This invention relates to microorganisms and their use in treating animal feed and silage. Background of the Invention

The use of enzymes and organisms can improve or enhance the performance of animals and the value of the feed the animals receive. For example, WO-A-9210945 discloses such a combination for use in enhancing the value of prepared silage, and WO-A-9617525 relates to enhancement of animal performance using microorganisms. The efficacy of combining the use of enzymes together with organisms producing volatile fatty acids (VFA's) is also described. In this case, better preservation of the silage, better animal performance and a reduction in effluent production were demonstrated. O-A-9503396 demonstrates that some advantages may accrue when a desired VFA profile is produced during the silage fermentation; it has also been found that this does not produce the desired reduction in heating, on opening the silage clamp.

The production of silage and the associated crop husbandry have over recent years developed to an extent that a number of different processes can be defined. These are: (i) the ensiling of young grass with particularly low dry matter, e.g. less than 25% (common in UK, Ireland and Scandinavia) , (ii) the ensiling of higher dry matter, more mature grasses (UK) , the ensiling of high dry matter but young grass achieved by wilting (Netherlands) ; and (iii) the ensiling of whole maize including stova and cob, usually at a dry matter concentration of about 35%, and whole crop cereals, e.g. wheat, at 45-50% dry matter.

Particularly in cases (ii) and (iii) , one major problem occurs on a regular basis. This is the phenomenon known as aerobic spoilage. This phenomenon is not well understood. Although there are many differing opinions, the process of aerobic spoilage can be divided into phases.

Thus, there is an initial phase in which yeasts and sometimes acetic acid bacteria start to respire the preserving organic acids. After an initial rise in pH, there is a secondary phase in which the activity of bacilli is apparent, and is associated with increasing temperature. A further phase includes activity of various microorganisms including fungi.

In those silages which contain a substantial content of dry matter, i.e. over 30%, the problem of spoilage is particularly acute. Spoilage is seen to a greater or lesser extent once a silage clamp is opened and exposed to air. Summary of the Invention

This invention is based at least in part on identifying the aerobic spoilage process as being closely related to heating in the clamp on exposure to the ingress of air. Subsequent examination of such silages showed high concentration of thermophilic Gram-positive organisms, yeasts and bacteria including bacilli and fungi. This apparently demonstrates the onset of a secondary fermentation, akin to that of composting (the primary fermentation being the ensiling process) . In this fermentation stage, yeast and moulds predominate. It appears that, in order to prevent spoilage, three main categories of organisms that need to be killed or suppressed are spore-forming bacteria, yeasts and fungi. To eliminate only one category may lead to the proliferation of the remaining categories, so that spoilage is not prevented. According to this invention, utility in the prevention of spoilage has been identified in materials that, at least in the first instance, inhibit microorganisms that initiate aerobic spoilage, notably yeasts and, at the surface of silage, fungi. An organism capable of doing this may also inhibit the development of other spoilage microorganisms, and may be identified by screening.

An organism of the species Lactobacillus buchneri , that meets this requirement has been deposited at the National Collection of Industrial and Marine Bacteria on 13th February 1996. Its accession number is 40788. As explained in more detail below, this organism has a surprising effect, different from and/or extending beyond that due to its ability to produde VFA's, such as acetic, propionic and lactic acids, that are normally produced in fermentation. The organism produces an antimicrobial substance or effect characterised by its ability to inhibit the growth of a variety of spoilage organisms, and which is stable at 80°C but inactivated at 120°C. It is reasonable to assume that any such substance (which may possibly be proteinaceous) may be produced by other organisms. Any one of ordinary skill in the art, provided with the information in this specification, will be able to identify whether any given organism, other than that which has been deposited, produces the same substance.

The substance may be isolated and purified by methods known to those of ordinary skill in the art. As such, it may be used directly to treat animal feed or silage. In other words, it may not be necessary to use a microorganism as such in the method of this invention. Description of the Invention For the purpose of illustrating the invention, an organism has been identified that is capable of producing an approximately normal silage fermentation together with a substance that can inhibit the other organisms thought to be associated with aerobic spoilage. Furthermore, when inoculated on whole crop wheat, maize and grass silage, this organism produces a silage that is well preserved and in which the onset of secondary fermentation associated with aerobic spoilage and heating is reduced or eliminated. The organism thus appears to be able to produce the inhibitory substance under the conditions of fermentation found in silages. A VFA profile as usually obtained in well-fermented silages (with or without the use of

inoculants including Lactobacillus plantarum) has been proven to be insufficient to prevent the occurrence of aerobic spoilage.

It appears that the inhibitory substance may be a secondary metabolite. Therefore, its full effect may not be seen if, when used in silage, that silage is opened too soon. The silage is preferably kept closed for at least 30 days, and more preferably for a longer period, e.g. at least 45 days. The optimum period will depend, e.g. inter alia , on the size of the silage mass, and the nature of the ensiled material.

Materials that are suitable for ensiling, in this invention, are any susceptible to aerobic spoilage. The material will usually contain at least 25% by weight dry matter. Such materials include rye or traditional grass, maize. Lucerne, wilted grass, wheat, barley or other whole crop cereal. The silage may be in bales (a form particularly susceptible to aerobic spoilage) . Alternatively, the invention may be used with any susceptible animal feed, whether solid or liquid, e.g. for pigs, poultry or ruminants.

The activity associated with this invention may be found in other strains of L. buchneri , in other species of Lactobacillus, e.g. L. kefir, L. parakefir and L. parabuchneri , and possibly also in other genera. This can be established by routine experimentation, on the basis of the information herein.

The activity is valuable in inhibiting the growth of various spoilage organisms. Examples of such organisms are Listeria organisms, Bacillus εpp. , Guillermondella selenospora, Trichoderma longibrachiatum, Aspergillus niger, Monascus, Penicillium roquefortii, Fusarium spp. _r and enteric bacteria such as Salmonella.

Selected organisms were grown in liquid culture and the fermentation medium was separated from the cells. This medium was then used in trials to ascertain if there was

present an inhibitory substance that could provide the desired effect, i.e. of heat-stable silage.

Experiments were conducted with three typical organisms that epitomise the spoilage process, i.e. a Bacillus , a yeast and a fungus. Several combinations of the VFA's lactate, acetate, propanoate were also tested. Only the extract from a small number of selected organisms produced the desired inhibitory result. This demonstrated that bacteria were capable of producing one or more substances inhibitory to the target organisms.

In order to demonstrate that such organisms could indeed prevent aerobic spoilage, further experiments were conducted. Organisms were cultured by conventional means and re-inoculated on to grass or whole crop wheat, in c. 10 kg batches, where the organism produced a fermentation not unlike that of a conventional ensiling fermentation, typical for the material ensiled. The nature of this fermentation is not fully understood, but the loss of dry matter (approx. 4%) as gas, presumably C0 ₂ , indicates that this may well be a partially heterofermentative fermentation. Silage prepared in this way proved to have a VFA profile approximating to that of a normal silage. There was no evidence that silages prepared by inoculating with this organism in combination with enzymes would give the desired results. However, silages prepared in this way were taken and placed in large plastic boxes, approximately 0.5 m x 0.25 m by 0.1 m. The boxes were approximately half-filled. Thermocouples were placed in the silages to record the onset of the heating phase of the aerobic spoilage. Each box was placed on an open lab with excellent ventilation.

The results of one such experiment are given in Example 1, below. The results indicate an apparently normal silage fermentation, but in a remarkable demonstration of the effects of the antimicrobial activity of the organism, the treated silage was thermostable, maintaining ambient temperatures (20°C) for a period of at

least ten days. In control and other treatments, the temperature rose to more than 35°C. Experiment 1

Whole crop wheat was treated with a formulation consisting of freeze-dried cells and enzymes. The formulation comprised 4.87% L . buchneri NCIMB 40788, 10.67% enzymes comprising ?-glucanase, xylanase and galactomannanase, 80.09% caster sugar, 3.13% Drimalan green BGE and 1.25% Sipernat 50S. These figures are based on a freeze-dried cell concentration of 3 x 10 /g L . buchneri . This pale green, free-flowing powder formulation was used by dissolving 150 g in 100 1 water, to treat 25 tonnes of forage, by spraying at an application rate of 4 litres per tonne. The viable count was 5 x 10 cfu/g whole crop wheat.

As the control, a conventional silage inoculant containing Pediococcus pentosaceus (NCIMB 12455) and iactoJaciilus plantarum (NCIMB 12422) was used. This is representative of organisms conventionally used in silage treatment, and is described below as the "normal inoculant".

The formulation was enclosed in fermentation barrels with vents to allow the normal fermentation process. The contents were examined after 130 days' analysis, and aerobic stability tests were then carried out.

Fermentation profiles

Treatment DM(%) pH CP(%) ME (MJ/Kg) NH ₅ (%)

Untreated 49. .6 4.1 12 10.6 0.11

Normal Inoculant 50. .3 3.9 10 10.7 0.12

Test organism 51. .0 4.0 10.4 10.1 0.11

Product profiles lactate acetate propionate ethanol

Untreated 41 1 0 6

Normal inoculant 45 5 0 3

Test organism 35 4 0 6

Temp, development (°C)

Treatment (hours) 0 50 100 150 200 250

Untreated 17 17 17 35 34 25

Normal inoculant 17 17 17 25 27 36 Test organism 17 17 17 17 17.5 18

These results, albeit on a small scale, demonstrate that the mechanism is not dependent on VFA profiles or VFA concentration. Experiment 2

Bigbale grass silage was prepared, and treated with a similar formulation as in Example 1, additionally comprising amylase, to provide 1 x 10 cfu/g grass, at an application rate of 1 x 10 g forage. Control silages and silages treated with the L. buchneri strain were examined and analysed after periods of 28, 63, and 113 days of ensiling. The composition of the bigbale silage was as follows:

Day 28 Day 63 Day 113

Dry matter (g/kg) 335 373 361 367 328 337

PH 4.48 4.57 4.42 4.49 4.40 4.47

Lactate (g/ g OM) 54 54 66 59 78 71

Acetate 11 13 15 19 21 27

Propionate 0.7 0.7 0.6 0.7 0.8 1.0

Ethanol 8 8 7 8 9 10

NH ₃ -N (%N) 9 8 11 10 12 12

Aerobic stability

(h) 138 178 187 222 150 >189

The results of Experiment 2 again demonstrate that the differences between the control and the invention are not simply due to the production of a particular VFA profile.

Moreover, there is an indication that the stability of the silage on exposure to air increases with time. This indicates a specific mode of action of the organism L . buchneri and others with similar activities.

From the above results, it would appear that the mechanism of action is the production of at least one antimicrobial substance by the selected micro-organisms during the fermentation process. In order to characterise

the nature of the antimicrobial substance, experiments were undertaken which demonstrated that the substance produced in the supernatant of fermentation broth of the selected organism did inhibit a wide range of micro-organisms. It was stable to heat treatment for 10 min at 80°C, but was inactivated by heat for 10 min at 120°C. The VFA's also present in the fermentation broth appeared to have negligible or no effects on the growth of selected spoilage organisms. Experiment 3

The mechanism of action was further demonstrated by a series of experiments in which culture supernatants were prepared after 24 and 48 hours' incubation in flask cultures. These supernatants were examined on specific organisms known to be active in the aerobic spoilage of fermentations such as those which occur in silage.

In this Experiment, attention was focused on the organisms falling into the yeast and mould categories. The effects of the supernatant were examined on the moulds, Fusarium oxysporum (IMI 236441) , Mucor racemosus (IMI 103730) , Aspergillus niger (IMI 096215) , Caldosporum herbarum (IMI 096220), Penicillum expansum (IMI 315902) and Trichoderma harzanium (IMI 275950) , and the yeasts Candida krusei (ATCC 62403) , Guilliermondella selenospora (ATCC 10629) , Debraryomyces hansenii (ATCC 9364) and Pichia subpellicosa (ATCC 16766) .

In each case, inhibition of the growth of the target organisms was demonstrated. In addition, those supernatants that were prepared from the broth following 48 hours' incubation generally showed the greater inhibitory activity than those incubated for 24 hours. This suggests that the antimicrobial effect is produced at a late stage in the cells' growth cycle, or that it accumulates with time, or that both these possibilities occur. The antimicrobial effect may thus accumulate at an increasing rate, with every generation. It is therefore desirable that, in the field, the fermentation should take place in an optimal

manner and that the fermentation should continue for such a time as shall provide for the maximal accumulation of the antimicrobial effect. The time taken for production of the antimicrobial effect in the field, will vary with the conditions and with the nature of the forage that is being ensiled. Experiment 4

To demonstrate the effect of time on the prevention of aerobic spoilage, grass silage was prepared as in Experiment 1. The aerobic stability was then determined. At 7 days, the stability of the silage was poor with heating occurring in line with the control. At 100 days, stability was assured.