Orobanche cumana
(sunflower broomrape)

O. cumana is an obligatory, non-photosynthetic root parasite. It is believed to have evolved relatively recently from forms of O. cernua parasitizing wild Asteraceae, in particular species of Artemisia, and transferring to cultivated Helianthus annuus (sunflower). O. cumana is thought to be native to eastern Europe (Russia) and has subsequently spread to most other sunflower growing regions of central and western Europe and Asia. The absence of O. cumana in sunflower growing regions of South America (for example Argentina) is believed to be associated with warmer winter temperatures not suitable for this species, rather than the seeds not being present. O. cumana can cause immense damage to cultivated sunflowers resulting in a significant decrease in yield. Despite resistant sunflower varieties being developed more virulent races of O. cumana have repeatedly evolved, or been selected, to overcome resistance. Thus, in spite of constant breeding efforts, losses continue in established sunflower growing areas and there is potential for it to invade new areas, wherever sunflower is grown.

O. cumana occurs widely from southern and eastern Europe eastwards through southern central Asia into China. From Russia, O. cumana has moved steadily westwards into most major sunflower growing countries. Curiously some major databases are out-of-date in this respect. For example, USDA-ARS (2016) and ITIS (2016) fail to indicate its occurrence in Turkey, Romania, Hungary, Italy, France and Spain. In is also now present in Tunisia. Conversely, a number of papers refer to its occurrence in Iran, but it has not been possible to trace a reliable record for this. In the Distribution Table, it is assumed that the species is native only in Russia and is introduced in all other countries.

O. cumana produces a large number of small, inconspicuous seeds which can remain viable for up to 10 years. The seeds are readily dispersed naturally by wind and water and can be accidentally introduced into a new area as a contaminant of seed, in particular H. annuus. O. cumana, like many other Orobanche species is listed and restricted under the phytosanitary regulations of most countries. However, detection of the seeds by visual inspection is near impossible.

O. cumana is host specific and is only found associated with cultivated H. annuus.

O. cumana is often associated with H. annuus and to some wild Helianthus species. Some reports of O. cumana on other wild hosts, including species of Artemisia are presumed to be a misnaming of the more typical O. cernua taxa.

Genetics

The chromosome number of O. cumana is 2n = 38 (Missouri Botanic Garden, 2016), however Musselman (1986) indicates that 2n = 24 may also occur.

Variation within the species has been much studied and Gagne et al. (1998) found two distinct groups; one corresponding to the East European countries, Bulgaria, Romania and Turkey and the other group corresponding to Spanish populations. Within Spain, two distant gene pools occur, one in Cuenca province and another in the Guadalquivir Valley, apparently deriving from separate introduction events. Different races occurred within each gene pool, suggesting that current races might have evolved through mutation from a common genetic background (Pieneda-Martos et al., 2013).

Reproductive Biology

O. cumana is generally considered to be autogamous (Gagne et al., 1998), but some studies have revealed that it can be partially allogamous (Rodríguez-Ojeda et al., 2013). Seeds are produced in very large numbers (up to 100,000 per plant) and remain viable in soil for many years, possibly 10 years or more.

Physiology and Phenology

O. cumana is an obligate parasite. The seed is minute (approximately 0.2 x 0.4 mm), from which only the radicle emerges and this can grow only a few mm long, needing to establish a connection to a host root within a few days of germination. Thus, germination depends on a chemical stimulus from the host root. The stimulus for most Orobanche species is one or more of the group of terpenoid lactones known as strigolactones. H. annuus produces a number of these compounds, including e.g. carlactone and heliolactone (Ueno et al., 2014) but O. cumana is much less responsive to these than to guaianolide sesquiterpene lactone and dehydrocostus lactone which are now considered to be the main stimulants exuded by sunflower (Joel et al., 2011). Seeds of O. cumana therefore only germinate when a host root is nearby but also require a moist environment (for several days) together with suitable temperatures. This preparatory period is known as conditioning or preconditioning. Studies with O. cernua found that conditioned seeds remain responsive to germination stimulants for a limited period beyond which secondary dormancy may be induced, especially at lower than optimal temperatures (Weldeghiorghis and Murdoch, 1997; Kebreab and Murdoch, 1999). The ability of the seeds to respond to germination stimuli also decreases gradually when the seeds dry and they then remain dormant until reconditioned (Timko et al., 1989; Joel et al., 1995). Germination of O. cumana has been reported to occur over a range from 4-32°C. However, Murdoch and Kebreab (2013) found an optimum for O. cernua of 26°C, while for O. cumana, Foy et al. (1991) found it to be 20°C, with much reduced germination at 15 or 30°C and Sauerborn (1989) reported an optimum at 15°C.

On contact with the host root, a haustorium, is formed and intrusive cells penetrate through the cortex to the vascular bundle to establish connection with the host xylem. The pectolytic activity of intrusive cells of the parasite is performed by pectin methyl esterase and another enzyme, possibly polygalacturonase, which cause complete degradation of cell wall pectins, allowing separation of the cells and smooth penetration by intrusive cells (Losner-Goshen et al., 1998).

O. cumana develops into a tubercle on the surface of the root, developing to a diameter of 5-20 mm. Secondary roots may develop on the tubercle and make separate contacts with the host root system. After several weeks, the tubercle develops one or more flowering shoots which emerges above the soil.

Species of Orobanche depend totally on their hosts for all nutrition, drawing sugars and nitrogen compounds directly from the phloem and also drawing most of their water from the host xylem. O. cumana therefore becomes an active sink, comparable to an actively growing part of the host plant itself, such that effects on the host are generally proportional to the biomass of the parasite. Thus, the mass of the parasite is reflected in a very similar loss in mass of the host crop (Hibberd et al., 1998; Hibberd et al., 1999). Photosynthesis is little affected (Grenz et al., 2008).

Although O. cumana behaves as an annual, the tubercle appears to retain some of the perennial characteristics of O. cernua, parasitising perennial wild hosts and producing new shoots after the first from that tubercle has matured and shed seed (Antonova et al., 2012).

Environmental Requirements

O. cumana apparently thrives on any soils where H. annuus is commonly grown, including coarse sands and heavier soils. Lozano-Cabello (1999) observed less O. cumana in soils of pH8 compared to pH6, but Miladinovic et al. (2012) concluded that soil texture, fertility and pH were not generally critical for its presence. However, low phosphorus content tended to encourage the growth of O. cumana. The absence from Argentine (in spite of assumed probability of accidental introduction) could be due to the warmer conditions of the Argentine winter (Miladinovic et al., 2012).

A study in China identified 234 different pathogens from O. cumana with 62% belonging to the genus Fusarium (114), 30% to Rhizotonia, 2% Pythium and 6% to others. Three Fusarium species identified were Fusarium oxysporum, F. solani and F. cerealis [Gibberella pulicaris] (Ding et al., 2012b). Pathogens identified in Armernia included F. lateritium [Gibberella baccata] (Taslakh'yan and Grigoryan, 1978). Other pathogens reported include Sclerotinia sclerotiorum (Ding et al., 2012a) Ulocladium botrytis [Alternaria botrytis] (Müller-Stöver et al., 2005) and Alternaria alternate, which causes the inhibition of sphinganine N-acyltransferase, a key enzyme in sphingolipid biosynthesis, leading to accumulation of toxic sphingoid bases (de Zélicourt et al., 2009)

An extract of F. verticillioides [Gibberella fujikuroi] growth medium caused complete mortality of O. cumana seedlings in vitro. The toxic metabolite was isolated and identified by spectroscopic methods as fusaric acid (Dor et al., 2009).

In a detailed review of insects attacking Orobanche species, Kroschel and Klein (1999) listed 40 phytophagous insects from 22 families. At least half of these were recorded from O. cernua (sensu lato). A mining fly, Phytomyza orobanchia was found to be present on O. cumana in most of the countries in which it occurs. In Hungary the larvae of P. orobanchia destroyed 37 and 69% of the seed capsules of infected plants in 1980 and 1982, respectively and similar rates of infestation are reported from many other regions. In addition, a moth Scotia segetum [Agrotis segetum] (Noctuidae) attacked >90% of O. cumana plants at 1-3 larvae/plant in a sunflower field (Lekic, 1970).

Visual detection of the seeds of O. cumana amongst crop seed is extremely difficult, however molecular techniques have been developed (Dongo et al., 2012). The results of this assay can be expressed in terms of the number of O. cumana seeds per kilogram of crop seeds and can help decisions regarding crop seed lot utilisation and commercialisation.

O. cumana (and O. cernua), are much less robust than O. crenata and are distinguished from the other closely related weedy species, O. minor, by the latter’s distinct veins in the corolla and wider-spreading lips. A key is provided by Parker (2013). They are distinguished from the related broomrapes Phelipanche ramosa and P. aegyptiaca by the absence of bracteoles and the lack of branching above ground

Morphological differences between typical O. cernua and O.cumana include: height 15-30 cm in O. cernua, 40-60 in O. cumana; flowers 10-20 mm in O. cernua, 20-30 mm in O. cumana; inflorescence dense in cernua, lax in cumana; filaments and anthers virtually glabrous in cernua, hairy in cumana. The flowers are conspicuously longer and more down-curved in O. cumana.

These differences are illustrated by Pujadas-Salva and Thalouarn (1998). They may also be distinguished on the basis of DNA markers, even from individual seeds (Joel et al., 1996) and on the basis of seed-borne oils with O.cumana containing much higher levels of linoleic acid than typical O. cernua (Pujadas-Salva and Velasco, 2000).

Due to the variable regulations around (de)registration of pesticides, your national list of registered pesticides or relevant authority should be consulted to determine which products are legally allowed for use in your country when considering chemical control. Pesticides should always be used in a lawful manner, consistent with the product's label.

Prevention

SPS Measures

O. cumana, like many other Orobanche species is listed and restricted under the phytosanitary regulations of most countries.

Control

Cultural Control and Sanitary Measures

In Spain, late sowings (from the end of March until the beginning of April) favour the enhanced expression of resistance of sunflower to O. cumana race F irrespective of seedbank and can be therefore recommended, under irrigation and together with the use of moderately resistant sunflower hybrids, as part of an efficient strategy on the control of C. cumana (Akhtouch et al., 2013). Similarly, in Israel, resistance of the sunflower cv. Sunbred-254 to O. cumana was enhanced when sown from January onwards than when sown before January (Ish-Shalom-Gordon et al., 1994). Conversely, in Romania, Grenz et al., (2008) found that delayed sowing combined with improved water and nitrogen supply were associated with increases in parasite number that neutralised the yield-boosting effects of irrigation and fertilisation at the highest infestation level.

Trap crops could help stimulate germination of O. cumana seeds and contribute to the reduction of the soil seed bank. Species identified in pot studies as possible trap crops include Panicum virgatum (An Yu et al., 2015), various species of Sorghum and Sudan grass varieties (Antonova et al., 2015), Secale cereale (rye) (Cimmino et al., 2015), Cannabis sativa (hemp) (Yu and Ma, 2014) and Zea mays (maize) (Ma et al., 2013) but this has not been reported in the field.

Physical/Mechanical Control

The shoots of O. cumana can be hand-pulled but the benefit is limited and often too late as most of the damage will already have been done.

Biological Control

Bedi et al. (1994) investigated the potential of Fusarium oxysporum f. sp. orthoceras isolated from diseases inflorescences of O. cumana in Bulgaria as a potential biocontrol agent. This pathogen has been studied further and has proven to be efficacious under greenhouse conditions when formulated as wheat-kaolin granules (Shabana et al., 2003; Dor et al., 2007). A combination with F. solani (a weak pathogen of O. cumana) isolated in Israel from O. aegyptiaca was found to be synergistic providing more effective control of O. cumana than either agent alone (Dor et al., 2006).

Other potential biocontrol candidates have included Aspergillus alliaceus (Aybeke et al., 2014) and Ulocladium botrytis [Alternaria botrytis] (Müller-Stöver et al., 2005). In spite of this there are no reports of the current use of fungi for biological control in the field.

The one insect to have been extensively studied as a possible biocontrol agent is the dipteran Phyomyza orobanchia which feeds on a number of species of Orobanche (Kroschel and Klein, 1999). In one study in Russia, P. orobanchia was exploited on over 30,000 ha, involving the release of 5-600 adults per ha and was estimated to have reduced seed production by 82-88%. Studies on other species of Orobanche achieved over 90% reduction but only when repeated for 3-4 years. However, since seed production is not completely prevented, the benefits of this agent are dubious. In addition to this P. orobanchia itself is severely affected by the hymenopterous parasites Chalcidoidea and Braconidae (particularly Opius occulisus) and also by Cladosporium cladosporioides and various species of Fusarium (Horváth, 1987).

Louarn et al. (2012) have demonstrated that the arbuscular mychorrhizal fungus Rhizophagus irregularis can significantly reduce infestation of sunflower by O. cumana, by directly and indirectly reducing its germination.

Chemical Control

Garcia-Torres et al. (1994) demonstrated the selectivity of imazethapyr, imazapyr and chlorsulfuron as pre-emergence herbicides for O. cumana but results where dependent upon good soil moisture. Imazapyr was later shown to be selective also as a post-emergence treatment (Garcia-Torres et al., 1995) but it appears imazethapyr may be the most useful, applied as two or three repeated post-emergence treatments (Kleifeld et al., 1998). The related imazapic is also effective (Aly et al., 2001). Demirchi et al. (2003) found a post-emergence application of imazapic + imazapyr to be most effective at the 6-8-leaf stage. Eizenberg et al., (2009) confirmed the efficacy of imazapic applied at the eight true leaf stage of the crop which prevented any further attachment of the parasite, but O. cumana already attached continued to develop and mature. Therefore there is great need for earlier application of herbicides at the time of first attachments. Ephrath and Eizenberg et al. (2010) established that optimum timing for application is 500 growing degree days from the time of sowing.

These herbicides are more fully safe and reliable if used in conjunction with sunflowers showing inherent imidazolinone resistance (Alonso et al., 1998). Introgression of genes underlying the herbicide tolerance trait from the original wild population to cultivated sunflower was successful and genetic stocks and breeding lines with imidazolinone herbicide resistance have been developed and released (Miller and Al-Khatib, 2002) for the development of commercial resistant hybrids. Some of the new lines have resistance to sulphylurea herbicides as well as the imidazolinones. In Turkey, best results have been obtained with a single foliar treatment with imazamox + imazapyr on imidazolinone-resistant sunflower plants at 8-10 true leaf stage. The treatment caused serious damage to susceptible sunflower plants but no damage was observed on the herbicide-resistant cultivars since they completely controlled O. cernua, resulting in a significant increase in sunflower seed yield (Demirchi and Kaya, 2009). Herbicide-tolerant sunflowers have gained a market share very quickly once the resistance trait was incorporated into high-performing hybrids. In some countries such as Turkey, Bulgaria and Romania, the sunflower area planted with such hybrids exceeded 25% in only three to four years after their introduction. This option is of particular value for treatment of confectionary sunflower varieties in which varietal resistance is not generally available.

Among other herbicides, trifluralin pre-planting has proved efficient in controlling O. cumana infestation in Romania (Jinga et al., 2009) and in Hungary (Horváth and Osztrogonác, 1991). Pre-emergence application of oxyfluorfen has also proved selective (Horváth and Osztrogonác 1991). Treatment of sunflowers glyphosate/ha 6-7 weeks after sowing successfully controlled O. cumana without damaging the crop, apart from slight yellowing of the leaves at the higher rate (Petzoldt and Sneyd, 1986)

Host Resistance

Control of O. cumana in cultivated H. annuus is largely based on using resistant cultivars (Molinero-Ruiz et al., 2015). The potential for resistant varieties was recognised in Russia around 1912 and there has been a continuous programme of research ever since (research conducted in Russia, Spain, Bulgaria, Romania, Yugoslavia, Turkey and France). Unfortunately there are different races of O. cumana and as new resistant varieties of H. annuus are developed, more virulent races of O. cumana appear. In the 1990s just five races of O. cumana were known (A, B, C, D and E with increasing levels of virulence) and corresponding resistance genes were identified (Or1, Or2, Or3, Or4 and Or5). These genes are all thought to be single dominant genes and there is extensive literature on resistance (Parker and Riches, 1993; Shindrova, 1994; Vranceanu and Pacureanum, 1995; Alonso 1998; Alonso, 1999; Lu YunHai et al., 1999).

In the 1990s, race F was recognised in Spain and this has now occurred in most other countries affected, while race G has more recently become apparent in several and race H now in Russia. Škoric et al. (2010) reported that although dominant genes were available for resistance to O. cumana races A to F, new virulence had then appeared in Romania, Russia, Turkey, Spain and perhaps in Ukraine but that two new restorer lines had been identified with resistance to the new races. In spite of races A to G already having been identified in many countries, there has been very little work to assess the similarity of those populations from across their distribution (Molinero-Ruiz et al., 2014). For races A to E there is a set of discriminatory lines of sunflower suggesting correspondence of these races across a geographical range. However, there is more confusion over races F and G which may not correspond in the same way. The resistance to these new forms may be recessive and not confer resistance to 'lower' virulence levels and the genes involved in these later races are also not straightforwardly dominant. The review gives an interesting table indicating the range of races currently recorded from all the affected countries. In the intensive breeding efforts to create new resistant varieties, wild species of Helianthus have been an important source of resistance genes (Christov et al., 2009).

A number of different resistance mechanisms are involved. There may be low exudation of stimulant, a barrier to penetration of the parasite, or failure of the parasite at various stages after connection is established, resulting from e.g. blockage of the vascular system (Molinero-Ruiz et al., 2015).

Resistance may also be influenced to some degree by environmental conditions. For example, resistance of the Ambar variety in Israel was effective at low temperatures, e.g. 18°C than at higher temperatures (Eizenberg et al., 2004).

The current situation is that varieties with adequate resistance are available for most regions but a continuous breeding and selection effort is required to keep pace with the development and spread of new races. In general, resistance is not available in confectionary varieties.

An interesting development has been the demonstration of induced resistance in H. annuus resulting from seed treatment with the benzothiadiazole compound known as 'BTH'. This treatment greatly reduces subsequent attack by O. cumana, apparently due to enhanced production of the phytoalexin scopoletin and/or hydrogen peroxide in the crop roots (Sauerborn et al., 2002).

IPM

Hershenhorn et al. (2006) proposed the integration of resistant lines with chemical and biological control plus sanitation for the control of O. cumana. Deep ploughing and seed cleaning have also been suggested as components of an integrated control programme (Petzoldt et al., 1994).

24/02/2016 Original text by:

Chris Parker, Consultant, UK