Effects of monensin level and roughage / concentrate ratio on ruminal fermentation in bovines

Nine rumen-fi stulated cows were used in a completely randomized design replicated three times to evaluate three monensin (0, 150 and 300 mg/animal/day) and three concentrate levels (25, 50 and 75%) in a 3x3 factorial arrangement. Twenty-one day subperiods were used, the fi rst sixteen for diet (Tifton hay+concentrate) adaptation. Monensin decreased DMI only in 50%-concentrate diet (P<0.05) and did not infl uence total VFAs concentrations. For ruminal pH, molar% of VFAs, and NH3-N concentration collected 7 times/day, the responses to high level of monensin were higher in low-forage diets (P<0.05). Responses to low level of monensin were higher in high-forage diets (P<0.05).


INTRODUCTION
For a long time nutritionists have tried to manipulate rumen fermentation.This can be done by increasing propionic production, decreasing methanogenesis or proteolysis and deamination in the rumen.Initially, this objective was done through diet manipulation.However, the discovery of rumen active molecules has brought great perspectives on agricultural sciences (Bergen and Bates, 1984).
Ionophores are a class of compounds that have obtained considerable success as additives.However, responses reached with the use of ionophores are variable.This can be explained partially by different experimental conditions (Galloway et al., 1993), as concentrate and ionophor level.This trial was carried out to study interaction between these two factors on monensin response.

MATERIAL AND METHODS
Nine dry Holstein cows (470 kg BW), fi tted with ruminal cannulas, were used in a completely randomized design replicated three times to determine the effects of different doses of monensin on different forage to concentrate ratio diets.Treatments were constituted of three monensin (Elanco) levels (0, 150 and 300 mg/animal/day) and three concentrate levels (25, 50 and 75%) in a 3 × 3 factorial arrangement.Diet contained Tifton 85 hay, maize grain, soyabean meal and minerals.
Each subperiod lasted 21 days.Cows were fed twice daily at 07.00 and 16.00 h.Until d 16, cows were fed 115% of their daily intake of the previous day.From d 17 to 21, cows were fed 100% of the mean daily intake obtained from d 12 to 16. Refusals and feed samples were collected every morning from d 17 until 21, pooled, and stored in sealed plastic bags at -10°C for later analysis.
On d 21, just prior to morning meal (0 h) and at 2, 4, 6, 8, 10, and 12 h postfeeding, samples containing 500 mL of ruminal fl uid were taken.Ruminal pH was determined using a portable pH meter.Following this, 2 mL of fl uid were added to 1 mL of sulphuric acid 1 N solution, and frozen at -20°C for later determination of ammonia N.After that, 50 mL of ruminal liquid were centrifugated at 2.000 × g for 15 min, and 1 mL of the supernatant was added to 0.2 mL of formic acid and frozen for later determination of VFAs.
For intake data, variance analysis separated effects of concentrate level, monensin level, interaction and experimental subperiod.Effects of main factors were separated in linear and linearity deviation (quadratic) effects.Interaction was separated in linear effect for both factors (Con L Xmon L ), linear effect for concentrate level and deviation for monensin level (Con L Xmon D ), deviation for concentrate level and linear effect for monensin level (Con D Xmon L ), and deviation effect for both factors (Con D Xmon D ).Ruminal parameters were analysed as described, but added of split-plot factor.The surface equation was obtained by multiple regression procedure (Roush et al., 1979).

RESULTS AND DISCUSSION
Concentrate and monensin levels interacted for molar percentage of C 2 , C 3 , C 4 and C 2 :C 3 ratio, although monensin had no effect on total VFAs concentration (Table 1).Using low monensin levels (ex.150 mg), as lower was the concentrate level of diet, the greater was the molar% of C 3 (increases of 2.85, 1.34 and -4.93 percentile units for 25, 50 and 75%-concentrate diets, respectively).However, MONENSIN LEVEL AND ROUGHAGE/CONCENTRATE RATIO using high doses of the product (ex.300 mg), the highest response to monensin was obtained with larger proportion of concentrate in the diet (3.51, 4.39 and 12.55 percentile units for 25, 50 and 75%-concentrate diets, respectively).Effects of monensin were lower in high-roughage diets and low doses of product would be enough to produce maximum responses.In high-concentrate diets, responses were larger than in roughage diets, although it was necessary higher doses to produce good responses in those conditions.Effects of monensin on ruminal pH were absent in high-roughage diets (-1.6 and 1.0% for 150 and 300 mg, respectively), lower in mixed diets (2.8 and 2.0%) and higher in high-concentrate diets (7.8 and 4.1%).For NH 3 -N, monensin decreased its concentration independent on concentrate level of diet.
Fibre level or monensin dose interacted for DMI.Concentrate level caused a curvilinear response, with larger intake for mixed diets, and monensin caused a linear response, decreasing DMI.In mixed diet, there was a decrease of 33.0% in DMI caused by monensin, which is high when compared to a decrease of 6.4% observed by Goodrich et al. (1984).

CONCLUSIONS
Responses to monensin vary with fi bre level and product dose.Response to monensin is lower in high-fi bre diet and lower dose is enough to produce maximum response.On the other hand, response to monensin is higher in lowfi bre diet and higher dose is necessary to produce maximum response.

Table 1 .
Dry matter intake and ruminal fermentation pattern obtained with treatments 1