Commercial pig diets are based on cereals and protein feed which hardly contain long-chain n-3 fatty acids(Reference Rooke, Sinclair and Edwards1). Actual requirements of the n-6 and n-3 families of fatty acids in pig diets and the ideal ratio of n-6:n-3(Reference Wiseman, Kyriazakis and Whittemore2) are still a matter of research. The n-6 and n-3 PUFA have distinct properties. It has been suggested that a high intake of n-6 PUFA, especially arachidonic acid, can contribute to inflammatory processes and predispose to or exacerbate inflammatory diseases(Reference Calder and Grimble3, Reference Calder4). Moreover, n-6 and n-3 PUFA are involved in glucose and lipid metabolism. A high n-6:n-3 ratio is considered to be a critical factor in both insulin resistance and atherosclerosis in studies in man and animal models(Reference Storlien, Kriketos, Calvert, Baur and Jenkins5).
A moderate replacement of safflower-seed oil with tuna oil prevented the development of insulin resistance at the whole-body level in Wistar rats(Reference Storlien, Kraegen, Chisholm, Ford, Bruce and Pascoe6). Moreover, in rats the addition of long-chain n-3 PUFA to the diet resulted in the amelioration of insulin resistance(Reference Storlien, Jenkins, Chisholm, Pascoe, Khouri and Kraegen7). Postpartum hypophagia in sows was shown to be related to decreased glucose tolerance and increased insulin resistance(Reference Weldon, Lewis, Louis, Kovar and Miller8). Therefore, it could be hypothesised that in the case of providing a diet for sows with a low n-6:n-3 ratio, insulin action and consequently postpartum feed intake would be improved, whereas diets with a high n-6:n-3 ratio would not exert these benefits.
During the past years, the effects of inclusion of n-3 and n-6 PUFA in sow diets on performance parameters and on accretion by the piglets have been studied. Piglet growth was improved when supplementing tuna oil at a level of 1·75 % at different periods during gestation(Reference Rooke, Sinclair and Edwards1). Supplementation of salmon oil to the sow diets during gestation and lactation at a level of 1·65 % reduced pre-weaning mortality(Reference Rooke, Sinclair, Edwards, Cordoba, Pkiyach, Penny, Finch and Horgan9). It was shown that the proportions of n-3 and n-6 PUFA in piglet brain and liver at birth depend on the fatty acid composition of the maternal diet, when oil sources were supplemented at a level of 1·75 % during pregnancy(Reference Rooke, Shanks and Edwards10). The low n-6:n-3 ratio used in these and in other studies(Reference Rooke, Bland and Edwards11) ranged between 0·9 and 4·3, while the high ratio ranged from 6·6 to 15·3. For the present study an inclusion level of 2 % was decided, thus obtaining a low (2·1) and a high (10·1) n-6:n-3 ratio.
The periparturient period in sows ranges from 4 d before to 3 d after parturition(Reference Nachreiner and Ginther12). Feeding sows ad libitum a lactation diet either from 4 d before the expected day of farrowing or from the farrowing day onwards, instead of the introduction of ad libitum feeding from 3 d after farrowing, resulted in higher daily feed intake and less mobilisation of body reserves by the sows(Reference Neil13). Furthermore, elevated rectal temperatures and increased incidence of agalactia (now described as postpartum dysgalactia syndrome(Reference Klopfenstein, Farmer, Martineau, Straw, Zimmerman, D'Allaire and Taylor14)) were reported in sows given feed ad libitum from farrowing compared with sows given the gestation feeding level until 3 d after farrowing(Reference Neil, Ogle and Anner15). Moreover, when sows were fed restrictively during the last 2 weeks of gestation, a higher feeding level was associated with elevated rectal temperatures and an increased incidence of agalactia(Reference Göransson16). The transition from a gestation to a lactation diet in sows commonly takes place when sows are transferred from gestation rooms to the farrowing crates. Depending on the study, the transition may occur at day 107(Reference Bontempo, Sciannimanico, Pastorelli, Rossi, Rosi and Corino17, Reference Fritsche, Huang and Cassity18), 108(Reference Lauridsen and Danielsen19), 109(Reference Govoni, Parmeggiani, Galeati, Penazzi, De Iasio, Pagotto, Pasquali, Tamanini and Seren20), 110(Reference Eder, Ramanau, Kluge and Spilke21) or 111(Reference Eder, Ramanau, Kluge and Spilke22) of gestation.
It has not yet been investigated in sows, fed restrictively, whether the time point of the transition from a gestation to a lactation diet during late gestation (earlier: 8 d or closer: 3 d to parturition), in combination with lactation diets differing in the n-6:n-3 ratio, would have an effect on parameters in sows related to performance, metabolism and inflammatory processes. Based on the properties of n-6 and n-3 PUFA, it could be hypothesised that a lactation diet with a high n-6:n-3 ratio would increase inflammatory parameters, compared with a lactation diet with a low n-6:n-3 ratio. Leptin would be an interesting parameter for investigation, since it could be indicative of energy balance in pigs(Reference Barb, Hausmana and Houseknecht23). It has been shown in gilts that acute changes in feed intake can affect leptin secretion(Reference Whisnant and Harrell24). In addition, it is interesting to investigate the association between maternal levels of leptin and insulin during the periparturient period with progeny performance, as poor glucose tolerance of sows during late gestation was reported to be a risk factor for the survival of pigs after birth(Reference Kemp, Soede, Vesseur, Helmond, Spoorenberg and Frankena25).
The present trial was set up to elucidate the consequences of the n-6:n-3 ratio of a lactation diet, a low v. a high ratio, in combination with the time point of transition from a gestation to a lactation diet in sows, either at 8 or 3 d before farrowing. Piglet viability and growth during lactation, sows' body condition and feed intake during lactation, parameters related to metabolism during the periparturient period and inflammatory parameters postpartum were monitored. Finally, associations between levels of leptin and insulin in sows during the periparturient period and progeny performance were also investigated.
Materials and methods
Study population and experimental design
The study was conducted in the Provimi Research Farm, Velddriel, The Netherlands at the end of 2006 and the beginning of 2007. In total, seventy-two pregnant hybrid sows (Topigs 20 breed; Dutch Landrace × Great York) were included. The experimental period started at day 107 of gestation and ended when the piglets were weaned at day 21 of lactation. The experimental protocol of the study was approved by the ethical committee of the Animal Sciences Group in Wageningen, The Netherlands.
Until day 107 of gestation, the sows were group-housed in a gestation unit and they were fed twice per d with a conventional gestation diet (4 kg/d). The composition (g/kg) of this feed was: DM, 892; ash, 67; crude protein, 126; crude fat, 34; neutral-detergent fibre, 197; metabolisable energy, 10 400 kJ/kg. Linoleic content was 11·53 g/kg, total n-3 PUFA was 1·66 g/kg, and DHA and EPA content was fairly low, i.e. 4·09 × 10− 5 g/kg and 3·47 × 10− 5 g/kg, respectively.
At day 107 of gestation, the pregnant sows were moved to the farrowing rooms, and they were randomly allocated using a 2 × 2 experimental design to four different treatment groups, namely f3, f8, s3, s8. Sows from the f and the s groups received a lactation diet that was supplemented with a fish oil product, or a sunflower-seed oil product, respectively. The final level of inclusion of fish oil was 2 %, while in the other group the level of inclusion of sunflower-seed oil was 1·5 % and soyabean oil was included at 0·5 % in order to achieve the final n-6:n-3 ratio. The n-6:n-3 ratios in these feeds were 2·09 and 10·13 for the f and s groups, respectively (Table 1). Half of the sows belonging either to the f group or the s group received the experimental lactation diets from day 107 of gestation (f8 and s8 groups) onwards (8 d before expected farrowing); the other half of the sows received the experimental lactation feed only from day 111 of gestation (f3 and s3 group) onwards, being 3 d before the expected day of farrowing (onset day groups). The latter sows continued to receive the conventional gestation feed between day 107 and day 111 of gestation. The allowance of feed before and after farrowing was according to the common practice in the herd. From day 107 of gestation onwards, the sows from all four groups received the same amount of feed, namely 3·3 kg/sow per d. An overview of energy and major nutrient supply in the four experimental groups before farrowing is given in Table 2. The herd practised a 3-week batch production system, and the seventy-two sows farrowed in three consecutive batches of twenty-four sows each. Within each batch, the sows were randomly assigned to the four experimental groups. The mean parity of sows per experimental group (f3 = 4·2; f8 = 4·4; s3 = 4·6; s8 = 4·3) was not significantly different between the groups (P = 0·570).
Table 1 Chemical composition of the two different feeds for the sows*
ND, not detected.
* The feeds were supplemented either with fish oil or with sunflower-seed oil and differed in the n-6:n-3 ratio.
Table 2 Overview of the energy and major nutrient supply (per d) in sows in every experimental group before farrowing*
f, Low n-3:n-6 ratio feed supplemented with fish oil; f3, group with feed f administered 3 d before parturition until weaning; f8, group with feed f administered 8 d before parturition until weaning; s, high n-3:n-6 ratio feed supplemented with sunflower-seed oil; s3, group with feed s administered 3 d before parturition until weaning; s8, group with feed s administered 8 d before parturition until weaning.
* The feed contained either a low (supplemented with fish oil; f groups) or high (supplemented with sunflower-seed oil; s groups) n-6:n-3 ratio and was provided from 8 or 3 d before parturition (onset day 8 or 3).
No feed was provided on the day of parturition (D0). From the first day of lactation onwards until weaning, the sows were fed twice per d, namely at 08.00 hours and again at 18.00 hours. The amount of feed that was provided daily was 3·0 kg on D1 and D2, 3·5 kg on D3, 4·0 kg on D4 and D5, 4·5 kg on D6 and D7, 5·0 kg on D8 and D9, and 5·5 kg from D10 until D21. The amount of feed consumed by every sow was recorded daily from D1 until D21. The feeding troughs were emptied and checked each evening.
Sows were weighed on a digital scale when they were placed in the farrowing crates (day 107 of gestation; D-8) and at weaning (D21). On these occasions, also backfat measurements using a digital backfat indicator (Renco Lean Meter) were made of every sow. Backfat levels were measured at three positions on each side of the sows: position A (position A left and position A right), between the crossing backside of the shoulder and the spine, 5 cm to the left and 5 cm to the right; position C (position C left and position C right), in the crossing between the last rib and the spine, 5 cm left of the spine and 5 cm right of the spine; position B (position B left and position B right), 5 cm left and 5 cm right of the spine in between points A and C.
Blood samples were taken from every sow daily from D-4 (day 110 of gestation) until parturition (D0). Two additional blood samples were taken after farrowing, namely at D1 and D3. The samples were taken before the morning meal by puncture of the jugular vein in vacuum tubes. The parameters analysed in serum were leptin, glucose, insulin, triiodothyronine (T3), thyroxine (T4), VLDL, while TNFα, IL-6 and serum amyloid A (SAA) were analysed in plasma. An overview of the examined parameters in blood is given in Table 3. The rectal temperature of the sows was recorded with a digital thermometer (accuracy 0·1°C) at the first day of lactation, before morning feeding and before blood sampling.
Table 3 Overview of the different parameters that were analysed in blood from the seventy-two sows included in the study*
× , Parameter was analysed; T3, triiodothyronine; T4, thyroxine; SAA, serum amyloid A.
* Blood samples were taken daily from 4 d before parturition (D-4) until parturition (D0), and on D1 and D3 after parturition.
Piglets were individually identified and weighed within 1–2 h after completion of birth, or during the next morning when sows farrowed during the night. The piglets were individually weighed at D1, D2, D7, D14 and D21. The accuracy of the digital scale was 0·005 kg. The number of stillborn, the number of piglets that died until weaning and the number of piglets weaned was recorded. Cross-fostering of piglets was allowed only during D1 and only in between sows of the same experimental group. There was not an effort to equalise the litters.
Analytical procedures
Insulin was analysed by BioSource INS-IRMA KIP1251 and KIP1254 methods (BioSource Europe S.A., Nivelles, Belgium). Thyroid hormones (T3 and T4) were analysed by using a specific RIA. Leptin was determined using a commercially available RIA test kit (Multi-Species Leptin RIA Kit, catalogue number XL-85K; Linco Research, Inc., St Charles, MO, USA). Glucose was analysed by an enzymic method (Gluco-quant) on a Roche/Hitachi Modular. TAG were analysed with TAG GPO-PAP (Roche/Hitachi, Vilvoorde, Belgium) Reagent 1, 11876023 216. VLDL was determined by the Friedewald equation (TAG/5). TNFα was analysed by a porcine TNFα immunoassay (Quantikine; R&D Systems, Minneapolis, MN, USA). IL-6 was analysed by a porcine IL-6 quantitative sandwich enzyme immunoassay technique (Quantikine; R&D Systems). SAA was analysed by a solid-phase sandwich ELISA kit (Tridelta Phase™; Tridelta PLC, Bray, Co. Wicklow, Republic of Ireland). All samples were analysed in duplicate and in the same array, in order to avoid inter-array variation.
Statistical analyses
All examined parameters in blood, feed intake of the sows, litter weight and litter growth, piglet weight and piglet growth, backfat and weight of the sows were analysed with two-way ANOVA at each time point separately using SPSS 15.0 (SPSS, Inc., Chicago, IL, USA). This type of analysis was selected to enable a clear view at which time point significant differences of the examined parameters between the experimental groups occurred. Specifically, individual piglet weights were pooled per litter, and afterwards the mean individual piglet weight and mean individual piglet growth were used for the statistical procedures. Survival analysis, including the effect of mortality level in the different groups and the time of death, was used to analyse the difference in survivability of the piglets. The type of feed (fish oil v. sunflower-seed oil-supplemented) and the duration of providing the experimental feed (onset day 3 v. diet day 8) were included as fixed factors in the models. Correlations between the examined parameters were investigated using Pearson's correlations. Mean values with their standard errors were calculated for each experimental group separately (f3, f8, s3, s8). The level of significance was at P ≤ 0·05 (two-sided test).
Results
Piglets
Survivability of piglets between the four treatment groups is shown in Fig. 1.
Fig. 1 Survival curves for piglet survivability during lactation from the four experimental diet groups in which sows were assigned. The feed contained either a low (supplemented with fish oil; f groups) or high (supplemented with sunflower-seed oil; s groups) n-6:n-3 ratio and was provided from 8 d (f8 (-–-–), s8 (– – –)) or 3 d (f3 (—), s3 (- - -)) before parturition. Differences in survivability were not significant (P>0·05). No losses of piglets occurred after day 10 of lactation and until weaning (day 21).
Mean individual piglet weights, piglet growth, litter weights and litter growth in the four treatment groups are presented in Table 4. Piglet weight on D0, D1 and D2 was higher for the onset 8 groups than onset 3 groups (P = 0·008, P = 0·004, P = 0·002 at D0, D1 and D2, respectively). Piglet growth between D0 and D2 was higher in the onset 8 groups than in the onset 3 groups (P = 0·048).
Table 4 Performance data (piglet and litter weight, piglet and litter growth) during lactation of the piglets from the sows of the four experimental groups†
(Mean values with standard errors from pooled data)
f, Low n-3:n-6 ratio feed supplemented with fish oil; f3, group with feed f administered 3 d before parturition until weaning; f8, group with feed f administered 8 d before parturition until weaning; s, high n-3:n-6 ratio feed supplemented with sunflower-seed oil; s3, group with feed s administered 3 d before parturition until weaning; s8, group with feed s administered 8 d before parturition until weaning; o, onset day effect.
*P < 0·05, **P < 0·01.
† The feed contained either a low (supplemented with fish oil; f groups) or high (supplemented with sunflower-seed oil; s groups) n-6:n-3 ratio and was provided from 8 or 3 d before parturition (onset day 8 or 3).
Sow body condition and feed intake
The mean backfat levels at day 108 of gestation and at weaning were not statistically significant for all positions measured. At position A, f3 = 28·3 mm; f8 = 28·5 mm; s3 = 28·5 mm; s8 = 28·5 mm (P = 0·921); at position B, f3 = 21·7 mm; f8 = 22 mm; s3 = 22 mm; s8 = 22 mm (P = 0·853); at position C, f3 = 20·7 mm; f8 = 20·8 mm; s3 = 20·7 mm; s8 = 20·9 mm (P = 0·908). The differences in backfat levels between day 108 of gestation and weaning are indicated in Fig. 2.
Fig. 2 Losses in backfat levels measured at day 108 of gestation and at weaning in the sows of the four experimental groups: (
), f3; (
), f8; (
), s3; (□), s8. The feed contained either a low (supplemented with fish oil; f groups) or high (supplemented with sunflower-seed oil; s groups) n-6:n-3 ratio and was provided from 8 d (f8, s8) or 3 d (f3, s3) before parturition. Backfat was measured at three different positions: position A (pA) between the crossing backside shoulder and spine; position B (pB) at an equal distance between pA and position C (pC); pC in the crossing between the last rib and spine. p all, Average of backfat levels at the three positions. Values are means, with standard errors represented by vertical bars. No significant differences were found between the four experimental groups on the same position (P>0·05).
The sows belonging in the s group consumed less feed than the sows belonging in the f group during the first 2 d of lactation (for D1, P = 0·003; for D2, P = 0·014) (Fig. 3).
Fig. 3 Feed intake of sows during the first 7 d of lactation in the two diet-type groups. Sows were receiving two types of lactation diets differing in the n-6:n-3 ratio (the low ratio diet supplemented with fish oil; the high ratio diet supplemented with sunflower-seed oil), administered from two time points onwards before parturition. (○), Sunflower-seed oil-supplemented group; (●), fish oil-supplemented group. Values are means for each experimental group, with their standard errors represented by vertical bars. Mean value was significantly different from that of the sunflower-seed oil-supplemented group: *P < 0·05, **P < 0·01.
Sow metabolism and immunology parameters
The results of the parameters related to sow metabolism and immunology are presented in Table 5 and Figs. 4 and 5. On D-4, sows in the onset 8 groups had higher insulin levels than those in the onset 3 groups (P = 0·021). On D1, sows in the s group had higher insulin levels than the f group (P = 0·029) (Fig. 4). Insulin on D-1 was overall negatively correlated with litter weight (Table 6).
Table 5 Mean values of glucose, triiodothyronine (T3), thyroxine (T4), VLDL and TAG in the sows of the four experimental groups around parturition (D0)†
(Mean values with standard errors from pooled data)
f, Low n-3:n-6 ratio feed supplemented with fish oil; f3, group with feed f administered 3 d before parturition until weaning; f8, group with feed f administered 8 d before parturition until weaning; s, high n-3:n-6 ratio feed supplemented with sunflower-seed oil; s3, group with feed s administered 3 d before parturition until weaning; s8, group with feed s administered 8 d before parturition until weaning; o, onset day effect; t, diet type effect; to, interaction between diet type and onset day.
*P < 0·05, **P < 0·01, ***P < 0·001.
† The feed contained either a low (supplemented with fish oil; f groups) or high (supplemented with sunflower-seed oil; s groups) n-6:n-3 ratio and was provided from 8 or 3 d before parturition (onset day 8 or 3).
Fig. 4 Leptin (A, C) and insulin (B, D) in sows (n 71) during the periparturient period of − 4 d before farrowing to 3 d after farrowing. Sows received two lactation diets differing in the n-6:n-3 ratio (the low ratio diet supplemented with fish oil (●); the high ratio diet supplemented with sunflower-seed oil (○)), administered from two time points onwards before parturition (8 d (△) or 3 d (▲) before farrowing; onset 8 or 3 groups). (A) Leptin in the two diet-type groups. (B) Insulin in the two diet-type groups. (C) Leptin in the two diet-day groups. (D) Insulin in the two diet-day groups. Values are means, with standard errors represented by vertical bars. Mean value was significantly different from that of the fish oil-supplemented group: *P < 0·05, **P < 0·01. Mean value was significantly different from that of the onset 3 group: †P < 0·005, ††P < 0·01.
Fig. 5 TNFα (A), IL-6 (B), serum amyloid A (SAA; (C)) and rectal temperature (D) of the sows (n 71) at the first (□) and third day (▧) of lactation. Sows were receiving two types of lactation diets differing in the n-6:n-3 ratio (the low ratio diet supplemented with fish oil; the high ratio diet supplemented with sunflower-seed oil), administered from two time points onwards before parturition (8 d or 3 d before farrowing; onset 3 or 8 groups). f3, Fish oil–onset 3; f8, fish oil–onset 8; s3, sunflower-seed oil–onset 3; s8, sunflower-seed oil–onset 8. Values are means, with standard errors represented by vertical bars. a,b Mean values within the same day of measurement with unlike letters were significantly different (P < 0·05).
Table 6 Associations (Pearson's correlation; r ) between sow insulin levels 1 d before farrowing (D-1) and leptin levels on the fourth and third day before farrowing (D-4 and D-3 respectively) from the four experimental groups with litter weight at birth (Lw 0) and on days of lactation (Lw 1, 2, 7, 14 and 21)
*P < 0·05, **P < 0·01.
T3 on D-4 was higher in the onset 8 than in the onset 3 group (P < 0·001). On D-1, the interaction between diet type and onset day was significant for T3 levels: in the f group T3 was higher in f3 than f8, but in the s group T3 was higher in s8 than s3 (P = 0·001). On D1, T3 was higher in the onset 8 than the onset 3 groups (P = 0·001). On D-4, T4 concentrations were higher in the onset 3 than the onset 8 groups (P = 0·046). On D-1, T4 levels were higher in the f than the s group (P = 0·048).
Leptin levels on D-4, D-3 and D-2 were higher in the s than the f groups (P = 0·047, P = 0·005, P = 0·031, respectively) (Fig. 4). Leptin on D-2 was higher in the onset 3 than in the onset 8 groups (P = 0·003) (Fig. 4). When leptin data were pooled across all treatment groups, significant negative correlations were found between leptin on D-4 and leptin on D-3 with litter weight and litter growth in different stages (Tables 6 and 7) . Furthermore, significant positive correlations existed between pooled leptin and insulin postpartum values on D1 (r 0·432; P = 0·001) and D3 (r 0·453; P = 0·002).
Table 7 Associations (Pearson's correlation; r) between sow leptin levels from all experimental groups on the fourth and third day before farrowing (D-4 and D-3 respectively) with litter growth (Lgr) between different stages of lactation
*P < 0·05, **P < 0·01.
On D-4, onset 3 groups had higher VLDL and TAG concentrations than the onset 8 groups (P < 0·001 for both parameters). On D-1, s groups had higher VLDL and TAG concentrations than the f groups (P = 0·015).
TNFα, IL-6 and SAA concentrations between the experimental groups on D1 and D3 are shown in Fig. 5. SAA on D3 was higher for the onset 3 group than the onset 8 group (P = 0·039). Rectal temperature on D1 was higher in the s than the f group (P = 0·041) (Fig. 5).
Discussion
The earlier transition from the gestation to a lactation diet (8 v. 3 d) led to an increase in neonatal piglet weight and subsequent growth. Although the increase in fetal weight during late gestation in sows reaches a plateau(Reference McPherson, Ji, Wu, Blanton and Kim26), it appears that the time point of diet change during the last days of gestation can affect piglet weight. This effect might be due to an increased transfer of energy and nutrients to the piglets in utero. The lactation diets used in the present study were superior in these elements comparing with the gestation diet.
In the present study, differences in feed consumption in sows during the first days of lactation were significant, even though sows of all groups were fed restrictively on an ascending scale and not on an ad libitum basis after farrowing. Feed consumption was lower when sows were fed the high n-6:n-3 ratio lactation diet. This may be attributed to the higher leptin levels during the periparturient period detected in this particular group in comparison with the other group. It has been described previously that intracerebroventricular treatment with leptin decreased feed intake in a dose-dependent manner in prepuberal gilts(Reference Barb, Yan, Azain, Kraeling, Rampacek and Ramsay27). The role of insulin in this phenomenon should not be excluded. Central insulin action is catabolic (reducing feed intake and body weight), whereas its peripheral action is anabolic(Reference Niswender, Baskin and Schwartz28). The high n-6:n-3 ratio diet induced higher insulin levels in sows at D1. Therefore, a peripheral insulin resistance may have occurred, while the central action of insulin may have remained unaltered.
After farrowing, an increase in insulin was triggered in all treatment groups. Increased insulin levels after parturition may play a role in partitioning nutrients to the mammary tissue(Reference Bauman and Currie29) at the onset of milk production. Yet, as mentioned, insulin was higher at D1 when sows consumed the high n-6:n-3 ratio diet. Results from male miniature pigs suggest that substitution of n-3 for n-6 PUFA in dietary lipids is associated with enhanced insulin sensitivity(Reference Behme30). In addition, fasting hyperinsulinaemia is present early in the process of diabetes and is thought to be a compensatory mechanism to maintain euglycaemia in the setting of insulin resistance(Reference Herman and Kahn31). Moreover, comparatively high levels of insulin in sows during lactation correspond to lower catabolic status and lower milk production(Reference Rojkittikhun, Einarsson, Uvnäs-Moberg and Edqvist32). It is therefore possible that in comparison with the low n-6:n-3 ratio groups, a short period of peripheral insulin resistance might have been developed after parturition in the high n-6:n-3 ratio groups.
T3 was increased and T4 was decreased at D-4 in the groups that already have shifted to the lactation diets (onset 8 group). This indicated a higher conversion of T4 into T3. Again at D1, T3 was higher in that experimental group. The secretions of the thyroid gland are involved with many metabolic functions including stimulating O2 consumption and protein synthesis by the mammary gland and concomitantly increasing milk yield(Reference Tucker and Larson33). Impaired thyroid action had been speculated to be linked with the occurrence of the postpartum dysgalactia syndrome. Normal postpartum sows had significantly higher thyroid cell heights than either the agalactic or cyclic sows(Reference Wagner34).
Leptin levels were affected by acute feeding events before farrowing, such as the transition from the gestation to the lactation diets. Between D-4 and D0, the time profile of leptin in sows receiving the lactation diets from 3 d pre-partum showed an increase and a subsequent spontaneous decrease. During the same period, the sows receiving the lactation diets from 8 d pre-partum already had the highest concentration of leptin. Of course, due to the absence of samples before D-4, such a profile in the sows in the 8 d group can only be hypothesised.
In the present study pre-farrowing levels of insulin and leptin in sows appeared to be associated with piglet performance. Similar findings were observed in rats(Reference Woodside, Abizaid and Walker35), in which exogenous leptin given centrally in rats was associated with reduced litter weight gain. In addition, poor glucose tolerance of pregnant sows was reported previously(Reference Kemp, Soede, Vesseur, Helmond, Spoorenberg and Frankena25) to be a risk factor for survival of pigs after birth. The precise mechanism underlying this relationship remains to be elucidated. In pigs, placental glucose transport occurs by a carrier-mediated facilitated diffusion involving several glucose transport isoforms, mainly GLUT1(Reference Ashworth, Kyriazakis and Whittemore36, Reference Doberenz, Birkenfeld, Kluge and Eder37). Most probably, pre-farrowing maternal levels of leptin and insulin affect these energy-dependent mechanisms and consequently affect litter weight at birth.
VLDL and TAG concentrations at D-4 were lower in the groups receiving already the lactation diets. This may indicate an increased rate of β-oxidation in the liver(Reference Doberenz, Birkenfeld, Kluge and Eder37). VLDL and TAG concentrations were increased at 1 d pre-partum by a high dietary n-6:n-3 ratio lactation diet. This suggests increased hepatic lipogenesis in these sows. Increased lipogenesis might not be related to changes in thyroid hormones, but might be more related to the lipogenic activity of the increased insulin.
The sows belonging to the high n-6:n-3 ratio group had significantly higher rectal temperatures. This can be attributed to the well-described pro-inflammatory properties of the n-6 fatty acids(Reference Calder and Grimble3, Reference Calder4). At D3, SAA levels were significantly higher in the onset 3 groups. Thus, SAA levels in sows postpartum can be affected by factors other than pathogenic challenges, as previously reported(Reference Zhu, Österlundh, Hultén and Magnusson38).
In conclusion, a lactation diet high in n-6:n-3 ratio was associated with higher leptin levels pre- and postpartum, a short-term peripheral insulin resistance on the first day of lactation and a decreased feed intake of sows on the first and second day of lactation. An increase in leptin concentration in the sows receiving the gestation diet until day 111 of gestation occurred after the acute elevation of the energy level, by offering these sows the lactation diet. Besides the effects on leptin, the time point of transition from a gestation to a lactation diet profoundly influenced piglet weight at birth and at subsequent stages. Any separate effect of n-6 and n-3 PUFA and pre-partum feeding strategy on insulin, thyroid hormones, leptin, rectal temperature, feed intake of sows in the periparturient period and the piglet growth in the early days of lactation should not be disregarded, but these effects should be considered as a whole. This approach should be a prerequisite to a better understanding of the pathogenesis of postpartum dysgalactia syndrome from a clinical nutrition approach. According to the present results, a diet low in the n-6:n-3 ratio (fish oil-supplemented diet) provided 8 d before the expected farrowing was the most appropriate combination.
Acknowledgements
Expenses for the laboratory analyses were covered by the scholarship programme awarded to G. A. P. by the Greek State Scholarship Foundation for postgraduate study abroad (www.iky.gr). G. A. P. was responsible for the study design, study performance, parameter analysis, data analysis and manuscript drafting. T. A. T. G. V. K. was responsible for the supervision of logistic arrangements, experimental diet development, and had a contribution in manuscript drafting. J. B. supervised the analysis of blood parameters and had a contribution in the manuscript drafting. S. V. W. participated in the study performance and in manuscript drafting. D. G. D. M. and G. P. J. J. evaluated the study design, also contributed to data analysis and manuscript drafting, and both are promoters of G. A. P. It is declared by the corresponding author that no conflict of interest exists for this manuscript.
