endophytic micro-organisms

The present invention relates to the production of plant seeds comprising endophytes.

In spite of limited arable land coupled with rising demand of a steadily increasing human population which could hit 9 billion by 2050, food supply is a global challenge making produc ^¬ tion of economically attractive and high quality food, free from unacceptable levels of chemicals, a dire need. The use of micro ^¬ organisms with the aim of improving plant growth and health is an important practice and necessary for agriculture.

During the past couple of decades, plant growth promoting rhizobacteria (PGPR) have received worldwide importance and ac ^¬ ceptance in agricultural practice. These microorganisms are the potential tools for sustainable agriculture because they not on ^¬ ly ensure the availability of essential nutrients to plants but also enhance the nutrient use efficiency.

Endophytic bacteria may in future be even more important than rhizosphere bacteria in promoting plant growth because they escape competition with rhizosphere microorganisms and establish a more intimate contact with plant tissues. In addition, the in ^¬ herent nature of certain endophytes to potentially colonize plants in a systematic manner provides a novel approach as a de ^¬ livery system to plants for various beneficial traits.

Bacterial mechanisms of plant growth promotion include bio ^¬ logical nitrogen fixation (BNF) , synthesis of phytohormones and vitamins, environmental stress relief, synergism with other bac ^¬ teria-plant interactions, inhibition of plant ethylene synthe ^¬ sis, as well as increasing availability of nutrients like phos ^¬ phorus, iron and other micro-elements, and growth enhancement by volatile compounds.

Numerous application strategies have been used for PGPR/endophytic bacteria at the experimental levels, ranging from seed treatment and soil application to stem injection and foliar spray. Seed treatment (soaking and embedding in carrier material) with bacterial inocula prior to sowing is the tradi ^¬ tional, most commonly used and easiest means of inoculation. Peat (carrier based mixing) inoculants have been the standard for the inoculation industry; however, several other commercial preparations have been marketed. Crop Genetics International Ltd. developed a seed inoculation technique by applying a pres ^¬ sure differential to infuse the bacterial suspension into im ^¬ bibed seeds and re-drying the seeds (US 5, 415, 672 A) .

To get benefits from bacterial inocula, it is crucial to ap ^¬ ply (technique and timing) bacterial inocula in a viable way. In addition, it is equally important that the microorganisms remain viable during several months of seed storage and are easily ac ^¬ tivated and colonize the plant environment. However, by using conventional methods (carrier based, liquid broth and soil ap ^¬ plication; see also: US 7,084,331 B2, US 7,906,313 B2, US 7,037,879 B2), the viability of bacteria is subjected to the hazards of drying, fertilizer contact, seed coat toxicity, in ^¬ compatible pesticidal and mineral additives. Besides this, sev ^¬ eral soil and environmental stresses affect the surviv ^¬ al/colonization efficiency of the inoculant strains. Bacterial population density, the host plant species, agronomic practices and climatic conditions are among the important factors for the success of biological plant fertilization. Examples for use of endophytes as plant growth enhancers, bio-pesticides, pathogen treatment or pest tolerance agents are disclosed e.g. in WO 00/29607 Al, WO 2011/117351 Al, WO 2010/115156 A2 , WO 2007/107000 Al, WO 2007/021200 Al, US 2012/144533 Al, US 4,940,834 A, CA 2562175 Al and WO 2011/082455 Al .

With current inoculation methods, however, colonisation of the plants with the desired endophytic microorganisms is diffi ^¬ cult and often not reproducible, which makes it difficult to ap ^¬ ply this technology on an industrial scale. For example, micro ^¬ organisms used in seed coating often do not survive well or are unable to colonize the plant (because the microorganisms on the outside cannot enter the seed or plant) . If the plant is mechan ^¬ ically (or otherwise) wounded to provide an entry, this puts the health of the seed, seedlings or plant at risk, because harmful microorganisms could enter the plant as well in an unprotected manner. Moreover, even if the microorganisms can colonise a given plant, there can be a natural loss of viability and the effi ^¬ ciency of colonization can be low. More complex inoculation techniques (e.g. by applying vacuum or pressure infiltration, inoculation by injection, etc.) are also causing risk for the plant and are - most importantly - not transferable to a large scale or industrial applicability and are therefore not effec ^¬ tive .

It is an object of the present invention to provide an im ^¬ proved method for producing seeds containing endophytic microorganisms. The method should provide seeds with a reproducible and defined composition of endophytic microorganisms enabling the growth of plants with the desired properties due to the presence of such endophytic microorganisms. It is another object to pro ^¬ vide methods for introducing endophytic microorganisms into plant seeds for microorganisms that are not or are not neces ^¬ sarily occurring in seeds.

Therefore, the invention provides a method for producing plant seed containing endophytic microorganisms which is charac ^¬ terised by the following steps:

contacting a flowering plant with a preparation of endophytic microorganisms, whereby the endophytic microorganisms enter the plant via the flowers and are conveyed to seed produced by the plant; and

obtaining the plant seed containing endophytic microorganisms from the plant.

The term "endophyte" means - in its broadest meaning - the location of an organism, with "endo" means "inside" and "phyte" means "plants". Therefore, endophyte - in its broadest meaning - refers to organisms that live within plants. Fungi and bacteria are the most common organisms associated with the term endo ^¬ phyte .

An important feature of endophytic microorganisms is that they occupy internal tissues of plants without causing substan ^¬ tive damage to their hosts. In many cases endophytes are respon ^¬ sible for conferring one or more advantages to the plant. For the present invention, an "endophytic microorganism" is defined in this usual way: as a microorganism that lives within a plant and is responsible for plant beneficial effects, for example tolerance to drought, metals, disease (e.g. increasing re ^¬ sistance to pathogens and parasites), and herbivory, and/or growth promotion and nutrient acquisition, production of phyto- hormones, antibiotics (protection against microorganisms being harmful for seeds and plants) or siderophores , pesticides; pro ^¬ moting biological nitrogen fixation, etc. (as some (of many) examples: chilling tolerance (Burkholderia) , salinity stress (Achrobacter, Azospirillum) , tolerance to drought (Burkholderia , Pantoea) , metals, disease [Bacillus , Pseudomons , Xanthomonas) , growth promotion ( Azospirillum, Azotobacter, Bacillus, Burkholderia, Enterobacter, Klebsiella, Pantoea and Pseudomonas) and nu ^¬ trient acquisition (Pseudomonas, Bacillus, Rhizobium, Micrococ ^¬ cus, Flavobacterium, Burkholderia, Achromobacter, Erwinia, and Agrobacterium) (US 7,906,313 B2)).

Endophytic organisms associated with plants are varied and complex. Endophytic microbes occupy a relatively privileged niche within a plant and frequently contribute to plant health or growth. Co-evolution may exist between endophytes and their host e.g. in resisting to environmental stresses. Endophytes have been targeted as valuable sources of new bioactive com ^¬ pounds. Endophytes inhabit plant tissues, particularly the so- called intercellular space, space between cells. Endophytic mi ^¬ croorganisms have been found in virtually every plant studied, where they colonize the internal tissues of their host plant and can form a range of different relationships including symbiotic, mutualistic, commensalistic and trophobiotic . Most endophytes appear to originate from the rhizosphere or phyllosphere ; howev ^¬ er, some may be transmitted through the seed. Endophytic micro ^¬ organisms can promote plant growth and yield and can act as bio- control agents. Endophytes can also be beneficial to their host by producing a range of natural products that are not only bene ^¬ ficial to the plant but could also be harnessed for potential use in medicine, agriculture or industry. In addition, it has been shown that they have the potential to remove soil contami ^¬ nants by enhancing phytoremediation and may play a role in soil fertility through phosphate solubilisation and nitrogen fixation. There is increasing interest in developing the potential biotechnological applications of endophytes for improving phy ^¬ toremediation and the sustainable production of non-food crops for biomass and biofuel production.

The method according to the present invention uses a com ^¬ pletely new and diligent way for introducing endophytes into plants which turned out in the course of the present invention to be very effective, yet without harming the plants or seeds during or after inoculation. Applying the microorganisms to the flowering plants uses a natural entry into the plant which in ^¬ troduces the endophyte in an efficient manner into the next gen ^¬ eration plant seeds. Within the course of the present invention it also turned out that when the microorganisms are applied to the plant at the time of flowering (e.g. by simple spraying), the microorganisms get entry when grain formation starts and es ^¬ tablish populations inside the seed. The method of the present invention can facilitate the crop productivity by enhancing ger ^¬ mination, seedling vigour and biomass in comparison with non- treated control. Moreover, the introduction of the beneficial microorganisms inside seed instead of external application by e.g. seed coating makes the inocula less susceptible to environ ^¬ mental perturbation and better compatible to chemical seed coat ^¬ ings (pesticides and herbicides) . Using bacterial colonized seeds, the plant growth and biomass are statistically similar as the conventional inoculation method e.g. exogenous seed soaking and soil inoculation (that are more laborious and less practica ^¬ ble in certain circumstances) .

Accordingly, the present invention provides a new concept of applying endophyte strains for improved plant growth and vitali ^¬ ty - the integration of the bacteria or fungus strain inside the plant seed. The microorganisms are e.g. sprayed on the parent flowering plants, enter the plants and colonize the emerging seeds. The microorganisms may also be applied by specific in ^¬ struments to the flower, e.g. by a spatula, a syringe or an in ^¬ oculating loop. Another preferred embodiment for administering the endophytes to the flower of a plant is performed by employ ^¬ ing pollen-feeding insects, preferably humble-bees, that carry the endophytic microorganisms. Such insects (besides humble-bees also honey-bees, butterflies, some wasp and fly species or other "pollinators" may be used) can even be provided from commercial sources and contacted with the endophytes before they are re ^¬ leased to contact the flowering plants. The microorganisms are preferably provided at a body part of these insects that has the highest probability to contact the flower of the plant (e.g. the legs or the ventral part of the body) .

By planting the internally colonized seeds the endophytes get activated and proliferate and colonize the offspring genera ^¬ tion plants. Internally colonized seeds may result (depending on the nature of the endophyte) in improved biomass production and plant vitality in the subsequent plant generation. The positive effects are at least comparable (if not improved) to that ob ^¬ served after external application of endophytes, but makes the inocula less susceptible to environmental perturbation and bet ^¬ ter compatible to chemical seed coatings (pesticides and herbi ^¬ cides) . With the present invention it is also possible to intro ^¬ duce endophytes into seeds that are not or are not necessarily present in seeds. Virtually any kind of endophytes can be intro ^¬ duced into seeds by the method according to the present, provid ^¬ ed that these endophytes have at least a basic power to estab ^¬ lish themselves in the seeds.

None of the prior art methods, especially not the methods disclosed in WO 00/29607 Al, WO 2011/117351 Al, WO 2010/115156 A2, WO 2007/107000 Al, WO 2007/021200 Al, US 2012/144533 Al, US 4,940,834 A, CA 2562175 Al and WO 2011/082455 Al, aim at provid ^¬ ing methods for providing seeds comprising selected endophytes. The main goal of these methods according to the prior art is al ^¬ ways the provision of the endophytes to the very plant treated and not - as in the present invention - to supply a mother plant with the endophytes of interest and to obtain endophyte contain ^¬ ing seeds from this mother plant for rising daughter plants already containing the endophytes and, optionally, passing the en ^¬ dophytes further to their own daughter generation. Accordingly, the technology provided with the present invention can provide seeds with completely novel characteristics, e.g. having a unique set-up of endophytes (for example by having one single endophyte species being predominantly present in the seeds (e.g. representing more than 50%, or more than 70% or even more than 80% of the total of endophytes in the seed) ) .

Suitable plants include both monocots and dicots (including eudicots) that can be colonized by the endophytic microorganisms according to the present invention. Of course, the plant has to be a flowering plant (angiosperm) in order to transfer the microorganisms to the plant in the course of the flowering phase. The resulting seeds contain the inoculated endophytes in an ef ^¬ ficient concentration. Plants grown from such seeds contain the endophytes and the beneficial properties of the endophyte can develop in the seeds or plants. Accordingly, the plants arising from such seeds - wherein the endophyte can express its benefi ^¬ cial function to the plant - may be at any stage of growth, in ^¬ cluding seeds, seedlings, or full plants. The present invention is therefore not simply about spraying the microorganisms to a given plant (or seed) in order to provide the beneficial endo ^¬ phytic effect to this plant, but it provides a method which safeguards presence of endophytes in the seeds generated from this plant and therefore for the next generations of the plant. This essentially differs from all inoculation strategies applied before (seed impregnation, spraying the microorganisms to the seeds, germs or the whole plants) , because the present method deals with the production of seeds which contain a reproducible endophyte set-up.

In a preferred embodiment, the target plant is a plant of the family Graminae (grasses) . The grass plants into which these endophytes are introduced may be any of the useful grasses be ^¬ longing to the genuses Agropyron , Agrostis , Andropogon , An- thoxanthum, Arrhenatherum, Avena , Brachypodium, Bromus , Chloris , Cynodon , Dactylis , Elymus , Eragrostis , Festuca , Glyceria , Hiero- chloe, Hordeum, Lolium, Oryza, Panicum, Paspalum, Phalaris , Phleum, Poa, Setaria, Sorghum, Triticum, Zea and Zoysia.

In a preferred embodiment, the target plant is selected from the wheats, including, Triticum monococcum, Triticum turgidum, Triticum timopheevi (Timopheev's Wheat) and Triticum aestivum (Bread Wheat) .

In another preferred embodiment, the target plant is a corn of the genus Zea. Zea is a genus of the family Gramineae (Poace- ae) , commonly known as the grass family. The genus consists of some four species: Zea mays, cultivated corn and teosinte; Zea diploperennis litis et at., diploperennial teosinte; Zea luxuri- ans (Durieu et Asch.) Bird; and Zea perennis (Hitchc.) Reeves et Mangelsd., perennial teosinte.

Other useful grasses which may be used on an industrial ba ^¬ sis are rye grasses and bluegrasses. Bluegrasses known in the art include Kentucky bluegrass, Canada bluegrass, rough meadow grass, bulbous meadow grass, alpine meadow grass, wavy meadow grass, wood meadow grass, Balforth meadow grass, swamp meadow grass, broad leaf meadow grass, narrow leaf meadow grass, smooth meadow grass, spreading meadow grass and flattened meadow grass.

In a preferred embodiment, the plants for which seeds are produced by the method according to the present invention are dicots, including eudicots such as tomato, watermelon, squash, cucumber, strawberry, pepper, soybean, peanut, Brassicaceae, especially rape, sunflower, sugar beet, cotton, alfalfa and ara- bidopsis .

Accordingly, the plant is preferably selected from the group of Graminae (grasses) , preferably grasses of the genuses Agropy- ron, Agrostis , Andropogon , Anthoxanthum, Arrhenatherum, Avena , Brachypodium, Bromus , Chloris , Cynodon , Dactylis , Elymus , Erag- rostis , Festuca , Glyceria , Hierochloe, Hordeum, Lolium, Oryza, Panicum, Paspalum, Phalaris , Phleum, Poa, Setaria , Sorghum, Triticum, Zea, especially Zea mays, cultivated corn and teosin- te, Zea diploperennis litis et at., diploperennial teosinte, Zea luxurians (Durieu et Asch.) Bird; and Zea perennis (Hitchc.) Reeves et Mangelsd., perennial teosinte. and Zoysia; cheats, preferably Triticum monococcum, Triticum turgidum, Triticum timopheevi (Timopheev's Wheat) and Triticum aestivum (Bread Wheat) ; preferably rye grasses and bluegrasses, especially Ken ^¬ tucky bluegrass, Canada bluegrass, rough meadow grass, bulbous meadow grass, alpine meadow grass, wavy meadow grass, wood mead ^¬ ow grass, Balforth meadow grass, swamp meadow grass, broad leaf meadow grass, narrow leaf meadow grass, smooth meadow grass, spreading meadow grass and flattened meadow grass; dicots, pref ^¬ erably eudicots, especially tomato, watermelon, squash, cucum ^¬ ber, strawberry, pepper, soybean, peanut, Brassicaceae, especially rape, sunflower, sugar beet, cotton, alfalfa and ara- bidopsis .

The method according to the present invention is specifical ^¬ ly suitable for providing seeds of transgenic plants. By the present invention, transgenic plants are obtainable that - be ^¬ sides their advantageous properties provided by the transgene - also contain "tailored" endophyte properties that can selective ^¬ ly be construed and provided by the present invention.

According to a preferred embodiment of the present method the endophytic microorganism is an endophytic bacterium, preferably selected from Burkholderia , Rhizobium, Bradyrhizobium, Mesorhizobium, and Sinorhizobium, Herbaspirillum, Azospirillum, Acetobacter, Arthrobacter, Bacillus , Paenibacillus , Streptomy- ces , Enterobacter _r and Pseudomonas , Pantoea and Enterobacter, especially Burkholderia phytofirmans .

According to another preferred embodiment, the endophytic microorganism is an endophytic fungus, preferably selected from Curvularia , Mycorrhiza , Pififmospora, Trichoderma , and Colleto- trichurn.

In a preferred embodiment according to the present inven ^¬ tion, contacting the flower of a plant with a preparation of endophytic microorganisms is performed via spraying the microor ^¬ ganisms at the time of flowering. Spraying is specifically use ^¬ ful in an industrial production method. Other methods include the inoculation by brushing, by an inoculating loop, by applying droplets, etc.; however, spraying can be easily automated, e.g. in glasshouse cultures.

Inoculation is done by applying the culture of the endophyte to the flowering plant. It is recommendable to safeguard condi ^¬ tions which are favourable to the microorganisms used. The mi ^¬ croorganisms are usually applied in suspension at a suitable concentration. Accordingly, it is preferred to contact the flower of a plant with a preparation of endophytic microorganisms by applying the microorganisms in a suspension of 10 ^s to 10 ¹⁰ cfu/mL, preferably of 10 ⁷ to 10 ⁹ cfu/mL, especially of 10 ⁸ to 10 ⁹ cfu/mL .

The seeds obtained by the present method can be treated like normal seeds. The beneficial properties (the endophytes) remain safely packed inside the seed preventing the exposure of hazards from outside (which usually causes damage to cultures which are applied when the seeds are only coated) . Accordingly, the seeds may be stored for considerable time without significant loss of their endophytic activity. Preferably, the plant seed obtained by the present method containing endophytic microorganisms from the plant is stored for at least 1 month, preferably for at least 3 months, especially for at least 6 months.

Also much longer storage times are, of course, possible for the seeds produced according to the present invention. It is therefore also preferred that the plant seed obtained by the present method containing endophytic microorganisms from the plant is stored for at least 12 months, preferably for at least 2 years, especially for at least 3 years.

The method according to the present invention is suitable for providing virtually any endophyte-containing seed, because the transfer of the microorganisms from the flower to the seed is a way with low hazard exposure (to plant and endophyte) . It is specifically suitable for producing seeds with an endophyte which is in principle known to naturally proliferate in plants, especially in the given plant, i.e. a "naturally obtainable en ^¬ dophyte". These endophytes are derivable from natural sources from the same plant type or from other plant types. According to a preferred embodiment, the endophytic microorganism is there ^¬ fore a naturally obtainable endophyte.

It is also possible to use the present method for providing seeds with artificially created or optimised microorganisms, e.g. recombinantly engineered bacteria or fungi; or strains which have been optimised by various culture techniques and/or selection rounds. Another preferred embodiment of the present invention is therefore to use a recombinantly produced bacterium as the endophytic microorganism.

As already mentioned, the seeds obtained by the present method can be further processed in usual ways. For example, it can be treated with various substances which further promote the plants to be produced from the seeds, e.g. by impregnating the seeds with growth promoting agents or other chemicals beneficial for plant health, such as herbicides, pesticides, antibiotics, etc.. It is, of course, also possible to provide a coating with further (or the same) endophytic microorganisms as the microor ^¬ ganism according to the present invention. According to a preferred embodiment of the present invention, the obtained plant seed containing endophytic microorganisms is therefore subjected to a seed impregnation step.

This invention also relates to the seeds obtainable by the method according to the present invention which show - compared to seeds according to the prior art - a unique endophyte set-up.

According to a preferred embodiment, the present invention provides seeds which can be grown to plants that are improved (compared to the wild type plants) with respect to stress toler ^¬ ance. "Stress" in this context may be an environmental stress, including, high temperature, drought, metals and metal ions, which cause a variety of plant problems and/or death, and abnor ^¬ mal pH (including both acidic and/or alkaline) . With the seeds produced by the present invention plants can be obtained that have reproducibly improved stress resistance, e.g. at least about a 5, 10, 20, 25 and 50% change in thermotolerance, at least about a 5, 10, 20, 25 and 50% change in drought tolerance, at least about a 5, 10, 20, 25 and 50% change in metal toler ^¬ ance, or at least about a 5, 10, 20, 25 and 50% change in pH tolerance (each as measured according to US 7,906,313 B2, and compared to controls without the method according to the present invention applied) .

According to a preferred embodiment, the seeds according to the present invention can be grown to plants with increased growth. Growth enhancement is generally measured as a comparison of plants cultured from seeds made according to the present in ^¬ vention with control plants lacking this endophyte composition. Differences in plant size, including leaf, root and stems are generally measured by weight, with increased growth being meas ^¬ ured as at least about an at least 2 % difference, preferably an at least 3 % difference (which can already be regarded as a very significant gain in yield. Even more preferred, in some instanc ^¬ es, a 5-10% difference between control plants and the plants grown from the seeds according to the present invention may be obtained, with at least about a 25% difference being specifical ^¬ ly preferred.

The method according to the present invention enables the creation of completely new seeds/endophyte combinations. One of the most significant properties of preferred seeds obtainable by the present invention is the possibility to provide predominant endophyte populations in the seeds. Normally, seeds containing endophytes contain a diverse population of many different endo ^¬ phytic microorganisms with usually more than 10 or even more than 20 different identifiable culturable strains (or even more than 30) (but none of these strains being predominant), the meth ^¬ od according to the present invention enables the production of seeds with a predominant species of endophytic microorganism. Accordingly, preferred seed preparations which are provided by the present invention contain seeds having an endophytic micro- organism population wherein more than 30%, preferably more than 40%, especially more than 50%, of the endophytic microorganisms represent the inoculant strain. This means that most (more than 50%, preferably more than 60 %, especially more than 70 %) of the seeds in the preparation contain more than 30%, preferably more than 40%, especially more than 50%, endophytic microorgan ^¬ isms comprising the inoculant strain.

It is even possible to provide a seed preparation containing seeds, wherein more than 60%, preferably more than 70 ~6 , more preferred more than 80%, especially more than 90%, endophytic microorganisms of a single species (the endophytic microorganism of the inoculant strain) . This enables the production of seeds containing e.g. more than 60%, preferably more than 70 % , espe cially more than 80%, of the applied endophytic strain (e.g. within a single field) .

A specific embodiment of the present invention is therefore a seed preparation obtainable by a method according to the pre ^¬ sent method.

According to a preferred embodiment, the present invention provides a seed preparation containing seeds having more than 30%, preferably more than 40%, especially more than 50%, of the endophytic microorganisms are Burkholderia phytofirmans, espe ^¬ cially Burkholderia phytofirmans PsJN (DSM17436) ; Pantoea sp. FD17 or Paenibacillus sp . S10., Actinobacter sp . S9, Bradyrhizo ^¬ bium sp . NC92 and Bradyrhizobium japonicum TAL379.

The present invention also provides seeds obtainable by the method according to the present invention with unique characteristics, e.g. with a predominant endophyte species as disclosed above. A preferred embodiment of the present invention is there ^¬ fore drawn to seeds, especially maize seeds, obtainable by a method according to the present invention, wherein the endophyt ^¬ ic microorganisms are preferably present in a population density of 10 ² to 10 ⁵ cfu/g fresh weight .

According to a preferred embodiment, the present invention provides maize seed obtainable by a method according to the pre ^¬ sent invention, preferably wherein the endophytic microorganisms are Burkholderia phytofirmans, especially in a population densi ^¬ ty of 10 ² to 10 ⁵ cfu/g fresh weight of seed. It is known that in maize, usually the viable population densities are much lower (for sweet corn, it was reported that such concentrations are below 10 ¹ cfu/g fresh weight (Kaga et al . Microbes Environ 24

(2009), 154-162)); in contrast thereto, the seeds according to this preferred embodiment contain at least 10 ² , preferably at least 10 ³ , especially at least 10 ⁴ , cfu/g fresh weight of one species, especially of Burkholderia phytofirmans (strain PsJN) . Accordingly, the endophyte concentration of such seeds contains a predominant strain, which is not the case in natural plants or plants having been inoculated with prior art inoculation methods .

The seeds according to the present invention provide a mar ^¬ ketable seed product containing a predetermined weight or volume of seeds with a uniform endophyte composition. For example, a marketable seed product containing at least 100 g seeds, prefer ^¬ ably at least 1 kg seeds, more preferred at least 5 kg seeds, especially at least 10 kg seeds, can be provided by the method according to the present invention that contains - as a whole product - more than 30%, preferably more than 40%, especially more than 50%, of a single species of an endophytic microorgan ^¬ ism, i.e. the inoculant strain. According to a preferred embodi ^¬ ment, the present invention provides a marketable seed product containing at least 100 g seeds, preferably at least 1 kg seeds, more preferred at least 5 kg seeds, especially at least 10 kg seeds, wherein - as a whole product - more than 50%, preferably more than 60 %, especially more than 70 % of a single species of an endophytic microorganism, i.e. the inoculant strain, are con ^¬ tained. According to an even more preferred embodiment, the pre ^¬ sent invention provides a marketable seed product containing at least 100 g seeds, preferably at least 1 kg seeds, more pre ^¬ ferred at least 5 kg seeds, especially at least 10 kg seeds, wherein - as a whole product - more than 75%, more preferably more than 80%, especially more than 90%, endophytic microorgan ^¬ ism of a single species (the endophytic microorganism of the in ^¬ oculant strain) are contained.

Such uniformity in endophytic composition is unique and is extremely advantageous for high-tech and/or industrial agricul ^¬ ture. It allows significant standardisation with respect to qualitative endophyte load of seed products. The term "marketa ^¬ ble seed product" means any commercially usable product contain- ing plant seeds in a suitable package (e.g. a box, a bag, an en ^¬ velope or any other container used for storing, shipping or offering plant seeds for sale) . Suitable volumes or weights are those that are currently used for plant seeds (i.e. the at least 100 g, at least 1, 5 or 10 kg; but also 25 or more, 40 or more, 50 kg or more, even 100 kg or more, 500 kg or more, 1 t or more, etc.). Suitable containers or packages are those traditionally used in plant seed commercialisation: however, also other containers with more sophisticated storage capabilities (e.g. with microbiologically tight wrappings or with gas- or water-proof containments) can be used. The amount of endophytes (qualita ^¬ tively and quantitatively) contained in the seeds or in the mar ^¬ ketable seed product as a whole can be determined by standard techniques in microbiology readily available to any person skilled in the art of plant endophyte analysis.

The invention is further described by way of the following examples and the drawing figures, yet without being restricted thereto .

Fig. 1 shows cob sheath, grain and cob interior colonization of Burkholderia phytofirmans strain PsJN in maize cvs Peso and Morignon (x-axis shows CFU/g dry weight);

Fig. 2 shows light microscopy images of a mature seed colo ^¬ nized by Burkholderia phytofirmans strain PsJN: :gusA; the blue colour is due to gusA-marked bacterial cells; strain PsJN is present inside the embryo (a,b) and in radicals (c) ; PsJN starts moving from embryo to germinated parts (c) ;

Fig. 3 shows Burkholderia phytofirmans strain PsJN recovery from the grain interior at different time intervals after har ^¬ vesting (DAH; Days after harvesting) ;

Fig. 4 shows the effect of Burkholderia phytofirmans strain PsJN colonized/non-colonized seeds on germination and seedling growth of maize (a,b,c); and

Fig. 5 shows the effect of Burkholderia phytofirmans strain PsJN colonized/non-colonized seeds on shoot growth of maize (a,b,c; 30, 45, 60 days after sowing).

Fig. 6 shows representative results of GUS-staining in pep ^¬ per treated with PsJN: : gusAHO 15 days p.i. GUS activity was found in all fruit parts including seeds.

Fig. 7 shows FISH analyses of pepper seeds colonized by B. phytofirmans PsJN using a general probe targeting eubacteria and a 23S rDNA probe specific for B. phytofirmans . Bacteria other than B. phytofirmans (eubmix-FITC) are indicated with a small arrow and B. phytofirmans PsJN is indicated with a longer arrow.

EXAMPLES

Example 1: Introducing Burkholderia phytofirmans strain PsJN into maize seeds

The concept of internal seed colonization with plant growth promoting microorganisms according to the present invention was tested with the endophytic bacterium Burkholderia phytofirmans stain PsJN and two plant varieties of maize. Strain PsJN was ap ^¬ plied by spraying female flowers with a suspension of 10 ⁸ -10 ⁹ cfu mL ^-1 . At maturity, PsJN cells were detected within maize seeds at viable population densities that ranged from 10 ² -10 ⁵ CFU g ^-1 fresh weight. Strain PsJN was not recovered from plants of the seed inoculation trial. After 12 months of storage 10 ² viable cells per g seeds were still recovered. Experiments were performed to determine the effects of internally colonized maize seeds on offspring plant biomass and vigor as compared to non-treated controls and external application of the same bacterial strain.

Experimental description

The present invention provides seeds having beneficial mi ^¬ croorganisms, especially bacteria, inside, enabling improved plant biomass equally over control as employing the same micro ^¬ organisms (in the present case: bacteria) exogenously to seeds. A variant of the bacterium Burkholderia phytofirmans strain PsJN chromosomally tagged with the beta-glucuronidase gene (gusA, re ^¬ porter gene for detection and monitoring of the strain by color formation) was used as a test strain in to maize cultivars (Peso and Morignon) . For this, series of experiments were performed and the experimental setup was divided into two categories (1 ^st and 2 ^nd year experiments) .

A) Evaluation of strain PsJN colonization potential in different tissues of maize plants (particularly grains) .

B) Follow-up evaluation of strain PsJN colonized seed and strain PsJN inoculation (exogenously) to improve plant productivity over control.

Growth of PsJN strain as bacterial inoculum

The bacterial strain was grown by loop-inoculating one single colony in LB broth amended with spectinomycin (100 yg mL ^-1 ) in 100 mL flasks. The bacterial culture was incubated at 28±2 °C for 2 days at 180 rpm in a shaking incubator. The bacterial inoculum was applied in two different ways i.e. seed soaking and spraying inoculum at flowering stage. Maize seeds were surface sterilized by dipping for 5 and 3 min in 70% ethanol and NaOCl following 3 washings with sterilized water. There were three treatments, 1) seed inoculation 2) specific spraying of flowers and 3) seed inoculation combined with flower inoculation. Plants grown from seeds treated with sterile culture broth only served as control. For inoculation, seeds of two maize cultivars were dipped for 3-4 hours in bacterial inoculum ( 10 ⁸ -10 ⁹ cfu mL ^-1 ) . Likewise, bacterial inoculum was specifically sprayed to the fe ^¬ male flower when the crop reached flowering stage. Seeds were sown in plastic trays filled with soil and 12 days old seedlings were transferred into 50 kg soil container (2 plants in each container) under wirehouse conditions.

Endophytic colonization by PsJN strain (particularly grain colonization)

The rhizosphere and endophytic colonization of root, stem and leaves by the gusA-labeled variant of B. phytofirmans strains PsJN was determined by plate counting using LB plates amended with 5-Bromo-4-chloro-3-indolyl β-D-glucuronide (X-glcA, 50 yg mL ^-1 ) , IPTG (50 yg mL ^-1 ) and the antibiotic spectinomycine (100 yg mL ^-1 ) . Root, stem and leaf samples were washed, surface sterilized (as described above) and used for PsJN recovery (col ^¬ onization) . For this, samples were crushed in 0.9% saline buff ^¬ er, subjected to oscillation in a pulsifier for 30 sec and dilution series were spread on agar plates. Beta-glucuronidase posi ^¬ tive cells appear blue on media containing X-glcA. The blue col ^¬ onies were counted after 3 days of incubation at 30°C and the original cell number per g plant tissue was calculated. Simi ^¬ larly, PsJN colonization was also observed from different cob parts i.e. sheath, grains and cob interior. The identity of the blue colonies was further confirmed by RFLP analysis of the 16S- 23S rRNA intergenic spacer region.

Follow-up experiments were performed in the 2 ^nd year to evaluate the

1. Viability, activation and colonization ability of strain PsJN colonizing maize seeds.

2. Effect of strain PsJN colonized seed on germination and seedling vigor compared to untreated control (plastic tray assay) .

3. Effect of strain PsJN colonized seed on plant biomass com ^¬ pared to untreated control (pot trials) .

Prior to the plant experiments, PsJN colonized seeds of both cultivars were tested to see whether PsJN cells are present and still alive inside. For this purpose, 20 seeds were imbibed in saline buffer for 2-3 days and subsequently crushed in 0.9% sa ^¬ line buffer, shaken for 45 second with a pulsifier and spread in dilutions on LB plates amended with X-glcA, IPTG and spectinomy- cin .

Bacterial inoculum was prepared as described above and three experiments were performed with four treatments i.e. control, seed inoculation with strain PsJN (exogenously) , PsJN colonized seeds (produced in 1 ^st year by spraying) , PsJN colonized seed + inoculation .

For testing the germination performance, seeds (45) were surface sterilized and inoculated as described earlier, and were sown in plastic tray (diameter 30 cm) with three replicates. Da ^¬ ta regarding time to start germination, mean germination time, time to 50% and final germination, germination index and energy, coefficient of uniform germination, and skewness were recorded of PsJN colonized over control.

Two pot experiments were performed to evaluate the perfor ^¬ mance of PsJN colonized seeds concerning plant biomass produc ^¬ tion as compared to control. Surface sterilized seeds were di ^¬ rectly sown in pots with soil (first pot trial) or alternatively sown in plastic trays, and after 10 days seedlings were trans ^¬ ferred to 5 kg pots (2 ^nd pot trial) . All plants were harvested after 60 days and data of plant height, number of leaves per plant and root-shoot biomass were recorded. The data were sub ^¬ jected to analyses of variance using SPSS software package ver ^¬ sion 19 (SPSS Ink, Chicago, IL) .

Results

Experiment A (1 ^st year) : Seed colonization by strain PsJN

The ability of strain PsJN to colonize maize cobs (cob sheath, cob interior and grains) was analyzed in plants treated by specific flower inoculation (by spraying) only or by seed inoculation (Figure 1) . Only inoculation of flowers resulted in internal colonization of seeds. Internal seed colonization by strain PsJN was observed in both cultivars and both flower inoc ^¬ ulation treatments. PsJN cells were detected in maize seeds at viable population densities that ranged from 10 ² -10 ⁵ CFU g ^-1 fresh weight .

Experiment Bl (2 ^nd year) : Viability, activation and colonization ability of strain PsJN colonizing maize seeds.

PsJN colonized seeds, recovered from the first year experi ^¬ ment were tested to see whether PsJN cells survive inside of dormant seed and have the ability to colonize the plants emerg ^¬ ing from the seeds what is very important as seeds may be stored for several months till planting. 10 ² viable cells were detected in two months old dormant seeds (Figure 1) . Imbibing in saline buffer for 2-3 days activated the 6 months old seeds and togeth ^¬ er with the seeds beginning to germinate PsJN started to prolif ^¬ erate resulting in a recovery of 10 ⁴ viable cells (Figure 4) . Sprouts the emerged of 420 day old seeds were colonized by 10 ⁵ PsJN cells and the bacteria was found all over inside the sprouts (Figure 1 and 2) .

Experiment B2 (2 ^nd year) : Effect of PsJN colonized seeds on ger ^¬ mination and seedling vigor as compared to untreated control

The data summarized in table 1 and 4 revealed that PsJN col ^¬ onized seeds showed significant improved germination ability. PsJN colonized seeds of both cultivars started to germinate 36- 48 hours early than the control. PsJN colonized seed showed al ^¬ most 100% final germination rate and required less mean germina ^¬ tion time as compared to the control seeds. Consequently, the colonized seeds have better germination index as compared to control .

Moreover PsJN colonized seeds of both cultivars showed sig ^¬ nificantly higher maize seedling biomass as compared to untreat- ed control seeds (Tables 2 and 5; Figures 3 and 4) but non- significantly higher seedling biomass as compared to seeds ex- ogenously inoculated with PsJN.

Experiment B3 (2 ^nd year) : Effect of PsJN colonized seed on plant biomass compared to untreated control (pot trials)

The data of the pot trials (Table 3 and 6) revealed that PsJN colonized maize seeds had a positive effect on plant bio ^¬ mass production comparable to seeds externally coated with PsJN cells with cv Morignon being more responsive than cv Peso in both treatments (Tables 3 and 6; Figure 5) . The PsJN colonized seeds showed 38% increase in plant biomass production and a sig ^¬ nificant increase in root biomass as compared to the control. Moreover, the number of leaves per plant was higher in plants of PsJN colonized seed as compared to the control.

Conclusions

• Burkholderia phytofirmans PsJN can be introduced into maize seeds by spraying cells onto flowers.

• Seed inoculation only does not allow colonization of maize seeds of the next generation.

• Strain PsJN can survive inside maize seeds for at least 12 months

• Seed-colonizing bacterial cells are rapidly activated, pro ^¬ liferate and colonize emerging sprouts

• Seed-colonizing PsJN shows substantial plant growth promo ^¬ tion

The present example therefore shows that the method accord ^¬ ing to the present invention enables an effective and reliable way to generate seeds with endophytes in a controlled and repro ^¬ ducible manner.

Example 2: Introducing B. phytofirmans PsJN and Enterobacter sp. FD17 into wheat and barley seeds

Experimental description

Seeds of wheat (Triticum spp . cvs Collada and Monsun) and barley (Hordeum vulgare L. cvs Victoriana and Totum) were sur ^¬ face sterilized by dipping for 5 and 3 min in 70% ethanol and NaOCl following 3 washings with sterilized water. Seeds were sown in plastic trays and 12 days old seedlings were transferred into 20 kg soil containers and grown under green-house condi ^¬ tions. The soil has been collected from an agricultural field in Tulln, Lower Austria, and sieved to remove plant material. Bac ^¬ terial strains ( gusA-labelled varients of B. phytofirmans PsJN and Enterobacter sp . FD17) were grown by loop inoculation in LB broth amended with spectinomycin (100 yg mL ^-1 ) in 100 mL Erlen- meyer flask. Bacterial cultures were incubated at 28±2°C for 2 days at 180 rpm in a shaking incubator. Bacterial inoculum was applied by spraying exclusively flowers. Control plants were treated with sterilized broth.

Endophytic colonization of wheat and barley seeds

Plants were harvested at ripening stage and seeds were col ^¬ lected. Seed colonization by the inoculant stains was determined by GUS-staining . Therefore, seeds were cut in two pieces and in ^¬ cubated in GUS-staining solution (1 mM EDTA, 5 mM potassium fer- ricyanide, 5 mM potassium ferrocyanide, 100 mM sodium phosphate, pH 7.0, 1 % Triton-X-100, 0.1 mg/mL X-Gluc predissolved in 5 yL/mg N, N-dimethylformamide, 0.1 % IPTG) directly after harvest ^¬ ing at 37°C for 20 hours. Afterwards, samples were rinsed with 70 % ethanol. The ethanol was then discarded and samples were fixed in paraformaldehyde solution (4 % paraformaldehyde dis ^¬ solved in PBS at 60°C with constant stirring until clarifying of the solution) overnight at 4°C. Finally, the fixed samples were rinsed 3 times in PBS and stored in the last rinse at 4°C until further processing. In parallel, seeds were manually crushed un ^¬ der sterile conditions and used for bacterial community DNA iso ^¬ lation employing standard procedures. The presence of the inocu ^¬ lant strains was confirmed by sequence analysis of the 16S-23S rRNA intergenic spacer region (IGS) of single clones and subse ^¬ quent comparison with those from the inoculants strains.

Results

Experiment A (1 ^st year) : Both seeds of wheat and barley were found to be internally colo ^¬ nized by the inoculants strains. Sequence analysis of the IGS- region confirmed the presence of Enterohacter sp . FD17 and B. phytofirmans PsJN.

Conclusions

• Burkholderia phytofirmans PsJN and Enterohacter sp . FD17 can be introduced into barley and wheat seeds by spraying cells onto flowers.

Example 3: Introducing B. phytofirmans PsJN into tomato and pepper seeds

Experimental description

The colonization behavior of Burkholderia phytofirmans PsJN during transmission from flowers to seeds was studied with toma ^¬ to (Solanum lycopersicum cv. Micro Tom and Matina) and pepper (Capsicum annuum cv. Feher) . Presence of PsJN was investigated at 3 different time points. Detection of bacteria in the seed interior of harvested samples was conducted by GUS-staining and microscopy on the one hand and strain-specific quantitative PCR on the other hand. For detection by visual observation of staining and microscopy, the gusA-labelled variant of the strain PsJN, Burkholderia phytofirmans PsJN : : gusAl 10 , was used in par ^¬ allel to the wild-strain that was detected via qPCR.

The ability of PsJN to survive in the seed and proliferate with the emerging seedling was studied in a subsequent germina ^¬ tion experiment. Hereby, the harvested seeds from the previously treated plants were sown and nursed for a certain period. After ^¬ wards the seedlings were examined regarding their presence of PsJN by GUS-staining and quantitative PCR of PsJN-specific genes .

The bacterial strains were grown by loop-inoculating one single colony in LB broth containing 0.1 % of the antibiotic spectinomycin in case of B. phytofirmans PsJN : : gusAl 10 and with ^¬ out antibiotics in case of the wild-type strain and incubated at 28°C on a shaker (160 rpm) overnight. The overnight culture was transferred to 500 mL Erlenmeyer flasks containing 250 mL liquid LB medium. They were incubated on a shaker (120 rpm) at 28°C for 2 days to allow for growth of bacteria. Subsequently, aliquots of 40 mL of the incubated medium containing the bacterial cul ^¬ ture were filled in 50 mL plastic tubes and centrifuged at 4500 rpm and 4°C for 10 minutes (Megafuge 40R, Heraeus, Hanau, Germa ^¬ ny) . Afterwards, the supernatant was discarded and the bacterial pellet re-suspended by vortexing in 20 mL PBS (0.2 g/L KC1, 1.44 g/L Na ₂ HP0 ₄ and 0.24 g /L KH ₂ P0 ₄ , in d¾0, pH 7.4, autoclaved) . The control suspension was treated accordingly. The aliquots of each bacterial suspension were then pooled in 500 mL Schott bot ^¬ tles. The concentration of the suspensions was measured by help of spectrophotometry (NanoDrop 1000 3.7.1., Wilmington, DE, USA) and adjusted to 10 ⁸ CFU/mL.

Specific inoculation of tomato and pepper flowers was con ^¬ ducted when the plants reached growth stage 61 - 63 on the BBCH scale (for tomato: first inflorescence: first flower open third inflorescence: first flower open; for pepper: first flower open - third flower open) (FELLER et al . , 1995b).

The bacterial inoculants and buffer only for the control were filled in a 50 mL glass pump spray bottle previously steri ^¬ lized with 70 % ethanol. The plants to be inoculated were spa ^¬ tially separated from the others to avoid contamination by drift. One single flower or 2 to 3 immediately adjacent flowers were sprayed with 675 μL of the inoculum. A filter paper was used to shield the surrounding plant parts such as leaves and stem from drift and take up surplus inoculum to avoid dripping on the soil. The treated inflorescences/flowers were marked with a twist tie to allow for later identification.

Six replicates of the inoculated plants were analyzed at 3 different developmental stages. Pepper samples were taken 3 days and 15 days after spraying as well as at full ripeness. The plant material (buds, flowers, fertilized flowers, developing fruits, immature fruits, ripe fruits and seeds) was cut with a sterile scalpel and subsequently incubated in GUS-staining solu ^¬ tion (1 mM EDTA, 5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, 100 mM sodium phosphate, pH 7.0, 1 % Triton-X-100, 0.1 mg/mL X-Gluc predissolved in 5 yL/mg N, -dimethylformamide, 0.1 % IPTG) directly after harvesting at 37 °C for 20 hours. Af ^¬ terwards, destaining was done by rinsing the samples with 70 % ethanol. The ethanol was then discarded and the samples fixed in paraformaldehyde solution (4 % paraformaldehyde dissolved in PBS at 60°C with constant stirring until clarifying of the solution) overnight at 4 °C. Finally, the fixed samples were rinsed 3 times in PBS and stored in the last rinse at 4°C until further processing .

Material of plants inoculated with PsJN wild-type and con ^¬ trol samples were immediately after harvest frozen in liquid ni ^¬ trogen and transferred for storage at -80°C. Afterwards, DNA was isolated using standard procedures and used as described above for Example 2.

Results

Experiment A (1 ^st year) :

Upon flower spraying B. phytofirmans PsJN colonized seeds and pericarp of fruits of tomato and pepper (Figure 6) . The col ^¬ onization process was monitored by GUS-staining and microscopy. The cell number of strain B. phytofirmans PsJN during transmis ^¬ sion from flowers into seeds was tested by TaqMan-quantitative PCR using primers and probe targeting a gene encoding glutamine synthetase. The amount of B. phytofirmans PsJN cells applied on one flower was roughly 10 ⁸ and the cell number calculated per mg plant material during the process of colonization dropped from about 3000 cells per flowers to a few dozen cells in seeds. Re ^¬ sults were confirmed by fluorescence in situ hybridization (Fig. 7) .

Conclusions

• Burkholderia phytofirmans PsJN can be introduced into tomato and pepper by spraying cells onto flowers. Table 1. Comparative performance of PsJN colonized seed and PsJN inoculation (exogenously) on germination of maize cv Peso (Data are average of three replicate)

tValues sharing similar letter (s) do not differ significantly P < 0.05, according to Duncan's Multiple Range Test.

iSeeds prepared by spraying PsJN inoculum (10 ⁸ -10 ⁹ cfu mL ^-1 )

÷Parent seed used for first year experiment

§Offspring seed produced from first year experiment

Table 2. Comparative difference of PsJN inoculated and PsJN colonized seed on biomass of maize cv Peso in plastic tray experiment (data are average of three replicate) .

tValues sharing similar letter (s) do not differ significantly P < 0.05, according to Duncan's Multiple Range Test.

ISeeds prepared by spraying PsJN inoculum (10 ⁸ -10 ⁹ cfu mL ^-1 )

Table 3. Comparative performance of PsJN colonized seed and PsJN inoculation (exogenously) on plant biomass of maize cv Peso under potted conditions (data are average of three replicate) .

tValues sharing similar letter (s) do not differ significantly P < 0.05, according to Duncan's Multiple Range Test.

ISeeds prepared by spraying PsJN inoculum (10 ⁸ -10 ⁹ cfu mL ^-1 )

Table 4. Comparative performance of PsJN colonized seed and PsJN inoculation (exogenously) on germination of maize cv Morignon

(data are average of three replicate) .

tValues sharing similar letter (s) do not differ significantly P < 0.05, according to Duncan's Multiple Range Test.

ISeeds prepared by spraying PsJN inoculum (10 ⁸ -10 ⁹ cfu mL ^-1 ) ÷Parent seed used for first year experiment

§Offspring seed produced from first year experiment

Table 5. Comparative performance of PsJN colonized seed and PsJN inoculation (exogenously) on seedling biomass of maize cv Morignon in plastic tray experiment (data are average of three rep ^¬ licate) .

tValues sharing similar letter (s) do not differ significantly P < 0.05, according to Duncan's Multiple Range Test.

ISeeds prepared by spraying PsJN inoculum (10 ⁸ -10 ⁹ cfu mL ^-1 ) Table 6. Comparative performance of PsJN colonized seed vs PsJN inoculation (exogenously) on plant biomass of maize cv Morignon under potted conditions (data are average of three replicate) .

tValues sharing similar letter (s) do not differ significantly P < 0.05, according to Duncan's Multiple Range Test.

iSeeds prepared by spraying PsJN inoculum (10 ⁸ -10 ⁹ cfu mL ^-1 )