Bruchus rufimanus, a univoltine seed weevil (bruchid), can cause severe seed yield and quality losses in faba beans restricting crop profitability and expansion. Chemical insecticide applications have been reported of low effectiveness and thus, growing genotypes tolerant to bruchid has been suggested as an alternative. Ten faba bean (Viciafaba L.) accessions belonging to three varieties (var. major (seven accessions), var. minor (two accessions), var. equina (one accession)) were tested under field conditions for two growing seasons. Agronomic and seed traits were determined in an attempt to associate any tolerance to bruchid with easily-assessible, highly-heritable characters in order to be used as indirect selection criteria. The genotypes varied in bruchid tolerance (percentage of bruchid emergence holes (BD), percentage of endoparasitoid (Triaspis thoracica) emergence holes and bruchid infestation level (BI = BD + PD)), agronomic traits and seed properties. The dark-colored, small- and medium-seeded accessions (var. minor and var. equina), commonly used for feed, had the lowest BI (4.21–8.17%) ranging below the limit of 10% set as the highest acceptable for using faba beans as feed. Large-seeded accessions (var. major), which had light-colored seed coat (testa) with yellow hue, showed BI from 11.80% up to 24.54%, far-above the limit of 3% for seeds used as food. Apart from the seed size and color, susceptible genotypes had more seeds per pod, less pods and less branches per plant, possibly offering an easy access to females for laying more eggs on the limited number of pods, albeit the more space and food (higher protein content per seed) they offer to the developing larvae. Phenols and tannins in seeds, a putative chemical defense mechanism against bruchid, did not associate with the percentage of bruchid- or endoparasitoid-damaged seeds. Concluding, certain plant architectural traits and seed properties related to bruchid infestation in faba beans can be used as useful tools to select tolerant genotypes.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Introduction
Faba bean (Vicia sativa L.), one of the founder crops of agriculture, draws interest the last decades as an indigenous substitution for soybean food and feed protein (Duc 1997; Duc et al. 2010; Ruisi et al. 2017). From an agronomic perspective, faba bean is incorporated mainly into cereal-based cropping systems providing agronomic and environmental benefits (Jensen et al. 2010; Ruisi et al. 2017).
Insect pests and especially seed weevil or bruchid (Coleoptera: Chrysomelidae, Bruchinae) can be a serious restrictive factor of faba bean grain yield and cropping expansion (Segers et al. 2021; Huber et al. 2023). Bruchids are monophagous or oligophagous species and faba bean is primarily infested by the cosmopolitan Bruchus rufimanus (Boheman, 1833), but it also hosts sporadically B. atomarius (Linnaeus, 1761) and B. dentipes (Herbst, 1783) (Kergoat et al. 2007; Titouhi et al. 2015).
Anzeige
Bruchus rufimanus is a univoltine beetle which terminates its reproductive diapause and exits overwintering refuges (e.g. under tree bark, soil cavities) at temperatures > 15 oC and photoperiod 16/8h (day/night) or even 12/12 h in the Mediterranean region (Medjdoub-Bensaad et al. 2018; Segers et al. 2021, 2022). Adults resume on flowering faba bean plants to feed on nectar and pollen; pollen consumption is necessary for the reproductive maturity of the females (Hamidi et al. 2021).
Females lay eggs at large numbers (up to 50–100 per female) on green pods starting as soon as the first pods are formed and continuing as long as pods remain green, no matter the growth stage; oviposition is heavier on lower pods (Hamani and Medjdoub-Bensaad 2015; Hamidi et al. 2021; Gailis et al. 2022; Huber et al. 2023). Oviposition and thus infestation levels are affected by weather conditions. Rainfall and high humidity deter oviposition or wash the eggs away, while high temperatures cause egg desiccation (Hamani and Medjdoub-Bensaad 2015; Carrillo-Perdomo et al. 2019; Segers et al. 2021). After hatching, young larvae enter the developing seeds where they consummate four instar stages consuming part of the cotyledons. After the end of pupation, before or after harvest, the adults emerge from the fully-matured seeds through rounded holes or remain in the seeds (Segers et al. 2021; Dell’Aglio and Tayeh 2023; Huber et al. 2023). It is noteworthy that the insect developmental stages are characterized by high rates of mortality (Seidenglanz and Huňady 2016).
The economic damage caused by the emerging bruchid adults can be high regarding either seed weight losses or quality degradation (Segers et al. 2021). Depending on the variety, seed weight losses, due to cotyledon consumption by larvae, have been estimated at 5.0–9.4% and the percentage of damaged seeds could be up to 37% (Titouhi et al. 2015; Seidenglanz and Huňady 2016; Segers et al. 2022). The large-seeded varieties and the lower pods show the highest damages (Kaniuczak 2004; Szafirowska 2012; Titouhi et al. 2015). On the other hand, seed quality standards dictate that infested seeds should not exceed 3% and 10% when faba beans are directed to human consumption and animal feeding, respectively (Segers et al. 2021; Dell’Aglio and Tayeh 2023).
Quality degradation regards the reduction of nutritional value due to the accumulation of larvae faeces and low marketability of the seeds owing to the aesthetic alteration by the adults’ emergence holes and the seed coat (testa) darkening (Tsialtas et al. 2019; Segers et al. 2021). Low quality of infested seeds has also to do with decreased germination rates because adults’ emergence holes make them vulnerable to soil pests and diseases (Kaniuczak 2004; Huber et al. 2023). Large seeds may bear multiple exit holes leading to proportionally higher reductions of germination rate (Khelfane-Goucem and Medjdoub-Bensaad 2016).
Anzeige
Interestingly, infestation by bruchids may have a positive impact on faba bean yield components in cases of absence of pollinators, but more work is needed to define whether the marketability of the seeds is negatively affected or not (Riggi et al. 2022).
What makes B. rufimanus a serious pest to faba beans, apart from its cosmopolitism and the high infestation levels, is the low effectiveness of the chemical sprayings to control the beetle. Limited options of pyrethroid insecticides, dense canopies in faba bean crops hindering spraying penetration in depth and limiting efficiency, high female fecundity and oviposition for long period and high mobility of the adults contribute to low efficiency of the chemical applications (Kaniuczak 2004; Segers et al. 2021; Huber et al. 2023).
Having the abovementioned in mind, selecting/breeding tolerant varieties comes up as a rational and possibly affordable option against bruchid infestation in faba beans (Bachmann et al. 2020). Previous works have revealed varietal differentiation in bruchid infestation levels, with small-seeded varieties to show less damaged seeds (Kaniuczak 2004; Szafirowska 2012; Titouhi et al. 2015; Seidenglanz and Huňady 2016; Carrillo-Perdomo et al. 2019). Seed size is a key trait in the classification of faba bean varieties and it is associated with their uses. Thus, Duc (1997) categorized faba beans to: a) large-seeded V. faba var. major with 1000-seed weight (TSW) higher than 1000 g, b) small-seeded V. faba var. minor (TSW < 500 g) and c) medium-seeded V. faba var. equina (500 g < TSW < 1000 g).
Dell’Aglio and Tayeh (2023) found that faba bean genotypes differed in flower attractiveness of feeding female adults, but this preference did not coincide with the preference for oviposition. Instead, females preferentially layed eggs on pods with more seeds and genotypes with more pods. Bruchids prefer oviposition on large seeds in an effort to secure more food and space for the developing larvae inside the swelling seeds (Sadakiyo and Ishihara 2012). Moreover, Szafirowska (2012) related faba bean plant architecture with cultivar susceptibility to bruchids. Agronomic traits like the number of branches, pod characteristics and plant height, along with seed size, are major, highly-heritable grouping traits used for the classification of faba bean genotypes (Terzopoulos et al. 2003; Toker 2004). Thus, any possible association of these easily assessible characters with bruchid infestation could be a useful tool for the selection of tolerant genotypes.
The research for tolerance to bruchids should take into consideration the genotypic differentiation in attracting the natural enemy of faba bean bruchid, the endoparasitoid Triaspisthoracica (Curtis, 1880) (Hymenoptera: Braconidae). The larvae of this wasp feed on the developing bruchid larvae increasing their mortality and contributing to bruchid control. Genotypes of V. faba and other Vicia species were found to differentiate in the percentage of bruchid larvae parasitism by T. thoracica (Seidenglanz and Huňady 2016; Tsialtas et al. 2018; Boulata et al. 2022).
Last, substances contained on testa or endosperm could be a kind of “chemical defense” against the bruchid larvae entering into or developing in the cotyledons. Phenolic compounds like tannins have been ascribed with such a defensive role (Barbehenn and Constabel 2011). In peas (Pisumsativum L.), high levels of phenolics were associated with darker seeds and colorful flowers (Troszyńska and Ciska 2002; Konstantopoulos et al. 2023), while in vetches (Vicia spp.), seeds with darker testa had higher levels of phenols and tannins and lower numbers of bruchid-damaged seeds (Tsialtas et al. 2018; Boulata et al. 2022). In this line, Szafirowska (2012) partially attributed the tolerance of faba bean cv. Makler to phenolic compounds (anthocyanins) of its violet-coloured testa.
The aims of the present work were to i) assess the bruchid infestation levels under Mediterranean conditions and ii) identify any agronomic traits that could be used as indirect criteria for selecting faba bean genotypes tolerant to bruchids.
Materials and Methods
Genetic Material and Experimental Set up
The experiment was conducted in the farm (40ο32’1’’Ν, 22ο59’2’’Ε, 0 m a.s. l.) of Aristotle University of Thessaloniki (AUTh), Thermi, Greece during 2014–2015 (2015) and 2015–2016 (2016) growing seasons (December to May). The soil was an alkaline clay (pH > 8.0) with organic matter 1.29%, NO3-N 37.5 mg kg−1, P‑Olsen 8.8 mg kg−1, and potassium (K) 97 mg kg−1 at 0–30 cm depth. The 2015 growing season was wetter and cooler than the 2016 growing season (428 mm vs. 253 mm and 10.8 °C vs. 11.9 °C). Focusing on spring (March to May), the 2015 growing season was drier (134 mm vs. 173 mm) and cooler (14.2 °C vs. 15.4 °C) compared to the 2016 growing season (Fig. 1).
Fig. 1
Mean monthly temperature and monthly precipitation during the growing season (December to May)
×
The genetic material tested were 10 V. faba (Vf) advanced lines derived from local material and preserved in the Laboratory of Agronomy, Faculty of Agriculture, AUTh. The lines belong to the varieties major (seven accessions), minor (two accessions) and equina (one accession). Table 1 presents some seed and pod characteristics distinguishing of the 10 Vf accessions.
Table 1
Variety, seed/pod characteristics and days after seeding of the first open flower stage (± standard error) of the 10 Viciafaba (Vf) accessions used in the study
Accession
Variety
Seed/pod characteristics
Anthesis
Vf1
Major
Very large seed, beige-brown testa, black hilum
124 (6)
Vf2
Major
Very large seed, cream-yellow testa, black hilum
119 (1)
Vf3
Major
Large seed, beige-brown testa, black hilum
119 (1)
Vf4
Major
Very large seed, beige-brown testa, white hilum
119 (1)
Vf5
Major
Large seed, cream-yellow testa, white hilum
118 (0)
Vf6
Major
Large seed, beige-brown testa, white hilum
119 (1)
Vf7
Major
Large seed, cream-yellow testa, black hilum
125 (6)
Vf8
Minor
Small seed, black testa, black hilum
136 (0)
Vf9
Minor
Small seed, black testa, black hilum, glabrous, thin pod wall
136 (0)
Vf10
Equina
Medium seed, black-brown testa, black hilum
130 (6)
The experimental layout was a Randomized Complete Block (RCB) design with three replications. Each plot (1 m2 area) consisted of four, 1‑m long rows, at 0.25 m separation. The seeding, at a rate of 24 seeds m−2, was done by hand on 5 December 2014 and 1 December 2015. The blocks were separated by a 1.5 m buffer zone. No fertilization, irrigation or chemical sprayings were applied. The weeds were removed by hand, when necessary.
Measurement of Agronomic Traits
In the 2015 growing season, the days after seeding of the first open flower stage (203 (1) stage, Knott 1990) were recorded per plot and means and standard errors were calculated per accession (Table 1).
At harvest maturity (410 stage, Knott 1990), five plants per plot were randomly selected from the two inner rows. Plant height (PH, cm), from the ground to the uppermost end of the main stem, was measured with a common ruler. Also, the number of branches (BpP) from basal nodes and the fertile pods (PpP, pods with at least one fully-developed seed) were counted. Then, the fertile pods were sorted out and 15 randomly selected pods per plot were used to measure the number of seeds per pod (SpP). All the fertile pods of the five plants selected per plot were threshed by hand and seeds were weighed. After measuring the seed moisture content using a Dramiński GMM grain moisture meter (Dramiński SA, Olsztyn, Poland), the seed yield per plant (SY, g plant−1) was calculated after normalizing the seed weight at 10% seed moisture. The vegetative parts of the five plants per plot, along with the infertile pods and the pod walls of the fertile pods, were dried at 75 oC till constant weight and used to calculate harvest index (HI, seed weight to total plant weight) on a dry weight basis.
Assessment of Bruchid and Endoparasitoid Infection Level
Fifty randomly selected seeds per plot were placed in plastic boxes (350 mL) and kept at room temperature (25/20 °C day/night temperature) for over three months until the emergence of the adult bruchids and endoparasitoids consummated. The adult bruchids were identified as B. rufimanus and the endoparasitoid wasp as T. thoracica (Kergoat et al. 2007; Bellifa and Chapelin-Viscardi 2021). Due to multiple and promiscuous (bruchid and endoparasitoid) emergence holes on seeds (Fig. 1S), the infestation levels were assessed by counting emergence holes and windows (unopened translucent larval hole, Huber et al. (2023)) of bruchids (BD) and parasitoids (PD) on the 50 seeds per plot and expressing the counts in percentage (%). The bruchid infestation level (BI) was calculated as the sum of BD and PD. The classification of the emergence holes was done by examining each seed by the naked eye or under magnifying lens (5×), when it was not easy to classify the holes (Tsialtas et al. 2018).
Testa Color, 100-seed Weight, Protein and Phenols Measurement
Fifty intact and healthy seeds per plot were used for the assessment of testa color using a CR-400/410 chroma meter (Konica Minolta, Kyoto, Japan). As many seeds as possible were placed in a petri dish and two measurements were taken per plot on randomly selected seeds. Chroma meter assessments were based on the International Commission on Illumination (CIE) color solid scale (L*, a*, b*): L* represents lightness (from black = 0 to white = 100), a* is for greenness and redness (red = positive value and green = negative value), and b* corresponds to blueness and yellowness (yellow = positive value and blue = negative value). The instrument was calibrated after five measurements using a white tile (L* = 96.9, a* = −0.04, b = 1.84).
The seeds used for chroma measurements were dried at 75 oC till constant weight and dry weights were used for 100-seed weight (HSW) calculations. Then, the dried seeds were ground to fine powder using an ultra-centrifugal mill ZM-1000 (Retsch GmbH, Haan, Germany) equipped with a 0.5 mm sieve. A subsample of ca. 1.0 mg per plot was analysed for nitrogen concentration (%N) using an elemental analyser (EuroEA 300, EuroVector SpA, Milan, Italy). Seed protein concentration (Prot) was calculated as the product %N × 6.25. Protein concentration was multiplied by the weight per seed (in mg) to estimate the seed protein content (SPC, mg seed−1).
For the extraction of phenolic compounds, a subsample of 0.2 g of ground material per plot was extracted with 4 mL 70% (v v−1) aqueous acetone using ultrasonic treatment for 15 min at room temperature. After centrifugation at 2200 rpm for 10 min, the supernatant was collected and the extraction was repeated one more time in a similar way. The supernatants were combined, vortexed and stored at −20 oC until chemical analysis. For total phenolic content (TPhe) measurements, 200 μL aliquots of phenolic extracts were reacted with 800 μL of Folin-Ciocalteu reagent (1:10, v v−1) and after 2 min, 2 mL of 75 g L−1 aqueous sodium carbonate solution was added and the volume was adjusted to 10 mL with distilled water (Singleton et al. 1999). After 1 h incubation, the absorption at 725 nm was recorded and the results were expressed as mg of gallic acid equivalents per g sample (mg GAE g−1). The difference in the total phenolic content values before and after adding polyvinylpolypyrrolidone (PVPP) represented the total tannin contents. Specifically, two aliquots of about 400 μL of the phenolic extract prepared as described above were placed into 2 mL tube. The PVPP (40 mg, 100 mg mL−1) was added to the former and the mixture stirred for 10 sec and incubated at 4 oC for 20 min. After centrifugation at 10,000 rpm (4 oC, 10 min), non-tannin phenolics (supernatant) determined as above by Folin-Ciocalteu method (Makkar et al. 1993). Total tannins (TTan) were calculated by subtracting non-tannin phenolics from total phenolics and the results were expressed as mg GAE g−1.
Statistical Analyses
Data were subjected to over-year analysis of variance (ANOVA) as Randomised Complete Block (RCB) design with faba bean accessions as main factor and three replications. The percentages of bruchid (BD) and parasitoid (PD) holes on seeds, along with their sum (BI), were subjected to arcsine transformation before the statistical analyses. Analyses were conducted using the statistical software M‑STAT (MSTAT‑C, version 1.41, Crop and Soil Sciences Department, Michigan State University, USA) and means were compared by the least significant difference (LSD) test at P < 0.05.
Pearson correlations between the traits were determined for the accessions over the two growing seasons, using the IBM SPSS software version 20 (IBM, Armonk, NY, USA). Principal component analysis (PCA) was performed. The first two principal components, PC1 and PC2, were derived from eigenvalue decomposition of the correlation matrix of the variables. Based on the eigenvectors values, those explaining most of the variation were used to print the respective scatterplots.
Results
Agronomic Traits
With the exception of plant height (PH; 74.83–87.73 cm), Vf accessions differed significantly in all the agronomic traits (Table 2).
Table 2
Agronomic traits (branches per plant (BpP), plant height (PH), pods per plant (PpP), seeds per pod (SpP), 100-seed weight (HSW), seed yield per plant (SY), harvest index (HI)) in 10 Viciafaba (Vf) accessions tested over two growing seasons. For the same column and factor (accessions, growing seasons), means labeled with the same letter did not differ significantly at 0.05 level of probability as indicated by the LSD test
BpP
PH
(cm)
PpP
SpP
HSW
(g)
SY
(g plant−1)
HI
Accessions
Vf1
2.13d
80.27a
5.2c
4.1bc
163.6ab
30.8bcd
0.639ab
Vf2
2.60cd
77.27a
7.3c
4.4ab
164.6ab
43.4a
0.676a
Vf3
2.23cd
75.97a
4.9c
4.6a
142.9cd
32.4bc
0.678a
Vf4
2.90bc
87.73a
6.4c
3.8 cd
168.4a
38.9ab
0.581b
Vf5
2.37cd
80.65a
6.5c
4.1bc
144.6bc
36.9abc
0.601b
Vf6
2.37cd
84.60a
8.4c
3.3de
122.6d
28.0cde
0.605b
Vf7
2.60cd
74.83a
5.9c
4.5ab
147.2bc
42.2a
0.455c
Vf8
3.53ab
77.70a
22.5a
3.1e
38.3f
22.1de
0.489c
Vf9
3.60a
77.40a
21.6a
3.1e
35.5f
20.3e
0.469c
Vf10
2.73cd
79.40a
12.6b
3.4de
85.9e
36.7abc
0.620ab
Growing seasons
2015
2.67a
81.22a
10.2a
3.6b
113.7b
30.5b
0.608a
2016
2.75a
77.94a
10.1a
4.0a
129.0a
35.8a
0.555b
CV (%)a
21.43
11.09
33.06
10.23
14.51
23.60
9.94
a Coefficient of variation
The two accessions of var. minor (Vf8 and Vf9) had the highest number of branches per plant (BpP; 3.53 and 3.60, respectively), while the Vf1 (var. major) had the lowest (2.13). The highest number of pods per plant (PpP) was also recorded in Vf8 and Vf9 accessions (21.6 and 22.5), followed by the var. equina (Vf10, 12.6). The accessions of var. major had the lowest PpP (4.9–8.4) with no significant differences among them. The number of seeds per pod (SpP) was lowest in the two accessions of var. minor (Vf8 and Vf9, 3.1), the accession of var. equina (Vf10, 3.4) and one accession of var. major (Vf6, 3.3). On the contrary, three accessions of var. major (Vf2, Vf3, Vf7) showed the highest SpP (4.4–4.6). The accessions of var. minor (Vf8, Vf9) were the smallest-seeded ones (HSW = 35.5–38.3 g), the Vf10 (var. equina) had moderate seed weight (HSW = 85.9 g), while three accessions of var. major (Vf1, Vf2, Vf4) showed the highest HSW (163.6–168.4 g). Seed yield per plant (SY) was lowest in the two accessions of var. minor (Vf8: 22.1 g plant−1 and Vf9: 20.3 g plant−1) and one accession of var. major (Vf6: 28.0 g plant−1), while four accessions of var. major (Vf2, Vf4, Vf5 and Vf7) and the accession of var. equina (Vf10) had the highest SY (36.7–43.4 g plant−1). The accessions of var. minor (Vf8 and Vf9) along with Vf7 (var. major) showed the lowest harvest index (HI = 0.455–0.489), whereas three accessions of var. major (Vf1, Vf2, Vf3) along with Vf10 (var. equina) had the highest values (0.620–0.678).
Regarding growing seasons, significant differences were found for SpP, HSW, SY and HI. With the exception of HI, higher values were recorded in the 2016 growing season compared to the 2015 growing season (Table 2).
Testa Color and Seed Protein and Phenolics
Testa color parameters L* and b* were lowest in var. minor (Vf8 and Vf9) and var. equina (Vf10) accessions, while no significant difference was evident among the accessions of var. major (Table 3). For a*, the lowest values were measured in the accessions of var. minor (Vf8: −0.54 and Vf9: −0.55) and var. equina (Vf10: 1.47) along with the Vf3 accession (1.30). Instead, the accessions Vf2, Vf4 and Vf6 (var. major) showed the highest values (3.28–4.93).
Table 3
Testa color parameters (L*, a*, b*), seed protein concentration (Prot), seed protein content (SPC), total phenolic compounds (TPhe) and total tannins (TTan) in 10 Viciafaba (Vf) accessions tested over two growing seasons. For the same column and factor (accessions, growing seasons), means labeled with the same letter did not differ significantly at 0.05 level of probability as indicated by the LSD test
L*
a*
b*
Prot
(%)
SPC
(mg seed−1)
TPhe
(mg GAE g−1)
TTan
(mg GAE g−1)
Accessions
Vf1
42.13ab
2.54b
20.97ab
28.91bcd
424.9a
5.42bc
3.26a
Vf2
40.47b
3.44ab
18.89b
26.99def
402.4a
6.12abc
3.93a
Vf3
46.90a
1.30bc
22.80a
25.32f
337.6bc
6.93a
3.93a
Vf4
41.48b
3.28ab
20.65ab
26.16ef
402.6a
6.76a
4.00a
Vf5
44.25ab
2.73b
21.77a
30.77ab
401.5a
6.36abc
3.95a
Vf6
41.73b
4.93a
20.09ab
25.96f
287.2c
7.20a
4.36a
Vf7
43.34ab
1.73b
21.62ab
28.03cde
378.7ab
6.72a
4.17a
Vf8
18.09c
−0.54c
1.51c
30.38ab
104.7e
6.97a
4.58a
Vf9
17.85c
−0.55c
1.54c
31.09a
99.4e
6.62ab
4.07a
Vf10
19.34c
1.47bc
3.22c
29.29abc
225.8d
5.26c
3.12a
Growing seasons
2015
42.07a
1.22a
16.96a
27.21a
274.0b
6.60a
3.92a
2016
29.04b
2.85a
13.65b
29.37a
339.0a
6.27a
3.95a
CV (%)a
11.70
91.83
15.41
6.19
16.59
16.16
20.27
a Coefficient of variation
Seed protein concentration (Prot) was highest in Vf9 (31.09%), Vf5 (30.7%), Vf8 (30.38%) and Vf10 (29.29%), while four accessions of var. major (Vf2, Vf3, Vf4 and Vf6) had the lowest Prot (25.32–26.99%). On the contrary, seed protein content (SPC) was highest in accessions of var. major (Vf1, Vf2, Vf4, Vf5, Vf7), ranging from 378.7 mg seed−1 (Vf7) up to 424.9 mg seed−1 (Vf1). The two accessions of var. minor had the lowest SPC values (Vf9: 99.4 mg seed−1, Vf8: 104.7 mg seed−1).
Total phenolics (TPhe) were lowest in the accession of var. equina (Vf10: 5.26 mg GAE g−1), followed by the Vf1 accession (5.42 mg GAE g−1), while the other accessions did not differ significantly. As of total tannins (3.12–4.58 mg GAE g−1), no significant difference was found among the accessions (Table 3).
Regarding growing seasons, L* (42.07 vs. 29.04) and b* values (16.96 vs. 13.65) were higher in the 2015 growing season compared to 2016, but for SPC (274.0 mg seed−1 vs. 339.0 mg seed−1), the opposite was evident (Table 3).
Bruchid and Endoparasitoid Damages on Seeds
While growing seasons had no significant effect on bruchid infestation level, accessions differed significantly (Table 4).
Table 4
Percentages (%) of bruchid holes (BD), parasitoid holes (PD) and bruchid-infestation level (BI = BD + PD)] in 10 Viciafaba (Vf) accessions tested over two growing seasons. For the same column and factor (accessions, growth seasons), means labeled with the same letter did not differ significantly at 0.05 level of probability as indicated by the LSD test
BD
PD
(%)
BI
Accessions
Vf1
16.55a
7.99abc
24.54a
Vf2
10.00bc
8.67abc
18.67ab
Vf3
14.00ab
9.67ab
23.67a
Vf4
13.33ab
9.33ab
22.67a
Vf5
9.16bcd
13.89a
23.05a
Vf6
8.78bcd
3.02de
11.80bc
Vf7
17.00a
6.33bcd
23.33a
Vf8
3.50de
1.50e
5.00c
Vf9
1.38e
2.83de
4.21c
Vf10
4.50cde
3.67cde
8.17bc
Growing seasons
2015
6.13a
6.47a
12.60a
2016
13.51a
6.91a
20.42a
CV (%)a
38.69
41.78
32.22
a Coefficient of variation
Small-seeded accessions Vf8 and Vf9 (var. minor) and the medium-seeded Vf10 (var. equina) had the lowest percentages of bruchid holes (BD), ranging from 1.38% (Vf9) up to 4.50% (Vf10). On the contrary, four accessions of var. major (Vf1, Vf3, Vf4, Vf7) showed the highest BD (13.33–17.00%). The small- and medium-seeded accessions also showed the lowest percentages of endoparasitoid holes (PD), from 1.50 to 3.67%. The Vf5 (var. major) had the highest PD (13.89%) followed by Vf1, Vf2, Vf3 and Vf4 (7.99–9.67%). Regarding bruchid infestation levels (BI = BD + PD), the two accessions of var. minor (Vf8, Vf9), the accession of var. equina (Vf10) and one accession of var. major (Vf6) showed the lowest BI values (4.21–11.80%). The other six accessions of var. major had the highest BI (18.67–24.54%) without significant difference among them (Table 4).
Relationships Between the Determined Traits
The PCA for the determined traits is presented in Fig. 2, while Table 1S gives the correlation coefficients (r) and the significance levels.
Fig. 2
Principal component analysis (PCA) for the determined traits in 10 V. faba accessions. BpP branches per plant, PH plant height, PpP pods per plant, SpP seeds per pod, HSW 100-seed weight, SY seed yield per plant, HI harvest index, L*, a*, b* testa color parameters, Prot seed protein concentration, SPC seed protein content, TPhe total phenolic compounds, TTan total tannins, BD percentage of bruchid emergence holes, PD percentage of parasitoid emergence holes, BI percentage of bruchid infestation (BD + PD)
×
Percentage of bruchid holes on seeds (BD) was negatively correlated with BpP (r = −0.73, P < 0.05) and PpP (r = −0.87, P < 0.01), but positively with SpP (r = 0.82, P < 0.01), HSW (r = 0.87, P < 0.01), L* (r = 0.86, P < 0.01), b* (r = 0.87, P < 0.01) and SPC (r = 0.86, P < 0.01). The same correlations were also significant, but stronger, for BI, which was additionally correlated with SY (r = 0.67, P < 0.05). On the contrary, the respective correlations for PD were weaker compared to those for BD. Moreover, PD was not correlated with BpP in contrast to BD and BI (Fig. 2, Table 1S).
Discussion
Selecting tolerant genotypes has been suggested as a putatively effective means to mitigate heavy damages by B. rufimanus since even chemical insecticide sprayings have been reported of low effectiveness (Kaniuczak 2004; Bachmann et al. 2020; Segers et al. 2021).
In the present study, there was significant genotypic variation among the faba bean accessions in bruchid and endoparasitoid emergence from seeds and finally in bruchid infestation level in accordance to previous works (Kaniuczak 2004; Seidenglanz and Huňady 2016; Dell’Aglio and Tayeh 2023). It is noteworthy that small- and medium-seeded accessions (var. minor and var. equina), commonly used for feed, showed infestation levels (4.21–8.17%) below 10%, the upper limit for accepting faba bean seeds as feed. However, the infestation levels (11.80–24.54%) of all the large-seeded accessions (var. major) were far-departed from the respective limit of 3% for faba beans used as food (Segers et al. 2021; Dell’Aglio and Tayeh 2023). In general, the tested faba bean accessions showed low to moderate levels of damages by bruchids compared to previous works, which reported percentages up to ca. 29% for var. minor and up to 37% for var. major (Kaniuczak 2004; Titouhi et al. 2015). It is interesting that an accession of var. major (Vf5) showed the highest percentage of seeds with parasitoid emergence holes (13.89%), but even higher levels (ca. 60%) have been found (Seidenglanz and Huňady 2016).
More seeds per pod and larger seed size (higher HSW) were related with higher seed yield per plant and were strongly related with increased susceptibility to bruchid and higher percentage of seeds with T. thoracica emergence holes. This comes to confirm Dell’Aglio and Tayeh (2023) who found that oviposition was higher in faba bean genotypes with more seeds per pod and pods per plant. Interestingly, oviposition preference did not coincide with a genotype’s flower attractiveness to females. The susceptibility of large-seeded faba beans compared to small-seeded genotypes has already been reported (Titouhi et al. 2015; Dell’Aglio and Tayeh 2023) and could be ascribed to the more space and food large seeds offer to developing bruchid larvae (Sadakiyo and Ishihara 2012). Supportive to this hypothesis is that bruchid infestation was strongly and positively associated with seed protein content (SPC), but seed protein concentration did not (Tsialtas et al. 2018, 2020). In contrast, Nikolova (2016) found that field pea genotypes with higher seed N and P concentrations were more susceptible to bruchids (B. pisorum (Linnaeus, 1758)). To the bruchid tolerance of small seeds, albeit not measured, possibly contributed the higher seed toughness which poses a mechanical impediment to the development of bruchid larvae (Fricke and Wright 2016; Tsialtas et al. 2018).
Confirming Szafirowska (2012), who stated that plant architecture may affect tolerance to bruchid infestation, susceptible accessions were those with less branches and pods per plant. A possible explanation is that the canopy of these genotypes is less complex, easily-accessible by females and offer oviposition opportunities thereby egg laying per pod is higher. Plant height, which did not differ among the 10 accessions, was not related to bruchid infestation in contrast to Dell’Aglio and Tayeh (2023), who found that shorter plants attracted more bruchids.
Seeds with light-colored testa (higher L*) and yellow hue (higher b*) were the most damaged by bruchid and its endoparasitoid. Szafirowska (2012) partially ascribed the tolerance of cv. Makler to the violet coloration of the seeds, which is indicative of high phenolic (anthocyanins) concentration. The association between dark testa color, high phenolics concentration and tolerance to bruchids has already been reported for other legume species, Vicia included (Teshome et al. 2015; Boulata et al. 2022; Rossi and Rodrigues 2023). It has been reported that the toxicity of organic compounds like phenolics and inorganic elements (e.g. Fe) was more severe for parasitoids than the bruchids (Tsialtas et al. 2020; Rossi and Rodrigues 2023). However, in the present study, testa color parameters (L*, a*, b*) did not relate with total phenols and tannins in seeds and possibly this mechanism of tolerance was not in works. Among four phenotypes of a pea landrace, Konstantopoulos et al. (2023) associated black hilum color with tolerance to B. pisorum. In faba beans, white hilum color is indicative of low levels of the putatively toxic alkaloid glycosides vicine and convicine (v-c). In our work, among the large-seeded, susceptible to bruchid accessions of var. major, hilum color seemed unrelated to bruchid infestation. Possibly, this had to do with that white hilum is not an absolute marker of low v‑c and black-hilumed faba beans may also contain low v‑c (Khazaei et al. 2019). Since we did not determine v‑c concentration in seeds, the possible association between hilum color and tolerance to bruchid should be an issue for study in the future.
Weather conditions (rainfall, humidity, temperatures) during the growing season impact both seed yield and bruchid infestation in faba beans through effects on oviposition volume, egg wash out and desiccation (Carillo-Perdomo et al. 2019; Segers et al. 2022). Seed yield was higher under warmer and wetter conditions in spring (March to May 2016), but bruchid infestation was not significantly affected by the growing season. In the perennial Lathyrus vernus, Östergård et al. (2009) found that infestation by B. atomarius was high in seasons with high seed production, a case not confirmed in the present study.
Conclusions
Faba bean accessions differed significantly in tolerance to bruchid, agronomic traits and seed properties. Specific agronomic and seed traits could be used as indirect criteria in selecting genotypes tolerant to bruchids. Genotypes with large seeds having light testa color and yellow hue, high number of seeds per pod, less dense canopies with less pods per plant were more susceptible to bruchid infestation. This preference of the insects has apparently to do with the offering of more space and food (higher protein content per seed) to the developing larvae of both bruchids and endoparasitoids. Seed phenolic concentration was not related to bruchid infestation level. Small- and medium-seeded accessions (var. minor and equina), commonly used for feed, showed damage levels < 10%, which is the upper limit for the use of faba bean seeds as feed.
Acknowledgements
We thank Mrs. A. Karypidou and Mrs. E. Fakoudi for their help during the course of field experimentation and Dr. D. Baxevanos, Hellenic Agricultural Organization-‘Demeter’, Institute of Industrial and Forage Crops, Larissa, Greece, for his help with the statistics.
Conflict of interest
I.T. Tsialtas and M. Irakli declare that they have no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
