Screening alternative antibiotics against oxytetracycline-susceptible and -resistant Paenibacillus larvae
Screening alternative antibiotics against oxytetracycline-susceptible and -resistant Paenibacillus larvae
n KOCHANSKY 0
A. KNOX 0
rk FELDLAUFER 0
Jeffery S. PETTIS 0
0 USDA ARS Bee Research Laboratory , Building 476, 10300 Baltimore Avenue, Beltsville, MD 20705-2350 , USA
- Since resistance of the causative organism of American foulbrood, Paenibacillus larvae subsp. larvae, to oxytetracycline (OTC) is becoming widespread in the United States, we began a search for effective alternative antibiotics. We investigated the sensitivity of P. l. larvae to 27 antibiotics, which were primarily ones already registered with the US Food and Drug Administration for agricultural uses. Bacterial resistance to OTC also conferred resistance to other tetracyclines, although the level of resistance varied. The most active antibiotics screened that are currently used in agriculture were erythromycin, lincomycin, monensin, and tylosin. Rifampicin was by far the most active antibiotic tested, but since it is used against tuberculosis, registration of this material for agricultural use is unlikely. Apis mellifera / Paenibacillus larvae larvae / antibiotic / resistance / American foulbrood
American foulbrood (AFB) and Euro
pean foulbrood (EFB) are two bacterial
diseases of honey bee brood, caused by
Paenibacillus larvae subsp. larvae (formerly
Bacillus larvae) and Melissococcus pluton,
respectively. AFB has traditionally been
controlled by burning infected colonies, and
this practice is still required in some areas.
In order to lessen this financial loss to
beekeepers, pharmacological methods for
foulbrood prevention and control were
The first medications tested were
synthetic antibacterials: the sulfa drugs,
Childers, 1944; Eckert, 1947; Reinhardt,
1947; Johnson, 1948; Katznelson and
Gooderham, 1949; Katznelson, 1950)
were effective against AFB, but their
stability and consequent residues in honey
caused problems, and the registration was
allowed to lapse in the 1970s (Shimanuki
and Knox, 1994).
Tests on antibiotics began in the late
1940s as well
(cited by Katznelson, 1950)
These early studies showed that
Aureomycin1 (chlortetracycline) was active, while
penicillin, chloramphenicol, streptomycin,
and others were considerably less active
against P. l. larvae. Oxytetracycline (OTC,
Terramycin1), usually as its hydrochloride,
has been used since the early 1950s for the
prevention and control of AFB and EFB
(Gochnauer, 1951; Katznelson et al., 1952)
It remains after many years the only
approved drug treatment for the foulbrood
diseases in the United States. Recently,
however, strains of P. l. larvae showing
resistance to OTC have been discovered in
Argentina (Alippi, 2000) as well as in
multiple areas in the United States
et al., 2000)
, and there is now general
concern about widespread resistance.
Other antibiotics have been investigated
as treatments for AFB. Tylosin (a macrolide
antibiotic) was first investigated for control
of AFB about thirty years ago
et al., 1970; Moffett et al., 1970; Peng
et al., 1996)
and was found to be much more
stable in sugar syrup than was OTC
(Kochansky et al., 1999)
(another macrolide) was first tested in 1955
(Katznelson et al., 1955; Katznelson, 1956;
see also Hitchcock, 1964)
, but reports on
its efficacy differed.
Katznelson et al. (1955)
Moffett et al. (1958)
found it to be
ineffective against AFB, but
found good efficacy against AFB, and it has
been reported to be effective against EFB
(Wilson and Moffet, 1957; Wilson, 1962)
Machova (1970) also tested other
antibiotics in a study of antibiotic sensitivity of
ten strains of P. l. larvae isolated from
various regions of Czechoslovakia. In addition
to the activity of erythromycin cited above,
good sensitivity was obtained to bacitracin
and the tetracyclines, with lower
sensitivity to 10 others.
A large number of P. l. larvae strains in
Japan were tested for antibiotic sensitivity
(Okayama et al., 1996)
another macrolide antibiotic, has been
studied as a result of its high activity in this
(Nakajima et al., 1998)
Ampicillin was another antibiotic with high
activity in vitro, but when tested in beehives
it gave high residues in honey but only very
low levels in larvae, casting doubt on its
utility in disease control
(Nakajima et al.,
In view of the increasing incidence of
OTC-resistant P. l. larvae in the US, we
embarked on a screening program to identify
additional antibiotics that might be useful
in the prevention and treatment of AFB. We
now report the results of laboratory tests
evaluating a variety of antibiotics for ability
to inhibit both OTC-resistant and
OTC-susceptible strains of P. l. larvae.
2. MATERIALS AND METHODS
Antibiotics (Tab. I) were purchased from
Sigma or Fluka and a stock solution
(200 mg/L in 50% methanol) was prepared
for each. Successive dilutions were prepared
to yield 60, 20, 6, 2, 0.6, and 0.2 mg/L
(rifampicin was also tested at 0.06 mg/L).
Antibiotic test disks (6.35 mm, No. 740-E,
1 Mention of trade names or commercial products in this article is solely for the purpose of
providing specific information and does not imply recommendation or endorsement by the US
Department of Agriculture.
Schleicher and Schuell), which had been
shown gravimetrically to absorb 0.02 g =
20 m L of water, were dipped in the antibiotic
solution to saturation and allowed to dry.
This resulted in a series of disks containing
4, 1.2, 0.4, 0.12, 0.04, 0.012, and 0.004 m g
of each antibiotic/disk.
Antibiotics were tested against P. l. lar
vae by our standard disk diffusion method
(Shimanuki and Knox, 1991)
, similar to the
method reported by
Feldlaufer et al. (1993)
The susceptible strain of P. l. larvae was
obtained from New Jersey and an
OTCresistant strain was obtained from
Minnesota. A stock spore suspension
(approximately 2 · 108 spores/mL) of each strain
was prepared by mixing 3–5 scales (the
dried remains of diseased honey bee larvae
containing the bacterial spores) with sterile
water (9 mL) in a screw-capped tube. Before
each use, the suspension was heat-shocked
at 80 °C for 10 minutes to kill any
nonsporeforming bacteria. For the bioassay,
0.2 mL of the stock suspension was spread
over the surface of freshly-prepared
brainheart infusion agar (BHI) plates (brain-heart
infusion, Difco Laboratories, Detroit, MI,
fortified with thiamine hydrochloride
(0.1 mg/L), 2% agar, and adjusted to pH 6.6
with hydrochloric acid). The
antibiotictreated disks were positioned in the center of
the BHI plates and the plates were incubated
at 34 °C in the dark. The diameters of the
zones of inhibition were measured after
72 hours. We did not make any correction
for disk diameter. If there was a visible zone
of inhibition around the disk, the total
diameter of that zone was recorded and the
compound was termed ‘active’ at that
concentration; if no such zone was visible, the
diameter was recorded as zero and the
compound was therefore ‘inactive’. Negative
controls consisted of disks treated with 50%
methanol, and commercial sensitivity test
disks (BBL, Becton Dickinson
Microbiology Systems, Cockeysville, MD) treated
with 5 m g of tetracycline were used as
The tests against tetracyclines in Table II
and in the general survey in Table III were
run only once, since only an indication of
the most active antibiotics was desired. The
tests of selected antibiotics against resistant
and susceptible AFB reported in Table IV
were replicated 3–4 (mostly 4) times and
are reported as mean ± standard error. Data
reduction and plotting were performed with
GraphPad Prism ver. 3.0 (GraphPad
Software, Inc. San Diego, CA).
Numbers in table are diameters of inhibition zones around the treated disks at the various concentrations, tested
against susceptible (New Jersey) P. l. larvae.
Amikacin, apramycin, colistin, kanamycin, neomycin, paromomycin, polymixin, spectinomycin, and
streptomycin were inactive at all concentrations tested.
3. RESULTS AND DISCUSSION
Since OTC is a member of the extended
tetracycline class of antibiotics, we tested
other members of the class against
susceptible and resistant P. l. larvae, with results
shown in Table II. While some tetracyclines
were more effective than others against
resistant P. l. larvae, in all but one case
tetracycline antibiotics were less effective against
resistant than against susceptible AFB
organisms, indicating considerable
crossresistance. Doxycycline was still
moderately active against resistant P. l. larvae, but
only minocycline showed no apparent loss
of activity against the resistant strain. It is not
unlikely that this cross-resistance would
increase rapidly under selection conditions,
and since minocycline has no approved
agricultural uses in the US, we did not
investigate tetracyclines further.
Twenty-one additional antibiotics (sum
marized in Tab. I) were screened against
susceptible P. l. larvae using the same
dosage series. The results are shown in
Table III. These antibiotics represented
8 classes. Aminoglycoside antibiotics were
either completely inactive at all
concentrations tested (amikacin, kanamycin,
neomycin, paromomycin, or streptomycin)
or active only at 4 m g/disk (gentamicin) or
0.4 m g/disk (tobramycin), and were not
tested further. Aminocyclitol antibiotics are
similar to aminoglycosides structurally, but
lack sugar groups bound to the rings. In this
class apramycin and spectinomycin were
investigated and both were found to be
inactive at all concentrations tested.
Chloramphenicol, in its own class, was only active
down to 1.2 m g/disk and was not tested
Of the three peptide antibiotics screened,
bacitracin was active down to 0.4 m g/disk,
while polymixin and colistin were inactive;
none was tested further. Vancomycin, a
glycopeptide antibiotic widely used in human
medicine against organisms resistant to
other antibiotics, was only active down to
0.4 m g/disk.
The remaining antibiotics fell into four
categories. The macrolides were represented
by tylosin, erythromycin, and oleandomycin,
the lincosaminides by lincomycin and
clindamycin, the polyether ionophores by
monensin, and the ansamacrolides by rifampicin.
Oleandomycin and clindamycin were only
active down to 1.2 and 0.4 m g/disk,
respectively, and were not tested further, but
all others were highly active (down to
0.04 m g/disk or below).
The five most active antibiotics from
these preliminary screens with the
susceptible strain of P. l. larvae were tested along
with OTC against susceptible and resistant
strains of P. l. larvae (Tab. IV). As expected,
OTC performed very poorly against resistant
bacteria. Monensin was the least active of
the five, down to 0.12 m g/disk.
Erythromycin, tylosin, and lincomycin were
active down to 0.04 m g/disk.
et al. (1996)
also found tylosin, er thromycin,
and lincomycin to be active, with generally
high activity from other macrolides,
including oleandomycin, which we found to have
poor activity. Okayama et al. found good
activity from the three tetracyclines they
tested (OTC, chlortetracycline, and
doxycycline), which suggests that none of their P.
l. larvae strains was OTC-resistant. As
before, rifampicin was outstanding, with
activity still present at 0.0012 m g/disk, the
lowest dose tested. For the initial screen,
extrapolation of the log (concentration) vs.
inhibition zone diameter line suggested a
minimum inhibitory concentration for
rifampicin of only 1.8 m g/L (36 pg/disk). In
all cases, these five selected antibiotics were
equally active against resistant and
susceptible foulbrood. With the exception of
rifampicin, all are listed in the FDA “Green
Book” (FDA-CVM, 1999) of drugs
approved for veterinary use.
Field tests are currently under way using
tylosin and lincomycin in colonies against
AFB. Preliminary results (to be further
described elsewhere) suggest that lincomycin
and tylosin are effective against resistant
AFB strains in the field in both Florida and
New Jersey, while we have not yet tested
erythromycin in the field.
Résumé – Criblage d’antibiotiques alter
natifs contre Paenibacillus larvae
résistant et sensible à l’oxytetracycline. Puisque
Paenibacillus larvae subsp. larvae
(anciennement Bacillus larvae), l’agent de la loque
américaine, est devenu résistant à
l’oxytetracycline (OTC) sur une grande échelle aux
États-Unis, nous avons recherché des
antibiotiques alternatifs qui pourraient
efficacement remplacer l’OTC. Nous avons étudié
principalement les antibiotiques déjà
autorisés par la Food and Drug Administration
pour des usages agricoles.
La méthode de diffusion sur disques a été
utilisée pour faire un premier criblage
d’antibiotiques à des concentrations
successives décroissantes à partir de 200 mg/L
(4 m g/disque) : 4, 1,2, 0,4, 0,12, 0,04 et
0,12 m g/disque. Seules les substances qui
étaient actives à une concentration au moins
égale à 6 mg/L (0,12 mg/disque) ont été
testées par la suite. Dans la plupart des cas la
résistance à l’OTC conférait, avec un niveau
variable, une résistance aux autres
tetracyclines. Seule la minocycline était active
contre les souches de P. l. larvae résistantes
et les souches sensibles. Les antibiotiques
testés les plus actifs (ceux actifs aux doses
£ 0,04 m g/disque) couramment utilisés en
agriculture étaient : l’érythromycine, la
lincomycine, la monensine et la tylosine. La
rifampicine, bien que non utilisée en
agriculture, était de loin l’antibiotique le plus
actif, avec une activité étant encore présente
à 0,0012 m g/disque mais, étant utilisée contre
la tuberculose, il est peu vraisemblable
qu’elle reçoive une autorisation pour un
usage agricole. La lyncomicine, la tylosine
et l’érythromycine étaient actives jusqu’à
0,04 m g/disque et la monensine était la
moins active avec une concentration
minimum d’inhibition de 0,12 m g/disque.
Apis mellifera / loque américaine /
antibiotique / résistance / Paenibacillus
Zusammenfassung – Prüfung von
alternativen Antibiotika gegen
Oxytetrazyklinempfindliche und -resistente Stämme
von Paenibacillus larvae larvae. Da sich
oxytetrazyklinresistente Stämme der
Amerikanischen Faulbrut in den USA
verbreiten, begannen wir mit der Suche nach
alternativen Antibiotika als wirksamen Ersatz.
Wir prüften vor allem Antibiotika, die
bereits für eine Nutzung im
landwirtschaftlichen Bereich zugelassen sind. In
Vorversuchen wurden Konzentrationen von
Antibiotika mit einer Diffusionsmethode unter
Verwendung von Antibiotika Testblättchen
erstellt, die von 200 mg/L (4
mg/Testblättchen) ausgehend in einer Reihe von
Verdünnungsstufen von 4, 1,2, 0,4, 0,12, und
0,012 mg/disk weitergeführt wurden. Nur
Substanzen, die bei einer Konzentration von
6 mg/L (0,12 mg/disk) und geringer
wirksam waren, wurden weiterhin getestet. Fast
immer war die OTC Resistenz mit
Resistenzen zu anderen Tetrazyklinen
verbunden, der Grad der Resistenz variierte, nur
Minozyclin erwies sich als gleich wirksam
gegen resistente und empfindliche Stämme
von P. l. larvae. Die wirksamsten und zur
Zeit in der Landwirtschaft genutzten
Antibiotika (wirksam bei 0,04 mg/disk) waren
Erythromyzin, Lincomyzin, Monensin und
Tylosin. Rifampicin, das nicht in der
Landwirtschaft genutzt wird, war das bei weitem
aktivste Antibiotikum im Test. Es war noch
in einer Konzentration von 0,0012 mg/disk
wirksam. Lincomyzin, Tylosin und
Monensin waren weniger wirksam als die anderen 4
(geringste Konzentration für eine Hemmung
war 0,12 mg/Scheibe).
Antibiotika / Resistenz / Amerikanische
Faulbrut / Paenibacillus larvae larvae
Alippi A.M. ( 2000 ) Is Terramycin® losing its effectiveness against AFB? Bee Biz 27-29 .
Eckert J.E. ( 1947 ) Use of sulfa drugs in the treatment of American foulbrood disease of honeybees , J. Econ. Entomol . 40 , 41 - 44 .
Feldlaufer M.F. , Lusby W.R. , Knox D.A. , Shimanuki H. ( 1993 ) Isolation and identification of linoleic acid as an antimicrobial agent from the chalkbrood fungus, Ascosphaera apis , Apidologie 24 , 89 - 94 .
Food and Drug Administration-Center for Veterinary Medicine ( 1999 ) FDA approved animal drug list ('The Green Book') . Drug Information Laboratory, College of Veterinary Medicine, Virginia Tech . University, Blacksburg, VA (available online at http://www.fda.gov/cvm/fda/greenbook.html). The online version of November 21 , 2000 was checked for latest registration status of all antibiotics .
Gochnauer T.A. ( 1951 ) Drugs fight foulbrood diseases in bees , Minn. Home Fm. Sci. 9 , 15 .
Haseman L. , Childers L.F. ( 1944 ) Controlling American foulbrood with sulfa drugs , Univ. Missouri Agric. Exp. Sta. Bull. 482 , 3 - 16 .
Hitchcock J.D. ( 1964 ) Effect of two erythromycin formulations on the longevity of the honey bee , J. Insect Pathol . 6 , 408 - 410 .
Hitchcock J.D. , Moffett J.O. , Lackett J.J. , Elliott J.R. ( 1970 ) Tylosin for control of American foulbrood disease in honey bees , J. Econ. Entomol . 63 , 204 - 207 .
Johnson J.P. ( 1948 ) Sulfa drugs for American foulbrood of honey bees: third report , J. Econ. Entomol . 41 , 314 - 318 , also earlier parts cited therein.
Katznelson H. ( 1950 ) The influence of antibiotics and sulfa drugs on Bacillus larvae, the cause of American foulbrood of the honeybee, in vitro and in vivo, J. Bacteriol . 59 , 71 - 79 .
Katznelson H. ( 1956 ) Stability of antibiotics in honey , Am. Bee J . 96 , 137 .
Katznelson H. , Gooderham C.B. ( 1949 ) Sulfathiazole in relation to American Foulbrood , Sci. Agric . 29 , 340 - 344 .
Katznelson H. , Arnott J. , Bland S.E. ( 1952 ) Preliminary report on the treatment of European foulbrood of honey bees with antibiotics , Sci. Agric . 32 , 180 - 184 .
Katznelson H. , Jamieson C.A. , Austin G.H. ( 1955 ) Further studies on the chemotherapy of diseases of the honeybee , Can. J. Agric. Sci . 35 , 189 - 192 .
Kochansky J. , Knox D. , Shimanuki H. ( 1999 ) Comparative stability of oxytetracycline and tylosin in sugar syrup , Apidologie 30 , 321 - 326 .
Machova M. ( 1970 ) Variations de la sensibilité aux antibiotiques souches de Bacillus larvae , Bull. Apic . 13 , 5 - 11 .
Miyagi T. , Peng C.Y.S. , Chuang R.Y. , Mussen E.C. , Spivak M.S. , Doi R.H. ( 2000 ) Verification of oxytetracycline-resistant American foulbrood pathogen Paenibacillus larvae in the United States , J. Invertebr . Pathol. 75 , 95 - 96 .
Moffett J.O. , Hitchcock J.D. , Lackett J.J. , Elliott J.R. ( 1970 ) Evaluation of some new compounds in controlling American foulbrood , J. Apic. Res . 9 , 39 - 44 .
Moffett J.O. , Wilson W.T., Parker R.L. ( 1958 ) The effect of Penicel, tetracycline and erythromycin on adult bees, brood-rearing, and honey production , Am. Bee J . 98 , 22 - 24 .
Nakajima C. , Okayama A. , Sakogawa T. , Okayama A. , Nakamura A. , Hayama T. ( 1997 ) Disposition of ampicillin in honeybees and hives , J. Vet. Med. Sci . 59 , 765 - 767 .
Nakajima C. , Sakogawa T. , Okayama A. , Nakamura A. , Hayama T. ( 1998 ) Disposition of mirosamycin in honeybee hives , J. Vet . Pharmacol. Therap. 21 , 269 - 273 .
Okayama A. , Sakogawa T. , Nakajima C. , Hayama T. ( 1996 ) Biological properties and antibiotic susceptibility of Bacillus larvae originated from American foulbrood of honeybee in Japan, J . Vet. Med. Sci. 58 , 439 - 441 .
Peng C.Y.-S. , Mussen E. , Fong A ., Cheng P., Wong G. , Montague M.A. ( 1996 ) Laboratory and field studies on the effect of the antibiotic tylosin on honey bee Apis mellifera L. (Hymenoptera: Apidae) development and prevention of American foulbrood disease , J. Invertebr. Pathol . 67 , 65 - 71 .
Reinhardt J.F. ( 1947 ) The sulfathiazole cure of American foulbrood: an explanatory theory , J. Econ. Entomol . 40 , 45 - 48 .
Shimanuki H. , Knox D.A. ( 1991 ) Diagnosis of honey bee diseases . US Department of Agriculture , Agriculture Handbook No. AH- 690 , 53 p. A revised edition of this publication was issued in 2000.
Shimanuki H. , Knox D.A. ( 1994 ) Susceptibility of Bacillus larvae to Terramycin, Am . Bee J. 134 , 125 - 126 .
Wilson W.R. ( 1962 ) Control of European foulbrood using two erythromycin formulations and yearly disease recurrence , Am. Bee J . 102 , 33 - 34 .
Wilson W.T. , Moffett J.O. ( 1957 ) The effect of erythromycin and other antibiotics on the control of European foulbrood of honeybees , J. Econ. Entomol . 50 , 194 - 196 .