The size of viable pollen is correlated with ploidy level, as evidenced by a polyploid series of Ocimum L. from Thailand

Abstract

The genus Ocimum (Lamiaceae) comprises approximately 65 aromatic species. The taxonomic classification of Ocimum is often very complicated due to centuries of selection and breeding, as well as spontaneous events, such as speciation, hybridization, and polyploidy. Our previous studies identified a ploidy-stable series of Ocimum species sharing the chromosome base number x=13 from Thailand: diploid O. americanum (2n=26), tetraploid O. basilicum (2n=52) and hexaploid O. africanum (2n=78). In this study, we measured and analyzed the sizes of viable pollen with the aim of characterizing the gametophytic features of this polyploid series. Samples were collected from three Ocimum species originating from two geographically distant locations in Thailand, Chiang Mai (northern) and Prachuab Khiri Khan (southwestern) provinces. A total of 29,418 pollen grains were analyzed. The results revealed positive and significant correlations between pollen size and ploidy level in O. americanum (43.9–63.5 µm), O. basilicum (52.9–73.6 µm), and O. africanum (57.7–80.8 µm) (Spearman’s rho=0.900, n=28,431, p＜0.001). An overlapping size distribution between nonviable and viable pollen grains was observed for all three species, probably indicating aneuploid viable pollen. Notably, polyploid pollen was also detected in the diploid species O. americanum, possibly because the pollen contains unreduced gametes, which can drive polyploidization. Despite the existence of polyploid and supposed aneuploid pollen grains, the viability of the pollen that formed in these species was exceptionally high: O. americanum (97.6–99.5%), O. basilicum (87.6–98.5%) and O. africanum (90.4–99.4%). These viable, unbalanced pollen grains should be capable of fertilization, resulting in aneuploidy and polyploidy.

The genus Ocimum L., generally known as basil, belongs to the Lamiaceae family. This genus comprises approximately 65 annual and perennial herbs and shrubs distributed worldwide; however, Ocimum species are thought to have occurred naturally in tropical Africa and were subsequently introduced to tropical Asia and America (Paton et al. 1999; Paton et al. 2004). Ocimum is one of the most important genera in the production of essential oils and is important to indigenous systems of medicine, and it provides fresh and dry herbs for the culinary market (Lawrence 1992; Suddee et al. 2005; Ben-Naim et al. 2015; Patel et al. 2015; Shahrajabian et al. 2020; Karataş 2022). Over the past several centuries, Ocimum has been cultivated from accessions obtained through backcross breeding for culinary and medicinal purposes and by local selection from natural hybrids and polyploids (Pyne et al. 2018); therefore, there is enormous genetic and phenotypic diversity among Ocimum species and accessions. Moreover, numerous morphological and chemical traits are developmentally and environmentally dependent, complicating the taxonomy of the genus (Paton and Putievsky 1996; Rewers and Jedrzejczyk 2016; Chowdhury et al. 2017; Srivastava et al. 2018).

The chromosome number is a discrete character that can help delineate species within a genus. For Ocimum species in Thailand, a polyploid series based on x=13 has been described (Lekhapan et al. 2019), and this is the subject of the present study. This series comprises diploid (2n=26) O. americanum L. (Common name: Hoary basil; Thai name: Maeng-Ka-Saeng), tetraploid (2n=52) O. basilicum L. (Common name: Sweet basil; Thai name: Ho-Ra-Phaa), and hexaploid (2n=78) O. africanum Lour. (Common name: Lemon basil; Thai name: Meang-Lak). Lekhapan et al. (2019) examined 33 accessions of these three Ocimum species from five geographically distant provinces in Thailand and found that all three of these species were ploidy stable; i.e., no within-species variation in the 2n numbers was detected.

In contrast, numerous records in the Chromosome Counts Database (CCDB) show that Ocimum possesses a large variability in chromosome numbers both in terms of intra- and interspecific variation (Rice et al. 2015). For example, O. americanum individuals with 2n=24, 26, 48, 52, and 64 have been described (Morton 1962; Vij and Kashyap 1976; Pushpangadan and Sobti 1982; Khosla 1995; Idowu and Oziegbe 2017; Ben-Naim et al. 2018), while O. africanum individuals with 2n=64 and 72 have been recorded (Khosla 1995; Paton and Putievsky 1996; Carović-Stanko et al. 2010; Ben-Naim et al. 2018). O. basilicum is also known for a high level of chromosome number variation, ranging from 2n=48 to 50, 52, 53, 54, 56, 60, 72, and 74 (Morton 1962; Mehra and Gill 1972; Khosla 1995; Paton and Putievsky 1996; Mukherjee et al. 2005; Carović-Stanko et al. 2010; Dhasmana 2013; Edet and Aikpokpodion 2014; Dash et al. 2017; Ben-Naim et al. 2018). These 2n numbers in Ocimum indicate frequent hybridization, polyploidization, aneuploidy, and dysploidy.

Polyploidy is known to affect cell size, although it is often circumstantial (Otto 2007; Clo and Kolář 2021; Heslop-Harrison et al. 2023). A positive relationship between the genome and cell size has been confirmed for functional sporophytic cell types, such as guard cells (Beaulieu et al. 2008), but for gametophytic cells, such as pollen, the correlation is not straightforward.

Most studies on Ocimum pollen involve the well-known basil species, O. basilicum. The equatorial diameter (size) of O. basilicum pollen has been described by several research groups. The equatorial diameter of O. basilicum from India was measured to be 53–64 µm (Kumari et al. 2022; Chiranjeevi and Chaya 2023), whereas a sample from Egypt exhibited a larger equatorial diameter of 66–71 µm (Azzazy 2016). Doaigey et al. (2018) examined pollen grains of O. basilicum from Saudi Arabia and found the equatorial diameter to be on the smaller side at 55.81 µm. Moreover, the size of the pollen of O. americanum was also highly variable. Arogundade and Adedeji (2009) reported that the pollen sizes of species from Nigeria, O. americanum, had pollen diameter of 79.01±1.09 µm, whereas two O. basilicum varieties had pollen diameters of 65.30±0.94 µm and 44.80±0.73 µm. Furthermore, the equatorial diameter of O. americanum from Pakistan was found to be 55.00–60.05 µm (Perveen and Qaiser 2003). This inconsistency in pollen diameter possibly reflects differences in cytotype caused by aneuploidy and polyploidy or differences in cultivar development in different locations, and differences due to taxonomic discrepancies. In the present study, we taxonomically verified the plant samples for pollen analysis and determined their chromosome numbers.

Although the Thai Ocimum polyploid series investigated here is ploidy stable, our previous studies of meiotic behavior and mitotic karyotypes (Lekhapan et al. 2019; Lekhapan et al. 2021) indicated active evolution, such as ongoing diploidization, in the polyploid genomes. Meiotic pairing irregularities were observed among the polyploid Ocimum species in this series, i.e., tetraploid O. basilicum and hexaploid O. africanum, indicating early stages of polyploidization. These two polyploid species were also found to be karyotypically unstable, indicating chromosomal rearrangements. Meiotic abnormalities, such as those in O. basilicum and O. africanum, are frequent in polyploid plants and possibly contribute to genetically unbalanced pollen and affect pollen viability (Ramsey and Schemske 2002), hence reducing plant fertility. In Ocimum breeding programs, it is important to evaluate fertility, most effectively by pollen, to determine the crossability of accessions with desirable characteristics, whether agronomical or end-use. Chromosomal rearrangements via meiotic pairing, on the other hand, enhance genotypic heterogeneity among progenies, hence increasing the genetic diversity of the population or the species (Lysák and Schubert 2013; Nibau et al. 2022). For Ocimum, high diversity increases the value of the genetic resources needed in breeding programs, and this diversity is worthy of further identification and conservation for future applications.

However, information about Ocimum pollen size and viability is still limited and insufficient to demonstrate a correlation between pollen size and ploidy level or to reveal the effect of meiotic abnormalities on pollen viability. Therefore, the objective of this study was to assess pollen size and viability in the x=13 polyploid series of Ocimum, which included diploid O. americanum, tetraploid O. basilicum, and hexaploid O. africanum.

Materials and methods

Plant materials

Whole plants of O. americanum, O. basilicum, and O. africanum were collected from Chiang Mai (northern) and Prachuab Khiri Khan (southwestern) provinces between 2019 and 2023 (Table 1). Sample collection sites were chosen because O. americanum existed only at these two locations. All plant materials were grown in pots in the greenhouse of the Department of Botany, Faculty of Science, Chulalongkorn University, Bangkok, Thailand. Voucher specimens were deposited at Professor Kasin Suvatabhandhu Herbarium, Chulalongkorn University (BCU).

Table 1. Sample collection sites of Ocimum species and accessions in Thailand.

Species	Accession no.	Herbarium no.	Collection site
O. americanum	OAMC001	17593	Chiang Mai, Yang Khram (330 MASL, 18°33′40.4″N, 98°47′10.9″E)
	OAMC002	17597
	OAMC003	17598
	OAMP004	17587	Prachuab Khiri Khan, Thap Sakae (12 MASL, 11°29′53.3″N 99°37′45.6″E)
	OAMP005	17588
	OAMP006	17589
O. basilicum	OBAC007	17583	Chiang Mai, Yang Khram (307 MASL, 18°34′44.3″N, 98°48′15.2″E)
	OBAC008	17582
	OBAC009	17584
	OBAP010	17591	Prachuab Khiri Khan, Thap Sakae (12 MASL, 11°29′55.6″N 99°37′31.9″E)
	OBAP011	17590
	OBAP012	17594
O. africanum	OAFC013	17585	Chiang Mai, Yang Khram (307 MASL, 18°34′44.3″N, 98°48′15.2″E)
	OAFC014	17586
	OAFC015	17599
	OAFP016	17592	Prachuab Khiri Khan, Thap Sakae (12 MASL, 11°29′55.6″N 99°37′31.9″E)
	OAFP017	17595
	OAFP018	17596

Mitotic chromosome preparations

The chromosome numbers of representative accessions of O. americanum, O. basilicum, and O. africanum were counted to determine the ploidy level. Actively growing root tips (1–2 cm) were harvested and pretreated with saturated para-dichlorobenzene at 27°C for 1–3 h, after which the materials were fixed in freshly prepared ethanol-acetic acid fixative (3 : 1 v/v) at 4°C for a minimum of 12 h. Somatic chromosome spreads were prepared following the enzymatic digestion method as described previously by Chow et al. (2020), with minor modifications. The fixed root tips were rinsed in enzyme buffer (citric acid and sodium citrate buffer), incubated in an enzyme mixture containing pectinase (10 U mL⁻¹; Merck no. 1.06021; Germany) and cellulase (16 U mL⁻¹; Merck no. 1.02321; Germany) at 37°C for 10–15 min, and subsequently squashed in 45% acetic acid. Cover slips were removed after freezing with dry ice, and the slides were air-dried. The preparations were counterstained with 1 µg mL⁻¹ 4,6-diamidino-2-phenylindole (DAPI) and observed using an Olympus BX51 fluorescence microscope, and the images were captured at ×1,000 magnification with an Olympus DP70 camera.

Analysis of pollen size and viability

For analysis, pollen grains were collected from three individual plants of each species and at each location. Fresh pollen samples were dissected from four anthers of a flower and mounted in a drop of 2% aceto-carmine on a microscope slide. All preparations were observed using an Olympus BX43 microscope, and images at ×200 magnification were captured with an Olympus DP27 camera using DP2-SAL camera control software (Tokyo, Japan). Pollen grains were considered viable when they appeared rounded and well-stained, according to Rathod et al. (2018). Pollen viability was calculated as the proportion of rounded and stainable pollen out of total pollen grains scored per flower. The diameter of stained spherical pollen grains was measured and subsequently used to establish a histogram of pollen diameter distribution following Bretagnolle (2001). In total, more than 700 pollen grains per sampled plant were recorded. Spearman’s rank order correlation coefficient was calculated using SPSS software (version 28.0.0.0) to statistically test the relationship between pollen size and ploidy level.

Results

Mitotic chromosome numbers

For each of the Ocimum species in the present study, the chromosome numbers obtained from the two collection sites, Chiang Mai and Prachuab Khiri Khan provinces, were identical. O. americanum is a diploid species with 2n=26, O. basilicum is a tetraploid with 2n=52, and O. africanum is a hexaploid species with 2n=78 (Fig. 1).

Fig. 1. Metaphase chromosome spreads of Ocimum.

Chromosomes were counterstained with DAPI and observed using a fluorescence microscope. (A) O. americanum OAMC001 (2n=26), (B) O. americanum OAMP004 (2n=26), (C) O. basilicum OBAC007 (2n=52), (D) O. basilicum OBAP011 (2n=52), (E) O. africanum OAFC014 (2n=78), and (F) O. africanum OAFP016 (2n=78). Scale bars=5 µm.

Pollen size and viability

A total of 29,418 pollen grains were analyzed. The pollen morphologies of O. americanum (OAMC002, OAMP006), O. basilicum (OBAC007, OBAP012), and O. africanum (OAFC014, OAFP016) based on light microscopy are shown in Fig. 2. The pollen grains examined, both viable (positively stained with aceto-carmine) and nonviable (negatively stained), appeared spherical, with rough and reticulate exine walls. The pollen diameter (µm) and pollen viability (%) of all accessions of O. americanum, O. basilicum, and O. africanum are shown in Table 2. Pollen grains are considered viable when they appear rounded and well-stained; in contrast, unstained pollen grains (often small and sometimes shrunken) are considered nonviable (Fig. 2). The pollen size distributions based on the grain diameters are illustrated in Fig. 4.

Fig. 2. Pollen grains of Ocimum.

Pollen grains were stained with 2% aceto-carmine. All preparations were observed using a light microscope. Pollen grains were considered viable when they appeared rounded and well-stained, in contrast, unstained pollen grains were considered nonviable, according to Rathod et al. (2018). (A) O. americanum OAMC002, (B) O. americanum OAMP006, (C) O. basilicum OBAC007, (D) O. basilicum OBAP012, (E) O. africanum OAFC014 and (F) O. africanum OAFP016. Arrows indicate nonviable pollen grains. Scale bars=100 µm.

Table 2. Pollen diameter and pollen viability of O. americanum, O. basilicum, and O. africanum sampled in Thailand.

Species	Accession number	Total number of pollen grains	Viable Pollen diameter (µm)		Nonviable Pollen diameter (µm)	% Viability	Chromosome number
			Normal pollen	Oversized pollen
O. americanum	OAMC001	770	47.8–63.5	—	35.6–40.4	99.4	2n=26
	OAMC002	1219	47.5–62.9	69.4	36.0–50.1	98.4	—
	OAMC003	1080	47.6–62.1	—	25.9–47.6	98.1	—
	OAMP004	923	50.2–62.7	73.0–75.1	39.4–47.9	99.4	2n=26
	OAMP005	1269	43.9–60.1	—	35.7–47.3	97.6	—
	OAMP006	1148	48.5–63.4	—	37.9–43.7	99.5	—
O. basilicum	OBAC007	3041	54.7–71.7	—	42.9–58.1	98.5	2n=52
	OBAC008	2909	55.1–73.6	—	46.4–59.3	99.3	—
	OBAC009	3456	54.1–69.6	—	42.1–56.2	87.6	—
	OBAP010	1840	56.0–71.6	—	40.6–56.8	98.4	2n=52
	OBAP011	1612	52.9–71.3	—	37.5–50.9	95.0	—
	OBAP012	2162	57.4–70.7	—	43.2–58.3	98.2	—
O. africanum	OAFC013	1052	61.5–79.4	—	54.3–68.2	97.7	2n=78
	OAFC014	1411	61.5–79.7	—	47.2–62.5	90.4	—
	OAFC015	1492	57.7–76.7	—	40.8–59.5	94.9	—
	OAFP016	1324	64.7–80.8	—	54.6–65.6	99.4	2n=78
	OAFP017	1237	63.0–80.1	—	58.2–65.4	99.4	—
	OAFP018	1473	63.9–78.8	—	51.8–68.3	97.8	—

For the diploid O. americanum (Figs. 2A, B, 4A, B; Table 2), the diameter of the viable pollen ranged from 47.5 to 63.5 µm for the accessions from Chiang Mai, with 98.1–99.4% pollen viability, whereas the diameter of the viable pollen ranged from 43.9 to 63.4 µm for the accessions from Prachuab Khiri Khan, with 97.6–99.5% pollen viability. The sizes of the unstained (nonviable) grains were 25.9–50.1 µm for the accessions from Chiang Mai and 35.7–47.9 µm for the accessions from Prachuab Khiri Khan. This means that the distribution of pollen sizes overlaps between viable and nonviable pollen grains in some O. americanum accessions (OAMC002, OAMC003, and OAMP005). Remarkably, ‘oversized’ but viable pollen grains with a diameter of 69.4 µm were found in the accession from Chiang Mai (0.08% occurrence in OAMC002) (Fig. 3A) and with a diameter range of 73.0–75.1 µm were found in the accessions from Prachuab Khiri Khan (0.22% occurrence in OAMP004) (Fig. 3B).

Fig. 3. Viable oversized pollen grains produced by diploid O. americanum..

(A) An oversized pollen grain with a diameter of 69.4 µm produced by O. americanum accession from Chiang Mai (OAMC002). (B) An oversized pollen grain with a diameter of 75.1 µm produced by O. americanum accession from Prachuab Khiri Khan (OAMP004). Arrows indicate oversized pollen grains. Scale bars=100 µm.

Among the tetraploid O. basilicum (Figs. 2C, D and 4C, D; Table 2), all three accessions from Chiang Mai exhibited a high percentage of pollen viability (87.6–98.5%). The diameter range of viable pollen grains from these accessions was 54.1–73.6 µm. The pollen viability among all the accessions of O. basilicum from Prachuab Khiri Khan was 95.0–98.4%, and the diameter range of the viable pollen was 52.9–71.6 µm, similar to that of the accessions from Chiang Mai. The diameter range of viable pollen of O. basilicum was apparently greater than that of diploid O. americanum. The unstained (nonviable) pollen grains found in the accessions of O. basilicum from Chiang Mai and Prachuab Khiri Khan had diameter ranges of 42.1–59.3 µm and 37.5–58.3 µm, respectively. Nonviable pollen is generally small; however, the larger end of its diameter distribution overlaps with the size distribution of viable pollen from both collection sites.

Similarly, compared with the other two species, the hexaploid O. africanum (Figs. 2E, F and 4E, F; Table 2) had the highest range of pollen diameters in the present study but no decrease in viability. The diameters of stained (viable) pollen grains of O. africanum accessions from Chiang Mai and Prachuab Khiri Khan were 57.7–79.7 µm and 63.0–80.8 µm, respectively. The diameter of the unstained pollen grains of the accessions from Chiang Mai ranged from 40.8 to 68.2 µm, while the diameter of the unstained pollen grains of the accessions from Prachuab Khiri Khan ranged from 51.8 to 68.3 µm, revealing an overlapping size distribution with the viable pollen. The percentages of pollen viability in O. africanum were 90.4–97.7 for the accessions from Chiang Mai and 97.8–99.4 for the accessions from Prachuab Khiri Khan.

Frequency distribution of pollen diameter and correlations between pollen size and ploidy level

For each species from each collection site, the pollen diameter data from all the accessions were combined (oversized pollen was excluded), and histograms of the pollen diameter distributions were established (Fig. 4). The six datasets analyzed here included 1) O. americanum accessions from Chiang Mai; 2) O. americanum accessions from Prachuab Khiri Khan; 3) O. basilicum accessions from Chiang Mai; 4) O. basilicum accessions from Prachuab Khiri Khan; 5) O. africanum accessions from Chiang Mai; and 6) O. africanum accessions from Prachuab Khiri Khan. The frequencies of pollen diameters in all six groups followed a normal distribution. In addition, when the pollen diameter data from all species and all accessions from the two collection sites were combined and used for constructing the frequency distribution histogram, the results showed that the pollen sizes of O. americanum, O. basilicum, and O. africanum were distributed in three different ranges with minor overlapping areas (Fig. 5). The results of the statistical analysis indicated that pollen size was positively and significantly correlated with ploidy level (Spearman’s rho=0.900, n=28,431, p＜0.001).

Fig. 4. Frequency distribution of viable pollen diameter in Ocimum species.

(A) O. americanum accessions from Chiang Mai and (B) Prachuab Khiri Khan, (C) O. basilicum accessions from Chiang Mai and (D) Prachuab Khiri Khan, and (E) O. africanum accessions from Chiang Mai and (F) Prachuab Khiri Khan.

Fig. 5. Range of size variability of pollen grains in Ocimum species.

O. americanum (), O. basilicum (), and O. africanum (). The pollen sizes of O. americanum, O. basilicum, and O. africanum were distributed in different ranges with minor overlapping areas.

Discussion

High pollen viability reflects ploidy stability

The chromosome numbers of O. americanum (diploid, 2n=2x=26), O. basilicum (tetraploid, 2n=4x=52) and O. africanum (hexaploid, 2n=6x=78) obtained in this study are consistent with our previous reports on the chromosome numbers of Ocimum species in Thailand (Lekhapan et al. 2019; Lekhapan et al. 2021), thus confirming that the different ploidy levels in this polyploid series share the same base number, x=13. The chromosome morphology of each species (Fig. 1) was also in good agreement with the karyotype described in the study by Lekhapan et al. (2021), both in terms of karyotypic composition and symmetry. The karyotypes of these Ocimum species revealed close genetic relationships in this polyploid series.

All species in the present study are relatively ploidy stable, as expected from our previous meiotic analysis (Lekhapan et al. 2019). This is supported by the pollen viability results. The six accessions of the diploid O. americanum, the tetraploid O. basilicum, and the hexaploid O. africanum showed viability percentages of 97.6–99.5, 87.6–99.3, and 90.4–99.4, respectively (Table 2). This finding is in good agreement with the calculated occurrence of bivalent meiotic pairs: 100% 13_II in O. americanum, 96.7–100% 26_II in O. basilicum, and 96.7% 39_II in O africanum (Lekhapan et al. 2019). Irregularities in chromosome pairing at meiosis often result in unbalanced and aborted gametes (Ramsey and Schemske 2002), leading to abnormal pollen development. The present study detected only a very small proportion of unstained, nonviable pollen grains, i.e., within 2.5% in the diploid species O. americanum but significantly more in the polyploid species, or approximately 1–12% in O. basilicum and 1–10% in O. africanum. High pollen viability is an advantage for the species itself, at least from the paternal side, as successful sexual reproduction increases genetic heterogeneity in the population. In Ocimum breeding programs, it is important to evaluate fertility, most effectively by pollen, to determine the crossability of accessions with desirable characteristics, whether agronomical or for end-uses.

Pollen size is significantly correlated with ploidy level

Pollen diameter was measured from light microscopy images of pollen grains from this Ocimum polyploid series with the aim of testing whether pollen size would be directly correlated with ploidy level. The pollen diameters of O. americanum, O. basilicum, and O. africanum were found to be in comparable ranges between the two collection sites (Table 2) which were geographically separated by approximately one thousand km (north–south). The two collection sites are completely different in terms of environmental conditions, such as altitude and the regional means of annual temperature and rainfall (Table 1). The environment does not seem to have any effect on pollen size. When the pollen diameter data from all studied species and accessions from both collection sites were combined and used for constructing the frequency distribution histogram (Fig. 5), it became apparent that the pollen sizes of O. americanum, O. basilicum and O. africanum were distributed in different ranges with minor overlapping areas. Therefore, the frequency distribution histogram of pollen diameters can be used to support taxonomic delineation among these three closely related Ocimum species.

Overall, the pollen size of this Thai Ocimum polyploid series, excluding oversized pollen, was positively and significantly correlated with ploidy level. The diploid O. americanum had the lowest range of pollen sizes (43.9–63.5 µm), the tetraploid O. basilicum had an intermediate range (52.9–73.6 µm), and the hexaploid O. africanum had the highest range of pollen sizes (57.7–80.8 µm) (Table 2). The increase in mean sizes was approximately 15% from diploid to tetraploid and approximately 11% from tetraploid to hexaploid (Fig. 5). Karlsdóttir et al. (2007) examined pollen from 70 ploidy-identified individual birch (Betula L.) trees/shrubs in natural woodlands and found that pollen grains from the tetraploid species (B. pubescens Ehrh.) were approximately 16% larger than those of their diploid counterpart (B. nana L.), which is similar to the increase in size from 2x to 4x in Ocimum in the present study. In Betula, the genome size of the tetraploid species is exactly double that of the diploid species (Anamthawat-Jónsson et al. 2010), but the increase in pollen size is much less. Nevertheless, the present study indicates a direct relationship among species of different ploidies at the microevolutionary level. Knight et al. (2010) reported that congeneric species exhibited a more positive trend than that expected by chance, and these results may reflect strong selection pressure for pollen to be small. This may also explain the smaller increase in pollen size in our Ocimum species at higher ploidy levels, such as from 4x to 6x (Fig. 5).

The range of pollen sizes of Thai O. basilicum in this study covers the highly diverse pollen size ranges of O. basilicum sampled from India, Egypt, Saudi Arabia, and Nigeria (Arogundade and Adedeji 2009; Azzazy 2016; Doaigey et al. 2018; Kumari et al. 2022; Chiranjeevi and Chaya 2023), while the pollen size of O. americanum sampled from Pakistan is in the same range as that of Thai O. americanum (Perveen and Qaiser 2003). Besides, the range of pollen sizes of O. africanum was first illustrated in this study.

The oversized pollen is likely polyploid

The oversized pollen grains found in the diploid O. americanum accessions OAMC002 and OAMP004 are comparable in size to the pollen of either O. basilicum or O. africanum (Table 2). As the meiosis of both the tetraploid O. basilicum and the hexaploid O. arfricanum is mostly normal, i.e., they form more than 96% bivalents, their pollen is diploid and triploid, respectively. Therefore, the observed oversized pollen grains produced by O. americanum could very well be polyploid pollen containing unreduced gametes.

The most common means of polyploidy, both auto and allopolyploidy, are indeed through the fertilization of unreduced gametes either with normally reduced (1n) or unreduced euploid gametes (Ramsey and Schemske 1998; Mason and Pires 2015). An extensive literature compilation and analysis by Kreiner et al. (2017) demonstrated that unreduced gametes are taxonomically widespread in plants and are likely a common mechanism underlying polyploid formation. In addition, the frequency of the oversized pollen grains in O. americanum relative to the total number of viable grains of this diploid species is 0.08–0.22%, which falls within the frequency estimated by Ramsey (2007); this value is approximately 0.1–2.0% of the gametes in a natural nonhybrid population or species. Regarding possible allopolyploidy in Ocimum, some studies suggest the presence of the O. americanum genome in polyploid species, such as O. basilicum and O. africanum (Pushpangadan and Sobti 1982; Suddee et al. 2005; Gonda et al. 2020; Vineesh et al. 2023). The union of reduced and unreduced gametes can generate polyploid offspring; hence, it could be hypothesized that polyploid Ocimum species, such as O. basilicum and O. africanum, may have formed via unreduced pollen production in diploid populations of O. americanum. Oversized pollen from diploid Ocimum species is reported here for the first time; that is, no other reports are available to date.

The overlapping distribution of pollen size between nonviable and viable pollen may serve as evidence of aneuploidy

An overlapping distribution of pollen sizes between nonviable and viable pollen grains was observed for O. americanum, O. basilicum, and O. africanum (Table 2). Nonviable pollen grains may contain incomplete chromosome complements and are therefore unlikely to mature into functional/stainable pollen (Tel-Zur et al. 2004). Viable pollen grains may also contain a certain level of incompleteness. The wide range of viable pollen diameters usually indicates different ploidy levels or chromosome number variations owing to meiotic abnormalities (Carputo et al. 1995; Wang et al. 2010; Kaur and Singhal 2019). Unbalanced anaphase separation can lead to a loss of chromosome(s) at one pole and a gain in chromosome(s) at the other pole, while chromosome lagging during anaphase could only mean chromosome loss. In most cases, euploid (1n) pollen grains are documented as the most common, but hypoeuploid and hypereuploid pollen grains exist (Ramsey and Schemske 2002). It is therefore reasonable to assume that the observed viable pollen grains that are in a comparable size range to the nonviable pollen grains in the present study could also be aneuploid and lose or gain chromosomes that are not composed of essential genes; hence, these grains have no adverse effect on the survival of aneuploid offspring. To evaluate the extent of aneuploid pollen production in these species, actual pollen cytotypes must be determined either by direct chromosome counting or by genome size estimation via flow cytometry.

Despite the existence of supposed aneuploid pollen, presumably as the result of meiotic abnormalities, the pollen viability of O. americanum, O. basilicum, and O. africanum remains considerably high (Table 2). These viable unbalanced but viable pollen grains are capable of fertilization, which might result in chromosome number variation, such as aneuploidy, in several Ocimum species over a large geographical range, as reviewed in the introduction. Ocimum americanum has 2n=24, 26, 48, 52, and 64 (Thai species has 2n=26); O. basilicum has 2n numbers anywhere from 48 to 74 (Thai species has 2n=52), whereas O. africanum has 2n=64 and 72 (Thai species has 2n=78). Studies such as those described here can help elucidate the mechanisms underlying chromosomal variation in this genus.

In Thailand, both the polyploid O. basilicum and O. africanum are cultivated for the culinary market, but the diploid species O. americanum is regarded by locals as a weedy plant; thus, extensive land use could lead to the disappearance of this wild species. Characterization studies such as the present study could help increase awareness of the importance of conserving wild genetic resources, which is important for the improvement of Ocimum cultivation in the long run.

Acknowledgments

This work was supported by the Science Achievement Scholarship of Thailand. The authors are very grateful to S. Suddee for sharing information on the distribution and characteristics of Ocimum species in Thailand. We would like to express our heartfelt gratitude to all the anonymous local people in Chiang Mai and Prachuab Khiri Khan provinces, Thailand, who provided us with plant materials from Ocimum species in the present study and provided us with traditional food during our field excursions.

Authors contributions

PL conducted the experiments, plant material collection, data collection, and data analysis and wrote the manuscript. PC supervised the development of the work and assisted with plant material collection, data collection, data analysis, and manuscript correction. KAJ revised the manuscript. All authors read and approved the final manuscript.

References

• Anamthawat-Jónsson, K., Thórsson, Æ. T., Temsch, E. M., and Greilhuber, J. 2010. Icelandic birch polyploids—The case of a perfect fit in genome size. J. Bot. 2010: 347254.
• Arogundade, B. and Adedeji, O. 2009. Pollen grain morphology of three species and a variety of Ocimum Linn. (Lamiaceae) in Southwestern Nigeria. J. Sci. Technol. 29: 1–7.
• Azzazy, M. 2016. Systematic importance of pollen morphology of some plants of Lamiaceae. Curr. Biot. 7: 5–10.
• Beaulieu, J. M., Leitch, I. J., Patel, S., Pendharkar, A., and Knight, C. A. 2008. Genome size is a strong predictor of cell size and stomatal density in angiosperms. New Phytol. 179: 975–986.
• Ben-Naim, Y., Falach, L., and Cohen, Y. 2015. Resistance against basil downy mildew in Ocimum Species. Phytopathology 105: 778–785.
• Ben-Naim, Y., Falach, L., and Cohen, Y. 2018. Transfer of downy mildew resistance from wild basil (Ocimum americanum) to sweet basil (O. basilicum). Phytopathology 108: 114–123.
• Bretagnolle, F. 2001. Pollen production and spontaneous polyploidization in diploid populations of Anthoxanthum alpinum. Biol. J. Linn. Soc. Lond. 72: 241–247.
• Carović-Stanko, K., Liber, Z., Besendorfer, V., Javornik, B., Bohanec, B., Kolak, I., and Satovic, Z. 2010. Genetic relations among basil taxa (Ocimum L.) based on molecular markers, nuclear DNA content, and chromosome number. Plant Syst. Evol. 285: 13–22.
• Carputo, D., Cardi, T., Frusciante, L., and Peloquin, S. J. 1995. Male fertility and cytology of triploid hybrids between tetraploid Solanum commersonii (2n=4x=48, 2EBN) and Phureja-Tuberosum haploid hybrids (2n=2x=24, 2EBN). Euphytica 83: 123–129.
• Chiranjeevi, M. and Chaya, P. 2023. Pollen characterization and medicinal importance of Ocimum species of Lamiaceae from Telangana region of South India. Int. J. Sci. Res. Arch. 8: 960–964.
• Chow, J., Puangpairote, T., Anamthawat-Jonsson, K., and Umpunjun, P. 2020. Karyotypic and molecular cytogenetic characterization of diploid and polyploid accessions of medicinal herbs in the genus Paris from northern Thailand. Sci. Asia 46: 1–11.
• Chowdhury, T., Mandal, A., Roy, S. C., and De Sarker, D. 2017. Diversity of the genus Ocimum (Lamiaceae) through morpho-molecular (RAPD) and chemical (GC-MS) analysis. J. Genet. Eng. Biotechnol. 15: 275–286.
• Clo, J. and Kolář, F. 2021. Short- and long-term consequences of genome doubling: A meta-analysis. Am. J. Bot. 108: 2315–2322.
• Dash, C., Afroz, M., Sultana, S., and Alam, S. 2017. Conventional and fluorescent karyotype analysis of Ocimum spp. Cytologia 82: 429–434.
• Dhasmana, A. 2013. Somatic chromosomal studies in Ocimum basilicum & Ocimum sanctum L. Int. J. Phytomed. 5: 330–340.
• Doaigey, A. R., El-Zaidy, M., Alfarhan, A., Milagy, A. E.-S., and Jacob, T. 2018. Pollen morphology of certain species of the family Lamiaceae in Saudi Arabia. Saudi J. Biol. Sci. 25: 354–360.
• Edet, O. U. and Aikpokpodion, P. O. 2014. Karyotype analysis of Ocimum basilicum in South-eastern Nigeria. Am. J. Plant Sci. 5: 126–131.
• Gonda, I., Faigenboim, A., Adler, C., Milavski, R., Karp, M.-J., Shachter, A., Ronen, G., Baruch, K., Chaimovitsh, D., and Dudai, N. 2020. The genome sequence of tetraploid sweet basil, Ocimum basilicum L., provides tools for advanced genome editing and molecular breeding. DNA Res. 27: 1–10.
• Heslop-Harrison, J. S. P., Schwarzacher, T., and Liu, Q. 2023. Polyploidy: Its consequences and enabling role in plant diversification and evolution. Ann. Bot. 131: 1–10.
• Idowu, J. and Oziegbe, M. 2017. Mitotic and meiotic studies on two species of Ocimum (Lamiaceae) and their F₁ hybrids. Bot. Lith. 23: 59–67.
• Karataş, İ. 2022. Production of rosmarinic acid and biomass from adventitious root cultures of Ocimum basilicum by optimization of medium components in airlift bioreactors. Plant Cell Tissue Organ Cult. 151: 235–251.
• Karlsdóttir, L., Thórsson, Æ. T., Hallsdóttir, M., Sigurgeirsson, A., Eysteinsson, T., and Anamthawat-Jónsson, K. 2007. Differentiating pollen of Betula species from Iceland. Grana 46: 78–84.
• Kaur, D. and Singhal, V. K. 2019. Meiotic abnormalities affect genetic constitution and pollen viability in dicots from Indian cold deserts. BMC Plant Biol. 19: 1–11.
• Khosla, M. K. 1995. Study of inter-relationship, phylogeny and evolutionary tendencies in genus Ocimum. Indian J. Genet. Plant Breed. 55: 71–83.
• Knight, C. A., Clancy, R. B., Götzenberger, L., Dann, L., and Beaulieu, J. M. 2010. On the relationship between pollen size and genome size. J. Bot. 2010: 171435.
• Kreiner, J. M., Kron, P., and Husband, B. C. 2017. Evolutionary dynamics of unreduced gametes. Trends Genet. 33: 583–593.
• Kumari, M., Prasad, A., Rahman, L. U., Mathur, A. K., and Mathur, A. 2022. In vitro germination, storage and microscopic studies of pollen grains of four Ocimum species. Ind. Crops Prod. 177: 1–12.
• Lawrence, B. M. 1992. Chemical components of Labiatae oils and their exploitation. In: Harley, R. M. and Reynolds, T. (eds.). Advances in Labiate Science. Royal Botanic Gardens Kew, London. pp. 399–436.
• Lekhapan, P., Anamthawat-Jónsson, K., and Chokchaichamnankit, P. 2019. Chromosome evolution based on variation in chromosome number and chiasma frequency in the genus Ocimum L. from Thailand. Cytologia 84: 199–206.
• Lekhapan, P., Anamthawat-Jónsson, K., and Chokchaichamnankit, P. 2021. Comparative karyotype analysis and chromosome evolution in the genus Ocimum L. from Thailand. Trop. Nat. Hist. 21: 27–40.
• Lysák, M. A. and Schubert, I. 2013. Mechanisms of chromosome rearrangements. In: Greilhuber, J., Dolezel, J., and Wendel, J. F. (eds.). Plant Genome Diversity: Physical Structure, Behaviour and Evolution of Plant Genomes. Springer Vienna, Vienna. pp. 137–147.
• Mason, A. S. and Pires, J. C. 2015. Unreduced gametes: Meiotic mishap or evolutionary mechanism? Trends Genet. 31: 5–10.
• Mehra, P. N. and Gill, L. S. 1972. Cytology of West Himalayan Labiate: Tribe Ocimoideae. Cytologia 37: 53–57.
• Morton, J. K. 1962. Cytotaxonomic studies on the West African Labiatae. J. Linn. Soc. Lond. Bot. 58: 231–283.
• Mukherjee, M., Datta, A. K., and Maiti, G. G. 2005. Chromosome number variation in Ocimum basilicum L. Cytologia 70: 455–458.
• Nibau, C., Gonzalo, A., Evans, A., Sweet-Jones, W., Phillips, D., and Lloyd, A. 2022. Meiosis in allopolyploid Arabidopsis suecica. Plant J. 111: 1110–1122.
• Otto, S. P. 2007. The evolutionary consequences of polyploidy. Cell 131: 452–462.
• Patel, R. P., Kumar, R. R., Singh, R., Singh, R. R., Rao, B. R. R., Singh, V. R., Gupta, P., Lahri, R., and Lal, R. K. 2015. Study of genetic variability pattern and their possibility of exploitation in Ocimum germplasm. Ind. Crops Prod. 66: 119–122.
• Paton, A., Harley, R. M., and Harley, M. M. 1999. Ocimum-an overview of relationships and classification. In: Hiltunen, R. and Holm, Y. (eds.). Basil: The Genus Ocimum. Harwood Academic Press, Amsterdam. pp. 1–38.
• Paton, A. and Putievsky, E. 1996. Taxonomic problems and cytotaxonomic relationships between and within varieties of Ocimum basilicum and related species (Labiatae). Kew Bull. 51: 509–524.
• Paton, A. J., Springate, D., Suddee, S., Otieno, D., Grayer, R. J., Harley, M. M., Willis, F., Simmonds, M. S. J., Powell, M. P., and Savolainen, V. 2004. Phylogeny and evolution of basils and allies (Ocimeae, Labiatae) based on three plastid DNA regions. Mol. Phylogenet. Evol. 31: 277–299.
• Perveen, A. and Qaiser, M. 2003. Pollen morphology of the family Labiatae from Pakistan. Pak. J. Bot. 35: 671–693.
• Pushpangadan, P. and Sobti, S. N. 1982. Cytogenetical studies in the genus Ocimum I. Origin of O. americanum, cytotaxonomical and experimental proof. Cytologia 47: 575–583.
• Pyne, R. M., Honig, J. A., Vaiciunas, J., Wyenandt, C. A., and Simon, J. E. 2018. Population structure, genetic diversity and downy mildew resistance among Ocimum species germplasm. BMC Plant Biol. 18: 69.
• Ramsey, J. 2007. Unreduced gametes and neopolyploids in natural populations of Achillea borealis (Asteraceae). Heredity 98: 143–150.
• Ramsey, J. and Schemske, D. W. 1998. Pathways, mechanisms, and rates of polyploid formation in flowering plants. Annu. Rev. Ecol. Syst. 29: 467–501.
• Ramsey, J. and Schemske, D. W. 2002. Neopolyploidy in flowering plants. Annu. Rev. Ecol. Syst. 33: 589–639.
• Rathod, V., Behera, T., Munshi, A., Durgesh, K., Jat, G., Krishnan, B., and Sharma, N. 2018. Pollen viability and in vitro pollen germination studies in Momordica species and their intra and interspecific hybrids. Int. J. Chem. Stud. 6: 32–40.
• Rewers, M. and Jedrzejczyk, I. 2016. Genetic characterization of Ocimum genus using flow cytometry and inter-simple sequence repeat markers. Ind. Crops Prod. 91: 142–151.
• Rice, A., Glick, L., Abadi, S., Einhorn, M., Kopelman, N. M., Salman-Minkov, A., Mayzel, J., Chay, O., and Mayrose, I. 2015. The chromosome counts database (CCDB) – A community resource of plant chromosome numbers. New Phytol. 206: 19–26.
• Shahrajabian, M. H., Sun, W., and Cheng, Q. 2020. Chemical components and pharmacological benefits of basil (Ocimum basilicum): A review. Int. J. Food Prop. 23: 1961–1970.
• Srivastava, A., Gupta, A. K., Sarkar, S., Lal, R. K., Yadav, A., Gupta, P., and Chanotiya, C. S. 2018. Genetic and chemotypic variability in basil (Ocimum basilicum L.) germplasm towards future exploitation. Ind. Crops Prod. 112: 815–820.
• Suddee, S., Paton, A. J., and Parnell, J. A. N. 2005. Taxonomic revision of tribe Ocimeae Dumort. (Lamiaceae) in continental South East Asia III. Ociminae. Kew Bull. 60: 3–75.
• Tel-Zur, N., Abbo, S., and Mizrahi, Y. 2004. Cytogenetics of semi-fertile triploid and aneuploid intergeneric vine cacti hybrids. J. Hered. 96: 124–131.
• Vij, S. P. and Kashyap, S. K. 1976. Cytological studies in some North Indian Labiatae. Cytologia 41: 713–719.
• Vineesh, S., Balaji, R., Tanuja., and Parani, M. 2023. The complete chloroplast genome of Ocimum americanum Linnaeus 1755 and phylogenetic analysis among the Lamiaceae family. Mitochondrial DNA B Resour. 8: 1077–1081.
• Wang, J., Kang, X., and Zhu, Q. 2010. Variation in pollen formation and its cytological mechanism in an allotriploid white poplar. Tree Genet. Genomes 6: 281–290.