Introduction
The poultry red mite, Dermanyssus gallinae, is currently the most economically deleterious ectoparasite of laying hens in Europe (Chauve, Reference CHAUVE1998). Research has shown that in the UK between 60% (Fiddes et al., Reference FIDDES, LE GRESLEY, PARSONS, EPE, COLES and STAFFORD2005) and 85% (Guy et al., Reference GUY, KHAJAVI, HLALELE and SPARAGANO2004) of commercial egg-laying premises may be infected with D. gallinae, with higher mite populations typically seen in free range systems compared to cage units (Guy et al., Reference GUY, KHAJAVI, HLALELE and SPARAGANO2004; Fiddes et al., Reference FIDDES, LE GRESLEY, PARSONS, EPE, COLES and STAFFORD2005). This is of particular concern given that conventional cages will be prohibited in the EU from 2012 under EU Directive 99/74/EEC and thus the proportion of hens housed in alternative systems such as free-range is likely to increase substantially. D. gallinae prevalence in laying flocks worldwide may vary even more than that reported for the UK, from 20 to 90% depending upon the country and production system considered (Sparagano et al., Reference SPARAGANO, PAVLIĆEVIĆ, MURANO, CAMARDA, SAHIBI, KILPINEN, MUL, VAN EMOUS, LE BOUQUIN, HOEL and CAFIERO2009). Evidence for higher infestation rates in free-range systems also appears to be country-dependant on a global scale, and factors such as flock/farm size may also be important in governing infestation rates (Sparagano et al., Reference SPARAGANO, PAVLIĆEVIĆ, MURANO, CAMARDA, SAHIBI, KILPINEN, MUL, VAN EMOUS, LE BOUQUIN, HOEL and CAFIERO2009).
D. gallinae feeds on blood, preferring to feed from birds for a brief period every few days during periods of darkness, while the majority of their life-cycle is spent off the host. Populations of D. gallinae are capable of rapid expansion under the amenable conditions found in most poultry houses, where Axtell (Reference AXTELL1999) observed that complete development from egg to adult can occur in as little as two weeks. The turnover of hens in laying systems typically exceeds 70 weeks, allowing ample time for D. gallinae populations to proliferate (Höglund et al., Reference HÖGLUND, NORDENFORS and UGGLA1995). Where infestations of D. gallinae are sufficiently severe, mite feeding may result in significant stress to hens with subsequent declines in bird condition, growth rates, egg quality (through increased shell thinning and spotting) and egg production (Chauve, Reference CHAUVE1998). In extreme cases, D. gallinae population levels may be so high that anaemia and even death of hens can result (Wojcik et al., Reference WOJCIK, GREYGON-FRANCKIEWICZ, ZBIKOWSKA and WASIELEWSKI2000; Cosoroaba, Reference COSOROABA2001). It has also been reported that D. gallinae may serve as a vector for poultry pathogens and possible zoonoses such as Salmonella spp. (Valiente Moro et al., Reference VALIENTE MORO, DE LUNA, TOD, GUY, SPARAGANO and ZENNER2009), suggesting a potential risk to human health.
Economic costs associated with both control and production losses due to D. gallinae have been estimated at €130 million annually for the EU egg industry (van Emous, Reference VAN EMOUS2005), where control of D. gallinae has typically been achieved by the use of synthetic contact acaricides including carbaryl, diazinon, dichlorvos and permethin. However, the continued use of these products may be hampered by issues of mite resistance (Beugnet et al., Reference BEUGNET, CHAUVE, GAUTHEY and BEERT1997; Kim et al., Reference KIM, YI, TAK J. and AHN2004, Fiddes et al., Reference FIDDES, LE GRESLEY, PARSONS, EPE, COLES and STAFFORD2005) and decreasing product availability as a result of tighter legislation regarding their use. Indeed, in Sweden there were no registered compounds available for the control of ectoparasites at the start of the last decade (Chirico and Tauson, Reference CHIRICO and TAUSON2002). Management of D. gallinae by conventional acaricides is further hampered by the behaviour of this pest. When not obtaining a blood meal, mites retreat to small cracks and crevices within the poultry housing system, where they are effectively protected from any acaricide that exerts its effect through contact toxicity.
With increasing reports of D. gallinae resistance to synthetic acaricides and changes in legislation and production practises, it is likely that in the future many more of the world's 2.8 billion laying hens (11.7% of which are located in the EU) (Axtell, Reference AXTELL1999) will suffer as a result of D. gallinae infestation. With this in mind, it is important to identify and evaluate new approaches for the management of D. gallinae in poultry systems. Indeed, in 2006 an international seminar was convened to allow D. gallinae researchers to discuss current and future research and highlight knowledge gaps in areas such as mite life cycle, production impact, monitoring and control (Mul et al., Reference MUL, VAN NIEKERK, CHIRICO, MAURER, KILPINEN, SPARAGANO, THIND, ZOONS, MOORE, BELL, GJEVRE and CHAUVE2009). Research in the areas of novel acaricides, immunological control, biological control and animal/premise husbandry are all areas in which research is being undertaken in an attempt to control D. gallinae. Whilst promising results have been reported in all these areas, work to combine management techniques in an integrated approach to D. gallinae control remains wanting. Research focusing on crop pests suggests that such an Integrated Pest Management (IPM) approach which combines control strategies can have numerous benefits over conventional control (i.e. the use of synthetic pesticides alone) (Dent, Reference DENT1995).
The aim of this paper was to review the main alternatives to synthetic D. gallinae control currently being researched and developed and to assess if these would lend themselves to combination with one-another, as well as conventional synthetic acaricides, to achieve an integrated approach to D. gallinae management.
Novel acaricides
bio-pesticides
Bio-pesticides can comprise either bacterial endotoxins or the bacteria themselves. The bacteria Bacillus thuringiensis, which has known pathogenicity in insects, has been shown to increase mortality in ticks (Hassanain et al., Reference HASSANAIN, GARHY, ABDEL-GHAFFAR, EL-SHARABY and MEGEED1997) and has been used for the control of mites of agricultural crops (van der Geest et al., Reference VAN DER GEEST, ELLIOT, BREEUWER and BEERLING2000; Dolinski and Lacey, Reference DOLINSKI and LACEY2007). Biological insect toxins are now commonly used as crop sprays and have even been incorporated into crop plants via genetic engineering, with B. thuringiensis (Bt) and Bt maize being undoubtedly the best-known examples. Spinosad, a natural product derived from fermentation of the micro-organism Saccharopolyspora spinosa, is another such biopesticide that has been used successfully against a range of insect pests, with corresponding research also finding that this product possesses relatively low toxicity to mammals and birds with 70-90% of beneficial insects also left unharmed by exposure (Anastas et al., Reference ANASTAS, KIRCHCHOFF and WILLIAMSON1999). Although the toxicity of spinosad to mites per se has been reported as being variable and/or reduced in comparison to insect species (Villanueva and Walgenbach, Reference VILLANUEVA and WALGENBACH2006; Holt et al., Reference HOLT, OPIT, NECHOLS and MARGOLIES2006), experiments using this product against the ectoparasitic cattle tick Boophilus microplus have produced some promising results, although repeat applications might be necessary according to the work conducted (Davey et al., Reference DAVEY, GEORGE and SNYDER2001, Reference DAVEY, MILLER, GEORGE and SNYDER2005). Pathogenic bacteria of arthropods could provide an alternative means of mite control. This suggests that spinosad, along with other ‘new-generation’ bio-pesticides, may warrant investigation as premise sprays to manage D. gallinae populations.
inert dusts
Inert dusts primarily include diatomaceous earths containing silica (although synthetic amorphous silica-based products also exist). They absorb lipids from the epicuticle of the exoskeleton of pest invertebrates, effectively resulting in death by dehydration. Whilst inert dusts have been well studied for their pesticidal potential and may provide an attractive alternative to synthetic chemicals for pest control, the quality of the raw material has been shown to be an important factor in the efficacy of the dust (Maurer and Perler, Reference MAURER and PERLER2006). There is therefore a need to evaluate the efficacy of inert dusts under laboratory conditions to ascertain which specific products are likely to have the greatest benefit in D. gallinae management. Work by Kilpinen and Steenberg (Reference KILPINEN and STEENBERG2009) evaluated a range of commercial inert dusts against D. gallinae under laboratory conditions and similarly reported variation in efficacy depending upon the product used, as well as with variations in humidity. High humidity levels (>85%) were shown to reduce the efficacy of inert dust products, suggesting that if used in poultry units with such high humidity levels, measures such as increased application rates would be necessary.
plant-derived products
Plant-derived products may offer an alternative to synthetic acaricides for managing D. gallinae populations and recent research in this field has produced some promising results (Kim et al., Reference KIM, YI, TAK J. and AHN2004, Reference KIM, NA, YI, KIM and AHN2007; Lundh et al., Reference LUNDH, WIKTELIUS and CHIRICO2005; Maurer et al., Reference MAURER, PERLER and HECKENDORN2009, George et al., Reference GEORGE, SPARAGANO, PORT, OKELLO, SHIEL and GUY2010). Several pesticides based on plant constituents are already used widely in certain areas of pest management (Isman, Reference ISMAN2006), including against pests of veterinary significance (George et al., Reference GEORGE, GUY, ARKLE, HARRINGTON, DE LUNA, OKELLO, SHIEL, PORT and SPARAGANO2008). Products based on extracts from the neem tree (particularly its seeds), for example, are commonly employed in pest management per se. Neem oil has been reported to have biocidal effects against some 200 species of arthropod pests (Choi et al., Reference CHOI, LEE, PARK and AHN2004), including ticks (Pathak et al., Reference PATHAK, MATHUR, LATHA and JOHN2004) and D. gallinae (Lundh et al., Reference LUNDH, WIKTELIUS and CHIRICO2005). Preliminary work by Kim et al. (Reference KIM, YI, TAK J. and AHN2004) tested 56 plant essential oils for their acaricidal effect on D. gallinae. Of these, oils from bay, cade, cinnamon, clove bud, coriander, horseradish, lime dis 5F, mustard, pennyroyal, pimento berry, spearmint and thyme all gave 100% mite mortality in contact toxicity tests at a concentration of 0.07 mg of oil per cm2. Further experiments by Kim et al. (Reference KIM, YI, TAK J. and AHN2004) showed that the acaricidal effect of selected essential oils was attributable to action in the vapour phase, an observation supported in later work from the same group (Kim et al., Reference KIM, NA, YI, KIM and AHN2007). In similar work by George et al. (Reference GEORGE, SPARAGANO, PORT, OKELLO, SHIEL and GUY2010), 50 plant essential oils were assessed for their toxic effect on D. gallinae. Twenty of the essential oils chosen gave greater than 80% mite mortality over 24 hours when used at a concentration of 0.14 mg/cm³, with one in every five of the 50 essential oils used, including thyme, tea tree and garlic, giving 100% mortality.
Vaccines
Observations of mite population levels in poultry flocks infested with D. gallinae tend to suggest that birds do not develop resistance to the mites (Nordenfors and Höglund, Reference NORDENFORS and HÖGLUND2000), while levels of IgY in mite-infested poultry flocks were not related to mite levels (Arkle et al., Reference ARKLE, GUY and SPARAGANO2006). Mite-induced immuno-modulation of the host is a possibility (Jackson et al., Reference JACKSON, FRIBERG, BOLCH, LOWE, RALLI, HARRIS, BEHNKE and and BRADLEY2009) and recent research suggests that D. gallinae both adapts its reproductive behaviour to the host (Harrington et al., Reference HARRINGTON, ROBINSON and SPARAGANO2010a) and modulates the host initial inflammatory responses (Harrington et al., Reference HARRINGTON, ROBINSON and SPARAGANO2010a,b). An understanding of the immunology of the D. gallinae-host relationship is important where vaccination is proposed as a control measure, yet despite the importance of D. gallinae in poultry production there is a lack of understanding of how the mite modulates and/or stimulates host immunity.
Vaccines provide attractive alternatives to insecticides for the same reasons as they are preferable to antibiotics, including lack of residues in foodstuffs, no withdrawal periods, no environmental contamination, avoidance of resistance in target populations and ease of administration (Shryock, Reference SHRYOCK2004). Until recently, there were few reports in the literature detailing the development of a D. gallinae vaccine or prospective vaccine candidates.
Immunisation of birds with somatic D. gallinae antigens has met with variable success. Arkle et al. (Reference ARKLE, HARRINGTON, DE LUNA, GEORGE, GUY and SPARAGANO2008) found no significant difference in D. gallinae mortality when fed in vitro on blood from birds immunised with D. gallinae proteins and controls. However, both Wright et al. (Reference WRIGHT, BARTLEY, NISBET, MCDEVITT, SPARKS, BROCKLEHURST and HUNTLEY2009) and Harrington et al. (2009a) reported a significant 7.5% or 50.6%, respectively, increase in in vitro mite mortality when fed blood spiked with egg-extracted antibodies from D. gallinae immunised birds. However, both authors used different methods of protein extraction; Harrington et al. (2009a) used a urea-based method, whilst Wright et al. (Reference WRIGHT, BARTLEY, NISBET, MCDEVITT, SPARKS, BROCKLEHURST and HUNTLEY2009) identified significant mite mortality using PBS-extracted proteins but reported no significant effect of urea-extracted D. gallinae proteins on mite mortality.
A genomics approach has also been undertaken to investigate D. gallinae vaccine candidates. Bartley et al. (Reference BARTLEY, NISBET, OFFER, SPARKS, WRIGHT and HUNTLEY2009) identified an orthologue of tick histamine release factor (HRF) in D. gallinae and suggested the protein could have a regulatory function in mites. Following production of the recombinant protein Dg-HRF-1, in vitro testing against D. gallinae demonstrated a significant 7% increase in mite mortality compared to controls. Harrington et al. (Reference HARRINGTON, CANALES, DE LA FUENTE, DE LUNA, ROBINSON, GUY and SPARAGANO2009b) immunised laying hens with recombinant proteins derived from ticks (Bm86) or mosquitoes (subolesin). Bm86 has well documented activity against Boophilus microplus (de la Fuente et al., Reference DE LA FUENTE, ALMAZÁN, CANALES, PERÉZ DE LA LASTRA, KOCAN and WILLADSEN2007), although non-Boophilus spp. activity is variable (Willadsen, Reference WILLADSEN2004), whilst subolesin has been shown to be highly conserved across a number of arthropods, including ticks, and can affect blood meal digestion and reproduction in ticks (de la Fuente et al., Reference DE LA FUENTE, ALMAZÁN, BLAS-MACHADO, NARANJO, MANGOLD, BLOUIN, GORTAZAR and KOCAN2006; Kocan et al., Reference KOCAN, MANZANO-ROMAN and and DE LA FUENTE2007). Harrington et al. (Reference HARRINGTON, CANALES, DE LA FUENTE, DE LUNA, ROBINSON, GUY and SPARAGANO2009b) demonstrated a significant 35.1% increase in D. gallinae mortality in subolesin-immunised versus control birds in an in vitro assay, whilst immunisation with Bm86 resulted in a non-significant, but numerical 23% increase in mite mortality.
The development of arthropod vaccines is notoriously difficult due to the limiting step of identification and characterisation of new protective antigens, and this has restricted the availability of commercial vaccines against arthropod ectoparasites (reviewed by Willadsen, Reference WILLADSEN2004). The recent research on D. gallinae antigens has demonstrated that there is potential to develop a vaccine to control the poultry red mite. However, the approaches undertaken have so far relied upon the concealed antigen approach, whereby the host immune system is not normally exposed to mite antigens, for example gut epithelial cell, and as such the parasite has not developed hot-immune avoidance mechanisms. This approach is not without pitfalls, most notably the necessity to repeatedly vaccinate animals due to the lack of stimulation of host immunity by these concealed antigens.
Biological control
natural enemies
Natural enemies in the form of predators and parasitoids can be of great benefit in controlling crop pests (Gurr et al., Reference GURR, WRATTEN and ALTIERI2004). Where naturally occurring populations are insufficient to provide pest management benefits, mass releases of artificially reared individuals can be used to reduce pest populations. Such releases are especially effective in enclosed systems (such as greenhouses), where the natural enemies are confined to the release site and so must necessarily feed/parasitize the target pest in order to survive. Whilst little work has been done to assess the potential of predators and parasitoids to control D. gallinae populations, the fact that hens are typically housed in at least partly enclosed systems suggests that such control might be achievable if suitable candidates are identified for mass rearing and release.
Research data exists that suggest that suitable candidate predators may exist for use against D. gallinae in poultry systems. Numerous authors have reported the occurrence of the predatory mite Cheyletus eruditus in poultry houses where this species has been observed feeding on juvenile D. gallinae (Lesna et al., Reference LESNA, WOLFS, FARAJI, ROY, KOMDEUR and SABELIS2009). Releases of C. eruditus have not provided control of D. gallinae in experiments to date, although additional research has identified a further two predatory mite species (Hypoaspis aculeifer and Androlaelaps casalis) with D. gallinae management potential (Lesna et al., Reference LESNA, WOLFS, FARAJI, ROY, KOMDEUR and SABELIS2009). It is possible that other predatory species aside from mites would also prey upon D. gallinae, although mass-rearing and release of species such as predatory beetles would be more difficult to achieve. Similarly, whilst parasitoids are often mentioned with regard to crop pest control, many of these species (such as the parasitic Hymenoptera) require additional resources to survive and reproduce (such as pollen and nectar) that would not be present within a poultry unit.
entomopathogenic fungi
Entomopathogenic fungi are currently used worldwide for control of a wide range of arthropod pests, particularly pests in protected crops or field crops. D. gallinae has been found to be susceptible to infection by isolates of each of the species Beauveria bassiana, Metarhizium anisopliae and Paecilomyces fumosoroseus when mites were inoculated with high doses of conidia (Steenberg and Kilpinen, Reference STEENBERG and KILPINEN2003). In subsequent work on the same project isolates from B. bassiana consistently proved to be most virulent and persistent over time, with up to 80% transmission when 5% of mites had been treated (Steenberg et al., Reference STEENBERG, KILPINEN and MOORE2006). Nevertheless, semi-field experiments showed that although the fungus was capable of reducing D. gallinae population growth, control levels were not satisfactory (Steenberg et al., Reference STEENBERG, KILPINEN and MOORE2006). This probably results from the fact that the time required to kill D. gallinae was sufficiently long to allow females to oviposit, where treatment was reported not to have effected oviposition (Steenberg et al., Reference STEENBERG, KILPINEN and MOORE2006). Tavassoli et al. (Reference TAVASSOLI, OWNAG, POURSEYED and MARDANI2008) reported that while three strains of M. anisopliae were effective in killing adult D. gallinae, pathogenicity against different mite developmental stage varied according to fungal strain as well as being dose and time dependent. Use of entomopathogenic fungi may also be dependent on maintaining adequate humidity levels to ensure fungal transmission and it remains to be seen if this method can be successfully developed and employed in the field for D. gallinae control.
Animal husbandry
Higher mite populations are typically seen in free range systems compared to cage units (Guy et al., Reference GUY, KHAJAVI, HLALELE and SPARAGANO2004; Fiddes et al., Reference FIDDES, LE GRESLEY, PARSONS, EPE, COLES and STAFFORD2005; Arkle et al., Reference ARKLE, GUY and SPARAGANO2006) probably as these systems provide both a more favourable habitat for the mites and a greater challenge for decontamination between flocks. Such systems often contain a large number of sites highly suited as D. gallinae refugia. When not obtaining a blood meal, mites retreat to these small cracks and crevices where they can easily survive unfed for many months in areas that are difficult to clean down between flocks. It should be possible to design poultry facilities that are less ‘mite-friendly’ than is currently the norm, although removing all refugia is clearly unrealistic. Recent research suggests that using a Hazard Analysis and Critical Control Point method (HACCP) can be potentially beneficial in preventing D. gallinae establishment in all types of hen housing where Mul and Koenraadt (Reference MUL and KOENRAADT2009) have adapted this technique from that first development by NASA in the 1960s. Forty-one potential infestation hazards are cited, such as introduction and spread of mites by birds, rodents and employees, with suggested corrective actions to minimise D. gallinae population establishment. The system has been trialled in Holland and the UK and described as a useful and feasible management tool by poultry farmers.
Mul and Koenraadt (Reference MUL and KOENRAADT2009) noted the potential of temperature manipulation to control D. gallinae, where heating the house between flocks to 55°C was suggested as a control measure. D. gallinae population development is known to be temperature dependant, where research suggests that the stage-specific survival of immature mites decreases rapidly outside the range of 10-37°C (Maurer and Baumgärtner, Reference MAURER and BAUMGÄRTNER1992). Tucci et al. (Reference TUCCI, PRADO and ARAÚJO2008) similarly reported high mortality of D. gallinae at 35°C indicating that this temperature had adverse effects on mite development. Complete mortality of D. gallinae can be expected at greater than 45°C and below -20°C (Nordenfors et al., Reference NORDENFORS, HÖGLUND and UGGLA1999). This suggests that temperatures of 55°C, which would be expensive to reach and maintain for any length of time, may be excessive for controlling D. gallinae and that similar levels of control might be achievable at a reduced temperature (although this might need to be maintained for longer).
It may be possible to control D. gallinae populations by manipulating lighting regimes within poultry units. Research suggests that short-cycle intermittent lighting regimes can markedly reduce D. gallinae numbers compared with more standard regimes, although lighting intensity (Zoons, Reference ZOONS2004; Stafford et al., Reference STAFFORD, LEWIS and COLES2006). This is likely caused by disruption to the mites’ normal nocturnal feeding behaviour, although exact reasons remain unclear. However, current EU legislation on hen welfare requires a statutory 8 hour dark period, making it difficult to envisage how intermittent lighting regimes could be employed in practise for D. gallinae control.
IPM potential for Dermanyssus gallinae control
Many of the different D. gallinae management approaches previously described would be amenable to use alongside one-another, as well as in combination with conventional currently used hygiene and chemical control methods. It is likely, for example, that a D. gallinae vaccine could, and depending upon the antigen it might be preferable, to use in conjunction with any other control method described. Similarly, the use of novel pesticides would unlikely be unaffected by advances in animal husbandry techniques and vice versa, with both being compatible with conventional D. gallinae control. Nevertheless, such compatibility would not be true of all the D. gallinae control methods described.
Novel pesticides may be just as toxic to invertebrate natural enemies as they are to the target pest. Inert dusts work via the invertebrate cuticle and so can be expected to exert a broad-spectrum effect on all invertebrates. Plant essential oils are thought to disrupt binding of invertebrate nerve cord proteins, specifically 3H-octopamine (according to work done with the American cockroach Periplaneta Americana) (Enan et al., Reference ENAN, BEIGLER and KENDE1998). As such, any product giving high levels of mortality in D. gallinae might similarly be expected to do so in predators and parasitoids (also possessing an invertebrate-specific octopaminergic nervous system), although some work exists to suggest that toxicity to invertebrates may depend on the species considered (Isman, Reference ISMAN2000; George et al., Reference GEORGE, SPARAGANO, PORT, OKELLO, SHIEL and GUY2009). Bio-pesticides may be more species specific and amenable to use alongside natural enemy releases. For example, research suggests that predatory invertebrate species may be relatively tolerant to spinosad, although parasitoids are at much greater risk from exposure (Williams et al., Reference WILLIAMS, VALLE and VIÑUELA2003). Use of invertebrate-specific bio-pesticides (and inert dusts) is also likely to be amenable with treating D. gallinae infestations with entomopathogenic fungi (although it is plausible that dusts could reduce premise humidity levels). The same may not be true for essential oil-based acaricides, however, where these products are often highly fungicidal. Similarly, animal husbandry techniques involving raising premise temperatures to levels lethal to D. gallinae between flocks would likely have an adverse effect on any beneficial organisms present making these techniques incompatible with one-another.
Conclusions
It remains to be seen how many of the new and emerging D. gallinae management techniques cited herein will be as useful in practice as preliminary results suggest. It will also be interesting to observe if any synergistic effect could be gained by combining any compatible techniques in an IPM approach to D. gallinae control. Even where techniques are amenable to combination with one-another, this will not automatically infer any pest management benefit in using multiple techniques together. In work by Maurer et al. (Reference MAURER, PERLER and HECKENDORN2009), for example, mortality of D. gallinae exposed to a diatomaceous earth was comparable with or without the addition of 2% pyrethrum extract. Whilst beyond the scope of this paper, as an integrated approach to pest management can also encompass the use of ‘conventional’ acaricides, the development of new synthetic products such as Phoxim (marketed as ‘ByeMite®’) should not be overlooked when considering IPM for the control of D. gallinae.
Although any consideration of IPM for use against D. gallinae remains in its infancy, according to pest management research in other areas, developing methods that could be combined in an integrated approach could be beneficial to D. gallinae management in the future. With continued research into new technologies and techniques for D. gallinae control an IPM approach to control this important poultry pest has the potential to be realised.
Acknowledgements
This work was partially supported financially by the European Commission through the STREP project "RESCAPE", contract no. 036018, under the 6th Framework Programme, priority 5, food quality and safety. The authors would also like to thank the John Oldacre Foundation (UK), the Yorkshire Agriculture Society (UK), the Biotechnology and Biological Sciences Research Council (UK), the Department for Environment, Food and Rural Affairs (UK), QLK2-CT-2001-01236 CHIMICO and also Food-CT-2006-035547 SAFEHOUSE, all of whom have contributed funding to the above research.
