Caryologia. International Journal of Cytology, Cytosystematics and Cytogenetics 73(3): 33-44, 2020 Firenze University Press www.fupress.com/caryologia ISSN 0008-7114 (print) | ISSN 2165-5391 (online) | DOI: 10.13128/caryologia-625 Caryologia International Journal of Cytology, Cytosystematics and Cytogenetics Citation: S. Bhadra, Z.-Q. Cai (2020) Kar- yological variability and chromosomal asymmetry in highland cultivars of Chenopodium quinoa Willd. (Amaran- thaceae). Caryologia 73(3): 33-44. doi: 10.13128/caryologia-625 Received: September 17, 2019 Accepted: April 27, 2020 Published: December 31, 2020 Copyright: © 2020 S. Bhadra, Z.-Q. Cai. This is an open access, peer-reviewed article published by Firenze University Press (http://www.fupress.com/caryo- logia) and distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All rel- evant data are within the paper and its Supporting Information files. Competing Interests: The Author(s) declare(s) no conflict of interest. Karyological variability and chromosomal asymmetry in highland cultivars of Chenopodium quinoa Willd. (Amaranthaceae) Sreetama Bhadra1,*, Zhi–Quan Cai1,2,* 1 CAS Key Laboratory of Tropical Forest Ecology, Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences, Mengla 666303, China 2 Department of Horticulture, Foshan University, Foshan 528000, China * Corresponding authors. E-mail: sreetama.bhadra@gmail.com; zhiquan.cai@126.com Abstract. Chenopodium quinoa Willd. is rapidly gaining importance worldwide as a superfood. However, structural diversity and asymmetry analyses of chromosomes of different cultivars of the species are largely understudied primarily owing to their small chromosomes. In this paper, karyomorphological investigations were performed on 21 cultivars of C. quinoa with varying seed morphology cultivated widely in the high- land regions of the Andes, which is the center of domestication of this species. Somatic chromosome number was found to be 2n= 36 in all cultivars with no occurrence of mixoploidy. Lengths of individual chromosomes varied between 0.63–6.53 μm, with their short arms ranging from 0.25–2.95 μm and long arms between 0.38–3.58 μm. Types of primary constriction ranged from median to sub–terminal. One pair of chro- mosome in each complement possessed a secondary constriction. Chromosome com- plements of all cultivars belonged to the asymmetry class 2B with an average asym- metry index value of 3.21±0.61. Values of intra- and inter-chromosomal asymmetry indices were 0.30±0.02 and 0.20±0.02 respectively across all cultivars. The average coef- ficients of variation of chromosome lengths was 19.77±2.11 and average centromeric index was 16.28±2.12. Arm ratio of the chromosomes varied from 0.34 to 5.76. The mean values of karyotypic asymmetry, symmetry index and karyotype asymmetry index percentage were 17.69±1.53, 147.69±5.54 and 58.71±0.86 respectively. Pearson correlation revealed strong correlation within inter– and intra–chromosomal asymme- try indices. Our analyses uncovered higher chromosomal variation in quinoa than pre- viously found with high inter–varietal similarities among the studied cultivars, revealed from scattered diagrams between asymmetry indices. Keywords: Chenopodium quinoa, chromosome, karyotype, chromosomal asymmetry, inter-varietal symmetry. INTRODUCTION Sustainable agricultural systems generally focus on traditional com- mercial food crops, which might not be adequate in the near future to fulfill the projected increase in global consumption due to increase in population. This, along with rapid climate change, has necessitated the investigation and 34 Sreetama Bhadra, Zhi–Quan Cai exploitation of alternative crop plants to complement the projected deficit (Graf et al. 2015). Recent research has focused on the South American crop Chenopodium qui- noa (quinoa), which the Food and Agriculture Organiza- tion of the United Nations have touted as a crop that can mitigate global food demand to a large extent (Graf et al. 2015). This plant’s adaptability to extreme agro–ecologi- cal conditions together with its balanced proportion of amino acids, carbohydrates, lipids, vitamins and miner- als, low glycemic acid content and its gluten–free nature, have resulted in expansion of its cultivation outside its native region (Ruiz et al. 2014; Pereira et al. 2019). Improvement programs of the crop include reduction in saponin content of its seed coat, increasing compactness of its inflorescence, high stress tolerance and resistance to pests and pathogens (Murphy et al. 2016; Jarvis et al. 2017). Success of these programs demand identification of existing genetic divergence in the species. An initial idea of genetic variability among plants can be perceived by analyzing cytological characters (Guerra 2008). Most crop plants have undergone recent polyploidization events (Renny-Byfield and Wendel 2014). Polyploidy results in variable chromosome numbers and the ensuing chromosomal changes like translocations and inversions give rise to structural variabilities (Leitch and Leitch 2008). Analyses and comparison of these 2 parameters, viz. chromosome number and structure, among related taxa is considered to be a reliable approach for quantitative interpretation of their similarity or diver- gence among related plants (Bennett and Leitch 2005). Generally considered an allotetraploid with a basic chro- mosome number x= 9 (Krak et al. 2016; Mandák et al. 2016), the reported somatic chromosome counts of qui- noa predominantly showed an invariable 2n= 36 (Kjell- mark 1934; Wulff 1936; Cárdenas and Hawkes 1948; Heiser 1963; Gandarillas and Luizaga 1967; Giusti 1970; Kolano et al. 2001, 2012; Rahiminejad and Gornall 2004; Bhargava et al. 2006; Palomino et al. 2008; Mandák et al. 2016). However, there are limited studies on variations in chromosome structure and asymmetry in the species mainly due to their small size, which makes common karyotyping of the species difficult (Kolano et al. 2001, 2012). Such studies, when present, were mostly limited to examining only 1 or 2 varieties restricting the possibility of understanding inter-varietal chromosomal differences, or, were based on assessments of insufficient asymme- try parameters (Kolano et al. 2001, 2012; Bhargava et al. 2006; Maughan et al. 2006; Palomino et al. 2008; Yang- quanwei et al. 2013). Earliest reports of domestication and cultivation of quinoa can be traced back to about 7,000 years ago to the Inca dynasty in South American Andes, especially near Lake Titicaca from Peru and Bolivia that constituted the ancestral highland ecotype of this species (Jarvis et al. 2017). Although variations in prominent morphological markers for commercial quinoa like seed size and colour, exist within the cultivated varieties of South America, it can be argued that, prolonged farming of a species at and around its center of domestication and diversity might expose them to the risk of genetic erosion, result- ing in low genetic variability (Gonzalo et al. 2019). Exist- ence of such genetic erosion, undesirable for germplasm improvement and conservation programmes, might be realized from cytological studies; however, chromosomal study of quinoa from this region is not forthcoming. The objectives of our work were (1) to analyze the karyotype structure of highland varieties of quinoa, (2) to determine the extent of karyotype differentia- tion among the varieties, (3) to evaluate chromosomal asymmetry in the varieties. We investigated 21 morpho- logically variable highland cultivars of quinoa, and, per- formed karyotypic analyses based on asymmetry param- eters that could help in elucidating worthwhile informa- tion on the inter–varietal divergence of quinoa. MATERIALS AND METHODS Materials For the purpose of this study, seeds of white, yellow, red and black–coloured varieties were selected from the highland ecotypes of quinoa, their sizes ranging from 1.5 to 3.0 cm in diameter. These cultivars were collected from Peru and Bolivia around Lake Titicaca (Figure 1). Seeds from at least 5 individuals of each cultivar were germinated on moist filter paper in Petri plates, kept in dark for 24–48 h at room temperature for cytological study. Growing root tips of freshly germinated seedlings were used for cytological analyses. Methods Actively growing 1.5–2 cm long root tips of seed- lings were harvested after 1 to 2 days of germination initiation, between 10 AM to 11 AM, for mitotic study. Pretreatment of the harvested root tips were performed in aqueous solution of 2 mM 8–hydroxyquinoline for 3 hours at 10°C after which the root tips were fixed over- night in Carnoy’s fixative at 10°C. Hydrolysis, staining of the root tips and slide preparation were done following the protocol of Sun et al. (2017). Squashed root tips were observed under Olympus BX51 microscope (Olympus, Germany) at a magnification of 1000X. Well scattered 35Karyological variability and chromosomal asymmetry in highland cultivars of Chenopodium quinoa metaphase plates were photographed with a mount- ed digital camera using the software ScopePhoto (ver. 3.0.12.444) (ScopeTek, Hangzhou, China). Chromosome numbers were counted from at least 50 metaphase plates prepared from 50 different seedlings for each cultivar of quinoa to ensure the uniformity of somatic chromosome numbers in all the plants. Among them, at least 30 best metaphase plates were selected for karyotype analyses with a minimum of 10 seedlings chosen per cultivar. Lengths of chromosome arms were measured from at least 5 best metaphase plates for each cultivar using Kar- yoType v2 software (Altınordu et al. 2016). The lengths of chromosomes and respective centromeric indices of each metaphase plate, obtained from this data, were used to describe the chromosome morphology following the nomenclatural system of Levan et al. (1964), and, to prepare their idiograms with OriginPro 2018 software (OriginLab Corporation, USA). Asymmetry of karyotypes of each cultivar were calculated using Stebbin’s classifica- tion (1971) and AI (karyotype asymmetry index), intra- chromosomal asymmetry indices using A1 (intra-chro- mosomal asymmetry index), CVCI (coefficient of variation of centromeric index), MCA (mean centromeric asymme- try), Syi (symmetric index), AsK% (Karyotype asymmetry index percentage) and AR (arm ratio), and inter-chromo- somal asymmetry using A2 (inter-chromosomal asymme- try index), CVCL (coefficient of variation of chromosome length) (for details see Bhadra and Bandyopadhyay 2015). Pearson correlation among these intra- and inter-chromo- somal asymmetry parameters was calculated in SPSS (ver- sion 21) software (München, Germany). RESULTS The chromosomes of studied cultivars of Chenopo- dium quinoa were small, though clearly visible with con- spicuous centromeric regions. Figure 1. (A) Map showing the collection area of the studied cultivars (B) Photographs of seeds showing morphological variation (bar=1 cm). 36 Sreetama Bhadra, Zhi–Quan Cai Table 1. Chromosome numbers, karyotype formulae, morphometric parameters and values of asymmetry indices studies in 21 Chenopo- dium quinoa varieties. Local variety name 2n Karyotype formula Short arm range (average) in µm Long arm range (average) in µm Total length range (average) in µm BL71 36 2M+22m+8Sm+2St+2m:Sm 0.42–1.77 (1.03±0.25) 0.80–2.64 (1.49±0.34) 1.35–4.24 (2.56±0.50) BL72 36 8M+14m+10m+2St+2m:Sm 0.42–1.92 (1.04±0.31) 0.50–3.02 (1.52±0.39) 0.92–4.19 (2.59±0.59) BL–D75 36 18m+14Sm+2St+2M:Sm 0.39–2.15 (1.04±0.26) 0.46–2.83 (1.55±0.40) 0.85–4.51 (2.62±0.57) BL–X3 36 6M+8m+16Sm+4St+2m:Sm 0.59–2.19 (1.21±0.28) 1.04–3.45 (1.91±0.36) 1.88–4.61 (3.16±0.42) RD14 36 6M+14m+12Sm+2St+2m:Sm 0.52–1.79 (1.10±0.25) 0.65–2.88 (1.69±0.44) 1.17–3.96 (2.82±0.50) RD31 36 4M+18m+10Sm+2St+2M:Sm 0.44–2.95 (1.13±0.33) 0.55–3.58 (1.71±0.42) 0.99–6.53 (2.88±0.66) RD36 36 4M+26m+4Sm+2m:Sm 0.40–1.34 (0.95±0.18) 0.57–2.12 (1.30±0.23) 0.97–3.95 (2.28±0.38) RD47 36 4M+20m+8Sm+2St+2m:Sm 0.40–1.83 (1.16±0.26) 0.46–2.82 (1.59±0.30) 0.86–4.41 (2.78±0.49) RD56 36 8M+20m+6Sm+2M:Sm 0.37–1.48 (0.99±0.24) 0.49–2.09 (1.41±0.29) 0.86–3.85 (2.44±0.48) RD58 36 8M+16m+8Sm+2St+2M:Sm 0.37–2.00 (1.13±0.37) 0.52–3.00 (1.60±0.44) 0.89–4.82 (2.76± 0.70) RD8 36 2M+16m+12Sm+4St+2m:Sm 0.33–1.93 (1.10±0.30) 0.60–3.66 (1.67±0.44) 0.93–5.45 (2.81±0.64) RD–D88 36 6M+16m+10Sm+2St+2m:Sm 0.42–1.85 (1.10±0.32) 0.50–2.96 (1.61± 0.35) 0.92–4.44 (2.75±0.58) RD–X2 36 2M+12m+18Sm+2St+2m:Sm 0.25–1.64 (1.00±0.24) 0.38–2.38 (1.52±0.30) 0.63–3.94 (2.55±0.47) WH32 36 6M+20m+6Sm+2St+2m:Sm 0.68–2.57 (1.18±0.32) 0.54–2.87 (1.77±0.40) 1.03–5.44 (2.99±0.58) WH82 36 8M+22m+4Sm+2m:Sm 0.48–1.81 (1.10±0.26) 0.58–2.52 (1.49±0.32) 1.06–4.18 (2.62±0.54) WH–X1 36 2M+18m+12Sm+2St+2m:Sm 0.35–1.42 (0.85±0.20) 0.38–2.42 (1.26±0.26) 0.73–3.20 (2.14±0.40) YEL59 36 8M+18m+6Sm+2St+2M:Sm 0.49–1.97 (1.16±0.28) 0.73–2.83 (1.63±0.36) 1.22–4.34 (2.82±0.54) YEL61 36 4M+16m+12Sm+2St+2m:Sm 0.42–1.66 (0.97±0.26) 0.47–2.91 (1.54±0.35) 0.92–3.98 (2.55±0.50) YEL63 36 6M+22m+4Sm+2St+2m:Sm 0.47–1.75 (1.06±0.29) 0.59–2.81 (1.59±0.42) 1.06–4.36 (2.69±0.59) YEL7 36 2M+22m+10Sm+2m:Sm 0.44–1.82 (1.16±0.26) 0.53–2.79 (1.70±0.36) 0.97–5.02 (2.92±0.54) YEL–X4 36 4M+24m+6Sm+2m:Sm 0.53–1.68 (1.07±0.24) 0.77–2.61 (1.63±0.31) 1.35–4.50 (2.74±0.42) Local variety name Centromeric index range (average) (i) Karyotype asymmetry class Arm ratio range (average) (AR) Value of relative chromatin (VRC) Total form percentage (TF%) Relative length percentage range (RL) BL71 18.79–49.76 (41.17±6.83) 2B 0.70–4.32 (1.51±0.06) 5.11±0.61 41.78±0.96 1.56–4.56 BL72 16.35–49.76 (40.52±7.72) 2B 0.69–5.12 (1.57±0.11) 5.19±0.64 41.35±1.06 0.86–4.06 BL–D75 19.43–49.70 (40.52±6.37) 2B 0.57–4.14 (1.54±0.13) 5.24±0.99 40.80±2.13 0.93–4.46 BL–X3 21.69–49.82 (39.40±7.63) 2B 0.41–3.61 (1.65±0.22) 6.32±0.88 39.76±3.18 1.91–3.64 RD14 14.79–50 (40.01±8.03) 2B 0.54–5.76 (1.62±0.11) 5.65±0.13 40.21±0.36 1.63–3.95 RD31 22.41–49.78 (40.13±6.44) 2B 0.61–3.46 (1.59±0.06) 5.76±1.52 40.33±0.93 1.20–6.74 RD36 31.22–49.76 (42.48±4.17) 2B 0.60–2.20 (1.38±0.02) 4.56±0.47 42.98±0.27 1.26–4.30 RD47 24.31–49.84 (42.36±5.77) 2B 0.77–3.11 (1.43±0.13) 5.56±0.56 42.77±1.88 0.96–4.02 RD56 25.73–49.84 (41.84±5.65) 2B 0.60–2.88 (1.46±0.09) 4.87±0.22 42.26±1.37 0.95–4.29 RD58 15.56–49.81 (41.44±7.46) 2B 0.68–5.42 (1.52±0.10) 5.53±0.89 42.08±0.96 0.89–4.57 RD8 18.16–49.32 (40.02±6.75) 2B 0.55–4.51 (1.59±0.21) 5.63±0.46 40.52±2.49 0.97–4.92 RD–D88 21.32–49.82 (40.35±7.40) 2B 0.39–3.69 (1.57±0.12) 5.49±0.81 41.34±1.44 1.10–5.26 RD–X2 20.92–50 (39.99±6.66) 2B 0.54–3.78 (1.59±0.28) 5.10±0.06 40.48±4.02 0.68–4.24 WH32 19.82–49.64 (40.52±7.08) 2B 0.53–4.04 (1.58±0.09) 5.98±0.45 40.63±1.43 0.88–4.65 WH82 25.00–49.72 (42.38±5.24) 2B 0.70–3.00 (1.38±0.04) 5.25±0.72 43.23±0.71 1.06–4.07 WH–X1 24.37–50 (40.97±5.53) 2B 0.44–3.10 (1.51± 0.20) 4.29±0.37 41.30±2.96 1.01–3.88 YEL59 22.09–49.77 (41.92±6.15) 2B 0.76–3.53 (1.46±0.10) 5.64±0.88 42.38±1.52 1.29–3.99 YEL61 18.03–49.69 (39.05±7.23) 2B 0.34–4.55 (1.66±0.16) 5.09±0.59 39.69±1.53 1.04–4.49 YEL63 17.07–49.83 (40.57±7.38) 2B 0.41–4.86 (1.56±0.10) 5.39±0.53 40.99±2.15 1.04–4.26 YEL7 25.69–49.60 (40.97±6.96) 2B 0.67–3.28 (1.51±0.05) 5.84±0.46 41.69±0.48 0.85–4.41 YEL–X4 18.03–50 (39.87±6.91) 2B 0.35–4.55 (1.62±0.41) 5.47±0.28 40.49±4.74 1.35–4.51 37Karyological variability and chromosomal asymmetry in highland cultivars of Chenopodium quinoa Local variety name Difference of relative length (DRL) Intra–chromosomal asymmetry index (A1) Inter–chromosomal asymmetry index (A2) Coefficient of variation of chromosome length (CVCL) Coefficient of variation of centromeric index (CVCI) BL71 2.52±0.54 0.29±0.02 0.19±0.03 19.46±2.92 16.61±1.14 BL72 2.72±0.30 0.29±0.03 0.22±0.03 22.48±2.72 18.61±1.68 BL–D75 2.92±0.15 0.31±0.06 0.22±0.04 22.35±3.63 15.79±1.83 BL–X3 1.49±0.30 0.33±0.07 0.14±0.03 13.71±3.18 19.58±4.55 RD14 2.18±0.06 0.31±0.01 0.18±0.03 17.65±2.80 20.08±3.43 RD31 4.65±0.95 0.33±0.02 0.23±0.05 23.28±4.90 16.04±2.22 RD36 2.54±0.30 0.26±0.01 0.17±0.01 16.75±1.13 9.80±0.38 RD47 2.48±0.18 0.26±0.05 0.18±0.01 17.78±0.12 13.66±2.14 RD56 2.62±0.38 0.28±0.03 0.20±0.01 19.76±0.73 13.56±3.10 RD58 3.07±0.14 0.27±0.01 0.25±0.02 25.17±2.28 18.04±4.42 RD8 3.06±0.34 0.32±0.06 0.23±0.02 22.78±2.36 17.10±4.25 RD–D88 3.11±1.01 0.30±0.04 0.21±0.05 21.48±5.02 18.46±3.91 RD–X2 2.90±0.59 0.32±0.10 0.18±0.03 18.42±2.72 16.81±2.54 WH32 2.73±0.74 0.31±0.04 0.20±0.03 19.47±2.67 17.54±2.46 WH82 2.56±0.27 0.25±0.02 0.20±0.03 20.35±3.37 12.35±0.94 WH–X1 2.29±0.46 0.30±0.08 0.19±0.02 18.78±1.74 13.68±3.97 YEL59 2.34±0.27 0.27±0.03 0.19±0.02 19.44±2.22 14.69±3.76 YEL61 2.75±0.95 0.34±0.03 0.20±0.05 20.14±4.61 18.63±5.77 YEL63 2.53±0.70 0.30±0.05 0.22±0.04 22.05±4.30 18.24±2.49 YEL7 2.89±0.60 0.29±0.02 0.18±0.02 18.32±2.18 14.65±0.44 YEL–X4 2.26±0.80 0.32±0.12 0.15±0.03 15.52±3.59 17.93±7.93 Local variety name Karyotype asymmetry index (AI) Mean centromeric asymmetry (MCA) Symmetric index (Syi) Karyotype asymmetry index percentage (AsK%) BL71 3.21±0.24 16.80±1.39 144.12±5.02 58.22±0.96 BL72 4.20±0.83 17.47±2.05 146.47±6.20 58.65±1.06 BL–D75 3.51±0.51 18.30±3.74 150.64±14.17 59.20±2.13 BL–X3 2.59±0.23 20.43±5.43 157.21±20.92 60.24±3.18 RD14 3.48±0.18 19.08±0.69 154.02±2.09 59.79±0.36 RD31 3.66±0.27 19.46±1.48 153.07±5.90 59.67±0.93 RD36 1.64±0.05 14.55±0.73 136.77±2.15 57.02±0.27 RD47 2.43±0.39 15.04±3.37 138.12±10.76 57.23±1.88 RD56 2.69±0.70 16.18±2.37 142.12±7.60 57.73±1.37 RD58 4.61±1.47 16.34±1.52 141.87±5.68 57.92±0.96 RD8 3.91±1.15 19.02±4.76 152.55±17.95 59.48±2.49 RD–D88 3.97±1.37 18.32±3.14 147.19±9.70 58.66±1.44 RD–X2 3.14±0.94 19.26±7.26 153.63±27.35 59.52±4.02 WH32 3.41±0.60 18.70±2.72 150.94±9.46 59.37±1.43 WH82 2.53±0.57 14.24±1.45 135.69±3.16 56.77±0.71 WH–X1 2.53±0.54 17.70±5.75 147.95±18.65 58.70±2.96 YEL59 2.81±0.45 15.52±2.37 140.52±9.14 57.62±1.52 YEL61 3.57±0.18 20.85±2.73 158.16±9.74 60.31±1.53 YEL63 4.09±1.36 17.98±3.48 149.65±12.98 59.01±2.15 YEL7 2.68±0.27 17.03±1.24 145.88±4.32 58.31±0.48 YEL–X4 2.66±0.84 19.28±8.95 154.89±32.99 59.51±4.74 38 Sreetama Bhadra, Zhi–Quan Cai Somatic cells of all the cultivars exhibited a uni- form chromosome number of 2n= 36 (Table 1, Figure 2). Lengths of chromosomes in all the cultivars varied from 0.63 μm to 6.53 μm, with the highest average chromo- some length being present in the cultivar BL–X3 and the lowest in WH–X1 (Figure 3 A–F). Short arms were 0.25 μm to 2.95 μm in length, while long arms showed a vari- ation of 0.38 μm to 3.66 μm. Centromeric indices ranged from 50% to 14.79% resulting in primary constrictions varying from median to sub-terminal types, though one cultivar did not possess any chromosome with median constriction and chromosomes of five cultivars did not exhibit sub-terminal constrictions. Number of chromo- somes with each type of primary constriction also var- Figure 2. Somatic metaphase chromosomes of Chenopodium quinoa (2n=36): (A) BL71 (B) BL72 (C) BL–D75 (D) BL–X3 (E) RD14 (F) RD31 (G) RD36 (H) RD47 (I) RD56 (J) RD58 (K) RD8 (L) RD–D88 (M) RD–X2 (N) WH32 (O) WH82 (P) WH–X1 (Q) YEL59 (R) YEL61 (S) YEL63 (T) YEL71 (U) YEL–X4 (bar= 5 μm). 39Karyological variability and chromosomal asymmetry in highland cultivars of Chenopodium quinoa Figure 3. Idiograms of some cultivars of Chenopodium quinoa: (A) RD14 (B) RD31 (C) RD56 (D) BL71 (E) BL72 (F) BL–X3 (G) Com- parative graphical representation of distribution of constriction types in the cultivars of Chenopodium quinoa (M= median constriction, m= nearly median constriction, Sm= sub–median constriction, St= sub–terminal constriction, sec= secondary constriction). 40 Sreetama Bhadra, Zhi–Quan Cai ied among the cultivars. All of the cultivars revealed 2 chromosomes possessing secondary constrictions, with one constriction in median or nearly median region and the other in sub–median region (Figure 3 G). The inter- and intra-chromosomal asymmetry indi- ces were calculated on the basis of chromosome lengths and centromeric indices. Stebbins classification (1971) placed the karyotype of quinoa in 2B category while AI ranged from 1.64 in RD36 to 4.61 in RD58. The comple- mentary indices A1 and A2 ranged from 0.25 to 0.34, and, 0.15 to 0.25 respectively. Values of CVCL and CVCI varied from 15.52–25.17, and, 9.80–20.08 respectively, while that of MCA was 14.24–20.85. Values of average AR ranged from 1.38 to 1.66, with the lowest recorded in YEL61 and highest in RD14. The values of Syi and Ask% ranged from 135.69 to 158.16, and, 56.77% to 60.31% respectively. DISCUSSION Domestication of plants has been impacted by poly- ploidy since the beginning of agriculture, with approxi- mately 40–70% of cultivated plants exhibiting poly- ploidy (Hilu 1993, Sattler et al. 2016). Polyploids, espe- cially allopolyploids show hybrid vigour with increased growth rates and higher productivity that favours their artificial selection over their diploid progenitors (Renny– Byfield and Wendel 2014). Considering Goldblatt’s (1980) assumption of polyploid determination, quinoa with the basic chromosome number of x= 9, is a tetraploid with 2n= 36 somatic chromosomes, a condition similar to the closely related species C. hircinum and C. berlandieri (Fuentes–Bazan et al. 2012; Mandák et al. 2016; Jarvis et al. 2017). Studies have endorsed its allotetraploid nature, hypothesizing possible hybridization between a North American and a Eurasian diploid species, whose iden- tities are yet unknown (Ward 2000; Jarvis et al. 2017). However, polyploid establishment and propagation is often hindered by irregular meiotic segregation of chro- mosomes that lead to chromosomal abnormality and reproductive sterility (Renny–Byfield and Wendel 2014). On the contrary, stability in the number of somatic chromosomes of this species, especially in those cutivars that were reported from Bolivia, Chile and Peru (Cárde- nas and Hawkes 1948; Gandarillas and Luizaga 1967; Kolano et al. 2012), and the apparent absence of evidenc- es of mixoploidy in the highland varieties investigated in the present study, both failed to give credence to spo- radic reports of mixoploidy (Kawatani and Ohno 1950; Gandarillas 1979; Wang et al. 1993). This could in effect perhaps lead to the high fertility of quinoa (Ward 2000) that, along with its self–pollinating nature, has facilitat- ed its extensive propagation, especially in new regions. Karyomorphological trait variations resulting from structural changes of chromosomes, primarily as a result of rearrangement of chromosomal parts including trans- location, inversion and/or deletion, is an important char- acter for understanding relationship among closely relat- ed taxa (Peruzzi and Altinordu 2014). However, no pub- lished record of structural details of somatic chromo- somes of quinoa was available until the beginning of this century. Present study recorded small somatic chromo- somes with higher length variations, between 0.63–6.53 µm, and the longest chromosome was almost double the length of previously recorded longest chromosome which was 3.30 µm (Kolano et al. 2001; Palomino et al. 2008; Yangquanwei et al. 2013). This difference can be attributed to (a) the variable cytological techniques used in these studies that affect the degree of chromosomal condensation (Palomino et al. 2008), and, (b) the use of only 1–2 varieties of quinoa in the previous studies that exempted any scope of understanding existing cytologi- cal variations in the species. The latter reason stated here might also be true for the limited variations observed in the position of primary constrictions in the past studies. Chromosomes with only nearly median constrictions were observed by Palomino et al. (2008) and Bhargava et al. (2006), with only a single variety revealing 2 chro- mosomes with sub-median constrictions being reported by Bhargava et al. (2006). This was in contrast with the presence of sub–terminal constrictions in about 75% of the varieties of quinoa investigated in the present study revealing higher degree of intra–chromosomal variations (Table 1, Figure 3G). The number of chromosomes pos- sessing secondary constrictions in each complement of all the cultivars in the present study, corroborated obser- vations of Bhargava et al (2006), though contrasting observations of 2 pairs of secondary constrictions were reported by Palomino et al (2008). However, some oth- er studies have reported absence of chromosomes with secondary constrictions (Cárdenas and Hawkes 1948; Gandarillas and Luizaga 1967; Giusti 1970; Gandarillas 1979). Chromosomes possessing secondary constriction had median or nearly median primary constriction and secondary constrictions in sub–median position. Gross chromosome morphology obtained in the present study, although showed significant variations in both chromo- some length and constrictions, did not reveal any char- acteristic pattern that can be related to the geography or seed morphology of the studied cultivars. Study of chromosomal characteristics forms the basis of cytotaxonomy that can be used to understand simi- larity or divergence among related species, and is based 41Karyological variability and chromosomal asymmetry in highland cultivars of Chenopodium quinoa on asymmetry indices calculated with the help of chro- mosome measurements (Levitzky 1931; Stebbins 1971). Since asymmetry is an expression of chromosomal mor- phology, it is important to compare karyotypes on appro- priate statistical grounds and choose correct statistical parameters (Peruzzi and Altinordu 2014). However, there is a lack of general consensus on the specific karyotype asymmetry parameter to be utilized which necessitates use of multiple available parameters to conclude about the asymmetry status of a taxa and its comparison with relatives. Inter-chromosomal asymmetry increases with increasing difference between the lengths of smallest and largest chromosomes of a complement, and, are assessed by calculating A2 and CVCL. Intra-chromosomal asym- metry increases with shift in centromeric position from median to terminal constriction in a complement. This is assessed by calculating the values of TF%, Syi, AsK%, A1, MCA and CVCI (Paszko 2006; Peruzzi and Eroğlu 2013). In the present investigation, the different cultivars of quinoa showed significantly higher asymmetry than former studies as was evident by the karyotype asym- metry indices (Table 1). Stebbins asymmetry class, where all the cultivars of the present study were classified as 2B karyotype, endorsed more asymmetric karyotype than 1A and 1B obtained from 7 cultivars by Bharga- va et al. (2006). AI, with an average value of 3.21±0.62 also indicate the same. However, calculation of these 2 indices are based on a combination of inter- and intra- chromosomal asymmetry, which has been suggested against in recent studies (Peruzzi and Eroglu 2013). TF% that decreased with increasing asymmetry ranged from 39.69–43.23% with an average of 41.29±0.86%, lower than previously reported values of 43.8–47.4% (Bhar- gava et al. 2006; Palomino et al. 2008) affirming higher asymmetry revealed in the present study. Similarly, eval- uation of AR that ranged from 0.34 to 5.76 in the pre- sent study, but showed much less variation (1.00–1.86) revealing lesser asymmetry in the previous studies also imply presence of higher variation among the cultivars assessed in the present study (Bhargava et al. 2006; Palo- mino et al. 2008). Other asymmetry indices examined in the present study also supported the above contention as was evident by the calculated Pearson correlation. Inter– chromosomal and intra–chromosomal asymmetry indi- ces revealed strong correlation within themselves (Table 2). TF% was found to be negatively correlated with other intra–chromosomal indices of A1 (r= –0.978), CVCI (r= –0.770), MCA (r= –0.991), Syi (r= –0.995) and Ask% (r= –1.000) justifying increasing intra-chromosomal asym- metry with decreasing value of TF%. A2 was positively correlated with CVCL (r= 0.992) indicating increasing value with increasing differences in chromosome size. However, inter– chromosomal asymmetry indices were weakly correlated with intra–chromosomal indices, jus- tifying the suggestion of not including both type of indi- ces in a single analysis (Peruzzi and Eroglu 2013). Both inter- and intra-chromosomal asymmetry are necessary to reveal correlation among related taxa. This is best represented by using bi-dimensional scatter plots that include one inter-chromosomal and one intra-chro- mosomal parameter in each axis of the plot (Peruzzi and Eroğlu 2013). Restricted distribution of cultivars in the scatter plots generated in the present study comparing the values of A1 versus A2, CVCI versus CVCL, and, MCA versus CVCL indicated high similarity among the culti- vars with respect to chromosomal asymmetry (Figure 4). The plot indicated that, although the study revealed con- sistent differences within the karyotype of each cultivar, the characteristic distinctions were inadequate for culti- var identification, and not related to their geographical origin and seed morphology. Overall, high inter- and intra-chromosomal variabil- ity in the quinoa cultivars examined here conclusively exhibited that much variation exists in this species that is yet to be explored exhaustively. During early domes- tication events, conspicuous divergence of domesticated Table 2. Values of Pearson correlation analysis for intra- and inter-chromosomal asymmetry indices   A1 A2 CVCL CVCI MCA TF Syi AsK A1 1 A2 –0.071 1 CVCL –0.073 0.992* 1 CVCI 0.667* 0.110 0.118 1 MCA 0.985* –0.067 –0.070 0.765* 1 TF –0.978* 0.070 0.079 –0.770* –0.991* 1 Syi 0.981* –0.121 –0.125 0.745* 0.990* –0.995* 1 AsK 0.977* –0.070 –0.079 0.770* 0.991* –1.000* 0.994* 1 *Correlation is significant at 0.01 level. 42 Sreetama Bhadra, Zhi–Quan Cai species from their wild relatives resulted from artifi- cial selection and controlled reproduction, especially cross–fertilization, by human intervention, that ended in genetic bottleneck, thereby reducing genetic diversity of cultivated plants with respect to their wild counter- parts (Tanksley and McCouch 1997; Cornille et al. 2014). Along with this, early farmers selected few individuals with more desirable traits for next generation cultiva- tion with little effort of conserving the unselected mate- rials that were being pushed to oblivion, thereby causing severe loss of genetic diversity and limiting the gene pool of present day domesticated plants (Pan et al. 2016). In recent years, cultivation of local varieties of quinoa has been largely neglected because of increased pressure on local farmers for production of certified high–yielding quinoa varieties in international quinoa market (Salazar et al. 2019). The inter–varietal similarity and stability of chromosomal characteristics observed in our studied quinoa cultivars might be an indication of the bottleneck experienced by the plants due to their long domestica- tion history of over 7,000 years (Jarvis et al. 2017; Sala- zar et al. 2019), that had allowed enough time to impart homogeneity and stability in their chromosomal struc- tures. Loss of diversity in local quinoa gene pool in this way will pose serious threat in future breeding programs for quality improvement that necessitates an imperative exploration of chromosomal variability (Jarvis et al. 2017; Salazar et al. 2019). This entails an imperative investiga- tion of inter–varietal chromosomal diversity in quinoa that would include chromosomal f luorescent staining techniques, like FISH, incorporating a large number of cultivars encompassing varieties growing on the distinct eco–geographical regions of quinoa cultivation, primarily from the highland and the coastal lowland ecotypes. It will provide a comprehensive basis for evaluation, selec- tion, conservation and maintenance of existing germ- plasm, thereby assisting in future breeding programs. ACKNOWLEDGEMENT We thank Zhang YH from Yunnan Normal Univer- sity in Kunming for providing us with the instrumenta- tion facility used in this study. FUNDING This work was supported by the grants from the National Natural Science Foundation of China [31670686], post–doctor program in XTBG and the ‘135’ Program of Chinese Academy of Sciences [2017XTBG–T02]. Figure 4. Scatter diagrams for the Chenopodium quinoa cultivars of (A) A1 against A2 (B) CVCI against CVCL (C) MCA against CVCL. 43Karyological variability and chromosomal asymmetry in highland cultivars of Chenopodium quinoa REFERENCES Altinordu F, Peruzzi L, Yu Y, He X. 2016. A tool for the analysis of chromosomes: KaryoType. Taxon. 65:586– 592. Bennett MD, Leitch IJ. 2005. Nuclear DNA amounts in angiosperms: progress, problems and prospects. Ann Bot. 95:45–90. Bhadra S, Bandyopadhyay M. 2015. Karyomorphologi- cal investigations on some economically important members of Zingiberaceae from Eastern India. Car- yologia. 68:184–192. Bhargava A, Shukla S, Ohri D. 2006. Karyotypic stud- ies on some cultivated and wild species of Cheno- podium (Chenopodiaceae). Genet Resour Crop Evol. 53:1309–1320. Cárdenas M, Hawkes JG. 1948. Número de cromosomas de algunas plantas nativas cultivadas por los indios en los Andes [Number of chromosomes of some native plants grown by the Indians in the Andes]. Rev de Agricul Bolivia. 4:30–32. Spanish. Cornille A, Giraud T, Smulders MJM, Roldan–Ruiz I, Gladieux P. 2014. The domestication and evolution- ary ecology of apples. Trends Genet. 30:57–65. Fuentes–Bazan S, Mansion G, Borsch T. 2012. Towards a species level tree of the globally diverse genus Che- nopodium (Chenopodiaceae). Mol Phylogenet Evol. 62:359–374. Gandarillas H. 1979. Botanica. Quinua y kaniwa. Cul- tivos Andinos [Botany. Quinoa and Kaniwa. Andean Crops]. In: Tapia ME, editor. Serie libros y materiales educativos [Educational books and materials series]. Colombia (Bogotá): Instituto Interamericano de Ciencias Agricolas; p. 20–44. Spanish. Gandarillas H, Luizaga J. 1967. Número de cromosomas de Chenopodium quinoa Willd. en radículas y rai- cillas [Number of chromosomes of Chenopodium quinoa Willd. in radicles and rootlets]. Turrialba. 17:275–279. Spanish. Giusti L. 1970. El género Chenopodium en Argentina: I. Números de cromosomas [The genus Chenopodium in Argentina: I. Numbers of chromosomes]. Darwini- ana. 16:98–105. Spanish. Goldblatt P. 1980. The index to plant chromosome num- bers–past and future. Taxon. 56:984–986. Gonzalo MJ, Díaz A, Dhillon NPS, Reddy UK, Picó B, Monforte AJ. 2019. Re–evaluation of the role of Indian germplasm as center of melon diversification based on genotyping–by–sequencing analysis. BMC Genomics. 20:448. Graf BL, Rojas–Silva P, Rojo LE, Delatorre–Herrera J, Baldéon ME, Raskin I. 2015. Innovations in health value and functional food development of quinoa (Chenopodium quinoa Willd.). Compr Rev Food Sci Food Saf. 14:431–445. Guerra M. 2008. Chromosome numbers in plant cyto- taxonomy: concepts and implications. Cytogenet Genome Res 120:339–350. Heiser CB. 1963. Numeración cromosomica de plantas ecuatorianas [Chromosome numbering of Ecuado- rian plants]. Ciencia y Nat. (Quito). 6:2–6. Spanish. Hilu KW. 1993. Polyploidy and the evolution of domesti- cated plants. Am J Bot. 80(12):1494–1499. Jarvis DE, Ho YS, Lightfoot DJ, Schmöckel SM, Li B, Borm TJA, Ohyanagi H, Mineta K, Michell CT, Saber N, et al. 2017. The genome of Chenopodium quinoa. Nature 542:307–312. Kawatani T, Ohno T. 1950. Chromosome numbers of genus Chenopodium, I. Chromosome numbers of Mexican Tea (Chenopodium ambrosioides L.), Ameri- can Wormseed (Chenopodium ambrosioides L. var. anthelminticum A. Gray and some allied species. Jap Jour Genet. 25:177–180. Kjellmark S. 1934. Einige neue Chromosomenzahlen in der Familie Chenopodiaceae [Some new chromo- some numbers in the family Chenopodiaceae]. Bot Notiser. 1934:136–149. German. Kolano B, Pando LG, Maluszynska J. 2001. Molecu- lar cytogenetic studies in Chenopodium quinoa and Amaranthus caudatus. Acta Soc Bot Pol. 70:85–90. Kolano B, Tomczak H, Molewska R, Jellen EN, Maluszyn- ska J. 2012. Distribution of 5S and 35S rRNA gene sites in 34 Chenopodium species (Amaranthaceae). Bot J Linn Soc. 170:220–231. Krak K, Vít P, Belyayev A, Douda J, Hreusová L, Mandák B. 2016. Allopolyploid origin of Chenopodium album s. str. (Chenopodiaceae): A molecular and cytogenet- ic insight. PLoS ONE. 11:e0161063. Levan A, Fredga K, Sandberg AA. 1964. Nomenclature for centromeric position on chromosomes. Hereditas. 52:201–220. Leitch AR, Leitch IJ. 2008. Genomic plasticity and the diversity of polyploid plants. Science. 320:481–483. Levitsky GA. 1931. The karyotype in systematics. Bull Appl Bot Gen Pl Breed. 27:220–240. Mandák B, Krak K, Víta P, Pavlíková Z, Lomonosova MN, Habibi F, Wang L, Jellen EN, Douda J. 2016. How genome size variation is linked with evolution within Chenopodium sensu lato. Perspect Plant Ecol Evol Syst. 23:18–32. Maughan PJ, Kolano BA, Maluszynska J, Coles ND, Boni- facio A, Rojas J, Coleman CE, Stevens MR, Fairbanks DJ, Parkinson SE, et al. 2006. Molecular and cyto- logical characterization of ribosomal RNA genes in 44 Sreetama Bhadra, Zhi–Quan Cai Chenopodium quinoa and Chenopodium berlandieri. Genome. 49:825–839. Murphy KM, Bazile D, Kellogg J, Rahmanian M. 2016. Development of a worldwide consortium on evolu- tionary participatory breeding in quinoa. Front Plant Sci. 7:608. Palomino G, Hernández LT, Torres EDLC. 2008. Nucle- ar genome size and chromosome analysis in Cheno- podium quinoa and C. berlandieri subsp. nuttalliae. Euphytica. 164:221–230. Pan Y, Wang X, Sun G, Li F, Gong X. 2016. Application of RAD sequencing for evaluating the genetic diver- sity of domesticated Panax notoginseng (Araliaceae). PLoS ONE. 11:e0166419. Paszko B. 2006. A critical review and a new proposal of kar- yotype asymmetry indices. Plant Syst Evol. 258:39–48. Pereira E, Encina–Zelada C, Barros L, Gonzales–Barron U, Cadavez V, Ferreira ICFR. 2019. Chemical and nutritional characterization of Chenopodium quinoa Willd. (quinoa) grains: a good alternative to nutri- tious food. Food Chem. 280:110–114. Peruzzi L, Eroğlu HE. 2013. Karyotype asymmetry: again, how to measure and what to measure? Comp Cytogenet. 7:1–9. Peruzzi L, Altinordu F. 2014. A proposal for a multivari- ate quantitative approach to infer karyological rela- tionships among taxa. Comp Cytogen. 8(4):337–349. Rahiminejad MR, Gornall RJ. 2004. Flavonoid evidence for allopolyploidy in the Chenopodium album aggre- gate (Amaranthaceae). Plant Syst Evol. 246:77–87. Renny–Byfield S, Wendel JF. 2014. Doubling down on genomes: polyploidy and crop plants. Am J Bot. 101: 1711–1725. Ruiz KB, Biondi S, Oses R, Acuña–Rodríguez IS, Antog- noni F, Martinez–Mosqueira EA, Coulibaly A, Cana- hua–Murillo A, Pinto M, Zurita–Silva A, et al. 2014. Quinoa biodiversity and sustainability for food secu- rity under climate change. A review. Agron Sustain Dev. 34:349–359. Salazar J, Torres ML, Gutierrez B, Torres AF. 2019. Molecular characterization of Ecuadorian quinoa (Chenopodium quinoa Willd.) diversity: implications for conservation and breeding. Euphytica. 215:60. Sattler MC, Carvalho CR, Clarindo WR. 2016. The poly- ploidy and its key role in plant breeding. Planta. 243:281–296. Stebbins GL. 1971. Chromosomal evolution in higher plants. London (UK): Edward Arnold Publishers Ltd. Sun W, Ma X, Zhang J, Su F, Zhang Y, Li Z. 2017. Karyo- types of nineteen species of Asteraceae in the Hengd- uan Mountains and adjacent regions. Plant Diversity. 39:194–201. Tanksley SD, McCouch SR. 1997. Seed banks and molec- ular maps: unlocking genetic potential from the wild. Science. 277:1063–1066. Wang S, Tsuchiya T, Wilson H. 1993. Chromosome stud- ies in several species of Chenopodium from North and South America. J Genet Breed. 47:163–170. Ward SM. 2000. Allotetraploid segregation for single– gene morphological characters in quinoa (Chenopo- dium quinoa Willd.). Euphytica. 116:11–16. Wulff HD. 1936. Die polysomatie der Chenopodiaceen. Planta. 26:275–290. Yangquanwei Z, Neethirajan S, Karunakaran C. 2013. Cytogenetic analysis of quinoa chromosomes using nanoscale imaging and spectroscopy techniques. Nanoscale Res Lett. 8:463.