coomassie blue. Representative pairs of gels from control and mutant cells, run simultaneously, are shown and the features reproducibly changed in mutant samples relative to controls are identified. (B) The relative intensities of staining for each feature that reproducibly changed were averaged for three gels in comparison to the intensity of the equivalent spot from control gels (defined as 1). The features are named according to the genes encoding the proteins identified by mass spectrometry (see dictybase.org and Table 1). In one case, no gene name is available and the identifying DDB number has been used (01 for DDB0187880). The statistical significance of differences between means was determined by one-way analysis of variance. If means were shown to be significantly different, multiple comparison by pairs was performed by using Tukey's test. Probability values of <0.05 were selected to indicate statistical significance and are marked with an asterisk. One of the features which showed different intensities on the two-dimensional gels for gskA– cells was DdCAD-1. This is an abundant and relatively well-studied protein from Dictyostelium for which there are a range of tools available (6). The relevant spot was absent in two-dimensional gel analysis of cad1– cells (42), confirming its identity (data not shown). Despite the differences apparent on two-dimensional gels, Northern blot analysis revealed no detectable differences in cad1 mRNA levels in the three strains (data not shown). Consistent with the equal mRNA levels, antisera against DdCAD-1 demonstrated no differences in total DdCAD-1 protein levels when DdCAD-1 from all three strains was extracted into SDS buffer on one-dimensional gels (data not shown). The change in the intensity of the feature corresponding to DdCAD-1 on two-dimensional gels could therefore represent a posttranscriptional modification or could be due to different solubility of DdCAD-1 in the gskA– cells.

 

Figure 3
View larger version (46K):
[in this window]
[in a new window]

  		FIG. 3. Representative features. (A) Noncoordinate regulation. Cropped images showing a control and two representative gels from mutant strains showing features which are not coordinately regulated in the two mutant strains. AlrA is up-regulated in zak1–Ax2 cells but down-regulated in gskA– cells, and DDB0187880 (DDB01) is down-regulated only in gskA– cells. (B) Coordinate regulation. Cropped images showing a control and two representative gels from mutant strains showing features which are coordinately regulated in the two mutant strains. H5 protein is present at higher levels (Bii) in the mutant strains than in the control strain, while DD7-1 is underrepresented (Bi) in both mutant strains. The feature labeled DD7-1 contains peptide sequences which correspond to this gene and a second, nearly identical, gene (DDB0190881) which lies directly adjacent to DD7-1 on chromosome 1. This seems likely to be the result of a recent gene duplication, and mass spectrometry could not distinguish between the two genes. The protein encoded by these two genes shows high homology to discoidin 1. (C) Quantitative real-time PCR was used to determine the relative levels of mRNA from the alrA and cinB genes under developmental conditions equivalent to those used to isolate the protein for two-dimensional gel analysis. The level in control cells (WT) was defined as 1 and the increase (n-fold) relative to this level in gskA– and zak1–Ax2 cells is shown. The averages of results from three independent experiments are shown with standard errors of the means. The change in alrA expression in zak1–Ax2 cells was shown to be statistically significant (P < 0.05) by a single sample t test (*). All other samples showed the same trend in that features were expressed at higher levels in gskA– and zak1–Ax2 cells than in control cells, although the biological variation among samples is such that the differences are just outside statistical significance (P < 0.08 for the other three).

 

View this table:
[in this window]
[in a new window]

  		TABLE 1. Features showing changes upon two-dimensional gel analysis of gskA– and zak1–Ax2 cells relative to control cells

 
(i) Noncoordinately regulated proteins. A number of proteins were misregulated in only one or the other of the two mutants, e.g., the metabolic enzyme aldoketo reductase (alrA) was underrepresented in the gskA– samples and overrepresented in the zak1–Ax2 samples (Fig. 2 and 3A).

(ii) Coordinately regulated proteins. The metabolic enzyme transketolase was down-regulated in the two mutants, as was the product of the DD7-1 gene, which encodes a homologue of the discoidin 1 proteins (Fig. 2 and 3B). In contrast, vegetative protein H5 (the product of the cinB gene) was up-regulated in both zak1–Ax2 and gskA– cells (Fig. 3B). Changes in spot intensities observed on protein gels could be the result of transcriptional or posttranscriptional events. Quantitative real-time PCR showed that the message encoding H5 was always present at higher levels in both mutant strains than in control cells, although there was considerable variation in the level of increase among biological samples. These data are consistent with the idea that the alteration in H5 protein levels is due at least in part to transcriptional changes (Fig. 3C).

Interestingly, although the AlrA protein level was significantly reduced in the gskA– cells, quantitative PCR revealed that the alrA mRNA level was not reduced and, if anything, may have actually been higher in both mutant strains (Fig. 3C). This finding suggests that GskA alone is responsible for a posttranscriptional event that leads to an overall reduction of the amount of AlrA protein in the spot in its absence.

Transcriptional targets regulated by Zak1 and GskA. In order to extend the analysis down to lower abundance limits, and also to investigate regulation at the RNA level, expression profiles of zak1–Ax2 and gskA– cells were compared with those of control cells. The microarray bears PCR products from approximately 8,600 genes, identified from the complete Dictyostelium genome sequence. Only genes for which unique PCR primer sets could be predicted were included. Gene prediction suggests that the genome contains around 12,000 genes. Subsequent detailed analysis has reduced the predicted number by around 1,500 genes by removing those derived from retrotransposons, pseudogenes, and those coding for very small predicted proteins (26). The array therefore likely represents around 80% of the total number of Dictyostelium genes.

The great majority of genes showed no difference between control and mutant strains. However, a number of genes were aberrantly expressed in similar ways in both zak1–Ax2 and gskA– cells (Fig. 4A and Table 2) and are candidates for transcriptional or posttranscriptional regulation in response to a GskA-Zak1 signaling pathway. The majority of genes (31 out of 37) coordinately altered in their expression in the two mutant strains were down-regulated, suggesting that, in general, the Zak1-GskA signaling pathway has a positive effect on specific gene transcription events.


Figure 4
View larger version (19K):
[in this window]
[in a new window]

  		FIG. 4. Microarray analysis of zak1–Ax2 and gskA– cells after 5 h of development in shaking suspensions. (A) Heatmap showing relative levels of expression of genes differentially regulated in zak1–Ax2 and gskA– cells relative to control cells. Genes with an adjusted P value of less than 0.05 in the comparison of either null strain with control cells were clustered according to the Euclidean distance between their expression levels and the average agglomeration method using the R package hclust. Expression values are represented (in rows) for each gene in each strain by color on a scale from blue (underexpressed in the mutant) through white (unchanged) to red (overexpressed) by using the heatmap.2 function of the R package gplots (39). (B) The levels of the mRNAs corresponding to the proteins identified as being differentially expressed in the zak1–Ax2 and gskA– cells by proteomic analysis were determined from the microarray data, relative to the levels in control cells (defined as 1). No data are available for DD7-1 as it was not present on the array. Genes for which the P values for the array data were ≤0.05 are marked with an asterisk. 01, DDB0187880; Ref, reference value.

 

View this table:
[in this window]
[in a new window]

  		TABLE 2. Numbers of genes misregulated in zak1–Ax2 and gskA– cells after 5 h of development in a shaken suspension

 
A total of 24 of the coordinately altered genes have functions that are known or can be reliably inferred. The genes that were differentially expressed in both mutant strains, whether they were up-regulated or down-regulated, encode proteins with a range of cellular functions (Table 3). Nine of the 24 are involved in metabolism, and six of the others have developmental functions. The microarray data were used to investigate the levels of mRNA encoding each of the proteins identified as being altered in the mutant strains in the proteomic analysis (Fig. 4B). There is, for most of the gene products, qualitative agreement between the microarray data and the proteomic data, but the mutant expression level/control expression level ratio is generally lower with the array data. If we take a standard cutoff unadjusted P value of 0.05 for the array data, then there is significant agreement between the results of the proteomic and array analyses for H5 (cinB), tkt, and DDB0187880 in gskA– and alrA in zak1–. In most other cases, there is qualitative agreement but the P value for the three experiments is greater than 0.05. In only one case is there significant disagreement between the proteomic and the array data. In the zak1–Ax2 strain, DDB0187880 was underexpressed according to the array data but was not seen to change on two-dimensional gels. Apart from this case, the fact that there is a good correlation between the changes in protein and mRNA levels is consistent with transcription being the major target of the Zak1-GskA pathway at this stage in development.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Functional analysis of genes showing coordinate alteration in gene expression in zak1–Ax2 and gskA– cellsa

 
The microarray analysis identified the family of discoidin 1 genes as being coordinately altered in zak1–Ax2 and gskA– cells. We confirmed this array result by Northern transfer, using conditions that detect all three discoidin 1 genes (Fig. 5). As predicted by the microarray analysis, the discoidin 1 genes were underexpressed in both the mutant strains during the first 6 h of development.


Figure 5
View larger version (37K):
[in this window]
[in a new window]

  		FIG. 5. Northern analysis of discoidin gene expression in zak1–Ax2 and gskA– cells. Control, zak1–Ax2, and gskA– cells were harvested, washed, and plated for development on filters. Cells were harvested at the times shown, and RNA was extracted and subjected to Northern analysis using probes specific for discoidin 1 (dsc) before stripping and reprobing with IG7 as a loading control.

 
The expression levels of the genes encoding both H5 (identified in the proteomic analysis) and discoidin 1 are commonly used markers of early development in Dictyostelium. We therefore used existing microarray data (36; see dictybase.org) to examine the expression profiles of all the genes identified by both analyses to see if any of the other potential targets also showed major changes in gene expression during early development (first 4 h). Data were available for all of the targets identified proteomically, and the expression of 8 out of 10 targets was altered by at least twofold during the first 4 h of development (Table 1). Expression profile data were not available for all of the targets identified by the microarray, but among the 21 genes for which data were available, 10 showed at least a twofold change in expression levels during this early stage of development (Table 3). Taken together, these findings support the conclusion that the unified Zak1-GskA pathway plays a role in regulating gene expression during the early stages of development, prior to the mound stage as previously thought.


   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Common and distinct targets for GskA and Zak1. The proteomic and the array analyses identified common targets of the two kinases GskA and Zak1, the common targets being defined as mRNAs or protein features which showed qualitatively similar patterns of misregulation in both mutants. It is not possible to determine whether the changes seen were direct or indirect effects of the loss of Zak1 or GskA. The relatively small number of changes identified and the fact that the analysis was carried out with cells prior to overt phenotypic alterations of the mutant strains might mean that this analysis may include direct targets. However, elucidation of the pathway linking Zak1-GskA to a single target would be necessary to demonstrate a direct link. This identification of a number of coregulated proteins and mRNAs significantly strengthens the case for a common pathway. However, both methods also identified targets specific to one or the other kinase. These two observations are entirely reconcilable with published data on the two mutant phenotypes. There are similarities between zak1– and gskA– cells, most notably in the loss of stalk cell repression by cAMP, but the developmental phenotypes of the null strains are not identical. Thus, in an Ax2 background zak1–Ax2 strains fail to complete development on filters, whereas gskA– cells culminate. Conversely, gskA– cells show a slightly increased rate of aggregation while zak1– cells aggregate normally. The independent effects of GskA suggest either that constitutive GskA activity plays a role in early development or that GskA activation can be regulated by other means. The small increase in GskA activity detected during the development of zak1– cells and the residual GskA tyrosine phosphorylation still detected in zak1– cells (21, 22) would support the latter conclusion.

Zak1 and GskA are required for correct regulation of transcription of many genes. The proteomic changes could, in principle, be due to transcriptional or posttranscriptional regulation, but the array data confirm that there are changes at the RNA level. The quantitative PCR data and the behavior of the genes encoding the targets identified on two-dimensional gels in the microarray analysis suggest that the majority of changes identified proteomically are transcriptional.

In the microarray analysis, a total of 168 changes were apparent in zak1–Ax2 cells, suggesting that Zak1 is active during the early stage of development. All the phenotypic changes described for zak1– cells were postaggregation, but the changes discovered here raise the possibility that these later phenotypic changes could be a consequence of earlier alterations in gene activity, rather than direct effects on late targets.

We identified around 150 genes whose expression at the RNA level is either up-regulated or down-regulated in gskA– cells. Again, these differences occur at an early stage of development, prior to overt cell type differentiation. A previous array analysis of gskA– cells, at the slug stage of development, identified a much smaller number of changes (30). However, this study utilized fewer genes, the mutant's developmental aberration made it possible to identify only transcripts which were under-expressed in the mutant, and the study imposed a threshold cutoff rather than employing a probabilistic analysis.

The array analysis identified 37 common targets for GskA and Zak1. Interestingly, most of the common targets of the Zak1-GskA pathway are down-regulated rather than up-regulated, suggesting that the pathway normally plays a positive role in controlling gene expression. However, as many of the targets identified may be indirect, this remains to be verified. The transcription factor(s) responsible has yet to be identified, but the above data suggest that, in most cases, an activator of transcription is turned on or an inhibitor is switched off. The only known target for GskA in Dictyostelium is STATa, and it is known to function as both a transcriptional repressor of ecmB (25) and an activator of cudA expression (13). Thus, the enhanced nuclear export of STATa, following its phosphorylation by GskA, could in principle facilitate the up- or down-regulation of some of the genes discovered here.

Although the cAMP receptors cAR3 and cAR4 have been implicated in GskA regulation later in development, it is not known which extracellular factors may be responsible for Zak1 or GSK-3 regulation at earlier stages. In this regard, it is of interest that the AlrA protein is misregulated in both mutant strains. AlrA has been previously identified on two-dimensional gels as a protein misregulated in cells with mutants of counting factor, a factor involved in regulating aggregate size in early development (35). Disruption of alrA causes development to arrest at the tipped-mound stage, similar to the zak1– phenotype (11), consistent with a link between Zak1 and AlrA.

The common pathway regulates gene expression in early development. H5 (CinB) is an esterase/lipase/thioesterase domain-containing protein used as a marker for growth as the abundance of its mRNA decreases as cells arrest growth and enter development (31). The proteomic analysis showed that the abundance of H5 was higher in both mutants than in controls, and the array and quantitative PCR analysis suggested that this difference was at the level of gene expression. In contrast, the discoidin 1 genes were underexpressed in the two mutants. These observations can be accommodated into the model of a unitary signaling pathway because the discoidin 1 genes, in contrast to the H5 gene, are activated rather than repressed during early development.

The discoidin 1 family is one of the best-characterized gene families, first expressed as cells grow to a high density in axenic medium and then further activated during early development. Discoidin 1 expression is induced in response to two density-sensing factors, conditioned medium factor and prestarvation factor, and is later repressed by pulses of extracellular cAMP during aggregation (5, 41). The promoter of the discoidin 1c gene has been dissected into separate regions that are needed for prestarvation factor-inducible expression and cAMP repression (28, 37).

In summary, the targets H5 and discoidin 1 both indicate that Zak1 and GskA have a role in transcriptional regulation at a much earlier stage in development than was previously thought. The discoidin 1 genes are a particularly attractive entry point for dissecting the pathway further because there is a very considerable body of information on their regulation.


   ACKNOWLEDGMENTS
 
Special thanks to Rob Kay for his invaluable input into the array, C. H. Siu for the anti-Cad1 antisera and cadA– cells, and the proteomics facility in Dundee for the mass spectrometry. We are indebted to our colleagues at dictyBase.

This work was supported by The Wellcome Trust (grant no. 063612). C.S. and T.A. are supported by Wellcome Trust grant no. 053640/Z to J.G.W. The microarray work (G.B., J.S., and A.I.) was also supported by funding from The Wellcome Trust (grant reference no. 064724).


   FOOTNOTES
 
* Corresponding author. Mailing address: Biochemistry Dept., Oxford University, South Parks Rd., Oxford OX1 3QU, United Kingdom. Phone: 44 1865 275737. Fax: 44 1865 275259. E-mail: pears{at}bioch.ox.ac.uk. Back

{triangledown} Published ahead of print on 3 November 2006. Back

{dagger} Present address: Universität Konstanz, FB Biologie, Universitätsstrasse 10, 78457 Konstanz, Germany. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Abe, T., J. Langenick, and J. G. Williams. 2003. Rapid generation of gene disruption constructs by in vitro transposition and identification of a Dictyostelium protein kinase that regulates its rate of growth and development.Nucleic Acids Res. 31:e107.[Abstract/Free Full Text]
  2. Araki, T., M. Gamper, A. Early, M. Fukuzawa, T. Abe, T. Kawata, E. Kim, R. A. Firtel, and J. G. Williams.1998 . Developmentally and spatially regulated activation of a Dictyostelium STAT protein by a serpentine receptor.EMBO J. 17:4018-4028.[CrossRef][Medline]
  3. Benjamini, Y., and Y. Hochberg. 1995. Controlling the false discovery rate—a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B 57:289-300.
  4. Bhat, R. V., J. Shanley, M. P. Correll, W. E. Fieles, R. A. Keith, C. W. Scott, and C. M. Lee. 2000. Regulation and localization of tyrosine216 phosphorylation of glycogen synthase kinase-3beta in cellular and animal models of neuronal degeneration. Proc. Natl. Acad. Sci. USA 97:11074-11079.[Abstract/Free Full Text]
  5. Blusch, J., S. Alexander, and W. Nellen. 1995. Multiple signal transduction pathways regulate discoidin I gene expression in Dictyostelium discoideum. Differentiation 58:253-260.[CrossRef][Medline]
  6. Brar, S. K., and C. H. Siu. 1993. Characterization of the cell adhesion molecule gp24 in Dictyostelium discoideum. Mediation of cell-cell adhesion via a Ca(2+)-dependent mechanism. J. Biol. Chem. 268:24902-24909.[Abstract/Free Full Text]
  7. Chubb, J. R., G. Bloomfield, Q. Xu, M. Kaller, A. Ivens, J. Skelton, B. M. Turner, W. Nellen, G. Shaulsky, R. R. Kay, W. A. Bickmore, and R. H. Singer.2006 . Developmental timing in Dictyostelium is regulated by the Set1 histone methyltransferase. Dev. Biol. 292:519-532.[CrossRef][Medline]
  8. Cole, A. R., F. Causeret, G. Yaeirgi, C. J. Hastie, H. McLauchlan, E. J. McManus, F. Hernandez, B. J. Eickholt, M. Nikolic, and C. Sutherland. 2006. Distinct priming kinases contribute to differential regulation of collapsin response mediator proteins by glycogen synthase kinase-3 in vivo. J. Biol. Chem. 281:16591-16598.[Abstract/Free Full Text]
  9. Dajani, R., E. Fraser, S. M. Roe, N. Young, V. Good, T. C. Dale, and L. H. Pearl. 2001. Crystal structure of glycogen synthase kinase 3 beta: structural basis for phosphate-primed substrate specificity and autoinhibition.Cell 105:721-732.[CrossRef][Medline]
  10. Ding, V. W., R. H. Chen, and F. McCormick.2000 . Differential regulation of glycogen synthase kinase 3beta by insulin and Wnt signaling. J. Biol. Chem. 275:32475-32481.[Abstract/Free Full Text]
  11. Ehrenman, K., G. Yang, W. P. Hong, T. Gao, W. Jang, D. A. Brock, R. D. Hatton, J. D. Shoemaker, and R. H. Gomer. 2004. Disruption of aldehyde reductase increases group size in Dictyostelium.J. Biol. Chem. 279:837-847.[Abstract/Free Full Text]
  12. Frame, S., and P. Cohen. 2001. GSK3 takes centre stage more than 20 years after its discovery. Biochem. J. 359:1-16.[CrossRef][Medline]
  13. Fukuzawa, M., and J. G. Williams. 2000. Analysis of the promoter of the cudA gene reveals novel mechanisms of Dictyostelium cell type differentiation.Development 127:2705-2713.[Abstract]
  14. Gentleman, R. C., V. J. Carey, D. M. Bates, B. Bolstad, M. Dettling, S. Dudoit, B. Ellis, L. Gautier, Y. Ge, J. Gentry, K. Hornik, T. Hothorn, W. Huber, S. Iacus, R. Irizarry, F. Leisch, C. Li, M. Maechler, A. J. Rossini, G. Sawitzki, C. Smith, G. Smyth, L. Tierney, J. Y. Yang, and J. Zhang. 2004. Bioconductor: open software development for computational biology and bioinformatics. Genome Biol. 5:R80.[CrossRef][Medline]
  15. Ginger, R. S., E. C. Dalton, W. J. Ryves, M. Fukuzawa, J. G. Williams, and A. J. Harwood.2000 . Glycogen synthase kinase-3 enhances nuclear export of a Dictyostelium STAT protein. EMBO J. 19:5483-5491.[CrossRef][Medline]
  16. Goldberg, J. M., G. Manning, A. Liu, P. Fey, K. E. Pilcher, Y. Xu, and J. L. Smith. 2006. The Dictyostelium kinome—analysis of the protein kinases from a simple model organism. PLoS Genet. 2:e38.[CrossRef][Medline]
  17. Harwood, A. J. 2000. Signal transduction: life, the universe and development. Curr. Biol. 10:R116-R119.[CrossRef][Medline]
  18. Harwood, A. J., S. E. Plyte, J. Woodgett, H. Strutt, and R. R. Kay. 1995. Glycogen synthase kinase 3 regulates cell fate in Dictyostelium. Cell 80:139-148.[CrossRef][Medline]
  19. Huang, H. J., D. Takagawa, G. Weeks, and C. Pears.1997 . Cells at the center of Dictyostelium aggregates become spores. Dev. Biol. 192:564-571.[CrossRef][Medline]
  20. Insall, R., and R. R. Kay. 1990. A specific DIF binding protein in Dictyostelium. EMBO J. 9:3323-3328.[Medline]
  21. Kim, L., A. Harwood, and A. R. Kimmel. 2002. Receptor-dependent and tyrosine phosphatase-mediated inhibition of GSK3 regulates cell fate choice. Dev. Cell 3:523-532.[CrossRef][Medline]
  22. Kim, L., J. Liu, and A. R. Kimmel. 1999. The novel tyrosine kinase ZAK1 activates GSK3 to direct cell fate specification. Cell 99:399-408.[CrossRef][Medline]
  23. Kooperberg, C., T. G. Fazzio, J. J. Delrow, and T. Tsukiyama. 2002. Improved background correction for spotted DNA microarrays. J. Comput. Biol. 9:55-66.[CrossRef][Medline]
  24. McManus, E. J., K. Sakamoto, L. J. Armit, L. Ronaldson, N. Shpiro, R. Marquez, and D. R. Alessi. 2005. Role that phosphorylation of GSK3 plays in insulin and Wnt signalling defined by knockin analysis. EMBO J. 24:1571-1583.[CrossRef][Medline]
  25. Mohanty, S., K. A. Jermyn, A. Early, T. Kawata, L. Aubry, A. Ceccarelli, P. Schaap, J. G. Williams, and R. A. Firtel. 1999. Evidence that the Dictyostelium Dd-STATa protein is a repressor that regulates commitment to stalk cell differentiation and is also required for efficient chemotaxis. Development 126:3391-3405.[Abstract]
  26. Olsen, R. 2005. How many protein encoding genes does Dictyostelium have?, p.265 -278. In W. F. Loomis and A. Kuspa (ed.), Dictyostelium genomics. Horizon Bioscience, Wymondham, United Kingdom.
  27. Plyte, S. E., E. O'Donovan, J. R. Woodgett, and A. J. Harwood. 1999. Glycogen synthase kinase-3 (GSK-3) is regulated during Dictyostelium development via the serpentine receptor cAR3. Development 126:325-333.[Abstract]
  28. Salger, K., and B. W. Wetterauer. 2000. Aberrant folate response and premature development in a mutant of Dictyostelium discoideum. Differentiation 66:197-207.[CrossRef][Medline]
  29. Sayas, C. L., A. Ariaens, B. Ponsioen, and W. H. Moolenaar. 2006. GSK-3 is activated by the tyrosine kinase Pyk2 during LPA1-mediated neurite retraction. Mol. Biol. Cell 17:1834-1844.[Abstract/Free Full Text]
  30. Schilde, C., T. Araki, H. Williams, A. Harwood, and J. G. Williams. 2004. GSK3 is a multifunctional regulator of Dictyostelium development. Development 131:4555-4565.[Abstract/Free Full Text]
  31. Singleton, C. K., C. E. McPherson, and S. S. Manning. 1988. Deactivation of gene expression upon the onset of development in Dictyostelium discoideum.Dev. Genet. 9:327-335.[CrossRef][Medline]
  32. Smyth, G. 2004. Linear models and empirical Bayes methods for assessing differential expression in microarray experiments.Stat. Appl. Genet. Mol. Biol. 1:article 3.
  33. Smyth, G. K., J. Michaud, and H. S. Scott.2005 . Use of within-array replicate spots for assessing differential expression in microarray experiments.Bioinformatics 21:2067-2075.[Abstract/Free Full Text]
  34. Strmecki, L., D. M. Greene, and C. J. Pears.2005 . Developmental decisions in Dictyostelium discoideum. Dev. Biol. 284:25-36.[CrossRef][Medline]
  35. Tang, L., T. Gao, C. McCollum, W. Jang, M. G. Vicker, R. R. Ammann, and R. H. Gomer. 2002. A cell number-counting factor regulates the cytoskeleton and cell motility in Dictyostelium. Proc. Natl. Acad. Sci. USA 99:1371-1376.[Abstract/Free Full Text]
  36. Van Driessche, N., C. Shaw, M. Katoh, T. Morio, R. Sucgang, M. Ibarra, H. Kuwayama, T. Saito, H. Urushihara, M. Maeda, I. Takeuchi, H. Ochiai, W. Eaton, J. Tollett, J. Halter, A. Kuspa, Y. Tanaka, and G. Shaulsky.2002 . A transcriptional profile of multicellular development in Dictyostelium discoideum.Development 129:1543-1552.[Abstract/Free Full Text]
  37. Vauti, F., P. Morandini, J. Blusch, A. Sachse, and W. Nellen.1990 . Regulation of the discoidin I gamma gene in Dictyostelium discoideum: identification of individual promoter elements mediating induction of transcription and repression by cyclic AMP. Mol. Cell. Biol. 10:4080-4088.[Abstract/Free Full Text]
  38. Walsh, E. P., D. J. Lamont, K. A. Beattie, and M. J. Stark. 2002. Novel interactions of Saccharomyces cerevisiae type 1 protein phosphatase identified by single-step affinity purification and mass spectrometry.Biochemistry 41:2409-2420.[CrossRef][Medline]
  39. Warnes, G., B. Bolker, and T. Lumley. 2006. gplots: various R programming tools for plotting data. R package version 2.3.1. http://cran.r-project.org/src/contrib/Descriptions/gplots.html.
  40. Watts, D. J., and J. M. Ashworth. 1970. Growth of myxameobae of the cellular slime mould Dictyostelium discoideum in axenic culture. Biochem. J. 119:171-174.[Medline]
  41. Williams, J. G., A. S. Tsang, and H. Mahbubani.1980 . A change in the rate of transcription of a eukaryotic gene in response to cyclic AMP. Proc. Natl. Acad. Sci. USA 77:7171-7175.[Abstract/Free Full Text]
  42. Wong, E., C. Yang, J. Wang, D. Fuller, W. F. Loomis, and C. H. Siu. 2002. Disruption of the gene encoding the cell adhesion molecule DdCAD-1 leads to aberrant cell sorting and cell-type proportioning during Dictyostelium development. Development 129:3839-3850.

Eukaryotic Cell, February 2007, p. 245-252, Vol. 6, No. 2
1535-9778/07/$08.00+0     doi:10.1128/EC.00204-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Other Versions of this Article:
EC.00204-06v1
6/2/245    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Strmecki, L.
Right arrow Articles by Pears, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Strmecki, L.
Right arrow Articles by Pears, C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Appl. Environ. Microbiol. Infect. Immun. J. Bacteriol.
Mol. Cell Biol. Microbiol. Mol. Biol. Rev. ALL ASM JOURNALS