Breeding programs primarily targeting growth and high meat yield have successfully transformed broiler lines in the last decades by substantially increasing growth (Zuidhof et al., 2014). While many physiological effects of breeding programs are known, such as changes to metabolism and the intestinal tract, including an increase of surface area (Mitchell and Smith, 1991; Zuidhof et al., 2014; Tallentire et al., 2016), it remains unclear if the gut microbiota has been modulated by breeding programs. The aim of this study is to address this point by comparing the cecum microbiota of a current modern commercial breeding dam line and a legacy broiler line, which has not undergone any selection since 1986 (Yair et al., 2017; Ben-Gigi et al., 2021).

It can be hypothesized that the gut microbiota is likely to be affected by physiological changes introduced in the host through breeding programs. These changes can include differences in retention time, affecting microbial clearance (Rougière and Carré, 2010). Differences in mucin expression levels can also affect the gut microbiota since bacteria use mucin as a binding site or as a nutrient (Cheled-Shoval et al., 2014). Other factors include reduced nutritional availability due to changes in host absorption (Croom et al., 1999; Schmidt et al., 2009), and changes in the regulation of components of the immune system, such as changes in secretion of antimicrobial peptides or IgA into the gut lumen (Qureshi and Havenstein, 1994; Schokker et al., 2015).

The gut microbiota can affect host growth. For example, different gut microbial communities can induce host obesity (John and Mullin, 2016) but can also reduce weight (von Schwartzenberg et al., 2021). One mechanism by which gut bacteria can positively affect growth is by converting indigestible fibers into short-chain fatty acids which the host can absorb and utilize (Krajmalnik-Brown et al., 2012). However, gut bacteria are also potential competitors and can reduce nutrient availability for the host (Romano et al., 2017). Finally, the gut microbiota can also affect the maturation and development of the host’s intestinal tract, thereby affecting its ability to utilize the feed (Hutsko et al., 2016; Dougherty et al., 2020). Thus, theoretically, there are multiple mechanisms by which modulation of gut microbiota composition during breeding programs may have supported or opposed the target of fast growth and higher meat yield.

Previous studies have examined the relationship between broiler lines and gut microbiota composition. A comparison of the cecal (Richards et al., 2019) and ileal (Richards-Rios et al., 2020) microbiomes of three fast growing commercial broiler lines up to day 42 revealed differences only on the day of hatch and day 3 of life. Thus, while some differences were observed early on, possibly because of different chick sources, the composition over time converged, likely because the chicks were raised together and exchanged gut microbes. This implies that fast growing commercial broiler lines are similar in their interaction with their gut microbiota at least until the age of 42 days. In comparison, studies comparing fast growing broiler lines to a historic line, or a line selected for slow growth, were able to identify differences in ileal and fecal bacterial communities. Lumpkins et al. (2010) identified composition differences between the ileal bacterial community of a historic line and two commercial modern broiler lines in the first 35 days of life, and Zhao et al. (2013) identified multiple composition differences in fecal samples at the age of 245 days of two divergent lines, selected for 54 generations for high or low body weight. Finally, a recent study comparing the ileal microbiota of four different lines of fast and slow growing broilers including the ancestral Jungle Fowl at the age of 56 days found unique signatures for the different lines and predicted microbiota functions (Emami et al., 2022). Here we extend these studies by comparing the cecum bacterial community of breeder dams from a commercial broiler line and a Legacy line. As the cecum is the site of bacterial fermentation in chickens, any relevant effect of the gut community on poultry growth would likely occur at the cecum. Furthermore, differences in adult breeders might be easier to detect because the microbiota has stabilized, they might reveal physiological differences that are relevant to younger birds, and they might affect the fertility as well as egg laying efficiency of breeders, including the nutrients deposited into the egg.

Thus, to examine the interaction of genetics and microbiota composition in the context of breeding programs, we raised modern Cobb breeder dams alongside dams from a Legacy line, which was kept as a relaxed line (without selection) from 1986 (Yair et al., 2017; Ben-Gigi et al., 2021) and compared their cecum bacterial communities.

To determine if breeding programs also modulated the gut microbiota, dams from the current Cobb commercial line, which targets high growth and meat yield, were compared with dams from a Legacy line that had not undergone targeted selection since 1986. Indeed, in the 35 years that have passed, the primarily growth targeted breeding program had a substantial effect on chicken size even under a feed restricted diet. This change in total weight is likely accompanied with multiple physiological changes that can affect gut microbiota community composition (Mitchell and Smith, 1991; Zuidhof et al., 2014; Tallentire et al., 2016).

To avoid confounding factors, we sampled Cobb and Legacy breeder dams that were housed together from day of hatch. This way, all dams were subjected to the exact same environmental conditions, including temperature, exposure to pathogens, feed (Havenstein et al., 2003), and handlers. When sampling age-matched animals that grow at different rates, any identified differences might be a result of the different developmental states. We avoided this confounder by sampling at an age when both breeds have reached peak laying performance and are well into adulthood. Last, by allowing the two lines to grow together for 37 weeks, we have allowed enough time to pass for multiple microbial cross contamination events between birds to occur. By removing these confounders, we have ensured that identified differences in microbiota composition are likely to be a result of gut environment differences due to differences in genetics.

The effects of breeding programs targeting growth on host physiology are still being assessed. Apart from the positive effects, including growth itself, a number of negative effects have been identified, including skeletal defects, metabolic disorders and altered immune function (reviewed by Zuidhof et al., 2014). Our results show a change in the cecum community composition between a current modern breeding line and the Legacy line. These results add to a previous report showing differences in the ileal bacterial community of a historic line and two commercial modern broiler lines (Lumpkins et al., 2010), and another report showing unique bacterial signatures for four different fast and slow growing broilers, including the ancestral Jungle Fowl (Emami et al., 2022). Thus, it can be concluded that genetic changes introduced during the breeding program resulted in a change in the gut bacterial community. These results raise an interesting question: are these genetic and physiologic differences between the two lines a direct result of a breeding program aimed at fast growth and meat yield, or did they happen by chance? The most prominent difference in the cecum community was that Akkermansia genus was a relatively high abundance member in the Cobb dams, while in the Legacy dams it was mostly absent (Figure 6). Interestingly, a high prevalence of Akkermansia was also found in Ross breeders (Figure 7). Assuming the breeding programs that gave rise to the current Cobb and Ross breeds are independent, these results might imply that Akkermansia in breeders is associated with fast growth and high meat yield phenotypes. Further research is required to establish this association.

Akkermansia bacteria are interesting as they are also found in humans and are studied as a future probiotic strain (Naito et al., 2018). These bacteria are mucin degraders and have been inversely correlated with metabolic disease in humans and mice (Everard et al., 2013; Naito et al., 2018). It was also correlated with high feed efficiency in layer chickens (Yan et al., 2017). Other bacterial orders which are different between Cobb and Legacy dams include Clostridiales, Lactobacillales, and Aeromonadales, which are more abundant in the ceca of Legacy dams. Higher levels of Lactobacillales in Legacy dams may result in reduced pH levels in the cecum, as these bacteria produce lactic acid which reduces the environment’s pH (O’Hanlon et al., 2013). This environmental change may inhibit pH-sensitive Akkermansia bacteria (Van Herreweghen et al., 2017). This is supported by the negative correlation observed between Lactobacillales and Akkermansia levels. Last, we identified a relatively large variation in Lactobacillales levels in the Legacy breed (Figure 6). In an attempt to explain this variability, we also noted a large variability in Bacteroidales levels (Figure 6). Indeed, we found a negative correlation between the two (Table 1). This negative correlation was previously observed, and it was suggested that these groups have an overlapping ecological niche based on their encoded carbohydrate utilizing functions (Ma et al., 2020).

Our results show specifically that Akkermansia bacteria colonized Cobb but not Legacy dams. One hypothesis for this difference is that breeding programs select not only for host genetics but also for specific bacteria. If some bacteria are vertically transmitted between generations, perhaps by surviving in or on the egg and colonizing the chicks (Ding et al., 2017; Lee et al., 2019; Shterzer et al., 2020) genetic drift processes in these bacteria could result in divergent strains that are specifically adapted to the selected chicken line. Indeed, it is known that different mouse strains harbor different gut microbial communities (Jacobson et al., 2018). However, most of the bacterial composition between Cobb and Legacy dams was similar, implying this was not true for most bacterial strains. Intentional exposure of newly hatched chicks to Akkermansia resulted in colonization at high levels (Kubasova et al., 2019). However, Akkermansia were not found in young individuals sampled from other sites. It could be expected that if Akkermansia bacteria were common to Cobb because they were carried on or in eggs, they would flourish by marketing age. Therefore, the differences we identified between Cobb and Legacy dams are not likely dependent on the origin facility or on vertical transmission, but on genetic differences between the lines. Moreover, any differences originating from origin facility likely disappeared through cross contamination, because both lines were housed in the same shed since day of hatch. This suggests the difference in Akkermansia colonization between Cobb and Legacy dams stems from physiological differences affecting the gut environment.

The identification of Akkermansia as differentially abundant between the two lines is of interest also because this group of organisms are known to degrade mucin. Theoretically, at least two options exist to explain this difference. One is that some of the genetic changes which have occurred during the growth directed breeding program regulate mucin secretion levels and/or composition. In this case, Akkermansia organisms might better colonize the industrial Cobb line because they find more suitable mucin in the cecum, which they can degrade and utilize as a nutrient source. The other option is that in the less feed-efficient Legacy dams, more nutrients pass the small intestine into the cecum, allowing the creation of a bacterial community which utilizes diverse nutrient sources. In comparison, the feed-efficient Cobb dams absorb most of the feed derived nutrients in the small intestine, leaving less feed derived nutrients to reach the cecum. In such an environment, mucin degraders feeding off cecum produced mucins will be more successful. In this scenario, host mucin genetics are likely not changed; rather genetic changes that affect feed-efficiency indirectly affect cecum composition. However, other mechanisms might also affect colonization of Akkermansia and other bacteria. For example, differences in metabolism causing changes in body temperature may affect colonization success (Tallentire et al., 2016). Thus, further research is needed to determine whether mucin levels or composition are indeed involved in the differential levels of Akkermansia between slow- and fast-growing lines.

The contribution of the gut microbiota to the mechanism of action of breeding programs is unknown. Our results show that cecum microbiota composition is different between the two groups of dams. As the cecum microbial community contributes to the digestion of nutritional fibers found in the feed that the chicken cannot digest by itself (Józefiak et al., 2004), it is possible that these differences contribute to the high feed efficiency of modern commercial breeds. However, it should be noted that we have identified these differences in mature dams and not in younger poultry, which are the main target of breeding programs. Furthermore, our results show that Akkermansia, which specifically colonize modern breeder dams, do not colonize modern broilers at least until marketing age. Therefore, it is unclear if the identified differences in the cecum bacterial community contribute to the fast growth of modern poultry lines. We have also shown that a major difference is that Akkermansia organisms colonize Cobb but not Legacy dams. Indeed, Akkermansia muciniphila has been correlated with better feed efficiency in layers (Yan et al., 2017). On the other hand, exposure of newly hatched chicks to A. muciniphila seems to have a minimal impact on growth (Zhu et al., 2020).

The identification of higher levels of Akkermansia in Cobb dams raises an interesting question: does Akkermansia cause Cobb dams to divert energy from egg to mucin production? If this is the case, strategies that will limit Akkermansia colonization might improve egg production.

In this study, we compared the composition of the gut microbiota of the current Cobb commercial breeder line and a Legacy line which has not undergone selection for 35 years. We were able to identify differences in the cecal bacterial community that were the result of genetic changes brought about by the broilers breeding programs. Specifically, Bacteria of the genus Akkermansia implicated in mucin degradation and associated with host metabolic health were a prominent member of the Cobb breeders’ cecum community, but were mostly absent from Legacy line dams. Inversely, Legacy dams had higher levels of Clostridiales, Lactobacillales and Aeromonadales. While we do not know if these differences also contribute to the fast growth of the current commercial line, by identifying these bacteria, we can now specifically target them for further study.