The fundamental processes of plant biology are controlled by a series of biochemical reactions collectively known as plant primary metabolism. Primary metabolites include sugars, amino acids, nucleotides and fatty acids and other compounds that are directly related to cellular homeostasis (Maeda, 2019). However, due to the complexity of ecological interactions, plants developed a more specialized metabolism focused on the production of a wide array of organic compounds, defined as secondary metabolites, that play a crucial role in processes such as tolerance to abiotic stresses, defense against herbivory and pathogens, attraction of pollinators and seed dispersal (Wink, 2018; Pott et al., 2019). Among the main products of plant secondary metabolism, volatile organic compounds (VOCs), metabolites with a low molecular weight and high vapor pressure at ambient temperatures, stand out by their importance in plant defense and reproduction (Picazo-Aragonés et al., 2020).

The largest group of plant VOCs are terpenes, lipophilic metabolites composed of one or more isoprene units (Gomes et al., 2021; Rosenkranz et al., 2021). Monoterpenes and sesquiterpenes are largely found as constituents of plant essential oils and have been extensively studied due to their biological implications (Tetali, 2019; Gomes et al., 2021). Due to their ecological functions, the structures of terpenes and other secondary metabolites are hypothesized to have been evolutionary molded to interfere with physiological processes of herbivores and microorganisms that competed with the plants for resources (Wink, 2018). Such features are fundamental for the current application of these compounds, since we learned how to explore this ecologically driven chemodiversity to improve our quality of life, applying plant essential oils, terpenes and other natural products in the formulation of perfumes, cosmetics, flavoring agents, antimicrobials, pesticides and repellents, among others.

The rising concern about the environmental and human health issues related to the use of synthetic chemicals as insecticides and insect repellents has increased the interest on the study of essential oils and related compounds as alternative methods of pest control (Nerio et al., 2010; Regnault-Roger et al., 2012; Chaaban et al., 2017). Among the most effective arthropod repellents of natural source are nepetalactones, volatile iridoid monoterpenes present in the essential oil of several Nepeta species (Lamiaceae), most notably N. cataria L., commonly known as catnip (Gomes et al., 2020a; Gomes et al., 2020c). Ethanolic extracts and essential oils from N. cataria as well as isolated nepetalactones have shown remarkable repellent effects against mosquitoes, bed bugs, ticks and other arthropods, with results comparable to the synthetic N,N-Diethyl-meta-toluamide (DEET), widely regarded as the gold standard of insect repellents (Reichert et al., 2019; Shi et al., 2021; González et al., 2022).

In addition to iridoid terpenes, catnip genotypes may also accumulate sesquiterpenes, mainly β-caryophyllene and caryophyllene oxide, as part of their aromatic profile (Gomes et al., 2020b). Those sesquiterpenes, although not as common as nepetalactones in arthropod repellent formulations, have been reported as potential natural ingredients for pest control strategies (Moraes et al., 2017; Plata-Rueda et al., 2018). In addition to bioactive aroma volatiles, N. cataria also produces a series of secondary metabolites characterized as polyphenols, such as apigenin, luteolin, caffeic acid and rosmarinic acid (Reichert et al., 2018), which have been reported as having direct influence on insect behavior (Onyilagha et al., 2012; Joshi et al., 2014; Khan et al., 2019) and can, therefore, be potentially used in integrated pest management systems. The study of those compounds, along with demonstrating sources of valuable natural products, can also provide insights into the function of secondary metabolites in plant ecophysiological interactions. The understanding of such interactions is important not only from an ecological perspective of understanding the evolutionary aspects of plant phytochemistry, but also from an agronomic perspective, as it can provide information on specific growing conditions to maximize the productivity of metabolites of industrial interest.

One of the main challenges associated with a steady supply of catnip secondary metabolites for industry is the scarcity of highly productive genotypes that offer a stable chemical profile and a consistent yield of aboveground biomass (Gomes et al., 2020a). Looking to address this issue, our team has recently developed two new catnip cultivars, CR3 and CR9, that show improved yields of biomass and essential oils rich in nepetalactones, and with superior productivity when compared to other commercially available catnip genotypes (Reichert et al., 2016; Simon et al., 2019). Although the chemical profile of most commercial catnip cultivars is also dominated by nepetalactones, both cultivars CR3 and CR9 are unique in their superior productivity due to higher essential oil contents and biomass productivity. Detailed comparisons of these new cultivars to other catnip genotypes in terms of essential oil productivity and composition are described by Reichert et al. (2016); Reichert et al. (2019) and Simon et al., 2019). In addition, Cultivars CR3 and CR9 were developed in part to have distinct chemical profiles, producing different nepetalactone isomers in their aromatic profile. While cultivar CR9 produces essential oils dominated by Z,E-nepetalactone, CR3 essential oil’s major compound is of E,Z-nepetalactone, followed by Z,E-nepetalactone (Reichert et al., 2016; Simon et al., 2019). Both cultivars show stability in terms of productivity of each of those isomers in similar cultivation conditions. However, the interactions of these newly developed genotypes with different agronomic practices have not yet been studied.

Due to their closely related evolutionary relationship, the productivity of secondary metabolites is strongly influenced by environmental conditions, including agronomic interventions (Gomes et al., 2021; Chrysargyris et al., 2022). Among the cultivation approaches that can affect the production of secondary metabolites in catnip, harvesting regimes have been reported to be of fundamental importance (Chalchat and Lamy, 1997; Mohammadi and Saharkhiz, 2011; Mohamed et al., 2018; Gomes et al., 2020b). Because catnip plants are perennial, successive harvests are a common practice for this aromatic crop and the effects of such practice on plant productivity and chemical profile have not been extensively studied. Although a few reports briefly discuss the effect of successive harvest on the terpene composition and biomass accumulation of some catnip genotypes (Mohamed et al., 2018; Gomes et al., 2020b), for the new cultivars CR3 and CR9 there have been no studies to date to determine the effects of this practice on the chemical profile in terms of both terpenes and polyphenols.

The present study aimed to characterize the aromatic profile and polyphenol accumulation of new catnip cultivars CR3 and CR9 and their hybrid (CR9×CR3) across four successive harvests as well as to assess if the effects of successive harvests on physiological and productive traits are genotype dependent. Moreover, this is the first description of the polyphenol composition of cultivars CR3 and CR9 as well as the first report on the chemical characterization of the polyphenol and aromatic profile of the self-pollinated progeny of their hybrid, CR9×CR3. The results are discussed in terms of ecophysiological interactions and potential applications of these highly productive genetic materials for the insect repellent industry and other markets.

In this study, we demonstrated a strong genotype-dependent response to successive harvests on the accumulation of secondary metabolites in novel catnip genotypes. However, for the productivity of aboveground biomass and total essential oil yield per plant (essential oil content multiplied by aboveground biomass), the interaction was not significant and only the main effect of harvest significantly affected those variables. Biomass accumulation appears to be determined by primary metabolism and the genes involved in those pathways are known to be more conserved, highly regulated, and connected in more complex gene networks when compared to genes involved in secondar metabolism pathways (Maeda, 2019; Moore et al., 2019). Results suggest that primary metabolite-related genes are, overall, less susceptible to drastic variations caused by environmental interactions. Additionally, N. cataria cvs. CR9 and CR3 were specifically developed through selective breeding to be highly efficient in terms of biomass accumulation, with productivities superior to other available genotypes (Reichert et al., 2016; Simon et al., 2019). Given the superior performance of those cultivars, we hypothesize that the lack of significant differences among genotypes for biomass accumulation in this study is due to species-related physiological limitations to further increases. Additional evidence to support this possibility is the observation that the hybrid CR3xCR9 seems to produce similar, but not surpass, the parent plant’s biomass accumulation, demonstrating that a limit to further biomass increase could be added by factors not yet addressed in current catnip breeding programs.

Despite significant differences on the aromatic profile and polyphenol contents, results from the accumulated yield over 4 harvests ( Table 4 ) show that the productive performance of these novel catnip genotypes is highly similar, with yields of about 500g of dried aboveground biomass and 3g of essential oil per plant. If multiplied by the plant density recommended by Park et al. (2007) for North American conditions, 80,742 plants per hectare, these new genotypes would yield about 242 kg of essential oil in 4 harvests, or about 60.5 kg per harvest per hectare. Those extrapolations are to be considered cautiously, as a high density of plants will likely reduce the individual productivity per plant, but they can be used to highlight the productive potential of these genetic materials. Such yield extrapolations are often higher than what growers often experience in commercial settings with larger field plots in which less than ideal conditions are the norm and where actual efficiencies in handling, harvesting and processing is less than what can often be achieved with small experimental plots.

As for the main effect of harvest on biomass and essential oil yield, it is possible to observe that the peak productivity occurs at the 3^rd successive harvest, first harvest after overwintering, after one year of establishment in the field and with the longest interval in between cuts. This emphasizes the importance of developing genotypes capable of surviving winter temperatures in North America since the peak productivity of biomass and essential oil occurs after one year in the field. Future studies should address possible harvest regimes to maximize production, such as not harvesting twice in the first year or, possibly, not harvesting at all, thinking on establishing a more productive stand for subsequent years. Similar approaches have been studied for other specialty crops such as stevia (Bihan et al., 2021) and blueberries (Strik and Buller, 2005).

For the essential oil content and composition as well as polyphenols, a clear genotype-dependent effect was observed, with contrasting responses mainly when comparing cultivars CR3 and CR9. The essential oil content in cultivar CR9 was the highest at the 4^th successive harvest, not following the pattern of decline observed for its biomass accumulation. For the hybrid CR9×CR3, the highest essential oil yield was observed in both 2018 harvests, also not following trends of biomass accumulation, which presented a steep decline in the second harvest of 2018 (4^th successive harvest). Such behavior illustrates the differences between responses of primary and secondary metabolism and the fact that secondary metabolites are more likely to vary according to cultivation practices and environmental conditions. It is possible to infer that the exposure to field conditions, including temperature stress, herbivory and the successive harvests hindered the ability of cultivar CR9 and hybrid CR9×CR3 to accumulate biomass after the 3^rd successive harvest and the highest accumulation of essential oils, terpenes and phenolics at that time may indicate their role as stress mediators. Accumulation of stress induced metabolites such as terpenes and phenolic compounds has been reported for other aromatic species subjected to successive harvests, such as basil (Ocimum basilicum L.), salad burnet (Sanguisorba minor Scop) and mints (Mentha arvensis L. and M. × piperita L.) (Ceccanti et al., 2019; Ciriello et al., 2021; Soltanbeigi et al., 2021). Although we did not directly assess herbivore incidence and abiotic stress, it is reasonable to infer that the longer the plants stay under the field conditions, the more opportunities for there to be exposure to biotic and abiotic stresses. In addition to direct herbivore interactions, one can also consider that the effect of cutting the plants during harvest can be in itself a stress perceived by the plant as similar to herbivory. In cotton plants (Gossypium hirsutum L.), for example, the accumulation of terpenes caused during herbivore-plant interactions has been shown to be a general wound response, which can be induced by mechanical wounding alone (Opitz et al., 2008). Previous research has also indicated that chemical defenses can be elicited in different plants species simply by the vibrations caused by insect feeding (Appel and Cocroft, 2014). Therefore, the accumulated effects of the two previous harvest can also be a strong contributor to the metabolite accumulation observed in harvests 3 and 4. Previous research on catnip have indicated that environmental stresses causing reduction on plant growth have also promoted increases on the contents of essential oil (Said-Al Ahl et al., 2016; Mohammadi et al., 2017), corroborating the results of the present study. The distinct response of cultivar CR3 further emphasizes the genotype-dependent nature of environmental interactions in catnip.

The contrast between the responses of cultivars CR3 and CR9 is also evident when considering the relative percentages of nepetalactones in the essential oil ( Table 3 ) and the estimated yield of Z,E- and E,Z-nepetalactone ( Figure 4 ). CR3 shows relatively stable chemical profiles across all harvests, with E,Z-nepetalactone dominating the essential oil. The changes on the yield of E,Z-nepetalactone per plant are, therefore, an expression of the trends of biomass accumulation and also reflect that CR3 produces more of that isomer than cultivar CR9 and CR9×CR3 hybrid. For cultivar CR9 and the hybrid CR9×CR3, the peak essential oil contents coincide with the harvests when Z,E-nepetalactone was present at the highest relative contents and, to some extent, also coincide with the highest yields of Z,E-nepetalactone (mg per plant). The hybrid CR9×CR3 has an aromatic profile more similar to that of CR9 than CR3. The enzymes involved in the production of different stereoisomers of nepetalactone have been recently identified as nepetalactol-related short-chain dehydrogenase enzymes (NEPS1, NEPS2 and NEPS3) (Lichman et al., 2019) and the differences observed among genotypes in the present study may be related to differential regulation of the expression and activity of those enzymes by environmental conditions. Greater knowledge about those enzymes will aid in breeding programs to explore the selection of more productive biosynthetic pathways of nepetalactones.

The biosynthesis and accumulation of nepetalactones in Nepeta species is known to take place in glandular trichomes on the leaf surface (Aničić et al., 2018) and changes on the amount (density) and morphology of such structures is known to correlate with the production of secondary metabolites in aromatic plants (Deschamps et al., 2006; Souza et al., 2016). Previous research has identified that glandular trichomes of different chemotypes of N. cataria differ in their structure and morphology (Kolalite, 1998), therefore, it is possible that part of the chemical contrast observed between cultivars CR3 and CR9 are correlated to differences in their glandular trichomes, a topic worthy of further research. Additionally, the density of glandular trichomes on the surface of N. cataria leaves is also known to be affected by environmental conditions such as saline stress (Lungoci et al., 2022) and fertilization (Paria et al., 2021). The investigation of trichome density and morphology in N. cataria and their correlation with agricultural practices and genotypes can be of fundamental importance for breeding programs as this is a trait directly associated with the productivity of compounds of interest in the species.

An interesting trend observed in the essential oil of CR9×CR3 was that in both harvests of 2017, sesquiterpenes, mainly caryophyllene oxide and (E)-β-caryophyllene, dominated the essential oil and, when the levels of those compounds were reduced (3^rd and 4^th successive harvests), Z,E-nepetalactone levels were high. A similar trend was observed for CR9, with the exception of the first harvest. Because sesquiterpenes are known to be involved in the regulation of lateral roots development and plant-herbivore interactions (Ditengou et al., 2015), one could infer that the synthesis of such compounds can have ecological advantages during the establishment of the plants in the field and, after which, the characteristic nepetalactone metabolic pathway can be fully expressed. The finding that CR3 presented smaller variation and smaller percentages of sesquiterpenes in its essential oil may indicate that this cultivar has a different root system, which allows it to establish faster than the other genotypes.

Previous studies have shown that some catnip genotypes can produce caryophyllene oxide and (E)-β-caryophyllene among the main components of their aromatic profiles (Gomes et al., 2020b; Said-Al Ahl et al., 2018). In addition to being involved with plant-environment interactions, such compounds have also been identified as potential tools for the use as natural repellents and/or insecticides (Moraes et al., 2017; Plata-Rueda et al., 2018) and can be included in integrated pest management systems including catnip extracts and nepetalactones. Furthermore, the constant presence of citronellol in all the studied genotypes, although not in high relative percentages, adds an active ingredient to the blend of terpenes that compose the essential oil of catnip, since this compound has been extensively reported to be a potent natural arthropod repellent (Mota et al., 2022). The essential oil blend containing several active ingredients for arthropod repellency, with different molecular weighs and biosynthetic pathways, and, possibly different modes of action, reinforces that the products from these new catnip genotypes can be part of an efficient strategy to slow down the development of resistance on pest populations and potentially to reduce the use of synthetic repellents.

A similar trend to that observed for sesquiterpenes occurred with the accumulation rosmarinic acid in CR9 and CR9×CR3, with higher accumulations when the levels of Z,E-nepetalactone were lower. Rosmarinic acid has been extensively described as a chemical component of catnip leaves and have been reported as the main responsible for the antioxidant properties of different extracts from this species (Dienaitė et al., 2018; Sharma et al., 2019). In addition to medical uses, rosmarinic acid has been identified as a potent biopesticide against aphids of agricultural importance (Khan et al., 2019), which further establishes the new catnip genotypes addressed in this study as sources of multiple phytochemical ingredients for use in pest control. Ecophysiologically, rosmarinic acid is known to be a mediator of plant environmental interactions and is recognized to be accumulated under biotic and abiotic stresses (Dewanjee et al., 2014; Mousavi and Shabani, 2019), which, in the context of CR9 and CR9×CR3 is possibly related to the period of adaptation to field conditions.

A comparable pattern to that of rosmarinic acid can be observed, although not as evident, is the accumulation of luteolin diglucuronide in CR9 and CR9×CR3, with peaks when those genotypes presented the highest concentrations of sesquiterpenes in their essential oils. This compound is not as frequently studied as other phenolic acids and flavones described in this study, but has been studied along with apigenin diglucuronide, regarding their anti-inflammatory and antinociceptive properties (Bueyuekokuroglu et al., 2008).

Similar to rosmarinic acid, caffeic acid has also been reported as an insecticide against pests of agricultural importance (Joshi et al., 2014) and is known to be involved in plant-herbivore interactions (Kundu et al., 2018). In the present study, for all genotypes, caffeic acid was found in higher amounts on the first harvests of each season. Those periods correspond to the highest values of solar radiation in the months when the plants have been harvested (July of 2017 and June of 2018) ( Table 1 ). Such information highlights the possible involvement of caffeic acid in catnip responses to radiation stress, as this compound has been previously described to stabilize the absorption of high energy radiations in mesophyll cells, preventing oxidative stress (Riaz et al., 2019).

Despite some minor variations regarding the specificities of each genotype, a general tendency of higher accumulation of caffeic acid, luteolin, luteolin glucoside, luteolin glucuronide, apigenin glucoside, apigenin glucuronide and apigenin diglucuronide were observed to occur in the first harvest of 2018 (3^rd successive harvest). In addition to the previously mentioned radiation-induced biosynthesis, the 3^rd successive harvest was also the first harvest after the plants overwintered in the field and the presence of these compounds may be a residual effect of the protection mechanisms put in place to protect those genotypes from oxidative damages caused by cold stress. Each of these phenolic compounds and their glycosides have been previously reported to be involved in antioxidant metabolism (Hojati et al., 2011; Šliumpaitė et al., 2013; Lee et al., 2015; Bian et al., 2017; Vukić et al., 2022). Another perspective is to understand consider such compounds as indicators of more recent stress, since after their peak, the plants presented a steep decline in biomass accumulation. Many of the phenolic compounds identified in the novel catnip genotypes have been reported to inhibit herbivory (Erhard et al., 2007; Cipollini et al., 2008; Goławska and Łukasik, 2012; Joshi et al., 2014), which may indicate that the plants were employing defense mechanisms to avoid herbivory, and possibly, as a response to the successive harvests. Further studies on the involvement of phenolic compounds and catnip primary and secondary metabolisms are of strategic importance, since those compounds can be used as breeding traits for stress resistance, for use in pest control programs or as indicators of the crop’s physiological status.

This is the first description of the polyphenol composition of the new catnip cultivars CR9 and CR3 and the first description of the polyphenols and aromatic profile of the hybrid CR9×CR3. The genotype-dependent responses to successive harvests emphasize the need for future studies to understand how these new, highly productive, genotypes interact with the environment to maximize the productivity of compounds of interest.

Biomass accumulation and essential oil yield (grams per plant) in new catnip cultivars CR3 and CR9 and their hybrid CR9x CR3 do not vary significantly as a function of genotype, but are strongly influenced by successive harvests, reaching a peak at the 3^rd successive harvest (first harvest after overwintering in the field) and with a decline at the 4^th successive harvest. In contrast, the chemical composition of the essential oil as well as the polyphenol accumulation present clear genotype-dependent responses to successive harvests. While cultivar CR3 shows E,Z-nepetalactone as the major component of its essential oil in all 4 harvests, CR9 and hybrid CR9×CR3 show an aromatic profile dominated by Z,E-nepetalactone at the 3^rd and 4^th successive harvests and, especially in CR9×CR3, essential oils dominated by sesquiterpenes at the 1^st and 2^nd harvests. For CR9 and CR9×CR3, rosmarinic acid and luteolin diglucuronide were at the highest contents at the 1^st and 2^nd harvest, while for CR3 the peak occurred at the 3^rd successive harvest. Apigenin was only detected in cultivar CR3 from 1^st harvest. Despite some variations due to genotype-specific responses, a trend can be observed with a higher accumulation at the 3^rd successive harvests of caffeic acid, luteolin, luteolin glucoside, luteolin glucuronide, and apigenin glucoside, glucuronide and diglucuronide. Future studies are necessary to unfold the interactions of these compounds with plant development and productivity. The novel catnip genotypes CR3, CR9 and CR9×CR3 show remarkable potential as sources of multiple natural compounds employed in insect repellent formulations and other industries.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

EG and JS: conceptualization. EG, HP, HRJ, WL, JS and BY: methodology. EG: Data analysis. EG and HP: original draft preparation. WL, BY, HRJ, QW and JS: review and editing. JS and QW: supervision and funding acquisition. All authors contributed to the article and approved the submitted version.