Creation of New Oregano Genotypes with Different Terpene Chemotypes via Inter- and Intraspecific Hybridization

Oregano is a medicinal and aromatic plant of value in the pharmaceutical, food, feed additive, and cosmetic industries. Oregano breeding is still in its infancy compared with traditional crops. In this study, we evaluated the phenotypes of 12 oregano genotypes and generated F1 progenies by hybridization. The density of leaf glandular secretory trichomes and the essential oil yield in the 12 oregano genotypes varied from 97–1017 per cm2 and 0.17–1.67%, respectively. These genotypes were divided into four terpene chemotypes: carvacrol-, thymol-, germacrene D/β-caryophyllene-, and linalool/β-ocimene-type. Based on phenotypic data and considering terpene chemotypes as the main breeding goal, six oregano hybrid combinations were performed. Simple sequence repeat (SSR) markers were developed based on unpublished whole-genome sequencing data of Origanum vulgare, and 64 codominant SSR primers were screened on the parents of the six oregano combinations. These codominant primers were used to determine the authenticity of 40 F1 lines, and 37 true hybrids were identified. These 37 F1 lines were divided into six terpene chemotypes: sabinene-, β-ocimene-, γ-terpinene-, thymol-, carvacrol-, and p-cymene-type, four of which (sabinene-, β-ocimene-, γ-terpinene-, and p-cymene-type) were novel (i.e., different from the chemotypes of parents). The terpene contents of 18 of the 37 F1 lines were higher than those of their parents. The above results lay a strong foundation for the creating of new germplasm resources, constructing of genetic linkage map, and mapping quantitative trait loci (QTLs) of key horticultural traits, and provide insights into the mechanism of terpenoid biosynthesis in oregano.


Introduction
Oregano is a perennial herb or semishrub of the genus Origanum and family Lamiaceae, which is mainly distributed and cultivated in the Mediterranean region, India, North America, and Mexico [1]. Only [2]. Oregano, also known as pizza grass, is widely used as a spice in southern Europe and the Americas to enhance the flavor of foods. The whole plant of oregano is used as a raw material for the extraction of essential oil, which is added to perfumes for bathing and sauna. Oregano has antibacterial, antioxidant, antiviral, anti-inflammatory, analgesic, and immune regulatory properties, and has, therefore, been used in traditional medicine to treat sunstroke, fever, acute gastroenteritis, dysentery, and other diseases [3,4]. Pure oregano essential oil is natural and has no side effects, which is an environmentally friendly and safe substitute for food additives [5], chemical preservatives [6], and feed additives [7][8][9]. materials with large phenotypic differences and high essential oil yield are usually selected for breeding. Molecular genetic analysis has been used to distinguish among the different oregano species and to study intra-and interspecific diversity [37][38][39][40][41]. We aim to develop SSR markers based on the oregano genome sequence to lay a foundation for subsequent molecular breeding.
In this study, through the evaluation of 12 oregano genotypes, we aimed to develop new chemotypes of oregano, which conclude high target terpenes in essential oil. Through increasing the contents of principal terpenes in essential oil, many hybrids will be developed by crossing different species, subspecies, and varieties of oregano. We will take into account the important traits, such as essential oil yield, glandular secretory trichome (GST) density, plant type (growth habit), and leaf and flower morphology. We speculate that our results will facilitate the breeding of new chemotypes or germplasm resources with higher terpene contents, which will lay a foundation for the breeding of new oregano varieties. Six crosses were performed using different oregano genotypes ( [Ovs], Ovv × Ovs, and Omh × Ovs), and F 1 progenies were obtained. SSR primers were developed based on the whole-genome sequence of O. vulgare (Sun et al., unpublished data) and used to verify the authenticity of 40 F 1 individuals resulting from the six crosses. Consequently, 37 F 1 lines were identified as true hybrids. This study will be very useful for the generation of new oregano varieties, the mechanism of terpenoid biosynthesis, molecular marker-assisted selection (MAS), construction of genetic linkage map, and mapping of quantitative trait loci (QTLs).

Phenotypic Evaluation of 12 Oregano Genotypes and Construction of F 1 Hybrids
Oregano has valuable medicinal and aromatic properties, with either erect or creeping growth habits. In this study, we evaluated the phenotypes of one wild Chinese and eleven European oregano accessions. All 12 oregano genotypes examined in this study were erect type (Figure 1a), but differed in leaf color, flower color, and biomass. The calyx of oregano is campanulate and is covered with glandular trichomes, while the corolla is longer and extends out of the calyx tube with four stamens. The flower phenotype of 12 oregano genotypes is shown in Figure 1b. Ovc, Ovc, and Olr have large flowers, purplish-red edges of calyx teeth, and deep pink corolla; Ovgs, Omh, and Ova produce pale pink flowers; Ovhs, Ovag, Ovs, Ovh, Ovt, and Ovv possess white flowers, but Ovs has pink anthers; Ov, Ovc, Olr, Ovs, and Ovh flowers possess both stigma and stamen which are male-fertile; and Ovgs, Omh, Ovhs, Ova, Ovag, Ovt, and Ovv have stigma but no stamen which are male-sterile.
GSTs are special structures that can secrete and store essential oils. The density of GSTs on leaves is an important factor affecting essential oil yield. Both the adaxial and abaxial surfaces of oregano leaves are covered with GSTs ( Figure 1c). GST density on the adaxial surface of leaves was the highest in Ovs and Ovhs, intermediate in Ovh and Ovv, and lowest in Ov and Olr (Figure 1d). Similarly, GST density on the abaxial surface of leaves was the highest in Ovs and Ovhs; intermediate in Ovh, Ova, Omh, and Ovt; and lowest in Olr, Ovgs, and Ovag. The GST densities of Ovs, Ovhs, Ovh, and Omh (1017, 900, 761, and 439 per cm 2 , respectively) were higher than those of other genotypes.
Next, cluster analysis was performed on the 12 oregano genotypes, based on 20 main terpenoids found in all genotypes. Results showed that the 12 oregano genotypes were grouped into four clusters ( Figure 2b). The first cluster included Ovv, Ovh, and Ovhs; because carvacrol was the main terpenoid in all three genotypes, these genotypes were defined as carvacrol-type. Olr, Ovs, Omh, and Ovgs grouped into the second cluster, with thymol as the main terpenoid, and were defined as thymol-type. Cluster three comprised Ov and Ovc, which contained higher contents of both germacrene D and βcaryophyllene; therefore, the chemotype of these genotypes was defined as germacrene D/β-caryophyllene-type. Similarly, cluster four contained Ova, Ovt, and Ovag, with linalool/β-ocimene-type chemotype.
Principal component analysis (PCA) was performed on the main terpenoids found in the essential oil of all 12 oregano genotypes. The results of PCA are shown in Figure 2c (composition score diagram of oregano essential oils) and Figure 2d (factor loading diagram). PCA divided the 12 oregano genotypes into four groups, consistent with the results of cluster analysis. Principal component 1 (PC1) explained 58.25% of the variation in essential oil composition of the 12 genotypes, and PC2 explained 21.34% variation. The factor loading diagram ( Figure 2d) showed that carvacrol, p-cymene, and 1-octen-3-ol were characteristic terpenoids of Ovv, Ovh, and Ovhs. Similarly, linalool and β-ocimene were the characteristic terpenoids of Ova, Ovt, and Ovag; β-caryophyllene and germacrene D were those of Ov and Ovc; and thymol and α-terpinene were those of Olr, Omh, Ovs, and Ovgs.

Development and Application of SSRs in Six Hybrid Combinations
Genome and annotation using high-fidelity (HiFi) and chromatin conformation capture (Hi-C) technologies revealed that O. vulgare (Chinese wild oregano) contains 15 chromosomes, with a total length of 641.87 Mb (Sun et al., unpublished data). The O. vulgare genome was highly repetitive. The annotated repetitive sequences (460.18 Mb) accounted for 71.70% of the entire genome. Additionally, a total of 394,276 tandem repeats were identified, accounting for 3.49% (22.38 Mb) of the whole genome (Sun et al., unpublished data). A total of 166,943 SSR loci were detected, of which 157,564 (94.38%) could be used for primer design, while 9379 SSR loci (5.62%) could not be used for primer design. Among these ten contigs, Contig000064 showed the highest frequency of SSR loci (439 SSR loci per Mb), and Contig000024 showed the lowest frequency (231 SSR loci per Mb) ( Figure 3a). Among the top ten contigs with the largest distribution of SSR sites, Contig000009 harbored the greatest number of SSR sites (5889), and Contig000093 showed the smallest number of SSR sites (3362) (Figure 3b). The majority of SSRs were trinucleotide repeats (12), followed by dinucleotide repeats (6) and tetranucleotide repeats (2)  A total of 300 primer pairs were designed to verify the authenticity of F 1 lines using SSR markers (Table S1). To test these primer pairs, polymerase chain reaction (PCR) amplification was first carried out on the parents of the six hybrid combinations shown in Table 2. Analysis of SSR polymorphisms between the parents revealed that one or two bands were generally amplified in the parents. After many repetitions, SSR primer pairs showing good specificity and clear and reproducible bands were selected for the authentication of hybrid progenies ( Figure 4). For example, the OvSSR176 marker showed codominance in the Omh × Ovc cross ( Figure 4a) and amplified specific bands in the female parent Omh (arrow 1) and male parent Ovc (arrow 2). Therefore, this primer could be used to screen the F 1 progeny of the Omh × Ovc cross. Similarly, OvSSR201 showed codominance in the Ovhs × Ovs cross ( Figure 4b) and produced specific bands in the female parent Ovhs (arrow 3) and male parent Ovs (arrow 4); OvSSR278 showed codominance in the Ovv × Ovh cross (Figure 4c; arrow 5 in the female parent Ovv and arrow 6 in the male parent Ovh). According to this method, 20 primer pairs were selected for screening the progeny of combination 1 (Omh × Ov; male parent Ovh). According to this method, 20 primer pairs were selected for screeni the progeny of combination 1 (Omh × Ov;    During genotyping, lines showing the bands of both parents or only one parent are true hybrids, while those showing non-parental bands are pseudohybrids. In combination 1 (Omh × Ov), for example, OvSSR083 amplified a specific band (arrow 1) from the female parent Omh and a specific band (arrow 2) from the male parent Ov. Among the seven hybrids, F1 lines 1, 2, 3, and 7 showed the amplification of bands from both parents. Genotyping using the OvSSR126 marker also revealed complementary bands in F1 lines 4 and 6. These results indicate that all six F1 lines are true hybrids. Moreover, nonparental bands appeared in the amplification results of F1 lines 4 and 6, indicating the appearance of new During genotyping, lines showing the bands of both parents or only one parent are true hybrids, while those showing non-parental bands are pseudohybrids. In combination 1 (Omh × Ov), for example, OvSSR083 amplified a specific band (arrow 1) from the female parent Omh and a specific band (arrow 2) from the male parent Ov. Among the seven hybrids, F 1 lines 1, 2, 3, and 7 showed the amplification of bands from both parents. Genotyping using the OvSSR126 marker also revealed complementary bands in F 1 lines 4 and 6. These results indicate that all six F 1 lines are true hybrids. Moreover, nonparental bands appeared in the amplification results of F 1 lines 4 and 6, indicating the appearance of new SSR loci and the abundance of genetic variation. However, F 1 line 5 showed only the maternal band when screened with all 20 pairs of SSR primers, which suggests that line 5 is a pseudohybrid (Figure 5a). In combination 2 (Ovv × Ovh), specific bands were amplified from the female parent Ovv (arrow 3) and male parent Ovh (arrow 4). Among the four Ovv × Ovh hybrids, F 1 lines 1, 2, and 3 showed both maternal and paternal specific bands, indicating that these lines were true hybrids; however, F 1 line 4 was a pseudohybrid since it produced no paternal band with any of the 23 SSR primer pairs (Figure 5b). Overall, among the 40 F 1 lines examined using SSR markers, 37 were true hybrids, and 3 were pseudohybrids ( Figure 5). SSR loci and the abundance of genetic variation. However, F1 line 5 showed only the maternal band when screened with all 20 pairs of SSR primers, which suggests that line 5 is a pseudohybrid (Figure 5a). In combination 2 (Ovv × Ovh), specific bands were amplified from the female parent Ovv (arrow 3) and male parent Ovh (arrow 4). Among the four Ovv × Ovh hybrids, F1 lines 1, 2, and 3 showed both maternal and paternal specific bands, indicating that these lines were true hybrids; however, F1 line 4 was a pseudohybrid since it produced no paternal band with any of the 23 SSR primer pairs (Figure 5b). Overall, among the 40 F1 lines examined using SSR markers, 37 were true hybrids, and 3 were pseudohybrids ( Figure 5). (a-f) Genotyping F1 lines of combination 1 (Omh × Ov) using the OvSSR083 marker (a), combination 2 (Ovv × Ovh) using the OvSSR185 marker (b), combination 3 (Omh × Ovc) using the OvSSR126 marker (c), combination 4 (Ovhs × Ovs) using the OvSSR165 marker (d), combination 5 (Ovv × Ovs) using the OvSSR184 marker (e), and combination 6 (Omh × Ovs) using the OvSSR120 marker (f). M, DNA Marker; ♀, female parent (indicated on the left in each combination); ♂, male parent (indicated on the right in each combination). Numbers represent F1 lines. Arrows 1, 3, 5, 7, 9, and 11 indicate characteristic bands of the female parents. Arrows 2, 4, 6, 8, 10, and 12 represent characteristic bands of the male parents. (a-f) Genotyping F 1 lines of combination 1 (Omh × Ov) using the OvSSR083 marker (a), combination 2 (Ovv × Ovh) using the OvSSR185 marker (b), combination 3 (Omh × Ovc) using the OvSSR126 marker (c), combination 4 (Ovhs × Ovs) using the OvSSR165 marker (d), combination 5 (Ovv × Ovs) using the OvSSR184 marker (e), and combination 6 (Omh × Ovs) using the OvSSR120 marker (f). M, DNA Marker; ♀, female parent (indicated on the left in each combination); ♂, male parent (indicated on the right in each combination). Numbers represent F 1 lines. Arrows 1, 3, 5, 7, 9, and 11 indicate characteristic bands of the female parents. Arrows 2, 4, 6, 8, 10, and 12 represent characteristic bands of the male parents.

Discussion
Among the 12 different oregano genotypes, Omh and Ovhs showed the highest essential oil yield (1.67% and 1.54%, respectively), followed by Ovh and Ovs (1.38% and 1.36%, respectively), while the essential oil yield of the remaining oregano genotypes varied from 0.17% to 0.65% (Figure 1e). In oregano, essential oil is synthesized in GSTs on the surface of stems, leaves, and flowers; thus, GSTs may be closely related to the yield and quality of essential oil [42]. Ovs, Ovhs, and Ovh showed relatively higher GST density on the adaxial and abaxial surfaces of leaves (1017, 900, and 761 per cm 2 , respectively) than the other genotypes (Figure 1d). We found that the higher the density of GSTs on leaves, the higher the essential oil yield of oregano. In addition, a certain correlation was observed between the yield of oregano essential oil and the extraction method. In this study, steam distillation was used to extract essential oil from whole oregano plants. However, essential oil could be extracted from the different parts of oregano plants (stem, leaf, and flower) separately to further compare and analyze its yield and terpene composition among the different organs. The yield of essential oil is also related to the harvest time, drying technique, and plant irrigation frequency [43,44]. The application of GC-MS and HS-SPME/GC-MS in this study was suitable for the determination of terpene compositions of oregano genotypes.
In this study, the 12 oregano genotypes showed variation in their essential oil contents and compositions (Figure 2a). For example, the main components of Omh essential oil were thymol (20.17%), terpinen-4-ol (14.90%), γ-terpinene (10.87%), and sabinene hydrate (10.71%). Ovh essential oil was mainly composed of carvacrol (35.59%), thymol (21.34%), and γ-terpinene (16.75%). The results showed that the dominant components in Ovh essential oil, carvacrol and thymol, were similar to those reported previously [22,27], but their proportions were slightly different. Similarly, the dominant components in Omh essential oil were similar to those reported previously [45], although their proportions were different. These differences may be related to environmental factors (such as temperature, humidity, and rainfall), harvest time, and location of plants [46]. This further proves that oregano is rich in genetic diversity, and its essential oil composition varies among the different varieties. In addition, essential oil extraction methods and plant growth environment also affect the composition and content of essential oil [47,48]. The thymol content of oregano essential oil determined in this study was slightly lower than that determined in previous reports, which may be used to set the distillation temperature. However, thymol and carvacrol are isomers, i.e., they can be reversibly converted into each other in plants under certain environmental conditions, which can also lead to differences in their contents in oregano essential oil [37].
The selection of parents is very important to achieve objectives of crossbreeding. To create excellent germplasm resources, materials with superior traits should be selected as parents. Essential oil yield and composition are important indicators of oregano quality. Understanding and mastering plant phenotypic diversity are of great significance for improving the utilization and creating of excellent germplasm resources. The focus of current research on oregano essential oil is on thymol and carvacrol, which are the main components of essential oil of most European oregano varieties [19][20][21][22]. However, the Chinese wild oregano used in this study contains the most abundant germacrene D and β-caryophyllene, which are reported to be important flavor compounds. Germacrene D exhibits antibiotic properties, repellent activity against insects and herbivores, insecticidal activity, and attractant or pheromone properties [18]. β-Caryophyllene, the main sesquiterpene in a variety of plant essential oils, is used as a fragrance in cosmetics and foods. It exhibits several important pharmacological properties including antioxidant, anti-inflammatory, anticancer, cardioprotective, hepatoprotective, antibacterial, immunomodulatory, and local anesthetic effects [17]. In this study, the chemotypes of European oregano accessions were mainly identified as thymol-and carvacrol-type, and these chemotypes are of important medicinal value. Therefore, oregano genotypes of different chemotypes were selected as parent materials to effectively aggregate these pharmacological properties into a single genetic background through crossbreeding.
Crossbreeding is an important approach for creating new germplasm resources because it can combine the favorable traits of parents into a single genetic background and enhance genetic diversity [35,36]. However, because of the long duration of the crossbreeding cycle, it is important to determine the authenticity of hybrids at an early stage. SSR markers are highly accurate, reproducible, and timesaving and are widely used to identify hybrid plants [31]. Therefore, crossbreeding and SSR markers can be applied together to improve the efficiency of germplasm development. In theory, a hybrid can be identified using one SSR primer pair [38][39][40][41]. However, in practice, the use of only one primer pair can lead to errors and have certain limitations. Therefore, to improve the identification accuracy, it is better to select multiple pairs of SSR primers for the simultaneous screening of a genotype. In this study, we observed the partial complementary bands of both parents and the generation of new bands in some F 1 lines ( Figure 5). It may be due to the high heterozygosity level of oregano, or the occurrence of homologous recombination or base mutation during meiosis. Thus, these markers could significantly improve the breeding efficiency of oregano by facilitating MAS, constructing F 2 progeny-based genetic linkage maps, and QTL mapping of key horticultural traits.
The chemotypes and contents of terpenes in the essential oils of 12 oregano genotypes were evaluated (Figure 2). The results showed that seven parental genotypes could be divided into three chemotypes: carvacrol-, thymol-, and germacrene D/β-caryophyllenetype. Similarly, the 37 hybrids could be divided into six chemotypes: sabinene-, β-ocimene-, γ-terpinene-, thymol-, carvacrol-, and p-cymene-type. Among these, the sabinene-, βocimene-, γ-terpinene-, and p-cymene-type chemotypes of hybrids were new, i.e., different from the chemotypes of their parents. Thus, this study shows that the identification of oregano hybrids using SSR markers is feasible. The application of SSR markers could significantly improve the breeding efficiency of oregano by providing a foundation for MAS.

Measurement of GST Density
GSTs were visualized using a stereomicroscope (Leica DVM6, Weztlar, Germany). The ImageJ software (http://rsb.info.nih.gov/ij) was used to count the number of GSTs and measure the leaf area. GST density was calculated based on three plants.

Extraction of Essential Oil
The essential oil from 12 oregano genotypes was isolated by steam distillation at 180-200 • C for 90 min. The essential oil yield (%) was calculated as volume (mL) of the isolated oil per 100 g of dried plant material. The isolated essential oil was dried using anhydrous sodium sulfate and stored at 4 • C until needed for further testing [49].

Analysis of Essential Oil by GC-MS
GC-MS analyses of 12 oregano essential oil samples was performed using Agilent 7890A-7000B gas chromatograph (Agilent, Palo Alto, CA, USA) equipped with Agilent 5975C MS detector (Agilent, USA). Volatiles were separated using the HP-5MS capillary column (30 m length, 250 m internal diameter [ID], 0.25 µm film thickness) and the following temperature program: 5 min at 60 • C, increase at the rate of 4 • C/min to 220 • C, increase at the rate of 60 • C/min to 250 • C, and hold at 250 • C for 5 min. The other parameters were as follows: injector and detector temperature, 250 • C; carrier gas, helium (He); flow rate, 1 m/min; split ratio, 1:10; acquisition range, 50-500 m/z in electron-impact mode; ionization voltage, 70 eV; and injected sample volume, 1 µL. The content of each compound (%) was determined based on the normalization of the GC peak areas. Identification of individual compounds was based on the comparison of retention indices (RIs), relative to a homologous series of n-alkanes (C7-C40), mass spectral data from the NIST library (v14.0), and data from scientific literature [50]. Retention indices were calculated using the following equation: where RT(x), RT(z), and RT(z + 1) for the retention time (RT) of the analyzed composition, previous n-alkane of analyzed composition and latter n-alkane of analyzed composition. Z for the number of carbons of the previous n-alkane of analyzed composition.

Hybridization Design
Considering essential oil composition and yield as the main breeding goals and taking into account other phenotypic traits, male-sterile genotypes (no or immature pollen) and male-fertile genotypes (with pollen) of oregano were used as the female parent and male parent, respectively. Different crosses were performed in 2020. Finally, six crosscombinations were selected for further analysis.

DNA Extraction
Leaves were collected from the plants of seven oregano accessions (Ovh, Ovv, Ovs, Ovhs, Ovc, Omh, and Ov) used as parents in six combinations (Omh × Ov, Ovv × Ovh, Omh × Ovc, Ovhs × Ovs, Ovv × Ovs, and Omh × Ovs), and 40 F 1 lines, immediately frozen in liquid nitrogen, and stored at −80 • C. DNA was extracted from the leaves using DNA Secure Plant Kit (Tiangen, Beijing, China). The concentration and quality of the isolated DNA were assessed by electrophoresis on 1% agarose gel and using 2.0 Fluorometer (Life Technologies, Carlsbad, CA, USA).

Genotyping Using SSR Markers
All PCRs were performed on the PCR system (Bio-Rad, Hercules, CA, USA) in a 10 µL reaction volume containing 2 µL (20 ng/µL) of genomic DNA, 3 µL of 2 × Taq PCR Master Mix II (Tiangen, Beijing, China), 2 µL of forward and reverse primer mixture, and 3 µL of double distilled water (ddH 2 O). The thermocycling conditions were as follows: 94 • C for 3 min; 6 cycles of 94 • C for 45 s, 55-65 • C for 1 min, and 72 • C for 1 min; 9 cycles of 94 • C for 45 s, 50-58 • C for 1 min, and 72 • C for 1 min; 19 cycles of 94 • C for 30 s, 50 • C for 30 s, and 72 • C for 1 min; and final extension at 72 • C for 5 min. Amplification products were analyzed by electrophoresis on 8.0% (w/v) denaturing polyacrylamide gel in TBE buffer for 1 h using the DYY-6C electrophoresis apparatus (Beijing Liuyi Instrument Factory, Beijing, China) under a constant voltage of 220 V. DNA fragments were then visualized by silver staining (Silver sequence staining reagents, Promega, Madison, WI, USA) and sized with 50 bp DNA ladder (Tiangen, Beijing, China) [51]. SSR primer sequences are listed in Table S1.

Analysis of Leaf VOCs by HS-SPME
The leaf VOCs of seven parental genotypes (Ovh, Ovv, Ovs, Ovhs, Ovc, Omh, and Ov) and 40 F 1 lines were detected by HS-SPME. Briefly, 0.25 g of fresh leaf powder was weighed and immediately placed into a 20 mL headspace vial (Agilent, Palo Alto, CA, USA) containing 20 µL of internal standard solution (1 mg/mL 3-octanol, Cas#589-98-o; Aladdin, Shanghai, China). The vials were sealed using crimp-top caps with TFE-silicone headspace septa (Agilent, Palo Alto, CA, USA). Subsequently, each vial was immediately incubated at 40 • C for 30 min. Then, a 100 µm thick polydimethylsiloxane (PDMS)-coated fiber (Supelco, Inc., Bellefonte, PA, USA) was exposed to the headspace for 30 min to absorb the volatiles. All VOCs on the PDMS-coated fiber were then analyzed by GC-MS using Model 7890A GC instrument and 7000B mass spectrometer (Agilent, Palo Alto, CA, USA), as described previously [52], under the following conditions: injector and transfer line temperature, 250 • C and 250 • C, respectively; column temperature, 50 • C for 3 min, followed by a gradual increase to 150 • C at 4 • C/min for 2 min and a final increase to 250 • C at 8 • C/min for 5 min; carrier gas, He; flow rate, 1 mL/min; and injection mode, splitless. The identity of VOCs was determined by comparing their retention times with those of authentic standards. Agilent MassHunter 5.0 was used to analyze the chromatograms and mass spectra. VOCs were identified by comparing the retention times of individual peaks and identified the mass spectra using the mass spectra in the NIST database v14.0 and the literature [50]. Retention indices were calculated using the equation in Section 4.4.

Statistical Analysis
All samples were prepared and analyzed in triplicate, and data were expressed as mean ± standard deviation. All statistical analyses were performed using the SPSS software (version 23.0; SPSS, Chicago, IL, USA). PCA and other types of analyses were performed online (https://www.metaboanalyst.ca). PCA plots were evaluated using Unscrambler X (version 10.4). To further distinguish chemotypes of essential oils, a supervised statistical treatment was run using Origin (version 2021). Histogram of the relative contents of terpenoids was generated using GraphPad Prism (Version 8.3.0.538) [53].

Conclusions
Oregano terpenoid compositions have numerous applications in pharmaceutical, food, feed additive, and cosmetic industries, owing to their antioxidant, antibacterial, antifungal, anti-inflammatory, antiviral, and immunological properties. In this study, we evaluated the phenotypes of 12 oregano genotypes and generated F 1 progenies by hybridization. SSR markers were developed based on unpublished whole-genome sequencing data of Origanum vulgare. These codominant primers were used to determine the authenticity of F 1 lines. The chemotypes and contents of terpenes in the essential oils of seven parental genotypes could be divided into three chemotypes: carvacrol-, thymol-, and germacrene D/β-caryophyllene-type. Similarly, the F 1 hybrids could be divided into six chemotypes: sabinene-, β-ocimene-, γ-terpinene-, thymol-, carvacrol-, and p-cymene-type. Among these, the sabinene-, β-ocimene-, γ-terpinene-, and p-cymene-type chemotypes of hybrids were new, i.e., different from the chemotypes of their parents. Thus, this study laid a strong foundation for the creation of new germplasm resources, the construction of a genetic linkage map, and mapping of QTLs of key horticultural traits, and provided insights into the mechanism of terpenoid biosynthesis in oregano.