Meta-analysis of dietary chitosan effects on performance, nutrient utilization, and product characteristics of ruminants

2.1 Database development

A database was developed to investigate the influence of dietary CHI addition on performance, nutrient utilization, and product characteristics of ruminants. Literature search was performed in Scopus database using the keywords “chitosan” and “cattle,” “cow,” “goat,” “sheep,” “ruminant,” or “rumen”. A total of 965 articles were initially obtained, and after further title and abstract evaluations, 31 articles were selected. Inclusion criteria for an article to be registered in the database were (1) the article was published in a peer-review journal, (2) English was used in the article, (3) the article reported dietary addition of CHI and its effects on ruminants, and (4) the experiment reported was performed directly on ruminants in vivo, that is, beef cattle, dairy cows, sheep, or goats. In vitro rumen experiments that evaluated the addition of CHI were not integrated in the database. We performed such in vitro meta-analysis of CHI, and the article had been published elsewhere (Harahap et al., 2020). After considering such criteria, 15 articles that composed of 21 experiments and 57 data points were included in the database (Araújo et al., 2015; De Paiva et al., 2017; Del Valle et al., 2017; Dias et al., 2017; Gandra et al., 2016; Garcia-Rodriguez et al., 2015; Goiri, Oregui, & Garcia-Rodriguez, 2010; Haraki et al., 2018; Henry et al., 2015; Kirwan et al., 2021; Mingoti et al., 2016; Seankamsorn et al., 2021; Vendramini et al., 2016; Zanferari et al., 2018; Zheng et al., 2021). Details of studies included in the meta-analysis is presented in Table 1. Number of articles reported the effects of dietary CHI addition on beef cattle, dairy cows, and sheep were six, seven, and two articles, respectively. These number of articles were not many since the use of CHI as an additive in ruminant diets is relatively new (the oldest article was published in 2010).

In all the papers, CHI sources were originated from commercial suppliers and their deacetylation degree ranged from 85 to >95%. Parameters integrated in the database were dry matter intake (DMI), milk production, milk efficiency (milk production per unit DMI), rumen fermentation (pH, ammonia [NH₃], total volatile fatty acids [VFA], acetate [C2], propionate [C3], butyrate [C4], valerate [C5], branched-chain volatile fatty acids [BCVFA], acetate to propionate ratio [C2/C3]), nutrient digestibility (dry matter digestibility [DMD], crude protein digestibility [CPD], neutral detergent fiber digestibility [NDFD], ether extract digestibility [EED]), nitrogen balance (N intake, N urine, N feces, N milk, N balance, N microbial, microbial protein), blood metabolite (glucose, cholesterol, blood urea nitrogen [BUN], aspartate aminotransferase [AST], gamma-glutamyl transferase [GGT]), and milk composition (milk fat, milk protein, milk lactose, milk urea nitrogen [MUN], stearic acid [C18:0], and vaccenic acid [t11 C18:1]. Parameter related to methane (CH₄) emission was not integrated in the database since it was originated from only one paper. Parameters related to beef cattle production such as average daily gain and carcass characteristics and composition were not reported in all papers and therefore could not be integrated as well.

Values for some parameters were reported in different measurement units. In such cases, calculations were performed in order to transform the values into similar units based on information provided in the articles. Since the present study integrated all experiments with different ruminant species, both small and large ruminants, parameters depended on body size were standardized by relating them to metabolic body weight (BW^0.75) or DMI. Statistical summary of the database is presented in Table 2.

2.2 Data analysis

Meta-analysis was based on the mixed model methodology (Sauvant et al., 2008; St-Pierre, 2001). CHI addition levels were considered as fixed effects, whereas different studies were treated as random effects. The statistical model employed was as follows:

where Y_ij is a dependent variable, B₀ is an overall intercept across all studies (fixed effect), B₁ is the linear regression coefficient of Y on X (fixed effect), B₂ is the quadratic regression coefficient Y on X (fixed effect), X_ij is the independent variable (CHI addition level), s_i is the random effect of study i, b_i is the random effect of study i on the regression coefficient of Y on X in study i, and e_ij is the unexplained residual error. Model statistics used were p value, root-mean-square error and coefficient of determination (R²). Significance was declared when p < 0.05 and a tendency was stated at 0.05 < p < 0.1. When a parameter was not significant at p < 0.05 nor had a tendency at p < 0.1, the quadratic mixed model above was altered to its corresponding linear mixed model. The statistical analysis was performed by using SAS software version 9.1 (SAS Institute Inc., Cary, NC, USA).

4.1 Effects CHI on performance, rumen fermentation, and nutrient digestibility

CHI is a natural, non-toxic and biodegradable biopolymer with antimicrobial properties that can be used to manipulate rumen microbial ecosystems and also can help to reduce the use of chemicals and avoid drugs resistance (Cushnie & Lamb, 2011; Jiménez-Ocampo et al., 2019; Kong et al., 2010; Vendramini et al., 2016; Zhan et al., 2017). The present study revealed that the addition of CHI to diets did not affect the DMI, milk production, and milk efficiency. The results of this study follow some previous studies reporting that the addition of CHI in the ruminant diets did not influence DMI (Araújo et al., 2015; De Paiva et al., 2017; Henry et al., 2015; Mingoti et al., 2016; Vendramini et al., 2016; Zanferari et al., 2018). In several studies, DMI and NDFI were found to decrease when supplemented together with whole raw soybean or soybean oil (Del Valle et al., 2017; Gandra et al., 2016; Haraki et al., 2018). Zanferari et al. (2018) also found a decrease on DMI by supplementing whole raw soybean + CHI, and they did not observe any effect on DMI in a soybean oil + CHI treatment at the same experiment. This evidence corroborates our result in this meta-analysis that CHI does not have a significant negative effect on nutrient intake but might be not recommended supplementing together with raw soybean ingredient. This was supported by previous studies suggesting that the negative effect on DMI was due to the alteration of ruminal fermentation caused by the diets composition rather than the direct effect of CHI. This proposed mechanism is the most widely accepted so far, similar to the ionophores effects (Del Valle et al., 2017; Zanferari et al., 2018).

This meta-analysis demonstrated that the addition of CHI had no effects on milk production and milk efficiency. It might be explained by the insignificance impact of adding CHI to DMI obtained in this study. The DMI is a main factor contributing to the production performance of various animal species including ruminants since it provides nutrients for tissue synthesis. In addition, it is generally expected that dietary CHI would increase productive performance of ruminants considering its positive effect on rumen fermentation, that is, increase propionate proportion, decrease acetate to propionate ratio, and alter rumen bacterial population (De Paiva et al., 2017; Harahap et al., 2020; Mingoti et al., 2016; Tong et al., 2020). However, such improvement may be diminishing probably due to short treatment and CHI maybe interacted with other feed components. In this regard, further investigation is important to conduct.

In this meta-analysis, no substantial change was observed on pH value when CHI was included to the ruminant diets, which was similar with most of previous published studies (Araújo et al., 2015; De Paiva et al., 2017; Del Valle et al., 2017; Dias et al., 2017; Goiri, Oregui, & Garcia-Rodriguez, 2010; Vendramini et al., 2016). However, another in vivo study reported that increasing level of CHI to the diet was associated with the increase of rumen pH (Zanferari et al., 2018). Belanche, Pinloche, et al. (2016) reported that CHI caused essential changes in the structure of rumen bacterial community, that is, decreasing the number of Firmicutes and Fibrobacter but increasing Bacteroidetes and Proteobacteria, which includes most amylolytic bacteria. However, increasing starch-degrading bacteria did not change rumen pH, possibly due to increasing proteolytic bacteria (Kirwan et al., 2021; Zanferari et al., 2018) that help to maintain the buffering capacity of the rumen environment. Additionally, the increase of buffering capacity could also be attributed to the mechanism of format production that can diffuse into the rumen liquid phase to form HCO₃− and H₂ and the formation of previous products (Leng, 2014).

Regarding the CHI effects on rumen fermentation, the present study did not find significant effect of dietary CHI on rumen ammonia concentration which was in agreement with previous in vivo studies (De Paiva et al., 2017; Del Valle et al., 2017; Vendramini et al., 2016; Zanferari et al., 2018) but different with our previous meta-analysis based on in vitro experiment. There have been several experiments reporting that CHI markedly influenced ammonia concentration. For instance, elevating rumen NH₃ concentrations was observed in beef steers received CHI at 100 mg/kg BW and 1,200 mg/kg DM, respectively (Araújo et al., 2015; Dias et al., 2017). Beier and Bertilsson (2013) have also reported that higher ammonia concentrations in the dietary CHI may be related to the conversion of amine groups (R-NH₂) to ammonia, due to CHI is a nitrogen compound that can be degraded by rumen microbes.

Several in vivo studies observed that dietary CHI did not affect total VFA concentration (Araújo et al., 2015; De Paiva et al., 2017; Del Valle et al., 2017; Dias et al., 2017; Goiri, Oregui, & Garcia-Rodriguez, 2010; Vendramini et al., 2016; Zanferari et al., 2018), which was confirmed in the present meta-analysis. However, the effects of the provision of CHI on molar VFA were previously reported, such as decreased the acetate (C₂) concentration (Araújo et al., 2015; Vendramini et al., 2016; Zanferari et al., 2018) and increased the propionate (C₃) concentration (De Paiva et al., 2017; Dias et al., 2017; Goiri, Oregui, & Garcia-Rodriguez, 2010; Vendramini et al., 2016), resulting in the decreased of C₂/C₃ ratio that was evidenced in the present meta-analysis as well as in our previous meta-analysis, whereas CHI addition linearly decreased C₂ while increasing C₃ production (Harahap et al., 2020). Araújo et al. (2015) reported that additional dietary CHI affected quadratically of increasing C₃ to a dose of 100 mg/kg BW and increasing butyrate (C₄) to a dose of 50 mg/kg BW. It could explain the occurrence of the ratio acetate to propionate reduction with dietary CHI. This present study reported no effect of dietary CHI to valerate (C₅) concentration, but different from others reported that dietary CHI reduced C₅ concentration on ruminal fermentation of lactating cows (Zanferari et al., 2018). The decrease in BCVFA concentration in the present study was similar to the previous in vivo research showing that the concentration of BCVFA decreased with dietary CHI (Goiri, Oregui, & Garcia-Rodriguez, 2010; Zanferari et al., 2018). The reduction of Gram-positive bacteria in the rumen caused an increase in propionate production (Gandra et al., 2016; Jiménez-Ocampo et al., 2019). Belanche, Pinloche, et al. (2016) reported that the higher propionate and lactate as fermentation products in response to CHI supplementation is attributed to the increase in amylase activity detected in cows fed CHI diets. When this happened, it could promote to increase efficiency of energy utilization that can be metabolized for growth through changes in the molar proportion of VFA (Blaxter, 1989; Jiménez-Ocampo et al., 2019). Propionate is the main precursor of glucose synthesis in ruminant, which is important as energy to support the growth. Higher propionate production at the same DMI in this study indicated a higher energy allocation for growth or production, which means a higher energy efficiency attributed to CHI supplementation.

The present meta-analysis provides strong evidences that dietary CHI significantly increased DMD and CPD but had no effect on NDFD, which are confirming previous studies with concentrated-based diets (Araújo et al., 2015; De Paiva et al., 2017; Del Valle et al., 2017; Dias et al., 2017; Gandra et al., 2016; Vendramini et al., 2016). Previous study examining the possible association effect of CHI and raw soybean resulted in lower DMD, NDFD, and EED (Zanferari et al., 2018), showing the detrimental effect of elevating fat content in the diets. Thus, it is noteworthy that CHI may be beneficial when provided without increasing dietary fat. In the rumen, CHI is associated with a decrease in protozoa populations because it is well known that CHI properties can impact the permeability of protozoa cells (Belanche, Ramos-Morales, & Newbold, 2016; Wencelová et al., 2013) due to interactions between polycationic CHI and electronegative charge on the microbial surface (Muxika et al., 2017). Considering that dietary CHI may cause a decrease in protozoa population with extending effect to reduce NDFD, appropriate dose and form may help to optimize CHI effects on ruminant production.

Dietary CHI increased in CPD, strengthen the evidence from meta-analysis based on in vitro experiment (Harahap et al., 2020). This result was also similar to previous studies (Araújo et al., 2015; De Paiva et al., 2017; Del Valle et al., 2017; Dias et al., 2017; Mingoti et al., 2016; Vendramini et al., 2016). The explanation regarding the increasing CPD is that CHI was proposed to increase intestinal permeability of the rumen, thus increasing digestibility of nutrients (Del Valle et al., 2017). In addition, it can also be related to the ionophores mechanism, whereas CHI can interact with rumen bacteria to promote deamination and proteolysis, making protein less degradable in the rumen but increase amino acids flow in the small intestine (Mingoti et al., 2016). Such protective mechanism of amino acids from ruminal degradation is plausibly associated with positive charges of CHI -NH₂₊ groups and with negative electrostatic charges with carboxyl in the amino acid group (Chiang et al., 2009). Apart from the possible antimicrobial effect on rumen microbes, CHI can also interact with nutritional feed and partially prevent rumen degradation to some extent similar to tannins (Jayanegara et al., 2017; 2020). It forms complexes with feed nutrients, especially proteins (Kondo et al., 2014).

4.2 Effects of CHI on N balance, blood metabolites, and milk composition

The addition of CHI in diets showed no effect on N metabolism as reflected with similar N outputs in urine, feces, milk, as well as N balance, microbial N, and microbial protein concentration. Increasing CPD with similar N output in the feces indicates a better N use by the ruminants. De Paiva et al. (2017) reported an increase in N milk in the dietary CHI treatment due to increased CPD. Increased microbial nitrogen production can be associated with an increase in feed intake and availability of ruminal ammonia nitrogen after supplying CHI to animals, due to the antimicrobial effect on rumen modulator (Goiri et al., 2009). Although some studies have reported a decreasing trend (Del Valle et al., 2017) and increasing trend on N intake (Vendramini et al., 2016), improvement of N use efficiency in ruminants has been the most widely accepted effect regarding the use of CHI as an additive.

Dietary CHI did not affect blood glucose concentration which is in agreement with most experiments with CHI treatment (De Paiva et al., 2017; Del Valle et al., 2017; Gandra et al., 2016; Vendramini et al., 2016; Zanferari et al., 2018). This study also demonstrated that CHI treatment did not alter BUN, SST, and GGT concentrations, giving evidence that CHI is safe without hepatotoxicity effects because these liver enzymes are the indicators of metabolic and health status when the value is within the normal range (Kaneko et al. (1997). However, because it has been reported that intraluminal propionate infusion can harm feed intake and milk fat concentration (Oba & Allen, 2003; Sheperd & Combs, 1998), more research is needed to rule out potential adverse effects of CHI when using higher inclusion rates. On the other hand, current investigation revealed that the addition of CHI increased totals cholesterol concentration which can be attributed to the protective effect of CHI on fatty acids, leading to increase cholesterol synthesis.

Generally, CHI provision for ruminants did not alter milk fatty acids (FA) contents especially C18:0 and t11 C18:1. In addition, CHI treatment also did not affect milk protein, lactose, and MUN concentration but at certain level decreased milk fat content in a curvilinear pattern (Table 4). These results are following previous studies reporting that dietary CHI did not affect milk protein and milk lactose (Del Valle et al., 2017; Garcia-Rodriguez et al., 2015; Mingoti et al., 2016; Vendramini et al., 2016; Zanferari et al., 2018). Regarding the CHI effect on milk fat concentration and FA profile, available literature is not sufficient to elucidate a clear mechanism toward the association between CHI and FA in the rumen. However, it was proposed that CHI exhibited a complete ruminal biohydrogenation inhibition, resulting in the higher monounsaturated fatty acid and some polyunsaturated fatty acid (PUFA) concentrations (Goiri, Indurain, et al., 2010). Another study in dairy cows supported the findings that CHI suppressed major bacteria involved in the ruminal biohydrogenation such as Butyrivibrio group and Butyrivibrio proteoclasticus (Zanferari et al., 2018). Bacteria involved in the lipolysis and FA ruminal biohydrogenation consisted of Anaerovibrio lipolytica that liberates fatty acids from their glycerol backbone, Butyrivibrio fibrisolvens that biohydrogenates PUFA into vaccine acids, and finally B. proteoclasticus that plays a role in the final steps of the biohydrogenation, that is, the conversion of vaccenic acid to stearic acid, saturated fatty acid C₁₈ (Jenkins et al., 2008; Lourenco et al., 2010; Toral et al., 2018; Vasta et al., 2019). Experiment performed using Next Generation Sequencing has reported that the addition of 5% of dietary CHI can suppress the relative abundance of both Anaerovibrio sp. and Butyrivibrio sp. in the rumen simulation technique system (Belanche, Pinloche, et al. (2016). At this point, CHI might be used to enhance beneficial fatty acids such as PUFA, omega-3 fatty acids, and CLA through the mechanism of modulating fatty acid metabolism in the rumen. However, in some cases, such microbial alteration might also lead to decrease DMI and fiber digestibility given the important role of Butyrivibrio group in fiber-degrading function in the rumen. In addition, milk fat might also be expected to decline as acetate to propionate ratio decreases because acetate is the main source of milk fat synthesis. Our present meta-analysis is consistent with this basic principle in dairy animals (Table 4), which was also in agreement with previous experiment when CHI fed to dairy cows with raw soybean meal (Zanferari et al., 2018; Zheng et al., 2021).

Taken together, the present meta-analysis proposed that CHI can be used as a natural rumen modifier since it is able to modify rumen fermentation towards a favorable direction, such as increasing propionate and decreasing acetate which are desirable for higher energy synthesis with potentially lower methane emission. Although CHI provision positively increased dry matter and CPD, however, it limitedly affects performance of ruminants.