MALACHITE GREEN

First draft prepared by

Dieter Arnold, Berlin, Germany
Bruno LeBizec, Nantes, France
and
Richard Ellis, Myrtle Beach, USA

IDENTITY AND PROPERTIES

The structural identity, some major physical-chemical properties and molecular characteristics of
malachite green salts, its carbinol base and its major metabolite leucomalachite green are summarised
in Table 1. Other important properties are briefly summarised in the below subsections.

Chemical properties

Solvolysis

Goldacre and Phillips (1949) investigated the solvolytic reaction of malachite green and the formation
of the carbinol at various pH values. For 2.7 x 10-4 M solutions at 25°C they found the following
degrees of ionization shown in figure 1.

Figure 1: Solvolysis of malachite green as function of pH

From these results a pK of 6.9 was calculated. At this pH the time to decline half-way from 100%
ionization to the equilibrium value is 2.1 hours. Velocity constants for the solvolytic reaction increase
with increasing pH. The carbinol base which is less soluble than the ionized form is the form in which
malachite green is probably taken up by fish. Therefore, the solvolytic equilibrium influences the
pharmacokinetics of bath treatments.
 2

Table 1: Physical-chemical properties of malachite green, its carbinol base and its metabolite leucomalachite green.

Substance name Malachite green Leucomalachite green Malachite green carbinol base
Total number: 82 Total number: 38 Total number: 20
Examples: Examples: Examples:
Methylene green Leuco malachite green Solvent Green- 1
Grenoble Green Malachite green leuco Malachite Green carbinol
Victoria Green Malachite green leuco base Malachite Green Carbinol base
Depositor-supplied
Aniline Green Tetramethyldiaminotriphenylmethane
synonyms (PubChem)
Benzal Green C.I. Basic Green 4, leuco base
China Green
Fast Green
Burma Green B
Diamond Green Bx
10309-95-2
Chemical Abstracts
(chloride: 569-64-2) 129-73-7 510-13-4
Registry number
(oxalate: 2437-29-8)
PubChem-CID 11295 67215 10521
209-322-8 (chloride)
EINECS 208-109-7
219-441-7 (oxalate)
[4-[(4-dimethylaminophenyl)-
4-[(4-dimethylaminophenyl)- bis(4-dimethylaminophenyl)-
IUPAC phenylmethylidene]-1-cyclohexa-2,5-
phenylmethyl]-N,N-dimethylaniline phenylmethanol
dienylidene]-dimethylazanium
Molecular formula C23H25N2+ C23H26N2 C23H26N2O
Formula weight 329.46 330.47 346.47
 3

Structure (propeller-
like) prepared with
Chemsketch 10.0

UV-VIS spectrum Not found Not found

λmax is 616.5 nm
H. Du, R. A. Fuh, J. Li, A. Corkan, J.
S. Lindsey. PhotochemCAD: A
computer-aided design and research
tool in photochemistry. Photochemistry
and Photobiology, 68, 141-142, 1998.

http://webbook.nist.gov/cgi/cbook.cgi?
ID=C129737&Units=SI&Mask=80#IR
IR spectrum Not found Not found
-Spec

Melting point [°C] 164 About 100 112.0 - 114.0

 4

(MG oxalate; Mallinckrodt-Baker (Sigma-Aldrich 2006)

2006,
http://www.jtbaker.com/msds/englishht
ml/m0286.htm)
Octanol/water partition
0.62 (MG chloride) Not found Not found
coefficient: logKow
40 g/l at 25°C (MG chloride)
Baughman, G.L., et al; Dye Solubilities
Solubility in Water in: Advances in Color Chemistry, No information available Low, exact data not found
Freeman, M., Peters, M.T., eds., NY,
NY: Elsevier (1993)
Leucomalachite green, carbinol and
Impurities found in Malachite green and mono-desmethyl
demethylated derivatives and 4-
commercial products1 leucomalachite green
DMABP

Malachite green is typically available as hydrochloride or oxalate salt. The hydrochloride may have been obtained through precipitation with Zinc chloride
and may be highly toxic to fish.

1
See for example LeGoff and Wood (2008)
Other physical-chemical properties

Adsorption

Adsorption characteristics of malachite green were studied in order to find ways to remove it from
waste waters. Traditional methods of wastewater cleaning only partly remove synthetic dyes such as,
malachite green and structurally related compounds. The adsorption on a variety of inexpensive and
more or less efficient inorganic and organic solid supports was tested as a method for the removal of
malachite green from water and wastewater. When testing the adsorption characteristics of malachite
green from aqueous solutions frequently only the kinetics of decolourisation of the aqueous solution
was measured. Garg, et al. (2003) studied the properties of chemically treated saw dusts (e.g.,
formaldehyde, sulphuric acid). Gong, et al. (2006) studied citric acid modified rice straw and soybean
hulls esterified with phosphoric acid. Hameed and El-Khaiary (2008) used rice straw-derived char.
Mittal (2006) determined the adsorption kinetics from waste water onto hen feathers. Janos, et al.
(2005) studied the sorption of malachite green from waters onto a naturally occurring kind of
weathered and oxidized young brown, acid stable coal called oxihumolite. Tahir and Rauf (2006) used
bentonite clay for the removal of malachite green from aqueous solutions. Wang and Ariyanto (2007)
studied adsorption of malachite green to natural zeoliths.

Carbon-based sorbents show excellent adsorption properties for a considerable number of synthetic
dyes. The preparation of carbon sorbents is generally energy consuming and large amount of carbon
sorbent is needed for the removal of dyes from large volumes of water. Consequently, according to
Forgacs, et al. (2004), the use of commercially available products is fairly expensive. Activated
carbon can remove malachite green from water. A system for removal of malachite green from waters
used for antifungal treatment in hatcheries was developed and described by Marking, et al. (1989).
The removal efficiency was significantly less than 100%. For solutions containing 2 mg/l of malachite
green an average of 23.4 mg of malachite green were adsorbed per gram of carbon.

Recent studies on the sorption using activated carbon with BET surface areas in the order of 1000
m2/g prepared from different sources of material using various models for the calculation of
adsorption isotherms (linearised and non-linear) and for studying the kinetics of adsorption.
According to some authors the adsorption of malachite green was best described using a pseudo-
second-order model with intra-particle diffusion of malachite green molecules within the carbon
particles as a rate-limiting step. The following materials (examples only) were used for the
preparation of activated carbon:

Authors Material used

Singh and Rastogi (2004) Used tea leaves
Rahman, et al. (2005) Rice husks
Başar (2006) Waste apricot
Onal (2006) Waste apricot
Onal, et al. (2006, 2007) Lignite,
Malik, et al. (2007) Groundnut shell waste.
Kumar (2006), Kumar and No information on source material
Sivanesan (2006)
Porkodi and Kumar (2007) Jute fiber.
Zhang, et al. (2008) Arundo donax root

Several groups determined thermodynamic parameters (ΔH, ΔG, and ΔS) and concluded that the
reaction was endothermic.

Malachite green is markedly biosorbed by activated sludge and reduces the rate of oxygen uptake by
activated sludge proportionally to its biosorption (Mihara, et al., 2005). The results of the above
studies suggest that malachite green and its metabolites and breakdown products may not be
 6

completely removed by wastewater treatment and may be present in sufficient amounts in effluents
from industry or waste water treatment plants or other sources to cause residues in wild fish. A group
of researchers has recently claimed – without providing convincing evidence - that they have
published the first example where malachite green was demonstrably taken up by eels caught
downstream from treatment plants in the lakes and rivers surrounding Berlin, Germany (Schuetze, et
al., 2008).

Photodegradation

Hydrogen peroxide has been frequently applied to decolorize synthetic dyes including malachite
green in waters in the presence of suitable photo catalysts. For example, micro porous solid material
prepared by precipitation of phosphotungstic acid and potassium ions, followed by calcinations was
proposed for this purpose (Chen, et al., 2006). The formation of active oxygen species such as the
radicals O2-, HO2 and OH are detected during the degradation of dye, and they are proposed to be
responsible for the degradation of dyes. In such systems CO2 and small organic acids are the main
reaction products.

The presence of catalysts such as TiO2 enhances the rate of photodecomposition of malachite green
under visible light. Both the superoxide anionic radical and the dye cationic radical are essential to the
mineralization of the dyes under visible light-induced photo catalytic conditions (Arpaç, et al., 2007).
Kominami, et al. (2003) used TiO2 nano-particles with various physical properties that had been
prepared by hydrothermal crystallization in organic media (HyCOM) and post-calcination, for photo
catalytic decomposition of malachite green in an aqueous suspension under aerated conditions.
Adsorptivity is a decisive factor for the initial bleaching of malachite green on this material which
follows pseudo zero-order kinetics. Chen, et al. (2007) have studied the reaction mechanism of
malachite green photo degradation on TiO2. They identified a series of N-demethylated intermediates
(mono-, di-, tri-, and tetra-demethylated malchite green) under basic reaction conditions of the
process. These degradation products are also known to be formed metabolically in bacteria and
animals (see below). Under acidic conditions, the whole conjugated chromophore structure of
malachite green was cleaved (Chen, et al., 2007). Hydrogen peroxide can effectively decolorize dye
wastewaters in the presence of Fe(III) – loaded ion exchange resin (Amberlite IRA 200). The
degradation process of Malachite green proceeds via demethylation and phenyl ring openings before
CO2 and small molecules are formed.

Binding to macromolecules

Malachite green can bind to macromolecules. Of interest is the binding to small artificial RNA
molecules (aptamers). The complex can exhibit interesting new properties, for example enzymatic
activities (Brackett and Dieckmann, 2006). In the binding process the RNA adapts to the ligand
(“adaptive binding”), but the ligand itself also undergoes conformational changes ("induced fit")
(Nguyen, et al., 2002; Nguyen et al., 2004). The crystal structure of such complexes has been studied
using tetramethylrosamine (TMR), a high-affinity analogue of malachite green (Baugh, et al., 2000).
The properties and a number of applications of such complexes have been published. One possible
use is the determination of malachite green itself, because aptamers are known which enhance
malachite green fluorescence by factors in the order > 1000.

It is long known that malachite green binds to DNA (Nordén, et al., 1978; Bhasikuttan, et al., 2007).
Cationic triarylmethane dyes also have complex-forming properties with proteins (Tacal and Ozer,
2004). A full discussion of all these properties is beyond the scope of this monograph.

Industrial uses of malachite green

Malachite green is used extensively as a dye for leather, wool, cotton, jute, paper, certain fibres, etc.
For such purposes it has been produced in large quantities and extremely variable qualities. About 10-
 7

15% of all dyes are directly lost to wastewater in the dyeing process (Parshetti, et al., 2006).
Frequently the purity of products used in biological studies has not been reported. In human medicine
the carbinol is/has been used as a wound antiseptic and as a treatment of mycotic skin infections, and
in staining of tissues and bacteria.

RESIDUES IN FOOD FROM AQUATIC SPECIES AND THEIR EVALUATION

Conditions of use in aquatic animals

Malachite green has been used as a fish fungicide in closed systems alone or in combination with
other chemicals such as formaldehyde for decades. Frequently reported concentrations are about 0.05
- 0.1 mg/kg. It is important to use zinc free preparations in order to avoid metal intoxications of the
fish. Therefore, the oxalate was most frequently used. Foster and Woodbury (1936) were reportedly
the first to introduce its use as fungicide and antiseptic. It took almost 47 years until researchers
considered for the first time the possibility that malachite green could be taken up by fish (Poe and
Wilson, 1983).

Malachite green has been used for the treatment of eggs of fish and crayfish. Malachite green is also
an effective topical and systemic antiprotozoal agent. Reported types of treatment of fish include dip
treatment, flush treatment, sustained culture treatment and application in feed. Extremely wide ranges
of concentrations and exposure times have been used (review by Alderman 1985).

In a review of historical uses of malachite green, Sudova, et al. (2007) discriminate between dip
treatments of 10-30 seconds duration and concentrations up to 100 mg/L to treat topical fungal
infections, short-term malachite green bath treatments of 60-90 minutes duration at 6.7 mg/L, and
long-term bath treatments of six days duration at 0.15 mg/L for salmonids and 0.5 mg/l for cyprinids.
They state that this type of treatment was used to control protozoan ectoparasites, particularly the
ciliated protozoan Ichthyophthirius multifiliis. Malachite green concentrations can be reduced in
multi-component baths, for example in combination with formaldehyde. Treatment with malachite
green can produce numerous side-effects in treated fish and fish eggs.

An important factor determining therapeutic and toxic effects is the temperature. Batch to batch
variation in concentration and purity of the dye and lack of standardization of test conditions have
been major confounding factors in the judgment of effectiveness of doses and exposure times.
Therefore, it goes beyond the scope of this monograph to make any conclusive statements and
comparisons about dosages and other conditions of use. Information on treatment conditions will be
given individually in connection with the discussion of pharmacokinetic and residue studies.

Malachite green is toxic to fish, in particular to small fry. Bills, et al. (1977) used standard laboratory
tests in order to determine the LC50 under various conditions of temperature, pH, and hardness of the
treatment bath and of duration of treatment. Fingerling fish of a great variety of species, weighing 0.5
to 1.5g were used for the tests. Increase in exposure time significantly increased the acute toxicity. In
short term-exposure (3 and 6 hours) of rainbow trout and channel catfish higher temperatures
increased the acute toxicity. At the longest tested exposure time (96 hours) the temperature effect
disappeared in rainbow trout. pH and hardness had no significant influence on acute toxicity. As an
example, some data obtained with rainbow trout were selected from the original paper and are
presented below in Table 2. The original work also provides the 95% conficence intervals of the LC50
which is not given in the below Table 2.
 8

Table 2: Acute toxicity of malachite green to rainbow trout

Incubation time [hours]

Temperature Water
pH 3 6 24 96
[°C] hardness
LC50 [mg/L
7 soft 7.5 >2 2.3 0.4 0.17
12 soft 6.5 >2 1.0 0.28 0.28
12 soft 7.5 1.4 0.8 0.36 0.25
12 very soft 8 2.0 0.8 0.36 0.29
12 soft 8 2.3 0.8 0.28 0.23
12 hard 8 2.3 1.4 0.35 0.29
12 very hard 8 2.4 0.8 0.28 0.25
12 soft 8.5 2.6 1.0 0.28 0.21
12 soft 9.5 >2 1.3 0.37 0.17
17 soft 7.5 1.4 0.6 0.57 0.28

PHARMACOKINETICS AND METABOLISM

Pharmacokinetics in Fish

Physiological facts: Relationship between carcass weight and organ weight of fish

Some basic physiological facts are necessary to understand the kinetic behavior of malachite green in
fish. Schmelzing and Claus (1990) found that in rainbow trout (Oncorhynchus mykiss) absolute organ
weights increased with increasing carcass weight while their weights as a proportion of carcass weight
decreased. Heart and liver weight were highly correlated with carcass weight, while the correlation
between spleen and carcass weight was moderate. According to Corti (1948), muscle meat, liver and
kidney make up 70.2, 1.2, and 0.77% of body weight in rainbow trout. The range of body weights
(n=7) was 54.1 to 82.3g and corresponding range of muscle meat weight ranged from 61% to 78% of
the body weights of the animals. The corresponding figures for eel were 80.9, 1.32, and 0.75 % for
muscle meat, liver and kidney, respectively.

Fish physiology: Uptake of hydrophobic compounds

Fish normally take up hydrophobic compounds via the gills by passive diffusion (Gobas, et al., 1986).
Rates of uptake can be a function of: water flow over the gills, blood flow through the gills, diffusion
through the aqueous stagnant layer along the gill epithelium, or diffusion through the gill membrane.
Hayton and Barron (1990) have summarized: “In general, for any particular chemical, only one of the
barriers is operative with the resistance offered by the others being negligible. The rate-limiting
barrier is determined by the physico- and biochemical properties of the substance: molecular size,
lipophilicity, binding to blood proteins and formed elements. The resistance of each barrier is affected
differently by variables such as temperature, molecular size, lipophilicity and body size of the animal.
When the resistance offered by the gill barriers is low, uptake may be controlled by transfer to storage
tissues, e.g., by blood flow to adipose tissue”.

When fish increase their water flow, e.g., with decreasing oxygen concentrations or other types of
stress, uptake per unit time frequently increases. Sijm, et al. (1994) studied the influence of blood and
water flows on the uptake of some hydrophobic compounds by rainbow trout. The fish used in their
experiments had an average weight of 110 ± 12g (n not given). The temperature was 12°C. For all
compounds studied the uptake rate constants increased with water flow between 0.045 and 0.52
l/min/kg body weight and remained constant at higher flow. The uptake rate constant was constant for
blood flow between 4.4 and 10 ml/min/kg body weight but doubled when the blood flow was
 9

increased from 10 to 20 ml/min/kg body weight. From their findings they deduce that water flow will
practically limit uptake of hydrophobic chemicals in fish weighing more than 5g.

Gill physiology

In a study on water flow and gas exchange at the gills of rainbow trout (Salmo Gairdneri) Erickson
and McKim (1990) developed a simple flow limited model for the exchange of organic chemicals at
fish gills. The mathematical model for the exchange of organic chemicals by fish gills was formulated
based solely on the limitations imposed by the flows of water and blood into the gills. The model
could be useful for approximate assessments of accumulation of organic chemicals by fish. For large
rainbow trout, the model was found to closely follow the magnitude and trends of observed gill uptake
rates over a range of octanol/water partition coefficient from 1 to 106 .

Davis and Cameron (1971) estimated the volume of water passing over the gills per unit time
(ventilation volume). The technique was direct measurement. For this purpose a rubber membrane
was stitched round the margin of the mouth of the fish in a way that it separated inspired and expired
water. 18 fish of a body weight of 210.3 ± 2.3 g were used at 8.6°C to determine ventilation volume
when the animals were quiet. The estimation was repeated 4-11 times with each animal. The lowest
individual estimate of ventilation volume observed in one fish was 22.0 ml/min; the average per fish
ranged from 26.0 to 49.0 ml/min. The overall average was 37.0 ±7.4 ml/min. The corresponding mean
ventilation rate was 74 breaths/min and the mean ventilator stroke volume was 0.5 ml/breath. When
the fish struggled or were disturbed maximum values rose as high as 162 ml/min in one animal
(average 88.2 ± 43.7).

Nichols, et al. (2004) developed a physiologically based toxico-kinetic model for dietary uptake of
hydrophobic organic compounds. Malachite green administered via the diet shows unsatisfactory
efficacy. Therefore, this model cannot be applied to the data discussed in this monograph.

Pharmacokinetic studies in rainbow trout

Alderman and Clifton-Hadley (1993) conducted a pharmacokinetic study in rainbow trout

(Onkorhynchus mykiss). The dye was administered through uptake from the water bath. The heavily
vascularized gill was assumed to be the principal site of malachite green uptake from solution under
these conditions

The fish used in the main pharmacokinetic experiments of the study were two separate groups of
rainbow trout with average body weights of 241 ± 33 g (n not given) for studies conducted at 16°C
and 199 ± 20g (n not given) used for studies conducted at 8°C, respectively. A third group of 30 fish
of 50g body weight was used for additional experiments in which residues in muscle were determined
in individual fish at 16°C (results not given). The pH was 7.6 and total hardness was 13.8° dH. Under
these conditions more than 95% of the malachite green was in the carbinol form. The purity of the
malachite green was tested by thin layer chromatography. Treatment solutions containing 1.6 mg/l
(4.86 µmoles/l) were prepared from a commercial liquid formulation 15 h before use in order that the
dye-carbinol equilibrium concentrations could become established. Treatment time was 40 minutes.

In the main experiments fluids and tissues of five fish per time point were pooled and stored frozen
for analysis. However, one graph of the publication shows individual kinetic data in serum obtained in
a separate experiment. Samples of serum and bile were allowed to defrost before analysis. After
dilution with buffer at pH 4.0 they were extracted for 24 h into pentan-1-ol (no information on the
partitioning of malachite green and its metabolites was provided). After 24 h, samples were
centrifuged at 2000 × g for 30 min, giving a clear pentan-1-ol supernatant. All other tissues were
allowed to defrost overnight and weighted composites were then blended in 2% pepsin adjusted to pH
2.0 with HCl. The blended samples were kept for 18h at room temperature (20°C) and were shaken
thoroughly several times in that time. The samples were then partitioned at room temperature at
approximately pH 4 into pentan-1-ol. Following the addition of pentan-1-ol the flasks were shaken
 10

vigorously before being left (with further shaking) for a further 24 h. Samples were then shaken
thoroughly again before centrifugation at 2000 × g for 30 min at 4°C.

Sample extracts were scanned at wave lengths from 540 to 700 nm. Peak absorption for the malachite
green dye ion in extracts was 625 nm. Reported recoveries at 10 mg/kg were 19, 82, 75, 68, and 60%
for serum, liver, kidney, muscle, and viscera, respectively. The spectrophotometer was calibrated
against extracts of malachite green obtained from representative fish tissues spiked with known
concentrations of malachite green and after equilibration subjected to the same extraction procedure
as described above for the experimental samples. The reported LOD was about 50 µg/kg (data not
shown). Some tissues presented general or occasional problems in residue determination including
interference of colored co-extracted substances. The paper exhibits a number of weaknesses:
• Concentration of the drug in water was not monitored during the experiment.
• The description of the experiment lacks precision. The exact relationship between
total weight of treated fish and weight of the bath is not given. One may speculate
that it not exceeded 24 x 5 x 0.241 kg of fish in 725 kg bath.
• Fat was not sampled.
• It was not determined to what extent leucomalachite green was extracted and whether
it was re-oxidized to the malachite green. Therefore, the kinetic data of this study can
probably not be interpreted.
• Insufficient information on method validation is provided. The study typically does
not provide information on results obtained with individual animals; thus no estimate
of the biological variability is possible.
• The authors report that malachite green appeared in the serum very rapidly, with
concentrations increasing steadily until the fish were removed from the dye (results
not shown). In fact, peak concentrations shown in figure 2 of the paper were in the
order of 13-13.5 mg/kg at both 8 and 16°C.
• The text states that peak concentrations in muscle were reached 90-120 min after the
end of exposure and gives values of 6.81 and 10.79 mg/kg for the peak concentrations
reached at 8°C and 16°C, respectively; however, the curve describing the influence of
exposure time never exceeds approximately 1 mg/kg for muscle.
• The text states that when a group of small rainbow trout was exposed to malachite
green and individual muscle residues examined at 24 h post-exposure, considerable
fish-to-fish variations were evident; however neither data nor an estimate of the
variance are provided.
• The legend to the figure describing the kinetics in bile and the corresponding text
state that bile could only be reliably collected for the first 40 h; however, the
corresponding graph shows data points for 48 and 72 hours.
• The analytical method used was inadequate.
• The calculated half-lives for serum and tissues cannot be compared because they are
either based on different kinetic models used for curve fitting, or - even when the
same model was used - the distribution in time of the data points covered different
phases of the kinetics.
• Some variations in treatment parameters and associated effects on results are
discussed without providing any data.
• Extrapolations (calculations and results not shown) far beyond the experimental time
points and orders of magnitude below the measured concentrations.

The following conclusions may be drawn from the results of the study: Uptake was lower at 8°C than
at 16°C; the initial rate of decrease of the optical density at 625 nm was higher at 16°C than at 8°C;
the maximum concentrations found in tissues were in the order of (all values in mg/kg):

Temperature/Tissue Serum Liver Kidney Muscle

8 °C 13 8 7.8
16 °C 13.5 16.5 34 10.8
 11

However, this statement is only valid if one can assume that the biotransformation of malachite green
in this experiment was a slow process compared to the rate of uptake or that leucomalachite green is
not picked up by the analytical method. If one assumes that malachite green was rapidly metabolized
to leucomalachite green and other molecules (as shown in other studies) which are all extracted, then
all results of this study could be meaningless numbers; if leucomalachite green was re-oxidized the
above given maximum concentrations could represent an estimate of total residue.

A table summarizing pharmacokinetic and residue data in aquatic species was found on the website of
the U.S. FDA. The table includes the work of Alborali, et al., published under the title: “The
persistence of malachite green in the edible tissue of rainbow trout (Oncorhynchus mykiss)” in the
Journal Rivista Italiana di Acquacoltura 32, 45-60. The summaries of the cited findings are given
here: Rainbow trout (Oncorhynchus mykiss) with a body weight of 60-80 g were exposed at 18 °C for
one hour to a solution of 1 mg/l of malachite green. Residues were determined by HPLC (no details
given). The following information contained in the FDA document is provided for gills, kidney,
muscle and skin:

“Gill: Residues decreased fairly rapidly during the first 320 days reaching 260 by d41, and below 1
ng/g after 7th month. T 1/2 in the 20 d range initially, with later slow decline in the 50d range.
Kidney: Max residue on d(ay) 30 = 1650 ng/g, declined to below 1 ng/g after 41 d.
Muscle: Residues high for 34 days - above 1,000 ng/g. Decreased to 200 ng at the 4th month, around
100 by 150 d, slowly declining to below 10 ng/g by 9th month. T1/2 in the 40-50 d range.
Skin: Residues decreased fairly rapidly during first 20 days, to about 2500 ng/g, d 50= approx 1000,
200 ng/g round the 4th month, below 1 ng/g at day 283. T1/2 in the 50 d range.”

Pharmacokinetic studies in channel catfish

The pharmacokinetics and metabolism of malachite green in channel catfish (Ictalurus punctatus)
were examined by Plakas, et al. (1996) after intravascular dosing or waterborne exposure. The
intravascular dosing solution contained 0.8 mg of 14C labeled dye cation per ml of 0.85% aqueous
NaCl solution corresponding to a specific activity of 0.925 MBq/ml or 1.16 MBq/mg. For waterborne
exposures, the initial dye concentration was 0.8 mg/l corresponding to a specific activity of 0.185
MBq/ml or 0.231 MBq/mg. The channel catfish were 0.5 to 0.7 kg. For the collection of blood and
urine the dorsal aorta and urinary bladder were cannulated.
ƒ Five animals were dosed intravascularly with [14C] malachite green at a dosage of 0.8 mg/kg
body weight. Blood specimens were collected 2.5, 5, 7.5, 10, 15, 20, 30, and 45 min and 1, 2, 4,
6, 8, and 10 h after drug administration.
ƒ Five animals were transferred to the dosing solution (0.8 mg/l). Blood specimens were taken at
15-min intervals during the 1-h exposure period. At the end of the dosing period, fish were
briefly rinsed in a water bath. Blood specimens were taken at 10, 20, 30, and 45 min and at 1, 2,
4, 6, 8, and 10 h after the end of the dosing period.
ƒ To determine the tissue distribution of malachite green and its residues after waterborne
exposure, groups of five animals were exposed to [14C]-malachite green solutions (0.8 mg/l) for
1 hour. Animals were killed and tissues collected immediately after dosing (designated 0 h) and
at 2, 4, 24, 96, 168, and 336 h (14 days). Additional animals dosed with unlabelled malachite
green were sacrificed after 28 and 42 days.

For HPLC determination of malachite green and leucomalachite green, plasma was extracted with
acetonitrile. Muscle was subjected to a more complex procedure involving extraction with acetonitrile
– acetate buffer, re-extraction, solvent partition, and SPE. HPLC fractions were subjected to post-
column oxidation to the malachite green ion. Mean extraction efficiencies for malachite green and
leucomalachite green from plasma and muscle were 85 - 95%. Mean recovery of total radioactivity
from muscle of treated fish was 88%, however, individual animal data were not provided.
 12

The mean of the concentrations of the plasma samples of five animals declined rapidly after
intravascular dosing. Simultaneously the corresponding concentrations of leucomalachite green
increased rapidly and reached an average maximum concentration of 0.875 μg/ml in the samples
taken at 0.75 h after dosing. At this time point the corresponding concentration of the parent malachite
green was 0.6 μg/ml. At ten hours the concentrations of leucomalachite green and of parent malachite
green were 0.20 and 0.05 μg/ml, respectively. The sum of these two compounds accounted for
approximately 70% of the total drug equivalents at each sampling time. The authors fitted a six
parameter (three exponential terms) equation to the data obtained at 14 time points of which nine
points were collected during the first hour after treatment. They estimated a terminal half life of 6.2
hours for malachite green.

During waterborne exposure at 21 oC total radioactivity and the concentrations of both malachite
green and leucomalachite green increased very rapidly to 2.77 and 1.56 μg/ml plasma, respectively.
Concentrations of malachite green then started declining immediately; however the peak
concentration of leucomalachite green was 2.36 μg/ml 1 hour after transfer of the fish to clean water.
The authors state that the decline followed a tri-exponential curve (results not shown) and estimated a
terminal half life of 4.7 h. After 10 h concentrations of malachite green in plasma had declined to the
limit of detection of 0.25 μg/ml. The authors state that the concentrations of leucomalachite green
were still 30 times higher. The half life of terminal depletion of leucomalachite green was not
estimated. No information on biological variability is available. The water bath conditions were such
that the ratio of the ionic and the carbinol form of malachite green was 6:10.

Malachite green and its metabolites were widely distributed and concentrated in the tissues.
Concentrations of radioactive residues were highest in the excretory tissues and fat and lowest in the
muscle and plasma. Concentrations of total residues exceeded the initial concentration of the water
bath in all tissues, with the following few exceptions: bile (0 hours), spleen (168 and 336 hours), and
muscle (336 hours). The authors’ results, originally in tabular form, are summarized in figure 2 below.
The variability of the results expressed as relative standard deviation of the mean of 5 data points
ranged from 0.07 to 0.76. Variability was lowest in muscle tissue and highest in skin and bile. The
variability increased over time. The high values of many standard deviations suggest that the residue
concentrations might not be normally distributed and that the averages calculated by the authors are
not the ideal parameters to show a central tendency. Concentrations of residues in skin were always
higher than in muscle. The highest ratio was 2.2, observed 24 hours after treatment.

The concentrations of malachite green and of leucomalachite green in plasma were 3.29 and 1.94
μg/ml, respectively, immediately after dosing. One day after dosing, malachite green levels were at
the LOQ while leucomalachite green levels were 0.11 μg/ml at day 14 after treatment.

In muscle, malachite green and leucomalachite green concentrations were 1.18 and 1.45 mg/kg,
respectively, at the end of the exposure period and 0.012 and 0.52 mg/kg 14 days after treatment.
Results obtained at other time points are not numerically given. The elimination of malachite green in
muscle appeared biphasic with a terminal half-life of about 67 h. Concentrations of leucomalachite
green were quantifiable for up to 42 days (0.02 mg/kg ). Unidentified metabolites eluting before
leucomalachite green during HPLC were found. The sum of the concentrations of these three
metabolites reached a maximum of 31.3% of the total residue at 24 hours after treatment. No detailed
pharmacokinetic information is provided for the water-borne exposure; however the authors state that
the half lives for malachite green and for leucomalachite green were 2.8 and 10 days, respectively.

The effect of pH of the exposure solution was studied at pH values of 6, 7, and 8. When catfish were
exposed to solutions of 0.8 mg/l of malachite green for one hour, uptakes increased significantly with
increasing pH, determined by the concentrations of malachite green and leucomalachite green in
plasma and muscle immediately after exposure. This may be due to the change in equilibrium
concentrations of the cation and the carbinol and in the rates of conversion of the cation to the
carbinol. The well designed and conducted study cannot be used for the derivation of MRLs because
statistical evaluations are not possible in the absence of individual animal data.
 13

Figure 2: Total residues of malachite green in tissues of channel catfish.

Pharmacokinetic studies in juvenile eels

Bergwerf, et al. (2004) exposed 450 juvenile eels (Anguilla anguilla) of an average weight of 4.1 g
(range 1.5 to 7.4 g) for 24 hours to malachite green in a water bath with temperature varying between
23.0 and 26.5°C and at pH values between 7.0 and 7.8 in 40 L water. A malachite green concentration
of 0.1 mg/l was intended. Fish were then cultured in malachite green free water at 21.6 to 27.5°C.
During this period pH varied between 6.0 and 7.8. Before, and at various time points during, and up to
100 days after treatment, ten fish and water were sampled (50 ml per time point). The fish did not
grow during the study.

The whole fish were cut in fine pieces and 2g of cut tissues were blended in buffer and extracted with
buffer/acetonitrile mixtures. Partitioning with dichloromethane was followed by solid phase
extraction. Brilliant green was added as internal standard before HPLC analysis. Water samples were
mixed with buffer, acetonitrile and internal standard prior to analysis. Two reversed phase columns
(phenyl-hexyl and C8) were used in series and were eluted with a mixture of 60% (vol/vol) acetonitrile
and 40% (vol/vol) of 0.05M ammonium acetate buffer, pH 4.5, at 0.6 ml/min. The eluate was
monitored at 620 nm after post column oxidation. Recoveries (n=36) were 61 ± 6% for malachite
green and 88 ± 10% for leucomalachite green. Results were recovery-corrected.

Analysis of water revealed that the starting bath concentrations were only 0.032 mg/L instead of 0.1
mg/l. This concentration further decreased exponentially during the experiment and fell below the
limit of detection at 12 hours. A definitive reason for the low initial concentration could not be found.
The further decrease can apparently be explained by the uptake of malachite green by the fish. One
can roughly estimate the total amount of malachite green taken up by the fish from the sum of the
concentrations of malachite green and leucomalachite green found at a given time point multiplied
with number and average weight of fish present in the bath at this time point plus the accumulated
amounts removed from the bath by sampling 10 fish at every earlier time point. The figure 3 shows
 14

the results of such a crude calculation. The total amount of drug initially present in the bath was
estimated as approximately 1.28 mg using the information given by the authors. The small
discrepancies are probably due to the crude estimates used in these calculations.

Figure 3: Estimated time course of the exhaustive uptake of malachite green in the exposure
experiment.

The density of fish in the small bath which absorbed the drug and the long exposure time probably
caused a total uptake of the malachite green present. Even the higher originally intended concentration
would have been too low for such an experiment. The figures 4a and 4b summarize the
pharmacokinetic results of the experiment. The symbols represent mean values. The bars show the
range. If a lower bar extends down to 0.2 µg/kg, the result of the analysis was <LOD. The LOD was
given as 0.2 µg/kg.
 15

Figure 4a: Concentration changes of malachite green and leucomalachite green during and
after bath exposure to 0.032 mg/L of malachite green.

Figure 4a shows the averages of the measurements obtained with 9-10 individual fish at every time
point and the range of the individual values. It is not clear how the averages have been calculated at
time points where “non-detects” occurred. The latest time point at which the results were not
influenced by “non-detects” was 480 hours for malachite green and 1487 hours for leucomalachite
green. Malachite green was quantified in the one or other fish up until day 62 following exposure and
leucomalachite green was found in the one or other fish over the whole 100 days observation period
following exposure. This finding is important in view of the very limited dose administered.
 16

Figure 4b: The exposure and early depletion part of figure 4a.

For graphical reasons figure 4b shows bars indicating ranges of individual observations only for
selected time points well spaced on the time scale. The figure shows that almost immediately with the
uptake of malachite green at the beginning of the exposure period also the concentration of
leucomalachite green starts increasing. Due to the circumstances described above, the data of this
study are neither useful for pharmacokinetic evaluations nor any further interpretations.

METABOLISM

Metabolism in Micro-organisms

Fungi

Some ligninolytic fungi have been found capable of decolorizing synthetic dyes. This is due to their
production of enzymes such as, laccase and Mn-peroxidase that enable these microorganisms to
oxidize a broad range of substrates. Studies have focused on the possible use of some model wood-
rotting white-rot species (Phanerochaete chrysosporium, Trametes versicolor, Pleurotus ostreatus
and others) for decolorization of synthetic dyes. Ligninolytic cultures of the white rot fungus
Phanerochaete chrysosporium were shown to metabolize crystal violet and malachite green to N-
demethylated metabolites catalyzed by lignin peroxidase. Extracellular fluid obtained from
ligninolytic cultures of this fungus, retained the activity provided that an H2O2-generating system was
supplied. Non-ligninolytic (nitrogen-sufficient) cultures also degrade crystal violet by another
mechanism without producing N-demethylated metabolites (Bumpus and Brock, 1988). Eichlerova, et
al. (2006a) investigated the dye decolorization capacity of two white rot fungi, Dichomitus squalens
 17

and Ischnoderma resinosum (Eichlerova, et al., 2006b). D. squalens showed high decolorizing
capacity. I. resinosum decolorized malachite green to a lower extent up to a concentration of 0.1 g/l.

A total of 26 white rot fungi from Argentina were tested for their ability to produce lignin-modifying
enzymes and decolorize industrial dyes (Levin, et al., 2004). Ten of the strains decolourised all tested
dyes including malachite green. The mycelia were grown on solid malt extract/glucose media
containing the dye. All ten strains produced laccase, lignin peroxidase and manganese peroxidase on
solid medium. White-rot fungi normally require a lignocellulose substrate. Their use in polluted water
streams or soils may be problematic since species of these typical wood colonizers do not exhibit
satisfactory growth and competitiveness under such conditions.

Litter-decomposing fungi differ from wood-rotting species with respect to their growth substrate,
forest litter and soil. They are characterized by higher C:N ratio and microbial activity. Laccase is the
most common ligninolytic enzyme among these organisms and Mn-peroxidase is produced only by
some species (Baldrian and Snijdr, 2006).

Cha, et al. (2001) performed biotransformation experiments of malachite green with cultures of
Cunninghamella elegans, a filamentous fungus which had previously been shown to enzymatically
catalyse N-demethylation and N-oxidation reactions of a number of chemicals. Metabolites were
analysed using HPLC-diode array and HPLC-MS methods. Malachite green was reduced to
leucomalachite green and also converted to N-demethylated and N-oxidized metabolites, including
primary and secondary arylamines. The mono-, di- and tri-desmethyl derivatives of malachite green
and the mono-, di-, tri-, and tetra-desmethyl derivatives of leucomalachite green were found in the
supernatant following removal of the mycelium. Malachite green N-oxide was only detected in the
mycelia. Identical patterns of metabolites were observed with malachite green and with
leucomalachite green as initial substrate. After prolonged incubation only reduced metabolites were
found suggesting that parent malachite green and N-demethylated metabolites were reduced by the
fungus. Microsomal fractions did not produce reduced metabolites in the absence of NADPH. The
cytochrome P450 inhibitor metapyrone completely inhibited the biotransformation reactions.

Intestinal bacteria

Henderson, et al. (1997) studied the metabolism of malachite green by intestinal microflora from
human, rat, mouse, and monkey fecal samples and 14 pure cultures of anaerobic bacteria
representative of those found in the human gastrointestinal tract. All complete microfloras were very
efficient in reducing malachite green to leucomalachite green (human and rhesus monkey intestinal
microfloras, C3H/HEN-MTV mouse intestinal microflora, and Fisher 344 rat intestinal microflora).
Of the bacteria commonly found in the human intestinal tract, Clostridium perfringens (ATCC 3624),
Escherichia coli (ATCC 25922), and Peptostreptococcus anaerobius (ATCC 27337) converted almost
all of the dye to the leuco derivative. The conversion was monitored with HPLC with diode array
detection and the structure was confirmed by mass spectrometry.

Baker’s yeast (Saccharomyces cerevisiae (MTCC 463) was also shown to effectively decolorize
malachite green, primarily through reductive pathways (Jadhav and Goindwar, 2006). A number of
other bacteria have been positively tested for decolorizing capacity of malachite green. A complete
review would go beyond the scope of this monograph.

Metabolism in Laboratory Animals

Rats and mice

In short term feeding studies, Culp, et al. (1999) have shown that MG is sequentially N-demethylated
to secondary and primary aromatic amines in rats and mice both before and after reduction to LMG.
Female and male B6C3F1 mice and Fischer 344 rats were fed up to 1200 mg/kg malachite green or
1160 mg/kg leucomalachite green for 28 days. The malachite green used was ≥ 94% pure. Impurities
 18

detected were leucomalachite green (1%) and demethlyated derivatives of malachite green (3.5%).
Leucomalachite green was ≥ 98% pure. Impurities detected were malachite green and mono-
desmethyl leucomalachite green. Livers were extracted using a modification of a published method
(Roybal, et al., 1995). The extracts were analysed by HPLC connected to a post-column oxidation
chamber and a photodiode array detector. Analyses using HPLC-APCI/MS also were performed. The
desmethyl derivatives were synthesized to confirm structures in the samples subjected to APCI/MS.

In HPLC-APC/MS analysis of liver extracts from rats treated with leucomalachite green the primarily
seen compounds were protonated leucomalachite green, protonated demethylated derivatives and the
molecular ions of malachite green N-oxide and demethylated N-oxide. A small, but measurable,
amount of malachite green was also found. At higher cone voltages additional collision-induced
diagnostic fragments were found that were formed following losses of dimethylaniline-, methyl-, or
phenyl- moieties. The appearance of these molecules was consistent with the fragmentation pathways
previously published for leucomalachite green (Doerge, et al., 1998a) who observed similar sequential
demethylation in a thyroid peroxidase-catalyzed reaction of leucomalachite green. A dose-related
increase in leucomalachite green and metabolites was observed in both rat and mouse liver extracts.

Similarly, HPLC-APC/MS analysis of liver extracts from rats treated with malachite green detected
the molecular ions for malachite green, its mono-, di-, tri-, and tetra-desmethyl derivatives, and
malachite green N-oxide. A small, but measurable, amount of leucomalachite green was also detected.
Higher cone voltages produced fragments consistent with those previously reported by Doerge, et al.,
(1998b). These authors incubated leucomalachite green with tyrosin peroxidase, iodide, and tyrosine
in the presence of an H2O2 generating system and obtained the mono-, di-, and tri—desmethyl
derivatives of leucomalachite green as well as malachite green and malachite green-N-oxide.
Concentrations of malachite green and metabolites increased with increasing dose.

The formation of both the symmetric and asymmetric di-desmethyl malachite green metabolite could
be demonstrated with the symmetrical isomer eluting first. When liver extracts were analysed using
HPLC/UV detection, leucomalachite green was the major product detected in rats fed leucomalachite
green (accompanied by small amounts of mono- and di-desmethyl-leucomalachite green) and
malachite green was the major product detected in the livers of rats and mice fed malachite green
(accompanied by mono- and di-desmethyl malachite green and leucomalachite green – and in the case
of rats mono- and di-desmethyl-leucomalachite green).
32
P-Postlabeling of liver DNA indicated the formation of a DNA adduct, or co-eluting adducts, that
increased with increasing dose, in rats and mice fed leucomalachite green or malachite green. Cho, et
al. (2003) mention that malachite green and the N-demethylated derivatives of malachite green and
leucomalachite green are capable of forming DNA adducts in vivo, with the binding being
consistently greater with the ionic MG derivatives.

Figure 5 below summarizes the structural elements of the N-desmethyl metabolites of malachite green
and leucomalachite green.
 19

Figure 5: Structures of malachite green and leucomalachite green.

R3 R3
N N
R4 R4

Structures of malachite H
green, leucomalachite
green, and demethylated
derivatives
+ R2 N
N R1
R2 R1
Malachite green Leucomalachite green
R1 R2 R3 R4 R1 R2 R3 R4
Parent molecule CH3 CH3 CH3 CH3 CH3 CH3 CH3 CH3
Desmethyl- CH3 CH3 CH3 H CH3 CH3 CH3 H
Di-desmethyl- CH3 H CH3 H CH3 H CH3 H
(symmmetric)
Tri-desmethyl- CH3 H H H CH3 H H H
Tetra-desmethyl- H H H H H H H H

Metabolism in Food Producing Animals

Fish

No systematic metabolism study has been performed in fish. However, many of the degradation
products formed either under physical-chemical conditions or in in vitro and in vivo studies in
bacteria, fungi and laboratory animals have also been found in fish. The potential presence of such
molecules in monitoring samples is largely ignored by analysts and only a few methods use protective
agents in order to prevent the breakdown of incurred residues during extraction and cleanup.

TISSUE RESIDUE DEPLETION STUDIES

Fish

Poe and Wilson (1983) reported in a short note to the journal “Progressive Fish Culturist” that they
had observed a green color developing after some storage time on the surface of the tissue of skinned
frozen fish. The color could be extracted using a procedure for lipid extraction and was identified as
malachite green by infrared and visible spectra. They conducted a small series of experiments with
channel catfish of body weights ranging from 0.3 to 0.7 kg in which they varied exposure conditions
(exposure levels and frequencies). Samples of visceral fat and the carcass were frozen. The green
color appeared typically in visceral fat more rapidly than in muscle. When Alderman published his
frequently cited article “Malachite green: a review” in 1985, uptake and residues had not yet become a
significant issue in the scientific literature.

Eggs and fry

Allen and Hunn (1986) reported that malachite green accumulates in the eggs of gravid female salmon
after treatment and is detectable in eggs and newly hatched fry.

Meinertz, et al. (2005) determined residues of [14C]-malachite green in eggs and fry of rainbow trout
after treatment of the eggs. The treatment method was flush treatment (a concentrated solution of
malachite green was added to an incubation unit and flushed through with fresh water). At the
beginning of the treatment the eggs were approximately 2 days old. Hatching began on day 25 and
 20

was completed on day 31. Six groups of 250 eggs each were treated in 500-ml glass test aquaria.
Before treatment water flow to the aquaria was stopped and the volume was drawn down to 265 ml.
Eggs in all aquaria were exposed simultaneously using a distribution manifold. The expected nominal
concentration was 1.0 mg/l. Water flow was re-established immediately after treatment. Treatment
was performed on days 0, 3, 6, 9, 12, 15, 18, 21, 24, and 31. Water samples were taken at every
treatment from one randomly selected aquarium immediately after addition of the treatment solution
and 2.5, 5, 10, 15, 20, 25, 30, and 60 minutes after water flow was re-established.

Ten eggs were sampled from each test aquarium immediately before each treatment through day 24.
Five were prepared for combustion analysis (92.8 to 95.8 combustion efficiency) and the remaining
five were used for analysis of malachite green residues by HPLC with post-column oxidation and
visible light detection. On day 31, 10 fry were sampled from each aquarium immediately before the
final treatment and 0, 6, 12, and 24 hours, and 2, 4, 7, 12, 17, 22, and 28 days after treatment. Mean
peak concentrations of chromatic malachite green in water were 0.37 mg/l and were slightly higher
than the radioactive concentration equivalents. The intended nominal concentration was not reached
and individual measurements were extremely variable at all time points.

Untreated eggs contained measurable concentrations of unlabelled malachite green. Pre-treatment

radioactive concentration equivalents in eggs and fry increased from day 0 to day 31 to a
concentration of 271 ± 42 (n=6) – without exhibiting saturation effects or reaching a steady state -
and declined to 55 ± 11 on day 28 after the final treatment. The efficiency of the analytical method to
extract radioactivity was 49 to 119 % (average 76%). Leucomalachite green was the predominant
residue. OD and radioactivity traces of HPLC separations showed one more polar unknown
compound in addition to the known compounds. If measurements in fry were corrected for growth,
the elimination half life was 9.7 days.

Kinetic depletion studies in fish

In addition to the studies described in the section of pharmacokinetics some other studies have been
performed. Bauer, et al. (1988) published an article “Uptake and excretion of malachite green in
rainbow trout”. They described an HPLC method for the determination of malachite green and
leucomalachite green. Recoveries were approximately 75% and all results were recovery corrected.
For the experiment they used 156 rainbow trout with an approximate body weight range of 200-300 g.
The total weight of all fish was 41.2 kg. The temperature was 9.7 ± 0.1°C. Treatment was performed
in a tank with 1000 l water to which 200 mg malachite green was added. The proportions of fish to
water corresponded to intensive aquaculture conditions for trout. Duration of treatment was 24 hours.
Samples for water analysis were taken every hour. Ten fish were sampled immediately after the end
of treatment and groups of six were sacrificed at all other sampling times. The last samples were taken
on day 143 after treatment. Homogenized tissue samples were frozen and stored at -30°C prior to
analysis. Two parallel smaller experiments were carried out in smaller tanks and with smaller fish
densities and a treatment concentration of 0.1 mg/l. for methodological studies.

The initial concentration in water of malachite green was 205 mg/m3 and decreased to 5 mg/m3 in 24
hours following an exponential term. Thus 97.6% of the malachite green disappeared and was
probably taken up by the fish because in one of the smaller experiments 80% of the malachite green
that had disappeared from the water bath was found in the fish. About 33% of the malachite green
which was taken up was found in muscle. At the end of the treatment period the total concentration of
malachite green plus leucomalachite green was 910 ± 243 µg/kg (n=10). The concentration of the
parent drug was 86.3 ± 54.4 µg/kg (n=6). On subsequent days the concentrations of the parent drug
rapidly decreased and the between fish variability increased.

A graph provided in the original paper shows that the decrease in the concentration of malachite green
did not follow a mono-exponential term. However, the group of data points describing the depletion
of the leucomalachite green comes closer to a log-linear curve. The concentrations measured in the
fatty tissue were very high. Therefore the authors determined the fat content of the muscle samples.
 21

For the muscle samples taken during the first 87 days classified according to fat content they found a
very high correlation between fat content and concentration of leucomalachite green in muscle and a
decrease of the rate of depletion of leucomalachite green in the groups of fish with the highest fat
contents. For fish with the highest fat content the elimination half life of leucomalachite green was
43.3 days. The almost complete uptake of the malachite green shows that the compartment fish was
still far from any saturation in this experiment.

Allen (1990) applied colorimetric analysis to samples of muscle, eggs and fry of malachite green
treated Atlantic Salmon (Salmo salar) and Chinook Salmon (Oncorhynchus tshawytscha). Fish had
been treated 10 to 47 times with a solution containing 1 mg/kg of malachite green oxalate for one
hour. Samples were obtained 1 to 18 days after the last treatment. Residues were extracted with a
mixture of 85% ethyl alcohol, 10% formalin and 5% acetic acid. Following extraction in the dark,
centrifugation and filtering absorbance at 615 nm was measured in the extracts. The method was not
validated. The author states that concentrations of residues in muscle of Atlantic salmon showed no
relation with the number of treatments and the concentrations in both species depended only on the
elapsed time after the last treatment. Since the methodology is inadequate for the determination of
malachite green the numerical results published by the author are most likely of little value.

A paper in Thai language (Amornchai Somletchaoen) for which only an English summary is available
reports on the persistence of malachite green in tilapia. Sixty juvenile tilapias of an average body
weight of 24.1 ± 6.8 g were exposed to malachite green at two therapeutic doses, 0.1 for 24 h and 0.2
for 1 h at a water temperature varying between 23.5 and 26.0°C. The fish were then transferred into
clean water and 3 fish were collected at 0, 6, 12, 24, 72, 120, 168 and 360 h post exposure for the
determination of malachite green and leucomalachite green residues in muscle tissues. A LC-MS-MS
method was used with LOD of 2µg/kg. Following treatment with the high therapeutic dose, highest
average concentrations of malachite green and leucomalachite green were 35.6 ± 5.8 and 32.2 ± 17.5
µg/kg, respectively. Malachite green depleted to 0.4 ± 0.15 µg/kg within 24 hours while
leucomalachite green was 1.5 ± 0.7 µg/kg after 120 hours. After treatment at the lower dose the
highest average concentration of malachite green and leucomalachite green 4.6 ± 1.8 and 30.6 ± 2.6
µg/kg, respectively. The concentration of malachite green was 2.0 ± 0.35 µg/kg at 72 h and not
detectable 168 h after treatment. Concentrations of leucomalachite green remained stable between 12
to 72 h after treatment. At 360 h after exposure, the average concentration was 3.5 ± 2.3 µg/kg.

A study investigating the metabolic profiles and residues of malachite green in trout tissues was
carried out for the United States Food and Drug Administration (Law, 1994). The study was
conducted in trout kept in tanks under the following conditions: water temperature (10±2°C), pH (6.0-
7.0), hardness (5-10 mg/l), and dissolved oxygen (9±2 mg/l). 14C-Labeled malachite green of a
radiochemical purity of 98% was used for the treatment. All experiments and analytical work was
carried out under decreased intensity room light. Concentrations in the exposure tanks were
maintained by a metering apparatus containing a 14C-MG stock solution at 800 mg/l and delivering 10
ml/min of this solution; the concentration of the treatment solution was 2 mg/kg.

Seventy-two randomly selected trout, each weighing about 350g, were divided into 3 groups of 24
fish and put into three 200-l continuous flow exposure tanks containing 2.0 mg/kg 14C-labeled MG
(actual concentrations 1.8 ± 0.2 mg/kg, 1.9 ± 0.3 mg/kg and 1.9 ± 0.2 mg/kg, respectively). A water
sample (5 ml) was withdrawn from the exposure tanks every 15 min during the 14C-labeled MG
exposure period. After a 1-h exposure, the fish were removed to a depuration tank containing flowing,
uncontaminated water. At specific time intervals during 14C-labeled MG exposure and depuration two
or three trout were removed randomly from each group of fish and sacrificed. The annexes to the
study report provided information on concentration of total radioactive residue in tissue homogenates
and ratio of malachite green to leucomalachite green concentrations in an organic extract. From these
data the concentrations of malachite green and leucomalachite green in the tissues were calculated.
The highest concentrations of residues were found in liver and kidney. Significant concentrations of
residues were also found in skin.
 22

The data of the study by Law (1994) representing the time period between the end of the treatment
and 505 hours post treatment were subjected to statistical analysis using one exponential term on the
basis of the natural logarithms of the residue contents for curve fitting. The following parameters
given in Table 3 were obtained by linear regression:

Table 3: Parameters of the linear regression analysis of kinetic residue depletion data in trout
muscle.
Parameter MG LMG
Intercept - 0.99747 - 0.01994
Slope - 0.02461 - 0.00352
Coefficient of correlation - 0.60012 - 0.73361
Residual variance 0.58312 0.52876

The kinetic data representing the concentrations of malachite green and leucomalachite green between
the end of the treatment and 505 hours were subjected to statistical analysis using one exponential
term on the basis of the natural logarithms of the residue contents. Depletion half lives were 28 hours
for malachite green and 197 hours for leucomalachite green. The kinetic parameters including the
variance of the data were used to perform estimates of dietary exposure to malachite green (see
below). The results obtained for muscle are summarised in Figures 6 and 7. The data are the same in
the two figures. In Figure 6 the time axis is given on a logarithmic scale in order to enable better
discrimination of the treatment phase and the phase after treatment.

Figure 6: Kinetics of malachite green and leucomalachite green in trout muscle.

 23

Figure 7: Kinetics in trout muscle – logarithmic time scale.

The following table of results of trials is compiled from the above mentioned publication of Sudova,
et al. (2007) that cites the data after Mitrowska and Posyniak (2005).

Table 4: A selection of treatment conditions cited after Mitrowska and Posyniak (2005).

Time Concentration in muscle

Average Duration Water
Fish pH of after tissue [μg/kg]
body of bath temperature
species bath treatment Malachite Leucomalachite
weight [hours] [°C]
[days] green green
62 2 139
4.1 80 28
Eel 24 25 6.9
100 < LOD 15
0.3 330 < LOD
14 12 518
Channel 600 21 7.1
62 < LOD 19
catfish
580 1 22 7.2 14 6 310
21 7.0 1 73 289
1350
12 7.8 5 15 230
Rainbow
40 1
trout 72 9.7 20
0.1 140
< LOD
144 13 300 2

Unsystematic small trials conducted in the context of method development studies

Some information on residue behavior was generated in a less systematic manner in the context of the
development of analytical methods. The below section summarized a selection of these studies.

Allen, et al. (1994) carried out recovery experiments in order to assess the performance of a method
and treated 6 adult rainbow trout (range of body weights was 1200-1500 g) in well water of pH 7.8
with 1 mg/l of 14C malachite green for 1h . Residues were determined by combustion analysis in
fillets (with skin left on) immediately after exposure and after 5 days withdrawal period. Both
fortified homogenates and homogenates from treated fish were extracted and analysed after cleanup
using HPLC, collection of radioactive fractions and liquid scintillation counting. The results are
summarised in Table 5.
 24

Table 5: Results of the study of Allen, et al. (1994).

Original concentration Found in

Composition of extract [%]
[mg/kg] extract
Days
Total
Material after Fortification
residue
treatment level with [mg/kg] %1 MG LMG Unknown
after
MG
combustion
Fish 1,
0.8 62 29 45 25
muscle
Fish 2,
0 1.3 1.1 85 26 49 24
muscle
Fish 3,
1.0 77 34 45 21
muscle
Fish 4, 10
0.5 3.0 46 51
muscle 0
Fish 5,
5 0.5 0.3 60 3.0 33 64
muscle
Fish 6,
0.3 60 1.8 40 59
muscle
Egg 1 85 84 7 9
homogenate 0.5 98 81 10 9
Fry
0.65 68 76 11 13
homogenate
Muscle
1 66 11 89
homogenate
1.
The values for the six incurred tissues are calculated from the data; they are not given in the
original paper.

The following results obtained with trout muscle are interesting:

ƒ Recoveries from incurred muscle tissues were lower than from fortified homogenates;
ƒ A significant fraction of the radioactive residue was of unknown structure. This fraction
increased from approximately 23% immediately after treatment to approximately 58% on day
five after treatment.
ƒ Malachite green added to homogenates was largely reduced to leucomalachite green.
ƒ The original solution of malachite green remained unchanged if processed by the same
procedure in the absence of tissue homogenate.

Andersen, et al. (2005) developed an analytical method based on liquid chromatography in which
leucomalachite green is oxidised prior to cleanup with 2,3-dichloro-5,6-dicyano-1,4-benzoquinone
(DDQ). Most of the DDQ is consumed by the sample extract. A high excess is needed; however,
excess oxidation and formation of breakdown products has to be avoided. The study included the
determination of residues in salmon that had been treated with malachite green. Fish (weight not
given) were treated in a water bath (volume not given) with 0.01 mg/l malachite green for one hour
(temperature not given). The fish were returned to clean water and sampled 2 and 4 hours after the
end of exposure. The concentrations (sum of malachite green and leucomalachite green) were found
in Table 6. The authors discuss the lower recoveries of spiked malachite green compared to
leucomalachite green as a possible explanation of the higher concentrations found at the later time
point. Mean recoveries for malachite green were approximately 71% over a concentration range of 1-
10 µg/kg. For the same concentrations of leucomalachite green recoveries averaged 95.4%.
 25

Table 6: Results of the study of Andersen, et al. (2005).

Hours after
Concentration [μg/kg]
treatment
2 29.8 35.2 36.3 35.3
4 48.0 47.6 36.9 52.5

The same group published a paper in which a similar method was used, but results were confirmed
with LC-MS (Andersen, et al., 2006). The method was validated for catfish, trout, tilapia, basa,
salmon, and shrimp. Incurred tissues were also analysed. For this purpose catfish, tilapia, trout, and
salmon were exposed to 0.01 mg/l malachite green in water for 1 h. After transfer to clean water
samples were analysed at a different time point for each species 16, 16.25, 16.5, and 24 h,
respectively, after the end of the treatment. The results are summarised below:

Table 7: Results of the study of Andersen, et al. (2006).

LC-VIS LC-MS
Fish Hours after
mean sd mean sd
species treatment
[μg/kg]
catfish 16 32.2 2.2 31.3 2.7
tilapia 16.25 1.9 0.13 2.1 0.3
trout 16.5 27.1 1.4 28.6 1.1
salmon 24 26.4 0.84 27.4 2

Bergwerf and Scherpenisse (2003) published a method for the determination of malachite green in
aquatic animals which is based on HPLC or LC-ESI-MS-MS (for confirmation). The mobile phase
was pumped through a pre-column oxidation reactor and effluent was oxidised in a post-column
reactor. The authors made the observation that the time interval between spiking of a homogenate and
further processing influenced the recovery of malachite green significantly, but had little influence on
recoveries of leucomalachite green.

Table 8: Results of the study of Bergwerf and Scherpenisse (2003)

Recovery [%]
Time [min] Malachite green Leucomalachite
green
1 81 100
15 63 96
30 60 93
60 54 93
120 45 95

Malachite green was apparently not reduced, but possibly degraded, since the chromatograms showed
satellite peaks next to malachite green. The method was optimised for leucomalachite green. Forty-
eight samples of trout, eel and salmon were collected at retail level and on fish markets.
Approximately 50% were tested positive for leucomalachite green.

Doerge, et al. (1998b) described LC methods for the simultaneous quantification of malachite green
and leucomalachite green using isotope dilution mass spectrometry. In addition they characterised
metabolites derived from malachite green and leucomalachite green found in catfish and trout. Mature
catfish of approximately 0.5 kg bw were exposed for one hour in a 40 l tank to 1 mg/kg malachite
green at 25°C and pH 7.2. The fish were briefly rinsed and transferred to fresh clean water. Fish were
killed 24 hours after dosing and fillets (skin removed) were blended and stored at -60°C prior to
 26

analysis. Leucomalachite and malachite concentrations in muscle of treated catfish were 1030 and 590
μg/kg, respectively. Trout were purchased in 1994-1995 from retail outlets in the UK. Blended tissues
were spiked with d5-leucomalachite green and 13C6-malachite green. Recoveries of the internal
standards were about 34 – 70% (n=12) for malachite green and 64-86 % (n=12) for leucomalachite
green. The concentrations of incurred residues ranged from 0.4-3.4 µg/kg for malachite green and 9-
96 µg/kg for leucomalachite green. Leucomalachite green was present at much higher concentrations
(range 12- to 38-fold).

Halme, et al. (2007) proposed an LC-ESI-MS/MS method for confirmation of residues of malachite
green and leucomalachite green in trout. D5-leucomalachite green was used as internal standard. They
analysed 34 fish monitoring samples of which eight contained malachite green residues. Only the
range of the results is given (0.35-1.34 µg/kg of leucomalachite green).

Roybal, et al. (1995) developed a method for the determination of malachite green and leucomalachite
green by SPE, HPLC, post-column oxidation and detection at 618 nm. In this context they analysed
catfish exposed to 1 mg/kg malachite green oxalate for 1h at 21°C and pH 7.0. The treated and rinsed
fish were place into separate aquaria equipped with activated carbon filters. Fish were sacrificed and
analysed at 0, 2, 4, 8, and 24 hours after placement in individual aquaria.

Table 9: Results of the study of Roybal, et al. (1995).

Concentration [µg/kg]
Hours after
Malachite green Leucomalachite green Replicates
treatment
mean s.d. mean s.d.
0 486 23.4 632 23.6 4
2 190 18.8 703 30.8 4
4 187 23.7 748 30.0 4
8 111 12.8 450 30.7 4
24 73.4 7.5 289 19.8 4

Scherpenisse and Bergwerff (2005) published a method for the determination of residues of malachite
green in finfish by LC-MS/MS. Recoveries for malachite green were very low in most fish matrices.
Recoveries for leucomalachite green were from 86 to 105%. They used the method to analyse
nineteen samples including pangasius, salmon, shrimps and trout bought in local shops. Residues
were found in three of the samples (trout 24 and 0.15 µg/kg, pangasius 7 µg/kg).

Turnipseed, et al. (2005) proposed an analytical method in which leucomalachite green is oxidised to
malachite green before the SPE extraction step of cleanup and final LC-MS determination. They used
the method to analyse two samples of treated salmon (10 μg/l for 1 hour). They found 34.6 μg/kg in a
fish 2h after treatment and 44.3 μg/kg in a fish 4 h after treatment.

ESTIMATION OF DAILY INTAKE

In the open literature, well conducted residue studies suitable to predict the concentration–time course
of residues of MG in fish are available for only two species, the rainbow trout and the channel catfish.
Only for trout were sufficient individual animal data available to perform a statistical evaluation.

Useful information on frequency of occurrence and levels of residues can primarily be obtained from
monitoring activities or from well supervised trials conducted under field conditions. The following
discussion analyses the selected data for estimation of exposure.

In the UK, approximately 400 trout samples were analyzed in three surveys between July 1993 and
March 1995. Sixty-seven samples contained malachite green at concentrations of 2-50 μg/kg. The
analytical method did not pick up leucomalachite green. In a survey of retail trout in 1996, malachite
 27

green was detected in 15 of 208 samples. In 1997 there was only one trout sample out of 137 that
contained malachite green. A change in methodology was introduced in 1997 and subsequently
malachite green and leucomalachite green were measured. When thirty-one randomly taken samples
of the 1997 survey of which 29 were negative were re-analyzed with the new method, seven became
positive. The new method was applied to the 27 samples taken in 1998. One contained both malachite
green and leucomalachite green. In five samples only leucomalachite green was found (COT, 1999).
Thus, the introduction of new methodology increased the number of positives. The individual results
of the non-compliant samples are given in the cited COT document.

The Veterinary Residues Committee established in 2001 in the UK published the results of all
statutory and non-statutory surveillance schemes. For non-compliant samples the individual numerical
values found are also given. The presentation of the data was initially such that it was not possible to
find out cases in which both residues were found in the same sample. The data are available on the
internet. When the results of the 2001-2006 plans were evaluated more than 2300 samples analyzed
for malachite green residues have been found. The main fish species covered were trout and salmon,
including imports. Occasionally the data are scheduled under “imported farmed fish”.

Another useful data set is reported from Denmark. Rasmussen (2007) reported on findings of
malachite green in fish in Denmark from 1988 to 2005. 446 plus 95 “targeted samples” were taken. 48
plus 82 “targeted samples” were positive. The author gives individual results for six samples and
ranges for the rest of the positives. Unfortunately the individual data were therefore largely not
available to the Committee.

Of 3277 samples selected from these reports, 222 samples were reported positive for malachite green
in the range from 0.2 to about 600 μg/kg fish muscle. For many of the samples it cannot be defined
what malachite green means (malachite green, leucomalachite green, or the sum of both or just a
number because the method was inadequate). Most likely the true concentrations were higher than the
results obtained and a significant fraction of the negatives were probably false negatives.

An exhaustive search all public sources of information for data on residues of malachite green is
impractical. Many fish species currently moving in international trade and commonly eaten in many
parts of the world are not covered. It is not known to what extent results of random sampling and of
biased sampling is mixed. It is not known whether any recognised sampling plan has been used in the
sampling of lots and which of the individual results given have been obtained from the same lot. The
currently published surveillance data are not transparent enough to use it for intake estimates.

The more systematically collected data of the UK from the above described activities can be used to
estimate a worst-case figure of upper limit of intake of malachite green resulting from illegal uses.
These were monitoring data (spanning from 1995 to 2006) published in the United Kingdom on the
occurrence of MG and LMG in fish muscle. If both substances were found in a sample a calculated
sum could be determined. The estimated mean level found in the positive samples was 30.7 μg/kg fish
muscle and the level at the 97.5th percentile to be 138 μg/kg. Assuming the daily consumption of fish
to be 300 g per person, the daily exposure to the sum of malachite green and leucomalachite green can
be calculated to be 9.2 and 41 μg/per person at the mean and 97.5th percentile, respectively. For a 60-
kg person, this would be equivalent to 0.15 μg/kg bw per day and 0.69 μg/kg bw per day,
respectively.

The Figure 8 below shows the frequency distribution of the levels found in the positive samples out of
the 3277 samples selected from the above mentioned programs.
 28

Figure 8: Non-representative frequency distribution of illegal residues of malachite green in

fish.

The assumption of consumption of 300g of fish contaminated with malachite green and
leucomalachite every day for a lifetime was used (a highly conservative assumption). In addition, it
was assumed that the concentrations of malachite green and leucomalachite would not change during
cooking of the fish.

The study by Law (1994) also was suitable to use for performing a dietary exposure estimate. Such an
estimate provides some information on the order of magnitude of likely human exposure to residues
of malachite green in case the drug would be authorized for treatments comparable to those performed
by the authors.

Depletion half-lives of 28 hours for malachite green and 197 h for leucomalachite green were
determined. The kinetic parameters, including the variance of the data, were used to calculate model
intakes for every day of 80 years of a human lifespan, assuming daily consumption of 300g of fish
muscle. For this purpose 29220 approximately log-normally distributed random numbers were
generated for each time point of interest ranging from the predicted value of the regression line minus
four times the residual variance to the same predicted value plus four times the residual variance.
These calculations were repeated for a number of assumed slaughter times of the fish, ranging from
1h (end of treatment) to 500 h. The results were expressed in mg malachite green/leucomalachite
green/kg of human body weight. The minima, maxima and several percentiles, including the median
of these estimated daily intakes, were calculated. The median was used for an assessment of chronic
intake. The median daily intake of leucomalachite green declined from 7.3 µg/kg bw at hour 1 to 0.87
µg/kg bw at 500 h. Results of an intake assessment for malachite green and leucomalachite green are
shown in Table 11 below.

The Committee considered that the assumption of consumption of 300 g of fish contaminated with
malachite green and leucomalachite green every day for a lifetime made these estimates highly
conservative. In addition, it was assumed that the concentrations of malachite green and
leucomalachite green would not change during cooking of the fish. However, that may not be the
case.
 29

Mitrowska, et al. (2007) investigated the stability of malachite green and leucomalachite green in
muscle of treated carp under various conditions of cooking. The initial concentrations of the residues
were approximately 200μg/kg. Leucomalachite green was much more stable than the parent
compound. Microwaving was the most effective way to partly destroy the incurred residues. The
authors published time curves of the degradation. The end results are summarised in Table 10.

Table 10: Stability of malachite green and leucomalachite green under various conditions
of cooking.
Temperature Duration % reduction
Sample Procedure
[°C] [min] MG LMG
Residues in carp Boiling, baking 15 54 0
muscle Microwave 1 61 40
Boiling water 100 0 0
10 49
Standard
150 90 97
solutions Cooking oil
120 0
210 10 97 18
Table 11: Results of an intake assessment for malachite green and leucomalachite green.

Part I – Estimated intake at various theoretical slaughter times of fish

1h 1.6h 2.4h 3.8h 5.9h 9.2h 14.3h 22.4h 34.9h 54.3h 84.7h 132.0h 205.8h 320.8h 500.0h
Intake of Malachite green [µg/kg bw per day]

min 0.2 0.2 0.2 0.2 0.2 0.2 0.1 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0
P50 1.8 1.8 1.7 1.7 1.6 1.5 1.3 1.1 0.8 0.5 0.2 0.1 0.0 0.0 0.0
P90 3.8 3.8 3.7 3.6 3.4 3.1 2.8 2.3 1.6 1.0 0.5 0.2 0.0 0.0 0.0
P95 4.7 4.7 4.5 4.4 4.2 3.8 3.4 2.8 2.0 1.3 0.6 0.2 0.0 0.0 0.0
P97.5 5.6 5.6 5.4 5.2 5.0 4.6 4.0 3.3 2.4 1.5 0.7 0.2 0.0 0.0 0.0
P99 6.9 6.9 6.6 6.5 6.1 5.7 5.0 4.1 2.9 1.8 0.9 0.3 0.0 0.0 0.0
max 15.6 16.2 21.1 15.2 14.2 14.1 11.8 9.8 10.4 3.8 2.0 0.6 0.1 0.0 0.0
Estimated intake of Leucomalachite green [µg/kg bw per day]

min 1.5 1.4 1.5 1.3 1.5 1.1 1.3 1.2 0.9 0.8 0.8 0.5 0.3 0.3 0.1
P50 7.3 7.3 7.2 7.1 6.9 6.7 6.4 6.0 5.5 4.8 4.0 3.3 2.5 1.7 0.9
P90 12.6 12.6 12.3 12.2 12.0 11.7 11.3 10.7 10.0 8.9 7.8 6.4 4.8 3.3 1.7
P95 14.7 14.7 14.5 14.6 14.1 13.8 13.3 12.6 11.7 10.7 9.3 7.7 5.8 4.0 2.1
P97.5 16.9 17.0 16.6 16.8 16.4 15.8 15.3 14.5 13.5 12.5 11.0 9.1 6.9 4.6 2.4
P99 19.7 19.5 19.6 19.7 19.1 18.9 18.3 17.3 16.0 15.0 13.4 10.9 8.3 5.6 3.0
max 48.5 42.4 36.0 39.7 36.2 34.6 38.0 38.9 31.5 32.0 30.3 22.6 16.6 14.3 6.4
Estimated intake of the sum of Malachite green and leucomalachite green [µg/kg bw per day]

min 1.7 1.6 1.7 1.4 1.7 1.3 1.4 1.3 1.0 0.8 0.8 0.5 0.3 0.3 0.1
P50 9.1 9.1 8.9 8.7 8.5 8.2 7.7 7.1 6.3 5.3 4.3 3.4 2.5 1.7 0.9
P90 16.4 16.4 16.0 15.8 15.4 14.8 14.1 12.9 11.6 9.9 8.3 6.6 4.9 3.3 1.7
P95 19.4 19.4 19.0 19.0 18.3 17.6 16.7 15.4 13.8 12.0 9.9 7.9 5.9 4.0 2.1
P97.5 22.5 22.5 22.0 22.0 21.4 20.4 19.3 17.8 15.9 14.0 11.7 9.3 7.0 4.6 2.4
P99 26.7 26.4 26.2 26.3 25.2 24.6 23.3 21.4 18.9 16.8 14.3 11.2 8.4 5.6 3.0
max 64.1 58.6 57.1 54.9 50.4 48.8 49.8 48.7 42.0 35.8 32.3 23.2 16.7 14.3 6.4
Note: For ease of reading and formatting the data, the table entries are rounded values using standard rounding techniques.
Results obtained from other surveys

Only a few examples of the type of information available from surveys can be given here in order to
facilitate the discussion of the limited usefulness of such results in the context of intake assessments.

Example 1: The Centre for Food Safety of the government of the Hong Kong special administrative
region frequently informs consumers about findings of noncompliant foods. In four separate reports
from December 2006 to November 2007, 29 positives were reported from 130 samples of varying sea
and fresh water samples collected at import and local markets at concentrations of 14 – 480 µg/kg.

Example 2: Reports on monitoring malachite green in aquatic species are available from Australia and
New Zealand (FSANZ, 2005). The 60 samples of 7 species of fish were from eight countries of
origin. The LOQ for malachite green was 2μg/kg using an LC-MS/MS method. The range of
malachite green concentrations was from 4 – 138 μg/kg.

Example 3: The Canadian total diet study (1993-2004) collected shrimp and fish samples of various
species from various sources to prepare 30 composite samples for analysis of residues of veterinary
drugs (Tittlemier, 2007). Fish were baked at 230 °C for approximately 10 minutes. Shrimp were
boiled in tap water. It is unlikely that malachite residues are stable under these conditions of sample
preparation. It would have been useful to include composite samples of raw fish. The composite
samples were frozen and stored at -20 °C until analysis in 2005. The report makes no statement on the
stability of residues over such long storage times at relatively high temperatures. The LC-MS/MS
methods used for the determination of malachite green and leucomalachite green had a limit of
detection of 0.15 μg/kg. Leucomalachite green was found in three of the composite samples
(freshwater fish 2002 and 2003, 0.95 and 0.73 μg/kg; shrimp 2002, 1.2 μg/kg).

Results obtained from residue depletion studies

Residue depletion studies are only available for the rainbow trout, the channel catfish and tilapia.
Only one of all the residue studies discussed above is suitable to predict the time course of residues of
malachite green in fish. Table 12 summarises some selected characteristics of four major kinetic and
residue studies. Despite the large differences in bath size the ratio of fish weight to bath weight is
similar; however, there are large differences in the amounts of drug available per fish. This is most
likely a critical factor in studies with prolonged exposure times. The figure is lowest for the study
with the longest exposure time.

Only one study, conducted by Law (1994), replaced the malachite green taken up from the bath by the
fish. The main argument against using the data of the Alderman and Clifton-Hadley study (1993) is
that the otherwise well designed study exhibited methodological deficits and used an analytical
method that was not valid for the purpose. The Plakas, et al. (1996) study is excellently designed and
conducted, but the individual fish residue data were not available. The Bauer, et al. study could have
been well used for an observation period similar to the exposure time in the other study; however,
individual data were also not available. Overall, only limited conclusions are possible for this study
because most of the malachite green was used up by the fish during exposure to the bath.

Table 13 highlights some possible impact on results of bath treatment studies regarding the study
designs. For the calculations in the table the physiological data for trout discussed in a previous
section were used. Three options regarding ventilation volume and breaths/min were calculated for
the Alderman and Clifton-Hadley study (1993), one alternative is given for the Bauer, et al. study
(1988). From the data presented the design of the Alderman and Clifton-Hadley study was largely
acceptable, however, it was not the case for the Bauer et al. study. This is further substantiated in
Figure 10.
Table 12: Selected characteristics of four kinetic residue studies.

weight of all fish in the

Duration of exposure
Average temperature

Estimated maximum
Initial concentration
Volume of the tank
used for treatment

of malachite green

Ratio bath weight

Initial amount of
drug available
Authors and year Species

to fish weight

[ of fish]
[mg/L]

[min]
[° C]

tank
[L]
Bauer, et al., 1988 Rainbow trout 1000 0.21 9.7 1440 41.2 24.27 4.98
16
Alderman and Clifton-Hadley, 1993 Rainbow trout 725 1.6 40 28.92 25.07 40.11
8
Plakas, et al., 1996 Channel catfish 100 0.8 21 60 3.5 28.57 22.86
Law, 1994 Rainbow trout 200 2.0 60 8.4 23.8 47.6

Table 13: Fish-physiological aspects of selected kinetic residue studies.

MG in tank [mg]

Stroke volume of
Tank volume [L]

Total amount of

Number of fish
Stroke volume

Exposure time

Total inspired
together [mL]
of breaths per
concentration

Total number

exposure [L]
water during
Breaths/min
Ventilation

Initial MG
mL/min
volume

[mg/L]

all fish
animal
[min]
[mL]

22 0.5 44 1.6 725 1160 40 1760 120 60 105.6

37 0.5 74 1.6 725 1160 40 2960 120 60 177.6
49 0.5 98 1.6 725 1160 40 3920 120 60 235.2
22 0.5 44 0.205 1000 205 1440 63360 150 75 4752
For a primitive modelling exercise it was assumed that the inspired water is completely cleared from
malachite green which means the amounts inspired with a stroke remain in the fish. On this basis the
following two graphs (Figure 10) were prepared modelling the situation of the Bauer, et al. study. The
data were generated by dissecting the whole uptake process into the number of elementary steps
dictated by the above given total number of breaths per animal.

Figure 10: Modelling of some aspects of the Bauer et al. (1988) study.

Left side: changes in the drug concentration in the bath; right side: uptake during exposure in a limited
bath volume and limited amounts of available malachite green.

The modelling on the left side of Figure 10 predicts that the bath volume is insufficient for an
exposure experiment of 1440 minutes duration involving 150 fish. The right modelling experiment
predicts that the uptake of malachite green by the fish will be limited by the amount of available drug
and therefore, could lead to the wrong interpretation that the uptake capacities of the fish were
saturated. The model predicts a final concentration of 2 mg/m3; the authors reported 5 mg/m3. Also the
predicted amounts of residues in the fish are in the same order of magnitude as experimentally
determined by the authors. Although the study has provided some remarkable results it is not
representative for the treatment of rainbow trout at 0.8 mg/l in a bath.

Similar calculations were performed using the information from the Alderman and Clifton-Hadley
study (1993) and are summarized in Figure 11 below. The graphs show that the concentration in the
water bath decreased by approximately 25% in the worst case model using the highest ventilation
volume. If one compares the predicted amounts taken up by the fish it appears that the scenario with
the lower ventilation volume fits better to the order of magnitude of initial tissue residue
concentrations given by the authors. Thus the study was well designed but suffered from a number of
weaknesses discussed in a previous section.

It was not possible to perform similar modelling with the information provided in the Plakas et al.
study (1996). However, this was also not necessary because the authors were prudent to use a fresh
bath for every set of five fish treated and they report that the concentration of malachite green in the
bath decreased by only approximately 15% during treatment.
 34

Figure 11: Modelling of some aspects of the Alderman and Clifton-Hadley study (1993).

Left side: Changes in bath concentration as function of strokes/min; Right side: Changes in drug
uptake as function of strokes/min.

METHODS OF ANALYSIS FOR RESIDUES IN TISSUES

Sample preparation

Animal tissues

Typically, residues of malachite green and leucomalachite green are extracted from (2 - 5 g)
homogenized animal (catfish, eel, rainbow trout, salmon, tropical prawn, turbot, carp, tilapia, tiger
shrimp) tissues (raw or cooked muscle samples). Amber flasks are generally used in the protocol to
avoid photo-degradation phenomena that would occur during sample preparation.

The typical protocols for malachite green and leucomalachite green extraction involve vortex mixing
or shaking in acetonitrile mixtures (extraction is generally performed over 3 to 15 minutes from 500 to
4000 rpm); the inclusion of anti-reductants and radical scavengers has been common practices. Several
acetonitrile buffer extractions protocols are reported: mixture of McIlvaine buffer (pH 3.0, 18.9 ml 0.2
M sodium hydrogen phosphate and 81.1 ml 0.1 M citric acid) and 12 ml acetonitrile (Bergwerff and
Scherpenisse, 2003; Dowling, et al., 2007); acidic (0.1% acetic acid) acetonitrile with NaCl
(Hernando, et al., 2006); 0.1M ammonium acetate pH 4.5 and acetonitrile (Andersen, et al., 2005,
2006; Tarbin et al., 2008; Hall, et al., 2008). Other mixtures such as hydroxylamine solution (25%),
0.5 ml of p-toluenesulfonic acid solution (1 M) and 5ml of acetate buffer (0.05 M, pH 4.5) are also
reported (Mitrowska, et al., 2005; Mitrowska, et al., 2007; Andersen, et al., 2008).

Purification is generally performed with SPE and /or liquid/liquid extraction with dichloromethane.
Clean up over SPE may be carried out over aromatic sulfonic acid solid-phase extraction columns
(Bergwerff and Scherpenisse, 2003; Anderssen, et al., 2008). Malachite green and leucomalachite
green are eluted with the following mixture: 2.5 ml 1.0 mg/ml methanolic ascorbic acid, 20 ml 50 mM
sodium perchlorate containing 25 mM sodium acetate and 25 mM 1-pentanesulfonic acid adjusted to
 35

pH 4.0 with acetic acid, and 27.5 ml acetonitrile (Bergwerff and Scherpenisse, 2003). Other elution
conditions have been reported on the same SPE cartridges: 90% (v/v) methanol, 5% (v/v) of 1mg/ml
ascorbic acid and 5% (v/v) of 25% (m/v) aqueous NH4OH (Scherpenisse and Bergwerff, 2005).
Purification is also reported on Strata SCX (strong cation-exchange) disposable columns with a
mixture containing acetonitrile and ammonium hydroxide (25%) (90/10) (Mitrowska, et al., 2005;
Tarbin, et al., 2008) or with citrate buffer/acetonitrile (Stubbings, et al., 2005).

Few papers report the development of Molecularly Imprinted Polymers (MIP) based SPE for selective
purification of malachite green from fish water and fish feed samples. Malachite green is used as
template, methacrylic acid (MAA) as the functional monomer and ethylene glycol dimethacrylate
(EGDMA) as the linking agent (Li, et al., 2008). Eighty percent cross reactivity with leucomalachite
green was observed. The use of malachite green as a template might however lead to a bleeding
phenomenon.

Solid-Liquid (SLE) extraction methods are also reported for the purification step using Bondesil-NH2,
40 µm particle size (Hernando, et al., 2006).

The literature also reports some protocols with an “in situ” quantitative oxidation of leucomalachite
green into MG by reaction with 2,3-dichloro-5,6-dicyano-1,4-benzoquinone after the acetonitrile
extraction step. Resulting total malachite green is then subsequently purified by solid phase extraction
with alumina and propylsulfonic acid phases (Anderssen, et al., 2005, 2006).

HPLC analysis - Screening tests

Current methods for the determination of malachite green and leucomalachite green in fish tissues or
water are based on liquid chromatography (LC), mainly with visible (VIS)/fluorescence (FLD) on-line
detections. The parent compound has λmax at 620 nm, whereas the leuco form has λmax at 265 nm,
making it difficult to determine malachite green and the leuco form using the same conditions. In
practice, the absorbance detector is set at 620 nm (Bergwerff and Scherpenisse, 2003; Mitrowska, et
al., 2005, 2007, 2008; Anderssen, et al., 2008) or 618 nm (Anderssen, et al., 2005; Stubbings, et al.,
2005) for malachite green detection while the fluorescence detector is set at λex = 265 nm and λem =
360 nm for leucomalachite green detection (Mitrowska, et al., 2005, 2007, 2008; Anderssen, et al.,
2008).

Chromatographic separation is reported on phenyl-hexyl analytical columns fitted with corresponding

guard columns; the mobile phase consisting in acetonitrile and acetate buffer (0.05 M, pH 4.5) (70:30,
v/v) in isocratic conditions (Mitrowska, et al., 2005, 2007, 2008). The use of reversed-phase analytical
chromatographic columns is also reported in this context - ODS-2 (Bergwerff and Scherpenisse,
2003), Alltima C18 (Anderssen, et al., 2005, 2008) with acetonitrile based mobile phases such as
mixture of sodium perchlorate containing pentanesulfonic acid and acetonitrile in a ratio 2:3 (v/v)
(Bergwerff and Scherpenisse, 2003) or ammonium acetate buffer/acetonitrile 50/50 (v/v) (Anderssen,
et al., 2005, 2008, Stubbings, et al., 2008).

Simultaneous LC-VIS determination of both forms is possible by post-column oxidation of

leucomalachite green to malachite green to convert the colorless leuco form into the chromophore
using cartridge containing lead(IV) oxide (PbO2) (Allen and Meinertz, 1991; Allen, et al., 1992;
Swarbrick, et al., 1997; Rushing. et al., 1995; Tarbin, et al., 1998; Bergwerff and Scherpenisse, 2003).
Post column oxidation protocols are also reported (Valle, et al., 2005). Electrochemical oxidation has
been used as an alternative to PbO2 (Rushing, et al., 1997). The determination of both compounds
together constitutes a good screening method to confirm the presence of this kind of residue, taking
into account that the combined signals will provide a gain of sensitivity. Detection limits reported for
LC-VIS measurements are around 1 µg/kg.

More recently, screening tests involving mass spectrometric detection have been reported for the
simultaneous measurement of malachite green and leucomalachite green. The sum of malachite green
 36

and leucomalachite green is determined by liquid chromatography coupled to atmospheric pressure

chemical ionisation mass spectrometry (LC-APCI-MS) after post column oxidation (Valle, et al.,
2005). Detection limit obtained on spiked salmon samples based on ion at m/z 313 is 0.15 µg/kg.
Typical recoveries are in the range 70-85%.

Screening tests involving Surface-Enhanced Raman microfluidic sensors have also been reported for
water analysis. This kind of biosensor allows fast and sensitive trace analysis of malachite green (Lee,
et al., 2007; Lucotti, et al., 2007). Malachite green molecules are adsorbed onto silver nanoparticles
while flowing along the polydimethylsulfoxane (PDMS) channel. A quantitative analysis of malachite
green is performed based on the measured peak height at 1615 cm-1 in its SERS spectrum.
Corresponding limit of detection was found around 1-2 µg/kg.

Finally, ELISA tests have also been developed for selective detection of malachite green and the
related triphenylmethane dyes in fish and fishpond water (LOD = 0.05 µg/L in water) (Yang, et al.,
2007). Performance characteristics are noted below.

LOQ: 0.49 µg/kg (malachite green, UV-VIS) (Andersen, et al., 2008)

LOD: 0.15 µg/kg (MS detection ion m/z 313 (MG+LMG)), 0.15 µg/kg (MG, UV-VIS,
(Andersen, et al., 2008); 1 µg/kg (MG + LMG, UV-VIS) (Bergwerff and Scherpenisse, 2003,
Andersen, et al., 2006); 1-2 µg/kg (SERS) (Lee, et al., 2007).
CCα: 0.15 µg/kg (MG UV-VIS), 0.13 µg/kg (LMG, FLD) (Mitrowska, et al., 2006)
CCβ: 0.37 µg/kg (MG UV-VIS), 0.32 µg/kg (LMG, FLD) (Mitrowska, et al., 2006)
Linearity: R2 = >0.995
Precision: RSD 7.7-10.9% (MG, UV-VIS) 7.7-8.4%(LMG, FLD) (Mitrowska, et al., 2006)
Accuracy: 60-64% (MG, UV-VIS), 89-92% (LMG, FLD) (Mitrowska, et al., 2006)
Recovery: 85-90% LMG (Andersen, et al., 2008; Mitrowska, et al. 2005), 60-70% MG
(Andersen, et al., 2008, Mitrowska, et al. 2005)

Confirmatory methods

For confirmatory purposes analytical procedures utilize detection by mass spectrometry (MS) with
liquid or gas chromatography, which does not demand post-column oxidation of leucomalachite green
(Turnipseed, et al., 1995a). However, either the PbO2 reactor or the “in situ” oxidations are used with
MS, because detection of malachite green is more sensitive comparing it with leucomalachite green
(Tarbin, et al., 1998; Bergwerff and Scherpenisse, 2003).

Gas chromatography coupled to mass spectrometry (GC-MS) analysis

GC-MS analyses were first developed in the mid 1990’s to provide confirmatory methods for
leucomalachite green in fish tissues (Turnipseed, et al., 1995b). Selected ion monitoring was
performed based on five diagnostic ions (m/z 330, 329, 253, 210 and 165).

Liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) analysis

Chromatographic separation is generally performed on phenyl phases using either a gradient of acidic
acetonitrile (0.1% FA)/ water or an isocratic mixture of acetonitrile/acetate buffer (70/30, v/v) as
mobile phases (Andersen, et al., 2006, 2008, Mitrowska, et al., 2008). C18 phases with 50mM
ammonium acetate/acetonitrile or acidic water/acetonitrile as eluents have also been reported
(Bergwerff and Scherpenisse, 2003; Scherpenisse and Bergwerff, 2005; Hernando, et al., 2006,
Tarbin, et al., 2008).

Atmospheric pressure chemical ionization coupled to ion trap (APCI-IT)

Atmospheric pressure chemical ionization coupled to ion trap has shown to be a very sensitive and
selective technique for the analysis of malachite green which is recovered under [M]+ charged species
 37

with a molecular ion at m/z 329 (Valle, et al., 2005). The use of ion trap as mass analyzer is reported
since it allows for full collection of product scan data, further increasing the analytical selectivity of
the method (Doerge, et al., 1998b). (MS acquisition program = MS2 scan of m/z 329, width 2 amu,
relative. collision energy 48-50%, activation Q = 0.25, activation time = 30 ms, mass range 150-350)
(Andersen, et al., 2006, 2008). The product ions include m/z 314 (M+-CH3), m/z 313 (M+-H-CH3),
m/z 285 (M+-NC2H6), m/z 251 (M+-C2H6), m/z 237 (M+-C6H5-CH3) and m/z 208 (M+-C6H5 NC2H6).
High collision energy is needed to obtain significant abundance of these ions.

LC-ESI+-QqQ

Analysis with liquid chromatography coupled to an electrospray ionization-triple quadrupole mass

spectrometer allows monitoring malachite green as [M]+ with the following transitions: 329.3>165.0,
329.3>208.0, 329.3>313.3 with associated collision energy ranging from 45 to 75 V; leucomalachite
green ([M+H]+ is monitored using the following transitions: 331>239 and 331>316 (Bergwerff and
Scherpenisse, 2003, Dowling, et al., 2007; Scherpenisse and Bergwerff, 2005; Mitrowska, et al., 2008;
Tarbin, et al., 2008).

LC-ESI+-TOF

Applications have also been reported with liquid chromatography coupled to time-of-flight mass
spectrometry with electrospray ionization (R = 9500 FWHM) for improved selectivity, especially with
regard to matrix effect (see figure 15) (Hernando et al., 2006). However, sensitivity performances of
the corresponding method are not compliant with the legislation since the minimum residue
performance level (MRPL) is exceeded.

LC-ESI+-LTQ

One application is reported with a linear ion trap as mass analyzer exhibiting similar performances as
observed with ion trap or triple quadrupole technologies (Wu, et al., 2007).

Isotopic internal standards (d5-MG and 13C6-LMG) are available and have been used to overcome
problems such as matrix suppression during electrospray ionisation (Hall, et al., 2008)

Performance characteristics of confirmatory methods

LOQ: 0.75 µg/kg (Andersen, et al., 2008)[6 ppb (MG), 3 µg/kg (LMG), TOF (Hernando, et
al., 2006)]
LOD: 0.2-0.25 µg/kg (Bergwerff and Scherpenisse, 2003, Andersen, et al., 2006, 2008) [2
µg/kg (MG), 1 µg/kg (LMG), TOF (Hernando, et al., 2006)]
CCα: 0.07-0.14 µg/kg (MG), 0.05-0.17 µg/kg (LMG) (Scherpenisse and Bergwerff, 2005;
Dowling, 2007) [8 µg/kg (MG), 38 µg/kg (LMG) TOF (Hernando, et al., 2006)][1.2 µg/kg
(MG) multiresidue (Tarbin, et al., 2008)]
CCβ: 0.15-0.23 µg/kg (MG),0.08-0.21 µg/kg (LMG) (Scherpenisse, 2005) [13 µg/kg (MG),
65 µg/kg (LMG), TOF (Hernando, et al., 2006)] [2.0 µg/kg (MG) multiresidue (Tarbin, et al.,
2008)]
Linearity: R2 > 0.995 in the range 0.5 – 10 µg/kg (Andersen, et al., 2008)
Accuracy: RSD 10%
Recovery: 85-100%

Extraction and Quantification in incurred samples

Although the ability to detect malachite green and leucomalachite green at regulated levels has been
dramatically improved by the use of LC-MS/MS and SPE clean-up procedures, the analysis of
malachite green in fish tissues remains a challenge, essentially due to issues surrounding extraction
and analyte stability. This issue has been reported recently in literature with a high accuracy method
 38

for quantification of malachite green and leucomalachite green in salmon using exact matching isotope
dilution mass spectrometry associated to longer extraction time (16h) (Hall, et al., 2008). Results
showed that whilst the total extraction and equilibrium of leucomalachite green was achieved in less
than 1h, further malachite green could still be extracted up to 16h. This highlights the difference in
chemical behaviour of the two analytes in fish matrix and the necessity for longer extraction time.
Further work could concentrate on improving the rate of release of malachite green from fish tissue
(e.g., enzymatic digestion). In particular the binding of malachite green to proteins might be an issue
in extraction efficiency.

Stability of the analytes in incurred samples

Degradation is reported as less than 10% after 12 months storage at -20°C, however, dramatic
degradation is observed for malachite green at room temperature (recoveries from 80 to 40 % in 2
hours), and little effects are observed on leucomalachite green. Malachite green and leucomalachite
green recoveries are strongly affected by freeze-thawing cycles and storages at +4°C and -20°C.

Degradation products

The metabolite leucomalachite green is not the endpoint of malachite green transformation and the
MRPL fixed at 2 µg/kg and the corresponding sum MG+LMG is an underestimate of the actual
presence of malachite green residues. Indeed, several studies have shown that malachite green and
leucomalachite green are de-methylated by systematic sequential oxidations (Culp, et al., 1999).
(Bergwerff and Scherpenisse, 2008) provide a tabular summary of the structures of residues of
malachite green identified in treated rainbow trout.

The degradations products may be formed in living fish organisms during enzymatic action but also
during photo-oxidative degradation in water (Mitrowska, et al., 2008; Bergwerff and Scherpenisse,
2008). Some identified degradation products in incurred rainbow trout or in water are shown in figure
16 [m/z 315 (N-demethyl-MG), m/z 301 (N,N-didemethyl-MG); m/z 317 (N-demethyl-LMG); m/z
303 (N,N-didemethyl-LMG)] (Mitrowska, et al., 2008; Bergrwerff and Scherpenisse, 2008). In
rainbow trout, these derivatives were found to represent about 20% of the total concentration of
malachite green residues. Since these demethyl derivatives are also expected, like malachite green and
leucomalachite green, to react with DNA, being thus potential carcinogens, the MRPL (2 µg/kg for
MG + LMG) (European Commission Decision 2004/25/EC) may therefore be subjected to future
revision.

Recent literature indicates that malachite green undergoes three main photolytic degradations under
natural sunlight irradiation: N-demethylation, hydroxylation and cleavage of the conjugated structure
forming benzophenone derivatives (Perez-Estrada, et al., 2008). More than 20 transformation products
have thus been identified. These processes involve hydroxyl radical attack on the phenyl ring, the
N,N-dimethylamine group and the central carbon atom. The Vibrio fischeri acute toxicity test showed
that the solution remains toxic after malachite green has completely disappeared. This toxicity could
be assigned, at least in part, to the formation of 4-(dimethylamine)benzophenone (D20), which is
considered ‘very toxic to aquatic organisms’ by current EU legislation.

Degradation of malachite green and leucomalachite green reported and explained by photo-oxidative
demethylation, might be prevented or at least reduced during sample preparation by addition of
ascorbic acid (Mitrowska, et al., 2005, 2008) or N,N,N’,N’-tetramethyl-1,4-phenylenediamine
dihydrochloride (TMPD) to the analytical matrix (Bergwerff and Scherpenisse, 2008).

Multi-residue method sensitivity

As multi-residue protocols are developed and more reported as a trend in scientific literature (Tarbin,
et al., 2008), performances (CCα = 1.2 µg/kg and CCβ = 2 µg/kg for malachite green in fish tissues)
they may not be compliant, in particular with the MRPL (2 µg/kg MG + LMG).
 39

Influence of processing

Effects of various cooking methods (boiling, baking, microwaving) on malachite green and
leucomalachite green have been investigated in incurred carp muscles as noted previously (Mitrowska,
et al., 2007). A decrease in concentration was observed for malachite green: - 54% after 15 min
boiling or baking; leucomalachite green was stable under these conditions. Microwaving induced a
loss of both malachite green and leucomalachite green after 1 min (- 61% and - 40%, resp.). Malachite
green also appeared to be degraded in cooking oil at 150 °C (50 % in 10 min).

REFERENCES

Alderman, D.J. (1985). Malachite green: a review. J. Fish Diseases, 8, 289-298.

Alderman, D. J., and Clifton-Hadley, R. S. (1993). Malachite green: a pharmacokinetic study in

rainbow trout, Oncorhynchus mykiss (Walbaum). J. Fish Diseases, 16, 297-311.

Allen, J.L. (1990). Residues of Malachite Green in Muscle, Eggs, and Fry of Treated Atlantic Salmon
and Chinook Salmon. Investigations in Fish Control 101. United States of the Interior, Fish and
Wildlife Service.

Allen, J.L., and Hunn, J.B. (1986). Fate and distribution studies of some drugs used in aquaculture.
Vet. Hum. Toxicol., 28 Suppl. 1, 21-24. (Only available as a PubMed abstract).

Allen, J.L., and Meinertz, J.R. (1991). Post column reaction for simultaneous analysis of chromatic
and leuco forms of malachite green and crystal violet by high performance liquid chromatography
with photometric detection. J. Chromatogr., 536, 217-222.

Allen, J.L., Meinertz, J.R., and Gofus, J.E. (1992). Determination of MG and its leuco form in
water. J. AOAC International, 75(4), 646-649.

Allen, L.J., Gofus, E., and Meinertz, J. R. (1994). Determination of malachite green residues in the
eggs, fry, and adult muscle tissue of rainbow trout (Oncorhynchus mykiss). J. AOAC International, 77,
553-557.

Amornchai Somjetlertcharoen (Undated). Persistence of malachite green in tilapia (Oreochromis

niloticus). Inland Aquatic Animal Health Research Institute, Inland Fisheries Research and
Development Bureau, Department of Fisheries. Jatuchak, Bangkok, Thailand. (Only summary in
English, Personal communication).

Andersen, W.C., Roybal, J.E., and Turnipseed, S.B. (2005). Liquid chromatographic determination
of malachite green and leucomalachite green residues in salmon with in situ LMG oxidation. J. AOAC
International, 88, 1292-1298.

Andersen, W.C., Turnipseed, S.B., and Roybal, J.E. (2006). Quantitative and confirmatory
analyses of malachite green and leucomalachite green residues in fish and shrimp. J. Agric. Food
Chem., 54, 4517-4523.

Andersen, W.C., Turnipseed, S.B., Karbiwnyk, C.M., Lee, R.H., Clark, S.B., Rowe, W.D.,
Madson, M.R., and Miller, K.E. (2008). Multiresidue method for trimethylmethane dyes in fish:
Malachite green, Crystal (Gentian) violet, brilliant green. Anal. Chim. Acta (in press).

Arpaç, E., Sayilkan, F., Asiltürk, M., Tatar, P., Kiraz, N., and Sayilkan, H. (2007). Photo catalytic
performance of Sn-doped and undoped TiO2 nanostructured thin films under UV and vis-lights. J
Hazard. Mater., 140, 69-74.
 40

Baldrian, P., and Snajdr, J. (2006). Production of ligninolytic enzymes by litter-decomposing fungi
and their ability to decolorize synthetic dyes. Enzyme and Microbial Technology, 39,1023–1029.

Başar, C.A. (2006). Applicability of the various adsorption models of three dyes adsorption onto
activated carbon prepared from waste apricot. J. Hazard. Mater., 135, 232-241.

Bauer, K., Dangschat, H., Knöppler, H.-O., and Neudegger, J. (1988). Archiv für
Lebensmittelhygiene, 39, 85-108.

Baugh, C., Grate, D., and Wilson, C. (2000). 2.8 Å crystal structure of the malachite green aptamer.
J. Mol. Biol., 301, 117-128.

Bergwerff, A.A., and Scherpenisse, P. (2003). Determination of residues of malachite green in

aquatic animals. J. Chromatogr. B., 788, 351-359.

Bergwerf, A.A., Kuiper, R.V., and Scherpenisse, P. (2004). Persistence of residues of malachite
green in juvenile eels (Anguilla anguilla). Aquaculture, 233, 55-63.

Bergwerff, A.A., and Scherpenisse, P. (2008). Determination of demethylated metabolites of

malachite green in finfish. Proceedings from the ERVI Conference,held at Egmond aan Zee, The
Netherlands, 19-21st May 2008.

Bhasikuttan, A.C., Mohanty, J., and Pal, H. (2007). Interaction of malachite green with guanine-
rich single-stranded DNA: Preferential binding to a G-quadruplex. Angew. Chem. Int. Ed Engl., 46,
9305-9307.

Bills, T.D., Marking, L., and Cahndler Jr., J.H. (1977). Malachite green: its toxicity to aquatic
organisms, persistence and removal with activated carbon. Investigations in Fish Control, 75, 1-6.

Brackett, D.M., and Dieckmann, T. (2006). Aptamer to ribozyme: the intrinsic catalytic potential of
a small RNA. Chembiochem., 7, 839-843.

Bumpus, J. A., and Brock, B. J. (1988). Biodegradation of crystal violet by the white rot fungus
Phanerochaete chrysosporium. Appl. Environ. Microbiol.,. 54, 1143–1150.

Cha, C.J., Doerge, D.R., and Cerniglia, C.E. (2001). Biotransformation of malachite green by the
Fungus Cunninghamella elegans. Appl. Environ. Microbiol. 67:4358-4360.

Chen, C., Wang, Q., Lei, P., Song, W., Ma, W., and Zhao, J. (2006). Photodegradation of dye
pollutants catalyzed by porous K3PW12O40 under visible irradiation. Environ. Sci. Technol., 40,
3965-3970.

Chen, C.C., Lu, C.S., Chung, Y.C., and Jan, J.L. (2007). UV light induced photodegradation of
malachite green on TiO2 nanoparticles. J. Hazard. Mater., 141, 520-528.

Cho, B.P., Yang, T., Blankenship, L.R., Moody, J.D., Churchwell, M., Beland, F.A., and Culp,
S.J. (2003). Synthesis and characterization of N-demethylated metabolites of malachite green and
leucomalachite green. Chem. Res. Toxicol., 16, 285-294.

European Commission (2004). 2004/25/EC: Commission Decision of 22 December 2003 amending

Decision 2002/657/EC as regards the setting of minimum required performance limits (MRPLs) for
certain residues in food of animal origin. Official Journal of the European Union L 6, 38-39.
 41

Corti, U.A. (1948). Die Matrix der Fische III. Prozentuale Fischorgangewichte. Aquatic Sciences –
Research Across Boundaries 11: 305-322.

COT (1999). Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment
(United Kingdom). Statement on surveillance for malachite green and leucomalachite green in farmed
fish. COT statement 1999/4.

Culp, S. J., Blankenship, L. R., Kusewitt, D. F., Doerge, D. R., Mulligan, L. T., and Beland, F.
A. (1999). Chem. Biol. Interact., 122, 153–170.

Davis, J. C., and Cameron, J.N. (1971). Water flow and gas exchange at the gills of rainbow trout,
Salmo Gairdneri. J. Exp. Biol., 54, 1-18.

Doerge, R., Chang, H.C., Divi, R.L., and Churchwell, M.I. (1998a). Mechanism for inhibition of
thyroid peroxidase by leucomalachite green. Chem. Res. Toxicol., 11, 1098–1104.

Doerge, D.R., Churxhwell, M.I., Genring, T.A., Pu, Y.M., and Plakas, S.M. (1998b). Analysis of
malachite green and metabolites in fish using liquid chromatography atmospheric pressure chemical
ionization mass spectrometry. Rapid Com. Mass Spectrom., 12, 1625-1634.

Dowling, G., Mulder, P.P., Duffy, C., Regan, L., and Smyth, M.R. (2007). Confirmatory Analysis
of malachite green, leucomalchite green, crystal violet and leucocrystal violet in salmon by liquid
chromatography coupled to tandem mass spectrometry. Anal. Chim. Acta, 586, 411-419.

Eichlerová, I., Homolka, L., and Nerud, F. (2006a). Synthetic dye decolorization capacity of white
rot fungus Dichomitus squalens. Bioresour Technol., 97, 2153-2159.

Eichlerová, I., Homolka, L., and Nerud, F. (2006b). Evaluation of synthetic dye decolorization
capacity in Ischnoderma resinosum. J. Ind. Microbiol. Biotechnol., 33, 759-766.

Erickson, R.J., and McKim, J.M. (1990). A simple flow limited model for the exchange of organic
chemicals at fish gills. Environ. Toxicol. Chem., 9, 159-165.

Forgacs, E., Cserhati, T., and Oros, G. (2004). Removal of synthetic dyes from wastewaters: A
review. Environment International, 30, 953– 971.

Foster, F.J., and Woodbury, L. (1936). The use of malachite green as a fish fungicide and antiseptic.
Prog. Fish. Cult., 18, 7-9. (Article only cited, but not reviewed in this monograph).

Food Standards Australia New Zealand (2005). Report on a survey of chemical residues in
domestic and imported aquacultured fish. A national survey conducted under the Coordinated Food
Survey Plan with participation by food regulatory agencies in all Australian States and Territories.

Garg, V.K., Gupta, R., Bala, Y.A., and Kumar, R. (2003). Dye removal from aqueous solution by
adsorption on treated saw dust. Bioresour. Technol., 89, 121-124.

Gobas, F.A.P.C., Opperhuizen, A., and Hutzinger, O. (1986). Bioconcentration of hydrophobic

chemicals in fish: relationship with membrane permeation. Environ. Toxicol. Chem., 5, 637-646.

Goldacre, R.J., and Phillips, J.N. (1949). The ionization of basic triphenylmethane dyes. J. Chem.
Soc., 1724-1732.

Gong, R., Jin, Y., Chen, F., Chen, J., and Liu, Z. (2006). Enhanced malachite green removal from
aqueous solution by citric acid modified rice straw. J. Hazard. Mater., 137, 865-870.
 42

Hall, Z., Hopley, C., and O’Connor, G. (2008). High accuracy determination of malachite green and
leucomalachite green in salmon tissue by exact matching isotope dilution mass spectrometry. J.
Chromatogr. B, Analyt. Technol. Biomed. Life Sci., 874(1-2), 95-100.

Halme, K., Lindfors, E., and Peltonen, K. (2007). A confirmatory analysis of malachite green
residues in rainbow trout with liquid chromatography–electrospray tandem mass spectrometry. Journal
of Chromatography B, Analyt. Technol. Biomed. Life Sci., 845(1), 74-79.

Hameed, B.H., and El-Khaiary, M.I. (2008). Kinetics and equilibrium studies of malachite green
adsorption on rice straw-derived char. J. Hazard. Mater., 153, 701-708.

Hayton, W.L., and Barron, M.G. (1990). Rate-limiting barriers to xenobiotic uptake by the gills.
Environ. Toxicol. Chem., 9, 151-157.

Henderson, A.L., Schmitt, T.C., Heinze, T.M., and Cerniglia, C.E. (1997). Reduction of Malachite
Green to Leucomalachite Green by Intestinal Bacteria. Appl. Environ. Microbiol., 63, 4099-4101.

Hernando, M.D., Mezcua, M., Suarez-Barcena, J.M., and Fernandez-Alba, A.R. (2006). Liquid
chromatography with time of flight mass spectrometry for simultaneous determination of
chemotherapeutant residues in salmon. Anal. Chim. Acta, 562, 176-184.

Jadhav, J.P., and Govindwar, S.P. (2006). Biotransformation of malachite green by Saccharomyces
cerevisiae MTCC 463. Yeast, 23, 315-323.

Janos, P., Sedivý, P., Rýznarová, M., and Grötschelová, S. (2005). Sorption of basic and acid dyes
from aqueous solutions onto oxihumolite. Chemosphere, 59, 881-886.

Kominami, H., Kumamoto, H., Kera, Y., and Ohtani, B. (2003). Photocatalytic decolorization and
mineralization of malachite green in an aqueous suspension of titanium(IV) oxide nano-particles under
aerated conditions: correlation between some physical properties and their photocatalytic activity.
Journal of Photochemistry and Photobiology A. Chemistry, 160: 99-104.

Kumar, K.V. (2006) Comparative analysis of linear and non-linear method of estimating the sorption
isotherm parameters for malachite green onto activated carbon. J. Hazard. Mater., 136, 197-202

Kumar, K.V., and Sivanesan, S. (2006). Pseudo second order kinetics and pseudo isotherms for
malachite green onto activated carbon: comparison of linear and non-linear regression methods. J.
Hazard. Mater., 136, 721-726.

Law, F.C.P. (1994). Total Residue Depletion and Metabolic Profile of Selected Drugs in Trout. U.S.
Food and Drug Administration, Contract number 223-90-7016 Study conducted at Simon Fraser
University, Burnby, British Columbia, Canada. Submitted by the United States Food and Drug
Administration.

Lee, S., Choi, J., Chen, L., Park, B., Kyong, J.B., Seong, G.H., Choo, J., Lee, Y., Shin, K.H., Lee,
E.K., Joo, S.W., and Lee, K.H. (2007). Fast and sensitive trace analysis of malachite green using
surface enhanced Raman microfluidic sensor. Anal. Chim. Acta, 8, 139-144.

LeGoff, T., and Wood, S. (2008). Production of malachite green oxalate and leucomalachite green
reference materials certified for purity. Anal. Bioanal. Chem., 391, 2035-2045.

Levin, L., Papinutti, L., and Forchiassin, F. (2004). Evaluation of Argentinean white rot fungi for
their ability to produce lignin-modifying enzymes and decolorize industrial dyes. Bioresour. Technol.,
94, 169-176.
 43

Li, Y.-H., Yang, T., Qi, X.-L., Qiao, Y., and Deng, A.-P. (2008). Development of a group selective
molecularly imprinted polymers based solid phase extraction of malachite green from fish water a,d
fish feed samples. Anal. Chim. Acta,. 624, 317-325.

Lucotti, A., Pesapane, A., and Zerbi, G. (2007). Use of a geometry optimized fiber-optic surface
enhanced Raman scattering sensor in trace detection. Appl. Spectrosc., 61(3), 260-268.

Malik, R., Ramteke, D.S., and Wate, S.R. (2007). Adsorption of malachite green on groundnut shell
waste based powdered activated carbon. Waste Manag. 27, 1129-1138.

Marking, L.L., Leith, D., and Warren, J. (1989). Demonstration of a system for removing malachite
green, Final report, Report to Bonneville Power Administration, Contract No. 1987BP36834, Project
No. 198742100, 45 pages (BPA Report DOE/BP-36834-1).

Meinertz, J.R., Stehly, G.R., Gingerich, W.H., and Allen, J.L. (2005). Residues of [14C]-malachite
green in eggs and fry of rainbow trout, Oncorhynchus mykiss (Walbaum), after treatment of eggs. J.
Fish Diseases, 18, 239–248.

Mihara, Y., Inoue, T., and Yokota, K. (2005). Relation between oxygen uptake rate and biosorption
of activated sludge against chemical substance. Yakugaku Zasshi., 125, 225-229. (Only abstract was
reviewed).

Mitrowska, K., Posyniak, A., and Zmudzki, J. (2005). Determination of malachite green and
leucomalachite green in carp muscle by liquid chromatography with visible and fluorescence
detection. J. Chromatogr. A, 1089, 187-192.

Mitrowska, K., Posyniak, A., and Zmudzki, J. (2007). The effects of cooking on residues of
malachite green and leucomalachite green in carp muscles. Anal. Chim. Acta,. 586, 420-425.

Mitrowska, K., Posyniak, A., and Zmudzki, J. (2008). Determination of malachite green and
leucomalachite green residues in water using liquid chromatography with visible fluorescence
detection and confirmation by tandem mass spectrometry. J. Chromatogr. A, 1207(1-2), 94-100.

Mittal, A. (2006). Adsorption kinetics of removal of a toxic dye, Malachite Green, from wastewater
by using hen feathers. J. Hazard. Mater., 133, 196-202.

Nguyen, D.H., DeFina, S.C., Fink, W.H., and Dieckmann, T. (2002). Binding to an RNA aptamer
changes the charge distribution and conformation of malachite green. J. Am. Chem. Soc., 124, 15081-
15084.

Nguyen, D.H., Dieckmann, T., Colvin, M.E., and Fink, W.H. (2004). Dynamics studies of a
malachite green-RNA complex revealing the origin of the red-shift and energetic contributions of
stacking interactions. J. Phys. Chem. B, 108, 1279 -1286.

Nichols, J.W., Fitzsimmons, P.N., and Whiteman, F.W. (2004). A physiologically based
toxicokinetic model for dietary uptake of hydrophobic organic Compounds by Fish. Toxicological
Sciences 77: 219-229.

Nordén B., Tjerneld, F., and Palm, E. (1978) Linear dichroism studies of binding site structures in
solution. Complexes between DNA and basic arylmethane dyes. Biophys. Chem. 8, 1-15.

Onal, Y. (2006). Kinetics of adsorption of dyes from aqueous solution using activated carbon
prepared from waste apricot. J. Hazard. Mater., 137, 1719-1728.
 44

Onal, Y., Akmil-Başar, C., Eren, D., Sarici-Ozdemir, C., and Depci, T. (2006). Adsorption kinetics
of malachite green onto activated carbon prepared from Tunçbilek lignite. J. Hazard. Mater., 128, 150-
157.

Onal, Y., Akmil-Başar, C., and Sarici-Ozdemir, C. (2007) Investigation kinetics mechanisms of
adsorption malachite green onto activated carbon. J. Hazard. Mater., 146, 194-203.

Parshetti, G., Kalme, S., Saratale, G., and Gowindwar S. (2006). Biodegradation of Malachite
Green by Kocuria rosea MTCC 1532. Acta Chim. Slov., 53, 492-498.

Perez-Estrada, L.A., Agüera, A., Hernando, M.D., Malato, S., and Fernadez-Alba, A.R. (2008).
Photodegradation of malachite green under natural sunlight irradiation: Kinetic and toxicity of the
transformation products. Chemosphere, 70, 2068-2075.

Plakas, S.M. El Said, K.R., Stehly, G.R., Gingerich, W.H., and Allen, J.L. (1996). Uptake, tissue
distribution, and metabolism of malachite green in the channel catfish (Ictalurus punctatus). Can. J.
Fish. Aquat. Sci., 53, 1427-1433.

Poe, W.E., and Wilson, R.P. (1983). Absorption of malachite green by channel catfish. Prog. Fish
Cult., 45, 228-229.

Porkodi, K., and Vasanth Kumar, K. (2007). Equilibrium, kinetics and mechanism modeling and
simulation of basic and acid dyes sorption onto jute fiber carbon: Eosin yellow, malachite green and
crystal violet single component systems. J. Hazard. Mater., 143, 311-327.

Rahman, I.A., Saad, B., Shaidan, S., and Sya Rizal, E.S. (2005). Adsorption characteristics of
malachite green on activated carbon derived from rice husks produced by chemical-thermal process.
Bioresour. Technol., 96, 1578-1583.

Rasmussen, A.R.R (2007). Restindhold af malakitgrønt i dambrugs fisk. Dansk Kemi, 88, 39-41.

Roybal, J.E., Pfenning, A.P., Munns, R.K., Holland, D.C., Hurlbut, J.A., and Long, A.R. (1995).
Determination of malachite green and its metabolite, leucomalachite green, in catfish (Ictalurus
punctatus) tissue by liquid chromatography with visible detection. J. AOAC International, 78, 453-
457.

Rushing, L.G., Webb, S.F., and Thompson Jr., H.C. (1995). Determination of leucogentian violet
and gentian violet in catfish tissue by high-performance liquid chromatography with visible detection.
J. Chromatogr. B, 674, 125-131.

Rushing, L.G., and Hansen, E.B. (1997). Confirmation of malachite green, gentian violet and their
leuco analogs in catfish and trout tissue by high-performance liquid chromatography utilizing
electrochemistry with ultraviolet-visible diode array detection and fluorescence detection. J.
Chromatogr. B, 700, 223-231.

Schmelzing, Th.O., and Claus, J. (1990). Relationship between carcass weight and organ weight of
cultured rainbow trout (Oncorhynchus mykiss). J. Appl. Ichthyology, 6, 65–72.

Schuetze, A., Heberer, T., and Juergensen, S. (2008). Occurrence of residues of the veterinary drug
malachite green in eels caught downstream from municipal sewage treatment plants. Chemosphere,
72, 1664-1670.

Scherpenisse, P., and Bergwerff, A. (2005). Determination of residues of malachite green in finfish
by liquid chromatography tandem mass spectrometry. Anal. Chim. Acta,. 529, 173-177.
 45

Sijm, D.T.H.M., Verberne, M.E., Pärt, P., and Opperhuizen, A. (1994). Experimentally
determined blood and water flow limitations for uptake of hydrophobic compounds using perfused
gills of rainbow trout (Onkorhynchus mykiss): Allomeric applications. Aquatic Toxicology, 30, 325-
341.

Singh, D.K., and Rastogi, K. (2004). Adsorptive removal of basic dyes from aqueous phase onto
activated carbon of used tea leaves: a kinetic and thermodynamic study. J. Environ. Sci. Eng., 46,
293-302.

Stubbings, G., Tarbin, J., Cooper, A., Sharman, M., Bigwood, T., and Robb, P. (2005). A multi-
residue cation exchange clean-up procedure for basic drugs in produce of animal origin. Anal. Chim.
Acta, 547, 262-268.

Sudova, E., Machova, J., Svobodova, Z., and Vesely, T. (2007). Negative effects of malachite green
and possibilities of its replacement in the treatment of fish eggs and fish: a review. Veterinarni
Medicina, 52, 527-539.

Swarbrick, A., Murby, E.J. and Hume, P. (1997). Post column electrochemical oxidation of leuco
malachite green for the analysis of rainbow trout flesh using HPLC with absorbance detection. J.
Liquid Chromatography and Related Technologies, 20, 2269-2280.

Tahir, S.S., and Rauf, N. (2006). Removal of a cationic dye from aqueous solutions by adsorption
onto bentonite clay. Chemosphere, 63, 1842-1848.

Tacal, O., and Ozer, I. (2004). Adduct-forming tendencies of cationic triarylmethane dyes with
proteins: metabolic and toxicological implications. J. Biochem. Mol. Toxicol., 18, 253-256.

Tarbin, J.A., Barnes, K.A., Bygrave, J., and Farrington, W.H. (1998). Screening and confirmation
of trimethylmethane dyes and their leuco metabolites in trout muscle using HPLC-Vis and ESP-LC-
MS. Analyst, 123, 2567-2571.

Tarbin, J.A., Chan, D. Stubbings G., and Sharman, M. (2008). Multiresidue determination of
triarylmethane and phenothiazine dyes in fish tissues by LC-MS/MS. Anal. Chim. Acta, 625, 188-194.

Tittlemier, S.A., van de Riet, J., Burns, G., Potter, R., Murphy, C., Rourke, W., Pearce, H., and
Dufresne, G. (2007) Analysis of veterinary drug residues in fish and shrimp composites collected
during the Canadian Total Diet Study, 1993–2004. Food Additives & Contaminants, 24,14-20.

Turnipseed, S.B., Roybal, J.E., Hurlbut, J.A., and Long, A.R. (1995a). GC-MS confirmation of
LMG in catfish (Ictalurus punctatus). J. AOAC International, 78: 971-977.

Turnipseed, S.B., Roybal, J.E., Rupp, H.R., Hurlbut, J.A., and Long, A.R (1995b). Particle beam
liquid chromatography-mass spectrometry of triphenylmethane dyes: application to confirmation of
malachite green in incurred catfish tissue J. Chromatogr. B, 670, 55-62.

Turnipseed, S.B., Anderson, W.C., and Roybal, J.E. (2005). Determination and confirmation of
malachite green and leucomalachite green residues in salmon using liquid chromatography/mass
spectrometry with no-discharge atmospheric pressure chemical ionisation. J. AOAC International, 88,
1312-1317.

Valle, L., Diaz, C., Zanocco, A.L., and Richter, P. (2005). Determination of the sum of MG and
LMG in salmon muscle by liquid chromatography-atmospheric pressure chemical ionisation-mass
spectrometry. J. Chromatogr. A, 1067, 101-105.
 46

Wang, S., and Ariyanto, E. (2007) Competitive adsorption of malachite green and Pb ions on natural
zeolite. J. Colloid Interface Sci., 314, 25-31.

Wu, X., Zhang, G., Wu, Y., Hou, X., and Yuan, Z. (2007). Simultaneous determination of MG
gentian violet and their leuco-metabolites in aquatic products by HPLC-linear ion trap mass
spectrometry. J. Chromatogr. A, 1172, 121-126.

Yang, M-C., Fang, J.M., Kuo, T-F., Wang, D-M., Huang, Y-L., Liu, L-Y., Chen, P-H., and
Chang, T-H. (2007). Production of antibodies for selective detection of malachite green and the
related triphenylmethane dyes in fish and fishpond water J. Agric. Food Chem., 55, 8851-8856.

Zhang, J., Li, Y., Zhang, C., and Jing, Y. (2008) Adsorption of malachite green from aqueous
solution onto carbon prepared from Arundo donax root. J. Hazard. Mater., 150, 774-782.