Chemical Mediation of Host Selection by a Biocontrol Agent

We identified the water-soluble compound from scale covers that stimulated oviposition behavior by A. melinus, as reported by R. F. Luck and N. Uygun [1].  This research was tedious because the compound occurs only at nanogram levels within the covers of scale insects, so it required the collection of thousands of covers to obtain even microgram quantities of material for chemical purification and identification.  None of this work would have been possible without the production of both scale insects and parasitoids in continuous culture by the laboratory of Robert F. Luck at UC Riverside.

Our research on this behavioral kairomone (e.g. a chemical messenger emitted by one species that benefits another species) can be separated into three components focusing upon the following questions:  1) what is the identity of the compound 2) is the compound a reliable cue for the quality of scale for utilization by A. melinus?  and 3) can we utilize the compound to improve the ability of A. melinus to provide biological control of the California red scale? 

Prior research showed the importance of the size of California red scale on the production of fecund, female A. melinus [2], but little was known about how A. melinus might identify and select larger scale hosts.  The first ~1 minute video below shows female wasps investigating potential scale hosts by lowering their antennae and passing them over the scale cover, then turning and repeating the behavior as though they were assessing the size of the scale cover.  This “drumming and turning” behavior [3] is followed by the female wasp inserting her ovipositor through the scale cover and depositing an egg on the scale body below (see also Fig. 1).  The second video shows the same drumming and turning behavior after an extract of scale covers was applied to a disk of filter paper, including an attempted oviposition through the filter paper disk.  This behavior was the basis of our bioassays for biological activity of extracts, fractions of extracts, and synthetic compounds.

Figure 1.  Adult female Aphytis melinus ovipositing through a cover of California red scale and depositing an egg on the scale body beneath the cover.

video

Figure 2. Close-up of California red scale covers and bodies.  The non-living covers have been removed from two scales to show the yellow bodies beneath.

Identification

Scale covers are not attached to scale bodies during the period of growth of the juvenile stages of the scales and are easily removed (Fig.2).  Thousands of scale covers were collected from scales reared on lemon fruit, extracted with aqueous methanol, fractionated by high performance liquid chromatography, and bioassayed.   Through several spectroscopic methods, principally proton nuclear magnetic resonance spectroscopy (Fig. 3), the biologically active compound was found to be a previously unknown ester of two ubiquitous components – the amino acid tyrosine, and the phenolic acid, caffeic acid - forming O-caffeoyltyrosine, or OCT [4, 5].  OCT was synthesized and found to be as biologically active as the natural compound [5]. 

Figure 3. ¹H NMR spectrum (in CD₃OD) and assignment of protons to the structure of O-caffeoyltyrosine.  Protons marked with asterisks were lost due to exchange.  From Millar & Hare 1993.

Is OCT a reliable indicator of the quality of California red scale for A. melinus?

Preliminary studies showed that A. melinus showed a narrow, curvilinear response to increasing concentrations of OCT (Fig. 4) and that highest OCT concentrations occurred in covers of newly-molted third-instar scales (Fig. 5).  The concentration of free OCT concentration then declines, presumably as it polymerizes and is incorporated into the scale cover [4].  Thus, by utilizing a precursor of California red scale’s cover and responding in a dose-dependent manner, A. melinus may facilitate the identification of the most suitable scale life stages for parasitization.  However, scale body size and scale cover size vary with the citrus cultivar and substrate on which the scales are reared, and scale cover size is more variable than scale body size [6].  This suggested that any environmental factors affecting scale cover growth will affect the concentration of OCT, possibly altering the susceptibility of those scales to attack by A. melinus

Figure 4.  Attractiveness of filter-paper disks treated with the indicated doses of O-caffeoyltyrosine to A. melinus.  Bars indicate the mean (+ standard error) percentage of drums and turns (filled bars) and oviposition probes (open bars) at each dosage.  From Hare et al. 1993.

Figure 5.  Concentration of O-caffeoyltyrosine in California red scale covers as a function of development in degree-days and size.  Open circles and dashed line: mean (+ SE) O-caffeoyltyrosine concentration; filled circles and solid line: mean (+ SE) scale cover area.  From Hare et al. 1993.

In a series of field experiments, both scale cover area and 0-caffeoyltyrosine concentrations were only qualitatively related to the body size of third instar scales, the most suitable stage for A. melinus [7, 8].  Scale cover area and 0-caffeoyltyrosine concentrations were reduced, relative to scale body size, when scale were reared on bark and leaves compared to fruits. Scale cover area and OCT concentration were also relatively reduced when scales were reared in mid-summer compared to spring and fall, and when reared on orange cultivars compared to lemon cultivars in the field (Fig. 6). Overall, scales appear to be chemically conspicuous to A. melinus for a short period of the time in which they are physiologically susceptible, but scales of similar quality for utilization by A. melinus may differ in conspicuousness to A. melinus due to variation in OCT concentrations.  Scales that minimize their cover size and maximize the incorporation rate of OCT into covers may minimize their conspicuousness to A. melinus. These results are consistent with an evolutionary tension between the emitter and receiver of kairomones such that although receivers such as A. melinus may be under strong selection pressure to recognize kairomones, emitters like California red scale should be under strong selection pressure to minimize the production or persistence of those kairomones.

Figure 6.  Mean + SE scale cover area and OCT concentration adjusted for scale body size for scales reared on  fruit, leaves, or bark of lemon or orange.  From Hare & Morgan 2000.

Can OCT be used to improve biological control of California red scale?

Aphytis melinus reared in commercial insectaries and released to control California red scale are reared on oleander scale, themselves reared on squash.  Oleander scale do not produce any OCT.  Like many wasps, the early experiences of adult A. melinus can greatly influence future foraging decisions, and the earliest experience of adult A. melinus reared on oleander scale is the absence of OCT.  Wasps that were removed from their hosts as pupae and allowed to emerge as adults isolated from their hosts retained a strong preference for California red scale regardless of rearing host (Fig. 7).  This preference was reduced if wasps were allowed to emerge from oleander scale, thus acquiring early adult experience with oleander scale (Fig. 8).  The preference for California red scale was restored by exposing wasps reared on oleander scale to synthetic OCT prior to bioassay [9].  Thus, it may be possible to improve the host recognition skills of A. melinus reared on oleander scale by exposing them to OCT prior to release.

Figure 7.  Oviposition probes (mean % + SE) by A. melinus on California red scale covers when wasps were reared on California red scale on lemons (black bars) or on oleander scale reared on either lemons (grey bars) or squash (open bars).  Naïve wasps were isolated from their hosts as pupae so they had no adult experience with their rearing host.  Experienced wasps were allowed to emerge naturally from their host and therefore acquired early adult experience with their host.  From Hare 1996.

Figure 8.  Oviposition probes (mean % + SE) by A. melinus on California red scale covers when reared on oleander scale.  “Oleander reared but Naïve” wasps were isolated as pupae and had no adult experience with oleander scale, California red scale, or OCT prior to bioassay.  “Oleander Emerged” wasps emerged from oleander scale and acquired early adult experience with oleander scale, then were exposed to filter paper disks treated with solvent prior to bioassay.  “Oleander Emerged and OCT-Conditioned” wasps also emerged naturally from oleander scale but then were exposed to OCT-treated filter paper disks prior to bioassay.  From Hare 1996.

We carried out a  field experiment using isofemale genetic lines of A. melinus carrying distinguishable electrophoretic alleles (Strains “A” and “B”) to determine the number of progeny from A. melinus reared on oleander scale only, and A. melinus reared on oleander scale but conditioned to OCT prior to release.  In half of the releases, Strain ‘A’ was conditioned to O-caffeoyltyrosine and Strain ‘B’ served as the control, while this pattern of conditioning was reversed for the other half of the releases.  Although there were differences in the rates of population growth of the two strains, whichever strain was the conditioned strain left 6 – 12% greater than expected number of progeny compared to the control strain in all comparisons (Fig. 9) [10].

Figure 9.  Mean (+ SE) percent Strain “A" and Strain “B" progeny of A. melinus recovered from a field conditioning experiment.  Horizontal lines (+ binomial SE) indicate the expected proportion of Strain “A” (38.4%) and Strain “B” (61.6%) in the absence of conditioning (see Hare et al. 1997 for more details).  Rearing Site abbreviations: LF, lemon fruit; OF, orange fruit; LL, lemon leaves; OF, orange leaves.  From Hare et al. 1997.

Although large numbers of wasps can be conditioned simply by spraying the parasitized oleander scales  on banana squash with a dilute OCT solution [11], the relatively imprecise recommendations for the number of A. melinus to release suggest that a ~10% increase in host finding by A. melinus may be overwhelmed by other factors affecting the interaction between California red scale and A. melinus in commercial citrus groves.  If release recommendations are improved and become more closely linked to densities of California red scale, then a ~10% improvement in host finding might prove to be economically beneficial.

1 Luck, R.F. and Uygun, N. (1986) Host recognition and selection by Aphytis species: Response to California red, oleander, and cactus scale cover extracts. Entomol. Exp. Appl. 40, 129-136.  DOI:  10.1111/j.1570-7458.1986.tb00493.x

2 Hare, J.D. and Luck, R.F. (1991) Indirect effects of citrus cultivars on life history parameters of a parasitic wasp. Ecology 72, 1576-1585.  DOI:  10.2307/1940957

3 Luck, R.F., et al. (1982) Host selection and egg allocation behaviour by Aphytis melinus and A lingnanensis: Comparison of two facultatively gregarious parasitoids. Ecol. Entomol. 7, 397-408.  DOI: 10.1111/j.1365-2311.1982.tb00682.x.

4 Hare, J.D., et al. (1993) A caffeic acid ester mediates host recognition by a parasitic wasp. Naturwissenschaften 80, 92-94.  DOI: 10.1007/BF01140427

5 Millar, J.G. and Hare, J.D. (1993) Identification and synthesis of a kairomone inducing oviposition by parasitoid aphytis melinus from California red scale covers. J. Chem. Ecol. 19, 1721-1736.  DOI: 10.1007/bf00982303

6 Hare, J.D., et al. (1990) Variation in life history parameters of California red scale on different citrus cultivars. Ecology 71, 1451-1460.   DOI: 10.2307/1938282.

7 Hare, J.D. and Luck, R.F. (1994) Environmental variation in physical and chemical cues used by the parasitic wasp, Aphytis melinus, for host recognition. Entomol. Exp. Appl. 72, 97-108.  10.1111/j.1570-7458.1994.tb01807.x

8 Hare, J.D. and Morgan, D.J.W. (2000) Chemical conspicuousness of an herbivore to its natural enemy: effect of feeding site selection. Ecology 81, 509-519. DOI: 10.2307/177444

9 Hare, J.D. (1996) Priming aphytis: behavioral modification of host selection by exposure to a synthetic contact kairomone. Entomol. Exp. Appl. 78, 263-269.  DOI: 10.1111/j.1570-7458.1996.tb00790.x

10 Hare, J.D., et al. (1997) Increased parasitization of California red scale in the field after exposing its parasitoid, Aphytis melinus, to a synthetic kairomone. Entomol. Exp. Appl. 82, 73-81.  DOI: 10.1046/j.1570-7458.1997.00115.x

11 Hare, J.D. and Morgan, D.J.W. (1997) Mass-priming Aphytis: Behavioral improvement of insectary-reared biological control agents. Biol. Control 10, 207-214.  DOI: 10.1006/bcon.1997.0565