 

 

. bpr‘ \ 5: 5‘:

4.3.74 "

‘~\.». 1': .' "
H" ’0

14‘3”?”

:4...”

x:'€,

’43"! {a}. .3 g; I}.
if ..

4
.l".l“

-l~
“Ag.
J Fr": .'

." ,- »r I
. ,3 W: ’3?" "I33? a:
@jﬁvfn'érru .414 73511"? ‘
{‘3'}?in I.“ {:51
. .vi _-‘, 'j-mf
1,4' 5??

 

4 .“
Risk

 

.§
#4:!"
”f.’ “1‘"

Indium-“Ln.

'4’. .‘.f;',';' '1')”.
II}?! “'3

V .
.rr‘l' {JUN
5+" '45“

.‘.,
19:35
‘ ' \

4

l

v
a:

I
,s

l
D"

I36.

 

.. ”in

21/5001 77

MI IICH IGAN STATE IBRA ARI IES
" 43312;: ”I

I III II II I IIIIIIIIIIIIIII IIII III II

063 3247

Ll:RARY
Michigan State
University

 

 

 

 

 

 

 

 

 

 

 

IIII

 

 

 

This is to certify that the

dissertation entitled

INTRODUCTION OF ANTIBIOTIC RESISTANT
MICROORGANISMS FROM CHICKENS T0 MINK

presented by

Michael Alan McKinney

has been accepted towards fulﬁllment
of the requirements for

_Ph.D. degree“, Avian Microbiology

 

 

0-12771

 

 

 

'bv1531—J RETURNING MATERIALS:
Piace in book drop to
”ﬁndings remove this checkout from
.—:—. your record. FINES W‘iII

be charged if book is
returned after the date
stamped beIow.

 

 

 

 

‘JAN 1 019%

APR 23 200E?"

 

 

 

INTRODUCTION OF ANTIBIOTIC RESISTANT
MICROORGANISMS FROM CHICKENS TO MINK
BY

Michael Alan McKinney

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Poultry Science

1988

ABSTRACT

INTRODUCTION OF ANTIBIOTIC RESISTANT
MICROORGANISMS FROM CHICKENS TO MINK

BY

Michael Alan McKinney

The use of antibiotics for growth promotion in food
animals has raised serious health concerns about the
increase of antibiotic-resistant, pathogenic bacteria due
to this practice. An increase in penicillin-resistant
bacteria in the human intestinal tract would make it
difficult to treat a bacterial infection in the intestinal
tract with penicillin.

Approximately one-half of all antibiotics made in the
0.8. are used in animal husbandry. Evidence indicates that
antibiotic-resistant bacteria increase in an animal's
intestinal tract when a growth promotant level of an
antibiotic is fed. The concern is that the
antibiotic-resistant bacteria will pass this resistance to
bacteria found in the intestinal tract of humans when they
eat antibiotic-fed animals.

The experiments presented in this dissertation were
conducted to study the effects on the intestinal microflora
of chickens after penicillin was fed at two different
levels,lO ppm or 100 ppm. Penicillin-resistant organisms
were observed in the intestinal microflora of the chickens

before and after feeding of the penicillin. The viscera of

Michael Alan McKinney

these chickens were fed to mink that had not been exposed
to penicillin in the recent past. Changes in the bacterial
counts of the feces of mink on different treatments were
noted. Penicillin resistance was monitored in the feces of
the mink.

The results indicate that the feeding of penicillin at
10 or 100 ppm to broilers raised in a clean environment did
not improve feed conversion or increase the level of
penicillin-resistant bacteria in the jejunum or cecum of
the broilers. The penicillin-resistant, aerobic bacteria
declined within three days after penicillin was withdrawn
from the feed of the chickens. The penicillin-resistant,
anaerobic bacteria did not decline within three days after
the withdrawal of penicillin. of the penicillin from the
broiler feed. The bacterial counts made on the feces of the
mink fed diets containing the viscera of penicillin-fed
chickens indicated that there was a significant increase in
the percent of penicillin-resistant organisms as compared
to the control group.

Similar plasmids were found in penicillin-resistant
bacteria isolated from an experimental mink diet and from
the feces of a mink. This indicates that a transfer of this
plasmid may have occurred in the intestinal tract of the

mink.

TO THE ONLY WISE GOD,

OUR SAVIOR... JUDE l:25(a)

1v

ACKNOWLEDGMENTS

I would like to thank Dr. Timothy S. Chang for his
encouragement and help in completing my degree. I would
also like to thank him for the time he spent with me
teaching me not only laboratory technic, but how to live
and work with people.

I would like to thank my committee members; M. Yokoyama,
R. Aulerich, R. Ringer, and J. Pestka. for their help with
my research and the writing of this dissertation.

I would also like to thank Dr. Hogberg the Chairman of
the Animal Science Department for the financial help that
he has provided to me.

The help of Tammy Myers and her enthusiasm for my
research is certainly appreciated.

The facilities and advice of Dr. Paul Coussens were of
invaluable help to me.

I am also grateful to Dr. Richard Balander for the time
he spent in helping prepare the slides for my presentation.

I want to thank the Reverend David Stephens, the
Reverend John Colegrove, and all the people who have prayed
for me during this ordeal.

Also, I want to thank my mother—in—law, Wilma Goers, for

her help in typing this dissertation.

I want to thank my mother, Fern, for her encouragement
and financial help.

Finally, I would like to acknowledge my wife, Kathy, and
my children, Sarah and Emily, for the many sacrifices that

they made in their lives so that I could finish this degree.

vi

TABLE OF CONTENTS

LIST OF TABLES ................................. ...... ix
LIST OF FIGURES .................. ......... .. ......... xiii
INTRODUCTION . ............. . .......................... l
OBJECTIVES............................................ 4
LITERATURE REVIEW .................. ........ .. ........ 6

Intestinal microflora of chickens ..... ........... 6
Digestive tract of the mink ............ .......... 9
The function and use of penicillin.......... ..... . 10
Resistance to antibiotics .................. ...... l3
Spread of resistance factors ............... ...... 18
Studies on bacterial plasmids . ................... 21

MATERIALS AND METHODS 0............OOOOCOOOOOOOO...... 23

Birds ........ ....... ......... ...... ... ........... 23
Mink ............................................. 26
Sampling ......................................... 29
Microbiological methods .......................... 30
Growth media used ................................ 32
Plasmid identification .. ....... ......... ........ . 35

RESULTS ......O.........OOOOOOOOOOOO ...... 0.... ....... 4O

Experiment one ............................. ..... . 40
Experiment two ................................... 43
Experiment three ................................. 55
Experiment four ............. ..... . ............... 65

DISCUSSION ........................................... 74
Experiment one ... ...... . ...... ................... 74
Experiment two 0 O O O O O O O O O O O O O O O O O O O 0 O O O O O O O O O O O 0 O O 7 5
Experiment three 0 I O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O 7 8
Experiment four 0 O O O O O O O O O O O O O O O O O O O O 0 O O O O O O O O O O O 0 8 0

CONCLUSIONS ................................ ..... ..... 82

v11

APPENDIX A, Formulation of M80 broiler feeds ......... 84
APPENDIX B, Solutions used for microbiological methods 85
APPENDIX C, Solutions used for plasmid identification 86

BIBLIOGRAPHY ......OOOOOOOCC......OOOOOOOOOO. ......... 87

viii

LIST OF TABLES

Table 1. Average body weight of broilers fed
different levels of penicillin. ............... .......

Table 2. Analysis of variance of the body weight of
broilers fed different levels of penicillin. .........

Table 3. Feed conversion of broilers fed different
levels of penicillin. ................................

Table 4. Analysis of variance of the feed conversions
of broilers fed different levels of penicillin. ......

Table 5. The effect of feeding penicillin to broilers
on total, aerobic, bacterial counts at two sites in
the broiler intestinal tract. ........................

Table 6. Analysis of variance for total, aerobic
bacteria at two sites in the intestinal tract of
brOilers fed peniCillin. ..........OOOOOOOOOOIOIOOOOOO

Table 7. The effect of feeding penicillin to broilers
on total, coliform counts at two sites in the broiler
intestinal tract. ...0.00.00.........OOOOOOOOOOOOOOOOO

Table 8. Analysis of variance for total coliforms at
two sites in the intestinal tract of broilers fed
peniCillin. ...0......00............OOOOOOOOOOOOI...0.

Table 9. The effect of feeding penicillin to broilers
on total, anaerobic, bacterial counts at two sites in
the broiler intestinal tract. ........................

Table 10. Analysis of variance for total, anaerobic
bacteria at two sites in the intestinal tract of
brOilers fed penicj-llin. .....OOOOOOOOOO00.0.00...O...

Table 11. The effect of feeding penicillin to broilers
on lactobacilli counts at two sites in the broiler
intestinal tract. .................OOOOOOOOOOOOOOO....

Table 12. Analysis of variance for lactobacilli at two
sites in the intestinal tract of broilers fed
peniCillin. ...0........IOOIOOOOOOOOOOOOOOO ...... 0....

.4-
#4

41

41

42

42

44

44

46

46

48

48

49

49

Table 13. The effect of feeding penicillin to broilers

on Clostridium perfringens counts at two sites in the

 

broiler intestinal tract. ............................

Table 14. Analysis of variance for Clostridium
perfringens at two sites in the intestinal tract of

 

 

broilers fed penicillin. .............................

Table 15. The effect of feeding a diet containing the

viscera of

penicillin-fed chickens to mink on total,

aerobic, bacterial counts found in the feces of mink.

Table 16. Analysis of variance for total, aerobic
bacteria found in the feces of mink fed a diet

containing
penicillin.

the viscera of chickens that were fed

Table 17. The effect of feeding a diet containing the
viscera of penicillin-fed chickens to mink on total,
coliform counts found in the feces of mink. ..........

Table 18. Analysis of variance for total coliforms
found in the feces of mink fed a diet containing the

viscera of

chickens that were fed penicillin. ........

Table 19. The effect of feeding a diet containing the
viscera of penicillin-fed chickens to mink on total,

anaerobic,
mink. .0...

bacterial counts found in the feces of

Table 20. Analysis of variance for total, anaerobic
bacteria found in the feces of mink fed a diet

containing
penicillin.

the viscera of chickens that were fed

Table 21. The effect of feeding a diet containing the
viscera of penicillin-fed chickens to mink on
lactobacilli counts found in the feces of mink. ......

Table 22. Analysis of variance for lactobacilli found

in the feces of mink fed a diet containing the viscera

of chickens that were fed penicillin. ................

Table 23. The effect of feeding a diet containing the
viscera of penicillin-fed chickens to mink on
Clostridium perfringens counts found in the feces of

 

mink. 0....

Table 24. Analysis of variance for Clostridium
perfringens found in the feces of mink fed a diet

 

containing
penicillin.

 

the viscera of chickens that were fed

X

50

50

52

52

53

53

54

54

56

56

57

57

Table 25. The percent of aerobic microorganisms
resistant to penicillin isolated from the jejunum of
chickens fed different levels of penicillin. .........

Table 26. Analysis of variance for the
penicillin-resistant, aerobic microorganisms isolated
from the jejunum of chickens fed different levels of
penicillin. ....... ....... ................ ..... . ......

Table 27. The percent of aerobic microorganisms
resistant to penicillin isolated from the cecum of
chickens fed different levels of penicillin. .........

Table 28. Analysis of variance for the
penicillin-resistant, aerobic microorganisms isolated
from the cecum of chickens fed different levels of
penicillin. ...................... ....... .. ..... . .....

Table 29. The percent of anaerobic microorganisms
resistant to penicillin isolated from the jejunum of
chickens fed different levels of penicillin. .........

Table 30. Analysis of variance for the
penicillin-resistant, anaerobic microorganisms
isolated from the jejunum of chickens fed different
levels of penicillin. . .......... ...... ...............

Table 31. The percent of anaerobic microorganisms
resistant to penicillin isolated from the cecum of
chickens fed different levels of penicillin. .........

Table 32. Analysis of variance for the
penicillin-resistant, anaerobic microorganisms
isolated from the cecum of chickens fed different
levels of penicillin. ........................... .....

Table 33. The percent of aerobic microorganisms
resistant to penicillin isolated from the feces of
mink that were fed the viscera of penicillin-fed
chickens. ............. ..... .. .......... . .............

Table 34. Analysis of variance for the
penicillin-resistant, aerobic microorganisms isolated
from the feces of mink that were fed the viscera of
penicillin-fed chickens. .............................

Table 35. The percent of anaerobic microorganisms
resistant to penicillin isolated from the feces of
mink that were fed the viscera of penicillin-fed
chickens. .... ....... . ..... .................. .........

59

59

60

62

62

Table 36. Analysis of variance for the
penicillin-resistant, anaerobic microorganisms
isolated from the feces of mink that were fed the
viscera of penicillin-fed chickens. ..................

Table 37. The amount of penicillin found in mink diets
containing viscera of penicillin-fed chickens. .......

Table 38. The percent of different aerobic,
penicillin-resistant, bacterial genera randomly
selected from the total, penicillin-resistant,
bacterial population. ................................
Table 39. Band sizes of two plasmids cut with EcoRI.

Table 40. Band sizes of two plasmids cut with BglII.

xii

66

67

67
72

72

LIST OF FIGURES

Figure 1. Photograph of an agarose gel electrophoresis
of two different plasmids and a standard plasmid. ....

Figure 2. Photograph of an agarose gel electrophoresis
of two plasmids cut with EcoRl. . ................. ....

Figure 3. Photograph of an agarose gel electrophoresis
of two plasmids cut with BglII. ....... ..... ..... .....

x111

69

7O

INTRODUCTION

Penicillin was discovered in the late nineteenth
century, although a practical use for it was not discovered
until 1927 (Betina, 1983). From 1927 until the early
1950's, penicillin was used only for treating bacterial
infections. At this time, it was discovered that the
feeding of low levels of penicillin to chicks caused an
increase in their body weight without increasing their feed
consumption (Heuser and Norris, 1952).

In 1951, only 110,000 kilograms of antibiotics or 16% of
all antibiotics produced were used in animal production.
The percentage of antibiotics used in all phases of animal
production has steadily increased until 1980 when almost
half (approximately 5.58 million kilograms) of all
antibiotics produced in this country were used for food
animals (Stallones gt 51., 1980; Levy, 1987). The
percentage use of antibiotics for food animals will
probably continue to increase. The concern of the public in
using antibiotics for livestock is the build up of
antibiotic-resistant bacteria in the intestinal tract of
animals which then may be spread to humans (Hays, 1981).

The use of antibiotics has lead to an increase of

antibiotic-resistant organisms in the intestines of farm

1

animals (Crawford and Shotts, 1982). If these
antibiotic—resistant organisms are passed from farm animals
to humans then a situation could arise where a person could
not be treated effectively with an antibiotic because the
pathogenic bacteria are resistant to the antibiotic. The
spread of antibiotic-resistant organisms from chickens to
chickens and from chickens to man has been demonstrated
(Levy gt al., 1976b), although the resistant organisms
appeared to be transient in nature in the second species of
animal.

The British government ordered a study on the use of
antibiotics in animal husbandry (Braude, 1978). Based on
this report the use of subtherapeutic levels of antibiotics
of human importance was prohibited in animals. The
percentage of organisms in animals that were resistant to
antibiotics was as great or greater four years after the
law had been passed. It has also been noted that the amount
of antibiotics used in the United Kingdom did not decline
after the passing of the law. This indicates that farm
animals may still be given the antibiotics illegally.

The United States, specifically the Food and Drug
Administration (FDA), gave notice in 1977 that it intended
to recommend to the 0.8. legislature to restrict or ban the
subtherapeutic use of penicillin and the tetracyclines for
growth promotion in animal feeds (Hays, 1981). This notice

still has not been presented to the U.S. legislature

because the FDA scientists and other research scientists
have not demonstrated that the antibiotics in question are
ineffective or unsafe. At this time, a law banning or
restricting the use of penicillin or the tetracyclines in
this country would be based on emotions and not on
scientific facts.

The spread of antibiotic resistance from animals to man
has been noted (Cohen and Tauxe, 1986). This transfer seems
to happen to people that are in contact with the animals
(Levy gt gt., 1976a). Holmberg gt gt. (1984) implicated the
transfer of antibiotic resistance from animal to man by the
consumption of hamburger, but direct transfer of the
antibiotic resistance was not proven. Gast and Stephens
(1988) demonstrated the transfer of antibiotic-resistant
bacteria from turkey to rats fed the turkey only when
antibiotics were being administered to the rats.

More research needs to be conducted on the transfer of
antibiotic resistance through the food chain. This research
study was designed to show if antibiotic-resistant bacteria
can be transferred from contaminated meat to the animal

which eats the meat in the absence of antibiotics.

OBJECTIVES

1. To determine if penicillin, when fed to broilers at 10
or 100 ppm in the feed, improved feed conversion when the

broilers were raised in a clean environment.

2. To determine whether the feeding of penicillin to
broilers at 10 ppm and 100 ppm changed the number of total
aerobic bacteria, total coliforms, total anaerobes,

lactobacilli, or Clostridium perfringens in the jejunum or

 

in the cecum of the broilers.

3. To elucidate the number of total aerobic bacteria,

coliforms, total anaerobes, lactobacilli, and Clostridium

 

perfringens found in the feces of mink.

 

4. To determine if the number of penicillin-resistant
bacteria decreased in the jejunum or cecum of broilers
within three days after penicillin was removed from the

feed.

5. To determine if the percent of penicillin—resistant
bacteria present in the feces of mink increased when the
mink were fed diets containing viscera of broilers that

were fed penicillin.

6. To determine if similar plasmids carrying
penicillin—resistant genes could be found in both the diet
of mink containing the viscera of chickens that were fed
penicillin and in the feces of the mink that ingested this

diet.

LITERATURE REVIEW

Intestinal microflora of chickens

Although the intestinal microflora of chickens has been
studied for years, the types and numbers of bacteria that
compose the microflora have not been completely elucidated
(Barnes gt gt., 1972; Beck, 1978). One of the reasons that
the microflora has not been elucidated is the tremendous
variation in the results of the research that has been
presented. One source of the variation of the types and
numbers of the bacteria reported is due to the sampling
procedure. For example, some research reported the numbers
of bacteria found in a fecal sample. This type of sample
gave an inaccurately low count of nonsporing, obligate
anaerobes due to the exposure of the bacteria in the feces
to air at the time of defecation. Extremely high counts of
coliforms were found in fecal samples due to the
reproduction of the coliform bacteria after the voiding of
the feces (Barnes and Goldberg, 1962). Also, a fecal sample
only indicates the bacteria present in the colon and not in
the small intestine or in the ceca of the bird. Another
problem that is encountered is deciding what the sample to

be analyzed should contain. Some studies used only

/
O

intestinal contents for enumeration and identification of
the bacteria (Salanitro gt gt., 1974) while other studies
used fecal samples (Barnes and Goldberg, 1962). Also, a
problem that was encountered is the length of time that a
sample is held prior to being plated onto growth media.
Reproduction of some of the sample bacteria or death of
other sample bacteria may have occurred during this period
(Barnes gt g1., 1972). Finally, the diet an animal was fed
and the time from the last meal until the animal was killed
influenced the kind and numbers of bacteria present in the
intestine (Smith, 1965). Due to these problems only a very
rough estimate has been made of the numbers of the
different bacteria found in the intestine.

Smith (1965) found that the bacterial population of a
fecal sample closely resembled the bacterial population in
the large intestine of the animal, although large
variations in bacterial population occurred in the stomach
or -small intestine with no noticeable change in the
bacterial population found in the fecal sample. It was also

reported by Smith (1965) that Escherichia, Clostridium,

 

 

Streptococcus, Lactobacillus, and Bacteroides were the most

 

 

 

common organisms in the small intestine and ceca. He found
that approximately one third of the birds had yeast present
throughout their intestine. Levy gt gt. (1976a) found that

Escherichia, Proteus, and enterococci were the predominant,

 

aerobic organisms, although the relative numbers of each

7-a

type of bacteria varied greatly. The numbers of the
different anaerobic organisms isolated from cecal drOppings

indicated that Lactobacillus, Clostridium, and Bacteroides

 

 

were the predominant, anaerobic, bacterial genera (Barnes
and Goldberg, 1962; Smith, 1965).

Escherichia coli was found to make up approximately 80

 

to 90 percent of all the enterobacteria in the chicken

intestine (Levy gt gl., 1976a; Mamber and Katz, 1985). E.

ggtt counts in the anterior region of the small intestine

ranged from 105 to 108 viable organisms per gram of sample
9

and ranged from 105 to 10 in the posterior region of the

small intestine. Clostridium perfringens counts ranged from

 

4 .
nondetectable to 10 organisms per gram of contents in the
anterior small intestine, but only from 102 to 105 in the

posterior portion of the small intestine. Total Clostridium

 

gpp. counts ranged from 102 to 16+ . Streptococcus gpp.

 

numbers ranged from 105 to 108 bacteria per gram with the

higher number of streptococci being counted during the

first three days of life and then decreased to the 103 to

105 level at two to three weeks. The lactobacilli were

found to be a slow colonizer in the small intestine with no

organisms being detected for the first two days after
6 8

hatching and then they increased to 10 to 10 organisms

per gram at two to three weeks after hatching. Bacteroides

 

was not detected at any time in the anterior or posterior

portions of the small intestine (Smith, 1965).

The number of bacteria were found to be greater in

number in the ceca with 10 to 10 E. coli per gram of
u
cecal contents. Streptococci gpp. counts ranged from 10 to

8

 

10 , and Clostridium perfringens numbers ranged from 102 to

108. The total Clostridium gpp. counts varied from 105 to

 

10 . The lactobacilli in three week-old birds stabilized

9

6
between 10 to 10 . The Bacteroides counts ranged from

 

nondetectable to 109 organisms per gram of contents. The
strict anaerobes appeared to take one to three days for
them to colonize in the ceca and then they increased
quickly (Barnes and Goldberg, 1962; Smith, 1965; Barnes gt
gt., 1972).

Changes in the intestinal microflora of chickens were
seen when the birds were fed rations supplemented with
penicillin. Mamber and Katz (1985) found that there was a
slight decrease in the number of E. ggtt with an increase

of Klebsiella when broiler chicks were fed a ration

 

containing 50 grams per ton of penicillin. When mice were
fed penicillin, at any level, there was a large decline of
lactobacilli accompanied by an increase in both
Gram-negative rods and enterococci (Dubos gt gt., 1963).
However, Lev gt gt. (1957) found that no changes occurred
in the number of lactobacilli, streptococci, or coliforms
when broiler chicks were fed a diet containing penicillin
at a level of 45.5 mg per kg of broiler feed. It was noted

that Clostridium pgrfringens was reduced to nondetectable

 

levels in birds when fed this diet. Anderson gt gt. (1951)
found that penicillin increased the number of coliforms and
lactobacilli while it reduced the number of enterococci.
Penicillin, when fed at 100 ppm to turkeys, was found to
increase the total number of anaerobic bacteria while it

significantly reduced the number of Clostridium gpp. This

 

study also found that the coliform numbers were
significantly reduced (Sieburth gt gt., 1951). Chang and
Murphy (1975) reported that chlortetracycline in the feed
at 50, 100, or 200 ppm had no significant effect on
anaerobe, lactobacilli, coliform, or enterococci numbers in
the chicken intestinal microflora. It was also found that
bacitracin methylene disalisylate significantly reduced the
numbers of anaerobes when it was added to the feed at high
levels (100 or 200 ppm) and also reduced the lactobacilli
and enterococci counts when added at all levels used in the

study.

Digestive tract of the mink

The mink (Mustela vison) is a monogastric carnivore. The

 

digestive system is similar to that of a human in that it
consists of an esophagus, a large stomach, and a small
intestine that is divided into the duodenum, jejunum, and
the ileum. The ileum makes up the last half of the small

intestine. The colon is wider than the rest of the

10

intestine and is approximately five centimeters long. The
mink has no appendix (Smith and Krasulak, 1979).

The basal diet of the mink on a dry weight basis
contains 45% protein. The food passage time of a mink is
approximately three hours (Bleavins and Aulerich, 1980).

The numbers and kinds of bacteria in the intestine of
the mink has not been reported in the literature. Also, the
effect of penicillin on the intestinal microflora of the

mink has not been studied (Aulerich, 1987).

The function and use of penicillin

The antibacterial substance produced by the mold,

Penicillium had been observed and used for treatment of

 

wounds before 1900. Fleming, who wrote his paper on the
observations of penicillin in 1929, is credited with the
discovery of a functional use for penicillin (Betina,
1983). He used it primarily in making selective media for

the isolation of Haemophilus, although he also tested it as

 

a topical powder for inhibiting infections in external
wounds on people. His observations indicated that
penicillin was effective in killing both Gram—positive and
Gram-negative cocci. It was not as effective in killing
Gram-negative rods due to the inability of Fleming to
purify and stabilize the penicillin. The drug was not used

routinely for chemotherapy until 1939 (Stewart, 1965;

11

Betina, 1983). By 1950, enough evidence had been gathered
that showed penicillin increased body weight in chickens
that farmers started using it commercially (Stallones gt gt
., 1980).

Over 11 million kilograms of antibiotics were produced
in the United States in 1979. Approximately one half of
these antibiotics were used in animals, particularly in
subtherapeutic amounts (Stallones gt gt., 1980). Nearly 80%
of all the poultry in this country are given antibiotics at
some point in their life, while only 60 to 75% of all the
other meat animals are fed antibiotics (Hays, 1981).

When a healthy animal is fed a low level of an
antibiotic, such as penicillin, the animal will gain more
weight on the same amount of feed than an animal not fed an
antibiotic (Nelson gt gt., 1963; Combs and Bossard, 1963;
Bird, 1969). There are three theories as to why antibiotics
increase feed conversion. They are:

l. Antibiotics cause a nutrient sparing effect. This
effect may be caused by the antibiotic killing or
inhibiting bacteria that metabolize essential nutrients,
such as vitamins or proteins, in the intestinal tract
before the animal can absorb them. The antibiotic may also
enhance the growth of bacteria that synthesize essential
nutrients needed by the animal. Many researchers doubt that
sparing vitamins would change the growth of poultry because

the known vitamins are provided in excess of dietary needs

12

for poultry, although eliminating bacteria that compete
with the bird for protein may reduce the protein
requirement for maximum growth rate (Bird, 1969).

2. Antibiotics, in low levels, control subclinical
infections by reducing the numbers of infectious or
toxin-producing bacteria. Research has been done that shows
a parallel relationship between improved weight gain in
poultry and the reduction of clostridia numbers in the
intestine by the use of penicillin. Also, ammonia levels
are reduced in animals being fed penicillin. When these
toxic compounds are reduced by the use of antibiotics, the
intestinal lining becomes thinner, similar to the
intestinal lining of gnotobiotic animals. The thinner,
intestinal lining permits an easier and more rapid
absorption of nutrients (Bird, 1969).

3. Antibiotics cause a metabolic change in an animal
(Hays, 1969; Wallace, 1970). The metabolic effect may be
explained by the control of subclinical disease, but
research shows that there is a change in water and nitrogen
excretion in animals being fed antibiotics which may affect
an animal's metabolic rate (Braude and Johnson, 1953).

Many scientists think that the most plausible theory of
the three listed is the elimination of subclinical disease.
The problem with this theory is that if the intestinal
microflora becomes resistant to antibiotics permanently

then the antibiotics should lose their effectiveness as

13

growth promotants. This has not been' demonstrated even
after 30 years of antibiotic use in farm animals. The
average improvement in feed efficiency attributed to
antibiotics in 1952 was three percent. In 1963, the feed
efficiency was still improved by three percent when
penicillin was fed to broilers (Bird, 1969). Other
researchers also indicate that there is still a significant
increase in feed efficiency when penicillin is fed to
animals (Menge, 1973; Dafwang gt gt., 1984).

Penicillin is used in both humans and animals to treat
bacterial infections. It is given both orally and by
injection. Even though penicillin is considered to be a
broad spectrum antibiotic, it is most effective against the
Gram-positive organisms. Penicillin is the antibiotic most
commonly used to treat bacterial infections in animals,
tetracyclines being the second choice (Stallones gt gt.,

1980).

Resistance to antibiotics

Gram-negative organisms are normally more resistant to
penicillin than the Gram-positive bacteria due to
differences in the cell wall. Penicillin exerts its
bactericidal effects by inhibiting a transpeptidase or a
carboxypeptidase in the bacterium, both of which are needed

for the incorporation of peptidoglycan oligomers into the

14

cell wall. Inhibition of the . transpeptidase or the
carboxypeptidase triggers the release of a peptidoglycan
hydrolase which hydrolyzes the covalent bonds between the
peptidoglycans in the cell wall and eventually causes lysis
of the bacterial cell (Tomasz, 1979). Peptidoglycan
molecules are present on both the Gram-positive and
Gram-negative bacteria. The peptidoglycan molecules are
randomly cross-linked in the cell wall of Gram-positive
bacteria while the Gram-negative bacteria have uniform
cross-links between the peptidoglycan molecules in their
cell wall. The Gram-negative bacteria have a highly
complex, outer membrane which retards the penicillin from
reaching the cell wall. They also have beta lactamases
located in the periplasmic space (the space between the
outer wall layer and the peptidoglycan layer). These
enzymes occur naturally in the bacteria and hydrolyze the
penicillin before it binds to its binding site. Even with
all these protective devices, Gram-negative bacteria can be
inhibited in growth by high concentrations of hydrolyzed
penicillin in the cell which causes an inhibition of the
binding sites for the peptidoglycan (Costerton and Cheng,
1975; Neu, 1986).

Besides having natural resistance, bacteria may develop
or acquire a resistance to antibiotics. This resistance may
come from a random mutation or from gene transfer from

another organism and involve changes either in the

15

penicillin, binding sites or in the production of beta
lactamases. Gene transfer may occur by transformation,
transduction, or by conjugation (Maas, 1986).

Transformation happens when a bacterium incorporates, at
special sites, free DNA that is present around the
bacterium into its own genome. The exogenous DNA must have
a homologous segment with some portion of the bacterium's
DNA to be inserted into the genome (Davis and Dulbecco,
1973). Transformation can be induced to happen tg ytttg by
treating bacteria with calcium chloride, adding the plasmid
DNA and then heat shocking the mixture. The transformation
is successful with one DNA molecule out of 10,000 (Maniatis
et al., 1982).

Transduction describes the transfer of genetic material
from one bacterium to another bacterium by a bacteriophage.
This occurs when a temperate phage infects an antibiotic
resistant bacterium. In most cases, this temperate phage
does not cause lysis of the cell, but if it does occur,
occasionally a piece of the bacterial genome is enclosed by
a phage coat. This is an infective phage and may recombine
with an antibiotic sensitive bacterium (Davis and Dulbecco,
1973).

Conjugation is the third and probably the most important
means of antibiotic resistance transfer. For conjugation to
occur, a donor bacterium comes into close proximity to a

recipient bacterium. A tube, called a sex pilus, which is

16

located on the donor bacterium, is connected to the
recipient bacterium. Once the pilus connects the two cells,
one strand of plasmid DNA migrates from the donor cell to
the recipient cell. The two cells then separate with both
cells carrying the plasmid DNA.

Plasmids usually control the conjugation process.
Plasmids are extrachromosomal DNA that are double stranded
and circular. They usually weigh from 40 x 106 to 200 x 105
daltons (Willetts, 1972). This is a small segment of DNA
when compared to the average bacterial chromosome that has
a molecular weight of 2 x 109 daltons. Approximately 33% of
the plasmid's genes are used for the transfer of the
plasmid, 20% of the genes are used for the replication of
the plasmid while the rest of the genes are used for
nonessential functions, such as antibiotic resistance
(Willetts, 1972).

There are two types of plasmids, conjugative and
nonconjugative plasmids. The conjugative plasmid can carry
up to ten antibiotic resistant genes while the
nonconjugative plasmid is very small and carries no more
than one or two antibiotic resistant genes. Both type of
plasmids can be transferred to other bacteria by
transduction or transformation, but the nonconjugative
plasmid cannot be transferred to another bacterium by
conjugation without the involvement of a conjugative

plasmid (Elwell and Falkow, 1986).

17

Resistance to antibiotics, by the intestinal microflora,
developed when animals were fed antibiotics (Kiser gt gt.,
1970; Levy gt gt., 1976b; Dawson gt gt., 1984). This
resistance developed whether an animal was fed a
therapeutic level or a subtherapeutic level of antibiotic.
The evidence seems to indicate that resistance increased to
a high level faster (one to two days) when using a
therapeutic level of an antibiotic than when using a
subtherapeutic level of an antibiotic. A study performed by
Langlois gt gt. (1984) found that 90% of the intestinal
coliforms of pigs became resistant to antibiotics within 35
days of being fed antibiotics, independent of the level of
the antibiotic in the diet. Only 58% of the fecal coliforms
in the control pigs were resistant to antibiotics.

Results vary concerning the length of time that
resistant organisms remain at high levels in the intestine
of animals after all antibiotics have been withdrawn from
the feed. Levy gt gt. (1976b) indicated that high levels of
antibiotic-resistant organisms remained in the intestinal
tract over ten weeks after antibiotics had been withdrawn.
Langlois gt gt. (1984) showed that there was a decline in
antibiotic-resistant organisms immediately after
antibiotics were withdrawn from an animal. This decline
continued until the percent of antibiotic resistant
organisms reached pretrial levels.

Recent research indicates that antibiotic-resistant

18

bacteria do not survive in nature as well as
antibiotic-sensitive bacteria. This is seen by a return to
an antibiotic sensitive state usually within two weeks
after the antibiotic treatment is discontinued in the
animal (Richmond, 1977; Langlois gt gt., 1984). The
inability of the antibiotic-resistant bacteria to survive
as well as the antibiotic-sensitive bacteria may be due to
the size of the resistant bacteria being larger than the
antibiotic-sensitive bacteria. The larger size may lengthen
the reproductive time of the bacteria which would give the
antibiotic-sensitive bacteria an advantage. Also, a change
in the cell wall that reduces the ability of the
antibiotic-resistant bacteria to adhere to the intestinal
lining of an animal may be involved (Nordstrom gt gt.,

1977).

Spread of resistance factors

There have been numerous studies done indicating that
antibiotic-resistant bacteria are spread from animals that
have been fed antibiotics or carry antibiotic-resistant
bacteria to animals that have not been fed antibiotics
(Levy gt gt., 1976b; Levy, 1978; Cherubin gt gt.,1980;
Holmberg gt gt., 1984).

Much of the older research that was done indicated that

antibiotic-resistant transfer did occur in animals

19

naturally, but the research was performed in a manner that
the animal was just a living test tube (Jukes, 1971). After
that statement was made, Levy gt gt. (1976b) showed that
antibiotic-resistant transfer was possible from one group
of animals to another group of animals if both groups of
animals were being fed antibiotics. The spread of the
resistance was facilitated by an animal caretaker. Cherubin
gt gt. (1980) reported that resistance to ampicillin in
1976-1978 dropped to 1965 levels in New York City, in
people, while the ampicillin resistance level increased in
cows in the northern part of New York State during the same
period. Cherubin suggested that this occurred because of
the lack of contact between these two areas. Holmberg gt gt
. (1984) gave the best evidence that antibiotic-resistant
bacteria were transferred between animals. This study
showed that people living 800 miles away from the beef that
they ate had the same antibiotic-resistant plasmid in their
intestinal microflora as the beef cattle raised in the
same area as the cattle that the people ate. The plasmid
was noted only in humans that were on antibiotic therapy. A
similar study was reported by Spika gt gt. (1987). A strain

of antibiotic-resistant Salmonella newport reached epidemic

 

proportions in California. A plasmid carrying the
antibiotic-resistant gene was isolated from this strain.
Examination of hamburger being sold in the area revealed

the presence of the same plasmid in bacteria isolated from

20

the hamburger. When the farms where the cows had been
purchased were inspected, the same plasmid was found in

other Salmonella gpp. This study indicated that antibiotic

 

resistance can be spread from animals to man, but the
authors never stated how long the resistant organisms
remained in the human intestine. It was noted that over 50
percent of the humans that had to be treated for the

Salmonella infection had ingested either penicillin or

 

tetracycline within a month of becoming ill.

Hummel gt gt. (1986) found that only when people came
into direct contact with animals being fed antibiotics did
they become resistant to the antibiotic in question. In
this study, a new antibiotic, nourseothricin, was
introduced into a new geographical area. The antibiotic was
used only for growth promotion purposes in pigs. Before the
use of nourseothricin in this area, no plasmid-mediated
resistance to this drug had been found in this locality.
After two years of use of this antibiotic in pigs, over 100
strains of bacteria isolated from humans were found to be
resistant to nourseothricin due to plasmids. No
plasmid-mediated resistance to this antibiotic was found in
areas not using nourseothricin during this same period of
time. The authors concluded that the feeding of antibiotics
in animals was not important for the increase of
antibiotic-resistant bacteria of importance to humans.

Two different conclusions have been presented by the

21

research reviewed in this thesis. They are that resistance
can be transferred from animals to man by plasmids, Spika
gt gt. (1987), or that little danger exists for the public
when animals are fed antibiotics, Kiser gt gt. (1970).
These two different conclusions may be explained by the
unknown condition of the intestinal microflora in the group
of animals to which the plasmid-mediated resistance is
passed. Any variance from the normal microflora in the
digestive tract can result in an increase of the transfer
of plasmid-mediated antibiotic resistance. Stress or
antibiotic therapy are just two examples of factors that
may cause a change in the intestinal microflora which may
allow the transfer of antibiotic resistance to occur

(Falkow, 1975; Elwell and Falkow, 1986).

Studies on bacterial plasmids

The study of the transfer of bacterial plasmids should
become easier in the future. Many rapid techniques are
being developed which will enable a researcher to screen
many bacteria for similar plasmids, quickly and accurately.
The ability to isolate plasmid DNA in small amounts in
three hours has been developed (Maniatis gt gt., 1982). The
use of simple agarose gels to separate the plasmid DNA
based on size can then be used. If two plasmids are found

to be of similar size then they can be cut with a

22

restriction endonuclease. If this procedure is done using
two different endonucleases and the fragments of the
plasmids resulting from the cuttings are similar, then it
can be concluded that the two plasmids are identical

(Farrar, 1983; Takahashi and Nagano, 1984).

MATER IALS AND METHODS

This research study was conducted in four separate
experiments. The first experiment involved the feeding of
penicillin to broiler chicks. The second experiment was the
feeding of poultry offal, that may have contained
penicillin-resistant bacteria, to mink. The microbiological
analysis of the intestinal microflora of both the mink and
the chickens comprised the third experiment. This
experiment was performed to determine any changes in the
intestinal microflora, such as increases or decreases in
bacterial families, and any change in penicillin-resistant
bacteria due to the feeding of penicillin. Identification,
isolation, and comparison of the plasmids obtained from
penicillin-resistant bacteria of the mink feces and the

chicken digestive tract made up the fourth experiment.

Birds

The 156 birds used in the first experiment were Hubbard
strain male broiler chicks that were purchased from
Fairview Farms, Remington, Indiana at one day of age. The
birds were shipped by mail to Michigan State University.

The chicks were taken to the M.S.U. Poultry Science

23

24

Research and Teaching Center and placed into previously
prepared pens. The birds were raised to seven weeks of age
(except the birds that were killed for intestinal
microflora sampling during the experiment) when they were
killed and processed. This experiment was conducted from
mid—August to the end of September.

Three groups of birds were used in this experiment.
Group one was fed diet one which contained no penicillin.
Group two was fed diet two which contained penicillin at a
level of 10 ppm. Group three was fed diet three which
contained 100 ppm of penicillin. Each group was divided
into two replicates. Each replicate contained 26 birds. The
two replicates of the same group were housed in the same
room with a wire partition to keep them apart. Each
replicate was called a pen.

A total of six birds were killed in each pen at timed
intervals during the experiment. The birds killed before
the seventh week were not included in the determination of
weight gain for the pen, but they were weighed so that feed
conversion for the pen could be calculated.

The birds were raised on solid cement floors with white
pine shavings for bedding. Each bird had 0.33 square meters
of floor space. Heat was supplied by one 1500 watt, red,
heat lamp placed in the center of each pen two feet above
the floor. The temperature was maintained at 32.2C
throughout the first week and then the temperature was

dropped 2.8C weekly until the birds were four weeks of age.

25

At this time, the heat lamp was removed. Ventilation was
supplied by a fan for each group. The fan was controlled by
both a thermostat and a timer. The air was drawn into the
pen from a center hallway through a blackout hood and
exhausted to the outside. One incandescent light of 60
watts was used in each room. The light schedule was 23
hours of light and one hour of darkness.

The three rooms used in this study were adjacent to one
another with a separate door leading from the central
hallway into each room. Strict isolation procedures were
maintained for all the rooms. Only the principal
investigator entered the rooms. Before entering any of the
rooms, boots, gloves, and a lab coat, that were kept by the
door, were put on by the investigator. The investigator
then washed his hands and boots in disinfectant when he
entered the room. The hands of the investigator and his
boots were also washed when the he left the room.

The diets used in this study were prepared by the M.S.U.
Feed Mill. A broiler starter (Appendix A) and a broiler
finisher (Appendix A) were prepared for each group. All of
the treatment groups were fed the nonmedicated starter diet
for three days and then were fed the experimental diets
until the birds were four weeks of age. At that time, the
starter diets were removed from all of the pens and the
finisher diets were placed into the feeders. The finisher
diets were fed for the last three weeks. The starter diets

contained 24% protein and the finisher diets contained 21%

26

protein. Diet one was the control diet, which was mixed
first, and contained no penicillin or other antibiotics.
Diet two had a level of 10 ppm of procaine penicillin
(Procaine Penicillin 100, Carl S. Akey Co., Lewisburg,
Ohio) mixed into it and diet three contained a procaine
penicillin level of 100 ppm. The procaine penicillin was
purchased in the form of a premix containing 132 gm
procaine penicillin per kg of premix and was added directly
to the diet.

The feed for each replicate was placed in a separate
container outside of the group's door. A separate scoop was
used for each diet. The feed was weighed when it was placed
in the container and weighed back on the days the chicks
were weighed to determine feed conversion. Feed and water

were available gg libitum.

Mink

Three treatment groups of mink were used in the second
experiment. Each group of mink was comprised of two
replicates. Each replicate contained two mink. All of the
mink used in this study were pastel females that were
approximately six months of age.

Each group was housed in a separate building at the
M.S.U. Mink Ranch. The buildings were open sided pole barns
with a space of 20 feet between the barns. The lighting and

temperature were both ambient. The mink were housed in

27

individual cages (30 x 75 x 45 cm) which were made of
galvanized hardware screen and had a wooden nest box (32 x
20 x 22 cm) connected on the outside that was accessible
from the cage.

The mink were fed a wet feed containing two thirds
viscera obtained from the chickens that were processed in
the first experiment and one third of a commercial pelleted
mink cereal (XK-40 Grower, XK Mink Foods, Thiensville,
Wi.). Water was added to the diet to make a workable
consistency. After mixing, the feed was placed into one kg
bags and frozen at -20C. The feed was kept frozen until two
days before the feed was needed. At that time, the feed was
placed in a 4C refrigerator to thaw and kept there until
the feed was used.

The experimental diets that were fed to the mink were
tested for the presence of penicillin using the A.O.A.C.
method of analysis for the determination of penicillin in
animal feeds (A.O.A.C., 1984). Brain Heart Infusion- (BHI)
Agar (Difco, Detroit, Mi.) plates were poured and allowed

to harden. Five ml of a Staphylococcus aureus culture (ATCC

 

25923) were added to 45 ml of BHI Agar that was preheated
to 48C. The g. aureus culture was grown overnight at 37C in
five m1 of BHI broth (Difco, Detroit, Mi.). Four ml of the
BHI Agar that contained the g. aureus culture were poured
onto the BHI Agar plates that had been poured previously.
Once the top layer hardened, five, 10 mm, stainless

steel, penicylinders (Fisher Scientific, Pittsburgh, Pa.)

28

were placed on top of the agar in each plate. Into cylinder
one, 250 ul of a 0.05 units/ml of penicillin (Gibco Labs.
Grand Island, N.Y.) was added. Cylinder two had 250 ul of a
0.20 units/ml of penicillin added. Into cylinder three, 250
ul of a 0.80 units/ml penicillin solution was added.
Cylinders four and five had 250 ul of a feed sample
solution added. All of the penicillin standards and the
feed samples were prepared according to the A.O.A.C. (1984)
method.

The mink were fed by placing the feed in a small can on
the floor of the mink's cage. The can was held in place by
a wire ring so the mink could not spill the feed. The mink
were fed daily and any feed remaining from the last feeding
was discarded. An acclimation period of one week was given
to the mink for them to become accustomed to the new feed
and the different style of feeding. Any of the animals that
did not adapt to the new feed were replaced with different
animals. Four days after the acclimation period was over,
feed consumption was measured for two consecutive days.

Fecal samples were collected from the mink before the
experimental feed was given, seven days after the
acclimation period ended, and seven days after the

experimental diets were withdrawn.

29

Sampling

Intestinal, microflora samples were taken from each
treatment group of chickens at 14, 28, and 46 days after
the birds were started on the experimental diet. One
treatment group of chickens was sampled at a time. Four
birds from each group (two birds from each replicate in the
group) were taken to the laboratory. One bird was killed by
cervical dislocation while the other birds were kept in a
cage in another room. The abdomen was opened and the
jejunum, which is the section of the small intestine from
the distal end of the duodenum to half the distance to
Meckel's diverticulum, was aseptically removed and placed
in a sterile petri plate and weighed. The ceca were then
removed aseptically and also placed in a sterile petri
plate and weighed. The organ was then cut medially and the
contents were removed by scraping the wall of the organ
once with the long side of a pair of forceps. The organ was
placed in a Waring blender while the petri plate and the
contents were weighed. The difference in the two weights
for each organ was multiplied by nine and this amount of
anaerobic, dilution salts (V.P.I., 1973) was added to the
Waring blender. The sample was mixed in the blender for one
minute. After the sample was mixed, one ml was pipetted
into nine m1 of anaerobic dilution salts and immediately
placed into the anaerobic glove box (Coy Manufacturing Co.,

Ann Arbor, Mi.). Another one m1 of sample was removed from

30

the blender and placed into nine m1 of phosphate buffered
saline solution (Appendix B) to be used for the aerobic
dilutions.

Samples of fresh mink feces were collected from
underneath the mink's cages by having the investigator wait
by the cage until the mink defecated. A fecal sample was
picked up with sterile forceps and placed in a sterile
petri plate. The feces were transported to the laboratory,
within 15 minutes after they had been collected, where they
were immediately prepared for plating. The samples were
weighed and put into a sterile 50 ml flask. One ml of the
sample was diluted one to ten with anaerobic dilution salts
and placed in the anaerobic glove box and another one ml of
the sample was diluted one to ten with phosphate buffered

saline to be used for the aerobic dilutions.

Microbiological methods

Aerobic dilutions of the feces and gut contents were
made using a one to ten serial dilution. The dilution
tubes contained phosphate buffered saline solution. The
serial dilutions were made with a pipette bulb. Anaerobic
dilutions were also a serial one to ten dilution. The
dilution blanks contained nine ml of anaerobic dilution
salts. The dilution blanks were kept in the anaerobic glove
box for 24 hours before they were used so that the

anaerobic dilution salt solution would be reduced. The

31

dilutions were accomplished by using a pipette pump in the
glove box.

The first set of samples, that was taken from the
chickens, was spread plated onto different agars. This
procedure involved pipetting 0.1 ml of a dilution on an
agar plate. The sample was then spread across the plate
with the use of a sterile bent Pasteur pipette. This method
used one agar plate for each dilution. Each sample was
plated using three to six dilutions. The drop plate
technique of Miles and Misra (1938), which used fewer petri
plates and less agar, was employed for the remaining
samplings. The drop plate method used one drop of a
bacterial dilution placed on an agar plate. The drop was
not spread on the agar plate. The method allowed five or
six dilutions to be placed on one plate (Davis, 1971). The
bacterial counts were similar whether the spread plate or
the drop plate method was used (Richmond and Chang, 1976).

The anaerobic glove box used in these experiments was
filled with a gas mixture that contained 10% carbon
dioxide, 10% hydrogen and 80% nitrogen. Anaerobic
conditions were maintained by using palladium pellets which
formed water from the oxygen in the glove box. The
palladium pellets had to be regenerated every three to four
days by heating them to 160C for two hours. The temperature
in the glove box was maintained at 37C and the relative
humidity was kept at 45% by the use of calcium chloride

granules. Hydrogen sulfide was eliminated by using a silver

32

sulfate solution (Appendix B). Once the sample was placed
into the glove box, all further anaerobic work was

performed in it.
Growth media used

Five different media were used for the enumeration of
bacteria from the intestinal tract of the chickens and the
feces of the mink. The media used were Brain Heart Infusion
Agar (BHI Agar), Eosin Methylene Blue Agar (EMB Agar),
Sulfite Polymixin Sulfadiazine Agar (SPS Agar), Rogosa SL
Agar (all from Difco, Detroit, Mi.) and Anaerobic Agar
(BBL, Cockeysville, Md.).

BHI Agar is a general purpose agar used for the
cultivation of aerobic bacteria. It is a good growth medium
even for fastidious organisms. This medium was used to
determine the total number of viable bacteria in the
samples.

EMB Agar is a differential medium that is recommended
for the isolation and detection of Gram-negative intestinal
bacteria. The lactose fermenters, which include the
coli-aerogenes group, appeared as colonies with dark
centers. The non-lactose fermenters appeared as colorless
colonies. The total number of coliforms in the samples were
determined using this medium (Murphy, 1975).

The determination. of total anaerobic bacteria in the

samples was accomplished by using Anaerobic Agar. This

33

medium is a non-selective, non-differential medium that is
useful for general purpose anaerobic bacteria growth
(Difco, 1953).

The number of Clostridium perfringens was enumerated

 

 

with the use of SPS Agar. This medium is selective for the

growth of Clostridium spps. and differential for Clostridium

 

 

perfringens. The Clostridium perfringens colonies appeared

 

 

 

black while the other organisms were white.

Rogosa SL Agar is a selective medium for the enumeration
of lactobacilli. The medium is very acidic which inhibits
the growth of most bacteria except for the lactobacilli
(Difco, 1968). The lactobacilli were grown in the anaerobic
glove box.

The anaerobic plates were placed in the anaerobic glove
box 24 hours prior to use to reduce the media and also to
check for sterility. The aerobic plates were placed in an
incubator for 24 hours to check for sterility. The placing
of the agar plates in the incubator or glove box for 24
hours also dried the media and reduced the smearing of the
sample drops when they were placed on the media. After the
aerobic plates were inoculated, they were incubated for 18
hours and then counted. The anaerobic plates were incubated
for 48 hours before they were counted. Any drop that
contained less than 50 colonies was counted.

One BHI Agar plate from each sample was replica plated
onto the same media containing 400 ug penicillin per ml of

media. One Anaerobic Agar plate from each sample was also

34

replica plated onto the same media containing 100 ug
penicillin per ml of media. The amount of penicillin that
was added to the media was determined by using a gradient
plate described by Carlton and Brown (1981). An aerobic
penicillin—resistant microorganism was streaked across a
BHI agar plate that had a penicillin-gradient level from
0.0ug/ml to 400 ug/ml. The highest level of penicillin on
which the penicillin-resistant organism grew was selected
to determine the resistance to penicillin in the rest of
the study. A lower penicillin level was selected for the
anaerobic microorganisms since the penicillin-resistant
anaerobes did not grow on media containing 400 ul/ml of
penicillin. The penicillin was added to the media after the
media was autoclaved and cooled to 50C to avoid destruction
of the penicillin. The BHI-penicillin Agar plates were
incubated for 24 hours after inoculation and the
Anaerobic-penicillin Agar plates were incubated for 48
hours in the anaerobic glove box before they were counted.
The BHI Agar plates and the Anaerobic Agar plates were
replicated using the method of Lederberg and Lederberg
(1951). The original plate was placed with the medium
exposed to the air. A sterile piece of velveteen wrapped
around a die was lightly touched to the surface of the
medium of the original plate. The velveteen was then
removed from the original plate and touched lightly to the
penicillin-containing plate. Only one transfer of the

colonies was needed to the velveteen for numerous replicas

35

to other plates.

Plasmid identification

The bacterial colonies used for plasmid identification
were isolated from the BHI-penicillin Agar plates. The
purpose of the plasmid identification was to determine if
any penicillin resistance was transferred by
extrachromosomal DNA from the bacteria in the chicken
viscera to the bacteria found in the feces of the mink that
ate the chicken viscera.

The purification of plasmid DNA was performed using a
modified procedure found in Maniatis gt gt. (1982). A pure
isolated culture was grown in five m1 of BHI Broth
containing 50 ul of ampicillin (Gibco Labs. Grand Island,
N.Y.) per ml for 18 hours. The tube was then centrifuged at
2800 RPM in a microcentrifuge model 235C (Fisher
Scientific, Pittsburgh, Pa.) for 10 minutes. The
supernatant was poured off and the pellet was resuspended
in 350 ul of a Sucrose, Tris, EDTA, Triton 100 solution
(STET, Appendix C). This solution was transferred to an
Eppendorf tube and 12.5 ul of a lysozyme (Sigma, St. Louis,
Mo.) solution (Appendix C) was added. Heat was added to
facilitate the action of the lysozyme and then the tube was
centrifuged in the microcentrifuge at 12,000 RPM for 10
minutes. The pellet was removed and discarded and 500 ul of

isopropanol (analytical grade) was added to the

36

supernatant. This mixture was placed in -20C freezer for 10
minutes to facilitate the precipitation of the DNA. The
tube was then centrifuged in the microcentrifuge at 12,000
RPM at 4C for 10 minutes. The supernatant was poured off
and the pellet was washed once with one ml of cold absolute
ethanol (EtOH). The sample was again centrifuged in the
microcentrifuge at 12,000 RPM at 4C for five minutes and
the supernatant was discarded. The precipitate was dried in
a dessicator. The dried pellet was resuspended in 200 ul of
a Tris EDTA (TE) solution, pH of 8.0 (Appendix C). The
sample was extracted using 200 ul of phenol (molecular
biology grade), then with 200 ul of a phenol and
formaldehyde solution (Appendix C) and finally, with 200 ul
of a formaldehyde solution (Appendix C). Each extraction
was shaken gently for 10 seconds, and then centrifuged in
the microcentrifuge at 12,000 RPM for three minutes. The
supernatant was kept and the pellet was discarded. Twenty
ul of a three molar sodium acetate solution was added to
the supernatant from the extraction process and the
solution was precipitated with one m1 of 95% cold EtOH.
The sample was placed in a -20C freezer for 15 minutes to
facilitate the precipitation. The tube was centrifuged in
the microcentrifuge at 12,000 RPM at 4C for 15 minute. The
supernatant was poured off and discarded. One ml of cold
95% EtOH was added to the pellet and centrifuged in the
microcentrifuge at 12,000 RPM at 4C for five minutes. The

supernatant was poured off and the pellet was placed in a

37

dessicator. The dried pellet was resuspended in 50 ul of TE
pH 8.0 and stored in a -20C freezer until needed.

The first test done with the sample was to determine by
gel electrophoresis if any plasmid DNA was present. The
liquid measurements were done using a 20 ul pipetman or a
200 ul pipetman (Gilson Co., France). Eight ul of sample
was mixed with one ul of a ribonuclease A (Sigma, ST.
Louis, Mo.) solution prepared according to Maniatis gt gt.
(1982). This mixture was heated at 37C for five minutes to
destroy any RNA that was present in the sample. At that
time 2.5 ul of a five-x loading buffer (Appendix C) was
added to the sample and heated at 65C for five minutes. The
sample was placed in a well in 0.7% agarose gel that
contained 0.01 mg/ml of ethidium bromide. A 0.7% agarose
gel was used because it gives an efficient separation of
DNA from 10,000 base pairs to 800 base pairs Maniatis gt gt
. (1982). The ethidium bromide was added as a stain. The
ethidium bromide intercalates with the DNA which causes the
bound ethidium bromide to fluoresce when exposed to
ultraviolet light.

The power for running the gel electrophoresis was set at
a constant 50 volts DC and run for one hour. After one hour
the gel was placed over a ultraviolet light and a picture
was taken and the bands of the different plasmids were
compared for similar weight or size by the distance that
the bands migrated through the gel.

The same plasmid samples were used for cutting by the

38

restriction endonucleases. Ten ul of sample were added to
seven ul of deionized water and two ul of a 10-x react
buffer three (BRL, Gaithersburg, Md.). One ml of the
restriction endonuclease was added to this mixture and
heated for two hours at 37C. One ml of the Ribonuclease A
solution was added at this time and kept at 37C for another
five minutes. Finally, five ul of the five-x sample buffer
was added and heated at 60C for five minutes. This sample
was then electrophoresed on a similar gel as the uncut DNA.

EcoRl (BRL, Gaithersburg, Md.) was the first restriction
endonuclease used to cut the plasmid DNA. Bgl II (BRL,
Gaithersburg, Md.) was the other restriction endonuclease
used to cut the plasmid DNA. EcoRl recognizes a five base
pair sequence and Bgl II recognizes a six base pair
sequence. EcoRl cuts at:

5' GAATTC 3'
3' CTTAAG 5'

Bgl II cuts at:

5' AGATCT 3'
3' TCTAGA 5'

Attempts were made to transform the isolated plasmids
into an E. ggtt strain number HBlOl. The E. ggtt had
previously been made competent by using the calcium
chloride method as described by Maniatis gt gt. (1982).

The data obtained from the first three experiments of
this study were analyzed using a split plot repeat
measurement. The analysis of variance was performed using

Bonferroni-t statistics (Gill, 1978). A level of P<0.05 was

39

used as the level of statistical significant difference
unless indicated differently.

A coefficient of variation was calculated for each
microbiological sample on each type of media. This is
listed in each table in the results. The coefficient of
variation is defined as the standard deviation divided by
the mean. The number of replicates used for determining the
coefficient of variation was ten. A coefficient of
variation less than 0.5 is good when working with animals

(Gill, 1978).

RESULTS

Experiment one

Summarized in Tables 1 and 2 are the effects of feeding
penicillin at 10 ppm or 100 ppm in a broiler diet as
compared to a diet containing no penicillin on the body
weights of broilers fed these diets. As noted in Table 2,
there was no significant difference in body weight between
any of the treatments by date. The average body weights of
the birds were within the expected range of body weights
for broilers of that age. The variation between treatments
was smaller than the variation between the birds within a
treatment.

The feed conversions for the birds in each pen are
listed in Table 3. The final feed conversions ranged from
1.77 grams of feed eaten for each gram of weight gained to
1.82 grams of feed eaten for each gram of weight gained.
These final feed conversions were better than expected for
commercial broilers. An usual feed conversion for a
commercial broiler flock would range from 1.85 to 1.95
grams of feed ingested for each gram of weight gained. The
analysis of variance, shown in Table 4, indicates that

there was no significant increase or decrease in feed

40

41

Table 1. Average body weight of broilers fed different
levels of penicillin#.

Age Pen Treatment 1 Treatment 2 Treatment 3
(days) 0 ppm 10 ppm 100 ppm
penicillin penicillin penicillin
”6"”?"’35;6TEI""'"""3633?;E'"""""ZS£6T35 “““
2 4010.72 4010.62 3910.40
14 1 42115.42 44816.53 44816.65
2 42116.02 45917.83 440110.58
28 1 1322118.28 1214117.46 1247117.52
2 l320120.30 l3511l7.28 1285123.l8
42 l 2356133.95 2285129.86 2380144.31
2 2349129.55 2400136.21 2340133.03

# Weight shown in gm 1 standard error.
n = 20.

Table 2. Analysis of variance of the body weight of
broilers fed different levels of penicillin.

Source
of
var. d.f. M.S. f value
Treatment 2 79.65 0.093
Error 1 3 849.45
Time x Trt. 6 746.35 0.443
Error 2 9 1684.94

42

Table 3. Feed conversion of broilers fed different levels
of penicillin#.

Age Pen Treatment 1 Treatment 2 Treatment 3
(days) 0 ppm 10 ppm 100 ppm
penicillin penicillin penicillin

14 1 1.31 1.21 1.22

2 1.30 1.20 1.24
28 l 1.63 1.73 1.77

2 1.59 1.69 l 64
42 1 2.20 2.16 2.11

2 2.15 2.15 2.14
final 1 1.81 1.81 1.82
feed
conv. 2 1.77 1.78 1.78

# gm feed per gm body weight gain.
n = 20 broilers in each pen.

Table 4. Analysis of variance of the feed conversions of
broilers fed different levels of penicillin.

Source
of
var. d.f. M.S. f value
Treatment 2 0.00 0.000
Error 1 3 0.00
Time x Trt. 4 0.0050 1.515

Error 2 6 0.0033

1+3

conversion between the three treatments on any weigh date.

Experiment two

Summarized in Tables 5 to 14 are treatment responses

over time and analysis of variance for total aerobes, total

coliforms, total anaerobes, lactobacilli, and Clostridium

 

perfringens in two locations of the intestinal tract of

 

chickens, the jejunum and the cecum. The bacterial counts
are displayed as log base 10 transformations based on per
gram of intestine (wet weight).

The total, aerobic, bacterial counts in the jejunum
(Table 5) fall within the expected range of 103 to 109.
Shown in Table 6 is the analysis of variance for the total
aerobes. As indicated, no significant difference was
observed between the treatments on the numbers of total
aerobes in the jejunum.

The bacterial counts of the microflora of the cecum were
much more consistent than the jejunum bacterial counts, as
seen in Table 5. The analysis of variance for total aerobes
in the cecum shows that there was no significant difference
in the total number of aerobic bacteria in the cecum due to
the feeding of penicillin at either level in this
experiment.

The coliforms that should make up approximately 90% of

the aerobic bacteria in the jejunum varied from 80% to 100%

44

mN.H Nh.H m H nouum

mm.o NH.H h~.H ma.m N pcwEummuB
wsam> m .m.z waam> m .m.: .um>
.m.© mo
mousom
EDUWU EDZDUMU

.cHHHHowch own muwawoun mo Donny HmcwummDCM
mcu cﬂ mmuﬂm 03» um mwumuomn oanoumm .Hmuou no“ mocMﬁum> mo mwmwamc< .m manna

. H.o u Esowo on» now cowumwum> mo ucwﬁOﬂmmmoo
.mm.o u Escsmmn on» now cowumﬁum> no ucmﬁoﬂmmmoo
.mumawoun ¢ u c

.uouuo pumocmum H msmmﬂu Hmcﬂumoucw may

mo unmﬂwz uma Emma Mom owumuomn mo mmoa 0H mmmn mm Ummmwumxm muczoo HH< *

mn.ouma.m mh.oH~m.oH mm.oHaq.oa oH.onH.v mm.oHH~.m mm.ouom.¢ ow

om.oumm.m am.oHHa.oH m¢.ouag.m mm.onG.m mm.ono.m -.owhm.q mm

av.owom.OH m~.oHHa.m mm.on~.m km.oHGv.q mm.cHH~.v oH.ono.¢ «a
chaaﬁohama cuaahonamm aHHHHoHama chaﬂﬁohcmm cﬂaaﬂohamm :ﬂHHhohcmm

Ema ooH Ema oa 5mm 0 5mm ooa Ema ca and o pwuumum comm

kucosﬂummxo

zoomo Eszonmn umumm mama

.*uomuu Hmcwumwucw umawoun on» cw woman 03» um muccoo
Hmwumuomn .oﬂooumm .Hmuou co mumaﬂoun o» cwaawowcmm mcﬁcmmm mo nommmm one .m manna

45

of the observed aerobic bacteria, which was close to what
was expected, see Table 7. No significant change was
observed in the coliform count for the jejunum, as
indicated in Table 8.

The coliforms made up 90 to 100 percent of the total,
aerobic, bacterial count of the cecum as shown in Table 7.
The number of coliforms found in the cecum agree with the
findings of Smith (1965) who found that coliform counts

2 to 108

vary from 10 . No significant change was observed
in the coliform counts in the cecum due to the feeding of
penicillin at 10 ppm or 100 ppm, as summarized in Table 8.
Shown in Table 9 are the total, anaerobic bacteria
counts obtained in this experiment for the broilers. The
anaerobic, bacteria counts in the jejunum were quite
variable between each animal but the average was consistent
at approximately 106 for any treatment or time. Barnes gt gt

. (1972) found 108

anaerobic bacteria in the small
intestine, which was 100 times greater than found in this
study, but intestinal contents were also included in those
counts which may have increased the counts. As indicated in
Table 10, no significant difference was found between the
three treatments on the anaerobic, bacterial counts in the
jejunum.

The total, anaerobic count in the cecum of the broilers

was significantly higher at days 14 and 28 in the birds

that were fed the diets containing penicillin than in the

46

l'l'I'l""l'|ll'l|"'llll'|"Ill't'l"l""‘l'""'l'||""""ll"l'|"|"|""'l"""':

Hm.H vm.H m H “Chum

ov.m mv.v mm.a NH.N N ucmEumwuB
wsam> m .m.: msaw> w .m.: .um>
.m.p «0
mousom
ZDUWU ZDZDhmh

.cwaawowcwm com mumawoun mo noon» Hmcwummucw
on» ca mmuﬂm o3u um manomwaoo Hmuou MOM mocwwum> mo mwmwamcc .m manna

.om.o u Esomo on» now coaumwum> mo ucmﬂOmemoo
.mm.o n Eccsﬂon on» now cowumaum> mo ucmﬁoﬂmmmoo

.muoawoun v u c

.uouum pumccmum H wsmmﬂu Hmcwumwucﬂ on»

NO Davao: #63 Emum umm wwumuomn mo mmoH ca mmmn mm cmmmmumxm mucsoo Had *

hm.ouom.m am.oumm.oa m~.oumo.OH oo.oH¢m.mv om.loH.m e¢.owmw.v mv

H~.owqa.m mm.owmm.OH m~.oHGo.m mm.owoa.m oa.oH~o.m «H.cho.v mm

mm.oH~h.m ~m.oHHo.m mo.owmm.m o~.oH¢H.v o~.ouq~.¢ oo.ou¢m.mv 4H
cgaaﬁoﬂcmm :HHHﬂohcmm :ﬂaanowcma cgaaﬂohcmm :HHHﬂoﬂcma cHHHHoﬁamm

Ema ooa Eda ea and o and OOH Eda ca 5mm o pmuumum comm

HchwEHummxm

:oomo zozanmn umuum mama

.*uomuu amcﬂumoucw uwawoun may cw mouﬁm 03» um
mucsoo Euomwaoo .Hmuou co mumawoun on :HHHHoﬁcmm mcwcmmm mo Dowmmm one .h wanna

47

birds that were fed a diet that contained no penicillin.
The results of the total anaerobic bacteria counts are
summarized in Tables 9 and 10.

The lactobacilli counted in the jejunum were slightly
lower than expected when compared to the work of Barnes gt
gt. (1972) who found 107 or Smith (1965) who found 108
lactobacilli per gram of intestine when compared to this
study's finding of 105, see Table 11. The analysis of
variance presented in Table 12 indicates that there was no
significant difference seen between treatments on the
lactobacilli counts in the jejunum.

The lactobacilli counted in the cecum from any of the

7
treatments were within the expected range of numbers of 10

to 109 (Smith, 1965). The counts are summarized in Table
11. The lactobacilli counts were not significantly changed
by the feeding of penicillin at 10 ppm or 100 ppm in this
study.

The Clostridium perfringens counts in the jejunum were

 

 

L;
not detectable at 10 , as shown in Table 13. No significant

change in the Clostridium perfringens numbers was

 

detectable between the three treatments in the jejunum, see
Table 14.

The Clostridium pgrfringgns counts obtained from the

 

cecum in this experiment, as summarized in Table 13, are
much higher than those of Barnes gt gt. (1972) found in

chickens of the same age. The feeding of penicillin did not

48

.mo.o Va um acmoﬁmﬂcmﬂm 1

:""""""|I"""|"I"|'|'|"""l""'I'"-II.""'l|"I|""""|"'I"I'|""'||"|

m H nouum

m ucoeummua

om.o

*Nm.~H Hm.a

msam> L .m.:
zoomo

mm.m

mv.o ~m.a

msam> L .m.:
zszasmn

.m.© mo

.cHHHHOHcoQ pom mumHHoun mo uomuu chHummucH
map cH mmuHm o3u um MHumuomn 0Hnoummcm .Hmuou now oOGMHnm> mo mHmemcﬁ .oH oHnma

oo.oHoo.~HA
«GH.oH~m.HH
mq.owmm.oH

oo.oHoo.mHA
«o>.owmm.aa
,m~.oHaH.HH

.mo.o vm um same man scum ucmummmgc saucmodmﬂcmhm 4
.NN.O N 59000 wnu Mom GOHHMHHMNV MO UCOHUHMMQOU

oo.oHoo.~HA
mh.oHom.m

mm.owmm.m

.G~.o u escsﬂmﬂ may you :oﬂumﬂum> Lo namgogommoo

.mumHHoun v u c

.uouum cumccmum H mommHu HmcHumoDCH on»
no uanms um: Emum Mom MHumuomn mo mmoH oH wmmn mm commmumxm mucsoo HH¢ *

m¢.oHaH.m 46.0Hmm.o
«m.oHaG.G «G.oHHm.G
h¢.oﬂmm.q mm.oum~.m

chﬂahoﬁcma
Ema COH

cﬁaahoﬁcwm
Ema oH

ZDUMU

cgaahoﬂcmd
Sam 0

ahaaﬂoﬁcmm cﬂﬁahohcmm
Ema ooH and 0H

SDZDHHH

hm.oHGc.o 64

mm.oump.o mm

oa.oHHH.¢ «a
aﬁaagohcmm

and o pwuumum comm

HousmEHuwmxw

uwumm mwmo

.*uomuu HmcHummucH umHHoun may :H mmuHm o3u pm mucsoo

HmHuwuomn .poouwmcm .Hmuou co mumHHoun on :HHHHoHch mchmmw

Lo nomomm was .m manna

I'l|"||'||""|'-"""'l‘l'"'I'l|"'"I'IIII"I'||'|""l||||'|"""'|'||"""""'"'

mm.m mm.m m H Mouum

O0.0 MO.~ hm.H mv.¢ m ucoEuooue
osHo> m .m.z osHo> w .m.z .uo>
.m.© mo
oousom
ZDUmU ZDZDOMH

.cHHHHoHcod pom muoHHoun mo poouu HmcHumoucH
onu EH mouHm 03» no HHHHoonouooH How oocoHuo> mo mmeHocd .NH oHnoe

.mH.O u Esooo on» How EOHuoHuo> mo ucoHonuooo
o/ .m~.O u Escsnon on» you EOHuoHuo> mo ucoHOHmwooo
n .muoHHoun w u c
.uouuo puoocoum H osmmHu HocHumoucH on»

NO uanoB uoz Eoum Mod oHuouoon mo mmoH OH omon mo commoudxo oucsoo HH4 *

H~.ouoa.a m~.oHHN.HH Hk.oﬂmm.m «q.cha.m ~m.oHHm.G >¢.oH~a.m mg

m~.oumo.m «m.oHo~.w mm.onH.m mm.chm.o «o.HHo~.G o~.onm.q mm

H4.ouom.a om.ono.m Go.HHGo.a mm.oH¢v.v om.oH~m.q oo.ouvm.mv «a
:HHHHoHEmd :Haanohcmd chaahogcmm :HHHHoHEmm cHHHnoﬂcmm :HHHHoncmd

Edd OOH Edd OH Edd O Edd OOH Edd OH Edd O wouuoum boom

HoucoEHuodxo

zoomo Suzanne gonna memo

.*uoouu HocHumoucH uoHHoun onu cH mouHm
03D no oucsoo HHHHoonouooH no muoHHoun on :HHHHOHcod mchoom mo uoowmo one .HH oHnoe

50

mm.m O0.0 m H Houum

O0.0 mo.m O0.0 O0.0 N ucoEuooue
osHo> w .m.: osHo> m .m.z .uo>
.m.© mo
oousom
EDUWU ZDZDUWU

.cHHHHoHcod com muoHHoun mo uoouu HocHumoucH on» EH
mouHm 03D no mnomcHumuod EchHuumoHU you oocoHuo> mo mHm>Hoc< .vH oHnoe

 

.em.o u Esooo on» now :oHuoHuo> mo ucoHonmooo

.muoHHoun v n c

.uouuo puopcoum H osmmHu HocHumoucH on»

mo uanoB uoz Eoum uod oHuouoon mo mOOH OH omon mo commoudxo mucsoo HH< *

so.on¢.p om.oHGm.m HG.onn.m oo.oH¢m.mv oo.cH¢m.mv oo.oH¢m.mv ow

ma.oﬂoh.oa Hm.onm.m m~.onm.h oc.cH¢m.mv oo.ova.mv oo.oH¢m.mv aw

m~.oumk.m Hm.oHHG.m m¢.oH>H.m oo.ouqm.mv oo.oH¢m.mv oo.oH¢m.mv 4H
:HHHHoncmd unannoncmd EHHHHogcmm :HHHHoncmd anaﬂgoncwd cHHHHoHcmm

Edd OOH Edd OH Edd O Edd OOH Edd OH Edd O couuoum coon

HoucoEHuodxo

zoomo zOzsnmn noumm memo

.wuoouu HocHumoucH uoHHoun on» EH mouHm 03» um mucooo
mcomcHumuod EsHoHuumoHo co muoHHoun o» cHHHHoHcod mcHooow mo pooumo one .MH oHnoe

 

51

significantly effect the number of the Clostridium

 

perfringens in the cecum of the broilers.

 

Summarized in Table 15 are the total, aerobic, bacteria

counts found in the mink feces. The average number of total

03

aerobes found in the mink feces was approximately 10 ,
regardless of the treatment group to which the mink
belonged. The analysis of variance in Table 16 indicates
that there was no significant difference between different
treatments.

The coliform counts obtained from the mink feces are
summarized in Table 17. The number of coliforms compares
with the number of coliforms found in the feces of cats
(Smith, 1961). The cat was the most similar animal to mink
that information could be found pertaining to bacterial
counts in the feces. The total aerobic population of the
mink feces was composed of approximately 75 % coliforms.
The analysis of variance for the coliform counts in the
mink feces is summarized in Table 18. As indicated, no
significant difference was found between treatments.

The total, anaerobic, bacteria counts shown in Table 19
were not as variable between treatments as the aerobic,
bacteria counts in the feces of the mink. The total

7 9

anaerobes varied from 10 to 10 . The analysis of variance
that is summarized in Table 20 shows that no significant
difference was found between treatments.

Summarized in Table 21 are the lactobacilli counts found

52

Table 15. The effect of feeding a diet containing the
viscera of penicillin-fed chickens to mink on total,
aerobic, bacterial counts found in the feces of mink#.

Days Treatment 1 Treatment 2 Treatment 3
:gperimental 0 ppm 10 ppm 100 ppm
feed penicillin penicillin penicillin
51'""""""§:3§;6:§§ """"" IIII"""""""TIZI’"
7 7.5810.53 7.9010.50 8.8110.83
7 days after 9.7010.82 8.4610.21 8.6210.89

put back on

normal feed

# All counts expressed as base 10 logs of bacteria per gram
wet weight of mink feces 1 standard error.

* n = 2 for day 0; n = 4 for the other two sampling dates.
coefficient of variation = 0.16.

Table 16. Analysis of variance for total, aerobic bacteria
found in the feces of mink fed a diet containing the
viscera of chickens that were fed penicillin.

Source
of
var. d.f. M.S. f value
Treatment 2 0.67 1.03

Error 1 3 0.65

53

Table 17. The effect of feeding a diet containing the
viscera of penicillin-fed chickens to mink on total,
coliform counts found in the feces of mink#.

Days Treatment 1 Treatment 2 Treatment 3
ggperimental 0 ppm 10 ppm 100 ppm
feed penicillin penicillin penicillin
51""'""""§T§;;3fg1 """""" 2222"""'""""IIII’"
7 6.6510.50 6.3310.40 7.4011.09
7 days after 8.2111.61 6.0210.90 6.2310.33

put back on

normal feed

# All counts expressed as base 10 logs of bacteria per gram
wet weight of mink feces 1 standard error.

* n = 2 for day 0; n = 4 for the other two sampling dates.
coefficient of variation = 0.22.

Table 18. Analysis of variance for total coliforms found in
the feces of mink fed a diet containing the viscera of
chickens that were fed penicillin.

Source
of
var. d f M.S. f value
Treatment 2 3.16 0.48

Error 1 3 6.63

54

Table 19. The effect of feeding a diet containing the
viscera of penicillin-fed chickens to mink on total,
anaerobic, bacterial counts found in the feces of mink#.

Days Treatment 1 Treatment 2 Treatment 3
:gperimental 0 ppm 10 ppm 100 ppm
feed penicillin penicillin penicillin
6"""""""§T§E§6TS§ """"""" IIII"""'""""IZIT"
7 7.2510.58 7.5210.54 8.2910.77
7 days after 8.0910.60 8.6610.40 7.6710.81

put back on

normal feed

# All counts expressed as base 10 logs of bacteria per gram
wet weight of mink feces 1 standard error.

* n = 2 for day 0; n = 4 for the other two sampling dates.
coefficient of variation = 0.49.

Table 20. Analysis of variance for total, anaerobic
bacteria found in the feces of mink fed a diet containing
the viscera of chickens that were fed penicillin.

Source
of
var. d f. M.S. f value
Treatment 2 0.38 0.48

Error 1 3 0.79

55

in the feces of the mink. The counts varied from 106 to 109
. This number relates closely to the number of lactobacilli
found in the feces of cats (Smith, 1961). The analysis of
variance between treatments for lactobacilli is shown in
Table 22. This table indicates that no significant
difference was found between the treatments.

The Clostridium perfringens counts are summarized in

 

Table 23. The Clostridium perfringens counts in treatment

 

one compare with the Clostridium perfringens counts Smith

 

(1961) found in the feces of cats. Treatment two and

treatment three had higher counts of Clostridium perfringens

 

than Smith (1961) found in the feces of cats. Although a
variation was seen between treatments, no significant

effect was observed, as indicated in Table 24.

Experiment three

In this study, an increase in resistance to penicillin
is defined as an increase in the percent of bacteria that
are able to grow in the presence of penicillin.

The effect of feeding penicillin on the resistance of
the aerobic, bacterial microflora in the chicken jejunum is
summarized in Table 25. The percent of
penicillin-resistant, aerobic bacteria found in the jejunum
was quite variable between the treatments and time, varying

from 0 to 8.2%. The variation was not related to treatment

56

Table 21. The effect of feeding a diet containing the
viscera of penicillin-fed chickens to mink on lactobacilli
counts found in the feces of mink#.

Days Treatment 1 Treatment 2 Treatment 3
ggperimental 0 ppm 10 ppm 100 ppm
feed penicillin penicillin penicillin
6"""""""ETI§;6TE§ """""" III-""-"""mIII'"
7 7.4710.45 7.5210.58 7.9410.61
7 days after 6.0710.41 8.6610.40 7.1511.04

put back on

normal feed

# All counts expressed as base 10 logs of bacteria per gram
wet weight of mink feces 1 standard error.

* n = 2 for day 0; n = 4 for the other two sampling dates.
coefficient of variation = 0.37.

Table 22. Analysis of variance for lactobacilli found in
the feces of mink fed a diet containing the viscera of
chickens that were fed penicillin.

Source
of
var. d.f. M.S. f value
Treatment 2 1.68 1.12

Error 1 3 1.50

57

Table 23. The effect of feeding a diet containing the
viscera of penicillin-fed chickens to mink on Clostridium
pgrfringens counts found in the feces of mink#.

 

Days Treatment 1 Treatment 2 Treatment 3
ggperimental 0 ppm 10 ppm 100 ppm
feed penicillin penicillin penicillin
6""""""'23T32§6T66 """""" III: """""""" III"-
7 <3.9410.00 <3.9410.00 5.4410.57
7 days after <3.9410.00 5.6111.11 4.3910.37

put back on

normal feed

# All counts expressed as base 10 logs of bacteria per gram
wet weight of mink feces 1 standard error.

* n = 2 for day 0; n = 4 for the other two sampling dates.
coefficient of variation = 0.25.

Table 24. Analysis of variance for Clostridium perfringens
found in the feces of mink fed a diet containing the
viscera of chickens that were fed penicillin.

 

Source
of
var. d.f M.S. f value
Treatment 2 3.26 0.23

Error 1 3 13.91

or to time effects, as indicated in Table 26. These data
show that no significant difference was found between the
treatments.

The percent of resistant, aerobic bacteria found in the
cecum of the birds did increase from approximately 3.0%
after 14 days of feeding the penicillin (to over 10% after
28 days of feeding the penicillin in the two treatment
groups (Table 27). The control group showed no increase in
penicillin-resistant bacteria in the cecum during the same
period of time. The high percent of penicillin-resistant
microorganisms indicated for treatment one at day zero was
composed of yeast cultures that were present in the birds
sampled at the start of the experiment. The increase of
penicillin-resistant bacteria in the cecum of the two
treatment groups being fed penicillin was not significant
(Table 28). The sampling taken at 46 days showed a large
drop in penicillin-resistant organisms when compared to the
28 day sample. This may reflect the withdrawal of
penicillin from the feed three days prior to the sampling.

The anaerobic bacteria found in the jejunum of the
chickens did not show a significant increase in the
percentage of penicillin-resistant bacteria, as summarized
in Table 29. The control group of birds had a variation
from 0.0 to 21.4% penicillin-resistant organisms at any
sampling time, while the treatment group fed penicillin at

10 ppm penicillin only had a variation of 5.3 to 10.0%

59

Table 25. The percent of aerobic microorganisms resistant
to penicillin isolated from the jejunum of chickens fed
different levels of penicillin.

Days Treatment 1 Treatment 2 Treatment 3
ggperimental 0 ppm 10 ppm 100 ppm
feed penicillin penicillin penicillin
’BE"'""""'IST£65636 """"" IIIZ""""'"""IZIT"
14 5.0015.00 0.0010.00 8.2518.25
28 0.0010.00 4.1814.17 4.2812.54
46 0.0010.00 0.8810.55 0.0010.00

All counts expressed as percent of total microorganisms 1
standard error.

Resistance expressed as growth on media containing 400

ug/ml penicillin.

# n = 2 for day 0; n = 4 for the other sampling dates.

Table 26. Analysis of variance for the
penicillin-resistant, aerobic microorganisms isolated from
the jejunum of chickens fed different levels of penicillin.

Source
of
var. d.f. M.S. f value
Treatment 2 37.51 1.02

Error 1 3 36.83

60

Table 27. The percent of aerobic microorganisms resistant
to penicillin isolated from the cecum of chickens fed
different levels of penicillin.

Days Treatment 1 Treatment 2 Treatment 3
:gperimental 0 ppm 10 ppm 100 ppm
feed penicillin penicillin penicillin
’6;""""""2'ETBSEEET66 """"" IIII"""""""TIZI‘"
l4 22.25119.37 3.0810.22 2.9011.16
28 0.4210.42 10.5314.68 16.9815.61
46 0.0010.00 0.3310.32 4.3013.52

All counts expressed as percent of total microorganisms 1
standard error.

Resistance expressed as growth on media containing 400

ug/ml penicillin.

# n = 2 for day 0; n = 4 for the other sampling dates.

Table 28. Analysis of variance for the
penicillin-resistant, aerobic microorganisms isolated from
the cecum of chickens fed different levels of penicillin.

Source
of
var. d.f. M.S. f value
Treatment 2 502.29 1.51

Error 1 3 332.61

61

penicillin-résistant organisms. The treatment group being
fed 100 ppm of penicillin in the feed had a variation from
16.2 to 35.2% of penicillin—resistant organisms. The
analysis of variance for these data is shown in Table 30.
The increase of penicillin-resistant organisms in the
jejunum was not statistically significant between
treatments.

The penicillin—resistant anaerobes isolated from the
chicken cecum did increase slightly in the birds being fed
penicillin as shown in Table 31. The slight increase in
penicillin-resistant organisms from the control group to
the treatment group fed 100 ppm of penicillin was not
significant (Table 32).

The percent of aerobic bacteria found in the feces of
mink that were resistant to penicillin at 400 ug/ml did
increase in the mink fed the viscera of the broilers that
had been fed penicillin. The control group of mink showed a
variation of l to 5% of resistant organisms while the
groups of mink fed the other two groups of broiler viscera
had a variation of resistant bacteria from 6.2 to 29%, as
summarized in Table 33. The increase in resistance in
treatment three as compared to treatments one and two was
significant at P< 0.05, as shown in Table 34.

The percent of anaerobic bacteria in the feces of
mink that were resistant to penicillin at 100 ug/ml was

high before any of the experimental chicken viscera was

62

Table 29. The percent of anaerobic microorganisms resistant
to penicillin isolated from the jejunum of chickens fed
different levels of penicillin.

Days Treatment 1 Treatment 2 Treatment 3
ggperimental 0 ppm 10 ppm 100 ppm

feed penicillin penicillin penicillin
‘61""""""’6T66§6T66 """"" :1.--_____-___--__::_-_
14 21.43121.43 8.0013.63 35.25122.35
28 8.5218.16 10.0315.90 16.2516.25

46 0.2510.25 5.3812.17 30.00111.97

All counts expressed as percent of total microorganisms 1
standard error.

Resistance expressed as growth on media containing 100

ug/ml penicillin.

# n = 2 for day 0; n = 4 for the other sampling dates.

Table 30. Analysis of variance for the
penicillin-resistant, anaerobic microorganisms isolated
from the jejunum of chickens fed different levels of
penicillin.

Source
of
var. d.f. M.S. f value
Treatment 2 383.51 0.76

Error 1 3 507.07

63

Table 31. The percent of anaerobic microorganisms resistant
to penicillin isolated from the cecum of chickens fed
different levels of penicillin.

Days Treatment 1 Treatment 2 Treatment 3
ggperimental 0 ppm 10 ppm 100 ppm
feed penicillin penicillin penicillin
'6E'"“""""6T66§6T66 """"" III"""'""""IIII'"
14 6.7513.12 0.0010.00 0.0010.00
28 7.0016.67 2.5010.65 2.5010.65
46 1.0011.00 2.0810.67 3.7512.17

All counts expressed as percent of total microorganisms 1
standard error.

Resistance expressed as growth on media containing 100

ug/ml penicillin.

# n = 2 for day 0; n = 4 for the other sampling dates.

Table 32. Analysis of variance for the
penicillin-resistant, anaerobic microorganisms isolated
from the cecum of chickens fed different levels of
penicillin.

Source
of
var. d f M.S. f value
Treatment 2 37.10 1.41

Error 1 3 26.33

64

Table 33. The percent of aerobic microorganisms resistant
to penicillin isolated from the feces of mink that were fed
the viscera of penicillin-fed chickens.

Days Treatment 1 Treatment 2 Treatment 3
ggperimental 0 ppm 10 ppm 100 ppm
feed penicillin penicillin penicillin
6;""""""'§T66_§§T66 """"" Iiim'mm""3333"“
7 l.0010.4l 7.0014.74 10.0016.47

7 days after 4.0312.64 6.2512.36 29.00112.29*

put back on

normal feed

All counts expressed as percent of total microorganisms 1
standard error.

Resistance expressed as growth on media containing 400

ug/ml penicillin.

# n = 2 for day 0; n = 4 for the other sampling dates.

* Significantly different from mean and treatment two at P<

0.05.

Table 34. Analysis of variance for the
penicillin-resistant, aerobic microorganisms isolated from
the feces of mink that were fed the viscera of
penicillin-fed chickens.

Source
of
var. d.f. M.S f value
Treatment 2 662.68 10.11*
Error 1 3 65.58

* Significant at P< 0.05.

65

fed. The percent of resistant bacteria in the feces in the
control group ranged from 11.5 to 23.9%. The mink in
treatments two and three fed showed an increase in
penicillin-resistant organisms with a range from 22.2 to
47.2% (Table 35). A significant increase in
penicillin-resistant microorganisms was found in treatment
three seven days after the experimental diet was withdrawn.
The data presented in Table 36 indicate that the increase
in resistant organisms was not significant at P< 0.05, but
the increase was significant at P< 0.10.

The experimental diets that were fed to the mink were
tested for the presence of penicillin. The data summarized
in Table 37 indicate that no penicillin was found in

detectable amounts in the diets.

Experiment four

A sample of 19, penicillin-resistant, bacterial colonies
from the aerobic, BHI-penicillin agar plates were randomly
selected for plasmid testing and comparison. Three of the
colonies were from the chicken cecum, five of the colonies
came from the diets prepared from the chicken viscera, and
11 of the samples were found in the feces of the mink fed
the experimental diets. Summarized in Table 38 are the
percent of different genera isolated from the plates. The

colonies were picked from isolated colonies and maintained

66

Table 35. The percent of anaerobic microorganisms resistant
to penicillin isolated from the feces of mink that were fed
the viscera of penicillin-fed chickens.

Days Treatment 1 Treatment 2 Treatment 3
:gperimental 0 ppm 10 ppm 100 ppm
feed penicillin penicillin penicillin
aim““"“3;j’;i;15i§;""“"'ZZZZ"“""“‘"ZZZZ """
7 11.5012.90 22.2514.31 22.5017.97

7 days after 18.001l.78 29.7513.01 47.25:11.92*

put back on

normal feed

All counts expressed as percent of total microorganisms 1
standard error.

Resistance expressed as growth on media containing 100

ug/ml penicillin.

# n = 2 for day 0; n = 4 for the other sampling dates.

* Significantly different from the mean P< 0.05.

Table 36. Analysis of variance for the
penicillin-resistant, anaerobic microorganisms isolated
from the feces of mink that were fed viscera from
penicillin-fed chickens.

Source
of
var. d f. M.S. f value
Treatment 2 813.80 7.24#
Error 1 3 112.37

# Significant at P< 0.10.

67

Table 37. The amount of penicillin found in mink diets
containing viscera of penicillin-fed chickens.

sample mm mg penicillin/kg sample
inhibition

standard one# l.0810.35 0.12

standard two 4.2310.28 0.50

diet 1, 0 ppm
penicillin <0.5010.00 <0.10

diet 2, 0 ppm
penicillin <0.5010.00 <0.10

diet 3, 0 ppm
penicillin <0.5010.00 <0.10

diet 4, 0 ppm
penicillin <0.5010.00 <0.10

diet 5, 0 ppm
penicillin <0.5010.00 <0.10

diet 6, 0 ppm
penicillin <0.5010.00 (0.10

# n = 6 for standards one and two; n = 2 for diets one to
six.
Numbers expressed as mean 1 standard error.

Table 38. The percent of different aerobic,
penicillin-resistant, bacterial genera randomly selected
from the total, penicillin-resistant, bacterial population.

Escherichia 5
Klebsiella

Enterobacter l
Edwardsiella 1
Proteus

Unknown

 

 

 

 

68

on BHI slants containing 50 ul/ml ampicillin (Gibco Labs.,
Grand Island, N.Y.). Thirteen of the 15

penicillin-resistant colonies tested by gel electrophoresis
showed evidence of having a plasmid.

Only two of the plasmids, that were isolated from
different colonies, were of the same size with each plasmid
containing approximately 6500 DNA base pairs as determined
by electrophoresis of the plasmid preparation in a 0.7%
agarose gel, see Figure l. The size of the plasmid was
determined by the comparison of the plasmid to a known
standard of lambda DNA cut with the restriction
endonuclease Hind III, (BRL, Gaithersburg, Md.).

One of the plasmids was isolated from an E. ggtt found
in the mink diet prepared from the viscera of the chickens
fed 100 ppm penicillin. The other plasmid was isolated from
an E. ggtt found in the feces of a mink fed a diet
containing viscera of broilers that were fed 100 ppm
penicillin.

Once the size of the two plasmids was determined to be
approximately equal, they were cut with two restriction
endonucleases. The two endonucleases used were EcoRl and
BglII. The base sequences where these two enzymes cut were
described in the Material and Methods. A picture of the
plasmids after cutting with EcoRl is shown in Figure 2. The
band sizes of the plasmids after cutting with EcoRl are

listed in Table 39. Plasmid one (isolated from the feces of

69

 

Figure 1. Photograph of an agarose gel electrophoresis of
two different plasmids and a standard plasmid.

Plasmid A isolated from mink feed containing viscera from
chickens fed 100 ppm penicillin.

Plasmid B isolated from mink fed a diet containing viscera
from chickens fed 100 ppm penicillin.

Plasmid S is standard lambda cut with Hind 3.

Gel is a 0.7% agarose gel run at 50 V for 60 minutes.

7O

 

Figure 2. Photograph of an agarose gel electrophoresis of
two plasmids cut with EcoRl.

Plasmid A isolated from mink feed containing viscera from
chickens fed 100 ppm penicillin.

Plasmid B isolated from mink fed a diet containing viscera
from chickens fed 100 ppm penicillin.

Plasmid S is standard lambda cut with Hind 3.

Gel is a 0.7% agarose gel run at 50 V for 75 minutes.

71

a mink) has three more bands than plasmid two (isolated
from the mink diet). The first band in plasmid one is not
plasmid DNA since it is larger than the plasmid DNA that
was isolated in the first run. The 4350 base pair band and
the 2500 base pair band found in plasmid one may just be an
artifact from an incomplete cut by the enzyme.

Listed in Table 40 are the band sizes found after
cutting the two plasmids with BglII. A picture of the
plasmids after cutting with BglII is shown in Figure 3. The
first band found in both runs is too large to be plasmid
DNA. The next three bands match both plasmids. The last
band seen in plasmid one is not seen in plasmid two
although this may be due to a quantity of DNA too small to
fluoresce enough to be seen.

The transformation experiments performed in this study

were inconclusive.

72

Table 39. Band sizes of two plasmids cut with EcoRl.

Band Plasmid 1* Plasmid 2**

number size in size in
base pairs base pairs

1 23000 6500

2 6500 2300

3 4350 1450

4 2500

S 2300

6 1500

* Plasmid l isolated from mink fed a diet containing
viscera from chickens fed 100 ppm penicillin.

** Plasmid 2 isolated from mink feed containing viscera
from chickens fed 100 ppm penicillin.

Table 40. Band sizes of two plasmids cut with BglII.

band Plasmid 1* Plasmid 2**

number size in size in
base pairs base pairs

1 7500 8000

2 6500 6500

3 5500 5500

4 3100 3100

5 1500

* Plasmid 1 isolated from mink fed a diet containing
viscera of chickens fed 100 ppm penicillin.
** Plasmid 2 isolated from mink feed containing viscera of

chickens fed 100 ppm penicillin.

73

 

Figure 3. Photograph of an agarose gel electrophoresis of
two plasmids cut with BglII.

Plasmid A isolated from mink feed containing viscera from
chickens fed 100 ppm penicillin.

Plasmid B isolated from mink fed a diet containing viscera
from chickens fed 100 ppm penicillin.

Plasmid S is standard lambda cut with Hind 3.

Gel is a 0.7% agarose gel run at 50 V for 75 minutes.

DISCUSSION

Experiment one

The use of penicillin for growth promotion has been
studied extensively. Most results indicate that broilers
will gain the same amount of weight on less feed when the
feed contains penicillin. The reason that this occurs
remains unclear. Three theories have been developed. One
theory suggests that the use of penicillin in the feed
causes a nutrient-sparing effect by killing bacteria in the
intestine that use essential nutrients needed by the
animal. Another theory suggests that penicillin causes a
metabolic change in an animal which allows a greater
percentage of nutrients to be used for growth. The third
theory suggests that the use of penicillin in low levels
kills or inhibits the growth of bacteria in the intestine
that cause subclinical infections. When these bacteria are
inhibited, the intestinal wall becomes thinner and the
nutrients are more easily absorbed (Murphy, 1975).

The lack of response of weight gain or feed conversion
in the broiler chicks fed penicillin in the first
experiment of this study supports the theory that
penicillin in the feed reduces subclinical infections by

74

75

killing undesirable bacteria in the alimentary tract. The
birds in the first experiment of the study were raised in
facilities that were cleaner than commercial facilities.
Also, strict quarantine procedures were enforced while the
birds were being raised. Finally, the birds were allowed
more floor space than in commercial facilities. This kept
the bedding dryer than normal. All of the above conditions
reduced stress in the birds and reduced the chance that a
subclinical infection would develop. These husbandry
methods support the findings of Lev gt gt. (1957) that
birds raised in "clean" facilities showed no growth
improvement when fed low levels of penicillin as compared

to birds not fed penicillin.
Experiment two

The results from experiment two showed that no
significant change in the numbers of aerobic bacteria
occurred in the jejunum of broilers which had been fed
penicillin at 10 ppm or 100 ppm. The variation in numbers
of aerobic bacteria in the jejunum may have masked any
changes in the number of bacteria due to the feeding of
penicillin. Marked variation in numbers of bacteria* is
commonly seen when working with intestinal bacteria. In
this experiment, an attempt was made to decrease the
variation in the numbers of bacteria by discarding the

intestinal contents and using the intestinal lining to

76

obtain the bacterial counts. When using this method, only
the bacteria that are closely associated with the lining of
the intestinal tract are counted.

The total, aerobic, bacterial counts in the jejunum
agree with the findings of Smith (1965). The coliform
counts were found to make up approximately 90% of the
total, aerobic, bacterial counts as expected.

The anaerobic bacteria in the jejunum increased through
the first four weeks of the bird's life and then leveled
off at approximately 106. No significant change in total,
anaerobic, bacterial numbers occurred when the birds where
fed penicillin.

The lactobacilli numbers did not change significantly
with the feeding of penicillin as found by Anderson gt gt.
(1951), but the lactobacilli reached high concentrations
faster in the treatments where penicillin was fed.

The Clostridium perfringens counts were below detectable
levels in the jejunum.

The bacterial counts were much higher in the cecum than
in the jejunum as expected. The total, aerobic, bacterial
counts in the cecum varied little between treatments or
time. The number of aerobic bacteria remained constant at

108 to 1010.

The coliform counts in the cecum of the
chicken made up between 80 to 100% of the total, aerobic,
bacterial counts. The research that has been published on

the effect of feeding penicillin to birds and the effect of

penicillin on coliform counts has indicated that the

77

coliform counts increased when penicillin is fed (Dubos gt
gt., 1963). Mamber and Katz (1985) saw no change in numbers
of enteric Gram-negative bacilli in chickens fed
antibiotics. A reason that this experiment did not show an
increase in coliform numbers may be due to the rearing
conditions discussed under experiment one.

The total, anaerobic, bacterial count in the cecum was
increased by the addition of penicillin to the feed at 10
or 100 ppm. This increase in anaerobic bacteria has been
noted before by Sieburth gt gt. (1951).

The sporadic and extremely variable counts of the
lactobacilli are typical of this genus (Dubos gt gt.,
1963). This variation makes a determination of a possible
differences in lactobacilli numbers between treatment
groups fed different levels of penicillin difficult.

The Clostridium perfringens numbers in the cecum did not
change significantly due to the feeding of penicillin. Lev
gt gt. (1957) reported that clostridia numbers may or may
not change when penicillin is fed to chickens.

The feeding of the broiler viscera from the different
treatments to the mink did not significantly change the
bacterial counts in the mink feces. This is not surprising
since the penicillin had been withdrawn from the chickens
three days prior to the processing of the birds. The mink
diets made with the broiler viscera showed no penicillin
present in detectable levels when tested.

Little information is available on the bacteria found in

78

the mink feces, but when compared to a cat, which is also a
carnivore, some similarities can be found. The number of
coliforms found in the mink feces was similar to the number
of coliforms found in cat feces. Both animals have
approximately 107 coliforms per gram of feces (Smith,
1961). The total, anaerobic population found in the mink
feces was only one percent of the total anaerobes found in
the cat feces. This difference may be due to inherent
differences in the species, but it may also indicate that
the method of collecting the mink feces may need to be done
in a shorter period, of time so that fewer obligate
anaerobes are killed due to the exposure to oxygen.

The lactobacilli counts made on the mink feces compares
closely with the number reported in the feces of cats. The

Clostridium perfringens counts in the mink feces were much

 

lower than what was reported for cat feces by Smith (1961).
The low counts may have been caused by exposure of the
bacteria to oxygen, but it may also have been caused by
keeping the fecal samples at 4C while the samples were

brought into the laboratory. Clostridium perfringens is

 

cold sensitive and a large number of them may have been

killed due to the temperature.

Experiment three

The results showed that no significant increase of

penicillin-resistant, aerobic bacteria occurred in the

79

jejunum of broilers that were fed either 10 or 100 ppm
penicillin in their diets. A significant difference in
resistant bacteria would be difficult to see due to the
large fluctuation of aerobic bacteria in the jejunum. Also,
the standard error was quite large. The large percentage of
resistant organisms seen at day 0 were found to be yeasts.

No significant increase in aerobic bacteria that were
resistant to penicillin at 400 ug/ml of media was seen in
the cecum of the chickens being fed ‘penicillin at either 10
or 100 ppm. The penicillin-resistant, aerobic bacteria
decreased within three days after the penicillin was
withdrawn from the feed, as seen from the 46 day sampling.
This indicates that the penicillin-resistant bacteria do
not survive as well as the wild-type bacteria. This may be
due to the inability of the resistant bacteria to adhere to
the intestinal lining as well as the wild-type bacteria
(Nordstrom gt gt., 1977). These results agree with the
results published by Langlois gt gt. (1984).

No significant differences in the percent of anaerobic
bacteria in the jejunum that were resistant to penicillin
were found.

The number of anaerobic bacteria resistant to penicillin
found in the cecum of the broilers was not significantly
different in the penicillin-fed treatment groups. The most
interesting feature of the resistant, anaerobic bacteria is
that they did not decrease within three days after the

penicillin was withdrawn from the broiler feed as did the

80

penicillin-resistant, aerobic bacteria. This may be
explained by the longer generation time of the anaerobic
bacteria.

The percent of aerobic bacteria resistant to penicillin
found in the feces of mink fed diets containing viscera of
chickens that had been fed penicillin increased over the
percent of the penicillin-resistant, aerobic bacteria found
in the feces of the mink in treatments one and two. The
percent of anaerobic bacteria resistant to penicillin found
in the feces of the mink increased significantly at P< 0.10
when the animals were fed diets containing the viscera of
chickens that had been fed penicillin. This level of
significance is important due to the variation in the
bacterial counts from the collection and handling methods
of the mink feces. The penicillin-resistant bacteria did
not decrease after the experimental diets had been removed.
The experimental feed was tested and no penicillin residue
was seen. The control diet of the mink was also tested and
no penicillin-resistant bacteria were found. These tests
show that any increase in penicillin-resistant bacteria
must be due to the presence of bacterial DNA from the

chickens.

Experiment four

The penicillin-resistant, aerobic bacteria isolated were

primarily Escherichia coli (58%), with a few Klebsiella

 

 

81

(5%), Enterobacter (16%), Edwardsiella (11%), and Proteus

 

 

(5%). Five percent of the isolated organisms could not be
identified. The plasmids isolated from the different groups
of birds, mink diets and the mink feces were all different
except for a plasmid isolated from a diet made of viscera
of chickens fed 100 ppm of penicillin and a plasmid
isolated from the feces of a mink that was fed a diet made
from viscera of chickens fed 100 ppm of penicillin. The
weights of these two plasmids were identical at 6500 base
pairs. When the plasmids were cut with the two restriction
endonucleases some differences were noted, but the patterns
of the cuts were similar for both restriction
endonucleases. This indicates that the plasmids may be
identical. The identical weight obtained with the first
electrophoretic analysis would have been used as conclusive
evidence that the plasmids were identical in the mid 1970's
as noted in the article by Meyers gt gt. (1976). The use of
two restriction endonucleases to further identify the
plasmids was suggested by Farrar (1983). The tests
performed in this study meet these two qualifications, but
further testing for homologous base pair sequences by tests
such as the Southern blot technique (Coussens, 1988) is
needed before a positive determination can be made.

The presence of a similar plasmid in the two different
samples does not indicate that a transfer of the plasmid
between two bacteria occurred. The bacteria in which the
plasmids were isolated may be the same bacterium but found

in two different locations.

CONCLUSIONS

1. The feeding of penicillin at 10 ppm or 100 ppm to
broilers does not improve feed efficiency if the birds are

raised in clean and stress-free facilities.

2. The feeding of penicillin to broilers at 10 ppm and 100
ppm does not significantly change the number of total
aerobic bacteria, total coliforms, total anaerobes,
lactobacilli, or Clostridium perfringens in the jejunum or

in the cecum of the broilers.

3. The coliform, lactobacilli, and Clostridium perfringens
counts in mink feces are similar to the coliform,

lactobacilli, and Clostridium perfringens counts taken in

 

the feces of cats.

4. The number of penicillin-resistant, aerobic bacteria
dropped in both the jejunum and in the cecum of chickens
fed penicillin within three days after the penicillin was
removed from the feed. The penicillin-resistant, aerobic
bacteria do not survive as well as the wild-type,
penicillin-sensitive bacteria without selective pressure.

The anaerobic bacteria resistant to penicillin in the

82

83

constant for at least three days after the penicillin is
removed from the feed. A longer withdrawal period needs to
be tested to see if the number of penicillin-resistant,
anaerobic bacteria will also decrease similar to the

aerobic bacteria.

5. The percent of penicillin-resistant, aerobic bacteria
increased significantly (P< 0.05) in mink feces when the
mink were fed the viscera of broilers that had been fed
penicillin. The percent of anaerobic bacteria resistant to
penicillin found in the feces of mink also increased
significantly (P< 0.10). The increase in resistance in the
mink feces was not dependent on the mink being exposed to

penicillin in this study.

6. Similar plasmids were found in both mink feed and in
mink feces. The plasmids from the two different samples had
a similar weight and were cut in a similar pattern by two
different restriction endonucleases. More testing needs to
be performed on the plasmids to determine if they are

identical.

APPENDICES

84

Appendix A

Formulation of MSU broiler feeds.

1. MSU broiler starter.

Corn 951.8 lbs
Soybean meal 44% 855.0 lbs
Calcium carbonate 23.0 lbs
Dicalcium phosphate 36.2 lbs
Salt 10.0 lbs
Vitamin-Mineral premix 10.0 lbs
Corn oil 110.0 lbs
Methionine 4.0 lbs
2. MSU broiler grower.

Corn 1066.8 lbs
Soybean meal 44% 741.6 lbs
Calcium carbonate 22.4 lbs
Dicalcium phosphate 36.2 lbs
Salt 10.0 lbs
Vitamin-Mineral premix 10.0 lbs
Corn oil 110.0 lbs

Methionine

2.0

lbs

85

Appendix B

Solutions used for microbiological methods.

1. Siver sulfafe solution

AgSOLp 0.5 gm

H Ou,1.0M 1.0 m1

g ycerol 100.0 m1

d H20 100.0 m1
2. Phosphate buffered saline

NagHPOh 1.236 gm
NaHgPOh 0.180 gm
NaCl 8.500 gm

H20 1000.000 m1

86

Appendix C

Solutions used for plasmid isolation.

1. STET

Tris 1.0M pH. 8.0 25 m1
EDTA 0.5M 50 ml
Triton X-100 25 m1
Sucrose 40 gm
dd. H O 400 ml
2. EDTA

Disodium ethylene diamine tetraacetate : 2H 0 186.1 gm
dd. H20 800.0 m1
NaOH 20.0 gm

3. Lysozyme solution

lysozyme 10.0 mg
Tris 250mM pH 8.0 1.0 m1
4. TE

Tris 1.0M pH 8.0 1.0 m1
EDTA 0.5M pH 8.0 0.2 ml
dd. H20 98.8 ml

5. Phenol:forma1dehyde solution

Phenol 25 m1
CHCl 24 m1
Isoamyl alcohol 1 m1

6. Five—x loading buffer

Bromophenol blue 0.25%
Xylene cyanol 0.25%
Ficoll(type 400) in H20 12.50%

7. Formaldehyde solution
CHC13 24 ml
Isoamyl alcohol 1 m1

BIBLIOGRAPHY

BIBLIOGRAPHY

Anderson, G.W., J.D. Cunningham and S.J. Slinger, 1951.
Effect of various protein levels and antibiotics on the
intestinal flora of chickens. Poultry Sci. 30:905.

A.O.A.C., 1984. Official Methods gt Analysis. 14th ed.
Williams, S. (ed.). Assoc. of Official Analytical
Chemists, Arlington. 1141 pp.

 

Aulerich, R.J., 1987. Personal communication.

Barnes, E.M. and H.S. Goldberg, 1962. The isolation of
anaerobic gram-negative bacteria from poultry reared
with and without antibiotic supplements. J. Appl. Bact.
25:94—106.

Barnes, E.M., G.C. Mead, D.A. Barnum and E.G. Harey, 1972.
The intestinal flora of the chicken in the period 2 to 6
weeks of age, with particular reference to the anaerobic
bacteria. Br. Poultry Sci. 13:311-326.

Beck, J.R., 1978. Uric acid-producing anaerobic bacteria of
the chicken cecum. Ph.D. Thesis. Michigan State
University, East Lansing, Mi. 144 pp.

Betina, V., 1983. In: The Chemistry, and Biology gt
Antibiotics. Elsevior Scientific Publishing. Amsterdam.
590 pp.

 

 

 

Bird, H.R., 1969. Biological basis for the use of
antibiotics in poultry feeds. In: Egg Egg gt Drugs tg
Animal Feeds. Nat. Acad. Sci., Washington D.C. Pub.
1679, 407 pp.

Bleavins, M.R. and R.J. Aulerich, 1980. Feed consumption
and food passage time in mink (Mustela vison) and
European ferrets (Mustela putorius furo). Lab. Animal
Sci. 31:268-269.

 

Braude, R., 1978. Antibiotics in animal feeds in Great
Britain. J. Animal Sci. 46:1425—1436.

Braude, R. and B.C. Johnson, 1953. Effect of aureomycin on
nitrogen and water metabolism in growing pigs. J.

87

88

Carlton, B.C. and B.J. Brown, 1981. Gene mutation. In:
Manual gt Methods for General Bacteriology. Gerhardt,
P., R.G.E. Murray, R.N. Costilow, E.W. Nester, W.A.
Wood, N.R. Krieg, and G.B. Phillips. (eds.). American
Society for Microbiology, Washington D.C. 524 pp.

Chang, T.S. and D.W. Murphy, 1975. The effect of
tetracycline and bacitracin on intestinal microflora.
Proceeding, A.V.M.A. Annual convention. July 14, 1975.
Anaheim.

Cherubin, C.E., J.F. Timoney, M. Sierra, P. Ma, J. Marr and
8. Shin, 1980. A sudden decline in ampicillin resistance
in Salmonella typhimurium. J. Am. Med. Assoc.243:439-442.

Cohen, M.L. and R.V. Tauxe, 1986. Drug-resistant Salmonella

in the United States: an epidemiologic perspective.
Science 234:964-969.

Combs, G.F. and E.H. Bossard, 1963. Comparison of growth
response of chicks to virginiamycin and other
antibiotics. Poultry Sci. 42:681-685.

Costerton, J.W. and K.J. Cheng, 1975. The role of bacterial
cell envelope in antibiotic resistance. J. Antimicrob.
Chemother. 1:363-377.

Coussens, P., 1988. Personal communication.

Crawford,L.M. and E.B. Shotts, 1982. Animal uses of
antibiotics as feed additives and the emergence of
antibiotic resistance. In: Egg Control gt
Antibiotic-resistant Bacteria. Stuart-Harris, C.H. and
D.M. Harris (eds.). Academic Press,Inc.,New York. 284 pp.

Dafwang, I.I., H.R. Bird and M.L. Sunde, 1984. Broiler
chick growth response to antibiotics, 1981-1982. Poultry
Sci. 63:1027-1032.

Davis, B.D. and R. Dulbecco, 1973. Bacterial and molecular
genetics. In: Microbiology. Davis, B.D., R. Dulbecco,
H.N. Eisen, H.S. Ginsberg, and W.B. Wood, Jr. (eds.).
Harper and Row, Hagerstown. 1562 pp.

Davis, J.G., 1971. Drop method for microbial counts.
Process Biochemistry. May, 31-33.

Dawson, K.A., B. Langlois, T. Stahly and G. Cromwell, 1984.
Some characteristics and antibiotic resistance of
anaerobic bacteria from the ceca and colons of pigs fed
chlortetracycline-containing and unmedicated diets.
Appl. Environ. Microbiol. 47:210-212.

89

Difco Laboritories, 1968. Supplementary literature. Difco
laboritories, Detroit, Michigan. 478 pp.

Difco Laboritories, 1953. Difco manual. Difco Laboritories,
Detroit, Michigan. 350 pp.

Dubos, R., R.W. Schaedler and R. Costello, 1963.
Composition, alteration, and effects of the intestinal
flora. Fed. Proc. 22:1322-1329.

Elwell, L.P. and S. Falkow, 1986. The characterization of R
plasmids and the detection of plasmid-specified genes.
In: Antibiotics tg Laboratory Medicine. Lorain, V.,
(ed.). Williams and Wilkins, Balt. 1259 pp.

 

Falkow, S., 1975. Infectious Multiple Drug Resistance.
Pion, London. 300 pp.

 

Farrar, W.E. Jr., 1983. Molecular analysis of plasmids in
epidemiologic investigation. J. Infec. Dis. 148:1-6.

Gast, R.K. and J.F. Stephens, 1988. Transmission of
drug-resistant Salmonella through the food chain. Ohio
Poultry Pointers. 26:9.

Gill, J.L., 1978. Design and Analysis gt Experiments tg the
Animal and Medical Sciences. Vol. 1. Iowa State
University Press. Ames. 409 pp.

 

 

 

Hays, V.W., 1969. Biological basis for the use of
antibiotics in livestock production. In: The Use gt
Drugs tg Animal Feeds. Nat. Acad. Sci. Pub. 1679, 407 pp.

 

Hays,V.W., 1981. Antibiotics in animal feeds. Council for
Agricultural Science and Technology. Report no.88. 79 pp.

Heuser,G.F. and L.C. Norris, 1952. Some results of feeding
antibiotics to chickens. Poultry Sci. 31:857-862.

Holmberg,S.D., M. Osterholm, K. Senger and M. Cohen, 1984.
Drug resistant Salmonella from farm animals fed
antimicrobials. N. Engl. J. Med. 311:617-622.

Hummel, R., H. Tschape and W. White, 1986. Spread of
plasmid mediated nourseothricin resistance due to

antibiotic use in animal husbandry. J. Basicmicrobiol.
26:461-466.

Jukes, T.H., 1971. Antibiotics and animal feeds - An
appraisal after twenty years of use. Proc. Aspects of
Infective Drug Resistance. 12 pp.

9O

Kiser, J.S., G. Kemp and H. Jarolmen, 1970. Medicated feeds
and transferable antibiotic resistance. Feedstuffs.
42(13):28.

Langlois, B.E., K. Dawson, T. Stahly and G. Cromwell, 1984.
Antibiotic resistance of fecal coliforms from swine fed
subtherapeutic and therapeutic levels of
chlortetracycline. J. Animal Sci. 58:666-674.

Lederberg, J. and E.M. Lederberg, 1951. Replica plating and
indirect selection of bacterial mutants. J. Bact.
63:399-406.

Lev, M., C.A.E. Briggs and M.E. Coates, 1957. The gut flora
of the chick 3. Differences in caecal flora between
infected, uninfected and penicillin-fed chicks. Br. J.
Nutr. 11:364-373.

Levy, S.B., 1978. Emergence of antibiotic resistant
bacteria in the intestinal flora of farm inhabitants. J.
Infect. Dis. 137:688-690.

Levy,S.B., 1987. Antibiotic use for growth promotion in
animals: Ecologic and public health consequences. J.
Food Protection. 50:616-620.

Levy,S.B., G.B. FitzGerald and A.B. Macone, 1976a. Changes
in intestinal flora of farm personnel after introduction
of a tetracycline-supplemented feed on a farm. N. Engl.
J. Med. 295:583-588.

Levy,S.B., G.B. FitzGerald and A.B. Macone, 1976b. Spread
of antibiotic resistant plasmids from chicken to chicken
and from chicken to man. Nature. 260:40-42.

Maas, W.K., 1986. Mutations to antibiotic resistance. In:
Antibiotics tg Laboratory Medicine. Lorain, V., (ed.).
Williams and Wilkins, Balt. 1259 pp.

 

Mamber, S.W. and S.E. Katz, 1985. Effects of antimicrobial
agents fed to chickens on some gram-negative enteric
bacilli. Appl. Environ. Microbiol. 50:638-648.

Maniatis, T., E.F. Fritsch and J. Sambrook, 1982. Molecular
Cloning‘ A Laboratory Manual. Cold Spring Harbor
Laboratory Publ. Spring Harbor, N.Y. 545 pp.

 

 

Menge, H., 1973. Lack of growth response of eight week old
broilers to certain antibiotics. Poultry Sci.
52:1891-1895.

91

Meyers,J.A., D. Sanchez, L.P. Elwell and S. Falkow, 1976.
Simple agarose gel electrophoretic method for the
identification and characterization of plasmid
deoxyribonucleic acid. J. Bact. 127:1529-1537.

Miles, A.A. and 8.8. Misra, 1938. The estimation of the
bactericidal power of blood. J. Hygiene. 38:732-749.

Murphy, D.W., 1975. Effects of antibiotics on intestinal
microflora, growth and feed efficiency of the chicken.
Ph.D. Thesis. Michigan State University, East Lansing,
Michigan. 126 pp.

Nelson, F.E., L.S. Jensen and J. McGinnis, 1963. Studies on
the stimulation of growth by dietary antibiotics.
Poultry Sci. 42:906-909.

Neu,H.C., 1986. Antibiotic inactivating enzymes and
bacterial resistance. In: Antibiotics tg Laboratory
Medicine. Lorain, V., (ed.). Williams and Wilkins, Balt.
pp.757-789.

 

 

Nordstrom, K., B. Engberg, P. Gustafsson, S. Molin and B.
Uhlin, 1977. In: Topics tg Infectious Diseases! Vol. 31
E Factors: Their Properties and Possible Control. Drews,
J. and G. Hogenauer, (eds.). Springer-Verlag, N.Y. 362

PP-

 

 

 

Richmond, D. and T.S. Chang, 1978. A comparison of
drop-plate and pour plate methods for bacterial
population counts of poultry anaphage (dehydrated caged
layer excreta). Poultry Sci. 57:293-295.

Richmond, M.H., 1977. The survival of R plasmids in the
absence of antibiotic selection pressure. In: Topics tg
Infectious Diseases, Vol. EL E Factors: Their Properties
ggg Possible Control. Drews, J. and G. Hogenauer,
(eds.). Springer-Verlag, N.Y. 362 pp.

 

Salanitro, J.P., I.G. Fairchilds and Y.D. Zgornicki, 1974.
Isolation, culture characteristics and identification of

anaerobic bacteria from the chicken cecum. Appl.
Microbiol. 27:678-687.

Sieburth, J.M., J. Gutierrey, J. McGinnes, J.R. Stern and
B.H. Schneider, 1951. Effect of antibiotics on
intestinal microflora and on growth of turkeys and pigs.
Proc. Soc. Exp. Biol. Med. 76:15-18.

Smith, A.A. and C.G. Krasulak, 1979. Manual of mink
anatomy. Burgess Publishing Co. Minneapolis. 117 pp.

92

Smith, H.W., 1961. The development of the bacterial flora
of the faeces of animals and man: The changes that occur
during ageing. J. Appl. Bact. 24:235-241

Smith, H.W., 1965. The development of the flora of the
alimentary tract in young animals. J. Path. Bact.
90:495-513.

Spika, J.S., S. Waterman, G. 800 Hoo, M. St. Louis, R.
Pacer, S. James, M. Bissett, L. Mayer, J. Chin, B. Hall,
K. Greene, M. Potter, M. Cohen and P. Blake, 1987.
Chloramphenicol resistant Salmonella newgort traced
through hamburger to dairy farms. N. Engl. J. Med.
316:565-570.

 

Stallones, R.A., E.R. Alexander, C.E. Antle, P. Gardner,
E.H. Kass, C.A. Keller, J.M. Lane, F.J. Massey,Jr.,R.H.
Rownd, P.R. Sheehe, and V.L. Tharp, 1980. The effects on
human health of subtherapeutic use of antimicrobials in
animal feeds. N.A.S. Washington D.C. 376 pp.

Stewart, G.T., 1965. In: The Penicillin Group gt Drugs.
Elsevier Publishing.Amsterdam. 212 pp.

 

Takahashi, S. and Y. Nagano, 1984. Rapid proceedure for
isolation of plasmid DNA and application to
epidemiological analysis. J. Clin. Micro. 20:608-613.

Tomasz, A., 1979. The mechanism of the irreversible
antimicrobial effects of penicillins: How the beta
lactam antibiotics kill and lyse bacteria. Ann. Rev.
Microbiol. 33:113-137.

V.P.I., 1973. Anaerobe Laboratory Manual. Holdeman, L.V.
and W.E.C. Moore (eds.). V.P.I. Anaerobe Laboratory,
Blacksburg. 132 pp.

 

Wallace, H.D., 1970. Biological responses to antibacterial
feed additives in diets of meat producing animals. J.
Animal Sci. 31:1118-1126.

Willetts, N.S., 1972. The genetics of transmissible
plasmids. Annu. Rev. Genet. 6:257-268.

"‘1i11011011“