1Department of Pathology, Bacteriology and Poultry Diseases,
Faculty of Veterinary Medicine, Ghent University, Merelbeke, Belgium
Summary
Several bacterial pathogens detected in poultry products can cause infections in humans. The control of transmission of these zoonotic agents through the food chain can be based on a post harvest or a pre-harvest strategy. Pre-harvest control measures for Salmonella are a legal obligation in the EU. Nutritional strategies constitute an important part of such pre-harvest control strategy. In this paper an overview is given of the tools that are available for control of poultry infections by zoonotic agents through the feed. These tools include measures for the reduction of bacterial load in finished feed and feed ingredients. They also include feed additives and feed ingredients that act essentially on the composition of the intestinal content, creating an intestinal environment that is deleterious to the zoonotic bacteria. Attention is paid to the mechanisms of action and to the advantages and disadvantages of the different tools.
Introduction
The importance of poultry products as a source of Salmonella infections in the food chain has attracted considerable interest over many years (Fernandez el al., 2000). Poultry products however can also be a source of other foodborne pathogens for humans, including Campylobacter spp. and possibly also Helicobacter pullorum, which is one of the more recently identified exotic agents. Also Listeria spp. and Clostridium perfringens can enter the human food chain through contaminated poultry carcasses. Yersinia, Hafnia and Bacillus spp. may become of greater importance than presently accepted (Waldroup, 1996). Finally Escherichia coli O157:H7 (VTEC) is yet another zoonotic agent which can cause severe human illness, although little information is available on the role of poultry in human outbreaks. Intestinal colonization and shedding of the pathogenic bacteria in the birds faeces is an important factor leading to the contamination and cross-contamination of the poultry products at slaughter and during processing of the meat. It is generally accepted now that decontamination of the carcasses can considerably reduce the contamination of the poultry meat products. The use of most carcass disinfectants, including chlorinated rinsing water, however is prohibited in the European community. Moreover, decontamination of fresh table eggs is very difficult. Therefore the control of transmission of zoonotic agents through the food chain in Europe presently is based not only on post-harvest measures, but also on prevention and monitoring/eradication during the live phase. These principles are laid down in directive EU-92/117 and subsequent amendments. Fortunately, the need to control the colonization of the live birds by the zoonotic agents in order to reduce the contamination of the poultry products rather than decontaminating the end products is gaining support also in the rest of the world, including North-America (Blankenship et al., 1993).
Nutritional strategies are part of an integrated control strategy
The European Union has requested every member state to draw a control plan for Salmonella in poultry, which has to be compatible with the general instructions of the directive EU-92/117. These instructions basically require a top-down approach, which means that measures are very strict in grand parent and parent flocks, and one expects these measures to be of benefit to the general sanitary status of the whole production chain. This top-down concept implies that zoonotic agents are transmitted vertically in the poultry population. To some extent, this is indeed the case for certain Salmonella serotypes. It is very well documented for Salmonella Enteritidis and many strains of Salmonella Typhimurium. We have also documented this to be the case for Salmonella Hadar (Desmidt et al., 1998). A limited number of other serotypes also have been classified as being vertically transmitted in poultry, including a.o. Salmonella Thompson, Salmonella Bertha, Salmonella Infantis. Other serotypes of Salmonella are not considered to be invasive and therefore less likely to be transmitted vertically, except perhaps through eggshell contamination. Whether Campylobacter spp. are vertically transmitted still is a matter of debate, although epidemiological data suggest they are not (Jacobs-Reitsma, 1994). Thus a top-down approach to the control of zoonotic agents, which does not include measures in the commercial layer and broiler flocks, is doomed with limited success. This has indeed been the experience recently in several EU member states. Therefore the newly amended control programs all include measures at all levels of the production chain. They also include a wide range of different types of measures, as it is commonly accepted now that each type of measure may have a more or less important effect on reducing the Salmonella contamination rate, but none has been shown to be sufficiently successful on its own. Moreover the original ambition of eradicating zoonotic agents from the animal production chain and thus producing animal products free from zoonotic agents has been considerably tempered. Plans now simply are designed to reduce the presence of specified zoonotic agents at the different levels of the production chain.
Complementarity of the nutritional strategy to other control strategies
Fortunately at the present time a wide panel of tools is available to reduce foodborne pathogens in poultry products. They have been overviewed for the meat production chain by Hafez (1999). Briefly these tools can be divided in pre-harvest, harvest and post-harvest measures. Pre-harvest measures include hygienic measures, feed and drinking water additives, and vaccinations. The measures that can be taken during catching and transport are essentially hygienic, while the post-harvest measures include both hygienic measures and the application of decontamination products on the meat. All of these measures are essentially meant to prevent the introduction of the zoonotic agents or to protect the bird from being persistently colonized by the zoonotic agent. They often are based on the assumption that most zoonotic agents enter the bird by oral uptake, which usually seems to be correct, even in the case of vertically transmitted Salmonella infections (oral uptake from the egg white or shell membranes during pipping).
It is clear that hygiene plays an essential role in the control of zoonotic agents at all levels of the production chain, be it in egg production as in meat production. Implementation of hygienic measures, especially pre-harvest, has mostly focused on the control of Salmonella. Epidemiological field studies in Campylobacter however have shown that most hygienic measures designed for the prevention and control of Salmonella also have a beneficial effect on the risk of Campylobacter contamination during the live phase (Brendtson, 1996). This may partly be explained by some similarities between Salmonella and Campylobacter in the way they persist. Indeed like Salmonella, Campylobacter can be carried and excreted for a long time by rodents (Brendtson et al., 1994),
Pre-harvest hygiene means hygienic handling of hatching eggs, including hatching egg sanitation, hygienic measures in the poultry houses, and feed hygiene.
Hygiene of feed ingredients
Salmonella-contaminated feed has been recognized as the most common source of infection in poultry flocks. Salmonella may be present in most processed feed ingredients on the international market (Sreenivas, 1998). Veldman et al. (1995) found 31% of 130 fish meal samples to be infected, as opposed to 4% of 83 meat and bone meal samples, 2% of 58 tapioca samples and 27% of 15 maize grit samples. Thus Salmonella contamination rate of feed ingredients varies widely depending on the type of ingredient. It varies also depending on the region of origin, and the conditions of harvesting and storage. A major handicap with respect to monitoring of Salmonella (and other zoonotic agents) in feeds and feed components is the uneven distribution of the agents in the material. This makes the sampling for bacteriological monitoring statistically unacceptable. Indeed, a representative sample of a population as a rule should include 10% of the population, in most sampling procedures however, only 1 kilo is taken and homogenized and only 10g is analysed. Thus heavily contamined feed ingredients probably will not be missed, but a low grade contamination can easily be missed. Fortunately the Salmonella serotypes found in feed ingredients and finished poultry feeds are mostly of the so-called "exotic" types. Out of 28 different serotypes isolated by Veldman et al. (1995) none were of the Enteritidis serotype, despite the occurrence of an epidemic caused by this serotype since 1987 in poultry and in man. Soil contamination of plant material is thought to be the major source of these exotic serotypes. The growth of these bacteria is inhibited by the growth of other micro-organisms such as yeast or lactic acid producing bacteria. This may be due to competition for nutrients or the inhibiting effect of products of their metabolism such as organic acids (see below regarding the effects of organic acids). Therefore Salmonella is never found in well-preserved silage. The so-called "exotic" or "feed-borne" serotypes however can multiply rapidly in moist feedstuffs where few competitive bacteria are present, either in the feed ingredients during wet post-harvest storage and transport in warm climate, or by condensation after processing in the feed mill under conditions of inadequate cooling. Nevertheless feed producers can not rely too heavily on cooling as a control procedure, since the exotic Salmonella serotypes can grow at temperatures as low as 4°C and maximum growth reportedly occurs at 46°C. Experiments with Salmonella typhimurium, another serotype regularly found in feedstuffs, have shown that, under dehydrating conditions, the bacterial cells become more heat resistant. Under these conditions a 40% survival rate was recorded after a treatment of 60 minutes at 100°C. The consequence of this is that, if Salmonella is present in raw material processing areas in the feed mill, it will persist after drying on surfaces of equipment. Dust particles also carry Salmonella and can be important routes for the spread of contamination. This is known to be a major route of recontamination of the finished feed.
Hygiene and decontamination of finished feeds
The extent of contamination often differs between the feed ingredients and the finished product. As an example, Veldman et al. (1995) reported on the incidence of Salmonella in poultry feeds and feed components between July 1990 and April 1991 in The Netherlands. They found 10% of 360, 10g samples of finished poultry feeds to be contaminated with Salmonella. Mash feeds were contaminated at a rate of 21%, versus 1.4 percent of pelleted feeds. This points to the value of pelleting as a hygienic control measure for feed. The success of conventional steam pelleting in reducing Salmonella contamination of finished feed however varies, depending on the moisture content and the temperature of the pellets. Further reduction of the Salmonella contamination is possible through steam conditioning prior to pelleting, by expanding and by high temperature - short time conditioning, which may result in a reduction to about 0.001% of the level in the untreated material (Sreenivas, 1998). Expanding in combination with pelleting reportedly is an effective means of reducing the Salmonella in a compound feed. The same author points to the well-known problem of recontamination of the finished feed during drying and storage.
One should realize however that Salmonella-free feed delivered at the poultry farm can be (re)contaminated during storage on the farm or in the feeders. Experiments carried out by us and other groups have shown that newly hatched chicks can be infected with less than 1 c.f.u. Indeed, these highly susceptible birds act like very efficient incubators. They start excreting within hours after inoculation (Desmidt et al., 1997), which results is rapid horizontal spread. (Cross) contamination in the feeders may come from infected rodents, from fecal contamination or from infections present in the crops of the birds. It is remarkable to note also that crop infection has been reported to increase during feed withdrawal not only for Salmonella, but also for Campylobacter (Byrd, 1999).
Finished poultry feeds as well as feed ingredients also can be decontaminated for the purpose of control of zoonotic agents by irradiation (Campbell et al., 1986). The ionizing radiation which may be employed in irradiating feed components and finished poultry feed is limited to gamma rays from the radionuclides 60Co or 137Cs, X-rays generated by machine sources operated at or below an energy level of 5 MeV, and electrons generated by machine sources operated at or below an energy level of 10 MeV. Electrons however are limited in their use due to their low penetration. Low-cost low-energy machines (1-2 MeV) can be applied to free-flowing feeds or feed components in a thin layer (Lapidot and Padova, 1978). High-energy machines (7-10 MeV) can be employed for shallow bags (25 kg). The amount of ionizing energy absorbed by the target material is a measure for the intensity of irradiation. The unit of absorbed dose is the Gray (Gy). One Gy is equal to the absorption of one joule per kg of feed. The dose employed is dependent on the level of initial contamination, the kind of contaminating organisms, and the purpose of the treatment. Control of the irradiation procedure is based on dosimetry (McLaughlin et al., 1989). Elimination of Salmonellae and other enteric bacterial pathogens of zoonotic importance, such as E. coli O157:H7, Yersinia enterocolitica, Listeria monocytogenes and Campylobacter spp. can be achieved at 4-6 kGy (Begum et al., 1989). This dose apparently does not cause significant chemical changes to the feed (Adler et ah, 1978). Lipids are attacked by free radicals which are formed upon irradiation, and peroxides and other oxidation products are formed under aerobic conditions in a way similar to the auto-oxidation process, however at the above mentioned dose of 4-6 kGy only small quantities of oxidation products of fat are formed in the feeds (Lapidot, 1979). Some vitamins however are susceptible to irradiation, depending on the composition of the feed (Hanis et al., 1985). Moreover irradiation of finished poultry feed in paper bags of 25 kg increases the cost by approximately 100%.
Antibiotics
Next to their curative effects, antibiotics are also given in feed to poultry for their growth promoting effects. Most antibiotic growth promoters act by modifying the intestinal flora, especially targeting gram-negative bacteria that are associated with poorer health and performance of the animal (Bedford, 2000). Indeed, many so-called non pathogenic bacterial species, especially gram-negative bacteria, can reduce feed conversion efficiency and growth in chickens due to competition with the host for the nutrients in the intestinal tract, degradation of host enzymes and reduction of the absorptive surface area. In some instances these bacteria can elicit an immune response which, as a side effect, causes reduction of appetite and catabolism of muscle protein. One of the mechanisms underlying these phenomena is the massive production of tumor necrosis factor alpha (TNF-α). Some of these intestinal bacteria are considered as opportunistic or facultative pathogenic bacteria. These include a.o. Clostridia, which have been associated with necrotic enteritis. For these reasons, antibiotics have been used for many years as growth promoting agents. Recently however, concerns arose about the use of antibiotics in livestock. Antibiotic feed additives were linked to the emergence of multiple drug resistant bacteria (Wray and Davies, 2000). The use of avoparcin in poultry for example has lead to vancomycin-resistant Enterococcus spp. in poultry and swine (Kruse et al., 1999; Eager et al, 1997). The presence of undesired antibiotic residues in meat and environmental contamination have largely added to the public concerns regarding the use of antibiotics in the feed. As a consequence at the end of June 1999 the majority of antibiotics as growth promoters in monogastric diets have been banned within the EU, with the exception of anticoccidial products (Bedford, 2000). Therefore no detailed description of antibiotics at growth promoting concentrations in poultry feed for the control of zoonotic agents will be given in this review. The use of therapeutic dosages of antibiotics for the prevention or treatment of Salmonella infections has been tested by many authors. Substantial scientific evidence indicates that antibiotic utilization may actually enhance enteric colonization and invasion by some antibiotic-resistant strains of Salmonellae (Smith and Tucker, 1975). More recent studies have shown that this is not the case for some antibiotics (Manning et ah, 1994). Indeed, treatment with chlortetracycline or monensin had no effect on subsequent invasion or colonization by Salmonella Enteritidis. In contrast to the previous studies, Bolder et al. (1999) recently showed that continuous administration of flavomycin (2g/ton) and salinomycin (60g/ton) in the feed actually reduced Salmonella and Clostridium shedding, whereas it did not affect Campylobacter shedding.
Prebiotics
Prebiotics have been defined by Gibson and Roberfroid (1995) as "non-digestible food ingredients that beneficially affect the host by selectively stimulating the growth and/or activity of one or a limited number of bacterial species already resident in the colon, and thus attempt to improve host health". Based on this definition, Russel (1998) formulated the criteria according to which a substance can be a prebiotic. First, prebiotics are always feed ingredients that are not digested by the host, not or little used and/or metabolized as they pass through the upper portion of the intestinal tract, so they can reach the flora of the large intestine. Secondly they have to be able to serve as a substrate for one or more bacterial species with a potentially beneficial effect on the host. Finally they have to be able to cause a shift in the microflora that improves the health of the host.
In principle only indigestible, fermentable feed components are prebiotics. Most of these prebiotics are carbohydrates. Besides carbohydrates however, there are other substances thought to have a positive effect on the intestinal flora and/or the intestinal environment, a.o. organic acids. Whether these belong to the prebiotics or not is not clear, therefore they will be treated separately (see below). The term prebiotics is mostly used however for indigestible carbohydrates. These carbohydrates are divided in groups based on their length: mono-, di-, oligo- and polysaccharides. Another possible classification is based on their source of origin: natural or synthetic saccharides (Iji and Tivey, 1998).
Classes of prebiotics:
The most important monosaccharide prebiotics consist of hexoses (glucose, fructose, galactose, mannose) and pentoses (ribose, xylose, arabinose). Some of these, like glucose and fructose, are naturally available. Galactose is available mostly under the disaccharide form of lactose. Mannose however is by far the most commonly used monosaccharide feed additive (Alien et ah, 1997). These monosaccharides can form the basis for enzymatically constructed oligo- or polysaccharides.
The most important naturally occurring disaccharides are sucrose, lactose and maltose. Isomerization products of these compounds can be used as prebiotics, e.g. lactulose (based on lactose). Lactose (Corrier et al., 1993), lactulose (Iji and Tivey, 1998) and lactosucrose (Terada et ah, 1994) reportedly have prebiotic effects in chickens. Lactose is not hydrolyzed or absorbed intact by the intestinal tract of chicks, and as much as 50% of ingested lactose in poultry diets may be excreted unchanged. Because of its lack of digestion and absorption, lactose passes into the lower segments of the intestine and caecum. The hydrolization of lactose that does occur is primarily the result of the microflora utilization of the disaccharide (Tellez et at., 1993).
Oligosaccharides are usually defined as glycosides that contain a limited number of hexose or pentose units. Sometimes they are used in their natural form but mostly they are obtained through enzymatic synthesis or hydrolysis. Synthetic oligosaccharides can be based on different hexose monosaccharides, for example glucose, fructose, galactose and mannose (Iji and Tivey, 1998). The synthetic route varies, depending on the direct polymerization of disaccharides or the fractionation of microbial cells to obtain the material from the cell wall. Polysaccharides, when fermented, also yield oligosaccharides (Iji and Tivey, 1998). Fructo-oligosaccharides (FOS) are the oligosaccharides most extensively studied in chickens with respect to their prebiotic effect and their activity against Salmonella (Fukata et al., 1999) and Campylobacter (Schoeni and Wong, 1994). FOS are reported to have little effect on Salmonella Typhimurium colonization when added at 0.375% to the diet of broiler chicks but, when fed at 0.75% in combination with competitive exclusion flora, fewer birds were colonized when compared to control groups (Bailey et al., 1991). Beta (2-1) fructanes, fructose polymers, such as inulin and their hydrolisation products, can also be used as probiotics (Roberfroid, 1999). Compounds with a polymerization length between 2 and 60 are called inulin (raftiline), while compounds with a higher polymerization length are called raftilose (Roberfroid and Delzenne, 1998). Neo-fructo-oligosaccharides (Neo-kestose, neo-nystose and neo-furanosylnystose) related to inulin, are synthetized enzymatically from sucrose using fructosyltransferase. Galacto-oligosaccharides are bifidogenic (increase the presence of bifidobacteria in the intestinal tract) in humans and can be produced industrially by the transgalactosidase activity of beta-galactosidases, however these products hitherto have not been tested in poultry (Durst, 1996).
Mannan-oligosaccharides (MOS) or mannose-based carbohydrates occur naturally in many products such as yeast cell walls or different gums. A commercial product is available for poultry which contains yeast cell wall fragments derived from Saccharomyces cerevisiae after centrifugation of a lysed yeast culture (Spring et al., 2000). Ceacal contents from hens fed these MOS protected chicks better from colonization with Salmonella Enteritidis (Fernandez et al. 2000a).
The most commonly used polysaccharides are pectins and guar gum (Ishihara et al., 2000).
Mechanisms of action of prebiotics:
Prebiotics can have a direct effect either by direct binding of pathogens or by increasing the osmotic value in the intestinal lumen. More often however their effects are indirect, mediated by products that are generated by the intestinal flora that uses the prebiotics for their own metabolism. Such metabolites include short-chain fatty acids, lactate, polyamines and bactericins.
Type 1 (F1) fimbriae are described in several bacterial enteric pathogens (Bengmark, 1998), including Salmonella (Oyofo et al., 1989). These type 1 fimbriae bind to several carbohydrate residues of glycoproteins, present on the surface of eukaryotic cells. They are needed for the adhesion of the bacteria to the mucosal surface, which is a prerequisite for the colonization of the host. Indigestible carbohydrates with mannose residues may bind the Fl fimbriae and therefore block the adhesion of the bacteria to the epithelial cells (Finucane et al., 1999). According to these authors yeast MOS induce mannose sensitive agglutination in 51% of E. coli strains tested and 53% of Salmonella strains tested. Among the Salmonella strains, 80% of the Enteritidis serotype and 67% of the Typhimurium serotype strains possessed these agglutinins. Prebiotics may constitute a substrate for the growth of intestinal flora. This multiplication of normal flora may inhibit the colonization with pathogenic bacteria. This phenomenon of inhibition is called "competitive exclusion" (see below). FOS have been shown to promote growth of Enterococcus faecium, Lactobacillus lactis and Pediococcus spp. in vitro. Moreover, Lactobacillus plantarum has been shown to possess mannose-sensitive receptors, an uncommon phenomenon in gram-positive bacteria. With these receptors it can compete for the same adhesion sites in the intestine as the gram-negative pathogens (Bengmark,
1998). Salmonella challenge in MOS supplemented broilers resulted in decreased numbers of Salmonella, but no differences in numbers of Lactobacilli, enterococci or anaerobes (Spring et al., 2000). Another possible mechanism of action of prebiotics is through a modification of the metabolic activity of normal intestinal flora. Saccharides, indigestible by the host, are fermented by the flora into volatile fatty acids (VFA), including acetate, propionate and butyrate, lactate and several gases including carbon dioxide, methane and hydrogen (Cummings, 1981). Administration of MOS in the feed however induced no change in lactate, VFA or ceacal pH in broilers (Spring et al., 2000).
There are some indications that prebiotics may stimulate the immune response. Supplementing FOS to the diet of mice increased the CD4 and CDS T-cell response (Pierre et al., 1999). Afzali and Devegowda (1999) observed an increase in antibody liter against Aflatoxin B1 in broiler breeders with aflatoxicosis after administering modified MOS. In humans prebiotics have been suggested to have several additional effects, including prevention of colon cancer, improving availability of minerals, improving intestinal vitamin synthesis and decreasing blood cholesterol. None of these effects have been demonstrated so far in poultry. Also the effect of prebiotics on performance in poultry is a matter of debate (Iji and Tivey, 1998). Prebiotics can be metabolized by the intestinal microflora into polyamines (putrescine, spermidine, spermine) which regulate synthesis of DNA, RNA and proteins and which can enhance colonic epithelial cell growth and differentiation (Roberfroid and Delzenne, 1998).
Probiotics
Probiotics are defined as single or mixed cultures of live microorganisms which, when applied to animals or to man, affect beneficially the host by improving the properties of the indigenous microflora (Havenaar et al., 1992). They can be undefined mixtures of bacteria, generally termed competitive exclusion (CE) products, or defined formulations. Some of these products can be added to the feed, but more often they are administered either through the drinking water or by direct spraying on the birds.
Competitive exclusion
Nurmi and Rantala (1973) originally developed the concept of CE. They stated that newly hatched chicks had little opportunity for rapid development of a normal intestinal microflora, due to the clean housing conditions in which the chicks were reared. Rapid transfer of normal flora from the hen to the chick is impossible in modern mass production systems and development of microflora is considerably delayed (Schneitz et al., 1992). By oral administration of a saline suspension of the contents of the alimentary tract from adult birds to newly hatched chicks, an adult-type microflora is established, which protects from Salmonella infection (Nurmi and Rantala, 1973).
This treatment is a prophylactic measure that aims at increasing the resistance of young chicks to Salmonella or other infections by fastening the development of normal gut microflora (Mead, 2000). It may also be used after antibiotic therapy to eliminate an existing (Salmonella) infection and, in this case, the CE flora restores the microflora (Seo et al., 2000). The first method of application was to incorporate the preparation in the drinking water. Uptake of drinking water in the first 24h after hatch however is very variable and the viability of the anaerobic strains in the water may be reduced. Moreover there is a delay between hatch and placement in rearing houses (Mead, 2000). Spray application in the hatchery has been designed to overcome these problems (Goren et al., 1984). The birds naturally preen themselves after being sprayed and thereby ingest the microflora. This is commonly enhanced by using a bright light (Schneitz el al., 1992). An attempt to use lyophilized flora encapsulated in alginate beads in pan feeders was unsuccessful (Corner et al., 1994). In ovo inoculation of CE products resulted in reduced hatchability due to microbial gas production or proteolysis (Cox and Bailey, 1993). The mechanism of action of CE flora may be explained in part by a bacteriostatic effect on Salmonella in the caeca (Impey and Mead, 1989). This may be due to the production of VFA by the bacteria (see below). Competition for receptor sites in the intestinal tract and competition between pathogens and native microflora for nutrients (Ha et al., 1994) also may play a role.
Apparently CE flora not only affords protection against various Salmonella serotypes, but also against other intestinal pathogens of possible zoonotic significance, including Campylobacter Jejuni (Hakkinen and Schneitz, 1999), Clostridium perfringens (Hofacre et al., 1998) and certain Escherichia coli strains (Stavric et al., 1992). Some commercial CE products also have been claimed to improve feed conversion efficiency and daily weight gain in broilers (Hofacre et al., 1998, Schneitz et al., 1998). Improvement in performance was shown to be associated with a decrease in the viscosity of the ceacal contents, resulting in better mixing and passaging of feed (Schneitz et al., 1998).
Defined cultures
Several research groups have isolated pure bacterial cultures from the ceacal contents of adult birds, in attempts to develop CE preparations of known composition. However, defined cultures were far less effective in protecting against Salmonella than the undefined CE products (Gleeson et al., 1989). One exception is Lactobacillus reuteri. This bacterium produces and secretes an intermediary metabolite, reuterin, which has antimicrobial activity against Salmonella, E. coli and Campylobacter (Mulder et al., 1997). In contrast to all other CE products and probiotic bacterial cultures, inoculation of Lactobacillus reuteri in ovo does not affect hatchability and it decreases the colonization by Salmonella and E- coli after hatch in both chicks and poults. Additionally, mortality due to in-hatcher exposure to E. coli or Salmonella is reduced. Use of antibiotics in ovo may preclude the co-administration of CE products, but gentamycin and Lactobacillus reuteri apparently are a compatible mixture when administered in ovo in separate compartments.
Short chain organic acids
VFA (acetic, propionic, butyric, isobutyric, valeric and isovaleric acid) are produced by the normal anaerobic intestinal flora as side products of their metabolism (Mead, 2000). The production of these VFA by the intestinal flora can be increased by adding prebiotics to the feed (Cummings, 1981). VFA or short chain fatty acids (SCFA), mainly acetic, propionic and butyric acid, can also be added directly to the feed. These acids not only exert an antibacterial effect in the intestine, but also in the crop (Hinton and Linton, 1988). In most experiments where single VFA were added to the feed, no protection against Salmonella was found (Hume et al., 1993). Commercial products however contain mixtures of propionic acid and formic acid that usually are added to the feed at the feed mill at a concentration of 1%. The acids can be coated onto a mineral granulate carrier and combined with FOS. This granulate also is mixed in the feed at the feed mill. It is claimed that the carrier matrix is colonized by lactic acid bacteria during gut transit, which helps to lower the pH of the hind gut.
SCFA are bacteriostatic and bactericidal in vitro for gram-negative bacteria, provided that there are sufficient undissociated acid molecules present and that they are in contact with the bacteria for sufficiently long time (Thompson and Hinton, 1997). Formic acid/propionic acid 1% in the feed reduces the ceacal pH (Waldroup et al., 1995), which increases the proportion of undissociated acids. The SCFA diffuse into the bacterial cell in undissociated form. Inside the bacterial cell, the acid dissociates, resulting in reduction of intracellular pH and anion accumulation (van der Wielen et al., 2000). Formic acid/propionic acid 1% decreases lactic acid concentration in the crop, suggesting lactic acid bacteria are killed. Although these lactic acid bacteria are thought to have a protective effect against Salmonella, the VFA reduce Salmonella colonization of the crop and the caeca. The decreased ceacal pH also has effects on protein synthesis, DNA synthesis, ionisation of minerals and enzyme catalyzed reactions (Tellez et al., 1993). Also the conversion of primary bile acids to secondary bile acids is inhibited by VFA at pH values below 6.5. These primary bile acids increase intestinal epithelial cell proliferation. They stimulate the migration of colonic epithelial ceils in vitro (Wilson and Gibson, 1997), which might be advantageous for epithelial repair. Unfortunately long exposure to VFA increases acid resistance of Salmonella under anaerobic conditions (Kwon and Ricke, 1998), suggesting that the beneficial effect of VFA might only be transient.
In vitro studies suggest that VFA might have an undesirable side effect possibly favouring the invasiveness of Salmonella and thus the development of the carrier state. Indeed at pH 6, all VFA induce hilA and invF expression (Durant et al., 2000). These genes are transcriptional regulators of the Salmonella pathogenicity island I, needed for invasion of host tissues. Moreover SCFA also induce spv expression in Salmonella Dublin (El-Gedailly et al., 1997). Spv genes are common in most Salmonella serovars and are involved in virulence, inducing systemic disease and macrophage survival (Libby et al., 2000).
Conclusions
This short overview shows that a large spectrum of tools is available to the nutritionist at present for the control of zoonotic (intestinal) pathogens in poultry. It is not meant to be complete, and undoubtedly there are more aspects to the nutritional strategies, indeed, for Campylobacter e.g. Fernandez et al. (2000b) clearly showed that the composition of the feed itself and supplementation with enzymes could influence the degree of colonization of the intestine. Supplementing wheat based diet for chicks with xylanase lowered the Campylobacter Jejuni counts. This is thought to be due to a reduced viscosity of the secreted mucins. Nutritional strategies implementing the above-mentioned tools become an increasingly important component of integrated prevention and control plans at the level of individual companies as well as nation-wide. Most of these tools undoubtedly can reduce the colonization and excretion of these zoonotic agents, however none can eliminate these pathogens. Therefore such tools should never be used on their own. Moreover one should realize, as mentioned above, that a number of the effects attributed to these tools only have been shown in laboratory animal models or in vitro, and remain to be proven in poultry under field conditions. Also any of these additives may slightly or considerably add to the cost of feed production, thus requiring a thorough cost - benefit analysis. Taking into consideration however the underestimated high level of contamination of the raw materials, it is clear that the feed industry has to take up its responsibility and its share in the struggle against these pathogens that are a permanent threat to consumer health.
From Proceedings of the "13th European Symposium on Poultry Nutrition", Blankenberge, Belgium.
