In-feed organic and inorganic manganese supplementation on broiler performance and physiological responses

Objective A trial was conducted to investigate the effects of supplemental levels of Mn provided by organic and inorganic trace mineral supplements on growth, tissue mineralization, mineral balance, and antioxidant status of growing broiler chicks. Methods A total of 500 male chicks (8-d-old) were used in 10-day feeding trial, with 10 treatments and 10 replicates of 5 chicks per treatment. A 2×5 factorial design was used where supplemental Mn levels (0, 25, 50, 75, and 100 mg Mn/kg diet) were provided as MnSO4·H2O or MnPro. When Mn was supplied as MnPro, supplements of zinc, copper, iron, and selenium were supplied as organic minerals, whereas in MnSO4·H2O supplemented diets, inorganic salts were used as sources of other trace minerals. Performance data were fitted to a linear-broken line regression model to estimate the optimal supplemental Mn levels. Results Manganese supplementation improved body weight, average daily gain (ADG) and feed conversion ratio (FCR) compared with chicks fed diets not supplemented with Mn. Manganese in liver, breast muscle, and tibia were greatest at 50, 75, and 100 mg supplemental Mn/kg diet, respectively. Higher activities of glutathione peroxidase and superoxide dismutase (total-SOD) were found in both liver and breast muscle of chicks fed diets supplemented with inorganic minerals. In chicks fed MnSO4·H2O, ADG, FCR, Mn balance, and concentration in liver were optimized at 59.8, 74.3, 20.6, and 43.1 mg supplemental Mn/kg diet, respectively. In MnPro fed chicks, ADG, FCR, Mn balance, and concentration in liver and breast were optimized at 20.6, 38.0, 16.6, 33.5, and 62.3 mg supplemental Mn/kg, respectively. Conclusion Lower levels of organic Mn were required by growing chicks for performance optimization compared to inorganic Mn. Based on the FCR, the ideal supplemental levels of organic and inorganic Mn in chick feeds were 38.0 and 74.3 mg Mn/kg diet, respectively.


INTRODUCTION
Manganese (Mn) is a bioactive element required for multiple postabsorptive physiologi cal processes in poultry, such as carbohydrate, lipid, and amino acid metabolism, as well as cartilage and bone development, and antioxidant defense [1]. For fastgrowing broilers, Mn deserves concern since its manipulation in practical feeds may attenuate the incidence of certain disorders associated with fast growth and, in turn, may protect against adverse impacts on revenue and profitability of the chicken meat industry. As a cofactor for glyco syltransferase, Mn acts by attaching glucosamines to a protein core during the synthesis of proteoglycans, an important constituent of bone organic matrix and cartilage [2]. Low concentrations of Mn in broiler feeds have been shown to decrease the activity of glycos yltransferase and, consequently, the synthesis of proteoglycans [3], which is commonly associated with the incidence of leg abnormalities in broilers fed Mn deficient feeds [4,5]. In turn, leg abnormalities in commercial flocks lead to deprivation of locomotive freedom and restricted access to feeds, and consequently to impairments in growth and feed efficiency, which on welfare and economic grounds is undesired in in tensive poultry rearing [6].
Manganese is also a component of Mncontaining superoxide dismutase (SOD), an essential antioxidant metalloenzyme responsible for free radical scavenging in mitochondria by catalyzing the dismutation of superoxide anions produced from the electron transport chain to molecular oxygen and hydrogen peroxide [7]. Mitochondria are the major oxida tive phosphorylation site where carbohydrates and fats are oxidized to produce energy for cell functioning [8]. Electron and proton leaks across the mitochondrial respiratory chain are believed to increase the generation of free radicals, which may reduce the respiratory chain function or even result in cellular apoptosis [7,8]. Apart from their consequences on the energetic efficiency of cells, intracellular oxidation of lipids and proteins caused by free radicals has a severe practical implication for the poultry meat industry, since damage in tissues affects relevant sensory traits of meat such as overall appearance, color, texture, and flavor [9]. Lu et al [5] dem onstrated that dietary Mn supplementation upregulated MnSOD gene expression, increased MnSOD activity, and reduced malondialdehyde content, a biomarker of lipid per oxidation, in leg muscles of growing chicks.
Manganese requirements of growing broilers are described by the National Research Council [10] based on leg abnor malities and growth as 60 mg/kg diet. This value is half the 120 mg Mn/kg recommended by Lu et al [5] few years earlier. Contrary to NRC [10], the referred authors [11] based their estimates on growth performance, tissue mineralization, and MnSOD activity in heart. As an attempt to reevaluate grow ing chick responses to supplemental Mn, Lu et al [4] found that neither growth performance nor meat quality traits were affected by the supplemental Mn levels under investigation (100 to 500 mg Mn/kg diet as Mn sulphate), indicating that adding 100 mg Mn/kg diet supports broiler carcass traits and meat quality. Even though relevant as references, these find ings fail to establish the minimum level of supplemental Mn to optimize chick responses.
From the mid2000s, organic trace minerals (OTM) re ceived considerable attention due to their potential benefits and advantages compared with inorganic salts. After decades of investments and research efforts, OTM have proved to be more bioavailable than carbonates, oxides, and sulphates tradi tionally used by industry [12]. One point commonly ignored in assessments of poultry responses to sources and concen trations of trace minerals, which could potentially lead to misinterpretations of outcomes, is the source of the other trace minerals provided in the supplement in experimental diets. Generally, trace mineral supplement used in assays conducted to determine ideal levels of a given OTM provide all the other trace minerals as inorganic salts. Although wide spread, such approach might not be the most accurate one to validate optimum levels of organic minerals for poultry. Firstly, because it fails in mimicking the conditions under which commercial flocks are reared. In intensive broiler rearing, Mn provided as inorganic salts has been typically supplemented with other trace minerals also in inorganic form, whereas Mn as proteinates or chelates has been offered in feeds with other OTMs. Therefore, assessment of the opti mal dietary concentrations of supplemental manganese for use by the poultry industry may be most appropriately accom plished in the dietary environment in which each manganese source will typically be used in practice, organic manganese with other organic minerals and inorganic manganese with other inorganic minerals.
Secondly, Mn interacts with other trace minerals in many physiological processes either antagonistically or synergis tically. A classic antagonistic interaction of Mn has been reported with iron (Fe). Both minerals share two common transport protein in intestine, the cellular importer divalent metal transporter 1 (DMT1) and the cellular exporter fer roportin 1 [13,14]. Evidences have suggested that higher amounts of Fe inhibit the expression of DMT1 in entero cytes, and lead to a depression of Mn uptake [15]. Because OTMs may be absorbed as peptides and/or amino acids in intestine, it seems reasonable to hypothesize that organic Fe in feeds, for example, could optimize the absorption and utilization of organic Mn by broilers. Therefore, chick responses to organic Mn could be potentially different, and lower levels could be required whether all the trace mineral sources are provided in organic form rather than inorganic form. We hypothesized that the supplemental level of organic Mn required to optimize growing broiler chick responses would be lower than inorganic supplemental Mn. There fore, we conducted an experiment to evaluate the effects of supplemental levels of Mn provided by organic and inor ganic trace mineral (ITM) supplements on growth, tissue mineralization, mineral balance, and antioxidant status of growing broiler chicks.

Animal care
Animals care and use procedures described in this section were in accordance with Brazilian Legislation on Animal Experimentation and Welfare, and the experimental protocol was approved by the Ethics Committee on the Use of Farm Animals (CEUAPUFV) of the Universidade Federal de Viçosa, (protocol 111/2014) before the initiation of the trial.

Birds and husbandry
A total of five hundred 1dold male Cobb 500 chickens ob tained from a local commercial hatchery were used in this study. During the first seven days of age, the birds were fed a prestarter diet based on corn and soybean meal formulated to meet or exceed nutritional recommendations of Rostagno et al [16], except for Mn, whose dietary concentration was provided at 43 mg Mn/kg of feed as manganese sulphate (MnSO 4 •H 2 O), which corresponded to 50% of the dose rec ommended by NRC [10]. The prestarter diet contained less Mn than typically used in commercial production to avoid excessive storage reserves of this mineral. Throughout the entire preexperimental period, chicks had free access to water and mash feed. On d 8 post hatch, chicks were housed in an environmentally controlled room and allotted into 49 cm× 27 cm×33 cm (length×height×width) plastic cages with raised wire floors until the end of the feeding assay. Chicks were al lowed ad libitum access to feed and demineralized water throughout the 10d experimental period via plastic feeders and cup drinkers. Photoperiod was set at 12 h natural light/ 12 h artificial light. Both temperature and humidity were set according to genetic guideline recommendations.

Performance, sampling, and chemical analysis
On d 17 posthatch, all chicks and feed leftovers from each experimental unit were weighed to determine the final BW, ADFI, average daily gain (ADG), and feed conversion ratio (FCR). From d 13 to 17 posthatch, trays covered with plas tic were placed underneath pens for total excreta collection. Total excreta were collected daily, weighed, and frozen (-20°C). At the end of the collection period, excreta from each pen were homogenized. On d 18 posthatch, one bird per cage (10 birds/treatment) was randomly selected and sacrificed by cervical dislocation. Subsequently, the breast, liver, and left and right tibia were collected and stored at -20°C. The bird had the breast muscle, liver, and left and right tibia col lected, stored at -20°C, and subsequently, were ether extracted for 4 h in a Soxhlet extractor. Posteriorly, these samples, as well the samples of excreta, were freezedried for 72 h at -80°C under 800 mbar of pressure (Liobras, São Carlos, SP, Brazil), ground in a stainless ball mill (Micro spray mill, RTE 350, TECNAL, Ourinhos, SP, Brazil), and finally analyzed in the atomic absorption spectrophotometer (Spctr AA800; Varian spectrometer, Harbor City, CA, USA) at Animal Nutrition Laboratory (Universidade Federal de Viçosa, Viçosa, MG, Brazil) to obtain the Mn, Cu, Zn, Fe, Ca, and P concentra tions as described by AOAC [17]. Manganese retention was calculated through the difference between the amount of Mn consumed and excreted and expressed as mg/kg of BW. The total superoxide dismutase (totalSOD) and glutathione peroxidase (GSHPx) activity in breast muscle, and liver tis sues were performed according to Walsh et al [18] using the kits of Randox Laboratories Ltd. (County Antrim, UK) Ransod and Ransel, respectively, through an automatic biochemical analyzer (Mindray BS200E, Shenzhen Mindray BioMedical Electronics Co., Shenzhen, China) following the manufac turer guidelines.

Statistical analysis
Data were analyzed as a completely randomized design un der an incomplete 2way (source×levels) factorial assay with inorganic and OTMs supplement without Mn supplementa tion (zero level) and Mn supplement levels as organic Mn in OTM supplement and inorganic Mn in ITM supplement. In this context, the traditional factorial analysis was updated to an incomplete factorial design [19]. This approach is easily accomplished by using common statements from PROC MIXED of SAS (Version 9.4, SAS Institute Inc., Cary, NC, USA) software. According to the previously mentioned analy sis, the significance (p<0.05) of source effect (only two levels) was evaluated through Ftest; whereas orthogonal contrasts were applied to perform the analysis between linear and quadratic responses of dependent variables in function of increasing Mn levels. Additionally, when Mn levels in each source were significant, means were compared using Tukey's multiple comparison test. Quadratic polynomial regression model is: where Y is the dependent variable, X is the dietary Mn con centration, and β0 is the intercept, β1 and β2 are the linear and quadratic coefficients, respectively. Additionally, perfor mance data were fitted to linear brokenline model (LBL) using NLMIXED procedure of SAS (Version 9.4, SAS Institute Inc., Cary, NC, USA) software to estimate the supplemental Mn level which optimized performance responses assessed.
The LBL model was expressed as: where Y is the dependent variable, X is the dietary Mn con centration, β0 is the value at the plateau, β1 is the slope and β2 is the Mn supplemental at the break point (level to opti mize the biological response). Statistical significance was considered as 0.05 for all executed analysis. The term "ten dency" was used for situations in which pvalue is between 0.05 and 0.10.

RESULTS
For both Mn sources under study, the analyzed values of Mn in experimental diets were close to those expected, as well as the concentrations of Fe, Cu, and Zn (Table 2).  (Table 4).

Manganese balance
Manganese supplemental levels and sources influenced the intake, excretion, retention, and balance of Mn in broiler chicks (

Manganese tissue concentrations
No interactive effects between Mn source and Mn supplemen tation levels (p>0.05) were noticed on Mn tissue concentration (    (p<0.05) in Mn concentration in liver and tibia and quadratic increase (p<0.05) in liver (Table 4).

Antioxidant enzyme activity
As detailed in Table 7, no interactive effects between source and supplemental Mn levels were noticed (p>0.05) on anti oxidant enzyme activity in chick liver and breast muscle (

Optimal supplemental Mn levels for organic and inorganic sources
Data of performance, Mn balance, and Mn tissue concentra tion were fitted to different regression models as summarized in

DISCUSSION
Establishing broiler requirements for Mn has been revealed to be particularly challenging, since estimates and biological responses may be affected by several factors, which include the source of Mn under study, the experimental basal diets used to produce treatments, the concentration, and source of the other trace minerals in experimental diets, as well as  the biological response used as a reference to determine the optimal level. In the current research, a semipurified diet based on dextrose, casein, and albumin was supplemented with Mn levels and sources to produce dietary treatments. As highlighted in the last revised edition of Nutrient Require ments of Poultry [10], estimates of trace mineral requirements of chicks fed semipurified diets are expected to be lower than those obtained from cerealbased diets due to poor or nonexistent presence of antinutritional factors such as phy tate and fiber. In this experiment, both sodium phytate and cellulose, as well as a microbial phytase, were added to ex perimental diets to simulate a commercial cerealbased diet. When investigating growing chick requirements for Mn using a semipurified dextrosecasein diet, Southern and Baker [20] estimated optimal Mn level at 14 mg Mn/kg diet.
In the current research, with the same type of ingredients, we estimated higher requirements, whose values were similar to literature. Such fact indicates fiber and sodium phytate added to experimental diets fulfilled the purpose of simu lating practical cerealbased diets. Our outcomes demonstrated that regardless of the sources assessed, performance traits were influenced by Mn supple mentation to the basal feeds, which proves the essentiality of this mineral for growing chicks. Even though chicks fed 25 mg supplemental Mn/kg diet achieved similar performance targets to chicks fed the highest Mn level under study, there was a linear improvement in performance responses as Mn supplementation increased, mainly in FCR (Table 3). In order to describe responses to supplemental Mn; and estimate the supplemental Mn levels required to optimize the performance traits, ADG, and FCR data were fitted to the LBL regression model. As summarized in   to reach maximal chick performance when Mn was provided as MnPro via the OTM supplement. So far as the mechanisms underlying Mn absorption are understood, Mn uptake occurs mainly in the upper small intestine by the transport protein DMT1 [21,22], whose expression in intestinal mucosa is modulated by Mn source and dietary level [13]. Previous findings have demonstrated that complexed or chelated or ganic Mn increases mRNA expression of DMT 1 in broiler chick small intestine compared with Mn sulphate [23], which explains, at least in part the fact that lower levels of organic Mn were able to reproduce similar performance to higher levels of inorganic Mn in the current research. Previous reports have suggested that Mn supplementation did not affect ADG and FCR of growing chicks, and that cerealbased diets containing 19 to 26 mg Mn/kg from cereal grains in the diet without a supplemental Mn source, could support performance objectives [1,3,4]. Our findings indicate that 25 mg supplemental Mn/kg diet was sufficient to support proper growth rates, which considering Mn content in basal diet, i.e. 6 mg Mn/kg, would be equivalent to 31 mg Mn/kg. Despite similarities, a comparison between our requirement estimates and those described in the refereed references cannot be made with confidence due to differences in the Mn sources. The bioavailability of Mn sources is limited in poultry, espe cially in cereals [10]. Wedekind et al [24] reported that only 9% of the Mn provided by a corn and soybeanbased diet supplemented with MnSO 4 •H 2 O (100 mg Mn/kg diet) was absorbed by broiler chicks. Yet, the authors reported that such rate was 2.8% when no Mn source was added to basal diet. Such findings may be clearly supported by our results. As detailed in Table 5, the amount of Mn retained in chicks relative to Mn intake, i.e. Mn balance, was 56% lower in birds fed basal diets compared with the balance in chicks fed the lowest supplemental Mn level of 25 mg/kg diet. Even though Mn balance differed among Mnsupplemented groups, the difference was narrow between the lowest and highest Mn balance (35.5% vs 36.5%), which suggests that dietary levels higher than those estimated for performance optimization were utilized by chicks and retained in the body, as our data for tissue mineralization show. As detailed in Table 8 [25] reported that chick liver Mn content increased up to 60 mg Mn/kg diet and remained constant up to 130 mg Mn/kg diet, regardless of the source assessed. When fitting liver Mn concentration data to LBL model, the breakpoints for maximum concentration were estimated at 39.3 mg Mn/kg diet considering both sources, and 33.5 and 43.1 mg Mn/kg diet for chicks fed MnPro and MnSO 4 •H 2 O supplemented diets, respectively. Liver is the primary organ responsible for regulating Mn body status through biliary excretion [26]. When provided in concentra tions that exceed physiological needs, Mn may be progressively accumulated in different organs and, beyond critical limits, be excreted to avoid toxicity [20]. Chicks fed higher levels of Mn than the requirements established herein for organic (38 mg Mn/kg diet) and inorganic (74.3 mg Mn/kg diet) stored Mn in extrahepatic tissues (e.g. breast and tibia). Breast muscle Mn content responded with increasing Mn deposi tion up to 75 mg Mn/kg diet, whereas Mn deposition in tibia continuously increased up to 100 mg Mn/kg diet, the highest level under study. Our findings support those reported by Yan and Waldroup [27] who reported that regardless of the source of Mn assessed (MnO, MnSO 4 •H 2 O, or amino acid chelated Mn), Mn concentrations in broiler tibia were grad ually increased up to 800 mg supplemental Mn/kg diet. Similarly, when investigating the Mn supplementation on broiler diets, Conly et al [25] noticed that Mn was continu ously deposited in the tibia of chicks fed tribasic manganese chloride or MnSO 4 •H 2 O supplemented diets (0, 30, 60, and 130 mg supplemental Mn/kg), achieving the greatest value at the highest level studied. Although Mn content in breast muscle and tibia were unaffected by Mn sources, the analysis of levels in each source suggests that Mn concentration in breast increased until 75 mg Mn/kg only when Mn was sup plemented as MnPro. According to polynomial regression model estimates, Mn concentration in breast reached its maxi mum values at supplemental Mn level of 67.5 mg/kg, and for chicks fed MnPro, maximum deposition was achieved at 62.3 mg/kg. We noticed that Mn concentrations in tibia and liver approximately doubled in chicks fed the highest supplemental Mn level compared with chicks fed diets without supplemen tal Mn. Such outcomes differ from those reported by Lu et al [5] that neither amino acid chelated Mn nor MnSO 4 •H 2 O af fected breast muscle Mn content of growing chicks fed diets supplemented at 100 and 200 mg Mn/kg, and from those re ported by Yan and Waldroup [27] who observed a higher Mn content in the tibia of chicks fed amino acid chelated Mn compared with chicks fed MnO and MnSO 4 •H 2 O. Our re sults suggest that Mn concentration in breast muscle are mainly influenced by proteinate Mn source, i.e. MnPro.
Although Mn levels did not affect antioxidant enzyme activity, we observed that, curiously, chicks fed inorganic minerals supplemented diets exhibited higher activity of GSHPx in breast muscle and liver, and higher totalSOD ac tivity in liver compared with chicks fed organic minerals. Organic minerals are potentially more bioavailable than the inorganic forms, so it was expected that they would support higher totalSOD activity. It is worth highlighting, however, that although trace minerals modulate the activity of anti oxidant enzymes like SOD, GSHPx, and catalase, they may also act as prooxidant agents [28,29]. Free iron and copper, for example, have been described as the major catalyzers of the production of free radicals such as hydrogen peroxide (H 2 O 2 ) and hydroxyl, which disrupt the redox balance in cells, and cause oxidative damage to tissues [30,31]. Because ITMs are affected by the variation of pH, they may reach some tissues such as gut mucosa and blood as reactive ions, which may potentially oxidize cytosolic structures and DNA of cells. Therefore, inorganic minerals may be potentially prooxidant compared with OTMs, whose chemical struc tures are more stable and not so easily dissociated. Even though GSHPx is an enzyme dependent on selenium and not Mn, its activity is expected to increase in response to the increase in totalSOD activity. Whereas SOD acts in a first level, in the dismutation of superoxide radical to H 2 O 2 , GSHPx ends the process by detoxifying H 2 O 2 .

CONCLUSION
To the best of our knowledge, this is the first research where the ideal level of organic Mn for broiler chicks was estimated using a supplement in which the other trace minerals (zinc, copper, iron, and selenium) were provided as organically complexed metals. As hypothesized, when compared with inorganic Mn, lower levels of organic Mn are required by growing chicks for performance optimization. Overall, esti mates of supplemental Mn levels required to optimize Mn balance and Mn tissue concentration in chicks were lower than those levels to optimize feed conversion ratio in chicks. Based on the feed conversion response, the ideal supplemental levels of organic and inorganic Mn for broiler chicks were 38.0 and 74.3 mg Mn/kg diet, respectively.

CONFLICT OF INTEREST
James Eugene Pettigrew contracts with Alltech, the supplier of the MnPro used in this experiment, to supervise this re search. We certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.