ADIASPORES OF Emmonsia parva var. crescens IN LUNGS OF SMALL RODENTS IN A RURAL AREA

Fischer O. A. : Adiaspores of Emmonsia parva var. crescens in Lungs of Small Rodents in a Rural Area. Acta Vet. Brno 2001, 70: 345–352. The purpose of this study was to compare the occurrence of Emmonsia parva var. crescens in populated and unpopulated habitats of a rural area (in a village with 615 inhabitants and in surrounding forests without human population) as indicated by findings of adiaspores in lungs of small rodents captured in these habitats. Adiaspores of E. parva var. crescens were found in lungs of 13 (9.6 %) out of 135 examined rodents. Four rodent species out of 7 were infected: the bank vole (Clethrionomys glareolus, 15.6 %), the house mouse (Mus musculus, 5.6 %), the wood mouse (Apodemus sylvaticus, 3.2 %) and the yellow-necked field mouse (A. flavicollis, 16.7 %). The highest prevalence was found in the A. flavicollis and in the C. glareolus in the village and in the forests, respectively. Intensity of the infection was low or moderate (below 100 adiaspores per animal). The highest prevalence of the infection was found in spring. The prevalence was not influenced by sex of the animals. Occurrence of both saprophytic stage of E. parva var. crescens and rodents can be enabled by plant cover of uncultivated soil in the village, where the same plant species as in the surrounding forests occur. Therefore the risk of infection for humans and animals is not limited only to forest (unpopulated) habitats. Sapronoses, emmonsiosis, Arvicolidae, Muridae, Czech Republic Emmonsiosis (adiaspiromycosis) is a world-wide distributed, largely neglected and underdiagnosed pulmonary disease of mammals, including humans (Hubálek et al. 1998). Only one causative agent of emmonsiosis, the fungus Emmonsia parva var. crescens (Emmons et Jel l ison 1960 van Oorschot1980), occurs in the Czech Republic (Dvofiák et al. 1973). The infection is widespread especially among rodents of the families Arvicolidae and Muridae (Hubálek et al. 1991) and more rarely in the families Sciuridae (Kfiivanec et al. 1976) and Cricetidae (Hubálek 1999). The source of infection is the saprophytic stage of fungus producing minute conidia. After inhalation of the conidia by mammalian host, large thick-walled spherules called adiaspores develop in host tissues, most often in lungs. Expanding adiaspore, a parasitic stage of the fungus, causes inflammatory reaction in lung tissue (Koìousek et al. 1971; Halouzka et al. 1989; Ooi and Lin 1996), which leads to collapse of the adjacent alveoli resulting, in case of massive infection, to respiratory distress or even lung failure (Hubálek 1999). Macroscopic pathomorphological changes are densely disseminated whitish nodules about 1 to 2 mm in diameter, described by Koìousek et al. (1971) in a case human emmonsiosis. After death of the infected host, the saprophytic stage can grow from the adiaspores released from dead host’s body (Kfiivanec and Otãená‰ek 1977). Although the main source of conidia is the soil (Dvofiák et al. 1973; Prokopiã and ·tûrba 1978), conditions of natural infection were not studied sufficiently. Because of danger for humans (Koìousek et al. 1971; England and Hochholzer 1993; Nuorva et al. 1997; de Montprévi l le et al. 1999) and close taxonomic relations of Emmonsia to other pathogenic fungi of the anamorphous genera Paracoccidioides, Blastomyces and Histoplasma (Peterson and Sigler 1998) emmonsiosis was studied by many biologists, mycologists and microbiologists, but no ACTA VET. BRNO 2001, 70: 345–352 Address for correspondence: Oldfiich Arno‰t Fischer Bohuslava MartinÛ 44 CZ 602 00 Brno Czech Republic Phone: + 420 5 4323 4315 http://www.vfu.cz/acta-vet/actavet.htm

Emmonsiosis (adiaspiromycosis) is a world-wide distributed, largely neglected and underdiagnosed pulmonary disease of mammals, including humans (H u b álek et al. 1998). Only one causative agent of emmonsiosis, the fungus Emmonsia parva var. crescens (Emmons et Jellison 1960, occurs in the Czech Republic (Dvofiák et al. 1973). The infection is widespread especially among rodents of the families Arvicolidae and Muridae (H u b álek et al. 1991) and more rarely in the families Sciuridae (Kfiivanec et al. 1976) and Cricetidae (H u b álek 1999). The source of infection is the saprophytic stage of fungus producing minute conidia. After inhalation of the conidia by mammalian host, large thick-walled spherules called adiaspores develop in host tissues, most often in lungs. Expanding adiaspore, a parasitic stage of the fungus, causes inflammatory reaction in lung tissue (Koìousek et al. 1971;Halouzka et al. 1989;Ooi and Lin 1996), which leads to collapse of the adjacent alveoli resulting, in case of massive infection, to respiratory distress or even lung failure (H u bálek 1999). Macroscopic pathomorphological changes are densely disseminated whitish nodules about 1 to 2 mm in diameter, described by Koìousek et al. (1971) in a case human emmonsiosis. After death of the infected host, the saprophytic stage can grow from the adiaspores released from dead host's body (Kfiivanec and Otãená‰ek 1977). Although the main source of conidia is the soil (Dvofiák et al. 1973;Prokopiã and ·tûrba 1978), conditions of natural infection were not studied sufficiently. Because of danger for humans (Koìousek et al. 1971;England and Hochholzer 1993;Nuorva et al. 1997;de Montpréville et al. 1999) and close taxonomic relations of Emmonsia to other pathogenic fungi of the anamorphous genera Paracoccidioides, Blastomyces and Histoplasma (Peterson and Sigler 1998) emmonsiosis was studied by many biologists, mycologists and microbiologists, but no complete information about all possible routes of infection and health hazard for humans and animals in densely populated areas, such as in villages and towns, is available. Most studies have been hitherto performed in exoanthropic habitats. Village ecosystem includes not only intensively cultivated arable fields, orchards and gardens, which are not suitable for growing of the saprophytic stage (H u bálek et al. 1998), but also uncultivated areas with weeds, such as nettle (Urtica spp.) and common elder (Sambucus nigra), and wet, shady places, enabling growth of the saprophytic stage of the fungus.
The aim of this study was to compare the occurrence of Emmonsia infection in village and forest habitats as indicated by findings of adiaspores in lungs of small rodents of the families Arvicolidae and Muridae.

Area under study
South-Moravian village Ketkovice near Brno (N 49 o 08' E 16 o 06', quadrat 6863 of the national faunal mapping grid) with 615 inhabitants is situated 35 km west of Brno and 5 km from west margin of Rosice-Oslavany Black Coal Basin in forest-arable hilly land at average elevation 433 m (340 -480 m) a.s.l. (Fischer 2000). The road from Rapotice to Oslavany passing through the village makes a frontier of nature reservation Oslava. The right (southwest) half of the village with adjacent pieces of land (fields and forests) belongs to the reservation, the left (northeast) one does not belong to this reservation. The climate is cold and dry, with average annual precipitation below 600 mm and drinking water deficits. The annual mean temperature is 7.5 o C (January -5 o C, July 20 o C).
The average year thickness of snow cover is less than 0.5 m. The air is polluted mostly by thermal powerplant and incinerator in Oslavany (7 km from Ketkovice). There are arable fields with potatoes, alfalfa, turnip, rye, wheat, barley, oat, intensively cultivated meadows, orchards and gardens around the village (Fischer 2000), but also uncultivated soil with weeds. The village is surrounded with spruce (Picea abies, Pinus silvestris) and deciduous (Carpinus betulus, Quercus spp.) forests in a distance 250 -1000 m from the margins of the village. Small brook (Ketkovick˘ potok) rises in the northern margin and flows through the centre of the village to the southeast. Another small brook (Balinka) rises 100 m from north-eastern margin of the village and flows far away to southeast. In the village, the rodents were caught in a house at the eastern margin of the village. The miner family house built in 1932 had three rooms, a shed, a former hen-house, two yards, a small alfalfa field and a small garden. The nearest forest margin is about 1000 m from the house (Fischer 2000). No poultry or domestic animals were kept in this house, but rabbits, poultry, cats, dogs, pigs and goats were kept in neighbouring barns, and a cow-shed with about 200 cows was at the western margin of the village, only 250 m from a forest. A stone marten (Martes foina Erxleben, 1777) lived in loft of the house at this time.

Collection and examination of rodents
Small rodents were caught in snap-traps in the village house and in surrounding forests from July 3,1999 to November 3, 2000. Captured rodents were determined, sexed and dissected. Macroscopic examinations of lungs were performed during dissections and special attention was given to any nodules in lung tissue. Whole lungs were preserved in 10% (v/v) water solution of formaldehyde. They were warmed in the formaldehyde solution to 80 o C for 30 minutes and after cooling to 40 o C prior the examination. Then they were exposed to 2% (w/v) water solution of sodium hydroxide NaOH for 3 h. Compression preparations of small pieces of lungs were examined microscopically at a standard magnification 32 ×. Usually 10 slides were prepared from the whole lungs of one animal. All adiaspores were measured. Only adiaspores with diameters above 70 µm were determined as E. parva var. crescens and counted. Intensity of infection was assessed as low (1 to 9 adiaspores), moderate (10 to 99 adiaspores), high (100 to 999 adiaspores) and very high (1000 or more adiaspores per animal) according to Hub álek (1999).

Plant species composition
Special attention was given to occurrence of the most abundant plants in both habitats of the study area (Faustus and Polívka 1976).

Statistics
Student's t-test was used to evaluate average numbers of adiaspores in animal species and mean diameters of adiaspores (Table 2). Differences in the prevalence of infection (Tables 1 and 3) were evaluated by the Fisher's exact test (Venãikov and Venãikov 1977).
The smallest adiaspore (83 µm) and the largest one (575 µm) were found in one M. musculus captured in the village in August and in a C. glareolus captured in the forest in June, respectively. Intensity of the infections was low to moderate (Table 2). Two moderate infections, 27 and 36 adiaspores per one host, were found in two A. flavicollis captured in the forests in March and April, respectively. A. flavicollis captured in the forests had significantly greater number of adiaspores than A. flavicollis captured in the village (P < 0.01). Average diameters of adiaspores were greater (P < 0.01) in A. flavicollis from the village (282.1 ± 34.5 µm) than in A. flavicollis from the forests (205.9 ± 46.6 µm).

Discussion
Hubálek et al. (1997) examined rodents of nine species in six localities near Ketkovice (quadrat 6863) in the vicinity of Moravsk˘ Krumlov. The most infected species were C. glareolus (20.9 %), A. sylvaticus (11.4 %) and A. flavicollis (11.3 %). They recorded also infection of M. arvalis (11.0 %) and one from five pine voles (M. subterraneus De Sélys-Longchamps, 1836). M. arvalis prefers fields and meadows. Most of individuals entering the house in Ketkovice originated from surrounding fields. Farmland habitats like arable fields, cultivated meadows, orchards and gardens are obviously less optimal than uncultivated habitats (woods, shrubby balks or windbreaks between fields) for the growth of Emmonsia in the soil. This resulted in a lower incidence of rodent emmonsiosis in the agrocenoses (Hubálek et al. 1995a). Growth requirements of saprophytic stage of E. parva var. crescens in nature have not been defined clearly. The most probable place of occurrence of saprophytic stage is wet soil in shady places (H u b álek et al. 1995b, 1998) and lairs of animals, especially rodents (Prokopiã and ·tûrba 1978). Hubálek et al. (1998) found a higher prevalence of the infection in adult rodents from windbreaks than in those from adjacent arable fields, 62.1 % and 8.2 %, respectively. Occurrence of Emmonsia was not influenced by water content and pH values of the soil, but significantly higher mean weight proportion of plant remnants was present in the soil from windbreaks than from fields.
Among 29 most abundant plant species growing in Ketkovice and surrounding forests were ten species occuring in both habitats. C. majus, U. dioica and D. carota var. silvestris are shadow-tolerating herbs (Faustus and Polívka 1976). R. fruticosus, R. canina, P. spinosa, C. avellana and S. nigra are bushes producing large amount of plant remnants and they provide small rodents food (Holi‰ová 1960; A b t 1992) and shelter. R. pseudoacacia and T. cordata are deciduous trees producing plant remnants (Faustus and Polívka 1976). Because these plants grow in both habitats, small rodent populations in village have similar living conditions as in the forests. In accordance with H u b álek et al. (1998), R. canina, S. nigra and T. cordata (growing also in the windbreaks) were found among plant species occurring in both habitats in Ketkovice near Brno. Nuorva et al. (1997) described emmonsiosis in a two-year-old Finnish girl of Caucasian origin. Her mother worked in a large garden shop and she was in contact with soil therefore.
Prokopiã and ·tûrba (1978) infected white laboratory mice (house mouse, M. musculus) by keeping them in a territory previously used by a colony of common voles (M. arvalis) spontaneously infected with E. parva var. crescens. Adiaspores 190 -210 µm in diameter were found in lungs of mice after four months of inhabiting of former vole lairs. Rodents from air-polluted areas were more infected than rodents from non-polluted areas (Jeãn˘ and Vojtûchová 1984;H u b álek et al. 1988).
The highest prevalence of infection occurs in spring and winter (H u b álek et al. 1995b). Although the age of rodents was not assessed in present study, it is well-known that adult animals are more often infected than juveniles (Jeãn˘ and Vojtûchová 1984;Hubálek et al. 1988Hubálek et al. , 1997. Low autumn prevalence can be explained by the fact that rodent populations include many juvenile animals in the autumn (Rajska-Jurgiel 2000).
Emmonsia infection was not influenced by sex of animals. The same results reported Hubálek et al. (1988;. Very low prevalence and intensity of infection in M. musculus can be explained by its synanthropy. Hemisynanthropic A. flavicollis was the most infected rodent species in the village and the second most infected species in the forests. The most infected species in the forests was the exoanthropic bank vole (C. glareolus). Similar relations were observed in mustelid carnivores of the family Mustelidae by Kfiivanec and Otãená‰ek (1977). Whereas exoanthropic pine marten (Martes martes Linnaeus, 1758) and steppe polecat (Putorius eversmanni Lesson, 1827) captured in the Czechoslovakia had a high prevalence, 72.2 % and 70.3 %, respectively, hemisynanthropic stone marten (M. foina) and dark polecat (P. putorius Linnaeus, 1758) had lower prevalence, 37.5 % and 30.6 %, respectively. Suitable living conditions for hemisynanthropic animals were indicated by presence of M. foina in house in Ketkovice during the study period.
No macroscopic pulmonary lesions were found in examined animals, because the intensity of infection was either low or moderate. H u b álek et al. (1988) found in lungs of one short-tailed vole (M. agrestis) from air-polluted area of Kru‰né hory (Ore Mountains, Bohemia) as many as 1130 adiaspores (very high intensity of infection), which undoubtedly influenced the health status of the animal.
Emmonsia infection was found not only in small rodents, but also in larger rodent species such as squirrel (Sciurus vulgaris Linnaeus, 1758) and muskrat (Ondatra zibethicus Linnaeus, 1758) in the Czech Republic (Kfiivanec et al. 1976;H u b álek 1999) and beaver (Castor fiber Linnaeus, 1758) in Sweden (Mörner et al. 1999). Kfiivanec et al. (1976) found infection in lungs of 36 (20.5 %) from 176 squirrels captured in many various parts of the Czechoslovakia. The diameters of adiaspores varied from 150 to 600 µm, but most frequent were diameters of 400 -500 µm. Adiaspores found in the beaver by Mörner et al. (1999) were 100 -200 µm large.
The size of adiaspores indicates a probable time of infection of the host. Adiaspores of E. parva var. crescens reach a diameter of 130 -230 µm within one month, that of 220 -420 µm in two months after infection (H u b álek et al. 1988). According to these data, most of rodents were infected probably in autumn or winter.
Emmonsiosis is a neglected and underdiagnosed disease, because there are no suitable diagnostic methods. Small nodules can be overlooked or mistaken for other pathomorphological changes, for instance miliary tuberculosis (Koìousek et al. 1971;Johnstone et al. 1993). Cultivation of Emmonsia spp. is difficult (Dvofiák et al. 1973;Kfiivanec and Otãená‰ek 1977) and serological dignostic methods are complicated by cross reactions with other soil fungi (H u bálek et al. 1998). Adiaspores can be easily demonstrated by histological staining methods. They are large, with a typical structure (Halouzka et al. 1989), and can be well stained with gallocyanine blue (Koìousek et al. 1971), PAS (Ooi and Lin 1996), Grocott (Jeãn˘ and Vojtûchová 1984;de Montpréville et al. 1999;Mörner et al. 1999) and hematoxylin and eosin (de Montpréville et al. 1999;Mörner et al. 1999). However, histological examinations are time consuming and expensive. Also reliable method of compression preparations of lungs is not performed at large scale.There is almost no information available about Emmonsia infection in game, pets, laboratory, domestic and farm animals therefore.
Village areas provide suitable living conditions for saprophytic stage of E. parva var. crescens and for possible transmitters of adiaspores, small rodents, and risk of the infection for humans and animals is not limited only to the forests.

SCREENING FOR PENICILLIN PLASMA RESIDUES IN CATTLE BY
In this study, we established a rapid prediction test for the detection of the cattle with violative tissue residues of penicillins. The recommended therapeutic doses of two penicillins, ampicillin (withdrawal time, 6 days) and amoxicillin (withdrawal time, 14 days), were administered to two groups of 10 cattle each. Blood was sampled and tested before drug administration and during the withdrawal period. The concentration of penicillins in plasma, determined by a semi-quantitative ELISA, was compared to that of internal standard (4ppb as penicillin G). The absorbance ratio of internal standard to sample (B/Bs) was introduced as an index to determine whether drug residues in cattle tissues are negative or positive. That means B/Bs ratio lower than 1 was considered residue positive and that higher than 1 negative.
All 10 plasma samples from non-treated cattle showed negative results for both penicillins. Both penicillins were detected in plasma samples of cattle treated until the 3rd day of withdrawal period.
The present study has shown that the semi-quantitative ELISA could be easily adapted for prediction of screening plasma residues for penicillin antibiotics (ampicillin and amoxicillin) in live cattle.

Penicillin ELISA, plasma, cattle contamination, live animal test
With the ever-growing world population, animal production practices have become more intensive and efficient, accompanied by increasing demands for drug treatment. Currently, approximately 80% of all food animals receive medication for part or most of their lives (Sternesjö et al. 1998). In the near future, nearly all animals bred in the world for food will receive chemotherapeutic and prophylactic agents of some type (Booth 1988). A survey of all violative carcasses in the United States in 1993 revealed that the most frequent drug residues were penicillin (20%), streptomycin (10%), oxytetracycline (10%), and sulfamethazine (9%) (Paige 1994). According to Canadian Animal Health Institute, penicillins were the most frequently detected residues in milk in most countries (Heeschen et al. 1996). Since 1986, Department of Veterinary Service, Ministry of Agriculture & Forestry, Korea has conducted National Residue Program (NRP) to sample meat and poultry for residue tests at the slaughtering establishments under its inspection authority and from import shipments at the port of entry. In 1997, a total of 45,000 samples comprising 10,000 beef, 23,000 pork, and 11,000 poultry meat were analyzed for five kind of antibiotics (penicillins and tetracyclines) and six sulfonamides. The results showed violative residues of tetracyclines, sulfonamides and aminoglycosides in beef, pork, and poultry meat.
A few cases of minor allergic reactions (e.g., skin rashes) in individuals previously sensitized to penicillin G residues in milk and meat have been documented, as well as strong evidence linking a widespread agricultural use of antibiotics to an increase in antibiotic resistance among the animal and human pathogens (Dewdney et al. 1984;Franco et al. 1990;H u b e r 1971;Kindred et al. 1993;Mitchell et al. 1995;Ormerod et al. 1987).
The demands for reliable, simple, sensitive, rapid and low-cost methods for detecting residues in foods continue to grow (M itchell et al. 1998;). Variety of enzyme immunoassays have been developed and adopted for detecting the generic groups of chemical residues in milk, urine, blood, and meat samples (Gardner et al. 1996;Szekacs 1994;Lee et al. 2000;. Enzyme-linked immunosorbent assay (ELISA) has become the most popular method for chemical residue detection in food due to its extreme sensitivity, simplicity, and ability to screen large number of samples (Clifford 1985;Gardner et al. 1996;Szekacs 1994;).
In the present study, we developed a live animal test to predict the tissue residues of penicillins (ampicillin and amoxicillin) in cattle by examining the concentration of drug in blood during the withdrawal period obtained by an ELISA technique.

Drug administration and samples
Ampicillin was administered intramuscularly to each of the 10 cattle at the rate of 11 mg per kg body weight per day for seven consecutive days, and amoxicillin twice (24 h interval) intramuscularly to each of 10 cattle at 15 mg per kg body weight. Blood samples were collected from all cattle before administration of the drugs and on days 1, 3, 5, 6, and 10 after the last ampicillin injection. From the cattle treated with amoxicillin, blood samples were collected on days 1, 3, 7, 10, and 14. Ten ml of blood from each cattle were collected in heparinized tubes and centrifuged at 4500 × g for 10 minutes to collect the plasma.

Preparation of standard curves
Stock standard solution of 1000 µg/ml of each ampicillin and amoxicillin were prepared using USP standards in saline. These stock solutions were further diluted with saline or blank serum to prepare 0, 1, 2, 5, 10, 20, 50, 100, 500, and 1000 µg/ml working standard solutions. Standard curves of each antibiotic were constructed using the standard solutions fortified into serum to determine the detection limit for the ELISA kit.

Analysis of penicillins in plasma
ELISA tests for β-lactams were applied to each plasma sample in duplicate using a modified methodology described by Boison et al. (1995), in which the manufacturer's protocol for milk screening was adapted for plasma screening. Briefly, 250 µl of the internal standard solution (equivalent to 4 ppb penicillin G) was pippetted into a test tube containing immobilized β-lactam antibodies. The plasma (250 µl, diluted 1 : 10 w/PBS) was pippetted into individually labeled tubes. An equal volume of tracer solution (enzyme conjugate, lyophilized horseradish peroxidase labeled β-lactam conjugate with preservative) was added, and the test tubes were incubated at room temperature for 3 minutes with continuous shaking. The excess sample and conjugate reactants were then washed out with saline. A colour developer (0.5 ml, enzyme substrate) made up of 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) and hydrogen peroxide in citrate buffer was added to the test tubes, and the mixture was incubated at room temperature for 3 minutes with continuous shaking. Dilute sodium dodecyl sulfate solution (0.5 ml) was added to each test tube to stop the reaction. The absorbance was read at the wavelength of 405 nm with a photometric detector (Idetek Reader, Awareness Technology, Inc., USA, operated in the 0.9 ratio mode) and compared with that of the internal standard (4 ppb). Samples with absorbance higher than that of the internal standard were considered to be negative (β-lactam drug free), and those with absorbance lower than that of the internal standard were considered as positive. In this analysis, no more than 5 samples were processed simultaneously, and the assay was completed within 10 minutes (Boison et al. 1995;Cullor et al. 1994).

Standard curves and detection limits
The standard curves of ampicillin and amoxicillin were constructed to determine the detection limits of each drug. The detection limits of ampicillin and amoxicillin were found to be lower than 1 ppb based on the B/Bo ratio of 0.8 in the ELISA system ( Figs. 1 and 2).

Live animal test for penicillins in plasma
Ampicillin. Results of plasma analysis are shown in Table 1. As the absorbance ratios of normal 10 cattle of the control group were higher than 1.0, that is, the concentrations of ampicillin in the diluted plasma (× 10) of this group were higher than 4 ppb, the control group was negative. On day 1 of withdrawal, 8 of the 10 samples were found positive. The number of positive samples on day 3 was 5. All samples showed negative reaction after day 5 of withdrawal (B/Bs ratio ≥1.0).
Amoxicillin. Results of plasma analysis are shown in Table 2. As the absorbance ratios of normal 10 cattle of the control group were higher than 1.0, that is, the concentrations of amoxicillin in the diluted plasma (× 10) of this group were higher than 4 ppb, the group tested negative. All samples tested positive on day 1 of withdrawal. On day 3, 4 of the 10 samples were positive. After day 5 of withdrawal, all samples showed negative reaction (B/Bs ratio ≥ 1.0).   Table 1 Depletion profile of ampicillin in plasma during withdrawal period * Blood was collected before administration of ampicillin. The drug was administered intramuscularly with 11 mg/kg body weight once daily for seven consecutive days, and blood samples were collected from cattle during the withdrawal period. Concentration of ampicillin in plasma was analyzed using a LacTek ELISA kit. B is absorbance of sample and Bs is absorbance of the internal standard (4 ppb). B/Bs ratio lower than 1.0 is considered positive and that higher than 1.0 negative.

Discussion
To prevent unwanted drug residues from entering the human food chain, both the government authorities and the industries have established extensive control measures (Sternesjö et al. 1998). A variety of rapid screening tests have been developed and applied for determining drug contamination of animal products on farms and slaughterhouses. The Swab Test On Premises (STOP), a nonspecific microbial inhibition test, has been used in abattoirs in the United States and Canada for over 10 years to screen for antibiotic residues in tissues from slaughtered animals (Korsrud et al. 1998). The test requires overnight incubation, and results are not ready until the following day. Sweeney et al. (1993) developed a model with urine of pigs to predict the number of days for sulfamethazine concentration to fall below 0.1 ng/g of tissue residues in various organs. This prediction model provided the practical basis for current Sulfa On Site (SOS) test in which swine urine is used for screening sulfonamide residues in animal tissue in federally inspected abattoirs of the United States, Canada, and Korea. With the correlation between residue levels in tissue and urine established, the urine residue is used as an indicator of sulfamethazine in animal tissue. Though, unlike STOP and the Live Animal Swab Test, the SOS test provides same-day results, and it detects only sulfonamides. Papich et al. (1994) conducted experiments to determine whether penicillin residues in the plasma of live animals can be used as a practical indicator of penicillin residues in tissues of food-producing animals at slaughter. According to the results, penicillin G in the plasma did not correlate with that in tissues. To determine whether commercially available rapid tests can be used as screening tests to indicate the presence of penicillin G in the plasma of live animals, Boison et al. (1995) analyzed plasma from healthy steers injected with procaine penicillin G only, and a combination using benzathine penicillin G with four commercially available tests (Brilliant Black reduction test, LacTek test, Charm Farm test, and Charm Test II receptor assay). When results of the four rapid tests were compared with the results of liquid chromatographic method, none of the rapid tests gave false-positive results. With the administered dosage taken into consideration, plasma concentration profiles of penicillin antibiotics in our study were similar to the above studies.
As the withdrawal time of a drug is established based on the tolerance level in tissue and elimination rate of the drug, and blood is a central pool of drug distribution to body compartments and elimination from tissues through biological fluids (Booth 1988), it may be help to predict the residue of drugs in tissue by examining the blood drug depletion profile Table 2 Depletion profile of amoxicillin in plasma during withdrawal period * Blood was collected before administration of amoxicillin. The drug was administered twice intramuscularly with 15 ?/kg body weight at intervals of 24 h and blood samples were collected from cattle during withdrawal period. Concentration of amoxicillin in plasma was analyzed using a LacTek ELISA kit. B is absorbance of sample and Bs is absorbance of the internal standard (4 ppb). B/Bs ratio lower than 1 is considered positive and that higher than 1 negative. during withdrawal period Boison et al. 1995;Lee et al. 2000;L e e et al. 2001). According to our results, the developed methods can be adapted easily to predetect residues of penicillin antibiotics (ampicillin and amoxicillin) in live cattle using diluted blood plasma (× 10) with the modified ELISA test kits. It is conceivable that the veterinary inspector in the abattoir may be able to use this method to screen for penicillin antibiotics in plasma from live cattle in holding pens prior to slaughter and obtain same-day results. Cattle that show positive can then be held in the pens until retest results come up negative before they are slaughtered.