Introduction

Aphids feed on the fluid in plant phloem for their survival. The high reproductive capacity makes many aphid species became serious agricultural pests in the world1. Aphids vary in their host range from oligophagous to polyphagous. The pea aphid, Acyrthosiphon pisum (Harris), is specialized on only some species of legumes and is considered as an oligophagous specie2; whereas some species, such as the green peach aphid, Myzus persicae (Sulzer), feeding on over 400 plant species in nearly 50 plant families, is a polyphagous species3.

Host plants of aphids display many defenses from genetic modifications to chemical changes, which could be directly harmful for aphids or attract natural enemies4. Meanwhile, aphids develop various adaptations such as using immune system to manage secondary metabolites and against its natural enemies5,6,7,8,9. In this typical tritrophic system, effects of host defense metabolites are normally negatively against aphids and positively benefit to aphid’s natural enemies4.

There is also another relationship existing among herbivores, their hosts and natural enemies. Many herbivorous insects can use plant-synthesised and normally insect-harmful chemicals to benefit themselves by ingesting and storing them in their body tissues or integument, which is known as sequestration to herbivorous pests10. More than 250 insect species in at least 6 orders have been recorded to sequester their hosts’ metabolites from at least 40 plant families10. For example, it has been found that pyrrolizidine alkaloids in leguminous plants can be acquired by a lepidopteran, Utetheisa ornatrix (L.) and then be carried to the moth’s eggs to protect them from the damage of an egg parasitoid, Trichogramma ostriniae Pang et Chen and a green lacewing, Ceraeochrysa cubana (Hagen)11,12,13. This host adaptation of insects is usually specific and observed in oligophagous species. The key chemicals in this relationship are alkaloids or some complex compounds such as phenolic glycosides, cyanogenic glycosides and glucosinolates and most of these chemicals are detected in relatively high amount in both host plants and herbivorous that consuming these plants10.

The broad bean, Vicia faba L., a host plant of A. pisum and other aphid species, has already been proved containing high quantity of L-DOPA (L-3,4-dihydroxyphenylalanine)14,15, which is a non-protein amino acid involving in many metabolizing of both animals and plants. In animals, L-DOPA is the precursor of the neurotransmitters dopamine, norepinephrine and epinephrine, which are collectively known as catecholamines. Many behaviuors of animals are modified by these neurotransmitters16,17,18. L-DOPA is also a key chemical involving in sclerotization and melanization of insects, influencing insect’s development and immunity9,19,20. In pharmacology, L-DOPA is used to treat Parkinson’s disease and the natural resource of the L-DOPA is present in legumes15. L-DOPA is biosynthesised via the shikimate pathway in plants as a secondary metabolite, the precursor of some alkaloids in several plants, like mescaline, berberine, morphine, heroin and papaverine; and most of them can negatively and directly affect animal nervous system21,22,23.

Legume plants contain substantial amounts of L-DOPA in the leaves and bean pods, such as V. faba, the velvet bean, Mucuna pruriens and the seeds of Cassia and Bauhinia trees, making them the most valuable commercial plants for the treatment of Parkinson’s disease14,15,22,24. The functions of L-DOPA in these plants on insects have still not been fully understood. Recent studies have shown L-DOPA as an important secondary metabolite for chemical defense against herbivores in plants25. Previous studies on a variety of bruchid species to the honey locust, Gleditsia triacanthos, seeds have shown the toxicity of L-DOPA26,27. Other evidence suggested the L-DOPA pathway acts as an anti-herbivore defense, reducing consumption by snails and isopods in the marine green alga, Ulvaria obscura28. Studies on the grain aphid, Sitobion avenae (Fabricius), indicated that the concentration of L-DOPA in wheat had a negative correlation to the intrinsic rate of natural increase rm29. In addition, it has been found that L-DOPA pathways in plants might play a role in response to stress conditions such as infection by bacteria, which might be involved in many aspects of development and growth of plants and might also modify carbohydrate metabolism in plants30.

Although L-DOPA is harmful to some herbivorous pest insects, it is also a key chemical involved in animal immune system9. For instance, aphids need to challenge natural enemies, including entomopathogens, parasitoids and predators and defense all of chemicals in host plants. These challenges have caused aphids to evolve numerous L-DOPA related defense mechanisms, such as clotting, encapsulation, phagocytosis and melanization, as well as initiate immune pathways for defense. Generally, L-DOPA was converted from tyrosine by tyrosine hydroxylase (tyrosine 3-monooxygenase, TH); it is a rate limiting step in synthesis of L-DOPA and dopamine in animals; and it could be inhibited by enzyme α-methyltyrosine (metirosine)5,6,7,9,31,32. Cellular encapsulation is a defense response against large foreign objects such as parasitoid eggs or bacteria that cannot be phagocytized by haemocytes and melanization is also involved in this process and L-DOPA is a key chemical involved7,31. The melanization reaction provides toxic chemicals like dopaquinone, which is also involved in encapsulation on the cuticle with cuticular proteins19,20,33. These reactions can cause capsules and display dark color on insect cuticle. Phenoloxidases (POs) are thought to be key enzymes in melanin synthesis which can hydroxylate monophenols and oxidise odiphenols to quinones and prophenoloxidase (proPO) is a modified form of the complement response found in some invertebrates. ProPO1 and proPO2 were found and studied in A. pisum in previous work9,33. Researches in human and other animals also showed the important relationship between melanization and ultraviolet (UVA) radiation-protection and melanics always are highly resistant to UVA radiation34,35,36,37. On the other hand, UVA irradiation can cause melanin pigment reorganization and generation in animal tissues to protect from damage of inner DNA38,39,40.

Considering the dramatically high amount of L-DOPA in broad bean with aphids on it, amino acid-compound-like L-DOPA has not been recorded in any insect sequestration. We hypothesized that L-DOPA might be stored and used by aphids in their metabolic system. Therefore, we attempted to investigate whether A. pisum could sequester L-DOPA from V. faba, and to determine how A. pisum uses this chemical to benefit itself. In this study, we analyzed the concentrations of L-DOPA and dopamine in two groups of A. pisum and their two host plants (the broad bean V. faba and the white clover Trifolium repens) and verified related genes in this pathway; and we then compared melanization and UVA protection between these two A. pisum populations fed on the two host plants.

Results

L-DOPA contents in A. pisum and two host plants

To understand the L-DOPA contents differences among our experimental plants, LC/MS results showed that the L-DOPA concentration in V. faba was more than 300 fold higher (49.750 ng/μg) than that in T. repens (0.157 ng/μg) (t = −22.571, df = 22, P < 0.0001) (Fig. 1A), which was also significantly higher (F = 102.607, df = 4, 38, P < 0.0001) than that in Capsicum annuum (0.184 ng/μg),Nicotiana tabacum (0.026 ng/μg) and Triticum aestivum (0.014 ng/μg) (Fig. S2A). The amount of L-DOPA in V. faba could reach 5% of total dry weight of the plant (Fig. 1A). EDTA extraction experiments showed about half amounts of L-DOPA (t = −13.791, df = 8, P < 0.0001, (Fig. 1D) were located in phloem which could be fed by aphid and no L-DOPA was detected in phloem of T. repens, C. annuum, N. tabacum and T. aestivum (Fig. S2D).

Figure 1
figure 1

L-DOPA contents in Vicia faba and Trifolium repens (A) and L-DOPA (B) dopamine (C) contents in Acyrthosiphon pisum reared separately on V. faba and T. repens; L-DOPA concentrations in phloem of V. faba were detected and shown in (D) L-DOPA amounts in leaves of V. faba were extracted by grinding as a positive control; survival rate of aphids treated with L-DOPA (20 mM) by feeding artificial diets. Survival rate of A. pisim fed on T. repens (nAD = 150, nDOPA = 150) was shown in (E) Survival rate of A. pisim fed on V. faba (nAD = 150, nDOPA = 150) was shown in (F).

Full size image

The contents of L-DOPA in the two groups of A. pisum were significantly different and the concentration of L-DOPA in the aphids fed on V. faba (43.170 ng/mg) was 25-fold higher than that A. pisum fed on T. repens (1.745 ng/mg) (t = −7.959, df = 7, P < 0.0001) (Fig. 1B). Dopamine concentrations in the two aphid groups were not significant different (2.982 and 4.151 ng/mg, respectively) (t = −2.000, df = 26, P = 0.056) (Fig. 1C).

The contents of L-DOPA and dopamine from three strains of M. persicae, a V. faba adapted-strain, a C. annuum adapted strain and a N. tabacum adapted strain and S. avenae and the bird cherry-oat aphid, Rhopalosiphum padi L., reared on T. aestivum were not significantly different (F = 72.968, df = 6, 38, P < 0.0001). Although the highest concentration of L-DOPA was found in A. pisum from V. faba, the amounts of dopamine were not significantly different among the aphids in all the other host plants (F = 2.301, df = 6, 42, P = 0.052) (Fig. S2B,C).

Toxicity of L-DOPA to Aphids

In L-DOPA toxicity analysis, A. pisum from V. faba had a 30% higher survival rate (>85%) (Fig. 1F) compared with that from T. repens (Fig. 1E). The aphids fed on an artificial diet without L-DOPA had a similar survival rate to that fed on the artificial diet with L-DOPA. The results of other experimental aphids showed that only M. persicae from V. faba could survive at a relatively low rate on the artificial diet containing L-DOPA while all other aphids could not (Fig. S3).

Concentrations of L-DOPA in different morphs and tissues of A. pisum

The L-DOPA concentrations in different aphid morphs showed significant differences (parthenogenetic female: t = 7.963, df = 7, P < 0.0001; sexual producing female: t = 11.425, df = 4, P < 0.0001; male: t = 7.310, df = 8, P < 0.0001; egg: t = 11.315, df = 4, P < 0.0001) (Fig. 2A). However, both male and parthenogenetic female contained relatively lower amounts of L-DOPA than the sexual female and eggs. The amounts of L-DOPA were significantly different among the aphid morphs (A. pisum from V. faba: F = 66.091, df = 3, 17, P < 0.0001, A. pisum from T. repens: F = 77.646, df = 3, 11, P < 0.0001). However, the eggs of the A. pisum from T. repens had the highest level of L-DOPA among the four aphid morphs.

Figure 2
figure 2

L-DOPA concentrations in different morphs of Acyrthosiphon pisum fed on two hosts, Vicia faba and Trifolium repens.

(A) L-DOPA concentrations; (B) Proportion of L-DOPA from each body part. Each value represents the mean ± SEM.

Full size image

Different tissues of the two A. pisum populations contained different amounts of L-DOPA (A. pisum from V. faba: F = 48.812, df = 3, 8, P < 0.0001, A. pisum from T. repens: F = 4.104, df = 3, 8, P = 0.049) (Fig. 2B). L-DOPA was mostly located in the head and thorax (55.11%). The distribution of L-DOPA in the A. pisum from V. faba was different from that of T. repens and the embryos (48.90%) contained the highest amount of L-DOPA among all tissues.

Effect of L-DOPA on the L-DOPA self-biosynthesis

L-DOPA concentration declined continuously after the aphids collected from V. faba had been transferred to the artificial diets (F = 13.910, df = 11, 28, P < 0.0001), but the concentration of L-DOPA in the transferred A. pisum was always higher than that in the aphids fed on T. repens than those fed on V. faba (Fig. 3A).

Figure 3
figure 3

(A) Declining trend of L-DOPA concentrations after Acyrthosiphon pisum on Vicia faba were transferred into an artificial diet and the last bar represents the L-DOPA concentrations in A. pisum fed on Trifolium Repens. (B) Relative transcript levels of TH from the aphids were reared on V. faba and the artificial diet; and (C) Relative transcript levels of TH from the aphids were reared on T. repens and the artificial diet. Each value represents the mean ± SEM.

Full size image

The endogenous production of L-DOPA in the aphids could be disrupted without L-DOPA assimilation. The transcription analysis showed that TH expression (up-regulation) in AD treatment was 20-fold higher than in the control (the aphids reared on V. faba) (t = −6.201, df = 4, P = 0.003) (Fig. 3B). On the contrary, TH expression in the A. pisum from T. repens was about 30% down-regulated in AD compared with that in T. repens (t = 0.048, df = 4, P = 0.048) (Fig. 3C).

Wound healing

In wound healing experiment with artificial diet, the aphids treated with α-methyltyrosine (A. pisum MT) had a lower proportion of melanized wounds than those fed with L-DOPA (0.5 h: P = 0.039, 1 h: P = 0.003; 1.5 h: P = 0.002; 2 h: P = 0.005) (Fig. 4A). In wound healing experiment with different hosts, The proportions of melanized wounds in the aphids fed on V. faba and T. repens were similar (Fig. 4B), although the amounts of L-DOPA were different (t = 5.145, df = 5, P < 0.0001) (Fig. 4C). The melanin in the aphids fed on V. faba was slightly darker than that in the aphids fed on T. repens in the first few hours (Fig. 4D–F). Cuticular color analysis showed significant differences in hue (t = −4.072, df = 34, P < 0.0001) and brightness (t = 6.524, df = 42, P < 0.0001) 1 hour after treatment, but not in saturation (t = 1.633, df = 34, P = 0.111) (Fig. 4F).

Figure 4
figure 4

(A) Difference in melanization efficiency of Acyrthosiphon pisum treated with L-DOPA and α- methyltyrosine; (B) Difference in melanization efficiency of A. pisum from Vicia faba and Trifolium repens; (C) L-DOPA contents from the aphids collected from two diets. Each value represents the mean ± SEM from independent determinations. (D) Wounds (pricking by glass needle) in melanization (arrows); (E) Eyes of embryos (arrow heads); (F) Color information (base on HSB, hue, saturation and brightness) in wound areas; and (G) Relative transcript level of proPO1 and proPO2 after treated by glass needle. Each value represents the mean ± SEM from independent determinations.

Full size image

The expression of melanization related genes (proPO1 and proPO2) were shown in Fig. 4G. The transcriptional levels of proPO1 and proPO2 in the wounded aphids were not different from the aphids in the control (proPO1: t = −0.196, df = 4, P = 0.335; and proPO2: t = −0.639, df = 4, P = 0.557) (Fig. 4G).

UVA Resistance

In UVA-resistances experiments. A. pisum from V. faba showed greater melanization (darker) (Fig. 5A) than that of A. pisum fed on T. repens (Fig. 5B). UVA treatments did not cause significant mortality at the beginning of treatment for the aphids that fed on the two different host plants. However, the mortality of A. pisum fed on T. repens was greater than those fed on V. faba 2 days after treatment (day1: t = 1.056; df = 6; P = 0.331; day 2: t = 2.826; df = 18; P = 0.009; day 3: t = 4.727; df = 7; P = 0.002) (Fig. 5C).

Figure 5
figure 5

Melanization level in 48 hours difference of Acyrthosiphon pisum after 365 nm Ultraviolet radiation (UVA radiation) and (A) for Vicia faba, and (B) for Trifolium repens. (C) Mortality was assessed 3 days after UVA radiation. Each value represents the mean ± SEM from independent determinations.

Full size image

The effects on L-DOPA concentrations after UVA radiation on the aphids fed on V. faba and T. repens differed dramatically (Fig. 6A). The aphids fed on V. faba contained more L-DOPA than those fed on T. repens. L-DOPA concentrations of the aphids from V. faba increased in the first 20 min, peaked in 30 min after treatment and slowly declined in the next 3 days. In contrast, the L-DOPA concentrations of the aphids from T. repens were low and stabilized for the first 12 h and then increased in the next 3 days.

Figure 6
figure 6

(A) Dynamic L-DOAP concentration during UVA radiation and 72 hours after treatment; (B) Relative transcript levels of TH after UVA radiation treatment to Acyrthosiphon pisum from two hosts in 72 hours; (C) Relative transcript levels of proPO1; and (D) proPO2 after UVA radiation treatment to A. pisum from two hosts in 72 hours. Each value represents the mean ± SEM.

Full size image

The expression levels of TH (Fig. 6B), proPO1 (Fig. 6C) and proPO2 (Fig. 6D) also changed over time. TH expression was relatively stable in the first day and changed dramatically thereafter (Fig. 6B). The expressions of TH between the aphids from the two host plants varied over time, especially 3 days after treatment. The expressions of proPO1 and proPO2 were slowly declining (Fig. 6C,D). The proPO1 expression levels between the aphids fed on V. faba and T. repens were similar at the first 12 h and then the expression level of the aphids fed on V. faba became higher than those fed on T. repens until 48 h (24 h: t = −4.108; df = 4; P = 0.015; 36 h: t = −8.355; df = 4; P = 0.001 and 48 h: t = −3.777; df = 4; P = 0.019) (Fig. 6C). However, the expression levels of proPO2 between the aphids fed on the two host plants were not significantly different, although they showed significantly higher in aphids fed on V. faba (t = 3.742; df = 4; P = 0.020) at 60 h (Fig. 6D).

UVA treatments caused deformity of aphid offspring (Fig. 7). Generally, the aphids fed on T. repens had a higher rate of deformity-descendants than those fed on V. faba, but these rates were only significantly different on day 4 (χ2 = 4.615, df = 1, P = 0.032) and on day 6 (χ2 = 4.279, df = 1, P = 0.039) (Fig. 7B).

Figure 7
figure 7

(A) Deformity-descendants laid by UV-treated mothers; (B) Deformity-descendants rate of Acyrthosiphon pisum from two hosts after UVA radiation treatment in 10 days.

Full size image

The contents of L-DOPA after the aphid host transfer, UVA and injury treatments are shown in Table 1. The contents of L-DOPA in the A. pisum that were originally reared on V. faba (high L-DOPA diet) and then were transferred to the same host (V. faba) differed significantly among the three treatments (F = 6.011, df = 2, 9, P = 0.022); the aphids on the same host had lower amounts of L-DOPA than the UVA-treated aphids, but had a similar amount of L-DOPA to the injured aphids, while the UVA-treated and injured aphids had a similar amount of L-DOPA. The contents of L-DOPA in the A. pisum that were originally reared on V. faba and then were transferred to T. repens (low L-DOPA diet) were not significantly different among the three treatments (F = 0.555, df = 2, 10, P = 0.591). Similarly, the contents of L-DOPA in the A. pisum that were originally reared on T. repens and then were transferred to V. faba were not significantly different among the three treatments (F = 0.080, df = 2, 10, P = 0.923). Again, the contents of L-DOPA in the A. pisum that were originally reared on T. repens and then were transferred to the same host (T. repens) were also not significantly different among the three treatments (F = 0.475, df = 2, 10, P = 0.636).

Table 1 L-DOPA contents analysis of Acyrthosiphon pisum in the injured, UVA-treated and untreated aphids after they were transferred from their original host plant (Vicia faba or Trifolium repens) to a different host plant (V. faba or T. repens).
Full size table

Discussion

Our results from the L-DOPA toxicity and concentration experiments confirmed that L-DOPA could be used as a defense chemical for V. faba against herbivores and L-DOPA has been proved to be sequestered by A. pisum. The results also showed that L-DOPA could be used by A. pisum as defense chemicals. The concentrations of L-DOPA varied in different morphs of A. pisum and the eggs had the highest concentration. L-DOPA can be transferred from one generation to the next and the newborn nymphs from A. pisum reared on V. faba had a high amount of L-DOPA. The results further indicated that L-DOPA might be used in wound healing of the aphids and could enhance UV-irradiation protection for the aphids as indicated by tissue melanization and L-DOPA assimilation would be enhanced as needed.

There are plenty of resistance mechanisms in legumes against herbivores41 and L-DOPA is harmful to some herbivores in some legumes and marine plants26,27,28. Our data proved that L-DOPA might have similar functions in V. faba. The mortality of A. pisum from T. repens showed that this aphid population had lost some of their adaptation and resistance to L-DOPA. A trace amount of L-DOPA was also detected in C. annuum, N. tabacum and T. aestivum, but these amounts are not high enough to cause any negative effect to herbivores.

Chemical sequestration in herbivores has been previously proved10. In our study, L-DOPA concentration (L-DOPA sequestration) in the tested aphids varied and this chemical sequestration was specific to A. pisum. A. pisum fed on V. faba contained the highest amount of L-DOPA. We found that in the A. pisum fed on V. faba, only a small part of L-DOPA was involved in dopamine metabolizing pathway and turned into dopamine for behavior modification or in melanization for cuticle development and immune system and similar results have been reported in the literature9,16,19,42. This phenomenon suggested that A. pisum could sequester L-DOPA and may have a L-DOPA throttle system to avoid extra L-DOPA over-converting into downstream products and protecting normal physiological activities from being disrupted. In the previous works of chemical sequestration, most of the sequestered chemicals are externally-derived and relatively large molecules such as toxic alkaloids and these chemicals mostly have specific functions10. Because L-DOPA originally exists in aphids and other animals, it might be more compatible to animal metabolic system. L-DOPA, as a non-protein amino acid showing both positive and negative effects, is built by a relatively simple molecule structure and it could be controlled or modified by adapted-aphids for wide uses.

Based on the preliminary work of L-DOPA and dopamine distribution in A. pisum collected from the two hosts (Fig. 2B), we found that the embryos had the most L-DOPA in the A. pisum fed on V. faba. This evidence revealed that A. pisum fed on V. faba could be born with a high level of L-DOPA and maternally transmitted to offspring during the embryo’s development. This indicates that L-DOPA adaptation system may be well developed at the very beginning of the A. pisum life-cycle. We found that most L-DOPA was located in the head in the A. pisum fed on T. repens (Fig. 2B) and the reason could be that most L-DOPA was synthesized by dopaminergic neurons in its brain without external assimilation. Sexual females and eggs contained more L-DOPA in the A. pisum fed on both V. faba and T. repens, indicating that L-DOPA might have specific functions in the eggs. Based on the tanning phenomenon in eggs, the high amount of L-DOPA in eggs could be used in melanization for absorbing heat and sclerotization and for preventing the attacks from natural enemies. These results were similar to Eisner’s work between pyrrolizidine alkaloids in leguminous plants and the moth U. ornatrix, which has been reported that some plant defense chemicals are concentrated and transferred from mothers to eggs for egg-protection11,12,13. However, more researches are needed to confirm the function of L-DOPA in aphid eggs.

After using an artificial diet to disturb the assimilation of L-DOPA in A. pisum, the biosynthesis of L-DOPA was enhanced by up-regulation of related gene TH. These results revealed that the aphids fed on V. faba would try to stabilize the amount of L-DOPA while its assimilation disrupted. The concentration of L-DOPA in 5 days after treatment revealed that L-DOPA in A. pisum was in a dynamic balance (Fig. 3A) and could decline without continuous accumulating. Compared with expression levels of TH of A. pisum fed on V. faba, TH of A. pisum fed on T. repens was quite different. On the contrary, the transcriptional level of TH in the aphid feeding on the artificial diet was down-regulated. We suspect that A. pisum in high L-DOPA diet conditions might have L-DOPA stabilize system and this functions may be lost if the aphids are transferred into low L-DOPA diet conditions after they adapted there.

Melanization and encapsulation in A. pisum have been previously reported6,7,8. The dark-wounds healing experiment showed different melanization efficiency with or without L-DOPA and these results verified the correlation between L-DOPA storage and melanization in the wounds (Fig. 4A). Melanization reactions in A. pisum wounds from the two host plants were relatively identical, but we observed that melanization in A. pisum fed on T. repens was not as dark as that in the aphids from V. faba at the very beginning of the experiment and this phenomenon might be caused by the differences in L-DOPA quantity. Stored L-DOPA could enhance melanization in pea aphid and on the other hand, A. pisum fed on T. repens might modify the melanization strategy in order to adapt to low L-DOPA situation. Our results also showed that L-DOPA might be related to wound melanization and healing. In an early study, Laughton7 analyzed phenoloxidase (PO) activity in A. pisum and found that neither wounding nor injection of bacteria significantly increased rates of PO activity in aphid haemolymph. In contrast, we did not find any expression difference between the two selected genes proPO1 and proPO2 at the transcriptional level (Fig. 4G). The possible explanation might be the absence of peptidoglycan receptor proteins (PGRPs), which activate the PO cascade in other insects following bacterial infection6. The mechanisms of melanization phenomenon in A. pisum need to be further studied.

Previous researches have already demonstrated that UVA irradiation can cause melanization in insect tissues38,39,40. The UVA resistance is always related to melanization level in many insect species34,35,36,37. The functions of the insect L-DOPA-dopamine pathway in melanic pigmentation suggest a connection between this physical (UVA) resistance and chemical (L-DOPA) concentration. Our results showed the dynamic changes of L-DOPA level (Fig. 6A) and related genes (TH) in A. pisum after UVA- radiation (Fig. 6B). The high level of L-DOPA could benefit A. pisum via melanization to enhance the UVA-resistance. We also found that the expressions of the two proPO genes decreased slowly during all the process in two hosts. Because the expression of proPO was stable, the down-regulation of these two genes could be caused by stressful environmental conditions and the relatively high proPO1 level (Fig. 6C) in A. pisum from V. faba demonstrated that A. pisum with higher L-DOPA had relatively better condition to protect them from UVA-irradiation. Because L-DOPA has multiple functions in pigmentation, neurotransmission and immune response9,19,20,33, we believed that the high amount of L-DOPA in A. pisum could have more advantages in pathogen resistance, wound healing and immunity to other attacks or damages.

We also found that the L-DOPA assimilation would be enhanced if they had L-DOPA resource when A. pisum from V. faba was in L-DOPA consumption situation (especially in UVA-irradiation, Table 1). On the contrary, A. pisum from T. repens did not present such reaction in this situation. Although we had transferred some treated individuals to L-DOPA rich diet (V. faba), they did not show any more assimilation in L-DOPA than control (Table 1). We believed that they could synthesize L-DOPA themselves to cope this situation (Table 1). This finding revealed that A. pisum from V. faba would be flexible in L-DOPA modification but A. pisum with the same genetic background from T. repens lost it.

Conclusion

At present study, we found that A. pisum fed on a high concentration L-DOPA diet could sequester L-DOPA. As far as we have known, this was the first record in chemical sequestration of toxic, non-protein amino acids in insects. A high amount of L-DOPA in A. pisum plays important roles in wound healing and physical resistance to UVA damage. Our finding showed that sequestrated chemical such as L-DOPA could be used in resistance and healing to physical damage and they even can enhance L-DOPA assimilation when needed. This phenomenon reveals that some herbivores could effectively manage harmful host-derived chemicals and integrate them into their own system for their benefits. Although we have known that L-DOPA sequestration has provided A. pisum some advantages in ecological competition, more researches are needed to elicit the mechanisms of L-DOPA sequestration and how L-DOPA storage and throttle system work. In addition, more works are needed to confirm these assumptions that L-DOPA in aphids can be used in parasitic and pathogen resistance.

Materials and Methods

Experimental insects and plants

A green strain of the pea aphid, A. pisum, was cultured on the broad bean (Vicia faba L., var. ‘Jinnong’) under long-day conditions (16L: 8D; 20 ± 1 °C) for more than 30 generations at the Key Laboratory of Applied Entomology, Northwest A&F University, China. The aphids were then reared on the white clover Trifolium repens (var. ‘Baisanye’) for more than 30 generations in a climate chamber under long-day conditions (16L: 8D; 20 ± 1 °C) before the experiments. All aphids were reared at low density (less than 30 individuals per plant) for more than three generations before experimented.

Two other aphid species, M. persicae and S. avenae, were used for comparison in L-DOPA concentration assay and toxicity testing as described in the Supplementary Information Fig. S2.

Transcription analysis

Experimental aphids were quickly-frozen by liquid nitrogen immediately after collection. RNA was extracted with RNAiso Plus (Takara, Japan). cDNA was synthesized using a PrimeScriptTM RT reagent Kit with gDNA Eraser (Takara, Japan). Quantitative real-time PCR (qRT-PCR) was performed with SYBR® Premix Ex TaqTM II (Takara, Japan) in an IQ-5 system (Bio-Rad, USA).

Tyrosine pathway regulation in A. pisum embryonic and early nymphal stage have been studied and most key genes in the phenylalanine, tyrosine and dopamine pathways have been identified42. The primers of those genes (TH and proPO for TH and PO, respectively) were designed by Primer-BLAST of NCBI online (http://www.ncbi.nlm.nih.gov/tools/primer-blast/index.cgi?LINK_LOC=BlastHome) and were shown in Table S2.

Toxicity of L-DOPA to Aphids

In this experiment, two artificial diets were used to rear aphids: one was a normal diet without L-DOPA based on the recipe of Febvay43 (coded as AD), the other one was the same diet but with 10 mM L-DOPA added (coded as AD + L-DOPA). Wingless A. pisum adults were collected from V. faba and T. repens, and were reared on each of the two artificial diets for one week. The same artificial diets and treatments were used for wingless M. persicae. In addition, S. avenae and R. padi were treated on another artificial diet based on the studies of Deng and Zhao44. The bioassay was shown in Fig. S3F. The survival rates in different treatments were assessed with five replications.

L-DOPA and dopamine extraction and assay

Ten grams of aphids were placed in a 1.5 ml centrifuge tube. The aphids were crushed by grinding into homogenate in 1 ml extraction solution (0.05 M HCl in 50% ethyl alcohol). The homogenate was then centrifuged at 20,000 g for 10 min at 4 °C45,46. Final samples were analyzed by an LTQ XL linear ion trap mass spectrometer (Thermo Scientific, Waltham, MA, USA). L-DOPA and dopamine were scanned and fragmented using data dependent MS/MS. Masses of precursor and product ions for each amino acid were described in Table S1. Full daughter scan MS Spectra and selected ion retention time (min) of dopamine and L-DOPA were shown in Fig. S1. All data were acquired and processed using Xcalibur 2.1 software (Thermo Scientific, Waltham, MA, USA). Quantification was achieved by external standard L-DOPA and dopamine mixture of known concentrations.

Concentration assay of L-DOPA in plants

Leaves of V. faba, T. repens, C. annuum, N. tabacum and T. aestivum were freeze-dried for 36 hours in lyophilizer (MSA3.6P, Sartorius, Germany). The samples were then moved into a drying oven (Thermo, USA) and dried for 2 hours at 60 °C. After that, 10 mg of treated samples were weighed precisely by a high precision electronic balance (LL3000, Thermo, USA) at room temperature and transferred into 50 ml plastic tubes and covered. All organized samples were then used for L-DOPA and dopamine extraction and assay.

To analyze relative L-DOPA concentrations in plants phloem, 100 g of leaves of V. faba, T. repens, C. annuum, N. tabacum and T. aestivum were cut into 5 pieces and extracted in 1 ml EDTA solution (5 mM, pH = 7.0) for 24 hours (20 ± 1 °C). Extracted solutions were then filtered into a new tube; and 100 mg leaves of V. faba were extracted by grinding as a positive control47,48,49. Each treatment was replicated six times.

Concentration assay of L-DOPA and dopamine in aphids

Wingless A. pisum and M. persicae (V. faba adapted) reared on V. faba. Wingless A. pisum on T. repens, M. persicae from C. annuum and N. tabacum, Wingless S. avenae and R. padi from T. aestivum were all collected and were used for comparison. Five replicates were used in each aphid species. The samples were then used for L-DOPA, dopamine extraction and further bioassay.

L-DOPA distribution in different morphs and tissues of A. pisum

The aphids from V. faba and T. repens were reared under short-day and low temperature conditions (8L: 16D; 18 ± 1 °C) to induce sexual adults. After two generations, males, sexual females and eggs were collected and used for L-DOPA extraction and bioassay. Ten individuals of wingless adult pea aphids from V. faba were dissected into four parts, head, gut, embryo and the remainder in PBS Buffer (pH = 7.2) under a stereomicroscope (JSZ-6, Jiangnan, Jiangsu Province, China). After the heads, guts and embryos were transferred, the remained body tissues were used with PBS buffer as one sample. All samples were then moved into centrifuge tubes with 0.5 ml extraction solution per tube and processed with the method as described above and then analyzed using HPLC/MS system. Each treatment was replicated three times.

Disruption of L-DOPA assimilation in A. pisum fed on V. faba

Third instar wingless A. pisum from V. faba were reared on the artificial diets (no L-DOPA added). L-DOPA concentration was monitored in 12-h intervals for 5 days. Each treatment had three replicates.

The related gene (TH) of L-DOPA synthesis pathway was also monitored. Third instar wingless A. pisum and T. repens from V. faba fed on the artificial diets. Aphid samples were collected 2 days after treatment. Three replications were prepared for RNA extraction and TH expression measurements.

Function Analysis of L-DOPA

Analysis of melanization efficiency in wounds

A. pisum from V. faba and T. repens and L-DOPA (20 mM in the artificial diet), α-methyltyrosine (TH inhibitor, 20 mM in the artificial diet) treated A. pisum (V. faba) for 5 days were prepared for abdomen-prick experiments. Each individual was pricked with 2 to 5 wounds depending on their size using a glass needle (P-97 Micropipette Puller, Sutter, CA, USA) in pulling program: Pull = 100, VEL = 100 and Time = 100. Badly-wounded individuals were removed and the remainders were reared on a leaf of the host plant in a plastic petri dish (φ35 mm). The aphids were monitored for 3 hours, images were captured every 30 min by digital camera (Olympus pen E-P5, Olympus Corporation, Tokyo, Japan; Lens: Olympus SZ61 stereomicroscope, Olympus Corporation, Tokyo, Japan), camera parameters were set in Manual (shutter speed: 1/200; aperture: depended on lens; ISO: 200; white balance: color temp, 5600K),. The numbers of wounds were counted every 30 min and the colors (hue, saturation and brightness) in the wounds were analyzed using Photoshop CS6 (Version 13.0 × 64, Adobe, California, USA)50.

The expression level of genes proPO1 and proPO2 related to melanization in A. pisum were measured 1.5 hours after the aphids had been mechanically damaged. Each treatment had replicated three times and 10 aphids were treated each time.

UVA treatment

After many attempts, 50 third instar wingless A. pisum nymphs from V. faba and T. repens were treated under UVA (365.0 nm, WFH-204B, Shanghai, China) irradiation for 1 h. The treated aphids were then moved to their original hosts for 72 h. The aphids were assessed for body-color, melanin and mortality. L-DOPA concentrations and expression levels of relative genes (TH, proPO1 and proPO2) of the aphids were measured in 10 min-intervals for 72 hours. Thereafter, the parameters were measured for every 12 hours.

Wingless A. pisum adults from V. faba and T. repens were placed under UV-irradiation for 60 min. Each replicate had 30 individuals. After treatment, the aphids were individually reared on a leaf of their original host, which was maintained in a plastic Petri dish (φ35 mm) with 1% agar in the bottom for moisturizing. The old leaves were replaced by new ones every two days. Deformity rates in their descendants were analyzed daily for 10 days. Each treatment was replicated three times.

L-DOPA assimilation analysis of A. pisum with wounds and UV-irradiation

Four instar wingless nymphs of A. pisum from V. faba and T. repens were collected for experiments. Twenty individuals were used for each treatment and starved for 6 hours before experiments as shown in Table 1. The treated aphids (5 wounds on each aphid for the injury treatment and 30 min UVA irradiation for the UVA treatment) were transferred into a new host. The samples were then collected for L-DOPA content analysis after 3 hours. Each treatment was replicated three times.

Statistical analysis

L-DOPA concentrations in plants and aphids were subjected to analysis of a variance (ANOVA); means were separated using Duncan test at P < 0.05. All other experimental data were analyzed using Student’s t-test, or Chi-square test (version 22; SPSS Inc., Chicago, IL, USA).

Additional Information

How to cite this article: Zhang, Y. et al. Pea aphid Acyrthosiphon pisum sequesters plant-derived secondary metabolite L-DOPA for wound healing and UVA resistance. Sci. Rep. 6, 23618; doi: 10.1038/srep23618 (2016).