Title of Invention


Abstract The present invention discloses a biologically pure bacterial culture of Meghasphera. elsdenii useful to prepare feed additives for ruminants. Said culture has a 16S ribosomal RNA sequence displaying at least 97.0% similarity to a 16S ribosomal RNA sequence of the M. elsdenii strain deposited at NCIMB, Aberdeen, Scotland, UK under number NCIMB 41125; and a growth rate of at least 0.7 h-1
This invention relates to biologically pure bacterial culture of Masphaera elsdenii
and its uses. This invention further relates to preparations and methods incorporating such strain. This invention also relates to feedstuffs for ruminants and a preparation and method for the prevention and treatment of lactic acidosis in ruminants.
Lactic acidosis
Lactic acidosis is a digestive disorder in ruminants that may occur when there is a sudden excess intake of readily fermentable carbohydrates, particularly when ruminants are switched from a diet of roughage to a high-energy or energy nch concentrate diet containing a high level of starch. The disorder is characterised by an accumulation of organic acids, especially lactic acid, in the rumen (Dawson & Allison, 1988). Studies have indicated that a gross imbalance between the numbers of lactic acid-producing bacteria and lactic acid-uitlising bacteria, brought on by a sudden increase in the proportion of readily fermentable carbohydrates in the diet is the main cause of the onset of lactic acidosis (Siyter, 1976)
Manipulating the rumen microbial population to prevent lactic acidosis by administenng material containing high numbers of lactate-utilising bacteria has been advocated for decades, but never practiced on a large scale. Manipulations to enhance lactate utilisation within the rumen has been achieved by administering rumen fluid from an already adapted animal (Allison eX a/., 1964; Braun eX al., 1992) and by administering pure or mixed bacterial cultures of lactate-utilisers (1,251,483 Wilker eX al., 1971; 3,857,971 Abdo & Cahilly, 1974; 4,138,498 Das, 1979; 5,380,525 Leedle et al., 1991; Hession & Kung, 1992; Robinson et al., 1992; Wiryawan & Brooker, 1995).
Some of these feed additives containing live bactenal cultures have been patented (1,251,483 Wilker et ai, 1971; 3,857,971 Abdo & Cahilly, 1974; 4,138,498 Das, 1979; 5,380,525 Leedle et al., 1991), but not commercialised extensively or at ail. In three of the patents (1,251,483 Wilker et al., 1971; 3,857,971 Abdo & Cahilly, 1974; 4,138,498 Das, 1979) the cultures were obtained from continuous culture fermenters with an initial inoculum of rumen fluid. However, the donor animals were not necessarily adapted to a high-concentrate diet. There is also no mention of pH tolerance for any of these cultures. In the other patent (5,380,525 Leedle et al., 1991) the cultures were isolated at pH 5.3 either directly or indirectly after enrichment from ruminants adapted to high-concentrate diets.
The incidence of sub-acute and acute acidosis in dairy cattle
Sub-acute rumen acidosis is a common and serious health and production problem in the dairy industry because diary cows are usually fed diets containing high levels of grains. Sub-acute and acute rumen acidosis are simply different degrees of the same problem. Acute rumen acidosis is more severe and physiological functions may be significantly impaired. The affected animal is depressed and usually ataxic, off-feed, with dilated pupils and an elevated heart rate. Diarrhoea will be obvious and the animal may become recumbent and die within 2 to 5 days after the insult (Nordlund, 1995). Acute acidosis is characterised by a dramatic reduction in ruminal pH (^ 5.0), a large increase in lactic acid concentration and a large decrease in protozoa (Nocek, 1997).
Signs of sub-acute rumen acidosis are very different from that of acute acidosis. Modern dairy management systems of group housing or group feeding make it difficult to recognise these symptoms because individual cows with these problems will usually not be noticed within a group. Herds with subacute rumen acidosis will present some or all of the following signs lammitis, intermittent diarrhoea, poor appetite or cyclical feed intake, high herd cull rates for poorly defined health problems, poor body condition in spite of adequate energy intake, abscesses without obvious causes and hemoptysis (coughing of
blood) or epistaxis (bleeding from the nose). Most of these signs are secondary to acidosis and most of them do not appear until weeks or months after the initial acidosis events. Contrary to feedlot cattle, dairy cows are kept for years and the management of acidosis is therefore of importance in increasing profits.
Chronic laminitis is perhaps the most consistent clinical sign of a herd with subacute rumen acidosis. Although the relationship between acidosis and laminitis is not completely understood, the association is widely recognised clinically and demonstrated in research trials (Kelly & Leaver, 1990; Manson & Leaver, 1988; Nocek, 1997). Furthermore, most dairy managers, veterinarians and nutritionists tend to underestimate or perhaps tolerate an abnormal incidence of laminitis and lameness in dairy herds. A survey in Minnesota demonstrated a mean incidence in lameness of 15% with a range of 0 - 33% (Nordlund, Garret & Oetzel, 1995). Studies in Europe have identified lameness as the third most costly health problem in dairy cows after mastitis and reproduction (McDaniel & Wilk, 1989). The management of acidosis is thus clearly of utmost importance.
A major symptom of sub-acute acidosis is decreased feed intake and decreased efficiency of milk production Sub-acute acidosis, because of difficulties in diagnosing the problem, tends to be dismissed as other problems, such as poor management, poor forage quality e
biggest economic sink to many dairy farmers because it is omni-present, particularly in high producing dairy herds.
Because of the high incidence of nutritional and metabolic disturbance amongst
high producing dairy cows, nutritional strategies for improving performance with
cereal based diets focus on the prevention of ruminal dysfunction by controlling
acid production or by stimulating more efficient microbial growth. At present,
feed additives play an important role in this regard (Hutjens, 1999). The use of
yeast culture strains that specifically stimulate the growth of lactic acid utilising
bacteria generates much interest and a recent survey indicated that yeast
cultures are being used in 33% of high producing Wisconsin herds. Results
from various studies suggest that the Yea Sacc strain 84170 appears to be
particulariy well suited for altenng ruminal fermentation and animal production
when used in high lactate silages and feeds high in concentrates (Dawson,
1995). Production results, however, are very inconsistent. In the USA the cost
for yeast culture supplementation is 4 - 6 cents per cow per day (Hutjens,
1999). lonophores, because of their ability to prevent the growth of important
lactic acid producers, can also play a role in managing sub-acute acidosis.
Although the cost is relatively low (1 - 2 US cents/cow/day, Hutjens, 1999)
there seems to be some resistance against the use of lonophores because of a
few recent cases of lonophore toxicity. Furthermore, lonophores have not been
registered in the USA for use in dairy cattle diets.
Experimentally, there have been several bactena that have potential as direct fed microbials (DFM) for ruminants, but have not been commercialised for a number of reasons. For example, Megasphaera elsdenii (ME) is the major lactate-utilising organisms in the rumen of adapted cattle fed high grain diets. | When cattle are shifted from high forage to high concentrate diet, the numbers of ME are often insufficient to prevent lactic acidosis. Kung and Hessian (1995) have shown that the addition of ME B 159 prevented accumulation of lactic acid during a challenge with highly fennentable carbohydrates. Robinson et al (1992) demonstrated that addition of a different strain of ME (407A) prevented lactic acidosis in steers.
Although the costs associated with subclinical ruminal acidosis are difficult to pinpoint, the potential costs to the dairy industry are huge (Hall, 1999). Donovan (1997) conservatively estimated the cost of subclinical acidosis to the US dairy industry at $500 million to $1 billion per year.
Elsden and Lewis (1953) first described a large, strictly anaerobic Gram-negative, fatty acid producing, non-motile coccus isolated from the mmen of sheep. However, the original isolate was lost before it had been characterised phenotypically in detail. An organism resembling the original strain was isolated from the rumen contents of sheep several years later by Elsden and his colleagues (Elsden et al., 1956). The characteristics of this organism did not fit the description of any known species at the time, but in view of the small
number of isolates studied, the authors refrained from assigning the organism to a new species and genus, but referred to it as LC. Gutierrez et al. (1959) encountered a similar organism in the rumens of bloating cattle and concluded that they fell Vvithin the definition of the genus Peptostreptococcus, proposing the creation of a new species P. elsdenii. Subsequently, Rogosa (1971) demonstrated that the LC-type isolates were Gram-negative and therefore should not be included in the genus Peptostreptococcus. He proposed transfer of P. elsdenii to a new genus Megasphaera and the new combination M. elsdenii, with the isolate LC1 of Elsden et al. (1956) as the type strain. M elsdenii is a strict anaerobe found mainly in the rumen of young animals and animals receiving high-concentrate diets in which lactate fermentation is particularly pronounced. The organism has also been isolated on occasion from the faeces of humans (Sugihara et al., 1974) and it ferments lactate to mainly butyrate, propionate, isobutyrate, valerate, CO2, H2 and sometimes trace amounts of caproate (Stewart and Bryant, 1988). Since M. elsdenii is not subject to catabolite repression by glucose or maltose as in Selenomonas, which is also a lactate utiliser occurring in the rumen, its contribution to lactate catabolism is particularly enhanced subsequent to feeding of soluble carbohydrates (Stewart and Bryant, 1988).
USA patent 3,956,482 (Hahn et al 1976) discloses a method of increasing milk production in ruminants including the steps of administering to the rumen of a lactating cow acetate producing micro-organisms consisting of a mixture of
0-4% M. elsdenii, 30-42% Streptococcus bovis, 3-10% Lactobacillus acidophilus, 12-20% Bifidobacterium adolescentis, 18-44% Bacteroides ruminicola and 3-12% Butyrivibrio fibrisolvens cultured and adapted to a nutrient medium.
A major disadvantage of the invention disclosed in the above patent is the relatively high percentage (between 30-42%) of Streptococcus bovis, which together with Lactobacillis is the leading cause of lactic acidosis in ruminants. The mixture further contains a relatively low percentage of M. elsdenii (0-4%) and the administration of the mixture would probably aggravate or initiate ruminal iactic acidosis rather than preventing or treating it. The mixture is further exposed to atmosphere so that most of the M. elsdenii perish. A mixture of microorganisms is furthermore much more difficult to control than a pure culture.
USA patent 4,138,498 (Das, 1979) discloses a feed additive for administration to ruminants to prevent or minimise lactic acidosis when ruminants are switched from a diet of roughage to starch, compnsing a bactenal culture of M. elsdenii admixed with an ingestibie animal feed additive M. elsdenii is strictly anaerobic and a disadvantage of the feed additive disclosed in this patent, over and above the disadvantages set out below, is that the M. elsdenii is exposed to atmosphere, leading to a rapid decline in the amount of viable cells available in the additive.

USA patent 5,380,525 (Leedle et al., 1991) discloses a biologically pure culture of M elsdenii NRRL-18624 and its use in the facilitation of the adaptation of ruminants from a roughage or normal pasture to a high-energy starch-rich diet. The culture suffers from the disadvantages set out below.
USA patent 5,529,793 (Garner et al., 1996) discloses a mixture of lactic acid producing bacteria and a lactate utilising bacteria such a M. elsdenii with a dry formulation or an animal feedlot diet for improving the utilisation of feedstuffs by a ruminant. (A disadvantage of this invention is that M. elsdenii is generally strictly anaerobic and the application thereof to dry feedstuffs would result in most of the cells'dying.
The applicants have evaluated the above strains of M. elsdenii and have deducted that they are generally not suitable for commercialisation and large scale preventative treatment of lactic acidosis in ruminants because of the following disadvantages of these strains namely they are not:
highly active and adapted to proliferate in the rumen of animals on high-concentrate diets;
capable of proliferating at relatively low pH values below pH 5 0 and as low as 4 5, charactensed as acute acidosis;
resistant to ionophore antibiotics commonly added to feedlot diets; and capable of preferentially using lactate as a carbon source even in the presence of soluble carbohydrates such as glucose and maltose.
Further disadvantages of these strains are that, generally, they: have a relatively low growth rate, i.e. less than 0.938 h-1 do not have the ability to grow on reducing sugars as well as on lactate; have a relatively low biomass output rate, i.e. less than 0.39 g (l.h)-1; are not ionophore resistant; and
produce predominantly propionate and butyrate and not predominantly acetate.
It is therefore an object of the present invention to provide a novel strain of M. elsdenii and its uses, and preparations and methods incorporating such strain with which the aforesaid disadvantages can be overcome or at least minimised
According to a first aspect of the invention there is provided a biologically pure bactenai culture of M. elsdenii having substantially the same 16S nbosomal RNA sequence as that of the M. elsdenii strain deposited at NCIMB, Aberdeen, Scotland, UK under number NCIMB 41125.
According to a second aspect of the invention there is provided a biologically pure bacteria! culture of the M. elsdenii strain deposited at NCIMB, Aberdeen, Scotland, UK under number NCIMB 41125.
The M. elsdenii strain in accordance with the first and second aspects of the invention is further characterised by its:
ability to utilise lactate very efficiently even in the presence of sugars,
resistance to lonophores;
relatively high growth rate;
capabiiity to produce predominantly acetate; and
capability to proliferate at relatively low pH values below 5.0 and as low
as 4.5.
According to a third aspect of the invention there is provided a composition for facilitating the adaptation of ruminants from a roughage-based diet to a high-energy concentrate-based diet, the composition consisting essentially of the bacterial culture of the first or second aspects of the invention.
According to a fourth aspect of the invention there is provided a method of facilitating the adaptation of ruminants from a roughage-based diet to a high-energy concentrate diet including the step of administering to the rumen of said ruminants an effective amount of a bacterial culture according to the first or
second aspects of the invention
According to a fifth aspect of the invention there is provided a feed-additive for ruminants comprising a carrier and an effective amount of a bactenai culture according to the first or second aspects of the invention.
Preferably the culture is disposed in an anaerobic container.
According to a sixth aspect of the invention there is provided a method for the treatment of ruminal lactic acidosis and prevention of any one or more of the following, namely ruminal lactic acidosis, rumenitis, ruminal lactic addosis induced laminitis, ruminal lactic acidosis induced bloat and liver abscesses, including the step of anaerobically administering to the rumen of a ruminant an effective amount of a bactenai culture according to the first or second aspects of the invention.
According to a seventh aspect of the invention there is provided a veterinary agent for the treatment of ruminal lactic acidosis and prevention of any one or more of the following, namely ruminal lactic acidosis, rumenitis, ruminal lactic acidosis induced laminitis, ruminal lactic acidosis induced bloat and liver abscesses, comprising an effective amount of a bactenal culture according to the first or second aspects of the invention
According to an eighth aspect of the invention there is provided a preparation for the treatment of ruminal lactic acidosis and prevention of any one or more of the following, namely mminal lactic acidosis, rumenitis, ruminal lactic acidosis induced iaminitis, ruminal lactic acidosis induced bloat and liver abscesses in ruminants composing:
an inoculum’s of a bacterial culture according to the first or second
aspects of the invention; and
a separate anaerobic growth medium, the components of the preparation being disposed in separate chambers of an anaerobic container which are anaerobically connectable to each other, thus to inoculate the growth medium with the culture anaerobically.
According to another aspect of the invention there is provided a method of achieving any one or more of the following improvements in ruminants namely:
increased milk production;
improved feedlot performance;
improved growth rate;
decrease in finishing time;
lower digestive morbidity and mortality;
lower incidence of lactic acidosis and related diseases,
improved feed conversion efficiency; and
capability to feed on relatively higher concentrate diets,

including the step of administering to the njmen of a ruminant an effective amount of a bactenal culture according to the first or second aspects of the invention.
Preferably the culture is administered anaerobically.
According to yet another aspect of the invention there is provided a method of isolating a biologically pure culture of a superior mminal microorganism in a relatively shorter time period than conventional methods, the method including the steps of:
obtaining a sample of ruminal fluids; and cultivating the sample on a pre-seiected growth medium, the method being characterised in that a plurality of parameters selected from the group comprising growth medium constituents, dilution rate, pH, temperature anti-microbial agents, gaseous environment, redox potential, lack of nutrients and challenging organisms, are pre-seiected to favour the superior rumen microorganism to the detnment of inferior rumen microorganisms.
The invention will now be descnbed in more detail below with reference to the below examples and the enclosed drawings wherein
Figure 1 is a graph of growth rates of lactate utilisers at various pH values;
Figure 2 is a graph of the growth rates (h-1) of the lactate utilising isolates,
on glucose medium at various pH values; and
Figure 3 is the phylogenetic tree of M elsdenii according to the present
In accordance with the present invention, organisms capable of utilising lactic acid were isolated directly from ruminants adapted to a high-concentrate diet. The objective was to select those cultures, with the best combination of characteristics for the purpose of application as mass-cultured, preserved inocula for prophylactic and/or therapeutic treatment of lactic acidosis.
For the lactate-utilising bacteria to be effective they must be highly active and adapted to multiplication in the rumen of animals on high-concentrate diets. The organisms should be able to multiply at pH values below pH 5.0. The selected isolates should also be resistant to lonophore antibiotics commonly added to feediot diets. Lactate should be preferentially used as a carbon source even in the presence of soluble carbohydrates such as glucose and maltose, which would be present in high proportions in high-concentrate diets.
1. Animals used during isolations
Samples of rumen contents from animals with a pre-selection for lactate utilising bacteria were chosen, namely lactatirg fistulated dairy cows at the Dairy Cow Nutrition unit, Irene, of the Agricultural Research Council (South
Afnca), as well as feedlot cattle of Chalmar Beef (Pretona, South Afnca) which were slaughtered at the end of their finishing periods. All the animals were adapted to high-concentrate diets, which increased the numbers of naturally occurring lactate utilising bactena.
2. Sample collection and preparation
Samples of rumen contents were collected from dairy cows at about 09h00, after the cows had been fed and milked. Samples of rumen contents from feedlot animals were obtained 15-30 minutes after the animals had been slaughtered. Plastic screw-cap sample bottles were filled to capacity with rumen fluid filtered through two layers of cheesecloth. The rumen fluid was transferred directly into the fermenter.
3. pH-auxostat
A New Brunswick Scientific Bioflo 1 continuous culture system was modified into a pH-auxostat by converting the pH-dosing pump to a medium addition pump. The pH was monitored with a Schott S23158 pH-electrode connected to a Digital Data Systems 302 pH-meter and titrator. A poorly buffered medium was added whenever the pH increased over the set value until the desired value was reached. The working volume of the culture vessel was 270 mi. The maximum dilution rate obtained for a given organism dunng auxostat cultivation
is a measure of the maximum growth rate of that organism during that condition.
4. Isolation of lactate utilising rumen bacteria via the auxostat
4.1 Growth conditions and medium
Filtered rumen fluid was used to fill the fermenter (270 ml) initially and the
titrator activated to add sterile medium (Medium 1) to the culture proportionally
to the increase in pH of the culture. Medium 1 was a semi-defined rumen fluid
free medium consisting of: Na-lactate (70%), 10 g/l; Peptone, 2 g/1; KH2PO4 1
g/l; (NH4)2S04 3 g/l; MgSO4.7H20 0.2 g/l; CaCl2.2H20 0.06 g/l; Vitamins
(Pyridoxolhydrochloride, 4 mg/1; Pyridoxamine, 4 mg/l; Riboflavin, 4 mg/l;
Thiaminiumchloride, 4 mg/l; Nicotinamide, 4 mg/i; Ca-D-pantothenate, 4 mg/l;
4- Aminobenzoic acid, 0.2 mg/l, Biotin, 0.2 mg/l. Folic acid, 0 1 mg/i and
Cyanocobalamin, 0.02 mg/l); NaaS. 9H2O, 0 25 g/l; Cysteine, 0.25 g/t; Antifoam,
0.07 ml/1 and Monensin, 10 mg/l. The Na-lactate and mineral solution were both
added to the reservoir bottle and autoclaved for 60 min. The peptone was
dissolved in 300 ml d.H20 and autoclaved separately in a 1.0 1 Schott bottle
with a bottom outlet fitted with Quick-fit glass connections. The vitamin solution
was filter sterilised beforehand as well as the two reducing agents Following
autoclaving, the reservir bottle was gassed with araerobic gas overnight and
the other constituents added separately after cooling The pH was adjusted to
the desired value with 5N HCI
Continuous cuituring followed until a pure culture was observed microscopically. A sample was taken from the fermenter with a sterile syringe, which was sealed and transferred into the anaerobic cabinet (Forma Scientific model 1024). One droplet of the culture was streaked out in a Petri dish containing Medium 1 solidified with 2% agar. Incubation followed at 39°C overnight and a single colony was transferred with a sterile needle and syringe into a fresh Medium 1 contained in a 30 ml serum bottle. After incubation at 39°C for 24 h the culture was transfen-ed to several slants containing Medium 1 and incubated overnight. These slants were stored above liquid nitrogen for long-term preservation.
4.2 Batch growth rates of isolates in fermenter
The growth rates of the isolates were verified using the batch cultivation technique and monitoring the increase in optica! density over time The natural log of the optical density (OD) was plotted against time and the linear part of the graph was used to determine the slope, which represented the maximum growth rate of the organism. Determination of the batch growth rate was performed in a chemostat culture, which was diluted with a sterile medium until a very dilute culture suspension was obtained and the medium supply cut off in order to start the batch growth. The advantage of using a chemostat culture for this work is that there is no lag phase since the cells are all viable and adapted to the medium
4.3 Analytical techniques
Volatile fatty acids were determined by gas chromatography with a Cade Erba GC4200 gas chromatograph with flame ionisation detector and a Tupelo 1-1825 column (Supeico Inc., Bellefonte PA, USA). Operating conditions were as follows: carrier gas, nitrogen, flame gases, hydrogen and air; column temperature 175°C; Injection port temperature 200°C. A Barspec data system (Barspec Systems Inc., Rehovot, Israel) was used for peak integration. Pivalic acid served as the internal standard. The utilisation of the D- and L-lactate isomers were determined enzymatically (Test combination 1112 821, Boehringer Mannheim GmbH, Mannheim).
5. isolations of bacteria via spread plate method
5.1 Culture media
The incubated rumen fluid lactate (IRFL) medium for the spread plate isolations consisted of 400 ml incubated clarified rumen fluid (Olumeyan et ai, 1986) from lucerne-fed sheep, 371 mi distilled water, 2 g peptone (Merck), 15 g agar, 100 mi 10% (w/v) sodium-D, L-lactate solution, 100 ml 0.04% (w/v) bromocresol purple solution and 25 ml mineral solution containing 40 g/1 KH2PO4; 120 g/!
(NH4)2S04; 8 g/l MgS04.7H20 and 2.4 g/1 CaCl2.2H20 Lactic acid (90%
w/v) was used to adjust the pH to 5.5 before autoclaving at 121°C for 25 minutes. After sterilisation the medium was cooled down in a 50°C water bath
while being gassed with an anaerobic gas mixture. Two millilitres of each of the reducing agents, Na2S.9H20 (12.5% w/v) and cysteine.HCI.H2O (12.5% w/v)
were added asepticaily. As IRFL medium is not completely selective for lactate-utilisers, bromocresol purple was incorporated to facilitate aetection of lactate-utilisers. When lactate is utilised there is a change in ionic balance in the immediate vicinity of the colony causing a pH increase. A rise in pH above 6.3 was indicated as a colour change from yellow to purple in the zone concentric with the culture.
Acid tolerance was determined on IRFL agar plates with initial medium pH values of 4.5, 5.0 and 5.5.
Resistance to lonophores was tested on IRFL agar plates containing 10 ppm of ionophores generally used in high-concentrate diets i.e. monensin (Sigma) and lasalocid (Sigma). Repression of lactate utilisation by soluble sugars was tested on IRFL agar plates supplemented with maltose or glucose at a final concentration of 10 g/1. A positive result i.e. a purple zone concentric with the colony indicated that the rate of base release due to lactate utilisation exceeded that of acid production from the added sugar. The isolates were also screened on IRFL agar medium without lactate, but to which glucose or maltose had been added at a concentration of 10 g/1, to determine utilisation of the two sugars.
Growth rates on maltose and glucose were determined on media similar to SDL medium, but in which the lactate was replaced with either 10 g/l glucose or maltose.
5.2 Spread plate isolations and screening
The samples for the spread plate isolations were diluted (Mackie & Heath, 1979) in an anaerobic cabinet. Spread plates of IRFL medium were prepared
with the 10-4 to 10-6 dilutions and incubated anaerobically at 39°C After 24 hours, well-spaced colonies showing a purple zone were transferred to IRFL liquid medium in 1.5 ml microtubes. The inoculated microtubes, which showed a colour change to purple within 16 hours, were screened for acid tolerance, (onophore resistance, catabolite repression and utilisation of glucose and/or maltose. Screening was done by replica plating (Lederberg & Lederberg, 1952) using a multipoint inoculator to inoculate 20 isolates onto a set of nine agar plates of different compositions described above.
5.3 Growth rate determinations
Growth was measured, in tnplicate, in SDL, SDG or SDM medium as an increase in turbidity at 578 nm Vials were incubated in a water bath at 39°C between readings. Readings were continued until the turbidity reached the limit of a satisfactory relationship with biomass The natural logarithms of optical
density (OD) were plotted against incubation time. The slope of the exponential growth phase that represents the specific growth rate was calcuiated by linear regression with the aid of a spreadsheet software package.
The cultures grown on SDL medium at pH 5.7 were then incubated further for a total of 24 hours after which 9 ml was preserved by the addition of one millilitre 10% (w/v) NaOH for analysis of end-products formed and utilisation of lactate isomers.
5.4 Growth physiological studies of plate isolates
Fermenter description. A continuous culture system was set up with three fermenters with a capacity of about 250 ml each. A single peristaltic pump was used to supply medium at different rates to the three fermenters Temperature of the cultures was maintained at 39°C. The medium and fermenters were gassed with 100% CO2 to maintain an anaerobic environment. The pH of the
cultures was maintained at pH 5 5 by addition of 20% (w/v) orthophosphonc acid as required. The dilution rate was set at 70%, 80% and 90% of maximum growth rate A sample of 80 ml was withdrawn asepticaily from each fermenter at steady state From this sample the dry mass of the cells and the residual lactic acid in the medium was determined. The biomass output rate, a product of the dilution rate and steady-state biomass, was calculated using the actual dilution rate and the dry mass figures. The growth yield coefficient, which is a function of the biomass concentration at steady state over the amount of
substrate utilised, was calculated using lactic acid residue and dry mass figures.
6. Trials with sheep to evaluate the ability of isolate CH4 to prevent ruminal lactic acid accumulation
In the first trial 12 ruminally-cannulated wether sheep (mean live weight ca 40 kg) were randomly divided into a treatment and a Control group, each comprising six animals. All animals were fed roughage ad lib for 21 days. On day 21 they were fasted for 11 hours prior to being offered 1 000 g of maize meal/animal and at the same time being dosed intra-ruminaliy with 300 g of maltose syrup/animal. One hour later all maize not yet consumed by each animal was packed directly into its rumen. Immediately thereafter animals in the treatment group were dosed intra-ruminally with 1 x 1011 cfu of CH4, whilst animals in the control group were similarly dosed with cell-free filtrate of CH4 preparation, i.e. CH4-free. Samples of rumen fluid were taken at two-hourly intervals, up to 12h post dosing, for detennination of rumen lactic acid concentration.
In the second trial another group of 12 ruminally-cannulated wethers, mean live weight 29 kg, with no previous exposure to concentrate feeding, were used.
They had random access to ground Eragrostis teff hay and a protein-mineral lick. Lambs were randomly divided into two groups of six animals each, namely a treatment and a control group. On the first day of the experiment (Day 1) all Iambs received the following diet ad lib: maize, 888; molasses, 69; urea, 17; limestone, 11; dicalcium phosphate, 6; salt, 4; ammonium sulphate, 4; mineral-vitamin premix with monensin, 1 (g/kg DM). On day one of concentrate feeding each animal in the treatment group received a dose of CH4 intra-ruminal at 12:00, i.e. 3h post feeding. Animals in the control group were similarly dosed with water. Rumen samples were obtained at various times on the day prior to the start of concentrate feeding {Day -1) and on the 1st 2nd, 3rd and 7th days of concentrate feeding, for determination of ruminal lactic acid concentrations.
7. Evaluation of isolate CH4 in high producing dairy cows
7.1 Cultivation of lactate utillser for the animal trial
A Braun Biostat B fermenter with a working volume of 10 litres was transformed into a chemostat using a Watson-Mariow 505S dosing pump equipped with a 55 rpm drive for transporting sterile medium from 50 litre stainless steel kegs. The working volume was kept constant by continuously transferring excess of culture above the 10-litre level of the fermenter via a dip tube and a peristaltic pump (Watson-Mariow 505S) to a 50 litre polypropylene Carboy that was cooled in a chest freezer. The delivery rate of this harvest pump was set to
approximately 120% of the medium supply pump. The excess of volume removed from the fermenter consisted of anaerobic gas from the headspace.
A tangential flow filter system (Millipore Pellicon) equipped with a Millipore HVMP 0.45 micrometer (15 ft2) filter and a Millipore Masterfiex Easy-Load peristaltic pump, was used for concentrating the culture.
The medium used was CSL4. The vitamin, reducing agent, mineral and trace element solutions were filter sterilised prior to addition to the medium reservoir. Following autoclaving, the reservoir was gassed with anaerobic gas.
The production approach was a staggered type of production. Two consecutive productions were performed, each producing enough cells for the treatment of one day's group of animals. The number of concentration steps was limited to one per production as each day's production was collected into a 50 litre vessel. The dilution rate of the culture was 0.4 h-1 and the "down time" between batches was 50 minutes. A backup run consisting of 45 litres was started prior to the first day's production, which also served to promote the chemostat culture into steady state.
7.2 Experimental animals
Sixty high producing dairy cows were blocked according to milk production during previous lactation and body weight and thereafter randomly allocated, within each block, to one of the following treatments: 1) Control diet 60%
concentrate; 2) Control diet 60% concentrate + CH4; 3) Control diet 70% concentrate; 4) Control diet 70% concentrate + CH4. Cows were dosed with organism CH4 at calving, 10 days post partum and 20 days post partum.
The following parameters were monitored:
1. dry matter intake daily;
2. milk production daily;
3. milk fat, protein and lactose weekly; and
4. body weight and condition score monthly.
8. Statistical analyses
Data were analysed by analysis of variance for a completely randomised block design using the program Genstat 5. Previous lactation milk productions were used as a covariate and milk production was reported as covariate adjusted values. Contrasts were used to determine the significance of difference among treatments as follows:
□ +CH4 or -CH4 (dosed vs. non-dosed)
□ Control diet 70% concentrate vs control diet 70% concentrate +CH4
□ Control diet 60% concentrate vs control diet 60% concentrate +CH4
Differences were declared significant at P 9. Effects of Megasphaera e/sdenii supplementation on animal health and feedlot performance
A homogenous group of 448 Bonsmara weaner steers (average initial live weight 215 kg) were randomly assigned to eight experimental treatments in a 2 X 2 X 2 factorial design, with factors (1) CH4 addition (yes or no); (2) ionophore addition (yes or no); and (3) roughage level (high or low). The feedlot diets used and the dietary regime followed were as follows:
Ingredient composition of experimental diets, at the end of the adaptation period (day 14 to end): (Table Removed)
Dietary adaptation regime followed from arrival, when animals were given roughage ad lib, until they were on the final feedlot diet:
(Table Removed)
Nutrient composition of the final feedlot diets used {% of as-fed):
(Table Removed)
Animals were kept in small expenmental feedlot pens. There were 7 pens per treatment and 8 animais per pen. Feedlot diets were fed once daily in the morning, at an ad-lib level. All steers were processed upon arrival (standard feedlot procedures) and fed only long roughage for a few days until they were dosed with either CH4 (treatment) or a similar amount of water (controls). During this dosing a 200 ml suspension of CH4 in medium was applied as an once-off oral drench to each treatment animal.
The CH4 culture was prepared by inoculating a 17.5-litre batch of sterile CSL6 medium (starting pH of 5.20) with 1000 ml of fresh inoculum of CH4, pumped directly from feimenter to Carboy container and incubated at 39°C overnight. The p H of the culture was 6.63 after cultivation and remained the same after 48 hrs. Counts were done on the CH4 culture, after incubation. A peristaltic pump was used to transfer the 200 ml dosage per os to the animal in 10 seconds from the 20-litre carboy.
CSL 6 medium as a 20 litre batch for cultivation of CH4.
(Table Removed)
Feed intake was determined for each pen (daily / weekly) and individual animal weights were determined (weekly / bi-weekly). These were used to calculate feed conversion ratio (per pen). Animals were observed daily and any animals showing signs of acidosis (diarrhoea, bloating, depression) were removed and treated before returning them to their respective pens right away.
Statistical analyses
Data were analysed using the program GenStat 5. Animals were blocked by weight group. The effects of CH4, ionophore and roughage level were tested by means of a 2 x 2 x 2 factorial design in an analysis of variance (ANOVA). The data was acceptably normal with homogenous treatment vanances. Treatment means were separated using Fishers' protected t-test least significant difference (LSD) at the 5 % level, provided that the F-probability from the ANOVA was significant at 5 %.
10. Identification of isolates using phylogenetics, based on 16S rRNA gene sequences.
10.1 Bacterial isolates and culture conditions.
M. elsdenii isolates CH4 and CH7, originally isolated from dairy cows (Wiederhold, 1994) were provided by the inventors. The type strain of M elsdenii, ATCC 25940, was obtained from the American Type Culture
Collection. The strains were cultivated in SDL medium as described previously and presumptively identified as Megasphaera e/scfen/7 (Wiederhold, 1994).
10.2 Amplification and sequencing of 16S ribosomal RNA genes.
Genomic DNA was extracted from bacterial cells using standard procedures (Ausubel et al., 1988). The primers used to amplify the 168 rRNA genes were selected from universally conserved regions in all eubacteria (Table 1). PCR was carried out using primers FD1 (covering positions 8 to 26) and R11 (positions 1384-1400). All target positions of pnmers used for amplification and sequencing refer to the E coli numbering system (Brosius et al., 1978). The PCR reaction mixture of 100 µl contained approximately 200 ng of DNA, 1 µM of each pnmer, 200 µM of each nucleotide (dATP, dCTP, dGTP and dTTP), 50mM KCi, 10mM Tris-HCl (pH 8.4) and 2.5mM MgCI and 2.5 U of Taq polymerase (Boehnnger Mannheim, Germany) The mixture was overlaid with liquid paraffin to prevent evaporation. The thermal profile consisted of 30 cycles of denaturation for 1 minute at 94°C, annealing at 45°C for 2 minutes and subsequent extension at 72°C for 3 minutes in a thermal cycler (Hybaid, U.K). Final extension was carried out at 72°C for 6 minutes. The homogeneity of the amplicons was analysed by agarose gel electrophoresis (Sambrook et al., 1989). The PCR product was excised from the gei and purified using the Wizard PCR Preps kit (Promega, U.S A.) as prescnbed by the manufacturer. Direct sequencing of double stranded PCR amplicons and subsequent

separation of sequencing reaction products on polyacrylamide gels were essentially carried out according to the protocol of Dorsch and Stackebrandt (1992). Sequencing primers are listed in Table 1.
Table 1: Primers used to amplify and sequence the 163 rRNA gene. Primer sequences have been published previously (Dorsch and Stackebrandt, 1992; Lane etal., 1985; Stackebrandt and Charfreitag, 1990; Hutson et al., 1993). A combination of these primers covered a total of 1419 nucleotides of the 16S rRNA gene.
(Table Removed)
Ail target positions for primers refer to E. coli numbering systems (Brosius et
a/. ,1978).
10.3 Data analysis.
The IBS rDNA sequences obtained were automatically aligned with sequences obtained from the Ribosomal Database Project (RDP; Maidak et al., 1996) using the alignment program CLUSTALW (Genetics Computer Group, 1991). Sequences in the profile were tnmmed in order to standardise with regard to the size of the sequences of each organism included in the alignment profile. A total of 1388 nucleotide sequence positions were included in the profile. Published sequences of a number of organisms occurring in the mmen were included in the alignment profile (Table 2). Ambiguous sequences in the alignment profile were manually aligned using the Genetics Data Environment (GDE) alignment editor (Smith, 1992). For infem'ng phylogenetic relationships, the program fastDNAmI (Olsen et al., 1994) was used, which is based on the maximum likelihood algorithm (Felsenstein, 1981). A phylogenetic tree was constructed using the program Treetool (GDE). Escherichia coli and
Acinetobacter caicoaceticus served as out groups in the construction of the tree.
Table 2: Organisms included in the alignment profile using the programs CLUSTALW. All the sequences were retneved from the RDP and the Genbank databases. (Table Removed)
RESULTS Auxostat Isolations
isoiation of lactate utilising bacteria in auxostat
Isolation 1: The rumen contents, obtained from cow 8710, filling the culture vessel of the fermenter were immediately exposed to fresh sterile selective medium when the auxostat was triggered by an increase in pH. Initial dilution rates were in the region of 0.53 h-1 for the first 2 hours at pH 5.30. During the foliowing two hours the dilution rate increased to 0.65 h-1 In order to increase the specificity of the isolation the pH was decreased to pH 5.0, which resulted in a decrease in dilution rate to 0.37 h-1 Cultivation was continued for a further 24 hours after which time only two morphological types could be detected in the enrichment culture. The dilution rate decreased slightly as the cultivation time increased after the initial 24-hour period and at the end of the isolation the dilution rate was only 0 33 h-1
A sample of the fermenter contents was streaked onto agar medium, in the anaerobic cabinet, and a single colony containing a pure culture was

transferred to agar slants and preserved over liquid nitrogen and this culture was denoted as isolate CH1.
Isolation 2: During this isolation from rumen contents of cow 8812 a dilution rate of 0.25 h-1 was observed for the first 24 hours and during the subsequent 24-hour period the dilution rate was between 0.34 and 0.41 h-1. After 48 hours of cultivation it was not clear whether a "pure" culture had been obtained and the cultivation proceeded for a further 24 hours. The dilution rate during this period was 0.41 h-1 and a pure culture was isolated from the fermenter via a colony from a Petri dish. This isolate was designated Isolate CH2.
Isolation 3: The dilution rate dunng this isolation penod decreased from, 0.28 to 0.21 h-1 over a 48-hour period. The isolate obtained from the rumen of cow 8708 was designated isolate CHS.
Isolation 4: Initial dilution rates were in the order of 0.38 h-1 but within 4 hours the dilution rate decreased to 0.276 h-1 and at the end of the 48-hour period the dilution was only 0.197 h-1 The isolate obtained during this isolation from the rumen content of cow 8826 was designated CH4
Isolation 5: At the end of the isolation penod a spore former was the dominant organism and the experiment was terminated.
Isolation 6: During this isolation the dilution rate decrease following the same pattern as for the other isolations and the final dilution rate was only 0.116 h-1 The isolate was obtained from rumen contents of feedlot cattle and was designated CH6
Isolation 7: The rumen contents used dunng this isolation were obtained from feedlot cattle. The dilution rates decreased from 0.142 to 0.106 h-1 during the first seven hours of isolation. The isolate obtained was designated CH7.
Medium modification for chemostat studies
A consistent decrease in dilution rates was observed dunng the isolation of the lactate utilisers, which indicated that the formulation of the medium was not optimal During the first 24 hours of isolation 7 the dilution rate decreased from 0.142 to 0.106 h-1 A pulse dose of 5 ml stenle mmen fluid was added directly into the fermenter and after 4 hours the dilution rate peaked at 0.408 h-1 Thereafter the dilution rate slowly decreased to 0.15 h-1 This "pulse and shift" technique demonstrated that the medium was nutritionally deficient
Another "pulse and shift" expenment with 1 ml vitamin solution resulted in a dilution rate peak of only 0.28 h-1 However, a larger vitamin pulse resulted in a dilution rate peak of 0.497 h-1 which was higher than with the rumen contents pulse Lactate utilisation reflected the same results namely, respective D- and
L-lactate isomer utilisation of 22 and 86% without extra vitamins, and with extra vitamins respectively 68 and 91%. Medium 1 listed in the Methods reflect the modified version with higher concentrations of the vitamins. During another "pulse and shift" experiment it was established that yeast extract increased the cell yields of the Isolates.
Growth rates of Megasphaera elsdenii A TCC 25940 and auxostat isolates vs. pH
Growth rates of the bacteria were determined with the pH-auxostat at various pH values between 4.5 and 6.5, using the modified lactate medium. These growth rates were checked against the values obtained during batch cultivation at the specific pH values and the average value was used
Megasphaera elsdenii ATCC 25940, the type strain, showed an increase in growth rate from pH 4.5 up to pH 6.0, followed by a rapid decrease in growth rate at pH 6.5 (Fig. 1). The maximum growth rate achieved by ATCC 25940 was 0.66 h-1 which corresponds to the reported growth rate of 0.6 h-1 by Therion et a/. (1981).
All the isolates outperformed ATCC 25940 as far as maximum growth rate was concerned, especially at pH values of 5.5 and below (Fig. 1) The maximum growth rates of the isolates all peaked at pH 5.5 with respective growth rates, (h-1) of 0.66, 0 93, 0.938 and 0.864 for isolates CH7, CH6, CH4 and CHS Of all
the isolates, CH4 proved to be the most acid tolerant with a growth rate of 0.389 h-1 at pH 4 5, and the organism with the second best acid tolerance, CH6, with a growth rate of only 0.19 h-1 at pH 4.5. A sharp decrease in growth rate between pH 5 5 and 6.0 was observed for three isolates, namely CH6, CH4 and CHS. Isolate CH7 had only a slight variation in growth rate between pH 5.0 and 6.0, which resembles ATCC 25940 between pH 5.5 and 6.5
Growth rates of auxostat isolates on glucose
The growth rates of three isolates were determined at pH 5.0, 5.5 and 6 0, using the fed-batch growth technique (Fig. 2). Growth rates were noticeably lower for all three isolates on glucose compared to lactate. The most promising isolate on lactate, namely CH4 achieved a maximum growth rate of only 0.25 h-1 at pH 5.5 on glucose, compared to 0.938 h-1 on lactate. Isolate CH7 achieved the highest growth rate (0.33 h-1) on glucose at pH 6.0 amongst the isolates.
Conversion of lactate by Isolate CH4
Isolate CH4 was cultivated at three chemostat dilution rates, namely 0 94, 0.83 and 0 75 h-1 on lactate medium. Dunng steady state, samples were taken and analysed for volatile fatty acids (VFA) and the utilisation of lactate determined. Batch cultivation was also conducted and sampies were taken at stationary phase Samples of the sterile medium were also analysed for VFAs and lactate
With an increase in dilution rate the relative production of fatty acids changed, nannely at low dilution rates more butyrate and valerate were produced and less propionate and acetate (Table 3). At the highest dilution rate very small amounts of butyrate were produced, with no valerate and only slightly nnore acetate and propionate. Lactate utilisation decreased as expected, with an increase in dilution rate. During the cultivation D = 0.75 more than 40% of the lactate was converted to VFAs and although lactate utilisation was high, a large proportion of the available energy was wasted. When CH4 was cultivated in batch it produced mainly acetate and propionate. The concentrations of VFAs produced during batch cultivation were much lower than expected and the only explanation would be that CH4 utilises the VFAs when lactate is depleted.
Table 3: Volatile fatty acids produced by isolate CH4 from lactate during chemostat cultivation at various dilution rates and during batch cultivation.
(Table Removed)
Spread plate isolations and screening
More than 800 colonies from nine rumen samples of four dairy cattle and two feedlot cattle were inoculated into iRFL liquid medium in microtubes. Of these 610 produced a colour change to purple in the medium within 16 hours of incubation. Nineteen of the screened isolates were chosen for further characterisation as they met the specifications required.
Four of the selected isolates, AW09; AW10; AW11 and AW12 were capable of growth at an initial pH of 4.5. The other fifteen isolates all grew at an initial pH of 5.0 and there was further selection here, as those cultures showing suitable characteristics, being the fastest growers at pH 5.0, were selected.
All nineteen isolates were resistant to the ionospheres monensin and lasalocid at concentrations of 10 ppm, utilised lactate in the presence of maltose and glucose and were capable of growth on both glucose and maltose. These nineteen isolates were all Gram-negative cocci (±1 8 micrometers) occurring in pairs or chains.
Physiological characterisation of isolates
All the AW isolates used both the D- and L-lactate isomers as both isomers were virtually completely utilised after incubation in SDL medium for 24 hours. Results indicated that the isolates composed a fairly uniform group and therefore only certain isolates were chosen for further charactensation. For five of the AW isolates the growth rates on glucose at pH 5.8 ranged from 0.38 to
1.05 h-1 with a mean of 0.66 h-1 (+/- 0.298). The AW isolates tested for VFA production from DL-iactate were found to produce acetic, propionic, n-butyric and n-valeric acids in the following ratio 2 : 1.5 ; 1 : 1.3. Some of the AW isolates produced trace amounts of methyl butync acid. The maximum biomass output rates obtained for the nine isolates ranged from 0.31 to 0.43 g (l.h)'\ AW15 had the highest biomass output rate and CH4 and AW01 were next in line with 0.39 g (l.h)-1 The yield of cell dry mass per gram of lactic acid utilised, ranged from 0.1 to 0.17 for the nine isolates
Presumptive Identification of isolates
The isolates obtained were presumptively identified and those meeting the morphological typing as strains of Megasphaera elsdenii were used for further characterisation.
Trials with sheep to evaluate the ability of isolate CH4 to prevent ruminal lactic acid accumulation
Results of the first sheep trial are shown in Table 4.
Table 4: Lactic acid concentration in rumen fluid of roughage-fed sheep suddenly challenged with concentrates, and being dosed intra-ruminally with CH4 (treatment) or placebo (control) at the same time. (Table Removed)
Results of the second sheep trial are shown in Table 5.
Table 5: Lactic acid concentration in rumen fluid of roughage-fed sheep suddenly changed onto a concentrate diet (ad lib), and being dosed intra-ruminally with CH4 (treatment) or placebo (control) 3h after first receiving the concentrate diet.
(Table Removed)
In both trials dosing with CH4 led to a marked and significant (P rumen. In companson ruminal lactic acid levels in CH4-treated animals remained more-or-less at pre-substrate-addition levels This clearly suggested that the lactic acid being produced was largely utilised by CH4
Evaluation of isolate CH4 in high producing dairy cows
The most relevant production data is presented in Tables 6 and 7. Data was analysed separately for all cows (15/treatment) and high producers 10/treatment), respectively.
Table 6: Effect of organism CH4 on productivity of lactating dairy cattle from calving to 80 days post partum (all cows). (Table Removed)
Treatment 1. Control diet, 70% concentrate + CH4 Treatment 2. Control diet, 70% concentrate - CH4 Treatment 3. Control diet, 60% concentrate + CH4 Treatment 4. Control diet, 60% concentrate - CH4
Table 7: Effect of organism CH4 on productivity of lactating dairy cattle from calving to 80 days post partum (high producers).
(Table Removed)
Effects of Megasphaera elsdenii supplementation on animal health and feedlot performance
The delivery of live CH4 bacteria per animal averaged 2 x 1011 colony forming units per dose per animal throughout the feedlot trials. The most relevant results are presented in Tables 8 to 11. The period of weeks three to five in the feedlot is normally considered the most critical in terms of dietary adaptation. During weeks one and two the diet still has a higher roughage content, which gradually decreases; Intake starts at a comparatively low level and builds up gradually. It is only from week three onwards that the diet is at Its lowest roughage level (and highest concentrate level) and intakes are rapidly getting higher. By the beginning of week six animals will normally be considered as adapted to the diet. Week three to five is really the critical period of adaptation of animals to the high-concentrate diet.
Table 8. Average daily feed intake (kg as-fed material per animal per day) for CH4-treatment vs control (no CH4 added), for various periods in the feedlot.
(Table Removed)
Overall feed intake was slightly (but not significantly) higher for CH4 than for control. Dunng week 3-5, for steers not receiving ionophore, CH4 had a significantly (P Table 9. Average daily gain (kg per animal per day) for CH4-treatment vs control (no CH4 added), for various periods in the feedlot. (Table Removed)
Overall average daily gain (ADG) during the critical period of week 3-5 was significantly (P=0.04) higher for CH4 than control treatments. During week 1-2 for all animals that did not receive CH4, the low-roughage treatment had a significantly (P Table 10: Feed conversion ratio (kg feed per kg gain) for CH4-treatment vs control (no CH4 added), for various periods in the feedlot (Table Removed)
Overall animals treated with CH4 had a significant (P=0.04) ca 5% improvement in feed conversion ratio (FCR) over control animals during week 3 - 5. During week 1 - 2 for all animals that did not receive CH4, the low-roughage treatment had a significantly (P=0.06) higher (less desireable) FCR than the high-roughage treatment (4.42 vs 3.83). However, for animals that did receive CH4, FCR was not significantly higher (less desireable) on low-roughage as compared to high-roughage diets (4 24 vs 4.25).
Table 11: Number of times (including multiple pulls of the same animal) and number of animals (multiple pulls of the same animal count only as one) that animals were pulled and treated for acidosis and bloat.
(Table Removed)
For the CH4 treatments only half the number of animals suffered from acidosis symptoms (one or more times) as compared to the controls. The same trend was observed if the total numbers of acidosis incidents was considered. It is also clear that acidosis was much more prevalent on the low-roughage diet and that CH4 treatment alleviated the problem on the low-roughage diet.
Identification of isolates using phylogenetics, based on 16S rRNA gene sequences.
Comparative sequencing results showed that our isolates, which are representatives of a larger phenotypically homogenous group, are between 97 and 99% similar (Table 12). Table 13 outlines positions of signature nucleotides suitable to distinguish the two recent Megasphaera elsdenii isolates and the ATCC strains from each other. Insertions and deletions accounted for 22% of nucleotide differences between the four strains. The major nucleotide sequence differences between the strains occur at nucleotide positions 529-536 and 1105-1120 (Table 13). The high sequence similarity displayed betA/een the different M. elsdenii strains is also consistent with their similar phenotypic characteristics (Wiederhoid, 1994), which is furthermore reflected by their tight phylogenetic dustenng. Strains of the species M. elsdenii share
only 91 to 92% sequence similarity with M. cerevisiae, and the two species form distinct clusters in the phylogenetic tree. The M. elsdenii cluster bifurcates into two monophyletic groups that evolved from the same ancestral taxonomic unit (ATU). The ATU from which M. cerevisiae evolved, however, predates the one from which the M elsdenii cluster evolved. The short branch lengths
between M. elsdenii strains (OTU s) and their respective ATU's also indicate that they have evolved more recently than the more deeply branched M cerevisiae.
Table 12: Sequence similarity matrix of 16S rDNA sequences of M. elsdenii and M cerevisiae. Sequence similarity values are based on a comparison of a total of 1388 unambigiously aligned nucleotide positions. The %G+C refer only to the respective aligned 16S rDNA sequences (Table Removed)
Table 13: Sequence signatures defining different M elsdenii isolates and strains (Table Removed)
a E. coll numbering system (Brosius et al,, 1978) bdenotes the type strain. An asterisk indicates where a gap was introduced dunng the alignment as a result of the occurrence of a nucleotide deletion or insertion at any one position of the sequences of the respective isolates and strains
The maximum likelihood method, which involves finding a tree, which gives the
highest probability of giving nse to the observed sequence data (Felsenstein,
1981), was used to infer a phylogenetic tree from the sequences included in the
alignment profile (Fig. 3). This method has the advantage above traditional
parsimony methods, which could lead to inference of erroneous trees if different
lineages evolve at unequal rates, in that it allows for evolutionary rates to differ
between different lineages (Felsenstein, 1981). Since tree topology is also
affected by the number of organisms used and the selection of the outgroups
(Stackebrandt and Ludwig, 1994; Stackebrandt and Rainey, 1995), a number
of apparently related and apparently unrelated organisms, occurring In the
rumen was included in an alignment profile This was subsequently used to
construct the tree. Although the phylogenetic tree was inferred from only nearly
complete (92%) 16S rRNA gene sequences, which could reduce resolution
between very closely related organisms (Utaker et ai, 1995; Li and Graur,
1991), the general topology of trees derived from either complete or partial
sequences has been shown to be in overall agreement with each other (Van
Camp etal, 1993, Vandamme etal, 1996).
Vandamme et al (1996) proposed that different isolates should be regarded members of the same species if they share more than 97% rRNA sequence homology, show phenotypic consistency and exhibit a significant degree of DNA: DNA hybridisation. Although the relationshiD between DNA similarity and
16S rRNA sequence homology between organisms is anything but linear, Fox
et al. (1992) proposed that effective 16S rRNA sequence identity should imply
that two organisms are members of the same species, since it would almost
certainly be validated by the DNA: DNA hybridisation Although 16S rRNA
sequence data alone may not be sufficient in all cases to define a species, it is
extremely useful in determining to which species a strain probably belongs,
once the relevant species is represented in a 16S rRNA sequence data base.
Strains with almost identical 16S rRNA sequences should be assigned to the
same "rRNA superspecies" or "rRNA species complex". It would thus be
appropriate to assign isolates CH4 and CH7, phenotypically presumptively
identified M. elsdenii strains, to the same rRNA species complex, which would
include reference strains ATCC 25940, the type strain of the species, and
ATCC 17752. Since the phylogenetic relationships of the respective isolates
are furthermore consistent with their phenotypic characteristics, these isolates
can be considered strains of the species Megasphaera elsdenii. The fact that
M. cerevisiae and M. elsdenii share only 92% 16S rDNA sequence homology,
confinns, together with genotypic and phenotypic data, the division of the genus
into two well resolved species.
Of the rumen bactena included in this study the ones which appear to be most closely related to the Megaspliaera duster are Selenomonas ruminantiunn and Quinella ovalis. The apparent phylogenetic relationship between Megasphaera elsdenii and Selenomonas ruminantium is consistent with some phenotypic and
genotypic charactenstics which the two species share, such as similar DNA G+C content (53-54%), anaerobic nature, chemoorganotrophic metabolism and utilisation of a similar range of substrates (Stackebrandt et a/., 1985; Stewart and Bryant, 1988; Haikara, 1992). The work of Stackebrandt et a/. (1985). who made use of the oligonucleotide cataloguing technique for phylogenetic inference between the species, supports this phylogenetic relationship. Selenomonas ruminantium on the other hand is also closely related to the relatively unknown Quinella ovalis, an organism that proliferates in the mmen when sugar nch diets are fed to the animal. These organisms, although not established in culture yet, share some physiological characteristics with the large selenomonads found in sheep (Stewart and Bryant, 1988). As expected, the most distant relatives of Megasphaera that occur in the rumen are those contained in the archaeal methanogen cluster, the members of which are believed to have appeared approximately 600 to 800 million years ago (van Soest, 1994; Woese, 1987). The evolutionary rates of these organisms are also slower than that of the Bacteria, and the pnmitiveness of the group is clearly reflected by the deeply branched methanogen cluster.
The recent divergence of the different M. elsdenii strains could possibly be attnbuted to a refinement of its phenotype in order to adapt to the highly selective conditions in the rumen. According to Woese (1987), the evolution of the phenotype of an organism is a process dunng which new or more efficient traits are gained in order to sun/ive in its particular niche Refinement would
result in the organisms being metabolicaily versatile, as is the case with Megasphaera. The slower evolving methanogens on the other hand are metabolicaily monotonous by comparison.
This study has demonstrated the suitability of 16S rDNA sequencing to distinguish between closely related strains of the species M. elsdenii. Furthermore, it provided a phyiogenetic framework for identification of recently isolated strains that have been characterised phenotypically. The framework would be of particular value in serving as a basis for the design of species and strain-specific probes intended for rumen ecological studies.
The incorporation of bromocresol purple to IRFL medium to facilitate detection of lactate-utilismg bacteria proved to be successful in the case of the faster-growing lactate-utilisers, which were of prime interest in this study. In the early stages of incubation these produced purple zones, concentnc • with the colonies, which contrasted clearly with the yellowish background of the agar medium However, on extended incubation the pH gradient surrounding the colonies dissipated due to diffusion of ions and the whole background became purple. Differentiation between colonies of slower-
growing Iactate-utilisers and those of organisms growing on other carbon sources present in the rumen fluid supplement then became difficult.
M. elsdenii is not the dominant lactate-utilising species in animals on high-concentrate diets (Mackie et ai, 1978; Mackie & Gilchrist, 1979; Mackie et al., 1984; van Gyiswyk, 1990), but there are a number of reasons why they may have predominated in the selection and screening procedures.
Some of the colonies screened consisted of Selenomonads and other morphological types. Most of these colonies were not chosen as there was not a positive indication ttiat lactate could be utilised in the presence of soluble sugars. Russell & Baldwin (1978) showed that M. elsdenii B159 used glucose, maltose and lactate, but not sucrose simultaneously in a multi-substrate medium. Marounek et al. (1989) showed that for four strains of M. elsdenii lactate was used more rapidly than glucose in media with both carbon sources.
Certain other laboratories, which have studied the possibility of inoculating ruminants on high-concentrate diets with lactate-utilising organisms to prevent an accumulation of lactate, have also worked with M. elsdenii strains (Das, 1979; Leedle et al., 1991; Robinson et al., 1992; Kung & Hession, 1995, Wiryawan & Brooker, 1995). It was not possible to compare the growth rates of the AW and CH isolates to strains in the literature to determine if they had faster growth rates and if they were more acid tolerant, as no results were
available in literature. The AW and CH isolates can however be compared to the type strain, M elsdenii ATCC 25940.
For the AW isolates the range of pH values at which growth was determined was not sufficient to detemnine the pH range for the isolates or the optimum pH for growth. However, it can be assumed that the optimum would be above pH 5.7 and the lowest pH would lie between pH 4.5 and pH 4.9 for all but four of the isolates, as there was no growth on IRFL plates at pH 4.5. This agrees with work done on the type strain of M. elsdenii ATCC 25940. The pH range for the type strain of M. elsdenii ATCC 25940 is pH 4.6 to 7.8 with the optimum for growth at pH 6.05 {Therion etal., 1982).
For the CH isolates the optimum pH for growth is between pH 5 and 6. In the range of pH values tested, the highest growth rates were found at pH 5.5. Both sets of isolates had higher growth rates on SDL medium than the type strain. The growth rates obtained for M. elsdenii ATCC 25940 in SDL medium is comparable to that obtained by earlier workers on lactate medium (Therion et al., 1982).
The growth rates of the isolates on lactate were higher than on glucose and maltose. This is in agreement with a previous study on M. elsdenii ATCC 25940, where growth rates on lactate medium between pH 5.0 and pH 6.5 was found to be higher than in glucose medium (Therion etal., 1982). Outside this
pH range growth rates on glucose were higher. A study on substrate preference in rumen bacteria reported that growth of M elsdenii B159 on lactate was slower than on glucose and maltose, however the pH of the media In the study was above pH 6.5, being between 6.75 and 6.9 (Russell & Baldwin, 1978).
The composition of femnentation end products on lactate medium has been detennined for four strains of M. elsdenii, including the type strain LC1 or ATCC 25940 (Marounek et ai., 1989). These results showed strain-to-strain variability in the proportions of fatty acids formed. Three of the strains produced little or no valeric acid while 22 mol% of the end-products of M elsdenii L8 was valenc acid (Marounek et ai, 1989). The nine AW isolates tested in the present study did not exhibit as much strain-to-strain variability as was the case in the strains tested by Marounek et al. (1989), but are similar to M. elsdenii L8, which was isolated from the rumen of a calf on a milk diet, as valeric acid was produced. CH4, however, produced the same fermentation end-products as the type strain.
From the point of view of maximum biomass output rate in SDL medium strain AW15 would be the organism of choice for larger scale production of cells for animal experiments with CH4 and AW01 being next in line. The time required to produce 100g dry mass of cells on SDL medium in a chemostat of 51 working volume would be 1.9 days for AW15 and 2.1 days for CH4 and AW01.
The growth rates of the selected isolates on lactate are high compared to the type strain of M. elsdenii. The isolates are acid tolerant and can grow at pH values below 5.0. They are resistant to ionophores, commonly added to feedlot diets, and can utilise both isomers of lactic acid even in the presence of glucose and maltose. The fermentation end products from lactate are VFA, which are an important energy source for the ruminant. Propionate production is especially important in the feedlot industry, as propionate is the main source of glucose for the ruminant tissues. The isolates, therefore, have the characteristics required for an effective product to combat lactic acidosis in ruminants.
Cultivation of the lactate utilisers was successful using a medium that did not contain any rumen fluid. The only modification to the original medium was the increase in the vitamin content and the addition of yeast extract to the medium. Bacteria survived remarkably well on this medium at 4°C for up to 20 days, when used as working cultures.
The technique of using a pH-auxostat for the enrichment of lactate-utilising rumen bacteria, with a predetermined combination of biochemical/physiological attributes, which would make them potentially highly suitable for preventing and combating lactic acidosis in feedlot animals was very successful. In most cases a fast-growing, morphologically homogeneous population became established

in the fermenter within two days after the start of a run. Subsequent tests on the cultures that were isolated from the femnenter contents by plating, confirmed that the cultures possessed the desired combination of charactenstics.
Presumptive identification of the isolates from the enrichments showed that all but one belonged to the species Megasphaera elsdenii.
Table 14: A comparison between the pH-auxostat isolation technique and a conventional spread plate screening technique. (Table Removed)
Since the isolates performed better than ATCC 25940 at pH values below 6.0, they would be more suited to the ruminal pH encountered by feedlot animals
that usually is below pH 6.0. Furthennore, CH4 proved to be the best suited isolate for trial experimentation on feedlot animals.
Cultivation of the lactate utilising isolates on glucose proved not to be a proposition due to the slow growth rates obtained compared to the growth rates obtained on lactate.
The thirty cows that were dosed with organism CH4 produced significantly more milk (P=0.10), had a higher average body weight (P=0.02) and body condition score (P=0.06). The milk fat percentage of cows receiving the 60% concentrate diet + CH4 were also significantly increased (3.57% vs. 3.23%).
Dairy animal trials
Dry matter intakes did not differ but milk production was significantly increased by 3.4 kg/d from 35.9 to 39.3 kg/d (P=0.06) when cows were fed the 70% concentrate diet and were dosed with organism CH4. Milk production tended to be increased (P=0.13) when all dosed cows were compared to all non-dosed cows. Body weight and condition score were increased (P=0.02) when high producing cows receiving the high concentrate diet were dosed with organism CH4. The dosing of cows receiving the 60% concentrate diet resulted in a significant increase in milk fat percentage (P=0 06) with a tendency towards
increased milk protein (P=0.11). iVlilk components play an important role in current milk payment schemes.
Dosing cows with organism CH4 significantly increased milk production and positively affected milk composition, body weight and body condition score. Dry matter intakes were not affected; therefore results suggest that dosing of cows with organism CH4 caused a more favourable rumen environment, which resulted in improved utilisation of nutrients.
Feedlot animal trials
The significant improvement In average daily gain (ADG) and feed conversion ratio (FCR) during the critical period in the feedlot (week 3 - 5), as well as the overall decrease in digestive disturbances, for CH4-treated animals as compared to controls, show that treatment of feedlot animals with CH4 can be effective in the following ways:
• It aided the adaptation from roughage to concentrate diets, as inherent in this experiment. CH4 also alleviated the poorer performance of the iow-roughage diet as compared to the high-roughage diet during the early adaptation stages.
• The application method used here for CH4 was effective in allowing its expression as intended.
• The use of CH4 was effective in preventing acidosis, as evidenced directly by less observed cases of acidosis and indirectly by improved
performance during the critical feedlot phase when acidosis is most likely to occur. Ail this was observed on diets and dietary regimes with an exceptionally high risk of acidosis (especially in the early feedlot stages), and was more pronounced where no ionophore was used.
• CH4 may be used as a veterinary agent to help in the prevention and / or treatment of acidosis, as evidenced by the sharp decline in acidosis cases for CH4 as compared to controls.
• CH4 may be used to improve feedlot performance including growth rate (and by implication to decrease time required for finishing) and feed conversion efficiency.
• CH4 can be used to allow the feeding of higher-concentrate diets, i.e. the use of less roughage, and to increase the rate of change from high roughage to high concentrate diets, i.e. also the use of less roughage. This is furttier supported by the observations, during the eariy feedlot phases, that the negative effect of low-roughage diets, as compared to the high-roughage diets, was largely alleviated when CH4 was added.
The applicants have further found that, in companson with the known strains of
M. elsdenii, the M. elsdenii CH4 strain is:
highly active and adapted to proliferate in the rumen of animals on high-concentrate diets,
capable of proliferating at relatively low pH values below pH 5.0 and as low as 4 5, characterised as acute acidosis,
resistant to ionophore antibiotics commonly added to feedlot diets; and
capable of preferentially using lactate as a carbon source even in the presence of soluble carbohydrates such as glucose and maltose.
Further advantages of this strain are that it:
has a relatively high growth rate, i.e. more than 0.938 h-1;
has the ability to grow on reducing sugars as well as on lactate;
has a relatively high biomass output rate, i.e. more than 0.39 g (l.h)-1;
is ionophore resistant;
produces predominantly acetate and not predominantly propionate and butyrate; and
has a unique 16S rRNA sequence and is therefore a new strain
The applicants have yet further found that animals challenged with maltose, fed directly into the rumen, or abruptly changed from a roughage to a high-concentrate diet, produced no measurable build-up of lactate in the rumen when inoculated with CH4
Furthermore, high producing dairy cows inoculated with CH4 has a 2.4 to 3.2 litres higher production of milk, than control animals not inoculated with CH4. The body condition score as well as body weight of the inoculated cows were statistically significantly higher than the control animals.
It will be appreciated that variations in detail are possible with a microorganism according to the invention and its uses without departing from the scope of the appended claims.

We claim
1. A biologically pure bacterial culture of Meghasphera. elsdenii useful to prepare feed additives for ruminants having a 16S ribosomal RNA sequence displaying at least 97.0% similarity to a 16S ribosomal RNA sequence of the M. elsdenii strain deposited at NCIMB, Aberdeen, Scotland, UK under number NCIMB 41125; and a growth rate of at least 0.7 h-1.
2. A biologically pure bacterial culture of M. elsdenii as claimed in claim 1 having a 16S ribosomal RNA sequence displaying 100% similarity to a 16S ribosomal RNA sequence of the M. elsdenii strain deposited at NCIMB, Aberdeen, Scotland, UK under number NCIMB 41125.
3. A biologically pure bacterial culture of M. elsdenii as claimed in claim 1 or claim 2, which utilises lactate at a rate of between 40% and 90% even in the presence of sugars.
4. A biologically pure bacterial culture of M. elsdenii as claimed in any one of claims 1 to
3, which is resistant to ionophores.
5. A biologically pure bacterial culture of M. elsdenii as claimed in any one of claims 1 to
4, which is capableof producing predominantly acetate.
6. A biologically pure bacterial culture of M. elsdenii as claimed in any one of claims 1 to
5, which is s capable proliferation at pH values as low as 4.5.
7. A veterinary treatment and/or prevention agent comprising a bacterial culture as
claimed in any one of claims 1 to 6.





157-delnp-2005-complete specification (as, files).pdf

157-delnp-2005-complete specification (granted).pdf




157-DELNP-2005-Description (Complete).pdf



















Patent Number 235743
Indian Patent Application Number 157/DELNP/2005
PG Journal Number 35/2009
Publication Date 28-Aug-2009
Grant Date 20-Aug-2009
Date of Filing 17-Jan-2005
# Inventor's Name Inventor's Address
PCT International Classification Number A61K 35/74
PCT International Application Number PCT/ZA2003/000093
PCT International Filing date 2003-07-15
PCT Conventions:
# PCT Application Number Date of Convention Priority Country
1 02/5742 2002-07-18 South Africa