UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

WASHINGTON D.C., 20460

	

	OFFICE  OF

	PREVENTION, PESTICIDES AND

	TOXIC SUBSTANCES

DP Barcode: 310732

PC Code:  006314

June 21, 2006

MEMORANDUM

SUBJECT:  	Antimycin A Drinking Water Characterization

TO:  		Lance Wormell, Risk Manager Reviewer

		Tom Myers, Risk Manager

		Christine Ollinger, HED

FROM:	Dirk F. Young, Ph.D. Environmental Engineer

THRU:	Betsy Behl, Chief ERB 4

		R. David Jones, Ph.D., Senior Agronomist

Background

	Antimycin A is a relatively minor-use piscicide, used primarily for
direct applications to remote mountain streams, lakes, ponds and fishery
operations. Several factors complicate the estimation of potential
drinking-water concentrations, among them non-strict application rates,
sparse fate data, and insensitive analytical methods.  Because of the
uncertainty associated with these problems, quantification of antimycin
potential exposure is highly speculative.  

.

Another difficulty is that there is comparatively little EPA-guideline
fate data available for evaluation of the fate of Antimycin A in the
environment.  The registrant has submitted only a hydrolysis and an
aerobic aquatic metabolism study, both of which the Office of Pesticide
Programs classified as ‘supplemental’, meaning that they supply some
useful information, but do not fulfill the data requirement.  There are,
however, non-guideline studies (primarily concerning hydrolysis rates)
that offer some insight into the fate of antimycin in the environment.
The combination of the submitted studies and the open literature studies
gives a wide range of degradation rates for antimycin, further
complicating the estimation of potential antimycin environmental
concentrations.

).  

	Although the above issues hinder the ability to calculate a
“standard” EECs in a manner similar to other pesticide assessments,
rough estimates that include the uncertainties associated with fate
properties and application rates are possible.  As expected by
considering the uncertainties, a large range of possible EECs are
realized as described below.

Antimycin Fate Properties

	Antimycin A is a complex of 4 structures that is reportedly soluble in
polar organic solvents (e.g. ethanol, acetone, chloroform) and
relatively insoluble in water (Walker et al., 1964).  There are only two
registrant-submitted studies—a hydrolysis study and an aerobic aquatic
study.  In the open literature, several hydrolysis studies are
available, but there are no direct studies that measure the sorption
properties of antimycin.  Some salient properties of antimycin are
summarized in Table 1.  When the fate studies are viewed together, some
conflicting information is apparent, making accurate estimates of
antimycin fate difficult.  

	The main mechanism of antimycin degradation is apparently base
hydrolysis, as alluded to in the open literature (Hussain, 1969; Walker
et al., 1964; Lee et al. 1971).  The supplemental registrant-submitted
hydrolysis study (MRID 46023101) also suggests that base hydrolysis is a
mechanism of degradation, but hydrolysis rates reported by the
registrant’s studies are significantly slower than the earlier open
literature studies, as shown in Table 1. 

 The registrant study ((MRID 46023101)  indicates that antimycin in an
aquatic environment at 25oC has a half life of about 15 days at a pH of
5, 3 days at pH 7, and 3 hours at pH 9.  Lee et al. (1971) reported much
faster degradation rates using yeast assays, with half lives of about
5.5 hours at pH 7, and 20 minutes at pH 9.5.  Hussain found degradation
half lives of about 46 hours for antimycin (A1) at pH 7.55 and about 2
minutes at pH 9.   Transformation products were not identified in the
registrant-submitted study or in Lee et al.  In an anecdotal report,
Walker et al. claimed that antimycin degrades by base hydrolysis and
produced a diagram showing the mechanism of degradation along with
assumed breakdown products. Hussain made similar claims as to the
degradation products. These products are blastmycic acid, antimycin
lactone, and antimycic acid. No quantitative information regarding the
relative concentrations of the transformation products is available. 

The only other registrant submitted environmental fate study is an
aerobic aquatic transformation study which indicated that antimycin A
degrades with a half life in the range of 23 to 47 days in pH 6.5 water.
This half-life appears to be substantially longer than the half-life due
to hydrolysis alone at this pH (see above and Table 1), which adds some
uncertainty to the half lives reported in this study as well as the
hydrolysis studies.  However, since the aerobic aquatic study was
conducted with sediment present, the longer half-life could be due to
high sorption of antimycin to sediment which may shield antimycin from
hydrolytic degradation (although this is purely speculative since
sorption studies are unavailable). 

Although the registrant did not provide sorption studies for antimycin,
indications from the aerobic aquatic study are that antimycin A does
sorb significantly to sediment.  Rough estimates of sorption can be made
from the aerobic aquatic study by considering the relative distribution
of antimycin in the water and the sediment as measured in that study. 
If equilibrium between the water and the sediment were assumed
throughout the aerobic aquatic study then Kd values would range from
about 1 to 88 ml/g or Koc in the range of 84 to 10000 ml/g, which
indicates that anticmycin A is mobile to slightly mobile in soil.

	In order to gain additional insight into antimycin, fate properties
were also estimated with ASTER.  ASTER (ASsessment Tools for the
Evaluation of Risk) was developed by the U.S. EPA Mid-Continent Ecology
Division, Duluth, MN (MED-Duluth) to assist regulators in performing
ecological risk assessments.  ASTER is designed to provide high quality
data for discrete chemicals either directly from associated databases or
from QSAR-based estimates. The QSAR system includes a database of
measured physicochemical properties such as melting point, boiling
point, vapor pressure, and water solubility as well as more than 56,000
molecular structures stored as SMILES (Simplified Molecular Input Line
Entry System) strings for specific chemicals.  Based on ASTER model
estimates, antimycin is not expected to be mobile in soil and sediment
(log Koc= 3.41) and the chemical has a relatively low potential for
bioconcentrating in aquatic organisms (bioconcentration factor
(BCF=350X).  Antimycin A is not likely to be persistent in the
environment and its low vapor pressure and Henry’s Law constant (2.42
x10-17 atm-m3 mol-1) limit its volatility.  Hydrolysis (t½ = 190 days)
appears to be a route of degradation for antimycin A.  

Table 1. Antimycin Fate Properties

Property	Value	Source

Molecular Wt	548.7





	Hydrolysis pH 5	15 days	MRID 46023101

Hydrolysis pH 7	3 days	MRID 46023101

Hydrolysis pH 9	3 hours	MRID 46023101

Hydrolysis pH 4.5 – 5.5	>7 hours	Lee et al. (1971)

Hydrolysis pH 7 – 8	5.5 hours	Lee et al. (1971)

Hydrolysis pH 8.5	40 minutes	Lee et al. (1971)

Hydrolysis pH 9.5	20 minutes	Lee et al. (1971)

Hydrolysis pH 10	6 minutes	Lee et al. (1971)

Hydrolysis pH 7.55(A1 only)	46 hours	Hussain (1969)

Hydrolysis pH 9 (A1 only)	2.2 hours	Hussain (1969)

Hydrolysis pH 10.1 (A1 only)	1 minute	Hussain (1969)

Hydrolysis (25 C) 	190 days	Estimated – ASTER, 2004

Water solubility (20 EC)	69 mg L-1	estimated - ASTER, 2004

Koc	2500 ml/g	estimated - ASTER, 2004

Koc	84 – 10000 ml/g	Estimated from aerobic study (see text)

Bioconcentration Factor (BCF)	350x	Estimated – PBT Profiler

Henry’s Law constant	2.42 x10-17 atm-m3 mol-1	estimated - ASTER, 2004

Vapor pressure (25 EC)	2.31x10-15 mm Hg	estimated - ASTER, 2004



Usage and Fate in Water	

(for pH < 8.5 and temperatures >60C).  The label suggests that the
actual concentration that is to be used should be confirmed with a
bioassay.  Because of the great latitude that the label gives, it cannot
be determined a priori what amount of antimycin may be applied to a
water body.  For both stream and lake applications of antimycin,
downstream movement of antimycin will occur; however the extent is
unknown.  In many cases, application of antimycin may coincide with the
use of livecars containing sensitive species that are placed downstream
of the antimycin application, although this is not a label requirement. 
Use of livecars with sensitive species would allow monitoring of the
dissipation of antimycin effectiveness but is it does not prevent
antimycin from proceeding downstream, and there is some evidence that
such downstream movement can be significant.

	In one study, Tiffan and Bergersen (1996) observed that antimycin was
100% effective at fish kills to at least 1.75 km down stream of a
Colorado creek (pH = 6.3, 9-15oC).  Because there was a 100% fish kill
at 1.75 km and no live cars placed downstream of this distance, the
effectiveness of antimycin probably proceeded much further downstream. 
In this creek, antimycin was applied to the creek at 8 ppb for 8 hours,
which is typical of an application.  The effectiveness of antimycin
observed by Tiffan and Bergersen (1996) is consistent with the fate
properties reported in Table 1, especially given the wide range of
values that the fate properties may attain. Although the velocities of
the streams were not reported by Tiffan and Bergersen (1996), it is
likely that the travel time over a 1.75-km distance was considerably
less than one half life, and thus it would be reasonable that antimycin
would have remained effective in this low pH stream; however, the actual
reason that antimycin remained effective in this stream system is
unclear (although Tiffan and Bergersen speculated that stream gradient
had an effect).  In other streams examined by Tiffan and Bergersen
antimycin was effective to at least 0.5 km downstream.

Estimates of Water Concentrations for Drinking Water

	When antimycin is applied to a water body (whether a lake, pond, or
stream) the most conservative acute concentration that could be used for
drinking water assessments is the application concentration.  The
maximum application concentration is ambiguous but is “roughly” 25
ppb.  However, because of the significant uncertainties regarding the
persistence and sorption properties of antimycin, temporal
concentrations trends and chronic concentrations prediction are also
full of uncertainty.  With this regard, ranges of estimates of chronic
concentrations can be made by using the full range of possible
degradation rates reported in Table 1.  Table 2 gives the range of
chronic concentrations that may result following a 25 ppb application of
antimycin to a water body and shows that chronic concentrations vary
considerably depending on the half life assumed for antimycin,,, with a
value of 4.6 ppb the most conservative based on the longest reported
half life (47 days).

.  Higher initial concentrations would result in proportionally higher
chronic concentrations.

Half life	Chronic Concentration (1 year average)

47 days	4.6 ppb

23 day	2.3 ppb

15 days	1.5 ppb

5 days	0.49 ppb

5.5 hours	0.022 ppb

20 minutes	0.0013 ppb



For human drinking water, antimycin may move downstream from the point
of application (outflow from a lake or a stream) and could enter a
drinking water reservoir.  Because of the uncertainties surrounding the
degradation of antimycin and the great variability in potential travel
times from the point of application to a reservoir, only first
approximations of antimycin concentrations in human drinking water can
be made.  The worst possible case would be the concentration of
antimycin at the point of application—an acute concentration of
“roughly” 25 ppb.  The worst possible case for a chronic
concentration would be the concentration derived using the longest
reported aquatic half life—“roughly” 4.6 ppb, depending on actual
application rate.  These “rough” estimates of 25 ppb and 4.6 ppb
chronic are the EFED-recommended acute and chronic concentrations for
use in human drinking water exposure assessments.   

The above recommended concentrations apply to both surface water and
groundwater. While applications of antimycin are made only to surface
waters, it is possible that antimycin could migrate to groundwater
(e.g., by  leaching beneath a pond or stream); however, the resulting
groundwater concentrations would be no higher than the source surface
water concentration.  Thus, the above recommendations are also
conservative estimates of groundwater concentrations.  

Note that these recommended concentrations are conservative with regard
to what is likely to actually occur as drinking water concentrations. 
In actual situations, antimycin likely will dissipate due to
hydrodynamic dispersion, mixing with side channels, and ultimate
dilution into a larger water body, however these factors are not readily
determined and would be highly localized.

References

Walker, C. R., R. E. Lennon, and B. L. Berger. 1964. Preliminary
observations on the toxicity of antimycin A to fish and other aquatic
animals. U.S. Bureau of Sport Fisheries and Wildlife, Investigations in
Fish Control No. 2 (Circular No. 186). 18 pp.

Lee, T.H., Derse, P.H., and Morton, S. 1971, Effects of Physical and
Chemical Conditions on the Detoxification of Antimycin. Transactions of
the American Fisheries Society 1971;100:13–17

Hussain, A., 1969. Kinetics and mechanisms of hydrolysis of antimycen
antimycin A in solution. Journal of Pharmaceutical Sciences 58:316:320.

Liu, W.C. EE van Tamelan, FM Strong , 1960  The chemistry of antimycen
antimycin A.  VIII Degradation of Antimycen Antimycin A. Journal of the
American Chemical Society 82:1652:1654

Tiffan, K.F. and EP Bergensen 1996. Performance of AntimycenAntimycin in
high-gradient Streams. North American Journal of Fisheries Management
16:465:468.

BR: Rates are set by species within label.

BR: A field sampling technique was developed by USGS, and was to be
field tested in 2006; it is referenced in the Assessment.

BR: Throughout the documents it is difficult to determine if they are
referring to ppb/mg/l of Fintrol or ppb/mg/l of antimycin A.  As
written, it appears that most of the documents are referencing ppb/mg/l
Fintrol.  Overall, all documents need to make this clear, since Fintrol
is not 100% anitimycin A.

BR: No concentrations of 25 ppb for trout are ever used.

