Physical Form of Diet Influence the Liver Function, Blood Biochemistry, and External Body Measurements in Broiler Chickens Exposed to Carbon Tetrachloride Toxicity

Introduction

With the expansion of the commercial broiler chicken industry, metabolic disorders have become emerging serious hazards in bird’s health and profitability (Savard et al., 2013). Fatty liver, among many other metabolic syndromes, has not received adequate attention in broiler research while it plays a significant role in the pathogenesis of many liver and kidney diseases and may directly cause low performance and a great economic loss (Zhang et al., 2018). It is well known that excessive lipid accumulation in the liver is promoted by enhanced mobilization of fatty acids from fat depots, de novo synthesis of fatty acids, dietary triglycerides, and a ceased fatty acid oxidation due to unfair influx vs. removal of lipids (Hong et al.,2019). Such biochemical facts are almost all anticipated in broiler flocks where birds are virtually raised with free access to high energy diets in confined circumstances without adequate physical activity. More complications arise when birds are maintained on diets contaminated with mycotoxins or other environmental contaminants (Murugesan et al., 2015; Mokubedi et al., 2019).

Carbon tetrachloride as an environmental toxicant has frequently been used as a promising method by several in vitro and in vivo experiments to study animal toxicity (Manibusan et al., 2007; Beheshti Moghaddam et al., 2016; Vahed et al., 2016). Toxicity of CCl₄realizes through its biotransformation by the cytochrome P₄₅₀ in the endoplasmic reticulum of the hepatocytes to produce free radicals dominated by trichloromethyl (CCl₃•) (Zhang et al., 2018). Moreover, Baradaran et al. (2019) reported remarkable down-regulation in the expression of CAT, GPx, and Mn-SOD hepatic genes in CCl₄-challenged broiler chicks.

Unfortunately, birds suffering fatty liver, irrespective of the relevant risk factors and etiology, are not diagnosed before clinical symptoms appear. Therefore, management of the promoting environmental and nutritional factors such as feed particle size (Shirani et al., 2018), feed physical form (Mohammadi Ghasem Abadi, et al. 2019), as well as contamination of diet with toxicants (Baradaran et al., 2019), must receive priority for the prevention of fatty liver in broiler farms. With the short lifetime of broiler chicken, appropriate non-invasive indications for diagnosis of birds suffering from the fatty liver are welcomed in broiler research and veterinary practice. A liver fat content beyond five percent is considered as fatty liver (Bedogni et al., 2014; Nicolas, 2019). However, access to perfect data on liver fat using noninvasive methods is not quite incredible but by autopsy of the expired birds. In the current study, we tried to explore the association between many external body dimensions as well as blood biochemical parameters with liver fat percentage hopeful to find a multivariate linear model to predict liver fat percentage in liver broiler chickens. Therefore, this study intended to investigate the effects of feed form on liver function, blood constituents, and certain external body measurements in broiler chickens exposed to CCl₄ toxicity and to provide multivariate linear models for prediction of liver fat percentage using the evaluated parameters.

Materials and methods

Animals and diets

Four hundred sixty-eight 10-day old female Arbor Acres broiler chicks at the highest proximity of size were chosen and used to examine the effects of six treatments in a 2 × 3 factorial arrangement with a completely randomized block design in six replicates of 13 birds each. The birds were selected from a flock consisting of 2000 straight run chicks which provided from a local hatchery and housed in a power-ventilated grow-out house where raised in a floor pen up to day 10 of age. During this early period, birds were grown on a pelleted starter diet (Table 1) and water ad libitum under a 23:1 light to darkness lightening regimen. All procedures carried out in this experiment were reviewed and approved by the Animal Care and Use Committee of Lorestan University, Khorramabad, Iran. The ambient temperature and relative humidity were kept at 32±1°C and 60±5%, respectively.

At the initiation of day 11, the selected birds were distributed into 36-floor pens (0.9 × 1.8 m) where they spent three days for acclimatization and then subjected to the experimental treatments up to day 42 of age. Treatments consisted of a grower diet (Table 1) presented in three physical forms (mash, crumble, and pelleted) and fed to birds with or without CCl₄ injection. For CCl₄ intoxication, birds were intraperitoneally injected with 0.5, 0.5, and 0.75 mL/kg body weight CCl₄ in olive oil at a ratio of 1: 1, v/v in days 14, 21, and 28 of age, respectively. The pelleting process was accomplished at a temperature of 90°C. The primary mash diet was pelleted and then pellets were crumbled in a roller mill, resulting in a crumbled diet. During the experimentation period, days 11 to 42, the chicks received a grower diet and water ad-libitum (Table 1).

Physical Form of Diet Influence the Liver Function, Blood Biochemistry, and External Body Measurements in Broiler Chickens Exposed to Carbon Tetrachloride Toxicity - Image 1

Performance data

Live body weight (BW) and feed intake (FI) were recorded in days 14 and 42 of age and data were utilized to generate daily weight gain (DWG), average daily feed intake (DFI), and feed conversion ratio (FCR). Mortality was recorded through the experiment upon occurrence. The European production efficiency index (EPEI) was calculated based on the method provided by Euribrid (1994).

EPEI (%) = (Body weight (Kg) × livability (%)) / (Age (days) × feed conversion ratio) ×100

Blood constituents

At the end of the experiment (Day 42), all the birds were weighed, killed by puncturing the jaguar veins and carotid arteries, scalded, de-feathered mechanically, and eviscerated manually. Individual samples of whole uncoagulated blood were collected and centrifuged at 1800 ×g for 15 min. The collected sera preserved at -20°C pending analysis. Concentrations of serum glucose (GLU), albumin (ALB), triglycerides (TG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-c), total protein (TP), total bilirubin (TBIL), direct bilirubin (DBIL), and the activity of serum alkaline phosphatase (ALP), aspartate aminotransferase (AST), Alanine aminotransferase (ALT) and lactate dehydrogenase (LDH) were determined using an auto-analyzer machine (Clima Ral. Co, Barcelona, Spain). The analyzer employed enzymatic procedures using SEPPIM Diagnostic Kits (SEPPIM S.A.S., Sees, France) based on the method reported by Elliott (1984).

Liver fat and liver health scoring

Livers from all the slaughtered birds were weighed and then macroscopically appraised for color and apparent health. Liver scores on a 4-point scale were assigned as follows: the most severe lesion and color alteration was given score 3, where normal = 0, as described by Trott et al. (2014) with a slight modification. After scoring, the liver samples kept at -4°C pending fat extraction based on the method devised by Folch et al. (1957). Briefly, about 1 g from each liver tissue weighed, added to chloroform/methanol (2/1) to a final volume of 20-times the volume of the tissue sample vortexed for one minute and allowed to stand with agitation for 2 h. The separated liquid was filtered through Whatman #1 filter paper into a 100-mL 54 graduated cylinder, and 5 mL of 7.3% Potassium chloride solution was added and blended. After phase separation, the top layer was carefully drained off. Total lipids are determined gravimetrically after evaporating the solvent. The samples were then dried, weighed, and total lipid weight was expressed as the percentage of liver fat against the total liver weight.

External body dimensions

Shank length (ShL), shank width (ShW), breast length (BrL), and breast width (BrW) were recorded using a digital caliper at the accuracy of 0.01 mm. Breast girth (BrG) was measured using a tape measure at the most plumpy and fleshy point at the accuracy of 0.5 cm. The breast angle (BrA) was determined using a breast angle meter at the accuracy of 1 degree. All measurements were taken at day 42 of age according to the procedure adopted by Khosravinia et al. (2006).

Fear test data

The tonic immobility reaction (fear test) was induced in all birds during day 40 of age based on the method described by Ghareeb et al. (2014). Briefly, an individual bird was placed on its back and restrained with one hand on its sternum for 45 seconds while its head and neck were kept with the other hand. Towards the end of the holding period, hand pressure was slowly lifted and the duration of tonic immobility reaction, as the time needed by the bird until righting and standing was recorded using a stopwatch.

Statistical analysis

Performance and blood-related data were subjected to a two-way analysis of variance using the GLM procedures of SAS 9.1 (SAS Institute, 2003). The following statistical model has used the effects of feed presentation form (mash, crumble, and pellet), CCl₄ injection, and their interactions.

Y_ijkl=µ+FF_i+TC_j+(FF×TC)_ij+B_k+ε_ijkl

where Y_ijkl is the variance associated with parameter; μ is the overall mean; FF_i is the feed physical form effect; TC_j is the CCl₄ administration effect; (FF×TC)_ij is the interaction effect; and e_ijkl is the experimental error effect. Multiple treatment comparisons were performed using the Tukey test (Kramer, 1956). The scores assigned to each bird for apparent liver health were statistically analyzed using PROC FERQ in the same statistical analysis software (SAS Institute, 2003) for their frequency. For all tests, the maximum likelihood for type-I error was considered at 5% (P < 0.05). Logistic regression analyses were undertaken to identify the best combination of parameters for predicting liver fat percentage using the forward, backward, and stepwise approaches in SAS modeling options. Forward and Backward elimination option with a threshold of P= 0.05 was adopted to pick up the covariates for the final model. We incorporated 17 predictor variables in each binary logistic regression analysis model. The body measurements including ShL, ShW, BrL, BrW, BrG, and BrA as well as a fear test (FT) and blood parameters including TG, TC, LDL, TP, TBIL, ALT, AST, LDH, ALB, TBIL, and DBIL were analyzed using multiple linear regression techniques (Draper and Smith, 1981).

Results

Mean DWG in the broilers fed on the pelleted diet increased by 7.03 g (19.08%) and 3.38 g (9.17%) compared with those received crumble and mash diets, respectively (P < 0.05; Table 2). Mean FCR and EPEI also improved in the pelleted diet-receiving birds than those offered to crumble diet in days 14 to 42 age (P < 0.05; Table 2). Broilers feeding with the mash diet showed a greater FI than those maintained on the crumble diet in days 14 to 42 of age. Mean weight gain was reduced in the birds fed on the crumble diet and exposed to CCl₄ toxicity at day 42 of age (P < 0.05).

The serum concentration of TC and LDL-c influenced by diet physical form × CCl₄ interaction (P < 0.05). The birds fed on the pelleted diet andreceived CCl₄injection exhibited an increased TC and LDL-c serum concentration compared with those grown on the crumble diet and exposed to CCl₄ toxicity (P < 0.05). The liver fat percentage in the birds receiving the pelleted diet was greater than those fed on the mash diet (P < 0.05). CCl₄ injection caused a significant increase in LFP compared with those that did not exposed the same challenge (P <0.05). Feeding with the pelleted diet concomitantwith CCl₄ administration caused a greater liver fat deposition compared to the other treatment combinations (P < 0.05; Table 3).

Serum activity of ALP, AST, ALT, and LDH was not modified by feed physical form and CCl₄ injection at day 42 of age (P > 0.05; Table 4). Serum AST activity, however, elevated in the CCl₄-injected birds when they fed with the crumble and pelleted diets (P 0.05; Table 4).

No significant differences were observed in serum concentrations of GLU, TP, TBIL, DBIL, and ALB in the birds subjected to the experimental treatments at day 42 of age (P > 0.05; Table 5). The serum concentration of GLU, however, increased in the broilers maintained on the pelleted diet and injected with CCl₄ and in those fed on the mash diet without CCl₄ injection compared with those receiving the pelleted diet without CCl₄ administration (P < 0.05; Table 5). The serum concentration of TP was greater in the birds receiving the pelleted diet and injected with CCl₄ (P < 0.05; Table 5). The birds which fed on the pelleted and mash diets with CCl₄ injection showed greater serum TBIL concentration than those receiving the pelleted diet without exposure to CCl₄ toxicity (P < 0.05; Table 5). The birds are grown on the pelleted diet and challenged with CCl₄ also demonstrated a greater concentration of DBIL than other broilers (P > 0.05; Table 5).

Apparent liver health, scored using a 4-grade scale, was affected by the physical form of the diet, CCl₄ injection as well as their interrelations effects (P 0.05; Table 6). For score 0, indicating a healthyliver, the relative frequency was greater (38.84%) in the birds maintained on the mash diet. The broilers receiving pelleted diet showed higher frequency for score 2 and 3 on day 42.

Mean BrG was greater in the broilers receiving the pelleted diet than those fed with the mash diet (P < 0.05; Table 7). Exposure to CCl₄toxicity resultedin a significant decrease in BrW, BrG, and BrA (P <0.05). Mean ShW, BrW, BrG, and BrA were alteredby feed presentation form × CCl₄ interaction, whereas ShW was greater in the broilers feeding with a crumble diet and exposed to CCl₄ toxicity and BrW increased in broilers receiving the pelleted and mash diets without CCl₄ exposure than those feeding with the mash diet with CCl₄ challenging (P < 0.05; Table 7). Mean BrG was affected by feed presentation form × CCl₄ interaction so that a greater in the birds fed with the pelleted diet and with no CCl₄ challenge (P < 0.05). The birds fed with the pelleted andcrumble diets with no CCl₄ intoxication showed a plump breast and wider BrA at day 42 of age. Fearfulness assayed by the tonic immobility test was not affected by either feed physical form, CCl₄ injection, or their interaction (P < 0.05; Table 7).

The tolerance statistic as the main multicollinearity diagnostic was greater than 0.1 for all parameters considered (Table 8). Liver fat percentage showed low phenotypic correlation coefficients (0.2<) with all the recorded parameters except for serum cholesterol and total protein concentration which demonstrated medium positive coefficient correlation of 0.27 and 0.21, respectively (Figure 1).

The best linear model for prediction of liver fat percentage achieved by the Forward modeling approach in SAS involved serum TC, LDL, TP, TBIL, LDH, BrA, and ShL with an R²=0.3011. The identical linear models generated by both Backward and Stepwise approaches included fear test but external body measurement was removed and their R² was lower than the previous model (0.25 in both) (Table 9).

Discussion

Almost all poultrymen and researchers believe that a pelleted feed increases weight gain and improves feed efficiency in broiler chickens (Chewning et al., 2012; Abdollahi et al., 2018). The outcome of the current study, confirm such general idea indicated by increased WG and EPEI and improved FCR in the pelleted feed-fed birds during the whole experimental period. In the line with our results, Mohammadi Ghasem Abadi et al. (2019) reported that growth performance is affected interactively by feed form, particle size, and pellet binder. Feeding pellet coarse with 3% pellet binder diets enhanced feed intake and subsequently gained more BW, which confirmed the importance of pellet physical quality. The primacy of a pelleted diet has been attributed to a reduced feed wastage, a decreased time for prehension, an increased realisability of diet ingredients, a declined microbial load among many other advantages (Zang et al., 2009; Abdollahi et al., 2019). However, because of such commercially-privileged merits, the possible disadvantages of a pelleted diet had not been received adequate interest in research. However, in many studies also, feeding pelleted diets did not show any advantage of diets in a crumble or mash form (Shirani et al., 2018).

Stressful conditions are an inevitable part of a broiler raising enterprise. As the birds are faced with hectic stimuli such as an extreme environment or contaminated feed with mycotoxins oxidative stress realize as an outcome (Lykkesfeldt and Svendsen, 2007). Intoxication using CCl₄ is a method of choice in animal toxicity models to mimic oxidative stress in experimental organisms (Manibusan et al., 2007; Behboodi et al., 2017). Our results showed that DWG reduced and serum concentration of many commonly accepted physiological stress indicators such as lipids and LFP increased in the birds subjected to CCl₄ toxicity compared with the unchallenged birds. Notably, adverse effects intensified in those birds which fed on the pelleted diet than those received the same diet in a crumble of mash form. These findings are in agreement with many previous studies demonstrating the negative effects of CCl₄ on broiler performance (Sonkusale et al., 2011; Khodadust et al., 2015). Adverse effects of toxins like CCl₄ may initiate in the gut by reducing the small intestine’s surface area for absorption (Wang et al., 2018), suppressing protein synthesis through impaired functionality of enzymes (Shuaib et al., 2010), or in a deeper perspective, disrupting function on many internal organs and metabolic pathways. In our study, feeding the pelleted diet worsens the CCl₄ toxicity, indicated by increased LFP and frequency of liver scores 2 and 3 in the birds treated with CCl₄ and fed on the pelleted diet. It was shown that during CCl₄ poisoning fat from adipose tissues is transmitted to the liver (Zhang et al., 2018), leading to fat accumulation, and ultimately resulting in tissue damage (Yalcin et al., 2017), results which has been confirmed by the current study. These findings are also consistent with Moawad (2007) report, who showed that CCl4 increased serum cholesterol and triglycerides via liver damage. In the CCl₄ treated broilers, the serum content of the total protein and albumin were decreased by 28% and 18%, respectively compared with the control birds. It was also shown that CCl₄ induces lipogenesis through the increased flowing of the acetate toward de novo lipogenesis. This process is facilitated via the transfer of acetate into hepatocytes followed by an enhanced blood lipid (Boll et al., 2001).

Several enzymes including ALT, AST, and ALP were shown to demonstrate a greater activity in hepatocytes (Jiang et al., 2015). Therefore, leakage of these enzymes into the bloodstream is anticipated in the pathological destruction of the liver cells (Parmar et al., 2012). In the current study, no significant elevation was observed in the serum activity of ALT, LDH, and ALP. However, Baradaran et al., (2019) inconsistent with our results recently reported that CCl₄ -treated broilers showed higher serum activity of ALP, AST, ALT, and GGT enzymes compared with birds in the control group, indicating CCl₄ -induced hepatotoxicity. Khorramshahi et al. (2014) also on the contrary to our results, reported that the Japanese quails treated with CCl₄ intraperitoneally exhibited an elevated serum concentration of ALP, AST, and ALT. Sonkusale et al. (2011) and Nateghi et al. (2013) also revealed that the administration of CCl₄ to broilers increased the serum activity of the hepatic enzymes. Gad et al. (2011) also reported that CCl₄ intoxication decreased serum concentrations of protein and albumin, but increased serum levels for lipid, verifying the adverse effects of CCl₄ on the liver function. The serum concentration of bilirubin, a major product form biotransformation of hemoglobin increases when a liver injury takes place or live damage lead to obstruction of the extrahepatic biliary ducts. Elevation in total serum bilirubin may result from a decreased uptake of and conjugation of bilirubin by the liver as a consequence of hepatic dysfunction (Sanjiv, 2002).

In the current study tonic immobility test as an important welfare criterion was not influenced by experimental treatments. We anticipated increased tonic immobility in the birds exposed to CCl₄ as Ghareeb et al. (2014) showed that the intoxication with deoxynivalenol (DON; known as vomitoxin) increased the underlying fearfulness and physiological stress responses in broiler chickens and resulted in a deteriorated welfare status evidenced by elevated plasma concentration of corticosterone, higher H/L ratio and a greater level of fearfulness. Similar reports have been appeared in the literature confirming adverse effects of diary toxins on fearfulness in birds (Roll et al., 2010; Nazar et al., 2012), however, our results may fail to show the same trend due to low CCl₄ doses injected or the administration route.

Whereas outcomes of our study showed a diverged effect of feed physical form on LFP and such diversity was widen by the administration of CCl₄, having a large set of data encouraged us to employ the data for the prediction of LFP in birds. Therefore, in the last part of the experiment, we tried to find a noninvasive way for the prediction of LFP by multivariate linear statistical models to help the clinicians and expert personals in the early finding of the birds suffering fatty liver. We considered 17 predictor parameters for the same purpose. In step 1, a multicollinearity test was conducted and results showed all considered variables were qualified to be included in multivariate linear models as predictor variables because tolerance exceeded 0.1 for all parameters. Tolerance, defined as 1/VIF (variance inflation factor), is applied in several studies to determine the degree of multicollinearity (Yoo et al., 2014). A tolerance value lesser than 0.1 is equivalent to a VIF of 10. It shows that the variable could be considered as a linear combination of other independent variables. As a simple and applicable rule, a variable with a VIF value greater than 10 may merit further investigation or incorporation in statistical models (Hair et al., 1995).

In the second step, Pearson correlation coefficients were calculated between all variables and LFP to provide an enhanced understanding of the relationship between the constituents of the models in the next step. No variable was found with a high positive correlation coefficient with LFP, indicating an uncertain prospect to achieve efficient linear models for the planned purpose. Results from this step confirm the previous relevant reports. In a clinical study, Portillo-Sanchez et al. (2015) reported that elevated liver enzymes do not show a significant correlation with liver histological grades of fatty liver. Moreover, declined and/or normalized serum concentration of liver enzymes following by an intervention cannot be a firm indicative criterion for improvement in liver histology. Gerzilov and Petrov, (2015) reported significant positive correlation coefficients between fatty liver weight and blood ALT, AST, and total cholesterol in both genders of chicken. They also demonstrated that blood serum triglycerides, which increased almost twice in liver steatosis, did not correlate with liver weight.

Despite the mainly weak association between considered variables and LFP, we tried to construct multilinear models. Therefore, considering all predictor variables and using three forward, backward and stepwise modeling approaches in SAS software we reached three multivariate linear models that did not satisfy us because their R2 could not exceed 0.31. Statistical models could be simple or greatly complicated, but they are always intended to improve our understanding of a system, using the existing data. The most important subsequent influence of response modeling was the theory suggested by Emmans (1981) to predict deliberate food intake in poultry and pigs, which offered a great increase the merit of estimating models by making food intake an output from, as opposed to input to, the growth model (Gous, 2007). In the same way, various biochemical parameters can be used to assess nutritional status including lipid profile and liver enzymes. In the current study, we also subjected our data to multivariable regression analysis to obtain decisive values for the association between the concerned noninvasive blood markers, body measurements, and liver fat. The same method was previously adopted by Takase et al. (2017) who predicted liver fat content in human cases by calculating the liver fat index as an alternate marker for fatty liver. Interestingly, tonic immobility time was chosen by two approaches as a significant predictor variable while it had a low correlation with LFP but a high tolerance (0.76). Probably, lack of collinearity with all other predictors in the models was the reason for the incorporation of the same variable. As far as fear is an important surrogate in welfare evaluation in poultry (Abe et al., 2013), fear testing may be a component of certification or labeling schemes that incorporate the evaluation of outcome-based welfare criteria on commercial farms. It was shown in several previous works that exogenous toxicants such as aflatoxins affect the tonic immobility duration as a behavioral indicator to stress in broiler chicken (Roll et al., 2010; Nazar et al., 2012).

Conclusion

It was concluded that a broiler diet in a pelleted form improves productive performance in broiler chickens but showed a greater potential to induce fatty liver in the same birds by increasing fat amassing in the liver. More the potential risk of fatty life will increase when feed ingredients are contaminated by environmental or microbial toxicants. In other words, exposure to CCl4 toxicity imposes adverse effects on broiler performance, blood biochemistry, and certain external body measurements, in particular, when birds are maintained on a pelleted diet. Using all 17 recorded predictor variables, despite no significant multicollinearity, we failed to introduce an efficient linear model for the prediction of LFP because R2 was not exceeded 0.31 in a model. Attempts have to be made to find more relevant predictors such as liver molecular mediators, cytokines, and transcription factors in serum or implementation of sophisticated nonlinear statistical models for the same purpose.

Acknowledgments

The authors wish to thank R. Karimirad, S. Azadbakht, O, Hasanvand, T. Pakzad, F. Delfan, and S.A. Nouraldinei, Dept. of Animal Sciences, Agriculture Faculty, Lorestan University, Khoramabad, Iran, University, Lorestan, Iran, for their technical help during the experimentation period.

This article was originally published in Poultry Science Journal 2020 8(2): 163-174. http://psj.gau.ac.ir. DOI: 10.22069/psj.2020.17869.1558.