Rumen microorganisms as providers of high quality protein

Ruminants obtain the majority of essential nitrogen from microbial degradation of dietary protein within the rumen and the subsequent increase in microbial biomass. Microbial protein generated by the increase in biomass is the main source of protein nitrogen for the animal. The supply of microbial protein may therefore be adversely affected by various environmental or bacterial factors. Polyphenolic plant products in tropical legumes can complex with dietary or bacterial protein and inhibit microbial growth. Cellulose or lignocellulose present in tropical grasses or other forages may restrict bacterial access to plant protein. The amino acid composition of bacterial protein may not always be adequate to support rapid animal growth and high levels of animal production.

We have recently isolated a phenolic acid-resistant microbial population from the rumen contents of feral goats. When introduced into the rumen of sheep, this population enhances protein supply and yields substantial increases in animal productivity on tannin- containing feeds. We have classified one organism from this mixture as Streptococcus caprinus and show that it can be maintained in sheep at a population of 10⁵ cfu/ml of rumen fluid.

Grazing ruminants receive most of their daily nitrogen requirement from amino acids released into the abomasum and small intestine by the degradation of bacterial protein. Very little plant protein is directly utilised as a nitrogen source but rather is deaminated in the rumen and used to support bacterial growth (Broderick et al 1991). For this reason, protein input into the rumen does not necessarily reflect amino acid absorption by the animal and essential amino acid supply can be a major limiting factor in animal production. Moreover, when feed is high in soluble plant protein, a large proportion of available nitrogen is not utilised by the bacteria, but is converted to urea by the animal and excreted in urine.

Many plants produce compounds that may be toxic, or at least reduce productivity in grazing ruminants. These include oxalic acid, fluoroacetate and a range of different tannin and related phenolic structures. Tannins are toxic, high molecular weight, hydrolysable or condensed polyphenols that exist in a wide variety of plant species. In grazing ruminants, high concentrations of tannins adversely affect nutrition by producing an insoluble proteintannin precipitate that is poorly digestible in the rumen and lower digestive tract, because it inhibits microbial enzymes involved in fibre degradation and produces an astringent taste (Kumar and Singh 1984). The combination of these factors inhibits growth and productivity of animals grazing tannin-rich forage (Clausen et al 1990; Provenza et al 1990). Conversely, low levels of tannins (< 5% dry weight) can benefit ruminants by protecting protein from bacterial deamination (Driedger and Hatfield 1972) and preventing bloat (Jones et al 1976).

Mulga (Acacia aneura), the most common tannin-containing plant in Australia, covers large expanses of South Central, Central and Northern Australia and has often been used as fodder for sheep. However, despite its high leaf protein content, Mulga is nutritionally poor due to the tannin content (11-14% dry weight). In contrast to domestic livestock, goats and other feral ruminants in arid areas of Australia thrive on Acacia. Microbial activity in the rumen of these animals may contribute to tannin resistance.

To determine whether feral ruminants that browse tannin-rich shrubs contain tanninresistant bacterial populations, rumen contents from feral goats were examined for the presence of bacteria that can grow in the presence of condensed tannins and can degrade a tannic acid-protein complex on a nutrient agar plate. Intra-ruminal inoculation experiments have also been carried out to determine whether crude mixtures of microorganisms from feral goats, or purified cultures can be transferred to sheep to confer tannin tolerance in these animals.

Tannin-protein clearing (TPC) bacterial strains were isolated from fresh rumen contents of feral goats browsing Acacia aneura. Three isolates, TPC 2.2, TPC 2.3 and TPC 10 have been classified as Streptococcus caprinus and deposited in the Australian Collection of Microorganisms, University of Queensland under ACM numbers 3968, 3969 and 3970 respectively. Modified BHI (Brain Heart Infusion, Oxoid) media contained (g/litre) cysteine, 0.5; glucose, 0.5; cellobiose, 0.5; soluble starch, 0.5; 1% hemin solution. 10ml (Hungate, 1966); 8% Na₂C0₃ solution, 50 ml. S.bovis strains and TPC isolates were maintained on modified BHI nutrient agar and incubated anaerobically at 39 °C (Hungate 1966).

Intraruminal inoculations of S.caprinus 10 were carried out using 200 ml of bacterial culture containing 2 x 10¹⁰ cfu, administered by stomach tube. Samples for analysis were obtained through rumen fistula and diluted in BHI media for determination of bacterial viability and clearing activity on BHI-tannin plates.

Dilutions of crude rumen contents from 9 feral goats were plated on 0.75% w/v tannic acidcontaining BHI agar plates. Several different types of colony were observed but the most abundant was large and mucoid, and produced clearing zones in the turbid tannic acid-agar plate. Several colonies were purified and three were selected for further examination. They were all facultatively anaerobic Gram positive cocci, occurring mainly in short chains. The MPN estimate for the goat ruminal population of this TPC type of bacterium was 2 x10⁵ - 2 x 10⁶ cfu/ ml of rumen fluid.

Table 1: Metabolic characteristics of S.caprinus and S.bovis

Reaction	Growth(1)			Growth(1)
	S.caprinus	S.bovis		S.Caprinus	S.bovis
Substrate			Substrate

(0.5% w/v)			(0.5% w/v)
Glucose	+	+	Fructose	+	+
Starch	+	+	Maltose	+	+
Cellobiose	+	+	Raffinose	+	+
Galactose	+	+	Inulin	+	+/-(2)
Glycerol	-	-	Rhamnose	-	-(2)
Mannitol	+	-	Sorbitol	-	-
Mannose	+	+	Inositol	-	-(2)
Sucrose	+	+	L-arabinose	-	-
Xylose	-	-	Tannic acid	-	-
Trehalose	+	-
Lactose	+	-
Biochemical
reaction(3)
ß-Galactosidase	+	-
ß-n-Acetyl-
glucosaminidase	+	-
Phenylalanine-
arylamidase	-	+

(1) Bacteria were inoculated into 10 ml cultures containing defined media plus 0.5% w/v of the carbon source under test.
(2) Growth data from Breed et al (1957).
(3) API Bacterial Identification Kit 32A.

TPC isolates were tested for their ability to grow on a number of carbon sources (Table 1) and acid and alcohol production were analyzed by GLC. All isolates gave similar results in these tests. L-lactate was the predominant acid produced although low levels of acetic acid and ethanol were also present. Results of metabolic tests carried out using API Bacterial Identification kit 32A were consistent with identification of all TPC isolates as members of the genus Streptococcus. In minimal media containing glucose, TPC isolates could utilise NH₄+, trypticase or casamino acids as a nitrogen source. All TPC isolates produced clear zones in tannic acid-nutrient agar whereas S.bovis did not do so under any culture conditions tested. These results suggest that the TPC isolates are members of the genus Streptococcus and we have named the species S.caprinus.

To determine the relative sensitivity of S.caprinus isolates to hydrolysable and condensed tannins, and whether the S.caprinus could utilise tannin as a sole carbon source, bacterial growth studies in vitro were carried out using tannic acid or Acacia -derived condensed tannin in the growth medium.

All S.caprinus isolates grew in the presence of tannic acid up to at least 2.5% w/v whereas growth of S.bovis was inhibited by concentrations greater than 0.25% w/v. A similar result was obtained when condensed tannin isolated from Acacia was present in the media. Neither S.caprinus isolates nor S.bovis would grow on defined media containing tannic acid or condensed tannin as a sole carbon source.

Lactic acid production was also measured as an end point of fermentation. In the presence of increasing tannic acid, the level of L-lactate detectable in the culture media paralleled the decline in bacterial growth. Complete inhibition of L-lactate production and S.caprinus growth was not observed until the tannic acid concentration was greater than 3% w/v. On bacterial plates, the pH of the clear zone around S.caprinus colonies was less than 5.0 compared with a pH of 6.8 in the uncleared tannic acid-agar and in zones around colonies of S.bovis grown on plates containing 0.5 % tannic acid.

Goats exist in Australia either as feral populations, or as domesticated flocks. To determine whether S.caprinus was widespread in feral and domesticated goat populations, rumen samples from feral and domesticated animals in South Australia, Central and Southern Queensland were examined. S.caprinus was demonstrated in at least 20 different feral goats browsing tannin-containing Acacia. However, domesticated goats showed no evidence of the species. Feral goats that had no previous history of exposure to tannin-rich plants, or had been domesticated on pasture, also failed to exhibit S.caprinus.

Rumen fluid from domestic sheep was examined for the presence of S.caprinus. No isolates were found. However, when 12 sheep were inoculated intraruminally with 200 ml of crude goat rumen fluid naturally containing S.caprinus, the organism established a stable population of 10⁴-10⁵ cfu/ml of rumen fluid provided that the animals were fed a diet of tannin-rich Acacia. After 20 days, the sheep were transferred to pasture. S.caprinus declined but eventually stabilised at 10³-10⁴ cfu/ml and remained at that level for several months. On return of the sheep to an Acacia diet, S.caprinus increased to 10⁵ cfu/ml. A similar result was obtained when a pure culture of S.caprinus strain l0 was administered to 2 sheep which were then maintained on Acacia for 20 days. These results demonstrate the capacity of S.caprinus to survive in the rumen of domestic livestock, in the absence of continual positive selection, and suggest that the species is a natural ruminal inhabitant.

Table 2: Production and metabolism changes for sheep inoculated with feral goat rumen fluid

Measurement	Pre-inoc		Post-inoc		Change	SE
Day	1-10	23-32	36-45	58-67		mean

DMI(g/d)	510a	524a	535a	611b	+17 %	17.1

N balance(g/d)	-0.61a	-0.55a	0.51ab	1.04b	1.59 g/d	0.402

N digestibility
(g/kg)	450a	417b	450a	463a	+11 %	8.8

Different subscripts across the table indicate statistical significance between pre-transfer and post-transfer periods (P<0.05).

To test the effect of intraruminal inoculation of S.caprinus into domestic sheep, samples of crude goat rumen fluid or purified S.caprinus cultures were introduced into sheep fed a diet of tannin-rich Acacia. Animals were tested for a range of ruminal parameters including dry matter intake (DMI), nitrogen balance, nitrogen digestibility and wool growth. Using crude goat rumen fluid, the results showed that inoculations were effective in increasing the ability of sheep to digest Acacia protein (Table 2). This increase was similar in magnitude to that achieved with traditional nitrogen supplements. Inoculated animals were shown to now support S.caprinus at a population of 10⁵ - 10⁶ cfu/ml of rumen fluid.

Table 3: Production and metabolic differences between sheep inoculated with purified bacterial cultures and goat rumen fluid (GRF)

Measurement	Uninoc.	S.bovis	S.capri.	GRF	SE
					mean

LW change (g/d)	-155a	-141a	-109ab	-56	15.9

DMI (g/d)	382	412	482	541	41.1

N balance (g/d)	-0.895	-1.197	-0.441	0.08	0.359

N digestibility
(g/kg)	368	303	410	390	26.2

DMD (g/kg)	341	357	353	351	7.6

Pure cultures of S.caprinus were also tested. On an Acacia diet, S.caprinus was maintained at a population of 10⁴ cfu/ml of rumen fluid for at least 19 days. Growth measurements showed that nitrogen balance in inoculated sheep was enhanced significantly during this period (Table 3).

These results demonstrate the presence of a novel streptococcal species, S.caprinus, in rumen contents of feral goats browsing tannin-rich Acacia. On the basis of metabolic and genetic characterisation, the organism is shown to be significantly different from previously described ruminal streptococci and should be classified as a separate species. Recent experiments (unpublished) indicate that this organism can also be isolated from the rumen of feral camels. There are at least two, and possibly more isolates of this species and the nearest non-ruminal relative appears to be the streptococci isolated from koala faeces and described as S.bovis biotype I. Growth experiments established that S.caprinus cannot utilise tannic acid as a sole carbon source. Other data (unpublished) demonstrate that a zone of clearing on a tannic acid-nutrient agar plate can be achieved by application of a drop of 50 mM L-lactic acid. The TPC effect on tannic acid-nutrient agar plates may therefore be a consequence of bacterial growth rather than utilisation of the tannic acid.

Intraruminal inoculation experiments show that rumen contents from feral goats contain microorganisms that mediate enhanced DMI and nitrogen balance in sheep fed a ration of tannin-rich Acacia. All inoculated animals were shown to have acquired a population of S.caprinus and the population was maintained for as long as the animal was fed the Acacia diet. When the diet was changed, the population of S.caprinus declined, but was reselectable when Acacia was reintroduced. These results suggest that the ability of feral goats to thrive on tannin-rich forage may be due to the protein-sparing effect and enhanced nitrogen balance of a highly adapted ruminal population. S.caprinus is one member of this population and may be used to enhance nitrogen balance in sheep browsing tannincontaining forage.