Influence of silver nanoparticles on growth and health of broiler chickens after infection with Campylobacter jejuni

Silver nanoparticles (AgNP) have gained much attention in recent years due to their biomedical applications, especially as antimicrobial agents. AgNP may be used in poultry production as an alternative to the use of antibiotic growth promoter. However, little is known about the impact of oral administration of AgNP on the gut microbiota and the immune system. The aim of the present study was to investigate the effects of AgNP on growth, hematological and immunological profile as well as intestinal microbial composition in broilers challenged with Campylobacter jejuni ( C. jejuni ).

The application of AgNP via the drinking water in the concentration of 50 ppm reduced broiler growth, impaired immune functions and had no antibacterial effect on different intestinal bacterial groups, which may limit the applicability of AgNP against C. jejuni in broiler chickens.

In broiler production, different kinds of antimicrobial agents are used for preventing and controlling diseases. Antimicrobials can affect the host intestinal flora, by reducing the colonization of intestinal bacteria, inhibiting the growth of pathogenic microorganisms, and enhancing the immune system, hence preventing diseases and improving animal performance [ 1 – 3 ]. However, the overuse of antimicrobial agents (antibiotics) promotes the emergence of antibiotic resistance in microorganisms [ 4 ], being harmful to animal and human health [ 5 , 6 ]. For example, resistance to ciprofloxacin in Camplylobacter jejuni ( C. jejuni ) isolated from Danish broiler meat increased significantly from 0% in 2009 to 17% in 2010 [ 7 ]. The use of all antibiotics as growth promoter has been prohibited in the European Union since 2006 [ 2 , 8 ]. Thus, there is a need to find alternatives to antibiotics in poultry production. When the conventional therapies, including antibiotics, anti-inflammatory agents, growth hormones, surgical interventions, and cytotoxic chemotherapies are ineffective in curing poultry infections, it is necessary to explore novel drug compounds. Nanoparticles have been emerging as one of the new treatment options, and their capability of penetrating normally intact physiologic barriers has reached a variety of molecular targets [ 9 , 10 ]. Recent studies on antibacterial materials such as various natural (oils, acids), inorganic antimicrobial agents such as metals (Ag, Au, Cu) and metal oxides (ZnO, SiO 2 , Fe 2 O 3 , TiO 2 ) have received increasing attention. Among metal nanoparticles, silver is one of the most promising components in several nanotechnology products. Currently, there are several consumer products containing various silver nanoparticles (AgNP) because of their antimicrobial properties [ 11 – 14 ]. AgNP have been shown to have a wide range of antibacterial activities against both Gram-positive and Gram-negative bacteria, including major foodborne pathogens [ 15 – 19 ]. At present, there is no study available on their antibacterial effect against C. jejuni being a leading cause of human gastroenteritis worldwide. It is mainly transmitted from contaminated chicken meat. Campylobacter infections are associated with the neurological disorder Guillain-Barre syndrome [ 20 , 21 ]. A common feature of C. jejuni causes enterocolitis and is involved in acute inflammatory response that can lead to tissue damage and may be responsible for many of the clinical symptoms [ 22 ]. Furthermore, antimicrobial resistance was observed in C. jejuni and Campylobacter coli [ 23 ]. Although the antimicrobial effect of AgNP has been studied extensively, the mechanism of antibacterial activity specific to bacteriostatic or bactericidal activity remains unclear. Studies had shown that AgNP upon contact with water can release Ag + ions from their surface [ 24 ]. Free Ag + has a potent antimicrobial effect, which destroys microorganisms immediately by blocking the cellular respiration and disrupting the function of bacterial cell membranes. This occurs when Ag + binds to tissue proteins, causing structural changes in the bacterial cell membranes which, in turn, cause cell death. Essential protein complexes of the bacterial electron transport chains are located on the cell exterior and, therefore, are manageable for inactivation by reactive silver ions. Ag + also binds and denatures the bacterial DNA and RNA, thus inhibiting cell replication [ 25 ]. Recently, evidence has been obtained suggesting that silver nanoparticles may modulate the phosphotyrosine profile of putative bacterial peptides that could affect cellular signaling and, therefore, inhibit the growth of bacteria [ 26 ]. In recent years, several studies have been focused on anti-inflammatory therapy and on molecules which could block pro-inflammatory pathways. Nanoparticles of Ag and Au are considered anti-inflammatory agents or components of anti-inflammatory molecules [ 27 , 28 ]. Moreover, in vivo studies with chicken embryos and quails showed that AgNP did not affect growth, development [ 29 ] and DNA oxidative damage to chicken embryos [ 30 ]. Results from toxicological assays have shown no in vitro cytotoxicity of AgNP (0.1, 0.5 and 1.0%) [ 17 ] but concentrations (2.5–50 μg/mL) of AgNP exert a cytotoxicity effect on human mesenchymal stem cells [ 31 ]. The most common health effects associated with chronic exposure to silver are a permanent grey or blue-grey discoloration of the skin (argyria). From the immunological perspective, it is known that phagocytosis of AgNP stimulates inflammatory signaling through the generation of reactive oxygen species (ROS) in macrophage cells, followed by the activated macrophage cell-induced secretion of TNF-α. The increase of TNF-α level causes damage to the cell membrane and apoptosis [ 32 ]. An inadvertent recognition of AgNP as a foreign particle by the immune cells may result in a multilevel immune respons

Ninety day-old male broiler chickens (Ross 308), obtained from a Danish commercial hatchery (DanHatch, Vrå, Denmark) were used. Upon arrival at the laboratory, the birds were wing-labeled, weighed and randomly distributed to two experimental groups: control, no AgNP, and provided with 50 ppm of AgNP in the drinking water. Furthermore, prior to the experiment antibacterial tests were done at in vitro level. The chickens were housed in six individual isolators with 15 birds per isolator. The room temperature was 32 °C at the beginning of the experiment and was gradually decreased according to the demands of the growing chickens. The humidity was maintained during the experiment. Over the entire experimental period, both experimental groups were fed the same diet as described in Table 1 . Drinking water was provided via nipple drinkers placed in the isolators. Birds were inspected every day and for every 2 days water and fed consumption were recorded. All birds had free access to food and water. Table 1 Composition of the diet for broiler chickens (g/kg) Ingredients Wheat 49.9 Maize 10.0 Rape seed (LL), grounded 4.00 Soybean meal (de-hulled, toasted) 29.4 Soybean oil 2.45 Calcium carbonate 0.90 Monocalcium phosphate 1.56 Sodium chloride 0.20 Natrium-bicarbonate 0.27 Lysine hydrochloride (100%) 0.25 DL-Methionine (100%) 0.35 Threonine (98%) 0.10 Vitamins and minerals (Vitfoss, Slut, 0.5%) 0.60 Total 100 At day 11, all chickens were weighed and cloacal swab samples of 5 chickens per isolator were taken and examined for the presence of C. jejuni colonies for initial observation. Subsequently, two birds were randomly selected from each treatment and killed, and then blood samples were collected for packed cell volume (PCV) determination before samples of liver tissue were collected for gene expression studies. Afterwards, the chickens were orally challenged with 0.5 mL of an overnight culture of C. jejuni (4 × 10 7 / mL per bird). The infection strain (DVI-sc181) has been isolated from infected commercial broilers and provided by the Technical University of Denmark, National Veterinary Institute. On days 15, 22 and 30, all chickens were weighed, and 5 chickens per isolator were randomly chosen. Blood samples were taken from the jugular vein and the chickens were immediately sacrificed by cervical dislocation. Liver, heart, spleen and bursa were collected, weighed and stored at -80 °C for further analysis. Ileum and caecum were collected to enumerate intestinal bacteria.

Before cervical dislocation, blood was collected by puncture of the jugular vein in disposable sodium heparinized hematocrit capillary tubes (Camlamb Ltd., Cambridge, UK). The tubes were filled up to two-thirds and sealed with cristaseal (Hawksley, Sussex, UK). The percentage of packed cell volume (PCV) was measured by using a micro-hematocrit reader (Hearaeus Reader, Osterode, Germany).

Liver tissue samples (approximately 30 mg) were homogenized in TRIzol® Reagent (Invitrogen, Carlsbad CA, USA, cat. no. 15596–018). Then, RNA clean-up was performed and total RNA was extracted according to the manufacturer’s protocol using SV Total RNA isolation system (Promega Corporation, Madison, WI, USA, cat. no. Z3105). Final RNA preparations were resuspended in RNase-free water and stored at -80 °C. The content of isolated RNA was quantified by UV-spectroscopy at 260 nm/280 nm with a NanoDrop® ND-1000 Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). Further quality was assessed by Agilent 2100 bioanalyser (Agilent RNA 6000 Nano kit, Waldbronn, Germany) and more than 6.5 RNA integrity number values were considered for cDNA preparation. Two mg of total RNA was reverse-transcribed using the cDNA Reverse Transcription Kit (Promega, cat. no. A3500) in a G-storm PCR System. After this, real-time PCR was performed with complementary DNA and gene-specific primer pairs (TAG, Copenhagen A/S, Copenhagen, Denmark) mixed with LightCycler®480 SYBR Green I Master mix (Roche Applied Science, Penzberg, Germany) in a LightCycler® 480 real-time PCR system (Roche Applied Science). The cycling conditions included an initial heat step at 95 °C for 5 min, denaturation at 95 °C for 10 s, annealing temperatures described in Table 2 for each of the primers and product elongation at 72 °C for 20 s for 45 cycles. Following amplification, the melting curve was conducted on each sample to ensure that a single product was obtained using three-segment cycle of 95 °C for 5 s, 65 °C for 1 min and 95 °C no hold for continuous-acquisition mode with a heating rate of 0.11 °C/s and 5 acquisitions per 1 °C). Characterization of the product size was, furthermore, confirmed by agarose gel electrophoresis and sequencing (Bechman coulter genomics Takeley, UK). Results were quantified based on the relative expression of the TNF-α and NF-kB genes versus the housekeeping genes GAPDH and ACTB using advanced relative quantification (Efficiency method) by Light Cycler 480 Software release 1.50 SP4. The gene expression experiment was repeated 2 times chemically with consistent results and three replicates were used per sample each time. Table 2 Genes and primers used in the study RNA target gene Primer sequences(5′–3′) PCR Product Gene bank Assess. No Annealing Temperature Amplicon (bp) GAPDH NM_204305.1 Forward TGCTGCCCAGAACATCAT 61 °C 199 Reverse ATCAGCAGCAGCCTTCAC ACTB NM_205518.1 Forward GTCCACCTTCCAGCAGATGT 60 °C 169 Reverse ATAAAGCCATGCCAATCTCG NF-kB M86930.1 Forward TTGCTGCTGGAGTTGATGTC 60 °C 167 Reverse TTGCTGCTGGAGTTGATGTC TNF-α NM_204267.1 Forward TTCAGATGAGTTGCCCTTCC 59 °C 150 Reverse TCAGAGCATCAACGCAAAAG GAPDH Glyceraldehyde-3-phosphate dehydrogenase, ACTB Actin beta, NF-kB nuclear factor kappa B, and TNF-α Tumor necrosis factor alpha

Enumeration of bacteria There were no differences between birds receiving AgNP and the control group (Fig. 1 ) with respect to the numbers of C. jejuni, lactic acid bacteria , Enterococci, Clostridium perfringens, Escherichia coli , and Lactose negative enterobacteria in the contents of caecum, ileum and feces of the birds. Fig. 1 Influence of AgNP administration in broiler chickens after challenge with C. jejuni on microbial profile in caeca, ileum and feces. At 15, 22, and 30 days of age, samples ( n = 30 for each time point) were used for microbial count. Error bars represent the mean and standard errors of 6 isolators each with 15 birds

The body weight of birds supplemented with AgNP was significantly lower than that of control bids (Table 4 ). Significant interactions between age and treatment were observed for the relative weight of bursa and spleen, where a lower relative bursa at age 15, and for spleen at age 30 were noticed for AgNP group. Relative organ weights of heart and liver were not different between the control birds and birds supplemented with AgNP. Table 4 Effects of silver nanoparticles (AgNP) on cumulative body and relative organ weight gain (g/kg) in C. jejuni infected broiler chickens Parameters (g/bird) Mean Pooled Treatment*Age Control AgNP SE p-value AWG Age 15 380 a 318 b Age 22 788 a 722 b Age 30 1424 a 1346 b 0.305 0.013 Heart Age 15 0.71 0.70 Age 22 0.62 0.66 Age 30 0.47 0.50 9.17 0.340 Bursa Age 15 0.25 a 0.20 b Age 22 0.21 0.22 Age 30 0.20 0.20 5.08 0.003 Spleen Age 15 0.09 0.11 Age 22 0.07 0.07 Age 30 0.07 a 0.05 b 2.31 <0.001 Liver Age 15 2.71 2.82 Age 22 2.52 2.56 Age 30 2.01 2.00 0.003 0.657 Abnormalities At age 30, big gall bladder was found from 2 out of 15 birds in AgNP group *Indicates interaction a, b values within rows with different superscripts are significantly different at p <0.05. SE – pooled standard error, AWG – average body weight gain. The relative organ weights (weight of organ/100 g live body weight). Values are mean of 6 isolators each with 15 birds

We observed significant interactions between treatment and age for the concentrations of IgA, IgG and IgM in plasma (Fig. 3 ). At day 30, the plasma concentration of IgA was higher in birds supplemented with AgNP than in control birds, whereas the opposite effect was observed at age 15 and 22 days ( p = 0.05, Fig. 3 a). At day 30, the concentration of IgG in plasma was significantly lower in birds receiving AgNP as compared to the control birds (Fig. 3 b). At age 15, plasma concentrations of IgM were higher in the AgNP group than in the control group (Fig. 3 c), whereas at day 22 and 30, the concentrations of IgM were lower ( p < 0.05). Fig. 3 Concentration of immunoglobulins (IgA, IgG and IgM) in chickens infected with C. jejuni Samples were analyzed at days 15, 22 and 30 of age; a ) IgA b ) IgG and c ) IgM. Error bars represent the mean values and standard errors of 6 isolators, each with 15 birds. Interactions between treatment and age effect significantly ( p < 0.05). * Indicates significant difference between control and AgNP ( p < 0.05)

In vitro antibacterial tests against C. jejuni showed a minimal inhibitory concentration of AgNP at the level of 40 and 50 ppm. AgNP have an effect on the target microorganism in the absence of chickens. In the present in vivo experiment, 50 ppm of AgNP was used because, prior to this experiment we carried out dose response studies to examine the antibacterial effect against C. jejuni of different concentrations of AgNP on chickens by chicken intestinal organ culture model (CIOC-model, Additional file 2 : Table S2) . The minimum inhibitory and minimum bactericidal concentration of silver nanoparticles on C. jejuni was determined by using the microtiter plate method (Additional file 3 : Figure S1), showed bactericidal effect with minimum concentration of 50 ppm. We have evaluated samples of ileum, caecum and feces because of the uniformity of the gut microbiota in these segments. The caecum has a slow digesta passage rate, allowing to harbor a complex microbiome that has considerable effects on host nutrition and health [ 38 ]. The small intestinal region of the ileum has received attention since it is the principal site of nutrient absorption and microflora [ 39 ]. Furthermore, we focused on test samples at various time points because the caecal and ileal microflora changes in relation to age and dietary treatments [ 38 , 39 ]. However, unexpectedly, AgNP did not change the microbial profile of caecum, ileum and feces of examined chickens (Fig. 1 ). These results are consistent with in vivo experiments with the microbial profile of young quails receiving hydrocolloids of AgNP administered with 5, 15 and 25 ppm [ 29 ]. Moreover, another study demonstrated that supplying AgNP did not effectively rescue Salmonella - mediated mortality in chickens but nanoscale silicate platelet (NSP) and its nanohybrid composite of AgNP/NSP effectively controlled the infection [ 40 ]. However, our present study was not similar with this study in terms of end point of examination, concentration of nanoparticles, bacterial infection time points and target bacterial species. The factors that could affect the present results might be the method of AgNP administration, although C. jejuni colonization in the intestinal tract depends on the number and diversity of environmental microorganisms, as well as influence of feedstuffs. The other possibilities could be the C. jejuni biofilm formation and adhesion to the host intestinal wall may provide protection against nanoparticles. The chicken intestinal mucus is able to attenuate C. jejuni virulence by inhibiting its ability to adhere and invade intestinal epithelial cells [ 41 ]. Furthermore, the antibacterial properties of AgNP could vanish inside the digestive tract environment of chickens, probably, due to AgNP agglomeration in the presence of low pH gastric acid. The obtained results may suggest that AgNP were gastro-sensitive, the stability and dispersion of AgNP in gastric acid is a critical factor for antibacterial activity. Once the particle enters a biological system, physical properties such as solubility, particle agglomeration, surface charge and particle-protein complex interactions might be different from those of the in vitro-measured properties. The loss of nanoparticles’ properties may be due to the low colloidal stability and the reduction of reactive surfaces may affect the efficacy for controlling pathogens in a solution [ 42 ]. The reduction in cumulative body weight and relative organ weights of the spleen and bursa of birds supplemented with AgNP might be due to the effect of AgNP blocking the intestinal absorption of actively transported sugars and amino acids, and decreased protein digestibility through the small intestine where mainly enzymatic activity is present. Another possible reason might be the fact that the birds undergo more cellular stress and excessive cellular interactions with AgNP. It appears that C. jejuni colonization may lead to weight loss in grower chickens. The present results are consistent with decreased body and organ weights in chickens treated with 25 ppm of AgNP at 42 days of age [ 43 ]. Moreover, the study of Asharani et al. [ 44 ] suggested that AgNP increase ROS production and interrupt ATP synthesis, leading to DNA damage and cell cycle arrest at G2/M phase. Park et al. [ 32 ] found that AgNP induce G1 phase arrest and a complete blockage of the S phase, with the induction of apoptosis. The AgNP (10 ppm) injected into fertilized eggs on days 5, 11 and 17 of incubation did not influence the development of embryos but decreased the number and size of lymph follicles in the bursa of Fabricius [ 45 ]. The PCV results indicated that the provision of AgNP did not influence the percentage of red blood cells (Fig. 2 ). On the other hand, it was reported that the oral administration of AgNP induced some changes in the red blood compartment, such as increased red blood cell count [ 46 ] and coagulation parameters [ 47 ]. The present results

Additional file 1: Table S1. Average daily water intake and the concentrations of silver nanoparticles (AgNP) when chickens were infected with C. jejuni. (DOCX 13 kb) Additional file 2: Table S2. Antibacterial effect of silver nanoparticles (AgNP) against C. jejuni by using chicken intestinal organ culture model (CIOC-model) . AgNP concentrations 10 ppm, 20 ppm were provided via drinking water, 40 ppm and 80 ppm were supplemented in intestinal organ culture. Bacterial count was done by standard plate count method. (DOCX 12 kb) Additional file 3: Figure S1. Effect of silver nanoparticles (AgNP) on C. jejuni at various concentrations, using broth microdilution method by microtiter plate. Bacterial density was measured by the plate counter. D20 to D70 ppm means diluted AgNP concentration from stock solution, C80 and C50 ppm means stock AgNP solution. Media, with bacteria were considered as negative and positive controls. The mean values of 3 repetitions. 1a) Bacterial growth (Optical density (OD)) difference = Average of bacteria - Average of media. 1b) Bacterial growth difference (%) = Bacterial growth difference / Average of bacteria. (DOCX 20 kb)

Data supporting the findings is found in the main paper and additional supporting files. Raw data files will also be shared by the corresponding author upon request.

All authors declare that they have no competing interests.