WO2005085417A2

WO2005085417A2 - Bacterial system for protein transport in eukaryotic cells

Info

Publication number: WO2005085417A2
Application number: PCT/DE2005/000330
Authority: WO
Inventors: Ulla Bonas; Daniela Büttner
Original assignee: Martin-Luther-Universität Halle-Wittenberg
Priority date: 2004-03-02
Filing date: 2005-02-26
Publication date: 2005-09-15
Also published as: DE102004010023A1; WO2005085417A3; DE112005001016A5

Abstract

The aim of the invention is to develop a system for the targeted transport of proteins into eukaryotic cells. The specific transport of proteins into eukaryotic cells is carried out using a type III secretion system, using bacteria strains that are mutated in hpaB or homogenous genes. A non-pathogenic strain is generated from a Gram-negative pathogenic bacteria strain, said non-pathogenic strain being mutated in the hpaB-gen or a homogenous gene which does not mediate any significant translocation of bacterial effector proteins. Over-expressed effector proteins or proteins containing the type 111 secretion signal of a non-effector protein can, however, be translocated into eukaryotic cells using the bacterially inherent TTSS. The inventive bacterial system is used to transport bacterial proteins into eukaryotic cells, in order to influence or modify cellular processes such as gene expression, growth, development and defence/resistance mechanisms.

Description

Bacterial system for protein transport in eukaryotic cells

The present invention relates to the targeted transport of proteins into eukaryotic cells by the bacterial type III secretion system (TTSS, "type III secretion system") using bacterial strains which are mutated in hpaB or homologous genes.

In most Gram-negative plant and animal pathogenic bacteria, the TTSS is essential for the infection and colonization of the eukaryotic host organisms. This protein transport system spans both bacterial membranes (inner and outer membrane) and is associated with an extracellular pilus. The TTSS mediates both the secretion of bacterial proteins into the extracellular medium and the translocation of so-called effector proteins into the eukaryotic host cell IM. In phytopathogenic bacteria, the TTSS is encoded by hrp ("hypersensitive response and ßathogenicity") genes, which are usually organized as clusters. Hrp genes have been used for many Gram-negative phytopathogenic bacteria such as Erwinia amylovora, Ralstonia solanacearum, Pathovare from Pseudomonas syringae and species of Xanthomonas described 121. Based on the arrangement of the genes and the frrp gene regulation, a distinction is made between bacteria of group 1, for example species of the genera Pseudomonas and Erwinia, and bacteria of group 2, which include Ralstonia solanacearum and species of the genus Xanthomonas 131 ,

In pathogenic bacteria, hrp genes are essential for bacterial pathogenicity and for inducing a hypersensitive response (HR) to resistant host or non-host plants. HR is a rapid, programmed death of plant cells and mostly occurs in connection with plant defense reactions. The induction of HR can be triggered by the direct or indirect recognition of individual bacterial effector proteins - also known as avirulence proteins (Avr proteins) - from corresponding plant resistance proteins / 4 /. The fact that HR is also induced in transient expression of individual Avr proteins in resistant plants was a first indication that Avr proteins are being transported into the plant cell. Many effector proteins of phytopathogenic bacteria were initially identified due to their avirulence activity in plants with corresponding resistance proteins.

The invention specified in claim 1 was based on the problem of developing a protein transport system which enables the specific transport of selected effectors into the plant cell. One possibility for this is the heterologous expression of Λtp gene clusters in non-pathogenic Gram-negative bacteria, which has already been described for Escherichia coli and Pseudomonas fluorescens strains (patent WO0002996 / US2002083489). Here, the /? Φ gene cluster of P. syringae pv. Syringae was expressed from the cosmid pHIR11 in both bacterial strains. If the P. syr / ^' ngae effector proteins AvrB or AvrPto are additionally expressed by introducing a second plasmid, the generated E. coli and P. fluorescens strains induce an AvrB or AvrPto-specific HR / 5,6 in the plant /. However, it was not possible to detect AvrB and AvrPto in the culture medium of P. syringae, E. coli (pHIR11) or P. fluorescens (pHIR11). In contrast, AvrB and AvrPto were secreted into the culture medium of E. co // strains that express the Λ / p gene cluster of Erwinia chrysanthemi III. In contrast to the P. syringae Λφ gene cluster, the E. chrysanthemi hrp genes are also expressed in complex medium and enable cell contact-independent protein secretion.

However, the expression of heterologous TTSS in non-pathogenic bacteria has so far only been possible using the Λφ gene clusters of Gram-negative bacteria from group 1 (P. syringae and E. chrysanthemi), since only here are the necessary regulatory proteins encoded within the gene cluster. However, the TTSS of group 1 bacteria cannot translocate effector proteins from group 2 bacteria. It has been shown that the effector protein AvrBs3 from X. campestris pv. Vesicatoria (group 2 bacterium) is not translocated in plant cells by the TTSS from P. syringae (belonging to group 1). For these analyzes, avrBs3 was expressed next to the /? Φ gene cluster from P. syringae in P. fluorescens and the resulting bacterial strain was infiltrated in AvrBs3-responsive plants.

A general disadvantage when using heterologously expressed Λφ gene clusters for the targeted transport of individual proteins is that non-pathogenic bacteria such as. B. P. fluorescens and E. co // - Laboratory strains may develop pathogenic properties on eukaryotic organisms by being equipped with a TTSS. This problem was solved according to the invention in that a non-pathogenic strain was generated from the gram-negative plant pathogenic bacterium Xanthomonas campestris pv. Vesicatoria, which is mutated in the hpaB (“Λφ-associated” B) gene. HpaB mutants do not translocate significant amounts of effector proteins, but are able to mediate the translocation of proteins that have the secretion signal of a so-called non-effector protein using their own TTSS.

The hpaB gene from X. campestris pv. Vesicatoria (GenBank accession AF056246) is located in the Λφ gene cluster downstream of hrpE1 and encodes a protein that controls the TTSS-dependent protein secretion and translocation. In X. campestris pv. Ves / cafot / a strain 85-10, the mutation of hpaB leads to the complete loss of the bacterial ability to grow on susceptible paprika plants and trigger disease symptoms or to induce HR on resistant paprika plants. In the X. campestris pv. Ves / cafo / va strain 85 *, the mutation of hpaB also leads to Loss of pathogenicity on susceptible plants, but still allows induction of a very weak, partial HR on resistant plants (see Figure 1). Strain 85 * is a derivative of strain 85-10 and contains hrpG ^* , which encodes a constitutively active form of the regulatory protein HrpG. HrpG is a member of the OmpR family of two-component regulators and uses the AraC-type-like transcription activator HrpX to activate the expression of the Λφ genes in plant 181. The use of? ΦG * mutants not only enables Λφ Gene expression in vitro, but also allows analysis of the type HI-dependent protein secretion in the culture medium. The described phenotypes of 7pass mutant strains in susceptible and resistant plants are due to a greatly reduced efficiency in the transport of effector proteins. Thus, in wϊro secretion analyzes, the effector proteins AvrBs3, AvrBsl and AvrBsT are only detectable in small amounts or not at all in the culture supernatant of Λpass mutants (see Figure 2). In contrast, the secretion of HrpF and XopA is not affected by the mutation of hpaB. HrpF and XopA do not belong to the group of effector proteins, ie they are not translocated into the plant cell. Since HrpF is essential for pathogenicity and interacts with membranes, this protein is the main component of the type Ill translocon 181. The mutation of xop> 4 leads to a reduction in bacterial pathogenicity on susceptible plants and to the induction of reduced HR on resistant plants / 9 /. XopA is therefore involved in the effector protein translocation. It has been shown that a fusion protein which contains the N-terminal 51 amino acids of XopA fused to AvrBs3Δ2 is secreted by strain 85 *, but does not induce AvrBs3-specific HR. AvrBs3Δ2 is an N-terminal deletion derivative of the effector protein AvrBs3, which is no longer secreted by the type III secretion system. AvrBs3Δ2, however, still induces HR in resistant plants after transient expression in the plant or after fusion to a secretion and translocation signal and is therefore used as a reporter protein for the analysis of type III-dependent protein translocation / 10,11 /. The fact that the Xopa _{_1-51} - AvrBs3Δ2 fusion protein * no HR induced after transport through the TTSS of stem 85 on AvrBs3-responsive plants shows that Xopa is not translocated from this bacterial strain. Both XopA and HrpF thus belong to the group of non-effector proteins, which refers to proteins that are secreted by the TTSS but not translocated. The secretion analyzes shown in Figure 2 show that the mutation of hpaB leads to a drastic reduction in the secretion of effector proteins, but does not impair the secretion of non-effector proteins. According to the invention, a “mutation of hpaB” characterizes all base pair exchanges and deletions in the hrp- Gene clusters in the area of the ftpass gene or in the area of Λpass associated promoter elements which prevent expression of the Λpass gene or lead to the expression of a non-functional gene product. According to the invention, the feature “inoperable” also designates all Λpass gene products which lead to a reduced efficiency in the in w ^' -ro secretion of effector proteins compared to the corresponding hpass wild-type strain. The mutations in are for the investigation of Λpass mutants to introduce the bacterial genome by means of homologous recombination with the aid of the suicide vector pOK1 / 12 /.

It has thus been shown that HpaB assumes an essential function during the transport of the effector protein by the TTSS. HpaB from X. campestris pv. Vesicatoria is a 162 amino acid protein with an isoelectric point of 4.3 and a leucine content of 13.6%. HpaB thus has typical characteristics of type III secretion chaperones, which rarely show sequence homology with each other, but are usually small, acidic and rich in leucine. Type III secretion chaperones contribute to the stability and / or the efficient secretion of one or more translocon or effector proteins / 13 /. HpaB as a general type III secretion chaperone enables the secretion of effector proteins, while it is not involved in the secretion of non-effectors. The complete loss of pathogenicity of frpaß mutants is due to the fact that the protein amounts of effectors which are still translocated in the absence of HpaB are insufficient for the colonization of susceptible plants and the induction of disease symptoms.

However, pass mutants are still able to transport significant amounts of effector proteins into the plant cell if they are overexpressed. X. campestris pv. Ves / caforia strain 85 * AhpaB (pDSF300), which overexpresses AvrBs3 under the control of a strong / ac promoter, induces HR in AvrBs3-responsive plants (see Figure 4). According to the invention, overexpression is to be understood as the gene expression of a "stronger" promoter compared to the wild-type promoter, the "strength" of the promoter determining the transcription rate of the gene to be expressed. Promoters are DNA sequences that enable the binding of the RNA polymerase and thus the mRNA synthesis. Expression vectors containing promoters can contain various combinations of transcription and translation initiation signals. According to the invention, overexpression is also understood to mean the expression of effector genes from the native promoter in an independently replicating expression vector, which leads to an increased number of copies of the corresponding gene compared to the wild-type strain. Examples of expression vectors which contain a replication signal for E. coli and X. campestris pv. Vesicatoria and a marker gene (antibiotic resistance gene) for selection and for Overexpression can be used are pLAFR, pDSK, pUFR and pBBR derivatives. The sequence to be expressed can, for example, be ligated into the vectors at suitable restriction sites. The plasmid obtained can be introduced into E. coli by electroporation and then brought into the desired X. campestris pv. Ves / cator / ^' a strains via triparental conjugation or electroporation. E. coli and X. campestris pv. Ves / cator / a strains containing the expression vector are selected on culture medium with appropriate antibiotics. Genes are cloned into expression vectors using known techniques / 14 /. The conjugation of expression plasmids in X. campestris pv. Vesicatoria has been described / 15 /.

It was also possible to show that, in addition to individual effector proteins whose corresponding coding sequences are overexpressed, Λpass mutants also translocate reporter proteins if they contain an N-terminal secretion signal from a non-effector protein. Strain 85 * AhpaB, which induces the fusion protein HrpF ₁ . ₂₀₀ -AvrBs3Δ2, which contains the N-terminal 200 amino acids of HrpF fused to AvrBs3Δ2, the AvrBs3-specific HR. An analogous result was obtained for strain 85 * Δ? Pass, which contains the fusion protein XopA ₁ . ₅₁ -AvrBs3Δ2, which contains the N-terminal 50 amino acids of XopA fused to AvrBs3Δ2. Both HrpF and XopA are normally non-effector proteins, ie they are secreted but not translocated in the presence of HpaB. Thus strain 85 * induces which HrpF ^ oo-AvrBs3Δ2 or XopA ₁ . ₅₁ -AvrBs3Δ2 expressed, no HR on AvrBs3-responsive plants, although both fusion proteins can be detected in the culture supernatant (see Figures 3 and 4). The translocation of both proteins in frpaß mutants shows that HpaB controls the type III-dependent protein translocation. HpaB thus inhibits the translocation of non-effectors into the host cell and, at the same time, promotes the transport of effector proteins as a general type III secretion chaperone.

According to the invention, a bacterial strain is therefore used which no longer has any pathogenic properties, but is capable of transporting effector proteins into the plant cell after their overexpression or fusion proteins which contain a type III secretion signal for this bacterial strain. The method according to the invention thus offers an inexpensive alternative to the use of heterologous protein transport systems which are based on the expression of the Λφ gene cluster of a group 1 bacterium in a non-pathogenic strain. A major advantage of the present invention is to have a bacterial strain available whose TTSS no longer translocates significant amounts of endogenous effector proteins. The The risk that Λpass mutant strains develop pathogenic properties is thus almost completely eliminated.

The solution according to the invention also extends the spectrum of bacterial transport systems and enables individual proteins to be introduced into eukaryotic cells by the TTSS of group 2 bacteria. This is an enormous advantage, because depending on the protein to be transported, only a certain type of TTSS guarantees an optimal translocation into the target cell. Thus, according to the invention, the TTSS of a strain of the same genus can be used for the targeted transport of effector proteins from bacteria of the genus Xanthomonas.

The present invention not only applies to the host plants of the Λpass mutants used, but can also be carried out with other dicotyledonous or monocotyledonous plants or other eukaryotic target cells such as, for example, yeast cells. The method according to the invention is used to transport bacterial proteins into eukaryotic cells in order to Modify cellular processes in a targeted manner. In general, all effector proteins can be used according to the invention which influence plant processes such as growth, development and defense / resistance mechanisms or which interfere with fundamental cellular processes in eukaryotic host cells. In addition to bacterial effector proteins, other proteins can also be transported into eukaryotic cells, provided they are fused to a functional bacterial type III secretion signal. Any protein or mRNA sequence that enables the secretion of a protein by the TTSS used is referred to as type III secretion signal. Suitable type III secretion signals are typically found in all proteins secreted by the TTSS within the N-terminal 50-100 amino acids. For example, the N-terminal 51 amino acids of XopA can be used according to the invention as a type III secretion signal for the transport of a protein without its own type III secretion signal. For the expression of fusion proteins, fusions between the coding sequence of the type III secretion signal and the protein to be transported are, for. B. prepared by ligation. For the fusion, the two sequences must be in the same reading frame. In addition, the ligation must be carried out in a manner that allows an optimal and stable expression of the encoded fusion protein. There may also be an intermediate area between the type III secretion signal and the gene to be fused. The resulting fusion construct can be expressed in an expression vector under the control of an endogenous promoter of the bacterium or under the control of a constitutive or inducible promoter in the expression plasmid. The expression and secretion of the constructed fusion protein can be determined using Secretion analyzes and using appropriate antibodies against epitopes of the protein to be analyzed can be checked.

Examples of applications for proteins that could be translocated by TTSS of Λpass mutants after fusion to a type III secretion signal are those proteins which intervene in important cellular processes and metabolic pathways and thereby, for example, promote growth in plants or resistance to pathogenic organisms or herbicides. In general, according to the invention, the transport of all those proteins that modify cell-specific or heterologously expressed proteins in eukaryotic cells or modify the DNA organization or gene expression is of interest.

Another possible application of the invention is the transport of proteins into the extracellular environment of a host cell. Proteins that either have their own type III secretion signal or are fused to a type III secretion signal that is functional in the bacterial system used can be used for this. However, it must be ensured that the proteins to be transported are secreted by the TTSS, but are not translocated into the plant cell. One way to do this is to use Λpass / ΛφF double mutants. The mutation in hrpF prevents the translocation of type III-dependent secreted proteins into the host cell, but does not impair the protein secretion by the TTSS into the extracellular medium 181 pass-mutants mediate secretion and translocation. In the event that the secretion and translocation signals are locally separated from one another, a signal can be generated by shortening or deleting in the area of the translocation signal, which in ftppass mutants mediates the secretion but not the translocation. In addition to the previously mentioned possible uses, the invention can also be used after in-culture cultivation of the bacteria for protein transport into the extracellular medium. Secreted proteins can be purified after the bacteria have been separated from the culture medium. For this purpose of use, the secretion of the protein to be transported must be guaranteed by the TTSS. The proteins can have their own type III secretion signal or can be fused to a heterologous type III secretion signal, which may be signals which, in hpaB mutants, only mediate secretion by the TTSS or else also translocation.

All pathogenic bacteria which express a pathφ gene cluster and hpaB or homologous genes can be used as starting material for the method according to the invention. The sequence of hpaB from X. campestris pv. Vesicatoria is available in the database. The invention includes the production of mutants corresponding to the Λpass gene in all Gram-negative bacteria that express hpaB or homologous genes. The invention further relates to the heterologous expression of Λrp gene clusters mutated in hpaB or a homologous gene in non-pathogenic bacteria. Since hpaB and homologous genes have so far only been found in group 2 bacteria, it should be noted when generating such a heterologous system that in addition to the hrp gene cluster, the regulatory proteins required for Λφ gene expression (which are not found in group 2 bacteria ftφ gene clusters are encoded) must be expressed or otherwise the expression of the Λφ genes must be guaranteed.

embodiments

1. Bacterial strains and plasmids £. coli DH5oc was purchased from Bethesda Research Laboratories, Bethesda, Md. The X. campestris pv. Ves / cator / a strains 85-10, 85 * and 82 * are described in references / 16 / and / 17 /. X. campestris pv. Ves / cafor / a strains were grown at 30 ° C in NYG medium / 18 /. The plasmid pBlueskript (II) KS used for cloning was purchased from Stratagene, Heidelberg, Germany. The expression vectors pDSK604 and pLAFR6 are described in Escolar et al. / 19 / or in Bonas et al. / 20 / described. Plasmids were introduced into E. coli by electroporation and into X. campestris pv. Vesicatoria by conjugation / 21 /. Antibiotics were used in the following concentrations: ampicillin, 100 μg / ml; Rifampicin, 100 µg / ml; Spectinomycin, 100 ug / ml and tetracycline, 10 ug / ml.

2. Plant material and infiltrations

The Early Cal Wonder (ECW), ECW-10R and ECW-30R 1221 pepper cultivars were grown as described in Bonas et al. / 23 / described and infiltrated with X. campestris pv. Ves / cator / a strains. Bacterial suspensions were infiltrated into paprika leaves in concentrations of 2x10 ⁸ bacteria / ml (corresponding to an optical density of 0.2 at a wavelength of 600 nm) in 1 mM MgCl ₂ . The occurrence of disease symptoms or HR was analyzed within a period of one to three days after infiltration. For investigations of the bacterial growth in the plant, bacterial suspensions in concentrations of 10 ⁴ cfu / ml in 1 mM MgCl _{2 were} infiltrated into leaves of the paprika cultivar ECW / 23 /. 3. Production of a Λpass mutant

To create a mutation in the Λpass gene, the 3.1 kb ΛφE region was cloned into the EcoRV and XΛol cleavage site from pBluescript (II) KS. A 420 bp fragment of hpaB was then cut out using Csp45l and the remaining construct was religated. The deletion of the Csp45l fragment leads to the deletion of amino acids 13 to 149 in the Λpass gene product. The resulting 2.7 kb insert was then cloned into the β HI / Sa / l sites of the suicide vector pOK1 / 12 / and the resulting hpaB deletion construct into the X. campestris pv. Ves / ^' cator / a strains 85- 10, 85 * and 82 * conjugated. The exchange of the wild-type gene for the deletion in hpaB by homologous recombination was carried out by means of PCR ("polymerase chain reaction") -based analyzes using the primers hpaB-for (5'- CGAATTCGTCCATGTCTCACCACAGATC-3 ') and hpaB-rev ( 5 '

CGAGCTCGGCGCGTAACCACAGATAGTT-3 ') checked. The resulting Λpass mutants were characterized by phenotypic analyzes. For this, bacterial suspensions were inoculated into leaves of susceptible and resistant plants. As shown in Figure 1, the deletion of hpaB results. in strain 85-10 for the complete loss of bacterial pathogenicity, ie the ability to grow on susceptible plants and trigger disease symptoms. Furthermore, the strain 85-10ΔΛpass is no longer able to induce HR in resistant plants. Strain 85 * AhpaB is also no longer pathogenic on susceptible plants, but still induces partial HR on resistant plants. Comparable phenotypes were observed when inoculation of strain 82 ^* AhpaB in leaves of corresponding susceptible and resistant plants.

4. Construction of AvrBs3Δ2 fusion proteins

To produce an HrpF-AvrBs3Δ2 expression construct, a 1.1 kb EcoRI fragment which contains the first 387 codons of hrpF was ligated into the EcoRI site of plasmid pDSF356F / 10 /. The resulting construct pDhrpFN356, which expresses HrpF ^ ₃₈ -AvrBs3Δ2, was conjugated into the X. campestris pv. Ves / cator / a strains 85 * and 85 * ΔΛpass. In order to produce a second HrpF-AvrBs3Δ2 expression construct, the first 200 codons of hrpF were PCR by means of the primers HrpF-for (5'- TACTGAATTCGCCTCTATGTCGCTC-3 ') and HrpF-200rev (5'-

CTGTCGAATTCGATCTTGCCGCCGCACTTG-3 ') amplified and ligated into the EcoRI site of pDSF356F. The resulting construct pDS200F356F expresses HrpF ^ oo-AvrBs3Δ2 and was also conjugated into the X. campestris pv. Ves / cafo / va strains 85 * and 85 * ΔΛpass. To make a XopA ₁ . ₅ rAvrBs3Δ2 expression construct, the first 51 codons of xopA and 680 bp of the region upstream of the translation initiation codon of xopA were PCR and the primers xopAfor (5'-CACCGTACCGTTGTTGTTGCGATG-3 ') and xopAGGGTCTC-3GGAVG (3') 'amplified (5). The resulting PCR product was ligated into the donor vector pENTR / D-TOPO (available from Invitrogen, Carlsbad, Calif.), Which contains aft sites, using a topoisomerase and then recombined into the target vector pL6GW356. pL6GW356 has an aftP reading frame cartridge fused to avrBs3A2 and was described in Noel et al. / 11 /. The resulting construct pL6xopA356 which Xopa ₁ - ₅₁ -AvrBs3Δ2 expressed, was conjugated into the X. campestris pv ves / cafoπ ^'a-85 strains * and 85 * AhpaB..

5. Type HI-dependent protein secretion in Λpass mutants

For the analysis of type III-dependent protein secretion in Λpass mutants, the X. campestris pv. Ves / cafor / ^' a strains 82 * and 82 * ΔΛpass were incubated in secretion medium. The protein total extracts were then separated from the culture supernatants and analyzed by SDS polyacrylamide electrophoresis and immunoblot using antibodies against HrpF and AvrBs3 124,251. HrpF was present in comparable amounts in protein total extracts and culture supernatants of both strains (see Figure 2). In contrast, AvrBs3 could not be detected in the culture supernatant from strain 82 * AhpaB, although it was detected in protein total extracts from both strains and in the culture supernatant from strain 82 *. To analyze the secretion of further effector proteins, avrBs1-c-myc and avrBs Tc-myc expression constructs / 19 / were conjugated into the X. campestris pv. Ves / cafoπa strains 85 * and 82 * and into the corresponding ftpass mutants. Since the proteins secreted by the TTSS under in v / fro conditions can only be detected immunologically in the culture supernatant, one has to rely on the availability of specific antibodies against epitopes of the proteins to be analyzed in secretion analyzes. For this reason, the effector proteins AvrBsl and AvrBsT were provided with a C-terminal c-myc epitope. Both AvrBsl-c-myc and AvrBsT-c-myc were detected in the total extracts of / ipaß wild-type and ftpaß mutant strains in comparable amounts by an anti-c-myc antibody (Amersham Pharmacia Biotech, Freiburg, Germany). Furthermore, both proteins were detected in the culture supernatant of the Λpass wild-type strain, however, were not or only weakly detectable in the culture supernatant of Λpass mutants (see Figure 2). In order to demonstrate that the detection of proteins in the culture supernatant is not due to lysis of the bacterial cells, the immunoblots were incubated with an antibody against the intracellular protein HrcN / 26 /, which was only detected in protein total extracts. The experimental observations described clarify that HpaB is involved in the secretion of effector proteins, while it presumably has no significant influence on the export of non-effector proteins by the TTSS. In addition to HrpF, the non-effector protein XopA was also detected in comparable amounts in culture supernatants from X. campestris pv. Ves cator / ^' a strain 85 * and 85 * ΔΛpass (see Figure 2). The hypothesis was also confirmed by secretion analysis of X. campestris pv. Ves / cator / a strains 85 ^* and 85 * AhpaB, which contained the fusion proteins HrpF- ^ oo-AvrBs3Δ2 and XopA ₁ . ₅₁ -AvrBs3Δ2 were confirmed. So both HrpF- |. _200- AvrBs3Δ2 as well as XopAι. ₅ ι-AvrBs3Δ2 detected in total extracts and culture supernatants of Λpass wild-type and Λpass mutant strains (see Figure 3). These results show that the N-terminal protein regions of HrpF and XopA contain secretion signals which mediate the Λpass-independent secretion of the AvrBs3Δ2 reporter protein. The type III-dependent secretion of the reporter protein AvrBs3Δ2 can therefore be carried out independently of the presence of HpaB.

6. Analysis of protein translocation in Λpass mutants

For the analysis of the protein translocation in Λpass mutants, AvrBs3 and AvrBs3Δ2 fusion proteins were expressed in the X. campestris pv. Ves / cafor / a strains 85 * and 85 * AhpaB and the bacteria were inoculated into leaves of the pepper cultivar ECW-30R. ECW-30R plants contain the resistance gene Bs3, which mediates the specific recognition of AvrBs3 and then induces defense reactions associated with HR / 22 /. Furthermore, the resistance protein Bs3 also recognizes AvrBs3Δ2 fusion proteins, provided that these are expressed transiently, for example, via i4groόacter / ^' um-mediated gene transfer in plant cells or are translocated by X. campestris pv. Vesicatoria. The presence of the nuclear localization sequences in AvrBs3 or AvrBs3Δ2 fusion proteins is essential for the detection by Bs3. It can be assumed that AvrBs3 and AvrBs3Δ2 fusion proteins have to be transported into the cell nucleus in order to induce the Bs3-specific HR. X. campestris pv. Ves / ^' cafo / va strain 85 * (pDS300F), which contains the avrßs3 expression construct / 27 /, induces the Bs3-specific HR on ECW-30R plants. In contrast, 85 * strains induce the HrpF ₁ . ₃₈₇ -AvrBs3Δ2, HrpF ₁ . ₂₀₀ -AvrBs3Δ2 or Xopa ₁ - ₅ expressing AvrBs3Δ2, disease symptoms ECW 30R plants. This shows that the N-terminal regions of HrpF and XopA have no export signal, which mediates the translocation of the AvrBs3Δ2 reporter in strain 85 * (see Figure 4). Become HrpF ^^ - AvrBs3Δ2, HrpF ₁ - ₂₀₀ -AvrBs3Δ2 or XopA ₁ . ₅₁ -AvrBs3Δ2, however, expressed in the Λpass mutant strain 85 * AhpaB, they induce the Bs3-specific HR after infiltration in ECW-30R plants. This shows that the N-terminal protein regions of HrpF and XopA are the Can translocate AvrBs3Δ2 reporter protein in the absence of HpaB. In addition to AvrBs3Δ2 fusion proteins, AvrBs3 is also translocated into the plant cell upon overexpression of plamide pDS300F from strain 85 * AhpaB. For example, strain 85 * Δ / 7pass (pDS300F) induces Bs3-specific HR in ECW-30R plants (see Figure 4). In summary, it has been shown that ftpaß mutants are able to translocate significant amounts of an effector protein into the plant cell if this effector protein is overexpressed. Amounts of protein that are sufficient to induce complete, non-partial HR in the corresponding resistant plant are termed significant here. In addition to effector proteins after overexpression, hpaB mutants can also translocate fusion proteins which contain the N-terminal protein region of a non-effector protein such as, for example, HrpF or XopA.

Explanation of the reference symbols

85-10 X. campestris pv. Ves / cator / a strain 85-10

85 * X. campestris pv. Ves / cafora strain 85 ^*

85-10Δ 7pass Λpass deletion mutant from X. campestris pv. Ves / cafor / a strain 85-

10

85 * AhpaB Λpass deletion mutant from X. campestris pv. Ves / cator / a strain 85 *

85-10Δ / 7φG non-pathogenic X. campestris pv. Ves / cafor / a strain 85-1 OAhrpG, with a mutation in the regulator gene hrpG known as watery lesions disease symptoms no visible disease symptoms hr partial HR (hypersensitive reaction)

wt wild type strain

AhpaB frpass-mutant strain

HrpF type III translocon protein, type III-secreted, not translocated

Wild-type strain

AvrBs3 effector protein, type III-dependent secreted and translocated HrcN intracellular protein of the TTSS, not secreted AvrBsl -c-myc effector protein with C-terminal c-myc epitope, type III-dependent secreted and translocated

AvrBsT-c-myc effector protein with C-terminal c-myc epitope, type III-secreted and translocated

XopA potential type III translocon protein, type III-secreted, not translocated in the wild-type strain

XopA ₁ . _51- AvrBs3Δ2 fusion protein between the N-terminal 51 amino acids of XopA and the deletion derivative AvrBs3Δ2, which is deleted in the N-terminal 152 amino acids

HrpF ₁ . ₃₈₇ AvrBs3Δ2 fusion protein between the N-terminal 387 amino acids of HrpF and the deletion derivative AvrBs3Δ2 HrpF ₁ . ₂₀₀ AvrBs3Δ2 fusion protein between the N-terminal 200 amino acids of HrpF and the deletion derivative AvrBs3Δ2 literature

IM Cornelis and Van Gijsegem (2000) Annu. Rev. Microbiol., 54, 735-774

121 Lindgren (1997) Ann. Rev. Phytopathol., 35, 129-152

/ 3 / Alfano and Collmer (1996) Plant Cell, 8, 1683-1698

14 / Dangl and Jones (2001) Nature, 411, 826-833

15 / Gopalan et al. (1996) Plant Cell, 8, 1095-1105

/ 6 / Pirhonen et al. (1996) Mol. Plant-Microbe Interact., 9, 252-260

171 Harn et al. (1998) Proc. Natl. Acad. Be. USA, 95, 10206-10211

181 Büttner and Bonas (2002) EMBO J., 21, 5313-5322

19 / Noel et al. (2002) J. Bact., 184, 1340-1348

/ 10 / Szurek et al. (2002) Mol. Microbiol., 46, 13-23

/ 11 / Noel et al. (2003) J. Bact., 185, 7092-7102

/ 12 / Huguet et al. (1998) Mol. Microbiol., 29, 1379-1390

/ 13 / Parsot et al. (2003) Curr. Op. Microbiol., 6, 7-14

/ 14 / Sambrook et al. (1989) Molecular Cloning: A Laboratory Manual, 2nd edition, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, USA

/ 15 / Bonas et al. (1989) Mol. Gen. Genet, 218, 127-136

/ 16 / Canteros (1990) Diversity of plasmids and plasmid-encoded phenotypic traits in Xanthomonas campestris pv. Vesicatoria. Dissertation. University of Florida

/ 17 / Wengelnik et al. (1999) J. Bacteriol., 181, 6828-6831

/ 18 / Daniels et al. (1984) EMBO J., 3, 3323-3328

/ 19 / Escolar et al. (2001) Mol. Plant Pathol., 2, 287-296

/ 20 / Bonas et al. (1989) Mol. Gen. Genet, 218, 127-136

/ 21 / Figurski and Helsinki (1979) Proc. Natl. Acad. Sci., 76, 1648-1652

/ 22 / Minsavage et al. (1990) Mol. Plant-Microbe Interact, 3, 41-47

/ 23 / Bonas et al. (1991) Mol. Plant-Microbe Interact, 4, 81-88

/ 24 / Rossier et al. (1999) Proc. Natl. Acad. Be. USA, 96, 9368-9373

/ 25 / Büttner et al. (2002) J. Bacteriol., 184, 2389-2398

/ 26 / Rossier et al (2000) Mol. Microbiol., 38,828-838

/ 27 / van den Ackerveken et al. (1996) Cell, 87, 1307-1316

Claims

claims

1. Bacterial system for protein transport in eukaryotic cells consisting of the type III secretion system and a / 7pass mutant bacterial strain, where hpaB denotes a gene in the Λφ gene cluster, or a bacterial strain which contains a mutation in a Λpassport homologous or functionally analogous gene ,

2. Transport system according to claim 1, characterized in that the organism used is a Gram-negative bacterium.

3. Transport system according to claim 1, characterized in that the organism used is cultivated in vitro.

4. Transport system according to claim 1, characterized in that the organism used is cultivated together with eukaryotic cells.

5. Transport system according to claim 1, characterized in that the organism used with plants e.g. is brought into contact by infiltration, dipping or spraying.

6. Transport system according to claim 1, characterized in that the mutation of hpaB or of a homologous gene prevents the expression of the respective gene product or modifies it so that its functionality is reduced or lost.

7. Transport system according to claim 1, characterized in that the mutation of hpaB or of a homologous gene leads to a reduction in the secretion of effector proteins by the type III secretion system.

8. Transport system according to claim 1, characterized in that hpaB or a homologous gene is mutated in the genome of the bacterial strain used.

9. Transport system according to claim 1, characterized in that the genes encoding the type III secretion system are expressed from an expression plasmid which contains a mutation in hpaB or a homologous gene.

10. Transport system according to at least one of claims 1 to 9, characterized in that the bacterial strain used contains a recombinant nucleic acid molecule in which an active promoter is operatively linked to a nucleic acid sequence which encodes an effector protein.

11. Transport system according to at least one of claims 1 to 10, characterized in that the coded effector protein contains a type III secretion signal which is functional in the bacterial strain used.

12. Transport system according to at least one of claims 1 to 9, characterized in that the bacterial strain used contains a recombinant nucleic acid molecule in which an active promoter is operatively linked to a nucleic acid sequence which is a fusion protein between a type III secretion signal and the protein to be transported coded.

13. Transport system according to at least one of claims 1 to 9 and 12, characterized in that the fusion protein can contain an amino acid sequence between the type III secretion signal and the protein to be transported.