Effective growth promoters are anti-inflammatory by nature, as is evident from omics and simple biomarkers


Previously, the beneficial effects of antimicrobial growth promoters (AGP) were attributed to their antibiotic characteristics, but neither with conclusive evidence, nor plausible mechanism.

Growth (and health) in production animals is reciprocal to (intestinal) inflammation, so it seems logical that the observed effects are rather based on an anti-inflammatory mechanism.

This hypothesis has been confirmed by omics techniques, and by using simple non-invasive biomarkers in vivo. It is now very easy to successfully select effective alternatives to AGP.


Animal protein is an important source of essential amino acids for humans. It is provided by a large livestock industry, which is projected to grow even further with increasing wealth and increasing numbers of the human population. This will put even further pressure on the costs of animal production, in particular for the monogastric high production species such as pigs and chicken, and prices for feed ingredients are expected to rise. As a consequence, feed efficiency will become even more important than ever before, also because of environmental reasons such as the reduction of phosphate (P) and nitrogen (N) pollution. In the past, antimicrobial growth promoters (AGP (which are essentially low dose antibiotics)) were widely used to increase feed efficiency and growth, but due to societal and political pressure and resulting legislation the use of AGP has been increasingly restricted and/or prohibited, which was instigated by the fear of transfer of bacterial antibiotic resistance to human pathogens, popularly known as superbugs. This has led to a search for non-antibiotic alternatives, based on the conventional theory that the action of AGP was mediated through the microbiota. This is extremely unlikely among others because of the fact that the dosages used were sub therapeutic, and it did not explain why not all antibiotics acted as AGP. Furthermore, alternatives based on the microbiota management hypothesis were not very effective, or at best gave quite inconsistent results as opposed to AGP.

A much more likely explanation is that antibiotics used as AGP worked through an anti-inflammatory mechanism, and indeed the most successful AGP, like the cyclines and macrolides, were demonstrated to have a direct anti-inflammatory effect both in vitro (in the absence of microbiota) and in vivo (Niewold, 2007), and there seems to be a perfect relationship between the latter property and successful use as AGP (Table 1). This is also consistent with the observation that the greatest costs for growth are due to immunological processes, in particular inflammation (Iseri and Klasing, 2013). In the absence of any challenge, an animal can grow to 100% of the genetic potential. In practice, this is of course hard to achieve, and pro-inflammatory processes lead to a growth retardation the magnitude of which depends on the circumstances. This growth retardation can be effectively remedied by supplying anti-inflammatory compounds such as AGP. It also explains the absence of effects of AGP in germ free animals or in animals raised under near optimal conditions such as they exist on many experimental farms, including the one at KU LEUVEN university. We therefore always introduce a challenge to mimic field conditions in order to see effects (Khadem et al., 2014).

In practice, production animals are exposed to many factors inducing a pro-inflammatory state. Apart from the pathogens, this can also be induced by (psychological) stress (Niewold, 2010), and by the consumption of high energy diets which leads to intestinal post-prandrial inflammation (Niewold, 2015a). Whatever the cause, inflammation is costly for production, for it causes reduction of appetite combined with increased catabolism of muscle. It also predisposes for certain (intestinal) pathogens, compounding the above problems. Overall, the result is decreased growth, decreased amino acid efficiency, increased N and P excretion, and associated environmental costs. The good news is, that one should be able to prevent this from happening by adding anti-inflammatory compounds to feed, for the clear reciprocal relationship between growth and inflammation. It is also fortunate that this is a testable hypothesis, stating that an effective growth promoter must have a direct anti-inflammatory effect on inflammatory cells in vitro, and show in vivo growth promotion accompanied by a down regulation of inflammatory parameters such as the acute phase proteins in the circulation, or lower expression of inflammatory parameters in the intestines.

Table 1. The relationship between the direct anti-inflammatory properties of antibiotics and their use as antimicrobial growth promoters (AGP), adapted after Niewold (2007).

How to select effective compounds

We routinely use the Raw 264.7 macrophage-like cell line to evaluate the influence of serial dilutions of compounds on the inflammatory response (measured as NO production) to lipopolysaccharide (LPS) stimulus. Furthermore, the response of the cells to the compounds without LPS stimulation is measured to check for the presence of pro-inflammatory constituents. To exclude artefacts due to cell death, cell viability is checked using simple tetrazolium salt methods. This test allows for fast prescreening of larger amounts of extracts and chemical compounds, in comparison with a known anti-inflammatory compound like oxytetracycline (Khadem et al., 2014). It delivers the effective dose 50 (ED50) of anti-inflammatory activity, and shows the presence of pro-inflammatory activity (Figure 1).

Figure 1. Establishing the anti-inflammatory potency of compounds in the Raw 264.7 cell assay. Oxytetracycline (closed circle) inhibits the LPS induced NO-response in a dose dependent way, whereas it does not elicit a response in cells without LPS stimulation (open circles). A phytobiotic Macleaya cordata extract contains anti-inflammatory activity in the presence of LPS (closed triangle), but also appears to contain pro-inflammatory activity towards unstimulated cells (open triangle) (unpublished results). The latter activity was removed by further purification of the extract used for in vivo trials (Khadem et al., 2014).

The results of this test can be translated to vivo, provided that the proximal intestinal uptake is low in order to achieve effective concentrations in the small intestine where the compounds exert their activity. This is the case for oxytetracycline (OTC) and Macleaya cordata extract (MCE), but short chain fatty acids like butyrate are absorbed already proximally, hence requiring stomach resistant encapsulation (van den Borne et al., 2015).

The selected anti-inflammatory compounds are subsequently tested in vivo and should show growth promotion accompanied by a down regulation of inflammatory parameters such as the serum acute phase proteins, lower levels of fecal inflammatory biomarkers, and lower expression of inflammatory parameters in the intestines. We have been able to demonstrate for OTC and MCE better growth in broilers accompanied by lower levels of acute phase proteins such as α1-acid glycoprotein (Khadem et al., 2014), and in pigs OTC promoted growth while inflammatory parameters in the serum proteome were down (Soler et al., 2016). These similar results in two different species strongly support the anti-inflammatory theory of growth promotion. It also shows the usefulness of inflammatory biomarkers (Table 2). Most biomarkers for intestinal health were usually obtained by invasive means from blood or at slaughter, it would however preferably be obtained in a minimally invasive or non-invasive way such as from feces. Fecal myeloperoxidase (MPO) is used in humans, and we use it successfully in pigs too (unpublished data). Unfortunately, MPO is not present in chicken, nor is IFABP or PAP/Reg3. Perhaps neopterin is a more likely candidate for universal use since it is identical in all species, and we are currently testing its potential. 

Table 2. Selected inflammatory biomarkers in pig (P) and chicken (C), after Niewold (2015b).

Final remarks

It is concluded that it is possible to select effective alternatives to AGP by rational design rather than by trial and error as it has been the approach in the past. Interestingly, it may also offer a plausible mechanism for the hitherto unexplained growth promotion by early-life antibiotics observed in children (Schwartz et al., 2016), and it is very striking that the most effective classes of antibiotics in the latter phenomenon are the cyclines and macrolides. This should ring a bell for those of us working in animal production, because the latter are the AGP’s of choice.

This is economically important, because of the more efficient use of (protein) resources, and hence more sustainable. In addition to that, it will reduce the environmental impact of manure (N, P) and antibiotics. The current problem is that a lot of so-called alternatives to AGP do not work because they are based on the wrong hypothesis. Farmers are tempted to find alternative ways of applying antibiotics for growth promotion such as by giving the same antibiotics therapeutically instead. This is still allowed. The problem is that this practice does not lead to any reduction in the amount of antibiotics used. Only if really effective alternatives to AGP are used, substantial reductions in antibiotic usage can be foreseen, and as demonstrated here, those can be selected and are proven to work.