Degradation and Metabolite Production of
Tylosin in Anaerobic and Aerobic
A. C. Kolz, T. B. Moorman, S. K. Ong, K. D. Scoggin, E. A. Douglass
Watershed contamination from antibiotics is becoming
Degradation half-lives for tylosin reported in the literature average
a critical issue because of increased numbers of confined animal-feeding
4 to 8 days in swine, calf, and chicken manure; 2 to 8 days in aqueous
operations and the use of antibiotics in animal production. To understand the
manure and soil–manure mixtures; and 10 to 40 days in surface-
fate of tylosin in manure before it is land-applied, degradation in manure
water simulation systems (De Liguoro et al., 2003; Ingerslev and
lagoon slurries at 228C was studied. Tylosin disappearance followed a
Halling-Sørensen, 2001; Ingerslev et al., 2001; Loke et al., 2000;
biphasic pattern, where rapid initial loss was followed by a slow removal
Teeter and Meyerhoff, 2003). However, most of these studies
phase. The 90% disappearance times for tylosin, relomycin (tylosin D), and
focused on the degradation of tylosin A in aerobic systems and fresh
desmycosin (tylosin B) in anaerobically incubated slurries were 30 to 130hours. Aerating the slurries reduced the 90% disappearance times to between
manure. Tylosin degradation in an aqueous manure mixture,
12 and 26 hours. Biodegradation and abiotic degradation occur, but strong
incubated under methanogenic conditions, produced several identi-
sorption to slurry solids was probably the primary mechanism of tylosin
fied degradates; however, analysis was hampered by interference
disappearance. Dihydrodesmycosin and an unknown degradate with
from components of the mixture (Loke et al., 2000). Degradation
molecular mass of m/z 934.5 were detected. Residual tylosin remained in
rates for tylosin A, in these systems, do not provide adequate
slurry after eight months of incubation, indicating that degradation in
information about the persistence of other forms of tylosin or the
lagoons is incomplete and that residues will enter agricultural fields. Water
appearance of degradates in lagoon slurry, where waste may reside
Environ. Res., 77, 49 (2005).
for six months before land application. Therefore, it is of interest to
tylosin, antibiotic degradation, anaerobic, aerobic, swine
understand the fate of common veterinary drugs, such as tylosin, in
lagoon storage, before the manure slurry is land-applied.
Typical anaerobic lagoons may have, as their design objectives,
a reduction in solids, nutrients, odor, and sludge volume (Zhang,
2001). Retention times for the lagoons will vary with the production
Antibiotic residues and increased numbers of antibiotic-resistant
phase of the facility and season. In some cases, anaerobic lagoons
bacteria have been reported near confined animal feeding operations
may be modified to include mechanical aeration, or an aerated
(CAFOs) and in agricultural watersheds (Campagnolo et al., 2002;
lagoon may follow as an additional stage for treatment of lagoon
Chee-Sanford et al., 2001; Haller et al., 2002; Kolpin et al., 2002).
effluent. It is of interest to determine how aeration of lagoons,
A major route for entry of veterinary pharmaceuticals into water-
a common modification, may change degradation rates.
sheds is through land application of animal biosolids and spills of
The objectives for the current research were to investigate the
animal waste at facilities using these drugs (Boxall et al., 2001;
disappearance of tylosin and to determine the production of degra-
Daughton and Ternes, 1999). Swine CAFOs often use antibiotics
dates in lagoon slurry. Lagoon slurries were collected from two
for therapeutic or growth-promoting purposes. Manure generated at
swine CAFOs, amended with tylosin and incubated under aerobic
CAFOs, containing excreted residues, is commonly stored in
and anaerobic conditions. The persistence of tylosin was also studied
earthen lagoons for several months before land application. Incom-
by incubating one of the spiked slurries anaerobically for eight
plete degradation of pharmaceuticals, in vivo and during manure
months. Sodium-azide amended slurry and filtered lagoon liquid
storage before biosolids are land-applied, could be a contributing
were included to distinguish biological from abiotic degradation or
factor to the presence of these drugs in waterways.
from sorption during the studies. Evaluation of solvent and solid-
In the year 2000, 92% of swine CAFOs reported using antibiotics
phase extraction systems was also done to optimize analyte recovery
in a nationwide survey, and the most commonly administered drug
for liquid chromatography-tandem mass spectrometry analysis.
was tylosin, a macrolide antibiotic (USDA, 2002). Between 4 and 5million pounds of tylosin and other macrolide antibiotics were soldannually in 2001 and 2000 (Vansickle, 2002). After oral administra-
Materials and Methods
tion of tylosin at a dose rate of 110 mg/kg, swine were found to
Slurry Collection and Characterization.
excrete up to 40% of the tylosin in antibiotically potent forms (Sieck
identified as an open anaerobic lagoon (OL) and covered anaerobic
et al., 1978). These forms include (in order of prominence) relo-
lagoon (CL), were collected from two swine CAFO lagoons, an
mycin (tylosin D), tylosin (tylosin A), dihydrodesmycosin, des-
open lagoon, and a covered anaerobic lagoon, respectively. The OL
mycosin (tylosin B), macrosin (tylosin C), and at least 10 other
slurry was a combination of settled solids and lagoon liquid col-
degradates in smaller quantities (Sieck et al, 1978; Teeter and
lected from below the liquid surface of the lagoon, while CL was
Meyerhoff, 2003). Different forms of tylosin are shown in Figure 1.
collected from a pump outlet of the covered lagoon. Collection took
microorganisms in the broth was observed after this period for signsof tylosin toxicity.
To assess the effectiveness of azide on
microbial inhibition, 50 g/L of sodium azide was added to CL slurryand incubated for 96 hours. Aliquots of 0.5 mL were plated ontoLuria Bertani agar to determine microbial growth.
Anaerobic degradation experiments were
conducted by transferring 20 mL of either CL or OL slurries toamber glass vials with Teflon-lined caps. Before the slurries weretransferred to the amber vials, the slurries were shaken for 10minutes on a reciprocating shaker. For both OL and CL slurries, twosets of vials (15 to 18 vials per set) were prepared. For one set ofvials, 400 lg of tylosin tartrate from aqueous stock solution wereadded to each vial giving a concentration of 20 mg/L. After additionof tylosin, the vials were capped, vortexed for 10 seconds, and the
Figure 1—Structure of various tylosin forms.
vial headspaces were immediately evacuated and filled with heliumgas (99.99% pure). The second set of vials was spiked with 1 g ofsodium azide (50 g/L) to inhibit biodegradation. After the addition
place in the summer, when the average temperature in the lagoons
of azide, the vials were vortexed and allowed to rest for 30 minutes
was approximately 258C. The slurries were collected in 68-L
at 228C. This was followed by spiking the vials with 400 lg of
(18-gal) rubber tubs, which were kept on ice during transport.
tylosin and evacuating the headspaces of the vials. In addition, a set
The slurries were homogenized and stored in 1.9-L (2-qt) glass
of vials were prepared with OL slurry, but spiked with 3900 lg of
jars at 48C until use.
tylosin (195 mg/L) to observe differences in disappearance rates and
Characterization tests were conducted to determine the pH, solids,
products at this higher concentration.
mineral, and carbon content. Dissolved organic carbon in the super-
A separate experiment using OL slurry was prepared under
natant was determined using UV-persulfate digestion with infrared
anaerobic conditions to study the persistence of residuals under long
detection. Organic and inorganic carbon solids content were deter-
term incubation (eight months). In this study, 240 lg of tylosin
mined by combustion and infrared spectrometry. Total solids were
tartrate was added (12 mg/L) to the vials. To inhibit biological
determined by weight difference after drying an aliquot of slurry at
activity, 0.08 g of azide per 20 mL of slurry (4 g/L) was added to
1058C. The oxidation–reduction potential (ORP) and pH of the
another set of vials. All vials were incubated at 22 6 18C. Triplicate
slurries were measured before and after incubation for the degradation
vials were sacrificed at each sampling event and measured for
experiments. All measurements were taken at 22 6 18C. The ORP
tylosin and degradates.
was measured using a platinum-tipped ORP probe filled with a 3-M
Tylosin was added to slurries rather than studying the disappear-
silver–silver chloride solution (Thermo Electron, Mississauga,
ance rates of tylosin residuals already present in slurries, for several
Canada). A hydrogen standard 1424 mV (mV EH) solution was
reasons. First, the detected tylosin residuals were too low to signifi-
used as a reference for the probe. The background concentrations of
cantly determine disappearance rates. Second, tylosin B and D were
tylosin in each source material were analyzed by liquid chromatog-
the persistent forms of tylosin found in the lagoons, which were not
raphy with tandem mass spectrometry (LC–MS–MS).
consistent with the forms excreted by swine, which were primarily
Tylosin Stability and Toxicity.
The stability of tylosin was
tylosin D and A.
monitored in several sterile matrices: (a) filtered lagoon liquids with
Aerated degradation experiments were pre-
50 g/L sodium azide, (b) Milli-Q purified water (Millipore, Billerica,
pared in a similar manner to the anaerobic experiments, except that
Massachusetts) with 50 g/L of sodium azide, and (c) Milli-Q water at
immediately after spiking the manure slurry with tylosin the vials
pH 7 and 9.2. The filtered lagoon liquids were obtained by centri-
were capped and a 16-gauge Teflon tube inserted and compressed
fuging the OL and CL slurries at 12 500 3 g for 30 minutes. The
air (breathing-air grade) bubbled at a continuous rate. A 22-gauge
lagoon liquids were decanted, and 50 g/L of sodium azide was added
needle was also inserted through the cap as a gas vent. Samples
before filtration. Milli-Q water with 50 g/L of sodium azide and
were incubated at room temperature (22 6 18C).
Milli-Q water without azide were adjusted with 0.1-M potassium
Three extraction solvent combinations
hydroxide (KOH) to pH 9.2, for comparison with the filtered lagoon
were evaluated for tylosin recovery and stability in CL slurry 24
liquids. All matrices were filter-sterilized with sterile 0.2-lm pore-
hours after spiking. The solvent combinations tested were methanol
size cellulose-acetate filters and dispensed into autoclaved glassware,
(Rabølle and Spliid, 2000), methanol–acetonitrile–0.1 M ascorbic
under a sterile laminar flow hood. Tylosin tartrate was spiked in all
acid (45:45:10, v:v:v) (Teeter and Meyerhoff, 2003), and acetoni-
matrices to a concentration of 20 mg/L. The proportion of forms in
trile–isopropyl alcohol (95:5, v:v) (Shang et. al., 2001). Extraction
aqueous tylosin tartrate stock solution was approximately 93% A,
recoveries, reported by the researchers, for the above extraction
5% D, 2% B, and 0.3% C. All containers were incubated in the dark
solvent combinations, from soil or manure-soil mixtures, ranged
at 22 6 18C, and aliquots were taken from the solutions periodically,
between 61% and 96%. Additional testing was conducted to deter-
under sterile conditions.
mine if the acetonitrile–isopropyl alcohol combination amended
The toxicity of tylosin to slurry microbes was assessed by spiking
with 5-M KOH (95:5:0.1, v:v:v) would be appropriate for use,
CL slurry with 20 mg/L of tylosin. After 72 hours of incubation at
because higher pH is often used to extract macrolides from tissues
228C, a 0.5-mL aliquot of slurry from the CL slurry assay was
(Fedeniuk and Shand, 1998). Adjusting the solvent to above pH 9.4
inoculated to 100 mL of a minimal salts broth. The broth was
was based on the assumption that tylosin's nonionized form may be
incubated at 258C for seven days, and the growth of the slurry
easier to extract and occurs at pH above its pKa of 7.4 (O'Neil et al.,
Water Environment Research, Volume 77, Number 1
2001). To evaluate each solvent combination, triplicate vials
50-min run. Column temperature was maintained at 408C. A cali-
containing 20 mL of CL slurries were spiked with 400 lg tylosin
bration curve for tylosin tartrate was prepared, showing a strong
tartrate (20 mg/L) and 1 g sodium azide (50 g/L) and incubated at
linear area response (R2 5 0.9997). Calibrations for tylosin A, B, C,
228C for 24 hours. The extraction procedure consisted of two cycles
and D were made by assuming an equal-area-concentration
of solvent extraction with sonication and shaking. To examine
response for all forms (Teeter and Meyerhoff, 2003). Retention
tylosin stability in the solvent, 400 lg tylosin tartrate and 1 g sodium
times for tylosin forms B, D, C, and A were 21.2, 21.5, 21.8, and
azide were added to 20 mL of each solvent combination, without
22.3 min, respectively. The detection limit for tylosin, with UV,
manure slurry, and were subjected to two cycles of sonication
under these settings, was approximately 500 lg/L. For the
degradation studies, HPLC analysis was not sufficient to determine
For the degradation studies, the slurry in each sacrificed vial was
tylosin C in the slurry assays. Liquid chromatograph-mass
transferred to a 50-mL polypropylene centrifuge tube. The amber
spectrometry was used to verify the presence of tylosin A, B, C,
vial was rinsed with 10 mL of Milli-Q purified water, and the rinse
D, and dihydrodesmycosin, and to determine the molecular mass of
water was transferred to the centrifuge tube. The slurry was centri-
degradates observed in some samples. The limit of detection for
fuged at 12 500 3 g for 30 minutes, resulting in a pellet weighing
tylosin A, B, C, and D forms were each approximately 4 lg/L. Mass
approximately 0.5 to 1 g and 20 to 30 mL of dark brown, trans-
spectra of tylosin forms and degradates were acquired by positive
lucent, aqueous supernatant. The supernatant, containing very small
and negative ion electrospray on an Agilent 1100 HPLC configured
amounts of particulate material, was decanted, measured, and
as above, with the addition of 0.1% acetic acid to acetonitrile,
retained. Then, 10 mL of acetonitrile–isopropyl alcohol (95:5 v:v)
coupled to a 1100 mass selective detector (MSD) ion trap LC–MS–
was added to each tube. The tubes were vortexed for 1 minute,
MS (Agilent, Palo Alto, California). The drying gas was operated at
sonicated (Bransonics Model 5200, Branson Ultrasonics Corp.,
a flowrate of 12 mL/min at 3508C. The nebulizer pressure was 50
Danbury, Connecticut) for 30 minutes, shaken for 1 hour at 270
psig, scanning mass from 50 to 1200 m/z. Quantification was based
cycles per minute on a reciprocating shaker, and then centrifuged at
on an external 15-point calibration from 8 to 800 mg/L of the base
12 500 3 g for 15 minutes. The solvent was decanted and retained
peak ion (protonated adduct of the molecular ion) of the analyte. For
separately from the aqueous supernatant. The procedure was re-
each compound, the protonated molecular ion, [M 1 H]1, and at
peated by adding another 10 mL of fresh solvent to the pellet and
least one confirming ion was acquired, based on MS–MS of the
the extract combined with the first 10 mL of solvent extract in an
base-peak and fragment ratios formed from the standards and
amber vial. The combined solvent extract was dried down to less
confirmed by published spectra (Van Poucke et al., 2003 and 2004).
than 5 mL by passing compressed air over the solvent at room
Ion suppression tests for the elution of four typical samples showed
temperature. The concentrated solvent extract was then transferred
no interference of tylosin A at the time window of interest.
to a 250-mL glass beaker, including 20 mL of Milli-Q (deionized)
Disappearance Rate Modeling.
A two-compartment, first-
water and 1 mL methanol rinse. The solvent extract was further
order model was used to describe the total tylosin disappearance.
diluted to 140 mL with Milli-Q water, and the pH was adjusted
This model assumes a pool of rapidly degrading tylosin (C1) and
above pH 9.4 with 0.5-M KOH. Likewise, 2-mL of the decanted
a pool of more persistent tylosin (C2).
aqueous supernatant was transferred to a 100-mL glass beaker anddiluted with 80 mL Milli-Q water. The pH of the dilution was raised
Ct ¼ C1ek1t þ C2ek2t
above 9.4 with 5-M KOH.
Both the decanted aqueous supernatant and the diluted solvent
extract were extracted separately using solid-phase extraction (SPE),under vacuum, at a flowrate less than 5 mL/min. Oasis Hydrophilic-
Ct 5 tylosin concentration (% of added) at time t;
Lipophilic Balance (HLB) cartridges (200 mg) (Waters, Milford,
C1 5 initial tylosin concentration (% of added) in pool 1;
Massachusetts) were used and primed with 5 mL methanol, followed
C2 5 initial tylosin concentration (% of added) in pool 2 (C1 1
by 4 mL 0.5-M KOH. Glass fiber filters (Fisher Scientific, Hampton,
New Hampshire) (1-lm pore-size) were inserted to prevent the SPE
k1 5 first-order rate constant (hour21) for pool 1;
frit from clogging. The cartridges were rinsed with three 1-mL
k2 5 first-order rate constant (hour21) for pool 2; and
aliquots of methanol–water–ammonium hydroxide (60:38:2, v:v:v)
to remove organic coextracted materials. Tylosin was eluted with
The model was fit to the experiment data using nonlinear
four 0.5-mL aliquots of acetonitrile–glacial acetic acid (98:2. v:v).
regression with SAS software (SAS, Cary, North Carolina). Note
The tylosin-containing acetonitrile eluants were dried down
that C2 was set as equal to the total added (100%) minus C1. If C2 or
completely, under nitrogen gas at 228C, and reconstituted in 0.5
k2 is not significantly different from zero, the model becomes a
mL of 0.01-M ammonium acetate (pH 6.8) to improve tylosin
simple first-order model.
stability before analysis. The tylosin recovered from the supernatantand the pellet was summed for each slurry sample.
Tylosin was analyzed using an Agilent 1100
Results and Discussion
Series high-pressure liquid chromatograph (HPLC) (Agilent Tech-
The characteristics of the two
nologies, Palo Alto, California). Injection volume of samples was
manure slurries are presented in Table 1. The pH for both manure
25 lL, and UV detection wavelength for all tylosin forms was
slurries were similar, ranging between 8.5 to 9.1, before incubation.
284.8 nm. A ZORBAX SB-C18 4.6 3 250 mm column was used
Background concentrations of tylosin B and D were approximately
(Agilent Technologies). The eluants used were 0.01-M ammonium
50 and 15lg/L in OL, respectively, and 1700 and 270 lg/L in the
acetate–glacial acetic acid buffer (pH 4.6) and acetonitrile, at a
CL slurry, respectively. The OL slurries had higher total solids,
constant flowrate of 0.5 mL per minute, starting with 10% acetonitrile
organic carbon, and dissolved organic carbon as compared to CL
increasing to 100% acetonitrile between 6 and 35 minutes for each
slurry, but the total carbon values were similar.
Table 1—Swine manure characteristics from an openanaerobic lagoon (OL) and a covered anaerobic lagoon(CL).
Tylosin B, D (lg/L)
Total solids (g/kg)
Organic carbon (g/kg)
Tylosin Stability and Toxicity.
Tylosin tartrate was stable for
at least one month when stored in Milli-Q water at pH 5.7 to 6.7at 22 6 18C. However, approximately 10% of added tylosin wasdegraded within the first 200 hours in Milli-Q water at pH 9.2.
Figure 3—Tylosin recovered (% of added, sum of tylosinA, B, and D) from (a) sterile OL lagoon liquid, (b)anaerobic OL slurry, and (c) aerated OL slurry. Error barsare 6 1 standard deviation. Initial spiked concentrationwas 20 mg/L as tylosin tartrate.
In the case of azide-amended Milli-Q water at pH 9.3, the amountdegraded was also approximately 10%.
Approximately 20 and 5% of added tylosin was degraded within
the first 72 hours for the CL and OL 0.2-lm filtered lagoon liquids,respectively (see Figures 2a and 3a). In azide-amended and un-amended Milli-Q water samples and both filtered lagoon liquids, anunknown degradate with molecular mass of m/z 934.5 appeared inan apparently base-catalyzed reaction (O'Neil et al., 2001; Paesenet al., 1995). The addition of sodium azide in the assays did notappear to influence tylosin stability.
The tylosin toxicity test showed that tylosin was not effective in
suppressing microbial viability at 20 mg/L. The minimal salts brothbecame turbid after one week of incubation, and actinomycetes,amoebae, fungi, and bacteria were observed by light microscopy.
Figure 2—Tylosin recovered (% of added, sum of tylosin
This is consistent with previous studies, where tylosin was found to
A, B, and D) from (a) sterile CL lagoon liquid, (b)
be active against gram-positive bacteria, but was only marginally
anaerobic CL slurry, and (c) aerated CL slurry. Error bars
active on gram-negative bacteria (Prescott, 2000).
are 6 1 standard deviation. Initial spiked concentration
Sodium azide was found to be effective at
was 20 mg/L as tylosin tartrate.
inhibiting microbial growth in the CL slurry, at a concentration of
Water Environment Research, Volume 77, Number 1
Table 2—Two-compartment model of tylosin disappearance in two manure slurries (CL and OL) under aerobic oranaerobic conditions, with or without sodium azide. Initial tylosin concentrations were 20 mg/L, except for OL anaerobichigh treatment, which received 195 mg/L of tylosin. Rate constants are given in hours with 95% conﬁdence intervals.
Manure or treatment
OL anaerobic high
a k2 not significantly different from zero.
b C2 percentage tylosin not significantly different from zero.
50 g/L. Colonies or growth were absent from plates incubated for
72 hours incubation. Percent recoveries of tylosin (sum of forms
96 hours with the azide-amended slurry.
A, B, and D) with time for CL and OL slurries spiked with 20 mg/L
Tylosin concentrations are reported as
of tylosin are presented in Figures 2b and 3b, respectively. In both
the sum of the A, B, and D forms recovered from the separate
CL and OL slurries, there was a rapid disappearance of tylosin (60
analysis of slurry supernatant and slurry solid phase. Recoveries are
to 85%) within the initial 24 hours, after which the disappearance
expressed as a percentage of the tylosin added to the samples, after
of tylosin slowed in both the CL and OL slurries.
correction for tylosin concentrations present in the slurry at the time
The two-compartment model (eq 1) was used to model the
of collection. The average recoveries from CL and OL slurry using
disappearance of tylosin in the OL and CL slurries. A first-order
acetonitrile–isopropyl alcohol were 99% (8% RSD, n 5 12) and 93%
model was initially used, but was found to overestimate the amount
(13% RSD, n 5 12), respectively, for vials sacrificed within 0.5
of tylosin in anaerobic samples during the first 24 hours and
hours of spiking. Recoveries for acetonitrile–isopropyl alcohol
underestimate the concentrations after 24 hours. The estimated
averaged approximately 41% after 24 hours in CL. Methanol and
tylosin disappearance rate constants, k1 and k2 and C1 and C2 for the
acidified methanol each gave an average 38% recovery after 24
anaerobic studies using OL and CL slurries, are summarized in
hours. Adding KOH to acetonitrile–isopropyl alcohol produced a
Table 2. The k1 rate constants for the azide-amended and un-
higher proportion of tylosin B to tylosin A in the extracts. Because
amended OL and CL slurries were not significantly different, at a
tylosin stability was desirable during extraction, the addition of KOH
confidence interval of 95%. This implies that the rapid disappear-
may not be appropriate. Of the three extraction solvents, acetonitrile–
ance of tylosin within the first 24 hours may be because of sorption
isopropyl alcohol (Shang et al., 2001) was chosen as the extractant
of tylosin to the solids. Substantial sorption of tylosin (90 to 99%)
also because of its use in the LC–MS–MS analysis. In addition, the
has been reported within 1 to 6 hours after spiking in soil and
SPE method outlined above for extracting tylosin forms and
manure mixtures (Ingerslev and Halling-Sørensen, 2000).
degradates was found to produce enhanced HPLC-baseline resolu-
The k1 rate constants for the OL slurries were two to three times
tion as compared to other similar methods, which used acidic or non-
larger than those for the CL slurries, indicating that tylosin
pH modified rinses (Kolpin et al., 2002; Teeter and Meyerhoff,
disappearance was more rapid and residues were lower in OL slurry.
2003). Oasis HLB cartridges provided recoveries of 96 to 105% from
Hydrophobic binding behavior has been related to sorption of tylosin
aqueous tylosin tartrate solutions and were resistant to pH changes
(Loke et al., 2002; Tolls, 2001). The OL slurry had higher total solids
up to pH 14, necessary to optimize the tylosin SPE procedure.
and organic carbon content than CL, which supports sorption as
After 72 hours of incubation, the pH of the
a process contributing to tylosin disappearance. Teeter and Meyerh-
slurries changed from 8.5 to 9.1 to approximately 8.2 to 8.5 and
off (2003) reported approximately 30% of the 14C-labeled tylosin
dropped further to pH 7.5 to 7.8 after eight months of incubation.
accumulated as bound (nonextractable) residues in manure.
The addition of azide decreased pH of slurries by approximately 0.2
In the azide-amended CL slurry, the k2 rate constant was 0.003
pH units. The ORP in slurries was between 210 and 280 mV EH
h21, and the estimated C2 value was 56%, showing that the tylosin
after 72 hours of incubation. The addition of azide resulted in a
continued to slowly disappear with time. For the unamended CL
further decrease of the ORP to between 290 and 2160 mV EH after
slurry, the k2 rate constant was 0.004 h21, with a lower C2 value of
making the disappearance follow a first-order model. The estimatedk1 rate constants and C1 and C2 values are presented in Table 2. Forthe azide-amended and unamended OL slurries, the k1 rate constantswere not significantly different. A similar assessment can be madefor the k1 rate constants for the CL slurries. However, the k1 rateconstants for OL were approximately 2 to 3 times larger than the CLslurries. Unlike the anaerobic studies, there was a lower amount ofresidual tylosin remaining at the end of the 72 hours. Less than 1%of the tylosin added remained after 12 days of aeration in OL slurry.
The 90% disappearance time for the OL aerated slurries was 12hours, as compared to 40 hours for the OL anaerobic slurries. ForCL, the 90% disappearance times were 26 and 310 hours for theaerobic and anaerobic slurries, respectively.
Disappearance of tylosin in manure slurries can be attributed to
biotic and abiotic degradation and to irreversible sorption (i.e., theformation of nonextractable bound residues). Assuming that theazide was effective in inhibiting microbial activity, the similar k1rate constants for azide-amended and unamended slurries indicatethat the portion of the initial tylosin disappearance attributed tobiodegradation was quite small. In addition, the magnitude ofabiotic degradation was small, with only 5 to 20% of the tylosindegraded in sterile filtered lagoon liquids. Therefore, much of thetylosin disappearance may be because of sorption, particularly in theanaerobically incubated slurries. Aeration increased slurry pH to 9.2
Figure 4—Tylosin forms A, B, D, and unknown degradate
to 9.3, which may have accelerated a base-catalyzed reaction and
(m/z 934.5) recovered at various times for OL lagoon
resulted in lower residuals in aerated slurries. Faster disappearance
slurry spiked with 195 mg/L tylosin and incubated
rates in aerated samples correspond well to findings by other
anaerobically. Average recovery is expressed as the
researchers in aerated and anaerobic surface water simulations and
percent of tylosin added. Degradate recovery is based
soil-manure slurries studies (Ingerslev and Halling-Sørensen, 2001;
on the UV peak response for tylosin.
Ingerslev et al., 2001). In these systems, aeration decreased tylosin'shalf-life by 30 days when compared with anaerobic incubation andeliminated detectable residuals in soil-manure slurries within 12 to
39%. The difference in the estimated C2 values for the two CL
slurries treatments indicated that there was less residual tylosin in the
The proportions of the different forms of
unamended CL slurry and may imply that biodegradation was
tylosin shifted through the tests, and some changes occurred
occurring in the unamended CL slurry. The OL slurries behaved in a
immediately. The proportion of tylosin forms added to the lagoon
similar manner to the CL slurries, with respect to the azide treatment,
slurries at the initiation of the experiments was approximately 93%
but the estimates of C2 were less in the OL slurry than in the CL.
A, 5% D, 2% B, and 0.3% C. Within 0.5 hours after spiking, the
Half-lives may be used to compare the differences in the various
average proportion of forms A, D, B, and C in OL slurry spiked at
treatments; however, because the disappearance of tylosin was very
195 mg/L was 77% A, 13% D, 9% B, and 0.2% C (Figure 4).
rapid, the time for 90% disappearance of tylosin was chosen as a
Similar changes in the proportions of tylosin forms were also noted
comparison. The estimated time necessary for 90% tylosin dis-
in OL and CL samples after spiking 20 mg/L of tylosin. Tylosin A
appearance was 40 and 310 hours for unamended OL and CL
decreased more rapidly than D, and both decreased more rapidly
slurries, respectively. The 90% disappearance times for OL and CL
than tylosin B. Tylosin B concentrations decreased in most slurries,
azide-amended slurries were 90 and 500 hours, respectively,
except in anaerobically incubated CL slurry, where concentrations
indicating that faster disappearance occurred in the unamended
of B doubled during the first 72 hours. Tylosin B was the most
predominant form in anaerobically incubated OL slurry, but this
For the anaerobic OL slurry treated with 195 mg/L tylosin, the k1
appears to be because of its increased relative persistence, as all
rate constant and the percent recovery were similar to that with
forms decreased in concentration. The production of tylosin B was
20 mg/L of tylosin. The estimated 90% disappearance time was
reported in aqueous anaerobically incubated manure mixtures and
also similar to that with 20 mg/L tylosin.
aerated soil manure slurries (Ingerslev and Halling-Sørensen, 2001;
The pH in the aerated slurries increased from
Loke et al., 2000). Tylosin D was also reported in these studies, but
9.1 to 9.3 after 24 hours of aerobic incubation and remained above
to a lesser extent than tylosin B. The highest relative proportion of
pH 9.0 after 12 days of aeration. Addition of sodium azide
tylosin D in the current research was in aerated CL slurry incubated
decreased the slurry pH by approximately 0.1 to 0.2 pH units. After
for 72 hours, with 36% A, 36% D, and 27% B. Aerobically in-
aeration, the ORP was approximately 1340 mV EH. The ORP
cubated OL slurry produced the smallest shifts in proportion of
decreased by approximately 10 mV EH in the aerated samples when
forms, with tylosin present as 93% A, 3% D, 4% B, and 0% C. After
azide was added.
72 hours, tylosin C was not detected in any of the samples.
As in the anaerobic studies, there was a rapid disappearance of
Dihydrodesmycosin was not detected in slurry-source materials, but
tylosin. The disappearance of tylosin was almost completely
increased to trace amounts during the first 72 hours of aerated and
described by the first term of the two-compartment model (eq 1),
anaerobic incubation. Dihydrodesmycosin eluted at 20.2 minutes on
Water Environment Research, Volume 77, Number 1
spectral fragmentation pattern for the degradate is compared withthat of tylosin A in Figure 5. The degradate also responded undernegative ionization mode, indicating that the compound hadamphoteric properties. The mass spectral fragmentation patternand peak UV absorbance of the degradate (285 to 290 nm) arestrongly suggestive of a relation to tylosin A plus m/z 18.
Studies in manure-lagoon slurries indicated that the majority of
tylosin was rapidly sorbed and degraded, with 90% disappearanceoccurring in less than five days. Both abiotic and biotic degradationwere apparent. However, disappearance in anaerobic slurries slowedafter 24 hours, and a residual pool of tylosin remained through theextended-anaerobic incubation. Aerating slurries significantly re-duced tylosin residuals, leaving less than 1% of the added tylosinafter 12 days. Tylosin B and D, which retain approximately 35 to80% of the antibiotic activity of tylosin A (Teeter and Meyerhoff,2003), were detected in slurry-source materials. Tylosin B and Dwere produced during the studies and remained after eight monthsof anaerobic incubation, in conventional-lagoon slurry. Dihydro-desmycosin and a degradate with undetermined antibiotic activityalso formed during the experiments and persisted for eight monthsafter tylosin addition. Based on these findings, residual tylosindegradates, with antibiotic activity, may be applied to agricultural
Figure 5—Liquid chromatography with tandem mass
fields with slurries and contribute to the detection of antibiotic
spectrometry fragmentation patterns of (a) tylosin A
residues near CAFOs.
(m/z 916.5), (b) unknown degradate (m/z 934.5), and(c) UV chromatograms of tylosin in CL slurry at 0.5 and
72 hours [in milliAbsorbance units (mAu)].
Funding for this work was provided by a grant from
the Iowa State Water Resources Research Institute, Ames, Iowa.
Mention of trade names or commercial products herein is solely for
chromatographs. Azide amendment did not significantly change the
the purpose of providing specific information and does not imply
proportion of tylosin forms recovered in samples during short-term
recommendation or endorsement by the U.S. Department of
Agriculture (Washington, D.C.).
In the long-term anaerobic tests using OL slurries, approximately
Angela C. Kolz is a research assistant and Say Kee
1% of the tylosin remained after eight months incubation. The
Ong is an associate professor in the Department of Civil,
tylosin forms detected after eight months were generally tylosin B
Construction and Environmental Engineering, Iowa State University,
and D. Comparison of the azide-amended and unamended slurries
Ames, Iowa. Thomas B. Moorman is a microbiologist, Kenwood D.
indicated that there was a greater amount of tylosin B, D,
Scoggin is a research technician, and Elizabeth A. Douglass is
dihydrodesmycosin, and an unknown degradate (described below)
a research technician at the USDA-Agricultural Research Service,
in the azide-amended slurry than in the unamended slurry.
National Soil Tilth Laboratory, Ames, Iowa. Correspondence should
A degradate, eluting at 20.6 mi-
be addressed to Say Kee Ong, 394 Town Engineering Building, Iowa
nutes, appeared within 12 hours after spiking in all unamended and
State University, Ames, Iowa 50011; e-mail: [email protected]
azide-amended anaerobic and aerobically incubated assays. Neither
Submitted for publication May 4, 2004; revised manuscript
aeration nor sodium azide affected the amount of degradate
submitted November 1, 2004; accepted for publication November 1,
production. The degradate compound was not detected in slurry-
source materials before tylosin addition or in tylosin tartrate
The deadline to submit Discussions of this paper is May 15,
standards at pH 7.0, but it did appear in sterile lagoon liquids and
water at pH 9.2. Recovery of tylosin forms A, D, B, and thedegradate are shown in Figure 4. Recovery of the degradate is basedon an assumption of equal UV area response as tylosin. Typical UV
chromatographs at 0.5 and 72 hours show the transformation of
Boxall, A. B. A.; Fogg, L.; Blackwell, P. A.; Kay, P.; Pemberton, E. (2001)
tylosin A, D, and B, and production of this degradate over time
Review of Veterinary Medicines in the Environment; Environment
Agency Report P6-012/8TR; U.K. Environment Agency: Bristol,
The base peak in the spectrum of the degradate was m/z 934.5,
Campagnolo, E. R.; Johnson, K. R.; Karpati, A.; Rubin, C. S.; Kolpin,
and a major fragment ion was found at m/z 772.5 when analyzed
D. W.; Meyer, M. T.; Esteban, J. E.; Currier, R. W.; Smith, K.; Thu,
with LC–MS–MS, under positive ion mode. A low-abundance
K. M.; McGeehin, M. (2002) Antimicrobial Residues in Animal Waste
fragment ion, at m/z 916.5, was also detected. Tylosin A's
and Water Resources Proximal to Large-Scale Swine and Poultry
characteristic fragment ion of its 916.5 [M 1 H]1 mass under
Feeding Operations. Sci. Total Environ., 299, 89–95.
positive ionization mode was 772.5 [M 1 H 2 C7H12O3]1,
Chee-Sanford, J. C.; Aminov, R. I.; Krapac, I. J.; Garrigues-Jeanjean, N.;
indicating the loss of the mycarose sugar. The MS–MS mass
Mackie, R. I. (2001) Occurrence and Diversity of Tetracycline
Resistance Genes in Lagoons and Groundwater Underlying Two Swine
Prescott, J. F. (2000) Lincosamides, Macrolides, and Pleuromutilins. In
Production Facilities. Appl. Environ. Microbiol., 67, 1494–1502.
Antimicrobial Therapy in Veterinary Medicine, 3rd Ed; Prescott, J. F.;
Daughton, C. G.; Ternes, T. A. (1999) Pharmaceuticals and Personal Care
Baggot, J. D.; Walker, R. D. (Eds.); Iowa State University Press: Ames,
Products in the Environment: Agents of Subtle Change? Environ.
Health Persp., 107 (Supplement 6), 907–938.
Rabølle, M.; Spliid, N. H. (2000) Sorption and Mobility of Metronidazole,
De Liguoro, M.; Cibin, V.; Capolongo, F.; Halling-Sørensen, B.;
Olaquindox, Oxytetracycline and Tylosin in Soil. Chemosphere, 40,
Montesissa, C. (2003) Use of Oxytetracycline and Tylosin in Intensive
Calf Farming: Evaluation of Transfer to Manure and Soil. Chemo-
Shang, D.; Dyck, M.; Dassie, N.; Gibbons, N.; Alleyne, C.; Nicolidakis, H.;
sphere, 52, 203–212.
Jia, X. (2001) Rapid Determination of Tylosin and Virginiamycin in
Fedeniuk, R. W.; Shand, P. J. (1998) Theory and Methodology of Antibiotic
Swine Manure by APcI LC/MS/MS. Proceedings of the 49th
Extraction from Biomatrices. J. Chromatogr., A, 812, 3–15.
Conference on Mass Spectrometry and Allied Topics, Chicago, Illinois,
Haller, M. Y.; Mu¨ller, S. R.; McArdell, C. S.; Alder, A. C.; Suter, M. J.-F.
May 27—31; American Society for Mass Spectrometry: Santa Fe, New
(2002) Quantification of Veterinary Antibiotics (Sulfonamides and
Trimethoprim) in Animal Manure by Liquid Chromatography-Mass
15 July 2004).
Spectrometry. J. Chromatogr., A, 952, 111–120.
Sieck, R. F.; Graper, L. K; Giera, D. D.; Herberg, R. J.; Hamill, R. L. (1978)
Ingerslev, F.; Halling-Sørensen, B. (2001) Biodegradability of Metronida-
14C Tylosin Tissue Residue Study in Swine. Unpublished report dated
zole, Olaquindox, and Tylosin and Formation of Tylosin Degradation
November 1978 from Agricultural Biochemistry, Lilly Research
Products in Aerobic Soil-Manure Slurries. Ecotoxicol. Environ. Safety.,
Laboratories, Greenfield, Indiana. Submitted to WHO by Lilly Research
Centre Ltd., Windlesham, Surrey, England; http://www.inchem.org/
Ingerslev, F. T.; Loke, M.-L.; Halling-Sørensen, B.; Nyholm, N. (2001)
documents/jecfa/jecmono/v29je08.htm (accessed 15 July 2004).
Primary Biodegradation of Veterinary Antibiotics in Aerobic and
Teeter, J. S.; Meyerhoff, R. D. (2003) Aerobic Degradation of Tylosin in
Anaerobic Surface Water Simulation Systems. Chemosphere, 44,
Cattle, Chicken, and Swine Excreta. Environ. Res., 93, 45–51.
Tolls, J. (2001) Sorption of Veterinary Pharmaceuticals in Soils: A Review.
Kolpin, D. W.; Furlong, E. T.; Meyer, M. T.; Thurman, E. M.; Zaugg, S. D.;
Environ. Sci. Technol., 35, 3397–3406.
Barber, L. B.; Buxton, H. T. (2002) Pharmaceuticals, Hormones,
U.S. Department of Agriculture (2002) Preventative Practices in Swine:
and Other Organic Wastewater Contaminants in U.S. Streams, 1999-
Administration of Iron and Antibiotics. U.S. Department of Agricul-
2000: A National Reconnaissance. Environ. Sci. Technol., 36, 1202–
ture: Fort Collins, Colorado.
Vansickle, J. (2002) Drug Use Drops Again. National Hog Farmer, 47,
Loke, M.-L.; Ingerslev, F.; Halling-Sørensen, B.; Tjørnelund, J. (2000)
Stability of Tylosin A in Manure Containing Test Systems Determined
Van Poucke, C.; De Keyser, K.; Baltusnikiene, A.; McEvoy, J. D. G.; Van
by High Performance Liquid Chromatography. Chemosphere, 40, 759–
Peteghem, C. (2003) Liquid Chromatographic-Tandem Mass Spectro-
metric Detection of Banned Antibacterial Growth Promoters in Animal
Loke, M.-L.; Tjørnelund, J.; Halling-Sørensen, B. (2002) Determination of
Feed. Anal. Chim. Acta, 483, 99–109.
the Distribution Coefficient (log Kd) of Oxytetracycline, Tylosin A,
Van Poucke, C.; Dumoulin, F.; De Keyser, K.; Elliott, C.; Ven Peteghem, C.
Olaquindox and Metronidazole in Manure. Chemosphere, 48, 351–361.
(2004) Tylosin Detection in Animal Feed by Liquid Chromatography-
O'Neil, M. J.; Smith, A.; Heckelman, P. E. (Eds.). (2001) The Merck Index,
Tandem Mass Spectrometry with Enzymatic Hydrolysis of the Tylosin
13th ed. Merck: Whitehouse Station, New Jersey.
Urea Adduct. J. Agric. Food Chem., 52, 2308–2306.
Paesen, J.; Cypers, W.; Pauwels, K.; Roets, E.; Hoogmartens, J. (1995)
Zhang, R. (2001) Biology and Engineering of Animal Wastewater Lagoons.
Study of the Stability of Tylosin A in Aqueous Solutions, J. Pharm.
University of California at Davis. http://ucce.ucdavis.edu/files/
Biomed. Anal., 13, 1153–1159.
filelibrary/5049/678.pdf (accessed 15 July 2004).
Water Environment Research, Volume 77, Number 1
MINISTERIO DE LA PRESIDENCIA DIRECCION ADMINISTRATIVA INFORME DE INGRESOS Y GASTOS FONDO PROGRAMA DEL DESPACHO SUPERIOR Del 1 de Julio de 2007 al 30 de Septiembre de 2007 Transferencias recibidas del Tesoro Nacional y otros Reembolso H.D. Jorge Alvarado 7894612 Depósito MEF No.7122110 Nota de crédito por reembolso de gastos médicos San Vicente de Paul de Colombia del el joven Gilberto Araúz por
Manual on LaserEmitters andFlight Safety Approved by the Secretary Generaland published under his authority First Edition — 2003 International Civil Aviation Organization Manual on LaserEmitters andFlight Safety Approved by the Secretary Generaland published under his authority First Edition — 2003