Genetic variation among biofortified and late blight tolerant potato (Solanum tuberosum L.) (mini tuber) production in Bangladesh

Md. Mushfiqur Rahman1*, Md. Nurul Amin1, Md. Harunor Rashid2, Md. Mazadul Islam2, Bimal Chandra Kundu2, Md. Mohi Uddin3 & E.H.M. Shofiur Rahaman4 1Breeder Seed Production Centre, 2Tuber Crop Research Centre, 3Breeder Seed Production Centre, Bangladesh Agricultural Research Institute, Debiganj, Panchagarh-5020, Bangladesh 4International Potato Center (CIP), Dhaka, Banani, Dhaka-1213, Bangladesh *Email: mushfiqsau1693@gmail.com


Introduction
Micronutrient deficiency is a major problem, which is often termed as hidden hunger. It affects over two billion people worldwide (1). Potato could contribute to a major role in food security for millions of people. Worldwide micronutrient malnutrition could be eradicated by serving biofortified potato and ensuring nutrient security. Biofortification is the development of micronutrient or vitamin-rich crops using the traditional crop improvement practices as well as modern biotechnology tools. It is a more sustainable and cost-effective method than the food supplementation, fortification and diet diversification (2). The most popular and earliest example of a success story of transgenic biofortification research is the development of Golden Rice or β-carotene rich rice. Golden rice transgenic lines have been under field trial in the Philippines (transgenic of RC-28), Bangladesh (transgenic of BRRI Dhan-29) and will certainly help to fight against iron deficiency (3). Potato is one of the most important tuber crops grown in Bangladesh for its high production, high nutritional values, easy digestibility and many industrial uses (4). According to a researcher (5), potato is considered as a promising candidate crop for feeding the hungry people of the world after rice and wheat. Genetic diversity and variability in a population of certain crop species is a prerequisite for an effective plantbreeding programme. A morphological tool for an efficient choice of parents is needed to initiate a hybridization programme. Principal Component Analysis (PCA) is a statistical method that attempts to describe the total variation in the multivariate sample using fewer variables than in the original data set (6). When a large number of desirable traits are involved, the application of PCA is recommended to facilitate the selection of the most relevant variables. The PCA identifies plant traits that distinguish the distinctness among the chosen genotypes for hybridization. A group of researchers (7), mentioned that these techniques aid in the classification of a population into groups of distinct orders based on similarities and thus assisting to choose parents for hybridization. The researcher and their associates (8), reported that PCA reduced the original five sensory attributes into two independent components, which accounted for 66% of the total variability in the data. The knowledge of genetic diversity helps to avoid duplicates in the collection provides a better classification and assist in breeding selection (9). The PCA has been used to partition observed agronomic variations in genotypes of many crops such as sweet potato landraces (7), rubber, rice, sesame and durum wheat (10)(11)(12)(13). Therefore, PCA is a very useful tool to classify genotypes of different crops. For a successful breeding programme, genetic diversity and variability are useful tool for an efficient choice of parents for hybridization to develop high-yield potential cultivars. In addition, multiple nutrition traits along with late blight tolerance will be possible to combine in a single cultivar along with high yield through hybridization (14). The purpose of this study was to investigate genetic divergence through PCA in biofortified and late blight tolerant CIP potato germplasm and the selection of desirable accession for hybridization in Bangladesh.

Materials and Methods
Forty-nine (thirty-nine high yielding biofortified and ten late blight tolerant) potato germplasm (Table 1) collected from Tuber Crop Research Centre (TCRC) of Bangladesh Agricultural Research Institute (BARI) and were evaluated during the 2019-20 growing season at Breeder Seed Production Centre (BSPC), Debiganj, Panchagarh, Bangladesh. The origin of those biofortified potato mini-tuber germplasm was collected from CIP, Bangladesh. The unit plot size was 3 m × 3 m with 3 replications. The plantlet was planted with a spacing of 40 cm × 25 cm during the first week of November 2019. Fertilizers were applied @ 325-220-250-120 kg per hectare of urea, TSP, MOP and gypsum respectively. Intercultural operations like earthing-up and weeding were done as and when required. The experimental plots were irrigated frequently to maintain adequate soil moisture and to keep the soil cool. For each accession, ten (10) randomly selected plants were used for the scoring of the characters. Data were collected on plant stand at 45 days after planting (DAP) and foliage coverage (%) at 45 DAP and plant height (cm), stem number per plant, number of tubers per plant, the yield of tubers per plant (gm) and the grade of tubers (%) were collected during harvest at 95 DAP. Quantification of morphological variability for Tuber skin color, tuber shape, flesh colour and tuber eye depth characters were done using the Shannon-Weaver Diversity Index (Supplementary Table 1 and Supplementary Table 2). The estimate of variability for each qualitative character was computed using the standardized Shannon-Weaver Diversity Index, designated as H' and has the formula: Where Pi is the proportion of the total number of genotypes belonging to the ith class. Following the work of groups of researchers (15), the same formula was applied to the quantitative characters.
Mean data for each character was subjected to multivariate analyses using FactoMineR, "Factoextra" package and "Performance Analytics" (16)(17)(18) for correlation in R Studio. The magnitude of genetic diversity among forty-nine genotypes was determined by using D 2 Mahalanobis genetic distance  (19). Hierarchical clustering using Tocher's method, as described by (20) was followed for the grouping of genotypes into distinct clusters.

Results and Discussion
A wide range of agronomic traits has been evaluated in potato germplasm collections for their possible use in the improvement of potato cultivars. Therefore, the account of characters association between the traits themselves and with the yield is very significant for the breeding materials subjected to selection for high yielding genotypes (Fig. 1). Stronger positive and positive correlations were found between tuber yield and plant height (r= 0.34), main stems per hill (r=0.13), the weight of tuber and yield (r=0.71), foliage coverage and tuber yield (r=0.32), plant vigor and plant height (r=0.56), plant vigor and foliage coverage (r=0.55), germination percentage and stem per hill (r=0.38), germination and foliage coverage (r=0.29), tuber count and stem per hill (r=0.40) (Fig. 1). These results revealed that any positive increase in such characters will boast the tuber yield. On the other hand, negative and significant correlations were found between tuber count and the weight of tuber (r=-0.61) (Fig. 1). The magnitudes of correlation coefficients were classified as based on asterisk marks (Fig. 1). Single asterisk marks were considered average; double asterisk marks were considered strong; triple asterisk marks were considered as very strong correlations. The positive correlation which means the change of the two traits be in the same direction (increase or decrease) and the negative correlation which mean the increase in the first trait combined with a decrease in the second trait (or reverse).
Another researcher (21) found a positive significant correlation between tuber weight and tuber count. We found a negative correlation between tuber weight and tuber count (r=-0.61). Other researchers (22) found similar results i.e., the significant correlation between tuber yield and tuber weight. PCA is commonly used to analyze phenotypic traits in crops and to select traits that contribute to genetic improvement (23 (24) found similar results. They reported that the first three PC accounted for 71% of the total diversity whereas the first PCA accounted for 29.01%. The first PC was more related to tuber yield per plant, tuber weight and the second PC was more associated with plant height, plant emergence/germination, plant vigor, foliage coverage, stems per hill and tuber count (Fig. 2). In addition, more variations were evident relatively in the traits, which were located on the first component (Fig. 2). The Germplasm located at quadrant four had wide variability (35.2%) and most of the germplasm present in into cluster III (Fig.  3). Q1 and Q2 had least divergence and present in mostly cluster I. Quadrant Q3 had moderate variability present in cluster II.
The cluster analysis divided genotypes based on similarity and provided a hierarchical classification (HCA). The results obtained following HCA were shown as a dendrogram (Fig. 4) in which three welldefined clusters were visible. Based on the hierarchical clustering on principal components the 49 genotypes were grouped into three different clusters (Table 3 and Fig. 3). Cluster II contained the maximum number of twenty-two genotypes followed by cluster III having sixteen genotypes and cluster I having eleven genotypes ( Fig. 3 and Table 3). On the basis of the cluster means, the important cluster was cluster II for percent germination, foliage coverage, stem per hill, plant height and plant vigor, cluster III for the weight of tuber and yield per plant. Cluster I had the lowest mean for germination percentage, foliage coverage, stem per hill, plant height, plant vigor, the weight of tuber, tuber count and yield per plant ( Table 4). The dendrogram was cut with the most distance from the other groups and 49 cultivars were included in three clusters (Fig. 4) (25), also found that the dendrogram obtained using highly variable morphological characters that separated 89 sweet potato genotypes into major clusters. About the distribution of tuber sizes, all the CIP clones showed a higher percentage of tubers size 25 mm and above (Table 5.). The quality of the seed depends on physiological age, uniformity and tuber size. The difference in tuber size is probably best explained by genetically different makeup of the TPS progeny. This is in agreement with (26). In addition, a researcher (27) reported that tuber size below 25 gm can successfully be used as seed tuber for next season, which gives the same potential yield as seed tubers (25-50 gm) of a standard cultivar. The Shannon-Weaver Diversity Index has a value ranging from 0 to 1, where 0 indicates the absence of diversity and 1 indicates maximum diversity (Supplementary Table 1 and Supplementary Table 2). The greater diversity in the studied materials will offer good scope for the improvement of potato through rational selection of parent's genotypes. Relatively more variations were evident in the traits, which were located on the first principal component. The above variables might be taken into consideration for the effective selection of parents during the hybridization programme. We selected CIP403, CIP404, CIP405, CIP413 and CIP445 accessions for their yield potentiality and tuber weight (Supplementary Fig. 1). The lines that are superior in terms of genetic diversity and agronomical properties during the improvement studies need to be selected (28). Similarly, another researcher (29) also selected nine genotypes among sixty-three potato genotypes in terms of maturity time, tuber shape plant height.  Thus, an improvement-breeding program involving such different cultivars may yield transgressive and heterotic segregants. Those selected clones will be used in the breeding programme after further evaluation.

Conclusion
The genetic variability among the CIP biofortified and late blight tolerant materials allows us to help the parental selection and pave the way to fortify and select blight tolerant potato for food security. The present study explored high level of morphological variation for market preferable tuber size, shape and yield among the accessions. Based on consumer preferences CIP403, CIP404, CIP405, CIP413 and CIP445 accessions present in cluster III were superior and will be used as parental material for fixation of heterosis in potato improvement programme.
administered through the Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) Fund for International Agricultural Research (FIA), contract number: 81219432. We also thank to International Potato Center (CIP), Lima, Peru for providing biofortified potato germplasm to TCRC, BARI for conducting the research.

Conflict of interests
Authors declare that no conflict of interests exists regarding the publication of this paper. Table 1. Morphological characters of biofortified potato varieties.