Sex Pheromone Biosynthetic Pathway in Spodopteru littoralis and Its Activation by a Neurohormone*

Deuterium-labeled fatty acids have been used to elucidate the sex pheromone biosynthetic pathway in Spodoptera littoralis. Label from palmitic acid was incorporated during the scotophase into all the pheromone acetates and their corresponding fatty acyl intermediates. (Z,E)-9,11-tetradecadienyl acetate, the major component of the pheromone blend, is synthesized from palmitic acid via tetradecanoic acid, which, by the action of a specific (E)-11 desaturase and subsequently a (Z)-9 desaturase, is converted into (Z,E)-9,11-tetradecadienoate. By further reduction and acetylation, this compound leads to the dienne acetate. Deuterated precursors applied to the pheromone gland during the photophase were also incorporated into the pheromone. The percentage of labeled (Z,E)-9,11-tetradecadienyl acetate relative to natural compound was significantly higher during the light period. Label incorporation from different intermediates into the pheromone was stimulated by injection of brain-subesophageal ganglion extract during the photophase. The influence of the pheromone biosynthesis-activating neuropeptide on the biosynthetic pathway is discussed.

Deuterium-labeled fatty acids have been used to elucidate the sex pheromone biosynthetic pathway in Spodoptera littoralis. Label from palmitic acid was incorporated during the scotophase into all the pheromone acetates and their corresponding fatty acyl intermediates. (Z,E)-9,11-tetradecadienyl acetate, the major component of the pheromone blend, is synthesized from palmitic acid via tetradecanoic acid, which, by the action of a specific (E)-11 desaturase and subsequently a (Z)-9 desaturase, is converted into (Z,E)-9,l l-tetradecadienoate.
By further reduction and acetylation, this compound leads to the dienne acetate. Deuterated precursors applied to the pheromone gland during the photophase were also incorporated into the pheromone.
The percentage of labeled (Z,E)-9,11-tetradecadienyl acetate relative to natural compound was significantly higher during the light period. Label incorporation from different intermediates into the pheromone was stimulated by injection of brainsubesophageal ganglion extract during the photophase. The influence of the pheromone biosynthesis-activating neuropeptide on the biosynthetic pathway is discussed.
Chemical elucidation of a large variety of sex pheromones from Lepidoptera (Insecta) has been followed in recent years with increasing interest on the biosynthesis of these specific sexual semiochemicals.
The analysis of pheromone gland composition in different species has revealed the occurrence of unusual fatty acids that have been proposed as precursors of pheromone components (l-4). Many lepidopteran sex pheromones are produced by limited P-oxidation steps in conjunction with desaturase systems (5,6). However, only in a few studies has a combination of different fatty acyl intermediates been used to demonstrate experimentally a pheromone biosynthetic pathway (e.g. [7][8][9]. The regulation of sex pheromone biosynthesis in Lepidoptera has also received special attention after Raina   biosynthesis-activating neuropeptide (ll), was able to induce pheromone production in neck-ligated Heliothis zea females. Pheromonotropic activity in brain-subesophageal ganglion extracts has been demonstrated in other moth species (10)(11)(12)(13)(14). However, the target site of PBAN' is yet to be confirmed.
Although there exist some indications that the peptide may act directly on the pheromone gland (15), Teal et al. (16) have recently shown that it exerts its action through the terminal abdominal ganglion. The control mechanism(s) of pheromone biosynthesis is also an unsolved question. Soroker and Rafaeli (15) (17) have suggested recently that the brain hormone possibly acts on some other tissue that supplies substrate to the pheromone gland for pheromone biosynthesis. In the present study, we use deuterium-labeled fatty acids to elucidate the biosynthetic pathway of Spodoptera littoralis sex pheromone.
With the aim of understanding the effect(s) of the PBAN on this process, label incorporation into pheromone intermediates and the final product was also analyzed during the photophase (when pheromone biosynthesis is not activated) and after stimulation with brain extracts known to contain PBAN in this species (14).
Likewise, methyl hexadecanoate and tetradecanoate were also present in the methanolyzed extracts. These compounds were found in the proportions indicated on Table I.
The identification of the above acetates and FAME was accomplished by capillary GLC and GLC-MS, by comparison of their retention times and mass spectra with those of authentic samples.
Virgin females close to the onset of their second scotophase were treated with different deuterated fatty acids, considered  to be putative intermediates in sex pheromone biosynthesis. The incorporation of deuterium was analyzed in fatty acyl moieties and acetates of pheromone gland extracts. The abundances of deuterated FAME relative to those of the natural compounds, as well as the amounts of labeled (Z,E)-9,11-C14:Ac, were determined for each precursor (Tables II and  III). [16,16,16-'HJHexadecanoic Acid-Analyses of sex pheromone extracts of females treated with d&16:acid revealed that label was incorporated into all the pheromone components studied (Fig. 1). Ions m/z = 255 and 199, corresponding to labeled (Z,E)-9,11-tetradecadienyl and tetradecyl acetates, respectively, appeared approximately 0.05 min before the natural compounds (ions m/z = 252 and 196, respectively). Likewise, ion m/z = 197 was detected at the retention times of deuterated (Z)-9-C14:Ac, (Z)-II-Cl4:Ac, and (E)-ll-C14:Ac. These ions were not observed in controls.
From the results described above, the sex pheromone biosynthetic pathway depicted in Fig. 3 is proposed, which is analyzed in detail under "Discussion."

Incorporation
of Labeled Intermediates during the Photophase Application of deuterated fatty acids was also performed during the photophase, when pheromone biosynthesis is normally not activated and only low levels of pheromone are detected.
GLC-MS analyses of pheromone gland extracts revealed that label from different precursors was also incorporated into fatty acyi intermediates and pheromone acetates during the photophase. The amounts of labeled (Z,E)-9,11-C14:Ac were not statistically different from those detected during the scotophase. However, the relative abundance of labeled pheromone with respect to the natural compound was twice as high during the photophase (Table II).
The relative amounts of label incorporation into different intermediates (FAME) were also determined and compared between photophase and scotophase. Only for the last intermediate of pheromone biosynthesis, (Z,E)-9,11-C14:acid, were significant differences found with all the precursors used (Table III).

Effect of Br-SOG Extract
The injection of Br-SOG extracts shortly after treatment with d&16:acid stimulated label incorporation into pheromone. However, the amounts of d3(Z,E)-9,11-C14:Me were significantly lower than in controls (saline-injected females) (Table IV).
Treatment with d3(Z,E)-9,11-C14:acid, prior to Br-SOG injection, also resulted in an increase in labeled pheromone and a decrease of this last precursor (Table IV).
The analysis of FAME from females treated with d&14:acid and further injected with Br-SOG extracts showed that the amounts of the labeled diene acyl intermediate were significantly lower in those animals than in controls (Table IV).  No significant difference was found in relation to the other fatty acyl intermediates of pheromone biosynthesis between controls and animals injected with Br-SOG extracts. DISCUSSION Several Cl4 acetates have been identified in the sex pheromone of 5'. littoralis females. In this context, Nesbitt et al. (28) reported the pheromone to be a mixture of Cl4:Ac, (Z)-9-C14:Ac, (E)-ll-C14:Ac, and (Z,E)-9,11-C14:Ac; later, (Z)ll-C14:Ac and (Z,E)-9,12-Cl4:Ac were also identified (29-31). The pheromone produced by the strain used in the present study consists of a mixture of the above acetates, except for the minor component (Z,E)-9,12-C14:Ac, which could not be detected. Likewise, the fatty acid composition we found in pheromone glands is in agreement with that reported by Dunkelblum and Kehat (3).
These researchers proposed that (Z,E)-9,11-tetradecadienyl acetate, the main pheromone component, could be synthesized from palmitic acid by A-11 desaturation and further P-oxidation. Specific (E)-11 desaturation of the thus formed (Z)-9-tetradecanoic acid would give rise to (Z,E)-9,11tetradecadienoic acid, the immediate precursor. Likewise, (2) and (E)-ll-tetradecanoic acids would be formed by the action of two specific (2) and (E)-11 desaturases on tetradecanoic acid, derived from P-oxidation of palmitic acid.
The results presented in this paper disagree with some of the above assumptions. Label from palmitic acid was incorporated into all pheromone acetates and intermediates above mentioned, however, d&Z)-ll-hexadecanoic acid did not produce d&Z,E)-9,11-C14:Ac. The only labeled compounds detected from this intermediate were (Z)-9-tetradecenyl acetate and its corresponding fatty acyl precursor, and only (2)-g-C14:Ac was labeled from d3(Z)-9-tetradecanoic acid treatments. These results indicate that whereas palmitic acid is All desaturated to (Z)-ll-hexadecanoic acid, which is then fioxidized to (Z)-9-tetradecanoic acid, this compound is not further A-11 desaturated, but is directly reduced and acetylated to (Z)-9-tetradecenyl acetate.
On the other hand, deuterated tetradecanoic acid led to the formation of both (Z) and (E)-ll-C14:Ac and their fatty acyl precursors but not to labeled (Z)-9-C14:Ac. These results indicate that C14:acid, formed by P-oxidation of palmitic acid, is probably the substrate of two specific A-11 desaturase enzymes, one of them introducing the E double bond and the other one giving rise to the Z isomer. Reduction and acetylation of both (Z) and (E)-ll-tetradecanoic acids would produce the corresponding (Z) and (E)-ll-C14:Ac. The existence of Z/E isomerases in the pheromone glands was ruled out, since females treated with dJZ)-ll-C14:acid did not produce labeled (E)-ll-tetradecenoyl moieties and vice versa. It is worth mentioning that only saturated C14:acid appears to be the substrate of these A-11 desaturases, since application of (Z)-9-tetradecanoic acid did not lead to the formation of the corresponding (Z,E)-9,11-Cl4 diene.
Our data show that palmitic acid is the precursor of (Z,E)-9,11-C14:Ac via tetradecanoic acid and not via (Z)-ll-C16:acid. Biosynthetic pathways of pheromones with conjugated diene structures have been proposed in both the silkworm (8) and the codling moth (9); in these species, the second double bond was suggested to be formed by a-oxidation of one of the monoene intermediate allylic positions, followed by a 1,4-elimination of water (9). In S. littoralis, however, (Z, E)-9,11-tetradecadienyl acetate appears to be synthesized by a sequential introduction of the two double bonds. Accordingly, the action of a specific (Z)-9 desaturase on the previously formed (E)-ll-tetradecanoic acid would produce (Z,E)-9,11-tetradecadienoic acid, which would be further reduced and acetylated. This assumption is supported by the fact that sex pheromone glands treated with Q(E)-ll-tetradecanoic acid produce labeled (E)-ll-tetradecenyl and (Z,E)-9,11tetradecadienyl acetates. Both the substrate and the stereoselectivity of this (Z)-9 desaturase enzyme are evidenced by the finding that insects receiving b(Z)-ll-tetradecanoic acid showed label incorporation into (Z)-ll-tetradecenyl acetate, but (Z,Z)-9,11-tetradecadienyl acetate was not identified. Additionally, neither (E,Z)-9,11-C14:Ac nor the (E,E) isomer was found in extracts from insects treated with b(Z)-ll-C14:acid and d(E)-ll-C14:acid, respectively. On the other hand, tetradecanoic acid does not seem to be a suitable substrate for this (Z)-9 desaturase, as concluded from the absence of detectable amounts of labeled (Z)-9-tetradecenyl acetate after treatment with d,C14:acid.
Whereas a E- (9) desaturase has been implicated in the biosynthesis of Cidia pomonella sex pheromone (9), the present evidence of the action of a (Z)-9 desaturase is, to our knowledge, an unprecedented example in the sex pheromone biosynthesis of Lepidoptera. In the light of the above results, a complex system of desaturases is proposed to be involved in the pheromone biosynthesis of S. littoralis: (a) a 6-11 palmitoyl-CoA desaturase, such as the one previously reported in Trichoplusia ni (5); (b) two specific (2) and (E)-11 desaturases, similar to those implicated in Argyrotaenia velutinana (32), introducing (2) and (E) double bonds in tetradecanoic acid, and (c) a specific (Z)-9 desaturase of (E)-llCI4:acid.
In our studies on the pheromonotropic activity of brain extracts in S. littoralis (14), we showed that the injection of brain-subesophageal ganglion homogenate at the onset of the scotophase restored pheromone production in decapitated females and stimulated pheromone biosynthesis during the photophase.
These results suggested that the enzymatic sys-  tern involved in pheromone biosynthesis could be activated at any time of the photoperiod cycle.
The second objective of the present study was to gain a better understanding of the biochemical events involved in such activation. In this context, we expected that the analysis of labeled pheromone and intermediates produced from different deuterated precursors during photophase and scotophase could give us some clue to this question. In these experiments, we found that the availability of precursors seems to be important for pheromone production, since application of deuterated intermediates during the photophase resulted in label incorporation into the pheromone, and no significant difference was detected in the amounts of labeled pheromone at both periods of time (Table II). However, since calling behavior and pheromone release occur after lights-off, it cannot be excluded that the amounts of labeled (Z,E)-9,11-Cl4:Ac produced during the scotophase were higher than those found. Accordingly, some release of labeled pheromone during calling could account for the lower percentage of label found during the scotophase (Table II). It is worth mentioning that titers of natural pheromone found in those females treated at the onset of scotophase (both controls and experimental animals) were considerably reduced as compared with those characteristic of second scotophase females. In order to see whether these low pheromone levels were due to insect handling, a group of females was treated as those in the experiments (C02-anesthetized and fixed under netting for 30 min), and another group was left undisturbed. The results showed that insect manipulation close to the scotophase may cause a a-fold reduction in pheromone production.
In order to obtain a more reliable determination of the amounts of labeled pheromone produced, pheromone biosynthesis was stimulated during the photophase by injection of brain extract, which does not induce calling behavior in the recipient females (14). In these experiments, a significant increase of label incorporation into (Z,E)-9,11-C14:Ac was observed from different deuterated intermediates. Titers of natural pheromone were also increased in those animals. These results suggest that, although the supply of precursors may be important for pheromone production, a rate-limiting step may be activated (directly or indirectly) by the PBAN. However, Br-SOG extracts have been used in the present experiments, and brain factors other than PBAN might well be affecting pheromone biosynthesis. Additionally, the study of FAME profiles in females treated with d&16:acid, d&14:acid, or &(Z,E)-9,11-C14:acid and further injected with Br-SOG extract showed no significant differences when compared with control animals (injected with saline), except in the amounts of the diene acyl precursor, which were significantly lower in Br-SOG-injected females (Table IV). These results suggest that the transformation of this intermediate into the pheromone might be affected by the PBAN, but pure peptide will be necessary to clarify this point.
In this context, we have observed that as indicated for other moth species, Hydraeeiu micucea (33), the last enzymatic step in S. 1ittoruEis pheromone biosynthesis, by which alcohols are converted into acetates, is not under hormonal control, since (Z,E)-9,11-tetradecadien-l-01 applied to the pheromone gland was converted to the corresponding acetate regardless of the time of application (photophase or scotophase) (data not shown).
The mechanism(s) by which the PBAN activates pheromone biosynthesis in Lepidoptera is still unknown. It has been indicated that in the redbanded leafroller moth, A. uelutinanu, the brain hormone regulates pheromone production by activating the supply of octadecanoyl and/or hexadecanoyl intermediates (17). In Heliothis urmigeru, however, it has been suggested that the PBAN is involved in the regulation of fatty acid biosynthesis in the sex pheromone gland (15). Results from the present study show that label incorporation from several precursors into the pheromone can be stimulated by Br-SOG extracts. Our data no not enable us to determine which is the precise pathway affected by the PBAN, but we suggest that the reduction step may be affected by the brain factor in S. littoralis.