Morpho-phytochemical screening and biological assessments of aerial parts of Iranian populations of wild carrot (Daucus carota L. subsp. carota)

This study investigate the chemical and phytochemical diversity among 118 individuals from five populations of Daucus carota L. subsp. carota. The collection areas, genotype numbers, soil characteristics, and climatic conditions of five populations are detailed in Table 1. The Daucus carota populations (DCPs: DCP1–DCP5) corresponding to Oshnavieh Road, Qarayi, Mavana, Band, and Gharib Hassan respectively. Specifically, DCP1 comprises genotypes G1 to G22, DCP2 includes G23 to G45, DCP3 spans G46 to G70, DCP4 covers G71 to G94, and DCP5 contains G95 to G118. The study site is situated in the West Azerbaijan province of Iran, near to the city of Urmia. Furthermore, detailed information about the study area including geographic maps, depicting slope, elevation, geological characteristics, and the exact locations is illustrated in Fig. 1a-f. The geographic map generated with ArcGIS 10.8 (ESRI) URL: https://support.esri.com/en-us/products/arcmap.

Detailed information about the study area including a and b) geographic maps c) Digital elevation model (DEM) d) slope, e) geological characteristics, and f) Exact locations of the five populations.

Morphological dimension analysis

Table 2 presents the average morphological characteristics for each population with their respective standard errors. Also different organs from collected samples are illustrated in Fig. 2. Plant height (PH) varied between 71.96 cm and 96.08 cm, with the tallest specimens recorded in the Gharib Hassan population (DCP5). The number of nodes on the main branch (NNMMB) was highest in both DCP3 and DCP5, reporting values of 6.44 and 6.13, respectively. Additionally, the number of flowers on the secondary stem (FNMSIN) ranged from 1.17 to 2.19, while the number of nodes on the secondary stem (NNMSIN) spanned from 1.15 to 3.21, with the maximum values observed in DCP5. DCP2 exhibited the highest internode length on the main branch (INNLMB) at 12.37 cm, along with a leaf length (LL) of 12.3 cm and a leaf width (LW) of 6.87 cm. The ratio of leaf length to leaf width (LL/LW) varied from 1.78 to 2.77, with the highest ratio observed in DCP4. Furthermore, additional morphological measurements were presented in Table 2. The broad ranges of these measurements highlight substantial morphological diversity among the populations studied. The wild carrot (Daucus carota subsp. carota) is primarily utilized for its root, which has been the focal point in breeding programs. However, recent attention to the functional aspects of plants has led to an increased exploration of other plant parts for supplementary studies. Given that the wild carrot originates from the southwestern regions of Asia like Iran, investigating the populations in these areas can aid in identifying diverse traits, thereby providing essential tools for the breeding of this species. To achieve this objective, it is crucial to maintain a broad range of plant diversity, as this facilitates the selection of suitable candidates for breeding initiatives. The existing diversity within this plant species is important, as it serves as a rich reservoir of genetic resources that can inform effective breeding strategies²⁵.

Different organs from collected samples of D. carota.

Essential oil (EO) content, compositions, chemo-typing and correlation

The essential oils (EOs) displayed a spectrum of colors, ranging from colorless to pale yellow, and each exhibited varying scent intensities. The primary reason for the observed differences in color among the EO samples is likely the alteration of their chemical compounds across the evaluated populations^26,27. The EO content varied among populations, ranging from 0.88 to 1.37 (% V/W) with the highest amounts was obtained in DCP3 (Table 3). This region has a relatively high elevation of 1803 m compared to other populations. The region receives an average annual precipitation of 400 mm, experiences a minimum relative humidity level of 57%, and has an average annual temperature of 10 °C. EO yields of D. carota subsp. carota, derived from the aerial parts, were found to be 0.7% from blooming umbels and 1.0% from ripe umbels (containing mature seeds) in samples collected from Portugal and Baunei, Italy²⁴. Additionally, a yield of 0.88% was reported from samples collected in Turkey²².

According to the GC and GC–MS analyses, a total of 23 distinct compositions were identified, which together accounted for approximately 88.49% to 95.5% of the EOs (Table 3). The main chemical groups identified in the EOs of Daucus carota subsp. carota included oxygenated sesquiterpenes (47.84 to 76.26%), hydrocarbon monoterpenes (2.33 to 37.37%), sesquiterpene hydrocarbons (2.22 to 5.69%), oxygenated monoterpenes (1.05 to 3.92%), and other compound categories (1.78 to 25.54%). The major compounds found in the EOs of the populations were carotol (46.64 to 74.03%), Daucene (1.38 to 3.92%), limonene (0.34 to 3.72%), and geraniol (0.49 to 2.91%). Notably, the highest concentration of carotol was recorded in the DCP4 population at 74.03%, followed closely by the DCP5 population at 73.61%. Carotol emerged as the dominant compound across the populations, specifically noted in DCP4 and DCP5. This finding is significant because carotol is known for its medicinal properties, including antioxidant and anti-inflammatory effects, which may enhance the pharmacological potential of these populations and inform future breeding programs for cultivating desirable traits²⁸. The presence of a predominant compound with a high percentage in the EOs of these populations may hold significant value for various pharmaceutical and food industries²⁹. The composition of major compounds (e.g., carotol) and other constituents varied significantly across the collected samples, a difference likely linked to their diverse geographical origins. This variability highlights a key advantage of plants from natural habitats. It supports the discovery of crucial chemotypes that can be targeted for domestication or breeding programs, serving as a valuable resource for agricultural development. Moreover, the extensive diversity of EO compounds has been documented in prior studies examining various plant species. The composition and quantity of essential oils in plants are influenced by a multitude of factors, including sexual dynamics, seasonal changes, ontogenetic stages, genetic variations, as well as ecological and environmental characteristics^27,30,31. For the categorization and clustering of five studied populations, the analyzed compounds in a two-way hierarchical clustering analysis were examined. Based on the results, carotol was found to have the most significant impact, followed by two compounds: butrnyl acetate and α-pinene, in clustering the populations. Accordingly, DCP1 was designated as chemotype I (carotol—butyl acetate), DCP3 as chemotype II (carotol—α-pinene), and populations DCP2, DCP4, and DCP5 were identified as chemotype III (with a very high percentage of carotol ranging from 64.03 to 74.03%) (Fig. 3). The carrot seed EO market size was valued at USD 2.45 billion in 2023 and is estimated to reach USD 3.2 billion by 2030, growing at a CAGR of 5% from 2024 to 2030 (https://www.verifiedmarketreports.com/product/carrot-seed-essential-oil-market/). Carotol, recognized in multiple studies as the primary compound in wild carrots, has drawn considerable attention from the cosmetics and food sectors. This interest stems from efforts to identify high-yield chemotypes with elevated concentrations of this valuable component, positioning it as a strategic target for commercial cultivation and industry applications³². In this regard, chemotypes characterized by only a few dominant constituents have attracted the attention of various industries³³. Two-Way Hierarchical Clustering Analysis serves as a robust method for identifying and classifying these chemotypes, allowing for a deeper understanding of the relationship between different populations and their chemical compositions. By focusing on these specific chemotypes, industries can optimize their products to meet consumer demands and explore new market opportunities³⁴.

Two-ways clustering analysis with two distinct dendrograms rows × columns respectively including essential oil constituent’s × Daucus carota populations (DCPs).

Figure 4 presents a detailed analysis of the correlations between various components of EOs, both quantitatively and visually. In this Figure, larger blue points indicate positive correlations, while larger red points show negative correlations. Specifically, E-β-caryophyllene exhibited a strong positive correlation with linalool (0.89), bornyl acetate (0.61), and geraniol (0.79). Conversely, it showed negative correlations with geranyl acetate (-0.63). Additionally, carotol, recognized as a primary constituent of the EO, displayed a significant negative correlation with daucene (-0.83), β-pinene (-0.65), (E)-methyl isoeugenol (-0.84), terpin-4-ol (-0.78), and p-cymene (-0.78). Further correlations among the various traits are also illustrated in Fig. 4.

Simple correlation analysis of 23 phytochemicals in Daucus carota Populations (DCPs): Dark blue indicates strong positive correlation (range 0 to + 1) and dark red indicates strong negative correlation (range 0 to -1).

Total phenolic content (TPC) and total flavonoid content (TFC)

The analysis of total phenolic content (TPC) and total flavonoid content (TFC) across various locations revealed significant variations (p < 0.05). Among the aerial parts (AP), the highest TPC was observed in sample DCP3 (54.81 mg of gallic acid equivalent (GAE)/g DW) (p < 0.05) (Table 4). Conversely, the lowest TPC (10.82 mg GAE/g DW) was measured in Band (DCP4). For the seed samples, the highest TPC was noted in Oshnavieh Road (DCP1) at 20.26 mg GAE/g DW, whereas Gharib Hassan (DCP5) showed the lowest TPC (4.09 mg GAE/g DW). The order of TPC (based on mg GAE/g DW) in aerial parts (AP) from the various populations, ranked from highest to lowest, is as follows: DCP3 (54.81) > DCP5 (27.10) > DCP2 (21.79) > DCP1 (13.80) > DCP4 (10.82). In contrast, the lowest TFC was found in Gharib Hassan (DCP5), with a value of 26.07 mg QE/g DW. More information about TPC and TFC are shown in Table 4. These results highlight the significant variability in flavonoid content among different populations and plant parts. This difference matches findings from other studies that indicate abiotic stressors, like temperature, drought, salinity, ultraviolet radiation and variations in soil nutrients found in native habitats, can boost phenolic production as a protective response³⁵. Phenolic compounds have been shown to possess antimicrobial and antioxidant properties, which play a crucial role in enabling plants to defend against infections caused by pathogens and pathogenic microorganisms. Furthermore, the presence of these compounds in plant tissues provides a protective mechanism against the harmful effects of reactive oxygen species (ROS). This dual functionality supports the overall health and resilience of plants in their environments³⁶. In resource-limited environments, plants strengthen their defensive mechanisms against environmental stressors while carefully regulating the synthesis of secondary metabolites. This strategic approach reflects their adaptive responses to ecological constraints for survival³⁷.

Antioxidant activity

The DPPH free radical scavenging activity for aerial parts (AP) was highest (53.25%) in samples from DCP5, while Band (DCP4) exhibited the lowest activity at 7.55% (Table 4). In the case of seed samples, DCP2 demonstrated the highest DPPH scavenging activity at 36.71%, whereas DCP5 had the lowest activity (7.31%). For the Ferric Reducing Antioxidant Power (FRAP) assay, Mavana (DCP3) showed the greatest antioxidant capacity for APs, with a measurement of 179.74 μmol Fe(II)/g dry weight (DW). In contrast, Gharib Hassan (DCP5) recorded the lowest FRAP value at 21.11 ± 0.34 μmol Fe (II)/g DW. For seed samples, the highest FRAP value was noted at Oshnavieh Road (DCP1), reaching 242.15 mg FeSO₄/g DW.

Antibacterial activity

The antibacterial activity of extracts derived from various populations of wild carrot was evaluated against two bacterial strains: Escherichia coli and Staphylococcus aureus. After a 24 h incubation period, the extracts demonstrated significant antibacterial effects against both evaluated strains (p < 0.01). The presence of these extracts resulted in the formation of inhibition zones, indicating an effective suppression of bacterial growth. As illustrated in Fig. 5a, the antibacterial effects of all evaluated populations against both bacterial strains showed a statistically significant difference compared to the negative control. However, the extracts exhibited less antibacterial activity than the positive control (Tetracycline 400 mg/ml) (Fig. 5b). For E. coli, the order of inhibitory effects was as follows: control⁺ (3.23) > DCP3 (1.24), DCP2 (1.15) > DCP1 (1.05), DCP4 (0.81) > control^– (0). Similarly, for S. aureus, the results showed an order of inhibitory effects as follows: control⁺ (2.8) > DCP5 (1.82), DCP3 (1.16), DCP1 (1.01) > DCP4 (0.84) > control^– (0). These findings indicate that while the extracts from wild carrot populations were effective in inhibiting bacterial growth, their efficacy was variable among the different populations studied. The variation in inhibition zone sizes observed in the study can be attributed to the differences in bacterial types, specifically E. coli and S. aureus, as well as the structural variances in their cell membranes³⁸. These findings indicate that extracts from wild carrot populations can effectively inhibit bacterial growth, though their effectiveness varies among different populations. This variability is likely due to differences in the chemical composition of the extracts, which may be influenced by environmental conditions and the genetic diversity within each population³⁹. Understanding these factors can help optimize the use of wild carrot extracts in antibacterial applications.

Antibacterial effects of vegetative extracts from five wild carrot populations (DCPs) against two strains of E. coli and S. aureus: a) column chart b) Experimental photos: larger inhibition zones indicate a stronger effect of the samples.

Correlations, and multivariate and regression analysis with combined data

The study aimed to examine the relationship between phytochemical traits and their antioxidant and antibacterial properties by using a Mantel Test correlation matrix (Fig. 6). This study systematically evaluated antioxidant and antibacterial activities in wild carrot tissues. Antioxidant capacity was quantified via FRAP (reducing power) and DPPH (radical scavenging) assays in both seeds and aerial parts, while antibacterial efficacy was tested against two Gram-positive and Gram-negative bacterial strains to assess broad-spectrum bioactivity. Additionally, with Mantel Test EO component parameters, morphological traits and phenolic compounds such as AP TFC and TPC, Seed TFC and TPC were assessed. The results indicated that only a limited number of evaluated traits significantly correlated with antioxidant and antibacterial properties at the studied bacterial strains. For antioxidant activity, two key traits including Seed TPC and AP TFC exhibited significant correlations, with Mantel’s P < 0.05 and Mantel’s r > 0.4. In relation to E. coli, no significant traits with high influence were observed. However, for the strain S. aureus, bulnesol from the essential oil showed a significant relationship with antibacterial properties (Mantel’s P < 0.05, Mantel’s r > 0.4). Notably, compounds like α-terpinyl acetate and AP TFC demonstrated stronger associations with the antibacterial properties of S. aureus. Most traits displayed limited isolated effects, which may be attributed to the synergistic effects of phytochemical compounds. This comprehensive approach sheds light on the intricate relationships between various biological traits and their potential applications in the industry⁴⁰.

Combined graph of Mantel test and Pearson correlation heatmap. A combined graph presenting the results of the Mantel test and Pearson correlation heatmap was created to elucidate the relationships among phytochemical and morphological attributes. The Pearson correlation analysis illustrates the correlations among all evaluated traits, with blue indicating positive correlations and red indicating negative correlations. The number of asterisks (*) represents the degree of statistical significance for each correlation. Additionally, the Mantel test was employed to assess the relationship between phytochemical and morphological attributes and their effect on bacterial genera, as well as to explore the association of antioxidant activity with these traits. In this context, a larger Mantel’s r value and a smaller Mantel’s p-value signify a stronger interaction between the indicators.

Correlation analysis is a valuable tool for exploring the relationships between various traits, with significant implications for plant breeding and domestication practices. Figure 6 provides a visual representation of Pearson’s correlation coefficients among key essential oil (EO) constituents and other attributes. In this figure, large dark blue points indicate a strong positive correlation, while large dark red points represent a strong negative correlation. Specifically, the analysis reveals several notable correlations: Bornyl acetate exhibited a negative correlation with MSD (p < 0.05); α-pinene showed a positive correlation with AP TPC (p < 0.05); Daucene was negatively correlated with FNMSIN (p < 0.05); limonene displayed a positive correlation with the ratio LL/LW (p < 0.05); β-pinene was positively correlated with AP TPC (p < 0.01); linalool was positively associated with both AP TFC (p < 0.05) and LL/LW (p < 0.05); and geranyl acetate showed a positive correlation with PH (p < 0.05). The simultaneous enhancement of one group of metabolites together with another is of significant interest to breeders, especially in the context of medicinal plants, where the goal is often to simultaneously increase all beneficial secondary metabolites. The positive relationship between α-pinene and β-pinene with the total phenolic content (TPC) in the aerial parts suggests a co-regulatory mechanism. This means that these terpenoids might work together to boost antioxidant capacity, potentially by sharing precursor molecules in the phenylpropanoid pathway⁴¹. This multi-trait breeding approach allows breeders to target multiple desirable attributes, thereby improving the overall quality and efficacy of the plants. As a result, breeders are increasingly focused on strategies that facilitate the parallel breeding of various traits, maximizing both yield and the therapeutic potential of these plants⁴². Additional correlations are illustrated in Fig. 6, where further details can be found.

Canonical Correspondence Analysis (CCA) is a multivariate statistical method utilized to investigate the relationships between two sets of variables. In this study, CCA was applied to analyze the associations between phytochemical and environmental attributes (climatic and soil factors). CCA was done in a sample of 118 genotypes sourced from 5 distinct populations. This type of analysis explains the relationships between traits by forming Canonical Correspondence (CCs), and the contribution of each canonical correspondence is determined by the Eigenvalues and the percentage of variance obtained. Table 5 illustrates each set of CCs along with their corresponding Eigenvalues and variance percentages. The formation of canonical sets involves the calculation of numerical values for each trait, where traits with high positive values correlate positively and traits with high negative values correlated with each other. This indicates that an increase in any of the related environmental traits is associated with an increase in the corresponding phytochemical values. In CC1, phytochemical components such as α-pinene, sabinene, bornyl acetate, carotol, and AP DPPH, which possess negative values greater than 0.4, showed canonical correlation with negative values of environmental traits such as relative humidity (%), calcium carbonate (%), silt percentage, exchangeable potassium, available phosphorus, zinc, copper, and calcium. This implies that an increase in any of these environmental traits is associated with an increase in the phytochemical traits correlated with them. The first canonical component (CC1), accounting for 63.24% of variance, played a primary role in explaining the variance of phytochemical compounds under different climatic conditions in the studied areas. Furthermore, CC1 displayed strong positive correlations with bulnesol (a promise compound of the DCP1 population) and clay percentage, alongside a partial but notable relationship with site height, implicating soil and climatic traits in shaping phytochemical variation. Additionally, CC2 explained 21.19% of the variance, where the components carotol (-0.47), α-pinene (-0.57), β-pinene (-0.55), and limonene (-0.65) showed positive correlations with site height (-0.80), pH (-0.56), and clay percentage (-0.77) due to their higher negative values. Also α-pinene and β-pinene exhibited an adverse correlation with annual mean temperature (+ 0.91), relative humidity (+ 0.71), electrical conductivity (EC) (+ 0.65), and magnesium content (+ 0.83). This suggests that an increase in these environmental traits in the studied areas is associated with a decrease in the aforementioned essential oils. CC3 accounted for 10.53%, while CC4 contributed 5.04% of the variance, representing a minor contribution in explaining the variations of phytochemical components related to environmental conditions. Considering that CC1 and CC2 cumulatively accounted for over 80% of the variance and environmental correlations with phytochemical compounds, a bi-plot of the first two canonical components was constructed. The green vectors represent environmental factors, which indicate a correlation with a phytochemical factor when they point towards it. EO 21, or carotol, positively correlated with phosphorus, exchangeable potassium, and zinc but exhibited a negative correlation with height and clay percentage. Both AP DPPH and AP TPC positively correlated with clay percentage, zinc, and copper, while showing a negative correlation with soil manganese. Additionally, further details on other environmental correlations with phytochemical components are illustrated in Fig. 7.

Canonical correspondence analysis (CCA) among the phytochemical, climatic and soil characterization, in the five studied Daucus carota populations (DCPs).

To further investigate the diversity of phytochemical and morphological traits among the populations, a cluster analysis was performed using R software with a circular clustering approach (Fig. 8). This analysis categorized the genotypes of all individuals from the five studied populations into three principal groups based on the traits examined. Specifically, within the analysis of 118 D. carota genotypes, individuals corresponding to the identified chemotypes were organized into three distinct clusters (Fig. 8). This finding indicates that the genotypes displayed less diversity within populations compared to the diversity between populations. Given that one of the regions is considered a center of origin for wild carrot in Iran, it appears that these wild carrot populations have adapted to their environments over the years. Environmental adaptation, driven by natural selection, has led to uniformity and the development of ecotypes that are well-suited to their surroundings. Wild carrot is self-pollinating; however, pollinators such as insects increase the rate of cross-pollination. This factor, beside natural selection, contributes to genetic exchange within populations and the selection of genotypes that favor adaptation to the natural environment, leading to a remarkable degree of uniformity.

Circular clustering of all combined attributes among 118 individuals of Daucus carota.

Stepwise regression is a statistical technique employed in linear regression analysis to identify a subset of relevant independent variables for inclusion in a model³¹. This method operates by systematically assessing the statistical significance of each independent variable, thereby determining its contribution to the overall model performance by selecting only those variables that significantly enhance its explanatory power. This analysis was conducted to assess the regression relationship between various phytochemical components and morphological traits. For instance, three morphological factors; BL, W, and RNM exhibited significant negative correlations with the compound α-pinene, with Standardized beta values of -0.502, -0.426, and -0.512, respectively (p < 0.001). Additionally, BNM displayed a positive correlation with α-pinene, with a Standardized beta value of 0.210. Therefore, these four morphological traits, with an r value of 0.835 and an r² value of 0.697, could significantly serve as morphological markers for the compound α-pinene. Given the presence of BL and W in the formation of linear regressions, these traits can be considered effective morphological markers for predicting phytochemical characteristics. However, the positive and negative values of Standardized beta indicate the respective negative and positive roles of these traits concerning the associated phytochemical values. Additionally, for the compound carotol, five morphological traits were significantly associated in the linear regression analysis, demonstrating a meaningful relationship with an r value of 0.6 (p ≤ 0.01). The positive Standardized beta values for three traits NNMSIN, MUD, and RNM were 3.544, 3.778, and 4.878, respectively, indicating a significant positive impact on the regression related to carotol. Conversely, the two traits BNM and MSD exhibited negative correlations in the context of the linear regression associated with Carotol, with beta values of -3.892 and -2.637, respectively, at a significance level of p ≤ 0.01. The regression models for other phytochemical traits, as detailed in Table 6, underscore the significant role of morphological markers in predicting phytochemical profiles. For instance, traits such as leaf length (BL) and width (W) consistently emerged as strong predictors and could serve as cost-effective alternatives for labor-intensive phytochemical analyses. This approach aligns with marker-assisted selection (MAS) strategies, enabling breeders to identify genotypes with desirable phytochemical traits based on easily measurable morphological characteristics⁴³. Integrating stepwise regression into plant breeding pipelines offers a viable and efficient alternative to traditional phytochemical screening methods. By identifying morphological traits that correlate strongly with target compounds, this approach reduces reliance on expensive analytical techniques, accelerating the selection of high-value genotypes. However, the observed trade-offs between specific traits (e.g., positive versus negative beta values) highlight the necessity for balanced selection strategies to mitigate any unintended consequences on plant growth or stress tolerance. Future research should seek to validate these regression models across diverse environments and genetic backgrounds to ensure their robustness. Additionally, exploring the underlying physiological and genetic mechanisms that drive the relationships between these traits and phytochemical profiles could further refine predictive models and enhance their applicability within breeding programs. Overall, this study emphasizes the potential of multivariate analysis, correlation and regression analysis as a powerful tool for linking morphological and phytochemical traits, paving the way for more efficient and targeted efforts in plant domestication.

The analysis of wild carrot populations revealed that the plant serves as a significant source of carotol. This compound, a type of oxygenated sesquiterpene, found in higher percentage in the studied populations. The importance of oxygenated sesquiterpenes has been recognized in various biological activities, including insecticidal, antioxidant, and allelopathic effects^44,45. Consequently, chemotypes that are rich in these compounds may represent promising candidates for industrial applications targeting these purposes^44,46,47. This study revealed the high contents of TFC, TPC, and antioxidant activity of the aerial parts (AP) and seed extracts were documented. This suggests their potential application as effective antioxidant agents^9,48. Therefore, the findings concerning the studied populations indicate that the plant holds considerable potential for using in breeding, facilitating its widespread cultivation. Notably, the DCP4 and DCP5 populations contain a high percentage of carotol, exceeding 73%, which can serve as a valuable source of this compound. Carotol is known for its diverse biological properties, including cytotoxicity, antioxidant effects, anti-inflammatory, antimicrobial, and antiviral activities^49,50,51.

The present study identified three valuable chemotypes, categorized as follows:

1. (I)

   Chemotype I: carotol—butrnyl acetate (DCP1).

2. (II)

   Chemotype II: carotol—α-pinene (DCP3).

3. (III)

   Chemotype III: carotol, with a high percentage ranging from 64.03 to 74.03% (DCP2, DCP4, and DCP5).

Studies on wild carrot have revealed significant geographic variation in the composition of essential oils across different plant tissues. In aerial parts sourced from Turkey, key compounds such as carotol (27%), elemicin (18.1%), and limonene (16%) were identified as major constituents¹⁹. Seeds collected from Montenegro at various maturity stages were characterized by high levels of β-bisabolene (32.3%) and 11-α-(H)-himachal-4-en-1-β-ol (27.9%)²⁰. In contrast, α-pinene (23.5%) emerged as the dominant compound in aerial parts of Moroccan D. carota²¹. Further analyses from Turkey samples highlighted carotol (1–74.6%) and β-bisabolene (0.9–62.4%) as key constituents, underscoring the chemical diversity from Turkey²². Algerian aerial parts, were rich in alismol (15.2%), (E)-β-caryophyllene (10.1%), myrcene (9.6%), α-humulene (9.5%), and β-ionone (5.2%)²³ Italian samples exhibited high concentrations of β-bisabolene (17.6–51.0%), while Portuguese samples were primarily composed of geranyl acetate (5.2–65.0%)²⁴. When compared to other geographic populations, the chemotypes identified in this study demonstrate both shared and unique traits. For instance, the prominence of carotol across all three chemotypes aligns with its widespread occurrence in D. carota from Turkey²². However, the specific combinations of carotol with butyl acetate (Chemotype I) or α-pinene (Chemotype II) highlight the influence of localized evolutionary pressures on secondary metabolite biosynthesis.

These chemotypes, exhibiting strong antioxidant and antibacterial activity, hold promise for development as new nutraceuticals and functional foods, contingent upon further supplementary studies. Additionally, to scale up production, domestication and breeding strategies should be considered. This study highlights the significant antibacterial properties of plant extracts from various populations, particularly DCP5 and DCP3, which exhibited enhanced levels of carotol, α-pinene, phenolic, and flavonoid compounds. The significant antibacterial effects of phenolic and flavonoid compounds have been reported in previous studies as well^52,53. The findings underscore the importance of these phytochemical traits in relation to environmental factors, as revealed by Canonical Correspondence Analysis (CCA) and scatter plots. By identifying the phytochemical characteristics that correlate with these environmental variables, we can establish valuable criteria for optimizing the cultivation and domestication of elite chemotypes of wild carrot. The demonstrated relationships between phytochemical and morphological traits through stepwise regression analysis offer a promising framework for reducing reliance on expensive analytical methods in the selection of plants for domestication and breeding, particularly within the context of marker-assisted selection (MAS)⁵⁴. Moreover the use of CCA, scatter plots, and cluster analysis provides important insights into the relationships among phytochemical, environmental, and morphological traits within the studied populations. These findings deepen our understanding of the diversity and interactions of these traits in the plant species under investigation⁵⁵. Future research should focus on investigating the specific mechanisms that contribute to the antibacterial activity of the identified compounds, as well as their interactions with various environmental factors. Expanding the scope of this study to encompass a wider range of populations and environmental conditions could further strengthen the reliability of the findings. One limitation of the current study is the potential variability in phytochemical profiles resulting from environmental fluctuations, which underscores the importance of conducting more extensive field trials to validate the results. Overall, these insights enhance our understanding of the diversity and interactions among phytochemical, environmental, and morphological traits in wild carrot species. This knowledge paves the way for more targeted strategies in breeding and cultivation, ultimately contributing to the development of elite chemotypes with desirable characteristics.