Colostrum immune quality of local sow breed in Benin : Growth , survival and acquisition of passive immunity in new-born piglet

1 Beninese Centre for Scientific Research and Innovation (CBRSI), 03 BP 1665, Republic of Benin. 2 Department of Animal Production and Health, Polytechnic School of Abomey-Calavi, 01 BP 2009, Cotonou, University of Abomey-Calavi, Republic of Benin. 3 International Centre for Research and Development of the Livestock in Subhumid zones (CIRDES), 01 BP 454, BoboDioulasso, Burkina Faso. 4 National Institute for Scientific Research, Research Center of Agonkanmey (CRA / INRAB), Abomey-Calavi, Republic of Benin. 5 Tropical Veterinary Institute, Department of Animal Production, Faculty of Medicine Veterinary, University of Liège 20 Bd de Colonster, B43, 4000 Liège, Belgique.


INTRODUCTION
The local pig breed of Benin is the one found in several West African countries.It is raised mainly in traditional farms and could ensure sustainable incomes for small farmers.Both mortality and growth of suckling piglets are important for profitability and survival of traditional farms.The traditional husbandry practices of these animals often lead to low survival and growth performance of piglets under sow.
Ingestion of adequate amounts of colostrum components (nutrients, immune factors and growth promoters) soon after birth is crucial for survival and growth performances of new-born piglets (Quesnel et al., 2012;Ferrari et al., 2014).Thus, colostrum production may be one of the causes of low reproduction performances.A priori in African indigenous pig, this hypothesis is not based on literature data, because to our knowledge, there is a lack of information concerning colostrum production and acquisition of passive immunity in this breed.By contrast, several studies on colostrum are available on improved pig breeds.Furthermore, little data exist on growth performance of suckling piglets from local breed of Benin in traditional breeding.It would be worth obtaining data on colostrum from local sow breed and its impact on survival and growth performance of piglet in traditional farms in Benin and consequently, to explore opportunities to improve viability of these farms.This study aims to evaluate traditional husbandry practices of pregnant sows, survival and growth of suckling pig for 42 days of age, immunoglobulins G (IgG) concentration in colostrum and their transfer to the piglets from local breed.

MATERIALS AND METHODS
This study has been approved by the Committee for Animal Experiments of the Department of Animal Production and Health of Polytechnic School of University of Abomey-Calavi of Benin.The methods used in this protocol involving animals were in accordance with the ethical standards of this institution.So, all applicable guidelines for the care and use of animals of this institution were followed during this study.

Animals
The study was performed from September 2014 to March 2015 in nine (9) traditional pig farms at south of Benin with twelve (12) local sows breed (parity1) followed from pregnancy to 42 days of lactation.During this follow up, the mode of animal management (feeding, housing) was inventoried.Boar used for servicing was generally bought or borrowed.At birth, piglets were identified, weighed and returned to the sow.They were weighed thereafter at 24 h and then weekly until 42 days of age.Their survival was also followed under the sow from birth to 42 days of life.In total, fifty-five (55) piglets born alive from the remaining 10 sows (2 sows died with loss of their piglets) were observed from birth to 42 days of age.

Sample collection and analyses
On 10 of the 12 sows, a sample of colostrum obtained from all teats was collected manually 24 h after the onset of farrowing.Colostrum ejection was induced with intravenous injection of 1 ml of oxytocin per sow.Colostrum was filtered through gauze tissue and stored in a continuous chain of cold.Three (03) piglets per litter were randomly selected for blood collection 7 days after birth.In total, on 30 of the 55 piglets, the blood samples were collected by anterior vena cava puncture using a 21G needle and vacuum tube (vacutainer).Blood were centrifuged and serum stored in a continuous chain of cold until analysis.
The amount of IgG in samples of sows colostrum and piglets serum was assessed by a direct enzyme-linked immunosorbent assay (ELISA) using porcine IgG Kit N ° CEA544PO from EUROMEDEX ® (France).The analyses were conducted at the International Centre for Research and Development of the Livestock in Subhumid zones (CIRDES) in Burkina Faso.

Statistical analysis
Two derivative data were calculated from weight: Average Daily Gain (ADG) and Relative ADG (RADG, that is, ratio of ADG to mean weight of the considered interval).Means and standard deviation for mortality rate, weight, ADG and RADG at different ages, concentrations of piglets serum IgG and sows colostrum IgG were calculated.Correlations between mortality rates, growth and immunity parameters were evaluated by using Pearson's coefficient.Binary logistic regression model was used to determine the relationships between piglets mortality from birth to 42 days as dependent variable, and both birth weight (BW) and weight gain (WG) at 24 h of life.The inclusion of sow effect in the model was not significant and thus it was not retained.Odds ratio for mortality from birth to 42 days are presented jointly with 95% confidence intervals.BW and WG at 24 h were compared between surviving and dead piglets by the Student's t-test.Associations between WG at different ages, BW and litter size were investigated using multiple regression analysis.In this case and in order to meet the conditions of the model validity (normality), the variables were transformed by the logarithm to base 10.The Effects of BW categories (BW < 500 g and BW ≥ 500 g) on the ADG, RADG at 42 days and on the concentrations of serum IgG were evaluated by analyses of variance (ANOVA) including the sow factor.The setting up of these two weight categories was based on the fact that the average BW of small West African piglets is 500 g in traditional breeding.

Pig farms management
The mode of animal management in the study farms was traditional.Pigs were confined in pens built of cement   bricks walls for most (6/9), wood and bamboo for some (3/6), with individual box for farrowing.The roofs were covered with salvage sheets metals or straw and the floor was bare, cemented or made of concrete.The byproducts used predominantly in feeding of sows were maize bran, palm kernel cake, kitchen wastes and wheat bran, sometimes malt brewery and mill waste.Sows were fed with a simple diet consisting of one of these byproducts, according to their availability.Moreover, they were daily fed with forages in all of farms and received no veterinary care.Forages and maize bran were used by all breeders (Figure 1).

Animal survival and numeric productivity of sows
During animal follow-up, two sows died at the second week after farrowing with loss of most of their piglets the first week of life and the remainder at the onset of the second week.Lack of appetite followed by gradual refusal to eat was the main symptom observed, starting from 3 rd and 5 th days after farrowing in the first and the second animals respectively.No neo or perinatal mortality was observed during farrowing of the 10 sows and the mean number of piglets born alive was 5.5 ± 1.96.The average pre-weaning (0-42 days) mortality rate of piglets was 14.8 ± 21.5% and the mean number weaned at 42 days was 4.9 ± 2.28.Half of the piglet deaths occurred in the first week of life.Table 1 shows the overall reproductive performance of sows.The model of binary logistic regression (Table 2) showed that the risk of mortality was affected by the piglet BW (P < 0.05) and WG at 24 h (P < 0.05).Piglets that died before 42 days of age had lower (P ˂ 0.05) average BW (338 ± 45 g versus 451.5 ± 14 g) and average WG 24 h after birth (-10 ± 22 g/day versus 44.9 ± 7.2 g/day) than surviving piglets (Table 3).

Growth of suckling piglets
Piglets average BW was 439 ± 106 g.The weight of piglets at 42 days of age was 2119 ± 720 g.Live weight increased steadily during the first two weeks of age, curbed the third weeks, to increase to a slower rate thereafter (Figure 2).This trend was the same as that observed for weekly ADG, with a higher gain obtained between 7 and 14 days (around 50 g/day), a fall between 14 and 28 days and a celling from 28 to 42 days of age reaching values just below 30 g/day the last week of measures.By contrast, RADG showed a continuous drop from 7 to 28 days and a celling from 28 to 35 days with a higher value obtained the first week (around 80 g/day.kg)and a lower between 21 and 42 days (below 20 g/day.kg).Piglets ADG until 7 days-old and until weaning at day 42 were respectively 49.7 ± 19.8 and 39.7 ± 15.7 g/day.The RADG 0-7 day and 0-42 day were respectively 78.4 ± 23 and 30 ± 4.3 g/day.kg.Intra-litter piglets average WG at 24 h of age was 37.8 ± 30.7 g with large inter-sows variability, ranging from a minimum of -20 g and a maximum of 88 g.The average litter WG at 24 h was 215 ± 218 g.Moreover, intra-litter coefficient of variation of BW was 18.54 ± 7.55%.Likewise, growth performances are also shown in Table 4.
Table 5 shows some regression models highlighting the  relationships between animal performance parameters.
Positive relationships between BW and WG 0-7 day (P <0.05), WG 0-28 day (P < 0.05) and WG 0-42 day (P < 0.001) were observed.Litter size was positively associated with WG 0-7 day (P < 0.05), but no association was observed between litter size and WG 0-28 day.The association between litter size and WG 0-42 day tended to be negative (P = 0,083).About one third of WG variation at 42 days could be explained by BW variation and litter size (R² = 35%, P < 0.001).The ADG 0-42 day was affected by the BW class (BW < 500 g and BW ≥ 500 g) (P < 0.005) and by the sow (P < 0.001) (Table 6).72% of the variation of WG 0-42 day was explained by the sow.Piglets whose average BW was less than 500 g had lower ADG 0-42 day than piglets whose average BW was higher or equal to 500 g (33.8 ± 2.1 g/day versus 48.3 ± 2.1 g/day, P <0.005).The RADG 0-42 day was not affected by the BW class (P = 0.932), but by sow (p < 0.001).Piglets of these two BW categories had, at 42 days of age, similar average RADG (30.4 ± 0.82 g/day.kgversus 30.5 ± 0.82 g/day.kg).Table 7 shows the significant correlations between animal performances.Both WG at 24 h and ADG 0-42 day were positively correlated with piglets BW, while ADG 1-42 day was correlated with WG 24 h.

Immune colostrum quality, IgG transfer and relationships with piglets reproductive performance
The average concentration of IgG in colostrum taken 24 h after the onset of farrowing was 21.8 ± 8.83 mg/ml, with a

Parameters and correlations r value P value
Piglets WG 24 h (g) and BW (g) 0.40 0.002 ADG 0-42 day (g/day) and BW (g) 0.55 0.000 ADG 1-42 day (g/day) and WG 24 h (g) 0.41 0.004 Colostrum IgG concentration (mg/ml) and intra-litter average WG of piglets at 24 h (g) 0.71 0.049 Piglets serum IgG concentration (mg/ml) and piglets weight at 7 days (g) 0.37 0.049 Piglets serum IgG concentration at 7 days (mg/ml) and ADG 0-7 day (g/day) 0.50 0.008 Piglets serum IgG concentration at 7 days (mg/ml) and ADG 0-28 days (g/day) 0.37 0.059 large variability between sows (Coefficient of Variation (CV) = 40.5%).A negative correlation was observed between intra-litter mortality rates and IgG concentration of colostrum, but it was not significant (P = 0.595).No relationship was observed between the IgG concentration in colostrum and intra-litter average ADG 0-7 day, 0-28 day and 0-42 day (with P value 0.433; 0.362 and 0.565 respectively).The correlation between IgG colostrum concentration and intra-litter average WG at 24 h of age was strong and significant (Table 7).By contrast, piglets serum IgG concentration was not correlated (P = 0.164) with average IgG colostrum concentration.The average concentration of serum IgG at 7 days of age was 2 ± 1.6 mg/ml with a very high variability between piglets of the same litter (average CV = 62.8%).
Figure 3 shows the distribution of piglets serum IgG concentration at 7 days of age.This was not affected by BW categories (P = 0.109) with however a trend for sow (P = 0.076).Moreover, it was not correlated with BW, but was positively correlated with ADG 0-7 day, as well as with weight at 7 days of age.No significant correlation was observed between this parameter and ADG 0-28 day and 0-42 day with however a trend for ADG 0-28 day (P = 0.059).

DISCUSSION
The observed husbandry practices in this study are similar to those reported within the traditional farming system of pig in Africa by Agbokounou et al. (2016a) through driving mode (straying, confinement), housing, health, reproduction and feeding of animals.From pregnancy to lactation, sows were kept in total confinement with a sustainable or precarious housing without any hygiene measures.They were fed with one by-product or scarcely more stemming from traditional or industrial food-processing, according to their availability.They received forages daily.The feeds used by the farmers of this study are similar to those observed within the traditional farming of local pig breed in Benin and in Africa in general (Agbokounou et al., 2016a).Forages, maize bran, palm kernel cake, wheat bran and kitchen wastes were the most ingredients used by these farmers.The values of digestible energy, crude protein (CP), Neutral Detergent Fiber (NDF) and crude cellulose reported for such by-products used in Benin and in Africa in general are respectively 3612 kcal/kg Dry Matter (DM), 197, 318 and 71 g/kg DM for maize bran, 3066 Kcal/kg DM, 146, 520 and 213 g/kg DM for palm kernel cake and 2510 Kcal/kg DM, 182, 436 and 100 g/kg DM for wheat bran (Agbokounou et al., 2016a).As for forages, CP and NDF concentrations reported are respectively between 88 and 324 g/kg DM and 279 and 688 g/kg DM (Kambashi et al., 2014).The unawareness of amounts of feed ingested by animals and use of by-products according to their availability did not allow evaluation of the satisfaction of nutritional needs of sows during their phases of pregnancy and lactation.In general, such sows receive enough fiber.The digestibility of energy should be thus limited by the high content in fiber, which could however promote intestinal health (Kambashi et al., 2014).In addition to these macronutrients, sows would have received enough minerals and vitamins because forages are characterized by high concentrations in some minerals (Ca, P, Na, Mg and Fe) and vitamins (A, D and E) (Lucas et al., 2006;Kambashi et al., 2014).
Regarding litter size, our observations are in agreement with findings of Youssao et al. (2008a) and Kiendrebeogo et al. (2012) which showed that small-sized West African local pig breed is less prolific in traditional conditions, with low litter size at birth ranging between 5 and 6 piglets.The litter size obtained in this study is lower than those reported by Agbokounou et al. (2016b) on Mukota breed of Zimbabwe (7 piglets), local pigs of Kenya (8 piglets), and of Tchad and Cameroon (7 piglets).Improving livestock farming conditions could increase this parameter.However, our values are similar to that observed on Nigerian local pigs in experimental station (between 5 and 6) (Ajayi and Akinokun, 2013).
The observed number of piglets weaned at 42 days was 5 and the average mortality rate from birth to 42 days of life was 15%.This value seems low compared to the 22% observed in traditional farms of Benin by Youssao et al. (2008a).However, the observed rate in the present experiment was obtained between birth and 42 days while the results of those authors give a mortality rate from birth to weaning.Weaning is performed at more than 8 weeks in traditional farms in Benin.At the experimental farm of the University of Abomey-Calavi in Benin, a lower rate at 12% has been obtained from birth to 42 days (Nonfon, 2005).The risk of mortality was affected by the piglets BW and WG at 24 h.For piglets with high BW and high WG 24 h after birth, the risk of mortality before 42 days was lower.Consequently, one randomly chosen piglet was around 0.99 times more likely to die than another whose BW or WG at 24 h was 1 g lower.Moreover, piglets that died between birth and 42 days of age had lower average BW and WG 24 h after birth than surviving piglets.These results confirm the results reported by Quiniou et al. (2012), which indicated that risk of piglets mortality increases when BW decreases.They are in agreement with findings of Le Dividich et al. (2004) and Devillers et al. (2011) who showed that piglets that died before weaning had lower BW, colostrum intake and WG at 24 h than piglets that survived until weaning.They also are in agreement with results obtained later by Ferrari et al. (2014) who observed that piglets from Large White × Duroc x Landrace that died before 42 days of age had lower BW, colostrum intake and serum IgG concentration 24 h after birth than surviving ones.According to Le Dividich et al. (2004), piglet WG from birth to 24 h of age is a marker of individual colostrum intake.According to this observation and in agreement with Ferrari et al. (2014) and Charneca et al. (2015), BW and colostrum intake are the major determinants of survival of local breed piglets under the sow.
The average BW obtained in the present study (439 g) is similar to those observed by Anugwa and Okwori (2008) and Youssao et al. (2008b), respectively on the small Nigerian pig in traditional breeding (500 g) and on Benin local pig in pilot farm of the Polytechnic School of Abomey-Calavi (540 g), but higher than that reported by Do Duc (2013) with Xao Va pig breed of Vietnam (200 to 250 g) characterized by small litter size (4-5 piglets).Considering this BW, the observed piglets weight and ADG at 42 days of age are slightly lower than those observed by Nonfon (2005) at the experimental farm (2660 g and 48 g/day at 42 days respectively).According to Étienne et al. (2000) and Nonfon (2005), ADG evolution would be an interesting indicator for milk production of local pig breed, and amount of milk produced by a sow is closely related to number of piglets and their average gain.Based on these results, the evolution of body weight and ADG observed in this study indicates that milk production of these sows could be more important the first two weeks with a peak between the 7 th and 14 th day.After a drop from the 14 th day, milk production would be stable between 28 and 42 days.Piglet milk intake would attain a peak between 7 and 14 days followed by a fall from the 14 th day and stability thereafter.These evolutions tend to be in agreement with that reported by Lachance (2010) on improved breeds with a peak at day 14 followed by a stability from the 22 nd to the 23 rd day.It is consistent with findings of Aguinaga et al. (2011), who showed that milk production by the Iberian breed (and consequently piglet milk intake) attained a peak at day 12 and leveled off thereafter.Additionally, the maximum of milk intake by improved breed piglets is obtained between the 2 nd and 4 th weeks of lactation (Nonfon, 2005).However, it must be highlighted that the steady decrease in RADG from the beginning of our experiment suggests that milk production is not really sustained all over the 42 days of study.
According to Ferrari et al. (2014), BW and colostrum intake were crucial to the development of improved piglets breed from birth to day 42.This is in agreement with results obtained in this study with local breed piglets of Benin.Indeed, a positive relationship was observed between BW and WG 0-42 day or ADG 0-42 day (P < 0.001).Moreover, the ADG 0-42 day was to some extent affected by BW categories (P < 0.005) and especially by sow (P < 0.001).According to these observations, growth of local breed piglet from birth to weaning is dependent on environmental factors (farming practices) and factors related to the sow, such as colostrum production and genetic potential.Similarly to what was observed for colostrum production of improved breed (Le Dividich, 2006), this study showed a large variability of colostrum production between sows.This is in agreement with the observed large variability between sows of the intra-litter average WG at 24 h (-20 to 87.5 g) which would be an indicator of the colostrum yield.Regarding influence of genetic traits on growth, local piglet breed of Benin would be genetically similar to Ashanti Black pig.Indeed Darfur Oduro et al. ( 2009) had shown that maternal heritability of this breed was higher than direct heritability for BW and pre-weaning ADG.They showed that these parameters were influenced by the same genes and maternal additive genetic effect is an important component of variation in growth traits.Piglets of this breed with genetic capacity for high BW have better genetic capacity for growth during lactation and sows with genetic capacity to give birth to heavier piglets have a high pre-weaning growth rate in their litters (Darfour-Oduro et al., 2009).Thus it appears that BW is the most important determinant of the pre-weaning growth of local piglets breed.
This study showed the same average RADGs between piglets with BW < 500 g and BW ≥ 500 g.This finding indicates that average weight from birth to day 42 of piglets with BW < 500 g could be much lower than the one of piglets with BW ≥ 500 g because the ADG of piglets with low BW was lower than the one of animals with high BW.Moreover, the effect of the colostrum intake on growth was deduced from positive correlation between WG at 24 h and ADG obtained between birth and day 42 (P <0.005).In addition to these results, the observed positive correlation between BW and WG at 24 h which is a marker of colostrum intake indicates that local breed piglets with better BW has a better WG at 24 h and therefore better colostrum intake.This result is consistent with that of Ferrari et al. (2014).Furthermore, only the variation of the WG from 0 to 7 days was positively explained both by litter size and birth weight (R 2 = 18%, P <0.005).By contrast, the relationship between the WG 0-42 days and litter size was not significant.Similar to what was observed for milk production of improved breed (Étienne et al., 2000;Ngo et al., 2012), it seems that the colostrum and milk production the first week by local sow breed increased with the litter size.This result is in contrast with findings of Foisnet et al. (2010) and Declerck et al. (2015), an indication that colostrum yield is independent on litter size.The result of the present study may be due to the fact that a high number of piglets stimulate all the teats, which would increase the number of functional ones.This higher number of functional teats coupled with the increased production of colostrum and milk from these sows the first two weeks, as indicated above, could explain this positive relationship between WG 0-7 day and litter size.The hypothesized decreased of milk production observed from the 14th day could explain the no significant relationship observed between WG 0-42 day and litter size.Finally, WGs from 0 to 42 days of suckling local breed piglets can be predicted by the equation: log 10 (WG 0-42d) = 0.079 -0.33log 10 (litter size) + 1.27 log 10 (BW), but with low reliability because the predictive value was 29%.This relationship must be studied with higher sample sizes to be confirmed.
Despite weak growth performance of its piglets before weaning, the local sow breed is characterized by adequate IgG concentration of its colostrum.The observed average IgG concentration of colostrum collected at 24 h after the onset of parturition (22 mg/ml) is numerically greater than those observed with Large White x Landrace by Charneca et al. (2015) (12 mg/ml), Devillers et al. (2011) (17 mg/ml) or Decaluwé et al. (2014) (18 mg/ml).It is slightly higher than the 20 mg/ml observed by Charneca et al. (2015) on the Alentejano sow, an Iberian breed from region of Alentejo in Portugal.The combined effects of genetic and environmental (feeding practices, hygiene) factors expected in this study could explain higher colostrum IgG level in these sows.Contrary to feeding practices observed with improved breed sows, animals of this study received daily a lot of fiber through forages and by-products from traditional and industrial food-processing.According to Lucas et al. (2006) green forages are rich in β-carotene and vitamin E. Furthermore, the structural polysaccharides, including β-glucans, are the main fibers of lignocellulosic biomass (Godin et al., 2011).According to Laws et al. (2009), increased maternal intakes of β-carotene and vitamin E could increase concentration of colostrum IgG.Moreover, Leonard et al. (2010) have reported that the immune modulating properties of β-glucans could increase concentration of colostrum IgG of the sow.According to these observations, daily forages intake by local sows breed could explain their higher colostrum IgG levels.In contrast, high dietary fiber could decrease by half IgA in colostrum sampled at 24 h (Loisel et al., 2013).Feed studies taking into account sow metabolic status are needed to confirm these results.Furthermore and similarly to what is observed for colostral IgG of improved sows breed (Le Dividich, 2006;Farmer and Quesnel, 2009), colostrum IgG concentration would be characterized by high variability between sows in local pig breed.The observed inter-sow coefficient of variation for colostrum IgG concentration was 41%.This value is close to that obtained with the Large White x Landrace sows (45%) and lower than that of Alentejano pigs (52%) for colostrum taken at 12 h after the onset of parturition (Charneca et al., 2015).
Although the high level of IgG in the colostrum from the sows, the transfer of these antibodies to piglets seems weak when compared to that observed in improved European piglets breed.The observed serum IgG level at 7 days of age (2 mg/ml) is very lower than those obtained at 7 days by Rooke et al. (2003) with Large -White x Landrace (8 mg/ml) and Yun et al. (2014) with Yorkshire x Landrace (7 mg/ml).According to Devillers et al. (2011), plasma IgG concentration of piglet is positively related to both colostrum intake and colostrum IgG concentration.Colostrum intake depends on production by sow and access to teats by newborns.Adequate transfer of IgG to piglet is the result of intake soon after birth of a sufficient amount of colostrum rich in this antibody.Furthermore, according to Charneca et al. (2015), decrease of serum IgG concentration observed from 48 h to 28 days with improved piglets breed is related to both the clearance of IgG, with a half-life of about 10 days, and the dilution effect associated with the increase of blood volume in that period.According to these considerations, when compared to improved breeds the low serum IgG level observed in this study may be related to lower colostrum production by local sows breed, lower colostrum intake, longer interval between birth and first sucking, or a more rapid clearance of IgG between 48 h and 7 days, for example, consecutively to infectious pressure or to high ability of piglets to clear IgG from the sow.According to Le Dividich et al. (2004), the litter WG at 24 h is a marker of colostrum production.The very low piglet WG at 24 h and the very low average litter WG at 24 h (215 g) observed in this study, as well as the decrease of RADG from the first week, could reinforce the thesis of low colostrum production by local sows breed.Studies need to be conducted to confirm these theories.The serum IgG concentration varied largely between piglets of the same litter and, in agreement with Cariolet et al. (2007), was not correlated with the colostrum IgG concentration.This variability could be related to the intra-litter variability of BW (CV = 19%) which is higher than the value of 15% observed with improved breeds by Quesnel et al. (2008) for litters size less or equal to 9 piglets.It also could be related to the interval between birth and first sucking and access to the teats competition.The hypothesis of intralitter variability of BW might not be evident, because piglets' serum IgG levels at 7 days of age were not affected by BW class.Moreover they were not correlated with BW.This observation is in agreement with the results of Ferrari et al. (2014), who observed no correlation between BW and serum IgG concentrations of piglets at 24 h, 10 and 20 days.By contrast, piglet weight at 7 days of age was positively correlated with serum IgG concentration at 7 days (P ˂ 0.05).Similarly, the correlation between ADG 0-7 day and serum IgG concentration at 7 days was positive (P ˂ 0.05), while this relationship did not exist between this serum IgG concentration and ADGs 0-28 day and 0-42 day.The same trend was observed with the colostrum IgG concentration at 24 h and ADGs.Indeed, a strong positive correlation was observed between colostrum IgG concentration at 24 h and piglets WG at 24 h (P ˂ 0.05).By contrast, no relation was observed between this colostrum IgG level and ADG (0-7 day, 0-28 day and 0-42 day).Serum IgG level appears thus important during the first stages of life in piglets from local breed in Benin.

Conclusion
In Benin, local sow breed in traditional breeding are subject to farming practices characterized by sustainable or precarious housing without any hygiene measures.Regarding feeding, they receive a significant part of nutrients by forages.Daily feeding with forages is a particularity of local pig breeding in Benin.Reproductive performance permitted by these farming practices remains weak and prolificacy, survival and growth performances of piglets under sow remains low.Birth weight and weight gain at 24 h are important indicators for survival and growth of local suckling piglets breed.Moreover, this study provides the first estimate of colostral IgG concentration and piglet serum IgG level of the local pig breed.Sows are characterized by adequate colostrum IgG concentration.However, the transfer or the retention of these antibodies at piglet level is low.Finally, feeding practices could be associated to high colostrum quality in local breed sow in Benin but the low immune transfer to piglets should be further investigated.

Figure 1 .
Figure 1.Feed used in local sow breed feeding in Benin.

Figure 3 .
Figure 3. Distribution of serum IgG concentrations of piglets.

Table 1 .
Reproductive performance of local sow breed in Benin.

Table 2 .
Mortality of piglets from local breed in Benin until 42 days of age according to BW and WG at 24 h.

Table 3 .
Average BW and WG 24 h after birth of dead and surviving piglets from local breed in Benin.
Figure 2. Evolution of live weight, ADG and relative ADG in local breed piglets in Benin.

Table 4 .
Growth performance of suckling newborn piglets from local breed in Benin.

Table 5 .
Linear models showing the relationship between weight gains at different ages, litter size and birth weight in the local piglets breed in Benin.

Table 6 .
Effect of sow and BW class on the ADG 0-42 day, RADG 0-42 day and serum IgG at 7 days in the local piglets breed in Benin.

Table 7 .
Correlations between different parameters of weight and serum IgG concentration in local breed piglets in Benin.