BATCH-CULTURES OF RUMEN MICROORGANISMS IN VITRO

Kopecny J., M. Mar~unek, S. Bartos: Batch-cultures of Rumen Microorganisms in Vitro. Acta vet. Brno, 50, 1981: 157 -178. The modified method of batch-culture of rumen microorganisms in vitro was developed. When checking this method we have found: The optimal ratio of urea and glucose in incubation medium was within the range 0.1-0.3 g N-ureaJg glucose. By adding of 0.003 % amorphous FeS in the inoculated medium more negative redox potential was always observed in the range close to in vivo conditions. The optimal microbial protein synthesis was achieved at an osmotic pressure 0.48 MPa. It follows from the results that osmolarity of buffer routine used is low. Enhanced utilisation of starch was obviously observed in the presence of maltose. By decreasing the glucose concentration to 2 gil buffer at the beginning of the incubation the lactate production was prevented. The isonitrogenic replacement of 25 % urea by amino acids increased the production of protein as well as VFA and YATP. Succinic acid, lactic acid and hydrogen were not present in significant concentrations which correspons to in vivo conditions. It follows from the results obtained that the developed batch-culture represents a suitable in vitro system for short-term rumen fermentation studies. Goats, sheep, fermentation, incubation, medium. Rumen microorganisms are able to convert plant forages to nutritionally valuable products. The cellulose degradation is the most important process. Besides this, microorganisms also synthesize most other saccharolytic enzymes. They further synthesize all essential amino acids for the host in such an amount that under optimal conditions the host is independent of their contents in the food. Besides this they produce sufficient amounts of vitamins of groups B, Hand K. From published results and theoretical studies it follows that the synthetic capacity of rumen microorganisms is utilized only partially. Thus, the research of their metabolism and possibilities of its modification are in the focus of interest in the animal production at the present time. The rumen fermentation after feeding may be divided depending on the decomposition of saccharides into two parts. At the first stage mono-, disaccharides and alpha-polymers of saccharides are prevalently hydrolyzed and metabolized. The length of this stage is usually of 5 to 10 hrs, depending on the fed substrate. The second stage comes after the first one, and is characterized by a decomposition of beta-polymers, i. e. particularly cellulose. This stage continues up to exhaustion of the substrate or up to next feeding. The two stages affect each other, however, they may be studied separately. Different approaches are used to studying the two stages. For the study of the decomposition of alpha-polymers of saccharides it is possible to use a short-term incubation in vitro. The metabolism of cellulolytic bacteria is more properly studied with the help of the continuous-culture. Advantages of the batch-culture are particularly simplicity, possibility of analyzing large amounts of samples and easy statistical evaluation of results (Lopper et al. 1966; Barr 1974; Maeng et al. 1976; Chen et al. 1976; Van Nevel and Demeyer 1977 and others). Since the end of the fifth decade, nobody studie4 in detail the method of batch-cultures or results were not published in spite of the fact that in the meantime many new factors affecting the rumen fermentation revealed. Thus, some authors employed improper conditions of the incubation and particularly

By adding of 0.003 % amorphous FeS in the inoculated medium more negative redox potential was always observed in the range close to in vivo conditions.The optimal microbial protein synthesis was achieved at an osmotic pressure 0.48 MPa.It follows from the results that osmolarity of buffer routine used is low.Enhanced utilisation of starch was obviously observed in the presence of maltose.By decreasing the glucose concentration to 2 gil buffer at the beginning of the incubation the lactate production was prevented.
The isonitrogenic replacement of 25 % urea by amino acids increased the production of protein as well as VFA and YATP.Succinic acid, lactic acid and hydrogen were not present in significant concentrations which correspons to in vivo conditions.It follows from the results obtained that the developed batch-culture represents a suitable in vitro system for short-term rumen fermentation studies.

Goats, sheep, fermentation, incubation, medium.
Rumen microorganisms are able to convert plant forages to nutritionally valuable products.The cellulose degradation is the most important process.Besides this, microorganisms also synthesize most other saccharolytic enzymes.They further synthesize all essential amino acids for the host -in such an amount that under optimal conditions the host is independent of their contents in the food.Besides this they produce sufficient amounts of vitamins of groups B, Hand K.
From published results and theoretical studies it follows that the synthetic capacity of rumen microorganisms is utilized only partially.Thus, the research of their metabolism and possibilities of its modification are in the focus of interest in the animal production at the present time.
The rumen fermentation after feeding may be divided depending on the decomposition of saccharides into two parts.At the first stage mono-, disaccharides and alpha-polymers of saccharides are prevalently hydrolyzed and metabolized.The length of this stage is usually of 5 to 10 hrs, depending on the fed substrate.The second stage comes after the first one, and is characterized by a decomposition of beta-polymers, i. e. particularly cellulose.This stage continues up to exhaustion of the substrate or up to next feeding.The two stages affect each other, however, they may be studied separately.Different approaches are used to studying the two stages.For the study of the decomposition of alpha-polymers of saccharides it is possible to use a short-term incubation in vitro.The metabolism of cellulolytic bacteria is more properly studied with the help of the continuous-culture.
Advantages of the batch-culture are particularly simplicity, possibility of analyzing large amounts of samples and easy statistical evaluation of results (Lopper et al. 1966;Barr 1974;Maeng et al. 1976;Chen et al. 1976;Van Nevel and Demeyer 1977 and others).Since the end of the fifth decade, nobody studie4 in detail the method of batch-cultures or results were not published in spite of the fact that in the meantime many new factors affecting the rumen fermentation revealed.Thus, some authors employed improper conditions of the incubation and particularly ratios of substrates, which could make the results of the experiment doubtful.When starting with the batch-culture in our laboratory, remarkable spread of results was often encountered and the results were sometimes mutually opposing.Thus, we decided to pay attention to a methodic treatment of the batch-culture of rumen microorganisms.In initial experiments we considered a combination of methods summarized in Table 1, where works by Maeng et al. (1976) and Barr (1974) served as a basis for this purpose.These authors were partially interested in the incubation method.A derived method was marked as "method I" (Marounek et al. 1979).When checking this method we used a number of modifications.They led to a modified "method II".Particular modifications and course of the fermentation by the modified method are discussed in the following chapters.

Animals and diets
In the experiments the rumen liquor was used of goats (1 to 2 years, 30 to 45 kg) and sheep (the same age, 50 to 65 kg).The animals were fed once daily by a constant diet consisting of 1 to 2 kg of hay, 0.5 kg of concentrate and 10 g of urea per animal and day.The animals received water ad libitum.All animals were provided with permanent rumen fistulas.

Sampling of the rumen liquor
The sampling was carried out before feeding, since the microorganisms, except for certain cellulolytic bacteria, are at the stationary stage of the growth that time.However, the cellulolytic bacteria are not remarkably manifested during a short-term incubation in vitro.The rumen liquor was sampled for the experiment from at least two animals by a tube, with the help of a vacuum pump (Doldel et al. 1969).
. Batch-culture method I 100-ml infusion flasks with Bunsen valves in stoppers were used for the incubation.Into each flask 0.1 g cellulose (cotton, ground paper Whatman 2) and 1 g soluble starch for microbiology (Lachema n. p.) or 1 g dried potato flakes were weighed.Further 30 ml buffer I were added, which contained 10 g glucose and 5 g urea per liter unless otherwise indicated.Then the infusion flasks were conditioned to 40 DC.After adding 15 ml of the rumen liquid the flasks were stoppered and placed in the RT-60 water bath (2 cycles/s) at 40°C.Into the infusion flasks a mixture of gases CO. and H. was introduced (20: 1).Residual amounts of oxygen from technical CO. were removed by leading the mixture through palladium catalyst.The atmosphere of infusion flasks was washed at a flow rate of 1 ml CO./s for 30 min to achieve a residual O. concentration below 0.1 ' Yo.Gases produced in the course of the incubation were released through Bunsen valves.The test incubation took 6 hrs.
Mter completing the incubation, -pH and rH were measured (Maluszyn'ska 1973) and the vitality of protozoa was observed.For the determination of proteins (by precipitation with 6 % trichloroacetic acid and kjehldalisation), ammonia (colorimetrically with Nessler agent), urea (Biotest Lachema Bmo), or of the incorporation of radioisotopes 10 ml of medium was preserved with 5 ml of 20 % trichloroacetic acid (TCA).The remaining portion was inactivated by adding HgCII and used for determining volatile fatty acids (Bartos et a1. 1960) and lactic acid (Marounek 1978).Gas was analyzed on a gas analyzer with a column filled with Porapak Q -(3.5 m long and 3 mm in diameter).One infusion flask served for testing one effect in one time interval.For each determination at least four parallel infusion ft8sks were used.Succinic acid was determined by the one-dimensional paper chromatography.Papers Whatman No.1 were used.Chromatograms were developed in the mixture of ethanol-ammonia-water.After air-drying at room temperature chromatograms were sprayed.with bromthymol blue solution.

Buffers
Buffer I.

Results
Testing and adjustments of batch-culture I 1.The effect of oxidation-reduction potential of the rumen liquor on the fermentation in vitro During the isolation of rumen microorganisms, where it is not provided that the microorganisms will be able to reduce the medium, a wide spectrum of reducing.agents is used, as e. g. cysteine (0.05 %), sodium sulphide (0.025), thiosulphate (0.003 %), thioglycolate (0.05), dithiothreitol (0.001 %) and their combinations.Amorphous ferrous sulphide was also recommended for the cultivation of anaerobic microorganisms (Brock and O'Dea 1977). .
In the course of preliminary incubations there was a remarkable spread of the redox potential (Eh) in individual infusion flasks under the same conditions of the cultivation.To reduce this spread, we tested the effect of amorphous ferrous sulphide on the course of Eh • After conditioning.amorphous ferrous sulphide was added into infusion flasks to achieve the final concentration in the inoculated medium of 0.003 %, and the redox potential (Eh) was compared with control samples.For results see Fig. 1.After the inoculation an enhanced redox potential was always observed -in a range of 100 to 200 mV (against hydrogen electrode).In one hour it is reduced by action of microorganisms as much as by 350 mV.At this stage of the fermentation no effect of ferrous sulphide was found.Its effect was obvious only in subsequent intervals.In samples without addition of FeS the redox potential increased after the second hour.By adding FeS a further increase of the redox potential was achieved in 12 hrs.The spread of Eh in measured samples was also reduced.2. The effect of the osmotic pressure of incubation medium on the fermentation of rumen microorganisms in vitro For testing purposes buffer I with the following concentrations was used: diluted 1: 1 (0.136 MPajl), non-diluted (0.269 MPajl), concentrated 1.5 x (0.404 MPajl), 2 x (0.53S MPajl) and 3 x (O.SOSMPajl).
For a comparison the effect was also established of McDougall artificial saliva (0.336 MPajl).The osmotic pressure of the rumen liquor varies between 0.56 and 0.62 MPaj1.It follows from the results obtained (Figs. 2 to 4) that the effect of the buffer osmolarity on the fermentation parameters of interest may be demonstrated.The differences between minimal and maximal values achieve 25 to 40 %.
The difference in pH was of 1.5 grades after completing the incubation, the increase being non-linear, since besides the buffering capacity the pH was also affected by the production of volatile fatty acids (Fig. 2).
The production of volatile' fatty acids was maximalum at an osmotic pressure of 0.48 MPa/1 (buffer II), which also holds against the McDougall artificial! saliva (P< 0.05) (Fig. 3).
. .The protein synthesis was •also most remarkable at higher osmotic pressure 0.48 PMa.There was a significant difference between buffers I and II (P < 0.05).
To provide optimal conditions for the growth of bacteria and protozoa, only buffer II was used in further incubations.
Under in vivo conditions the ascertained optimal osmotic pressure of the rumen liquor occurs in animals before feeding.Mter feeding the osmotic pressure increases and it is likely that, similarly as under in vitro conditions, it leads to a decreased production of volatile fatty acids and proteins.These assumptions were checked indirectly by applying inorganic salts into the food (Barry et a1. 1977;Thomson 1978).These authors found, that long-term additions.ofKCl into food decrease the production of volatile fatty acids and exerts no effect on the ratio of particular volatile fatty acids, i. e. C 2 -C S (Barry et a1. 1978).Thomson et al. (1978) found a negative correlation between the production of propionate and diluting rate after the application of 5 to 10 % salt into the diet.
In continuous-culture of rumen liquor a drop of the DNA synthesis was found after the KCIO a application.These nucleic acids correlate with the content of microbial proteins.Simultaneously the production of volatile fatty acids, CO2 and CUt was reduced.The redox potential of the medium was increased.The authors attributed this effect to a toxic effect of KCIO a (Barry et a1. 1978).
To achieve optimal conditions for the growth of rumen microorganisms, in the modified method only buffer II was used (osmotic pressure = 0.4 MPa/I).

Optimal ratio of urea and glucose during the incubation in vitro
An optimal course of pH during the incubation can be achieved by suitable buffering capacity of buffer or suitable ratio of N-compounds and saccharides.Since the buffer concentration cannot be varied within wide concentration ranges, it is necessary to search for an optimal ratio of saccharides and nitrogen -containing compounds.The fermentation products of nutrients (particularly ammonia and VF A) have an opposite effect on pH.With respect to the fact that there is a matter of a short-term incubation the largest portion of volatile fatty acids is produced by a decomposition of glucose.Starch participates only partially and celluloses serve for maintaining vital cellulolytic bacteria in medium.The optimal ratio N: glucose was established in an experiment, where at a constant urea concentration (5 gj1) the glucose level was adjusted to 0,5, 10,33 and 66 g glucosej1 buffer (Fig. 5).It follows from the graph that optimal results were achieved at glucose concentrations of 10 to 25 gj1 buffer, i.e. ratio N: glucose 0.1 to 0.3 Since for in vivo conditions there is a typical drop of pH in 6 hrs, for further work the ratio N : glucose of 0.15 was chosen.So, the optimal ratio of urea and glucose was established to 12 g N-ureajlOO g glucose.This is in agreement with the used ratio of glucose and urea in certain in vitro methods (Van Nevel and Demeyer 1977;Barr, 1974;Chen et al. 1976).However, these authors did not use in cultures any other saccharides, so that the final ratio Njsaccharides was high in their experiments and did not cor- respond to requirements of rumen microorganisms for their optimal growth.Unfortunately, these authors do not report the final pH during the incubations, so that comparison with our experiments is impossible.Under in vitro conditions the optimal ratio Njsaccharides was determined by Zaki EI Din and EI Shazly (1969) -2.35 g Nj100 g starch.In vivo the optimal value varies about 1.6 to 2.25 g Nj100g DOM (Rofler and Satter 1975;Satter and Slyter 1974).In the case of our method the optimal ratio was established to 1.92 g Nj100 g saccharides (glucose, starch and cellulose) which corresponds to an optimal ratio of substrates found in vivo.

Effect of the formulation of saccharides on rumen microorganisms in vitro
Starch is a basic substrate during short-term incubation of the rumen liquor.Experiences of authors (Durand et al. 1976;Minato and Suto 1976) however, indicate that there are rather large differences between different types of starch.Thus, we compared soluble starch, potato starch and corn starch in two experiments, where these starches served as the only one source of saccharides.The finding of volatile fatty acids at the end of the incubation was the highest in the case of the soluble starch, next to it there is the corn starch and the potato starch was manifested as the worst.(Expressed relatively: 100-92.4-82.4%).Thus for the experiments we recommend soluble starch and, by contrast to this, we do not recommend potato starch.Besides this we searched for a suitable mono-or disaccharide in a mixture with starch, through the decomposition of which the rumen microorganisms could gain energy from the very beginning of the incubation.
In this experimental series the urea concentration was reduced to 0.5 gj1 and starch was added in an amount of 0.5 gjinfusion flask.For concentrations of the other saccharides see Tables 2 and 9.The 32p concentration was of 2.44 KBqjml medium.
In the experiments no significant differences in the production of volatile fatty acids and pH values were ascertained, however, the samples containing maltose exerted a higher production of volatile fatty acids and reduction of pH connected with this.(Table 2).The protein production, as determined by the 32p incorporation, was also higher after ;ldding maltose.The effect of the other saccharides on the 32p incorporation was insignificant.In a mixed sample of cellobiose and maltose the effect of disaccharides was not additive.A drop was found of the 32p incorporation as well as of the volatile fatty acid production.
Most authors use glucose as a basic saccharide for in vitro cultures (Maeng etal. 1976;Chen et a1. 1976;Owens and Issacson 1977 and others).However, there was an objection (Van Nevel and Demeyer 1977) that an addition of glucose can result in an increased production of lactic acid on account of volatile fatty acids as observed by Maeng and Baldwin (1976).In the elaborated method the increased production of lactate (20 to 50 mmoljl) was often found in experiments, where the initial glucose concentration was higher than 3 gjl buffer and always when pH of medium dropped below pH = 5.2 (Marounek 1978(Marounek , 1980)).By decreasing the glucose concentration to 2gjl buffer, at the beginning of the incubation the lactate accumulation was prevented and pH was not decreased below 5.5.
In further experiments the effect of xylane (ICN Pharmaceuticals, Inc.), glucose and equimolar mixture cellobiose + maltose was observed.As compared to starch, after adding other saccharide an enhanced production of volatile fatty • :P < 0.05 acids was observed.There were no significant differences in the protein production, however, an enhanced proteosynthesis was observed in samples with additions ofxylane and mixture cellobiose + maltose (Table 3).
Cellobiose with maltose are main products of the hydrolysis of cellulose and starch.Their effect was followed in vitro in the absence of further saccharides according to the following schematic diagram: For results see Figs. 6 and 7.
The mixture of both disaccharides does not affect the volatile fatty acids production (Fig. 8) . .By contrast to this, the protein synthesis is affected (Fig. 9).

_ I,S
The results of experiments testing the effect of disaccharides and starch on the growth of rumen microorganisms are in agreement with those by Barr (1974), who demonstrated a relative possibility of replacing usual monosaccharides when affecting the proteosynthesis.Enhanced proteosynthesis in the presence of maltose obviously resulted probably from a high selection of amylolytic bacteria during the incubation.An addition of xylane makes a growth of a wider spectrum of bacteria possible, thus favourably affecting the utilisation of available substrates (Williams et al. 1969;Belasco 1956).
.. The last experiments were supposed to elucidate a mutual effect of cellobiose and maltose on the growth of microorganisms.Besides glucose, this is a matter of main intermediate products of the polysaccharide decomposition.The ascertained effect of an addition of cellobiose to maltose in medium corresponds to results of a study by Russel et al. (1979).These authors, when using five pure cultures of bacteria, also ascertained an inhibition effect of cellobiose in the presence of maltose on the bacteria growth.Henderickx and Martin (1963) by contrast to this report found that the proteosynthesis of rumen microorganisms is higher in the presence of cellobiose than that in the presence of maltose.In all probability, the effect of cellobiose is one of control mechanisms between amylolytic and cellulolytic bacteria in the rumen.
By addition of cellobiose to maltose the proteosynthesis of microorganisms was reduced, however, the production of volatile fatty acids was unaffected.From this it results that the utilisation of A TP is different when combining both substrates;  It can be seen from the table that through the decomposition of cellobiose the microorganisms gain less A TP than through the decomposition of maltose.In a mixture of both disaccharides, the inhibition effect of cellobiose on the maltose utilisation was manifested again.It would be suitable to check the utilisation of A TP under these conditions on pure cultures of bacteria and to establish in which bacteria the Y ATP is affected.
With respect to the results mentioned it was decided to use glucose in in vitro cultures up to the final elucidation of the interaction of cellobiose and maltose on the fermentation.

Urea concentration in the course of the in vitro incubation
The original concentration of urea in medium was high and ammonia released by the hydrolysis achieved concentrations which are not typical under in vivo conditions.Thus the in vivo fermentation was followed at an urea concentration of 0.5 gil buffer.At this concentration an equilibrium is established in the incubated medium (Fig. 10) where the released ammonia is proportional to the consumption for the synthesis of microbial proteins, i. e. 100 to 300 mg N-NH2/1 and its concentration varies within physiological ranges.Simultaneously with the adjustment of the initial urea concentration the glucose concentration was reduced in such a way that the optimal ratio N/saccharides was adhered to.By this decrease a possibility was almost precluded of increasing the lactate production.The inhibitory effect of high ammonia concentration was also prevented, which is indicated by Owens and Isaacson (1977).The reduced concentration of nitrogen sources is lower than that used by the other authors (Table 7), however, it yielded good results in cultures.

Choice of nitrogen source for the fermentation
The fermentation activity of in vitro system expressed by the volatile fatty acid production is affected not only by the amount and type of supplied saccharides, but also by the source of N. In a further experiment we compared as sources of nitrogen urea (35 mg N/45 ml) and isonitrogenic amount of an ammonium salt (NH 4 CI), casein hydrolysate, yeast autolysate and combination of urea with yeast autolysate (17.5 mg N-urea + + 17.5 mg N-yeast autolysate/45 ml).The amount of saccharides (starch, cellulose, glucose in usual concentrations) was identical in all cases.Regarding the volatile fatty acid production as well as protein production in the incubation medium ammonium chloride is a least suitable source of nitrogen (52.2 ± 4.2 mmol VFA/l).Urea is insignificantly better (56.9 ± ± 2.1).Amino acids in the form of casein hydrolysate and yeast autolysate stimulate the fermentation (61.4 ± 6.5 and 64.5 ± 8.1 mmol VFA/l, respecti vely).
The differences as against NH 4 CI are significant.The combination of urea and yeast autolysate (67.5 ± 3.5 mmol/l) was shown to be the best.The proteosynthesis follows a similar trend.A reason for this is obviously in a dependence of certain bacteria on amino acids -peptides, demonstrated in a number of papers.
An addition of urea to the source of amino acids and peptides is, however, advantageous from the standpoint of a rapid delivery of ammonia into the culture medium.It removes the dependence of bacteria requiring ammonia as a source of nitrogen on bacteria deaminating amino acids.

The effect of peptides and amino acids on the fermentation in vitro
In short-term cultures of the rumen liquor, the amylolytic bacteria, protozoa and bacteria utilizing intermediate metabolic products contribute essentially to the fermentation.They are dependent on the amount of amino acids in medium more than other microorganisms.Thus, the isonitrogenic replacement of 25 % urea by casein hydrolysate or potato flakes which served as a natural source of plant proteins was tested.The casein hydrolysate is advantageous in that it is not precipitated with 6.5 % TCA and that it does not affect the determination of the proteosynthesis.
One liter of buffer contained 0.5 g urea and 0.36 g casein hydrolysate CImuna n. p. Sarisske Michafany) and control flasks contained buffer with 0.65 g urea/I.The glucose level was the same in both cases -2 gil buffer.In isotope experiments the 32p activity was of 0.925 MBq/45 ml medium, Ci. e. infusion flask).The same amounts of starch and potato flakes were weighed -1 g/45 ml medium.The potato flakes in the infusion flask contained 30 mg amino acids.
For results of the experiment see Figs. 11-14.An addition of amino acids as well replacement of starch by potato flakes increases the production of proteinas well as incorporation of1 4 C-acetate into lipids and incorporation of 32P-phosphate into microorganism biomass.The increase of proteosynthesis determined by the kjeldahlisation and incorporation of 14C-acetate was significant.The spread of results during the determination 0.7 of the proteosynthesis with the ... o.~ help of the 32P-phosphate incor-g .,.
• 0. acid concentration are useless.The effects of the amino acid addition on the pH and volatile fatty acid production are summarized in Table 4.Samples with amino acids yield a higher production of volatile fatty acids ...  source of nitrogen.These extreme conditions do not occur in in vivo processes.Even animals on a synthetic diet are passaging mucins of saliva into the rumen and also the desquamation of rumen epithelium serves as a source of proteins for rumen microbes.It was demonstrated in our experiments, that a 25 % replacement of urea by amino acids or proteins is sufficient for providing an optimal growth of microorganisms as Maeng and Baldwin(1976).However, these authors found that an addition of amino acids had no effect on the volatile fatty acid production.This may be explained by that in their experiments there was an enhanced production oflactate, which sometimes exceeded the production of the other volatile fatty acids (Maengland Baldwin 1976c).In our experiments the production oflactate was low (3-8 mmol/l/6 hrs) and amino acids exerted an effect on the production of volatile fatty acids.The addition of amino acids increased significantly the fermentation rate, which is in agreement with the found nutritional requirements of amylolytic bacteria (Bryant and Robinson 1961, 1962, 1963).The utilisation of A TP for the synthesis was also increased by adding amino acids.As far as urea served as a single source of nitrogen, Y ATP was of 11.46 g b. s./mol ATP.The addition of amino acids resulted in an increase to 15.75 g b. s./mol ATP, i. e. by 37 %.A similar result was obtained by Maeng and Baldwin (1976).In the absence of starch in the medium amino acids increased Y ATP by 33 %.
It follows from the results presented, that in short-term incubations, where a maximal growth of microorganisms is required, it is necessary to supply a part of the nitrogen source in the form of amino acids.

The effect of light on the in vitro fermentation
This effect was tested in one experiment.One half of incubation flasks were painted black to prevent the admitance of light to the incubated rumen liquor.The concentrations of glucose, 1 kg hay 0.5 kg concentrate 10 g urea urea and casein hydrolysate were of 2, 0.5, and 0.36 g/l buffer, respectively.
Due to a small spread the differences were significant (P < 0.05).Changes of pH and production of volatile fatty acids and protein were followed.In a 6-hr incubation the effect of light was manifested as an inhibition of the production of protein as well as fatty volatile acid: The light exerts bactericidal or bacteriostatic effects on most heterotrophic bacteria, partiCularly the pathogenic ones (Hampl 1964).In our experiments it was demonstrated that it also acted on rumen microorganisms.The ascertained inhibition effect on the fermentation was small (about 5 %) however for establishing optimal conditions it is more suitable to eliminate the effect of the light in the course of the incubation.

Modified Method of the Incubation of Rumen Microorganisms in Vitro
The modified method was checked by a 24-hrs incubation of the rumen liquor.Sampling as well as inoculation of the rumen liquor were performed similarly as in the original method.Into an infusion flask 0.5 g of soluble starch (Lachema Bmo) and 0.1 g cellulose (cotton or ground paper Whatman No.2) were weighed.Then 30 ml buffer II were added and the flasks were heated for 15 min to 40 DC.The glucose concentration was of 2 gil buffer, urea concentration 0.5 gil buffer and casein hydrolysate concentration 0.36 gil buffer.Mter heating amorphous FeS was added, whose final concentration in medium was of 0.003 %.
The incubation time was 24 hrs.
For the course of pH see Fig. 15.For a comparison a course is presented of the pH of rumen liquid of rams, from which the inoculum was obtained (Fig. 16).The drop of pH in initial hours under in vivo conditions is more rapid than that during the incubation.In both cases the course is in the same range of pH.The decrease of pH after the 12th hr of the incubation under in vitro conditions results from . the volatile fatty acid accumulation.These acids are resorbed in the rumen.The vitality of protozoa was followed microscopically during the in vitro incubation.It is indicated in the graphs.A decrease of the protozoa vitality occurs only in the second half of the incubations.In 6 hrs, i. e. time of the incubation, the vitality of protozoa remained unaltered.
The course of the oxidationreduction potential of the incubation medium can be seen in Fig. 17.From the initial value of 180mV it decreases in one hr to -74mV, up to the second hr it drops to -160 mV and it remains constant up to 12th hr.Then there is a moderate increase of Eh to -100 mV in 24 hrs.The measured values were calculated with respect to rH units (Fig. 18).Even here the course was similar.The final drop of r H after 24 hrs of incubation results from a decrease of pH.
The volatile fatty acid production may be controlled by an amount of added saccharides.Under mentioned conditions their production is shown in a graph (Fig. 19).In first 10 hrs the volatile fatty acids originate from the decomposition of glucose (at most 17 mmol/l) and starch (at most 160 mmol/l).The decreased production of fatty volatile acids in an interval of 12 to 24 hrs results from decreasing the number of microorganisms (Fig. 24).The maximal rate of the volatile fatty acid production is of about 12.6 mmol/hr/l medium.The concentration of individual volatile fatty acids in the course of the incubation is in Fig. 20 and besides this, in Fig. 21 there is a production of individual volatile fatty acids.In first 10 hrs of the incubation the rate of the propionate production is the same as that of the acetate production.The molar ratio of individual volatile fatty acids in the course of the incubation was as follows: The rate of the urea hydrolysis varied about 500 mgN-urea/hr/1 medium during the in vitro incubation (Fig. 22).In one hour 98 % urea was decomposed to ammonia and CO 2 , These results also correspond to the dynamics of ammonia in the medium incubated (Fig. 23).In the first hour of the incubation there was a maximum concentration of ammonia -295 mg N -NH3/1.It was continuously consumed for the synthesis of microbial proteins.Mter 8th hour of incubation the ammonia level is already decreasing below the concentration optimal for the proteosynthesis.This decrease also resulted in decreasing the production of volatile fatty acids and proteins.After 10th hour of incubation there is a moderate increase of the ammonia concen--0.1 '" tration.This increase obviously results from the decomposition of microbial proteins, amino : acids or amines.By suitable ::> choice of initial concentration of urea and amino acids an equilibrated dynamics was achieved.
The synthesis of total• proteins was followed in a time interval of 0 to 24 hrs (Fig. 24).In first two hours the production of proteins stagnated.This corresponds to a latent stage of the growth of microorganisms.From the second to seventh hour the exponential stage was observed.In intervals 2 to 4 hrs and 4 to 7 hrs the rate of the protein synthesis varied about 48.3 and 51.3 mg N-proteins/hr/l medium, respectively.Mter the seventh hour the protein synthesis was stopped.The decreasing level of ammonia exerted the main effect on the limitation of the proteosynthesis.In 7 hrs the protein concentration in medium reached 85 % of the initial concentration in the rumen liquor.
The utilisation of A TP by microorganisms in medium was as follows (Fig. 25).! In the initial hour the utilisation of A TP is concentrated to the adaptation to new environment and with respect to the decrease of protein the Y ATP is of zero value.Between the first and second hours the microorganisms start to grow and Y ATP increased to 13.4 g d. b./mol ATP.In the exponential stage of the growth the Y ATP increases to 21.8.Then the concentration of protein is decreased.The growth of bacteria after the 12th hour is limited by the substrate and this also corresponds to a low Y ATP --2.5 g d. b./mol ATP.In our experiments, num.bers of bactei"ia in the course of 6-hrs incubation were established.The increase in three groups of bacteria -amylolytic, proteolytic and cellulolytic -was also determined.For the cultivation of amylolytic bacteria medium according to Bryant and Burkey (1953) was used, the group of proteolytic bacteria was cultivated in medium according to Fulghum and Moore (1963) and the cellulolytic group in medium Hungate (1950).
These groups of bacteria were cultivated in solid agar medium on Petri dishes.The dishes were put into an anaerostat, where the anaerobic atmosphere was formed by a mixture of gases (95 % CO2 + H2).
The determination of the number of the grown colonies was performed after 5 days in amylolytic and proteolytic bacteria and after 7 days in the cellulolytic group.

Group Amylolytic Proteolytic Cellulolytic
Number of bacteria before incubation after 6-hr incubation 5.0 ± 1.0 X 10 5 3.7 ± 1.3 X 10 8 4.2 ± 0.6 X 10 5 1.5 ± 0.3 X 10 7 7.1 ± 2.0 X 10 6 2.4 ± 0.9 X 10 9 The rumen liquor was taken from experimental animals just before feeding.This corresponds to the initial representation of individual groups: the largest increase in biomass was found in cellulolytic bacteria, smaller increase was observed in amylolytic and proteolytic bacteria.In six hours of the incubation there was the most remarkable increase in the amylolytic group, about twice as high as that in the cellulolytic group.The smallest growth was noted in proteolytic bacteria.

Discussion
By adjusting the "method I" conditions were achieved, under which the in vitro fermentation yields results comparable to those obtained during an in vivo fermentation.
Besides the first hour, the oxidation-reduction potential in the course of the incubation is within physiological ranges.The initial increase obviously results from the contamination with oxygen and from sudden addition of a large amount of the substrate.The redox potencial was tentatively expressed in terms of rHo However, some authors do not recommend this expression for anaerobic media (Jacob 1970).The expression of redox potential in rH units presented in our case smaller spread of results than that in Eh units.Similar results were obtain• ed by Barry et al. (1977) and Marounek (1980).For this reason, during the incubation of the rumen liquid in vitro it is possible to recommend the expression of the redox potential with the help of rHo The results of the dependence of rH on time also show a physiological course.They differ from Broberg's (1957) results by about two units of rHo In spite of this increase there is no obvious disturbance of the activity of rumen microorganisms.The vitality of metanogens, i. e. bacteria which are most sensitive to the increase in r H was demonstrated in the course of the in vitro incubation.
The production of volatile fatty acids in the followed experiments is similar to that under in vivo conditions.The ascertained production rate of fatty volatile acids -12.6 mmol/hr/l medium -corresponds to the rate of the production in the rumen -5 to 18 fatty volatile acid/hr/l rumen liquor (Van Der Walt and Briel 1976;Marounek 1978).The increased production of propionate is characteristic for concentrate feeding (Ryan 1964;Church 1970).For a given composition of substrates the production of particular fatty volatile acids is normal.The dynamics of the production of fatty volatile acids in the course of the incubation did not correlate with the protein production and so it cannot be recommended as an indicator of the bacteria growth.
In the course of the incubation no lactate production was found.From this it is possible to consider that neither Streptococcus bovis nor other bacteria producing lactate are selected as in the case of a method used by Maeng and Baldwin (1976).
The maximal rate of the hydrolysis of urea in the experiments performed varied about 500 mg N-urea/hr/l medium.This corresponds to the hydrolysis rate in the rumen (Sharma and Shukla 1973;Henderickx 1976;Mahadevan et al. 1976Mahadevan et al. , 1977)).The ammonia concentration in medium (at most 300 mg N -NH31) cannot moderate the growth of microorganisms (Owens and Isaacson 1977) and its level is the same to the 8th hour of incubation as that in the rumen (Church 1970).
The synthesis of the microbial protein was satisfactory in the incubated medium.The initial stagnation represents the lag-phase of the microorganism growth.It is obviously affected by a change of the environment and initial high redox potential of the medium (Malek 1953).A similar stagnation was also observed by Maeng and Baldwin (1976) who sampled rumen microorganisms for the incubation in log-phase of the growth, i. e. 2 hrs after feeding.With respect to the adaptation of microorganisms to a new environment, it is unsuitable to use the batch--cultures for following the proteosynthesis of microorganisms in a time interval shorter than 2 hrs (Barr 1974).The rate of the growth of microorganisms in the log-stage is in an agreement with the found rate of the growth in vivo (Walker et al. 1975;Church 1970) as well as in vitro (Nikolic et al. 1975;Henderickx 1976).
There was a high utilisation of ATP for the growth during the incubation.Some authors refuse the batch-culture 'just for low utilisation of A TP for synthetic processes, or for high consumption of ATP in the basal metabolism (Hespell 1979).The consumption of the basal energy -Me -is to the largest extent dependent on the diluting capacity of the medium.With respect to the fact that the batch culture has continuously a zero diluting rate, the growth of microorganisms depends on the initial dilution of the inoculum.In our case the rumen liquor was diluted during inoculation three times which, when related to 10-hrs interval of incubation, represents 20 % dilutionjhr.
In the exponential stage the value Y ATP achieves the growth of 21.8 g d. b.jmol ATP.However this was not achieved in all the incubations.The YATP values usually vary between 13 and 23g d. b.jmol ATP, which are results comparable with those obtained in continuous incubations (Hespell and Bryant 1979).Similar Y ATP values with the batch culture were also obtained by Maeng and Baldwin (1976).
Our results indicate that the obtained parameters of the batch fermentation in vitro in 6 hrs are comparable with parameters of the fermentation obtained in vivo.It is obvious, that in 6 hrs particularly amylolytic organisms and organisms utilizing intermediate products contribute to the fermentation.This fact should always be considered when evaluating the incubation results.

Fig. 4 .
Fig. 4. Effect of osmotic pressure on protein synthesis during incubation of rumen microorganisms in vitro.

Fig. 5 .
Fig. 5. Effect of glucose on pH during batch-culture •of rumen microorganisms in vitro.Dash lines show the limits of pH in vivo GPnditions.

Fig. 6 .
Fig. 6.Effect of cellobiose and maltose on VFA production during incubation of rumen microorganisms in vitro.
Fig. 7. Effect of cellobiose and maltose on protein synthesis during incubation of rumen microorganisms in vitro

Fig. 8 .
Fig. 8. Effect of mixture of cellobiose and maltose on VFA production during in vitro incubation.

Fig. 9 .
Fig. 9. Effect of cellobiose and maltose on~ microbial protein production during in vitro incubation.

Fig. 10 .
Fig. 10.Effect of urea on ammonia level during incubation of rumen microorganisms in vitro.
Fig. 11.Effect of amino acids on 3.p incorporation into rumen microorganisms during incubation in vitro.
Fig. 12.Effect of amino acids on protein synthesis of rumen microorganisms in vitro.

Fig. 14 .
Fig. 14.Effect of starch and dry potato meal on 32p incorporation from phosphate into rumen microorganisms in vitro.

Fig. 17 .
Fig. 17.Changes of Eh during incubation of rumen microorganisms in vitro.

Table 1
Comparison of ditferent batch cultures of rumen n>icroorganisms in vitro I I

Table 2
Effect of saccharides on starch fermentation by rumen microorganisms in vitroVal ues are means of four observations.

Table 3
Effect of mono-and disaccharides on starch fermentation by rllmfn microorgBnUms in vitroValues are means of ten obsel vations.

Table 4
Effect of amino acids on pH and VFA production during incubation of rumen microorganisms in vitro a) pH (average of 6 observation) After the incubation (7 hrs) the residual atmosphere was washed with 40 % NaOH and so CO 2 was removed.The remaining gases (CH 2 and H 2 ) represented 50 to 10 % of the original gas volume.