Arabidopsis thaliana plants expressing human beta-defensin-2 are more resistant to fungal attack: functional homology between plant and human defensins
- First Online:
- Cite this article as:
- Aerts, A.M., Thevissen, K., Bresseleers, S.M. et al. Plant Cell Rep (2007) 26: 1391. doi:10.1007/s00299-007-0329-4
- 198 Views
Human beta-defensin-2 (hBD-2) is a small antimicrobial peptide with potent activity against different Gram-negative bacteria and fungal/yeast species. Since human beta-defensins and plant defensins share structural homology, we set out to analyse whether there also exists a functional homology between these defensins of different eukaryotic kingdoms. To this end, we constructed a plant transformation vector harbouring the hBD-2 coding sequence, which we transformed to Arabidopsis thaliana plants, giving rise to A. thaliana plants indeed expressing hBD-2. Furthermore, we could demonstrate that this heterologously produced hBD-2 possesses antifungal activity in vitro. Finally, we could show that hBD-2 expressing A. thaliana plants are more resistant against the broad-spectrum fungal pathogen Botrytis cinerea as compared to untransformed A. thaliana plants, and that this resistance is correlated with the level of active hBD-2 produced in these transgenic plants. Hence, we demonstrated a functional homology, next to the already known structural homology, between defensins originating from different eukaryotic kingdoms. To our knowledge, this is the first time that this is specifically demonstrated for plant and mammalian defensins.
KeywordsArabidopsis thalianiaBotrytis cinereaHuman defensinPlant defensin
Dahlia merckii antimicrobial protein
Concentration causing 50% inhibition of fungal growth
Matrix attachment region
Post-transcriptional gene silencing
Raphanus sativus antifungal protein
All living organisms, ranging from microorganisms to plants and mammals, have evolved mechanisms to actively defend themselves against pathogen attack. The most sophisticated of those mechanisms, the adaptive immune system, is only elaborated in higher vertebrates (reviewed by Matsunaga and Rahman 1998), and deploys antibodies and killer cells to recognize and eliminate specific invaders, respectively. Innate immunity, on the other hand, is a much more widespread, ancient defence strategy involving, among other responses, the production of antimicrobial peptides (Boman 1995). In the innate immune response, only one peptide class has been shown so far to be conserved between plants, invertebrates and vertebrates, being defensins (Thomma et al. 2002). It was hypothesized that defensins from these different eukaryotic kingdoms arose from a common ancestral gene (Thomma et al. 2002). More recently, Mygind and co-workers (2005) isolated the first defensin from a fungus, being plectasin from the saprophyte Pseudoplectania nigrella. Plectasin has primary, secondary and tertiary structures that closely resemble those of defensins found in spiders, scorpions, dragonflies and mussels (Mygind et al.2005). The identification of a defensin in a lower eukaryote further evidences the widespread distribution of the defensin class of peptides over different kingdoms.
Defensins were first identified as a family of peptides in rabbits (Selsted et al. 1984) and subsequently in other higher vertebrate species, including humans. Up to now, three types of defensins have been characterized in mammals, being α-, β- and θ-defensins. While the former two types appear to generally occur in mammals, θ-defensins were thus far only found in rhesus monkey leukocytes (Tang et al. 1999; Trabi et al. 2001). The α- and β-type defensins differ in size and arrangement of cysteines within their sequences (Lehrer et al. 1993; Selsted et al. 1993). Mammalian α-defensins do not comprise an α-helix, whereas mammalian β-defensins combine an α-helix with a triple-stranded antiparallel β-sheet (Hoover et al. 2000, 2001; Sawai et al. 2001).
One of these β-defensins, human beta-defensin-2 (hBD-2) was first isolated as a small basic (41 amino acid residues) peptide from the skin of patients with psoriasis (Harder et al. 1997). Subsequently, hBD-2 was demonstrated to be expressed by epithelial cells lining various tissues facing the external environment (reviewed in Yang et al. 2004). hBD-2 has bactericidal activity against many Gram-negative bacteria, including Pseudomonas aeruginosa, but was found to only have bacteriostatic activity against the Gram-positive bacterium Staphylococcus aureus. Additionally, hBD-2 is active against the yeast Candida albicans (Harder et al.1997). The antimicrobial mode of action of hBD-2 is still not clear. In a model for the antibacterial mechanism proposed by Hoover and co-workers (2000), hBD-2 molecules form positively charged octamers that neutralize the anionic lipid headgroups of the bacterial membrane. This neutralization disrupts the integrity of the lipid bilayer, causing membrane permeabilization. In contrast, little experimental data on the antifungal mechanism of action of hBD-2 is available.
The first plant defensins, originally termed γ-thionins, were isolated from wheat and barley grains at the beginning of the 1990s (Colilla et al. 1990; Mendez et al. 1990), but later it was demonstrated that they are widespread over the plant kingdom (reviewed in Thomma et al. 2002). Plant defensins are small (45–54 amino acid residues), basic, cysteine-rich peptides that possess antimicrobial activities at micromolar concentrations (reviewed in Thomma et al. 2002; Lay and Anderson 2005). They are active against a broad-range of phytopathogenic fungi (such as Fusarium culmorum and Botrytis cinerea) and human pathogens (such as C. albicans), but are non-toxic to either mammalian or plant cells (Terras et al. 1992; B. P. A. Cammue, unpublished results). The global fold of plant defensins comprises a cysteine-stabilized αβ motif (CSαβ motif) consisting of an α-helix and a triple-stranded β-sheet, organized in a βαββ architecture and stabilized by four disulfide bridges (Fant et al.1998). As such, plant defensins are not only structurally homologous to the β-class of human defensins (Hoover et al. 2000, 2001; Sawai et al. 2001), but also to the insect defensin-like peptide heliomicin (Lamberty et al. 2001). Regarding amino acid composition, the plant defensin family is quite diverse with sequence conservation restricted to eight structurally important cysteines (reviewed in Thomma et al. 2002; Lay and Anderson 2005). Plant defensins induce membrane permeabilization in susceptible fungi through a specific interaction with high-affinity binding sites on fungal cells, identified as complex sphingolipids (Thevissen et al. 1999, 2000a, b, 2004). Experimental evidence points towards a role of the plant defensins in defending the host plant from fungal attack (reviewed in Thomma et al. 2002, 2003).
Since hBD-2 is structurally homologous to plant defensins (Thomma et al. 2002), the question arises whether hBD-2 could also be functionally homologous to plant defensins. To this end, we investigated whether the hBD-2 gene could be expressed in Arabidopsis thaliana and whether such overexpression could increase the plant’s resistance to fungal attack, as is reported for plant defensins (Terras et al. 1995; Wang et al. 1999; Gao et al. 2000; Kanzaki et al. 2002; Park et al. 2002; Li et al.2003).
Materials and methods
Biological material and growth conditions
Seeds of the A. thaliana post-transcriptional gene silencing (PTGS) mutant sgs2 (Mourrain et al. 2000) were kindly provided by Dr. Vaucheret (INRA, Versailles, France). Plants were grown on soil in a growth chamber (22°C day-time temperature and 18°C night-time temperature with a 12 h photoperiod at a photon flux density of 100 μ/Em2/s). Fungi were maintained and spores harvested as previously described (Broekaert et al. 1990). Fungal strain used: B. cinerea Korea (Brouwer et al. 2003).
Construction of plant transformation vector pCMPG7057
A pUC18 vector containing the coding sequence of wild-type hBD-2 (kindly provided by Professor Cassiman, Department of Human Genetics, K.U.Leuven, Heverlee, Belgium) was used to amplify hBD-2 using hBD-2 specific primers AFP464 (5′-TTACACCATGGTGAATCGGTCGGTTGCGTTCTCCGCGTTCGTTCTGATCCTTTTCGTGCTCGCCATCTCAGATATCGCATCCGTTAGTGGAGGTATAGGCGATCCTGTTAC) and AFP465 (5′-ATTGAGCTCTTATCATGGCTTTTTGCAGCATTTTG). The sequence of the DmAMP1 leader peptide (Osborn et al. 1995) (underlined in primer sequence) was incorporated in primer AFP464. The resulting PCR product was cloned in the plant transformation vector pFAJ3184 (De Bolle et al. 2003) using a modular vector system (Goderis et al. 2002), giving rise to vector pCMPG7057. The vector pCMPG7057 was transferred into Agrobacterium tumefaciens GV3101 (pMP90) (Zambryski et al. 1983) by electroporation.
Arabidopsis thaliana (Columbia 0) sgs2 plants were grown as described (De Bolle et al. 2003). Plants were transformed using the A. tumefaciens—mediated floral dip transformation method (T0 generation) (Clough and Bent 1998). Primary transformants (T1 generation) were selected on a sand/perlite mixture subirrigated with water containing phosphinotricin (5 mg/l, Basta®, Bayer, Leverkusen, Germany). T2 generation plants were obtained from seeds from T1 generation plants.
PCR-analysis on genetic transformants
PCR-analysis was done on the primary transformants to confirm the insertion of the transgene hBD-2. Therefore, primers AFP464 and AFP465 amplifying the coding region of hBD-2 were used.
Protein extraction from A. thaliana leaves for ELISA assays
Leaf material from 4 to 5-week-old A. thaliana plants was collected in a screw-capped centrifuge tube and frozen in liquid nitrogen. Proteins were extracted as described in François et al. (2004). Total soluble protein content was determined according to Bradford (1976) using bovine serum albumin as a standard.
ELISA assays (enzyme linked immunosorbent assays) were set up as competitive type assays essentially as described by Penninckx et al. (1996). Coating of the ELISA microtiterplates (Greiner, type 655092) was done with 50 ng/ml hBD-2 (Peptides International, Louisville, KY, USA) in coating buffer (15 mM Na2CO3, 35 mM NaHCO3, pH 9.6). Primary antiserum (Peptides International) was used as a 1,000-fold diluted solution in 0.3% (w/v) bovine serum albumine (Sigma, St. Louis, MO, USA) in phosphate buffered saline (PBS) (140 mM NaCl, 3 mM KCl, 2 mM KH2PO4, 8 mM Na2HPO4, pH 7.4) containing 0.05% (v/v) tween 20. The heterologous hBD-2 concentration was measured relative to a twofold dilution series of native hBD-2 (Peptides International), which was used as a standard.
Partial purification of heterologously produced hBD-2 from pCMPG7057 transformed A. thaliana plants
To prepare a crude leaf extract, fully expanded rosette leaves from pCMPG7057 transgenic A. thaliana plants (75 g leaf material) were homogenized under liquid nitrogen and extracted overnight at 4°C with icecold 50 mM TRIS (pH 7), containing a mixture of protease inhibitors [(1 mM thioureum, 1 mM phenylmethylsulfonyl fluoride, 0.02 mM pepstatin A and 5 mM ethylenediaminetetraacetic acid (EDTA) (pH 8)]. The homogenate was cleared by centrifugation (30 min at 10,000×g) and the supernatant is referred to as the crude leaf extract. This crude leaf extract was filtrated using a Whatman paper filter. Subsequently, proteins of the crude leaf extract were precipitated by addition of solid ammonium sulphate (80% saturation). The precipitate was dissolved in demineralized water and dialysed against demineralized water. Subsequently, the leaf extract was fractionated by ion exchange chromatography (IEC). IEC was performed by passing the leaf extract over an anion exchange matrix (Q-Sepharose Fast Flow, Amersham Bioscience Benelux, Roosendaal, The Netherland), equilibrated with 50 mM TRIS (pH 7.5). The anion exchange matrix was eluted with 50 mM TRIS (pH 7.5). The resulting fraction was collected, dialysed against demineralized water, lyophilysed, dissolved in Milli-Q water and assessed for the presence of proteins cross-reacting with hBD-2 antibodies (hBD-2 CRPs) with ELISA.
In vitro antifungal activity assay
Antifungal activity was measured by microspectrophotometry as described in Broekaert et al. (1990). Spores of the test fungus B. cinerea Korea were suspended in 12 g/l potato dextrose broth (Difco, Detroit, MI, USA), supplemented with tetracycline (10 μg/ml). This suspension was dispended by aliquots of 80 μl into wells of a 96-well microtiter plate (Greiner, type 655101) containing 20 μl of the sample to be analysed. After 48–72 h of incubation at 25°C, growth of the fungus was evaluated microscopically, using an inverted microscope. All samples were corrected to contain equal amounts of total protein using a BCA protein assay as described by Smith et al. (1985).
Plant inoculations and disease quantification
A total of 1,000 rosette leaves from 160 5-week-old soil-grown A. thaliana high- and low hBD-2 expressor plants (with an average high- and low-production level of active hBD-2 of total soluble protein, respectively) and a total of 500 rosette leaves from 80 5-week soil-grown A. thaliana control plants (expressing the GUS construct pMAR-p35S-uidA) were inoculated with a 5 μl droplet of a suspension of 5 × 105B. cinerea Korea spores/ml in 12 g/l potato dextrose broth (Difco). Plants were incubated at 24°C/18°C with a 12 h photoperiod at high humidity. Disease symptoms were scored by determining the ratio of decayed leaves/total number of inoculated leaves. Statistical analysis of the results was performed using SAS statistical software, Mixed models Procedure, Tukey–Kramer adjusted.
Construction of plant transformation vector pCMPG7057 and transfer to A. thaliana
Characterization of transgenic A. thaliana plants expressing the hBD-2 construct
In a first instance, selected transgenic plants were confirmed for the presence of the hBD-2 gene by PCR-analysis using an hBD-2 specific primer combination (data not shown). Subsequently, in order to demonstrate the presence of heterologously produced hBD-2 in the A. thaliana transgenic lines and to identify hBD-2 containing fractions in subsequent hBD-2 purification steps (see further), we optimised an hBD-2 specific competitive enzyme-linked immunosorbent assay (ELISA). Using this ELISA, proteins cross-reacting with hBD-2 antibodies (hBD-2 CRPs) could clearly be demonstrated in extracts from randomly selected T1 generation pCMPG7057 transformed A. thaliana plants, previously demonstrated to be PCR-positive for the hBD-2 gene, in contrast to extracts from PCR-negative or control transgenic plants (data not shown). The latter consisted of sgs2A. thaliana plants transformed with a pCMPG7057 analogue (pMAR-p35S-uidA; Butaye et al. 2004) in which the hBD-2 coding region is replaced by that of the β-glucuronidase (GUS) gene of E. coli.
Partial purification and in vitro antifungal activity of hBD-2 heterologously produced in pCMPG7057 transgenic A. thaliana plants
In order to purify the heterologously produced hBD-2, 75 g leaf material from pCMPG7057 transgenic T2 plants, demonstrated by ELISA to contain hBD-2 CRPs, were pooled and used to prepare a crude leaf extract. The crude leaf extract was partially fractionated by IEC as described in materials and methods, using the developed hBD-2 ELISA for selection of hBD-2 CRPs containing fractions. As a reference a similar purification protocol was applied on extracts from control, pMAR-p35S-uidA transgenic, plants. Partially purified hBD-2 from hBD-2 expressing plants was further assessed for its in vitro antifungal activity against B. cinerea, and compared to that of the corresponding fraction similarly prepared from control plants. The latter displayed a basic antifungal activity against B. cinerea, typically observed in A. thaliana extracts due to the presence of preformed antimicrobial components (Tierens et al. 2001). As compared to this basic endogenous activity partially purified hBD-2 fractions from hBD-2 expressing plants could be clearly distinguished by an activity against B. cinerea that was at least fourfold more potent (data not shown). Comparison of this additional hBD-2 expression linked activity with the in vitro antifungal activity of native hBD-2 on B.cinerea (the concentration causing 50% inhibition of fungal growth (IC50) being 4.7 μg/ml) allowed us to deduce the average level of active hBD-2 produced in the hBD-2 expressing plants as 0.21% of total soluble protein, which correlated with the ELISA results. Whether hBD-2 act on B. cinerea in a fungistatic or fungicidal way remains to be clarified.
pCMPG7057 transgenic A. thaliana plants exhibit reduced susceptibility towards the necrotrophic fungus Botrytis cinerea
Plant defensins are generally believed to be part of the plant’s innate immunity system. Moreover, several research groups have reported an increased resistance to fungal diseases of plants overexpressing different types of plant defensins (Terras et al. 1995; Wang et al. 1999; Gao et al. 2000; Kanzaki et al. 2002; Park et al. 2002; Li et al.2003). A similar protective effect has been reported for plants expressing plant defensin-like proteins from insects (Banzet et al. 2002; Langen et al. 2006). In the present study we demonstrated for the first time that plants expressing a human defensin (hBD-2), which is structurally homologous to plant defensins (Thomma et al. 2002), can also become more resistant to fungal attack, thereby extending the earlier observed structural similarity to a functional homology between these defensins of two different eukaryotic kingdoms. To this end, we transformed A. thaliana plants with a vector harbouring the hBD-2 coding sequence resulting in transgenic plants heterologously expressing hBD-2. We could demonstrate that partially purified hBD-2 from these plants possesses in vitro antifungal activity to the broad-spectrum pathogen B. cinerea. Finally, we demonstrated that in planta overexpression of the human defensin hBD-2 can confer protection against the same pathogen and that the degree of protection can be linked with the relative production level of the hBD-2 gene in planta.
In order to generate hBD-2 transgenic plants, we expressed the hBD-2 gene in A. thaliana using an in-house developed PTGS-MAR expression system for high heterologous in planta expression (Butaye et al. 2004). This approach resulted in average production levels of 0.21% of total soluble protein, which are, however, still inferior to those obtained by Butaye et al. (2004) for the marker protein GUS (on average 2.5%). On one hand, this could be due to the small size of hBD-2 (4.3 kDa) as compared to the tenfold higher molecular weight of GUS (Butaye et al. 2004). It is generally accepted that expression of smaller proteins is less efficient and that, for example, fusion of peptides to larger protein modules can give rise to higher production levels (Okamoto et al. 1998; Hondred et al.1999; François et al.2002b). On the other hand, we have very recently demonstrated that the use of this PTGS-MAR system could also result in production levels of up to 3% (of total soluble protein) for A. thaliana plants overexpressing their own plant defensins AtPDF1.1 and AtPDF1.3 (J. Sels et al. 2006, submitted data) which have a similar molecular weight as hBD-2. Therefore, another reason for the observed relatively low production levels of hBD-2 in A. thaliana might rely on the fact that the hBD-2 expression construct in this study was not specifically adapted for optimal codon usage in planta. Such an optimization may indeed result in enhanced production levels of the heterologous protein as was demonstrated for the production of human insulin-like growth factor-1 (hiGF-1) in transgenic rice and tobacco plants (Panahi et al.2004).
In the present study, we have demonstrated that the expression of the human defensin hBD-2 in A. thaliana plants could result in increased resistance to the pathogenic fungus B. cinerea. Moreover, this protection seems to be correlated with the levels of transgenically produced hBD-2. Indeed, while average production levels of 0.65% of total soluble protein were found to be sufficient for statistically significant increased resistance against B. cinerea at all time points, lower average hBD-2 levels down to 0.05% were only effective in the early stages of infection when disease pressure is low. In accordance with this range of production levels, we earlier showed in transgenic tobacco that production levels of 0.1–0.2% of the radish plant defensin RsAFP2 were necessary for significant increased resistance to infection by the foliar pathogen Alternaria longipes (Terras et al. 1995). Similar values of production levels are reported for protection of transgenic tobacco against fungal attack by heterologous expressed Chinese cabbage defensin (Park et al. 2002) and heliomicin (Banzet et al. 2002).
Expression of human genes in plants has been reported previously (Kwon et al. 2003; Bode et al. 2004; Geyer et al. 2005). However, to our knowledge, there is only one report describing in planta expression of a human gene for enhanced protection against fungal attack. Takaichi and Oeda (2000) demonstrated that expression of a human lysozyme gene in carrots confers protection against the powdery mildew fungus Erysipheheraclei. In the present study we demonstrated that overexpressing in planta of a human defensin (hBD-2) with structural homology to the family of plant defensins can also confer increased protection against fungal infection as is reported in literature for plant defensins (Terras et al. 1995; Wang et al. 1999; Gao et al. 2000; Kanzaki et al. 2002; Park et al. 2002; Li et al.2003). Moreover, it was recently demonstrated that another class of structurally related proteins originating from insects can reduce fungal disease when expressed in plants (Banzet et al. 2002; Langen et al. 2006). This indicates that structurally homologous peptides from different kingdoms can also share a functional homology.
Whether these homologies of hBD-2 with other defensins are also reflected in their antifungal working mechanism remains unclear at the moment. During the past decade we have focussed on the mode of antifungal activity of different plant defensins and of a defensin-like peptide from insects. In summary, these defensins appear to interact with specific complex sphingolipids in the fungal membrane (Thevissen et al. 2000b, 2004), resulting in membrane permeabilization and cell growth arrest (Thevissen et al. 1996, 1999). Moreover, very recently, we could demonstrate a causal link between the antifungal action of RsAFP2 and RsAFP2-induced generation of reactive oxygen species in susceptible fungi, pointing to a putative induction of apoptosis by RsAFP2 (Aerts et al. 2006). Whether comparable intracellular signalling pathways are involved in the mode of antifungal activity of hBD-2 is currently investigated. Preliminary experiments suggest that hBD-2 is not interacting with one of the complex sphingolipids earlier demonstrated to be crucial for binding of the plant or insect defensins. It is expected that the heterologous in planta production system described in the present manuscript will allow us to purify hBD-2 in larger quantities for further research on its antifungal mode of action.
The authors wish to thank Dr. Hervé Vaucheret (INRA, Versailles, France) for providing seeds of the sgs2-mutant and Prof. Cassiman (K. U. Leuven, Belgium) for the kind gift of the pUC18 vector containing the hBD-2 coding sequence. This work was supported by grants from the Fonds voor Wetenschappelijk Onderzoek (FWO)-Vlaanderen to B. C (G028804N and G.0405.05). A. A. acknowledges the receipt of a pre-doctoral grant of the Instituut voor de Aanmoediging van Innovatie door Wetenschap en Technologie in Vlaanderen (IWT-Vlaanderen). K. T. acknowledges the receipt of a post-doctoral fellowship of the Industrial Research Fund (IOF) of the Katholieke Universiteit Leuven. I.F. acknowledges the receipt of a post-doctoral grant of the IWT-Vlaanderen. The authors wish to thank Dr. Stijn Delauré for his assistance with the infection assays and statistical analysis.