Polymorphism of Selected Microsatellite Markers against Breeding Performance Indexes of Polish Large White and Polish Landrace Sows

Abstract

The study aimed to analyze 12 microsatellite markers located in the areas of quantitative trait loci related to litter size in 82 sows, including 45 Polish Large White and 37 Polish Landrace sows, kept on a farm in southwest Poland. Breeding documentation provided data on the total number of piglets born per litter and the numbers of live-born, stillborn and weaned piglets; the corresponding percentage values were also calculated. DNA isolation was performed from 15–20 hairs taken from live animals, and the markers used in the research were divided into four multiplexes. The influence of genotype at a given locus on the reproductive traits was analyzed only for genotypes represented by at least five animals. The results of the analyses for all the sows (treated as the maternal component and not as distinct breeds) showed statistically significant (p ≤ 0.05) differences between the genotypes and the values of the analyzed traits for markers S0008, SW160, SW245, SW714, SW1125 and SW2411. Among these markers, the breed-dependent analysis also showed significant differences in the Polish Large White breed for markers S0008, SW160 and SW1125, and in the Polish Landrace breed for SW245. Additionally, the breed-dependent analysis found significant differences in markers SW903 and SW1808 for the Polish Large White sows, and S0064, SW472 and SW903 in Polish Landrace sows. SW903 was the only marker for which differences in the analyzed reproductive traits differed between genotypes in both breeds (still analyzed separately), although in terms of different traits. The above results indicate the usefulness of microsatellite markers in researching the differentiation of litter size indicators. Although both breeds belong to the maternal component, they showed significant differences in terms of markers. This may suggest difficulties in finding universal (that is, working well for various breeds of the maternal component) markers, indicating the need to look for breed-specific markers, something that calls for further research into numerous animals.

1. Introduction

Marker-Assisted Selection (MAS)—that is, selection with the use of molecular markers—has become a valuable tool in breeding work, enabling the search for Quantitative Trait Loci (QTL) in the genome [1]. One type of molecular markers is microsatellite sequences (MS), also known as short tandem repeats (STR). These consist of simple nucleotide sequences of two to six base pairs (bp) repeated 5 to 20 times, which translate into a fragment of about 100–200 bp [2,3,4]. These markers are evenly distributed in the genome of animals, are inherited according to Mendel’s laws, and show a high degree of polymorphism; they also can be analyzed using the polymerase chain reaction (PCR) [5].

Microsatellite markers have been widely used in research; for example, in the construction of precise genetic maps [6], QTL search for production [7,8] and reproductive traits [9,10,11,12], research on animal affinity and origin [13] and identification of individuals [14] and species [15]. They have also been used in breeding for the assessment of the genetic variability of a population [16] and its structure [17].

In recent years, however, studies on relationships between a specific polymorphic variant of a given STR marker, and importantly from an economic point of view, performance characteristics of animals, have become significantly more frequent. Some researchers have looked for relationships between the genotype of a given marker and average daily milk yield and fat content [18,19], or the number of somatic cells in cow’s milk [20]. Moliński et al. [21] pointed to the possibility of using microsatellite sequences as an effective tool to increase the mass of pectoral muscles, while reducing abdominal and subcutaneous fat in duck carcass muscle. Zatoń-Dobrowolska et al. [22], in turn, investigated the relationship between the polymorphism of microsatellite markers and body weight and selected morphometric features in foxes kept in farm conditions.

Research aimed at searching for microsatellite markers related to important production and reproductive traits has also been conducted in pigs. The presence of dinucleotide repeats in the desmin gene transcript has been showed to influence pork quality [23]. Examining the polymorphism of STR markers in the first intron of adipocyte fatty acid binding protein (A-FABP) and in the third intron of the leptin receptor gene (LEPR), Chmurzyńska et al. [24] found statistically significant differences in production characteristics between different genotypes of sows of the Polish Large White breed, the Polish Landrace breed, and the 990 synthetic line. Another research team, analyzing the microsatellite polymorphism within insulin-like growth factor 1 (IGF1), found significant relationships between individual genotypes and the total number of piglets born per litter, the number of live-born piglets and gestation length [25].

This study aimed to: (i) analyze 12 STR markers located within the peak of quantitative traits loci related to litter size (total number of piglets born in a litter, numbers of live-born and stillborn piglets, and number of weaned piglets) in Polish Large White and Polish Landrace sows, with both breeds constituting the maternal component in trade crossings; and to (ii) identify those markers that can be used in breeding programs to improve selected litter size-related reproductive traits in sows.

2. Materials and Methods

2.1. Material

The study included 82 unrelated (based on breeding information) sows belonging to two breeds: Polish Large White (45 animals) and Polish Landrace (37), both considered the maternal component, kept on a farm located in southwest Poland. The sows were selected randomly from among sows of third to sixth farrowings; in total, 407 farrowings were analyzed. The study included only those sows in which estrus synchronization, litter equalization procedures and cross-fostering were not used. In all sows, artificial insemination was used, each sow having been inseminated twice during the estrus stage. After successful insemination, females were kept in group pens up to the 90th day of gestation, and then moved to individual farrowing pens.

Data on the breeding performance in terms of the total number of piglets born in a litter, the numbers of live-born and stillborn piglets, and the number of weaned piglets (all determined per litter) were obtained from breeding documentation. These data were used to calculate the percentages of live-born piglets per litter (as the number of live-born piglets to the total number of piglets born in the litter), stillborn piglets per litter (as the number of stillborn piglets to the total number of piglets born in the litter) and weaned piglets per litter (as the number of weaned piglets to the number of live-born piglets in the litter).

Note that for the sake of simplicity, wherever we reference a number of piglets (whether in total, live-born, stillborn or weaned), we mean the corresponding number per litter. This also applies to percentage measures; for example, the percentage of stillborn piglets in fact represents the percentage of stillborn piglets per litter.

2.2. Methods

2.2.1. DNA Isolation

For DNA isolation, 15–20 hairs collected from each live animal were used. Genetic material was isolated using the Sherlock AX genomic DNA isolation kit (A&A Biotechnology, Gdańsk, Poland), according to the standard procedure described in the kit protocol [26]. The isolates obtained were suspended in 50 µL of TE buffer and placed in a refrigerator (4 °C) for 24 h, in order to dissolve the DNA in the buffer, after which their quantitative (NanoDrop 2000 spectrophotometer, Thermo Scientific, Gdańsk, Poland)) and qualitative (2% agarose gel with ethidium bromide) assessment was conducted.

2.2.2. Amplification of Microsatellite Sequences

The study used 12 polymorphic microsatellite markers (

Table 1. Characteristics of microsatellite markers: name, primer sequences, and author.

No.	Marker	Forward/Reverse Primers	Primer Sequence	Author
1	S0008	F	GAGGCAGTGTGTTCTATTCA	[27]
		R	RGCCATGTGTAAAGTGTTGCT
2	SW472	F	AAAATGAACCCTCTCCAGTTTC	[27]
		R	TCTGAACACTACAGCCCGC
3	SW460	F	ATTGCACACCTATCTCTATGCG	[27]
		R	AATCTCCATGTGCCGCAG
4	SW1083	F	CCTTGCTGGCCTCCTAAC	[27]
		R	CATACTCCAAAATTTCTATGTTGA
5	S0064	F	TGAGCTGGAGGTTAGCTACC	[27]
		R	TGTCAGAAAGACTGCTTGCG
6	SW245	F	TGGTGCTAGCAGAACCTGTG	[27]
		R	AAACCTGGCAACACAGCAG
7	SW903	F	TTTCTTTGACAGTTGTGCAAGG	[27]
		R	TGAACTACAGCAGCGACCTG
8	SW1125	F	TAGATGTATATACTTCCATGTGTG	[27]
		R	ATGTTGAGCTCTTAATTTTATACA
9	SW1808	F	CCAAAAAAGTGGACTGTAAGCC	[28]
		R	TACGGATGGATGGAGACAGG
10	SW2411	F	CCTGGACTCATTCTTGCTTTG	[28]
		R	TTCCTATTCTGTCCTGCCTTG
11	SW160	F	TCTTCCTTGTCATACATGCCC	[27]
		R	ACTAGACAGCCAGGGTGGG
12	SW714	F	ATCTCCTTGTTAGAACTTGTGTGTG	[27]
		R	GAGATGAATATGGGGAAAATGAAC

) divided into four multiplexes, taking into account the hybridization temperature and the type of fluorescent marker used (FAM or HEX) (

Table 2. Characteristics of microsatellite markers divided into individual multiplexes, the type of marker used, and the temperature of primer attachment.

Multiplex	Marker	Chromosome	Amplicon Length [bp]	Fluorescent Marker	Hybridization Temperature (°C)
I	S0008	1	177–191	FAM	60
	SW472	7	95–111	FAM
	SW460	3	159–199	HEX
	SW1083	7	117–147	HEX
II	S0064	7	93–160	FAM	58
	SW245	14	106–130	HEX
	SW903	11	195–201	HEX
III	SW1125	14	117–141	FAM	60
	SW1808	18	106–147	HEX
	SW2411	16	196–216	FAM
IV	SW160	3	128–132	HEX	60
	SW714	4	145–169	FAM

). They were then amplified using the multiplex PCR method using Multiplex PCR Kit (Qiagen^®, Germantown, MD, USA).

The mix of primers used in the PCR reaction was prepared by taking 1.5 µL of each primer included in each multiplex, and then supplementing it with the TE buffer to a volume of 50 µL. PCR reactions were performed in a volume of 10 µL, using a Bio-Rad thermocycler. The reaction mixture included 5 µL Qiagen^® Multiplex PCR Kit, 2 µL nuclease-free DEPC water, 1 µL primer mix, and 2 µL template DNA. The cycling profile was as follows: 95 °C for 15 min, followed by 30 cycles of 94 °C for 30 s, 58–60 °C for 90 s, 72 °C for 1 min, and finally 60 °C for 25 min.

2.2.3. Genotyping of PCR Products

After qualitative assessment, PCR reaction products were sent to GENOMED S.A., in order to be read as fluorograms using the GeneScan procedure. The results obtained were then read using the GeneMarker 2.4.2 program (Soft Genetics LLC^®, State College, PA, USA), to establish the genotype of the individual within each marker.

2.3. Statistical Analysis

Statistical analysis was performed in R version 3.4.4 software [29]. Two types of analyses were conducted: One analyzed all the sows together (i.e., the breed was ignored), while the other analyzed the sows by breed. Summary statistics were determined using the pastecs R package [30]. The variables’ distributions were tested against the normality distribution using the Shapiro-Wilk test. The homogeneity of the traits’ variance in the analyzed groups was verified using the Bartlett test.

To analyze the influence of genotypes at a given locus on the considered reproductive traits, only genotypes represented by at least five animals were considered. Since the traits violated ANOVA assumptions of normality and variance homogeneity, for this analysis the non-parametric Kruskal-Wallis test was used, with a post-hoc procedure of pairwise comparisons implemented in the pgirmess package [31] of R [29]. Principal component analysis (PCA) for microsatellite markers was applied using the adegenet [32], ade4 [33,34,35] and factoextra [36] R packages.

3. Results

3.1. Analysis of Genetic Markers across All Sows

Table 3. Summary statistics of the traits related to litter size in the sows analyzed, across both breeds studied.

Trait	Mean	Median	Min	Max	SD	CV (%)
Total no. of born piglets	14.05	14.00	10.67	19.33	1.48	10.56
Number of live-born piglets	12.88	13.00	7.00	17.00	1.48	11.47
Percentage of live-born piglets	92.32	94.08	41.18	100.00	0.08	8.64
Number of stillborn piglets	1.18	1.00	0.00	8.00	1.14	96.63
Percentage of stillborn piglets	6.10	7.67	0.00	58.82	0.07	101.00
Number of weaned piglets	11.20	11.19	8.20	14.60	1.45	13.00
Percentage of weaned piglets	89.60	87.50	62.76	100.00	0.08	9.21

SD—standard deviation; CV—coefficient of variation.

presents summary statistics of the reproductive traits related to litter size, across all sows studied. Out of the 12 microsatellite markers analyzed, for six markers we found statistically significant (p ≤ 0.05) differences between genotypes in terms of the traits related to litter size: the total number of live-born piglets (SW2411), the percentage of live-born piglets (S0008, SW245, and SW714); the number and percentage of stillborn piglets (S0008, SW245, and SW714), the number of weaned piglets (S0008, SW160, SW245, SW714, SW1125, and SW2411), and the percentage of weaned piglets (S0008, SW160, and SW1125). We will discuss detailed results for only three of these markers: S0008, SW245 and SW714, because they were related with the most analyzed traits (

Table 4. Reproductive performance indicators in sows, across both breeds studied, depending on their genotype and the analyzed microsatellite markers.

Marker	Genotype	n	Trait
			Mean	Median	Min	Max	SD	CV (%)
Percentage of live-born piglets
S0008	183/185	22	92.32 ab	95.41	41.18	100.00	0.12	13.28
	185/185	8	96.02 a	96.88	86.67	100.00	0.05	4.87
	185/187	24	89.68 b	90.62	73.05	100.00	0.07	7.06
	185/193	11	94.65 ab	95.82	87.50	100.00	0.04	3.88
SW245	129/129	11	91.23 a	91.22	84.64	97.06	0.04	4.32
	129/131	19	90.86 ab	94.91	41.18	100.00	0.14	15.80
	131/131	12	94.46 ab	94.85	90.14	100.00	0.04	3.33
	131/133	10	97.43 b	97.00	95.05	100.00	0.02	1.76
SW714	143/151	12	91.65 ab	91.10	84.86	87.00	0.04	4.28
	151/151	18	88.21 a	89.31	73.05	100.00	0.07	7.75
	151/155	16	94.68 b	95.17	87.50	100.00	0.03	3.30
	151/167	8	90.17 ab	91.96	73.77	98.61	0.09	9.94
	155/157	7	96.86 b	97.33	91.67	100.00	0.03	3.62
Number of stillborn piglets
S0008	183/185	22	1.15 ab	0.75	0.00	8.00	1.66	143.74
	185/185	8	0.60 a	0.45	0.00	2.00	0.71	117.65
	185/187	24	1.51 b	1.35	0.00	5.00	1.03	68.47
	185/193	11	0.86 ab	0.67	0.00	2.00	0.61	71.21
SW245	129/129	11	1.49 a	1.41	0.50	2.80	0.74	49.58
	129/131	19	1.34 ab	0.87	0.00	8.00	1.95	145.22
	131/131	12	0.82 ab	0.75	0.00	1.62	0.50	60.72
	131/133	10	0.39 b	0.45	0.00	0.67	0.26	66.97
SW714	143/151	12	1.27 ab	1.28	0.60	2.50	0.58	45.72
	151/151	18	1.88 a	1.75	0.00	5.00	1.18	62.95
	151/155	16	0.84 b	0.75	0.00	2.00	0.52	60.83
	151/167	8	1.13 ab	1.20	0.25	2.00	0.68	59.77
	155/157	7	0.53 b	0.40	0.00	1.25	0.57	108.31
Percentage of stillborn piglets
S0008	183/185	22	7.84	4.79	0.00	58.82	0.12	155.34
	185/185	8	3.98	3.12	0.00	13.33	0.05	117.28
	185/187	24	9.80	9.38	0.00	26.95	0.06	61.63
	185/193	11	5.35	4.18	0.00	12.50	0.04	68.66
SW245	129/129	11	9.67	8.78	0.03	15.61	0.04	45.13
	129/131	19	9.14	5.09	0.00	58.85	0.14	156.96
	131/131	12	5.51	5.15	0.00	9.86	0.03	57.06
	131/133	10	2.57	3.00	0.00	4.35	0.17	66.55
SW714	143/151	12	8.35	8.90	0.03	15.14	0.04	46.92
	151/151	18	11.79	10.69	0.00	26.95	0.07	57.95
	151/155	16	5.32	4.83	0.00	12.50	0.03	58.65
	151/167	8	7.75	8.04	0.00	13.73	0.05	63.82
	155/157	7	3.49	2.67	0.00	8.33	0.04	108.39
Number of weaned piglets
S0008	183/185	22	11.29 ab	11.58	8.43	13.33	1.35	12.03
	185/185	8	12.34 a	12.29	11.00	13.33	0.75	6.12
	185/187	24	10.59 b	10.24	5.58	13.50	1.40	13.31
	185/193	11	12.44 a	12.33	10.87	14.60	1.05	8.49
SW245	129/129	11	10.67 ab	10.40	8.58	12.33	1.40	13.13
	129/131	19	11.98 a	12.00	8.42	14.60	1.49	12.41
	131/131	12	10.49 b	10.56	9.55	12.00	0.71	6.80
	131/133	10	12.24 a	12.30	9.80	13.50	1.22	9.98
SW714	143/151	12	10.49 a	10.60	8.58	13.50	1.23	12.10
	151/151	18	10.24 a	10.14	8.43	12.00	1.03	10.60
	151/155	16	12.00 b	12.00	10.00	14.60	1.15	9.63
	151/167	8	11.50 ab	12.20	8.75	13.50	1.98	17.22
	155/157	7	12.07 b	12.00	11.00	12.75	0.69	5.71
Percentage of weaned piglets
S0008	183/185	22	88.86 ab	92.26	74.03	95.49	0.06	7.18
	185/185	8	94.10 a	94.93	84.62	100.00	0.06	5.93
	185/187	24	85.04 b	85.39	62.76	96.67	0.09	10.37
	185/193	11	91.35 ab	92.26	78.75	96.25	0.04	5.48

^a,b different letters indicate statistically different means at p ≤ 0.05; SD—standard deviation; CV—coefficient of variation.

). The information on the other markers that showed significant relation to some of the traits is presented in Table S1.

Sows with genotypes 185/185 (S0008), 131/133 (SW245) and 155/157 (SW714) gave birth to the fewest stillborn piglets per litter (0.60, 0.39 and 0.53, respectively), which translated into the highest percentage of live-born piglets (96.0, 97.4 and 96.9%, respectively) and the lowest percentage of stillborn piglets (4.0, 2.6 and 3.5%, respectively). Sows with genotypes 131/133 (SW245) and 155/157 (SW714) also reared the most numerous litters. In the case of marker S0008, most piglets were weaned from sows of genotypes 185/193 and 185/185 (12.4 and 12.3, respectively). The lowest reproductive performance indicators were achieved by sows with genotypes 185/187 (S0008), 129/129 (SW245) and 151/151 (SW714).

For markers showing statistically significant relationships between the genotype and the reproductive traits studied (SW2411, S0008, SW245, SW714, SW160, and SW1125), we applied principal component analysis (PCA). However, here we show only the results for marker SW245, for which we observed the greatest between-genotype variation in terms of the traits studied; this analysis included the 52 animals with this genotype.

The PCA analysis for maker SW245 showed that the three first components (PC1–PC3) explained 99.3% of the variation in the data (

Table 5. PCA for marker SW245: Eigenvalues, variance explained, cumulative variance explained, and correlation coefficients with the original traits of the first three principal components (PCs).

Trait	PC1	PC2	PC3
Total no. of born piglets	0.08	0.99	−0.04
Number of live-born piglets	−0.41	0.91	0.07
Percentage of live-born piglets	−0.95	−0.11	0.24
Number of stillborn piglets	0.94	0.23	−0.23
Percentage of stillborn piglets	0.96	0.11	−0.25
Number of weaned piglets	−0.75	0.49	−0.45
Percentage of weaned piglets	−0.56	−0.42	−0.71
Eigenvalue	3.75	2.31	0.89
Variation (%)	53.59	33.03	12.72
Cumulated variation (%)	53.59	86.62	99.34

). The first principal component explained over 50% of the variation and was positively related to the number and percentage of stillborn piglets, and negatively related to the percentage of live-born piglets and the number of weaned piglets. Explaining over 30% of the variation in the data, the second principal component was positively correlated with the total number of piglets per litter and the number of live-born piglets. The third principal component was negatively correlated with the percentage of weaned piglets.

The analysis of PC1 against PC3 obtained for marker SW245 (Figure 1) showed clear differences in the number and percentage of stillborn piglets between sows with genotypes 131/133 and 129/129, corresponding to the results presented in Table 4.

3.2. Analysis of Genetic Markers in Sow Breeds

3.2.1. Polish Large White

Table 6. Summary statistics of the traits related to litter size in the Polish Large White sows.

Trait	Mean	Median	Min	Max	SD	CV (%)
Total no. of born piglets	14.06	14.00	11.40	19.33	1.60	11.40
Number of live-born piglets	12.68	12.80	7.00	17.00	1.61	12.75
Percentage of live-born piglets	91.00	92.60	41.18	100.00	0.09	10.21
Number of stillborn piglets	1.40	1.28	0.00	8.00	1.35	96.17
Percentage of stillborn piglets	9.20	7.42	0.00	58.82	0.09	101.40
Number of weaned piglets	10.70	10.40	8.20	13.50	1.33	12.47
Percentage of weaned piglets	85.00	86.30	62.76	100.00	0.09	10.18

SD—standard deviation; CV—coefficient of variation.

shows summary statistics of the traits related to litter size in Polish Large White sows. For this breed, five markers showed statistically significant differences between genotypes (p ≤ 0.05), for the following traits: the total number of piglets born in the litter (SW903), the percentage of live-born piglets (S0008), the number and the percentage of stillborn piglets (S0008), the number of weaned piglets (S0008, SW1125, and SW1808), and the percentage of weaned piglets (SW160 and SW1808). Here, we will only show the analyses for the markers associated with at least two reproductive traits (

Table 7. Reproductive performance indicators of Polish Large White sows, depending on their genotype and the analyzed microsatellite markers.

Marker	Genotype	n	Trait
			Mean	Median	Min	Max	SD	CV [%]
Percentage of live-born piglets
S0008	183/185	13	91.01 a	94.14	41.18	100.00	0.15	16.81
	185/187	17	88.98 b	89.68	73.05	100.00	0.06	7.24
Number of stillborn piglets
S0008	183/185	13	1.33 b	0.83	0.00	8.00	2.07	155.39
	185/187	17	1.71 a	1.50	0.00	5.00	1.14	66.42
Percentage of stillborn piglets
S0008	183/185	13	9.00 a	5.86	0.00	58.82	0.15	170.19
	185/187	17	11.02 b	10.31	0.00	26.92	0.06	58.48
Number of weaned piglets
S0008	183/185	13	11.54 b	12.20	9.60	13.00	1.25	10.87
	185/187	17	10.32 a	10.14	8.58	13.50	1.22	11.85
SW1808	132/132	11	11.90 a	12.22	9.67	13.00	1.07	9.03
	150/152	15	10.03 b	9.85	9.17	11.14	0.63	6.26
Percentage of weaned piglets
SW1808	132/132	11	93.11 a	93.99	84.62	100.00	0.04	4.60
	150/152	15	83.51 b	84.21	69.04	94.18	0.07	8.13

^a,b different letters indicate statistically different means at p < 0.05; SD—standard deviation; CV—coefficient of variation.

); Table S2 shows the results for the other significant markers.

The sows with genotype 183/185 (S0008) gave birth to fewer stillborn piglets per litter than did sows with genotype 185/187 (1.33 versus 1.71), and thus the former had a lower percentage of stillborn piglets (9.0 versus 11.0%) and a higher percentage of live-born piglets (91.0 versus 89.0%). Homozygous females of genotype 132/132 compared to heterozygous females of genotypes 150/152 (SW1808) weaned more piglets per litter, but also showed a greater percentage of weaned piglets by nearly 10 percentage points (Table 7).

For the five markers for which we detected statistically significant influence of the genotype on the reproductive traits analyzed, we applied PCA. Below, however, we report only the results for marker SW1808, which showed the greatest influence of the genotype on the traits studied.

PCA for marker SW1808 showed that the first three principal components accounted for as much as 99.6% of the variation in the data (

Table 8. PCA for marker SW1808 in Polish Large White sows: Eigenvalues, variance explained, cumulative variance explained, and correlation coefficients with the original traits of the first three principal components (PCs).

Trait	PC1	PC2	PC3
Total no. of born piglets	−0.11	−0.97	0.19
Number of live-born piglets	−0.67	−0.73	0.08
Percentage of live-born piglets	−0.94	0.22	−0.19
Number of stillborn piglets	0.92	−0.31	0.18
Percentage of stillborn piglets	0.95	−0.22	0.19
Number of weaned piglets	−0.59	−0.13	0.79
Percentage of weaned piglets	0.03	0.61	0.79
Eigenvalue	3.48	2.07	1.39
Variation (%)	49.67	29.61	19.86
Cumulated variation (%)	49.67	79.28	99.14

). Explaining nearly 50% of the variation, PC1 was negatively correlated with the percentage of live-born piglets, and positively correlated with the number and percentage of stillborn piglets. PC2 explained almost 30% of the variability, which was due to its negative correlation with the total number of piglets born per litter and the number of live-born piglets. PC3 was positively correlated with the number and percentage of piglets weaned per litter.

Figure 2 graphs PC1 compared to PC3 for marker SW1808. This analysis showed clear differences in the number and percentage of weaned piglets between sows of genotypes 132/132 and 150/152 (in the context of marker SW1808), an observation corresponding with the results presented in Table 7.

3.2.2. Polish Landrace

Table 9. Summary statistics of the traits related to litter size in the Polish Landrace sows.

Trait	Mean	Median	Min	Max	SD	CV (%)
Total no. of born piglets	14.02	14.00	10.67	17.33	1.34	9.58
Number of live-born piglets	13.13	13.00	9.71	15.67	1.27	9.65
Percentage of live-born piglets	94.00	95.40	73.77	100.00	0.06	6.08
Number of stillborn piglets	0.90	0.75	0.00	3.15	0.74	82.07
Percentage of stillborn piglets	5.81	4.61	0.00	21.05	0.04	80.82
Number of weaned piglets	11.79	12.00	8.42	14.60	1.37	11.66
Percentage of weaned piglets	90.60	92.30	76.14	100.00	0.06	6.70

SD—standard deviation; CV—coefficient of variation.

shows summary statistics of the traits related to litter size in Polish Landrace sows. The analysis of genotypes within this breed for the four markers showed statistically significant differences (p ≤ 0.05) for the following traits: the percentage of live-born piglets (S0064 and SW472), the number of stillborn piglets (S0064, SW245, and SW472), and the percentage of weaned piglets (SW903). As we did for the Polish Large White breed, we present here detailed results only for the markers associated with at least two reproductive traits (

Table 10. Reproductive performance indicators of Polish Landrace sows, depending on their genotype and the analyzed microsatellite markers.

Marker	Genotype	n	Trait
			Mean	Median	Min	Max	SD	CV (%)
Percentage of live-born piglets
SW472	89/91	19	94.34 a	95.39	84.64	100.00	0.04	4.18
	91/91	11	97.17 b	97.06	91.67	100.00	0.03	2.80
S0064	107/107	20	93.40 a	95.17	78.95	100.00	0.5	5.16
	107/109	10	98.44 a	97.78	97.06	100.00	0.01	1.47
Number of stillborn piglets
SW472	89/91	19	0.94 a	0.75	0.00	2.80	0.67	70.92
	91/91	11	0.47 b	0.50	0.00	1.25	0.44	94.00
S0064	107/107	20	1.04 b	0.75	0.00	3.14	0.73	69.58
	107/109	10	0.22 a	0.33	0.00	0.40	0.21	92.07
Percentage of stillborn piglets
SW472	89/91	19	5.85 a	4.61	0.00	15.36	0.04	63.93
	91/91	11	2.99 b	2.94	0.00	8.33	0.03	96.49
S0064	107/107	20	6.69 b	5.01	0.00	21.05	0.05	71.68
	107/109	10	1.56 a	2.22	0.00	2.91	0.01	92.66

^a,b different letters indicate statistically different means at p < 0.05; SD—standard deviation; CV—coefficient of variation.

). Information on the other markers of significant importance is presented in (Table S3).

Genotypes 91/91 (SW472) and 107/109 (S0064) were associated with high percentages of live-born piglets (97.2 and 98.4%, respectively), and with the low numbers of stillborn piglets (0.47 and 0.22, respectively) and stillborn percentages (3.0 and 1.6%, respectively).

In PCA for marker S0064, which showed the greatest variation in the values of the studied traits, the first three principal components explained as much as 99.9% of the variation in the data (

Table 11. PCA for marker S0064 in Polish Landrace sows: Eigenvalues, variance explained, cumulative variance explained, and correlation coefficients with the original traits of the first three principal components (PCs).

Trait	PC1	PC2	PC3
Total no. of born piglets	−0.05	−0.99	0.02
Number of live-born piglets	−0.53	−0.84	0.12
Percentage of live-born piglets	−0.95	0.24	0.21
Number of stillborn piglets	0.92	−0.33	−0.19
Percentage of stillborn piglets	0.95	−0.24	−0.20
Number of weaned piglets	−0.71	−0.56	−0.41
Percentage of weaned piglets	−0.39	0.32	−0.86
Eigenvalue	3.58	2.36	1.04
Variation [%]	51.17	33.70	14.98
Cumulated variation [%]	51.17	84.87	99.85

). PC1 explained over 50% of the variability, which was due to its relation with four traits: negative correlation with the percentage of live-born piglets and the number of weaned piglets, and positive correlation with the number and the percentage of stillborn piglets. Explaining over 30% of the variation, PC2 was negatively correlated with the total number of piglets per litter and the number of live-born piglets. PC3 was negatively correlated with the percentage of weaned piglets per litter.

Figure 3 shows PC1 plotted against PC2. The analysis showed clear differences in the number and percentage of weaned piglets per litter between Polish Landrace sows of genotypes 107/107 and 107/109 in marker S0064; this observation corresponds with the results presented in Table 10.

Marker SW903 was the only microsatellite marker for which we observed significant differences in the litter-related reproductive traits between genotypes in both breeds. In Polish Large White sows, this difference was related to litter size; while in Polish Landrace sows the difference was related to the number of weaned piglets per litter. Interestingly, this marker did not show significant differences between genotypes in the analysis conducted across the two breeds.

4. Discussion

The current national breeding program for Polish Large White and Polish Landrace sows—both considered the maternal component—is based mainly on classic selection methods related to phenotypic traits (number of teats—16, age of first farrowing—340 days, number of live-born piglets in a litter—14, number of piglets on their 21st day of life—13, average daily gain—min. 680 g, and meat content at the level of 58%) [37]. Molecular biology techniques, however, make it possible to extend classic selection methods with methods related to the use of genetic markers (Marker Assisted Selection, MAS). Enabling the assessment of individuals of both sexes at a very young age, MAS is of particular importance to species characterized by long intergenerational intervals, and also to the process of improving traits that are revealed later in the life of the animal, including reproductive traits. Currently, however, the national breeding program for Polish Large White and Polish Landrace sows relies on MAS only to a minimal extent; it only assumes that these females cannot be carriers of the stress sensitivity gene RYR1^T [37].

One of the types of markers used in the selection of MAS are microsatellite sequences, which due to their even distribution in the genome, inheritance according to Mendel’s laws, and above all a high degree of polymorphism and the ease of analyzing with the use of PCR, surpass other types of molecular markers, including single nucleotide polymorphism (SNP) markers [5]. Even though microsatellite markers are often considered neutral molecular markers, much research in the last dozen or so years has analyzed relationships between specific alleles or genotypes within a given marker, and production and reproductive traits of various species of farm animals. Such studies have been carried out in cattle [18,19,20], sheep [38], goats [39], poultry [21,40] and pigs. Most studies in pigs have focused on production traits related to meat quality [23,24], but studies on reproductive traits have also been conducted [25,41,42].

In our analysis of sows of both breeds considered together (as a maternal component), we found the greatest influence of female genotype within the microsatellite marker on reproductive traits related to litter size for markers S0008, SW245 and SW714; the genotype-dependent traits were the percentage of live-born piglets, the number and the percentage of stillborn piglets, and the number of weaned piglets per litter. We also analyzed the two breeds separately. Marker S0008 also showed significant differences in Polish Large White sows for these traits, except for the number of weaned piglets. These results are difficult to analyze using the available literature, since—to the best of our knowledge—these are the first results showing relationships between reproductive traits of sows and the polymorphism of the abovementioned markers.

Tribout et al. [43], however, studied the area between markers SW245 and SW1125 on SSC14. They mapped QTL for the number of stillborn piglets per litter. Their results corresponded to our study, since when we analyzed the two breeds together, we showed that sows of genotype 131/133 within the marker SW245 gave birth to greater than one stillborn piglet per litter less than did sows of genotypes 129/129 and 129/131 (0.34, 1.49 and 1.39, respectively). This directly translated into a higher percentage of live-born piglets per litter, and a lower percentage of stillborn piglets observed in sows of genotype 131/133. We observed similar results within this marker in the number of stillborn piglets for the Polish Landrace breed. The analysis of all sows together did not show a significant relationship between the genotype of the SW1125 marker and the number of stillborn piglets, but it did show statistically significant differences (p ≤ 0.05) in the number and the percentage of weaned piglets. Sows with the 135/139 genotype weaned smaller litters than did sows with either the 113/125 or the 121/125 genotype (9.33 versus 12.08 and 11.83, respectively); they also had a lower percentage of weaned piglets (81.6 versus 88.5 and 92.50%, respectively), although similar to that of sows with genotype 139/139 (82.0%). This may suggest that sows with the 139 bp allele are characterized by lower fertility and weaker maternal instincts.

The abovementioned research team [43] also identified a potential QTL related to the number of live-born piglets per litter on SSC16, located between markers S0111 and SW2411; and on SSC18, located in the vicinity of marker SW1808. The results of our re-search indicate that within marker SW2411, sows with genotypes 176/202 and 202/202 gave birth to more piglets per litter than did females with genotype 176/198 (13.76, 13.65 and 12.09, respectively). Considering that all analyzed sows—except for those with genotype 176/198 bp—had the 202 bp allele, we can assume that the shorter length of the microsatellite marker decreases the perinatal survival of piglets. Lugovoy et al. [44] con-firmed this finding, indicating that in the case of marker SW951 (SSC10), sows with the 128 bp allele gave birth to more piglets per litter than did sows with the 122 and 120 bp alleles (11.90, 11.07, 10.80 piglets, respectively). In addition, sows that had the 113 bp allele (marker SW24, SSC17) weaned an average of 9.51 piglets per litter, while those with the 93 bp allele weaned only 7.96 piglets per litter. Li et al. [42] presented similar results. They found that with the increase in the number of base pairs within a microsatellite marker, the number of weaned piglets per litter increased (9.20 for 98/98 homozygotes, versus 10.62 for 118/118 homozygotes). This observation partially agrees with our results, because homozygous 202/202 (SW2411) sows weaned the largest litters. On the other hand, their result disagrees with our results obtained for marker SW1808, for which sows with the 132/132 genotype gave birth and weaned larger litters than those with the 150/152 genotype. Note, however, that the small number of the analyzed sows and the demonstration of statistically significant differences for only the Polish Large White breed call for further and broader analyses of the SW1808 marker.

Marker SW903 was the only one for which we detected a significant effect on reproductive traits for both breeds. This may indicate this marker’s usefulness in breeding involving a maternal component. For Polish Large White sows, the effect was related to the total number of piglets per litter, while for Polish Landrace sows, to the number of weaned piglets. This result corresponds somewhat with the results of Tribout et al. [43], who indicated the area between markers SW1415 and SW903 on SSC11 as a probable QTL for the number of stillborn piglets per litter. Cassady et al. [45] mapped a putative QTL for the number of live-born piglets per litter on the same chromosome, at a position between 61 and 81 cM, which includes the location of marker SW903.

Quantitative trait loci for litter size during the first farrowing were detected on SSC7 in the region between markers S0025 and S0064 [46], a result corresponding to that by Tribout et al. [43], who also indicated QTL for the number of live-born piglets per litter on the same chromosome, but in the region between markers S0383 and S0064. We showed that heterozygous Polish Landrace sows within the marker S0064 gave birth to fewer stillborn piglets than did homozygous females, which directly translated into a higher percentage of live-born piglets and a lower percentage of stillborn piglets. The obtained results seem promising, because in the area between markers SW1354 and SW1369, where the analyzed marker S0064 is contained, there is a locus of the properdin gene, which has been shown to be related to litter size [47] and QTLs for the weight of the ovaries [12] and ovulation rate [10].

Li et al. [42] mapped a QTL for the number of stillborn piglets per litter on SSC7 at position 59 cM. This result agrees with our studies for the marker SW472, located at position 58.9 cM, for which in the case of Polish Landrace sows, we showed a significant relationship between sow genotype, the number and percentage of stillborn piglets, and the percentage of live-born piglets. Lugovoy et al. [44] analyzed the polymorphism of the S0101 marker, which is also located on the SSC7 at position 139.4 cM. They found no statistically significant differences in litter size traits between sows with different genotypes. However, they showed that individuals with 209 and 213 bp alleles weaned, on average, more piglets per litter than did the other sows.

5. Conclusions

The results of the research confirmed that litter-size indicators depend on the female genotype within the analyzed microsatellite loci and allowed us to identify markers significantly related to reproductive traits among Polish Large White and Polish Landrace sows, with both breeds representing the maternal component. Among these markers, we were able to select those alleles and genotypes which performed better in terms of the reproductive traits related to litter size. The separate analyses for the two breeds additionally showed breed-specific sequences that were related to the traits studied. The between-breed differences in significant markers may result from the genetic history of the breeds, because although both breeds belong to the maternal component, they were produced based on different breeds, with other different breeds used to improve them. However, due to the relatively small number of sows used in our study, further analysis is required; but such research could be limited only to markers for which we detected significant between-genotype differences and should use a greater number of subjects. Such an approach would allow for a better association of markers with reproductive traits and would also enable the identification of specific alleles that should be included in breeding with the aim of improve breeding efficiency focused on breeds representing the maternal component in production crossbreeding.