Assessment of Morphological Diversity, Yield Components, and Seed Biochemical Composition in Common Bean (Phaseolus vulgaris L.) Landraces

Abstract

The common bean (Phaseolus vulgaris L.) is a staple legume crop for the Balkan Peninsula, mainly used for food. A large range of landraces well adapted to the local climate are maintained by farmers. The aim of this study was to estimate in field conditions the variability in morphology and seed biochemical composition of fourteen local common bean genotypes. Sixteen morphological and three biological characteristics were evaluated. Considerable morphological variation was found among genotypes. Thirteen genotypes possessed a determinate growth habit, and one of them an indeterminate one. Plant weight without pods, total weight, and the number and weight of pods per plant displayed the highest variation coefficient (CV%) with 54.5, 44.2, 45, and 37.6%, respectively. According to the seed shape, the variation was among kidney, cuboid, and oval. Seed energy value varied from 339 to 347 kcal/100 g, the amount of protein from 21.8 to 27%, lipids content from 1.6 to 2.5%, carbohydrates from 54 to 60%, ash from 4.6 to 5.4%, dietary fibers from 3.3 to 5.9%, tannins from 14 to 21%, phenols from 1.3 to 17.2 mg/g, and antitrypsin activity from 1.2 to 3.1 units/mg FW. Genotypes were classified according to the earliness, plant and seed characteristics, and yield. Genetic material was discerned useful for future research and breeding purposes.

1. Introduction

The common bean (Phaseolus vulgaris L.) is one of the most widely distributed food legumes, grown mainly for its green pods and dry seeds [1,2,3]. Data presented by the Food and Agriculture Organization (FAO) for 2023 estimate the total world production as 28,505,529 tons, with a harvested area of approximately 37,750,753 hectares globally [4]. Common bean, as a dry grain, is cultivated extensively in the five continents, the main producers being countries from Asia, followed by Africa and Americas. The biggest producers of common bean in Europe are the countries near the Mediterranean basin, like Italy, France, and Spain, as well as the region of the West Balkans [4]. Vavilov identified southern Mexico and Central America as the center of origin for different crops, among them various beans [5]. The common bean was domesticated in two different geographical locations in the Americas [6]. Gepts et al. [7] reported on two large gene pools identified based on phaseolin seed protein variation. The domestication occurred independently in two regions: the Mesoamerican gene pool was found in Mexico and Central America to Panama, while the Andean gene pool was differentiated in Peru, Chile, Bolivia, and Argentina approximately 8000 years ago [8]. In Europe, 67 to 71% of the common bean landraces have been reported to belong to the Andean gene pool [9,10]. Due to several introductions from the Americas, direct exchanges between countries, and hybridization events between the two gene pools, Europe is nowadays considered as a secondary diversification center for Phaseolus vulgaris, harboring high genetic variability [9,10].

The common bean is broadly adapted to environments with moderate growing temperatures ranging from 15 °C to 27 °C and tolerates temperatures up to 29.5 °C, while higher temperatures (35 °C) and rainfed conditions during flowering and pod formation stages cause large abortion of blossoms and young pods, which leads to the greatest number of pods without well-formed grains, and consequently lower pod weights and fewer seeds per plant, resulting in reduced yield [11,12,13]. During the vegetation period, Phaseolus plants require an optimum quantity of rainfall between 350–500 mm and low relative humidity in order to avoid conditions favorable for bacterial or fungal diseases [14]. High temperatures and dry conditions during the vegetation cycle lead to a decreased final quantity and quality of common bean yield production. Legumes play an important role in improving soil fertility through N fixation, decrease utilization of synthetic ammonium fertilizers, and contribute to diminishing the greenhouse effect, a leading cause of global warming [12,13].

Pulses in general, and the common bean in particular, are multipurpose crops providing many benefits to farmers as a source of protein, animal feed, and cash income and in improving soil fertility through N fixation [14]. The common bean is a staple food legume crop globally and it is well known for its nutritional and health benefits [12,13,14,15,16]. The consumption of pulses in Europe and the EU has increased in recent years; however, there is a significant variation between countries, with major consumption of grain legumes focused on Spain, France, and the UK, with 60% of the total pulse consumption in the EU [17]. Dry beans are a cheap source of protein compared to protein in foods of animal origin and can be eaten in various ways as main or side dishes [18]. The common bean is characterized by its high nutritional value as a major source of protein (15–30%), carbohydrates (60–70%), dietary fibers, vitamins, and minerals such as K and Mg [19,20]. Compared to other legume species, the common bean has a relatively high content of protease inhibitors and tannins [21], as well as various phenolic compounds [22], known as antinutrients.

European common bean landraces are well adapted to the local edapho-climatic conditions and present high variability for most agronomical traits, especially the yield and its components [9,23,24,25] and pest and disease resistance [24], and are important for food security. However, this local biodiversity has been threatened by genetic erosion due to the modernization and globalization of food production, the commercialization of only a few highly productive varieties, the depopulation of many territories, and the ongoing climatic changes [26,27]. Many local bean varieties remain neglected or underutilized and constant efforts are made for collection, ex situ conservation in genebanks, and characterization of local plant populations in order to prevent the loss of genetic variation, to preserve as much as possible local diversity of this crop and increase the utilization of plant genetic resources [23,28]. The national collection of Phaseolus germplasm at IPPG-Sadovo, Bulgaria consists of about 2700 accessions with local and foreign origin [29]. In the recent decade, breeders, researchers, and farmers have presented increasing interest in local genetic materials, most of which has been preserved in genebanks [26]. To ensure better use of a local population germplasm, it needs to be characterized, evaluated, and documented, with open access for the users.

The aim of this study was to assess variations among fourteen selected accessions of Phaseolus vulgaris originating from the northern, central, and south eastern part of the Bulgarian country, grown under open field conditions, comparing morphology and yield-related components, as well as seed biochemical composition.

2. Materials and Methods

2.1. Plant Material

Twelve accessions and two cultivars with local origin were included in this study.plant material was collected by expeditions throughout the country and conserved in an ex situ collection in the genebank at the Institute of Plant Genetic Resources (IPGR) (

Table 1. Passport information of the Phaseolus genotypes included in this study.

No	Accession	Origin	Biological Status	Source
1	A9E0312	Silistra	Local population	IPGR genebank
2	Ustrem	DAI-Gen. Toshevo	Cultivar	DAI-Gen. Toshevo
3	A9E0430	Obretenik village, Ruse	Local population	IPGR genebank
4	A9E0543	Totleben village, Pleven	Local population	IPGR genebank
5	B1E0485	Velingrad	Local population	IPGR genebank
6	A9E1100	Bisser village, Haskovo	Local population	IPGR genebank
7	A9E0588	Umarevtsi, village, Lovech	Local population	IPGR genebank
8	B0E0048	Dimitrovgrad	Local population	IPGR genebank
9	Blyan	DAI-Gen. Toshevo	Cultivar	DAI-Gen. Toshevo
10	A8E0139	Klisura	Local population	IPGR genebank
11	A9E0460	Oryahovitsa village, Pleven	Local population	IPGR genebank
12	A9E1009	Dragushinovo village, Samokov	Local population	IPGR genebank
13	A9E0150	Sevlievo	Local population	IPGR genebank
14	C0E0309	Sadovo	Local population	IPGR genebank

). Accessions were chosen based on the results of preliminary trials with 60 genotypes of common beans provided by the IPGR and carried out in previous years. Seven accessions originated from the northern part of the country, five accessions were collected from the central and southern parts, and two cultivars were received by the Dobrudzha Agricultural Institute (DAI), General Toshevo, located in the northeastern part of the country.

The 60 accessions were preliminary screened for seed antitrypsin activity and protein SDS electrophoretic profiles. Populations were grouped into genotypes with low, intermediate, and high trypsin-inhibitory activity in seeds. Five distinct protein band profiles on electrophoregrams were established. Other parameters to be considered were the seed color, form, and size and the place of each accession’s collection. Thus, the sample size was reduced to 14 accessions to be studied in the field trials while keeping high local diversity.

2.2. Field Environmental Conditions

2.2.1. Location and Soil Characteristics

This experiment was conducted at the experimental field of the IPGR, Sadovo, Bulgaria during a period of three years from 2021 to 2023. Sadovo is situated in the central southern country region, at geographical coordinates 24°57′ E longitude and 42°09′ N latitude, and 153 m a.s.l. Plants were grown on a cinnamon forest soil with a neutral pH, without irrigation. Phaseolus plants could be infested by aphids or bacterial diseases. We applied standard plant protection in each phase of plant development against different pathogens and insects.

2.2.2. Weather Conditions

The mean monthly temperatures differed during the three experimental years, as in April and May in 2021 the T °C was higher compared with the same months in 2022 and 2023. The average and maximum temperatures during flowering and pod setting stages are important for grain formation and the quality and quantity of the final grain yield. The highest value of mean T °C in June was registered in 2022 with 21.5 °C, while the same month in 2021 and 2023 was 1.2 °C lower (Figure 1, Figure 2 and Figure 3). The average maximum T °C during the vegetation cycle in May ranged from 19.5 °C in 2023 to 24.2 °C in 2022, while in June the maximum average T °C was registered with 27.7 °C in 2022, as the same month in 2021 and 2023 showed a maximum average T of 26.5 °C. In July, the values of maximum temperature were over 30 °C, as follows: 32.0, 31.2, and 32.7 °C in consecutive experimental years. The monthly amount of rainfall differed by month and by year (Figure 1, Figure 2 and Figure 3). The biggest amount of rainfall from April to July was accumulated in 2023 with a value of 279.9 mm, while the quantity of rainfall in 2022 was 238.8 mm and in 2021—191.5 mm.

2.3. Experimental Design

Field trials were carried out in a Randomized Complete Block Design (RCBD) with three independent replications on plots of 5.6 m² and a 1 m distance between plots. Each accession was sown manually with 50 seeds per row, with a 2 m length in four rows of plots and 0.7 m interrow. Seeds were sown in favorable climatic conditions when the soil T °C was higher than 8 °C and there was availability of soil moisture. The seeds were sown in the first half of April between the 6th and 12th during experimental years and plants were matured and harvested in the last decade of July. The accessions were grown without herbicides and fertilizers under rain-fed conditions and were regularly hand-cleaned of weeds. Phaseolus plants could be infested by aphids (Aphis fabae Scop.) or bacteria, such as the common bacterial blight (Xanthomonas axonopodis pv. phaseoli and X. axonopodis pv. phaseoli var. fuscans), bacterial wilth (Curtobacterium flaccumfaciens pv. flaccumfaciens), halo blight (Pseudomonas syringae pv. phaseolicola), and anthracnose (Colletotrihum lindemuthianum). We applied fungicides (Kuprantol Duo 200 mL/ha, with active substances copper oxychloride—245 g/kg and copper hydroxide—244 g/kg) and insecticides (Sumi Alfa 5EK—registered trademark, 2500 g/ha, with active substance of Esfenvalerat 50 g/kg) during flowering and pod setting phases against insects and pathogens.

2.4. Morphological and Agro-Biological Traits

During the period of the field experiment, 4 agro-biological and 19 morphological traits were observed, following the International Phaseolus vulgaris descriptors [30] (

Table 2. Descriptor list of morphological and agro-biological characteristics of common bean genotypes.

No	Character	Abbreviation	No	Character	Abbreviation
1	Days to emergence	DE	13	Weight of pods/plant	WPPl
2	Days to flowering	DFl	14	Pod length	PL
3	Duration of flowering	DurFl	15	Pod width	PW
4	Days to maturity	DMat	16	Pod thickness	PT
5	Plant height	PH	17	Number of seeds/pod	NSP
6	Length of medium leaf	LMl	18	Seed length	SL
7	Width of medium leaf	WMl	19	Seed width	SW
8	Weight of plant	WPl	20	Seed thickness	ST
9	Number of branches	NBr	21	Weight of seeds/plant	WSPl
10	Weight of plant without pods	WPlWP	22	Weight of 100 seeds	W100S
11	First pod height	FPH	23	Yield per plot	Yplot
12	Number of pods/plant	NPPl

).

The harvest was performed 74 to 80 days after germination of the accessions, at 95% maturity of the pods. Ten randomly selected plants from the middle row of each plot were used for biometrical measurements, such as PL, PW, PT, NSP, SL, SW, and WSPl (Table 2), after harvesting from the field. Vegetative, inflorescence, and fruit data were collected in the field. Plant height was measured from the soil surface to the tip of the stem at the mid-grain feeling stage.

2.5. Seed Biochemical Analyses

Analyses were performed on seed material from field trials (three independent extractions per accession from finely ground seed material). Proximates and energy value were estimated according to AOAC standard methods in seeds ground to a fine powder. Water content in air-dried seeds was determined according to the AOAC [31]. The Kjeldahl method was applied to estimate the crude protein content on the basis of seed nitrogen content and a factor of 6.25 [32]. Lipid content of the seeds was determined gravimetrically after extraction with n-hexane in a Soxhlet apparatus for 8 h and removal of the solvent by vacuum evaporation [33]. Carbohydrate content was estimated after hydrolysis of starch into glucose according to [34]. Gravimetry was used for assessment of insoluble fiber [31]. The ash content was evaluated after incineration of the samples for 6 h at 550 °C [31]. The content of tannins was determined according to [34], and the energy value following [35]. For estimation of soluble sugar and starch, free amino acid, and phenol content, 80% ethanol extracts of seed flour (50 mg in 1 mL) were made and spectrophotometric methods were applied according to [36]. Soluble sugar content in the supernatant was analyzed directly by Anthrone reagent, and starch content was determined by the same method in the pellet (after hydrolysis with 30% perchloric acid); in both cases using a standard curve with glucose [37]. Amino acids were assessed by ninhydrin reagent using a standard curve made of equimolar quantities of L-proline and L-glycine [38]. Phenols were determined applying Folin–Ciocalteu reagent and a standard curve with caffeic acid, according to [39]. The method of Kakade et al. [40] was used for measuring the trypsin-inhibitory activity (TIA) with substrate benzoyl-DL-arginine-p-nitroanilide in protein extracts from 20 mg seed flour with 0.8 mL distilled H₂O. One unit of TIA was defined in the extract volume as giving 50% inhibition of trypsin activity.

2.6. Statistical Analysis

Two-way analysis of variance (ANOVA), a correlation matrix and Principal Component Analysis (PCA) were used for analysis of the collected data with statistical software Statistical Package for the Social Sciences (SPSS 19) [41]. Means were compared by the three Least Significant Difference (LSD) values at the 0.05, 0.01, and 0.001 probability levels (p) [42]. To identify the similarity and proximity of the studied Phaseolus genotypes, hierarchical cluster analysis was applied [43]. The Euclidian distance among groups was used as a measure of genetic similarity. Quantitative traits were expressed using descriptive analysis with mean, minimum, and maximum values, range of variation, and coefficient of variation (%). Biochemical data were analyzed applying multifactor ANOVA (StatGraphicsPlus 2.1.software), and statistically significant differences at p ≤ 0.05 are designated by different letters.

3. Results

3.1. Agro-Biological Data

Field trials were sown at different dates during the period of experiment, as follows: in 2021 on 12th April, in 2022 on 6th April, and in 2023 on 11th April. Numbers of days to emergence differed from 16 to 22 days depending on T °C and amount of rainfall and its distribution, particularly in April after the seed sowing date. The least number of days to emergence (18.6) belonged to accession A9E0150 and the longest period of (21.9) days to emergence was registered with accession COE0309 (Figure 4). The minimum number of days (44) to reach 50% flowering was registered in the last mentioned accession, COE0309, while the maximum days (49.1) for 50% flowering was observed in accession BOE0048. The average number of days to reach flowering stage was 46.9. Most of the accessions entered into the mass flowering stage in the first and second decade of June. The average maximum temperatures in June 2021 were 26.5 °C, and the highest maximum temperatures were measured in third decade with 31.5 °C (Figure 1). In June 2022 and 2023, the average monthly temperatures were 27.7 and 26.5 °C, respectively. The maximum temperatures during the last two years were 29.5 and 29.6 °C (Figure 2 and Figure 3).

Duration of flowering (from mass flowering to 50% stopped flowering) continued from 26.6 to 32.3 days, with a mean value of 28.6 days. Accessions with a 30 and more days duration of flowering were A9E1009, A9E0150, and COE0309. Days to maturity ranged from 73.7 to 78.7 days, with a mean value of 76.1. The three accessions showing the highest grain yield production had a vegetative cycle of 75, 75.2, and 76.7 days to maturity. According to the duration of the vegetative period, accessions were grouped in two main categories. Eight accessions were early matured, with a lower number of days to maturity than average, and six accessions with late maturation needed more days than average (76.1 days). Variation of this trait was the lowest among all phenological observations, with CV 2.1%.

3.2. Variation in Agro-Morphological Traits

Plant height ranged from the shortest accession, A9E0150, with PH 35.3 cm, to the tallest accession, A9E0312, with PH 67.2 cm, with a mean value of 48.3 cm (

Table 3. Morphological traits of the studied Phaseolus accessions, averaged for the three consecutive years 2021–2023.

Cat. No	H of Plant (cm)		Weight of Plant (g)		Number of Branches		Number of Pods/pl		Weight of Pods/pl (g)		Number of Seeds/Pod		Weight of Seeds/pl (g)		Size of Pod				Weight of 100 Seeds (g)		Harvest Index (%)
															Length. (cm)		Width. (cm)
	M	±D	M	±D	M	±D	M	±D	M	±D	M	±D	M	±D	M	±D	M	±D	M	±D	M	±D
A9E0312	67.2	26.0 **	45.3	15.9 **	2.9	0.5 **	15.5	4.4 **	20.6	4.0 *	5.4	0.9 **	13.8	1.7	10.5	1.4 **	0.9	0.1 **	22.5	−11.1 **	30.6	−8.5 **
Ustrem St	41.2		29.4		2.4		11.0		16.6		4.5		12.1		9.1		0.8		33.6		39.2
A9E0430	55.9	14.7 **	28.9	−0.5	2.5	0.1	12.0	1.0	13.9	−2.7	5.4	0.9 **	10.4	−1.7	10.1	0.9 **	0.9	0.1 **	24.9	−8.7 **	35.8	−3.3
A9E0543	41.1	−0.1	20.9	−8.5 *	2.3	−0.1	8.0	−3.1 **	12.3	−4.3 **	3.6	−0.9 **	8.7	−3.4 **	10.6	1.4 **	0.9	0.0	33.7	0.1	41.5	2.3
B1E0485	38.7	−2.5	45.1	15.7 **	2.3	−0.2	18.9	7.9 **	21.3	4.8 **	5.2	0.7 **	15.4	3.3 **	9.1	−0.2	0.8	0.0	20.2	−13.4 **	34.2	−5.0 *
A9E1100	40.4	−0.8	27.4	−2.0	2.6	0.2	9.5	−1.5	14.5	−2.1	4.2	−0.3	9.1	−3.0 *	10.7	1.5 **	0.9	0.0	33.7	0.1	32.8	−6.3 **
A9E0588	40.6	−0.6	16.8	−12.6 **	2.0	−0.5 **	7.8	−3.2 **	10.6	−6.0 **	4.2	−0.3	7.9	−4.1 **	9.4	0.3	1.0	0.2 **	36.8	3.2 **	47.2	8.0 **
BOE0048	51.8	10.6 **	51.5	22.1 **	2.9	0.5 **	18.8	7.7 **	21.9	5.4 **	5.7	1.2 **	15.4	3.4 **	9.4	0.3	0.7	−0.1 **	20.0	−13.6 **	30.6	−8.5 **
Blyan	37.6	−3.5	22.3	−7.1 *	20	−0.4 **	8.0	−3.1 **	11.4	−5.2 **	3.9	−0.7 *	7.5	−4.6 **	10.7	1.6 **	1.0	0.2 **	32.3	−1.3	33.4	−5.7 **
A8E0139	47.6	6.4 *	33.9	4.5	2.2	−0.2	12.3	1.3	18.5	1.9	4.6	0.1	13.5	1.4	11.1	2.0 **	1.0	0.1 **	31.3	−2.3 *	40.2	1.0
A9E0460	62.9	21.7 **	19.6	−9.8 **	2.3	−0.1	10.0	−1.0	10.6	−5.9 **	5.9	1.4 **	7.6	−4.5 **	10.2	1.1 **	1.0	−0.1 **	24.1	−9.5 **	37.9	−1.2
A9E1009	35.3	−5.8	11.4	−18.0 **	2.2	−0.2	6.0	−5.1 **	5.8	10.8 **	4.7	0.2	4.1	−7.9 **	10.8	1.7 **	0.8	−0.0	20.2	−13.4 **	36.1	−3.1
COE0309	42.1	0.9	26.6	−2.8	2.1	−0.3 **	6.3	−4.7 **	8.8	−7.7 **	5.7	1.2 **	6.6	−5.4 **	11.8	2.7 **	0.9	0.1 **	36.6	3.0 **	25.1	−14.0 **
A9E0150	55.3	14.2 **	23.0	−6.4	2.3	−0.1	9.8	−1.2	10.9	−5.7 **	4.9	0.4	7.5	−4.5 **	10.2	1.0 **	1.1	0.2 **	26.9	−6.7 **	33.9	−5.2 *
Min	35.3		20.9		2.0		5.9		5.8		3.6		4.1		9.1		0.7		20.0		25.1
Max	67.2		51.5		2.9		18.9		21.3		5.7		15.4		11.8		1.1		36.8		47.2
Mean	48.3		29.8		2.4		11.3		14.4		4.9		11.3		10.3		0.9		28.3		35.6
R	31.9		30.6		0.9		9.0		15.5		2.1		11.3		2.7		0.4		16.8		22.1
CV%	13.7		18.6		7.8		25.9		24.9		9.0		26.9		7.2		10.4		14.8		15.2

Significant difference compared to the standard variety Ustrem at p ≤ 0.05 *; p ≤ 0.01 **; M—mean value; ±D—difference regarding the mean value of the standard; CV—coefficient of variation; R—range of variation.

). The highest weight of plant was obtained from accession BOE0048 (51.5 g), with the highest number of primary branches (2.9) and the highest weight of pods per plant, while the lowest weight of plant was measured from accession A9E0543 (20.9 g), with a mean value of this trait of 29.8 g. Number of branches ranged between 2.0 and 2.9 with a mean value of 2.4. Variability of each morphological trait was shown by the coefficient of variation (CV%) (Table 3).

Number of primary branches was found to have the second lowest coefficient of variation (CV 7.8%), after the lowest variation in length of pod (CV 7.2%). The maximum number of pods per plant was achieved by accession B1E0485 with 18.9 pods, followed by accession BOE0048 with 18.8 pods per plant, while accessions A9E1009 and COE0309 produced the lowest number of pods per one plant with 5.9 and 6.3, respectively. The coefficient of variation of this trait was the highest among all studied morphological characteristics, with a value of 25.9% (CV%), followed by weight of pods per plant with CV 24.9%. Six accessions (43%) produced pods with higher weight than average, while the weight of pods of the remaining eight accessions (57%) were lower than the average value of 14.4 g. Number of seeds per pod varied from 3.6 to 5.7 with a mean value of 4.9 seeds per pod. The highest number of seeds per pod belonged to accession BOE0048 with the maximum value of 5.7. Accession A9E0543 formed pods with the lowest number of seeds per pod (3.6).

The variability of number of seeds per pod was comparatively low compared with the previous yield components, with a coefficient of variation (CV%) of 9.0%. Weight of seeds per plant ranged from 4.1 to 15.4 g. The two genotypes BOE0048 and B1E0485 were significantly the most productive with 15.4 g seeds per plant, followed by two other genotypes, A9E0312 and A8E0139, with 13.8 g and 13.4 g, respectively. This trait showed the highest coefficient of variability with 26.9% (CV%) among the analyzed quantitative traits (Table 3). Regarding characterization of pod size, based on pod length (excluding the beak), the maximum value of 11.8 cm was found in genotype COE0309, with an average of 10.3 cm. The pod length was found to be the least variable morphological characteristic, with CV 7.2%. The pod width varied between 0.7 cm (BOE0048) and 1.1 cm (A9E0150), with an average of 0.9 cm. Pod width showed higher variability compared with pod length, with CV 10.2%. The average weight of 100 seeds for all the studied accessions was 28.3 g, with a range from 20.0 to 36.8 g. Accessions with small seed size were 43% of the total, with a weight of 100 seeds less than 25 g, while accessions with medium seed size consisted of 57%, with a weight of 100 seeds between 25 g and 40 g. Accession BOE0048 was found to have the smallest seeds among the studied accessions, with 20.0 g W100S, while the biggest seeds belonged to genotype A9E0588 with 36.8 g W100S. The average value of the harvest index (HI) based on this three-year experiment was 35.6%. The harvest index of three accessions (A9E0588, A9E0543, and A8E0139) exceeded the control cultivar “Ustrem”. Ten accessions expressed lower reproductive efficiency, with HI < 39.2%, compared with the standard cultivar “Ustrem” (Table 3). The lowest harvest index was shown by the accession COE0309, with a value of 25.1%, and the highest value of harvest index was registered with accession A9E0588, with a value of 47.1%. The range of the harvest index was 22% and the mean value of HI (%) for the studied accessions, averaged over the three years of the experiment, was obtained with a value of 35.6%. There were significant differences between the accessions A9E0312, A9E1100, A9E0588, BOE0048, and “Blyan” compared to the standard cultivar “Ustrem” at p ≤ 0.01 **, while significant differences at p ≤ 0.05 * appeared between accessions B1E0485, COE0309, and A9E0150.

3.3. Phenotypic Correlation

The positive and negative correlations were revealed by Pearson’s correlation analysis based on 23 quantitative traits in the common bean accessions (

Table 4. Correlation among different quantitative characteristics in Phaseolus vulgaris accessions.

	DE	DFl	DurFl	DMat	PH	LMl	WMl	WPl	NBr	WPlWP	FPH	NPPl	WPPl	PL	PW	PT	NSP	SL	SW	ST	WSPl	W100S	Yplot
DE	1	−0.47	0.51	−0.5	−0.4	0.49	−0.3	−0.536 *	−0.2	−0.344	−0.603 *	−0.577 *	−0.502	0.843 **	−0.218	0.163	0.096	0.503	−0.684 **	0.023	−0.487	0.345	−0.35
DFl		1	−0.3	0.63 *	0.41	0.29	0.06	0.357	−0.16	0.491	0.051	0.278	0.072	−0.571 *	0.096	0.134	−0.105	−0.439	0.136	0	0.041	−0.773 **	−0.13
DurFl			1	0.08	−0.1	0.34	−0.5	−0.342	−0.48	0.135	−0.347	−0.757 **	−0.712 **	0.678 **	0.308	0.653 *	−0.546 *	0.613 *	−0.788 **	0.01	−0.715 **	0.216	−0.715 **
Dmat				1	0.31	0.31	−0.1	0.206	−0.24	0.392	0.093	0.067	−0.103	−0.379	0.102	0.101	−0.392	−0.25	−0.225	−0.19	−0.124	−0.676 **	−0.37
PH					1	−0.3	−0.2	0.602 *	0.257	0.622 *	−0.239	0.317	0.375	−0.271	0.558 *	0.244	0.243	−0.474	0.149	−0.04	0.325	−0.307	−0.37
LMl						1	0.03	−0.208	−0.19	0.096	−0.132	−0.328	−0.447	0.229	−0.304	0.255	−0.252	0.292	−0.686 **	0.095	−0.48	−0.275	−0.37
WMl							1	0.348	0.484	0.119	0.457	0.508	0.456	−0.549 *	−0.378	−0.540 *	0.307	−0.267	0.374	0.496	0.474	0.162	0.543 *
WPl								1	0.623 *	0.831 **	0.181	0.826 **	0.813 **	−0.632 *	0.222	−0.152	0.443	−0.783 **	0.346	0.118	0.787 **	−0.155	0.198
NBr									1	0.288	0.337	0.682 **	0.762 **	−0.322	−0.277	−0.466	0.701 **	−0.489	0.294	0.168	0.755 **	0.159	0.417
WPlWP										1	−0.004	0.407	0.355	−0.381	0.363	0.263	0.088	−0.49	−0.082	0.151	0.32	−0.277	−0.27
FPH											1	0.343	0.277	−0.607 *	−0.231	−0.19	−0.174	−0.045	0.486	0.229	0.28	0.024	0.621 *
NPPl												1	0.959 **	−0.771 **	−0.112	−0.566 *	0.627 *	−0.879 **	0.643 *	0.035	0.962 **	−0.152	0.617 *
WPPl													1	−0.643 *	−0.015	−0.525	0.688 **	−0.796 **	0.652 *	0.048	0.993 **	0.035	0.607 *
PL														1	−0.02	0.281	−0.128	0.621 *	−0.697 **	−0.22	−0.633 *	0.322	−0.5
PW															1	0.701 **	−0.221	0.018	0.134	0.288	−0.066	0.268	−0.35
PT																1	−0.558 *	0.506	−0.352	0.27	−0.587 *	0.104	−0.614 *
NSP																	1	−0.656 *	0.298	0.135	0.717 **	0.162	0.371
SL																		1	−0.457	0.093	−0.814 **	0.338	−0.33
SW																			1	0.182	0.658 *	0.121	0.758 **
ST																				1	0.044	0.551 *	0.131
WSPl																					1	0.05	0.640 *
W100S																						1	0.243
Yplot																							1

*: correlation is significant at the 0.05 level (2-tailed); **: correlation is significant at the 0.01 level (2-tailed). Significant correlations are in bold. Abbreviations: days to emergence (DE), days to flowering (DFl), duration of flowering (DurFl), days to maturity (DMat), plant height (PH), length of medium leaf (LMl), width of medium leaf (WMl), weight of plant (WPl), number of branches (NBr), weight of plant without pods (WPlWP), first pod height (FPH), number of pods/plant (NPPl), weight of pods/plant (WPPl), pod length (PL), pod width (PW), pod thickness (PT), number of seeds/pod (NSP), seed length (SL), seed width (SW), seed thickness (ST), weight of seeds/plant (WSPl), weight of 100 seeds (W100S), yield per plot (Yplot).

).

Several traits related to yield and yield components showed strong positive and negative correlations, as is shown in Table 4. A strong significant and positive correlation was observed between the weight of plant without pods and weight of plant (0.831 **). The number and weight of pods per plant were in strong positive correlation with the weight of plant (0.826 ** and 0.813 **) and with number of branches (0.682 ** and 0.762 **). Strong positive correlation was obtained between weight of pods per plant and number of pods per plant (0.959 **). The length of pod was in strong positive correlation with number of days to emergence (0.843 **) and duration of flowering (0.678 **) and had a strong negative association with number of pods per plant (−0.771 **). A medium-strength negative correlation was observed between pod length and days to flowering (−0.571 *) and width of medium leaf (−0.549 *), as well as between pod length and weight of plant (−0.632 *) and between pod length and first pod height (−0.607 *) and weight of pods per plant (−0.643 *). A medium-strength positive correlation was established between pod width and plant height (0.558 *). Pod thickness had a medium-strength positive correlation with duration of flowering (0.653 *) and a strong positive one with the pod length (0.701 **). The number of seeds per one pod was in strong positive correlation with number of branches (0.701 **) and weight of pods per plant (0.688 **). Seed length depended positively on duration of flowering (0.613 *) and pod length (0.621 *), but was strongly and negatively correlated with weight of plant (−0.783 **) and the number and weight of pods per plant (−0.879 **; −0.796 **), and had a medium-strength association with number of seeds per one pod (−0.656 *). Seed width had strong negative correlations with days to emergence (−0.684 **), duration of flowering (−0.788 **), length of medium leaf (−0.686 **), and length of pod (−0.697 **). Positive medium-strength associations were established between seed width and number and weight of pods per plant (0.643 * and 0.652 *). Weight of seeds per plant was in negative association with duration of flowering (−0.715 **), weight of plant (0.787 **), number of branches (0.755 **), number and weight of pods per plant (0.962 ** and 0.993 **), and number of seeds per pod (0.717 **). Negative strong correlations were observed between weight of seeds per plant and seed length (−0.814 **). The size of seeds expressed by the weight of 100 seeds had strong negative dependencies from number of days to flowering (−0.773 **) and number of days to reach maturity stage (−0.676 **). The seed yield per plot showed a strong positive correlation with seed width (0.758 **) and medium-strength correlations with medium leaf width (0.543 **), first pod height (0.621 *), number and weight of pods per plant (0.617 *; 0.607 *), and seed width (0.758 **). A negative correlation to seed yield per plot was observed with duration of flowering (−0.751 **). The trait with only one medium-strength correlation was seed thickness, which was associated with weight of 100 seeds (0.551 *) (Table 4).

The influence of the factors year, genotype, and the interaction of year * genotype on the weight of seeds per plant (

Table 5. Influence of year, genotype, and the interaction of year * genotype on seed weight per plant.

Source of Variation	SS	df	MS	F	F Crit	ŋ.%	Sign.
Year	225.3	2	112.6	12.2	7.5	11.4	***
Genotype	854.4	13	65.7	7.1	3.1	43.4	***
Year * Genotype	113.3	26	4.4	0.5	2.5	5.8	n.s.
Error	776.5	84	9.2			39.4
Total	1969.5	125				100.0

SS—sum of squares; df—degree of freedom; MS—variance; F—F experimental; F crit—F critical; ŋ.—force of influence of the factor (%); n.s.—non significant, ***—significance proven at α = 0.001.

) indicated that the strongest factor was the genotype (ŋ-43.4%), followed by the factor year (ŋ-11.4%) and the interaction of year * genotype (ŋ-5.8%). The influence of the factors genotype and year were statistically significant at α = 0.001, while the third factor, the interaction of year * genotype, was statistically non-significant. The results from analysis of variance showed that influence of factors year and genotype on yield per plot were almost equal (ŋ-32.6% and ŋ-32.7%), followed by the interaction of year * genotype (ŋ-20.4%) (

Table 6. Influence of factors year, genotype, and the interaction of year * genotype on yield per plot.

Source of Variation	SS	df	MS	F	F Crit	ŋ.%	Sign.
Year	5,893,823.8	2	2,946,911.9	95.4	7.5	32.6	***
Genotype	5,910,807.0	13	454,677.5	14.7	3.1	32.7	***
Year * Genotype	3,696,186.5	26	142,161.0	4.6	2.5	20.4	***
Error	2,595,728.5	84	30,901.5			14.3
Total	18,096,545.8	125				100.0

SS—sum of squares; df—degree of freedom; MS—variance; F—F experimental; F crit—F critical; ŋ —force of influence of the factor (%); ***—proven at α = 0.001.

). The influence of the three factors year, genotype, and year * genotype were statistically significant at α = 0.001.

3.4. Hierarchical Cluster Analysis

Fourteen common bean accessions were assessed using the complex comparison of twenty three traits. The dendrogram shows the genetic similarities among all studied accessions (Figure 5).

There were five major clusters with different numbers of genotypes, with genotype COE0309 forming a separate group and A9E0312 and Ustrem clustering together. The first cluster was divided in two sub-clusters; each contained three accessions. The first sub-cluster included A9E0543, A8E0139, and BOE0048. The most distinctive trait for the last three accessions was a higher yield per plot. The second sub-cluster also contained three accessions, A9E0430, A9E0588, and A9E1100. They differed to other accessions in seed length, seed width, and seed thickness and high values of weight of 100 seeds and yield per plot. One accession and the cultivar “Blyan” were located individually, aside from the two sub-clusters. The accession B1E0485 was differentiated with a greater number of days to reach maturity compared to the rest of the accessions from the first cluster. Accession B1E0485 showed a high plant height, high value of weight of plant, and higher number of branches. This accession had the highest number of pods and weight per plant. Cultivar “Blyan” was placed aside of the two sub-clusters A and B, situated after the accession B1E0485. The two last accessions were similar in some of their quantitative traits, like days to maturity and characteristics related to pods and seeds. Cultivar “Blyan” showed a higher first pod height of 20.2 cm above the soil surface compared with accession B1E0485 with 18.9 cm, and a higher weight of 100 seeds—28.9 g against 25.5 g. The yield per plot of cultivar “Blyan” was lower (624.1 g) compared with the accession B1E0485 (669.1 g). Accessions with a narrow genetic base were shown by their coefficients of similarity, as follows: A9E0543 with A8E0139 with a coefficient of 6.168, accession A9E0430 was genetically narrow with A9E0588 based on the coefficient of similarity of 15.238, and A9E0430 with accession A9E1100 with a coefficient of 16.082, as shown in

Table 7. Accessions with genetically narrow base.

Coefficients	Genetically Narrow
6.168	A9E0543	A8E0139
15.238	A9E0430	A9E0588
16.082	A9E0430	A9E1100

.

Intercluster coefficients showed the maximum value of genetic distance between accession A9E0460 and accession BOE0048, followed by accession A9E0460 and accession A9E0543, and accessions A9E460 and A9E0430 (

Table 8. Accessions with high genetic distance.

Coefficients	Genetically Distance
606.0	A9E0460	BOE0048
580.7	A9E0460	A9E0543
557.8	A9E0460	A9E0430

). The second cluster contained five accessions divided into two sub-clusters, and one accession was located individually. The first three accessions showed similarities in several morphological traits and they differed from other accessions with the lowest yield per plot. The accession A9E0312 and cultivar “Ustrem” showed similar yield per plot and related yield components. The accession COE0309 was situated individually and differed from others in the group with the shortest vegetation cycle, with the longest pods and the biggest number of seeds per pod.

3.5. Principal Component Analysis

To understand better the diversity among the studied genotypes and to be more useful for breeders, the data collected were analyzed by Principal Component Analysis (PCA). This analysis allows an opportunity to observe the distribution of the genotypes and characteristics in two figures according to their parameters (Figure 6 and Figure 7).

The results showed that the first component (PC1) explained 42.1% of the total variation, the second (PC2) 20.1%, and third (PC3) 11.7%. The three components together explained 73.9% of the total variation of the accessions with relevant traits (

Table 9. Distribution of the total variation among PCA components.

Component	Total	% of Variance	Cumulative %
1	9.3	42.1	42.1
2	4.4	20.1	62.2
3	2.6	11.7	73.9

).

The characteristics responsible for the separation in the first component were days to emergence, duration of flowering, length and width of medium leaf, weight of plant, number of branches, number and weight of pods per plant, pod length, pod thickness, number and weight of seeds per plant, seed length, seed width, and yield per plot (

Table 10. Loadings of morphological traits in the first three principal components (PC1, PC2, and PC3) derived from Principal Component Analysis (PCA) of common bean genotypes.

			Component
Traits	var No	PC1	PC2	PC3
DE	var1	−0.668	−0.387	0.367
DFl	var2	0.228	0.777	−0.373
DurFl	var3	−0.822	0.164	0.209
DMat	var4	0.012	0.750	−0.450
PH	var5	0.296	0.691	0.473
LMl	var6	−0.424	0.220	−0.270
WMl	var7	0.607	−0.246	−0.226
WPl	var8	0.760	0.461	0.321
NBr	var9	0.748	−0.158	0.323
WPlWP	var10	0.312	0.745	0.300
FPH	var11	0.464	−0.171	−0.579
NPPl	var12	0.970	0.100	0.073
WPPl	var13	0.944	−0.016	0.269
PL	var14	−0.808	−0.332	0.365
PW	var15	−0.123	0.531	0.310
PT	var16	−0.592	0.495	0.278
NSP	var17	0.618	−0.244	0.542
SL	var18	−0.804	−0.302	−0.091
SW	var19	0.857	−0.429	−0.198
WSPl	var20	0.951	−0.064	0.251
W100S	var21	−0.104	−0.625	0.470
Yplot	var22	0.689	−0.550	−0.254

Abbreviations: days to emergence (DE), days to flowering (DFl), duration of flowering (DurFl), days to maturity (DMat), plant height (PH), length of medium leaf (LMl), width of medium leaf (WMl), weight of plant (WPl), number of branches (NBr), weight of plant without pods (WPlWP), first pod height (FPH), number of pods/plant (NPPl), weight of pods/plant (WPPl), pod length (PL), pod width (PW), pod thickness (PT), number of seeds/pod (NSP), seed length (SL), seed width (SW), seed thickness (ST), weight of seeds/plant (WSPl), weight of 100 seeds (W100S), yield per plot (Yplot). The bolded values determine which factor the traits fall into.

).

For the second component, the most important characteristics were days to flowering, days to maturity, plant height, weight of plant without pods, pod width, and weight of 100 seeds. For the separation in the third component, the first pod height was recognized. The genotypes A9E1009, COE0309, and A9E0460 were of special interest due to their high values along the first, second, and third components (Figure 6). The genotype COE0309 was characterized by the smallest number of days to flowering and the shortest cycle to reach maturity. This is an advantage in plant growth as the high temperatures and low air humidity during flowering and pod setting stages may be avoided and consequently more pods and well matured seeds can be produced by the end of vegetation period. The genotype COE0309 was differentiated from others with the highest weight of 100 seeds (var21). The accession A9E1009 possessed one of the highest number of pods per plant, the smallest number of seeds per pod, and the lowest weight of 100 seeds. These characteristics were situated in the opposite side of the accession A9E1009 in PC1.

The morphological characteristics with high importance for yield productivity, i.e., the number and weight of pods per plant, number of seeds per pod, and weight of seeds per plant, were situated closely in PC1. Accessions A9E0543, BOE0048, and A8E0139 were situated in the same component and differed from the others by their high yield per plot, high weight of 100 seeds, weight of seeds per plant, and weight of pods per plant. Accessions A9E0460, A9E0150, and COE0309 were situated closely and differed from the others by the lowest yield productivity, low number of pods per plant, weight of seeds per plant, and number of seeds per pod.

3.6. Seeds’ Biochemical Composition

Moisture content in the seeds showed non-significant variations in the range of 11.0–12.2%. Seed energy value was between 339 and 347 kcal/100 g (on average 343.4 ± 1.9 kcal/100 g) (

Table 11. Moisture content, main nutraucetical compounds, and energy value in seeds from the studied accessions.

Accession	Mois- ture %	Proteins %	Oils %	Ash %	Fibres %	Tannins %	Carbohyd- Rates %	kcal/ 100 g
A9E0312	11.0 ± 0.8 a	25.6 ± 0.2 f	2.2 ± 0.03 e	5.1 ± 0.03 e	5.4 ± 0.1 fg	16.0 ± 0.1 b	56.3 ± 0.7 bc	347 ± 4 c
Ustrem	12.1 ± 0.7 a	23.4 ± 0.4 d	2.4 ± 0.07 f	4.8 ± 0.1 cd	5.3 ± 0.1 def	17.4 ± 0.1 cd	57.4 ± 1.3 cde	345 ± 3 abc
A9E0430	11.9 ± 0.7 a	22.7 ± 0.1 bc	2.5 ± 0.07 f	4.8 ± 0.1 d	3.3 ± 0.2 a	19.1 ± 0.1 fg	58.2 ± 0.5 d–g	346 ± 3 bc
A9E0543	11.9 ± 0.7 a	23.5 ± 0.2 d	2.1 ± 0.1 de	4.6 ± 0.1 ab	4.8 ± 0.3 cde	18.0 ± 0.6 de	58.1 ± 0.5 c–f	344 ± 2 abc
B1E0485	12.2 ± 0.5 a	23.4 ± 0.3 d	1.9 ± 0.1 bc	5.1 ± 0.03 e	5.7 ± 0.3 fg	16.5 ± 0.5 bc	57.6 ± 0.2 cde	340 ± 1 ab
A9E1100	11.9 ± 0.7 a	24.2 ± 0.1 e	2.2 ± 0.1 e	4.6 ± 0.1 a	4.8 ± 0.1 c	21.5 ± 0.3 i	57.2 ± 0.6 cde	345 ± 3 abc
A9E0588	12.2 ± 0.7 a	25.4 ± 0.2 f	2.5 ± 0.1 f	4.7 ± 0.01 abc	4.7 ± 0.1 cd	17.1 ± 0.1 cd	55.3 ± 0.4 ab	345 ± 3 abc
BOE0048	11.6 ± 0.6 a	24.5 ± 0.2 e	2.2 ± 0.1 e	5.1 ± 0.03 e	5.6 ± 0.3 fg	20.2 ± 0.1 h	56.7 ± 0.7 bcd	344 ± 3 abc
Blyan	11.8 ± 0.5 a	22.2 ± 0.2 ab	1.7 ± 0.07 ab	4.6 ± 0.03 a	5.7 ± 0.3 fg	13.6 ± 0.6 a	59.9 ± 0.7 g	343 ± 3 abc
A8E0139	12.2 ± 0.5 a	21.8 ± 0.2 a	1.7 ± 0.07 ab	4.7 ± 0.0 abc	5.3 ± 0.2 ef	18.4 ± 0.2 ef	59.7 ± 0.6 fg	341 ± 2 abc
A9E0460	11.7 ± 0.6 a	27.0 ± 0.2 g	1.9 ± 0.07 cd	5.3 ± 0.1 f	3.9 ± 0.1 b	21.3 ± 0.2 i	54.2 ± 0.7 a	342 ± 3 abc
A9E1009	11.9 ± 0.8 a	23.0 ± 0.1 cd	1.6 ± 0.03 a	4.7 ± 0.03 abc	5.9 ± 0.1 g	19.8 ± 0.1 gh	58.9 ± 0.9 efg	341 ± 3 abc
A9E0150	11.6 ± 1.0 a	26.6 ± 0.2 g	1.7 ± 0.03 a	5.4 ± 0.1 f	3.3 ± 0.3 a	16.6 ± 0.4 bc	54.4 ± 0.9 a	339 ± 3 a
COE0309	11.7 ± 0.7 a	22.2 ± 0.1 ab	1.9 ± 0.03 cd	4.7 ± 0.03 abc	3.6 ± 0.3 ab	14.0 ± 0.2 a	59.6 ± 0.6 fg	345 ± 3 abc

Results are mean values and standard deviations of three replicates. Different superscript letters denote significant differences at p ≤ 0.05.

). The crude protein content varied between 21.8% and 27%, being lowest in A8E0139 and highest in A9E0460, with a mean value of 23.94 ± 1.37%. The lipid content was between 1.6% and 2.5%, lowest in A9E1009 and highest in A9E0430 (mean value of 2.02 ± 0.26%). The ash content varied between 4.6–5.4% (on average 4.83 ± 0.23%), lowest in A9E1100 and cv Blian and highest in A9E0150. The fiber content was 3.3–5.9% (4.77 ± 0.76%), lowest in A9E0150 and A9E0430 and highest in A9E1009. Tannin content in seeds varied from 13.6 to 21.5% (17.81 ± 1.94%), lowest in cv Blyan and highest in A9E1100. The carbohydrate content was between 54.2% and 59.9% (57.4 ± 1.5% on average), lowest in A9E0460 and highest in cv Blian.

Most of the seeds’ carbohydrate content is in the form of starch reserves. In our study, seed starch content varied between 580.5 and 795.7 mg of starch per g FW, on average 689.4 ± 37.1 mg·g⁻¹ (

Table 12. Soluble sugars, starch, phenols, and free amino acids content and trypsin-inhibitory activity in the bean seeds.

Accession	Sol.Sugars mg·g−1FW	Starch mg·g−1FW	Phenols mg·g−1FW	Amino Acids mg·g−1FW	TIA U·mg−1FW
A9E0312	49.9 ± 1.9 abc	580.5 ± 25.1 a	1.3 ± 0.10 a	3.6 ± 0.27 abc	1.2 a
Ustrem	63.2 ± 10.7 a–f	633.2 ± 70.1 ab	1.6 ± 0.03 a	3.3 ± 0.15 abc	2.4 bc
A9E0430	62.9 ± 6.9 b–f	678.2 ± 30.50 a–d	1.4 ± 0.03 a	3.8 ± 0.62 abc	1.9 b
A9E0543	64.4 ± 6.2 b–f	692.8 ± 58.30 a–d	1.4 ± 0.01 a	3.9 ± 0.45 bc	2.3 bc
B1E0485	51.2 ± 5.1 a–d	691.6 ± 48.0 a–d	1.6 ± 0.14 a	2.9 ± 0.34 ab	1.5 a
A9E1100	76.4 ± 8.6 fg	696.9 ± 45.50 a–d	1.6 ± 0.09 a	3.9 ± 0.68 bc	2.4 bc
A9E0588	64.1 ± 5.4 b–f	652.9 ± 39.3 abc	1.6 ± 0.05 a	4.5 ± 0.14 c	3.1 c
BOE0048	46.9 ± 5.4 ab	686.2 ± 12.30 a–d	1.6 ± 0.01 a	3.3 ± 0.42 abc	1.7 ab
Blyan	43.7 ± 5.7 a	687.9 ± 41.10 a–d	1.3 ± 0.15 a	2.6 ± 0.25 a	1.7 ab
A8E0139	56.8 ± 5.9 a–e	770.8 ± 40.0 cd	17.2 ± 0.1 d	3.3 ± 0.48 abc	1.4 a
A9E0460	73.0 ± 0.6 e–g	748.3 ± 16.5 bcd	4.4 ± 0.35 b	8.6 ± 2.13 d	2.4 bc
A9E1009	72.2 ± 10.2 d–g	681.3 ± 30.00 a–d	6.1 ± 0.60 c	7.4 ± 0.17 d	1.4 a
A9E0150	89.1 ± 11.7 g	795.7 ± 38.9 d	2.2 ± 0.25 a	7.4 ± 0.49 d	2.1 b
COE0309	69.9 ± 1.5 c–g	655.1 ± 34.30 a–d	2.2 ± 0.11 a	3.6 ± 0.05 abc	3.0 c

Results are mean values and standard deviations of three replicates. Different superscript letters denote significant differences at p ≤ 0.05.

), lowest in accession A9E0312 and highest in A9E0150 followed by A8E0139. Soluble sugars accounted for 8.4% of carbohydrates with an average content of 63.12 ± 9.62 mg·g⁻¹, being lowest in BOE0048 and Blyan and highest in A9E0150 and A9E1100. Free amino acids content was between 2.55.and 8.58 mg·g⁻¹, on average 4.43 ± 1.45 mg·g⁻¹, lowest in cv Blyan and B1E0485 and highest in A9E0460, A9E1009 and A9E0150. The content of phenols in the seeds of the populations and cultivars under study varied in the range of 1.3–17.2 mg per g FW, with accession A9E0312 having the lowest value and accession A8E0139 the highest value, followed by A9E1009 and A9E0460, which differed significantly from the rest of the accessions in amount of phenolic compounds. Trypsin-inhibiting activity in bean seeds varied from 1.2 to 3.1 units per mg FW; lowest in A9E0312, A9E1009, and A8E0139 and highest in A9E0588 and COE0309.

Despite the observed significant differences among accessions in seed biochemical composition, the nutritive quality in all studied common bean accessions corresponded to the range expected for this plant species.

4. Discussion

4.1. Morphological Trait Diversity

Evaluating common bean genotypes under natural conditions for different morphological, agronomical, and biochemical traits is of main importance in order to identify genotypes as a source for desirable characteristics for breeding or direct use for production. Many researchers have studied diversity in common bean germplasm collections with the main purpose of finding potential genotypes or lines with useful traits for bean improvement breeding programs [44,45,46,47,48]. Long et al. [49] studied the diversity index of quantitative and qualitative traits among 115 accessions from 27 counties in Chongqing province, China and reported the highest values of the diversity index of seed-related traits. Jan et al. [50] characterized morphologically a set of 109 accessions from the Western Himalayas and reported wide variation for plants, leaves, flowers, pods, and seeds, and at the same time the authors analyzed seeds from 60 selected genotypes for micronutrient and macronutrient content and found substantial variation of these nutrients. The variation of plant height and the number of primary branches per plant were determined mostly by their inherent variability [51]. Our results showed variation with medium CVs (%) on plant height, weight of plant, weight of 100 seeds, and harvest index (10–20%), and the highest variation was found in number of pods per plant, weight of pods per plant, and weight of seeds per plant (>20%). The wide diversity of 88 common bean landraces with Iberian origin was reported, wherein the genotypes were grouped based on their similarities of 17 quantitative characteristics with 13 clusters, which can be considered as a diverse genetic source to widen the genetic base of this crop in Europe [52]. Results from our experiment showed two main groups of accessions with two sub-clusters each and three accessions were situated individually aside the groups. The most differentiated characteristics were yield per plot, pod length, pod width, and 100 seed weight.

Evaluation of germplasm conserved ex situ in genebanks and in situ/on farm is a valuable source of information for better utilization of plant genetic resources in breeding improvement activities. Characterization of 100 genotypes of the active collection of the common bean of the Federal University of Viçosa based on 10–20 morphological descriptors classified them in groups of genetic similarities, and at the same time the authors supposed that traits such as seed gloss, seed uniformity, and growth habit may be important for genetic improvement and consumer preferences, but they are less relevant to the morphological characteristics [53]. The collection of 121 diverse common bean accessions from the University of Guelph was genotyped, confirming the association between yield and number of yield-related traits, including seed weight [54]. Brazilian researchers found significant differences between 67 accessions for 12 quantitative traits, where the coefficient of variation (CV%) ranged between 3.69 to 17.48%, as 10 descriptors reported a low CV% (less than 10%) and only two descriptors had medium CVs (10 to 20%), regarding number of pods (NPP) and number of seeds per pod (NSP). The authors supposed that these results could be due to the fact that the plant trial was conducted in a greenhouse [55]. In our study, the variation of number of pods per plant was one of the highest among all studied quantitative morphological traits, with 25.9%, while the number of seeds per pod showed a low variation, presenting with 9.0% CV.

4.2. Seed Biochemical Composition

Grain legumes in general, and common bean in particular, are valued as protein-rich, healthy functional food having a low glycemic index [19,56,57,58]. The biochemical analysis of bean seed composition gave percentages of crude protein, lipids, carbohydrates, ash, and fiber, as well as energy values, similar to those reported by other authors for Phaseolus vulgaris [59,60,61], along with significant differences among the compared genotypes. Seeds of the two varieties included in our study differed in proximate composition—cv Ustrem was lower in carbohydrates and starch content but with more lipids, whereas cv Blyan was higher in carbohydrates and lower in proteins and lipids; however, both cultivars were rich in fibers and low in tannins (Table 11). Among accessions, contrasting genotypes could be discerned regarding the main seed components; for example, with higher carbohydrate but lower protein content (Blyan, A8E0139, and COE0309) and the opposite—high in protein content but lower in carbohydrates (A9E0312, A9E0460, and A9E0150). A negative correlation between crude protein and carbohydrate/starch content in common bean seeds has been reported [62,63]. Contrasting accessions could be also found for the amount of oil in seeds (Ustrem, A9E0430, and A9E0588—high, whereas A9E1009 and A9E0150—low), for ash content (A9E0460 and A9E0150—high, A9E1100 and Blyan—low), and for tannins (particularly high in A9E0460 and A9E1100, low in COE0309, Blyan, and A9E0312). Similar high variability has been found in bean seed biochemical composition by other authors [19,58,64]. Combined effects of environment and genotype on nutritional traits variation are established [58,65], protein content being of intermediate heritability, while oil content has very low heritability [58]. Landraces, compared to cultivars selected mainly for higher yield, possess superior nutritional value [23,66].

Bean seeds also contain varying amounts of bioactive substances, such as phenols, tannins, and protease inhibitors [67,68,69,70]. A strong correlation has been documented between phenolic compounds and antioxidant activity in bean seeds [59,60,70]. Common bean varieties with high antioxidant potential could be particularly useful for human health improvement [68]. The significant anti-inflammatory effects of some phenol compounds are due to their antioxidant and free radical scavenging properties. Moreover, it has been shown that some phenolic compounds produced by common bean exert antifungal effects, decreasing aflatoxin production by Aspergillus flavus [71]. In our study, three bean accessions (A8E0139, A9E0460, and A9E1009) had significantly higher amount of phenols in seeds (Table 12). Tannins have protein-precipitating properties, which diminish the nutritive value of foods; considerable variation has been found comparing bioactive substances in common bean genotypes [72]. Protease inhibitors could also have antinutrient properties because of digestive enzymes’ inhibition, thus impeding the nutrient absorption in the gastrointestinal tract [67]. In our study, trypsin-inhibiting activity in common bean seeds was lowest in accessions A9E0312, A8E0139, and A9E1009 and highest in A9E0588 and COE0309. The knowledge on the biochemical composition of local populations can be used in breeding programs aiming at better common bean seed qualities [73].

5. Conclusions

The quantitative and qualitative traits of plants, inflorescence, pods, and seeds were used to assess the differences among the accessions. With the presence of global climate change and high daily temperatures starting sometimes from May and June, the flowering and pod formation phases are of special importance. Based on the number of days to maturity, the accessions were categorized as early and late maturing. The variability of each morphological trait was shown by the coefficient of variation (CV%). Accession BOE0048 showed the highest value of weight of plant, weight of pods per plant, and number of seeds per pod. The maximum number of pods per plant was registered by accession B1E0485, followed by accession BOE0048. The biggest seeds belonged to genotype A9E0588 with 36.8 g W100S. Harvest index of the accessions A9E0588, A9E0543, and A8E0139 exceeded that of the cultivar “Ustrem”, used as a standard. Strong significant and positive correlations were revealed between the weight of pods and number of pods per plant, weight of seeds, and number of pods per plant. The dendrogram built based on genetic similarities among all studied accessions grouped them in two major clusters divided in four sub-clusters with different numbers of genotypes. The most distinctive trait for accessions A9E0543, A8E0139, and BOE0048 grouped in the first sub-cluster was higher yield per plot.