Phytochemistry 71 (2010) 10321039

Contents lists available at ScienceDirect

Phytochemistry
journal homepage: www.elsevier.com/locate/phytochem

Molecules of Interest

Sorgoleone
Franck E. Dayan a,*, Agnes M. Rimando a, Zhiqiang Pan a, Scott R. Baerson a,
Anne Louise Gimsing b, Stephen O. Duke a
a
b

United States Department of Agriculture, Agricultural Research Service, Natural Products Utilization Research Unit, P.O. Box 8048, University, MS 38677, USA
The Danish Environmental Protection Agency, Strandgade 29 , DK-1401 Kbenhavn K, Denmark

a r t i c l e

i n f o

Article history:
Received 8 January 2010
Received in revised form 9 March 2010
Available online 10 April 2010
Keywords:
Allelopathy
Lipid quinone
Lipid resorcinol
Fatty acid desaturase
Polyketide synthase
O-Methyltransferase
P450 monooxygenase
Mode of action
Compartmentalization
Soil fate
Mineralization

a b s t r a c t
Sorgoleone, a major component of the hydrophobic root exudate of sorghum [Sorghum bicolor (L.)
Moench], is one of the most studied allelochemicals. The exudate also contains an equivalent amount
of a lipid resorcinol analog as well as a number of minor sorgoleone congeners. Synthesis of sorgoleone
is constitutive and compartmentalized within root hairs, which can accumulate up to 20 lg of exudate/
mg root dry weight. The biosynthesis pathway involves unique fatty acid desaturases which produce an
atypical 16:3 fatty acyl-CoA starter unit for an alkylresorcinol synthase that catalyzes the formation of a
pentadecatrienylresorcinol intermediate. This intermediate is then methylated by SAM-dependent
O-methyltransferases and dihydroxylated by cytochrome P450 monooxygenases. An EST data set derived
from a S. bicolor root hair-specic cDNA library contained all the candidate sequences potentially
encoding enzymes involved in the sorgoleone biosynthetic pathway. Sorgoleone interferes with several
molecular target sites, including inhibition of photosynthesis in germinating seedlings. Sorgoleone is
not translocated acropetally in older plants, but can be absorbed through the hypocotyl and cotyledonary
tissues. Therefore, the mode of action of sorgoleone may be the result of inhibition of photosynthesis in
young seedlings in concert with inhibition of its other molecular target sites in older plants. Due to its
hydrophobic nature, sorgoleone is strongly sorbed in soil which increases its persistence, but experiments show that it is mineralized by microorganisms over time.
Published by Elsevier Ltd.

1. Introduction
1.1. Nomenclature consideration
Sorgoleone normally refers to 1 (2-hydroxy-5-methoxy-3-[(80 Z,
110 Z)-80 ,110 ,140 -pentadecatriene]-p-benzoquinone [CAS 10501876-6]), one of the major components of the oily substance exuding
from the roots of sorghum (Sorghum bicolor (L.) Moench) (Fig. 1).
However, the nomenclature surrounding sorgoleone requires some
clarications because this term can also include the reduced form
of 1 and a number of structurally related lipophilic p-benzoquinones present in small amounts in the oily exudate (see Section 2)
(Chang et al., 1986; Netzly et al., 1988). This latter use of the term
sorgoleone should be avoided because the exudate also contains an
equivalent amount of non-quinoid lipid resorcinols (Dayan et al.,
2009; Erickson et al., 2001; Fate and Lynn, 1996). Adding further
confusion, 1 has mistakenly been called xenognosin (Fate and
Lynn, 1996) although the name xenognosin was already associated
with a modied chalcone isolated from tragacanth (Astragalus spp.)

* Corresponding author. Tel.: +1 662 915 1039; fax: +1 662 915 1035.
E-mail addresses: franck.dayan@ars.usda.gov, fdayan@olemiss.edu (F.E. Dayan).
0031-9422/$ - see front matter Published by Elsevier Ltd.
doi:10.1016/j.phytochem.2010.03.011

gum extracts (El-Feraly and Hufford, 1982). In this review, the term
sorgoleone will refer exclusively to 1, unless specically indicated
otherwise.

1.2. Sorgoleone discovery and allelopathy

Sorgoleone was rst discovered by investigators searching for
secondary metabolites involved in triggering the germination of
the obligate parasitic plant Striga asiatica (witchweed) (Chang
et al., 1986; Netzly et al., 1988). It was later demonstrated that 1
was not a chemical cue involved in the complex chemical communication between parasitic weeds and their host plants (Chang
et al., 1986; Fate and Lynn, 1996; Hauck et al., 1992; Hess et al.,
1992). However, germination of Striga hermontica seed was directly
related to the concentration of the dihydrosorgoleone (2), the reduced form of 1. Other scientists suggested that 1 may be involved
in allelopathic potential of sorghum. Indeed, sorghum has been
known to cause soil sickness and affect the growth of other crops
in rotation systems (Breazeale, 1924; Einhellig and Rasmussen,
1989; Forney et al., 1985; Putnam et al., 1983). Early studies reported that several classes of water-soluble compounds, such as
the phenolics, may be responsible for these allelopathic properties,

F.E. Dayan et al. / Phytochemistry 71 (2010) 10321039

1033

Fig. 1. Structure of sorgoleone and related quinoid and resorcinolic lipids present in the sorghum root hair exudate.

but these compounds are ubiquitous in plants and are not likely to
be involved in the unique activity observed in sorghum (Alsaadawi
et al., 1986; Lehle and Putnam, 1983; Panasiuk et al., 1986). On the
other hand, 1 is an uncommon lipid benzoquinone with herbicidal
activity produced exclusively by sorghum species (Netzly and Butler, 1986). It suppresses the growth of a large number of plant species, but it is most active on small-seeded plants (Barbosa et al.,
2001; de Souza et al., 1999; Einhellig and Souza, 1992; Forney
et al., 1985; Netzly and Butler, 1986; Panasiuk et al., 1986). Sorghum is sometimes used in integrated pest management systems
as a green manure or as a cover crop to suppress weed populations
(Weston, 1996) or as a crop residue in no-tillage farming (Alsaadawi and Dayan, 2009) (see Section 5).
2. Structure elucidation and analogs
As mentioned above, 1 is a major constituent of the root extract
of S. bicolor. Sorgoleone and dihydrosorgoleone (2) were identied
and distinguished from each other in the 1H NMR spectrum of the
whole extract. The quinone form (1) accounts for approximately
half of the exudate (Czarnota et al., 2003b; Netzly et al., 1988).
GCMS analysis of individual droplets collected with PDMS probes
(Fig. 2) indicated that the remainder of exudate consists of the resorcinol 10 (Dayan et al., 2009; Erickson et al., 2001). The extract
also contains several congeners of 1, but in much lower quantities
(Kagan et al., 2003) (Fig. 1).

The major and three minor benzoquinone constituents of the

exudate were isolated by HPLC, identied by electron impact mass
spectrometry, and were referred to as sorgoleone-358 (1), sorgoleone-360, sorgoleone-362, and sorgoleone-386, respectively, based
on the observed molecular ions (Netzly et al., 1988). The identication of these phytotoxins as benzoquinones was also supported by
1
H NMR, reportedly showing the signal at d 5.9 (s, 1H) ppm, which
is characteristic for the quinoid proton; cf. d 6.3 (s, 1H) ppm for the
hydroquinone proton.
Successful isolation of sorgoleone analogs with varying number carbons and double bonds in the aliphatic side chain was
achieved by argentation chromatography (Kagan et al., 2003).
Sorgoleone-360 (3), sorgoleone-362 (4), sorgoleone-364 (5), and
sorgoleone-386 (6) were puried and full structural characterization was performed by 1- and 2-dimensional NMR spectroscopy,
revealing the number and positions of the double bonds in the
lipid chain. Sorgoleone-364 has a fully saturated aliphatic 15carbon chain. Interestingly, sorgoleone-360, sorgoleone-362, and
sorgoleone-386 did not have the terminal double bond that is
a characteristic feature of sorgoleone-358 (1). These four sorgoleones inhibited PSII similarly in an assay using spinach thylakoid membranes, suggesting that the aliphatic side chain is not
essential for activity. Barbosa et al. (2001), in their investigation
of the phytotoxic activity of 1, 5 and a synthetic analog (2-acetoxy-5-methoxy-3-(pent-1-yl)-1,4-benzoquinone (7) likewise concluded that the activity resides in the quinone moiety. In support

1034

F.E. Dayan et al. / Phytochemistry 71 (2010) 10321039

Fig. 2. Solid-phase microextraction of the oily droplets accumulating at the tip of

sorghum root hairs using a PDMS probe mounted on a micromanipulator. (A) Before
contact with root hair with oily droplet still present; (B) probe in contact with the
root hair; and (C) after contact with the root hair showing that the oily droplet was
collected by the PDMS ber. Bars in AC represent 20 lm. With permission from
Dayan et al. (2009).

of these observations, two analogs having the same C15, trienoic

side chain as sorgoleone-358, namely, 5-ethoxy sorgoleone (a
natural analog, 8) and 2,5-dimethoxy-sorgoleone (a synthetic

analog, 9) were shown to be less active as PS II inhibitors than

sorgoleone-358 (Rimando et al., 1998).
While the phytotoxic activity of sorgoleone appears to be associated with the quinone moiety, resorcinolic compounds having
the characteristic three double bond, terminal methylene lipid side
chain as sorgoleone, i.e., 4,6-dimethoxy-2-[(80 Z, 110 Z)-80 ,110 ,140 pentadecatriene]resorcinol (10) and 4-methoxy-6-ethoxy-2-[(80 Z,
110 Z)-80 ,110 ,140 -pentadecatriene]resorcinol (11) also inhibited PSII
(Rimando et al., 2003). Compound 10 was more active than sorgoleone-358 in inhibiting the growth of lettuce seedlings. Compound 10 incubated with commercially available fungal
polyphenol oxidase was not converted to a quinone, and kept its
activity, indicating that the quinone functionality is not a strict
structural requirement for phytotoxicity. Compound 10 exists as
a major constituent together with the sorgoleones in the root extract of S. bicolor DeKalb hybrid SX17, as seen in the thin layer
chromatogram of the root extract (Rimando et al., 2005).
It appears therefore that there are two groups of phytotoxic
compounds in the root exudates of S. bicolor: the sorgoleones (predominantly 1) and the resorcinolic lipids (predominantly 10).
These two groups can be distinguished by GCMS. We have reported that the sorgoleones show a base peak at m/z 168 formed
by the loss of a tetradecyl side chain (Fig. 3a). This stable fragment
is consistent with a proposed fragmentation of a quinone that has
similar substitutions as that of sorgoleone (Wong et al., 1985).
However, other groups have reported different fragment ions.
Chang et al. (1986) observed peaks at m/z 167/168/169 and suggested that these ions resulted from benzylic cleavage giving the
ions m/z 167 representing Q (quinone), m/z 168 representing QH+
(quinone + H) and m/z 169 representing QH
2 (quinone + 2H)
(Fig. 3b). Fate and Lynn (1996) also observed the ions m/z 168/
169. However, these researchers suggested that these ions originated from a hydroquinone (Fig. 3c). The resorcinolic lipids (exemplied by 10) shows a base peak at m/z 183 with the loss of a
tetradecyl side chain. The existence of a major component with
m/z 362 has been reported under specic GCMS conditions (Erickson et al., 2001), but the exact nature of this compound is unclear
since our work established that sorgoleone-358 (1) is the most

Fig. 3. (a) Major fragment of sorgoleone-358 (m/z 168) obtained from loss of 14-carbon chain, (b) adapted from Chang et al. (1986) and based on Beaumont and Edwards
(1969) proposed fragmentations of quinones, and (c) adapted from Fate and Lynn (1996).

F.E. Dayan et al. / Phytochemistry 71 (2010) 10321039

abundant lipid benzoquinone in the root exudate of sorghum (Kagan et al., 2003).
3. Biosynthesis, compartmentalization, and production
3.1. Biosynthesis
The biosynthesis of 1 was elucidated using retrobiosynthetic
NMR analysis of root exudate collected from sorghum seedlings labeled with 13C-labeled substrates (Fig. 4) (Dayan et al., 2003; Fate
and Lynn, 1996). The pathway involves the convergence of two
metabolic routes, namely the fatty acid and polyketide synthase
pathways. Incorporation of 2-13C-D-glucose (green in Fig. 4) in
the lipophilic tail of 1 demonstrated that the pathway initiates
with fatty acid synthase and fatty acid desaturases producing an
unusual D9,12,15-16:3-CoA intermediate with a terminal vinyl
group. This CoA derivative serves as the starter unit of a specialized
polyketide synthase (PKS) that can also utilize the malonyl-CoA labeled via the incorporation of 2-13C-D-glucose to form the ring of 1
(green in Fig. 4). Interestingly, labeling with 1-13C-acetate (blue in
Fig. 4) resulted incorporation of 13C in the ring only, which suggests
that the PKS involved in the synthesis of 1 is in a different cellular
compartmentalization than the fatty acid synthase. The lipid resorcinol intermediate is subsequently acted upon by a SAM-dependent O-methyltransferase that acts on one of the hydroxy groups
(red on Fig. 4), and a P450 monooxygenase to produce the reduced
(hydroquinone) form of 1 (Fig. 4).
Our research team subsequently began the systematic molecular
and biochemical characterization of the biosynthesis of 1 (see this
section and Section 4). The synthesis of the C16 acyl-CoA by the action of FAS is ubiquitous to plants. Similarly, the formation of the
D9,12 double bonds are fairly common. However, the addition of

1035

the terminal D15 vinyl double bond characteristic of the tail of 1 is

catalyzed by a unique class of fatty acid desaturase (Pan et al.,
2007). We have established a likely role for two desaturases (DES2
and DES3) in the de novo synthesis of D9,12,15 fatty acids via heterologous expression in yeast (Pan et al., 2007). As such, this step could
be considered the rst committed step in the synthesis of 1. However, the next step in 1 and lipid resorcinol biosynthesis is catalyzed
by a type III polyketide synthase, which truly commits the fatty acid
product to this lipid resorcinol route. The puried root hair preparation readily converted long chain acyl-CoA starter units to their
corresponding lipid resorcinols. Optimum activity was with
decanoyl-CoA, which yielded a 5-n-nonyl-resorcinol. The isolated
root hair preparation also had high S-adenosyl-L-methioninedependent O-methyltransferase activity, which catalyzes the next
step in the biosynthesis pathway. Several 5-alkyl-resorcinols were
readily methylated, but the optimum activity was with 5-n-pentyl-resorcinol. The last biochemical step, catalyzed by a hydroxylase
(putatively a P450 monooxygenase), produces dihydrosorgoleone.
The in situ hydroxylase activity was low relative to the other enzymes in the pathway, but still detectable in isolated root hairs.
Thus, sorghum root hairs possess the entire metabolic machinery
necessary for the biosynthesis of 1. Analysis of individual oily droplets collected with PDMS probes (Fig. 2) revealed that the exudate
consists of a 1:1 ratio of 1 and its dimethylated resorcinol analog
(Dayan et al., 2009; Erickson et al., 2001; Fate and Lynn, 1996). This
analog is the result of the dimethylation of the 5-pentadecatriene
resorcinol (see branch point on Fig. 4).
3.2. Compartmentalization
One of the unique aspects of the biology of 1 is that this lipid
benzoquinone is produced in specialized root hair cells. Indeed,

Fig. 4. Biosynthesis of sorgoleone showing the incorporation of 13C-labeled substrates in the carbon backbone adapted from Dayan et al. (2003). 2-13C-D-glucose (green)
incorporated in carbon atoms 1, 3 and 5 of the quinone head, as well as carbon atoms 20 ,40 ,60 ,80 ,100 ,120 , and 140 of the tail. 1-13C-acetate (blue) incorporated in carbon atoms 1
and 5 of the quinone head but not in the lipid tail. 13C-methyl-L-methionine (red) incorporated in the 3-methoxy group. Notice the branching point resulting from the double
methylation of the 5-pentadecatrienylresorcinol intermediate leading to the dimethylated resorcinol end-product. The names of the enzymes involved in the pathway are
shown in bold capital letters. FAS = fatty acid synthase, SAD1, DES2 and DES3 = fatty acid desaturases introducing the D9,12,15 double bonds, ARS = acyl-resorcinol synthase
(polyketide synthase), OMT3 = O-methyltransferase involved in sorgoleone biosynthesis, and P450 = putative P450 monooxygenase.

1036

F.E. Dayan et al. / Phytochemistry 71 (2010) 10321039

The production of the exudate from the root hairs is independent of the stages of root development (Dayan, 2006), reaching
ca. 20 lg of exudate/mg root dry weight after which the production appears to be suppressed (Dayan et al., 2009). However, more
exudate is produced following gentle washing of the roots with
water, suggesting that the biosynthesis of lipid benzoquinones
and resorcinols is a dynamic process, and that the exudate is involved in some feedback inhibition mechanism (Dayan et al.,
2009). At this time, the regulatory mechanism of root exudate production is unknown, but probably involves a feedback mechanism
when either 1 or one of its biosynthesis intermediates begins to
accumulate. This mechanism may be necessary to protect sorghum
from potential autotoxicity caused by the accumulation of 1.
While the production is mostly constitutive, it is sensitive to
temperature, with optimum levels at 2535 C. The amount of exudate was lower at colder temperatures, and reduced by 95% at
40 C (Dayan, 2006). Also, the production appears to be stimulated
by the presence of other plants (Dayan, 2006), but more research
needs to be performed to quantify the extent of this response.

Two fatty acid desaturase-like sequences were identied within

the EST data; one of which encoded an enzyme capable of converting palmitoleic acid to hexadecadienoic acid (DES2), and a second
capable of converting hexadecadienoic acid into hexadecatrienoic
acid (DES3) (Pan et al., 2007). These results were obtained by heterologous expression in Saccharomyces and provided strong evidence for the involvement of DES2/3 in the synthesis of 1 in
planta. Additional studies led to the identication of an OMT,
which, based on substrate specicity, gene expression, and molecular modeling studies, also appears likely to participate in the 1
pathway in planta (Baerson et al., 2006). This OMT, designated
OMT3, was found along with DES2 and DES3 to represent some
of the most abundant sequences expressed in sorghum root hair
cells.
More recently, we have also pursued the identication of polyketide synthases capable of catalyzing the formation of the 5-pentadecatrienylresorcinol pathway intermediate utilizing the
hexadecatrienoic acid substrate generated by the sequential action
of DES2 and DES3 on palmitoleic acid. Type III PKS enzymes capable of generating alkylresorcinols represent a new subfamily referred to as alkylresorcinol synthases (ARS) and have recently
been identied in Azotobacter vinelandii and Neurospora crassa
(Funa et al., 2007, 2006). Among the ve PKS-like sequences identied in the sorghum root hair ESTs, two accepted a diversity of
fatty-acyl-CoA starters, and RNAi-mediated inhibition of their
expression in transgenic sorghum resulted in the loss of detectable
levels of 1 (Cook et al., 2010). In total, genes likely representing
three of the four enzyme classes required for the biosynthesis of
1 have now been identied (Fig. 4), and we anticipate completing
the remaining cytochrome P450 gene characterization work in the
near future.
From an evolutionary biology perspective, it is of particular
interest that the complete biosynthesis pathway of 1 appears
exclusive to Sorghum spp. and has not been identied in any other
Poaceae member (Baerson et al., 2008b). The ability to produce
alkylresorcinols and related phenolic lipids is, on the other hand,
widespread throughout the plant kingdom, and these compounds
have been proposed to serve as a means of defense against microbial pathogens (e.g. Kozubek et al., 2001). The identication of
higher plant ARS (acyl-resorcinol synthase) enzymes such as those
identied from sorghum and rice (Cook et al., 2010) creates new
opportunities for the exploration of phenolic lipid biosynthetic
pathways in plants. The current paucity of higher plant ARS examples however makes inferences into the emergence of this PKS subfamily circumspect at best. It will therefore be of signicant
interest for future studies to determine whether ARS enzymes, like
stilbene synthases and other non-chalcone synthase PKSs (Jiang
et al., 2008), are polyphetic in origin or, alternatively, evolved from
a common ancestor.

4. Genetics and evolution of pathway

5. Biological activities and signicance

As mentioned in Section 3, the biosynthetic pathway resides in

root hair cells. Furthermore, given the copious amounts of 1 produced by this cell type, it was postulated that all the genes involved in the biosynthesis of 1 would be among the most highly
expressed in these specialized cells. Therefore, a root hair-specic
expressed sequence tags (EST) library was generated to identify
potential candidate desaturase (DES), polyketide synthase (PKS),
O-methyltransferase (OMT), and cytochrome P450s for follow-up
recombinant enzyme studies. A relatively modest dataset consisting of 5468 50 -sequenced ESTs were obtained from these efforts,
which after cluster analysis revealed seven unique DES-like sequences, ve polyketide synthase (PKS)-like sequences, 12 OMTlike sequences, and 11 P450-like sequences (Baerson et al., 2008a).

The allelopathic potential of 1 is strongest on small-seeded

weeds (de Souza et al., 1999; Einhellig and Souza, 1992; Netzly
and Butler, 1986; Nimbal et al., 1996a). Large seeded weeds tend
to be less sensitive to 1. This is also the case for many soil-applied
commercial herbicides. Larger plants may avoid the herbicidal effect by having lower absorption and translocation, faster metabolic
degradation of the allelochemical, or simply by their roots rapidly
growing beyond the zone of the sorghum rhizophere where 1
accumulates.
As the phytotoxic exudate is released directly in the soil, its action is similar to a pre-plant incorporated herbicide. One factor that
may prolong the persistence of 1 in soil is the fact that it may be
released continually from the roots during the growing season of

the presence of oily droplets exuding from the tip of sorghum root
hairs was rst reported by Netzly and Butler (1986) and Yang et al.
(2004) later demonstrated that the biosynthesis of 1 is intrinsically
linked to the presence of living root hairs. Observation of the ultrastructure of sorghum root hairs revealed these cells are rich with
mitochondria, endoplastic reticulum and numerous vesicles, indicating that these specialized cells are highly physiologically active
(Czarnota et al., 2003a). Subsequent studies performed with isolated root hair preparation demonstrated that mature sorghum
root hairs contain the entire genetic material and biochemical
machinery required for the production of this bioactive benzoquinone (Baerson et al., 2008a; Czarnota et al., 2003a; Dayan et al.,
2007; Pan et al., 2007) (see Sections 3.3 and 4).
Compartmentalization of highly specialized secondary metabolic pathways in trichomes are common in plants. Since trichomes are dened as specialized cells derived from the
epidermis, root hairs are, from a morphological standpoint, classied as trichomes (Werker, 2000). Interestingly, recent advances in
the eld of molecular biology have corroborated their somatic similarity by demonstrating that root hairs are under some of the same
genetic controls as leaf, stems and ower trichomes e.g., (Kellogg,
2001). While the synthesis of bioactive secondary metabolites is
often associated with trichomes on the surfaces of leaves and
stems, sorghum root hairs are a unique example of root localized
trichomes that behave as natural pesticide factories (Dayan and
Duke, 2003).

3.3. Production

F.E. Dayan et al. / Phytochemistry 71 (2010) 10321039

sorghum. This slow release of de novo synthesized 1 may sustain

its concentration in soil over a much longer time than that typically
resulting from a single application of a herbicide.
The molecular target sites affected by 1 include photosynthetic
and mitochondrial electron transport (Einhellig et al., 1993; Gonzalez et al., 1997; Nimbal et al., 1996b; Rasmussen et al., 1992;
Rimando et al., 1998), the enzyme p-hydroxyphenylpyruvate dioxygenase (Meazza et al., 2002), and root H+-ATPase and water uptake (Hejl and Koster, 2004). However, the in vivo mechanism of
action of 1 and its analogs is not well understood. In particular, 1
is a potent inhibitor of PSII in isolated chloroplasts, being as effective as diuron at inhibiting photosynthetic electron transport as
diuron (I50 ca. 100 nM) (Gonzalez et al., 1997; Nimbal et al.,
1996a). Its efcacy is not reduced in triazine-resistant pigweed
(Dayan et al., 2009) because the common mutation of Ser264 to
Gly or Ala in PSII causes resistance to triazines, but not to the quinone inhibitors (Oettmeier et al., 1982). However, Hejl and Koster
have shown that applying 1 to the foliage of mature plants did not
affect photosynthesis (Hejl and Koster, 2004). Furthermore, they
questioned the ability of this highly lipophilic natural herbicide
to be taken up by roots and translocated to the foliage where it
must enter the chloroplast and inhibit PSII in the thylakoid membrane (Hejl and Koster, 2004). The problems posed by the spatial
separation between the location of 1 exudation (soil) and its putative site of action (foliage) as a PSII inhibitor has been investigated
recently. As reported by Hejl and Koster (2004), 1 had no effect on

1037

the photosynthesis of older plants, whether applied to the roots or

foliage (Dayan et al., 2009). However, it inhibits photosynthesis in
germinating seedlings and plants 4 days or younger (Fig. 5A). Sorgoleone is not translocated acropetally in older plants (Fig. 5B), but
can be absorbed through the hypocotyl and cotyledonary tissues.
Therefore, the mode of action of 1 may be the result of inhibition
of photosynthesis in young seedlings in concert with inhibition
of its other molecular target sites in older plants (Dayan et al.,
2009).
6. Soil behavior of sorgoleone
6.1. Sorption of sorgoleone
Sorgoleone is a very hydrophobic compound with a log Kow of
6.1 (Trezzi et al., 2006). Due to its hydrophobicity, it is expected
to sorb strongly to soil, especially to the organic matter but also
to other hydrophobic components in soils. These properties are
ideal for biological activity of a somewhat persistent allelochemical
that must be retained in the root zone of germinating target species. A few studies have investigated the sorption of 1 in soil. Sorption of 1 to an ultisol in a methanol system resulted in a linear
isotherm and a methanol:water (60:40) system could be tted by
either two linear isotherms or the Freundlich isotherm (Demuner
et al., 2005; Trezzi et al., 2006). These results suggests that 1 has
a higher afnity for the soil in the hydrophobic methanol system
than in the less hydrophobic methanol:water system perhaps
caused by conformational changes in the soil organic matter
(Demuner et al., 2005; Trezzi et al., 2006).
The Kd values for 1 ranged from 19.7 to 91.3 L kg1 which correspond to Koc values of 11345254 L kg1. These values are much
lower than what could be expected from a compound with a very
high Kow. Trezzi et al. (2006) argues that the relatively low Kd and
Koc values could be due to ionization of the quinone part of the 1
molecule that has a pKa of 6.0. In a recent study the effect of pH
on sorption of 1 was investigated, but due to variation in the data
it was difcult to draw any conclusions. However, there was a tendency towards lower sorption at the highest pH value tested (pH
8.1) (C.N. Thomsen, personal communications). This was to be expected as this pH is well above the pKa value of 1. Extracting 1 from
the soil with 0.01 M CaCl2 resulted in linear isotherms, supporting
the hypothesis of sorption to the organic matter by partitioning. Kd
values of 1 ranged between 185.4 and 343.9 L kg1, the lowest Kd
value being at the highest pH. As expected, Kd values measured
in the aqueous system are higher than those in the more hydrophobic systems (Trezzi et al., 2006).
6.2. Degradation of sorgoleone

Fig. 5. (A) Effect of sorgoleone (j) and atrazine (.) on photosynthetic electron rate
of velvetleaf leaf tissues of different age. Measurements at 4 and 7 days old plants
were done on cotyledon discs. All other time points were done on leaf discs. The
samples were incubated for 6 h on 100 lM sorgoleone or 33 lM atrazine prior to
analysis. (B) Picture of a representative velvetleaf seedling used in the 14C-ring
labeled sorgoleone root uptake experiment (left) and the autoradiogram showing
the lack of acropetal transport of sorgoleone (right). With permission from Dayan
et al. (2009).

Sorgoleone has some persistence in soil. Detectable levels of 1

were measured seven weeks after incorporation (Czarnota et al.,
2001), and the soil half-life of 1 has been estimated at 10 days
(Demuner et al., 2005). There are no detectable levels of 1 60 days
after incorporation (Demuner et al., 2005). Mineralization of 1 in
Danish and American soils has been investigated using radiolabeled forms of the natural quinone (Gimsing et al., 2009). The potential for degradation was present even in soils where the
microorganisms have never encountered 1 before. However, American soils previously exposed to sorghum had the greatest potential and up to 30% of 1 was mineralized after 60 days. The
mineralization was also dependent on the initial concentration,
with greater mineralization at the higher concentration. The study
also showed that different parts of the molecule are not mineralized at the same rate. The methoxy group is quickly mineralized
whereas mineralization of the other parts of the molecule was

1038

F.E. Dayan et al. / Phytochemistry 71 (2010) 10321039

slower. Microorganisms in the American soils can use 1 as a carbon

source, which means a more rapid and more complete degree of
mineralization and reduced risk of persistence.
The effect of pH on 1 mineralization was investigated by adjusting the pH of a Dundee Silt loam to 4.0, 4.6, 5.6, 7.1, and
8.1. Mineralization of 1 decreased at pH 4.0 compared to the
soil samples with higher pH values and increased at pH 8.1. Differences in sorption and hence bioavailability may partly explain
the differences in mineralization, but the differences may also be
due to differences in the microbial community. 16S rRNA based
DGGE indicated the presence of a sorgoleone-degrading microorganism at pH 8.1. Sequencing of a band from DGGE-gels suggested that Pseudomonas bacteria might be involved in enhanced
1 degradation in the pH 8.1 soil.
7. Concluding remarks
Sorgoleone is perhaps the most thoroughly studied allelochemical, even though our knowledge of other allelochemicals such as
m-tyrosine and benzoxazinoids is growing rapidly (Dayan and
Duke, 2009). But, for no other allelochemical is such a complete
understanding of the biochemical pathway, genetics, enzymology,
cell biology, behavior in soil, and mode of action in soil available.
Still, there is much left to do. The capability of manipulating the
production of sorgoleone either up or down in sorghum to explore
its biological role is now within reach. Will eliminating its biosynthesis completely eliminate the allelopathic properties of Sorghum
species? Will this solve the soil sickness problem with sorghum?
Will increasing production of sorgoleone increase the allelopathic
trait for sorghum, or will it cause autotoxicity? These and other aspects of the production and role of sorgoleone in sorghum are now
within our grasp.
References
Alsaadawi, I.S., Dayan, F.E., 2009. Potentials and prospects of sorghum allelopathy in
agroecosystems. Allelo. J. 24, 255270.
Alsaadawi, I.S., Al-Uqaili, J.K., Alrubeaa, A.J., Al-Hadithy, S.M., 1986. Allelopathic
suppression of weed and nitrication by selected cultivars of Sorghum bicolor
(L.) Moench. J. Chem. Ecol. 12, 209219.
Baerson, S.R., Dayan, F.E., Rimando, A.M., Pan, Z., Cook, D.D., Nanayakkara, N.P.D.,
Duke, S.O., 2006. A functional genomics approach for the identication of genes
involved in the biosynthesis of the allelochemical sorgoleone. In: Rimando,
A.M., Duke, S.O. (Eds.), Natural Products for Pest Management. ACS Symposium
Series, vol. 927, pp. 265276.
Baerson, S.R., Dayan, F.E., Rimando, A.M., Nanayakkara, N.P.D., Liu, C.-J., Schrder, J.,
Fishbein, M., Pan, Z., Kagan, I.A., Pratt, L.H., Cordonnier-Pratt, M.-M., Duke, S.O.,
2008a. A functional genomics investigation of allelochemical biosynthesis in
Sorghum bicolor root hairs. J. Biol. Chem. 283, 32313247.
Baerson, S.R., Rimando, A.M., Pan, Z., 2008b. Probing allelochemical biosynthesis in
sorghum root hairs. Plant Signal. Behav. 9, 667670.
Barbosa, L.C.A., Ferreira, M.L., Demuner, A.J., da Silva, A.A., de Cassia Pereira, R., 2001.
Preparation and phytotoxicity of sorgoleone analogues. Quim. Nova 24, 751
755.
Beaumont, P.C., Edwards, R.L., 1969. Constituents of the higher fungi. Part IX.
Bovinone, 2,5-dihydroxy-3-geranylgeranyl-1,4-benzoquinone from Boletus
(Suillus) bovinus (Linn. ex Fr.) Kuntze. J. Chem. Soc. C, 23982403.
Breazeale, J.F., 1924. The injurious after-effects of sorghum. J. Am. Soc. Agron. 16,
689700.
Chang, M., Netzly, D.H., Butler, L.G., Lynn, D.G., 1986. Chemical regulation of
distance: characterization of the rst natural host germination stimulant for
Striga asiatica. J. Am. Chem. Soc. 108, 78587860.
Cook, D.D., Rimando, A.M., Clemente, T.E., Dayan, F.E., Nanayakkara, N.P.D.,
Schrder, J., Pan, Z., Noonan, B.P., Duke, S.O., Baerson, S.R., 2010.
Alkylresorcinol synthases from Sorghum bicolor involved in the biosynthesis
of the allelopathic benzoquinone sorgoleone. Plant Cell., doi:10.1105/
tpc.109.072397.
Czarnota, M.A., Paul, R.N., Dayan, F.E., Nimbal, C.I., Weston, L.A., 2001. Mode of
action, localization of production, chemical nature, and activity of sorgoleone: a
potent PSII inhibitor in Sorghum spp. root exudates. Weed Technol. 15, 813
825.
Czarnota, M.A., Paul, R.N., Weston, L.A., Duke, S.O., 2003a. Anatomy of sorgoleonesecreting root hairs of Sorghum species. Int. J. Plant Sci. 164, 861866.
Czarnota, M.A., Rimando, A.M., Weston, L.A., 2003b. Evaluation of root exudates of
seven sorghum accessions. J. Chem. Ecol. 29, 20732083.

Dayan, F.E., 2006. Factors modulating the levels of the allelochemical sorgoleone in
Sorghum bicolor. Planta 224, 339346.
Dayan, F.E., Duke, S.O., 2003. Trichomes and root hairs: natural pesticide factories.
Pestic. Outlook 4, 175178.
Dayan, F.E., Duke, S.O., 2009. Biological activity of allelochemicals. In: Osbourn, A.E.,
Lanzotti, V. (Eds.), Plant-derived Natural Products: Synthesis, Function, and
Application. Springer, New York, NY, pp. 361384.
Dayan, F.E., Kagan, I.A., Rimando, A.M., 2003. Elucidation of the biosynthetic
pathway of the allelochemical sorgoleone using retrobiosynthetic NMR
analysis. J. Biol. Chem. 278, 2860728611.
Dayan, F.E., Watson, S.B., Nanayakkara, N.P.D., 2007. Biosynthesis of lipid
resorcinols and benzoquinones in isolated secretory plant root hairs. J. Exp.
Bot. 58, 32633272.
Dayan, F.E., Howell, J.L., Weidenhamer, J.D., 2009. Dynamic root exudation of
sorgoleone and its in planta mechanism of action. J. Exp. Bot. 60, 21072117.
de Souza, C.N., de Souza, I.F., Pasqual, M., 1999. Extrao e ao de sorgoleone sobre
o crescimento de plantas. Cinc. Agrotechnol. 23, 331338.
Demuner, L.A.J., Barbosa, C.A., Luiz, S., Chinelatto, L.S., Reis, C., 2005. Sorption and
persistence of sorgoleone in redyellow latosol. Quim. Nova 28, 451455.
Einhellig, F.A., Rasmussen, J.A., 1989. Prior cropping with grain sorghum inhibits
weeds. J. Chem. Ecol. 15, 951960.
Einhellig, F.A., Souza, I.F., 1992. Phytotoxicity of sorgoleone found in grain sorghum
root exudates. J. Chem. Ecol. 18, 111.
Einhellig, F.A., Rasmussen, J.A., Hejl, A.M., Souza, I.F., 1993. Effects of root exudate
sorgoleone on photosynthesis. J. Chem. Ecol. 19, 369375.
El-Feraly, F.S., Hufford, C.D., 1982. Synthesis and carbon-13 nuclear magnetic
resonance assignments of xenognosin. J. Org. Chem. 47, 15271530.
Erickson, J., Schott, D., Reverri, T., Muhsin, W., Ruttledge, T., 2001. GCMS analysis of
hydrophobic root exudates of Sorghum and implications on the parasitic plant
Striga asiatica. J. Agric. Food Chem. 49, 55375542.
Fate, G.D., Lynn, D.G., 1996. Xenognosin methylation is critical in dening the
chemical potential gradient that regulates the spatial distribution in Striga
pathogenesis. J. Am. Chem. Soc. 118, 1136911376.
Forney, D.R., Foy, C.L., Wolf, D.D., 1985. Weed suppression in no-till alfalfa
(Medicago sativa) by prior cropping with summer-annual forage grasses.
Weed Sci. 33, 490497.
Funa, N., Osawa, H., Hirata, A., Horinouchi, S., 2006. Phenolic lipid synthesis by type
III polyketide synthases is essential for cyst formation in Azotobacter vinelandii.
Proc. Nat. Acad. Sci. USA 103, 63566361.
Funa, N., Awakawa, T., Horinouchi, S., 2007. Pentaketide resorcylic acid synthesis by
type III polyketide synthase from Neurospora crassa. J. Biol. Chem. 282, 14476
14481.
Gimsing, A.L., Blum, J., Dayan, F.E., Locke, M., Sejer, L.H., Jacobsen, C.S., 2009. Mineralization of the allelochemical sorgoleone in soil. Chemosphere 76, 10411047.
Gonzalez, V.M., Kazimir, J., Nimbal, C., Weston, L.A., Cheniae, G.M., 1997. Inhibition
of a photosystem II electron transfer reaction by the natural product sorgoleone.
J. Agric. Food Chem. 45, 14151421.
Hauck, C., Muller, S., Schildknecht, H., 1992. A germination stimulant for parasitic
owering plants from Sorghum bicolor, a genuine host plant. J. Plant Physiol.
139, 474478.
Hejl, A.M., Koster, K.L., 2004. The allelochemical sorgoleone inhibits root H+-ATPase
and water uptake. J. Chem. Ecol. 30, 21812191.
Hess, D., Ejeta, G., Butler, L.G., 1992. Selecting sorghum genotypes expressing a
quantitative biosynthetic trait that confers resistance to Striga. Phytochemistry
31, 493497.
Jiang, C., Kim, S.Y., Suh, D.Y., 2008. Divergent evolution of the thiolase superfamily
and chalcone synthase family. Mol. Phylogenet. Evol. 49, 691701.
Kagan, I.A., Rimando, A.M., Dayan, F.E., 2003. Chromatographic separation and
in vitro activity of sorgoleone congeners from the roots of Sorghum bicolor. J.
Agric. Food Chem. 51, 75897595.
Kellogg, E.A., 2001. Root hairs, trichomes and the evolution of duplicate genes.
Trends Plant Sci. 6, 550552.
Kozubek, A., Zarnowski, R., Stasiuk, M., Gubernator, J., 2001. Natural amphiphilic
phenols as bioactive compounds. Cell. Mol. Biol. Lett. 6, 351355.
Lehle, F.R., Putnam, A.R., 1983. Allelopathic potential of sorghum (Sorghum bicolor):
isolation of seed germination inhibitors. J. Chem. Ecol. 9, 12231234.
Meazza, G., Schefer, B.E., Tellez, M.R., Rimando, A.M., Nanayakkara, N.P.D., Khan,
I.A., Abourashed, E.A., Romagni, J.G., Duke, S.O., Dayan, F.E., 2002. The inhibitory
activity of natural products on plant p-hydroxyphenylpyruvate dioxygenase.
Phytochemistry 59, 281288.
Netzly, D.H., Butler, L.G., 1986. Roots of sorghum exude hydrophobic droplets
containing biologically active components. Crop Sci. 26, 775778.
Netzly, D.H., Riopel, J.L., Ejeta, G., Butler, L.G., 1988. Germination stimulants of
witchweed (Striga asiatica) from hydrophobic root exudate of sorghum
(Sorghum bicolor). Weed Sci. 36, 441446.
Nimbal, C.I., Pedersen, J.F., Yerkes, C.N., Weston, L.A., Weller, S.C., 1996a.
Phytotoxicity and distribution of sorgoleone in grain sorghum germplasm. J.
Agric. Food Chem. 44, 13431347.
Nimbal, C.I., Yerkes, C.N., Weston, L.A., Weller, S.C., 1996b. Herbicidal activity and site
of action of the natural product sorgoleone. Pestic. Biochem. Physiol. 54, 7383.
Oettmeier, W., Masson, K., Fedtke, C., Konze, J., Schmidt, R.R., 1982. Effect of
different photosystem II inhibitors on chloroplasts isolated from species either
susceptible or resistant toward s-triazine herbicides. Pestic. Biochem. Physiol.
18, 357367.
Pan, Z., Rimando, A.M., Baerson, S.R., Fishbein, M., Duke, S.O., 2007. Functional
characterization of desaturases involved in the formation of the terminal

F.E. Dayan et al. / Phytochemistry 71 (2010) 10321039

double bond of an unusual 16:3D9,12,15 fatty acid isolated from Sorghum bicolor
root hairs. J. Biol. Chem. 282, 43264335.
Panasiuk, O., Bills, D.D., Leather, G.R., 1986. Allelopathic inuence of Sorghum bicolor
on weeds during germination and early development of seedlings. J. Chem. Ecol.
12, 15331543.
Putnam, A.R., DeFrank, J., Barnes, J.P., 1983. Exploitation of allelopathy for weed
control in annual and perennial cropping systems. J. Chem. Ecol. 8, 10011010.
Rasmussen, J.A., Hejl, A.M., Einhellig, F.A., Thomas, J.A., 1992. Sorgoleone from root
exudate inhibits mitochondrial functions. J. Chem. Ecol. 18, 197207.
Rimando, A.M., Dayan, F.E., Czarnota, M.A., Weston, L.A., Duke, S.O., 1998. A new
photosystem II electron transfer inhibitor from Sorghum bicolor. J. Nat. Prod. 61,
927930.
Rimando, A.M., Dayan, F.E., Streibig, J.C., 2003. PSII inhibitory activity of resorcinolic
lipids from Sorghum bicolor. J. Nat. Prod. 66, 4245.
Rimando, A.M., Kagan, I.S., Dayan, F.E., Czarnota, M.A., Weston, L.A., 2005. Chemical
basis for weed suppressive activity of Sorghum. In: Petroski, R.J., Tellez, M.R.,

1039

Behle, R.W. (Eds.), Semiochemicals in Pest and Weed Control, vol. 906.
American Chemical Society, Washington, pp. 5970.
Trezzi, M.M., Vidal, R.A., Dick, D.P., Peralba, M.C.R., Kruse, N.D., 2006. Sorptive
behavior of sorgoleone in ultisol in two solvent systems and determination of
its lipophilicity. J. Environ. Sci. Health Part B Pestic. Food Contam. Agric.
Wastes 41, 345356.
Werker, E., 2000. Trichome diversity and development. Adv. Bot. Res. 31, 135.
Weston, L.A., 1996. Utilization of allelopathy for weed management in
agroecosystems. Agron. J. 88, 860866.
Wong, S.-M., Pezzuto, J.M., Fong, H.H.S., Farnsworth, N.R., 1985. Isolation, structural
elucidation and chemical synthesis of 2-hydroxy-3-octadecyl-5-methoxy-1,4benzoquinone (irisoquin), a cytotoxic constituent of Iris missouriensis. J. Pharm.
Sci. 74, 11141116.
Yang, X., Owens, T.G., Schefer, B.E., Weston, L.A., 2004. Manipulation of root hair
development and sorgoleone production in sorghum seedlings. J. Chem. Ecol.
30, 199213.