Heritability of udder morphology and colostrum quality traits in swine¹

ABSTRACT

The heritability of udder quality traits, defined as morphology and colostrum IgG concentration at farrowing, was estimated together with the genetic and phenotypic correlations of these traits with other production and reproduction criteria. Udder morphology traits were recorded in 988 Meidam sows and colostrum samples were collected from 528 sows. Teat length, teat diameter (DIA), interteat distance within the same row (SAMER), and teat distance from the abdominal midline (AML) were recorded to the nearest millimeter. For each sow, a record was also made of udder development score (DEV), the proportion of teats oriented perpendicular to the udder, and the proportion of nonfunctional teats. Colostrum IgG concentration was estimated with a Brix refractometer. Heritability of udder morphology traits varied from high (h² = 0.46 for teat length and h² = 0.56 for DIA) to moderate (h² = 0.37 for SAMER, h² = 0.22 for AML, h² = 0.25 for DEV, h² = 0.3 for the proportion of nonfunctional teats, h² = 0.1 for the proportion of teats oriented perpendicular to the udder, and h² = 0.35 for colostrum IgG concentration). The SAMER was negatively genetically correlated with the number of stillborns (genetic correlation [r_g] = −0.48) and positively genetically correlated with the number of piglets born alive (r_g = 0.69), with the opposite for the trait AML (r_g = −0.40 for number of piglets born alive and r_g = 0.40 for stillborns). The highest genetic correlation with productive traits was estimated between AML and ADG during rearing (r_g = 0.42), although this had a negative phenotypic correlation (r_p; −0.11). Teat length was also moderately correlated with ADG (r_g = 0.27). Backfat thickness at 100 kg was positively correlated with DIA and the total number of teats present in both rows (r_g = 0.28 and r_g = 0.36, respectively) and negatively correlated only with DEV (r_g = −0.22). The same results were found for the phenotypic correlation between backfat thickness at end of test and the total number of teats present in both rows (r_p = 0.03). Udder quality traits can be included in the breeding goal and appropriately weighted with other important traits in the breeding objectives to enhance maternal performance.

colostrum quality, genetic correlation, heritability, phenotypic correlation, sow, udder morphology

INTRODUCTION

The past 30 yr have seen increasingly rapid advances in the field of swine genetic selection for efficient feed conversion, high growth rate, carcass leanness, and sow prolificacy. Unfortunately, these traits often have unfavorable genetic correlations with piglet mortality around farrowing (Canario et al., 2007; Tribout et al., 2010) and with sow behavior (Rauw et al., 1998; Rauw, 2007; Canario et al., 2012, 2014). Rydhmer (2000) reviewed sow selection criteria and stated that the present genetic increase in number of piglets born must be accompanied by improvements in sow maternal ability traits in order to increase overall production and welfare. Traits considered beneficial for production and welfare are often similar across livestock sectors, although the most valuable traits vary depending on the intended use of the animal and the species in question. It has been established that some udder characteristics in dairy cattle are controlled by both genetic and environmental factors (Hickman, 1964). Moreover, interest has increased in udder traits in sheep (Marie-Etancelin et al., 2005; Casu et al., 2006) to consider functional traits such as milkability. In pig production, udder morphology has not received the same attention, despite its fundamental role in nursing piglets. To date, only teat number (Hirooka et al., 2001; Ding et al., 2009) and teat functionality (Chalkias et al., 2013) are traits included in swine breeding programs. The heritability of udder morphology and the way that these traits relate to important traits for production (e.g., growth rate, backfat thickness) and reproduction (e.g., gestation length, litter size) have never been subject to genetic evaluation. The objective of this work was, therefore, to estimate the heritability of udder morphology traits and colostrum quality and their correlation with other production and reproduction traits.

MATERIAL AND METHODS

Animals

The study was approved by the Newcastle University Animal Welfare and Ethical Review Body and was carried out at the ACMC Ltd. breeding company (Beeford, UK) from July 2014 to February 2015. The study population consisted of 988 Meidam (crossbreed Large White × Meishan) sows. Animals were moved from the group gestation house to the farrowing unit at 110 d after insemination, where they were kept in individual crates equipped with a feeder and drinker. Ambient temperature in farrowing rooms averaged 21°C. No specific interventions were applied in the study; feed, environment, and management were maintained as standard commercial practice. Data were collected from 10 farrowing batches; in total, 988 sows of different parities (230 sows in parity 1, 196 sows in parity 2, 141 sows in parity 3, 170 sows in parity 4, 68 sows in parity 5, 70 sows in parity 6, and 113 sows with more than 6 parities [parity 7, 8, 9, and 10]) were observed in 3 units (321 sows at unit B, 339 sows at unit F, and 327 sows at unit U).

Udder Traits

All these traits were recorded on 1 occasion per sow, at 1 to 3 d prior to farrowing, from the upper row of teats while the sow was in a lying down posture. The total number of teats present in both rows, teat length (LEN), teat diameter (DIA), interteat distance within the same row (SAMER), and teat distance from the abdominal midline (AML) were recorded with a ruler and a caliper (DIA), and measurements were taken to the nearest millimeter (Table 1). The methodology used for measuring morphological parameters was that described by Balzani et al. (2016a,c). For each sow, a record was also made of udder development score (from 1 to 3, where 1 was defined as the udder not being developed and mammary glands being not defined, 2 was defined as the udder begin well developed but the mammary glands being not clearly distinct, and 3 was defined as the udder being well developed and mammary glands being clearly distinct), of the proportion of teats not oriented perpendicular to the udder (score of 0 or 1, where 0 was defined as the teat not being orientated perpendicular to the mammary gland and 1 was defined as the teat being orientated perpendicular to the mammary gland), and of the proportion of nonfunctional teats (A score of 0 was defined as a nonfunctional teat with the milk channel not working, including teats that were blind [teats that were injured early in the life of the pig and remained as a small protuberance], inverted [the top of the teat, or even the entire teat, is inverted to form a crater], very damaged [teat injured to the point where milk ejection is not possible], or supernumerary [small teats in between 2 normal teats]. A score of 1 indicated a functional teat.).

Table 1.

Description of reproduction, production, udder morphology, and colostrum quality traits recorded

Trait	Description
Reproduction
SAF	Sow age at current farrowing (d)
GLEN	Sow gestation length (d)
NBA	Number of piglets born alive
LW0	Litter weight at birth excluding stillborns (kg)
TB	Total number of piglets born
LS10	Litter size at 10 d after farrowing
LW10	Litter weight at 10 d after farrowing (corrected for fostering weight [kg] + weight of fostered piglet out − weight of fostered piglet in)
LiveD	Total number of piglets born alive that died during the first 10 d of age
STB	Number of stillborn piglets
Udder morphology
COL	Colostrum IgG concentration (Brix percentage)
SAMER (SD)	Interteat distance within the same row (mm; and SD of this measure)
LEN (SD)	Teat length (mm; and SD of this measure)
DIA (SD)	Teat diameter (mm; and SD of this measure)
AML (SD)	Teat distance from the abdominal midline (mm; and SD of this measure)
OR	Proportion of teats oriented perpendicular to the udder
NoFun	Proportion of nonfunctional teats
TT	Total number of teats present in both rows
DEV	Udder development score (1 = not developed to 3 = fully developed)
Production
ADGgt	ADG during gilt test 30 to 100 kg (g)
BFAT	Backfat thickness at end of test (mm at P2)

Litter Traits

A descriptive list of recorded phenotypic traits is reported in Table 1. After the last piglets were born, litter weight at birth excluding stillborns was recorded. Litter size at birth was recorded for the total number of piglets born, the number of piglets born alive, and number of stillborn piglets (STB). Cross-fostering was permitted during the whole suckling period in order to ensure animal welfare. This practice was applied very often in all the units. During the first 10 d after farrowing, the number of dead piglets in each litter was recorded (total number of dead piglets per nursing sow). Litter size and weight at 10 d after farrowing were recorded, and this value was corrected for fostering (plus number/weight of fostered piglets out or minus number/weight of fostered piglets in) to give a value for total maternal investment by the sow over this period.

Sow Traits

The observation day, sow age at current farrowing, sow parity number, and sow gestation length were recorded. Records taken from gilt test data gathered during the rearing period included ADG and backfat thickness at end of test at 100 kg. A full list of trait descriptions is reported in Table 1.

Colostrum Sample Collection.

Samples of colostrum were collected when freely available immediately before or early during parturition, without the use of oxytocin. The operator quietly approached the sow and obtained a sample by using hand pressure exerted approximately in the center of the mammary gland. A 15-mL sample of colostrum was collected by sampling from all the teats located in the upper row and, when possible without disturbing the sow, also from the teats in the lower row of the udder. The sample was collected and stored in a sterile pot (30 mL Polystyrene Universal container; Starlab, Milton Keynes, UK) labeled with identity of the sow, sampling time, and date. After the samples were drawn, they were frozen and stored at −20°C until further analysis.

Colostrum Sample Analyses.

A Brix refractometer (MA871 digital; Obione, La Valette, France) was used to estimate colostrum IgG concentration as described in Balzani et al. (2016b). Briefly, at the start of each set of analyses, it was calibrated with distilled water before proceeding with the sample analysis. A drop of well-mixed whole colostrum was then placed on a refractometer prism and the Brix score (%) was recorded.

Statistical Analyses.

Phenotypic correlations between udder morphology traits were calculated using a single trait measurement per animal (calculated as the arithmetic mean of the teat-specific measurements taken across the whole udder, where relevant). Phenotypic correlations were calculated as pooled within class (parity number and the batch, year, and month effect of the day of the observation). The correlations were also estimated between udder quality traits and sow productive (ADG and backfat thickness at end of test) and reproductive traits (sow gestation length, total number of piglets born, number of piglets born alive, STB, litter weight at birth excluding stillborns, total number of piglets born alive that died during the first 10 d of age, and litter size and weight at 10 d after farrowing). Genetic analysis of heritability of udder traits and genetic correlations with productive and reproductive traits were estimated for the same traits as the phenotypic analyses.

For genetic analysis, udder morphology measurements and reproductive traits were modeled as a function of parity number and the batch, year, and month effect of the day of the observation. Productive traits were modeled as a function of the batch, year, and month effect of the day of the performance test. Sow was included as a random effect in the analyses. The genetic parameters were estimated using single-trait animal models in VCE (Variance Component Analysis) and PEST (Model- independent Parameter Estimation and Uncertainty Analysis). The mixed model equations used were as follows:

for udder morphology measurements and reproductive traits and

for productive traits, in which y is the vector of observations for sow udder morphology traits; μ is the overall mean; BSM-OD is the fixed effect of batch, year, and month of the day of the observation; BSM-TEST is the fixed effect of batch, year, and month of the day of the performance test; p is the fixed effect of sow parity number; and the vectors of random effects consisted of animal-specific effects (a; modeled in terms of additive genetic effects with appropriate assumed covariances to allow for pedigree relationships between sows) and residual (e) effects.

RESULTS

Descriptive statistics of udder morphology and reproductive and productive traits of the sows are shown in Table 2. For female reproduction traits, the heritability estimates were quite low (0.03–0.1), whereas heritability estimates for productive traits were moderate. Heritability estimates for udder morphology traits were moderate to high (0.22–0.53; Table 2). The distribution of stillborns was skewed (566 had no stillborn piglets, 198 sows had only 1 stillborn, 128 sows had 2, 56 sows had 3, 24 sows had 4, 12 sows had 5, and only 3 sows had more than 6), which could potentially violate some of the assumptions of the VCE analysis with respect to this trait. However, models that would allow for this appropriately are not available in VCE. Sixty-four percent of the sows had all teats oriented perpendicular to the mammary gland and 57% of the sows had all teats functional.

Table 2.

Descriptive statistics of sow udder morphology and reproduction and production traits together with corresponding heritability estimates

Traits1	No.2	Mean	SD	Minimum	Maximum	h2 (SE)
Reproduction
SAF	988	738.6	338.87	315.0	1,794.0	–
GLEN	988	115.5	1.55	110	125	0.01 (0.01)
NBA	988	11.7	3.11	1.0	21.0	0.06 (0.04)
TB	988	12.5	3.16	1.0	21.0	0.06 (0.04)
LW0	988	16.7	4.42	0.87	28.96	0.11 (0.03)
LS10	715	10.9	1.68	4.0	16.0	0.05 (0.03)
LW10	607	21.49	8.84	1.0	48.0	0.11 (0.04)
LiveD	691	0.88	1.54	0.0	14.0	0.04 (0.02)
STB	987	0.8	1.21	0.0	8.0	0.03 (0.04)
Udder morphology
COL	528	25.5	3.50	15.1	35.0	0.35 (0.07)
SAMER	987	104.5	14.45	62.2	154.0	0.37 (0.06)
SAMER SD	987	21.62	7.94	4.8	74.8	0.18 (0.03)
LEN	986	16.1	3.00	7.0	31.3	0.46 (0.04)
LEN SD	986	2.81	2.21	0.0	11.34	0.42 (0.1)
DIA	987	10.5	1.70	4.6	16.2	0.53 (0.02)
DIA SD	987	1.77	1.38	0.1	9.4	0.11 (0.03)
AML	987	61.2	10.88	30.0	105.0	0.22 (0.04)
AML SD	987	17	4.95	5.0	47.8	0.19 (0.03)
OR	987	0.9	0.21	0.0	1.0	0.14 (0.03)
NoFun	987	0.04	0.05	0.0	0.4	0.22 (0.05)
TT	987	15.6	1.12	12.0	19.0	0.42 (0.02)
DEV	987	2.24	0.51	1.0	3.0	0.25 (0.04)
Production
ADGgt	863	448.5	169.39	45.9	755.1	0.25 (0.04)
BFAT	732	12.1	3.79	2.3	32.7	0.30 (0.05)

¹SAF = sow age at current farrowing; GLEN = sow gestation length; NBA = number of piglets born alive; TB0 = total number of piglets born; LW0 = litter weight at birth excluding stillborns; LS10 = litter size at 10 d after farrowing; LW10 = litter weight at 10 d after farrowing; LiveD = total number of piglets born alive that died during the first 10 d of age; STB = number of stillborn piglets; COL = colostrum IgG concentration; SAMER = interteat distance within the same row; LEN = teat length; DIA = teat diameter; AML = teat distance from the abdominal midline; OR = proportion of teats oriented perpendicular to the udder; NoFun = proportion of nonfunctional teats; TT = total number of teats present in both rows; DEV = udder development score; ADG = ADG during gilt test 30 to 100 kg (g); BFAT = backfat thickness at end of test.

²Number of observations.

Estimates of Genetic and Phenotypic Correlations

The estimates of genetic and phenotypic correlations between udder morphology traits are presented in Table 3. Almost all udder morphology traits showed medium to high genetic correlations with each other. The strongest genetic correlation (r_g) and phenotypic correlation (r_p) were found between teat dimensions (LEN and DIA; r_g = 0.55, r_p = 0.53, P < 0.001); DIA was also correlated with udder size traits (SAMER, r_g= 0.52, r_p = 0.23, P < 0.001; AML, r_g = 0.25, r_p = 0.13, P < 0.001; and udder development score, r_g = 0.31, r_p = 0.25, P < 0.001). Udder dimensions AML and SAMER were also correlated (r_g = 0.54). Colostrum IgG concentration was genetically correlated with SAMER (r_g = 0.54). Negative phenotypic correlations with total number of teats (present in both rows) were found for SAMER (r_p = −0.52, P < 0.001) and, with low negative correlation, for teat dimensions (LEN, r_p = −0.07, P = 0.04, and DIA, r_p = −0.11, P < 0.001).

Table 3.

Estimated genetic (above the diagonal line) and phenotypic (below the diagonal line) correlations between udder morphology traits (SE in parentheses)

Morphology traits1	LEN	DIA	AML	SAMER	NoFun	DEV	OR	TT	COL
LEN		0.55 (0.06)	−0.06 (0.06)	0.21 (0.05)	−0.04 (0.11)	0.04 (0.09)	−0.47 (0.09)	−0.01 (0.05)	−0.16 (0.06)
DIA	0.53 (0.05)		0.25 (0.12)	0.52 (0.06)	−0.21 (0.09)	0.31 (0.07)	−0.21 (0.05)	0.13 (0.08)	0.07 (0.06)
AML	0.04 (0.03)	0.13 (0.04)		0.54 (0.11)	0.68 (0.09)	0.29 (0.16)	0.13 (0.11)	−0.08 (0.18)	0.15 (0.11)
SAMER	0.05 (0.03)	0.23 (0.07)	0.15 (0.13)		−0.04 (0.1)	0.12 (0.12)	−0.09 (0.10)	−0.74 (0.05)	0.41 (0.06)
NoFun	−0.18 (0.01)	−0.27 (0.01)	0.12 (0.01)	0.06 (0.01)		0.25 (0.1)	0.17 (0.02)	0.4 (0.06)	−0.12 (0.10)
DEV	0.15 (0.01)	0.25 (0.01)	0.15 (0.03)	0.16 (0.03)	0.03 (0.00)		0.48 (0.1)	0.07 (0.06)	0.1 (0.07)
OR	−0.15 (0.01)	−0.01 (0.00)	0.14 (0.02)	0.03 (0.01)	−0.03 (0.00)	0.05 (0.03)		−0.42 (0.14)	−0.43 (0.11)
TT	−0.07 (0.01)	−0.11 (0.01)	0.02 (0.01)	−0.52 (0.09)	0.10 (0.01)	−0.06 (0.01)	−0.03 (0.00)		−0.17 (0.12)
COL	0.03 (0.02)	0.09 (0.01)	0.04 (0.05)	0.03 (0.06)	−0.03 (0.00)	0.10 (0.02)	−0.05 (0.01)	−0.01 (0.01)

¹LEN = teat length; DIA = teat diameter; AML = teat distance from the abdominal midline; SAMER = interteat distance within the same row; NoFun = proportion of nonfunctional teats; DEV = udder development score; OR = proportion of teats oriented perpendicular to the udder; TT = total number of teats present in both rows; COL = colostrum IgG concentration. Estimates significantly different from 0 are in bold.

The estimated genetic correlations between reproduction traits and udder morphology traits are presented in Table 4. Although most of the estimates were not significantly different from 0, some significant correlation estimates were observed. The SAMER showed a genetic association with almost all the considered reproductive traits. The highest correlations estimated were positive between SAMER and litter size at birth (0.89) and negative between SAMER and STB (−0.48)

Table 4.

Estimated genetic correlations between reproduction and udder morphology traits¹ (SE in parentheses)

	Morphology traits
Reproduction traits	LEN	DIA	AML	SAMER	OR	NoFun	DEV	TT	COL
GLEN	0.08 (0.10)	0.17 (0.11)	0.36 (0.28)	−0.14 (0.23)	−0.01 (0.11)	−0.03 (0.11)	−0.36 (0.1)	0.42 (0.16)	−0.41 (0.18)
NBA	−0.1 (0.15)	0.04 (0.04)	−0.4 (0.19)	0.69 (0.08)	−0.03 (0.24)	0.17 (0.15)	0.42 (0.34)	−0.47 (0.08)	0.54 (0.13)
TB	−0.33 (0.20)	−0.24 (0.21)	−0.26 (0.23)	0.89 (0.13)	−0.37 (0.3)	−0.3 (0.22)	0.13 (0.17)	−0.46 (0.08)	0.55 (0.29)
LW0	−0.02 (0.09)	0.14 (0.16)	0.69 (0.12)	0.78 (0.15)	0.10 (0.08)	0.09 (0.15)	0.04 (0.11)	−0.7 (0.13)	0.45 (0.11)
LS10	−0.15 (0.1)	−0.5 (0.09)	0.22 (0.11)	0.38 (0.11)	−0.15 (0.15)	−0.16 (0.17)	−0.52 (0.1)	−0.03 (0.11)	−0.2 (0.11)
LW10	0.58 (0.17)	−0.13 (0.11)	−0.56 (0.15)	−0.37 (0.13)	−0.17 (0.08)	−0.51 (0.13)	0.16 (0.18)	0.5 (0.24)	−0.11 (0.11)
LiveD	−0.13 (0.07)	0.45 (0.14)	−0.21 (0.22)	0.87 (0.09)	−0.41 (0.13)	−0.07 (0.11)	0.46 (0.17)	−0.57 (0.18)	0.55 (0.17)
STB	−0.2 (0.2)	−0.27 (0.18)	0.4 (0.2)	−0.48 (0.12)	−0.01 (0.27)	−0.12 (0.22)	0.16 (0.25)	0.52 (0.25)	−0.44 (0.24)

¹LEN = teat length; DIA = teat diameter; AML = teat distance from the abdominal midline; SAMER = interteat distance within the same row; OR = proportion of teats oriented perpendicular to the udder; NoFun = proportion of nonfunctional teats; DEV = udder development score; TT = total number of teats present in both rows; COL = colostrum IgG concentration; GLEN = sow gestation length; NBA = number of piglets born alive; TB0 = total number of piglets born; LW0 = litter weight at birth excluding stillborns; LS10 = litter size at 10 d after farrowing; LW10 = litter weight at 10 d after farrowing; LiveD = total number of piglets born alive that died during the first 10 d of age; STB = number of stillborn piglets. Estimates significantly different from 0 are in bold.

Table 5 presents the estimated phenotypic correlation between udder morphology traits, colostrum quality, and reproductive traits. The phenotypic correlations between udder quality traits and sow reproductive traits followed the same pattern as the genetic correlations.

Table 5.

Estimated phenotypic correlations between reproduction traits and udder morphology traits¹ (SE in parentheses)

	Morphology traits
Reproduction traits	LEN	DIA	AML	SAMER	OR	NoFun	DEV	TT	COL
SAF	0.06 (0.2)	0.01 (0.07)	0.23 (0.7)	−0.04 (0.3)	−0.65 (0.21)	0.17 (0.05)	0.08 (0.1)	0.08 (0.15)	0.33 (0.72)
NBA	−0.01 (0.01)	0.04 (0.01)	0.06 (0.04)	0.14 (0.07)	0.04 (0.01)	−0.01 (0.00)	0.01 (0.00)	−0.02 (0.01)	0.10 (0.04)
TB	−0.02 (0.01)	0.03 (0.01)	0.05 (0.04)	0.14 (0.08)	−0.01 (0.00)	0.01 (0.00)	0.00 (0.00)	−0.01 (0.00)	0.07 (0.04)
LW0	0.03 (0.02)	0.05 (0.01)	0.08 (0.06)	0.17 (0.1)	−0.02 (0.00)	0.03 (0.00)	0.06 (0.01)	−0.03 (0.01)	0.10 (0.05)
LS10	−0.01 (0.01)	−0.08 (0.02)	−0.02 (0.02)	0.02 (0.02)	−0.04 (0.00)	−0.04 (0.00)	−0.03 (0.01)	0.03 (0.01)	0.04 (0.02)
LW10	−0.04 (0.02)	−0.14 (0.01)	0.02 (0.05)	−0.03 (0.02)	0.03 (0.00)	−0.03 (0.00)	−0.01 (0.01)	0.07 (0.01)	0.01 (0.05)
LiveD	0.04 (0.01)	0.11 (0.02)	−0.00 (0.01)	0.08 (0.04)	−0.010.00)	0.03 (0.00)	0.01 (0.00)	−0.04 (0.01)	0.07 (0.03)
STB	−0.02 (0.01)	−0.03 (0.01)	−0.03 (0.02)	−0.00 (0.01)	−0.07 (0.00)	0.00 (0.00)	−0.03 (0.00)	0.04 (0.01)	−0.10 (0.03)

¹LEN = teat length; DIA = teat diameter; AML = teat distance from the abdominal midline; SAMER = interteat distance within the same row; OR = proportion of teats oriented perpendicular to the udder; NoFun = proportion of nonfunctional teats; DEV = udder development score; TT = total number of teats present in both rows; COL = colostrum IgG concentration; SAF = sow age at current farrowing; NBA = number of piglets born alive; TB0 = total number of piglets born; LW0 = litter weight at birth excluding stillborns; LS10 = litter size at 10 d after farrowing; LW10 = litter weight at 10 d after farrowing; LiveD = total number of piglets born alive that died during the first 10 d of age; STB = number of stillborn piglets. Estimates significantly different from 0 are in bold

The estimated genetic correlations between production traits and udder morphology traits are presented in Table 6 and the phenotypic correlations are presented in Table 7. Only a few traits were significantly, but weakly, correlated.

Table 6.

Estimated genetic correlations between production traits and udder morphology traits¹ (SE in parentheses)

	Morphology traits
Production traits	LEN	DIA	AML	SAMER	OR	NoFun	DEV	TT	COL
ADG	0.27 (0.11)	0.14 (0.11)	0.42 (0.14)	−0.06 (0.18)	0.00 (0.01)	0.01 (0.02)	0.03 (0.11)	−0.03 (0.06)	0.26 (0.16)
BFAT	−0.02 (0.07)	0.28 (0.08)	0.15 (0.13)	−0.13 (0.1)	0.00 (0.02)	0.01 (0.07)	−0.22 (0.06)	0.36 (0.07)	0.01 (0.02)

¹LEN = teat length; DIA = teat diameter; AML = teat distance from the abdominal midline; SAMER = interteat distance within the same row; OR = proportion of teats oriented perpendicular to the udder; NoFun = proportion of nonfunctional teats; DEV = udder development score; TT = total number of teats present in both rows; COL = colostrum IgG concentration; ADGgt = ADG during gilt test 30 to 100 kg (g); BFAT = backfat thickness at end of test. Estimates significantly different from 0 are in bold.

Table 7.

Estimated phenotypic correlations between production traits and udder morphology traits¹ (SE in parentheses)

	Morphology traits
Production traits	LEN	DIA	AML	SAMER	OR	NoFun	DEV	TT	COL
ADG	0.02 (0.07)	0.07 (0.10)	0.11 (0.33)	0.00 (0.00)	0.01 (0.01)	−0.03 (0.01)	0.08 (0.02)	0.01 (0.03)	−0.03 (0.12)
BFAT	0.00 (0.00)	0.02 (0.01)	−0.01 (0.01)	0.00 (0.01)	−0.02 (0.00)	0.01 (0.00)	−0.06 (0.1)	0.03 (0.01)	0.04 (0.03)

¹LEN = teat length; DIA = teat diameter; AML = teat distance from the abdominal midline; SAMER = interteat distance within the same row; OR = proportion of teats oriented perpendicular to the udder; NoFun = proportion of nonfunctional teats; DEV = udder development score; TT = total number of teats present in both rows; COL = colostrum IgG concentration; ADG = ADG during gilt test 30 to 100 kg (g); BFAT = backfat thickness at end of test. Estimates significantly different from 0 are in bold

DISCUSSION

This is the first study to estimate the heritability of sow udder morphology traits and colostrum quality and their correlation with important reproductive and productive traits. A recently developed methodology for large-scale evaluation of sow udder morphology (Balzani et al., 2016a,c) and colostrum quality (Balzani et al., 2016b) was adopted in this study, using a combination of measurements and linear scores. Similar methods have been used to evaluate udder morphology in dairy sheep (Labussière et al., 1981; Fernandez et al., 1995; de la Fuente et al., 1996; Casu et al., 2006; Huntley et al., 2012), cattle (Seykora and McDaniel, 1985; Kuczaj, 2003; Zwertvaegher et al., 2011), and goats (Horak and Gerza, 1969; Wang, 1989). Heritability estimates of udder morphology traits found in this study on sows were moderate to high (0.11 to 0.53).

The heritability of total number of piglets born and litter size at 10 d (although this latter measure was greatly influenced by cross-fostering) in this study involving the Meidam breed was similar to the estimated values for Landrace (0.10 and 0.09, respectively) and Yorkshire (0.12 and 0.10, respectively) found in the study of Nielsen et al. (2013), but estimated genetic parameters for number of piglets born alive and total number of piglets born and stillborn were slightly lower compared with the values found by Canario et al. (2007). In this study, the results show both positive and negative genotypic and phenotypic correlations between udder quality traits and productive and reproductive traits.

Teat length and DIA were highly heritable. Similar results regarding the inheritance of teat size (length and diameter were considered 2 distinct traits) have been reported by Mavrogenis et al. (1988) for Chios sheep and by Seykora and McDaniel (1985) for first lactation Holstein cattle. The estimates of Gootwine et al. (1980) for the Assaf sheep and Horak and Gerza (1969) for Cigaja and Valaska sheep were much lower (0.04 to 0.21) but based on linear scores rather than precise measurement. These 2 teat dimensions were also genetically and phenotypically correlated with each other. Likewise, LEN had a positive genetic correlation with teat form, placement, and position also in cattle (r_g = 0.54 to 0.82; Vukasinovic et al., 1997), where these traits are often evaluated together as teat size, defined as a combination of LEN and teat circumference (Kirschten, 2001; Bunter and Johnston, 2014; Bradford et al., 2015).

There was a high positive genetic correlation between LEN and the sow milking ability, expressed as litter weight at 10 d, although the robustness of this variable is questionable because the high number of cross-fosterings. However, all the other traits were negatively genetically correlated with the sow milking ability, apart from total teat number. A positive genetic correlation between LEN and sow ADG during gilt testing suggests that LEN may have been selected for in conjunction with this production trait. A negative genetic and phenotypic correlation was recorded between the DIA and the number and weight of piglets alive at 10 d of age, which is associated with the positive genetic correlation with the number of piglets born alive but dead before 10 d. This result might suggest that teat size is linked with piglet mortality. Large teats may be more difficult to suckle and impair early colostrum intake. Altogether, these outcomes agree with the previous size-related finding on the association between parity and teat size (Balzani et al., 2016c) and confirm the hypothesis of Vasdal and Andersen (2012) that older sows with bigger udder size have impaired teat access, which negatively influences piglet survival.

Estimates of the genetic correlation between the AML and SAMER and with the proportion of nonfunctional teats were high. This result might suggest that these traits are all associated with larger udder size. As expected, the SAMER was negatively genetically correlated with the total number of teats, and between these 2 traits, there was also a negative phenotypic correlation. There was a positive genetic correlation between interteat distances and sow reproductive traits (total born and born alive) and an associated negative genetic correlation between those traits and the total teat number. Interteat distance within the same row was also positively genetically correlated with litter size at 10 d and number of piglets born alive but dead before 10 d and negatively correlated with the STB. In contrast, the trait total teat number was negatively genetically correlated with the number of piglets born alive but dead before 10 d and positively correlated with the number of stillborns. The heritability of total teat number was similar to the average values reported in the literature, which range from 0.10 to 0.42 (Pumfrey et al., 1980; Rydhmer, 2000; Chalkias et al., 2013). The genetic correlations between teat number and sow reproductive traits are in accordance with data obtained by Pumfrey et al. (1980), but there is inconsistency in the literature. In contrast with our results, Allen et al. (1959) found that total teat number was correlated with litter size at birth, whereas there have been reports of both positive (Korkman, 1947) and negative (Pumfrey et al., 1980) genetic correlations between teat number and litter size at weaning. One interesting finding in the current study was that the STB was negatively correlated with the SAMER and highly positively correlated with the total teat number. Canario et al. (2007) stated that STB is positively correlated with number of total piglets born (0.58). Therefore, it can be suggested that selection for an increment in SAMER, which may facilitate early suckling by newborn piglets (Balzani et al, 2016c), will increase litter size and reduce the number of stillborns but have a negative effect on total teat number. Another interesting result was the genetic correlation between SAMER and the colostrum immunoglobulin estimate. This result, accompanied by the genetic correlation between SAMER and litter size, suggests that colostrum immunoglobulin could be related to increased litter size. Colostrum immunoglobulin was genetically and phenotypically correlated with the number and weight of piglets born alive, in contrast to the study of Quesnel (2011), who found no evidence of a relationship between litter size and colostrum quality. Considering the relationship between interteat distance and teat number and the complex genetic relationships between these 2 traits and mortality and stillbirth incidence, careful decisions need to be made in order to select for number of teats to reduce piglet mortality.

The estimated heritability value for the proportion of nonfunctional teats matches that observed in earlier studies (0.32 [Long et al., 2010] and 0.29 [Chalkias et al., 2013]). The positive correlation between number of teats and nonfunctional teats suggests that single-trait selection for increased teat number could increase the number of nonfunctional teats, in line with the results of Long et al. (2010) and Chalkias et al. (2013). The proportion of nonfunctional teats was highly but negatively genetically correlated with litter weight at 10 d. This might suggest that sows that have a high proportion of nonfunctional teats cannot provide enough good quality teats to allow the offspring to thrive until 10 d of age. As suggested by Long et al. (2010) and Chalkias et al. (2013) and the results of the current study, it is possible to state that adding to the genetic selection scheme a negative weighting on the number of nonfunctional teats will reduce piglet mortality. Chalkias et al. (2013) showed that nonfunctional teat number recorded in males at 3 wk of age was negatively genetically correlated with side fat thickness recorded at the age of 100 kg live weight. This differs from the findings presented here, where backfat thickness was recorded at gilt performance testing, with a significant genetic correlation found with the total teat number but not with the proportion of nonfunctional teats.

When considering the overall impact of udder morphology on piglet survival, the number of piglets born alive but dead before 10 d of age was positively genetically correlated with DIA, SAMER, and udder development score and negatively correlated with the total number of teats. These results are in line with those of previous studies in cattle (Bunter and Johnston, 2014) and suggest that increasing udder size impairs teat access, with an impact on piglet mortality. Moreover, litter weight at 10 d was positively genetically correlated with LEN but negatively genetically correlated with all the other traits. The inconsistency of these results at 10 d may be due to the intense cross-fostering applied to the litters in the study farms, where less-robust piglets may be given to sows considered to have better quality udders, and these relationships need to be further investigated in a situation where no cross-fostering after initial litter equalization occurs.

Conclusion

These findings, as the first of their type, will doubtless be much scrutinized, but there are some dependable conclusions for the heritability of udder morphology traits and their importance for piglet survival and performance. All udder morphology traits measured in this study were moderately to highly heritable and with some important correlations with reproductive and productive traits. These traits should be included in the breeding goal and weighted appropriately with other important traits in the breeding objectives to enhance optimal genetic progress. In further research, the use of these data to create scores to evaluate udder morphology in sows by grouping together some morphology traits, as has been previously done in cattle and sheep, will facilitate simpler data collection and allow larger databases for further interpretation of results.

Footnotes

1

We are grateful to BPEX (British Pig Executive) for funding this project. Thanks at the technicians at AFRD, Newcastle University, and ACMC.

© American Society of Animal Science 2016