Genetic diversity of maize inbreds
In the present study, all the 52 markers were successfully amplified in 30 maize inbred lines (Fig. 1). A total of 189 alleles were generated across the 30 genotypes with a mean of 3.63 alleles per locus (Table 1). The allele size of the 52 SSR markers ranged from 50 bp (phi 036) to 295 bp (p 101049) which revealed a high level of genetic diversity among the loci. Several researchers, viz., Hung et al. (2012) [117 alleles, 3.26 average alleles/ locus], Yu et al. (2012) [60 alleles, 2.73 average alleles/ locus], Devi et al. (2017b) [203 alleles, 2.69 average alleles/ locus] observed similar result while working on genetic diversity of maize genotypes. The high number of alleles per locus is most likely attributable to the higher genetic diversity of the lines. The PIC of the markers refers to the relative value of each marker regarding the amount of polymorphism exhibited, ranged from 0.17 (umc 1622) to 0.76 (umc 1153) with an average value of 0.49 demonstrating the high discriminatory power of these markers among the 30 genotypes (Table 1). Out of the 52 SSR markers, 24 SSR loci (46.15%) were found to be more informative by having of PIC value of ≥ 0.50, depicting the higher capability of these loci to differentiate between the genotypes, determine genetic diversity, gene introgression, genetic conservation, etc. Genetically diverse genotypes will have higher value of PIC, whereas closely related lines will have lower value (Muthusamy et al. 2015; Devi et al. 2017b). The PIC value of 0.46, 0.31 and 0.40 was also observed by Hung et al. (2012), Hao et al. (2015) and Devi et al. (2017b), respectively. The number of alleles per locus (K) ranged from 2 to 8 with an average of 3.634 ± 1.428. The bnlg1031 marker mapped at 8.06 bin showed the highest number of alleles (8). The value of Expected Heterozygosity (HExp) ranged from 0.19 to 0.80, whereas the Observed Heterozygosity (HObs) ranged from 0 to 0.89, indicating that further generations of inbreeding are required to stabilize the inbreds. The presence of high heterozygosity in some of the loci like umc 1354 (0.89), bnlg 2162 (0.79), bnlg 1031 (0.73), umc 2064 (0.57), bnlg 565 (0.39), bnlg 2291 (0.34), umc 2252 (0.33), bnlg 1496 (0.29), bnlg 1601 (0.27), etc. indicated the needs for further inbreeding in order to stabilise the lines. Hung et al. (2012) also reported high heterozygosity in his study among waxy maize inbreds.
Table 1
Genetic diversity analysis for 52 SSR markers in 30 genotypes
Sl No. | Locus | Bin location | K | HObs | HExp | PIC |
1 | umc 1354 | 1.00 | 6 | 0.89 | 0.71 | 0.66 |
2 | umc 1685 | 1.01 | 2 | 0.03 | 0.50 | 0.37 |
3 | umc 2225 | 1.02 | 4 | 0.07 | 0.59 | 0.50 |
4 | bnlg 2238 | 1.04 | 5 | 0.07 | 0.66 | 0.59 |
5 | umc 2064 | 1.07 | 4 | 0.57 | 0.75 | 0.69 |
6 | umc 1622 | 1.12 | 3 | 0 | 0.19 | 0.17 |
7 | bnlg 1064 | 2.03 | 3 | 0.2 | 0.56 | 0.47 |
8 | umc 2007 | 2.04 | 5 | 0.03 | 0.68 | 0.61 |
9 | umc 2252 | 2.05 | 3 | 0.33 | 0.44 | 0.37 |
10 | bnlg 2077 | 2.07 | 4 | 0.15 | 0.75 | 0.69 |
11 | umc 2085 | 2.08 | 2 | 0 | 0.51 | 0.37 |
12 | P 101049 | 2.10 | 5 | 0.07 | 0.68 | 0.61 |
13 | phi 036 | 3.04 | 5 | 0.21 | 0.74 | 0.67 |
14 | bnlg 1601 | 3.05 | 4 | 0.27 | 0.72 | 0.66 |
15 | umc 2050 | 3.07 | 2 | 0.14 | 0.46 | 0.35 |
16 | umc2174 | 3.08 | 2 | 0.05 | 0.34 | 0.28 |
17 | bnlg 1496 | 3.09 | 6 | 0.29 | 0.74 | 0.68 |
18 | umc 1008 | 4.00 | 2 | 0 | 0.51 | 0.37 |
19 | umc 2410 | 4.02 | 3 | 0.03 | 0.62 | 0.54 |
20 | umc 1117 | 4.04 | 3 | 0 | 0.67 | 0.58 |
21 | bnlg 2291 | 4.06 | 3 | 0.34 | 0.52 | 0.45 |
22 | bnlg2162 | 4.08 | 5 | 0.79 | 0.77 | 0.71 |
23 | umc 2289 | 4.10 | 5 | 0.13 | 0.64 | 0.56 |
24 | umc 1305 | 5.00 | 4 | 0.07 | 0.35 | 0.32 |
25 | bnlg 565 | 5.02 | 5 | 0.39 | 0.76 | 0.71 |
26 | umc 2304 | 5.05 | 4 | 0.18 | 0.47 | 0.41 |
27 | umc 1225 | 5.08 | 5 | 0.1 | 0.79 | 0.74 |
28 | umc 1686 | 5.03 | 2 | 0 | 0.50 | 0.37 |
29 | umc 1153 | 5.09 | 6 | 0 | 0.80 | 0.76 |
30 | umc 1002 | 6.00 | 2 | 0 | 0.49 | 0.36 |
31 | umc 2318 | 6.05 | 2 | 0 | 0.44 | 0.34 |
32 | umc 1257 | 6.02 | 3 | 0 | 0.55 | 0.45 |
33 | umc 1912 | 6.06 | 5 | 0 | 0.70 | 0.64 |
34 | bnlg 1740 | 6.07 | 5 | 0.22 | 0.63 | 0.58 |
35 | umc 1642 | 7.00 | 4 | 0 | 0.61 | 0.51 |
36 | umc 2325 | 7.01 | 2 | 0 | 0.48 | 0.36 |
37 | mmc 0411 | 7.03 | 3 | 0.22 | 0.56 | 0.45 |
38 | umc 1154 | 7.05 | 2 | 0.04 | 0.40 | 0.31 |
39 | umc 1543 | 7.04 | 2 | 0 | 0.34 | 0.27 |
40 | umc 1932 | 7.02 | 2 | 0 | 0.46 | 0.35 |
41 | umc 2042 | 8.01 | 2 | 0 | 0.50 | 0.37 |
42 | bnlg 1194 | 8.02 | 4 | 0.36 | 0.72 | 0.65 |
43 | bnlg 1031 | 8.06 | 8 | 0.30 | 0.73 | 0.68 |
44 | umc 1607 | 8.07 | 2 | 0 | 0.41 | 0.32 |
45 | bnlg 1012 | 9.04 | 3 | 0 | 0.58 | 0.49 |
46 | bnlg 1091 | 9.05 | 4 | 0.12 | 0.58 | 0.49 |
47 | umc 1380 | 10.00 | 3 | 0.13 | 0.56 | 0.46 |
48 | umc 1576 | 10.02 | 2 | 0 | 0.47 | 0.36 |
49 | umc 1962 | 10.03 | 5 | 0.23 | 0.44 | 0.40 |
50 | umc 1506 | 10.05 | 4 | 0 | 0.62 | 0.54 |
51 | bnlg2190 | 10.06 | 5 | 0.07 | 0.65 | 0.58 |
52 | umc 2021 | 10.07 | 3 | 0.17 | 0.55 | 0.46 |
Mean ± SD | | 3.634 ± 1.428 | 0.139 ± 0.193 | 0.574 ± 0.137 | 0.493 ± 0.146 |
The heterozygosity percentage of the inbred lines at 52 SSR markers was studied which revealed the range of heterozygosity from 0 (SCM 15/9LC) to 27.45 (GP H 6) (Table 2). The graphical representations of heterozygosity in 30 maize inbreds are depicted in Fig. 2. Out of 30 inbred lines, 13 lines showed heterozygosity ≤ 10%, which means 43% of the lines attained more than 90% homozygosity. Another 3 inbred lines viz. GP (H) 38, MN 1 and SCM15/9LC showed heterozygosity less than 5%, which revealed a high level of homozygosity in these lines. These three inbred maize lines may be considered as pure lines which will be useful for development of mapping populations, genetic mapping, etc. Analysis of molecular variance (AMOVA) is an efficient way to assess overall distribution of diversity within and among populations (Belalia et al. 2019). In the present study, the AMOVA result revealed a higher molecular variance within individuals of population than among individuals of population. Similar findings were reported by Prasanna (2012), Silva et al. (2015) and Belalia et al. (2019).
Table 2
Population structure group and heterozygosity (%) of 30 early inbred lines
Sl No. | Genotypes | Heterozygosity (%) | Q1 | Q2 | Q3 | Structure groups |
1 | MZ 44 (c) | 14.28 | 0.006 0 | 0.042 0 | 0.952 | G3 |
2 | MZ 13(a) | 8.0 | 0.144 | 0.011 | 0.845 | G3 |
3 | MZ 34 (a) | 13.04 | 0.196 | 0.006 | 0.798 | G3 |
4 | GP(H) 1 | 14.58 | 0.024 | 0.017 | 0.959 | G3 |
5 | GP(H) 6 | 27.45 | 0.112 | 0.036 | 0.852 | G3 |
6 | MZ 45 (A) | 10.20 | 0.947 | 0.018 | 0.035 | G1 |
7 | MZ 21 (a) | 8.51 | 0.077 | 0.009 | 0.914 | G3 |
8 | MN 17 | 8.69 | 0.609 | 0.344 | 0.046 | AD |
9 | GP(H)-2 | 5.12 | 0.036 | 0.009 | 0.956 | G3 |
10 | MZ 41 ( A) | 6.60 | 0.986 | 0.006 | 0.008 | G1 |
11 | MZ 46 (B ) | 16.67 | 0.905 | 0.011 | 0.084 | G1 |
12 | GP (H)38 | 4.25 | 0.988 | 0.004 | 0.008 | G1 |
13 | Kolasib1(2) | 20.41 | 0.961 | 0.029 | 0.01 | G1 |
14 | MN 30 | 6.12 | 0.022 | 0.003 | 0.975 | G3 |
15 | MN 12 | 6.52 | 0.988 | 0.008 | 0.004 | G1 |
16 | MZ 42 | 8.33 | 0.956 | 0.007 | 0.037 | G1 |
17 | MZ 40 | 14.00 | 0.956 | 0.006 | 0.038 | G1 |
18 | MZ 70 | 20.41 | 0.983 | 0.004 | 0.013 | G1 |
19 | MZ 24 | 10.41 | 0.965 | 0.028 | 0.007 | G1 |
20 | MZ 2 | 11.53 | 0.991 | 0.005 | 0.004 | G1 |
21 | MN 29 | 10.0 | 0.005 | 0.902 | 0.092 | G2 |
22 | MN 1 | 3.92 | 0.005 | 0.992 | 0.003 | G2 |
23 | SCM 15/9LC | 0 | 0.012 | 0.948 | 0.039 | G2 |
24 | MZ 56 (a) | 6.38 | 0.022 | 0.972 | 0.006 | G2 |
25 | MZ 48 | 14.89 | 0.006 | 0.986 | 0.007 | G2 |
26 | MZ 29 | 8.33 | 0.003 | 0.017 | 0.981 | G3 |
27 | MN 13 | 12.76 | 0.01 | 0.951 | 0.038 | G2 |
28 | MN 3 | 14.0 | 0.003 | 0.994 | 0.003 | G2 |
29 | GP(H) 28 | 7.84 | 0.008 | 0.931 | 0.061 | G2 |
30 | MZ 22 | 6.38 | 0.207 | 0.759 | 0.034 | G2 |
AD: Admixture |
Genetic Relationships Analysis Among Early Maize Inbreds
The genotypic data of 52 SSR markers were used to estimate genetic distance matrix and a corresponding dendrogram was developed. Cluster analysis of 30 early maize inbred lines developed from local landraces was conducted based on genetic dissimilarities from SSR data using neighbour- joining method. The genetic dissimilarity between the genotype pairs ranged from 0.40 to 0.64 with a mean value of 0.57. Cluster diagram grouped the 30 maize genotypes into three distinct major clusters viz., A, B and C (Fig. 3). Cluster A comprised of 3 genotypes viz., GP(H) 2, GP(H) 28 and SCM15/9LC and was further sub divided into 2 sub- groups (A1; GP(H) 2 and A2; GP(H) 28 and SCM15/9LC). Another cluster, B holds the highest number of genotypes i.e. 24 with three sub- groups viz. B1, B2 and B3. B1, B2 and B3 comprised of 13, 3 and 8 genotypes, respectively. There were only 3 genotypes under cluster C which comprised of inbreds MZ 21(A), MN 30 and MZ 29.
To established genetic relationships among the 30 maize inbred lines based on 52 SSR markers, the PCoA (Principal Coordinate Analysis) using GenAlEx 6.502 software was constructed. The genotypic PCoA revealed that the lines were distributed in all the four quadrangles, showing high degree of genetic variability. A scatter plot generated from the PCoA analysis showed that the first two components accounted for 9.82% and 8.23% of the genetic variation which resulted in a total genetic variation of 18.05% (Fig. 4). Similar works on genetic relationship analysis in maize were also reported by Choudhary et al. 2015, Muthusamy et al. 2015 and Devi et al. 2017b.
Population Structure Analysis And Allelic Patterns Across Populations
The genotypic data of 52 SSR markers were also used to determine the population structure in 30 maize inbred lines using STRUCTURE software. Using STRUCTURE HARVESTER program, the peak plateau of ad hoc measure ΔK was detected at K = 3, which indicated that the entire population can be grouped into three subgroups (Fig. 5).
Population structures of 30 genotypes based on 52 SSR markers are shown in Fig. 5. Different colour within group indicates the proportion of shared ancestry with other group which has the same colour with the admixture. The population structure group of 30 genotypes based on inferred ancestry values were also estimated (Table 2). The percentage of individuals in each population belonging to each group were estimated based on Q-matrix (QST) for each value of K and the cut off probability for assigning to a group was assumed to be 70%. A set of 29 genotypes out of 30 showed the estimated membership coefficients of more than 70% of its genome fraction value and were therefore assigned to a specific group. Whereas, only one genotype was classified as admixture since it showed the estimated membership coefficients of less than 70%. This revealed that almost all the genotypes under study are not admixture. In the present study, only one inbred was under the category of admixture, indicating that it is of mixed pedigree, possibly because of introgression of gene(s) while developing the inbred line (Yin et al., 2019). Group 1 consisted of 11 genotypes out of which 9 (MZ 45 (A), MZ 41 (A), MZ 46 (B), Kolasib 1 (2), MZ 42, MZ 40, MZ 70, MZ 24 and MZ 2) genotypes have originated from Mizoram and only 2 (GP(H) 38 and MN 12) genotypes from Manipur. Group 2 gathered 9 genotypes out of which 6 genotypes (MN 29, MN 1, SCM15/9LC, MN 13, MN 3 and GP (H) 28) originated from Manipur and only 3 genotypes (MZ 56 (A), MZ 48 and MZ 22) have origin in Mizoram. Group 3 has 9 genotypes out of which 5 are originated in Mizoram (MZ 44 (C), MZ 13 (A), MZ 34 (A), MZ 21 (A) and MZ 29) and 4 from Manipur (GP (H) 1, GP (H) 6, GP (H) 2 and MN 30.
In depth knowledge of genetic diversity and population structure of maize would have a big impact on maize improvement program (Liu et al., 2003). The clustering pattern of all the genotypes based on marker information was found to have a close correlation with their pedigree information. The 30 early inbreds were developed from several populations collected from two north- eastern States of India i.e., Manipur and Mizoram. The lines developed from same source population from Manipur fall under same cluster/ sub- cluster while those developed from Mizoram also fall under same cluster/ sub- cluster. For instance, lines developed from populations collected from Manipur viz. GP (H) 2, GP (H) 28 and SCM 15/9LC were together in same cluster i.e., A. Likewise, lines like MZ 2, MZ 70, MZ 24, MZ 42, MZ 40, Kolasib 1(2), MZ 46(B), etc. developed from populations collected from Mizoram clustered together in same sub- group viz., B1. Lines derived from the populations from Manipur were found to be closely placed in genotypic PCoA, while that from Mizoram occupied the nearby axes in the PCoA. To conduct the analysis of molecular variance (AMOVA), the 30 maize inbred lines were grouped into two: Manipur (developed from Manipur collection) and Mizoram (developed lines using Mizoram collection). AMOVA revealed highly significant variance among populations and within individuals of population (Table 3). The overall genetic differentiation indicated that 6% of total variance is due to differences among populations, while 94% of total molecular variance is accounted by within population molecular variance (Fig. 5).
Table 3
Results of analysis of molecular variance (AMOVA) between the two populations
Source | df | SS | MS | Est. Var. | Percentage of variation |
Among Pops | 1 | 0.316 | 0.316 | 0.010 | 6 |
Within Pops | 28 | 4.751 | 0.170 | 0.170 | 94 |
Total | 29 | 5.067 | | 0.180 | 100 |
df: degree of freedom; SS: sum of squares; MS: mean squares; Est. Var. = estimated variance (P value > 0.001) |
The mean effective numbers of allele (Ne) were estimated to be 2.31 and 3.99 for Manipur (MAN) and Mizoram (MIZ) population, respectively (Table 4). The number of alleles (allelic richness) in a population is an important and fundamental measure of genetic variation in a population, and is a useful statistic for identifying populations for conservation (Kalinowski 2004). The Shannon’s information index (I) was calculated as 0.96 and 1.58 for MAN and MIZ population, respectively. Among 189 alleles generated across the 30 genotypes, 3 private alleles (number of alleles unique to a single population) were detected in SSR marker umc 1354. These private alleles were revealed in 3 genotypes originated from Mizoram. The mean values of population parameters such as allelic richness/ number of different alleles (Na), number of different alleles with a frequency > = 5% (Na FrEq. >= 5%), expected Heterozygosity (He) and Unbiased expected Heterozygosity (uHe) were also estimated (Table 4). The results of this study provides useful information to maize breeders for selection of inbred lines and for making promising cross combinations based on genetic distance and clustering.
Table 4
Mean Allelic Patterns Across Populations
Mean values | | |
Population | MAN | MIZ |
Na | 3.000 | 6.000 |
Na FrEq. >= 5% | 3.000 | 6.000 |
Ne | 2.305 | 3.986 |
I | 0.958 | 1.575 |
No. Private Alleles | 0.000 | 3.000 |
He | 0.566 | 0.749 |
uHe | 0.593 | 0.772 |
Na = No. of Different Alleles |
Na (Freq > = 5%) = No. of Different Alleles with a Frequency > = 5% |
Ne = No. of Effective Alleles = 1 / (Sum pi^2) |
I = Shannon's Information Index = -1* Sum (pi * Ln (pi) |
No. Private Alleles = No. of Alleles Unique to a Single Population |
No. LComm Alleles ( < = 25%) = No. of Locally Common Alleles (FrEq. >= 5%) Found in 25% or Fewer Populations |
No. LComm Alleles ( < = 50%) = No. of Locally Common Alleles (FrEq. >= 5%) Found in 50% or Fewer Populations |
He = Expected Heterozygosity = 1 - Sum pi^2 |
uHe = Unbiased Expected Heterozygosity = (2N / (2N-1)) * He |
Agro- Morphological Diversity And Marker- Trait Association
The statistical parameters (mean, standard deviation and correlation coefficients) related to agro- morphological traits of the lines are presented in Tables 5 and 6.The plant height (cm) ranged from 172.00- 286.67 with a mean and standard deviation of 225.88 and 29.68, respectively. The ear height (cm) was recorded between 44.50–170.00 with a mean and standard deviation of 104.18 and 27.70, respectively. For number of kernel rows per ear, the minimum range was 8.83 and maximum was 16.67 with a mean and standard deviation of 12.45 and 1.73, respectively. The number of kernel per row was estimated between 9.50- 54.72 with a mean and standard deviation of 20.78 and 8.16, respectively. The cob length (cm) ranged from 8.75- 17.00 with a mean value and standard deviation of 12.99 and 2.25, respectively. The cob diameter (cm) and 100 kernel weights (g) had recorded minimum value of 1.30 and 14.62 and maximum value of 3.32 and 32.42, respectively with a mean and standard deviation of 2.46 and 22.12 and 0.47 and 4.50, respectively. A positive correlation was found between some important agronomic traits (Table 6). The highest and positive significant correlation was observed between PH and EH (r = 0.76), followed by CD and KW (r = 0.46), PH and KW (r = 0.38), CL and KW (r = 0.37) and EH and KRPE (r = 0.37).
Table 5
Descriptive statistics of seven traits in the 30 early maize inbreds used in the study
Statistic | PH | EH | KRPE | KPR | CL | CD | KW |
Minimum | 172.000 | 44.500 | 8.833 | 9.500 | 8.750 | 1.300 | 14.623 |
Maximum | 286.667 | 170.000 | 16.667 | 54.720 | 17.000 | 3.320 | 32.423 |
Mean | 225.876 | 104.177 | 12.455 | 20.782 | 12.993 | 2.465 | 22.119 |
Standard deviation | 29.683 | 27.695 | 1.726 | 8.158 | 2.247 | 0.472 | 4.498 |
Table 6
Estimation of correlation coefficients at seven agronomic traits in the 30 early maize inbred lines used.
Traits | EH | KRPE | KPR | CL | CD | KW |
PH | 0.758*** | 0.093 | -0.023 | 0.342 | 0.263 | 0.379* |
EH | | 0.371* | 0.074 | 0.075 | 0.188 | 0.175 |
KRPE | | | 0.233 | 0.004 | 0.290 | -0.186 |
KPR | | | | 0.292 | 0.183 | 0.005 |
CL | | | | | 0.265 | 0.374* |
CD | | | | | | 0.463** |
*, ** & *** Significance at P value 0.001, 0.01 & 0.05 respectively |
In the present study, significant marker- trait associations were conducted between the SSR markers and the 7 agro- morphological traits (Table 7).The lowest R2 value (0.06) was detected in umc 1257, mmc 0411, umc 1607 and umc 1002, which was associated with CL, KRPE, CD and KW and the highest (0.21) was in umc 2325, which was associated with KW. The PH was associated with 5 markers viz., umc 1275, umc 1912, mmc 0411, umc 1932 and umc 1962 with R2 value ranged from 0.08–0.13. For EH, the R2 value was ranged from 0.08–0.14 with 3 markers associated with the trait viz., umc 1912, mmc 0411 and umc 1962. The KPR was associated with only 2 markers umc 1153 (0.08) and umc 2318 (0.08).The CL was found to be associated with 4 markers viz., umc 1257, mmc 0411, umc 1031 and umc 1506 withR2 value ranged from0.06- 0.12. In case of KRPE, 4 markers were associated with the trait with R2 value that ranged from 0.06–0.20. Likewise, in CD and KW, 3 markers were linked with these traits with R2 value ranging from 0.06–0.07 and 0.06–0.21, respectively. The SSR markers identified in the present study may serve as useful molecular markers for selecting important agronomic traits in maize breeding.
Table 7
Association analysis of seven agronomic traits with 52 SSR markers in 30 early maize inbred lines
Sl No. | Traits | Markers | R2 values |
1 | Plant Height (PH) | Umc1275 | 0.08 |
Umc1912 | 0.13 |
Mmc0411 | 0.11 |
Umc1932 | 0.08 |
Umc1962 | 0.12 |
2 | Ear Height (EH) | Umc1912 | 0.14 |
Mmc0411 | 0.08 |
Umc1962 | 0.10 |
3 | Number of kernel per row (KPR) | Umc1153 | 0.08 |
Umc2318 | 0.08 |
4 | Cob Length (CL) | Umc1257 | 0.06 |
Mmc0411 | 0.06 |
Umc1031 | 0.12 |
Umc1506 | 0.10 |
5 | Number of kernel rows per ear (KRPE) | Mmc0411 | 0.06 |
Umc2042 | 0.09 |
Umc1962 | 0.20 |
Umc201 | 0.10 |
6 | Cob Diameter (CD) | Umc1607 | 0.06 |
Umc1543 | 0.07 |
Umc2325 | 0.07 |
7 | Kernel Weight (KW) | Umc1002 | 0.06 |
Umc1257 | 0.07 |
Umc2325 | 0.21 |
A high range of variation was found for all the seven agronomic traits which revealed a high level of variation in these lines. The marker- trait association study showed that many markers are associated with the agronomic traits studied and these markers will be useful for selecting lines having good agronomic characteristics. Moreover, these results will be useful in selection of promising inbred lines for use as parents in hybrid breeding program and in marker assisted selection of maize.