In vitro dry matter disappearance using roe deer inocula from summer and winter

In vitro dry matter disappearance (IVDMD) of 12 forages were determined using ruminai inocula from 10 European roe deer (Capreolus capreolus L.) collected in summer and winter. There was significant difference in the ability of winter and summer inocula to digest winter and summer forages respectively. Each of the 6 summer forages had a significantly higher IVDMD in ruminai inocula of animals collected in summer versus winter. However, no significant difference in IVDMD of winter versus summer inocula was observed for each of the 6 winter forages. These results suggest adaption, although limited, by ruminai microorganisms in roe deer to winter forages or a potential problem in standard in vitro laboratory procedures when using animals on a high-fiber diet as inocula donors.


Introduction
The European roe deer has been classified as a concentrate selector (Hofmann et al. 1976) dependent upon consumption of high quality forage to meet nutritional requirements.When snow depth exceeds 0.5 m, roe deer are apparently forced to browse twigs and buds from deciduous and evergreen trees (Cederlund et al. 1980) low in digestibility.
Studies with domestic ruminants have shown that the composition of rumen microbial populations adapt to dietary changes which enables fermentation of the major components of the diet (Hungate 1966).Rumen microbial adaptations to seasonal dietary changes have been suggested as an important nutritional strategy for wild ruminants facing a seasonal environment (Felber 1968;Thomas and Kroeger 1980;Trudell et al 1980;Cederlund and Nystrøm 1981;Boomker 1984).This study tests the hypothesis that the ruminai microbial population of European roe deer adapts to seasonal dietary changes.This was accomplished by assessing the ability of summer and winter collected ruminai inocula to digest summer versus winter forages.Summer forages, low in fiber, were expected to be more completely digested by ruminai inocula collected during summer than by ruminai inocula collected during winter.Winter forages, high in fiber were expected to be digested more completely by winter collected inocula than by summer collected inocula.

Study area
The study was conducted in southeastern Norway, in the vicinity of the Agricultural University of Norway.The area was within the boreonemorale zone (Abrahamsen et al. 1977).Long-term mean temperature in January and July were -5.2 and 16.8°C respectively and average snow depth in February was 35 cm, with a range of approximately 0 to 110 cm (Heldal 1975).

Methods
In vitro dry matter disapperance (IVDMD) of 12 typical forages of roe deer were determined using inocula from all portions of the rumen of 5 roe deer collected in June 1987 and 1988 and 5 animals collected in February 1987.One in vitro trial was run immediately after each collection.Two of the animals collected in summer were road-kills.
Samples of typical summer forages of roe deer in Fennoscandia (Huseby 1976;Cederlund et al. 1980) consisted of leaves from 4 herbs and 2 deciduous trees, and typical winter forages (Hagen 1958;Cederlund et al. 1980) consisted of green shoots from 1 dwarf shrub, needles from 2 evergreen trees and current-year growth of stems (<2 mm in diameter) from 3 deciduous trees.Forages were collected in June 1986 and February 1987.
Plant samples were dried at 50°C and ground to pass through a 1 mm mesh sieve.Representative samples of whole rumen contents were taken, freeze-dried to a constant weight and ground.Forages and rumen samples were analyzed for neutral detergent fiber (NDF), acid detergent fiber (ADF), and acid detergent lignin (ADL) (Van Soest and Wine 1967).
A two stage in vitro dry matter disappearance (IVDMD) technique (Tilley and Terry 1963) was used.Immediately after being sacrified or reported as road-killed (within 1 hour after estimated death) ruminal fluid from an animal was strained through 4 layers of cheesecloth and added to preheated buffer solution (McDougall 1948).Inocula was maintained at 39°C and flushed with CO2 for 15 minutes.Duplicates of all plant materials, three blanks and 3 different standards in triplicates were incubated in each run.Mean in vivo dry matter digestibility of standards by sheep were 80.5 %, 73.4 % and 68.5 % (unpublished data).Deer were grouped in two distinct groups according to time of collection and were a representative subsample of the population.Pairedsample t-test (Zar 1984) was performed to test average difference in IVDMD between the two groups of animals, for summer forages and winter forages.In addition Student's t-test (Zar 1984) was performed to test differences in mean IVDMD between the two groups of animals, for each of the 12 forages separately.Student's t-test was also carried out to test differences in chemical characteristics (i.e., hemicellulose, cellulose and lignin) of rumen contents sampled in summer and winter and of summer and winter forages.Significance levels were set to P<0.05 for all tests.

Results and discussion
The summer forages ranged between 50-70 % IVDMD, and winter forages ranged between 30-40 % (Fig. 1).Seasonality in forage quality was reflected in the significantly higher cellulose and lignin content of winter forages than summer forages (Table 1) and significantly greater cellulose and lignin content of rumen content from winter compared to summer collected animals (Table 2).The hypothesis of microbial adaptation to dietary changes in roe deer therefore may be tested by assessing the ability of summer and winter collected inocula to digest summer versus winter forages.
Seasonal adaptations of rumen microorganisms resulted in positive effects on digestibility of forages both in summer and winter.There was significant difference in the ability of winter and summer inocula to digest winter and summer forages respectively.The IVDMD of each of the 6 summer forages was significantly higher with ruminal inocula from deer collected in summer versus winter (Fig. 1).However, no significant difference in IVDMD of winter versus summer inocula was observed for each of the 6 winter forages (Fig. 1).Faster fermentation rates in summer than in winter have been reported in various northern and temperate cervids (e.g., Felber 1968;Short 1971;Langer 1974).These differences were most pronounced for highly digestible forages and was probably an effect of higher quality forages and higher food intake in summer which can support a larger microbial population.However, different forages may approach the asymptotic maximum in vitro digestibility at different times (e.g., Troelsen and Hanel 1966;Milchunas et al. 1978;Trudell et al. 1980), and a given forage may approach the same end point in different ways depending on the microbial environment (Tilley and Terry 1963, Troelsen and Hanel 1966, Trudell et al. 1980).Normally, 48 hours after incubation the in vitro fermentation process is complete (Tilley and Terry 1963;Milchunas et al. 1978).End-point digestibility does not measure rate of digestion, but rate of digestion is related to in vivo rate of passage.Total mean retention time for roe deer fed a high quality diet was about 20 hours versus 30 hours when fed a low quality winter diet high in fiber (Holand unpublished data).Further, there are time lags observed for bacterial populations to adapt to a switch in substrate quality (Van Soest 1982).Both rate of digestion and lag times in microbial population adaptation are factors not accounted for in end-point in vitro digestion.Therefore, my estimates of the ability of roe deer to adapt to seasonal dietary changes is probably conservative.
The pool size of cellulolytic organisms may limit the initial breakdown of diets rich in fiber (Van Soest 1982).Roe deer have a sparse ruminal population of cellulolytic bacteria compared to larger cervids (Cervus elaphus and Dama dama) (Briiggeman et al. 1967;Prins and Geelen 1971).Based on in vitro techniques, Prins and Geelen (1971) record that cellulose fermentation by European roe deer is low, in fact the lowest in vitro cellulose disapperance rate out of 11 ruminants (Prins et al. 1984).In contrast, Dissen and Hartfiel's (1985) in vivo studies indicate a relatively high digestibility of crude fiber by roe deer, however the forages tested were low in crude fiber compared to a typical winter browse diet.The more likely results are by Drozdz (1979) who found an in vivo digestibility coefficient of natural winter food (30-40 % crude fiber) of 40 %; again indicating low cellulolytic activity.In both groups of studies proximate food analysis was used to assess fiber and this technique does not given a uniform composition of crude fiber (Robbins 1983) and therefore makes comparison difficult.
Variations in the rate of cellulolysis with the changes in diet are well documented (Prins et al. 1984).There are some indications that IVDMD of cellulose-rich plant material is highest during winter in European roe deer (Felber 1968;Cederlund and Nystrøm 1981).The higher IVDMD of winter forages in inocula collected during winter may indicate higher cellulolytic activity.However, no evidence of higher activity was found considering that for each of the 6 winter forages the cell soluble fraction was higher than its IVDMD.This makes a sound biological interpretation difficult, but may suggest some component of the soluble fraction is prevented from being digested, perhaps by plant compounds (White^ and Trudell 1980;Person et al. 1980).
Microbes that digest fiber adhere to the particulate phase of the rumen content (e.g., Akin and Amos 1975;Patterson et al. 1975).Greater amounts of fiber in the rumen contents collected in winter than in summer may have resulted in the discarding of a greater proportion of cellulolytic microorganisms from rumens collected in winter.This may have contributed to the small differences in IVDMD of winter forages using ruminal inocula collected in summer versus winter.Agitation of rumen contents from animals on high fiber diets may be necessary to dislodge cellulolytic microorganisms before filtering and use as in vitro inocula (Milchunas and Baker 1982).
The inoculum preparation stage of in vitro disappearance trials using wild ruminants as donors may influence disappearance coefficients, because of difficulties in maintaining the microbial activity of the ruminal fluid (Schwartz and Nagy 1972).Prins and Geelen (1971) found that storing rumen fluid up to 2 hours did not significantly change cellulolytic or amylolytic activi-ty.However, data from Milchunas and Baker (1982) indicated that if inoculum temperatures dropped to 29 °C or if incubation was delayed for 2 hours fermentation was significantly reduced.In this study, incubations were started within 1.5 hours after the animals were sacrified or estimated road-killed, to maintain viable inocula.Fermentation was evidenced by sustained gas production.
Although a conclusion possibly limited due to laboratory technique, my results based upon standard IVDMD techniques does not conclusively support the contention that a high degree of rumen microbial adaptation to diet is an important mechanism by which roe deer improve their survival in harsh winter conditions.

Table 1 . Mean composition of cell wall fraction (% of dry matter
) (±SD) of summer (n = 6) and winter (n = 6) forages used as substrate for in vitro trials.* Seasonal significant difference (P < 0.05).

Table 2 . Mean composition of cell wall fraction (% of dry matter) of rumen samples (±SD) of roe deer used as donors of inocula.
* Seasonal significant difference (P < 0.05).