Euphytica (2014) 198:243–254 DOI 10.1007/s10681-014-1097-1
Diversity evaluation of morphological traits and allicin content in garlic (Allium sativum L.) from China Haiping Wang • Xixiang Li • Di Shen Yang Oiu • Jiangping Song
•
Received: 6 September 2013 / Accepted: 5 March 2014 / Published online: 20 March 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract China has a long history in garlic cultivation and is the biggest country of garlic production in the world. 375 accessions of garlic from 23 provinces and areas in China has been collected and preserved in national germplasm repository for vegetatively propagated vegetables in Beijing since 2002. However, the genetic background and diversity of garlic from China has not been well characterized. In this study, 212 of 375 accessions of garlic were evaluated based on 29 morphological traits and allicin content. Cluster, principal compound, principal ordinates, Shannon diversity index and Pearson correlation analysis were used. The results showed that the garlic clones from China had a widely diversity among all traits. Principal component analysis showed the cumulative proportion of the first eight components explained 71.35 % of total morphological variation in all accessions. Germplasm cluster
H. Wang (&) X. Li (&) D. Shen Y. Oiu J. Song Institute of Vegetables and Flowers, Chinese Academy of Agricultural Sciences, 12 Nandajie Zhongguancun, Haidian District, Beijing 100081, China e-mail:
[email protected] X. Li e-mail:
[email protected] D. Shen e-mail:
[email protected] Y. Oiu e-mail:
[email protected] J. Song e-mail:
[email protected]
analysis whether based on 29 morphological traits or bulb yield-related traits could distinguish all germplasm. All accessions were divided into two groups with bolting and non-bolting respectively, or into five subgroups with different traits based on 29 morphological traits. Principal coordinate analysis based on eight bulb related traits divided all accessions into 6 groups. Yield among the accessions ranged from 1.60 to 16.78 t/ha, and three accessions yielded above 15 t/ha. Pearson analysis suggested bulb yield was significantly positively correlated with bulb weight (r = 0.99), bulb diameter (r = 0.73), bulb height (r = 0.53), clove number (r = 0.52), leaf width (r = 0.52). Allicin content ranged from 0.81 to 3.01 %. Pseudostem diameter was found to be significantly positively correlated with allicin content but with the low correlation coefficient (r = 0.23). The result will obviously be helpful for breeder and researchers to comprehensively understand the genetic background of the collection and more easily select the target accessions, especially those with high yield and allicin content. Keywords Garlic Diversity Morphological traits Allicin
Introduction Garlic (Allium sativum L.) has been cultivated for at least 5,000 years presumably having originated in Central Asia (Vavilov 1951) and has been spread west,
123
244
south and east (Etoh et al. 2001). Currently, garlic as a second biggest crop after onion among allium species is grown all over in the world. Asia is the main principle area, where, China is the biggest producer, occupying 3/4 plus of the total production in the world. Although garlic is an asexually propagated crop, and reproduces only by vegetative way, a large-scale diversity of different ecotypes has been established over time in various areas of cultivation (Baghalian et al. 2005; Bradley et al. 1996; Avato et al. 1998). Different ecotypes display great morphological diversity in bulb and leaf size, color and shape, scape presence and height, and flower color, fertility, and bulbil (topset) development in inflorescence (Pooler and Simon 1993). So, evaluation of garlic genetic resources both by morphological traits or molecular makers will make us to understand the variation between accessions and select out those with our interested character for breeding program. Garlic is mainly used as spice and flavoring agent for foods. Garlic is also cultivated for its medicinal properties attributable to sulfur compound including allicin (Taucher et al. 1996). Allicin was proved as the most important composition which had been used for human, animal and plants to fight against kinds of diseases (Tucak et al. 2009). Plant and animal studies published indicated that allicin has potential antibacterial and anti-fungal properties and positive pharmaceutical effect against atherosclerosis, fat deposition, lipoprotein unbalance, hypertension (Fry et al. 2005; Fujisawa et al. 2008; Lawson 2009; Lawson and Gardner 2005; Chen et al. 2011; Dugan et al. 2011; Kamkar et al. 2011). So, the allicin content was the most important quality trait for garlic. The amount of allicin in garlic is highly variable due to agronomic parameters (Mayeux et al. 1988; Baghalian et al. 2006) and genetic factors as well (Baghalian et al. 2006; Wang et al. 2010). Therefore, the allicin content evaluation of garlic genetic resource will be valuable for high allicin content variety improvement. The collection and conservation of garlic germplasm has long history in China. The national-wide collection started since 2002. 375 accessions of garlic from 23 provinces and areas in China has been collected and preserved in national germplasm repository for vegetatively propagated vegetables. However, the genetic diversity based on morphological traits and allicin content were not understood well yet. The objectives of the study are to analyze the variation in important morphological characteristics and the
123
Euphytica (2014) 198:243–254
allicin content of garlic in China, to understand the genetic background based on the morphological traits and the relationship between the allicin content and morphological traits. The results will benefit the further research at molecular level, and definitely will provide lot information of the garlic collection in China for researchers, breeders and producers. Materials and methods Plant material and field experiments Totally, 212 accessions of garlic germplasm from different areas in China were used in the study (Fig. 1). Field experiments were conducted during 2006–2009 using complete randomized block design with three replicates. Each clone planted in a plot of 4.5 m2 by hand in four rows with 20 cm between and 10 cm within lines at the vegetable research center of international agricultural high and new technology industrial park, Chinese academy of agricultural sciences. Yield (t/ha) were estimated by the three replicates. Morphological traits were measured at various growth stages according to the descriptors for garlic developed by the international plant genetic resources institute (IPGRI and GR 2001) and descriptors and data standard for garlic (Li and Zhu 2006). Totally, 21 quantitative and eight qualitative traits were investigated, and ten plants from each replicate were measured for quantitative traits and three times were set according to different mature time of different accessions (Table 1). The bulb and bolt are the two main consume organs for garlic in China. Thus, the traits related with bulb, like bulb height, bulb
Fig. 1 Geographic distribution of 212 accessions of garlic germplasm
Euphytica (2014) 198:243–254
245
Table 1 Quantitative and Qualitative traits investigated in the experiments Investigating time
Quantitative traits
Qualitative traits
About 20 days before bulb harvest (May 9–29)
Plant height, plant width, leaf length, leaf width, leaf number, pseudostem height, pseudostem diameter
Plant type, leaf posture, leaf color, leaf wax
Time when bolt was harvested (May 10–30)
Bolt length, basal diameter of bolt, mid-diameter of bolt, bolt weight, spathe length, spathe width
Bolting
About 20 days after harvest (May 29–June 9–19)
Bulb height, bulb diameter, bulb weight, clove height, clove diameter, clove number, basal plate thickness, basal plate diameter
Bulb type, bulb color, clove type
diameter, bulb weight, clove diameter, clove number, basal plate thickness, basal plate diameter, and traits related with bolt, like bolt length, basal diameter of bolt, mid-diameter of bolt, bolt weight, spathe length, and spathe width were very important to help us to evaluate the quality of bulb and bolt.
Allicin determination Five bulbs and one clove from each bulb from each of 212 accessions were randomly selected in three replications 40 days after harvest in 2009. Cloves from each replicate were peeled for removal of the dry protective leaves and kept in -20 °C for 3–4 h, and then chopped into small slices immediately. Soon, the garlic slices were kept in -80 °C for 3–7 h, and then frozen to dry and finally ground to powder. Allicin content analysis was done at supervision and testing center for vegetable quality, Ministry of Agriculture,
Table 2 Basic statistic information for 21 quantitative traits Maximum (cm)
Minimum (cm)
Range (cm)
Mean ± SD (cm)
CV (%)
H0
Plant height
90.50
27.50
63.00
58.19 ± 9.64
16.56
1.41
Plant width
70.10
9.01
61.09
35.82 ± 12.24
34.18
1.45
Leaf length
74.00
1.84
72.16
40.57 ± 8.88
21.88
1.42
Leaf width
4.74
0.53
4.21
1.67 ± 0.53
31.87
1.34
Leaf number
10.30
4.67
5.63
7.28 ± 0.97
13.39
1.47
Pseudostem height
42.34
11.55
30.79
27.14 ± 6.43
23.68
1.47
2.55
0.48
2.07
1.12 ± 0.34
30.44
1.40
Pseudostem diameter Bulb height Bulb diameter Bulb weightb
6.54
2.10
4.44
3.52 ± 0.5
14.19
1.4
7.61 59.90
2.23 4.11
5.38 55.79
3.8 ± 0.67 21.86 ± 10.4
17.67 47.57
1.39 1.38
Clove height
4.26
1.68
2.58
2.85 ± 0.47
16.57
1.47
Clove diameter
5.05
0.85
4.20
1.78 ± 0.46
25.68
1.37
Clove numbera
12.60
1.00
11.60
6.42 ± 2.04
31.72
1.48
1.55
0.20
1.35
0.5 ± 0.17
34.23
1.45
Basal plate thickness Basal plate diameter Bolt length
3.33
0.48
2.85
1.46 ± 0.38
25.86
1.39
68.04
0.48
67.56
43.06 ± 12.49
29.02
1.67
Basal diameter of bolt
1.41
0.25
1.16
0.53 ± 0.14
26.59
1.45
Mid-diameter of bolt
1.21
0.25
0.96
0.43 ± 0.12
28.51
1.46
Bolt weight
51.50
2.67
48.83
8.48 ± 5.58
65.84
1.52
Spathe length
37.10
0.70
36.40
21.27 ± 5.12
24.06
1.59
Spathe width
2.51
0.38
2.13
0.9 ± 0.36
39.92
1.54
a
Unit is ‘‘number’’
b
Unit is ‘‘g’’
123
246
Euphytica (2014) 198:243–254
China according to the method described by previous study (Wang et al. 2010). Data analysis ArcView 32 software was used to reveal the distribution of all accessions from China. Agronomical data was analyzed using SPSS software. Basic statistics including mean, maximum, minimum and coefficient of variation (CV) were applied to compare genetic variation of quantitative traits among accessions.
The five grades (1, 3, 5, 7, 9) were given for each quantitative trait according to the divided points (-1.2818S), (-0.5426S), (?0.5246S) and (?1.2818S) (S, standard error). In order to combine quantitative and qualitative traits for cluster analysis and principal component analysis (PCA), a series of number was given to each grade for qualitative traits. Shannon diversity index was calculated by the equation H0 = –Rpi ln (pi) (i, trait grade; pi, the frequency of the sample within a certain grade). Ward’s cluster analysis and PCA procedures were performed by software
Table 3 Basic statistic statistics for eight qualitative traits Percentage
H0
66
31.13
0.83
2
133
62.74
3 1
13 34
6.13 16.04
Semi-drooping
2
134
63.21
Erect
3
44
20.75
Traits
Categories
Series of number
Plant type
Erect
1
Semi-erect Spreading Drooping
Leaf posture
Leaf color
Leaf wax
Bolting
Bulb type
Bulb color
Clove type
123
Accessions
Yellowish green
1
3
1.42
Light green
2
4
1.89
Green
3
72
33.96
Dark green
4
133
62.74
Absent
1
0
Little
2
84
39.62
Medium
3
117
55.19
0
Much
4
11
5.19
No bolting
1
33
15.57
Incomplete bolting
2
1
Bolting
3
178
84.43
Flatly spherical
1
101
47.64
Nearly spherical High spherical
2 3
47 64
22.17 30.19
White
1
13
6.13
Light yellow
2
25
11.79
Light red
3
7
3.30
Purplish red
4
8
3.77
Brown
5
9
4.25
Purplish stripe
6
150
70.75
Regular multi-fan
1
1
0.47
Regular two-fan
2
28
13.21
Regular single-fan
3
121
57.08
Single
4
22
10.38
Irregular
5
40
18.87
0.91
0.79
0.85
0.43
1.05
1.04
1.16
Euphytica (2014) 198:243–254
NTSYS3.10PC based on the 21 quantitative traits and 8 qualitative traits. Pearson correlation coefficients (r) among 29 morphological traits and bulb yield were calculated automatically by software SPSS13.0. To clearly understand the relationship among quantitative traits which were relative with bulb yield, principal coordinate analysis was used based on eight traits including bulb height,bulb diameter, bulb weight, clove height, clove diameter, clove number, basal plate thickness, and basal plate diameter. To understand the distribution of allicin content among accessions, cluster analysis was performed based on unweighted pair-group method with arithmetic means (UPGMA) and the cluster tree diagram was put out by software DSP 9.5.
247 Table 4 Eigenvalues and their proportions Principles
Eigen value
Proportion
1
7.70
26.56
2
3.61
12.46
39.02
3.61
3
2.23
7.70
46.73
2.23
4
1.97
6.81
53.53
1.97
5
1.58
5.46
58.99
1.58
6
1.29
4.45
63.44
1.29
7
1.203
4.14
67.58
1.20
8
1.09
3.77
71.35
1.09
9
0.99
3.44
74.78
10
0.90
3.11
77.89
11
0.78
2.68
80.58
12
0.75
2.59
83.17
13
0.65
2.24
85.410
14
0.55
1.91
87.32
Results
15 16
0.49 0.45
1.69 1.55
89.01 90.57
Phenotypic variation analysis based on basic statistics
17
0.42
1.45
92.01
18
0.34
1.16
93.17
19
0.32
1.09
94.26
20
0.29
1.00
95.27
21
0.26
0.88
96.15
22
0.23
0.78
96.93
23
0.19
0.67
97.60
24
0.19
0.65
98.25
25
0.16
0.56
98.81
26
0.15
0.50
99.31
27
0.09
0.31
99.62
28
0.09
0.30
99.92
29
0.02
0.08
100.00
Maximum, minimum, range mean CVs and coefficient traits of 21 quantitative traits among 212 garlic clones showed wide variation (Table 2). Among 21 traits, descending sorting of the CVs was as follows: bolt weight, bulb weight, spathe width, basal plate thickness, plant width, leaf width, clove number, pseudostem height, bolt length, mid-diameter of bolt, basal diameter of bolt, diameter of basal plate, clove diameter, spathe length, pseudostem height, leaf length, bulb diameter, clove hight, plant height, bulb height, and leaf number. As in result, the leaf number did not change so much with CV of 13.39 %. However, there was wide variation for weight of bolt and bulb with CV of 65.84 and 47.57 % respectively, which means there would be big potential possibility to select elite germplasm with high yield of bulb and bolt from the collection. All the Shannon’s diversity index (H0 ) of the quantitative traits were above 1.30 from the lowest of 1.34 for leaf width to the highest of 1.67 for bolt length. For qualitative traits, almost all grades but absent of leaf wax for 8 traits were present among 212 accessions of garlic (Table 3). The Shannon’s diversity index (H0 ) of qualitative traits were from the lowest of 0.43 for bolting to the highest of 1.05 for clove type.
Cumulative (%) 26.56
Extracted principles 7.72
Variation dissection by principal component analysis Principal component analysis of 21 quantitative and 8 qualitative traits revealed that eight principal components with eigenvalue above 1 explained 71.35 % of total morphological variation. Besides, the eigenvalues of first three principles were obviously higher than the others (Table 4). The first principal component was comprised of plant height, plant width, leaf length, leaf width, leaf number, pseudostem height, pseudostem height, bolt length, basal diameter of bolt, mid-diameter of bolt, bolt weight, spathe length, spathe width, and bolting, that the traits descriptive for aboveground part
123
248
Euphytica (2014) 198:243–254
8N076 8N233 8N078 8N123 8N218 8N042 8N060 8N245 8N140 8N023 8N178 8N241 8N272 8N013 8N038 8N026 8N139 8N263 8N085 8N125 8N100 8N181 8N030 8N208 8N175 8N211 8N324 8N039 8N043 8N036 8N200 8N155 8N273 8N156 8N172 8N168 8N031 8N260 8N025 8N261 8N259 8N268 8N265 8N194 8N258 8N327 8N219 8N264 8N093 8N120 8N094 8N102 8N047 8N065 8N061 8N121 8N220 8N115 8N267 8N186 8N222 8N071 8N195 8N104 8N064 8N067 8N041 8N045 8N154 8N069 8N185 8N232 8N269 8N255 8N275 8N126 8N130 8N189 8N190 8N066 8N001 8N010 8N027 8N170 8N037 8N079 8N017 8N223 8N040 8N072 8N032 8N035 8N201 8N236 8N099 8N256 8N257 8N197 8N034 8N191 8N326 8N238 8N239 8N240 8N150 8N152 8N205 8N147 8N173 8N207 8N151 8N148 8N149 8N188 8N359 8N322 8N325 8N320 8N212 8N244 8N321 8N364 8N362 8N358 8N365 8N248 8N249 8N296 8N315 8N217 8N242 8N044 8N237 8N313 8N028 8N046 8N074 8N129 8N271 8N167 8N274 8N141 8N254 8N146 8N169 8N091 8N097 8N192 8N004 8N183 8N073 8N112 8N124 8N128 8N118 8N270 8N113 8N107 8N180 8N101 8N109 8N108 8N116 8N117 8N266 8N221 8N323 8N246 8N002 8N206 8N122 8N314 8N202 8N352 8N302 8N303 8N367 8N084 8N157 8N231 8N234 8N089 8N215 8N304 8N330 8N329 8N182 8N318 8N351 8N355 8N298 8N360 8N363 8N361 8N193 8N309 8N198 8N127 8N179 8N307 8N310 8N209 8N308 8N019 8N306 8N024 8N016 8N145 8N312 8N096 8N003 8N070
Fig. 2 The cluster tree based on 29 morphological traits
of pant, which explained 26.56 % of the total variation;the second principal component was comprised of bulb height, bulb diameter, bulb weight, clove height, clove number, clove number, pseudostem diameter, that traits relative with bulb yield, which explained 12.46 % of the total variation; the third principal component was comprised of only one trait descriptive for leaf pose, which explained 7.07 % of the total variation;the fourth principle component was comprised of clove width, and bulb type were traits descriptive for bulb type, which explained 6.81 % of total variation; the fifth principle component was comprised of clove structure, leaf color, and leaf wax were relative with clove structure and leaf color, which explained 5.45 % of total variation; the sixth principle component was comprised of basal plate was relative with the position of basal plate, which explained 4.45 % of total variation; the seventh principle component was comprised of plant type explained 4.14 % of total variation; the eighth principle component was comprised of bulb color explained 3.76 % of total variation. It indicated that these traits were helpful to classify the garlic germplasms. Cluster analysis based on 29 morphological traits Cluster analysis revealed that 212 accessions of garlic were clustered into two groups (Fig. 2). The first group marked by ‘‘A’’ mainly included the accessions which did not bolt, and the second group marked by ‘‘B’’ mainly included the accessions which bolted. Furthermore, the first group could be divided into two subgroups, the first subgroup marked by ‘‘A1’’ mainly covered the accessions with high spherical bulbs and semi-drooping leaves, and the second subgroup
123
marked by ‘‘A2’’ mainly was comprised of the accessions with dark green leaves and erect plant type and erect leaves. The second group was clustered into three subgroups, the first subgroup marked by ‘‘B1’’ mainly included those accessions with bolt length from 36.28 to 50.00 cm, the second subgroup marked by ‘‘B2’’ mainly included those accessions with clove diameter from 1.53 to 1.80 cm and the other accessions were grouped into the third subgroup marked by ‘‘B3’’. Correlation analysis among 29 morphological traits and bulb yield In the correlation matrix (Table 5), most of the traits showed high correlation with each other, which indicate that some traits could be selected for variety improvement programs to save time and labors. Bulb yield was found to be strongly positively correlated (r = 0.99) with bulb weight and bulb diameter (r = 0.71). Obviously, if there is bigger bulb diameter or higher weight bulb, higher yield could be expected. Clove diameter was moderately correlated with bulb clove number(r = 0.45), which indicates that accessions with big bulb diameter may not necessarily produce more cloves. Principal coordinate analysis based on bulb relative quantitative traits Principal coordinate analysis based on traits relative with yield clustered 212 accessions of garlic into six groups (Fig. 3). The first group included four accessions of garlic with smaller clove diameter, smaller pseudostem diameter, moderate weight bulb, and bulb
a
a
0.01
0.21a
0.08
0.36a
Allicin
Bulb yield
-0.03
0
0.01
Bulb color
Clove type
-0.18
Bulb type
0.32a
0.38a
-0.01
-0.16b
0.07
0.20a
Bolting
-0.04
-0.02
0.38a
0.09
Leaf wax
0.11
0.14b
-0.09
0.16b
Leaf color
0.02
0.18
0.19
0.46
0.34a
0.41a
-0.05
0.05
0.17b
a
0.06
0.03
-0.06
0.15
Leaf posture
Plant type
0.13
Spathe width
a
a
b
0.48
a
a
0.30
0.25a
0.28a
Bolt weight
Spathe length
a
0.37a
0.31a
0.52a
0.02
-0.06
-0.05
-0.13
0.24a
0.1
0.13
-0.06
0.24
0.32
0.23
a
0.55a
0.52a
0.47a
0.12
Mid-diameter of bolt
0.43a
0.38a
0.18a
0.29a
0.38a
0.24a
0.40a
0.17b
0.31a
0.33a
0.41a
0.24a
Basal plate diameter
0.22a
0.26a
0.33
a
0.06
0.52a
0.54a
0.39a
0.75a
0.30a
0.57a
Leaf width
Basal diameter of bolt
0.18a
0.22
a
-0.13
0.37a
0.38a
0.25a
0.54a
0.46
a
0.45a
0.67a
Leaf length
Bolt length
0.19a
0.15b
0.22a
Clove number
0.14
Basal plate thickness
0.26
Clove diameter
a
b
a
a
-0.21
0.20a
0.36a
Bulb weight
-0.20
0.30a
Bulb diameter
Clove height
0.15b
0.23a
0.34a
Bulb height
0.39a
0.46a
Pseudostem diameter
0.36
0.68
a
0.32
Pseudostem height
Leaf number
a
a
a
0.43
0.49a
a
0.45
0.64a
Leaf width
0.72a
0.47a
Leaf length
Plant width
Plant width
Plant height
0.48a
0
0.04
0.08
-0.17
b
0.12
0.07
0.06
-0.07
0.11
0.11
0.23
a
0.24a
0.24a
0.19a
0.11
0.32a
0.21a
0.33a
0.11
0.14
b
0.48a
0.47a
0.35a
0.54a
0.40a
Leaf number
Table 5 Pearson Correlation Coefficients between traits
0.42a
0.13
0.1
0.06
-0.24
a
0.23a
0.08
0.18a
0.08
0.03
0.09
0.25
a
0.11
0.18a
0.20a
0.36a
0.41a
0.33a
0.26a
0.17
b
-0.15
b
0.42a
0.43a
0.31a
0.27a
Pseudostem height
a
0.44a
-0.06
-0.06
0.01
-0.04
0.08
0.22
a
0.09
-0.01
0.12
0.18
a
0.14
b
0.39a
0.37a
0.33a
-0.01
0.25a
0.19a
0.21a
0.30
0
0.44a
0.43a
0.33a
Pseudostem diameter
0.53a
0.1
-0.05
0.03
-0.12
-0.06
0
0.11
0.02
0.06
-0.05
-0.07
0.13
0.06
0.02
-0.02
0.31a
0.26a
0.22a
0.28
a
0.07
0.53a
0.60a
Bulb height
a
a
b
0.71a
0.06
0.02
0.03
-0.25
a
0.14b
0.07
0.17b
-0.16b
0.18
0.14
0.09
0.34a
0.28a
0.22a
0.18a
0.42a
0.28a
0.45a
0.27
0.02
0.71a
Bulb diameter
0.19
0.99a
0.08
-0.04
-0.01
-0.25
a
0.08
0.21
a
0.21a
-0.16b
a
0.08
0.07
0.26a
0.23a
0.19a
0.12
0.30a
0.09
0.52a
0.11
0.26a
Bulb weight
0.26a
-0.04
-0.01
-0.1
-0.04
-0.19a
0.13
-0.08
-0.1
0.05
-0.05
-0.21
a
0.01
-0.09
-0.12
-0.21a
-0.31a
-0.17b
-0.04
-0.03
Clove height
0.11
0.07
-0.03
0.04
0.07
-0.09
0.16
b
0.09
0.09
0.12
0.1
-0.12
0.38a
0.21a
0.17b
0.02
0.14b
0.06
-0.44a
Clove diameter
a
0.52a
0
0.13
-0.08
-0.22
a
0.25a
-0.03
0.17b
-0.17b
0.12
0.11
0.24
0.1
0.20a
0.18b
0.16b
0.36a
0.14b
Clove number
0.09
0.07
-0.04
0.14b
-0.09
0.22a
-0.17b
0.01
0
-0.06
0.07
0.21
a
0.06
0.13
0.11
0.15b
0.33a
Basal plate thickness
a
0.29a
0.1
0.04 0.07
0.23a
0.08
-0.27a
0.70a
-0.09
-0.04
-0.08
0.09
0.57a
0.66a
0.56a
0.72a
0.75a
Bolt length
0.11
0.04
-0.1
0.34a
-0.13
0.12
0.05
0.07
0.19
0.33
a
0.20a
0.30a
0.28a
0.33a
Basal plate diameter
Euphytica (2014) 198:243–254 249
123
123
-0.01
-0.03
0.75
Leaf color
Leaf wax
Bolting
Allicin
b
a
a
a
Correlation is significant at the 0.05 level
Correlation is significant at the 0.01 level
0.17b
-0.1
Clove type
Bulb yield
0.04
-0.06
Bulb color
-0.19
-0.09
Leaf posture
Bulb type
0.23
Plant type
a
a
a
0.22a
-0.07
-0.06
0.04
-0.21
0.73
-0.03
-0.02
-0.12
0.24
0.81
a
a
a
0.70a
Spathe length
0.76
0.65a
0.79a
Bolt weight
Spathe width
0.86a
0.96a
Mid-diameter of bolt
Mid-diameter of bolt
Basal diameter of bolt
Table 5 continued
0.24a
-0.04
-0.02
-0.06
-0.16
b
0.43
a
0.09
0.04
-0.12
0.26
a
0.75a
0.38a
Bolt weight
0.05
-0.03
-0.05
0.11
-0.13
0.75
0.06
-0.08
-0.01
0.03
-0.16
b
0.63
a
-0.05
-0.20a a
-0.09
-0.12
0.25a
Spathe width
-0.08
0.01
0.06
0.50a
Spathe length
b
0.18a
-0.13
0.01
0.03
-0.09
0.16
0.07
0.01
-0.68a
Plant type
b
-0.15b
0.12
0.01
0
0.14
-0.1
-0.09
0
Leaf posture
a
0.23a
-0.02
0.12
-0.13
-0.19
-0.1
0.14b
Leaf color
0.21a
-0.03
0.04
-0.13
0.02
-0.19a
Leaf wax
0.07
-0.03
-0.02
0.11
-0.16b
Bolting
-0.25a
-0.03
-0.14b
0.05
Bulb type
-0.02
-0.03
-0.32a
Bulb color
-0.04
0
Clove type
0.09
Allicin
250 Euphytica (2014) 198:243–254
Euphytica (2014) 198:243–254
251
Fig. 3 The scatter plot of the first and second principal coordinates by the principal coordinates analysis
Table 6 Bulb yield distribution of the 212 accessions Yield range (t/ha) C15
Accessions 3
Percent (%) 1.42
12–15
20
9.43
6–12
101
47.64
4–6
58
27.36
2–4
28
13.21
\2
2
0.94
with more clove number. Six accessions of garlic were included in second group, which are with lighter bulb, short clove. In the third group, there were 136 accessions with moderate bulb weight and bulb diameter. There were 64 accessions in the fourth group with bulb weight over 16 g and moderate bulb diameter. In the fifth group, there were six accessions with wide clove width, and less clove number. There were six accessions in the sixth group with light bulb weight and less clove number. Compared with the other groups, the fourth group possessed most of accessions with higher bulb weight and big bulb diameter, and was potential to select high yield candidate varieties. According to the yield, 212 accessions of garlic were clustered into six groups (Table 6), 3, 20, 101, 58, 28, 2 accessions for each group with yield over
15, 12–15, 6–12, 4–6, 2–4 t/ha, and lower than 2 t/ha respectively. There were 1.42 % of total accessions, three accessions with yield above 15 t/ha. Allicin content distribution and its relationship with morphological traits According to cluster analysis based on allicin content (Fig. 4), 212 accessions of garlic were clustered into five groups. The first group included 10 accessions of garlic with higher allicin content from 2.12 to 3.01 %. The second, third, and fourth groups included 82, 9 and 18 accessions with moderate allicin content from 1.61 to 2.07 %, from 1.56 to 1.59 %, and from 1.51 to 1.55 % respectively. In the fifth group, there were 93 accessions of garlic with lower allicin content from 0.49 to 1.49 % (Table 7). In correlation analysis (Table 5), among 29 traits, only pseudostem diameter was found to be significantly positively correlated with allicin content. However, the correlation coefficient (r = 0.23) was very low.
Discussion Morphological traits diversity of crop germplasm resources plays a significant role for breeding program. The variation found in qualitative traits are more useful in germplasm identification and developing new varieties, and quantitative traits are of direct
123
252
Euphytica (2014) 198:243–254 b Fig. 4 Cluster tree diagram based on allicin content of 212
accessions of garlic
IV Table 7 Allicin content distribution of garlic accessions among different groups
V
II
I III
123
Groups
Range of allicin content
Accessions
I
2.21–3.01
10
II
1.61–2.07
82
III
1.56–1.59
9
IV
1.51–1.55
18
V
0.49–1.49
93
agronomic interest (Panthee et al. 2006). Generally, diversity coefficient (H0 ) and CVs are considered as important criterion to evaluate the variation of the traits among samples and within them. In our study, a large collection comprising of 212 accessions were evaluated in more detail than previous studies. Both qualitative and quantitative traits of 212 accessions of garlic showed wide variation, especially in some quantitative involved in plant growth and bulb development. Besides, some traits relative to bolt, bolt length, basal diameter of bolt, middiameter of bolt, bolt weight, spathe length, spathe width for instance, which were seldom studied in previous studies were evaluated in our experiment. Because the bolt is the second consume organ of garlic after bulb, the information in this study will benefit breeders and researchers who are interested in bolt. Bulb yield is the most important trait for garlic, and has been widely evaluated by couples of researchers in previous studies (Jabbes et al. 2012; Baghalian et al. 2006). Studying diversity of 31 garlic landraces from Tunisia, Jabbes et al. (2012) found that the yield was highly influenced by the following traits: weight of the clove, the weight and the diameter of the bulb, the number of leaves per plant and the stem length. Baghalian et al. (2005) found a significant positive correlation between clove and bulb mean weight as well, negative correlation between clove mean weight and clove number. Raju et al. (2013) got similar findings based on 56 garlic genotypes. Mishra et al. (2013) found that 20 promising garlic genotypes differed significantly as to the different morphological attributes In this
Euphytica (2014) 198:243–254
experiment, bulb yield was also found to be strongly positively correlated with bulb weight and bulb diameter. Cluster analysis based on morphological traits performed on genetic resources make breeder easier to understand the germplasm and select the potential material among a large-scale accessions with high efficiency, which were studied on the genetic resources of many crops (Piyusha and Jaiswal 2013; Sadegh et al. 2013), horticulture crops include pepper (Sisaphaithong 2009), cowpea (Simango and Lungu 2010), watermelon (Xiao 2012). Some similar research could also be referenced. Menezes Sobrinho et al. (1999) conducted a study to 89 garlic germplasms of Brazil and found 13 clusters. Lallemand et al. (1994) also evaluated 65 garlic accessions and found six clusters on the basis of morphological traits. In this study, germplasm cluster analysis whether based on 29 morphological traits or bulb yield-related traits can distinguish all germplasm and will make breeder comprehensively understand the genetic background of the collection and more easily select the target asscessions with high yield and different component traits from different groups. We clustered 212 accessions of garlic into five groups based on 21 quantitative and 8 qualitative traits, which will help breeder easily to select out the potential materials in different groups. For instance, breeder interesting in bolt production could search genotypes from group B, and the information would also benefit the study on true seed production and future sexual breeding programs. According to Principal coordinate analysis, 212 accessions of garlic were clustered into six groups. Research or breeder might quest high yield genotypes from the fourth group which possessed accessions with higher bulb weight and big bulb diameter. Besides, three accessions yielded above 15 t/ha were selected out which will benefit the high yield variety improvement. Allicin content was both affected by geographical and genetic factors (Baghalian et al. 2005). In this study, the allicin content among 212 accessions of garlic grown in same environment ranged from 0.81 to 3.01 %, which could be mainly ascribed to influence by genetics. Although the allicin content in different studies are not comparable, several authors (Baghalian et al. 2005; Camargo et al. 2004; Mirzaei et al. 2007; Gonza´lez et al. 2009; Soto Vargas et al. 2010) agreed that the genetic variation of allicin content existed
253
extensively, which could be employed for garlic quality improvement. We clustered all accessions of garlic into five categories according to the allicin content. At same time, we found that allicin content was significantly positively correlated with pseudostem diameter, which might help to screen the high allicin content garlic genotypes from large germplasm collection. However, the correlation coefficient (r = 0.23) was unfortunately very low, the validity between them need further study. Acknowledgments The research was supported by national science foundation for young scientists of China (31000910), key laboratory of horticultural crops genetic improvement,ministry of agriculture, China national R&D special fund for public welfare industry (200903018-03), and the National Key Technology R&D Program from the Ministry of Science and Technology of China (2013BAD01B04-8), and key laboratory of horticultural crops genetic improvement,ministry of agriculture. We also very appreciated their help in data analysis from all members in the Philipp W. Simon’s lab of USDA-ARS Vegetable Crops Research Unit and Department of Horticulture, University of Wisconsin USA.
References Avato P, Miccolis V, Tursi F (1998) Agronomic evaluation and essential oil content of garlic (Allium sativum L.) ecotypes grown in southern Italy. Adv Hortic Sci 4:201–204 Baghalian K, Ziai SA, Naghavi MR, Badi HN, Khalichi A (2005) Evaluation of allicin content and botanical traits in Iranian garlic (Allium sativum L.) ecotypes. Sci Hortic 103(2):155–166. doi:10.1016/j.scienta.2004.07.001 Baghalian K, Naghavi MR, Ziai SA, Badi HN (2006) Postplanting evaluation of morphological characters and allicin content in Iranian garlic (Allium sativum L.) ecotypes. Sci Hortic 107(4):405–410. doi:10.1016/j.scienta.2005.11.008 Bradley K, Rieger MA, Collins G (1996) Classification of Australian garlic cultivars by DNA fingerprinting. Anim Prod Sci 36(5):613–618 Camargo A, Masuelli R, Burba J (2004) Characterization of argentine garlic cultivars for their allicin content. In: IV international symposium on edible alliaceae, vol 688, pp 309–312 Chen C-H, Chou T-W, Cheng L-H, Ho C-W (2011) In vitro antiadenoviral activity of five allium plants. J Taiwan Inst Chem Eng 42(2):228–232. doi:10.1016/j.jtice.2010.07.011 Dugan FM, Hellier BC, Lupien SL (2011) Resistance to Penicillium allii in accessions from a national plant germplasm system allium collection. Crop Prot 30(4):483–488. doi:10. 1016/j.cropro.2010.12.021 Etoh T, Watanabe H, Iwai S (2001) RAPD variation of garlic clones in the center of origin and the westernmost area of distribution. Mem Fac Agric Kogoshima Univ 37:21–27 Fry FH, Okarter N, Baynton-Smith C, Kershaw MJ, Talbot NJ, Jacob C (2005) Use of a substrate/alliinase combination to
123
254 generate antifungal activity in situ. J Agric Food Chem 53(3):574–580. doi:10.1021/jf048481j Fujisawa H, Suma K, Origuchi K, Seki T, Ariga T (2008) Thermostability of allicin determined by chemical and biological assays. Biosci Biotechnol Biochem 72(11): 2877–2883 Gonza´lez RE, Soto VNC, Sance MAM, Camargo AB, Galmarini CR (2009) Variability of solids, organosulfur compounds, pungency and health-enhancing traits in garlic (Allium sativum L.) cultivars belonging to different ecophysiological groups. J Agric Food Chem 57(21): 10282–10288 IPGRI E, GR A (2001) Descriptors for Allium (Allium spp.). International Plant Genetic Resources Institute, Rome Jabbes N, Arnault I, Auger J, Al Mohandes Dridi B, Hannachi C (2012) Agro-morphological markers and organo-sulphur compounds to assess diversity in Tunisian garlic landraces. Sci Hortic 148:47–54. doi:10.1016/j.scienta.2012.08.013 Kamkar B, Koocheki A, Mahallati MN, Teixeira da Silva JA, Moghaddam PR, Kafi M (2011) Fungal diseases and inappropriate sowing dates, the most important reducing factors in cumin fields of Iran, a case study in Khorasan provinces. Crop Prot 30(2):208–215. doi:10.1016/j.cropro. 2010.11.007 Lallemand J, Messian C, Briand F, Etoh T (1994) Delimitation of varietal groups in garlic (Allium sativum L.) by morphological, physiological and biochemical characters. In: I international symposium on edible alliaceae vol 433, pp 123–132 Lawson LD (2009) Allicin bioavailability from alliinaseinhibited garlic. Plant Med 75(4):451–451 Lawson LD, Gardner CD (2005) Composition, stability, and bioavailability of garlic products used in a clinical trial. J Agric Food Chem 53(16):6254–6261. doi:10.1021/ jf050536? Li X, Zhu D (2006) Descriptors and data standard for garlic (Allium sativum L.). China Agriculture Press, Beijing Mayeux P, Agrawal K, Tou J-S, King B, Lippton H, Hyman A, Kadowitz P, McNamara D (1988) The pharmacological effects of allicin, a constituent of garlic oil. Agents Actions 25(1–2):182–190 Menezes Sobrinho JAd, Charchar JM, Araga˜o FA (1999) Morphological characterisation of garlic germplasm by cluster, principal components and canonic variables analysis. Horticultura Brasileira 17(2):96–101 Mirzaei R, Liaghati H, Damghani AM (2007) Evaluating yield quality and quantity of garlic as affected by different farming systems and garlic clones. Pak J Biol Sci 10(13):2219–2224 Mishra R, Prasad K, Pandey S, Gupta R (2013) Evaluation of garlic accessions for growth, yield, quality and resistance
123
Euphytica (2014) 198:243–254 to stemphylium blight disease in northern India. VegetosInt J Plant Res 26(2):291–296 Panthee DR, Kc RB, Regmi HN, Subedi PP, Bhattarai S, Dhakal J (2006) Diversity analysis of garlic (Allium sativum L.) germplasms available in Nepal based on morphological characters. Genet Resour Crop Evol 53(1):205–212. doi:10.1007/s10722-004-6690-z Piyusha S, Jaiswal JP (2013) Assessment of genetic diversity in bread wheat (Triticum aestivum L. em. Thell), genotypes based on agro-morphological traits using Mahalanobis D2 statistic. Environ Ecol 31(2A):679–682 Pooler M, Simon P (1993) Characterization and classification of isozyme and morphological variation in a diverse collection of garlic clones. Euphytica 68(1–2):121–130 Raju P, Jain PK, Avneesh G, Sasode DS (2013) Morphological variability and character association in diverse collection of garlic germplasm. Afr J Agric Res 8(23):2861–2869 Sadegh M, Jafari AA, Rosna Mat T, Jamilah Syafawati Y, Azani S (2013) Genetic diversity analysis of agro-morphological and quality traits in populations of sainfoin (Onobrychis sativa). Aust J Crop Sci 7(7):1024–1031 Simango K, Lungu D (2010) Evaluation of physiological and morphological traits conferring drought tolerance in cowpea. RUFORUM working document series no. 5, pp 441–445 Sisaphaithong T (2009) Genetic diversity and clustering of tomato Lycopersicon lycopersicum (L.) Karsten accessions in Laos based on morphological traits (vol 34, pg 12, 2009). Philipp J Crop Sci 34(3):112 Soto Vargas VC, Gonza´lez R, Sance M, Burba J, Camargo A (2010) Genotype-environment interaction on the expression of allicin and pyruvic acid in garlic (Allium sativum L.). Revista de la Facultad de Ciencias Agrarias, Universidad Nacional de Cuyo 42(2):15–22 Taucher J, Hansel A, Jordan A, Lindinger W (1996) Analysis of compounds in human breath after ingestion of garlic using proton-transfer-reaction mass spectrometry. J Agric Food Chem 44(12):3778–3782 Tucak M, Cupic T, Popovic S, Stjepanovic M, Gantner R, Meglic V (2009) Agronomic evaluation and utilization of red clover (Trifolium pratense L.) Germplasm. Notulae Botanicae Horti Agrobotanici Cluj-Napoca 37:206–210 Vavilov NI (1951) The origin, variation, immunity and breeding of cultivated plants. Soil Sci 72(6):482 Wang H, Li X, Liu S, Jin S (2010) Quantitative determination of allicin in Allium sativum L. bulbs by UPLC. Chromatographia 71(1–2):159–161 Xiao G (2012) Review of watermelon genes for control of morphological traits and resistance traits. Journal of Hunan Agricultural University 38(6):567–573. doi:10.3724/sp.j. 1238.2012.00567
