Synthesis of water-soluble 9,10-anthraquinone analogues with potent cyanobactericidal activity toward the musty-odor cyanobacterium Oscillatoria perornata

Article abstract of DOI:10.1021/jf072836s

A series of water-soluble 9,10-anthraquinone analogues were prepared and evaluated for their selective toxicity toward Oscillatoria perornata, which grows in catfish production ponds and causes musty off-flavor in channel catfish (Ictalurus punctatus). Water-soluble mono- and dicationic salts were prepared by conjugating various small amines directly or through a methylene or ethylene bridge to the 9,10-anthraquinone nucleus. One of the dicationic salts, 2-[N-(1′-methyl-4′-N,N-diethylaminobutyl) aminometyl]anthraquinone diphosphate, exhibited very high water solubility and potent selective toxicity toward O. perornata. However, the tendency of this compound to potentially bind to suspended sediments may be the reason for its limitations in controlling O. perornata in catfish production ponds. The monocationic salt, 2-[N-(1′-methylethyl)]aminomethyl]anthraquinone monophosphate, showed good solubility and high selective toxicity toward O. perornata. Neutral water-soluble analogues prepared by conjugating terta- or pentaethylene glycol directly or by a methylene bridge to the 9,10-anthraquinone nucleus had less activity than the parent compound.

Full text of DOI:10.1021/jf072836s

1002 J. Agric. Food Chem. 2008, 56, 1002–1007
Synthesis of Water-Soluble 9,10-Anthraquinone
Analogues with Potent Cyanobactericidal Activity
toward the Musty-Odor Cyanobacterium Oscillatoria
perornata
§
N. P. DHAMMIKA NANAYAKKARA*,† AND KEVIN K. SCHRADER
Thad Cochran National Center for Natural Products Research, School of Pharmacy, Research Institute
of Pharmaceutical Sciences, University of Mississippi, University, Mississippi 38677, and Natural
Products Utilization Research Unit, Agricultural Research Service, U.S. Department of Agriculture,
University of Mississippi, University, Mississippi 38677
A series of water-soluble 9,10-anthraquinone analogues were prepared and evaluated for their
selective toxicity toward Oscillatoria perornata, which grows in catfish production ponds and causes
“musty” off-flavor in channel catfish (Ictalurus punctatus). Water-soluble mono- and dicationic salts
were prepared by conjugating various small amines directly or through a methylene or ethylene bridge
to the 9,10-anthraquinone nucleus. One of the dicationic salts, 2-[N-(1′-methyl-4′-N,N-diethylami-
nobutyl)aminometyl]anthraquinone diphosphate, exhibited very high water solubility and potent
selective toxicity toward O. perornata. However, the tendency of this compound to potentially bind to
suspended sediments may be the reason for its limitations in controlling O. perornata in catfish
production ponds. The monocationic salt, 2-[N-(1′-methylethyl)]aminomethyl]anthraquinone mono-
phosphate, showed good solubility and high selective toxicity toward O. perornata. Neutral water-
soluble analogues prepared by conjugating terta- or pentaethylene glycol directly or by a methylene
bridge to the 9,10-anthraquinone nucleus had less activity than the parent compound.
KEYWORDS: Algicide; anthraquinone; channel catfish; Ictalurus punctatus; 2-methylisoborneol; Oscil-
latoria perornata
INTRODUCTION
Copper-based products and diuron are limited in their usefulness
in controlling cyanobacteria due to their accumulation in the
environment, lack of selectivity toward noxious cyanobacteria,
and the small margin of safety between phytotoxic concentra-
tions and ichthyotoxic concentrations (1).
The cyanobacterium (blue-green alga) Oscillatoria perornata
[Skuja] (previously designated Oscillatoria cf. chalybea) grows
in channel catfish (Ictalurus punctatus) production ponds in the
southeastern United States and produces the musty-odor com-
pound 2-methylisoborneol (MIB) (1). MIB is rapidly absorbed
into the flesh of catfish and imparts a “musty” flavor to the
catfish, rendering them unpalatable and unmarketable. In the
United States, musty “off-flavor” problems in the channel catfish
production industry can cost producers as much as $25–65
million annually, depending on the frequency of O. perornata
blooms (2). One management approach for the mitigation of
musty off-flavor in farm-raised channel catfish is the application
of chemicals to production ponds to kill or prevent the growth
of undesirable cyanobacteria. Currently, only copper-based
compounds, such as copper sulfate and chelated copper com-
pounds, and the herbicide diuron [under limited-use registration
by the U.S. Environmental Protection Agency (USEPA)] are
approved by the USEPA for use in fish production ponds.
As part of a research program to discover natural product-
based algicides for the selective control of cyanobacteria (blue-
green algae), a large number of natural products and their
analogues have been evaluated (3, 4). Among the compounds
tested so far, optimum activity and selectivity were shown by
9,10-anthraquinone and 2-methyl-9,10-anthraquinone (3). How-
ever, due to their extremely low solubility in water these two
compounds were found to be ineffective as selective algicides
to control the abundance of O. perornata in catfish production
ponds.
In an effort to increase the utility of anthraquinones as
potential selective algicides, several structural modifications to
the anthraquinone nucleus were carried out to retain the activity
and increase the solubility in water. This study describes the
synthesis of a large number of water-soluble 9,10-anthraquinone
analogues (Figure 1) and the evaluation of these analogues for
their toxicity toward O. perornata using a laboratory bioassay.
* Author to whom correspondence should be addressed [telephone
(662) 915-1019; fax (662) 915-1006; e-mail dhammika@olemiss.edu.]
† University of Mississippi.
§ U.S. Department of Agriculture.
10.1021/jf072836s CCC: $40.75
2008 American Chemical Society
Published on Web 01/19/2008

Anthraquinone Analogues with Cyanobacterial Activity
J. Agric. Food Chem., Vol. 56, No. 3, 2008 1003
using a Hewlett-Packard 8452A spectrometer. 1H NMR and 13C NMR
spectra were recorded on Varian Mercury-400BB (400 MHz for 1H
NMR and 100 MHz for 13C NMR), or Bruker Avance DRX-500 (500
MHz for 1H NMR and 125 MHz for 13C NMR) spectrometers, run in
CDCl3 with TMS as an internal standard. High-resolution MS (HRMS)
were obtained by direct probe using Bruker Bioapex-FTMS with
electro-spray ionization.
General Procedure for the Preparation of Phosphate Salts of
Alkylamino-9,10-anthraquinone Analogues. A mixture of appropriate
haloanthraquinone [2-chloromethylanthraquinone or 2-bromomethy-
lanthraquinone for compounds 1-3 and 6-16; 2-chloroanthraquinone
or 2-bromoanthraquinone for compounds 4 and 5; 2-(1′-bromoethyl)
anthraquinone for compounds 17 and 18; 1-bromomethylanthraquinone
for compound 19] (1 mmol) and an excess amount of requisite amine
(5 mL), in the absence (for compounds 4 and 5) or presence (for
compounds 1-3, 6-19) of cosolvent acetonitrile (8 mL), was stirred
at refluxing temperature for up to 1 h (for compounds 1-3, 6-19) or
at 100 °C for 2 h (for compounds 4 and 5). The cosolvent was
evaporated and the reaction mixture was partitioned between HCl (1
M, 50 mL) and dichloromethane (50 mL). The aqueous layer was then
basified (pH 10) with 20% sodium hydroxide solution and extracted
with ethyl acetate. This organic layer was washed with water, dried
over anhydrous sodium sulfate, and evaporated. The product obtained
was dissolved in methanol and then precipitated as a phosphate salt by
treating with phosphoric acid. The precipitate was filtered off and dried
under vacuum.
2-[N-(1′-Methyl-4′-N,N-diethylaminobutyl)aminomethyl]anthraqui-
none (1). Physical and spectroscopic data were reported earlier (7).
2-[N-(3′-N,N-Dimethylaminopropyl)aminomehyl]anthraquinone (2).
1H NMR δ 1.68 (2H, quintet, pentet, J ) 6.9 Hz, 2′-CH2), 2.20 (6H,
s, N(CH3)2), 2.31 (2H, t, J ) 7.0 Hz, 3′-CH2), 2.67 (2H, t, J ) 6.9 Hz,
1′-CH2) 3.93 (2H, s, 2-CH2), 7.74–7.77 (3H, m, 3,6,7-H), 8.19–8.27
(4H, m, 1,4,5,8-H); HRMS, m/e 323.1727 (M + H+, C20H22N2O2, calcd
323.1738).
2-[N-(3′-N,N-Dimethylaminopropyl]aminomehyl]anthraquinoneDiphos-
phate. Anal.: C, 46.61; H, 5.30; N, 5.27; P, 11.78%. Calcd for
C20H22N2O2 ·2H3PO4: C, 46.34; H, 5.44; N, 5.40; P, 11.95%.
2-[N-(3′-N,N-Diethylaminopropyl)aminomehyl]anthraquinone (3).
NMR δ 0.88 (6H, t, J ) 7.2 Hz, -N-(CH2CH3)2), 1.57 (2H, m, 2′-
CH2), 2.37 (4H, t, J ) 7.1, -N-(CH2CH3)2), 2.42 (2H, t, J ) 7.2 Hz,
3′-CH2), 2.57 (2H, t, J ) 6.7 Hz, 1′-CH2), 3.79 (2H, s, 2-CH2), 7.61
(3H, m, 3,6,7-H), 8.02–8.11 (4H, m, 1,4,5,8-H); HRMS, m/e 351.2038
(M + H+, C22H27N2O2, calcd 351.2067).
2-[N-(3′-N,N-Diethylaminopropyl)aminomehyl]anthraquinone Diphos-
phate. Anal.: C, 48.22; H, 5.84; N, 4.97; P, 11.45%. Calcd for
C22H27N2O2 ·2H3PO4: C, 48.36; H, 5.90; N, 5.13; P, 11.34%.
1-[N-(3′-N,N-Diethylaminopropyl)amino]anthraquinone (4). NMR
δ 0.96 (6H, t, J ) 7.1 Hz, N-(CH2CH3)2), 1.57 (2H, quintet, J ) 6.9
Hz, 2′-CH2) 2.47 (6H, q, J ) 7.1 Hz, 3′-CH2, N-(CH2CH3)2), 3.17
(2H, q, J ) 6.7 Hz, 1′-CH2), 6.84 (1H, d, J ) 8.2 Hz, 2-H), 7.31 (1H,
t, J ) 8.2 Hz, 3-H), 7.37 (1H, d, J ) 8.1 Hz, 4-H), 7.57 (2H, m, 6,
7-H), 8.07 (2H, m, 5, 8-H), 9.55 (1H, t, NH); HRMS, m/e 336.1872
(M + H+, C21H24N2O2, calcd 336.1838).
1-[N-(3′-N,N-Diethylaminopropyl)amino]anthraquinone Phosphate.
Anal.: C, 57.68; H, 6.45; N, 6.41; P, 7.07%. Calcd for C21H24N2
O2 ·H3PO4: C, 58.06; H, 6.26; N, 6.45; P, 7.13%.
1-[N-(3′-N,N-Dimethylaminopropyl)amino]anthraquinone (5). NMR
δ 1.80 (2H, quintet, J ) 7 Hz, 2′-CH2), 2.20 (6H, s, N(CH3)2), 2.35
(2H, t, J ) 7.0 Hz, 3′-CH2), 3.22 (2H, q, J ) 6.8 Hz, 1′-CH2), 6.89
(1H, d, J ) 8.2 Hz, 2-H), 7.35 (1H, t, J ) 8.1 Hz, 3-H), 7.41 (1H, d,
J ) 6.6 Hz, 4-H), 7.60 (2H, m, 6, 7-H), 8.10 (2H, m, 5, 8-H), 9.58
(1H, t, NH); HRMS, m/e 308.1541 (M + H+, C19H20N2O2, calcd
308.1525).
Figure 1. Structures of compounds 1-23.
1-[N-(3′-N,N-Diethylaminopropyl)amino]anthraquinone Phosphate.
Anal.: C, 55.98; H, 5.84; N, 6.77; P, 7.48%. Calcd for C19H20N2
O2 ·H3PO4: C, 56.16; H, 5.70; N, 6.89; P, 7.62%.
MATERIALS AND METHODS
2-[N-(1′-Methylethyl)aminomethyl]anthraquinone (6). Physical and
spectroscopic data were reported earlier (7).
General Synthesis. Melting points (uncorrected) were recorded on
an Electrothermal 9100 instrument. UV spectra were obtained in CHCl3,

1004 J. Agric. Food Chem., Vol. 56, No. 3, 2008
Nanayakkara and Schrader
2-[(Methylamino)methyl]anthraquinone (7). NMR δ 2.48 (3H, s,
N-CH3), 3.92 (2H, s, 2-CH2), 7.76–7.80 (3H, m, 3,6,7-H), 8.22–8.29
(4H, m, 1,4,5,8-H); HRMS, m/e 252.0997 (M + H+, C16H13NO2, calcd
252.1024).
2-[(Methylamino)methyl]anthraquinone Monophosphate. Anal.: C,
55.36; H, 4.67; N, 3.94; P, 9.08%. Calcd for C16H13NO2 ·H3PO4: C,
55.02; H, 4.62; N, 4.01; P, 8.87%.
2-[(Ethylamino)methyl]anthraquinone (8). NMR δ 1.12 (3H, t, J )
7.1 Hz, NCH2CH3), 2.26 (2H, q, J ) 7.1 Hz, NCH2CH3), 3.90 (2H, s,
2-CH2), 7.70–7.73 (3H, m, 3,6,7-H), 8.13–8.22 (4H, m, 1,4,5,8-H);
HRMS, m/e 266.1153 (M + H+, C17H16NO2, calcd 266.1181).
2-[(Ethylamino)methyl]anthraquinone Monophosphate. Anal.: C,
56.49; H, 5.22; N, 3.89; P, 8.24%. Calcd for C17H16NO2 ·H3PO4: C,
56.20; H, 4.99; N, 3.86; P, 8.53%.
(CH2)2), 2.41 (4H, t, J ) 7.4 Hz, 1′-(CH2)2), 3.65 (2H, s, 2-CH2),
7.69–7.72 (3H, m, 3,6,7-H), 8.15–8.22 (4H, m, 1,4,5,8-H); HRMS, m/e
350.2077 (M + H+, C23H27NO2, calcd 350.2118).
2-[N,N-(Dibutyl)aminomethyl]anthraquinone Monophosphate. Anal.:
C, 61.84; H, 6.57; N, 3.33; P, 7.11%. Calcd for C23H27NO2 ·H3PO4: C,
61.74; H, 6.76; N, 3.13; P, 6.92%.
2-[1′-(Isopropylamino)ethyl]anthraquinone (17). NMR δ 0.97, 1.02
(6H, 2×d, J ) 6.3 Hz, NHCH(CH3)2), 1.36 (3H, d, J ) 6.6 Hz, 2′-
CH3), 2.60 (1H, septet, J ) 6.2 Hz, NHCH(CH3)2), 4.06 (1H, q, J )
6.6 Hz, 1′-CH), 7.72–7.78 (3H, m, 3,6,7-H), 8.19–8.28 (4H, m, 1,4,5,8-
H); HRMS, m/e 294.1512 (M + H+, C19H20NO2, calcd 280.1494).
2-[1′-(Isopropylamino)ethyl]anthraquinone Monophosphate. Anal.:
C, 58.11; H, 5.76; N, 3.37; P, 7.71%. Calcd for C19H19N2O2 ·H3PO4:
C, 58.31; H, 5.67; N, 3.58; P, 7.91%.
2-[N-(Propyl)aminomethyl]anthraquinone (9). NMR δ 0.93 (3H, t,
J ) 7.0 Hz, NHCH2CH2CH3), 1.56 (2H, hextet, J ) 6.9 Hz,
NHCH2CH2CH3), 2.62 (2H, t, J ) 6.9 Hz, NHCH2CH2CH3), 3.95 (2H,
s, 2-CH2), 7.77–7.79 (3H, m, 3,6,7-H), 8.22–8.30 (4H, m, 1,4,5,8-H);
HRMS, m/e 280.1321 (M + H+, C18H18NO2, calcd 280.1259).
2-[N-(Propyl)aminomethyl]anthraquinone Monophosphate. Anal.: C,
57.19; H, 5.45; N, 3.82; P, 7.98%. Calcd for C18H18NO2 ·H3PO4: C,
57.30; H, 5.36; N, 3.71; P, 8.21%.
2-[N-(1′,1′-Dimethylethyl)aminomethyl]anthraquinone (10). NMR δ
1.19 (9H, s, NC(CH3)3) 3.90 (2H, s, 2-CH2), 7.77–7.82 (3H, m, 3,6,7-
H), 8.23–8.31 (4H, m, 1,4,5,8-H); HRMS, m/e 294.1483 (M + H+,
C19H20NO2, calcd 294.1494).
2-[N-(1′,1′-Dimethylethyl)aminomethyl]anthraquinone Monophos-
phate. Anal.: C, 58.11; H, 5.83; N, 3.52; P, 7.69%. Calcd for
C19H19NO2 ·H3PO4: C, 58.31; H, 5.67; N, 3.58; P, 7.91%.
2-[N-(Cyclopropyl)aminomethyl]anthraquinone (11). NMR δ 0.43
(4H, m, -CH2-CH2-), 2.17 (1H, m, NHCH), 4.01 (2H, s, 2-CH2),
7.75–7.81 (3H, m, 3,6,7-H), 8.23–8.32 (4H, m, 1,4,5,8-H); HRMS, m/e
278.1155 (M + H+, C18H16NO2, calcd 278.1181).
2-[N-(Cyclopropyl)aminomethyl]anthraquinone Monophosphate.
Anal.: C, 57.74; H, 5.09; N, 3.51; P, 8.21%. Calcd for C18H16NO2 ·
H3PO4: C, 57.60; H, 4.83; N, 3.73; P, 8.25%.
2-[N-(Pentyl)aminomethyl]anthraquinone (12). NMR δ 0.78 (3H,
t, J ) 6.7 Hz, 5′-CH3), 1.19 (4H, m, 3′- and 4′-CH2), 1.44 (2H, m,
2′-CH2), 2.53 (2H, t, J ) 7.4 Hz, 1′-CH2), 3.82 (2H, s, 2-CH2),
7.63–7.68 (3H, m, 3,6,7-H), 8.06–8.14 (4H, m, 1,4,5,8-H); HRMS, m/e
308.1619 (M + H+, C20H22NO2, calcd 308.1645).
2-[N-(Pentyl)aminomethyl]anthraquinone Monophosphate. Anal.: C,
59.53; H, 6.25; N, 3.66; P, 7.46%. Calcd for C20H22NO2 ·H3PO4: C,
59.26; H, 5.97; N, 3.46; P, 7.64%.
2-[N-(1′-Methylbutyl)aminomethyl]anthraquinone (13). NMR δ 0.89
(3H, t, J ) 7.1 Hz, 4′-CH3) 1.08 (3H, d, J ) 6.2 Hz, 1′-CH3), 1.33
(2H, m, 3′-CH2), 1.46 (2H, m, 2′-CH2), 2.68 (1H, J ) 6 Hz, 1′-NHCH),
3.89, 3.97 (2H, 2×AB doublets, J ) 14.2 Hz, 2-CH2), 7.74–7.77 (3H,
m, 3,6,7-H), 8.20–8.28 (4H, m, 1,4,5,8-H); HRMS, m/e 308.1619 (M
+ H+, C20H22NO2, calcd 308.1645).
2-[N-(1′-Methylbutyl)aminomethyl]anthraquinone Monophosphate.
Anal.: C, 59.11; H, 5.86; N, 3.37; P 7.51%. Calcd for C20H22NO2 ·
H3PO4: C, 59.26; H, 5.97; N, 3.46; P, 7.64%.
2-[1′-{3′′-(N,N-Diethylamino)propylamino}ethyl]anthraquinone (18).
NMR δ 0.94 (6H, t, J ) 7.1 Hz, -N(CH2CH3)2), 1.35 (3H, d, J ) 6.6
Hz, 2′-CH3) 1.58 (2H, quintet, J ) 6.9 Hz, 2′′-CH2), 2.34–2.56 (8H,
m, 1′′, 3′′-CH2, -N(CH2CH3)2), 3.89 (1H, q, J ) 6.5 Hz, 1′-CH),
7.71–7.77 (3H, m, 3,6,7-H), 8.17–8.24 (4H, m, 1,4,5,8-H); HRMS, m/e
364.2168 (M + H+, C23H28N2O2, calcd 364.2151).
2-[1′-{3′′-(N,N-Diethylamino)propylamino}ethyl]anthraquinone Diphos-
phate. Anal.: C, 49.16; H, 5.87; N, 4.87; P, 11.36%. Calcd for
C23H28N2O2 ·2H3PO4: C, 49.29; H, 6.11; N, 5.00; P, 11.05%.
1-(N,N-Diethylaminomethyl)anthraquinone (19). NMR δ 1.08 (6H,
t, J ) 7.0 Hz, N(CH2CH3), 2.61 (4H, q, J ) 7.0 Hz, N(CH2CH3), 4.23
(2H, s, 1-CH2), 7.72–7.76 (4H, m, 2,3,6,7-H), 8.25–8.38 (3H, m, 4,5,8-
H); HRMS, m/e 294.1476 (M + H+, C19H20NO2, calcd 294.1494).
1-(N,N-Diethylaminomethyl)anthraquinone Monophosphate. Anal.:
C, 58.54; H, 5.82; N, 3.60; P, 7.76%. Calcd for C19H19N2O2 ·H3PO4:
C, 58.31; H, 5.67; N, 3.58; P 7.91%.
Preparation of 2-Ethylaminoanthraquinone Analogues. 2-(Hy-
droxyethyl)anthraquinone. A mixture of 2-methylanthraquinone (2.2
g, 10 mmol), paraformaldehyde (150 mg, 5 mmol), and KOH (50 mg)
in DMSO (25 mL) was heated at 110–120 °C under nitrogen for 96 h.
Three more portions of paraformaldehyde (150 mg, 5 mmol each) were
added at 24 h intervals. The reaction mixture was poured into water
and then filtered. The precipitate obtained was chromatographed over
silica gel and elution with dichloromethane:hexane (1:1) yielded the
unreacted 2-methylanthraquinone. Further elution with dichloromethane/
methanol (99:1) yielded 2-(hydroxyethyl)anthraquinone (700 mg): NMR
δ 3.05 (2H, t, J ) 7.2 Hz, CH2CH2OH), 3.99 (2H, dt, J ) 7.2, 1.2 Hz,
CH2CH2OH), 7.66 (1H, br d, J ) 8.0 Hz, H-3), 7.77–7.80 (2H, m,
6,7-H), 8.14 (1H, br s, 1-H), 8.21 (1H, d, J ) 8.0 Hz, 4-H), 8.26–8.30
(2H, m, 5,8-H); HRMS, m/e 253.0865 (M + H+, C16H13O3, calcd
253.0855).
Synthesis of 2-(Bromoethyl)anthraquinone. A mixture of 2-(hy-
droxyethyl)anthraquinone (400 mg, 1.6 mmol), triphenylphosphine (620
mg, 2.4 mmol), and carbon tetrabromide (660 mg, 2 mmol) in
dichloromethane (15 mL) was stirred at 0 °C. After 1 h, solvent was
evaporated and the residue was chromatographed over silica gel. Elution
with ethyl acetate/hexanes (5:95) yielded 2-(bromoethyl)anthraquinone,
which was crystallized from chloroform/hexanes to yield pale yellow
crystals (310 mg): NMR δ 3.34 (2H, t, J ) 7.2 Hz, CH2CH2Br), 3.65
(2H, t, J ) 7.2 Hz, CH2CH2Br), 7.64 (1H, dd, J ) 8.0, 1.6 Hz, H-3),
7.75–7.80 (2H, m, 6,7-H), 8.14 (1H, d, J ) 1.6 Hz, 1-H), 8.26 (1H, d,
J)6.4 Hz, 4-H), 8.25–8.31 (2H, m, 5,8-H); HRMS, m/e 315.0023 (M
+ H+, C16H12BrO2, calcd 315.0027).
2-[N,N-(Diisopropyl)aminomethyl]anthraquinone (14). NMR δ 1.03
(12H, d, J ) 6.5 Hz, N-CH(CH3)2) 3.03 (2H, septet, J ) 6.5 Hz,
N-CH(CH3)2), 3.78 (2H, s, 2-CH2), 7.75–7.88 (3H, m, 3,6,7-H),
8.21-8.31 (4H, m, 1,4,5,8-H); HRMS, m/e 322.1767 (M + H+,
C21H24NO2, calcd 322.1802).
2-(Isopropylaminoethyl)anthraquinone (20). A mixture of 1-bromo-
ethylanthraquinone (300 mg) and isopropylamine (1 mL) in acetonitrile
(10 mL) was refluxed for 6 h. The reaction mixture was evaporated
under vacuum and the residue was sonicated with chloroform and then
filtered to yield 2-[ethylamino-N-(1′-methylethyl)]-9,10-anthraquinone
hydrobromide as an off-white crystalline compound. This compound
was partitioned between 0.5 M sodium hydroxide and ethyl acetate,
and the organic layer was dried and evaporated to yield 20: NMR δ
1.03 (6H, d, J ) 6.4 Hz, NCHCH3), 2.81 (1H, septet, J ) 6.4 Hz),
2.93 (4H, s, CH2CH2N), 7.60 (1H, dd, J ) 8.0, 1.6 Hz, H-3), 7.73–7.80
(2H, m, 6,7-H), 8.10 (1H, d, J ) 1.6 Hz, 1-H), 8.19 (1H, d, J ) 6.4
Hz, 4-H), 8.20–8.28 (2H, m, 5,8-H); HRMS, m/e 294.1511 (M + H+,
C19H20NO2, calcd 294. 1494).
2-[N,N-(Diisopropyl)aminomethyl]anthraquinone Monophosphate.
Anal.: C, 60.22; H, 6.20; N, 3.39; P, 7.51%. Calcd for C21H24NO2 ·
H3PO4: C, 60.14; H, 6.25; N, 3.34; P, 7.39%.
2-[N,N-(Diethyl)aminomethyl]anthraquinone (15). NMR δ 1.03 (6H,
t, J ) 7.1 Hz, -N(CH2CH3)2), 2.52 (4H, t, J ) 7.1 Hz, -N(CH2CH3)2)
3.67 (2H, s, 2-CH3) 7.71–7.80 (3H, m, 3,6,7-H), 8.17–8.24 (4H, m,
1,4,5,8-H); HRMS, m/e 294.1473 (M + H+, C19H20NO2, calcd
294.1488).
2-[N,N-(Diethyl)aminomethyl]anthraquinone Monophosphate. Anal.:
C, 58.12; H, 5.69; N, 3.71; P, 7.72%. Calcd for C19H20NO2 ·H3PO4: C,
58.31; H, 5.67; N, 3.58; P, 7.91%.
2-[N,N-(Dibutyl)aminomethyl]anthraquinone (16). NMR δ 0.83 (6H,
t, J ) 7.2 Hz, 4′-(CH3)2) 1.27 (4H, m, 3′-(CH2)2), 1.43 (4H, m, 2′-

Anthraquinone Analogues with Cyanobacterial Activity
J. Agric. Food Chem., Vol. 56, No. 3, 2008 1005
2-(Isopropylaminoethyl)anthraquinone Monophosphate. The residue
obtained was dissolved in methanol (10 mL), treated with H3PO4 (1
mL), diluted with ether (10 mL), and left overnight at 4 °C. The
crystalline product was filtered off and dried to yield an off-white solid.
Anal.: C, 58.07; H, 5.85; N, 3.38; P, 7.86%. Calcd for C19H19NO2 ·
H3PO4: C, 58.31; H, 5.67; N, 3.58; P, 7.91%.
Table 1. Bioassay Results of Anthraquinone Analogues
test organism
O. perornata
LOECa (nM) LCICb (nM)
S. capricornutum
compd
LOEC (nM)
LCIC (nM)
1
2
3
4
5
6
7
8
10
100
100
100
1000
100
10000
1000
100
1000
1000
1000
1000
1000
1000
100
10000
10000
10000
100000
100000
10000
10000
10000
10000
1000
10000
100000
100000
100000
10000
100000
100000
100000
10000
10000
10000
10000
100000
100000
100000
10000
10000
10000
10000
100000
100000
100000
1000000
100000
1000000
100000
100000
10000
100000
>100000
>100000
>100000
General Procedure for the Preparation of 1- and 2-Pentaethyl-
eneglyco-9,10-anthraquinone. A mixture of 1-chloroanthraquinone
(2.5 g) or 2-chloroanthraquinone and potassium carbonate (2.5 g) in
pentaethyleneglycole (15 mL) was heated for 3 h at 140 °C, poured
into cool water, and filtered. The resulting solid was chromatographed
over silica gel and elution with dichloromethane:methanol (96:4) to
yield1-or2-[2-[2-[2-[2-(2-hydroxyethoxy)ethoxy]ethoxy]ethoxy]ethoxy]
anthraquinone. This compound was crystallized from ether to yield
yellow crystalline compounds.
1-[2-[2-[2-[2-(2-Hydroxyethoxy)ethoxy]ethoxy]ethoxy]ethoxy]anthra-
quinone (21). NMR δ 3.44 (2H, m, 10′-CH2), 3.50–3.58 (12H, m,
4′,5′,6′,7′,8′,9′-CH2), 3.73 (2H, m, 3′-CH2), 3.88 (2H, t, J ) 4.9 Hz,
2′-CH2), 4.16 (2H, t, J ) 4.7 Hz, 1′-CH2), 7.20 (1H, br d, J ) 8.4 Hz,
2-H), 7.51 (1H, br t, J ) 8.0 Hz, 3-H), 7.58 (2H, m, 6,7-H), 7.73 (1H,
dd, J ) 7.7, 0.8 Hz, 4-H), 8.01 (1H, dd, J ) 7.6, 1.4 Hz, 8-H) 8.04
(1H, dd, J ) 7.8, 1.5 Hz, 5-H); HRMS, m/e 445.1867 (M + H+,
C24H29O8, calcd 445.1862).
2-[2-[2-[2-[2-(2-Hydroxyethoxy)ethoxy]ethoxy]ethoxy]ethoxy]anthra-
quinone (22). NMR δ 3.21 (3H, s, -OCH3), 3.39 (2H, m, 8′-CH2),
3.46–3.62 (10H, m, 3′,4′,5′,6′,7′-CH2), 3.77 (2H, t, J ) 4.6 Hz, 2′-
CH2), 4.11 (2H, t, J ) 4.6 Hz, 1′-CH2), 7.04 (1H, dd, J ) 8.5, 2.6 Hz,
3-H), 7.42 (1H, d, J ) 2.5 Hz, 1-H), 7.56 (2H, m, 6,7-H), 7.95 (1H, d,
J ) 8.6 Hz, 4-H), 8.01 (2H, m, 5,8-H); HRMS, m/e 445.1856 (M +
H+, C24H29O8, calcd 445.1862).
10000
1000
10
1000
1000
1000
1000
1000
1000
100
1000
100
1000
1000
10000
100
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
1000
100
1000
1000
10000
1000
1000
10000
100000
100000
1000
1000
10000
100
10000
>100000
>100000
>100000
a Lowest-observed-effect concentration (lowest concentration to inhibit growth).
b Lowest-complete-inhibition concentration.
2-Methylene-[2-[2-[2-(2-hydroxyethoxy)ethoxy]ethoxy]ethoxy]anthra-
quinone (23). A mixture of 2-chloromethylanthraquinone (2.5 g) and
barium hydroxide (5 g) in pentaethyleneglycol (15 mL) was sonicated
for 10 min and then stirred for 3 h at room temperature. The reaction
mixture was poured into cool water and filtered. The solid obtained
was chromatographed over silica gel and then eluted with dichlo-
romethane/ethanol (96:4) to yield 2-methylene-[2-[2-[2-(2-hydroxy-
ethoxy)ethoxy]ethoxy]ethoxy]anthraquinone. This compound was crys-
tallized from ether to yield an off-white crystalline compound: NMR
δ 3.60 (2H, m, 6′-CH2), 3.67–3.72 (14H, m, 1′,2′,3′,4′,5′,6′,7′-CH2),
4.73 (2H, s, 2-CH2), 7.75–7.85 (3H, m, 3,6,7-H), 8.24–8.32 (4H, m,
1,4,5,8-H); HRMS,m/e 415.1768 (M + H+, C23H27O7, calcd. 415.1756).
Bioassay of Water-Soluble Anthraquinone Analogues. An isolate
of the cyanobacterium O. perornata was obtained from a water sample
collected from a Mississippi catfish pond (5). An isolate of the green
alga S. capricornutum was obtained from Dr. J. C. Greene, USEPA,
Corvallis, OR, and S. capricornutum was used as a representative of
green algae in the bioassay to determine selective toxicity of the
compounds. Green algae (Division Chlorophyta) are the preferred type
of phytoplankton in catfish aquaculture ponds due to their lack of
production of earthy and musty off-flavor compounds and because they
are better oxygenators of the water compared to cyanobacteria (6). Each
culture was maintained separately in continuous, steady-state growth
using the conditions outlined in Schrader et al. (7) to provide a source
of cells growing at a fairly constant rate.
The rapid bioassay of Schrader et al. (7) was used to evaluate the
water-soluble anthraquinone analogues for selective toxicity toward O.
perornata. Stock solutions of each pure compound were made in
deionized water at concentrations of 4.0, 40.0, 400.0, 4000.0, 40000.0,
400000.0, and 4000000.0 nM. Stock solutions of water-soluble an-
thraquinones were micropipetted into wells (50 µL of anthraquinone
solution per well) in a 96-well microplate (Costar Corp., Cambridge,
MA) containing 150 µL of either cyanobacterial or unialgal culture
material from continuous cultures. Deionized water was added to control
wells (50 µL per well). Final test concentrations were 1.0, 10.0, 100.0,
1000.0, 10000.0, 100000.0, and 1000000.0 nM. Three replications were
used for each concentration of each pure compound and for the control;
bioassay experiments were repeated. Microplates were placed in a
growth chamber held at 29 ( 1 °C and were illuminated continuously
by fluorescent lights (40 W, cool white) at a photon flux density of
21-27 µEinstein/m2/s. Absorbance measurements of each well were
measured at 650 nm at 24 h intervals for 4 days using a Packard model
SpectraCount microplate photometer (Packard Instrument Co., Meriden,
CT). Mean values and standard deviations of absorbance measurements
were calculated and graphed to determine the lowest-observed-effect
concentration (LOEC) and lowest-complete-inhibition concentration
(LCIC). The 96 h IC50 (50% inhibition concentration) values of the
most promising compounds (8) were compared to the 96-h IC50 of
anthraquinone, with determination of 95% confidence limits for each
value.
RESULTS AND DISCUSSION
In previous studies, a few water-soluble anthraquinone
analogues were evaluated for their algicidal activity (3).
Anthraquinone-2-sulfonic acid, the anionic (sodium) salt of
anthraquinone-2-sulfonic acid and anthraquinone-1,5-disulfonic
acid disodium salt have high water solubility. These compounds
had significantly low algicidal activity compared to 9,10-
anthraquinone. Bioassay results indicated that water-soluble
analogue 1 is highly effective and selective toward O. perornata
(8). The side chain of this compound is the same side chain in
the antimalarial drug chloroquine and contains an asymmetric
center. These results guided us to prepare and evaluate water-
soluble cationic salt analogues of 9,10-anthraquinone. The
condensation of 1-chloro- or 2-chloromethylanthraquinone with
3-(diethylamino)propylamine, 3-(dimethylamino)propylamine,
or 2-amino-5-(diethylamino)pentane yielded a series of diami-
noanthraquinone analogues (2–5). The treatment of a methanolic
solution of these amines with phosphoric acid (85%) yielded
highly water-soluble dicationic anthraquinone analogues. Their
algicidal activities are presented in Table 1. Compounds 2 and
3 also had high water solubility, however, less activity than 1.
Compounds 4 and 5, in which the side chain is directly attached
to the anthraquinone ring, were less active. Because of the
reported mutagenic activity of 2-aminoanthraquinone, its ana-
logues were not prepared.
Previous pond studies conducted with compound 1 at an
application rate of 2.0 µM (1.15 mg/L) found levels of MIB

1006 J. Agric. Food Chem., Vol. 56, No. 3, 2008
Nanayakkara and Schrader
Table 2. 96-h IC50 Values of Anthraquinone and Compounds 1 and 6
of compounds 17 and 18 are provided in Table 1. The
introduction of a methyl group to the “2-CH2” group was found
to reduce the algicidal activity.
a
IC50 (nM) for test organism
compd
O. perornata
S. capricornutum
To investigate the effect of the position of methyl amino
group on the algicidal activity, 1-aminomethylanthraquinone
analogues were prepared. Bromination of 1-methylanthraquinone
with N-bromosuccinimide in the presence of azobisisobutyroni-
trile yielded 1-bromomethylanthraquinone in good quantities.
Condensation of 1-bromomethylanthraquinone with secondary
amines such as isopropylamine or 2-amino-5-diethylaminopen-
tane yielded a deep-blue compound, which tentatively was
identified as the product resulting from the condensation of the
secondary amino group with the adjacent carbonyl. The reaction
of 1-chloromethylanthraquinone with diethylamine yielded the
expected compound 19. This compound had moderate algicidal
activity toward O. perornata.
To investigate the effect of the length of the carbon chain on
algicidal activity, the 2-ethylaminoanthraquinone analogue was
prepared and evaluated. Condensation of 2-methylanthraquinone
with paraformaldehyde in the presence of base yielded 2-etha-
nolanthraquinone. Bromination of 2-ethanolanthraquinone and
subsequent condensation with isopropyl amine yielded 2-[ethyla-
mino-N-(1′-methylethyl)]-9,10-anthraquinone (20). This com-
pound also had only moderate algicidal activity toward O.
perornata.
A series of neutral water-soluble anthraquinone analogues
were prepared by attaching polyethylene glycol of different
ethylenedioxy chain lengths. These compounds were prepared
by reacting either 1-chloroanthraquinone, 2-chloroanthraquinone,
or 2-chloromethylanthraquinone with the respective polyethylene
glycol. The water solubility of the analogues increased with the
length of the ethylenedioxy chain, and those chains with four
or more ethylenedioxy units (e.g., 21, 22, and 23) were found
to be substantially water-soluble. Algicidal activities of these
compounds are provided in Table 1. Although some of these
compounds had low LOEC values, the LCIC values were
relatively high, indicating that they had only marginal algicidal
activity toward O. perornata.
AQb
1
79 (4.8)
63.0 (6.1)
6.3 (1.2)
5012.0 (144.4)
5012.0 (144.4)
5623.0 (842.4)
6
a IC50 ) 96-h 50% inhibition concentration; 95% confidence limits are in
parentheses. b AQ ) anthraquinone; unpublished observations.
were reduced by 93.6% after 18 h and remained low for 8 days
(8). However, subsequent pond studies at lower concentrations
(e0.58 mg/L) found that compound 1 was ineffective (8). The
efficacy of this compound may have been dependent on the
amount of suspended sediments in the test ponds. This pos-
sibility was inferred on the basis of the dicationic characteristics
of compound 1. Compound 1 may have a tendency to bind to
suspended sediments and organic matter in the water column
and settle to the bottom of the pond, thereby resulting in the
loss of activity toward noxious species of cyanobacteria. Similar
effects have been observed with other dicationic aquatic
herbicides such as diquat (9).
To overcome this potential drawback, a series of monoami-
noanthraquinone analogues 6–16 were prepared and evaluated
for selective algicidal activity. These analogues were readily
prepared by reacting 2-chloromethylanthraquinone with the
appropriate amine. The results of the evaluation of these
analogues are provided in Table 1. The monoamino analogues
were less water soluble than the diamino analogues, but their
solubilities were higher than the LCIC observed for them. These
analogues had similar activities except for compounds 6, 13,
and 15, which all had greater toxicity. Of these analogues,
compound 6 had the best activity. The LOEC and LCIC
observed for this compound were similar to those observed
for compound 1. However, determination of the IC50 against
O. perornata indicated that compound 6 was 10 times more
active than 1 and >10 times more active than anthraquinone
(Table 2).
Three efficacy studies performed with compound 6 using
limnocorrals placed in catfish ponds indicated that it was very
effective against O. perornata (8). At rates of 0.3 µM (125 µg/
L), 6 dramatically reduced the abundance of O. perornata and
levels of MIB 2 days after application. During this time, the
abundance of nontarget organisms such as green algae and
diatoms increased dramatically. These studies showed that
compound 6 was also effective against Anabaena circinalis and
that geosmin (a compound responsible for an earthy off-flavor
in catfish) levels were reduced (8). Previous studies have
identified a strain of A. circinalis that was isolated from a river
providing municipal drinking water to be a producer of geosmin
(10). However, there has not yet been verification of geosmin
production by an isolate of A. circinalis from a catfish pond.
Analysis of the pond water and suspended sediments indicated
that compound 6 has very low tendency to bind to sediments.
The half-life of 6 in the water column of ponds was 19 h, making
it much less persistent than diuron (8). Compound 6 appears to
undergo oxidative deamination to yield anthraquinone-2-car-
boxylic acid, which has low algicidal activity toward O.
perornata (unpublished observations). Replacement of one of
the protons adjacent to the primary amine group is known to
retard oxidative deamination. In an effort to decrease the rate
of the conversion, analogues with a methyl group on the
2-methylene group were prepared and evaluated. These com-
pounds were readily prepared by condensing 2-(1′-bromoethyl)
anthraquinone with the corresponding amine. Algicidal activity
Our results indicate that certain water-soluble anthraquinone
analogues that were synthesized during this study have high
potential for use as selective algicides toward undesirable species
of cyanobacteria found in catfish aquaculture ponds. Compound
6 appears to be the most promising among this group of patented
quinones (11). Additional studies are needed to further evaluate
the most promising compounds to determine their environmental
fate in catfish ponds, toxicity toward nontarget organisms, and
potential accumulation in the flesh of channel catfish.
ACKNOWLEDGMENT
We sincerely thank Dr. Chuck Dunbar for recording mass
spectra and Ramona Pace and Dewayne Harries, USDA-ARS,
for technical assistance in this research.
LITERATURE CITED
(1) Boyd, C. E.; Tucker, C. S. Pond Aquaculture Water Quality
Management; Kluwer: Norwell, MA, 1998.
(2) Tucker, C. S. Off-flavor problems in aquaculture. ReV. Fish. Sci.
2000, 8, 45–88.
(3) Schrader, K. K.; de Regt, M. Q.; Tidwell, P. R; Tucker, C. S.;
Duke, S. O. Selective growth inhibition of the musty-odor
producing cyanobacterium Oscillatoria cf. chalybea by natural
compounds. Bull. EnViron. Contam. Toxicol. 1998, 60, 651–658.
(4) Schrader, K. K.; Harries, M. D. Compounds with selective toxicity
toward the musty-odor cyanobacterium Oscillatoria perornata.
Bull. EnViron. Contam. Toxicol. 2001, 66, 801–807.

Anthraquinone Analogues with Cyanobacterial Activity
J. Agric. Food Chem., Vol. 56, No. 3, 2008 1007
(5) van der Ploeg, M.; Dennis, M. E.; de Regt, M. Q. Biology of
Oscillatoria cf. chalybea, a 2-methylisoborneol producing blue-
green alga of MS catfish ponds. Water Sci. Technol. 1995, 31,
173–180.
(6) Paerl, H. W.; Tucker, C. S. Ecology of blue-green algae in
aquaculture ponds. J. World Aquacult. Soc. 1995, 26, 109–131.
(7) Schrader, K. K.; de Regt, M. Q.; Tucker, C. S.; Duke, S. O. A
rapid bioassay for selective algicides. Weed Technol. 1997, 11,
767–774.
Oscillatoria perornata, in catfish aquaculture ponds. J. Appl.
Aquacult. 2003, 14, 149–154.
(10) Rosen, B. H.; MacLeod, B. W.; Simpson, M. R. Accumulation
and release of geosmin during the growth phases of Anabaena
circinalis (Kütz.) Rabenhorst. Water Sci. Technol. 1992, 25 (2),
185–190.
(11) Schrader, K. K.; Nanayakkara, N. P. D. Selective algaecides for
control of Cyanochloronta. U.S. Patent 6,949,250, 2005.
(8) Schrader, K. K.; Nanayakkara, N. P. D.; Tucker, C. S.; Rimando,
A. M.; Ganzera, M.; Schaneberg, B. T. Novel derivatives of 9,10-
anthraquinone are selective algicides against the musty-odor
cyanobacterium Oscillatoria perornata. Appl. EnViron. Microbiol.
2003, 69, 5319–5327.
Received for review September 24, 2007. Revised manuscript received
November 19, 2007. Accepted November 30, 2007. This work was
supported in part by U.S. Department of Agriculture ARS Specific
Cooperative Agreement 58-6408-2-009.
(9) Schrader, K. K.; Tucker, C. S. Evaluation of diquat as a potential
algicide for controlling the musty-odor-producing cyanobacterium,
JF072836S