Feasibility of classical biological control of Sonchus oleraceus in Australia

Abstract

Sonchus oleraceus L. (Asteraceae), an annual species native to Eurasia and northern Africa, is among the most widely distributed plant species on Earth. In Australia, S. oleraceus, is a common weed in disturbed areas such as crop fields, pastures, gardens and roadsides. In agricultural settings, it can dominate fallows and cultivated fields where it uses stored soil moisture and reduces crop yield. This weed has also developed herbicide resistance, predicating the need for alternative management solutions. In this context, we undertook field surveys and preliminary host range studies in the native range of S. oleraceus to determine the feasibility of developing classical biological control solutions for Australia. Fifty-nine phytophagous arthropod species were recorded and nine pathogenic fungi were recovered from disease symptoms. Four arthropod species were selected for initial host-specificity testing based on information available in the literature. Preliminary host-specificity tests were also performed with representative isolate(s) of six of the pathogenic fungi. All these candidate agents were shown in the tests to affect key native species in Australia in the same subtribe as S. oleraceus (i.e. Sonchus hydrophilus and Actites megalocarpus). The results of our investigations suggest that classical biological control may not be a feasible option for the management of S. oleraceus in Australia, and that alternative integrated weed management tactics may need to be pursued to mitigate the impacts of this weed.

Full text

Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=cbst20
Biocontrol Science and Technology
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/cbst20
Feasibility of classical biological control of Sonchus
oleraceus in Australia
Vincent Lesieur, Mireille Jourdan, Thierry Thomann, Mélodie Ollivier,
Johannes Tavoillot, Louise Morin & S. Raghu
To cite this article: Vincent Lesieur, Mireille Jourdan, Thierry Thomann, Mélodie Ollivier, Johannes
Tavoillot, Louise Morin & S. Raghu (2021): Feasibility of classical biological control of Sonchus
oleraceus in Australia, Biocontrol Science and Technology, DOI: 10.1080/09583157.2021.1936451
To link to this article: https://doi.org/10.1080/09583157.2021.1936451
View supplementary material
Published online: 06 Jun 2021.
Submit your article to this journal
View related articles
View Crossmark data

RESEARCH ARTICLE
Feasibility of classical biological control of Sonchus oleraceus
in Australia
Vincent Lesieur
a,b
, Mireille Jourdan
a
, Thierry Thomann
a
, Mélodie Ollivier
b
,
Johannes Tavoillot
b
, Louise Morin
c
and S. Raghu
d
a
CSIRO Health and Biosecurity, European Laboratory, Montferrier sur Lez, France;
b
CBGP, Montpellier
SupAgro, INRAE, CIRAD, IRD, Univ Montpellier, Montferrier-sur-Lez, France;
c
CSIRO Health and Biosecurity,
Black Mountain Science and Innovation Park, Acton, Australia;
d
CSIRO Health and Biosecurity, Ecosciences
Precinct, Brisbane, Australia
ABSTRACT
Sonchus oleraceus L. (Asteraceae), an annual species native to
Eurasia and northern Africa, is among the most widely distributed
plant species on Earth. In Australia, S. oleraceus, is a common
weed in disturbed areas such as crop fields, pastures, gardens
and roadsides. In agricultural settings, it can dominate fallows
and cultivated fields where it uses stored soil moisture and
reduces crop yield. This weed has also developed herbicide
resistance, predicating the need for alternative management
solutions. In this context, we undertook field surveys and
preliminary host range studies in the native range of S. oleraceus
to determine the feasibility of developing classical biological
control solutions for Australia. Fifty-nine phytophagous arthropod
species were recorded and nine pathogenic fungi were recovered
from disease symptoms. Four arthropod species were selected for
initial host-specificity testing based on information available in
the literature. Preliminary host-specificity tests were also
performed with representative isolate(s) of six of the pathogenic
fungi. All these candidate agents were shown in the tests to
affect key native species in Australia in the same subtribe as
S. oleraceus (i.e. Sonchus hydrophilus and Actites megalocarpus).
The results of our investigations suggest that classical biological
control may not be a feasible option for the management of
S. oleraceus in Australia, and that alternative integrated weed
management tactics may need to be pursued to mitigate the
impacts of this weed.
ARTICLE HISTORY
Received 12 March 2021
Accepted 25 May 2021
KEYWORDS
Arthropods; classical
biological control; common
sowthistle; host-specificity;
natural enemy surveys; plant
pathogens
Introduction
The genus Sonchus (Asteraceae), especially species with a Palearctic origin, comprises
many successful invaders in several parts of the world. Sonchus arvensis L. and
Sonchus asper (L.) Hill are major weeds in Canada (Hutchinson et al., 1984; Park
et al., 2012; Skinner et al., 2000). Sonchus oleraceus L., an annual species native to
© 2021 Informa UK Limited, trading as Taylor & Francis Group
CONTACT Vincent Lesieur vincent.lesieur@csiro.au CSIRO Health and Biosecurity, European Laboratory, 830
avenue du Campus Agropolis, Montferrier sur Lez 34980, France
Supplemental data for this article can be accessed at https://doi.org/10.1080/09583157.2021.1936451.
BIOCONTROL SCIENCE AND TECHNOLOGY
https://doi.org/10.1080/09583157.2021.1936451

Eurasia and northern Africa (Boulos, 1973), is among the most widely distributed plant
species on Earth, across temperate, Mediterranean, subtropical, and tropical regions
(Pyšek et al., 2017). In Australia, S. oleraceus was first reported at the time of the first
British settlement in the late eighteenth century (Fensham & Laffineur, 2019), and is
now widespread within the country, where it is routinely referred to as common sowthis-
tle (McCarren & Scott, 2013).
Sonchus oleraceus can persist across diverse habitats and climates (Hutchinson et al.,
1984; Widderick et al., 2004). It is a common weed in disturbed areas such as fields, pas-
tures, gardens and roadsides (Hutchinson et al., 1984) (Supplementary Materials Figure
S1). It flowers throughout the year and is a prolific seed producer –one plant can produce
up to 200,000 seeds (McIntosh, 2018). Those seeds can germinate all year round depend-
ing on soil moisture availability (Widderick et al., 2004; Widderick et al., 2010).
The weed can dominate in fallows and cultivated fields (Figure 1), where it uses stored
soil moisture, reduces crop yield and contaminates harvested grain with seeds and green
matter (Widderick et al., 1999). Grower surveys identified S. oleraceus as one of the main
problem weeds in the western and northern grain producing regions of Australia
(Walker et al., 2005; Osten et al., 2007). Conservation tillage enhances density of the
weed by creating a more favourable environment for seeds to germinate over a wider
period of time (Widderick et al., 2010). The development of resistance to routinely
used herbicides, has made populations difficult to manage in cropping systems
(Adkins et al., 1997; Manalil et al., 2017; St John-Sweeting, 2011). Further, S. oleraceus
is an alternate host for insects that can transmit viral diseases to commercial crops
(Moreno & Fereres, 2012; Stubbs et al., 1963). It also constitutes a reservoir for major
agricultural insect pests (e.g. the cotton bollworm, Helicoverpa armigera (Hübner))
(Gregg et al., 2019; Gu et al., 2001; Rajapakse & Walter, 2007) and for some diseases
(e.g. downy mildew of lettuce) (Vieira & Barreto, 2006).
Figure 1. Geographic map of sampling locations of natural enemies of Sonchus oleraceus in the native
range. Dots represent each sampling site (n= 122).
2V. LESIEUR ET AL.

A recent study showed that S. oleraceus populations in Australia outperform
counterpart populations in the native range by producing larger plants (Ollivier
et al., 2020); potentially indicating that it is under less pressure from natural
enemies in the invaded range (Keane & Crawley, 2002;Liu&Stiling,2006). While
S. oleraceus already hosts several natural enemies in Australia, their collective
impact is insufficient to keep the plant under control. Field surveys conducted in
south-western Western Australia, south-east Queensland, coastal South Australia
and Victoria in 2006–2008 identified a novel species of eriophyid mite, Aceria
thalgi Knihinicki, likely native to Australia and possibly New Zealand (Knihinicki
et al., 2009; McCarren & Scott, 2013). This mite causes curling and rolling of leaves
(McCarren & Scott, 2013). These surveys also widely recorded the rust fungus,
Miyagia pseudosphaeria (Mont.) Jørst., probably of Eurasian origin, on S. oleraceus
(McCarren & Scott, 2017). The eriophyid mite and the rust fungus were shown to
be specifictoSonchus spp. (McCarren & Scott, 2013;2017)andwhiletheycouldposs-
ibly be augmented at regular intervals to increase their impact in Australia, this
approach would likely be unsustainable in the long term. A more recent study, com-
bining a literature review and field surveys, identified 21 phytophagous arthropod
species, mostly generalists of exotic origin, able to feed and develop on S. oleraceus
in Australia (Ollivier et al., in press). There may be scope to complement effects of
these natural enemies already present on S. oleraceus in Australia by introducing
additional, highly specialised herbivores and pathogens from the native range. Such
enhanced biological control could greatly contribute to landscape-scale management
of S. oleraceus in Australia.
A biological control program for S. arvensis in Canada was undertaken in the
1970–1980s (McClay & Peschken, 2001), with surveys for potential insect agents
conducted in central Europe (Schroeder, 1974). Since there are no native species
in the subtribe Sonchinae in Canada (Bremer, 1994), it was possible to introduce
three biological control insect agents specifictothegenusSonchus:Cystiphora
sonchi Vallot, Liriomyza sonchi Hendel and Tephritis dilacerata Loew (McClay &
Peschken, 2001). The situation in Australia is different because there are two
native species in the same subtribe as S. oleraceus,Sonchus hydrophilus Boulos
and Actites megalocarpus (Hook. f.) Lander (Kim et al., 2004;Kimetal.,2007). Con-
sequently, a higher level of host-specificity is required for natural enemies to be con-
sidered for introduction as biological control agents in Australia. Preliminary
surveys for natural enemies of S. oleraceus were undertaken in southern France in
2004–2005 where the climate closely matches that of areas where the weed occurs
in Australia (Scott & McCarren, 2012). Several candidate agents were found but
their specificity was not assessed.
In this paper, we report on results from additional field surveys of natural enemies of
S. oleraceus conducted across a wide range of climates and habitats in Europe and in
Morocco, northern Africa. Results of initial host-specificity testing with the prioritised
putative specialised natural enemies, on a subset of key closely related species to
S. oleraceus, are also presented. Based on the results of these investigations, we discuss
the feasibility of developing classical biological control solutions for S. oleraceus within
the Australian context.
BIOCONTROL SCIENCE AND TECHNOLOGY 3

Materials and methods
Survey area
Native range field surveys for natural enemies of S. oleraceus were conducted between
March 2017 and March 2020, mainly in spring and autumn. The selection of areas sur-
veyed was based on bioclimatic models developed in parallel that identified northern
Africa and southern Europe as most climatically similar to regions where S. oleraceus
is a problem in Australia (Kriticos et al., in press). A genetic analysis also identified
this broad region as one of the likely sources of S. oleraceus populations in Australia
(Encinas-Viso et al., in press). Consequently, candidate agents collected from this
region were likely to be well adapted to utilise S. oleraceus in Australia. Surveys were
thus concentrated in southern Portugal, southern Spain, southern France, northern
Italy, the Balkans and Greece. Surveys were also performed in western Morocco and
the Canary Islands, in the Macaronesian region offthe west coast of northern Africa.
This archipelago displays a high diversity of the genus Sonchus, with several endemic
species (Kim et al., 1999; Kim et al., 2007). In addition, short surveys were conducted
in northern France, Belgium, the Netherlands and Germany. Overall, 122 sites were sur-
veyed (Figure 1).
Some sites were visited more than once across the years, especially in southern Spain,
Morocco and within the vicinity of Montpellier, southern France. In sites near Montpel-
lier, surveys were conducted regularly during the year to gather phenology data on the
natural enemies present on S. oleraceus.
Field collections and sample processing
Sonchus oleraceus grows predominantly in disturbed areas. Therefore, only a small
number of collections of natural enemies or arthropod-damaged/diseased plants were
made in relatively natural and undisturbed habitats. Collections were primarily per-
formed in ruderal areas (e.g. roadsides, uncultivated fields) and within cropping
systems (e.g. cultivated fields, fallows after harvesting, field margins). Autumn surveys
focused on the rosette stage to collect natural enemies that attack plants before
flowering while all plant stages were surveyed in spring-summer. Time spent searching
for natural enemies and damaged plants varied from a few minutes to 1–2 h depending
on the number of plants present at the site. Wherever possible, collections were also made
on the closely related species, S. asper and Sonchus tenerrimus L.
Arthropods
At each site, plants of different sizes and growth stages (rosette, flowering, seeding) were
inspected for arthropods on both above- and below-ground parts. Adults found on plants
were collected and placed in 95% ethanol in a vial. Plants were transported to the labora-
tory for dissection of stems and roots to carefully inspect for the presence of feeding
insects. All insects found at larval stages were placed in screened cages with potted
S. oleraceus plants or into Petri dishes with fresh plant material of S. oleraceus within con-
trolled-temperature (CT) rooms at 19–25°C, until they reached the adult stage for
identification. In some instances, dug-up plants from the surveyed sites were trans-
planted into potting mixture in pots and placed in screened cages in CT rooms to
4V. LESIEUR ET AL.

minimise disturbance for the development of larvae and maximise chances of adult
emergence.
Fungi
Plants were examined for disease symptoms in the field and samples of diseased tissues
were collected and placed in an herbarium press between layers of absorbent paper,
which was regularly changed. Wherever possible, one or more accessions of each bio-
trophic fungus found was purified by initiating a culture on S. oleraceus plants with
spores from a single sorus for the rust fungi and from one sporulating leaf spot for
downy and powdery mildew fungi. Inoculated plants were exposed to the same con-
ditions as those outlined in the host-specificity testing section below. Spores from
these cultures were collected and either directly used in the tests or stored in vials at –
80°C. Wherever necessary, several cycles of multiplication of the fungi, through repeated
inoculations of plants, were undertaken to obtain sufficient spores for DNA extraction
and host-specificity tests.
Leaf or stem material with well-defined necrotic lesions were placed onto moist filter
paper contained in 10-cm glass Petri dishes in a CT room with temperature ranging
between 19°C and 23°C and a 12-h photoperiod (LED lights) for 2–3 days. The material
was examined for fungal fruiting bodies at regular intervals with a stereoscopic micro-
scope. Fungi were tentatively identified to the genus level with morphological characters,
and only genera that comprise plant pathogenic species were isolated. For one or more
accessions of each fungus, including accessions from different sites, several conidia were
removed with a sterile needle and placed on a film of solidified 2% water agar (WA) on a
microscopic slide contained in a Petri dish, which was incubated for 18‒24 h in the CT
room (same conditions as above). A piece of agar with a single germinated conidium was
removed aseptically with a sterile scalpel and placed on the surface of potato dextrose
agar (PDA) contained in a 10-cm plastic Petri dish. Plates were incubated (as above)
until colonies were well developed. A colony of each single-conidium isolate on PDA
was cut into small pieces which were stored in vials in sterile saline solution at 4°C
and in a milk–glycerol solution at –80°C.
The pathogenicity of the fungal isolates recovered from necrotic symptoms was tested
by inoculating healthy S. oleraceus plants following the methods detailed below for host-
specificity testing. In instances where the inoculated plants showed symptoms identical to
those observed in the field, the causal agent was isolated using the method above and
identified morphologically. The Koch’s postulates were fulfilled if the causal agent corre-
sponded to the isolate used for inoculation (Agrios, 1997; Berner & Bruckart, 2005).
Identification of natural enemies
Arthropods
Identifications were obtained by a combination of morphological and DNA-based mol-
ecular approaches (see Table 1). Expert taxonomists for different insect groups were con-
sulted for morphological identifications of adults collected in the field or emerged from
the collected material. We identified most of the sampled taxa by using both morphologi-
cal characters and molecular barcode. However, in cases where no adult emerged, thus
limiting capacity to perform morphological identifications, identifications relied on
BIOCONTROL SCIENCE AND TECHNOLOGY 5

Table 1. Phytophagous arthropod species collected on Sonchus oleraceus during the surveys.
Order/Family Species Frequency
a
Feeding site on
plant
b
Type of damage
Damaging
stage
c
Recorded host
range
d
Identification
e
Acari
Tetranychidae
Tetranychus urticae (Koch, 1836) C FB, FH, L, S Chewing L, A Polyphagous Morph.
Thysanoptera
Thripidae Limothrips cerealium (Haliday, 1836) R L, F Sucking L, A Polyphagous Morph.
Tenothrips frici (Uzel, 1895) C L, F Sucking L, A Oligophagous Both
Hemiptera
Aphididae Hyperomyzus lactucae (Linnaeus 1758) C FB, FH, L, S Sucking L, A Polyphagous Both
Pemphigus bursarius (Linnaeus, 1758) O R Sucking L, A Polyphagous Morph.
Uroleucon sonchi (Linnaeus, 1758) C FB, FH, L, S Sucking L, A Oligophagous Both
Aleyrodidae Aleyrodes proletella (Linnaeus, 1758) C L Sucking L, A Polyphagous Morph.
Bemisia tabaci (Gennadius, 1889) C L Sucking L, A Polyphagous Morph.
Aphrophoridae Philaenus spumarius (Linnaeus, 1758) C L, S Sucking L, A Polyphagous Both
Coccidae Coccus hesperidum (Linnaeus, 1758) R S Sucking L, A Polyphagous Morph.
Parthenolecanium corni (Bouché, 1844) R S Sucking A Polyphagous Morph.
Margarodidae Icerya purchasi (Maskell, 1878) O L, S Sucking L, A Polyphagous Morph.
Pseudococcidae Vryburgia brevicruris (McKenzie, 1960) O R Sucking L, A Polyphagous Morph.
Heteroptera
Coreidae Coreus marginatus (Linnaeus, 1758) R FB, FH, S Sucking L, A Polyphagous Morph.
Lygaeidae Nysius cymoides(Spinola, 1837) C FB, FH Sucking L, A Polyphagous Both
Lygaeus equestris (Linnaeus, 1758) C FB, FH, S Sucking L, A Polyphagous Morph.
Spilostethus pandurus (Scopoli, 1763) C FB, FH, S Sucking L, A Polyphagous Morph.
Pentatomidae Carpocoris mediterraneus (Tamanini, 1958) O FB, FH, S Sucking L, A Polyphagous Morph.
Dolycoris baccarum (Linnaeus, 1758) C FB, FH, S Sucking L, A Polyphagous Both
Odontotarsus robustus (Jakovlev, 1884) R FB, FH, S Sucking L, A Oligophagous Morph.
Rhopalidae Corizus hyoscyami (Linnaeus, 1758) C FB, FH Sucking L, A Polyphagous Morph.
Liorhyssus hyalinus (Fabricius, 1784) C FB, FH Sucking L, A Polyphagous Both
Stictopleurus punctatonervosus (Goeze, 1778) C FB, FH Sucking L, A Oligophagous Morph.
Hymenoptera
Cynipidae sp. R S Gall-making L NA Morph.
Tenthredinidae Cephaledo neobesa (Zombori, 1980) O L Chewing L Oligophagous Morph.
Coleoptera
Curculionidae Dichromacalles diocletianus (Germar, 1817) R R, S Boring L Polyphagous Both
Dichromacalles dromedarius (Boheman, 1844) R L, S Boring L, A Polyphagous Both
Lixus punctiventris(Boheman, 1835) C L, S Boring L, A Oligophagous Both
Cetoniidae Oxythyrea funesta (Poda, 1761) O F Chewing A Polyphagous Morph.
Cerambicydae Agapanthia villosoviridescens (DeGeer, 1775) O S Boring / Chewing L, A Polyphagous Both
Phalacridae Tolyphus sp. O* FB, FH Seed-feeding L NA Both
6V. LESIEUR ET AL.

Melolonthidae Hoplia africana (Escalera, 1914) O* F Chewing A Polyphagous Morph.
Nitidulidae Brassicogethes aeneus (Fabricius, 1775) C F Chewing A Polyphagous Both
Lepidoptera
Erebidae Eublemma cochylioides (Guenée, 1852) O FB, FH Chewing L Oligophagous Morph.
Geometridae Gymnoscelis rufifasciata (Haworth, 1809) R†FB, FH, L Chewing L Polyphagous Mol.
Noctuidae Cornutiplusia circumflexa (Linnaeus, 1767) C†FB, FH, L Chewing L Polyphagous Mol.
Hecatera dysodea (Denis & Schiffermüller,
1775)
C FB, FH Chewing L Oligophagous Both
Helicoverpa armigera (Hübner, 1808) R FB, FH Chewing L Polyphagous Both
Mamestra brassicae (Linnaeus, 1758) R FB, FH, L Chewing L Polyphagous Both
Spodoptera exigua (Hübner, 1808) R L Chewing L Polyphagous Morph.
Trichoplusia ni (Hübner, 1803) O FB, FH, L Chewing L Polyphagous Both
Thysanoplusia orichalcea (Fabricius, 1775) O†FB, FH, L Chewing L Polyphagous Mol.
Tortricidae Acroclita sonchana (Walsingham, 1908)O†L, S Mining / Chewing L Specialist Both
Acroclita subsequana (Herrich-Schäffer, 1851) O* L, S Mining / Chewing L Polyphagous Both
Cacoecimorpha pronubana (Hübner, 1799) C L, S Mining / Chewing L Polyphagous Both
Clepsis pallidana (Fabricius, 1776) R L, S Mining / Chewing L Polyphagous Both
Diptera
Agromyzidae Chromatomyia horticola (Goureau 1851) C L Mining L Polyphagous Both
Liriomyza sonchi (Hendel, 1931) O L Mining L Oligophagous Morph.
Liriomyza strigata (Meigen, 1830) O L Mining L Polyphagous Morph.
Ophiomyia cunctata (Hendel, 1920) C L Mining L Oligophagous Both
Napomyza lateralis (Fallén, 1823) C S Mining L Oligophagous Both
Nemorimyza maculosa (Malloch, 1913) O†L Mining L Oligophagous Both
Cecidomyidae Cystiphora sonchi (Vallot, 1827) C L Gall-making L Specialist Both
Drosophilidae Gitona distigma (Meigen, 1830) O FB, FH Seed-feeding L Polyphagous Morph.
Syrphidae Cheilosia latifrons (Zetterstedt, 1843) O R, S Mining L Unknown Both
Tephritidae Campiglossa producta (Loew, 1844) C†FB, FH Gall / Seed-
feeding
L Oligophagous Both
Ensina sonchi (Linnaeus, 1767) C FB, FH Seed-feeding L Oligophagous Both
Tephritis formosa (Loew, 1844) C FB, FH Gall / Seed-
feeding
L Oligophagous § Both
Trypanea amoena (Frauenfeld, 1857) R FB, FH Seed-feeding L Oligophagous Morph.
a
C = common (30 or more individuals collected, and present at five or more sites); O = occasional (5–29 individuals collected, or present at fewer than five sites); R = rare (fewer than five indi-
viduals collected). †Only collected in Canary Islands; * only collected in Morocco.
b
R = roots; S = stem; L = leaf; FB = flower bud; F = flower; FH = flower head.
c
A = Adult; L = Larva.
d
Polyphagous = Generalists; Oligophagous = Restricted to Asteraceae; Specialist = Restricted to Sonchus spp. § doubtful, potentially specialist to Sonchus spp.
e
Morph. = morphological identification; Mol. = molecular identification; Both = morphological and molecular identifications.
BIOCONTROL SCIENCE AND TECHNOLOGY 7

DNA sequencing. Conversely, only morphological identification was sometimes per-
formed due to the low number of specimens collected, and/or the low quality of DNA
extracted.
For identification based on DNA sequencing, total genomic DNA was extracted from
the entire insect (except for adult Lepidoptera for which only one leg was used) using
DNeasy
®
Blood & Tissue kit (Qiagen, Hilden, Germany) following the manufacturer’s
protocol without destruction of the specimens, to allow subsequent examination of mor-
phology if necessary. The DNA of cytochrome c oxidase subunit I (COI) was amplified
using the universal primer cocktail designed by Germain et al. (2013) and Cruaud et al.
(2010), from the classic primers designed by Folmer et al. (1994). Additional details on
primers and PCR conditions can be found in the Supplementary Material. The purified
PCR products were sequenced in both directions by Eurofins Genomics (https://www.
eurofinsgenomics.eu/). Consensus sequences were generated from forward and reverse
DNA sequence reads and manually edited with Geneious 10.2.3 (https://www.
geneious.com). Consensus sequences were compared to those in GenBank (https://
www.ncbi.nlm.nih.gov/genbank/) and the Barcode Of Life Database (BOLD, http://
www.boldsystems.org/index.php). The analysed sequences were deposited in GenBank
and accession numbers are provided in Supplementary Material (Table S4). DNA and
specimens are held in frozen and dried collections, respectively, at the Centre de Biologie
pour la Gestion des Populations (CBGP), Montferrier sur Lez, France.
Fungi
One or more purified isolates of the biotrophic fungi and all fungal isolates recovered
from necrotic lesions for which Koch’s postulates were fulfilled during pathogenicity
tests were sequenced. DNA sequencing was performed to confirm, where applicable, ten-
tative morphological identifications to the genus level previously obtained by examining
vouchers of diseased material or cultures on agar. Spores from isolates of the biotrophic
fungi stored at –80°C and mycelium from actively growing cultures of the facultative
parasitic fungi on PDA plates were used for DNA extraction. DNA was also extracted
from traces of spores recovered from a few leaves of a single S. oleraceus plant infected
by a rust fungus at the aecial stage in the field.
Total genomic DNA of the fungi was extracted using FastPrep-24 5G and DNeasy
®
Plant Mini Kit (Qiagen) following the manufacturers’protocols. A set of several
primers to amplify the full or part of the nuclear internal transcribed spacer (ITS)
region (ITS1, 5.8S, ITS2), the nuclear ribosomal large subunit (LSU) of the nuclear ribo-
somal DNA or the gene region BrRxLR11 was used depending on the fungus. Details on
primers used and PCR conditions are provided in Supplementary Material. The sequen-
cing of both strands was performed by Genoscreen (https://www.genoscreen.fr)or
ADNid (http://www.adnid.fr/). DNA sequences were manually edited and aligned with
Chromas lite v.2.6.6 and Clustal omega programs (Madeira et al., 2019) and a consensus
sequence were established. The Mega Basic Local Alignment Search Tool (MegaBLAST)
of NCBI was used to retrieve from GenBank highly similar sequences to those of the
different fungal isolates. DNA sequences were deposited in GenBank (Table 2). Dried
specimens or frozen cultures of each fungal/isolate, except the unpurified rust fungus
at the aecial stage for which no material was available, are held at the CSIRO European
Laboratory (Montferrier sur Lez, France).
8V. LESIEUR ET AL.

Selection of promising candidates
Arthropods
The herbivorous arthropods found on S. oleraceus during surveys were prioritised for
initial host-specificity testing based on their potential host range, information available
in the literature and the damage observed on S. oleraceus in the field.
The fruit flyTephritis formosa Loew (Diptera: Tephritidae) was prioritised because it is
considered specific to the genus Sonchus in the literature (Berube, 1978; White, 1988).
Another fruit fly, Campiglossa producta Loew (Diptera: Tephritidae) that is known to
develop on a wide range of Cichorieae, including Sonchus spp. in Europe (mainland)
(Namin & Nozari, 2015; White, 1988), was prioritised. This fly was only found in Tener-
ife Island (Canary Islands, Spain), despite our extensive sampling of S. oleraceus capitula
in Morocco and mainland Europe. An endemic fauna and flora occur on these islands
(Kunkel, 1976). This population of the fly found in the Canary Islands might be a
cryptic species, or a divergent population with a different behaviour to that of popu-
lations in the European mainland, justifying the prioritisation of this population as a can-
didate agent for further testing. Due to the lack of information in the literature on the
host range of the syrphid fly, Cheilosia latifrons Zetterstedt (Diptera: Syrphidae)
(Reemer & Claussen, 2008; Speight, 2011), it was also prioritised for initial testing. Cysti-
phora sonchi (Diptera: Cecidomyiidae) was prioritised due to its reported specificity to
the genus Sonchus, however, results of preliminary host-specificity testing with this
species have already been published by Lesieur et al. (2020) and are not included in
this paper.
Fungi
The selection strategy employed for fungi was slightly different. Initial host-specificity
tests were performed with representative isolate(s) of the fungi for which pathogenicity
was confirmed on S. oleraceus in laboratory tests (i.e. Koch’s postulates fulfilled), and the
biotrophic fungi purified and cultured on S. oleraceus,except the rust fungus Miyagia
pseudosphaeria and powdery mildew fungus. These fungi were not tested because the
former is already present and widespread across Australia (McCarren & Scott, 2017)
and the latter belongs to a group of fungi that has never been used in classical biological
control programmes due to their general lack of specificity. It is noteworthy that for some
of the fungi, tests were performed before their identity was confirmed with DNA sequen-
cing. The pathogens tested were two biotrophic fungi (downy mildew and rust species),
and four facultative parasitic fungi causing necrotic leaf lesions on S. oleraceus.
Initial testing for host-specificity
Prioritised candidate agents were tested on the target weed S. oleraceus (positive control)
and one or two closely related species in the subtribe Sonchinae that are native to Aus-
tralia: S. hydrophilus and A. megalocarpus. Seeds of S. oleraceus accessions from France
(Montferrier sur Lez) and/or Australia (Canberra) were used, while S. hydrophilus and
A. megalocarpus were grown from seeds collected at Wilmington (South Australia)
and Augusta (Western Australia), respectively. Seeds were sown in standard horticultural
grade compost (Neuhaus Humin Substrat N6; Klasmann-Deilmann GmbH, Geeste,
BIOCONTROL SCIENCE AND TECHNOLOGY 9

Germany) contained in free-draining pots (9 × 9 cm wide and 10 cm deep) and depend-
ing on the tests, plants were transplanted into larger pots (3 l). Plants were maintained in
a non-heated glasshouse with natural light. Plants at the rosette stage (1–2-months after
planting) were used in tests with fungi while plants at different growth stages were used
for tests with arthropods depending on the biology of the candidate agent selected.
Arthropods
The three prioritised candidate agents tested have a different biology, therefore, only
general statements on the methodology used are made here. More details on the
methods used with each species are provided in the Supplementary Material. Both no-
choice and choice tests were performed. Tests with C. latifrons included both non-
target plant species, while T. formosa and C. producta were only tested on
A. megalocarpus because of unavailability of flowering S. hydrophilus plants at that time.
Tests were conducted under laboratory conditions using potted plants and insects
sourced from established laboratory colonies, or under semi-natural conditions using
plants transplanted in an outdoor garden and exposed to a naturally occurring popu-
lation of the insects. All tests, except for C. latifrons, were performed by exposing
plants to sexually mature adults to evaluate oviposition and subsequent larval develop-
ment. The genus Cheilosia is difficult to rear in captivity (Grosskopf et al., 2002); only
unfertilised eggs of C. latifrons could be obtained from different trials performed and
it was thus not possible to initiate a laboratory rearing colony. Therefore, to overcome
this issue, no-choice tests were performed by transferring field-collected eggs on plants
to evaluate the capacity of C. latifrons to complete larval development in the test plant
species. Choice tests with this candidate agent were conducted with adults obtained
from field-collected larvae to estimate the oviposition preferences of female
C. latifrons. Depending on the insect species tested, no-choice tests and choice tests con-
sisted of 3–6 replicates and 5–18 replicates, respectively (Supplementary Material).
Tephritis formosa and C. latifrons originated from the vicinity of Montferrier sur Lez
(43.68429° N; 3.87581° E, France), while C. producta was sourced from a laboratory
colony initiated with naturally infested flower heads of S. oleraceus collected at Icod de
los Vinos (28.37425° N; –16.67754° E, Canary Islands, Spain). Consequently, all tests
conducted with C. producta were performed in a containment laboratory (CSIRO –
European Laboratory, Montferrier sur Lez, France).
Depending on the biology of the candidate agents, relative performance on test plants
was measured as infestation rate and/or survival rate. Infestation rates were expressed as:
the percentage of infested capitula for T. formosa and C. producta and the total number of
eggs laid per plant for C. latifrons (only in choice test). Survival rates were expressed as:
the percentage of adults emerged/total number of individuals produced (i.e. dead larvae,
dead pupae and adults emerged) per capitula for T. formosa and C. producta and the per-
centage of successful development from eggs to the pupal stage (assessment only possible
in no-choice test) for C. latifrons. Statistical analyses were conducted using R version
3.5.3 (R Core Team, 2018). Infestation and survival rates were modelled as a binomial
response variable or as a Quasi-Poisson distribution, using a generalised linear model
(GLM) with a logit link function, with test plant species as a fixed effect, except for the
survival rate of C. latifrons for which the effect of plant species was assessed by Fisher
10 V. LESIEUR ET AL.

exact test. Additional details on statistical analyses can be found in the Supplementary
Material.
Fungi
One or two isolates of each prioritised fungus were included in different tests, each com-
prising three replicates of each of the plant species (with some exceptions where one
replicate of one of the two S. oleraceus accessions, and 4 or 8 replicates of the non-
target species were used). For biotrophic fungi, fresh spores were collected from uredinia
or sporulating lesions on the cultures maintained on S. oleraceus plants in the laboratory.
Urediniospores were mixed with talcum powder (1:19 ratio) and applied with a fine hair
paint brush onto the adaxial surface of fully expanded leaves of the test plants. Conidia
were suspended in a solution of 0.1% Tween
®
80 (Merck Schuchardt, Hohenbrunn,
Germany) in sterile demineralised water and sprayed until run-offonto fully expanded
leaves of each plant using a hand-held sprayer.
Each facultative parasitic fungus was grown on PDA plates by transferring colonised
pieces of agar from the stock culture in saline solution and incubating plates in a CT
room (conditions as above). Once plates were fully colonised by the fungus, a suspension
of conidia and/or mycelial fragments was prepared by (i) scraping the surface with a
sterile scalpel to produce a slurry, (ii) gently crushing the slurry with a pestle and
mortar, (iii) adding sufficient amount of the water Tween
®
80 salution to dilute the
slurry and (iv) filtering through two layers of cheesecloth. The suspension of conidial
and/or mycelial fragments was applied with a fine hair paint brush onto the adaxial
and abaxial surface of fully expanded leaves.
To verify the viability of the inoculum used, a microscope slide coated with WA was
dusted with the spore-talcum powder mixture or brushed with the conidial and/or myce-
lial fragments suspension and placed in a Petri dish within the same CT room as inocu-
lated plants. Germination of spores and/or mycelial fragments was assessed after 24 h
using a light microscope.
Inoculated plants (also sprayed with sterile demineralised water when inoculated
with spore-talcum powder mixture) were placed in humid chambers (lidded clear
plastic boxes containing water to a depth of 2–3 cm) within CT rooms with a 12-h
photoperiod for 1 (rust fungus), 2–3(downymildewfungus)or1–4 (facultative para-
sitic fungi) days. The temperature in the CT rooms containing the humid chambers
fluctuated between 17°C and 18°C (downy mildew fungus) or 19°C and 23°C (rust
fungus and facultative parasitic fungi). Once removed from the humid chambers,
plants were placed on the bench of another CT room (12-h photoperiod; 18–26°C).
Five days later, plants inoculated with conidia from the downy mildew fungus were
transferred back to the humid chambers in the original CT room until the end of
the test to encourage symptoms development and induce sporulation. All inoculated
plants were regularly examined over a 5‒6 weeks period and disease symptoms
recorded.
To confirm that the facultative parasitic fungi inoculated on plants were the cause of
the necrotic lesions observed, leaves and/or pieces of stem with lesions were excised and
placed on moist sterile filter paper contained in 10-cm glass Petri dishes. Dishes were
incubated in a CT room (12-h photoperiod; 18–26°C) until sporulation was observed
on the lesions. Fruiting bodies and conidia were mounted on microscope glass slides
BIOCONTROL SCIENCE AND TECHNOLOGY 11

with acetic blue and examined with a light microscope to confirm the identity of the
fungus.
Results
Natural enemy surveys and identification
Arthropods
A total of 59 phytophagous arthropod species were collected on S. oleraceus during the
surveys (Table 1). Only one mite species was collected, while insects from seven different
orders were recovered, with the highest representation from the orders Diptera (5
families, 13 species) and Lepidoptera (4 families, 13 species). Most of the species collected
are polyphagous or oligophagous (n= 38, 65% and n= 16, 27%, respectively), and only
few are recorded as specialist on the genus Sonchus (n= 2, 3%). The host range of
three insects collected could not be determined because they could not be identified to
species level or due to the lack of information in the published literature. Most of the
insects found are ectophagous sucking or chewing species (n= 22, 37% and n= 19,
32% respectively), while the endophagous guild is dominated by mining insects (n=
11, 19%). The most damaged part of the plant was the flower heads, with 39% of the
total number of species found suspected to use these as feeding resource. No severe
damage was observed on roots and only a few species were suspected to be associated
with this feeding niche.
Of the four species prioritised for testing, T. formosa was commonly found in infested
capitula of S. oleraceus and S. asper across most areas surveyed in mainland Europe (38
sites) but was not recovered in the Canary Islands and Morocco. This species was
common in spring. Campiglossa producta, another fruit fly infesting capitula, was only
collected in four sites in the Canary Islands on S. oleraceus. The haphazard sampling
of S. oleraceus capitula performed in the islands indicated that the infestation rates
were quite high (data not shown). Cheilosia latifrons, with larvae that mined the stems
and root-crowns, was commonly found around Montpellier (France) (11 sites) in early
spring and autumn on both S. oleraceus and S. asper. The species was also collected in
one site in northern Spain on S. oleraceus and S.tenerrimus (only one plant infested
for the latter). Cystiphora sonchi, which caused galls on leaves, was widespread across
the sites surveyed (21 sites) and was found infesting S. oleraceus,S. asper and S.
tenerrimus.
Fungi
The rust fungus M. pseudosphaeria was commonly observed across the sites surveyed and
easily identified morphologically by its characteristic row of brown paraphyses surround-
ing each uredinium (Gäumann, 1959; Wilson & Henderson, 1966)(Table 2). The
sequence obtained for the ITS2-LSU region (1410 bp) of the representative isolate, col-
lected in the Canary Islands (Spain), was identical to several sequences of
M. pseudosphaeria collected on S. oleraceus in different countries deposited in
GenBank (Supplementary Material, Table S5).
The powdery mildew fungus on S. oleraceus was also common across sites, but tenta-
tive identification with morphological characters could not be performed because the
12 V. LESIEUR ET AL.

Table 2. Pathogenic fungi recovered from disease symptoms on Sonchus oleraceus during the surveys.
Order/Family Name
a
Symptoms on
leaves Frequency
b
Isolate source locality
c
GenBank no.
Closest sequence match in GenBank (%
similarity)
d
Capnodiales
Mycosphaerellaceae Ramularia sp. Necrotic lesions R Anaktorio, Greece MW666029 Ramularia helminthiae (100%)
Septoria sp. Necrotic lesions O Montferrier-sur-Lez, France MW666030 Septoria sonchi (99.80%)
Triadou, France MW666031
Erysiphales
Erysiphaceae Golovinomyces sp. Powdery
mildew
C Baillarguet, France MW666032 Golovinomyces sonchicola (100%);
Golovinomyces cichoracearum (100%)
Peronosporales
Peronosporaceae Bremia sp. Downy mildew O Saint Gély du Fesc, France MW666049,
MW654507
Bremia sonchicola (DC6/ITS2, 100%; BrRxLR11-
F/ BrRxLR11-R, 100%)
Triadou, France MW666050
MW654508
Pleosporales
Didymellaceae Didymella sp.
(=Phoma-like)
Necrotic lesions O Itea, Greece MW666033 Didymella rosea (100%)
Brignoles, France MW666034
Mauguio, France MW666035
Quissac, France MW666036
La Crau 2018, France MW666037
La Crau 2019, France MW666038
Rochessadoule, France MW666039
Pleosporaceae Alternaria sp. Necrotic lesions C San Cristóbal de La Laguna, Canary
Islands, Spain
MW666040 Alternaria sonchi (100%)
Livadia, Greece MW666041
Guérande, France MW666042
Quissac, France MW666043 Alternaria alternata (100%); Alternaria
tenuissima (100%)
(Continued)
BIOCONTROL SCIENCE AND TECHNOLOGY 13

Table 2. Continued.
Order/Family Name
a
Symptoms on
leaves Frequency
b
Isolate source locality
c
GenBank no.
Closest sequence match in GenBank (%
similarity)
d
Pucciniales
Coleosporiaceae Coleosporium sp. Rust O ’s-Hertogenbosch, Netherlands MW666044 Coleosporium sonchi (99.68-100%)
Sainte Etienne de Montluc, France MW666045
Le Haut Brivin, France MW666046
La Baule, France MW666047
Tremeac, France MW666048
Pucciniaceae Miyagia
pseudosphaeria
e
Rust C Tacoronte, Canary Islands, Spain MW666051 Miyagia pseudosphaeria (100%)
Aecidium sp. (aecial
stage)
Rust R Corsica, France MW666052 Uromyces junci (99.42%)
a
Name given based on morphological identification and/or sequencing, except for the rust fungus at the aecial stage for which insufficient information was available to give it a name with some
certainty.
b
Number of sites where the fungal species was found: C = common (> 30 sites); O = occasional (5–30 sites); R = rare (< 5 sites).
c
Isolates from the localities in bold were used in host-specificity tests (Table 4).
d
Based on sequences retrieved from GenBank using MegaBLAST of NCBI (see Supplementary Material).
e
Miyagia is a small genus distinguishable with its palisade-like paraphyses in its sori and only one species M. pseudosphaeria has ever been reported on Sonchus.
14 V. LESIEUR ET AL.

sexual stage was not present. The ITS sequence (591 bp) of the representative isolate,
which was collected on S. oleraceus in southern France, was identical to three sequences
of Golovinomyces sonchicola U. Braun & R.T.A. Cook from S. oleraceus or S. asper and to
two sequences of Golovinomyces cichoracearum (DC.) V.P. Heluta from S. oleraceus or
Rosa gallica L. in GenBank (Supplementary Material, Table S5).
A second rust fungus at the uredinial stage, but with uredinia morphologically
different to those of M. pseudosphaeria, was only found at one site in The Netherlands
and four sites in western France (Brittany region). The ITS sequences (626–669 bp) of
purified isolates from these sites were most similar to two sequences of Coleosporium
sonchi Lev. (current name Coleosporium tussilaginis (Pers.) Lév.) in GenBank, each col-
lected on S. asper and S. arvensis (Supplementary Material, Table S5).
A rust fungus at the aecial stage, which is not informative for morphological identifi-
cation of rust fungi, was only found at a single site, in Corsica (France) (Table 2). The
ITS2-LSU region was successfully sequenced (1390 bp) and a part of this sequence
(516 bp) was found to be 99.42% similar to a GenBank sequence of the heteroecious
rust fungus Uromyces junci (Desm.) Tul. (Supplementary Material, Table S5).
A downy mildew fungus with morphological characteristics corresponding to the
genus Bremia was occasionally found during the surveys. The two isolates of the
fungus sequenced, both collected in southern France, had similar sequences for the
two markers used; ITS1 (630 and 644 bp) and the BrRxLR11 gene (575 and 578 bp)
(Table 2). These ITS1 and BrRxLR11 sequences were identical to sequences of Bremia
sonchicola (Schltdl.) Sawada from S. oleraceus, S. asper and/or S. arvensis lodged in
GenBank (Supplementary Material, Table S5).
Of the 31 isolates of different fungi recovered from necrotic lesions and tested for
pathogenicity on S. oleraceus,Koch’s postulates were fulfilled for 14 isolates, belong-
ing to 4 different genera (Table 2). Alternaria sp. was easily recognisable with mor-
phological characters and found at more than 30 sites surveyed. The ITS sequences
(565 bp) of three Alternaria sp. isolates collected in Greece, western France and the
Canary Islands were identical to each other and to a sequence of Alternaria sonchi
Davis from S. asper in GenBank (Supplementary Material, Table S5). The ITS
sequence (566 bp) of a fourth Alternaria sp. isolate recovered from S. oleraceus at a
site in southern France, with smaller spores than the other three isolates, was identical
to four GenBank sequences of Alternaria alternata (Fr.) Keissl. recovered from
Festuca roemeri (Pavlick) E.B. Alexeev or Panax ginseng C.A.Mey and to one sequence
of Alternaria tenuissima (Kunze) Wiltshire from Rubus idaeus L. (Supplementary
Material, Table S5).
Two other facultative parasitic fungi morphologically identified to the genus level, Sep-
toria sp. and Ramularia sp., were occasionally and rarely found during the surveys,
respectively. The two representative isolates of Septoria sp. sequenced, collected in
southern France, had identical ITS sequences (536 bp) which were highly similar
(99.80%) to sequences of Septoria sonchi Sacc. collected on S. asper available in
GenBank (Supplementary Material, Table S5).The ITS sequence (531 bp) of the single
Ramularia sp. isolates recovered during our survey, from a site in Greece, was identical
to a GenBank sequence of Ramularia helminthiae Bremer & Petr. isolated from Picris
echioides L. (Supplementary Material, Table S5).
BIOCONTROL SCIENCE AND TECHNOLOGY 15

Seven other isolates with a Phoma-like morphology recovered from necrotic lesions
were found at several sites across two countries (France, Greece) during our surveys
(Table 2). The ITS sequences (537 bp) of these isolates were identical to each other
and to a sequence of a Didymella rosea from S. arvensis deposited in GenBank
(Supplementary Material, Table S5).
Initial testing for host-specificity
Arthropods
Tephritis formosa was observed to mate and lay eggs on both S. oleraceus and
A. megalocarpus in choice and no-choice tests. The larvae produced similar button-
shaped galls in capitula of each species that prevented flowers opening (Figure 2). On
Figure 2. Tephritis formosa. (a) Adult (female); (b) Infested flower head of Sonchus oleraceus (black
arrow); (c) Infested flower head of Actites megalocarpus (black arrow); and (d) Results of larval
feeding of Tephritis formosa in Sonchus oleraceus flower head (larva and pupa are present in the
flower head).
16 V. LESIEUR ET AL.

both plant species, larvae destroyed all seeds produced within the capitula. In the no-
choice test, the infestation rate (i.e. mean percentage of infested capitula per plant)
observed was highest on S. oleraceus (Table 3). The survival rate of the fly was high
for both species in that test but it was statistically lower on S. oleraceus.
In the choice test performed in outdoor conditions, T. formosa adults were observed
resting or laying eggs on both plant species. The infestation rate on A. megalocarpus
was low and significantly lower than on S. oleraceus (Table 3). The infestation rates
on plants during the first series of tests (χ
2
= 12.929; df =1;p< .001, see Supplementary
Material) were higher than in the second series, while the survival rates of the insect
were constant across the two series (χ
2
=1.601; df =1; p= .205) and significantly
higher on S. oleraceus.
As observed with T. formosa,C. producta mated and laid eggs on both S. oleraceus
and A. megalocarpus during both tests and larvae produced similar button-shaped
galls in the capitula (Figure 3) and consumed seeds. In the no-choice test, infestation
rate was significantly higher on S. oleraceus than on A. megalocarpus,althoughitwas
still high on the latter (Table 3). Survival rate was high and equivalent on both plant
species.
In the choice test, C. producta infestation rates were high on both plant species, but the
insect infested significantly more capitula of S. oleraceus than of A.megalocarpus
(Table 3). There was no statistical difference in survival rates on each of the plant species.
In the no-choice test, larvae of C. latifrons were observed tunnelling the stem
(Figure 4) and completing their development on all plant species tested (S. oleraceus,
S. hydrophilus and A. megalocarpus). There was no statistical difference in survival
rates of the insect on the different species (Table 3).
During the oviposition choice test, females laid their eggs on all growth stages of the
three plant species. However, as observed in different trials to attempt to initiate a lab-
oratory colony, none of the eggs hatched, or only unfertilised eggs were laid. No prefer-
ence for laying was observed for a specific growth plant stage (F= 2.605; df =3;p= .104)
or species (Table 3).
Fungi
The inoculum of each of the fungal/isolates used in the host-specificity tests readily
germinated on the film of WA on microscope slides used to confirm viability. Both
accessions of S. oleraceus and the two non-target native species inoculated with the
various isolates developed disease symptoms during the tests (Table 4). In most cases,
symptoms on A. megalocarpus and S. hydrophilus were like those observed on
S. oleraceus (Figure 5). Only slight differences in disease incidence (expressed as the per-
centage of inoculated leaves that became infected) were observed between
A. megalocarpus and S. hydrophilus.Sonchus hydrophilus was generally more susceptible
than A. megalocarpus, to the Septoria sp., Didymella sp. and Alternaria sp. isolates,
although for both plant species several leaves inoculated with the latter two fungi did
not develop symptoms. Fewer leaves of the French accession of S. oleraceus inoculated
with Ramularia sp. or Didymella sp. developed disease symptoms compared to those
of the Australian accession. All facultative parasitic fungi were reisolated from the necro-
tic lesions that developed on the different species by the end of the tests.
BIOCONTROL SCIENCE AND TECHNOLOGY 17

Table 3. Results of host-specificity tests performed with the prioritized arthropod candidate agents on the target weed Sonchus oleraceus and non-target species
native to Australia, Sonchus hydrophilus and Actites megalocarpus. Results for the leaf-gall midge, Cystiphora sonchi are reported in Lesieur et al. (2020).
Candidate agent Plant species
Infestation (±SE)
a
Survival (±SE)
b
Choice test Statistic; p-valueNo-choice test Statistic; p-value Choice test Statistic; p-value
No-choice
test Statistic; p-value
Tephritis formosa S. oleraceus 39.26% (± 4.72) χ
2
(df = 1) = 31.74;
<.01
13.41% (± 2.41) χ
2
(df = 1) = 39.77;
<.01
94.01%
(±1.54)
χ
2
(df = 1) = 3.93;
<.05
93.62%
(±2.24)
χ
2
(df = 1) = 21.24;
<.01
A. megalocarpus 21.41% (± 4.98) 3.58% (± 0.97) 97.04%
(±1.77)
83.06%
(±5.41)
Campiglossa
producta
S. oleraceus 37.32% (± 4.89) χ
2
(df = 1) = 13.97;
<.01
48.99% (± 5.02) χ
2
(df = 1) = 5.80;
<.05
92.59%
(±2.48)
χ
2
(df = 1) = 0.99;
>0.05
97.71%
(±1.25)
χ
2
(df = 1) = 2.52;
>.05
A. megalocarpus 18.93% (± 4.12) 31.64% (± 4.54) 95.56%
(±3.11)
92.74%
(±3.81)
Cheilosia latifrons S. oleraceus NT
c
1.50 (± 0.65) Fisher’s exact test 66.7% F
(2;16)
= 2.89; >.05 NT
c
S. hydrophilus NT
c
5.50 (± 2.86) >0.05 83.3% NT
c
A. megalocarpus NT
c
1.88 (± 0.74) 66.7% NT
c
a
T. formosa and C. producta = percentage of infested capitula; C. latifrons = the total number of eggs laid per plant (only possible in choice test).
b
T. formosa and C. producta = the percentage of adults emerged / total number of individuals produced (i.e. dead larvae, dead pupae and adults emerged) per capitula; C. latifrons = the per-
centage of successful development from eggs to pupal stage (only possible in no-choice test).
c
NT = not tested.
18 V. LESIEUR ET AL.

Discussion
Despite extensive surveys throughout the native range of S. oleraceus and the many
natural enemies recorded on the plant, none were found to be specific enough to the
Figure 4. Cheilosia latifrons. (a) Larva mining the root-crown; (b) Typical damage at the insertion of
leaves on the stem; and (c) adult (female).
Figure 3. Campiglossa producta. (a) Adults in copula; (b) Results of larval feeding (black arrows indicate
the presence of larvae); and (c) Infested flower head of Sonchus oleraceus.
BIOCONTROL SCIENCE AND TECHNOLOGY 19

weed to have potential as a biological control agent in Australia. Most of the arthropod
species collected during the surveys are generalist and several are reported as agricultural
pests. For instance, Philaenus spumarius L., the meadow spittlebug, common in spring on
Figure 5. Typical disease symptoms observed on the target weed Sonchus oleraceus and two native
species in Australia, Sonchus hydrophilus and Actites megalocarpus following inoculation with each of
the fungi tested for host-specificity. (a) Coleosporium sp., (b) Bremia sp., (c) Septoria sp., (d) Didymella
sp., (e) Alternaria sp., and (f) Ramularia sp.
20 V. LESIEUR ET AL.

S. oleraceus, is known to be an effective vector in Europe of Xyllela fastidiosa, a Gram
negative bacterium that causes important disease in numerous crops and ornamental
plants (Cruaud et al., 2018). This demonstrates that S. oleraceus acts as a reservoir for
insect pests, not only in Australia (Ollivier et al., in press), but also in its native range.
Nonetheless, our surveys did identify putative specialist natural enemies of
S. oleraceus, insects and fungi, that were investigated further. Unfortunately, all these
specialists could also attack S. hydrophilus and/or A. megalocarpus, two species native
to Australia that are closely related to S. oleraceus within the subtribe Sonchinae (Kim
et al., 2004).
The risk of non-target attack by a candidate biological control agent is generally linked
to how closely the non-target plants are related to the target weed (Hinz et al., 2019).
Sonchus oleraceus is of major concern in cropping systems in Australia, primarily
Table 4. Results of host-specificity tests performed with the prioritized fungal candidate agents on the
target weed Sonchus oleraceus and non-target species native to Australia, Sonchus hydrophilus and
Actites megalocarpus.
Name Isolate source locality Plant species
a
Number of
replicates
b
Total number of
inoculated leaves
across replicates
Disease
incidence (%
infected leaves)
Ramularia sp.Anaktorio, Greece S. oleraceus FR 1 6 83
S. oleraceus AU 3 15 100
S. hydrophilus 3 15 100
A. megalocarpus 3 15 100
Septoria sp.Montferrier-sur-Lez,
France
S. oleraceus FR 3 19 100
S. oleraceus AU NT
c
NT NT
S. hydrophilus 3 14 100
A. megalocarpus NT NT NT
Triadou, France S. oleraceus FR 3 17 100
S. oleraceus AU NT NT NT
S. hydrophilus 3 25 100
A. megalocarpus 321 95
Bremia sp. Saint Gély du Fesc,
France
S. oleraceus FR 3 15 100
S. oleraceus AU NT NT NT
S. hydrophilus 3 15 100
A. megalocarpus 3 15 100
Triadou, France S. oleraceus FR 3 15 100
S. oleraceus AU NT NT NT
S. hydrophilus 3 15 100
A. megalocarpus 3 15 100
Didymella sp. Brignoles, France S. oleraceus FR 3 15 53
S. oleraceus AU 1 6 100
S. hydrophilus 318 33
A. megalocarpus 318 33
Mauguio, France S. oleraceus FR 3 18 61
S. oleraceus AU 1 6 100
S. hydrophilus 318 56
A. megalocarpus 318 44
Alternaria sp. San Cristóbal de La
Laguna, Canary
Islands, Spain
S. oleraceus FR 3 15 100
S. oleraceus AU NT NT NT
S. hydrophilus 315 93
A. megalocarpus 315 40
Coleosporium
sp.
s-Hertogenbosch, The
Netherlands
S. oleraceus FR 4 23 100
S. oleraceus AU 8 47 100
S. hydrophilus 4 32 100
A. megalocarpus 4 32 100
a
FR = French accession; AU = Australian accession.
b
Number of plants tested.
c
NT = not tested.
BIOCONTROL SCIENCE AND TECHNOLOGY 21

because it invades fields during the fallow phase and is difficult to control since it has
developed resistance to commonly used herbicides (Adkins et al., 1997; McCarren &
Scott, 2013; Widderick et al., 2010). It grows in diverse habitats including those where
the two native Australian species in the Sonchinae used in our tests exist (http://www.
ala.org.au). For the candidate agents prioritised, we adopted a rapid screening method
for host-specificity tests by only including the target weed and its two closest native Aus-
tralian species. This method allowed us to perform an initial assessment of 10 candidate
agents in a short period. Applying this method ensures that only the most promising can-
didate agents are progressed further and subjected to comprehensive (and costly) host-
specificity testing, generally in quarantine laboratories in the country where biological
control is to be implemented (Lesieur et al., 2020; Morin, 2020; Morin et al., 2006).
The prioritisation of four insect species, belonging to Diptera, for further investi-
gations was based on an assessment of their specificity using information from the litera-
ture and field observations. The gall-forming tephritid flyT. formosa, which lays eggs
within the young flower buds and the larvae feed on the developing seeds, is reported
to be specific to the genus Sonchus (White, 1988). Because of its specificity, a closely
related species, Tephritis dilacerata, was introduced in Canada for the biological
control of S. arvensis (Berube, 1978; Peschken, 1979). In both choice and no-choice
tests, T. formosa was able to attack the non-target A. megalocarpus and survival rates
of its progeny were high. Moreover, the choice tests performed outdoor provided
robust results because they relied on flies from the naturally occurring population
seeking and selecting suitable hosts, thus minimising the likelihood of oviposition on
plants outside their host range due to artificial conditions (Schaffner et al., 2018; Shep-
pard et al., 2005). While S. oleraceus was preferentially selected by the fly compared to
A. megalocarpus in those tests, it still developed on the latter. This result indicated
that this native plant species could be impacted by the fly if released in Australia and
thus it was deemed unsuitable for biological control.
The other tephritid fly, C. producta, has a similar biology to T. formosa. While
C. producta is widely distributed through the Palearctic region (Merz, 1992; Namin &
Nozari, 2015; White, 1988), we did not find it at sites in Morocco and mainland
Europe, but only on Tenerife Island in the Canary Islands, Spain. The species is reported
to have been reared from capitula of a wide range of Cichorieae, including Sonchus spp.
(Merz, 1992; Namin & Nozari, 2015; White, 1988). However, no information on host-
range was reported for the specimens previously collected in the Canary Islands
(Merz, 1992). Since the Canary Islands is an evolutionary centre of diversification for
the genus Sonchus (Kim et al., 1999; Kim et al., 2007), we hypothesised that this popu-
lation of the fly was different (e.g. cryptic species, different behaviour) to populations
in the European mainland recorded in the literature. However, the results reported
here tend to suggest the contrary. Sequences of the specimens we collected perfectly
matched sequences of C. producta available in GenBank and BOLD from specimens col-
lected in mainland Europe and Israel. Moreover, the host range of the population col-
lected in the Canary Islands seems not to be restricted to S. oleraceus. In both choice
and no-choice tests, C. producta completed development on A. megalocarpus, albeit at
a lower rate than on S. oleraceus in choice tests. Therefore, C. producta is deemed unsui-
table as a candidate biological control agent for S. oleraceus in Australia.
22 V. LESIEUR ET AL.

The taxonomy of the hoverflyC. latifrons is still unresolved (Reemer & Claussen, 2008;
Speight, 2011). The specimens found mining the stems and root-crowns of S. oleraceus
were morphologically identified as C. latifrons. However, sequences of these specimens
did not perfectly match with sequences for this taxon in databases but were closer to
sequences of Cheilosia marokkana, which has been synonymised (based on morphology)
with C. latifrons (Speight, 2011). Cheilosia latifrons might be a cryptic species complex,
with members specialised on different host plant species in different parts of its range
(Reemer & Claussen, 2008; Speight, 2011). The biology of this fly is also poorly documen-
ted (Speight, 2011). Leontodon spp. and Hypochaeris spp. (both Asteraceae) might be part
of the host range (Reemer & Claussen, 2008; Stuke & Carstensen, 2002) but further
studies are needed to better understand host associations of C. latifrons sensu lato
(Schmid & Grossmann, 1996). During our surveys, eggs and larvae were found on
S. oleraceus and S. asper, approximately in equal proportion but not on other Asteraceae
plants present at the same site (data not shown), justifying the prioritisation of this fly for
further investigations. No differences in infestation and survival rates of C. latifrons
between test plant species were observed in our choice and no-choice tests. In the
choice tests, females produced only unfertile eggs. If this pattern reflects how the flies
would deposit fertile eggs in non-artificial conditions, this suggests that the tested
plant species are equivalent host for C. latifrons.
The midge C. sonchi was introduced in Canada in the 1980s as a biological control
agent against S. arvensis after its specificity to the genus Sonchus was confirmed in lab-
oratory tests (Peschken, 1982; Peschken et al., 1989). A reevaluation of its specificity
was required for the Australian context considering that plant species native to Australia
which is closely related to S. oleraceus had not been included in the previous testing. The
host-range of the midge was confirmed to be narrow, restricted to species in the subtribe
Sonchinae (Lesieur et al., 2020), but still, this insect was not sufficiently specificto
warrant further consideration for biological control of S. oleraceus in Australia.
Rust fungi are widely used in weed biological control given their typical high level of
host specificity and efficacy (Morin, 2020). Unfortunately, none of the rust fungi found
during the surveys proved promising. Miyagia pseudosphaeria is already present in Aus-
tralia on a range of Sonchus spp., including the native species S. hydrophilus (McCarren &
Scott, 2017). In a previous study performed under glasshouse conditions, this rust fungus
was shown to infect S. oleraceus,S. asper and to a lesser extent S. hydrophilus, but did not
develop on A. megalocarpus (McCarren & Scott, 2017). Another rust fungus found
during the surveys, at a single site in Corsica, was at the aecial stage and is thus referred
to as Aecidium sp. Sequencing indicated that it has a high similarity with the heteroecious
rust fungus, Uromyces junci. This species’life cycle is known to alternate between plants
in the Asteraceae (aecial hosts) and Juncaceae (telial hosts) families (Farr & Rossman,
2020). The third rust fungus found on S. oleraceus, with uredinia different to those of
M. pseudosphaeria, was tested for host specificity before sequencing was performed,
which revealed it was a Coleosporium sp. All isolates tested infected both
S. hydrophilus and A. megalocarpus. Most of the 100+ species in the genus Coleosporium
are heteroecious, with spermogonia and aecia on needles of Pinus spp. and uredinia and
telia on leaves of dicot species (Farr & Rossman, 2020; Helfer, 2013). Heteroecious rust
fungi are not usually considered for classical biological control of weeds because exten-
sive testing would be required to demonstrate that the selected fungus does not pose a
BIOCONTROL SCIENCE AND TECHNOLOGY 23

threat to species related to the target weed, as well as those related to the alternate host(s)
in a completely different family (Morin, 2020). Furthermore, infection of the alternate
host(s) by the candidate agent may not be acceptable in the country where biological
control is to be implemented. Consequently, the Coleosporium sp. found on
S. oleraceus was not investigated further.
The five other pathogenic fungi found on S. oleraceus during our surveys and tested
for host-specificity also infected the two native plant species to variable degrees. While
sequences of isolates of all fungi tested (ITS, ITS2-LSU, ITS1 and/or BrRxLR11)
closely matched sequences of named species in GenBank, we refrained from adopting
these names and only use genus-level names in this paper. Detailed morphological exam-
ination, sequencing of additional gene regions and phylogenetic analyses would be
required to confirm the species identity of these fungi, as, for example, has been empha-
sised for species within the Bremia genus (Choi et al., 2011; Choi et al., 2017). Despite
S. sonchi and A. sonchi being recorded in Australia, we decided to test isolates of Septoria
sp. and Alternaria sp. recovered from S. oleraceus in our surveys because they could be
different species. We did not explore further their taxonomy because these fungi were
found to infect the native species tested and were thus rejected as a candidate biological
control agent for S. oleraceus in Australia.
Of the 59 fungal pathogens recorded as occurring on S. oleraceus in the literature
review we performed (data not shown), we were particularly interested in Entyloma
sonchi Vánky reported on S. asper in Europe (Vánky, 1983; Vánky, 1994).This interest
was based on the fact that one of the most successful fungal pathogen used for the bio-
logical control of a weed belongs to this genus –Entyloma ageratinae Barreto and Evans
on Ageratina riparia (Regel) R. King and H. Robinson (Barton et al., 2007). The vicinity
of the area in Brittany, France where two collections of E. sonchi (originally identified as
Entyloma calendulae Cif. var. bullulum) were made in the 1940s (Vánky, 1983; Vánky,
1994) was intensively surveyed at the same time of the year of the previous collections
but without success.
None of the prioritised candidate agents tested in this study were found to be specific
enough to be pursued further for S. oleraceus biological control in Australia. Three
insect species found on S. oleraceus that were not tested due to the small number of
individuals found and failure to initiate laboratory colonies, Tolyphus sp. (Coleopera:
Phalacridae), an unidentified gall wasp (Hymenoptera: Cynipidae) and the tortricid
moth Acroclita sonchana Walsingham (Lepidoptera: Tortricidae), might warrant
further investigations. While the two former species may not be promising considering
the minor damage they were associated with on S. oleraceus in the field, A. sonchana
might have more potential based on observations of significant feeding of leaves and
stems by larvae. Acroclita sonchana is an endemic species to the Canary Islands that
has only previously been reported to feed on leaves of the endemic plant species
Sonchus gummifer Link, Sonchus leptocephalus Cass. and Sonchus congestus (Klimesch,
1987; Walsingham, 1908).
Any top-down regulation of S. oleraceus by its natural enemies in its native range
appears to be exerted primarily by generalist species. Schroeder (1974) suggested that
the ecology of Sonchus may be crucial in understandings its natural enemy community.
Most of the weedy Sonchus spp. are ruderal species and dependent of disturbed con-
ditions; they do not tend to be good competitors and do not persist if the disturbance
24 V. LESIEUR ET AL.

is not maintained. These conditions might be suboptimal for the development of special-
ised host-associations by arthropods and fungi. When focusing on S. oleraceus, the lack of
highly specific natural enemies may further be explained by its genetic origin. This
species is an amphidiploid that is postulated to be the result of a combination of
S. asper and S. tenerrimus (Cho et al., 2019; Stebbins et al., 1953). In our survey, most,
if not all, the arthropod species detected on S. oleraceus were also found on S. asper
and, to a less extent, on S. tenerrimus. The ability of all putative specialist candidate
agents prioritised and tested to use or infect multiple species within the subtribe Sonch-
inae may, in part, reflect this. Collectively, our investigations suggest that classical bio-
logical control may not be a feasible option for the management of S. oleraceus in
Australia, and that alternative integrated weed management tactics may need to be
pursued to mitigate the impacts of this weed.
The candidate agents tested in this study, however, might be promising for other parts
of the world, such as in North and South America where no native Sonchinae are present
(Kim et al., 2004). While testing of other plant species related to S. oleraceus would be
necessary to thoroughly assess risks, the candidate agents prioritised in this study
might be found to be suitable for biological control in those other geographic areas.
Acknowledgements
The authors are grateful to Marie-Stéphane Tixier and Jean-Francois Martin (Montpellier
SupAgro) for their support throughout the programme. The authors thank Jenny Guerin
(South Australian Seed Conservation Centre), Kathryn Batchelor (CSIRO) and Denzel Murfet
(private citizen) for seeds of the native Australian species, S. hydrophilus and A. megalocarpus,
used in our testing. The authors also thank the Dirrecíon General de Proteccíon de la Naturaleza
(Gobierno de Canarias), Consejería de Agricultura, Pesca y Medio Ambiente (Junta de Andalucia)
and Direcció General de Polítiques Ambientals I Medi Natural (Generalitat de Catalunya) for
authorisations of collections. Pr. Mohamed Ghamizi (Cadi Ayyad Univ.) organised collection
authorisations for our surveys in Morocco and Yassine Fendane assisted with collections there.
The authors thank Armelle Cœur d’Acier, Jean Claude Streito, Eric Pierre, Bruno Michel from
the UMR CBGP, INRAE (France), Jean Marie Ramel and Valérie Balmes (ANSES-LSV,
France), and Michel Martinez for their help in morphological identification of insects. The
authors are grateful to EBCL (USDA-ARS, France) and particularly to Marie-Claude Bon for
giving us access to the molecular biology laboratory to perform our molecular investigations of
fungi. Chloé Malik, Paul Vaast, Manon Hervé and Maëva Miranda are gratefully acknowledged
for assistance with initial testing for host-specificity. Thanks are extended to colleagues
Kumaran Nagalingam, Gavin Hunter and X anonymous reviewers for constructive feedback on
previous versions of the manuscript.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Funding
This project was supported by AgriFutures Australia (Rural Industries Research and Develop-
mentCorporation), through funding from the Australian Government Department of Agriculture,
Water and the Environment, as part of its Rural Research and Development for Profit program
(PRJ-010527).
BIOCONTROL SCIENCE AND TECHNOLOGY 25

ORCID
Vincent Lesieur http://orcid.org/0000-0003-3374-6004
Mélodie Ollivier http://orcid.org/0000-0002-8379-5771
Louise Morin http://orcid.org/0000-0002-9515-2255
S. Raghu http://orcid.org/0000-0001-5843-5435
References
Adkins, S. W., Wills, D., Boersma, M., Walker, S. R., Robinson, G., McLeod, R. J., & Einam, J. P.
(1997). Weeds resistant to chlorsulfuron and atrazine from the north-east grain region of
Australia. Weed Research,37(5), 343–349. https://doi.org/10.1046/j.1365-3180.1997.d01-56.x
Agrios, G. (1997). Plant pathology. Academic Press.
ALA, Atlas of Living Australia. (2020). http://www.ala.org.au. Accessed March 8, 2021
Barton, J., Fowler, S. V., Gianotti, A. F., Winks, C. J., De Beurs, M., Arnold, G. C., & Forrester, G.
(2007). Successful biological control of mist flower (Ageratina riparia) in New Zealand: Agent
establishment, impact and benefits to the native flora. Biological Control,40(3), 370–385. https://
doi.org/10.1016/j.biocontrol.2006.09.010
Berner, D., & Bruckart, W. (2005). A decision tree for evaluation of exotic plant pathogens for clas-
sical biological control of introduced invasive weeds. Biological Control,34(2), 222–232. https://
doi.org/10.1016/j.biocontrol.2005.04.012
Berube, D. (1978). The basis for host plant specificity in Tephritis dilacerata and T. Formosa [Dipt:
Tephritidae]. BioControl,23(4), 331–337. https://doi.org/10.1007/BF02373049
Boulos, L. (1973). Révision systématique du genre Sonchus L. s.l. IV. Sous-genre I. Sonchus.
Botaniska Notiser,126, 155–196.
Bremer, K. (1994). Cladistics and classification. Timber Press.
Cho, M.-S., Kim, J. H., Kim, C.-S., Mejías, J. A., & Kim, S.-C. (2019). Sow thistle chloroplast
genomes: Insights into the plastome evolution and relationship of two weedy species,
Sonchus asper and Sonchus oleraceus (Asteraceae). Genes,10(11), 881. https://doi.org/10.
3390/genes10110881
Choi, Y.-J., Thines, M., Runge, F., Hong, S.-B., Telle, S., & Shin, H.-D. (2011). Evidence for high
degrees of specialisation, evolutionary diversity, and morphological distinctiveness in the genus
bremia. Fungal Biology,115(2), 102–111. https://doi.org/10.1016/j.funbio.2010.10.010
Choi, Y.-J., Wong, J., Runge, F., Mishra, B., Michelmore, R., & Thines, M. (2017). BrRxLR11 –a
new phylogenetic marker with high resolution in the downy mildew genus Bremia and related
genera.Mycological Progress,16(2), 185–190. https://doi.org/10.1007/s11557-016-1258-9
Cruaud, A., Gonzalez, A.-A., Godefroid, M., Nidelet, S., Streito, J.-C., Thuillier, J.-M., Rossi, J.-P.,
Santoni, S., & Rasplus, J.-Y. (2018). Using insects to detect, monitor and predict the distribution
of Xylella fastidiosa: A case study in corsica. Scientific Reports,8(1), 15628. https://doi.org/10.
1038/s41598-018-33957-z
Cruaud, A., Jabbour-Zahab, R., Genson, G., Cruaud, C., Couloux, A., Kjellberg, F., Van Noort, S.,
& Rasplus, J. Y. (2010). Laying the foundations for a new classification of Agaonidae
(Hymenoptera: Chalcidoidea), a multilocus phylogenetic approach. Cladistics,26(4), 359–
387. https://doi.org/10.1111/j.1096-0031.2009.00291.x
Encinas-Viso, F., Bovill, J., Morin, L., Raghu, S., Knerr, N., Roux, C., & Broadhurst, L. (in press).
Population genomics reveal multiple introductions and admixture of Sonchus oleraceus in
Australia. [Manuscript submitted for publication]. Evolutionary Applications.
Farr, D., & Rossman, A. (2020). Fungal databases, US national fungus collections. USDA. https://nt.
ars-grin.gov/fungaldatabases/
Fensham, R., & Laffineur, B. (2019). Defining the native and naturalised flora for the Australian
continent. Australian Journal of Botany,67(1), 55–69. https://doi.org/10.1071/BT18168
Folmer, O., Black, M., Hoeh, W., Lutz, R., & Vrijenhoek, R. (1994). DNA primers for amplification
of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates.
Molecular Marine Biology and Biotechnology,3(5), 294–299.
26 V. LESIEUR ET AL.

Gäumann, E. (1959). Die Rostpilze Mitteleuropas. Beiträge zur Kryptogamenflora der Schweiz (Vol.
12). Buchdruckerei Buehler & Co.
Germain, J.-F., Chatot, C., Meusnier, I., Artige, E., Rasplus, J.-Y., & Cruaud, A. (2013). Molecular
identification of Epitrix potato flea beetles (coleoptera: Chrysomelidae) in Europe and North
America. Bulletin of Entomological Research,103(3), 354–362. https://doi.org/10.1017/
S000748531200079X
Gregg, P. C., Del Socorro, A. P., Le Mottee, K., Tann, C. R., Fitt, G. P., & Zalucki, M. P. (2019). Host
plants and habitats of Helicoverpa punctigera and H. Armigera (Lepidoptera: Noctuidae) in inland
Australia. Austral Entomology,58(3), 547–560. https://doi.org/10.1111/aen.12349
Grosskopf, G., Smith, L. A., & Syrett, P. (2002). Host range of Cheilosia urbana (Meigen) and
Cheilosia psilophthalma (Becker)(Diptera: Syrphidae), candidates for the biological control of
invasive alien hawkweeds (Hieracium spp., Asteraceae) in New Zealand. Biological Control,
24(1), 7–19. https://doi.org/10.1016/S1049-9644(02)00011-7
Gu, H., Cao, A., & Walter, G. (2001). Host selection and utilisation of Sonchus oleraceus
(Asteraceae) by Helicoverpa armigera (Lepidoptera: Noctuidae): a genetic analysis. Annals of
Applied Biology,138(3), 293–299. https://doi.org/10.1111/j.1744-7348.2001.tb00114.x
Helfer, S. (2013). Coleosporium in Europe. Mycotaxon,124(1), 87–99. https://doi.org/10.5248/124.87
Hinz, H. L., Winston, R. L., & Schwarzländer, M. (2019). How safe is weed biological control? A
global review of direct nontarget attack. The Quarterly Review of Biology,94(1), 1–27. https://
doi.org/10.1086/702340
Hutchinson, I., Colosi, J., & Lewin, R. A. (1984). The biology of Canadian weeds. 63. Sonchus asper
(L) Hill and Sonchus oleraceus.Canadian Journal of Plant Science,64(3), 731–744. https://doi.
org/10.4141/cjps84-100
Keane, R. M., & Crawley, M. J. (2002). Exotic plant invasions and the enemy release hypothesis.
Trends in Ecology & Evolution,17(4), 164–170. https://doi.org/10.1016/S0169-5347(02)02499-0
Kim, S.-C., Chunghee, L., & Mejías, J. A. (2007). Phylogenetic analysis of chloroplast DNA matK
gene and ITS of nrDNA sequences reveals polyphyly of the genus Sonchus and new relationships
among the subtribe Sonchinae (Asteraceae: Cichorieae). Molecular Phylogenetics and Evolution,
44(2), 578–597. https://doi.org/10.1016/j.ympev.2007.03.014
Kim, S.-C., Crawford, D. J., Francisco-Ortega, J., & Santos-Guerra, A. (1999). Adaptive radiation
and genetic differentiation in the woody Sonchus alliance (Asteraceae: Sonchinae) in the Canary
islands. Plant Systematics and Evolution,215(1-4), 101–118. https://doi.org/10.1007/
BF00984650
Kim, S.-C., Lu, C. T., & Lepschi, B. J. (2004). Phylogenetic positions of Actites megalocarpa and
Sonchus hydrophilus (Sonchinae: Asteraceae) based on ITS and chloroplast non-coding DNA
sequences. Australian Systematic Botany,17(1), 73–81. https://doi.org/10.1071/SB03019
Klimesch, J. (1987). Beitrage zur Kenntnis der Microlepidopteren-Fauna des Kanarischen
Archipels. 9 Beitrag: Tortricidae, Cochylidae. Vieraea,17, 297–322. https://doi.org/10.1016/j.
biocontrol.2021.104667
Knihinicki, D. K., McCarren, K.L., & Scott, J. K. (2009). A new species of Aceria (Acari: Eriophyidae)
damaging sowthistles, Sonchus spp. (Asteraceae), in Australia with notes on Aceria sonchi (Nalepa,
1902). Zootaxa,2119(1), 23–38. https://doi.org/10.11646/zootaxa.2119.1.2
Kriticos, D. J., Ireland, K. B., Morin, L., Kumaran, N., Rafter, M. A., Ota, N., & Raghu, S. (in press).
Integrating ecoclimatic niche modelling methods into classical biological control programmes.
[Manuscript submitted for publication]. Biological Control.https://doi.org/10.1016/j.biocontrol.
2021.104667
Kunkel, G. (1976). Biogeography and ecology in the Canary islands. Junk.
Lesieur, V., Thomann, T., Ollivier, M., & Raghu, S. (2020). Making host specificity testing more
efficient: Exploring the use of abridged test plant lists. Journal of Applied Entomology,144(6),
546–551. https://doi.org/10.1111/jen.12760
Liu, H., & Stiling, P. (2006). Testing the enemy release hypothesis: A review and meta-analysis.
Biological Invasions,8(7), 1535–1545. https://doi.org/10.1007/s10530-005-5845-y
Madeira, F., Park, Y. m., Lee, J., Buso, N., Gur, T., Madhusoodanan, N., Basutkar, P., Tivey,
A. R. N., Potter, S. C., Finn, R. D., & Lopez, R. (2019). The EMBL-EBI search and sequence
BIOCONTROL SCIENCE AND TECHNOLOGY 27

analysis tools APIs in 2019. Nucleic Acids Research,47(W1), W636–W641. https://doi.org/10.
1093/nar/gkz268
Manalil, S., Werth, J., Jackson, R., Chauhan, B. S., & Preston, C. (2017). An assessment of weed
flora 14 years after the introduction of glyphosate-tolerant cotton in Australia. Crop and
Pasture Science,68(8), 773–780. https://doi.org/10.1071/CP17116
McCarren, K. L., & Scott, J. K. (2013). Host range and potential distribution of Aceria thalgi (Acari:
Eriophyidae): A biological control agent for Sonchus species. Australian Journal of Entomology,
52(4), 393–402. https://doi.org/10.1111/aen.12041
McCarren, K. L., & Scott, J. K. (2017). Host range and potential distribution of the rust fungus,
Miyagia pseudosphaeria, a biological control agent for Sonchus species. Australasian Plant
Pathology,46(5), 473–482. https://doi.org/10.1007/s13313-017-0509-9
McClay, A., & Peschken, D. (2001). 80 Sonchus arvensis L., Perennial Sow-thistle (Asteraceae).
Biological Control Programmes in Canada, 1981-2000, 416.
McIntosh, P. (2018). Sowthistle biology –management & resistance status. Retrieved October 22,
2019, from https://grdc.com.au/resources-and-publications/grdc-update-papers/tab-content/
grdc-update-papers/2018/12/sowthistle-biology-management-and-resistance-status
Merz, B. (1992). The fruit flies of the Canary Islands (diptera: Tephritidae). Insect Systematics &
Evolution,23(2), 215–231. https://doi.org/10.1163/187631292X00308
Moreno, A., & Fereres, A. (2012). Virus diseases in lettuce in the Mediterranean basin. Advances in
Virus Research,84, 247–288. https://doi.org/10.1016/B978-0-12-394314-9.00007-5
Morin, L. (2020). Progress in biological control of weeds with plant pathogens. Annual Review of
Phytopathology,58(1), 201–223. https://doi.org/10.1146/annurev-phyto-010820-012823
Morin, L., Evans, K. J., & Sheppard, A. W. (2006). Selection of pathogen agents in weed biological
control: Critical issues and peculiarities in relation to arthropod agents. Australian Journal of
Entomology,45(4), 349–365. https://doi.org/10.1111/j.1440-6055.2006.00562.x
Namin, S. M., & Nozari, J. (2015). Fruit flies of the genus Campiglossa (Diptera, Tephritidae) in
Iran, with the key to species. Vestnik Zoologii,49(3), 251–260. https://doi.org/10.1515/vzoo-
2015-0026
Ollivier, M., Kazakou, E., Corbin, M., Sartori, K., Gooden, B., Lesieur, V., Thomann, T., Martin, J.-
F., & Tixier, M. S. (2020). Trait differentiation between native and introduced populations of the
invasive plant Sonchus oleraceus L. (asteraceae). NeoBiota,55,85–115. https://doi.org/10.3897/
neobiota.55.49158
Ollivier,M.,Labouyrie,M.,Raghu,S.,Tavoillot,J.,Tixier,M.S.,&Lesieur,V.(in press).
Characterising the herbivore community on Sonchus oleraceus (Asterales: Asteraceae) in
its invaded range in Australia. [Manuscript submitted for publication]. Austral
Entomology.
Osten, V. A., Walker, S. R., Storrie, A., Widderick, M., Moylan, P., Robinson, G. R., & Galea, K.
(2007). Survey of weed flora and management relative to cropping practices in the north-eastern
grain region of Australia. Australian Journal of Experimental Agriculture,47(1), 57–70. https://
doi.org/10.1071/EA05141
Park, K. W., Kolkman, J. M., & Mallory-Smith, C. A. (2012). Point mutation in acetolactate
synthase confers sulfonylurea and imidazolinone herbicide resistance in spiny annual sow-
thistle Sonchus asper (L.) hill. Canadian Journal of Plant Science,92(2), 303–309. https://doi.
org/10.4141/cjps2011-159
Peschken, D. (1979). Host specificity and suitability of Tephritis dilacerata [Dip.: Tephritidae]: A
candidate for the biological control of perennial sow-thistle (Sonchus arvensis)[compositae] in
Canada. BioControl,24(4), 455–461. https://doi.org/10.1007/BF02374185
Peschken, D. (1982). Host specificity and biology of Cystiphora sonchi [Dip.: cecidomyiidae], a
candidate for the biological control of Sonchus species. Entomophaga,27(4), 405–415. https://
doi.org/10.1007/BF02372063
Peschken, D., McClay, A., Derby, J., & DeClerck, R. (1989). Cystiphora sonchi (Bremi) (diptera:
Cecidomyiidae), a new biological control agent established on the weed perennial sow-thistle
(Sonchus arvensis L.)(Compositae) in Canada. The Canadian Entomologist,121(9), 781–791.
https://doi.org/10.4039/Ent121781-9
28 V. LESIEUR ET AL.

Pyšek, P., Pergl, J., Essl, F., Lenzner, B., Dawson, W., Kreft, H., Weigelt, P., Winter, M., Kartesz, J.,
Nishino, M., Antonova, L. A., Barcelona, J. F., Cabesaz, F. J., Cárdenas, D., Cárdenas-Toro, J.,
Castaño, N., Chacón, E., Chatelain, C., Dullinger, S., …Kleunen, M. v. (2017). Naturalized alien
flora of the world. Preslia,89(3), 203–274. https://doi.org/10.23855/preslia.2017.203
Rajapakse, C. N. K., & Walter, G. H. (2007). Polyphagy and primary host plants: Oviposition pre-
ference versus larval performance in the lepidopteran pest Helicoverpa armigera.Arthropod-
Plant Interactions,1(1), 17–26. https://doi.org/10.1007/s11829-007-9003-6
R Core Team. (2018). R: A language and environment for statistical computing. R Foundation for
Statistical Computing. https://www.R-project.org
Reemer, M., & Claussen, C. (2008). Het grijs gitje Cheilosia griseiventris sensu Stubbs & Falk (1983)
in Nederland (Diptera: Syrphidae). Nederlandse Faunistische Mededelingen,28,59–62.
Schaffner, U., Smith, L., & Cristofaro, M. (2018). A review of open-field host range testing to evalu-
ate non-target use by herbivorous biological control candidates. BioControl,63(3), 405–416.
https://doi.org/10.1007/s10526-018-9875-7
Schmid, U., & Grossmann, A. (1996). Eiablage von Cheilosia latifrons (Zetterstedt, 1843) (Díptera,
Syrphidae) an Leontodón autumnalis L. Volucella,2(1/2), 86–87.
Schroeder, D. (1974). The phytophagous insects attacking Sonchus spp.(Compositae) in Europe.
Miscellaneous Publications Commonwealth Institute of Biological Control.
Scott, J. K., & McCarren, K. (2012). Sonchus oleraceus L. –sowthistle. In J. Cullen, M. Julien, & R.
McFadyen (Eds.), Biological control of weeds in Australia (pp. 563–567). CSIRO Publishing.
Sheppard, A., Van Klinken, R., & Heard, T. (2005). Scientificadvances in the analysis of direct risks
of weed biological control agents to nontarget plants. Biological Control,35(3), 215–226. https://
doi.org/10.1016/j.biocontrol.2005.05.010
Skinner, K., Smith, L., & Rice, P. (2000). Using noxious weed lists to prioritize targets for devel-
oping weed management strategies. Weed Science,48(5), 640–644. https://doi.org/10.1614/
0043-1745(2000)048[0640:UNWLTP]2.0.CO;2
Speight, M. (2011). Species accounts of European Syrphidae (diptera), glasgow 2011. Syrph,65,
285.
Stebbins, G. L., Jenkins, J. A., & Walters, M. S. (1953). Chromosomes and phylogeny in the com-
positae, tribe cichorieae. Madroño,12,65–81.
St John-Sweeting, R. S. (2011). Dispersal and genetic variability of Sonchus oleraceus L. in relation
to its resistance to ALS-inhibiting herbicides [Ph.D. dissertation]. School of Agriculture, Food
and Wine, University of Adelaide.
Stubbs, L., Guy, J., & Stubbs, K. (1963). Control of lettuce necrotic yellows virus disease by the
destruction of common sowthistle (Sonchus oleraceus). Australian Journal of Experimental
Agriculture,3(10), 215–218. https://doi.org/10.1071/EA9630215
Stuke, J., & Carstensen, L. (2002). Die Biologie und Morphologie der dritten Larvenstadien von
vier mitteleuropäischen Cheilosia Arten. Beiträge für Entomologie,52, 283–307.
Vánky, K. (1983). Ten new species of Ustilaginales [fungi, description, smut, plant hosts].
Mycotaxon,18(2), 319–336.
Vánky, K. (1994). European smut fungi. Gustav Fischer Verlag.
Vieira, B., & Barreto, R. (2006). First record of Bremia lactucae infecting Sonchus oleraceus and
Sonchus asper in Brazil and its infectivity to lettuce. Journal of Phytopathology,154(2), 84–87.
https://doi.org/10.1111/j.1439-0434.2006.01064.x
Walker, S. R., Taylor, I. N., Milne, G., Osten, V. A., Hoque, Z., & Farquharson, R. J. (2005). A
survey of management and economic impact of weeds in dryland cotton cropping systems of
subtropical Australia. Australian Journal of Experimental Agriculture,45,79–91.
Walsingham, T. D. G. (1908). Microlepidoptera of Tenerife. Proceedings of the Zoological Society
London.
White, I. M. (1988). Tephritid flies: Diptera: Tephritidae (Vol. 10). Royal Entomological Society.
Widderick, M., Sindel, B. M., & Walker, S. R. (1999). Distribution, importance and management of
Sonchus oleraceus (common sowthistle) in the northern cropping region of Australia. 12th
Australasian Weeds Conference, Hobart, Tasmania, Australia.
BIOCONTROL SCIENCE AND TECHNOLOGY 29

Widderick, M., Walker, S., & Sindel, B. (2004). Better management of Sonchus oleraceus
L. (common sowthistle) based on the weed’s ecology. 14th Australian Weeds Conference,
Wagga Wagga, New South Wales, Australia.
Widderick, M. J., Walker, S. R., Sindel, B. M., & Bell, K. L. (2010). Germination, emergence, and
persistence of Sonchus oleraceus, a major crop weed in subtropical Australia. Weed Biology and
Management,10(2), 102–112. https://doi.org/10.1111/j.1445-6664.2010.00370.x
Wilson, M., & Henderson, D. M. (1966). British rust fungi. Cambridge University Press.
30 V. LESIEUR ET AL.