Cowpea chloroplastic ATP synthase is the source of multiple plant defense elicitors during insect herbivory

Abstract

In cowpea (Vigna unguiculata), fall armyworm (Spodoptera frugiperda) herbivory and oral secretions (OS) elicit phytohormone production and volatile emission due to inceptin [Vu-In; (+)ICDINGVCVDA(-)], a peptide derived from chloroplastic ATP synthase gamma-subunit (cATPC) proteins. Elicitor-induced plant volatiles can function as attractants for natural enemies of insect herbivores. We hypothesized that inceptins are gut proteolysis products and that larval OS should contain a mixture of related peptides. In this study, we identified three additional cATPC fragments, namely Vu-(GE+)In [(+)GEICDINGVCVDA(-)], Vu-(E+)In [(+)EICDINGVCVDA(-)], and Vu-In(-A) [(+)ICDINGVCVD(-)]. Leaf bioassays for induced ethylene (E) production demonstrated similar effective concentration(50) values of 68, 45, and 87 fmol leaf(-1) for Vu-In, Vu-(E+)In, and Vu-(GE+)In, respectively; however, Vu-In(-A) proved inactive. Shortly following ingestion of recombinant proteins harboring cATPC sequences, larval OS revealed similar concentrations of the three elicitors with 80% of the potential inceptin-related peptides recovered. Rapidly shifting peptide ratios over time were consistent with continued proteolysis and preferential stability of inceptin. Likewise, larvae ingesting host plants with inceptin precursors containing an internal trypsin cleavage site rapidly lost OS-based elicitor activity. OS containing inceptin elicited a rapid and sequential induction of defense-related phytohormones jasmonic acid, E, and salicylic acid at 30, 120, and 240 min, respectively, and also the volatile (E)-4,8-dimethyl-1,3,7-nonatriene. Similar to established peptide signals such as systemin and flg22, amino acid substitutions of Vu-In demonstrate an essential role for aspartic acid residues and an unaltered C terminus. In cowpea, insect gut proteolysis following herbivory generates inappropriate fragments of an essential metabolic enzyme enabling plant non-self-recognition.

Figure 1.

HPLC purification of Vu-^GE+In from cowpea-derived armyworm OS. A previously uninvestigated SCX HPLC fraction (3–4 min; Schmelz et al., 2006) inducing E production in cowpea leaves was sequentially purified by: A, RP-C18; B, GF; and C, NP chromatography. UV traces (λ = 200 Å) are overlaid on an arbitrary scale. Active fractions (*) were sequentially collected, desalted, evaporated, and resolubilized in water for bioassays. C, Final purification resulted in fractions (*) used for MS, Edmund N-terminal sequencing, and confirmation by comparison with synthetic peptides.

Figure 2.

Purified OS contain a mixture of proteolysis products related to inceptin. LC/MS selected [M + H]⁺ m/z ion traces of inceptin-related peptides copurified with Vu-In and Vu-^GE+In from armyworm OS. A, NP-LC/MS of the isolated Vu-In natural product reveals the predominant m/z [M + H]⁺ ion 1,119.5 (Vu-In) and trace amounts of 1,048.5, 1,248.5, and 1,305.5, representing Vu-In^−A, Vu-^E+In, and Vu-^GE+In, respectively. B, Analysis of purified Vu-^GE+In demonstrates the primary m/z [M + H]⁺ ion 1,305.5 and trace amounts of 1,248.5 (Vu-^E+In) and 1,119.5 (Vu-In).

Figure 3.

Activity of inceptin-related peptides requires conservation of the C terminus. Average (n = 4, ±SEM) dose responses of induced E production in cowpea leaves stimulated by the four biochemically characterized inceptin-related peptides Vu-In^−A, Vu-In, Vu-^E+In, and Vu-^GE+In. EC₅₀s of Vu-In, Vu-^E+In, and Vu-^GE+In were calculated to be 68, 45, and 87 fmol leaf⁻¹, respectively. Symbols for treatments are denoted in the legend.

Figure 4.

Activity and elicitor abundance in HPLC-purified OS from armyworm larvae collected at different times after consuming GST-GmIn8H precursor proteins. A and C, Induced E production in cowpea leaves stimulated by fractions of NP-HPLC-purified OS originally collected at 2 and 6 h after the initiation of larval feeding on GST-GmIn8H. B and D, LC/MS analysis of m/z [M + H]⁺ ion relative abundance of 1,105.5 (Gm-In), 1,234.5 (Gm-^E+In), and 1,291.5 (Gm-^GE+In) in samples corresponding to A and C, respectively.

Figure 5.

Inceptin exhibits preferential persistence in armyworm larvae OS. Average (n = 4, ±SEM) picomoles per larvae of inceptin-related fragments Gm-In, Gm-^E+In, Gm-^GE+In, and Gm-In^−A present in OS collected at times designated as 0, 2, 4, 6, and 8 h following the complete consumption of GST-GmIn8H proteins (4.8 μg). Symbols for individual peptides analyzed are denoted in the legend.

Figure 6.

OS from spinach-fed larvae rapidly loses E-inducing activity. Average (n = 4, ±SEM) induced E production in cowpea leaves treated with OS collected from larvae 0, 1, 2, or 4 h after being removed from cowpea and spinach plants. Unlike cowpea, spinach harbors a trypsin-sensitive Lys (K) within the predicted inceptin sequence [⁺ICDINGKCVDA⁻]. A damage plus water-only treatment was included as a negative control. Different letters (a and b) represent significant differences (all ANOVA Ps < 0.001, Tukey's test corrections for multiple comparisons [P < 0.05]).

Figure 7.

Sequential induction of phytohormones and volatiles induced by cowpea-derived armyworm OS. Average (n = 5, ±SEM). A, JA. B, E. C, SA. D, DMNT levels in undamaged control cowpea leaves or those damaged and treated with 1 μL armyworm OS collected from larvae feeding on either cowpea roots or shoots. Symbols for treatments are denoted in the legend. Shoot-derived OS contained 937 ± 54 fmol μL⁻¹ (n = 4, ±SEM) of total active inceptin-related peptides (Vu-In, Vu-^E+In, and Vu-^GE+In), while corresponding root-derived OS peptides were undetectable (<10 fmol μL⁻¹). Within figures, different letters (a–e) represent significant differences (all ANOVA Ps < 0.001, Tukey's test corrections for multiple comparisons [P < 0.05]).

Figure 8.

Ala substitution of inceptin confirms essential role of the C terminus for activity. Vu-In was sequentially substituted with Ala from the N terminus (A1) to the penultimate C-terminal residue (A10). C and N termini of Vu-In were also examined by either adding (Vu-In^+A) or removing residues (Vu-In^−A, Vu-^I−In). Average (n = 4, ±SEM) induced leaf production of: A, E (1 h); B, SA (4 h); and C, DMNT (4 h) in cowpea leaves treated as either undamaged controls (Con), damage + 5 μL water (Dam), or damage plus 4 pmol of peptide in 5 μL water. Within figures, different letters (a–f) represent significant differences (all ANOVA Ps < 0.001, Tukey's test corrections for multiple comparisons [P < 0.05]).

Figure 9.

Simplified proposed model for inceptin-related peptides in mediating interactions between armyworm larvae and cowpea. (1) Larvae consume cowpea leaves and produce digestive fragments of cATPC; (2) plants indirectly perceive attack when inceptin-related peptides recontact the wounded leaf surface and bind a putative receptor; (3) multiple signaling pathways are activated that include the phytohormones JA, E, and SA; (4) biochemical defense responses are induced, including protease inhibitor transcripts (PI), phenylpropanoids (cinnamic acid [CA]), and volatiles including MeSA and DMNT (Schmelz et al., 2006); and (5) insect-induced plant volatiles are released, providing reliable cues that can facilitate attraction of natural enemies.