The C3 domain of Pasteurella multocida toxin is the minimal domain responsible for activation of Gq-dependent calcium and mitogenic signaling

doi:10.1110/ps.083445408

Journal List > Protein Sci > v.17(5); May 2008

Protein Sci. 2008 May; 17(5): 945–949.

doi: 10.1110/ps.083445408.

PMCID: PMC2327272

The C3 domain of Pasteurella multocida toxin is the minimal domain responsible for activation of G_q-dependent calcium and mitogenic signaling

Leila R. Aminova, Shuhong Luo, Yuka Bannai, Mengfei Ho, and Brenda A. Wilson

Department of Microbiology, University of Illinois at Urbana-Champaign, Urbana, Illinois 61801, USA

Received January 8, 2008; Revised February 9, 2008; Accepted February 9, 2008.

This article has been cited by other articles in PMC.

Abstract

The large 1285-amino-acid protein toxin from Pasteurella multocida (PMT) is a multifunctional single-chain polypeptide that binds to and enters eukaryotic cells and acts intracellularly to promote G_q and G_12/13 protein-dependent calcium and mitogenic signal transduction. Previous studies indicated that the intracellular activity domain responsible for PMT action was located within the C-terminal 600–700 amino acids. In this study, we have exogenously expressed a series of N- and C-terminal PMT fragments directly in mammalian cells and have used the dual luciferase reporter system to assay for toxin-mediated activation of calcium-calcineurin-NFAT signaling (NFAT-luciferase) and mitogenic serum response signaling (SRE-luciferase). Using this approach, we have defined the last 180 amino acids, which encompass the C3 domain in the crystal structure, as the minimum domain sufficient to activate both NFAT and SRE signaling pathways.

Keywords: dermonecrotic toxin, NFAT, SRE, calcium mobilization, G_q protein

The protein toxin from Pasteurella multocida (PMT) experimentally induces all of the major symptoms of a number of economically important zoonotic diseases in wild animals and domestic livestock and pets (Foged 1992; Arashima and Kumasaka 2005; Harper et al. 2006; Wilson and Ho 2006), including moderate to severe progressive atrophic rhinitis, bite wound dermonecrosis and abscesses, chronic respiratory infection, and decreased overall stature and weight gain. PMT is a 1285-amino-acid protein that binds to and enters mammalian cells via receptor-mediated endocytosis (Rozengurt et al. 1990; Pettit et al. 1993; Dudet et al. 1996). PMT has dramatic and differential effects on differentiation and proliferation of fibroblasts, bone cells, epithelial cells, and cardiomyocytes (for review, see Wilson and Ho 2004, 2006). In fibroblasts and osteoblasts, PMT acts intracellularly to enhance calcium signaling pathways, including inositol phospholipid hydrolysis, intracellular Ca²⁺ mobilization, increased protein kinase C (PKC)-dependent phosphorylation, and calcineurin-dependent nuclear factor of activated T cells (NFAT) activation (Staddon et al. 1990, 1991; Wilson et al. 1997; Sabri et al. 2002; Aminova and Wilson 2007; Luo et al. 2007), as well as mitogenic pathways, including increased tyrosine phosphorylation, MAPK (Erk1/2, p38) activation, and DNA synthesis (Rozengurt et al. 1990; Dudet et al. 1996; Lacerda et al. 1996; Mullan and Lax 1996; Seo et al. 2000; Wilson et al. 2000).

The primary intracellular target of PMT responsible for activation of the phospholipase C β1 (PLCβ1) pathway is the free, monomeric α subunit of the G_q protein (Wilson et al. 1997; Zywietz et al. 2001; Wilson and Ho 2004). PMT-induced stimulation of PLCβ1 by Gα_q leads to the hydrolysis of phosphatidylinositol 4,5-bisphosphate to inositol 1,4,5-trisphosphate and diacylglycerol. Accordingly, the release of these second messengers stimulates Ca²⁺ mobilization and activates PKC-dependent phosphorylation and calcineurin-dependent NFAT activation. PMT stimulation of mitogen-activated protein kinase occurs upstream via G_q/11-dependent transactivation of the epidermal growth factor receptor (Seo et al. 2000). PMT also initiates cytoskeletal rearrangements, including focal adhesion assembly and stress fiber development (Dudet et al. 1996; Lacerda et al. 1996; Ohnishi et al. 1998). At least part of the mitogenic effect of PMT occurs through a G_q-independent pathway (Zywietz et al. 2001). Rho activation by PMT has been reported to be the result of indirect activation through G_12/13 (Orth et al. 2005) or through tyrosine kinases (Sagi et al. 2001). Our recent study revealed two additional pathways affected by PMT, namely, down-regulation of Notch1 and stabilization of β-catenin during adipocyte differentiation (Aminova and Wilson 2007).

There is now strong evidence that the G_q-activating domain of PMT is localized to the C terminus, since site-specific mutations at H1205, H1223, and C1165 abolish intracellular activity when the mutant proteins are introduced into cells via microinjection (Pullinger et al. 2001; Baldwin et al. 2004) or electroporation (Busch et al. 2001; Orth et al. 2003). The recently reported crystal structure of the C-terminal 575–1285 residues revealed three subdomains (see Fig. 1), designated as C1, C2, and C3 domains (Kitadokoro et al. 2007). The shortest active fragment in this study was identified as residues 569–1285. According to this study, the C1 domain (residues 575–719) may be responsible for membrane targeting, the C2 domain (residues 720–1104) has no obvious function, and the C3 domain (residues 1105–1285) possesses a putative Cys-His-Asp catalytic triad, which when the C1165 is released from a disulfide bond (C1159-C1165) through mutation of C1159 to Ser sterically aligns to that of a papain or cysteine protease-like triad. However, the authors were not able to demonstrate any proteolytic activity.

Figure 1.

Structure of C-PMT. Shown is a ribbon diagram of the C terminus of PMT (C-PMT, residues 575–1285): C1 (blue, 575–719), C2 (green, 720–1104), and C3 (red, 1105–1285). Lines indicate location of the starting residue (yellow (more ...)

The minimal domain of PMT responsible for intracellular activation of G_q-dependent signaling has been the subject of some debate. One group reported that a C-terminal fragment of PMT consisting of residues 720–1285, when microinjected into cells, induced cellular morphology changes similar to full-length PMT, while a smaller fragment consisting of residues 849–1285 showed no activity (Pullinger et al. 2001). Consistent with the shortest active fragment of 569–1285 reported in the crystallographic study (Kitadokoro et al. 2007), another group found that a fragment of PMT consisting of residues 581–1285, when electroporated into cells, increased intracellular inositol phosphate levels, but a fragment consisting of residues 701–1285 was not biologically active (Busch et al. 2001). However, none of these studies demonstrated intracellular activity in smaller fragments consisting of only the C3 domain.

The methods used in previous studies to define the minimal intracellular activity domain relied on expression and purification of the recombinant truncated proteins from Escherichia coli, which assumed that these mutant proteins folded properly in E. coli without proteolytic degradation and were stable enough to withstand subsequent purification steps. In each study, the choice of truncation site was arbitrary and only a few relatively large truncation mutants were examined. In addition, these studies also assumed that the techniques used to introduce the truncated proteins into the cells (i.e., microinjection or electroporation) would not perturb the functional readout of the assay, such as cytoskeletal changes, calcium, or other signaling.

Our approach reported here circumvented these disadvantages by expressing a series of N- and C-terminal toxin fragments (the locations of the truncation sites are indicated in Fig. 1) and directly expressing them in mammalian cells using mammalian expression vectors. To determine the intracellular activity of these proteins, we used two different downstream luciferase reporter systems: one based on PMT activation of calcium signaling (Wilson et al. 1997; Aminova and Wilson 2007; Luo et al. 2007), leading to stimulation of calcineurin and NFAT (Crabtree and Olson 2002; Im and Rao 2004; Aminova and Wilson 2007), and the other based on PMT activation of mitogenic signaling (Seo et al. 2000; Wilson et al. 2000; Sagi et al. 2001; Orth et al. 2005), leading to stimulation of serum response element (SRE) through several mitogenic signaling pathways, including those regulated by the Ras, Rho, G_q, and G_12/13 families of GTPases (Hill et al. 1995; Fromm et al. 1997; Seo et al. 2000; Sagi et al. 2001).

Results and Discussion

Inclusion of GFP- or FLAG-tags as markers for visualization allowed us to determine via Western blot analysis the level of expression and stability of the expected full-length and mutant toxin proteins exogenously expressed in the mammalian cells (see Supplemental material). Full-length PMT intracellularly expressed in CHO-K1 cells caused a 20-fold increase in NFAT-luciferase reporter activity and a 10-fold increase in SRE-luciferase reporter activity (Fig. 2). Whereas truncations in the C terminus of PMT (1–568g, g1–1132f, and g1–1256f) showed no activation of these two signaling pathways (Fig. 2, left plots), two larger N-terminal deletion fragments f506–1285 and f576–1285, encompassing the putative catalytic regions reported previously (Busch et al. 2001; Pullinger et al. 2001; Kitadokoro et al. 2007), possessed NFAT-luciferase and SRE-luciferase activities comparable to that of full-length toxin (f1–1285 or g1–1285f) (Fig. 2, right plots). In addition to these fragments, we tried to express mutants starting from residues 710, 720, 771, 917, and 1005 (Supplemental Fig. 1); however, all of these constructs showed significant degradation, except fragment f1005–1285, which expressed poorly, and all of these constructs exhibited little or no NFAT or SRE activity (only data for f771–1285 and f1025–1285 are shown in Fig. 2). Interestingly, while a bacterially expressed fragment encompassing residues 701–1285 showed no activity by electroporation (Busch et al. 2001), a smaller bacterially expressed fragment starting from residue 720 exhibited mitogenic activity upon microinjection into cells (Pullinger et al. 2001). However, an analogous fragment f720–1285 expressed in mammalian cells was unstable and showed little or no intracellular activity (data not shown).

Figure 2.

Luciferase-reporter assays for PMT activity. Shown are plots of NFAT-luciferase (A) and SRE-luciferase (B) activities after transient transfection with dual NFAT-luciferase or SRE-luciferase reporter plasmids and plasmids expressing full-length PMT on (more ...)

Truncated mutants starting from residues 1025, 1066, 1085, 1103, or 1105 to the C terminus were expressed as stable proteins (Supplemental Fig. 1) and showed ~50% of NFAT or SRE-luciferase activity compared with that of full-length toxin (Fig. 2, right plots). As shown in Figure 1, the smallest active domain of PMT (f1105–1285) corresponded to the C3 domain defined in the crystal structure (Kitadokoro et al. 2007). Any further deletion into the C3 domain resulted in loss of activity (Fig. 2, right plots) even though protein was expressed (Supplemental Fig. 1). Thus, this C-terminal 180 amino acid fragment was sufficient to activate both NFAT and SRE signaling pathways, strongly supporting that the biological activity of PMT responsible for activation of both of these signaling pathways is confined primarily to this region. Activation of the NFAT and SRE signaling pathways was specific to PMT activity, since point mutations previously shown to abolish PMT activity, such as C1165S, likewise ablated the activity of the toxin proteins in both assays (Supplemental Fig. 2). The partial activity exhibited by the C3 domain (residues 1105–1295), compared with the full activity of the larger fragments containing domains C1-C3 (residues 576–1285), suggests that the entire C1-C3 domain is important for full toxin activity and that although C3 possesses the minimal intracellular biological activity, the C1 and C2 domains may contribute to proper intracellular localization, interaction with other potential protein cofactors, and/or protein conformation necessary for full activity.

In summary, we have used exogenous expression of a series of PMT truncations and substitution mutants in mammalian cells to define the minimal intracellular activity domain of PMT. Our results demonstrated that the last 180 amino acids, the C3 domain in the crystal structure (Fig. 1), define the minimal activity domain sufficient to stimulate both calcium-dependent (calcineurin-NFAT) and mitogenic (SRE) signaling pathways. With this information, it may now be possible to more easily ascertain the as yet elusive biological activity of PMT.

Materials and Methods

Plasmid constructs

Deletion mutants were constructed by fragment exchange or PCR cloning using previously described PMT constructs in pET21 vectors (Novagen) (Wilson et al. 1999). Two mammalian expression vectors were used for cloning: pEGFP-C1 (Clontech) or pcDNA3 (Invitrogen) with a 3×FLAG sequence inserted into KpnI/BamHI restriction sites. Corresponding PMT fragments were amplified by PCR using PCR master mix (ABgene) and specific primers (IDT) and cloned into pEGFP or pcDNA3 vectors. The single mutation C1165S was introduced into wild-type PMT fragment (1005–1285) by PCR-based site-directed mutagenesis using published primers (Ward et al. 1998; Orth et al. 2003). After sequence verification, the corresponding mutations were introduced into the indicated toxin constructs by fragment exchange using standard methods.

Cell culture

CHO-K1 cells (ATCC, CCL-61) were cultured at 37°C and 5% CO₂ in F-12 medium, consisting of F-12 Nutrient Mixture (GIBCO) supplemented with 10% heat-inactivated bovine growth serum (BGS, HyClone), pH 7.4, 100 units/mL penicillin G, and 100 μg/mL streptomycin. For transient transfection, cells were plated into 24-well plates at 70%–80% confluence 24 h prior to transfection.

Luciferase assay

The effect of external treatment with rPMT on calcium signaling was assayed using a stable CHO-K1 cell line containing NFAT-luciferase reporter (Aminova and Wilson 2007). To study the effect of expression of PMT or PMT deletion mutants on NFAT-luciferase or SRE-luciferase activity in mammalian cells, CHO-K1 cells were transiently transfected with a combination of pNFAT-luc and pRenilla-TK vectors (pGL 7.4 hRluc/TK, Promega) or pSRE-luc (Stratagene) and pRenilla-TK vectors in a ratio of 10:1 and vector expressing recombinant protein (50 ng/well) using Lipofectamine 2000 reagent (Invitrogen). Cell extracts were prepared and analyzed 24 h after transfection using the Dual Luciferase Assay System (Promega), according to manufacturer's protocol. Luminescence was measured using a Synergy-HT multi-detection microplate reader (BioTek), and results were reported as relative light units. All cell extracts used for luciferase assay were analyzed by Western blot to verify expression of protein of interest. In all experiments reported here, data are expressed as the mean ± SD of results from at least three independent experiments performed in triplicate.

Electronic supplemental material

The Supplemental materials provide additional details on the following: (1) results of Western blots of CHO-K1 cells showing expression levels of toxin mutant proteins (Supplemental Fig. 1), (2) methods used for determination of mammalian expression of toxin proteins using immunoblotting, and (3) results of point mutations previously shown to abolish PMT activity on NFAT and SRE luciferase activity (Supplemental Fig. 2).

Acknowledgments

We thank Miranda Bertram and Tracy Chong for technical assistance with construction of some of the mutants. This work was supported by NIH/NIAID grant AI038396 (to B.A.W.).

Footnotes

Supplemental material: see www.proteinscience.org

Reprint requests to: Brenda A. Wilson, Department of Microbiology, University of Illinois at Urbana-Champaign, 601 South Goodwin Avenue, B128 CLSL, Urbana, Illinois 61801, USA; e-mail: bawilson/at/life.uiuc.edu; fax: (217) 244-6697.

Abbreviations: rPMT, recombinant Pasteurella multocida toxin; NFAT, nuclear factor of activated T cells; SRE, serum response element; PLC, phospholipase C; PKC, protein kinase C.

Article published online ahead of print. Article and publication date are at http://www.proteinscience.org/cgi/doi/10.1110/ps.083445408.

References

Aminova, L.R., Wilson, B.A. Calcineurin-independent inhibition of 3T3-L1 adipogenesis by Pasteurella multocida toxin: Suppression of Notch1, stabilization of β-catenin and pre-adipocyte factor 1. Cell. Microbiol. 2007;9:2485–2496. [PubMed]
Arashima, Y., Kumasaka, K. Pasteurellosis as zoonosis. Intern. Med. 2005;44:692–693. [PubMed]
Baldwin, M.R., Lakey, J.H., Lax, A.J. Identification and characterization of the Pasteurella multocida toxin translocation domain. Mol. Microbiol. 2004;54:239–250. [PubMed]
Busch, C., Orth, J., Djouder, N., Aktories, K. Biological activity of a C-terminal fragment of Pasteurella multocida toxin. Infect. Immun. 2001;69:3628–3634. [PubMed]
Crabtree, G.R., Olson, E.N. NFAT signaling: Choreographing the social lives of cells. Cell. 2002;109(Suppl):S67–S79. doi: 10.1016/S0092-8674(02)00699-2. [PubMed]
Dudet, L.I., Chailler, P., Dubreuil, J.D., Martineau-Doize, B. Pasteurella multocida toxin stimulates mitogenesis and cytoskeleton reorganization in Swiss 3T3 fibroblasts. J. Cell. Physiol. 1996;168:173–182. [PubMed]
Foged, N.T. Pasteurella multocida toxin. The characterisation of the toxin and its significance in the diagnosis and prevention of progressive atrophic rhinitis in pigs. APMIS Suppl. 1992;25:1–56. [PubMed]
Fromm, C., Coso, O.A., Montaner, S., Xu, N., Gutkind, J.S. The small GTP-binding protein Rho links G protein-coupled receptors and Gα₁₂ to the serum response element and to cellular transformation. Proc. Natl. Acad. Sci. 1997;94:10098–10103. [PubMed]
Harper, M., Boyce, J.D., Adler, B. Pasteurella multocida pathogenesis: 125 years after Pasteur. FEMS Microbiol. Lett. 2006;265:1–10. [PubMed]
Hill, C.S., Wynne, J., Treisman, R. The Rho family GTPases RhoA, Rac1, and CDC42Hs regulate transcriptional activation by SRF. Cell. 1995;81:1159–1170. [PubMed]
Im, S.H., Rao, A. Activation and deactivation of gene expression by Ca²⁺/calcineurin-NFAT-mediated signaling. Mol. Cells. 2004;18:1–9. [PubMed]
Kitadokoro, K., Kamitani, S., Miyazawa, M., Hanajima-Ozawa, M., Fukui, A., Miyake, M., Horiguchi, Y. Crystal structures reveal a thiol protease-like catalytic triad in the C-terminal region of Pasteurella multocida toxin. Proc. Natl. Acad. Sci. 2007;104:5139–5144. [PubMed]
Lacerda, H.M., Lax, A.J., Rozengurt, E. Pasteurella multocida toxin, a potent intracellularly acting mitogen, induces p125FAK and paxillin tyrosine phosphorylation, actin stress fiber formation, and focal contact assembly in Swiss 3T3 cells. J. Biol. Chem. 1996;271:439–445. [PubMed]
Luo, S., Ho, M., Wilson, B.A. Application of intact cell-based NFAT-β-lactamase reporter assay for Pasteurella multocida toxin-mediated activation of calcium signaling pathway. Toxicon. 2008;51:597–605. [PubMed]
Mullan, P.B., Lax, A.J. Pasteurella multocida toxin is a mitogen for bone cells in primary culture. Infect. Immun. 1996;64:959–965. [PubMed]
Ohnishi, T., Horiguchi, Y., Masuda, M., Sugimoto, N., Matsuda, M. Pasteurella multocida toxin and Bordetella bronchiseptica dermonecrotizing toxin elicit similar effects on cultured cells by different mechanisms. J. Vet. Med. Sci. 1998;60:301–305. [PubMed]
Orth, J.H., Blocker, D., Aktories, K. His1205 and His1223 are essential for the activity of the mitogenic Pasteurella multocida toxin. Biochemistry. 2003;42:4971–4977. [PubMed]
Orth, J.H., Lang, S., Taniguchi, M., Aktories, K. Pasteurella multocida toxin-induced activation of RhoA is mediated via two families of Gα proteins, Gα_q and Gα_12/13 . J. Biol. Chem. 2005;280:36701–36707. [PubMed]
Pettit, R.K., Ackermann, M.R., Rimler, R.B. Receptor-mediated binding of Pasteurella multocida dermonecrotic toxin to canine osteosarcoma and monkey kidney (vero) cells. Lab. Invest. 1993;69:94–100. [PubMed]
Pullinger, G.D., Sowdhamini, R., Lax, A.J. Localization of functional domains of the mitogenic toxin of Pasteurella multocida . Infect. Immun. 2001;69:7839–7850. [PubMed]
Rozengurt, E., Higgins, T., Chanter, N., Lax, A.J., Staddon, J.M. Pasteurella multocida toxin: Potent mitogen for cultured fibroblasts. Proc. Natl. Acad. Sci. 1990;87:123–127. [PubMed]
Sabri, A., Wilson, B.A., Steinberg, S.F. Dual actions of the Gα_q agonist Pasteurella multocida toxin to promote cardiomyocyte hypertrophy and enhance apoptosis susceptibility. Circ. Res. 2002;90:850–857. [PubMed]
Sagi, S.A., Seasholtz, T.M., Kobiashvili, M., Wilson, B.A., Toksoz, D., Brown, J.H. Physical and functional interactions of Gα_q with Rho and its exchange factors. J. Biol. Chem. 2001;276:15445–15452. [PubMed]
Seo, B., Choy, E.W., Maudsley, S., Miller, W.E., Wilson, B.A., Luttrell, L.M. Pasteurella multocida toxin stimulates mitogen-activated protein kinase via G_q/11-dependent transactivation of the epidermal growth factor receptor. J. Biol. Chem. 2000;275:2239–2245. [PubMed]
Staddon, J.M., Chanter, N., Lax, A.J., Higgins, T.E., Rozengurt, E. Pasteurella multocida toxin, a potent mitogen, stimulates protein kinase C-dependent and -independent protein phosphorylation in Swiss 3T3 cells. J. Biol. Chem. 1990;265:11841–11848. [PubMed]
Staddon, J.M., Barker, C.J., Murphy, A.C., Chanter, N., Lax, A.J., Michell, R.H., Rozengurt, E. Pasteurella multocida toxin, a potent mitogen, increases inositol 1,4,5-trisphosphate and mobilizes Ca²⁺ in Swiss 3T3 cells. J. Biol. Chem. 1991;266:4840–4847. [PubMed]
Ward, P.N., Miles, A.J., Sumner, I.G., Thomas, L.H., Lax, A.J. Activity of the mitogenic Pasteurella multocida toxin requires an essential C-terminal residue. Infect. Immun. 1998;66:5636–5642. [PubMed]
Wilson, B.A., Ho, M. Pasteurella multocida toxin as a tool for studying G_q signal transduction. Rev. Physiol. Biochem. Pharmacol. 2004;152:93–109. [PubMed]
Wilson, B.A., Ho, M. Pasteurella multocida toxin. In: Alouf J.E., Popoff M.R., editors. The comprehensive sourcebook of bacterial protein toxins. 3rd ed. Academic Press; Boston, MA: 2006. pp. 430–447.
Wilson, B.A., Zhu, X., Ho, M., Lu, L. Pasteurella multocida toxin activates the inositol triphosphate signaling pathway in Xenopus oocytes via G_qα-coupled phospholipase C-β1. J. Biol. Chem. 1997;272:1268–1275. [PubMed]
Wilson, B.A., Ponferrada, V.G., Vallance, J.E., Ho, M. Localization of the intracellular activity domain of Pasteurella multocida toxin to the N terminus. Infect. Immun. 1999;67:80–87. [PubMed]
Wilson, B.A., Aminova, L.R., Ponferrada, V.G., Ho, M. Differential modulation and subsequent blockade of mitogenic signaling and cell cycle progression by Pasteurella multocida toxin. Infect. Immun. 2000;68:4531–4538. [PubMed]
Zywietz, A., Gohla, A., Schmelz, M., Schultz, G., Offermanns, S. Pleiotropic effects of Pasteurella multocida toxin are mediated by G_q-dependent and -independent mechanisms. Involvement of G_q but not G₁₁ . J. Biol. Chem. 2001;276:3840–3845. [PubMed]

Articles from Protein Science : A Publication of the Protein Society are provided here courtesy of
The Protein Society