Delay of flower senescence by bacterial endophytes expressing 1-aminocyclopropane-1-carboxylate deaminase

Authors


Correspondence

Bernard R. Glick, Department of Biology, University of Waterloo, 200 University Avenue West, Waterloo, ON, Canada N2L 3G1. E-mail: glick@sciborg.uwaterloo.ca

Abstract

Aims

The ability of 1-aminocyclopropane-1-carboxylate (ACC) deaminase-containing plant growth-promoting bacterial (PGPB) endophytes Pseudomonas fluorescens YsS6 and Pseudomonas migulae 8R6, their ACC deaminase minus mutants and the rhizospheric plant growth-promoting bacterium Pseudomonas putida UW4 to delay the senescence of mini carnation cut flowers was assessed.

Methods and Results

Fresh cut flowers were incubated with either a bacterial cell suspension, the ethylene precursor ACC, the ethylene inhibitor l-α-(aminoethoxyvinyl)-glycine or 0·85% NaCl at room temperature for 11 days. Levels of flower senescence were recorded every other day. To verify the presence of endophytes inside the plant tissues, scanning electron microscopy was performed. Among all treatments, flowers treated with wild-type ACC deaminase-containing endophytic strains exhibited the most significant delay in flower senescence, while flowers treated with the ACC deaminase minus mutants senesced at a rate similar to the control. Flowers treated with Ps. putida UW4 senesced more rapidly than untreated control flowers.

Conclusion

The only difference between wild-type and mutant bacterial endophytes was ACC deaminase activity so that it may be concluded that this enzyme is directly responsible for the significant delay in flower senescence. Despite containing ACC deaminase activity, Ps. putida UW4 is not taken up by the cut flowers and therefore has no effect on prolonging their shelf life.

Significance and Impact of the Study

The world-wide cut flower industry currently uses expensive and potentially environmentally dangerous chemical inhibitors of ethylene to prolong the shelf life of cut flowers. The use of PGPB endophytes with ACC deaminase activity has the potential to replace the chemicals that are currently used by the cut flower industry.

Introduction

Flower senescence is a sequence of events that ends with the death of the flower. These events are plant family and species specific (Van Doorn 2002) and generally include petal in-rolling, loss of petal colour, petal wilting, shedding of flower parts and gradual fading of the blossom (Tripathi and Tuteja, 2007). Senescence in flowers can be broadly classified as being either ethylene sensitive or ethylene insensitive. Generally, monocotyledonous plants are more ethylene insensitive, while dicotyledonous plants show a greater sensitivity towards ethylene (Woltering and Van Doorn 1988; Van Doorn 2001). Woltering and Van Doorn (1988) further categorized ethylene-sensitive flowers into five subclasses from not sensitive to highly sensitive. Some of the common members of highly ethylene-sensitive flowers include zinnia, carnation, rose and geranium (Woltering and Van Doorn 1988). Ethylene is a key stress regulatory plant hormone. Under normal conditions, plants produce only low levels of ethylene, typically conferring beneficial effects on plant growth and development; however, in response to various stresses, there is often a significant rise in the endogenous ethylene production that has adverse effects on plant growth (Abeles et al. 1992) and is thought to be responsible for senescence in flowers (Woltering and Van Doorn 1988; Nayani et al. 1998).

Application of chemical inhibitors of plant ethylene production, such as silver thiosulphate (STS) (Veen and van de Geijn 1978), cyclic olefin norbornadiene (NBD) (Reid and Wu 1992) and l-α-(aminoethoxyvinyl)-glycine (AVG) (Nayani et al. 1998), has been reported to prolong the shelf life of ethylene-sensitive flowers. However, the use of different chemicals has a variety of drawbacks. For example, NBD has a foul odour and is carcinogenic (Reid and Wu 1992), treatment of flowers with AVG and STS increases the cost, and above all, treating flowers with high concentration of chemicals is potentially phytotoxic and environmentally hazardous (Abeles et al. 1992).

The ability of some rhizospheric plant growth-promoting bacteria (PGPB) in slowing the senescence of isolated carnation flower petals has previously been demonstrated (Nayani et al. 1998). In plant tissues, 1-aminocyclopropane-1-carboxylate (ACC) is the immediate precursor of ethylene and is synthesized from S-adenosylmethionine by the action of the enzyme ACC synthase. Chemical inhibitors of ethylene production (such as AVG) normally target this pyridoxal phosphate dependant (Yu et al. 1979) enzyme (Penrose and Glick 1997). Under physical or chemical stress conditions, ACC synthase is induced with the result that more S-adenosylmethionine is converted to ACC (Penrose and Glick 1997), consequently increasing the level of endogenous ethylene. ACC deaminase-containing PGPB (Glick et al. 1998, 2007a,b; Glick 2004) can sequester and subsequently cleave ACC, thus preventing deleterious concentrations of ethylene from accumulating inside plant tissues.

The use of ACC deaminase-containing PGPB endophytes provides a number of potential advantages compared to rhizospheric PGPB with the same activity, especially when the organisms are used in a field or other nonlaboratory setting (Barac et al. 2004; Rashid et al. 2012). In this work, we have investigated the ability of two different ACC deaminase-containing PGPB endophytes, their ACC deaminase minus mutants and one rhizospheric bacterium, Pseudomonas putida UW4, to delay the senescence of mini carnation (Dianthus caryophyllus) flowers.

Materials and methods

Plant material

Mini carnation plants (D. caryophyllus) were grown from seed and maintained in a green house within the Department of Biology, University of Waterloo. The flowers were cut at full bloom; the stems were trimmed to a uniform length of approx. 8 cm and then processed immediately.

Bacterial strains and growth conditions

Two recently isolated bacterial endophytes (Rashid et al. 2012) and their ACC deaminase minus mutants were used in this study. Pseudomonas fluorescens YsS6 and Pseudomonas migulae 8R6 were initially isolated from tomato plants grown in soils collected from France and Canada, respectively. Both strains possess ACC deaminase, promote root elongation in canola (Brassica campestris) seedlings, produce siderophores, synthesize indoleacetic acid (IAA) and solubilize phosphorus (Rashid et al. 2012). The acdS minus mutants were constructed by the insertion of a tetracycline resistance gene at position 237 in the acdS gene of Ps. migulae 8R6 and at position 323 in the acdS gene of Ps. fluorescens YsS6 as described by Sun et al. (2009) for the bacterial endophytic PGPB Burkholderia phytofirmans PsJN. It was confirmed that the acdS minus mutants no longer exhibit significant ACC deaminase activity: wild-type Ps. fluorescens YsS6 had an ACC deaminase activity of 7·42 μmol μg−1 h−1 and its ACC deaminase minus mutant activity of 0·11 μmol μg−1 h−1, while wild-type Ps. migulae 8R6 and its ACC deaminase minus mutant had activity levels of 6·27 and 0·03 μmol μg−1 h−1, respectively. The apparent residual enzyme activity in the two mutants (in both cases approx. 1% of the wild-type value) is deemed to be insignificant and is within the margin of error of this assay.

A root-colonizing (rhizospheric) PGPB, Ps. putida UW4 (Glick et al. 1995; Hontzeas et al. 2005), was also used to treat the mini carnation cut flowers.

All bacterial strains were grown in 15 ml of tryptic soy broth (TSB) (Bacto™ Becton, Dickinson and Company, Sparks, MD, USA) with appropriate antibiotics at 30°C for 24 h. Overnight cultures were washed once with 0·85% NaCl, centrifuged at 4500 g for five minutes and then diluted with 0·85% NaCl to an absorbance of 0·15 ± 0·02 at 600 nm.

Chemicals

The compounds ACC and sodium chloride (NaCl) were purchased from Calbiochem®; Merck KGaA, Darmstadt, Germany, and AVG was purchased from Sigma-Aldrich (Steinheim, Germany). All three chemicals were dissolved and diluted in milli-Q water. Stock solutions of ACC (0·5 mol) and AVG (0·25 mol) were prepared and then stored in small aliquots at −20°C.

Flower senescence experiment

Each flower was placed in a separate 13 × 100-mm glass test tube filled with 8 ml of either diluted bacterial suspension, 1 × 10−4 mol ACC solution, or 1 × 10−4 mol AVG solution; then, the tube was placed in a tube rack and incubated at room temperature (21–23°C for 11 days. A set of 105 flowers was used for each treatment. Flowers treated only with 0·85% NaCl were used as a negative control. Flower senescence was recorded on a scale from zero to four where zero is a freshly cut flower, and four is a completely senesced flower (Fig. 1).

Figure 1.

Levels of senescence of different coloured mini carnation flowers on a scale from 0 to 4, no senescence or fresh flower (a) to 1 (b), 2 (c), 3 (d) and 4 (e) completely senesced flower.

Scanning electron microscopy

Different portions of treated flowers were examined for the presence or absence of bacteria inside the tissue. A 1–1·5 cm piece of the flower stem or flower thalamus was removed, and the interior portion of the tissue was exposed by cutting it lengthwise; it was then fixed in a 2·5% glutaraldehyde solution in 0·1 mol phosphate buffer (pH 7·0) overnight at 4°C and then washed twice with phosphate buffer. After dehydration in a series of acetone solutions (20, 50, 70, 95, 100%), the specimens were dried to critical point in liquid CO2, mounted on a stud, coated with gold and examined under a scanning electron microscope (s570; Hitachi, Mito City, Japan) at 15 kV accelerating voltage (Matzk et al. 1996). The entire experiment was repeated three times with two samples for each treatment (i.e. control flower, flower treated with wild-type strain, flower treated with mutant strain).

Results

Flowers treated with the ACC solution died earlier than any of the other treated flowers, that is, by day 6 of the treatment, control flowers developed senescence symptoms and died by day 8, while the AVG treatment delayed the onset of senescence by around 1 day compared to the control (Fig. 2). Pseudomonas putida UW4-treated flowers senesced surprisingly quickly, that is, at a rate between the control and the ACC-treated flowers (Fig. 2).

Figure 2.

Senescence of flower from untreated carnation, served as control, and carnation treated with 1-aminocyclopropane-1-carboxylate, l-α-(aminoethoxyvinyl)-glycine and Pseudomonas putida UW4. Each point represents a mean (±SE; n = 105) value of senescence level.

Both wild-type endophytes (Ps. fluorescens YsS6 and Ps. migulae 8R6) delayed flower senescence to approximately the same extent, providing an extra 2 days of protection as compared to AVG and about 3–4 days of protection when compared to no treatment (control) (Figs 2 and 3). When the cut flowers were treated with ACC deaminase minus mutants of Ps. fluorescens YsS6 and Ps. migulae 8R6, the flowers behaved in a manner similar to control flowers. That is, unlike their wild-type counterparts, the mutant bacterial strains failed to delay flower senescence compared to the untreated flowers (control) (Fig. 3).

Figure 3.

Senescence of flower from untreated carnation, served as control, and carnation treated with suspension of Pseudomonas fluorescens YsS6 wild-type, Ps. fluorescens YsS6 1-aminocyclopropane-1-carboxylate (ACC) deaminase minus mutant, Ps. migulae 8R6 wild-type and Ps. migulae 8R6 ACC deaminase minus mutant. Each point represents a mean (±SE; n = 105) senescence level value.

Scanning electron micrographs confirmed the presence of both endophytes as well as their ACC deaminase minus mutants within the stems of the cut flowers (Fig. 4). On the other hand, no bacteria could be seen in the flower thalamus of the endophyte-treated plants (Fig. 4). The untreated control plants did not harbour any bacteria in either the stem or the thalamus.

Figure 4.

Scanning electron micrographs of treated flower's parts viewed on day 8 of the treatment. No bacteria were found in the flower thalamus (a) and stem (b) of control (untreated) flowers. The thalamus of flowers treated with both wild-type bacterial endophytes (c) and mutant bacterial endophytes (e) showed no bacterial presence. Rod shaped bacteria were present in flower stems treated either with wild-type bacterial endophytes (d) or mutant bacterial endophytes (f).

Discussion

Carnation is a typical ethylene-sensitive flower that produces ethylene through an autocatalytic pathway and enters into rapid flower senescence (Rahemi and Jamali 2011). Ethylene production is influenced by a number of factors inside the tissue of an ethylene-sensitive flower. The presence of auxins (e.g. IAA) positively regulates ethylene production by stimulating transcription of genes encoding the enzyme ACC synthase that synthesizes ACC, immediate precursor of ethylene (Penrose and Glick 1997). A bacterium that possesses the enzyme ACC deaminase can utilize and cleave ACC into α-ketobutyrate and ammonia and thus facilitates a reduction in the level of ethylene in the plant associated with that bacterium (Glick 1995, 2005).

Previously, our laboratory demonstrated that ACC deaminase-containing plant growth-promoting rhizobacteria are able to delay the senescence of carnation flower petals to a significant level (Nayani et al. 1998). In that study, (i) only rhizospheric PGPB were utilized, and (ii) senescence was delayed only when the carnation petals were removed from the flower, and the individual petals were incubated in solutions of the PGPB. Importantly, in that study, rhizospheric bacteria did not have any effect on the senescence of cut flowers per se, only on isolated flower petals. This is presumably because the rhizosphere bacteria were not taken up by the cut flowers. These experiments demonstrated that root-colonizing bacteria are able to delay the carnation flower senescence only if an exchange of plant-produced ACC from flower tissue to the bacteria can occur; in that case, a wound created by excising the petal from the flower provided a channel for this interaction. On the other hand, when root-colonizing bacteria were sprayed on whole flowers or the stems of cut flowers were incubated in a suspension of these organisms, flower senescence was not affected. This is because of the fact that rhizospheric bacteria could not establish an association with the complete flower and could not sequester and cleave ACC so that they had no influence on senescence events.

On the basis of the observations that bacterial endophytes can be taken up internally by plants and establish a relationship that is generally beneficial to those ‘colonized’ plants, and bacteria with ACC deaminase activity can act as a sink for plant ACC, it was hypothesized that ACC deaminase-containing bacterial endophytes might effectively delay flower senescence. To test this hypothesis, in the present study, cut mini carnation flowers were treated with the bacterial endophytes, Ps. fluorescensYsS6, Ps. migulae 8R6, and their respective ACC deaminase minus mutants. In these experiments, it was first demonstrated that cut mini carnation flowers behaved as expected when treated with either the ethylene precursor ACC or the ethylene inhibitor AVG (Fig. 2). That is, treatment with ACC hastened flower senescence, while treatment with AVG prolonged the lifetime of the mini carnation flowers compared to the control. When flowers were incubated with a suspension of wild-type ACC deaminase-containing bacterial endophytes, flower senescence was delayed to an even greater extent (Fig. 3) than when AVG was added (Fig. 2). Flowers treated with ACC deaminase minus mutants of the same bacterial endophytes senesced at a rate that was essentially the same as the control flowers and significantly faster than when the flowers were treated with the wild-type endophytic strains (Fig. 3). This observation is consistent with the suggestion that ACC deaminase is the key factor that wild-type bacterial endophytes utilize to delay flower senescence.

Scanning electron micrographs of treated flower tissues revealed the presence of rod-shaped bacteria (Fig. 4d,f) in the flower stems in all instances where the flowers were treated with a bacterial endophyte regardless of whether a wild-type or mutant strain was utilized. The fact that bacteria were never found in tissues of the flower thalamus indicates that these endophytes do not colonize the reproductive part of the plant.

Cut flowers treated with Pseudomonas putida UW4 senesced earlier than the control flowers (Fig. 2). As a consequence of the lack of a direct association of this bacterium with either the flower or the stem, Ps. putida UW4 did not affect flower senescence. However, Ps.putida UW4 is able to synthesize IAA that may be taken up by the flower stems and thereby possibly elevate the plant ethylene level by stimulating transcription of the enzyme ACC synthase, resulting in an increase in the rate of flower senescence. Thus, a comparison of the behaviour of cut flowers treated with endophytes as compared to rhizosphere-binding bacteria indicates that only endophytic bacteria (that contain ACC deaminase) are able to delay flower senescence.

The use of ACC deaminase-containing plant growth-promoting endophytes to delay the senescence of cut flowers may be an attractive prospect for the flower industry. The delay of cut flower senescence engendered from plants treated with ACC deaminase-containing plant growth-promoting endophytes at the seedling stage is an intriguing possibility. However, given the necessity of establishing stable endophytic colonization of each plant, it remains to be determined experimentally whether this is a viable approach. Some commercially important flowers (such as rose, carnation, zinnia and geranium) exhibit very high sensitivity to ethylene (Woltering and Van Doorn 1988), and treatment with naturally occurring PGPB endophytes may extend their shelf life to a significant extent without the use of potentially problematic chemicals.

Acknowledgements

We are very thankful to Dale Weber for help with scanning electron microscopy and to Lynn Hoyles for growing and continuously providing fresh mini carnation flowers. This work was supported by grants from the Natural Sciences and Engineering Research Council of Canada to B.R.G. and T.C.C. S.A. is the recipient of scholarship from the Higher Education Commission (H.E.C.), Pakistan.

Ancillary

Advertisement