Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

The Dark Side of the Salad: Salmonella typhimurium Overcomes the Innate Immune Response of Arabidopsis thaliana and Shows an Endopathogenic Lifestyle

  • Adam Schikora,

    Affiliation Unité de Recherche en Génomique Végétale, Institut National de la Recherche Agronomique/Centre National de la Recherche Scientifique/University of Evry Val d'Essonne, Evry, France

  • Alessandro Carreri,

    Affiliation Department of Plant Molecular Biology, Max F. Perutz Laboratories, Vienna, Austria

  • Emmanuelle Charpentier,

    Affiliation Department of Microbiology and Immunobiology, Max F. Perutz Laboratories, Vienna, Austria

  • Heribert Hirt

    Affiliations Unité de Recherche en Génomique Végétale, Institut National de la Recherche Agronomique/Centre National de la Recherche Scientifique/University of Evry Val d'Essonne, Evry, France, Department of Plant Molecular Biology, Max F. Perutz Laboratories, Vienna, Austria


Salmonella enterica serovar typhimurium contaminated vegetables and fruits are considerable sources of human infections. Bacteria present in raw plant-derived nutrients cause salmonellosis, the world wide most spread food poisoning. This facultative endopathogen enters and replicates in host cells and actively suppresses host immune responses. Although Salmonella survives on plants, the underlying bacterial infection mechanisms are only poorly understood. In this report we investigated the possibility to use Arabidopsis thaliana as a genetically tractable host system to study Salmonella-plant interactions. Using green fluorescent protein (GFP) marked bacteria, we show here that Salmonella can infect various Arabidopsis tissues and proliferate in intracelullar cellular compartments. Salmonella infection of Arabidopsis cells can occur via intact shoot or root tissues resulting in wilting, chlorosis and eventually death of the infected organs. Arabidopsis reacts to Salmonella by inducing the activation of mitogen-activated protein kinase (MAPK) cascades and enhanced expression of pathogenesis related (PR) genes. The induction of defense responses fails in plants that are compromised in ethylene or jasmonic acid signaling or in the MKK3-MPK6 MAPK pathway. These findings demonstrate that Arabidopsis represents a true host system for Salmonella, offering unique possibilities to study the interaction of this human pathogen with plants at the molecular level for developing novel drug targets and addressing current safety issues in human nutrition.


Salmonella enterica serovar typhimurium (S. typhimurium) is a facultative endopathogen and the causative agent of various human diseases ranging from enteritis to typhoid fever. It is responsible for salmonellosis, which is the most frequent food-borne disease with around 1.5 billion yearly infections world wide (WHO). Disease in mammals occurs after consumption of contaminated food or water. Systemic infection can follow, which depends on the ability of the bacteria to survive the harsh conditions of the gastric tract before crossing the intestinal epithelium. Owning to its importance, the mechanism of Salmonella invasion of human cells is under intense study. Two Salmonella pathogenicity islands (SPI-1 and SPI-2) encoding structural elements of type III secretion systems (T3SS) and effectors injected into the host cells, are necessary for entry and proliferation within mammalian cells [1]. Infection occurs in several well-organized and adjusted steps. The first one includes docking of Salmonella to the epithelial cell and injection of SPI-1 encoded effectors, which suppress the host immune system and modify the actin and tubulin cytoskeleton [2]. Endocytosis is the second step and requires formation of Salmonella Containing Vacuoles (SCVs). Retainment of SCVs in the host cytoplasm is assured by SPI-2 encoded effectors [3] and strains not capable to sustain intact SCVs are avirulent [4].

Salmonella-contaminated vegetables and fruits were recently identified as a widespread source of human infection [5]. Although diverse plant species support growth of Salmonella [6], the underlying molecular mechanisms of the Salmonella–plant interaction are largely unknown.

The plant immune system functions at different levels during pathogen attack. Pathogen-associated molecular patterns (PAMPs) and effectors injected into plant cells, trigger activation of defence mechanisms [7], [8], [9], [10]. PAMPs are recognized by plant receptors and trigger a defence mechanism referred to as “basal” defence [11]. On the other hand, pathogens have developed mechanisms to overcome detection by injecting effectors into plant cells, which interfere with signalling cascades and thereby abolish basal defence response. In some cases, pathogen effectors can be recognized by plant resistance proteins (R proteins), triggering a hypersensitive response (HR) and thereby limit pathogen infection [12]. Systemic acquired resistance (SAR) represents another level of defence resulting in resistance to a broad spectrum of pathogens throughout the plant [13]. SAR requires signal molecules activating the expression of pathogenesis-related (PR) genes and salicylic acid (SA), jasmonic acid (JA) and ethylene (ET) are involved in regulating their expression. These molecules play important roles in the induction of plant defence responses upon pathogen attack and are implicated in different forms of resistance [14], [15]. While SA-dependent pathways are generally important in defence against biotrophic pathogens and lead to hypersensitive response (HR) and/or local resistance [13], JA and ET pathways seem to be involved in defence mechanisms against herbivore attack and necrotrophic pathogens [16].

Mitogen-activated protein kinase (MAPK) cascades play a crucial role in mediating defence responses to plant pathogen attack [17]. In Arabidopsis, MPK3 and MPK6 are activated by different bacterial elicitors [18], [19] and trigger enhanced expression of PR genes [7], [20]. Silencing of MPK6 results in compromised resistance to plant pathogens [21]. MEKK1, MKK4/5, MPK3/6/4 were identified downstream of the FLS2 receptor recognizing bacterial flagellin [7] and are also involved in regulating reactive oxygen species (ROS) production [22], [23], [24]. Recently, MKK3 was shown to have a dual function; activating either MPK6 in response to JA signalling [25], or MPK7 in response to ROS and plant pathogens [26].

Comparing the defence response to classical plant pathogens, the response to human pathogens is much less understood. The human pathogen S. aureus triggers SA-dependent defence responses in Arabidopsis plants [27] and its spreading can be diminished by SA treatment. In addition, the SA-depleted isochorismate synthase defective ics1 mutants and transgenic Arabidopsis lines, overexpressing the bacterial SA hydroxylase (NahG), are hypersensitive to S. aureus infection [27]. Recently, it was reported that treating plants with the ethylene precursor ACC, significantly diminishes colonization of Medicago sativa and Arabidopsis by Klebsiella pneumoniae or Salmonella [28].

The present study shows that the human pathogen S. typhimurium triggers the activation of plant immune responses including enhanced transcription of PR genes. We show that Salmonella can overcome plant defence mechanisms and enter and proliferate inside various Arabidopsis tissues, causing wilting and chlorosis as disease symptoms. Among different possible immune responses of Arabidopsis, the ET- and JA-dependent signalling pathways were found of major importance for inducing defence responses during S. typhimurium infection. Moreover, the JA-mediating MKK3-MPK6 MAPK pathway was identified to be essential for restricting Salmonella proliferation and disease in Arabidopsis plants. Our results indicate that Arabidopsis can serve as a valuable model system to study Salmonella-host interaction, suggesting the necessity to modify agricultural practices to improve food safety


Salmonella is pathogenic to Arabidopsis

Three different types of experiments were performed to assess the question whether Salmonella can actively invade, proliferate and spread through plants. To test whether Salmonella is capable to proliferate inside plants, whole rosettes of Arabidopsis thaliana Col-0 wild-type plants were vacuum-infiltrated with S. typhimurium wild-type strain 14028s and the internal bacterial population was then counted over the following four days. The number of colony forming units (cfu) in discs excised from infiltrated leaves drastically increased over the first 2 days before reaching plateau levels (Fig. 1a). Two weeks after dipping the plants in bacterial solutions, invasion of Salmonella into Arabidopsis caused a defined disease phenotype as revealed by severe chlorosis and wilting of leaves (Fig. 1c–d). To evaluate whether Salmonella is also able to actively recognize and invade Arabidopsis, liquid media in which Arabidopsis seedlings were immerged, were inoculated with bacteria. Over the following days, the bacterial population inside seedlings was monitored after surface-sterilization and homogenisation of the seedlings (Fig. 1b). A 40-fold increase in cfu of internal bacteria was observed over a period of 2 days (Fig. 1b). Altogether these results indicate that Salmonella can actively invade and proliferate in Arabidopsis plants and cause disease.

Figure 1. Salmonella can invade and grow in Arabidopsis cells and induce plant innate defence mechanisms.

a–b, Three weeks old A. thaliana plants were vacuum-infiltrated (a) and 14 days old A. thaliana seedlings were incubated in medium inoculated (b) with S. typhimurium wild type 14028s. Proliferation of Salmonella was analyzed over a period of 4 (a) and 2 (b) days by determination the cfu of internal bacteria in homogenates of leaf discs (a) or seedlings (b). c–d; Disease symptoms after dipping in Salmonella. Three weeks old A. thaliana plants were dipped for 5 min in S. typhimurium 14028s solution and macroscopic changes were observed over a 2 weeks period, c: before dipping, d: 14 days after dipping in bacterial solution. e–h, Salmonella enters plant cells. GFP marked S. typhimurium 14028s were localized in A. thaliana root cells after 3 (e) or 20 h (f, g). h; 3 h after infection, cell culture protoplasts contain bacterial cells. Images were taken using a confocal microscope with 488 nm excitation and 505–530 nm emission filters. i, Treatment with Salmonella induces MPK3 and MPK6 kinase activities. Two weeks old A. thaliana seedlings were treated with either 10 mM MgCl2 or S. typhimurium 14028s. Endogenous MPK6 was immunoprecipitated from total protein extraction. Myelin basic protein (MBP) was used as substrate. A, activity; P, protein amounts were detected by Western blotting with antibodies specific for MPK3/6. j, Treatment with Salmonella induces defence responses. 14 days old A. thaliana seedlings were treated with S. typhimurium 14028s. Quantitative RT-PCR analysis was performed on 5 µg reverse transcribed total mRNA; clathrin, ubiquitin4 and actin2 were used for normalization.

Most pathogenic bacteria can spread through susceptible host plants very rapidly. We investigated whether Salmonella can also systemically infect plants. For this purpose we selectively infected roots with Salmonella and analyzed its presence in leaves two weeks later. While bacteria were growing on and in Arabidopsis roots two weeks after infection, we were not able to detect Salmonella in leaf homogenates from these plants (data not shown). Alternatively, to determinate whether Salmonella can spread from leaf-to-leaf, we syringe-infiltrated single leaves and monitored the presence of bacteria in non-infiltrated neighbouring leaves two weeks later. Although Salmonella effectively infected single leaves, no bacteria were detected in non-infiltrated neighbouring leaves (Fig. S1). These data indicate that S. tyiphimurium is not able to spread to other non-exposed parts of the Arabidopsis plant, suggesting that Salmonella cannot migrate via the xylem (root-to-shoot) nor via the phloem (leaf-to-leaf) systems.

From the epidemiological standpoint, it is important to know the duration of Salmonella persistence in infected plants. We therefore determined how long Salmonella could stay alive in Arabidopsis plants. For this purpose, whole rosettes of three weeks old Arabidopsis plants were vacuum-infiltrated and the presence of internal bacteria was monitored over the period of one month. Although the infiltrated leaves died within 5 days, apical meristems survived the infection and produced new leaves. One month after infiltration, newly formed leaves contained significant albeit much lower Salmonella cfu, possibly indicating that meristems are poor Salmonella targets or that meristematic tissues are particularly resistant to bacterial infection (Fig. S2).

Endopathogenic lifestyle of Salmonella in Arabidopsis

In animals and humans, Salmonella actively enters epithelial and other cells in order to replicate and spread through the organism. We focused our attention on the question whether, similar to the situation in mammals, Salmonella can also invade plant cells. For this purpose, S. typhimurium was transformed with a plasmid constitutively expressing green fluorescent protein (GFP) [29]. Three hours post-inoculation of liquid medium with immerged seedlings, GFP-marked Salmonella were localized inside root hairs (Fig. 1e) and at 20 hours post-inoculation, inside rhizodermal cells (Fig. 1f–g). At this time point, large numbers of motile bacteria were observed inside single host cells, confirming that Salmonella can proliferate in plant cells. To our knowledge, this is the first report of an infection of the plant cytoplasm by a human enteropathogen. Salmonella was also found to form biofilm-like structures on the surface of roots and leaves, preferentially colonizing regions around emerging lateral roots and wounded tissues (data not shown). To extend our in planta observations to the single cell level, inoculation experiments were also performed using Arabidopsis protoplasts. Probably due to the high sucrose concentration in the medium, infection rates in this system were low. However, protoplasts were clearly invaded by GFP-marked Salmonella within 3 hours (Fig. 1h). Optical Z-stacking allowed the exact positioning of the GFP-marked Salmonella to the cytoplasmic compartment of infected protoplasts (Fig. S3). These data demonstrate that Salmonella has the ability to enter and proliferate inside plant cells.

Arabidopsis induces defence responses upon infection with S. typhimurium

Under normal conditions, plants react to pathogens by activating various defence responses, thereby diminishing the proliferation of bacteria, fungi or viruses. We tested whether plants recognize Salmonella as a pathogen and actively induce defence responses in order to limit infection, as observed with other bacterial plant pathogens [30]. In Arabidopsis, a variety of PAMPs were shown to activate the MAPKs: MPK3 and MPK6 [18], [19], followed by the transcription of a number of PR genes [7], [20]. To determine whether these defence responses also occur upon S. typhimurium infection, we used MPK3 and MPK6 specific antibodies to analyze the activation of these kinases in seedlings exposed to Salmonella. The activities of both MPK3 and MPK6 strongly increased at 15 min after contact with the bacteria (Fig. 1i). To examine whether recognition of Salmonella results in activation of other MAPKs, the activities of 9 MAPKs, representing all 4 phylogenetic groups of the MAPK family in Arabidopsis [31], were analyzed after infection of protoplasts with S. typhimurium. From the nine investigated kinases, only MPK3 and MPK6 showed activation, confirming our previous in planta results. Activities of two other MAPKs; MPK2 and MPK17 were reduced (Fig. 2). Interestingly, Arabidopsis appeared to sense multiple Salmonella PAMPs, because S. typhimurium 14028s infection of an A. thaliana fls2–17 mutant, which is defective in flagellin perception [20], still resulted in MPK6 activation (Fig. 3).

Figure 2. Activation of MAP kinases transiently expressed in protoplasts.

MPK3 and MPK6 are activated and MPK2 and MPK17 are inhibited upon infection with Salmonella. Nine of 20 MAPKs were expressed in cell culture derived protoplasts for 24 h and protoplasts were exposed to bacteria for 20 min. All MAPKs were expressed as HA-tagged versions and immunoprecipitated for 2 h with 2 µl HA specific antibody and 25 µl protein A-Sepharose beads. Kinase activity assays were performed as described above.

Figure 3. Arabidopsis responds to multiple Salmonella-derived PAMPs.

Plants can recognize several Salmonella-derived molecules, and this recognition leads to defence mechanisms. We tested whether Arabidopsis, in order to activate its response to Salmonella, relies on the most prominent flagellum and/or other bacterial effectors. The native MPK6 can be activated by S. typhimurium wild type 14028s in A. thaliana fls2–17 mutant, which is not able to recognize the flagellar protein flagellin. This result shows that the interaction between plants and Salmonella involves other signalling molecules, probably lipopolysaccharide (LPS), lipoproteins or other effectors. Kinase activity assays were performed on equal amounts of total protein extracted from 14 days old plants immunoprecipitated with MPK6 specific antibody. The A. thaliana wild type Ler was used as control in the experiment with the A. thaliana fls2–17 mutant. Flg22 is a 22 amino-acid peptide derived from flagellin.

To study the role of the plant innate immune system during Salmonella infection in more detail, we turned our attention to the SA-, JA-, and ET-dependent signalling pathways. To clarify their role during infection of Arabidopsis with Salmonella, quantitative RT-PCR analysis of marker genes for SA-, JA- and ET-dependent defence gene expression was performed. PDF1.2, encoding an antifungal defensin, is commonly used as a marker for JA- and ET-induced resistance [32]. In A. thaliana Col-0 infected with Salmonella, expression of the PDF1.2 gene increased rapidly during the first 6 h (Fig. 1j). Expression of other JA-regulated genes, such as PR2 and PR4 [33] as well as the SA marker PR1 gene were also induced during Salmonella infection (Fig. 1j, Fig. 4). These results show that Arabidopsis reacts to Salmonella infection by activating multiple defence pathways of the innate immune system.

Figure 4. Quantitative RT-PCR analysis of JA-inducible PR2 expression upon infection with Salmonella and Pseudomonas syringae.

Transcript amounts of PR2 (pathogenesis related2), similarly to PDF1.2, are induced by MeJA treatment. Transcript levels accumulate also after contact with Salmonella, pointing to a central role of the JA pathway in response to this bacterium. Moreover, the induction follows the kinetics of induction caused by the plant specialized pathogen P. syringae. Q-RT-PCR was performed as described above.

JA- and ET-dependent signalling pathways are involved in resistance against S. typhimurium

It was previously shown that lack of SA promotes and that treatment with the ethylene precursor ACC lowers Salmonella colonization of plants [28]. Therefore, we investigated whether defects in any of the three signalling pathways affect Salmonella's ability to infect Arabidopsis plants. We performed infection experiments with SA depleted NahG plants [34], ein2 (ethylene insensitive 2) [35] and coi1 (coronatine insensitive 1) [36], [37] mutant plants compromised in either the ET or JA signalling pathways, respectively. Liquid media with immerged seedlings were inoculated with S. typhimurium for 48 hours before determination of cfu of the internal bacteria. When compared with wild type A. thaliana Col-0 plants, coi1–16 and ein2-1 plants contained 35 and 50 times more Salmonella, respectively (Fig. 5e). In contrast, only a small difference in Salmonella infection levels was detected between wild type and NahG plants (Fig. 5e), indicating that the JA and ET pathways are important for resistance against S. typhimurium. Comparing the disease symptoms in coi1–16, ein2-1 and NahG plants challenged with Salmonella by dipping plants for 5 min into a bacterial solution, the JA-signalling pathway appeared to be of primary importance for Arabidopsis resistance to Salmonella infection. coi1–16 plants revealed severe senescence symptoms and wilting two weeks after dipping (Fig. 5c–d). Although ein2-1 plants showed even higher proliferation levels of Salmonella than coi1–16 plants (Fig. 5e), these plants did not shown enhanced disease symptoms (Fig. 5a–b). In contrast, NahG plants showed similar Salmonella proliferation rates and symptoms as Col-0 wild type plants (data not shown).

Figure 5. Jasmonate and ethylene signalling pathways are necessary for defence response to Salmonella.

a–d, Disease symptoms in JA and ET insensitive mutants ein2 (a,b) and coi1–16 (c,d) plants dipped for 5 min in S. typhimurium 14028s solution; (a, c) before dipping, (b, d) 14 days after dipping.e, coi1–16 and ein2 are susceptible to Salmonella invasion. 14 days old A. thaliana wild type Col-0, coi1–16 and ein2 mutants or NahG transgenic seedlings were inoculated with S. typhimurium 14028s for 48 h, and cfu of internal bacteria in the seedling homogenates were determined. f, Activities of native MPK6. MPK6 activity was poorly induced in coi1–16 plants after Salmonella treatment. Kinase activity assays were performed as described above. g–h, Quantitative analysis of JA-regulated PDF1.2 (g) and SA-dependent PR1 (h) expression upon Salmonella infection. Induction of PDF1.2 expression was abolished in coi1–16 and delayed in ein2. Expression of PR1 was not induced in coi1–16 or NahG plants. Quantitative RT-PCR analysis was performed on 5 µg reverse transcribed total mRNA. clathrin, ubiquitin4 and actin2 were used for normalization.

To better understand the different signalling events, we monitored native MPK6 kinase activities in coi1–16, ein2-1 and NahG plants upon Salmonella exposure. The activation pattern of MPK6 in ein2-1 plants was very similar to that observed in wild type A. thaliana Col-0 plants and strong transient MPK6 activation was observed at 15 min after bacterial contact (Fig. 5f). NahG plants also showed a similar activation pattern of MPK6, although total activities were generally lower (Fig. 5f). In contrast, the level of MPK6 activation was poorly above background and the activation peak was shifted towards 30 min in coi1–16 plants (Fig. 5f).

To monitor another level of the Arabidopsis innate immune system, the coi1–16, ein2-1 and NahG plants were also analysed for the expression of different defence response marker genes upon Salmonella treatment. In comparison to wild type Arabidopsis plants, coi1–16 mutants showed no enhanced accumulation of PDF1.2 and PR1 transcripts (Fig. 5g–h). In ein2-1 mutants, PDF1.2 and PR1 transcript accumulation was delayed upon Salmonella infection (Fig 5g–h). Salmonella treatment of NahG plants resulted in accumulation of PDF1 transcripts, but no induction of PR1 expression was detected (Fig. 5g–h). These results reveal that the Arabidopsis SA, ET and JA pathways differentially contribute to disease development and the induction of defence responses upon Salmonella infection.

The MKK3-MPK6 cascade plays a crucial role in resistance to Salmonella

Of the Arabidopsis mutants, coi1–16 was compromised in MPK6 activation, all defence responses and moreover showed enhanced disease symptoms (Fig. 5). These results identify the JA signalling pathway to be of outstanding importance in Salmonella-Arabidopsis interaction. COI1 is an F-box protein acting in an E3 ubiquitin ligase containing complex, which degrades the repressors of JA-responsive genes [38], [39]. The JA-dependent activation of MPK6 is consistent with recent data, suggesting that full activation of MPK6 by JA requires mitogen-activated protein kinase kinase 3 (MKK3) [25]. To test this model during Arabidopsis-Salmonella interaction, we analyzed the Salmonella-induced activation of native MPK6 in mkk3 mutants (Fig. 6a). Like in mpk6 knock-out plants, no activation of MPK6 was detected in mkk3 mutants, revealing that Salmonella induces MPK6 by MKK3 activation. A corollary of these results suggests that MPK6 might play an important role in defence against Salmonella infection. In agreement with this hypothesis, mpk6 plants were significantly more susceptible to Salmonella infection, showing a 17-fold higher cfu after 24 hours of culture in liquid medium inoculated with Salmonella (Fig. 6b). Moreover, mpk6 plants showed enhanced Salmonella proliferation levels when the bacteria were vacuum-infiltrated into leaves (Fig. 6c). These data indicate that the MKK3/MPK6 module of the JA signalling pathway plays an important role in restricting Salmonella infection in Arabidopsis.

Figure 6. Salmonella induction of MPK6 depends on MKK3.

a, Kinase activity assays on native MPK6 immunoprecipitated from mkk3 and mpk6 mutants treated with 10 mM MgCl2 or S. typhimurium 14028s. Myelin basic protein (MBP) was used as substrate. A, activity; P, protein amounts were detected by Western blotting with antibodies specific for MPK6. b, Infection assays. 14 days old seedlings incubated in MS/2 medium inoculated with S. typhimurium 14028s during 2 days, cfu of internal bacteria were determined from plant homogenates. c, Proliferation of bacteria in infiltrated plants is faster in mpk6 mutants. A S. typhimurium 14028s solution was used to vacuum-infiltrate 3 weeks old leaves of mpk6 or Col-0, cfu of internal bacteria were determined in homogenates of leaf discs.


The results presented in this report demonstrate that the human pathogen Salmonella enterica serovar typhimurium (S. typhimurium) may also be a true plant endopathogen. This bacterium can actively invade various tissues, enter into and proliferate in cells of Arabidopsis. Although infection of Arabidopsis triggers immune responses similar to those known from other plant pathogens, S. typhimurium is able to overcome the host defence mechanisms and multiply in Arabidopsis plants. Using a panel of plant defence mutants, the JA signalling pathway was identified to be of major importance for resistance of Arabidopsis plants to Salmonella infection. Overall, these results demonstrate that Arabidopsis can be used as a valuable genetic host model system to study Salmonella pathogenesis in plants.

Salmonella is a plant endopathogen

Infection of animals or humans with S. typhimurium is dependent on various factors. The ability of S. typhimurium to enter mammalian host epithelial cells is tightly controlled by SPI-1 encoded factors and results in the formation of Salmonella containing vacuoles (SCVs), which are necessary for survival and subsequent systemic bacterial infection [4] [1]. The survival of S. typhimurium outside of the host organism is less well understood, but Salmonella can clearly adapt to different external conditions including low pH or high temperature [40]. Salmonella was not only shown to persist in soil for as long as 900 days after inoculation [41], but these bacteria can also survive in a variety of fruits, explaining why mangos and tomatoes are often causally linked to human salmonellosis. Although in these cases, food contamination is rather thought to occur during post-harvest processing, a number of recent reports indicated soil-grown vegetables as the source of human Salmonella infection [5]. The presence of Salmonella in soil might mostly originate from animal-derived contamination such as organic fertilizers or slaughter waste. The question arising now is whether infection of plants by Salmonella in the soil occasionally occurs via passive and nonselective mechanisms or whether infection underlies active bacterial infection processes employed by Salmonella. Our results, as well as other recently published data [42], demonstrate that Arabidopsis and a number of other plants have to be considered as true hosts for Salmonella. Inoculation experiments suggest that in the cases of Medicago truncatula and Arabidopsis, Salmonella is able to recognize plants as a suitable host and actively enters root tissues (Fig. 1b and [42]). Furthermore, when infiltrated into leaves, Salmonella are also able to proliferate in this environment (Fig. 1a and [28]). Most reports so far suggested the presence of Salmonella in the apoplast (the interspace between cells) [5], or in the form of biofilms, as shown for parsley [43]. Using GFP-marked bacteria, we show here that, Salmonella also inhabits the cytoplasmic compartment (Fig. 1e–h). Only after three hours post inoculation Salmonella were present inside root hairs, and 17 hours later also non-root hair rhizodermal cells were infected. By confocal microscopy, bacteria were observed to be moving in the cytoplasm. It is presently unclear whether Salmonella are present in endocytotic vesicles comparable to the SCVs in animal host cells. Further studies are also necessary to clarify the exact mechanism how Salmonella can penetrate through cell walls and enter into plant cells. Our present findings may however provide an explanation for the common S. typhimurium infection after consuming raw, contaminated fruits or vegetables, even after washing or surface sterilization. As shown by our investigation, bacteria present inside plant tissues and/or inside plant cells are resistant to these kinds of treatments and other precautions are required to fulfil current food safety standards.

Importance of the JA-dependent signaling pathway for defence gene induction

The first event in plant-pathogen interaction is recognition of the pathogen by the plant. Although a number of PAMPs were already identified, only a few receptors have so far been identified. FLS2 [20] and EFR [44] (receptors for flg22 and EF-Tu, respectively) are closely related LRR receptor kinases from the LRR-XII subfamily of receptor-like kinases. In the presence of the respective elicitors, both receptors trigger the activation of downstream kinases and defence responses. A nonproteinaceous binding site for harpinPsph PAMP was identified in tobacco plasma membranes that is required for triggering MAP kinase activation [9]. Activation of MAPK cascades is an essential step to induce defence reactions in response to pathogen attack. Several MAPKs are activated by different plant pathogenic bacteria as well as by treatment with several PAMPs [7], [18], [19], [44]. Salmonella attack of Arabidopsis plants also results in the activation of MPK3 and MPK6 with a similar kinetics (Fig. 1i), showing a transient activation maximum at 15 min. Since MPK3 and MPK6 are implicated in various pathways, the respective signalling complexes rather than the MAP kinases themselves, are thought to provide the necessary signalling specificity. Since a recent report reveals the MAPK kinase MKK3 to be involved in activating MPK6 in response to JA [25], we investigated a possible involvement of this MAPKK in triggering MPK6 activation in response to Salmonella infection. MPK6 activation was totally compromised when mkk3 mutants were treated with Salmonella, revealing that Salmonella-induced activation is mediated by MKK3 (Fig. 6a). In agreement with a role of MKK3 and MPK6 in the defence response against Salmonella, we recently found that MKK3 is also involved in defence against bacterial and fungal pathogens and becomes activated by reactive oxygen species [26]. Although MPK6 can also be activated by other MAPKKs besides MKK3 [7], [45], our observation that MPK6 could not be activated to any extent in mkk3 mutant plants speaks against the involvement of other MAPKKs in Salmonella-induced activation of the MPK6 pathway. The role of MPK6 is also underscored by the fact that mpk6 mutant plants were significantly less resistant to Salmonella attack, allowing infection to occur significantly faster and the development of higher Salmonella levels inside Arabidopsis plant tissues (Fig. 6b–c).

Arabidopsis infected by Salmonella initiates transcription of a number of defence genes, including the antifungal defensin gene PDF1.2 [32] and the PR2 and PR4 pathogenesis related genes (Fig. 1j, Fig. 4). The transcription of these genes is generally upregulated in response to pathogens as well as by JA and ET [33]. The well-studied SA-regulated PR1 gene was also upregulated when plants came into contact with Salmonella (Fig. 1j). Together, these data indicate that Salmonella attack induces in Arabidopsis a complex defence response similar to that observed upon attack by other typical plant pathogens [12].

In a simplified view, defence reactions of plants can be divided into SA- and JA/ET-dependent immune responses. SA-dependent responses (further subdivided into NPR1-dependent and NPR1-independent reactions) are important in defence against biotrophic pathogens [13] whereas JA and ET are mainly involved in responses against herbivores and necrotrophic pathogens [16]. We used Arabidopsis plants blocked in these three signalling pathways (NahG, coi1–16 and ein2-1) in order to determine which if any of these pathways is relevant for resistance against infection by Salmonella. The coi1–16 mutant is defective in an F-box protein required for degradation of repressors of JA-responsive genes [38], [39], and is highly susceptible to Salmonella attack (Fig. 5d). The Arabidopsis coi1–16 mutants were not able to induce activation of MPK6 kinase, nor transcription of any of the tested pathogenesis-related marker genes (Fig. 5f–h), indicating that the JA signaling pathway is absolutely required to induce downstream defence reactions against Salmonella. In contrast to Iniquez [28] who reported that NahG plants are more susceptible to Salmonella, we found slightly enhanced proliferation rates (Fig. 5e), but this may be due to different experimental procedures. Finally, ET was also identified to play a role in Arabidopsis defence against Salmonella. Although the ET signalling ein2-1 mutant plants showed delayed expression of defence genes that was correlated with enhanced proliferation rates similar to those observed in coi1–16 mutants, Salmonella infected ein2-1 plants did not produce enhanced disease phenotypes (Fig. 5b). However, this phenomenon does not seem to be a particularity of the interaction between Salmonella with Arabidopsis plants. Upon infection with pathogenic Pseudomonas syringae, ein2-1 mutant plants showed similar bacterial growth rates as wild type Col-0, plants but only minimal chlorophyll loss [46], indicating that marker gene expression and disease phenotype are not necessarily causally related.

Overall our findings give rise to the following human health considerations. The present work provides an explanation why commonly used practices like washing or surface sterilization of Salmonella-infected plants fail to prevent human infection, suggesting that current safety regulations should undergo a serious reconsideration. Alternative strategies to ensure food safety might be achieved by controlling the quality of organic fertilizers and water for irrigation. Moreover, breeding crops and fruits for improved defence signalling could also help to prevent Salmonella infection already at the plant level. If plants can be infected with Salmonella and thus have be considered as intermediate hosts to spread human diseases, then other human endopathogens could possibly use similar strategies. We therefore also suggest to investigate the potential of other human and animal endopathogenic bacteria to use plants as alternative hosts.

Materials and Methods

Plants and bacterial strains

Arabidopsis thaliana plants were grown on half MS medium, pH 5.4, 0.7% w/v agar at 22°C under 16 h light conditions (60 µM m−2s−1 light) or in soil at 22°C under 16 h light conditions. Escherichia coli XL1blue (host for cloning experiments) and S. typhimurium LB5010 (for generation of recombinant strains), 14028s (wild type), were cultured in LB medium aerobically at 37°C with antibiotic selection, when required.

GFP marked bacteria

The gfpmut2 [29] gene encoding green fluorescent protein (forward: ttgaattcgttaactaactaattaatttaagaaggagatatgag, reverse: aaaaaaaagcttccgtctggacatttatttgtatag) was cloned downstream of the constitutive rpsM promoter (forward: tctagagaaaggctacggccgttaattgg, reverse: gttaacgccaggatggctttagaacggg) and upstream of rrnB-based transcriptional terminators in a high-copy number replicating vector pEC75 (composed of a pUC19-based origin of replication and the pUC19-ampicillin resistance gene). The GFP plasmid was introduced in S. typhimurium 14028s. Constitutive expression of GFP in the recombinant strain throughout the growth in vitro was verified using Zeiss LSM 510 META confocal microscope.

Infection of plants with S. typhimurium

Bacteria used in infection experiments were grown to late-logarithmic phase. All infections were performed using a bacterial solution with a density of 3×108 cfu/ml. Plants were either cultivated in half MS medium without sucrose and inoculated with bacteria (inoculation experiments) or vacuum-infiltrated with a bacterial solution in 10 mM MgCl2 (infiltration experiments). At specific times after infection, plants were surface sterilized in 50 mM PBS pH 7.4 supplemented with 1% v/v bleach, 0.1% w/v SDS and 0.2% v/v Tween20 for 2 min and washed 4 times (2 min each) in H2O. Seedlings (inoculation experiments) or 0.7 cm3 excised leaf discs (infiltration experiments) were homogenized in 20% glycerol in 10 mM MgCl2, and appropriate dilutions of homogenates were plated on selective LB agar media in triplicates. The cfu number was calculated per mg fresh weight per seedling or per leaf disc

Kinase assays

Kinase activity assays were performed on extracts from 14 days old seedlings treated with Salmonella for 0, 15, 30 or 60 min. Cleared cell extracts with equal amounts of proteins were subjected for 2 h immunoprecipitation with 2 µl MPK specific antibodies and 25 µl protein A-Sepharose beads. The kinase reactions of immunoprecipitated proteins were performed at 24°C for 30 min in 15 µl kinase buffer containing 10 µg myelin basic protein (MBP), 0.1 mM ATP and 2 µCi γ32P-ATP. Phosphorylation levels of MBP were analyzed with a Phosphorimager.

Quantitative RT-PCR

Transcript levels were analyzed using the Eppendorf system (realplex and RealMaster Mix). 5 µg of total RNA extract were used for reverse transcription. PCRs were run at 95°C, 15 sec: denaturation; 60°C, 15 sec: annealing and 68°C, 15 sec: elongation, for 40 cycles. RNA concentrations were normalized using transcript levels of clathrin, ubiquitin4 and actin2 genes, and values were plotted as a function of non-infected control samples.

Supporting Information

Figure S1.

Salmonella cannot spread between different Arabidopsis organs. Single leaves from soil grown 3 weeks old A. thaliana wild type Col-0 were infiltrated with a bacterial solution. The plants were cultivated for two additional weeks and cfu of internal bacteria in both infiltrated and non-infiltrated leaves were determined. S. typhimurium 14028s was not detected in the non-infiltrated organs. In contrast, bacteria were still present in infiltrated leaves.

(0.05 MB PDF)

Figure S2.

Salmonella are present in newly formed Arabidopsis leaves even one month after infiltration. a–b, A. thaliana wild type Col-0 plants were infiltrated with S. typhimurium 14028s strain and incubated for an additional month in control growing conditions. Infiltrated leaves died within 5 days (arrowheads in b), however, newly formed leaves were present (arrows in b). cfu number was calculated in discs excited from those leaves two days after infiltration or from newly developed leaves one month after infiltration (a). As control non-infiltrated plants were used.

(0.34 MB PDF)

Figure S3.

Z-stacking through an Arabidopsis protoplast infected with GFP-marked Salmonella cells. One µm optical sections of two infected protoplasts (a–b) were taken using a LSM 510 META confocal microscope and reassembled with the Zeiss LSM Software. 488 nm excitation and 505–530 nm emission filters were used.

(0.45 MB PDF)


Authors would like to thank Sergio de la Fuente van Bentem, Bénédicte Sturbois and Thorsten Nünberger for comments on this manuscript.

Author Contributions

Conceived and designed the experiments: HH EC AS. Performed the experiments: AS AC. Analyzed the data: AS. Contributed reagents/materials/analysis tools: EC. Wrote the paper: HH EC AS.


  1. 1. Holden DW (2002) Trafficking of the Salmonella Vacuole in Macrophages doi:10.1034/j.1600-0854.2002.030301.x. Traffic 3: 161–169.
  2. 2. Guignot J, Caron E, Beuzon C, Bucci C, Kagan J, et al. (2004) Microtubule motors control membrane dynamics of Salmonella-containing vacuoles. Journal of Cell Science 117: 1033–1047.
  3. 3. Waterman SR, Holden DW (2003) Functions and effectors of the Salmonella pathogenicity island 2 type III secretion system doi:10.1046/j.1462-5822.2003.00294.x. Cellular Microbiology 5: 501–511.
  4. 4. Fields PI, Swanson RV, Haidaris CG, Heffron F (1986) Mutants of Salmonella typhimurium That Cannot Survive within the Macrophage are Avirulent 10.1073/pnas.83.14.5189. PNAS 83: 5189–5193.
  5. 5. Brandl MT (2006) Fitness of Human Enteric Pathogens on Plants and Implications for Food Safety doi:10.1146/annurev.phyto.44.070505.143359. Annual Review of Phytopathology 44: 367–392.
  6. 6. Jablasone J, Warriner K, Griffiths M (2005) Interactions of Escherichia coli O157:H7, Salmonella typhimurium and Listeria monocytogenes plants cultivated in a gnotobiotic system. International Journal of Food Microbiology 99: 7–18.
  7. 7. Asai T, Tena G, Plotnikova J, Willmann MR, Chiu W-L, et al. (2002) MAP kinase signalling cascade in Arabidopsis innate immunity. 415: 977–983.
  8. 8. Fellbrich G, Romanski A, Varet A, Blume B, Brunner F, et al. (2002) NPP1, a Phytophthora-associated trigger of plant defense in parsley and Arabidopsis doi:10.1046/j.1365-313X.2002.01454.x. The Plant Journal 32: 375–390.
  9. 9. Lee J, Klessig DF, Nurnberger T (2001) A Harpin Binding Site in Tobacco Plasma Membranes Mediates Activation of the Pathogenesis-Related Gene HIN1 Independent of Extracellular Calcium but Dependent on Mitogen-Activated Protein Kinase Activity 10.1105/tpc.13.5.1079. Plant Cell 13: 1079–1093.
  10. 10. Navarro L, Zipfel C, Rowland O, Keller I, Robatzek S, et al. (2004) The Transcriptional Innate Immune Response to flg22. Interplay and Overlap with Avr Gene-Dependent Defense Responses and Bacterial Pathogenesis 10.1104/pp.103.036749. Plant Physiol 135: 1113–1128.
  11. 11. Gomez-Gomez L, Felix G, Boller T (1999) A single locus determines sensitivity to bacterial flagellin in Arabidopsis thaliana doi:10.1046/j.1365-313X.1999.00451.x. The Plant Journal 18: 277–284.
  12. 12. Jones JDG, Dangl JL (2006) The plant immune system. 444: 323–329.
  13. 13. Durrant WE, Dong X (2004) SYSTEMIC ACQUIRED RESISTANCE doi:10.1146/annurev.phyto.42.040803.140421. Annual Review of Phytopathology 42: 185–209.
  14. 14. Beckers GJM, Spoel SH (2006) Fine-Tuning Plant Defence Signalling: Salicylate versus Jasmonate. Plant Biology 1–10.
  15. 15. Halim VA, Vess A, Scheel D, Rosahl S (2006) The Role of Salicylic Acid and Jasmonic Acid in Pathogen Defence. Plant Biology 307–313.
  16. 16. Zimmerli L, Stein M, Lipka V, Schulze-Lefert P, Somerville S (2004) Host and non-host pathogens elicit different jasmonate/ethylene responses in Arabidopsis doi:10.1111/j.1365-313X.2004.02236.x. The Plant Journal 40: 633–646.
  17. 17. Nakagami H, Pitzschke A, Hirt H (2005) Emerging MAP kinase pathways in plant stress signalling. Trends in Plant Science 10: 339–346.
  18. 18. Desikan R, Hancock JT, Ichimura K, Shinozaki K, Neill SJ (2001) Harpin Induces Activation of the Arabidopsis Mitogen-Activated Protein Kinases AtMPK4 and AtMPK6 10.1104/pp.126.4.1579. Plant Physiol 126: 1579–1587.
  19. 19. Nuhse TS, Peck SC, Hirt H, Boller T (2000) Microbial Elicitors Induce Activation and Dual Phosphorylation of the Arabidopsis thaliana MAPK 6 10.1074/jbc.275.11.7521. J Biol Chem 275: 7521–7526.
  20. 20. Gomez-Gomez L, Boller T (2000) FLS2: An LRR Receptor-like Kinase Involved in the Perception of the Bacterial Elicitor Flagellin in Arabidopsis. Molecular Cell 5: 1003–1011.
  21. 21. Menke FLH, van Pelt JA, Pieterse CMJ, Klessig DF (2004) Silencing of the Mitogen-Activated Protein Kinase MPK6 Compromises Disease Resistance in Arabidopsis 10.1105/tpc.015552. Plant Cell 16: 897–907.
  22. 22. Ichimura K, Casais C, Peck SC, Shinozaki K, Shirasu K (2006) MEKK1 Is Required for MPK4 Activation and Regulates Tissue-specific and Temperature-dependent Cell Death in Arabidopsis 10.1074/jbc.M605319200. J Biol Chem 281: 36969–36976.
  23. 23. Nakagami H, Soukupova H, Schikora A, Zarsky V, Hirt H (2006) A Mitogen-activated Protein Kinase Kinase Kinase Mediates Reactive Oxygen Species Homeostasis in Arabidopsis 10.1074/jbc.M605293200. J Biol Chem 281: 38697–38704.
  24. 24. Suarez-Rodriguez MC, Adams-Phillips L, Liu Y, Wang H, Su S-H, et al. (2007) MEKK1 Is Required for flg22-Induced MPK4 Activation in Arabidopsis Plants 10.1104/pp.106.091389. Plant Physiol 143: 661–669.
  25. 25. Takahashi F, Yoshida R, Ichimura K, Mizoguchi T, Seo S, et al. (2007) The Mitogen-Activated Protein Kinase Cascade MKK3-MPK6 Is an Important Part of the Jasmonate Signal Transduction Pathway in Arabidopsis 10.1105/tpc.106.046581. Plant Cell 19: 805–818.
  26. 26. Doczi R, Brader G, Pettko-Szandtner A, Rajh I, Djamei A, et al. (2007) The Arabidopsis Mitogen-Activated Protein Kinase Kinase MKK3 Is Upstream of Group C Mitogen-Activated Protein Kinases and Participates in Pathogen Signaling 10.1105/tpc.106.050039. Plant Cell: tpc.106.050039.
  27. 27. Prithiviraj B, Bais HP, Jha AK, Vivanco JM (2005) Staphylococcus aureus pathogenicity on Arabidopsis thaliana is mediated either by a direct effect of salicylic acid on the pathogen or by SA-dependent, NPR1-independent host responses doi:10.1111/j.1365-313X.2005.02385.x. The Plant Journal 42: 417–432.
  28. 28. Iniquez LA, Dong Y, Carter HD, Ahmer BMM, Stone JM, et al. (2005) Regulation of Enteric Endophytic Bacterial Colonization by Plant Defenses. Molecular Plant Microbe Interaction 18: 169–178.
  29. 29. Cormack BP, Valdivia RH, Falkow S (1996) FACS-optimized mutants of the green fluorescent protein (GFP). Gene Flourescent Proteins and Applications 173: 33–38.
  30. 30. Chisholm ST, Coaker G, Day B, Staskawicz BJ (2006) Host-Microbe Interactions: Shaping the Evolution of the Plant Immune Response. Cell 124: 803–814.
  31. 31. Ichimura K, Shinozaki K, Tena G, Sheen J, et al. MAPK Group (2002) Mitogen-activated protein kinase cascades in plants: a new nomenclature. Trends in Plant Science 7: 301–308.
  32. 32. Penninckx I, Eggermont K, Terras F, Thomma B, Samblanx G, et al. (1996) Pathogen-Induced Systemic Activation of a Plant Defensin Gene in Arabidopsis Follows a Salicylic Acid-Independent Pathway 10.1105/tpc.8.12.2309. Plant Cell 8: 2309–2323.
  33. 33. Jung C, Lyou S, Yeu S, Kim M, Rhee S, et al. (2007) Microarray-based screening of jasmonate-responsive genes in Arabidopsis thaliana. Plant Cell Rep 26: 1053–1063.
  34. 34. Heck S, Grau T, Buchala A, Metraux J-P, Nawrath C (2003) Genetic evidence that expression of NahG modifies defence pathways independent of salicylic acid biosynthesis in the Arabidopsis-Pseudomonas syringae pv. tomato interaction doi:10.1046/j.1365-313X.2003.01881.x. The Plant Journal 36: 342–352.
  35. 35. McGrath R, Ecker JR (1998) Ethylene signalling in Arabidopsis: events from the membrane to the nucleus. Plant Physiol Biochem 14: 119–126.
  36. 36. Xie D-X, Feys BF, James S, Nieto-Rostro M, Turner JG (1998) COI1: An Arabidopsis Gene Required for Jasmonate-Regulated Defense and Fertility 10.1126/science.280.5366.1091. Science 280: 1091–1094.
  37. 37. Xu L, Liu F, Lechner E, Genschik P, Crosby WL, et al. (2002) The SCFCOI1 Ubiquitin-Ligase Complexes Are Required for Jasmonate Response in Arabidopsis 10.1105/tpc.003368. Plant Cell 14: 1919–1935.
  38. 38. Chini A, Fonseca S, Fernandez G, Adie B, Chico JM, et al. (2007) The JAZ family of repressors is the missing link in jasmonate signalling. 448: 666–671.
  39. 39. Thines B, Katsir L, Melotto M, Niu Y, Mandaokar A, et al. (2007) JAZ repressor proteins are targets of the SCFCOI1 complex during jasmonate signalling. 448: 661–665.
  40. 40. Samelis J, Ikeda JS, Sofos JN (2003) Evaluation of the pH-dependent, stationary-phase acid tolerance in Listeria monocytogenes and Salmonella Typhimurium DT104 induced by culturing in media with 1% glucose: a comparative study with Escherichia coli O157:H7 doi:10.1046/j.1365-2672.2003.02013.x. Journal of Applied Microbiology 95: 563–575.
  41. 41. Nicholson F, Groves S, Chambers B (2005) Pathogen survival during livestock manure storage and following land application Bioresource Technology 96: 135–143.
  42. 42. Dong Y, Iniguez AL, Ahmer BMM, Triplett EW (2003) Kinetics and Strain Specificity of Rhizosphere and Endophytic Colonization by Enteric Bacteria on Seedlings of Medicago sativa and Medicago truncatula 10.1128/AEM.69.3.1783-1790.2003. Appl Environ Microbiol 69: 1783–1790.
  43. 43. Lapidot A, Romling U, Yaron S (2006) Biofilm formation and the survival of Salmonella Typhimurium on parsley. International Journal of Food Microbiology 109: 229–233.
  44. 44. Zipfel C, Kunze G, Chinchilla D, Canierd A, Jones JDG, et al. (2006) Perception of the Bacterial PAMP EF-Tu by the Receptor EFR Restricts Agrobacterium-Mediated Transformation. Cell 125: 746–760.
  45. 45. Meszaros T, Helfer A, Hatzimasoura E, Magyar Z, Serazetdinova L, et al. (2006) The Arabidopsis MAP kinase kinase MKK1 participates in defence responses to the bacterial elicitor flagellin doi:10.1111/j.1365-313X.2006.02888.x. The Plant Journal 48: 485–498.
  46. 46. Bent AF, Innes RW, Ecker JR, Staskawicz BJ (1992) Disese Development in Ethylene-Insensitive Arabidopsis thaliana Infected with Virulent and Avirulent Pseudomonas and Xanthomonas Pathogens. Mol Plant Microbe Interact 5: 372–378.