Plant material (tubers) of E. byunsanensis, E. lobulata, E. pinnatifida, E. sibirica, E. stellata, and E. tanhoensis was collected during field investigations in Russia, China, Japan and South Korea during 2018–2020. The list of the samples examined is presented in Table 1. Herbarium specimens were deposited in the E and NS herbaria (herbarium acronyms according to Thiers 2019, continuously updated).

The comparative karyotype analysis was conducted for 22 populations: one of E. byunsanensis and E. lobulata, four of E. pinnatifida, three of E. sibirica, five of E. stellata, and eight of E. tanhoensis (Table 1). Somatic chromosomes of Eranthis were studied from root tip cells. Tubers were germinated in wet moss at ~ 15 °C for 2–4 weeks. Newly formed 1–2 cm long roots were excised and pretreated in 0.5% colchicine solution at 15 °C for 3–4 h. Roots were fixed in a mixture of 96% ethanol and glacial acetic acid (3:1). Root tips were stained with 1% aceto-haematoxylin, and the karyotypes were investigated by the squash method (Smirnov 1968). Chromosomes were counted in 30–100 mitotic cells for each population (a more detailed study was conducted for E. sibirica and E. tanhoensis). Mitotic metaphase chromosome plates were studied using an Axio Star microscope (Carl Zeiss, Munich, Germany) and photographed using an Axio Imager A.1 microscope (Carl Zeiss, Munich, Germany) with AxioVision 4.7 software (Carl Zeiss, Munich, Germany) and AxioCam MRc5 CCD-camera (Carl Zeiss, Munich, Germany) at 1000× magnification in the Laboratory for Ecology, Genetics and Environmental Protection (Ecogene), National Research Tomsk State University (Tomsk, Russia). KaryoType software (Altınordu et al. 2016) was used for karyotyping, and Adobe Photoshop CS5 (Adobe Systems, USA) and Inkscape 0.92 (USA) were used for image editing.

Karyotype formulas were derived, based on measurements of the photographed mitotic metaphase chromosomes. The measurements were performed on 4–12 metaphase plates per population. We used 2–6 metaphase plates per population with the most condensed chromosomes to calculate mean karyomorphological parameters. The degree of chromosome condensation was estimated from the total haploid length of the chromosome set. The symbols used to describe the karyotypes corresponded to those coined by Levan et al. (1964): m = median centromeric chromosome with arm ratio (r) of 1.0–1.7 (metacentric chromosome); sm = submedian centromeric chromosome with arm ratio of 1.7–3.0 (submetacentric chromosome); st = subterminal centromeric chromosome with arm ratio of 3.0–7.0 (subtelocentric chromosome); t = terminal centromeric chromosome with arm ratio of 7.0 and more (acrocentric chromosome); T = chromosome without obvious short arm (telocentric chromosome). Mean values of arm ratio (r), centromeric indices (CI), mean chromosome length (CL), and relative chromosome length (RL) for each chromosome pair and total haploid length (THL) were determined. In addition, we calculated the Coefficient of Variation of Chromosome Length (CV_CL; Paszko 2006), Coefficient of Variation of Centromeric Index (CV_CI; Paszko 2006), and Mean Centromeric Asymmetry (M_CA; Peruzzi and Eroğlu 2013).

To determine the karyological relationships among taxa, we carried out a multivariate PCoA (Principal Coordinate Analysis) using Gower's general coefficient of similarity, including six basic karyomorphological parameters (2n, x, THL, M_CA, CV_CL, and CV_CI) in the data matrix (Peruzzi and Altınordu 2014), by plotting every single metaphase. Then, we also subjected the same data matrix to LDA (Linear Discriminant Analysis) to test the diagnosability of the six species on karyomorphological grounds. Finally, we tested the Spearman correlation between THL and 1Cx for each species, using mean data. To perform PCoA, LDA and correlation tests, the software Past 4.06b (Hammer et al. 2001; Hammer 2021), freely available online, was used.

Flow cytometry with propidium iodide (PI) staining was used to determine the absolute DNA content. In this study, we have determined this parameter in representatives of four Eranthis species: E. byunsanensis, E. lobulata, E. pinnatifida and E. stellata from 10 different populations (Table 1). Silica-gel-dried leaf material (0.5–1.0 cm²) was chopped with a sharp razor blade in a 1 ml cold nuclei extraction buffer composed of 50 mM Hepes, 10 mM sodium metabisulphite, 10 mM MgCl₂, 0.5% polyvinylpyrrolidone, 0.1% bovine serum albumin, 0.3% Tween 20, 0.2% Triton X-100, 50 μg/ml RNase, 1 μg/ml β-mercaptoethanol and 50 μg/ml propidium iodide (PI). The samples were filtered through 50 μm nylon membranes into sample tubes and incubated in the dark at 4 °C for 15 min. The samples were measured using a Partec CyFlow PA flow cytometer equipped with a green laser at 532 nm wavelength. The absolute nuclear DNA content, the 2C-value according to Greilhuber et al. (2005), was calculated as the ratio of the mean fluorescence intensity of the sample nuclei to that of the external standard multiplied by the total nuclear DNA content of the standard. The possible effect of secondary metabolites on the binding of the intercalating dye was evaluated by measuring the fluorescence of Allium fistulosum L. leaf samples prepared as described above, but with the addition of the supernatant from Eranthis samples centrifuged without PI. The samples were measured three times at 10 min intervals. If no variation in the average values of the detection channels was observed for the A. fistulosum peak, the effect of secondary metabolites was considered negligible. The 1Cx-value (monoploid DNA content sensu Greilhuber et al. 2005) was calculated by dividing the 2C-value by the ploidy level of the species. The species, used as external standards, were Zamioculcas zamiifolia Engl., 2C = 48.35 pg and Vicia faba L. "Inovec" 2C = 26.90 pg (Doležel et al. 1992; Skaptsov et al. 2016). We used the Statistica 8.0 software (StatSoft, Inc.), Flowing Software 2.5.1 (Turku Centre for Biotechnology) and CyView software (Partec, GmbH) for data analyses. Flow cytometry was performed at the Laboratory for Bioengineering of the South-Siberian Botanical Garden, Altai State University (Barnaul, Russia).

Karyotypes

Karyomorphometric data, microphotographs of metaphase plates and idiograms for the studied species are presented in Tables 2, 3 and Figs 2, 3.

 Eranthis lobulata

Notes

The somatic and basic chromosome numbers in E. lobulata, endemic to China, are 2n = 16 and x = 8, respectively (Table 1; Fig. 2A). Five pairs of chromosomes (I–V) are metacentric, and three pairs (VI–VIII) are submetacentric, subtelocentric and acrocentric (Tables 2 and 3; Fig. 3). A pair of submetacentric chromosomes exhibits a secondary constriction. We also found a single small B chromosome in some cells. These Bs are metacentric, about 2.5 μm long. The karyotype formula of E. lobulata is 2n = 2x = 16 = 10m + 2sm^sat + 2st + 2t + 0–1B.

Table 2.

Download as

CSV

XLSX

Karyomorphological parameters in white-flowered Eranthis sect. Shibateranthis.

Species	Chromosome pair	CL (µm)	r	CI	RL (%)	Type
E. lobulata	I	8.46 ± 0.42	1.07 ± 0.04	0.48	7.80	m
	II	8.19 ± 0.31	1.16 ± 0.09	0.46	7.55	m
	III	7.43 ± 0.30	1.17 ± 0.07	0.46	6.85	m
	IV	7.38 ± 0.16	1.36 ± 0.10	0.42	6.80	m
	V	7.00 ± 0.29	1.28 ± 0.05	0.44	6.45	m
	VI	6.11 ± 0.15	2.05 ± 0.09	0.33	5.63	smsat
	VII	5.05 ± 0.21	5.04 ± 0.51	0.17	4.66	st
	VIII	4.62 ± 0.24	8.35 ± 0.84	0.11	4.26	t
E. stellata (pop. 2)	I	9.61 ± 0.34	1.07 ± 0.04	0.48	7.84	m
	II	9.29 ± 0.31	1.07 ± 0.04	0.48	7.58	m
	III	8.85 ± 0.39	1.06 ± 0.03	0.49	7.22	m
	IV	8.31 ± 0.42	1.06 ± 0.04	0.49	6.78	m
	V	7.89 ± 0.16	1.33 ± 0.07	0.43	6.44	m
	VI	6.21 ± 0.25	2.00 ± 0.19	0.33	5.06	sm
	VII	6.13 ± 0.40	2.14 ± 0.18	0.32	5.00	smsat
	VIII	5.01 ± 0.34	7.86 ± 0.38	0.11	4.08	t
E. tanhoensis (pop. 12)	I	8.68 ± 0.36	1.09 ± 0.05	0.48	8.74	m
	II	8.56 ± 0.41	1.23 ± 0.06	0.45	8.62	msat
	III	8.16 ± 0.29	1.07 ± 0.05	0.48	8.21	m
	IV	7.73 ± 0.35	1.07 ± 0.05	0.48	7.78	m
	V	6.63 ± 0.46	1.37 ± 0.11	0.42	6.67	m
	VI	5.72 ± 0.46	1.92 ± 0.14	0.34	5.76	sm
	VII	4.19 ± 0.38	2.34 ± 0.15	0.30	4.22	sm
E. sibirica (pop. 15)	I	9.51 ± 0.24	1.08 ± 0.04	0.48	2.88	m
	II	9.47 ± 0.29	1.03 ± 0.02	0.49	2.87	m
	III	9.20 ± 0.06	1.17 ± 0.03	0.46	2.78	m
	IV	9.13 ± 0.13	1.10 ± 0.06	0.48	2.76	m
	V	9.00 ± 0.11	1.05 ± 0.02	0.49	2.72	m
	VI	8.91 ± 0.14	1.39 ± 0.12	0.42	2.70	m
	VII	8.88 ± 0.07	1.20 ± 0.03	0.45	2.69	m
	VIII	8.87 ± 0.16	1.05 ± 0.03	0.49	2.68	m
	IX	8.67 ± 0.10	1.08 ± 0.05	0.48	2.62	m
	X	8.47 ± 0.09	1.27 ± 0.09	0.44	2.56	m
	XI	8.44 ± 0.15	1.07 ± 0.03	0.48	2.55	m
	XII	8.14 ± 0.13	1.16 ± 0.02	0.46	2.46	m
	XIII	7.71 ± 0.04	1.18 ± 0.09	0.46	2.33	m
	XIV	7.46 ± 0.15	1.35 ± 0.15	0.43	2.26	m
	XV	7.26 ± 0.21	1.70 ± 0.06	0.37	2.20	sm
	XVI	7.10 ± 0.04	1.28 ± 0.03	0.44	2.15	m
	XVII	6.89 ± 0.05	1.61 ± 0.05	0.38	2.08	m
	XVIII	6.45 ± 0.31	1.70 ± 0.08	0.37	1.95	sm
	XIX	5.36 ± 0.23	1.97 ± 0.09	0.34	1.62	sm
	XX	5.24 ± 0.25	1.74 ± 0.03	0.37	1.59	smsat
	XXI	5.08 ± 0.34	2.29 ± 0.14	0.30	1.54	sm
E. byunsanensis	I	8.59 ± 0.19	1.05 ± 0.03	0.49	8.55	m
	II	8.13 ± 0.31	1.06 ± 0.04	0.49	8.09	m
	III	7.65 ± 0.13	1.07 ± 0.04	0.48	7.61	m
	IV	6.18 ± 0.09	1.40 ± 0.05	0.42	6.15	m
	V	5.68 ± 0.21	1.19 ± 0.05	0.46	5.65	m
	VI	5.44 ± 0.19	5.22 ± 0.30	0.16	5.41	st
		5.19 ± 0.08	1.74 ± 0.05	0.37	5.17	sm
	VII	5.20 ± 0.13	5.64 ± 0.19	0.15	5.17	st
	VIII	3.52 ± 0.07	4.06 ± 0.37	0.20	3.50	st
E. pinnatifida (pop. 21)	I	9.24 ± 0.18	1.12 ± 0.02	0.47	8.58	m
	II	8.63 ± 0.24	1.08 ± 0.06	0.48	8.02	m
	III	8.25 ± 0.31	1.13 ± 0.03	0.47	7.66	m
	IV	6.62 ± 0.12	1.37 ± 0.07	0.42	6.15	m
	V	6.24 ± 0.26	2.77 ± 0.14	0.27	5.80	sm
	VI	5.88 ± 0.18	2.38 ± 0.07	0.30	5.46	sm
	VII	5.04 ± 0.11	1.95 ± 0.13	0.34	4.68	sm
	VIII	3.92 ± 0.09	3.10 ± 0.29	0.24	3.64	stsat

Notes: CL – chromosome length, mean value ± standard deviation; r – arm ratio, mean value ± standard deviation; CI – centromeric index; RL – relative chromosome length; m – metacentric chromosome; sm – submetacentric chromosome; st – subtelocentric chromosome; t – acrocentric chromosome; ^sat – chromosome showing secondary constriction.

Figure 2. 

Mitotic metaphase plates of white-flowered Eranthis sect. Shibateranthis A E. lobulata, 2n = 16 B E. stellata (pop. 2), 2n = 16 C E. stellata (pop. 6), 2n = 16 D E. tanhoensis (pop. 12), 2n = 14 E E. tanhoensis (pop. 10), 2n = 14+0–8B (arrows point at B chromosomes) F E. sibirica (pop. 15), 2n = 42 G E. byunsanensis, 2n = 16 (arrows point at the heteromorphic chromosome pair) H E. pinnatifida (pop. 21), 2n = 16 I E. pinnatifida (pop. 20), 2n = 16 (arrows point at heteromorphic chromosome pair). Scale bars: 10 μm. Microphotographs by E.Yu. Mitrenina.

Figure 3. 

Haploid idiograms of white-flowered Eranthis sect. Shibateranthis species. I–VIII – chromosome pairs; m – metacentric chromosome; sm – submetacentric chromosome; st – subtelocentric chromosome; t – acrocentric chromosome; T – telocentric chromosome; B – B chromosome.

The highest level of interchromosomal asymmetry, estimated via CV_CL, was found in E. byunsanensis and E. pinnatifida. The highest levels of intrachromosomal asymmetry (M_CA) and heterogeneity in centromere position (CV_CL) were found in E. lobulata and E. byunsanensis, while E. sibirica had the most symmetric karyotype (Table 3). We analyzed 27 accessions (metaphase plates) by PCoA (cumulative variance explained by the first two axes: 81.31%). No overlap among species was evident (Fig. 4). Indeed, LDA correctly attributed objects (accessions) to the six species in 100% of cases (jackknifed).

Figure 4. 

PCoA (Coordinate 1, 65.31% of variance explained vs. Coordinate 2, 16% of variance explained) based on six karyological parameters of white-flowered Eranthis sect. Shibateranthis species.

The absolute nuclear DNA content for 23 studied populations of six species of Eranthis is presented in Table 1. There was no significant correlation (r = 0.51, p = 0.29) between mean 1Cx values and total haploid (monoploid) chromosome length (THL) in these species. Indeed, for instance, while the 1Cx value is the smallest in the hexaploid E. sibirica, THL in this species is higher than in the diploid E. tanhoensis, sharing the same basic chromosome number x = 7 (Table 3).

According to our results and other data (Kurita 1955; Tak and Wafai 1996; Gömürgen 1998; Yuan and Yang 2006), chromosome sets of different species of Eranthis share some common features, albeit showing some species-specific peculiarities, which allow a clear-cut distinction among species based on karyo-morphological features according to LDA (see also Fig. 4). The traits of the karyotype within each species are sufficiently stable. However, in some cases, polymorphism was observed in the chromosome morphology, for instance, in E. pinnatifida. The karyotypes of E. byunsanensis and E. lobulata were described here for the first time.

The chromosomes of Eranthis belong to the Ranunculus-type (Langlet 1932). The karyotypes of E. lobulata and E. stellata are similar. Both species show a chromosome pair with a very small, not always visible, short arm. In the other four studied species of Eranthis, no chromosome of this type was found. The secondary constrictions in E. lobulata and E. stellata are localized in the short arms of submetacentric chromosome pairs. Different localization of secondary constrictions in these species (Fig. 3) is possibly due to a paracentric inversion. Previously, the karyotype of E. stellata from China (Jilin Province) was studied by Yuan and Yang (2006). These authors described its formula as 2n = 2x = 16 = 10m + 2sm + 2st + 2T. In contrast to our data, they assigned pair VII to subtelocentric rather than submetacentric chromosome type. Our data show that the arm ratio of this chromosome pair is 2.14 ± 0.18 (Table 2), congruent with a sm chromosome-type. They also did not find a short arm in pair VIII and referred it to T-type (telocentric chromosomes). We found short arms in this VIII chromosome pair, which led us to classify it as chromosomes of t-type (acrocentric chromosomes).

Two species, endemic to Siberia, E. sibirica and E. tanhoensis, show atypical dysploid basic chromosome number for Eranthis (x = 7) and exhibit hexaploid (2n = 42) and diploid (2n = 14) cytotypes, respectively (Erst et al. 2020b). Since there are different definitions of the term "basic chromosome number (x)" concerning polyploids (Peruzzi 2013), we clarify that, in the study, we mean, "x" as "chromosome number found in the gametes of their diploid relatives", according to Darlington (1958). A recent phylogenetic study (Xiang et al. 2021) found that E. tanhoensis and E. sibirica are closely related species that formed separate groups with basic chromosome number x = 7 within the North Asian clade of Eranthis. The same basic chromosome number x = 7 with 2n = 14 was previously reported in the genus Eranthis for E. stellata (Starodubtsev 1985), albeit this author does not provide any microphotograph of the metaphase plate. We re-analyzed plants from the same area (pop. 3; Russia, Primorsky Krai, Malaya Sedanka River), but we found a somatic chromosome number 2n = 16. At the same time, previous studies on E. sibirica reported 2n = 16 (Gnutikov et al. 2016, 2017) and 2n = 32 (Krogulevich 1976) chromosomes. However, the diploid plants described in these studies apparently refer to the recently described species E. tanhoensis. Some populations of this species show B chromosomes that researchers may have identified as regular chromosomes. In addition, a pair of metacentric chromosomes show large satellites, which, when using the squash method, are sometimes detached and can be misidentified as small telocentric chromosomes. Based on the large amount of material analyzed and on careful analysis of chromosome morphology, we conclude that the basic chromosome number of the studied populations of E. tanhoensis and E. sibirica is x = 7. However, we do not rule out the possible occurrence of different cytotypes in plants from Siberia.

The karyotypes of the two related species, endemic to Korea and Japan, also show peculiar features. Eranthis byunsanensis has a heteromorphic pair of chromosomes (VI). Unfortunately, we had material from a single population of this species. Therefore, we cannot conclude whether this feature is characteristic of the whole species or just a heterozygous chromosomal mutation. Eranthis pinnatifida has another feature that distinguishes it from other diploid species: four pairs of isobrachial chromosomes and four pairs of heterobrachial chromosomes. Our results concerning this species are consistent with the data published by Kurita (1955). Among the four E. pinnatifida populations studied, one population (pop. 20) shows a heteromorphic pair of chromosomes. In this case, we are sure that this mutation is just a polymorphic variant.

Carta et al. (2020) estimated x = 7 as the most likely ancestral basic chromosome number in Ranunculaceae. However, we hypothesize that, in Siberian species, E. sibirica and E. tanhoensis, the basic chromosome number evolutionarily reduced from x = 8 to x = 7 and not vice versa. This hypothesis is because most of the tribe Cimicifugeae members (i.e., Actaea, Anemonopsis, Beesia, Cimicifuga, Souliea and closely related Helleborus; Wang et al. 2009) have x = 8 (Rice et al. 2015). In addition, it has been established that Eranthis originated in East Asia and then dispersed to the west Qinghai-Tibetan Plateau and Mediterranean regions (Xiang et al. 2021). East Asian Eranthis species (i.e., E. byunsanensis, E. lobulata, E. pinnatifida, and E. stellata) have x = 8. According to a recent phylogenetic study (Xiang et al. 2021), E. sibirica and E. tanhoensis are a derived group within the North Asian clade of Eranthis with non-canonical basic chromosome number x = 7 for the tribe Cimicifugeae.

The karyotypes of the two related species E. stellata and E. tanhoensis, with 2n = 16 and 2n = 14 chromosomes, respectively, are similar concerning five metacentric (I–V) and two submetacentric (VI–VII) chromosome pairs and differ by the presence of acrocentric pair (VIII) in E. stellata. It is well known that the basic chromosome number can change (dysploidy) due to chromosome rearrangements, fusion or fission of some chromosomes of the set and chromosome loss (Shubert 2007; Guerra 2008; Escudero et al. 2014). Dysploidy can establish powerful crossing barriers between sympatric taxa, as it disturbs regular chromosome pairing and bivalent formation at meiosis, drastically reducing hybrid fertility. These processes can result in the formation of new species (Grant 1981; Levin 2002; Baltisberger and Hörandl 2016). Such restructuring is known, for example, in the evolution of Arabidopsis thaliana (2n = 10) from A. lyrata (2n = 16) (Koch and Kiefer 2005). A similar case of descendant dysploidy was revealed for other Brassicaceae (Lysak et al. 2006) and plants from other families (Levin 2002). For Ranunculaceae, a decrease in the basic chromosome number from x = 8 to x = 7, caused by chromosome rearrangements, is known within Ranunculus (Baltisberger and Hörandl 2016) and Anemone (Mlinarec et al. 2012).

The shift to x = 7 in Eranthis possibly led to reproductive isolation of the populations with a new cytotype and, ultimately, speciation. We assume that further isolation of E. tanhoensis and E. sibirica was associated with polyploidization of the latter species. However, the type of polyploidy (i.e., autopolyploidy or allopolyploidy) has to be determined for this species. The karyotype of E. sibirica is similar to that of E. tanhoensis in chromosome morphology (metacentric and submetacentric chromosomes only), and they differ from the karyotypes of other related species. The organization of E. sibirica karyotype with 2n = 42 seems functionally diploid. The chromosomes are grouped in pairs (Fig. 3) and not in groups of 6. It is known that the size and shape of homologous chromosomes may change in the course of the diploidization process following polyploidization, i.e., due to the genome downsizing. Repetitive DNA sequences, both non-coding and coding, gene duplicates may be eliminated from the genome, resulting in changes in the karyotype parameters (Leitch and Bennett 2004; Mandáková and Lysak 2018; Wang et al. 2021).

A distinguishing feature of E. tanhoensis is the presence of small Bs in some of its populations (pops 11 and 14). Sporadic Bs were previously detected in individual cells only in E. lobulata (Erst et al. 2019). In some representatives of E. tanhoensis, up to 8 Bs could be observed in many cells. Bs are often found in representatives of Ranunculaceae and other families (Rice et al. 2015). However, their origin and possible adaptive and/or evolutionary roles are still poorly understood (Datta et al. 2016; Dhar et al. 2019). It is generally accepted that Bs are formed from A chromosomes in different ways. The most convincing case was the fully documented origin of a nascent B in trisomic Plantago lagopus L. from a supernumerary. This origin was associated with chromosome fragmentation, specific DNA sequence amplification, the addition of telomeric repeats and centromeric misdivision (Dhar et al. 2002). Bs could also escape as small centric fragments following unequal translocation and a reduction in chromosome number (Jones et al. 2008). Bs in E. tanhoensis may be preserved fragments of the lost ancestral pair VIII. The presence of Bs in the genome increases the adaptive capabilities of the population to adverse environmental conditions (Datta et al. 2016), which can be quite relevant for plants growing in this climatic zone.

The Kew list of DNA C-values contains only one C-value for white-flowered Eranthis (i.e., 1C = 8.20 pg) determined by Zonneveld et al. (2005) for E. pinnatifida. In the present study, we determined the genome size for six white-flowered Eranthis species. According to our data, the Japanese E. pinnatifida has an average 1Cx = 9.80 ± 0.33 pg. It is the lowest absolute nuclear DNA content among the studied diploids. A lower 1Cx value (9.26 ± 0.25 pg) was found only in the polyploid E. sibirica. Remarkably, a closely related diploid species, such as E. tanhoensis, shows 1Cx = 12.48 ± 0.25 pg. According to the genome downsizing theory, an increase in the ploidy level leads to a decrease in the size of the monoploid genome. The loss of DNA in polyploids is a widespread phenomenon occurring in many plant groups (Shaked et al. 2001; Leitch and Bennett 2004; Adams and Wendel 2005). In the present study, Eranthis stellata exhibited the highest 1Cx-value of 15.93 ± 0.32 pg and the highest total haploid length as well. However, it is interesting to note that, in this system, we found no significant correlation between 1Cx and THL, as otherwise commonly found in plants (Levin 2002; Peruzzi et al. 2009), where this correlation typically exceeds r = 0.8. This inconsistency could be explained by different condensation degrees of the studied chromosomes. Nonetheless, it also may suggest differences in chromosomes width and volume (Kramer et al. 2021), not addressed in this study.