Probiotic supplementation as an alternative to antibiotics in broiler chickens

One-day-old ROSS 308 broiler chickens were hatched at the Malec Poultry Hatchery (Góra Kalwaria, Poland) and then delivered to the National Veterinary Research Institute (NVRI) in Puławy, Poland. All chickens were reared at the animal research facility of the NVRI.

Four groups of 20 Ross 308 broiler chickens originating from young parent stock were housed for six weeks in BSL 3 animal facilities at the NVRI. Two replicates per supplementation were conducted, and therefore in total 160 birds were used for the studies. In the experiment, four different feeding schemes were evaluated. The first group was fed a normal diet, and neither antibiotic nor supplementation was used. This was the control group, C.

The second group was comprised of chickens maintained on a normal diet but also given supplementation with probiotics, prebiotics and vitamins. The probiotic strains were Enterococcus faecium, Lactobacillus casei, Lactobacillus plantarum and Pediococcus acidilactici, and they were all used at a final concentration of 1.5 × 10¹¹ colony-forming units (CFU)/mL. Additionally, supplementation was provided of prebiotic fructooligosaccharides and the prebiotics inulin, maltodextrin and oligofructose at a final concentration of 250,000 mg, of vitamins B1 at 350 mg, B2 at 250 mg, B3 niacinamide at 2,000 mg, B5 panthenol at 1,200 mg, B6 at 320 mg, B12 at 1,000 μg, C at 30,000 mg, K3 at 300 mg, and folic acid at 3,000 mg, and of glucose, and silica. The broilers took these supplements in drinking water on the following scheme of three cycles: the first on the 2^nd, 3^rd, 4^th and 5^th days of life, the second on the 17^th, 18^th and 19^th days of life, and the third on the 31^st, 32^nd and 33^rd days of life. This was the pro- and prebiotic group, P.

The third group were the chickens fed the normal diet and antimicrobials. The drugs were enrofloxacin 500 mg/g (15 mg/kg body weight), given on the 1^st, 2^nd, 3^rd and 4^th days of life, fortamox (0.03 g/kg body weight) on the 14^th, 15^th and 16^th days of life, and trimsulfasol on the 24^th, 25^th and 26^th days of life in doses as follows: 10 mg/kg body weight, and 1.5 mL/L in drinking water, respectively. This group was referred to as the AO group.

The fourth group consisted of chickens provided with a normal diet with probiotic, prebiotic and vitamin supplementation, and also administered antibiotics. This group was designated P&AO.

The base diet was the normal diet for broiler chickens until six weeks old. The daily feed intake was 100 g/kg body weight. The concentration of live bacteria in the probiotic and prebiotic preparation was verified. The broiler chickens were examined on the 1^st, 21^st and 42^nd days of life.

The average weekly gain for each group of chickens was monitored and birds were weighed individually on day 1, and then weekly until the end of the experiment on the 42^nd day of life. The experiment was conducted in duplicate. Chickens were provided ad libitum access to feed and water during the 42-day trial.

Chickens were vaccinated at the hatchery according to the standard vaccination schedule for broiler chickens. On day 14 after hatching, the birds were vaccinated with the infectious bursal disease Nobilis Gumboro D78 IBD vaccine (Intervet International, Boxmeer, the Netherlands) in accordance with the standard vaccination scheme.

One week after vaccination, 21-day-old chickens from all groups were challenged with an infectious bronchitis virus (IBV) field strain designated G052/16 Ip CAS/SPF 2016-09-20 CT:15 IBV/Var2 with an EID₅₀ = 10^5.0. The strain was obtained from the NVRI collection. The experimental challenge was conducted at the NVRI animal facilities in a high containment unit.

After the disease challenge model, the birds were observed clinically. The immune status as antibody levels, all parameters connected with the presence of pathogens, body mass index, health parameters and mortality rate were assessed in individual birds. Anatomopathological examinations of dead and euthanised birds were conducted.

Samples were collected from each of the four experimental groups. Birds were euthanised by cervical dislocation after isoflurane administration in accordance with the protocol submitted to the local animal ethics committee. Anatomopathological examinations were conducted during which the hearth, liver, spleen, gizzard, intestines, lungs, and kidneys were collected and changes in them were described.

Serum samples taken from one-day-old and three- and six-week-old chickens were tested for the presence of antibodies against specific viral antigens: chicken anaemia virus (CAV), avian orthoreovirus (ARV), infectious bursal disease virus (IBD), infectious bronchitis virus (IBV) and fowl adenovirus (FAdV). All samples were tested with ELISA immunoassays dedicated to these poultry pathogens in accordance with the manufacturers’ instructions. The assays used were an IDEXX CAV Ab Test (IDEXX Laboratories, Westbrook, ME, USA); ID Screen Avian Reovirus Indirect, ID Screen IBD Indirect and ID Screen Infectious Bronchitis Indirect (all Innovative Diagnostics, Grabels, France); and FAdV Group 1 Antibody test (BioChek, Reeuwijk, the Netherlands). Sera from chickens were also tested for the presence of antibodies against the antigens Mycoplasma gallisepticum and M. synoviae with an IDEXX MG/MS Ab Test (IDEXX Laboratories).

Changes in the internal organs were assessed according to the following scale: no visible changes; visible minor changes in the form of hyperaemia; medium-grade hyperaemia of the mucous membrane with minor petechiae observed; and strongly expressed hyperaemia of the mucous membrane, marked petechiae, and enlargement and swelling of internal organs.

Microbiological identification was performed using 5% horse blood agar, MacConkey agar, and brilliant green agar culture media (Condalab, Madrid, Spain). Bacterial colony identification was performed using the standard method for mass spectrum profile (MSP) identification with a matrix-assisted laser desorption/ionisation–time-of-flight Biotyper (Bruker Daltonics, Bremen, Germany).

Statistical analysis was carried out using the Mann–Whitney U test and, where possible, Student’s t-test. The normality of distributions was tested with the Shapiro–Wilk test, and the homogeneity of variance with Levene’s and the Brown–Forsythe tests. The relationships described were statistically demonstrated at the assumed significance level of α = 0.05. Analyses were performed using Statistica version 13 (TIBCO Software, Palo Alto, CA, USA).

The clinical signs observed in infected chickens included depression, ruffled feathers, coughing, and tracheal and respiratory lesions. The mortality rate in chickens from the control group vaccinated and infected with the IBV/Var2 G052/16 strain was 10% at 7 and 12 days post infection (dpi). No mortality was observed in chickens from the P, P&AO or AO groups. All birds from groups P, P&AO and AO responded as expected to pathogen contact following appropriate immunisation (Table 1).

There were no macroscopic changes in the liver, while the kidneys were clearly enlarged and hyperaemic in the AO and P&AO groups. In these groups, no erosions or mucosal lesions were noted. No macroscopic changes were observed in other internal organs. In the P&AO and AO groups, inflammatory changes were observed in the duodenum, manifested by hyperaemia of the mucous membrane and oedema of the liver and kidney ranging in severity in 8 out of 10 birds in the AO group and in 6 out of 10 in the P&AO group. In all chickens, various degrees of change related to congestion of the mucous membrane of the small intestine were observed; these changes observed in the control group may have indicated the influence of the stress factor. In the AO and P&AO groups, significantly more marked inflammatory changes were observed. These changes could have been caused by antibiotics. In two chickens from the control group characteristic inflammatory changes in the digestive tract have been observed. In the control group, the birds were smaller, with lower production parameters, and lesions in the trachea and respiratory system were noted until the 7^th day after challenge. Data are presented in Table 2.

Serological examinations were conducted on serum samples obtained from all 20 birds per group. The level of maternal antibodies was determined to evaluate the parental flock vaccination programme. The vaccination of the experimental birds and their immunity were also assessed on the basis of the obtained results. The possibility of the occurrence of health problems was determined by comparing the chickens across groups. Mean antibody titres (arithmetic and geometric) and the coefficient of variation were calculated and documented. The results obtained for antibodies against infectious anaemia virus suggested that the chickens maintained maternal antibodies up to two weeks of age. The highest levels were found in the P and P&AO groups.

The results of tests for antibodies against infectious bursal disease virus demonstrated levels of maternal antibodies in one-day-old chickens of all groups. No clinical symptoms of IBD infection were observed in the examined birds however still a few positive serological results concerning characteristic IBD antibodies presence were recorded.

Noteworthy is the high average titre and high coefficient of variation (52.2%) in the six-week-old birds from the P&AO and AO groups. However, the remaining birds had satisfactory levels of antibodies and the applied immunoprophylactic programmes appeared to be effective.

A high coefficient of variation (83.0%) applied to the anti-infectious bronchitis virus antibody titres in the P group in the six-week-old chickens. An even higher coefficient of variation was found in one-day-old chickens in the C group, where neither probiotics nor antibiotic were supplemented or administered, and this was 85.3%. Production of the antibodies in the P&AO group was induced by contact with the field virus.

The presence of specific antibodies to avian reoviruses was also reported, which play a major role in protecting chickens against reovirus infection. There was an even level of maternal antibodies in one-day-old chickens of all tested groups, where the coefficient of variation of 21.5% in the P&AO group compared to one of 29.5% in the AO group.

The flocks were also tested for mycoplasma infection. The presence of specific antibodies to M. synoviae and M. gallisepticum was not correlated with the degree of immunity. Hence, the antibodies did not protect against infection, clinical signs or transmission. No specific antibodies against M. gallisepticum were found in any of the chicken groups tested. In contrast, the presence of M. synoviae antibodies was found in group C in one-day old chickens of this age in seven tested samples. Antibodies to M. synoviae were also found in one sample from group P and one from P&AO, as well as in seven samples from the AO group. In three-week-old and six-week-old chickens, no antibodies were found. Equal maternal antibody titres to poultry adenoviruses were also noted across the groups in one-day-old chickens.

The obtained results suggested that chickens from the P and P&AO groups showed a stronger vaccination protection index against pathogens commonly existing in poultry flocks, which was also closely correlated with a higher rate of weight gain in the P&AO group.

Escherichia coli and Enterococcus faecium were detected in microbiological studies carried out on the livers, yolk sacs and hearts collected from one-day-old chickens from the P group. Escherichia coli growth was abundant in samples of the yolk sacs of P group birds. Growth of Escherichia coli and Enterococcus hirae was found in liver and heart samples collected from three-week-old chickens in this group. Escherichia coli, Streptococcus alactolyticus and Aerococcus viridans were detected in liver samples from the six-week group of chickens treated with the antibiotics (the AO group), and internal organs collected from one-day-old chickens in this group showed the presence of Enterococcus casseliflavus, Enterococcus hirae, Enterococcus faecium and Escherichia coli, while no increase in bacterial flora was found in the heart samples. The presence of Enterococcus hirae and faecium was also confirmed in the yolk-sac samples in P group. Streptococcus alactolyticus and Staphylococcus cohnii were isolated in samples from the livers of three-week-old birds, and the former along with Staphylococcus xylosus was found in heart samples. From liver samples from six-week-old chickens, Proteus mirabilis was isolated, and Escherichia fergusonii and Candida rugosa were detected in heart samples from AO group.

In the control group of chickens after one day of life, Corynebacterium stationis, Streptococcus alactolyticus and Enterococcus casseliflavus were detected in the liver. In the heart, Escherichia coli and Streptococcus alactolyticus were confirmed. In the third week, bacterial growth of Escherichia coli and Enterococcus hirae was found only in liver samples. In the heart samples collected from six-week-old chickens, besides the bacteria species Enterococcus casseliflavus, the species durans was also found. In liver samples, Enterococcus gallinarum and Acinetobacter lwoffii were confirmed.

In one-day-old chickens supplemented with probiotics, prebiotics, vitamins and antibiotics – the P&AO group – the growth of bacteria was found only in liver samples and the species were Enterococcus hirae and faecium and Escherichia coli. In liver samples taken from six-week-old chickens, the following bacterial species were found: Staphylococcus hominis and cohnii, Streptococcus alactolyticus, Corynebacterium spp., and Escherichia coli. Only Escherichia coli was identified in heart samples in six-week-old chickens.

The average body weights of chickens at 7, 14, 21, 28, 36 and 42 days of life were analysed by the Mann–Whitney U, Shapiro–Wilk, Levene’s, Brown–Forsythe and Student’s t-test. Statistically significant differences were noted for the individual groups P, AO, and P&AO when compared with C, with a P-value <0.0500 being considered to be significant. The reduction of weight gain was significant from 28 dpi until the end of the experiment (P <0.0500) (Fig. 1). In the comparison of group P with group C, the day 1 P-value was 0.0001, the day 21 P-value was 0.0001, the day 28 comparison produced 0.0007 and for day 36 it was 0.0006. This analysis was by Mann–Whitney U test (continuity corrected). The condition takes into account group AO and group P&AO.When the AO group average body weights were compared with those of group C, a P-value of 0.0002 emerged for day 21, one of 0.0002 was also calculated for day 28, the value was unchanged for day 36, and using Student’s t-test for the day 42 comparison the P-value was 0.0338. Except the noted exception on day 42, the P-values were found in a Mann–Whitney U test (continuity corrected). The condition takes into account group P and group P&AO. The P&AO group was also analysed against group C. Body weights measured on day 1 were statistically different, shown by a P-value of 0.0002; those on day 21 also were and the resulting P-value was 0.0011, weights on day 28 gave a P-value of 0.0022 and those on day 36 returned 0.0028. As in other groups’ analyses, a Mann–Whitney U test (continuity corrected) was applied. The condition takes into account group P and group AO.

Fig. 1

Based on the Shapiro–Wilk test, the hypothesis of normal distribution should be rejected in the case of day 1 body weights in groups AO, P&AO and C; day 28 body weights in group P&AO; and day 36 values in group AO.