EFFECT OF MONENSIN AND ITS METABOLITES IN BROILER LITTER ON SHEEP CONSUMING THE BROILER LITTER

.1 due to monensin. It has also been ' I ABSTRACT Two trials w~e conducted to determine the effect of monensin in broiler litter on sheep receiving the broiler litter in their diets. Broiler litter from chickens fed monensin as a coccidi<lStat, and from chickens receiving no coccidiostat, was included at a level of 30% in 2 sheep diets. In a further 2 treatments, monensin (15 mg kg-I) was added to each of the 2 diets to give a 2x2 factorial experimental design. In the first trial, copper (20 mg kg1 feed) was added to the diets. These lambs were fed individually at a sligndy.re$trjcted leyelof intake. No dif.. ferences between treatments were observed in feed intake, average daily gain or efficiency of feed utilisatio~ or in the. concentrations of zinc, iron and manganese in the liver, glutathione perbxidase in erythrocytes and creatine kinase concentrations in the. plasma. Hepatic copper con· tent and copper retention in the livers o(the.sheep receiving the added monensin were signfficantly higher (P <0,05 and < 0,01 respectively) than in those not receiving added monensin. The aspartate tran· saminase and alkaline phosphatase concentrations in the plasma of these sheep were also higher (P < 0,05) than in those not consuming added monensin. In the second trial, the lambs were group-fed according to treatment and received the diets on an ad lib basis. The mean intakes of the groups receiving the diets with the added· monensin, were lower than the intakes by the other groups. It was concluded that the monensin metabolites in broiler litter had no measurable effect on any of the parameters of monensin activity investigated in sheep> when such litter was induded in sheep diets.


INTRODUCTION
Monensin sodium (Elanco Product Company, Indianapolis, USA) is used widely as a coccidiostat in the diets of broilers, at a maximum recommended rate of inclusion of 120 mg kg-I feed 6 • Monensin is also effective as an ionophor in the diets ofsheepll, at recommended levels of 15 to 22 mg kg-I, and of cattle at a maximum of 33 mg kg-l feed b • As an ionophor, the feeding of monensin can have various beneficial effects on the production of ruminants.One of its most important effects is an improvement in efficiency of feed utilisation in animals on high energy diets.This is quite often accompanied by a decrease in feed intake, without a concurrent decrease in bodymass gains 3 • Further beneficial effects are the control of coccidiosis in calves, control of feedlot disorders such as a reduction in the incidence of bloat 2 and acidosis especially during the period of adaptation to high concentrate diets 8 .
Various metaboiic changes due to the consumption of monensin have been observed 3 • Of relevance to the present study are the changes in the metabolism of minerals.Van Ryssen & Barrowman l8 found that monensin increased the retention of dietary copper in the liver of sheep.Anderson et al.I measured an improvement in the selenium status of sheep reported that the metabolism of sodium, potassium, magnesium, calcium, phosphorus and zinc in ruminants is altered by monensin9 10 14.
At intakes higher than the recommended levels, monensin can have detri-I mental effects on animals.A reduced feed intake is one of the first symptoms of an excessive monensin intake ll 12.At high intakes monensin can be very toxic.The LD50 of monensin for sheep is 11,9 mg kg-I and for cattle 26,4 mg kg-I livemass l2 .The LDI for a 400 kg bovine is estimated to be 2 210 mg monensin l2 .Symptoms of chronic toxicity are anorex• ia, skeletal muscle weakness, ataxia, decreased ,.gain in mass and eventuall} death.Primary target tissues are the skeletal and cardiac muscles l5 .This is evi• dent from elevated plasma concentrations . of enzymes such as creatine kinase, which is usually associated with the breakdown of muscle tissues It was demonstrated with the use of radio-active markers that the monensin ingested by poultry, is excreted almost quantitatively via the faeces in the form of metabolites 6 • The monensin is meta' bolised in the body of the bird and reaches the digestive tract in the bile.The presence of these metabolites in faeces could not be detected with the use of the standard microbial test for monensin and showed a very low biological activity>.The question therefore arises whether the metabolites of monensin in the broiler litter could have the same detrimental effect on animals as does the unmetaboIised compound. .

Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2011)
Two trials were conducted to establish whether excreted monensin or its metabolites in broiler litter, have the same measurable and metabolic effects on sheep as dietary monensin when sheep consume the litter.It is assumed that if litter is included in a sheep diet at a level of 30%, a relatively high level of these metabolites should still be• present in the diet.

MATERIALS AND METHODS
Two groups of broilers were reared for 6 weeks in experimental rearing pens containing wood shavings as bedding material.One group received a diet containing monensin as a coccidiostat at a level of 100 mg kg-1 feed while the other group received no monensin.At the end of the rearing period, the litter from these pens was removed separately, sifted and sun-dried.
Two feeding trials with sheep were conducted .inwhich these 2 types of litter were included in their experimental rations.At least 3 weeks before the onset of the trials, all the sheep were vaccinated against botulism and dosed with a widespectrum anthelmintic.
Trial 1: Liver biopsies were done on SA Mutton Merino wethers, (n=38) ca 9 months of age, with a mean body mass of 29 kg.Six wethers with liver copper concentrations deviating the most from the mean of the group, were slaughtered as a pre-experimental group to determine the relationship between liver mass and empty body mass.Using a 2x2 factorial design, the remaining 32 wethers were allocated to 4 treatment groups in a ran-domised block according to the' copper concentration in their liven •.The treatments were: i) a control group (C) receiving a diet containing 30% broiler litter from the broilers which did not receive the coccidiostat in their diet; ii) the control diet (as for Group C) but with added monensin (15 mg kg-1 feed) (C +); iii) a group on the test diet (T) including the litter (at 30%) from the chickens receiving monensin, and iv) the test diet fortified with monensin (15 mg kg-1 feed) (T+).The experimental diet consisted of 300 kg broiler litter, 380 kg hominy chop, 50 kg fishmeal, 180 kg veld hay, 80 kg molasses meal, 5 kg lime and 5 kg salt per 1000 kg.Copper (as copper sulphate) was added to the rations at 20 g per 1000 kg.For the first 2 weeks of the trial, monensin was included at half the intended level.This practice enabled the lambs to adapt to monensin feeding and allowed an opportunity to minimise mortalities if treatments should prove to be toxic.The sheep were fed individually.In order to keep the metabolic changes in the body as comparable as possible, it was planned to keep feed intake per sheep as constant as possible.Therefore allocations were adjusted weekly, according to the intakes of the groups receiving the added monensin.
Feed intakes and body mass changes were recorded weekly.The sheep were observed closely for ill-health and any other symptoms of toxicity.Blood samples were collected in heparin at va-rious stages of the trial for mineral and enzyme assays.In order to minimise any effect of the liver biopsy on enzyme levels in the plasma, the first collections were only done at 36 d after the' onset of the trial.On Day 77 the lambs were slaughtered, after exsanguination.Carcass arid fresh liver masses were determined.Liver samples were taken, dried at 80°C and stored for further analysis.
Trial 2: In a group feeding trial, ewe lambs were allocated randomly to simil~r experimental treatment groups as in Trial 1.However, the added copper was omitted from the diets.The lambs (5 per group) received their diets on an ad lib feeding basis in order to measure voluntary feed intake.The trial lasted for 45 days.Feed intakes per group and individual body masses were recorded weekly.
Atomic absorption spectrophotometry was used to determine the concentrations of copper, zinc, iron and manganese in the liver and feed.Sodium and potassium concentrations in plasma were obtained using flame photometry.Glutathione peroxidase (EC 1.11.1.9)was assayed at 25°C using the coupled enzyme procedure as modified by Whanger et aJ.20.Boehringer Mannheim standard kits (Boehringer Mannheim GmbH Diagnostica, West Germany) were used to estimate the aspartate transaminase (EC 2.6.1.1),alkaline phosphatase (EC 3.1.3.1) and creatine kinase (EC 2.7.3.2) concentrations in plasma.The monensin content of the diets was estimated using an anti-microbial growth assay;.
To calculate hepatic copper retention at the end of the trial, the following procedure was carried out: An estimate of individual liver masses at the onset of the trial was made from the body mass of each lamb and the ratio ~f UJ ~oth the boxes designated B, contain the means of the 2 basic treatments, viz.with or, without monensin as a coc-:dlOstat in Bl, and with or without added monensin in B2.The differences within each box can be compared with e LSD's, e.g. in B2 the difference is 151 therefore less than 156 and not significant at P = 0,05.
Where the means within the basic treatments were large, being significant or approaching significance, the means are presented in subsequent tables.Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2011) liver mass to body mass as obtained from the pre-experimental slaughter group.The copper content of the livers was calculated from this estimated liver mass and the copper concentration in the biopsy samples.This was subtracted from the copper present in the liver at the end of the trial (copper concentration x liver mass) to calculate hepati~::"copper accumulation.The percentage dietary copper retained in the livers was obtained from total copper intake and hepatic copper accumulation.
The 2x2 factorial analyses were performed on the data with the aid of the computer programme Genstat (Lawes Agricultural Trust, 1980).To demonstrate the interpretation of results in a 2x2 factorial statistical design, a description is presented in Table 1 which can be used to evaluate the proceeding tables.In these tables the original treatment means, each from 8 sheep, are supplied and compared with the use of the LSD (least significant difference).Where combined means related to one of the 2 basic treatments were important, the means of the 16 sheep are presented with the relevant LSD's, e.g. in Tables 5 and 6.

Diet, feed intake and performance
The concentrations of monensin and minerals in the experimental diets are presented in Table 2.
Due to restricted feeding, no difference in feed intake among groups was observed in Trial 1.In Trial 2 the differences in Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2011)     feed intake between the groups receiving the metabolites and those without the metabolites (Treatments T versus C and I T + versus C +) were negligible, while those receiving the added monensin.
(Treatments C + and T + ) consumed substantially less feed compared to the other 2 groups respectively (Table 3).It was not possible to compare the in~akes in Trial 2 statistically, because group feeding was applied.
The mean final body masses, average daily body and carcass gains and efficiency offeed utilisation (kg feed consumed divided by kg gain) calculated for body and carcass mass gains (initial carcass masS estimated from live mass at onset of trial and dressing percentagl; of pre-experimental slaughter group), did not differ significantly between treatments in Trial 1 (Table 3).In Trial 2, sheep on the T treatment showed a higher (P < 0,05) gain in body mass than sheep on the other treatments (Table 3).

Blood analyses
The concentration of sodium (149, 148, 149 and 151 mmol 1-1) and potassium (4,7; 4,8; 5,0 and 4,9 mmoll-I ) in plasma• for treatments C, C +, T and T + respectively, did not differ significantly between treatments.The differences between treatments per collection in erythrocyte glutathione peroxidase activity and plasma creatine kinase, were not significant (Table 4).However, both glutathione peroxidase and creatine kinase levels increased between the col1ection on Day 36 and that on Day 77.When extra monensin was added to the diets (Treatments C + and T +), the mean concentration of aspartate transaminase and alkaline phosphatase (Table 5) in plasma were significantly (P < 0,05) higher than in the 2 treatments without, the added monensin.

Liver
Mean iron, zinc and manganese concentrations in the livers did not differ significantly due to treatments.These concentrations for treatments C, C +, T and T + were: 228, 224, 241 and 224 mg iron kg-I dry matter (DM); 169, 163, 182 and 159 mg zinc kg-I DM and 13,9; 14,2; 13,3 and 13,9 mg manganese kg-I DM respectively.d The livers of the groups receiving aded monensin in the diets (C + and T + ) had a higher (P < 0,05) total hepatic cop-Per Content and copper accumulation ~able 6) than the groups without the adc ed monensin.The percentage dietary opper retained in the livers of the groups :~vin?added monensin was signifi-.gr Y hlgher (P < 0,01) than for the other oups (Table 6).
The mean liver mass and liver mass expressed as a percentages of carcass mass, did not differ significantly between treatments.Fresh•liver mass as a percentage of carcass mass in treatments C, C +, T and T + was 3,2; 3,3; 3,3 and 3,6% respectively.

DISCUSSION
Accepting that metabolites from monensin cannot be detected in poultry manure 6 , the concentration of un metabolised monensin in the litter of the birds receiving the coccidiostat, can be calculated from the monensin in diet T (based on the 30% litter in the experimental diets).This amounted to 23 mg monensin per kg litter.Some of this might have originated from feed spilled onto the bedding material.Donoho 6 reported that chickens excreted less than lO%•ofthe oral1y-administered monensin as "parent" monensin via the faeces.The presence of 3 mg monensin per kg feed in the Control diet, is probably a false positive reading which showed up in the anti-microbial growth assay for monensin.Such a false positive reading, plus sampling errors, might have reduced the reliability of the monensin concentrations presented.
Evidence of an effect of monensin on the sheep was observed in Trial 2, in the treatments where monensin was added (Treatment C + and T +) to the diets.Feed intake was not restricted in this trial• and voluntary feed intake in these 2 treatments was lower than in the others.This is in accordance with most other studies 3 • Monensin metabolised by the broilers, did not depress the feed intake of the sheep.In both trials, efficiency of feed utilisation did not differ between treatments.An improved efficiency of feed utilisation is considered to be one of the main beneficial effects of monensin in ruminants 3 , although this is quite often not observed in sheepls.The significantly higher gain in mass of the sheep in Treatment T of Trial 2 is difficult to explain.In Trial 1, the average daily gain in Treatment T was also higher, though not statistically different, from that in Treatment C.
The increased copper content and copper retention in the livers of the sheep receiving the added monensin, is further evidence of an effect of monensin on mineral metabolism.This agrees with the observation by Van Ryssen & Barrowman 1s that monensin enhanced hepatic copper retention in sneep.From this study it is clear that the metabolites of monensin in the diets had no effect on copper metabolism while the added monensin increased copper accumulation in the liver.Significantly higher plasma concentrations of aspartate transaminase and Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2011) A glutathione peroxidase activity in ex-cesS of 40 nmol NADPH mg-I haemoglobin indicates that the animal is consuming sufficient selenium to meet its requirements l .Therefore, the high levels of this enzyme in the present study, preclude the possibility that a selenium deficiency could accentuate the effects of monensin toxicosisl 9 • Anderson et al.I reponed that monensin improved the selenium status of sheep.On Day 36, in the present trial, the groups receiving the added monensin (C+ and T+) had higher glutathione peroxidase levels than treatments C and T respectively, although the differences were not statistically significant.On Days 58 and 77 differences between treatments were minimal, which is in agreement with Costa's4 observation that monensin has no effect on the selenium status of animals.Whanger et a1. 20observed a continuous increase with time in the blood ghitathione peroxidase concentrations of sheep receiving high levels of selenium in the diet.From the increases in enzyme ~oncentrations between Days 36 and 17, It can be concluded that the sheep diets Contained more than sufficient selenium to meet their requirements.
The fact that creatine kinase concentrations in plasma did not differ significantly between treatments may indicate that monensin, whether derived from litter or from addition to the diet at recommended levels, had no damaging or toxic effects on muscle tissue I3 IS.The relatively high creatine kinase levels measured in all treatments at the end of the trial is difficult to explain.
From this investigation, it may be concluded that the monensin fed to poultry as a coccidiostat had been metabolised to such an extent that it became metabolically inactive.It showed no effect in any of the parameters of monensin activity measured in the sheep.These results suggest that the feeding ofmonensin as a coccidiostat to poultry does not pose any risk to sheep consuming the litter as pan of their diet.This suppons the evidence of Donoh0 6 that the metabolism of monensin in the body of poultry results in the destruction of most ofits biological activity.Funhermore, the results suggest that statements in the popular press 5 17 that coccidiostats which are excreted in the litter of broilers can cause deaths in ruminants, especially if the same product is included in the ruminant rations as an ionophore, are too generalised.They may possibly apply to certain coccidiostats, but should exclude monensin.

Table 1 :
An illustration of the interpretation ofa single 2x2 factorial analysis (Copper concentration in Table6used as example) ference between any pair of these can be calculated from the LSD's, e.g. the difference between C and C + is 207, therefore less than 220 and not significant.These means with the respective LSD's are presented in a column in SUbsequent tables.Differences between 2 treatments can be compared.

Table 2 :
The concentration of monensin and minerals in experimental diets containing broiler litter (C), broiler litter with monensin added (C + ), broiler litter from chickens which received mon~nsin (T), and broiler litter from chickens which received monensin with monensin added (T +).Copper sulphate was added to all rations

Table 5 :
Plasma concentrations of aspartate transaminase and alkaline phosphatase in sheep on diets containing broiler litter (C), broiler litter with monensin added (C +), broiler litter from chickens which received monensin (T), and broiler litter from chickens which received monensin with monensin added (T + )

Table 6 :
Hepatic copper concentrations in sheep which were fed broiler litter (C), broiler litter 98 with monensin added (C + ), broiler litter from chickens which rec~ived monensin (T), and broiler litter from chickens which received monensin with monensin added (T + ) *means to compare with or without added monensin **LSD -least significant difference ***means to compare with or without monensin as a coccidiostat not significantly different 0038-2809 JI S.Afr.vet.Ass.(1991)62(3):94Reproduced by Sabinet Gateway under licence granted by the Publisher(dated 2011)