The influence of supplementary feeding during suckling and later weaning on growth performance, incidences of post-weaning diarrhoea and immunity of piglets

There is an urgent call for alternative strategies to the use of antibiotics for treatment of diarrhoea during the first 14 days of the post-weaning period. The alternative strategies should be directed at enhancement of the robustness of pigs in terms of body weight at weaning, and contribute to proper functional development and maturation of the intestine before and after weaning. Therefore, both nutrition and physiological age are of high interest. We hypothesised that a liquid nutritional supplement during the suckling period


Introduction
Weaning is a stressful event associated with reduced feed intake and marked changes in the structure and function of the gastrointestinal tract (Kelly et al., 1991;Pluske et al., 1997).The immunological state of piglets at weaning is characterised by a gap in the protective immune system, caused by depletion of maternal immunity and their own under-developed components of innate immune function.This state can lead to dysfunction of the intestinal immune system making piglets highly susceptible to enteric pathogens after weaning (Adam et al., 2017).Moreover, the abrupt transition of the piglets from sow milk to solid feed creates a positive environment for colonization and proliferation of pathogenic strains of Escherichia coli (E.coli), the main pathogen involved in generation of post-weaning diarrhoea (PWD) (Fairbrother et al., 2005;Madec et al., 2000;Rhouma et al., 2017).Outbreaks of PWD have serious economic consequences for pig production worldwide due to production losses associated with sudden death and poor growth of the piglets.This leads to a high frequency of antibiotic usage in the post-weaning period, which provokes antibiotic resistance among pathogens (Gresse et al., 2017).
Zinc oxide (ZnO) was a common antimicrobial feed additive used widely in the European Union (EU) as an alternative to antibiotics, aiming to prevent and control PWD outbreaks in weaned pigs.However, widespread application of high doses of dietary ZnO (veterinary zinc) raised serious environmental, economic and public health concerns (Gresse et al., 2017).As a consequence, the European Commission withdrew all marketing authorizations for veterinary medicines containing ZnO in June 2022 (Directive 2001/82/EC, 2017).This could potentially increase the usage of antibiotics in the conventional pig production sector.Therefore, alternative strategies to ZnO and antibiotics usage are urgently needed to enhance the robustness of pigs before weaning and to prevent the outbreaks of PWD.In this context we studied dietary intervention (milk replacer and liquid feed) in combination with enhanced weaning age because such a strategy may form a more gradual transition from sow's milk to solid feed and by that avoid the abrupt diet change at weaning (feeding exclusively the solid plant-based feed) as common in practise; and it is a well-known fact that on litter basis only an average around 50% of piglets eat solid feed before weaning (Bruininx et al., 2001).
Many studies aiming to identify the alternative dietary strategies to ZnO and antibiotics usage have focused on the post weaning period and the risk of developing diarrhoea caused by E. coli infection, and studies such as reduction of dietary protein level and manipulation with amino acids composition (Engelsmann et al., 2023;Heo et al., 2008;Lynegaard et al., 2021) seem to be convincing too consider a dietary strategy post-weaning.Less knowledge is however available on the influence of dietary treatments of the suckling piglet to develop robustness of the piglet before weaning, because such studies has primarily been carried out per se using neonatal piglets or preterm piglets and inclusion of such feed additives as prebiotics (de Vries et al., 2020;Wu et al., 2020), probiotics (Hansen et al., 2022;Menegat et al., 2018), organic acids, and feeding dairy by-products (Thymann et al., 2012) and milk replacers (de Greeff et al., 2016).However, these studies have not addressed the impact on robustness.According to Munoz et al. (2022) robustness is defined as the ability to maintain, cope, and withstand the influence of adverse events.In this context, the weaning from the dam could be considered as an external disturbance of the piglet, where the evaluation of physiological functions associated with productivity and health preand post-weaning will allow for assessing piglets' robustness.Therefore, the robustness before and after weaning in the present study was assessed by using the following markers of production performance such as weight gain before and after weaning, feed intake during the first week after weaning, and estimation of the body composition and the diarrhoea related measurements (faecal consistency score, faecal dry matter content and the excretion of E.coli toxins) and immunological measurements (concentration of immunoglobulins and specific antibody titre) were chosen in this context as health indicating parameters.
Previous findings have indicated that piglets' nutrition during the suckling period influencing piglets' gastro-intestinal health and maturation before and after weaning (Madec et al., 1998).Thus, the usage of supplementary milk replacer (Azain et al., 1996;King et al., 1998;Wolter et al., 2002) and skim milk (Dunshea et al., 1999) during the suckling period had increased weaning weight, which contributed to rapid growth after weaning (Cabrera et al., 2010) and lowered risk of PWD occurrence (Madec et al., 1998); however, intake of supplementary feed during suckling period is highly variable across and within litters (Kuller et al., 2007).While dry creep feed is a frequently applied nutritional strategy during the suckling period (Bruininx et al., 2004;Byrgesen et al., 2021), sparse scientific knowledge is available regarding the influence of liquid nutritional supplementation during the suckling period on pigs' performance in the post-weaning period and development of PWD (Canibe and Jensen, 2003;Xu et al., 2023).Moreover, a number of studies demonstrated that the later weaning (weaning after day 28 of age) is associated with improved development and functionality of the gastro-intestinal tract (Massacci et al., 2020), reduced stress level among weaned pigs (van der Meulen et al., 2010) and improved growth (Nielsen et al., 2022).Thus, functional development and maturation of the intestine, which is necessary for prevention of the infectious diseases during the post-weaning period could be accomplished by early nutritional interventions during suckling period and increased weaning age.Therefore, the present study was designed to investigate the possible improvement of piglets' robustness prior to weaning measured in terms of growth performance, body fat content, immune parameters and the prevention of PWD.
It was hypothesised that the combination of increased weaning age and the provision of liquid nutritional supplementation during the suckling period would enhance piglets' robustness assessed as growth performance, health and immunity before and after weaning.

Material and methods
The experiment was carried out with permission from The Danish Animal Experimentation Inspectorate, and the animal experiment was conducted according to a license obtained by the Danish Animal Experiments Inspectorate, Ministry of Food, Agriculture and Fisheries, Danish Veterinary and Food Administration (license number: 2018-15-0201-01484).The animals used in the experiment were followed by proper veterinary surveillance throughout the experiment.Housing and rearing of the animals were in compliance with Danish laws and regulations for humane care and use of animals in research.The health of the animals was monitored daily, and illness was treated by trained personnel.

Animals and facilities
The experiment involved two blocks of 24 sows (Landrace × Yorkshire) with a congruent parity number (parity 1 and 2), thus sows in first block had parity 1 (n = 24) and sows in the second block had parity 2 (n = 24); and their progeny n = 720 piglets [(Landrace × Yorkshire) × Duroc], born at the experimental facility at Aarhus University, AU Viborg, Research Center Foulum.The experiment was carried out in fall 2018 and spring 2019, so the sows involved in first block were used again in block 2, beside a minor replacement of 5 sows.This experiment was a part of large study investigating the impact of feeding interventions of gestating sows (Feyera et al., 2021) and behavior (Kobek-Kjeldager et al., 2021b) during suckling and early post-weaning periods.Animals were kept under controlled environmental conditions (temperature and humidity, depending on age).In the farrowing unit, sows (in farrowing crates) and piglets were housed in individual in pens (2.2 m × 2.4 m) with a partly slatted floor, installed with heating lamps for the piglets and with ad libitum access to water.The ambient temperature was 21 • C.There was a rope in every pen, and sows were provided with moderate amounts of straw as nest building material prior to farrowing.After farrowing, sawdust was provided for piglets.All farrowings were supervised, and assistance by skilled personnel was provided if necessary.All litters were standardized to 15 piglets, in order to meet the study design criteria of behavioral study running in parallel with the present study, which was investigating udder competition behavior of piglets born in large litters.A live weight of 800 g was considered as a threshold value for litter standardization; thus, piglets below 800 g were euthanized, and no cross-fostering sows were used in the experiment.Once litters have been standardized, and no piglets were moved between the litters afterwards until weaning at day 24 and day 35 of age respectively.The main criterion during the litter standardization procedure was the following: the piglets should be kept with their own dams.After weaning, piglets were kept in groups of 25-28 piglets per weaning pen, and the groups were composed of 2 litters of the same treatment (combination of feeding strategy and weaning age).The weaner pens had 2/3 of slatted floor and 1/3 of concrete floor that were covered (two-climate pen).The pens had a feed trough, a drinking nipple, and a water trough that was automatically refilled.Wood shavings were provided in the weaner pens.

Experimental design
The experiment was arranged as a 2 × 2 factorial block design with 2 blocks of 24 sows each.The two factors were the nutritional treatment with or without supplementation (milk replacer/wet feed) and the weaning age (day 24 v day 35 of age).The experimental treatments (n = 4) comprised each combination of the two factors.Initially, 12 litters were provided with milk/feed while the other 12 litters only had access to suckle the sow.Within each nutritional supplement (milk/feed+/-) half of the litters were weaned at day 24 of age, and other half at day 35 of age.Thus, 12 litters weaned at day 24 (n = 12), WEAN24+ (n = 6) and WEAN24-(n = 6) treatments with +/-nutritional supplementation, and 12 litters weaned at day 35 (n = 12), WEAN35+ (n = 6) and WEAN35-(n = 6) treatments with +/-nutritional supplementation, respectively.In the second block (parity 2), due to health issues 2 sows and their litters had to be withdrawn from the experiment 2 days prior to planned weaning, therefore the respective treatments WEAN24-and WEAN35included only 5 sows each.
Litters subscribed to the treatments with inclusion of nutritional supplementation were from day 2 of age supplied with milk replacer (Pigipro 1 Milk Care, 3S, Schills, The Netherlands) by an automatic system (Babydos Bopil, Sønderborg, Denmark) with a feeding trough (semicircle with radius of 10 cm), where no activation was needed from the piglets.On day 12, the milk replacer was automatically changed to liquid feed, consisting of a Danish standard starter meal diet (Vestjyllands Stjerne Care, Ringkøbing, Denmark) mixed with water immediately before automatic feeding.The liquid feed was then supplied until weaning (day 24/day 35).The system was set to the highest frequency, which was approximately four feedings per hour, to achieve ad libitum access to the nutritional supplement.
The milk replacer was based on whey protein powder, and the liquid feed was based on soya protein concentrate, heat-treated corn and dehulled oats.Samples of the dietary treatments were analyzed for the nutritional composition as presented in Table 1.After weaning, the piglets were provided with the same feed (Vestjyllands Stjerne Care, Ringkøbing, Denmark) as given from day 12 until weaning, but in dry form, until the end of the study (day 49 of age).The sow diet was a standard diet composed as recommended for lactating sows (Tybirk et al., 2018) based on soya bean meal, barley and wheat, and was delivered by a commercial feed plant (Vestjyllands Andel, Ringkøbing, Denmark).

Registrations and sample collections
The individual weight of piglets was recorded at birth, every week during the suckling period and at weaning.After weaning, pigs were weighed individually on days 3, 7, 14 and at the end of the experiment (day 49 of age).Feed intake was recorded every day during the first week post-weaning.
Blood and faecal samples were collected from pigs at weaning and at the end of the experiment, blood samples were collected from all piglets in block 1 and from selected piglets in block 2; and no faecal samples was collected in block 1.In block 2, based on the results of individual colostrum intake (Vodolazska et al., 2023a), six piglets (three with high colostrum intake and three with low colostrum intake) per litter were selected for blood samples and faecal samples; blood samples were collected at day 3, 24, 35, and 49 of age, and faecal samples were collected at day 3, 7, and 14 post-weaning.Faecal samples were collected from the rectum, using a sterile cotton swab dipped in glycerine.Samples were stored at -80 • C for further analyses.
Blood samples were collected from the jugular vein in EDTAcontaining vacutainers for whole blood haematology, and blood samples intended for later analyses were collected in heparinised vacutainers (Vacuette, Greiner Bio-One GmbH, Kremsmünster, Austria).Further, plasma was separated by centrifugation at 1 850 × g and stored at -20 • C until being analyzed for concentrations of immunoglobulins (IgG, IgM and IgA).
To determine the body composition of piglets at the end of the experiment, the deuterium dilution technique was employed and carried out as described by Theil et al. (2002).In total, 306 piglets at day 49 of age were selected (2 pens per experimental treatment) and enriched with deuterium oxide (D 2 O) in the neck by an intramuscular injection of 20% D 2 O solution at a rate of 0.5 g per kg live weight.To calculate the mass of the D 2 O infusate injected in the piglet, the weight of the syringe was recorded before and after injection.Blood was sampled before injection of D2O solution from jugular vein into serum vacuum tubes (5 mL Vacuette Serum; Greiner Bio One International GmbH, Kremsmünster, Austria) to measure the background level, and again 2 h after injection to be able to calculate the deuterium dilution space and the total body water of each enriched piglet.The samples were centrifuged at 1 850 × g for 10 min at 20 • C. Serum was harvested and stored at − 20 • C until analysis.Calculations of piglet body composition were performed as described by Hojgaard et al. (2020).

Analytical methods
Immediately after collection of the blood samples, the haematological analysis was performed as a diagnostic health-monitoring tool, using a haematology analyzer (IDEXX ProCyte Dx, IDEXX Laboratories, United States).The following haematological parameters were obtained: total leucocytes, neutrophils, lymphocytes, monocytes, eosinophils, erythrocytes, haematocrit, haemoglobin, reticulocytes, thrombocytes, the mean cell volume, the mean corpuscular haemoglobin and the mean corpuscular haemoglobin concentration.
In order to evaluate the occurrence of PWD at the post-weaning period, the consistency of faeces was evaluated by applying the faecal consistency score as described by Carstensen et al. (2005) where 1: hard, dry and cloddy, 2: firm, 3: soft with shape, 4: soft and liquid, 5: watery and dark, 6: watery and yellow.Score >3 was defined as a clinical sign of diarrhoea.The dry matter content of the faeces was determined by freeze-drying (ScanVac Coolsafe 55, Labogene ApS, Lynge, Denmark) the faecal samples to a constant weight (in a vacuum oven at an operating pressure of 0.4 mbar for 48 h).Further, the faeces samples were analyzed for specific toxins produced by enterotoxigenic E. coli, such as the heat-stable E. coli enterotoxin (STb) and heat-labile E. coli enterotoxin (LT2) as well as all E. coli forms in the faeces.Quantification of E. coli and the enterotoxins LT2 and STb in faeces samples were analyzed with qPCR according to the procedure described by Forootan et al. (2017).
Plasma concentrations of IgA, IgG and IgM were measured as representative of humoral immune status using a commercial kit (pig ELISA quantification kit; Bethyl Laboratories, Montgomery TX).For determination of titres of specific antibodies in piglets' plasma, E. coli O149:F4 strain 9910045-1 (Jensen et al., 2013) and E. coli O138:F18 strain 9910297-2 STM (Frydendahl et al., 2003) were used for coding the plates, followed by ELISA.Titre values are reported as arbitrary values (i.e., the last dilution [× 10 2 ] that gave a positive reaction) as described by Lauridsen and Jensen (2005).

Statistical analyses
Statistical analyses were performed using the software R v. 4.1.2(R Core Team, 2022), including packages nlme, glmmTMB and emmeans, at a significance level of 0.05 and with 0.05 < P ≤ 0.10 declared as 'trends'.Fixed effects were generally tested by chi-squared likelihood ratio tests after maximum likelihood estimation, while final estimates and standard error of the mean (SEM) in linear mixed effects models (LME) were obtained by restricted maximum likelihood estimation (REML).Assumptions of normality and variance homogeneity were checked and if not fulfilled, relevant heterogeneity was included, or a transformation (square root or log) was applied.Only data from second parity were available for average daily gain (ADG), feed intake and IgA after weaning, and for faecal excretion of all E.coli and LT2, STb toxins and dry matter in faeces.
Body weight was modeled longitudinally by LME using days 2, 7, 14 and 21 of age for both weaning ages, day 24 for WEAN24 and day 35 for WEAN35.Parity and sow within parity were included as random effects, the model allowed variance heterogeneity over days and repeated measures within pigs were handled by a continuous-time auto-regressive covariance structure of order 1 (corCAR1).Fixed effects were day (categorical), weaning age and milk replacer/wet feed, 2-and 3-way interactions among these, but interactions that include weaning age and day, only for days 2 to 21 (by use of indicator functions).Interactions that include both weaning age and milk replacer/wet feed factors were not significant but kept in agreement with the experimental design to allow for individual estimates.
Concentrations of IgG, IgA and IgM in piglets' plasma measured at day 3, body weight at day 49 and ADG 0-7, 0-14 and 7-14 days after weaning were analyzed by LME with weaning pen and sow as random effects.For ADG 0-7 days and 7-14 days, the variance was allowed to depend on weaning age and, moreover, IgG, ADG 0-14 and ADG 7-14 days were square root transformed.Fixed effects were weaning age, milk replace/wet feed and their interaction, and as a covariate also the scaled weight at weaning (standardised within treatment group).For the analyses of the body components water, fat and ash, mixed effects beta regressions were used, while the protein was analyzed using LME, with weaning pen and sow within weaning pen as random effects and milk replacer/wet feed and weaning age as fixed effects.The concentrations of IgG, IgA and IgM and of specific antibodies to E.coli O138 and E.coli O 149, in piglets' plasma post-weaning were analyzed after logtransformation by LME as described for body weight, except fixed effects here consisted of day (categorical 24/35/49), weaning age, milk replacer/wet feed and all interactions among these.The model for IgA only included random effect of sow (second parity).Individual faeces score was analyzed by a mixed effects logistic regression where the score >3 was considered as diarrhoea.
Pairwise comparison among treatments was adjusted for multiple comparisons by the Tukey-Kramer method and significant differences are indicated by lower case letters.When relevant, results will be presented after back-transformation to the original scale.

Growth performance, body composition and feed intake
The increased weaning age was positively associated with piglets' body live weight at weaning (Table 2), as average body weight at weaning was 6.93 kg and 6.77 kg for WEAN24-/+ and 9.59 kg and 9.46 kg for WEAN 35-/+ respectively.No significant difference was obtained between the groups with or without nutritional supplementation with regard to weights prior to and at weaning.In spite of this, the provision of nutritional supplement was positively associated with greater average daily gain in the first week after weaning (P = 0.04).Moreover, increased weaning age was positively associated with greater average daily gain observed during the first two weeks after weaning (P < 0.001) and the mean body content of fat at day 49 of age (Fig. 1).The analysis of piglets' body composition indicated that the mean body content of fat and protein was greater among the pigs weaned at day 35 of age compared with pigs weaned at day 24 (P < 0.001), as the mean body content of fat was 11.5% and 14.2% for WEAN35-/+, and 3.83% and 4.59% for WEAN24+/-, respectively.
Main results obtained for feed intake after weaning were influenced by a significant interaction between weaning age and day after weaning (P < 0.001, Fig. 2), i.e., feed intake was positively associated with increased weaning age.Furthermore, litters weaned at day 35 and provided with nutritional supplement during the suckling period had the greatest average feed intake after weaning (P < 0.001, Fig. 2).

Faecal dry matter
Faecal dry matter content was influenced by nutritional supplement provided prior to weaning, weaning age and day of sampling (Table 3).The pattern observed for all experimental groups indicated a decreased faecal dry matter content during the first week after weaning (P = 0.002).Furthermore, on average, higher faecal dry matter content was observed among the piglets provided with nutritional supplement during the first two weeks after weaning (P = 0.05).In general, piglets weaned at day 24 demonstrated higher faecal dry matter content compared with piglets weaned at day 35 (P < 0.001).Furthermore, the faecal dry matter content below 20% was reflected in faecal consistency score >3 and it was defined as a clinical sign of diarrhoea; on individual level.The results for the faecal consistency score indicated that the probability of having score >3 was highest for pigs weaned at day 35 of age (P = 0.03) without nutritional supplementation (P = 0.04).However, there was no difference between the groups with regard to frequency of medical treatment after weaning (data not shown).

Haematological profile at weaning
All haematological parameters analyzed at weaning were affected by piglets' weaning age (Table 4).Furthermore, the values observed for eosinophils, haematocrit and reticulocytes were positively associated with provision of nutritional supplement before weaning.Thus, those parameters were elevated in piglets provided with nutritional supplement (P = 0.04, P = 0.02 and P = 0.007 for eosinophils, haematocrit and reticulocytes, respectively).Additionally, the total white blood cells, lymphocytes, neutrophils to lymphocytes ratio, haematocrit, haemoglobin and reticulocytes were affected by body live weight of piglets.

Concentrations of immunoglobulins A, G and M in piglets' plasma
The concentrations of IgG, IgA and IgM in piglets' plasma at day 3 of age were elevated compared to other days of sampling, and, in addition, this pattern was observed for all combinations of weaning age and nutritional supplement (Table 5).For concentrations of IgG and IgM (P < 0.001) there was significant interaction between weaning age and day of sampling, thus elevated concentrations of these immunoglobulins were observed in older piglets.Concentration of IgA was not affected neither by weaning age nor by nutritional supplement (Table 5).

Specific antibodies against E. coli O138 and O149 serotypes
With regard to the measurement of E. coli O138 antibodies in plasma of piglets, there were significant interactions between day of sampling and weaning age (P < 0.001) and between day of sampling and The faecal dry matter content below 20% was reflected in faecal consistency score >3 and it was defined as a clinical sign of diarrhoea; on individual level. 1 WEAN24-/WEAN24+ weaned at day 24 without (-) or with (+) nutritional supplementation; WEAN35-/WEAN35+ weaned at day 35 without (-) or with (+) nutritional supplementation.Nutritional supplementation -milk replacer provided from day 2 shifted to wet feed at day 12 of age.
2 Days of sampling in the post-weaning period.
3 P-values obtained from type-2 test; P-values are reported for weaning age (WEAN) day 24 and day 35; nutritional supplement +/-(SUP), and their interaction and day of sampling.a-c Values within a row with different superscripts differ significantly and reflecting the difference between weaning ages after.
nutritional supplement (P = 0.04; Table 6).Thus, the lowest arbitrary values were found in plasma of piglets at weaning (day 24 and day 35), and the values increased after weaning.Remarkably, the provision of nutritional supplement prior to weaning resulted in numerically higher arbitrary values of E. coli O138 antibodies at day 49 of age for both weaning ages: day 24 and day 35 respectively.For E. coli O149 antibodies in serum of piglets, there was a 3-way interaction between day of sampling and weaning age and nutritional supplement (P = 0.05; Table 6), however, the patterns were similar to those described above with regard to E. coli O138 antibodies.

Faecal excretion of total E. coli, STb and LT2 toxins
For faecal samples collected after weaning the excretion of total E. coli and STb and LT2 toxins were associated with an interaction between weaning age and day of sampling (Table 7).The amount of total E. coli DNA detected in faeces was markedly greater in the youngest piglets and, in general, decreased with age (P < 0.001).The STb toxin detected in faeces showed the opposite pattern and increased with age (P < 0.001).The LT2 toxin excretion in the faeces was greatest at weaning for both weaning ages (day 24 and day 35) and in general increased with age (P = 0.02).

Discussion
This study showed that increased weaning age contributed to greater weaning weight, and that nutritional supplementation during the suckling period improved feed intake and growth performance of the piglets in the post-weaning period.These findings were expected, as previously reported results showed that increased weaning age in combination with creep feeding during the suckling period contributed to greater weight at weaning and better piglet performance after weaning (Bruininx et al., 2004;Byrgesen et al., 2021;Cabrera et al., 2010;Faccin et al., 2020;Kobek-Kjeldager et al., 2021b;Turpin et al., 2016;van der Meulen et al., 2010).Considering that weaning-associated anorexia immediately after weaning is a well-known predisposing factor of PWD (Rhouma et al., 2017), the obtained increase in post-weaning feed intake of the piglets weaned at day 35 calls for further research aiming to enhance newly weaned piglets' feed intake before and immediately after weaning in order to prevent the development of the condition favorable for generation of PWD.However, the results of the present experiment have demonstrated that the supplemental nutrition during the suckling period in combination with increased weaning age may ameliorate the abrupt decline in post-weaning feed intake of the piglets; contrary to no supplemental nutrition during suckling and weaning age at 24.
The specific antibodies against E.coli measured in plasma were used as an indicator of immunological activity of pigs in the present study, however, they may also reflect the natural exposure of piglets to E. coli due to their presence in the barn environment.Specific antibodies against E. coli would start to increase around 4 days post inoculation in pigs challenged experimentally with E.coli (Luise et al., 2019), and it should also be reflected in diarrhoea.When considering piglets weaned at day 35 without nutritional supplementation, the results of the faecal score showed higher proportion of piglets with clinical signs of diarrhoea, based on faecal score and faecal dry matter content.This observation was however not supported by the levels of the specific antibodies against E.coli in plasma of those piglets, as the highest arbitrary values were observed in piglets' plasma weaned at day 35 and provided with nutritional supplement.It should be also emphasised that no difference was detected in numerosity of medical treatments applied in all experimental groups.Therefore, the most likely explanation is that lack of dietary experience and abrupt change from milk to solid feed among the piglets weaned at day 35 resulted in looser faeces.These results are in agreement with those obtained by Callesen et al. (2007) who observed lower faeces score among piglets weaned at 33 days.In addition, our study demonstrated that provision of nutritional supplement prior to weaning contributes to higher dry matter content in faeces in the first week after weaning.
As an indicator of the weaning-associated exposure of piglets to pathogenic E. coli, the excretion of LT2 and STb toxins in faeces was measured, and the observed increase of excretion of these toxins with progression of post-weaning period was also reflected in the increase in specific E. coli antibodies in plasma of the piglets.Remarkably, piglets weaned at day 35 showed higher arbitrary values of specific E. coli antibodies and plasma IgG and IgM concentration, which implies a more active immune response in these piglets.This result was expected considering that piglets weaned at a higher age have a more mature immune system than piglets weaned at younger ages (Stokes, 2017).
Furthermore, the litters provided with nutritional supplement during the suckling period have shown the greatest levels of the antibodies against E.coli after weaning.This observation could be ascribed to the fact that provision of a nutritional supplement during the suckling period affected the gut microbiota of those piglets, and such changes may in fact influence the gut immune system, and may be the driver of the development of the active immune system (Bailey et al., 2005).In addition, the recent report of Vodolazska et al. (2023b) has showed a significant impact of investigated factors (nutritional supplement and weaning age) on total short-chain fatty acids concentration in pigs intestine (reported result is from the same animal experiment as presented in the current study).
Furthermore, considerable difference was observed in haematological profiles of piglets weaned at day 24 compared to those weaned at day 35 of age.This finding is in agreement with expectations as age and physiological state are the most important factors determining the values of blood parameters in animals (Czech et al., 2018).Therefore, changes observed in red blood and white blood cell parameters were age related and associated with increase of piglets' body mass, development and maturation of their organs.However, the results for blood parameters at weaning indicated that piglets in the group WEAN35-were to some degree more anaemic, compared to other groups.The possible explanation is that these piglets were characterised by the greatest mean body live weight at weaning.Therefore, considering that both observed parameters (haematocrit and haemoglobin) are dependent on plasma volume (Nguyen et al., 2020) and assuming that organs responsible for red blood cells formation and maturation were somewhat compromised in this period, the signs of anaemia could appear.On other hand the results of the research of Collard (2009), Ziegler et al. (2014), Dong et al. (2020) who studied iron requirements of both human infants and newborn pigs have showed that the rapid growth during the early stage of life requires an increased production of red blood cells and consequently the iron which contained in systemic iron reserves, which are limited in young humans and animals and often the iron present in maternal milk cannot cover fully their demands.Thus, the observed anaemia among the piglets in group WEAN35-could be explained by this physiological phenomenon.However, it is necessary to emphasise that all haematological parameters for every group obtained in this experiment were within the normal physiological range and at a similar level to those reported by Czech et al. (2018).Surprisingly, no beneficial impact of the nutritional supplement was observed with regard to growth performance before weaning in the groups of same weaning age.However, it is necessary to emphasise that the range of beneficial effects reported in the literature have been addressed to the feeding of milk replacer as the supplemental nutrition before weaning (Dunshea et al., 1999;King et al., 1998;Pustal et al., 2015), while in our study milk replacer was provided in a short period of time (10 days), and replaced with the wet feed afterwards.In addition, the lack of information regarding pigs' individual feed intake during suckling period make the interpretation of the absence of pronounced effect of nutritional supplementation complex.The feed intake could not be obtained during the suckling period, due to technical characteristics of the automatic feeding system used in the experiment (system could not perform the weighing of the feed that remain in the trough).Based on the evidence found in the literature, the quantitative measure of the consumption of supplementary feed by suckling piglets in some cases requires the specific equipment for weighing the feed.In their study Pajor et al. (1991) obtained the individual creep feed consumption of piglets by calculation of the loss of weight of the feed dispenser before and after the feeding bout.Another option described in literature is usage of digestibility markers, such as 1% chromic oxide (Kuller et al., 2007) or 1% red ferric oxide (Byrgesen et al., 2021), but this method provides a visual assessment of the feed intake, as mentioned markers colored the piglet feces if the creep feed was eaten, and the amount of consumed feed reflected in the intensiveness of the faces color (green color for chromic oxide or red color when ferric oxide was used).However, with regard to our experiment, the results of the conducted in parallel study using litters and sows from the present experiment have indicated that almost all piglets (98%) ingested nutritional supplement (Kobek-Kjeldager et al., 2021b).
The prolonged suckling period for piglets weaned at day 35 resulted in considerably higher body fat depots contrary to those weaned at day 24.The obtained results could be explained by a lower nutritional value of the provided wet feed compared to sow milk, which is considerably rich in fat (30 to 40% on dry matter basis) as shown in Lauridsen and Danielsen (2004).This indicates, that more studies focused on development of the nutritional strategies meeting the optimal energy and nutrient requirements of suckling piglets is needed.
Most scientific literature has investigated the possible effects of provision of milk replacer without changing to wet feed during the suckling period (Amdi et al., 2021;Greeff et al., 2016;Kobek-Kjeldager et al., 2021a); and to our knowledge only one previous study has investigated the effect of liquid creep in comparison to dry creep feed on feed disappearance, number of eaters and intestinal enzymatic development (Byrgesen et al., 2021).The results reported Byrgesen et al. (2021) showed that provision of liquid feed during suckling period had no significant beneficial effect on piglets' growth and gut maturation when compared with dry creep feed.Although the studies cannot be compared as such due to a difference in experimental design and set up.
Our results showed that provision of the supplementary feeding during the suckling period increased feed intake and faecal dry matter content during the first week after weaning, which is considered as beneficial impact with regard to gut development and maturation in the early post-weaning period.However, the disability of obtaining the information of individual feed intake during suckling make the interpretation of this result to some extent complex and call for more studies.

Conclusion
As expected, this study indicated that increased weaning age improved weight at weaning.Furthermore, the increased weaning age contributed to higher fat percentage and hence energy depots at day 49 of age, and had a significant impact of immunological, haematological parameters as well as feed intake after weaning.However, none of the interventions (supplementary feeding or weaning age) applied in the present experiment had significant effect on preventing the occurrence of PWD during the first week after weaning.Overall, our study contributes to the existing knowledge with regard to feeding programs during suckling period aiming to improve piglets' robustness before weaning.Furthermore, the results of the present study have important implications in the context of improving piglets' weight at weaning and early feed intake and fat depots after weaning.However, more studies are needed to reveal the impact of interventions applied in this study on maturation and mucosal immunity of piglets' intestine in early postweaning period.In context of future studies in should be emphasised that precise assessment of individual feed intake per se is required in order to link this information to the gut health of the individual piglet.

Table 1
Analyzed chemical composition of milk replacer and feed used in the experiment.
1Pigipro 1 Startprovided from day 2 to day 12 postpartum; powder was mixed with water in the ratio of 150 g: 1 L.2 Vestjyllands Stjerne Careprovided as liquid feed from day 12 to weaning and solid feed from weaning to the end of the study (day 49 of age).D.Vodolazska et al.

Table 2 .
Effect of nutritional supplement (milk replacer/ wet feed) and weaning age (day 24 vs. day 35) on body weight in the entire experimental period, and ADG in the postweaning period.Nutritional supplementation -milk replacer provided from day 2 shifted to wet feed at day 12 of age Values are presented as least squares means and SEM.Presented SEM is a greatest value of all calculated for the particular day.

Table 4 .
Effect of nutritional supplement (milk replacer/ wet feed) and weaning age (day 24 vs. day 35) on haematological parameters of piglets at weaning.
1 WEAN24-/WEAN24+ weaned at day 24 without (-) or with (+) nutritional supplementation; WEAN35-/WEAN35+ weaned at day 35 without (-) or with (+) nutritional supplementation.Nutritional supplementation -milk replacer provided from day 2 shifted to wet feed at day 12 of age.Values are presented as least squares means and SEM.Presented SEM is a greatest value of all calculated for the particular blood parameter.2P-valuesobtained from type-2 test; P-values are reported for weaning age (WEAN) day 24 and day 35; nutritional supplement +/-(SUP), and their interaction.3Presented values are back-transformed from natural logarithm.* Values affected by piglets' body weight.

Table 5
Effect of nutritional supplement (milk replacer/wet feed) and weaning age (day 24 vs. day 35) on concentrations of immunoglobulins A, G and M (IgA, IgG and IgM) in plasma of piglets at day 3, 24, 35, 49 of age.Nutritional supplementation -milk replacer provided from day 2 shifted to wet feed at day 12 of age.Values are presented as least squares means and SEM.Presented SEM is a greatest value of all calculated for the particular immunoglobulin within the same day of sampling.
2Days of age and sampling. 3ata only from parity 1. 4 P-values obtained from type-2 test; P-values are reported for weaning age (WEAN) day 24 and day 35; nutritional supplement +/-(SUP), and their interaction, and interaction between weaning age (WEAN) and day of sampling (Day); three ways interaction (Day of sampling × SUP × Weaning age) is not significant for all immunoglobulins; day 3 was analyzed separately, therefore P-value is not available for WEAN × Day.a-dValues within a rows and columns of each immunoglobulin with different superscripts differ significantly after Tukey correction for multiple testing; for IgA letters indicate the difference between sampling day as effect of day is significant (P < 0.001).* Presented values are back transformed from square root (IgG) and natural logarithm (IgM).

Table 6
Effect of nutritional supplement (milk replacer/ wet feed) and weaning age (day 24 vs. day 35) on specific antibodies to E.coli O149 and E.coli O138 serotypes in piglets' plasma at weaning and the end of the experiment (day 49 of age).Nutritional supplementation -milk replacer provided from day 2 shifted to wet feed at day 12 of age.Values are presented as least squares means and SEM.Presented SEM is a greatest value of all calculated for the particular serotype.2Days of age and sampling. 3P-values obtained from type-2 test; P-values are reported for interactions between weaning age (WEAN) day 24 & day 35 and day of sampling (Day); nutritional supplement +/-(SUP) and day of sampling (Day); weaning age (WEAN) day 24 & day 35 and nutritional supplement +/-(SUP); three ways interaction (Day of sampling × SUP × Weaning age); for O149 P = 0.24 (WEAN) and P = 0.61 (SUP); for O138 O149 P = 0.24 (WEAN) and P = 0.61 (SUP).a-g Values within a rows and columns of each serotype with different superscripts differ significantly after Tukey correction for multiple testing.

Table 7 .
Effect of nutritional supplement (milk replacer/wet feed) and weaning age (day 24 vs. day 35) on faecal excretion of total E. coli, STb and LT2 toxins 1 in the postweaning period.Nutritional supplementation -milk replacer provided from day 2 shifted to wet feed at day 12 of age.Values are presented as least squares means and SEM.3 P-values obtained from type-2 test; P-values are reported for weaning age (WEAN) day 24 and day 35; nutritional supplement +/-(SUP), and their interaction, and interaction between weaning age (WEAN) and day of sampling (Day); three way interaction (Day of sampling × SUP × Weaning age) significant P = 0.002 for all E.coli and it is not significant for LT2 and STb.
4Days of age and sampling.a-dValueswithin a row with different superscripts differ significantly after Tukey correction for multiple testing.D.Vodolazska et al.