Hydroperoxide lyase depletion in transgenic potato plants leads to an increase in aphid performance

Abstract

Hydroperoxide lyases (HPLs) catalyze the cleavage of fatty acid hydroperoxides to aldehydes and oxoacids. These volatile aldehydes play a major role in forming the aroma of many plant fruits and flowers. In addition, they have antimicrobial activity in vitro and thus are thought to be involved in the plant defense response against pest and pathogen attack. An HPL activity present in potato leaves has been characterized and shown to cleave specifically 13-hydroperoxides of both linoleic and linolenic acids to yield hexanal and 3-hexenal, respectively, and 12-oxo-dodecenoic acid. A cDNA encoding this HPL has been isolated and used to monitor gene expression in healthy and mechanically damaged potato plants. HPL gene expression is subject to developmental control, being high in young leaves and attenuated in older ones, and it is induced weakly by wounding. HPL enzymatic activity, nevertheless, remains constant in leaves of different ages and also after wounding, suggesting that posttranscriptional mechanisms may regulate its activity levels. Antisense-mediated HPL depletion in transgenic potato plants has identified this enzyme as a major route of 13-fatty acid hydroperoxide degradation in the leaves. Although these transgenic plants have highly reduced levels of both hexanal and 3-hexenal, they show no phenotypic differences compared with wild-type ones, particularly in regard to the expression of wound-induced genes. However, aphids feeding on the HPL-depleted plants display approximately a two-fold increase in fecundity above those feeding on nontransformed plants, consistent with the hypothesis that HPL-derived products have a negative impact on aphid performance. Thus, HPL-catalyzed production of C6 aldehydes may be a key step of a built-in resistance mechanism of plants against some sucking insect pests.

Oxylipins (1) are oxygenated derivatives of fatty acids that, in many cases, are involved in plant defense reactions either as modulators of gene expression, e.g., the plant hormone jasmonic acid (JA; refs. 2 and 3), or as direct deterrents to pests and pathogens, e.g., divinyl ethers (4). Oxygenation of fatty acids proceeds through distinct enzymatic activities, most of which reside in chloroplasts, and gives rise to a series of diverging metabolic pathways that eventually yield a large array of different oxygenated and nonoxygenated derivatives (5). In one of these reactions, lipoxygenases (LOX) introduce molecular oxygen to unsaturated fatty acids, such as linoleic and linolenic acids, to yield either 9- or 13-hydroperoxides. In potato, 9-LOX are prevalent in tubers, roots (6, 7), and in healthy leaves (8), whereas 13-LOX are induced in the leaves on mechanical damage (7). Further down the LOX pathway (9), 9-hydroperoxides may become substrates for divinylether synthase, resulting in the production of divinyl ethers such as colneleic and colnelenic acids. These compounds accumulate in the leaves of potato plants infected with tobacco mosaic virus or Phytophthora infestans, and have inhibitory activity toward this fungal pathogen (4). On the other hand, 13-hydroperoxides may serve as substrates for two alternative enzymatic activities, one of which, allene oxide synthase, ultimately leads to formation of the plant hormone JA that accumulates on wounding and activates defense-gene expression (10). Alternatively, hydroperoxide lyase (HPL) cleaves 13-hydroperoxides to produce traumatin (12-oxo-dodecenoic acid), the wound hormone involved in healing of damaged tissues (11), and the aliphatic aldehydes hexanal or 3-hexenal, from 13-hydroperoxy linoleic or linolenic acids, respectively. These volatile aldehydes are components of the aroma of the fruits in several plant species and are, in addition, responsible for the cut-grass odor produced when leaves are crushed (12). No other definite physiological role has been assigned to these C6-aldehydes, and thus this branch of the LOX pathway is considered to be a disposal route for the removal of toxic hydroperoxides. However, hexanal and 3-hexenal are known to have a potent antimicrobial effect (13) and to reduce aphid fecundity in vitro (14), and thus are thought to be involved in plant defense responses to pests and pathogens. Recently, it has been proposed that these aldehydes may regulate defense-gene expression as well (15).

In tomato, a close relative to potato, an HPL has been characterized to specifically utilize 13-hydroperoxides (16), and a cDNA-encoding HPL has been isolated from tomato leaves (17). In the work described here, a potato HPL cDNA has been isolated and characterized, allowing a comparison of gene expression to enzyme activity during the course of plant development and on mechanical damage. The isolated cDNA also has been used to modulate HPL activity levels in transgenic potato plants with the aim of elucidating the in vivo role of HPL and derived products in plant defense reactions.

Materials and Methods

Plant Materials and Transformation.

Potato plants (Solanum tuberosum cv. Desiree) were grown in the greenhouse and transformed as described (18). Plants were wounded and treated with JA as described (18). Transgenic lines were vegetatively propagated as explant cuttings from sprouting tubers.

Substrate Synthesis and HPL Activity Assays.

Two hydroperoxy-linolenic acids (13-HPOT and 9-HPOT, respectively) were synthesized as described (19).

Two methods were used for in vitro determination of HPL activity present in leaves and tubers of greenhouse-grown potato plants. Hydroperoxide decomposition was monitored by following the decrease in absorbance at 234 nm (19). In this spectrophotometric assay, the activity of all hydroperoxide-degrading enzymes is determined. A typical assay consisted of 1 ml of phosphate buffer (0.1 M, pH 6.8), 5 μl of either 5 mM 13-HPOT or 5 mM 9-HPOT, and 25–50 μl crude extracts obtained as described (16). The second method is based on the direct analysis of formed aldehydes by headspace-gas chromatography (19). For this purpose, a similar enzymatic reaction was carried out in sealed 12-ml glass vials for 1 min and stopped by addition of 125 μl 12 M HCl.

To determine HPL activity in vivo, 24 discs were cut with an 8-mm cork borer from different equivalent leaves of potato plants and were stored immediately in liquid N₂ until assaying. In vivo HPL activity was measured in quadruplicate by placing the leaf discs in a 12-ml vial containing 2.1 ml of phosphate buffer (0.2 M, pH 6.8) and 1 mM 13-HPOT at 1°C. The vial was sealed, incubated at 30°C for 30 min, and the reaction stopped by lowering the pH to 1.3 with 12 M HCl. Resulting aldehydes were determined as above by headspace-gas chromatography.

The endogenous volatiles present in the leaves were determined as above, but leaf discs were placed in a 12-ml vial containing 2.1 ml of phosphate buffer (0.2 M, and the pH adjusted to 1.3 with 12 M HCl) without incubation at 30°C.

Synthesis of an HPL cDNA.

Reverse transcription (RT; GIBCO/BRL) reaction was performed on total potato leaf RNA (10 μg), using HPL-3 primer (5′-CATCACA/TAG/AC/TGGCTGA/GTAACCAC-3′) designed to match the bell-pepper HPL sequence (20), for 1 h at 37°C. One-tenth of the RT reaction was used for PCR amplification with primers HPL-2 (5′-GTC/GGCC/GGTT/ACTT/GGAC/TGTCAAGTC-3′) and HPL-3, using the following cycling conditions: 4X (30 sec, 94°C; 15 sec, 40°C; 1 min, 72°C) and 35X (30 sec, 94°C; 30 sec, 53°C; 1 min, 72°C). Two bands were obtained with molecular weights of ≈950 and 900 bp, and were blunted with the Klenow fragment of DNA polymerase and cloned subsequently in the SmaI site of pBluescript II SK(−) (Stratagene). Sequence analysis revealed that only the larger PCR product, PHPL1, had high similarity to other HPLs. The second band has not been analyzed further. The 5′ and 3′ ends of the cDNA were obtained with the Marathon cDNA amplification kit (CLONTECH). By using HPL-5 (5′-GGCTTACTCCACCAGTGCCAAGTC-3′) and a nested primer HPL-7 (5′-GACTTTAAGCTGAGCTCACATG-3′), a band of ≈450 bp was obtained that was confirmed to be the 3′ end of potato HPL by sequence analysis. Similarly, the 5′ end was obtained by using HPL-4 (5′-CCGGGGAGACTGATGGATCGGCGCCG-3′) and a nested primer HPL-6 (5′-GGAGTGCAGGAAGAAGAGAAGCTTCC-3′).

For plant transformation, the 950-bp HPL cDNA fragment was cloned in antisense orientation under the control of the cauliflower mosaic virus 35S promoter and the octopine synthase 3′ terminator in a BIN19 vector (18).

RNA Analysis.

Total RNA extraction and Northern blot analysis with ³²P-labeled cDNA probes have been described (7). For analysis of HPL expression, a double-stranded HPL cDNA probe, which detected both sense and antisense transcripts, was used. Prosystemin cDNA was provided by C. A. Ryan (Washington State University, Pullman, WA). Allene oxide synthase was obtained by PCR amplification of total RNA from wounded leaves of potato (G.V. and J.J.S.-S., unpublished results). Other cDNAs used as probes are as described (7, 21).

Insect–Plant Interactions.

A laboratory clone of Myzus persicae was cultured for several generations on potato plants in a growth chamber under a 16/8-h light/dark photoperiod at 24 ± 1°C with a relative humidity of 85 ± 10%. For fecundity experiments, adult apterous aphids (less than 24-h-old) were confined singly on each of the four to five fully expanded upper leaves of wild-type and HPL-depleted potato plants by using leaf clip-on cages (22). Aphids were inspected every 3 days during the first 9 days of reproduction, and the nymphs produced by each adult were counted and removed. The position of the cage was changed to other leaflets of the same leaf after each inspection to avoid local plant damage. For nymph-development experiments, one adult apterous aphid was confined in a clip cage on the third upper leaf of potato plants, as described previously. After 24 h, the adult was removed, and cohorts of 5 newly born nymphs were retained to determine by daily inspection of caged aphids the time needed to develop into adults. For population-increase experiments, five adults were placed on each plant, which were confined by using whole-plant cages. After 5 days, progeny production and the number of the initial adults remaining on the plants were scored. All experiments were performed by using 11–12 plants per genotype. Plants were held in a growth chamber at the environmental conditions described previously and were arranged in a completely randomized block design.

Results and Discussions

13-HPL Activity in Potato Leaves.

HPL cleaves fatty acid hydroperoxide products of LOX activity. Two distinct 13-LOX genes have been shown to be expressed in the leaves of potato plants, and their expression is induced on wounding and/or treatment with JA (7). However, 9-LOX is the predominant specific activity in the leaves of nonstressed potato plants (8). To determine the fate of the fatty acid hydroperoxides produced in potato leaves, and the specificity of the ensuing reactions, extracts of nonwounded leaves were analyzed for the presence of fatty acid hydroperoxide-cleaving activity by monitoring hydroperoxide consumption. In addition, (E)-2-hexenal production was determined as a specific diagnostic of 13-HPL activity [the actual product, (Z)-3-hexenal, is isomerized readily to the more stable (E)-2 isomer during preparative procedures]. Although 9-hydroperoxides were rather stable in these crude extracts (not shown), 13-HPOT was degraded readily, with a consumption of 1,393 ± 84 nmol 13-HPOT/min/gram fresh weight. (E)-2-hexenal production was 1,454 ± 83 nmol/min/gfw, consistent with HPL accounting for essentially all 13-HPOT degrading activity present in these leaf extracts. These values are close to the 3,000 nmol/min/gfw HPL activity reported for tomato leaves (16). The data obtained suggest that an HPL activity is present in potato leaves with a marked substrate preference toward 13-hydroperoxide isomers.

A number of genes encoding 13-HPL have been reported in several plant species (for a comparison of some of the available HPL sequences, see ref. 17). Degenerate primers were designed to match conserved regions and were used to amplify an internal cDNA fragment from total RNA of potato leaves. Sequencing of the fragment revealed a high degree of similarity to HPLs from other plant sources, indicating that it was derived from potato HPL RNA indeed. Because the amplified fragment did not span the whole coding region, rapid amplification of cDNA ends (RACE)-PCR was used to produce the 5′ and 3′ ends to obtain a full-length cDNA. The amino acid sequences deduced for potato and tomato HPLs are over 90% identical.

Southern-type experiments performed under stringent conditions showed HPL hybridization to a number of bands in the potato genome (not shown), consistent with the restriction sites present in the cDNA sequence. Thus, HPL is likely a single-copy gene.

HPL Expression in Potato Plants.

The HPL cDNA was used as a hybridization probe to analyze HPL gene expression throughout the development of potato plants and on wounding and JA treatment. HPL is expressed in the leaves of healthy potato plants, with young leaves having the highest expression levels; HPL transcripts are barely detectable in older leaves (Fig. 1A). For equal amounts of total RNA, young leaves accumulate at least 10 times more HPL mRNA than older ones, as determined by densitometry of Northern blot autoradiograms (not shown). This developmentally regulated expression of HPL also has been reported to occur in the tomato (17), although the differences observed between young and old tissues were in this case less pronounced. However, these differences in transcript accumulation are not reflected in HPL activity, which is nearly identical in young and old leaves of the potato plant (a ± 10% variation in activity based on soluble protein content; not shown). This result suggests that, once synthesized, the HPL enzyme is rather stable.

Figure 1.

HPL gene expression in potato plants. HPL mRNA accumulation was determined by Northern blotting techniques, using an internal cDNA fragment as probe (HPL). Hybridization to the proteinase inhibitor II gene (PIN2) was used to monitor wound-induced gene expression. Ethidium bromide staining of rRNA was used to verify even loading of the gel. (A) Total RNA was isolated from the leaves of a healthy potato plant, starting from the top (lane 1, immature leaf) and sequentially following to the bottom (lane 7, approaching senescence). (B) Second (L2) and fifth (L5) leaves (from top to bottom) were wounded and harvested at different times thereafter (indicated in hours above the corresponding lanes). Nonwounded leaves (0) were collected as controls. (C) Total RNA was isolated from second leaves (L), and flowers at different developmental stages: small (lane 1), green (lane 2), and colored (lane 3) buds; and open (lane 4) and senescent (lane 5) flowers. RNA from potato berries (B) was included in the analysis also.

The oxylipin pathway is a component of the plant's response to stress, including the response to mechanical damage to the plant tissue (23). HPL-derived aldehydes confer the cut-grass odor to crushed leaves (12), whereas the HPL-derived oxoacid traumatin was identified as a “wound hormone,” triggering cell division at wound sites (9). Therefore, HPL expression was analyzed in wounded leaves of potato. In young leaves, HPL was expressed at similarly high levels in intact and damaged leaves at different times after wounding (Fig. 1B). In the older leaves, where the basal level of expression is low, a transient wound-induced increase in HPL-transcript accumulation could be detected peaking 4 h postwounding (Fig. 1B). In either case, HPL enzymatic activity remained nearly constant, both in young and old leaves, throughout the time course analyzed (not shown), suggesting that wound-dependent aldehyde and oxoacid production is likely caused by the mixing of enzyme and substrates resulting from the de-compartmentalization that occurs in the damaged tissues.

HPL is expressed in flowers also, with expression levels increasing along floral development and declining on senescence (Fig. 1C). As suggested for tomato (17), HPL expression in potato flowers may be related to the production of volatile compounds that serve as pollinator attractants. HPL is expressed in the berries also (Fig. 1C, lane B). In tomato, HPL is expressed in the fruits (17) where hexanal and 3-hexenal are major contributors to aroma (24). In contrast, HPL transcripts were below the limit of detection in potato tubers of different ages (not shown). Consistent with this, HPL activity in the tubers ranged from 44.5 ± 4.1 nmol/min/gfw (in small tubers) to 48.1 ± 3.7 nmol/min/gfw (in large tubers) by using 13-hydroperoxy linolenic acid as substrate, some 30 times lower than in the leaves. No HPL activity could be measured by using the 9-hydroperoxide as substrate. Potato tubers are known to have high levels of LOX activity, which has essentially a 9-LOX specificity (25). Thus it is likely that 9-hydroperoxy fatty acids are the most abundant isomers formed in tubers. Although the 13-HPL activity determined may suffice to cleave the residual 13-hydroperoxides produced in tubers, the absence of a 9-HPL activity suggests that 9-hydroperoxides have to be disposed in a different tuber-specific manner that does not require this enzymatic activity.

Antisense-Mediated Depletion of HPL in Transgenic Potato Plants.

Thus, HPL is expressed constitutively in the leaves and throughout floral development of potato plants. To elucidate the role of HPL and its products, transgenic potato plants with reduced HPL levels were generated by expression of an antisense RNA. To this end, an HPL cDNA fragment was placed in an antisense orientation under the control of the cauliflower mosaic virus 35S promoter and was used to transform potato. Over 60 independent transgenic lines were obtained. Because of their difference in size (the antisense RNA is derived from a 1-kb internal HPL fragment), the accumulation of HPL sense and antisense RNA could be monitored simultaneously. From all transgenic plants examined, three lines showing the largest reduction in endogenous HPL mRNA levels (lines H1, H4, and H57) were subject to further analyses.

A strong reduction of HPL mRNA was observed in both young and old leaves of the transgenic lines analyzed (Fig. 2), indicating that the antisense-mediated HPL depletion was effective within the range of expression levels present in these developmental stages.

Figure 2.

HPL gene expression in HPL-antisense plants. A radioactive cDNA probe was used to detect simultaneously sense (s) and antisense (as) HPL transcripts in total RNA extracted from second (L2) and sixth (L6) leaves (as in Fig. 1) of nontransformed potato plants (WT), and from three independent transgenic potato lines (H1, H4, and H57) expressing HPL-antisense RNA. The smaller hybridizing bands that can be observed in the lanes corresponding to the transgenic lines comigrate with the rRNAs, and are likely a result of trapped HPL-antisense transcripts.

Also, HPL activity was reduced much in these three transgenic lines. Crude extracts prepared from the leaves of the transgenic line with lowest HPL mRNA accumulation (H57) produced 43.46 ± 19.29 nmol (E)-2-hexenal/min/gfw, 3% of the value determined for extracts of nontransformed plants. Perhaps because of a lower sensitivity of the spectrophotometric assay, little, if any, degradation of 13-HPOT could be detected in H57 extracts, confirming that this 13-HPL constitutes a major route of 13-HPOT removal in potato leaves. Indeed, the leaves of H57 plants accumulate significantly higher levels of fatty acid hydroperoxides than wild-type plants (26). An in vivo HPL activity assay was performed with leaf discs from antisense lines. All three transgenic lines analyzed showed reduced levels of HPL activity, compared with nontransformed plants (Table 1). Again, line H57 showed the lowest HPL activity with an (E)-2-hexenal production of only 6% of the nontransformed controls, which was in close agreement with the data obtained with plant extracts (see above). As expected, HPL-depleted plants also produced less hexanal than the wild-type plants (not shown).

Table 1.

In vivo activity in nontransformed and HPL-depleted potato plants

	Wild type	H1	H4	H57
Activity	187.4  ± 13.4	34.5  ± 9.0	60.5  ± 7.9	10.6  ± 5.4
% of WT	100	18.4	32.3	5.7

Values are results from four independent determinations [in nmol (E)-2-hexenal/24 discs/30 min ± standard error] from leaf discs of nontransformed plants (wild-type) and the HPL-antisense lines H1, H4, and H57. The residual HPL activity present in the transgenic lines also is given as percentage (% of WT) relative to nontransformed plants. 

Volatile analyses have allowed the elucidation of the fate of fatty acid hydroperoxides produced in HPL-deficient plants. Endogenous levels of HPL-derived volatiles were determined in the leaves of nontransformed potato plants and in the HPL-deficient transgenic line H57. (E)-2-hexenal was the more abundant HPL-derived volatile in nontransformed leaves, whereas hexanal accumulated to only 3% of that value (Fig. 3), probably because of the much smaller pool of its linoleic acid precursor present in potato leaves (27). In addition, a higher specificity of HPL toward linolenic acid also may account for the values obtained. Leaf discs from the H57 transgenic line showed a marked reduction in both hexanal and (E)-2-hexenal content (Fig. 3), 54% and 23%, respectively, of the values observed in nontransformed plants. Qualitatively similar results were obtained with leaf discs from the transgenic lines H1 and H4 (not shown), indicating that HPL depletion has a clear impact on the levels of volatiles accumulating in the leaves. However, the values obtained for the transgenic lines, H57 for instance, are perhaps higher than would be expected given their residual HPL activity. Although both hexanal and (E)-2-hexenal are highly volatile, they have been shown to accumulate to some extent in the cytosolic cell fraction of the tomato fruits (28). From these results, it may be speculated that C6 aldehydes may form complexes, perhaps with specific proteins (29), thus preventing their evaporation. The concentration of the interacting protein(s) may be a rate-limiting factor to control the levels of endogenous volatiles contained in the leaves. Reducing the production of these aldehydes would thus result in an imbalance with more proteins available for complex formation.

Figure 3.

Endogenous levels of 13-hydroperoxide derivatives. Volatiles present in the leaves of nontransformed potato plants (WT) and a transgenic antisense line showing the largest depletion in HPL activity (H57) were compared by head-space GLC analysis. All HPL-derived compounds [hexanal, (E)-2-hexenal; and (Z)-3 and (E)-2-hexenols] are reduced in the HPL-antisense H57 line, whereas levels of ethyl vinyl ketone (EVK) and (Z)-2-pentenal and pentenols (derived of LOX activity on 13-hydroperoxides) are increased. The identity of the peaks was determined by comparison of retention times with authentic standards and GC/MS determination of their structures (not shown).

Ethyl vinyl ketone, 1-penten-3-ol, and (Z)-2-pentenal, all products of LOX cleavage of 13-fatty acid hydroperoxides (30), accumulated to much higher levels in the HPL-depleted transgenic lines (Fig. 3 and data not shown). These results suggest that LOX themselves are recruited for fatty acid hydroperoxide disposal when HPL activity is not available. The in vivo relevance, if any, of this alternative mechanism for fatty acid hydroperoxide disposal is not known at this moment.

Thus, the HPL-depleted transgenic lines are impaired markedly in the production and accumulation of HPL-derived volatiles. It has been suggested that these volatiles may participate in the plant defense reaction as inducers/modulators of defense-gene expression (15). The HPL-antisense lines provided a good experimental system to elucidate the role of these compounds in vivo. To this end, the leaves of H57 plants were wounded and the accumulation of transcripts from defense-related genes was analyzed and compared with nontransformed plants. The expression of allene oxide synthase (AOS) and LOX H1 and H3 genes, involved in oxylipin metabolism, prosystemin (SYS), from which the systemic wound signal peptide systemin is derived, and the wound-inducible proteinase inhibitor II (Pin2) and leucine aminopeptidase (LAP) genes, was examined. As shown in Fig. 4, little variation in the pattern of transcript accumulation can be observed between nontransformed plants and the HPL-depleted H57 line, for all of the genes analyzed. Although a larger accumulation of LAP, Pin2, and prosystemin transcripts may be observed in the H57 plants in this particular experiment, it has not been reproduced consistently in others, and thus may be attributed to small differences in the conditions of the plants used. Similar results were obtained with lines H1 and H4 (not shown). Although the expression of other genes not included in this analysis might be subject to regulation by HPL-derived compounds, the results obtained herein suggest that hexanal/hexenal production and accumulation are not required for gene activation on wounding.

Figure 4.

Defense gene expression in nontransformed and HPL-depleted plants. Second and third leaves of nontransformed (WT) and HPL-depleted (H57) potato plants were wounded and harvested at different times after wounding (indicated in hours above the lanes). Small differences in the levels of sense HPL RNA (HPL) were observed on wounding, which may be related to slight differences in the ages of the leaves used. The accumulation of transcripts from wound-induced genes, allene oxide synthase (AOS), leucine aminopeptidase (LAP), lipoxygenases H1 (LOXH1) and H3 (LOXH3), proteinase inhibitor II (PIN2), and prosystemin (SYS), also was analyzed by Northern blotting. Equal RNA loading was verified by ethidium bromide staining of rRNA.

Detailed examination along development and tuber formation revealed no obvious phenotypic differences between nontransformed and HPL-depleted plants grown in the greenhouse.

Reduced Aldehyde Production Leads to an Increase in the Performance of Aphids Feeding on HPL-Depleted Plants.

Oxylipins are important contributors to the plant defense response to insect pests (31). Aliphatic aldehydes, hexanal, and 3-hexenal in particular, have been shown to have a negative impact on the fecundity of adults of Myzus nicotianae that were fed on tobacco leaves exposed to these compounds (14). To ascertain in vivo the role of HPL-derived aldehydes on plant–pest interactions, the fecundity, nymph development, and population increase attained by green peach aphids (M. persicae) feeding on wild-type and HPL-depleted potato plants were determined. As shown in Table 2, adults of M. persicae placed on the leaves of the different transgenic lines with reduced HPL levels (H1, H4, and H57) exhibit 15–31% higher fecundity than those placed on the leaves of nontransformed potatoes. There were also significant differences in time required for nymph development, which was 5–9% faster in aphids feeding on transgenic lines as compared with those feeding on nontransformed plants. Furthermore, when five adult individuals of M. persicae were allowed to proliferate on potato plants, a larger larval progeny was derived from aphids feeding on the HPL-depleted transgenic lines. In the case of line H57, aphids gave rise to nearly 90% more nymphae than those feeding on nontransformed plants. The significantly smaller number of adults recovered on wild-type plants at the end of the experiment together with their lower fecundity may account for a larger larval progeny obtained on the HPL-depleted transgenic lines, a population increase that was perhaps higher than expected by the differences in fecundity. All three transgenic lines gave similar results in the different bioassays, suggesting that the better performance of the M. persicae aphids observed was a result of the HPL-depletion engineered in the transgenic lines, and it is likely a consequence of the reduced aldehyde levels present in these transgenic lines.

Table 2.

Performance of M. persicae on nontransformed (WT) and HPL-depleted potato plants

	Aphid fecundity, nymphs/day*	Nymphal development, days†	Population increase‡
			Nymphs/ plant	Adults/ plant
WT	3.2  ± 0.2 a	6.5  ± 0.1 a	35.3  ± 3.5 a	2.2  ± 0.3 a
H57	3.8  ± 0.1 b	6.2  ± 0.1 b	66.5  ± 6.5 b	3.3  ± 0.4 b
H4	3.9  ± 0.1 b	5.9  ± 0.1 c	61.8  ± 5.9 b	4.0  ± 0.3 b
H1	4.2  ± 0.2 b	6.1  ± 0.1 b	61.9  ± 5.9 b	3.3  ± 0.4 b

All experiments were performed by using 11–12 plants per genotype. Column means followed by the same letter are not significantly different from each other (Student–Newman–Keuls test, P ≤ 0.05). 

^*Mean aphid fecundity (± standard error) of single adult apterous aphids during the first 9 days of reproduction (4–5 aphids per plant). 

^† Mean developmental time needed (± standard error) for newly born nymphs to develop into adults (cohorts of 5 nymphs per plant). 

^‡ Five adults were placed on each plant and confined by using whole-plant cages. After 5 days, the progeny (nymphs/plant) and the number of the initial adults remaining on the plants (adults/plant) were scored (mean ± standard error). 

Conclusions

The isolation of a cDNA encoding the potato 13-HPL has allowed a detailed characterization of HPL expression in potato in relation to HPL enzymatic activity. The results obtained indicate that posttranscriptional mechanisms may contribute to the regulation of HPL activity levels in the aerial parts of potato plants. Among them, protein stability and/or the presence of enzyme inhibitors may be included. Thus, it is likely that hexanal and 3-hexenal production is determined by substrate availability to HPL rather than by the abundance of HPL activity, which seems to be constitutively present. This regulatory mechanism may be particularly relevant in the production of the cut-grass odor in crushed leaves.

HPL depletion in transgenic plants has shown that the identified gene is largely responsible for 13-fatty acid hydroperoxide degradation in healthy potato leaves, and hexanal and 3-hexenal production is derived essentially from its activity. Data from the HPL-depleted transgenic potato lines strongly suggest that the constitutive activity of this branch of the oxylipin biosynthetic pathway influences plant defense toward sucking insects whose proliferation is restricted severely by the products of HPL activity. The feeding habits of the attacking insects thus seem to be key determinants of their interaction with plants and the opposing defense responses involved. Although chewing insects trigger an induced response, signaled by JA production and based on activation of proteinase inhibitor genes (18), HPL-derived volatiles, themselves toxic, are part of a defense mechanism that is deployed constitutively in the foliage of the plant. In both cases, rather than aiming toward a full eradication of attacking pests, the arms race in plant–pest coevolution has culminated in a situation where plants instead contrive to limit pest proliferation.

Abbreviations

HPL

hydroperoxide lyase

JA

jasmonic acid

LOX

lipoxygenase

HPOT

hydroperoxy-linolenic acid