http://orcid.org/0000-0002-2559-6379
Figures
Abstract
This study investigated the associations between abundance of rumen ciliate protozoa and the proportion of the main bioactive fatty acids related to rumen biohydrogenation, as 18:0, t10-18:1, t11-18:1, c9,t11-18:2, 18:3n-3 and 18:2 n-6, in rumen and meat of growing lambs, using data derived from 3 production experiments. A global correlation analysis and a linear regression analysis considering the effect of the experiment were performed. Ten of the 86 lambs involved in the experiments did not present ciliate cells in rumen liquor and the remaining lambs presented an average of 1.35 × 106ciliates / ml rumen liquor. From the nine genera of ciliates identified, Entodinium was the most abundant, averaging 1.17 × 106 cells / ml of rumen liquor. A large variation among lambs was observed for both rumen concentration and community structure of ciliates. Rumen t11-18:1 (P < 0.001) and meat deposition of t11-18:1 (P < 0.001) and of c9,t11-18:2 (P < 0.001) increased linearly with total ciliates, whereas the t10/t11 ratio in rumen (P = 0.002) and in meat (P = 0.036) decreased linearly. Entodiniomorphids seems to be strongly related with meat deposition oft11-18:1 and c9,t11-18:2 and with the reduction of the trans-10 shifted pathway. Completeness of RBH decreased linearly with Holotrichs (P = 0.029), Entodiniomorphids (P = 0.029), Isotricha (P = 0.011) and Epidinium (P = 0.027) abundances. Rumen 18:0 also decreased linearly with increasing counts of total ciliates (P = 0.015), Holotrichs (P = 0.020), Entodiniomorphids (P = 0.010) and Isotricha (P = 0.014). Rumen protozoa were positively linked with the deposition of healthy bioactive FA and simultaneously negatively associated with the occurrence of trans-10 shift.
Citation: Francisco AE, Santos-Silva JM, V. Portugal AP, Alves SP, B. Bessa RJ (2019) Relationship between rumen ciliate protozoa and biohydrogenation fatty acid profile in rumen and meat of lambs. PLoS ONE 14(9): e0221996. https://doi.org/10.1371/journal.pone.0221996
Editor: Aneta Agnieszka Koronowicz, University of Agriculture in Krakow, POLAND
Received: May 29, 2019; Accepted: August 19, 2019; Published: September 6, 2019
Copyright: © 2019 Francisco et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This study had financial support through the grants ALT20-03-0145-FEDER-000040 from Alentejo2020 and UID/CVT/276/2019 from Fundação para a Ciência e Tecnologia. Individual fellowships S. Alves (SFRH/BPD/76836/2011), from Fundação para a Ciência e Tecnologia are acknowledged. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
By rumen biohydrogenation (RBH), the dietary unsaturated fatty acids (UFA) are converted by rumen microbiota to saturated fatty acids (SFA) which negatively impact to human health. However, RBH is not complete and a variable proportion of UFA submitted to RBH end up as a isomerized or partially hydrogenated FA, hereafter named biohydrogenation intermediate (BI), that are absorbed and incorporated in ruminant´s milk and meat fat[1]. Some of them, present healthy properties such as vaccenic (t11-18:1) and rumenic (c9,t11-18:2) acids. The fibrolytic bacteria, which are believed to be the main responsible for RBH pathways that generate t11-18:1 and c9,t11-18: 2 as the main octadecaenoic and octadecadienoic BI[2], are very sensitive to low ruminal pH[3]. Diets rich in quickly fermentable carbohydrates, like starch, led to low rumen pH, impacting the ecosystem equilibria and the balance between microbial populations. Such conditions favour the amilolytic populations associated to the occurrence of a shift in RBH pathways where the major BI becomes t10-18:1 instead of t11-18:1 (i.e. trans-10 shift) [4]. Avoiding the trans-10 shift is a main goal in the development of strategies to improve nutritional value of ruminant products since t10-18:1 is not a precursor of c9,t11-18:2 in tissues [1] and is considered as an unhealthy FA [5]. Moreover, in dairy ruminants, trans-10 shift has been related to the milk fat depression syndrome [6].
Global warming and climate changes concerns have been the trigger for intensive research on mitigating methane production by ruminants [7, 8]. Symbiotic relationships between rumen protozoa and methanogenic archea involving the H2 transfer from protozoa to methanogenese have been described. Defaunation (i.e. elimination of protozoa from rumen) is one of the strategies proposed to reduce methane emissions by ruminants [7, 8]. However, this approach may have implications in other mechanisms of rumen metabolism.
Ciliate protozoa, especially Entodiniomorphids, directly interfere on rumen metabolism of non-structural carbohydrates by engulfing the starch granules [9] and thus slow down starch fermentation and promote a more stable and higher rumen pH [10, 11]. Moreover, by predating amylolyticbacteria, ciliate protozoa also reduce the contribution of those bacteria to rumen starch digestion[9]. Thus, protozoa act as a stabilizing agent of starch fermentation in the rumen of grain-fed ruminants[8]. Besides that, Entodiniomorphids are also responsible for a direct reduction of lactic acid concentration, due to their high efficiency in removing the lactic acid from rumen environment [12] which also contributes to prevent pH values below 6.0, which is referred by [3] as the critical value for the activity of cellulolytic bacteria.
Rumen ciliate protozoa can comprise up to half of the rumen biomass and up to 75% of microbial lipids present in the rumen [9]. Their role on rumen lipid metabolism is still not totally understood [8], but it is generally accepted that ciliate protozoa are not directly involvedto RBH [2]. However, they actively contribute to lipolysis, namely the cellulolytic species as Epidinium sp. [13, 14] and influence the RBH by different mechanisms. Ciliate protozoa ingest and directly incorporate dietary PUFA in their cellular membranes, protecting them from hydrogenation by bacteria; they extensively prey and engulf several species of biohydrogenating bacteria and have symbiotic relationships with others [2, 10, 15–17].[18] reported that when compared with faunated lambs, protozoa-free lambs presented higher levels of SFA, including stearic acid (18:0) in the intramuscular fat (IMF). This may suggest that the presence of rumen protozoa may contribute to reduce RBH completeness (i.e. the proportion of dietary biohydrogenated UFA that end up as totally saturated FA) and consequently, the deposition of SFA in edible product. Moreover, ciliate membranes are rich in BI such as t11-18:1 and c9,t11-18:2[19, 20] and the levels of PUFA in protozoa cells are proportionally higher when compared to rumen bacteria [13, 15]. Despite the controversy over the mechanisms by which rumen protozoa accumulate t11-18:1, c9,t11-18:2 and PUFA in their membranes, due to their large contribution to rumen biomass, these microorganisms may represent an important reservoir of such FA.
Therefore, the main goal of the present work was to study the relationships between ciliate protozoa in rumen and the proportionsof the BI t10-18:1, t11-18:1, c9,t11-18:2, the occurrence of trans-10 shift, and the proportion of PUFA in rumen content and in meat of lambs fed with complete diets.
Material and methods
Animal experiments
Data used in the present study were obtained from three independent productive experiments performed with growing lambs, conducted by our research team in 2017 and 2018. All the experiments were performed at Estação Zootécnica Nacional, Instituto Nacional de Investigação Agrária e Veterinária (EZN-INIAV), located at Vale de Santarém, Portugal. They were conducted in the framework of the ValRuMeat research project to evaluate the impact of the diet on the nutritional value of meat from ruminants fattened in intensive production systems, particularly on the levels of t11-18:1 and c9, t11-18:2 in and the occurrence of the trans-10 shift.
A total of 86 lambs were used in the three experiments and all procedures followed the Directive 2010/63/EU which regulates the use of production animals in animal experiments, and were approved by the Organ Responsible for the Animal Welfare of INIAV I.P. (ORBEA-INIAV). The diets were formulated by our research group and prepared in the Feed Compound Unit of EZN-INIAV. Animal management, experimental conditions, slaughter, sample collection and analytical procedures were similar in the three experiments and were reported by [21]. The crossbred Merino Branco lambs, were reared by their dams on extensive pasture until weaning, with about 60 days of age. At that time, lambs were transported to EZN-INIAV, housed and randomly assigned to the individual pens and to the treatments. The diets were supplemented with 60 g/kg DM of soybean oil, to increase the supply of PUFA, mainly linoleic acid, in rumen. All the diets were presented in the ground form, except the diets with hay in Experiment 2. Feed was offered daily at 9:00 am at a rate of 110% of ad libitum intake and lambs had permanent access to clean water. Lambs were weighed every week and the duration of the trials was of 6 weeks, after 1 week of adaptation to the experimental conditions. In Table 1 are presented the summaries of the objectives and of the treatments in the three experiments. The detailed ingredient and chemical composition of the diets are presented in S1, S2 and S3 Tables.
Sample collection and laboratory analysis methods
Lambs had feed and water available until they were transported in groups of 3–4 to the experimental abattoir of EZN-INIAV, where they were slaughtered. Abattoir facilities are located at the campus of EZN-INIAV, about 400m from the lambs´ barn. Lambs waited a maximum of 30 minutes until they were stunned and exsanguinated.
Rumen content sampling was performed immediately after slaughter in the abattoir. The whole rumen content of each lamb was immediately collected, homogenized and a sample was strained through 4 layer of cheesecloth resulting in an aliquot with about 80 ml of rumen liquor. The pH of rumen content was immediately measured using a pH meter (Metrohm 744) and an aliquot of 2 ml was immediately preserved with 2 ml of 10% formalin solution and stored at 2°C until microscopic examination and analyses of the rumen ciliate protozoa. A representative sample of whole rumen content was also collected from each lamb and frozen, freeze-dried, milled and stored at −20°C until FA analysis. Also a sample of the liquid fraction of ruminal content was collected and frozen until volatile fatty acids analysis. Longissimus thoracis (Lt) muscle from all the carcasses were sampled at the third day after slaughter and the meat samples were frozen, freeze died and stored at −20°C until FA analysis, according to [22].
The analysis of ciliate protozoa was performed individually by microscopic counting as described by [23] and using a Brand® counting chamber Blaubrand® Neubauer Improved with 0·100 mm depth (BR7178110 Sigma-Aldrich, Portugal) and a LeitzLaborlux K binocular light microscope (Leitz, Germany). Ciliate cell numbers were determined in duplicate for each sample and the identification at the genus level was made based on protozoa morphology, according to [24]. Protozoa community structure was also accessed by the type of rumen protozoa population (A B or O), according to [24]. Microscopic counting of protozoa was performed since it allows a better accuracy when relating the rumen concentration of total ciliates and of individual genera with rumen parameters and meat FA[8].
Fatty acid methyl esters (FAME) from freeze-dried rumen content samples were directly transesterified by reaction with sodium methoxide in methanol (0.5 M) at 50°C for 10 min followed by reaction with HCl in methanol (2.5 M) at 80°C for 15 min. Afterwards, thin-layer chromatography was used to separate FAME (fraction 1) from the fractions containing oxo- and hydroxy-groups and dimethylacetals (fraction2) as described by [25]. Methyl heneicosanoate (21:0) (internal standard) was used for quantification at concentrations: 1 mg for FAME and 50 g for oxo- and hydroxy-FAME. FAME were quantified by GC with flame ionization detection (GC-FID) using a Shimadzu GC 2010-Plus (Shimadzu, Kyoto, Japan) equipped with aSLB-IL111 (100 m × 0.25 mm, 0.20 m film thickness, Supelco, Bellefonte, PA, USA) capillary column. The chromatographic conditions were as follow: injector and detector temperatures were set at 250°C, helium was used as the carrier gas at1 ml/min constant flow and two different oven temperature programs were used to analyze the two fractions. For the FAME fraction, the GC oven was maintained at 168°C for 43 min, afterwards the temperature was increased at 2°C/min to 220°C, and kept at this temperature for 10 min.
Intramuscular lipids were extracted from Lt muscle as described by [26]. Fatty acids were transesterified according to[27], using sodium methoxide in methanol, followed by hydrochloric acid in methanol (1:1 v/v). Fatty acid methyl esters were analyzed using a Shimadzu GC2010Plus chromatograph (Shimadzu, Kyoto, Japan), equipped with a flame-ionization detector and fused silica capillary column (SP-2560 (100 m × 0.25 mm internal diameter × 0.20 μm film thickness, Supelco, Bellefonte, PA, USA)). The injector and detector temperatures were 250°C and 280°C, respectively. The initial oven temperature of 50°C was held for 1 min, increased at 50 °C/min to 150°C and held for 20 min, increased at 1°C/min to 190°C and then increased at 2°C/min to 220°C and held for 40 min. Helium was used as carrier gas at a flow rate of 1 ml/min and the split ratio was 50:1. Nonadecanoic acid (19:0) was used as internal standard to quantify muscle lipid FA methyl esters. Fatty acids were identified by comparison of the FAME retention times with those of authentic standards (FAME mix 37 components from Supelco Inc., Bellefont, PA, USA) and by comparison with published chromatograms [28, 29].
Calculations
The biohydrogenation completeness (BC) (%), reports the extent of the biohydrogenation of dietary c9-18:1, 18:2 n-6 and 18:3 n-3 and was estimated considering the maximum 18:0 content in rumen, assuming a complete biohydrogenation of the C18 FA from diet as proposed by [30]. The calculations were as follow:
where 18:0r, is the 18:0 in rumen as percentage of total C18 FA; c9,18:1d/r is the c9-18:1 in diet or rumen as percentage of total C18 FA; 18:2n6d/r: is the 18:2n−6 in diet or rumen as percentage of total C18 FA;18:3n3d/r is the 18:3n−3 in diet or rumen as percentage of total C18 FA; 18:0d is the 18:0 in diet as percentage of total C18 FA.
Statistical analysis
Initially, using a pool of the three experiments (n = 86 individual data), it was performed a global Spearman correlation analysis to test the existence of relationships between the concentrations (log10 cells/ml rumen liquor) of: 1) Total Ciliates; 2) Holotrichs (ciliates from the Order Vestibuliferida); 3) Entodiniomorphids (ciliates from de Order Entodiniomorphida) and 4) individual genera of ciliates that occurred in more than 15% of the rumen samples (Isotricha, Dasytricha, Entodinium, Polyplastron, Diplodinium and Epidinium) with: 1) FA proportions in rumen content; 2) rumen pH and 3) FA proportions in intramuscular fat (IMF) of Lt muscle. The Spearman rank order correlation coefficients were determined with STATISTICA®10software (StatSoft Inc., 2010).
In addition and in order to confirm if the relationships found with the correlation analysis were maintained when the experiment was considered in the model, a linear regression analysis was performed using the SAS MIXED procedure (SAS Institute Inc. Cary, NC). The relationships between concentration (as log10 cells/ml) of rumen ciliate protozoa (Total ciliates, Holotrichs, Entodiniomorphids and ciliate genera presented in more than 30% of rumen samples: Isotricha, Entodinium and Epidinium) and the FA in rumen and Lt muscle were evaluated with a linear model that included the concentration of ciliate protozoa as fixed effects and the experiment as a discrete random effect. An analysis of variance to evaluate the effect of the type of protozoa population in rumen and Lt muscle FA was performed, using the MIXED procedure of SAS, and when the effect of protozoa population was significant, the least square means were compared, using a pairwise Tukey’s comparison test. In statistical models, the variance was accommodated in the model using the covariance structure with the best fit to the data. Statistical significance was set for a level of P< 0.05, but regression parameters for models with 0.05≤P≤0.06 were also considered.
Results
Rumen ciliate protozoa densities
The individual results on total ciliate protozoa abundance and on ciliate genus relative abundance are presented in Figs 1 and 2, respectively. From the total of 86 lambs sampled in the study, 10 of them (n = 4 in Experiment 1, n = 3 in Experiment 2 and n = 3 in Experiment 3) did not present protozoa cells in the rumen liquor samples (0 cells/ml rumen liquor) and therefore were considered as defaunated. The remaining lambs presented, in average, 1.35 × 106 ciliates per ml of rumen fluid, and a large individual variability was observed among lambs (Fig 1).
Total density of ciliates (arithmetic mean and standard deviation) in the 86 lambs rumen samples, grouped by treatments and by the experiments.
Relative abundances for each ciliate genus identified in the 86 lambs rumen samples grouped by treatments and by experiments.
Nine ciliate genera were identified: three Holotrichs (Isotricha, Dasytricha and Charonina) and six Entodiniomorphids (Entodinium, Diplodinium, Polyplastron, Diploplastron, Epidinium and Ophryoscolex) (Fig 2). Entodiniomorphids were observed in the rumen of all faunated lambs (n = 76), and represented 95% of total ciliates present. Globally, the amilolytic genus Entodinium was the most abundant, averaging 1.17 × 106cells / ml (n = 71 lambs), followed by the celulolytic Epidinium (5.34 × 105cells / ml) (n = 23 lambs), and the mainly soluble sugars consumer Isotricha (1.30× 105cells / ml) (n = 36 lambs). The least abundant genera were Ophyroscolex (6.50 × 103 cells / ml) (n = 2 lambs) and Diplodinium (2.01× 104 cells / ml) (n = 14 lambs). The distribution of ciliates genera also presented a high variability, even for lambs fed the same diet (Fig 2).
Characterization of rumen variables and muscle FA
The mean, standard deviation, minimum and maximum values obtained in each one of the three experiments whose data was used and for the variables considered for correlation and regression analyses (rumen ciliate protozoa concentration, total FA content and FA proportions of rumen content, RBH completeness, rumen pH, FA content and FA proportions of Lt muscle) are presented in Tables 2–4 for Experiments 1, 2 and 3, respectively. Santos-Silva, Francisco (21) reported the results concerning the effect of the diets of Experiment 1 on Lt muscle fatty acids. The detailed effect of diets in rumen and muscle FA, in rumen pH and in RBH completeness for Experiments 2 and 3 will be published elsewhere.
Globally, it was observed a high variation for each phenotype among the three experiments and the 16 diets used. In Experiment 1, were observed the highest values for t10-18:1, t10,c12-18:2 and t10/t11 ratio and the lowest values for t11-18:1 and c9,t11-18:2. In Experiment 3 were observed the highest values for 18:0 and RBH completeness and the lowest values for 18:2 n-6, 18:3 n-3 and PUFA. In Experiment 2 were observed the highest values for rumen c9-18:1 and for rumen and muscle 18:3 n-3, and intermediate values for the other phenotypes. The lowest values for rumen pH were observed in Experiment 1 and the highest in Experiment 3.
Correlations between ciliate protozoa and rumen variables and longissimus thoracis muscle fatty acids
The significant Spearman´s rank correlations between ciliate protozoa concentration (as log10cells/ml), rumen FA and pH and, biohydrogenation completeness, performed with a pool of the data from the three experiments, are shown in Fig 3. Several moderate and strong correlations, between -0.71 and 0.55, were observed between protozoa concentration and FA composition of the whole rumen content and rumen pH. Total ciliates were mainly negatively correlated with the t10,c12- 18:2 (ρ = -0.71; P < 0.001), 18:2 BI (ρ = -0.65; P < 0.001), t10-18:1 (ρ = -0.64; P< 0.001) and t10/t11 ratio (ρ = -0.61; P< 0.001). Holotrichs showed the strongest negative correlations with t10,c12- 18:2 (ρ = -0.48; P < 0.001), 18:2 BI (ρ = -0.47; P < 0.001), 18:3 n-3 (ρ = -0.35; P = 0.002) and t10/t11 ratio (ρ = -0.31; P = 0.006) and Entodiniomorphids mainly negatively correlated with t10,c12- 18:2 (ρ = -0.68; P < 0.001), t10-18:1(ρ = -0.64; P < 0.001), 18:2 BI (ρ = -0.63; P < 0.001) and t10/t11 ratio (ρ = -0.62; P < 0.001). Concerning the ciliates individual genera, the strongest negative correlations were observed between Entodinium and t10/t11 ratio (ρ = -0.63; P < 0.001), t10-18:1 (ρ = -0.62; P < 0.001), t10,c12-18:2 (ρ = -0.60; P < 0.001), and 18:2 BI (ρ = -0.57; P < 0.001) and between Isotricha and t10,c12-18:2 (ρ = -0.56; P < 0.001).
The strongest Spearman´scorrelations are represented by large circles, while the weakest correlations are represented by small circles. The colors and their intensity on the bar scale denote the nature of the correlation, with the darker shade of blue indicating more positive correlations (close to 1) and the darker shade of red indicating more negative correlations (closer to -1) (n = 86). RBH, Rumen biohydrogenation; BI, Biohydrogenation intermediates; SFA, Saturated fatty acids; PUFA, Polyunsaturated fatty acids.
Total ciliates were mainly positively correlated with rumen pH (ρ = +0.55; P < 0.001) and with t11-18:1 (ρ = +0.38; P < 0.001). As observed for the total ciliates, both Holotrichs and Entodiniomorphids showed the highest positive correlations with rumen pH (ρ = +0.37; P < 0.001 and ρ = +0.52; P < 0.001, respectively) and with t11-18:1 (ρ = +0.25; P = 0.027 and ρ = +0.38; P< 0.001, respectively). In relation to ciliates individual genera, the strongest positive correlations were observed between Entodinium and rumen pH (ρ = +0.46; P < 0.001), Isotricha and rumen pH (ρ = +0.41; P < 0.001) and Entodinium and t11-18:1 (ρ = +0.40; P < 0.001).
Results concerning the correlation analysis among ciliates concentration in rumen and the FA present in Lt muscle of the 86 lambs are shown in Fig 4. In general, the correlations that were observed between protozoa and FA in rumen were also observed with the same FA in muscle, but with lower coefficients. Several moderate and strong correlations, between -0.67 and 0.56, were observed between rumen ciliates and the muscle FA in study. Total ciliates were mainly negatively correlated with t10,c12-18:2 (ρ = -0.68; P < 0.001), n-3 PUFA (ρ = -0.60; P < 0.001), 18:3 n-3 (ρ = -0.57; P < 0.001) and t10/t11 ratio (ρ = -0.45; P < 0.001). Holotrichs had the strongest negative correlations with and t10,c12- 18:2 (ρ = -0.42; P < 0.001), 18:3 n-3 (ρ = -0.35; P = 0.002) and n-3 PUFA (ρ = -0.30; P = 0.007) and Entodiniomorphids were mainly negatively correlated with t10,c12- 18:2 (ρ = -0.65; P < 0.001), n-3 PUFA (ρ = -0.58; P < 0.001), 18:3 n-3 (ρ = -0.54; P < 0.001) and t10/t11 ratio (ρ = -0.45; P < 0.001). Considering ciliates individual genera, the strongest negative correlations were observed between Entodinium and t10,c12- 18:2 (ρ = -0.61; P < 0.001), n-3 PUFA (ρ = -0.51; P < 0.001), t10/t11 ratio (ρ = -0.46; P < 0.001) and 18:3 n-3 (ρ = -0.45; P < 0.001), between Polyplastron and t10/t11 ratio (ρ = -0.40; P < 0.001) and Isotricha and 18:3 n-3 (ρ = -0.40; P< 0.001) and t10,c12-18:2 (ρ = -0.37; P < 0.001). The strongest positive correlations between total ciliates and muscle FA were observed with t11-18:1 (ρ = +0.55; P < 0.001) and with c9,t11-18:2 (ρ = +0.55; P < 0.001). Also for Entodiomorphids these were the main positive correlations observed (ρ = +0.56; P < 0.001, for both t11-18:1 and c9,t11-18:2). For Holotrichs, only a significant and positive correlation was found, and it was between the individual genus, Isotricha and n-6 LC-PUFA (ρ = +0.25; P = 0.029). Entodinium was the ciliate genus that strongly correlated with the muscle FA in study. The highest positive correlations were observed between Entodinium and c9,t11-18:2 (ρ = +0.56; P < 0.001) and t11-18:1 (ρ = +0.55; P < 0.001).
The strongest Spearman´s correlations are represented by large circles, while the weakest correlations are represented by small circles. The colors and their intensity on the bar scale denote the nature of the correlation, with the darker shade of blue indicating more positive correlations (close to 1) and the darker shade of red indicating more negative correlations (closer to -1) (n = 86). SFA, Saturated fatty acids; cis-MUFA, cis monosaturated fatty acid; PUFA, Polyunsaturated fatty acids; n-6 PUFA, 18:2 n-6 + 20:4 n-6 + 22:4 n-6;LC n-6 PUFA, 20:4 n-6 + 22:4 n-6; n-3 PUFA,18:3 n-3 + 20:5 n-3 + 22:5 n-3 + 22:6 n-3; LC n-3 PUFA, 20:5 n-3 + 22:5 n-3 + 22:6 n-3.
Regression analysis between ciliates concentration, rumen variables and meat fatty acids
Tables 5 and 6 show the regression coefficients with significant effect for rumen ciliates and rumen and meat FA, obtained with models that included the experiment as a random effect. In rumen (Table 5), the 18:0 (P = 0.015), the t10-18:1 (P = 0.058), the t10/t11 ratio (P = 0.002) and the 18:2 BI (P = 0.021) decreased linearly with the concentration of Total ciliates. Stearic acid (18:0) (P = 0.020), 18:2 BI (P = 0.002), SFA (P = 0.011) and RBH completeness (P = 0.029) decreased linearly with Holotrichs and 18:0 (P = 0.010), t10-18:1 (P = 0.062), t10/t11 ratio (P < 0.001) and RBH completeness (P = 0.029) decrease linearly with Entodiniomorphids. At the genus level, it were also observed linear decreases of 18:0 (P = 0.014), 18:2 BI (P = 0.016), SFA (P = 0.054) and RBH completeness (P = 0.011) with Isotricha, of t10/t11 ratio (P < 0.001) with Entodinium and of Total FA (P = 0.020), c9-18:1 (P < 0.001), t10-18:1 (P = 0.062), t10/t11 ratio (P < 0.001), 18:2 BI (P = 0.016) and RBH completeness (P = 0.027) with Epidinium. Linear increases of the t11-18:1 (P <0.001) and SFA (P <0.001) were observed for Total ciliates in rumen. The t11-18:1 also increased linearly with Entodiniomorphids (P = 0.011), Isotricha (P <0.001), Entodinium (P = 0.018) and Epidinium (P = 0.023). Linear increases of c9,t11-18:2 with Isotricha (P = 0.041), of 18:1 BI with Isotricha (P = 0.017), Entodinium (P = 0.003) and Epidinium (P = 0.004) and of rumen pH with Holotrichs (P = 0.025) were also observed.
In meat (Table 6), the c9-18:1(P = 0.040), the 18:2 n-6 (P = 0.001), the t10/t11 ratio (P = 0.036), the n-3 PUFA (P <0.001), the n-3 LC-PUFA (P = 0.024) and the PUFA (P = 0.033) decreased linearly with the concentration of Total ciliates. Oleic acid (c9-18:1) (P = 0.035), 18:3 n-3 (P = 0.001), SFA (P = 0.046), n-3 PUFA (P < 0.001) and n-6 PUFA (P = 0.005) decreased linearly with Holotrichs. Stearic acid (P = 0.005), c9-18:1(P = 0.009), 18:2 n-6 (P < 0.001), n-3 PUFA (P <0.001), n-6 PUFA (P = 0.009), n-3 LC-PUFA (P = 0.024) and PUFA (P < 0.001) decreased linearly with Entodiniomorphids. At the genus level, it were also observed linear decreases of c9-18:1(P = 0.014) and SFA (P = 0.054) with Isotricha, of c9-18:1 (P = 0.011), 18:2 n-6 (P = 0.003), n-3 PUFA (P <0.001), n-6 PUFA (P = 0.019), n-3 LC-PUFA (P = 0.054) and PUFA (P = 0.021) with Entodinium and of c9-18:1(P < 0.001) with Epidinium. Vaccenic acid (t11-18:1) (P <0.001) and of the c9,t11-18:2 (P <0.001) increased linearly with Total ciliates, Entodiniomorphids and Entodinium.
Effect of the type of protozoa population on rumen and meat FA
The results of the effect of the type of the protozoa population on the FA content of rumen are presented in Table 7. The type of population influenced the individual proportion of 18:0 (P = 0.003), c9-18:1 (P = 0.004), 18:3 n-3 (P< 0.001), t10-18:1 (P = 0.024) and the sum of SFA (P = 0.010). The A type population resulted in the lowest value for 18:0, low proportions of t10-18:1 and SFA and in the highest value for 18:3 n-3 proportion and high c9-18:1. The B type population resulted in the lowest proportion of 18:3 n-3, intermediate proportions of c9-18:1, t10-18:1 and SFA and high proportion of 18:0. The O type population resulted in the low proportion of c9-18:1, intermediate proportion of 18:3 n-3 and high value for 18:0, t10-18:1 and SFA proportions.
Table 8 shows the results of the effect of the type of ciliates population on the FA of Lt muscle. The type of population influenced the individual proportions of 18:3 n-3 (P = 0.009), t10-18:1 (P = 0.009), the t10/t11 ratio (P = 0.006) and the sums of SFA (P< 0.001) and of n-3 PUFA (P = 0.011). The A type population resulted in the lowest value for t10-18:1 and low values of t10/t11 ratio and n-3 PUFA, in the highest SFA and in a high proportion of 18:3 n-3. The B type population, resulted in the lowest proportion of 18:3 n-3, low SFA and n-3 PUFA proportions, intermediate t10/t11 ratio and a high proportion of t10-18:1. The O type population resulted in low proportion of SFA, in the highest value for n-3 PUFA and in high 18:3 n-3, t10-18:1 and t10/t11 ratio in muscle.
Discussion
Rumen ciliate protozoa are usually present on rumen at numbers ranging between 104 to 106 cells/ml [14] but host diet strongly affects their numbers which can range from zero (defaunation) up to 5 × 106 cells/ ml [31]. Dietary lipids rich in PUFA or in lauric acid, excerpt toxic effects on rumen protozoa, and may be used as supplement in ruminant diets to reduce methane emissions [7, 8, 32, 33] or as promoters of t11-18:1 and c9,t11-18:2 in ruminant products [1]. All the diets used in the three experiments were supplemented with 6% of soybean oil. Despite this, 88.4% of the lambs were faunated, with protozoa numbers ranging between 9.0 × 103cells/ ml and 1.2 × 107cells/ml, which suggest that in some samples the concentration of protozoa was even above the upper value referred by [14].
The abundance of ciliate protozoa is highly influenced by rumen pH [34, 35] and values below 5.5 above 15 h/day generally cause the elimination of protozoa [34]. In the present study, we have no data regarding the daily fluctuations of rumen pH. The rumen pH, was determined only once, immediately after the slaughter and the lambs were not submitted to pre-slaughter fasting. The values ranged between 5.25 and 7.67and correlated positively with rumen ciliates, which is in line with the literature [34, 35].
Usually 5 to 6 genera of rumen ciliates are observed in each faunated rumen [36] but in this study most cases (n = 67) only presented 1 to 3 ciliate genera in rumen. Host diet plays a crucial role not only in the abundance but also in the community structure of rumen protozoa [31, 37, 38]. The ciliate genera composition observed probably reflects the balance between starch, soluble sugars and structural carbohydrates available in the rumen. Our data also show a strong variability among individuals, which is in accordance to [39] and [38]. A high variability among lambs was observed for both the concentration of ciliates and for the genera composition and frequently, animals fed the same diet presented different ciliate densities and community structures.
The relationships between rumen protozoa and lipid metabolism in rumen are still not fully clarified [8]. However, independently of the mechanisms involved, our results suggest that rumen ciliates may modulate the RBH, by favouring its first steps and reducing the last step, which results in the accumulation 18:1-BI, as it was suggested by[18]. In fact, the present results show negative relationships between the concentration of ciliates and the proportions of PUFA, 18:2-BI, 18:0 and RBH completeness in rumen and positive relationships between Isotricha, Entodinium and Epidinium genera and 18:1-BI in rumen. Moreover, also the c9-18:1 in muscle decreased linearly with the protozoa concentration, independently of the ciliates genera. In lipid supplemented ruminants, the c9-18:1 is derived mostly from endogenous desaturation of rumen derived 18:0 [1], thus the negative relationship of muscle c9-18:1 with protozoa is consistent with a reduced rumen outflow of 18:0.
Rumen concentration and muscle deposition of t11-18:1 and c9,t11-18:2 increased linearly. These results confirm the previous observations that faunated lambs tended to have higher proportion of t11-18:1 in the muscle when compared to the defaunated lambs [18]. Rumen protozoa accumulate high levels of the t11-18:1 and c9,t11-18:2 in their membranes [19, 20, 40, 41] and its association with t11-18:1 and c9,t11-18:2 in meat and milk was hypothesized [18, 40, 42]. Our results are in line with that hypothesis, showing a clear positive association between ciliates concentration in rumen and the deposition of t11-18:1 and c9,t11-18:2 in meat.
The incorporation into the membranes of the ciliate protozoa of both t11-18:1 and c9, t11-18:2, allows them to escape to the subsequent hydrogenation, increasing the amount of those AF flowing from the rumen, becoming available for absorption and deposition in the tissues animals. The amount and rate of biomass protozoa flow from rumen is being under intense debate [8], due to the selective retention of these microorganisms within the rumen[42–44]. It seems that the retention of ciliates in rumen is not similar for every protozoa group[8]. While Holotrichs have a strong chemotaxis toward sugars and after a meal migrate to the ventral reticulo-rumen, avoiding to be washed out, Entodiniomorphids have a moderate chemotaxis toward glucose and peptides and are uniformly distributed in ruminal content do not show the same affinity to the rumen wall as Holotrichs, and thus flow out of the rumen with the particulate matter [8, 45, 46]. The results of the present study are compatible with that dynamics of protozoa in rumen. The strong positive relationships between the Entodiniomorphids, namely the genus Entodinium, and t11-18:1 and c9,t11-18:2 in meat suggest a high availability of t11-18:1 for duodenal absorption and deposition in tissues. The positive relationships that were observed between Holotrichs and t11-18:1 and c9,t11-18:2 in rumen, were not maintained for meat.
Another mechanism that may contribute to explain the relationships among ciliates and t11-18:1 and c9,t11-18:2 is their effect as stabilizing agents of rumen fermentation of non-structural carbohydrates, contributing to the maintenance of rumen pH values above 6.0, which are compatible to the growth and activity of the cellulolytic biohydrogenating bacteria [2]. Holotrichs metabolize preferentially soluble readily available carbohydrates [47], which absorb and accumulate within the cell, preventing its rapid fermentation by bacteria [9, 47]. Our results suggest that genus Isotricha may be particularly relevant in this context because it was the only for whom linear regression was significant. Moreover, Entodiniomorphids, particularly from Entodinium genus, engulf and metabolize particulate matter, and are responsible by the removal of large amounts of starch and feed particles from rumen liquor [9]. Due to their fast up take of starch granules, they reduce the availability of substrate to the amilolytic bacteria, allowing a better control of starch fermentation and reducing rumen pH fluctuations [9].
As far as we know, this is the first study reporting the relationships between rumen protozoa and the occurrence of the trans-10 shifted RBH pathway evaluated by the t10-18:1 and t10/t11 ratio in rumen content and in meat. The proportion of t10-18:1 and the t10/t11 ratio showed negative and significant correlations with total ciliates and total Entodinomorphids and the regressions analysis reinforced these results. Considering individual genera, the correlations with t10-18:1 and with the t10/t11 ratio were also generally negative. However, when linear regression analysis was applied to rumen data, only for Entodinium and Epidinium the regression coefficients were significant. These results suggest that, in the set of genera considered, these two Entodinomorphids are the most associated to the reduction of the trans-10 shifted pathway in the rumen.
In addition, the ANOVA results suggest that the presence of the protozoa type A population is related to a low t10-18:1 content and to a t10/t11 ratio in the rumen and in meat. The Type A protozoa population is characterized by the presence of the genus Polyplastron, which has a large cell size and a high predatory and digestive activities for most carbohydrates [48]. These characteristics may contribute to impair rumen the conditions that favour the t10 shifted pathway.
This study was performed with data from three previous experiments of our group and was not specifically designed to evaluate the effects of protozoa in lipid rumen metabolism. However, our results suggest a close relationship between the concentration of ciliate protozoa in rumen and the accumulation of 18:1 BI in rumen and meat, namely t11-18:1 and reducing the probability and intensity of the trans-10 shift in rumen, what positively impacts the nutritional value of ruminant´s meat. The effects of diet composition on rumen ciliates concentration and community structure should be considered when nutritional strategies are planned to improve the nutritional value of ruminant products. More research is needed to clarify and to explore the relationships observed between ciliate community and RBH pathways and completeness, including other genera that were not explored in this study.
Conclusions
The relationships between protozoa community and lipid rumen metabolism was associated with the type of ciliates and Entodinomorphids were more linked with RBH than Holotrichs.
Rumen protozoa were positively linked with t11-18:1 and c9,t11-18:2 and negatively related with the t10-18:1, with trans-10 shift, with RBH completeness and with 18:0.
The concentration of ciliate protozoa in rumen was positively related to the t11-biohydrogenation pathway and to a healthier fatty acid profile in lamb meat.
Supporting information
S1 Table. Ingredients, chemical composition and fatty acid (FA) profile of the experimental diets from the Experiment 1.
https://doi.org/10.1371/journal.pone.0221996.s001
(PDF)
S2 Table. Ingredients, chemical composition and fatty acid (FA) profile of the experimental diets from the Experiment 2.
AP0, alfalfa pellets and 0% barley grain; AP35, alfalfa and 35% of barley grain; AP65, alfalfa pellets and 65% of barley grain; AP100, alfalfa pellets and 100% of barley grain; AH0, alfalfa hay and 0% of barley grain; AH35, alfalfa hay and 35% of barley grain; AH65, alfalfa hay and 65% of barley grain; AH100, alfalfa hay and 100% of barley grain.
https://doi.org/10.1371/journal.pone.0221996.s002
(PDF)
S3 Table. Ingredients, chemical composition and fatty acid (FA) profile of the experimental diets from the Experiment 3.
A0.5, alfalfa as forage source and 0.5% of SB as buffer; A2.0, alfalfa as forage source and 2.0% of SB as buffer; ALV, alfalfa as forage source and 0.5% of SB plus 1.5% Levucell® SC20 as buffer; R0.5 –ryegrass as forage source and 0.5% of SB as buffer; R2.0, ryegrass as forage source and 2.0% of SB as buffer.
https://doi.org/10.1371/journal.pone.0221996.s003
(PDF)
References
- 1. Bessa RJB, Alves SP, Santos-Silva J. Constraints and potentials for the nutritional modulation of the fatty acid composition of ruminant meat. Europ J Lipid Sci Technol. 2015;117:1325–44.
- 2. Jenkins TC, Wallace RJ, Moate PJ, Mosley EE. Board-invited review: Recent advances in biohydrogenation of unsaturated fatty acids within the rumen microbial ecosystem. J Anim Sci. 2008;86(2):397–412. Epub 2007/11/29. pmid:18042812.
- 3. Martin SA, Jenkins TC. Factors affecting conjugated linoleic acid and trans-C18:1 fatty acid production by mixed ruminal bacteria. J Anim Sci. 2002 80(12):3347–52. pmid:12542176
- 4.
Griinari JM, Bauman DE. Biosynthesis of conjugated linoleic acid and its incorporation into meat and milk in ruminants. In: Yuraweczmp MP, Mossoba MM, Kramer JKG, Pariza MW, Nelson G, editors. Advances in conjugated linileic acid research. Champagin, IL: American Oil Chemists Society Press; 1999. p. 180–200.
- 5. Aldai N, Renobales M, Barron LJR, Kramer JKG. What are the trans fatty acids issues in foods after discontinuation of industrially produced trans fats? Ruminant products, vegetable oils, and synthetic supplements. Europ J Lipid Sci Technol. 2013;115:1378–401.
- 6. Shingfield KJ, Griinari JM. Role of biohydrogenation intermediates in milk fat depression. Europ J Lipid Sci Technol. 2007;109(8):799–816.
- 7. Hook SE, Wright A-DG, McBride BW. Methanogens: Methane Producers of the Rumen and Mitigation Strategies. Archaea. 2010;2010:11. pmid:21253540
- 8. Newbold CJ, de la Fuente G, Belanche A, Ramos-Morales E, McEwan NR. The Role of Ciliate Protozoa in the Rumen. Frontiers in Microbiology. 2015;6:1313.
- 9.
Williams AG, Coleman GS. The rumen protozoa. In: Hobsen PN, Stewart CS, editors. The rumen microbial microsystem. Dorbrecht, Netherlands: Springer; 1997.
- 10. Nagaraja TG, Titgemeyer EC. Ruminal Acidosis in Beef Cattle: The Current Microbiological and Nutritional Outlook. J Dairy Sci. 2007;90:E17–E38. https://doi.org/10.3168/jds.2006-478. pmid:17517750
- 11. Noziere P, Ortigues-Marty I, Loncke C, Sauvant D. Carbohydrate quantitative digestion and absorption in ruminants: from feed starch and fibre to nutrients available for tissues. Animal. 2010;4(7):1057–74. Epub 2010/07/01. pmid:22444609.
- 12. Newbold CJ, Williams AG, Chamberlain DG. The in vitro metabolism of d,l-lactic acid by rumen microorganisms. Journal of Science and Food Agriculture. 1987;38(1):9–18.
- 13.
Harfoot CG, Hazelwood GP. Lipid metabolism in the rumen. In: Hobson PN, editor. The rumen microbial ecosystem. London, UK: Elsevier Science Publishing; 1997. p. 382–426.
- 14.
Nagaraja TG. Microbiology of the rumen. In: Millen D, De Beni Arrigoni M, Lauritano Pacheco RD, editors. Ruminology. Switzerland: Springer International Publish.; 2016. p. 39–61.
- 15. Huws SA, Kim EJ, Kingston-Smith AH, Lee MR, Muetzel SM, Cookson AR, et al. Rumen protozoa are rich in polyunsaturated fatty acids due to the ingestion of chloroplasts. FEMS microbiology ecology. 2009;69(3):461–71. Epub 2009/07/09. pmid:19583786.
- 16. Boeckaert C, Morgavi DP, Jouany JP, Maignien L, Boon N, Fievez V. Role of the protozoan Isotricha prostoma, liquid-, and solid-associated bacteria in rumen biohydrogenation of linoleic acid. Animal. 2009;3(7):961–71. Epub 2009/07/01. pmid:22444816.
- 17. Kim EJ, Huws SA, Lee MRF, Scollan ND. Dietary Transformation of Lipid in the Rumen Microbial Ecosystem. Asian-Australasian Journal of Animal Sciences. 2009;22(9):1341–50.
- 18. Yanez-Ruiz DR, Williams S, Newbold CJ. The effect of absence of protozoa on rumen biohydrogenation and the fatty acid composition of lamb muscle. British Journal of Nutrition. 2007;97(5):938–48. Epub 2007/03/27. pmid:17381978.
- 19. Devillard E, McIntosh FM, Newbold CJ, Wallace RJ. Rumen ciliate protozoa contain high concentrations of conjugated linoleic acids and vaccenic acid, yet do not hydrogenate linoleic acid or desaturate stearic acid. British Journal of Nutrition. 2007;96(4):697–704. Epub 03/08.
- 20. Sultana H, Miyazawa K, Kanda S, Itabashi H. Fatty acid composition of ruminal bacteria and protozoa, and effect of defaunation on fatty acid profile in the rumen with special reference to conjugated linoleic acid in cattle. Animal Science Journal. 2011;82(3):434–40. Epub 2011/05/28. pmid:21615837.
- 21. Santos-Silva J, Francisco A, Alves SP, Portugal APV, Dentinho MT, Almeida AM, et al. Effect of dietary neutral detergent fiber source on lambs growth, meat quality and biohydrogenation intermediates. Meat Sci. 2019;147:28–36. pmid:30196198
- 22. Francisco A, Alves SP, Portugal PV, Pires VMR, Dentinho MT, Alfaia C, et al. Effect of feeding lambs with a tanniferous shrub (rockrose) and a vegetable oil blend on fatty acid composition of meat lipids. Animal. 2016:1–13. pmid:27306827
- 23. Dehority BA. Evaluation of subsampling and fixation procedures used for counting rumen protozoa. Applied and Environment Microbiology. 1984;48(1):182–5.
- 24.
Hogimoto K, Imai S. Atlas of rumen microbiology. Tokyo: Japan Scientific Societies Press; 1981.
- 25. Alves SP, Santos-Silva J, Cabrita ARJ, Fonseca AJM, Bessa RJB. Detailed Dimethylacetal and Fatty Acid Composition of Rumen Content from Lambs Fed Lucerne or Concentrate Supplemented with Soybean Oil. PLOS ONE. 2013;8(3):e58386. pmid:23484024
- 26. Jerónimo E, Alves SP, Prates JAM, Santos-Silva J, Bessa RJB. Effect of dietary replacement of sunflower oil with linseed oil on intramuscular fatty acids of lamb meat. Meat Sci. 2009;83:499–505. pmid:20416670.
- 27. Alves SP, Raundrup K, Cabo Â, Bessa RJB, Almeida AM. Fatty Acid Composition of Muscle, Adipose Tissue and Liver from Muskoxen (Ovibos moschatus) Living in West Greenland. PLOS ONE. 2015;10(12):e0145241. pmid:26678792
- 28. Alves SP, Bessa RJB. Comparison of two gas-liquid chromatograph columns for the analysis of fatty acids in ruminant meat. Journal of Chromatography A,. 2009;1216::5130–9. pmid:19446820
- 29. Vahmani P, Rolland DC, Gzy IKE, Dugan MER. Non-conjugated cis/trans 18:2 in Beef Fat are Mainly Δ-9 Desaturation Products of trans-18:1 Isomers. Lipids,. 2016;51:1427–33 pmid:27853932
- 30. Alves SP, Francisco A, Costa M, Santos-Silva J, Bessa RJB. Biohydrogenation patterns in digestive contents and plasma of lambs fed increasing levels of a tanniferous bush (Cistus ladanifer L.) and vegetable oils. Anim Feed Sci Technol. 2017;225:157–72.
- 31.
Church DC. Digestive Physiology and Nutrition of Ruminants: Nutrition: O & B Books; 1979. 452 p.
- 32. Guyader J, Eugene M, Noziere P, Morgavi DP, Doreau M, Martin C. Influence of rumen protozoa on methane emission in ruminants: a meta-analysis approach. Animal. 2014;8(11):1816–25. Epub 2014/07/31. pmid:25075950.
- 33. Toprak NN. Due fats reduce methane emission by ruminants. Animal Science Papers and Reports. 2015;33(4):305–21.
- 34. Franzolin R, Dehority BA. Effects of ruminal PH and feed intake on defaunation in sheep fed high concentrate diets. Rev Bras Zootecn. 1996;25(6):1207–15.
- 35. Dehority BA. Effect of pH on viability of Entodinium caudatum, Entodinium exiguum, Epidinium caudatum, and Ophryoscolex purkynjei in vitro. Journal of Eukaryotic Microbiology. 2005;52(4):339–42. Epub 2005/07/15. pmid:16014011.
- 36.
Bohatier J. The rumen protozoa: taxonomy, cytology and feeding behaviour. In: Jouany J, editor. Rumen microbial metabolism and ruminant digestion. Paris, France: INRA; 1991. p. 27–38.
- 37. Coleman GS. The role of rumen protozoa in the metabolism of ruminants given tropical feeds. Tropical Animal Production. 1979;4(3):199–213.
- 38. Henderson G, Cox F, Ganesh S, Jonker A, Young W. Rumen microbial community composition varies with diet and host, but a core microbiome is found across a wide geographical range. Scientific Reports [Internet]. 2015; 5:[14567 p.]. pmid:26449758
- 39. Weimer PJ. Redundancy, resilience, and host specificity of the ruminal microbiota: implications for engineering improved ruminal fermentations. Frontiers of Microbiology. 2015;6:296. Epub 2015/04/29. pmid:25914693.
- 40. Or-Rashid MM, Odongo NE, McBride BW. Fatty acid composition of ruminal bacteria and protozoa, with emphasis on conjugated linoleic acid, vaccenic acid, and odd-chain and branched-chain fatty acids. J Anim Sci. 2007;85(5):1228–34. Epub 2006/12/06. pmid:17145972.
- 41. Cièslak A, Miltko R, Belzecki G, Szumacher-Strabel M, Michalowski T. Rumen Ciliates Entodinium caudatum, Eudiplodinium maggii and Diploplastron affine: a Potential Reservoir of Unsaturated Fatty Acids for the Host. Acta Protozoologica 2009;48 (4):335–40.
- 42. Huws SA, Lee MR, Kingston-Smith AH, Kim EJ, Scott MB, Tweed JK, et al. Ruminal protozoal contribution to the duodenal flow of fatty acids following feeding of steers on forages differing in chloroplast content. British Journal of Nutrition. 2012;108(12):2207–14. Epub 2012/03/02. pmid:22377337.
- 43. Dehority BA, Tirabasso PA. Factors Affecting the Migration and Sequestration of Rumen Protozoa in the Family Isotrichidae Journal of General Microbiology. 1989;135:539–48.
- 44. Hook SE, Dijkstra J, Wright AD, McBride BW, France J. Modeling the distribution of ciliate protozoa in the reticulo-rumen using linear programming. J Dairy Sci. 2012;95(1):255–65. Epub 2011/12/24. pmid:22192205.
- 45. Diaz HL, Karnati SKR, Lyons MA, Dehority BA, Firkins JL. Chemotaxis toward carbohydrates and peptides by mixed ruminal protozoa when fed, fasted, or incubated with polyunsaturated fatty acids. J Dairy Sci. 2014;97(4):2231–43. https://doi.org/10.3168/jds.2013-7428. pmid:24534499
- 46. Hook SE, France J, Dijkstra J. Further assessment of the protozoal contribution to the nutrition of the ruminant animal. Journal Theoretical Biology. 2017;416:8–15. Epub 2016/12/23. pmid:28007554.
- 47. Williams AG. Rumen holotrich ciliate protozoa. Microbiological Reviews. 1986;50(1):25–49. pmid:3083220
- 48. Jouany JP, Zainab B, Senaud J, Groliere CA, Grain J, Thivend P. Rôle of the rumen ciliate protozoa Polyplastron multivesiculatum, Entodinium sp. and Isotricha prostoma in the digestion of a mixed diet in sheep. Reprod Nutr Dev. 21(6A):871–84. pmid:6818640