Potential of Lactobacillus plantarum IBB3036 and Lactobacillus salivarius IBB3154 to persistence in chicken after in ovo delivery

Abstract The aim of this study was to characterize and compare selected Lactobacillus strains originating from different environments (cow milk and hen feces) with respect to their applicative potential to colonize gastrointestinal track of chickens before hatching from an egg. In vitro phenotypic characterization of lactobacilli strains included the investigation of the important prerequisites for persistence in gastrointestinal tract, such as a capability to survive in the presence of bile salts and at low pH, enzymatic and sugar metabolic profiles, adhesion abilities, and resistance to osmolytes, temperature, and antibiotics. Regarding the resistance of lactobacilli to most of the various stress factors tested, the milk isolate Lactobacillus plantarum IBB3036 showed better abilities than the chicken feces isolate Lactobacillus salivarius IBB3154. However, regarding the acidification tolerance and adherence ability, L. salivarius IBB3154 revealed better characteristics. Use of these two selected lactobacilli isolates together with proper prebiotics resulted in the preparation of two S1 and S2 bioformulations, which were injected in ovo into hen Cobb500 FF fertilized eggs. Furthermore, in vivo tests assessing the persistence of L. plantarum IBB3036 and L. salivarius IBB3154 in the chicken gastrointestinal tract was monitored by PCR‐based classical and quantitative techniques and revealed the presence of both strains in fecal samples collected 3 days after hatching. Subsequently, the number of L. salivarius IBB3154 increased significantly in the chicken intestine, whereas the presence of L. plantarum IBB3036 was gradually decreased.


| INTRODUC TI ON
Lactic acid bacteria (LAB) are important autochthonous residents in the chicken gastrointestinal tract (GIT) (Musikasang, Tani, H-kittikun, & Maneerat, 2009). GITs of birds are colonized by beneficial bacteria in nests when progeny (offspring) are exposed to hen and nest materials (de Oliveira, van der Hoeven-Hangoor, van de Linde, Montijn, & van der Vossen, 2014). Inasmuch as modern poultry farms prevent chicken contact with hens, the alimentary canals of newly hatched birds could be colonized by bacteria including pathogens present in a hatchery (de Oliveira et al., 2014).
Mead (Mead, 1997) showed that lactobacilli inhabit the small intestine and caeca of chickens 1 week after hatching. It has been noted that the colonization of GIT with nonpathogenic microorganisms creates a protective surface against pathogens (Mahroop Ra, Raja, & Mohamed Im, 2009). In ovo administration of probiotic bacteria may lead to GIT colonization before the first contact with environmental microorganisms (Villaluenga, Wardeńska, Pilarski, Bednarczyk, & Gulewicz, 2004). The initial microbe inoculation may protect the gut of the chick from pathogenic bacteria by competitive exclusion mechanisms, which are based on bacterial interactions mediated by competition for mucosal adhesion sites and nutrients (Majidzadeh Heravi et al., 2011). Moreover, the initial delivery of desirable bacteria is crucial for processes referring to the development and maturation of the immune system (Di Bartolomeo, Startek, & Van den Ende, 2012). Lactobacilli are natural inhabitants of GIT and are able to ferment various sugars (hexoses, pentoses, and disaccharides) to lactic acid. The lactate-dependent strong acidification, in addition to bacteriocin production, can influence anti-pathogen properties of the protective surface to be more efficient and ultimately may prevent the proliferation of undesirable bacteria and fungi. Thus, these phenomena are involved in maintaining the microbiological balance in the intestines (Mahroop Ra et al., 2009). The impact of lactobacilli on the natural balance of the microflora occurs when strains exhibit competitive exclusion abilities, attachment to epithelial cells of the intestine, fermentation of a broad spectrum of sugars including complex sugars, enhancement of the immune system and resistance to inner digestive track conditions such as a low pH and the presence of bile salts (Ding & Shah, 2007). Among the evaluation of probiotic safety, it should be ascertained that any given probiotic strain is not at significant risk with regard to transferable antibiotic resistance genes (Ramirez-Chavarin, Wacher, Eslava-Campos, & Perez-Chabela, 2013). Nurami and Rantal (Nurmi & Rantala, 1973) reported a phenomenon of prevention against Salmonellae infections by oral administration of gut microorganisms in chickens. Moreover, the results of some other studies have evidenced that probiotics may represent an alternative for antibiotics (Majidzadeh Heravi et al., 2011). Additionally, desirable bacteria improve digestive processes and, in parallel, might have an impact on the nutrient intake and growth of farm chickens . As a result, desirable bacteria can have a positive impact on the feed conversion ratio (FCR), resulting in better meat (weight) gain.
To increase the chance of survival and the persistence of bacteria in GIT, some complex sugars are used as prebiotics (Bednarczyk et al., 2016). Prebiotics are defined as non/low-digestible food ingredients, which pass undigested into the lower gut where they become available for some colonic bacteria but are not utilized by the majority of the bacteria present in the colon. Lactulose, galactooligosaccharides, fructooligosaccharides, inulin and its hydrolysates, maltooligosaccharides, and starch are the most-known prebiotics.
The ability to ferment prebiotics is mainly associated with the bifidobacteria but also some lactobacilli (Grajek, Olejnik, & Sip, 2005). Since the role of LAB in the feed conversion efficacy and health of the birds was recognized, Lactobacillus strains have been extensively studied and used as synbiotics, which refer to a mixture of probiotic and prebiotic compounds (Ehrmann, Kurzak, Bauer, & Vogel, 2002;Patterson & Burkholder, 2003). Their synergistic effects can potentiate the implantation and persistence of desirable bacteria in animal GIT.
Many bioprotective formulations are based on probiotic strains isolated from the natural intestinal microflora of animals. The aim of this study was to characterize selected Lactobacillus strains originating from different environments (milk and hen feces) regarding their applicative potential as a beneficial bacteria administered in ovo. Various criteria for the selection of a beneficial candidate as well as the assessment of bacterial growth in response to the presence of prebiotic compounds were investigated, resulting in the preparation of two bioactive formulations, which were injected in ovo into Cobb500 FF fertilized eggs. The persistence of the two selected Lactobacillus strains in the gut was monitored, using a PCR-based quantitative method showing the presence of both strains in 3-day fecal samples and the considerable species-dependent variability over that time.

| Growth conditions, isolation, and identification of Lactobacillus strains
The bacterial strains used in this study are listed in Table 1.
Lactobacillus salivarius IBB3154 was isolated from chicken stool and identified as described previously (Kobierecka et al., 2015).

| Enzymatic activity and sugar fermentation profiles
The enzymatic activity and sugar fermentation patterns were assessed at least in duplicate using the API ® 50 CHL and API ® ZYM kits, respectively, following the manufacturer's (BioMerieux, France) protocol. API ® ZYM tests 19 different enzymes, whereas API ® 50 CH detects the fermentation of 49 different carbon sources. The resulting sugar fermentation pattern was inspected following anaerobic incubation at 37°C after 48 hr.

| Growth and spot assays under acid, bile salts, sodium chloride, and temperature stresses
The acid tolerance of L. plantarum IBB3036 and L. salivarius IBB3154 was investigated in NaCl 0.9% adjusted with 5 M HCl to pH 2.5, 3.5 and 7 (used as a positive control). Briefly, overnight (o/n) bacterial cultures grown in MRS were centrifuged (6000 g, 7 min), and the cell pellet was resuspended in saline to avoid the introduction of residual sugars into the test solution.
Subsequently, each strain was diluted 100-fold in test tubes containing saline solution adjusted to different pH values and incubated (37°C; 1.5-3 hr).
To evaluate bile salt tolerance, L. plantarum IBB3036 and L. salivarius IBB3154 were grown overnight in MRS broth. Subsequently, each strain was diluted 100-fold in test tubes containing MRS supplemented with different concentrations (0, 0.3, 1%) of ox gall (Sigma-Aldrich, USA) and incubated (37°C; 1.5-3 hr). In both tests, samples obtained at two time points (1.5 and 3 hr) were collected for enumeration. Viable cell counts were determined by plating on solid MRS and calculated by comparing the common logarithm of the final count after the time points, with different medium supplementation (0%, 0.3%, 1% of ox gall or pH 2.5, 3.5 and 7) and with the common logarithm of the final plate count in pure MRS broth.
Lactobacillus survival was expressed as "%" and calculated as follows: Na/Nb × 100, where Na = log colony-forming units (CFU)/ml after TA B L E 1 Bacterial strains and oligonucleotides used in this study incubation and Nb = log CFU/ml before incubation. Each experiment was carried out in three independent repetitions.
To assess the effects of osmolarity, 3 μl of an o/n culture of each Lactobacillus strain was spotted on plates with MRS-agar broth (positive control) and MRS-agar containing 2, 4, 6 or 8% (w/v) NaCl

| Adhesion assay to bare polystyrene (PS) under static conditions
Bacteria were evaluated for their adhesion ability to bare polystyrene in 96-well microtiter plates (Thermo Fischer Scientific Nunc A/S, Denmark) using the technique described by (Christensen, Baldassarri, & Simpson, 1995) and modified by Radziwill-Bienkowska   Bacteria were spread on LMS-agar surface with a swab soaked with the final cell suspension. Subsequently, the strips were laid on the LMS-agar-bacteria surface and incubated at 37°C for 48 hr. The MIC was determined from the inhibition curves that intersected the scale on a strip.
The experiment was conducted in a commercial hatchery (Piast Grupa, Lewkowiec, Poland). At three time points (3 days and 3 and 6 week after hatching), fecal samples were collected and used for DNA isolation and qPCR.
Injection in ovo, hatchability and analyses of embryo mortality were conducted at UTP (Dunislawska et al., 2017). Reactions were performed with an initial denaturation step all results to one, which was common for all genes and DNA concentrations. The amount of each target gene was calculated by the ΔC t method with a geometric mean of two common genes as a reference (Vandesompele et al., 2002). In this approach, two primer pairs

| Isolation and identification of Lactobacillus species
L. salivarius IBB3154 was isolated from chicken stool and identified as described previously (Kobierecka et al., 2015). and β-gentiobiose) was species or strain-dependent ( Figure 1). None of the strain utilized 2-ketogluconate potassium, 5-ketogluconate A comparison of the two tested species revealed that all L. plantarum used a wide range of simple and more complex carbohydrates F I G U R E 1 Carbohydrate assimilation capacities and enzyme activities among Lactobacillus plantarum and L. salivarius strains. Enzyme activity and sugar fermentation ability and efficacy is indicated by different color and size tetragons. The presented carbohydrate assimilation pattern is based only on carbon sources, for which the utilization was variable among species

| Adhesion properties
The adhesion microtiter PS plate assay is a simple method for the preliminary assessment of strain adhesion ability, which usually positively correlates with its adhesiveness to biotic surfaces, as Remarkably, this strain was even 20% more adherent than control highly adhesive L. rhamnosus GG.

| Osmotic stress and temperature resistance
In this study, the isolated Lactobacillus strains showed a similar capacity to survive under an elevated concentration of NaCl, but none of the isolates could grow in 8% NaCl. The strongest resistance (up to 6%) was characteristic of L. plantarum IBB3041, IBB3039 and IBB3036, whereas L. salivarius IBB3154 and L. plantarum IBB3075 revealed a slightly weaker resistance, demonstrating tolerance up to 4% NaCl ( Figure 3).  (Figure 3).

| Low pH and bile salt tolerance
The study of acid tolerance showed that cells of both L. plantarum IBB3036 and L. salivarius IBB3154 strains were vital after an incubation period of 3 hr, at pH 3.5 and pH 2.5. In the case of L. plantarum IBB3036, no loss of viability was detected over 3 hr of incubation at pH 3.5 (10 9 CFU/ml). However, the same strain showed a loss of viability following exposure to pH 2.5 (>10 3 CFU/ml). The number of L. salivarius cells decreased slightly after a 3-hr incubation at pH 3.5 (from 3 × 10 8 to 7 × 10 7 CFU/ml). Nevertheless, L. salivarius F I G U R E 2 Adhesion of Lactobacillus strains to bare PS microtiter plates. Standard deviations (±) from at least three independent experiments are shown as error bars 0 0.5 1 1.5 2 2 .5 3 3 .5 4 4 .5 5

| Persistence in chicken GIT after in ovo administration
All newly designed lactobacilli primer pairs were analyzed, using NCBI/Primer-BLAST (https://www.ncbi.nlm.nih.gov/tools/primerblast/). The primer pair specificity was validated using DNA isolated from fecal samples from the chick control group and from chicks from small family-owned farms. No positive signals were obtained by PCR for either DNA isolated from fecal samples from small familyowned farms or fecal samples from the control group.
When typing using the classical PCR approach, L. plantarum IBB3036 was detected in the feces samples of 3-and 21-day chickens with a frequency of 60% and 10%, respectively, and it was not detected in the feces of 42-day chickens. In contrast, PCR analysis of the microbiome of chickens that had been hatched from eggs injected in ovo with S1 revealed L. salivarius

| D ISCUSS I ON
Although a reasonable number of well-characterized probiotic strains are commercially available worldwide, screening for novel strains is still of great interest in general and also from a poultry production perspective. In this process, lactobacilli may be potentially applied as an alternative to antibiotic growth promoters (already banned in the EU) to improve animal health, growth performance and food safety. Strains possessing unique and particular features that may enable health benefits could be derived from intestine as well as the dairy environment. Some authors have promoted the importance of an intestinal origin as a selective criterion for the search for desirable strains, but the expert FAO-WHO (2006) panel suggested that probiotic activity is more relevant than the source of the bacteria. In fact, the evaluation and selection of dairy LAB isolates for potential use as probiotics has been previously described (Leite et al., 2015;Zheng et al., 2013). Thus, in addition to LAB strains of intestinal origin, those derived from milk could also be used as a source for obtaining novel probiotic strains. Considering the abovementioned information, the aim of this study was to evaluate the beneficial properties of a few L. planatrum milk isolates and subsequently test the selected strain persistence in the chick digestive tract after in ovo injection in synbiotic form. As a reference for the chick intestine, a Lactobacillus isolate was used.
First, we compared the sugar metabolism profile of LAB isolates with a special focus on sugars, which are omnipresent in chicken GIT, and two selected prebiotics (RFOs and Bi 2 tos). Oligosaccharides and higher oligosaccharides are the major carbohydrates in foodrelated and intestinal habitats (El-Fallal, Abou, El-Sayed, & Omar, 2012; Roberfroid & Slavin, 2000) and their utilization appears to be an ubiquitous feature of lactobacilli (Gänzle & Follador, 2012); however, in this study, it was only characteristic of L. plantarum isolates. The observed wide spectrum of sugars utilized by L. plantarum strains may be associated with a high versatility of this species as it is commonly found in many different ecological niches (Siezen & van Hylckama Vlieg, 2011) and is reflected in the more than 3-Mb L. plantarum genome, which is among the largest ones of peptide substrates (Mudryk & Podgorska, 2006).
NaCl is an inhibitory substance, which by acting on the loss of cells membrane and tension turgor pressure, may negatively influence the growth of some bacteria (Liu, Asmundson, Gopal, Holland, & Crow, 1998). The ability to resist its presence is one of the essential natural attributes of desirable strains because to survive and grow in the gastrointestinal tract, bacteria must adapt to an environment with an osmolarity equivalent to 0.3 M NaCl (Chowdhury, Sahu, & Das, 1996). These study results suggest that all analyzed strains could withstand the NaCl osmolarity in the animal GIT and could resist technological processing because they were all tolerant to high, up to 4%-6%, NaCl concentrations, which can be present in pelleted or dried animal food.
Adhesion to mucosa is essential for bacterial persistence in the host. The ability to fix itself to the mucosal layer allows bacteria to avoid being swept away by peristaltic movement (Fernandez, Boris, & Barbes, 2003). It is also postulated that highly adhesive probiotic bacteria have the greatest beneficial effects on host health and should, at least transiently, colonize the host gut (FAO-WHO 2006).
Among the lactobacilli tested during this study, the greatest adherence was noted in strains of intestinal origin, whereas this ability was considerably reduced in milk isolates. According to the data available in the literature, efficient adhesiveness of intestinal isolates does not seem to be the rule because many strains of this origin do not display any ability to adhere (Kobierecka et al., 2017), however, many strains isolated from environments other than the intestine can be effective binders (Douillard et al., 2013).
Tolerance to bile salts and a low pH is essential for bacteria to survive in GIT because they are most commonly orally delivered in a feed system. In comparison to mammals, the alimentary canal of chickens is shorter. The time required for feed to pass through the entire gastrointestinal tract can be as short as 2.5 hr (Duke, 1977;Jin, Ho, Abdullah, & Jalaludin, 1998 The common usage of antibiotics in the poultry industry was banned (Mahroop Ra et al., 2009); however, knowledge concerning antimicrobial susceptibility is crucial to avoid the introduction of exogenous sources of resistance determinants. According to our results, L. plantarum IBB3036 and L. salivarius IBB3154 are not considered to harbor antibiotic resistance determinants, although they have higher levels of resistance to vancomycin and streptomycin.
It is well known that Lactobacillus strains have an intrinsic resistance to vancomycin due to the absence of the appropriate cell wall precursor target. Remarkably, a higher level of resistance to aminoglycoside antibiotics is also considered to represent an intrinsic resistance in lactobacilli and is attributed to the lack of cytochromemediated electron transport (Hummel, Hertel, Holzapfel, & Franz, 2007).
Although most Lactobacillus strains only transiently colonize chicken gastrointestinal chicken tracts (Spivey, Dunn-Horrocks, & Duong, 2014;Stephenson, Moore, & Allison, 2010), the results of this study indicate that L. salivarius IBB3154 was able to persist and even to significantly increase in number in the chicken intestine after a single in ovo inoculation and within the entire production period. In contrast, L. plantarum IBB3036 was detectable at low levels on the third day of the chicks' life and was virtually undetectable during the last stage of the experiment (day 42). Such a severe discrepancy between both strains was unexpected because they were characterized by similar probiotic features. Moreover, in some aspects, such as the range of sugar metabolism and tolerance to bile salts, NaCl and elevated temperatures, L. plantarum IBB3036 showed better capabilities than L. salivarius IBB3154 to withstand the harsh environmental conditions of the chicken GIT. Regarding two other probiotic features, tolerance to acidification and adherence ability, L. salivarius IBB3154 showed better performance. However, the higher acid resistance of L. salivarius IBB3154 cannot be a reason for the pronounced persistence of this strain because the bacteria were introduced by in ovo injection and thus were not exposed to the acidic conditions of the chick GIT. Thus, according to our findings, the main reason for the lack of L. plantarum IBB3036 persistence may be its poor adherence ability, which led to progressive strain removal by peristalsis and/or its exclusion by other resident gut microorganisms (such as, presumably, L. salivarius IBB3154). In contrast, the strong adherence abilities of L. salivarius IBB3154 allowed its survival and extensive proliferation in the chick GIT.
Confirmation of whether this phenomenon is indeed important for L. salivarius IBB3154 adherence abilities and/or other mechanisms requires further studies. Nevertheless, L. salivarius IBB3154 seems to have potential for modulation of the GIT microbiota due to the adherence ability and in ovo delivery into the chicken embryo, which extends the effective time of action to the prehatching period.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.