DOI: 10.1111/jpn.12307
ORIGINAL ARTICLE
Effect of high dietary zinc oxide on the caecal and faecal shortchain fatty acids and tissue zinc and copper concentration in pigs is reversible after withdrawal of the high zinc oxide from the diet € ckler1 and A. Zeyner4 €sing2, B. Dobenecker3, K. No P. Janczyk1, K. Bu 1 Unit for Molecular Diagnostics, Genetics and Pathogen Characterisation, Department of Biological Safety, Federal Institute for Risk Assessment Berlin, Germany 2 Chair of Nutrition Physiology and Animal Nutrition, Faculty of Agricultural and Environmental Sciences, University of Rostock, Rostock, Germany 3 Animal Nutrition and Dietetics, Department of Veterinary Science, Ludwig-Maximilians-University Munich, Oberschleißheim, Germany, and 4 Group Animal Nutrition, Institute of Agricultural and Nutritional Sciences, Martin-Luther-University Halle-Wittenberg, Halle (Saale), Germany
Summary Zinc oxide (ZnO) used in high (‘pharmacological’) levels to prevent diarrhoea in pigs is assumed to reduce copper (Cu) in tissues and inhibits large intestinal microbial fermentation. To test it, German Landrace pigs were weaned on d28 of age and fed diets containing either 100 (LowZinc, LZn, n = 10) or 3100 mg ZnO/kg (HighZinc, HZn, n = 10). The mixed feed (13.0 MJ ME, 18.5% crude protein) was based on wheat, barley, soya bean meal and maize. After 4 weeks, the HZn group was further fed 100 mg ZnO/kg for another 2 weeks. Caecal contents, faeces and tissues were collected after 4 weeks (n = 5 and n = 10 respectively) and 6 weeks (n = 5 and n = 5 respectively). Faeces and caecal content were analysed for dry matter (DM), pH, ammonia, lactic acid (LA) and shortchain fatty acids (SCFA) on native water basis. ANOVA was performed to elucidate significant differences at p < 0.05. No diarrhoea occurred. After 4 weeks, the caecal contents’ pH increased (p < 0.001) and butyric (p < 0.05) and valeric acid (p < 0.01) decreased in the HZn group in comparison with LZn. In faeces, a decrease of acetic (p = 0.009), butyric (p = 0.007) and valeric acid (p = 0.046), as well as reduced acetic:propionic acid (A: P) ratio (p = 0.025) was observed in the HZn group in comparison with LZn. Faecal ammonia decreased in HZn (p = 0.018). No differences (p > 0.05) were recorded in caecal contents after 6 weeks. In faeces, acetic acid remained lower in the HZn group in comparison with LZn (p = 0.006), as did the A:P ratio (p = 0.004). Zn concentration in liver, kidneys and ribs, and Cu concentrations in kidneys increased in HZn. Withdrawal of ZnO resulted in reversibility of the changes. The effect on butyric acid should be discussed critically regarding the energetic support for the enterocytes. High Zn and Cu tissue concentrations should be considered by pet food producers. Keywords intestinal microbial activity, pig, short-chain fatty acids, zinc oxide, copper Correspondence Pawel Janczyk, Karl-Heinrich-Ulrichs-Strasse 8A, 10787 Berlin, Germany. Tel: +49-1708095390; Fax: +49-32221007198; E-mail: [email protected] Received: 30 January 2014; accepted: 27 June 2014
Introduction Zinc (Zn) is a component of over 300 enzymes in the mammalian cells and thus an essential micronutrient (Poulsen and Larsen, 1995; Terr es et al., 2001). Growing pigs (up to 20 kg body weight) require approximately 100 mg Zn/kg dry matter (DM) of feed, and later approximately 80 mg Zn/kg DM of feed (Gesellschaft f€ ur Ern€ ahrungsphysiologie, 2006). Because the availability of Zn occurring naturally in feedstuffs may often be reduced by many factors (Smith et al., 1962; Pond et al., 1985;
Poulsen and Carlson, 2001), supplementation of diets with Zn by way of both organic and inorganic sources has become a standard (Richards et al., 2010). Furthermore, supplementation of dietary zinc oxide (ZnO) at high levels (2000–3000 mg/kg) has been used as prophylactic measure against postweaning diarrhoea and to improve growth performance of weaning pigs (Poulsen, 1995; Wang et al., 2010). Next to its direct effect on the Zn homeostasis in the host, it is believed that ZnO acts locally in the intestinal lumen affecting the microbiome (Pieper et al., 2011), but its exact growth-promoting
Journal of Animal Physiology and Animal Nutrition 99 (Suppl. 1) (2015) 13-22 © 2015 Blackwell Verlag GmbH
13
P. Janczyk et al.
Dietary zinc and intestinal microbial activity in pigs
and antidiarrhoea mode of action still remains unclear. Enteric bacteria can be inhibited by increased Zn concentrations (Surjawidjaja et al., 2004), and effects on the composition of ileal core microbiome have been reported (Vahjen et al., 2011). High concentrations of Zn were shown to affect the functionality and the microbial community of soils (Bewley and Stotzky, 1983). So far there are little data on functional changes of the large intestinal microbiome caused by increased ZnO levels in the diet of pigs (Højberg et al., 2005), but no data were found on their reversibility after removal of the excess of ZnO. Few studies investigated the concentrations of Zn in the tissues after feeding high amounts of ZnO (Hahn and Baker, 1993; Poulsen, 1995; Shell and Kornegay, 1996; Jensen-Waeren et al., 1998; Case and Carlson, 2002; Rincker et al., 2005). However, no data could be found for tissue concentrations of Zn when ZnO was fed in high amounts for a period of time with the following reduction of ZnO level in the diet. Excessive Zn in the diet is made responsible for a reduction of the copper (Cu) bioavailability (Hill et al., 1983). Increased dietary ZnO could therefore result in a decrease of Cu in tissues of weaned pigs. Tissue (especially edible as muscles) concentration of Zn and Cu is important for the calculations of dietary intake of these micronutrients, both for humans and especially for pets, as most of the animal by-products such as liver and kidney are taken for pet food production. This study aimed to evaluate the effect of a withdrawal period for 2 weeks following the 4 weeks high-Zn nursery period with large intestinal pH, short-chain fatty acids, lactate and ammonia, as well as tissue Zn and Cu concentrations as the response criteria. Material and methods Animals and treatment
Twenty German Landrace piglets of both sexes were weaned at 28 days of age (8.5 1.0 kg body weight), transported to the experimental facility and allocated to pens, two pigs each (male and female). The piglets received a commercial creep feed from 2 weeks of age (Turbostart, Trede & von Pein, Itzehoe, Germany) that contained 19.0% of crude protein, 7.5% of crude fat, 3.0% of crude fibre, 5.5% of crude ash, 0.7% Ca, 0.6% P, 0.25% Na, 1.5% Lys, 0,5% Met and 14.8 MJ ME/kg. The feed containing the following supplements (calculated values): 22 500 I.U. Vitamin A (E672), 2000 I.U. Vitamin D (E671), 150 mg Fe (FeII-sulphate), 1.0 mg I (potassium iodide), 0.3 mg Co 14
(alcalic Co-II-carbonate), 150 mg Cu (Cu- II-sulphate), 70 mg Mn (Mn-III-oxide), 0.3 mg Se (sodium selenite), 100 mg Zn (Zn sulphate). The study was planned and performed before the ban of the use of Co as supplement for pigs in the EU (European Community, 2013). Five pens formed one group. Wheat–barley–soya– maize diet (Table 1) was prepared to contain 150 mg Zn/kg diet that is the maximal allowed level of Zn according to the EU legislation at that time (European Community, 2003), by adding 100 mg of ZnO per kg of diet (CAS: 1314-13-2; Sigma-Aldrich Chemie GmbH, Germany). This feed was fed to the low-zinc (LZn) group (control). The experimental group received the same diet, but the Zn level was increased by adding 3100 mg ZnO/kg – high zinc group (HZn).
Table 1 Composition of basal and experimental diets used in the study Item [g as fed]
LZn
HZn
Wheat Barley Soya bean meal Maize meal Calcium carbonate Monocalcium phosphate Mineral mix* Salt Lysine-HCl Methionine Soybean oil ZnO Total Nutrients (as calculated) Dry matter [%] Metabolisable energy [MJ] Crude protein [%] Starch [%] Fibre [%] Crude fat [%] Crude ash [%] Lysine [%] Methionine [%] Methionine + Cysteine [%] [mg/kg DM] as analysed Zn Cu Fe
384 303 234 10 18 18 13 2 2 1 15 0.1 1000
384 303 234 7 18 18 13 2 2 1 15 3.1 1000
87.9 13.0 18.5 37.6 3.5 3.4 8.1 1.15 0.35 0.7 179 33 308
87.9 13.0 18.5 37.6 3.5 3.4 8.1 1.15 0.35 0.7 2353 31 308
*Mineral mix contained per kg: semolina bran (36%), NaCl (33.6%), MgO (10.5%), vit. A (600 000 IU), vit. D3 (120 000 IU), vit. E as alpha-tocopherol acetate (8000 mg), vit. K3 as menadione-NaHSO3 (300 mg), vit. B1 as thamin-HCl (250 mg), vit. B2 as riboflavine (250 mg), vit. B6 as pyridoxol-HCl (400 mg), vit. B12 (2000 lg), nicotin acid (2500 mg), folic acid (100 mg), biotine (25 000 lg), Ca-D-panthotenate (1000 mg), choline-Cl (80 000 mg), MnO (6000 mg), FeCO3 (5000 mg), CuSO4 9 5H2O (1000 mg), CoSO4 9 7H2O (30 mg), Ca(IO3)2 (45 mg), Na2SeO3 (35 mg).
Journal of Animal Physiology and Animal Nutrition © 2015 Blackwell Verlag GmbH
P. Janczyk et al.
To investigate the reversibility of potential changes caused by the high content of ZnO, after 4 weeks of treatment, the HZn group received the same feed as LZn for another 2 weeks. Feed was fed in meal form (88% DM) and offered mixed with some water twice daily for an hour as semi ad libitum, to avoid refusals. Water was provided ad libitum via nipple drinkers. Ambient temperature was kept at 25 1 °C for the first 4 weeks, and then reduced to 22 1 °C, with humidity 30–55% and light regime of 12 h light and 12 h darkness. Pigs were weighed once a week before the morning feeding. Feed intake was recorded daily on dry matter basis for each pen. Average daily gain (ADG), average daily feed intake (ADFI) and feed conversion ratio (FCR) were calculated. Faeces were collected after 4 and 6 weeks of treatment, directly from the rectum, and kept cooled on ice until further processing. At these time-points, five pigs from each group were euthanized by overdose of pentobarbital under general azaperon (Stresnil; Janssen Animal Health, Neuss, Germany) and ketamine (Ketamin 10%; Bremer Pharma GmbH, Warburg, Germany) anaesthesia; the feed was withdrawn 2 h before the euthanasia. Total digesta from caecum was collected. The whole liver, both kidneys, left ham and X-XIIth ribs were dissected for analysis of Zn and Cu concentrations. The animal experiment was approved by the local authority (Landesamt f€ ur Gesundheit und Soziales, LAGeSo, Berlin) under the accession number G 0349/ 09. Analysis of digesta and faeces
Faeces and caecal digesta were analysed for DM, pH, ammonia, lactic acid (LA) and short-chain fatty acids (SCFA), namely acetic, butyric, propionic, valeric, and capronic acid and their isoforms. DM was determined by freezing fresh digesta or faeces at 20 °C with final lyophilisation. For other analyses, fresh samples were diluted 1:3 with sterile distilled water (20 g sample, 60 ml water) and homogenised in BagMixer (Interscience, Saint Nom, France) at maximum speed for 30 s. Forty-five millilitre of the homogenate was transferred into 50 ml-Falcon tube and pH was measured using Digital-pH-Meter 646 (Knick, Berlin, Germany). Further, the homogenates were centrifuged at 4000 g for 10 min, at 20 °C. Supernatant was collected and stored at 20 °C before further analyses. Within the thawed samples, the contents of LA and SCFA were analysed by HPLC and gas chromatography, respectively, as described (Hackl et al., 2010). The content of ammonia was determined by the Journal of Animal Physiology and Animal Nutrition © 2015 Blackwell Verlag GmbH
Dietary zinc and intestinal microbial activity in pigs
modified microdiffusion method (Voigt and Steger, 1967). All concentrations were calculated as mmol of substance per litre of native water content in caecum digesta/faeces according to Zeyner et al. (2004). The ratio of acetic to propionic acid (A:P) was calculated. Zinc and copper analysis
Dissected organs, ham and ribs were weighed and homogenised using a kitchen mill (Moulinex, SEB S.A., Ecully Cedex, France). The homogenised tissues were put into aluminium bowls and frozen at 20 °C. They were subsequently lyophilised, packed hermetically and sent to the analytic laboratory. The determination of Zn and Cu in the homogenised and lyophilised tissues was performed in an acetylene flame using atomic absorption spectrometry (PerkinElmer Inc., Waltham, Massachusetts, USA) after wet hydrolysis with HNO3 in a microwave (Milestone Inc., Shelton, Connecticut, USA). Statistical analysis
Levene test was applied for testing the homogeneity of variance of all traits. Effect of different ZnO levels on the body weight of the pigs was analysed as analysis of variance (ANOVA) with repeated measures with pig as experimental unit. The effect of ZnO on DM, pH, ammonia, LA and SCFA, as well as on tissue Zn and Cu concentrations, were tested by t-test to elucidate which differences were responsible for the observed effects. The calculations were performed using SPSS for Windows version 12.0.2 (The Appache Software Foundation; IBM Corp., Armonk, NY, USA). Mean values with pooled standard error of the mean (pSEM) are provided in the tables. Differences for the traits within groups and between both time-points were analysed performing also a t-test. When t-test was performed, the Bonferroni correction was applied. Differences between mean trait values were considered significant at p < 0.05. Tendency for differences was considered at 0.05 < p ≤ 0.1. Results The mean analysed Zn concentration in the feed was 180 mg/kg DM (160 mg/kg feed) in LZn and 2400 mg/kg DM (2250 mg/kg feed) in HZn (Table 1). The Cu concentration in the feed was analysed to be 33 and 31 mg/kg DM in LZn and HZn respectively (Table 1). All piglets were in good condition throughout the time of the treatment; no diarrhoea occurred. 15
P. Janczyk et al.
Dietary zinc and intestinal microbial activity in pigs
There was no effect of ZnO on the body weight. However, the pigs from HZn weighed numerically more than pigs from LZn from d53 onwards (Table 2). During the 2nd and 3rd week, the HZn gained more than LZn (p < 0.05). ADFI was higher in the HZn during the 2nd week (p < 0.05). The FCR was higher in HZn during the 1st week and lower in the HZn during the 3rd and 5th week (Table 2). After 4 weeks of the experimental period, pH of the caecal digesta was greater in the HZn in comparison with LZn (p < 0.001). Whereas (iso-)butyric and valeric acid decreased in the HZn (p < 0.05), propionic acid tended to decrease (0.05 < p < 0.1) but lactic and acetic acid were not affected there (p > 0.1). The con-
SEM, standard error of the mean. *ADG was calculated using pig as experimental unit. ADFI and FCR were calculated using pen as experimental unit. †LZn received diet containing 100 mg ZnO/kg diet for 6 weeks; HZn group received diet with 3100 mg ZnO/kg diet for 4 weeks, then the same diet as LZn group for another 2 weeks. ††statistically significant differences at P < 0.05 presented in bold.
tents of DM and ammonia were neither affected (p > 0.1). A weak negative correlation was calculated between the pH and the total SCFA in caecum (r2 = 0.461), and the strongest correlation was observed between the butyric acid concentration and pH (r2 = 0.572). In faeces, acetic and (iso-)butyric, (iso-) valeric acid decreased in the HZn (p ≤ 0.01). Faecal propionic acid concentration was not affected (p > 0.1). The A:P ratio in caecum was greater in HZn in comparison with LZn (p < 0.05). In faeces, an opposite was observed – A:P ratio was lower in HZn (p < 0.05). Faecal ammonia concentration decreased in HZn (p < 0.05). No correlation between the total SCFA and pH was detected (r2 = 0.147). All traits are summarised in Table 3. After the change of diet in the HZn group to the feed administered to the LZn group and feeding the animals for two consecutive weeks, no differences were recorded in the caecal digesta between the groups. In faeces, again, acetic acid was lower in the HZn in comparison with LZn (p < 0.01). Similar to the results obtained after 4 weeks, in faeces, the A:P ratio was lower in HZn in comparison with LZn (p < 0.01). No further differences were recorded (Table 4). Comparison of the data obtained after 4 and 6 weeks of feeding within the groups revealed no differences for LZn except for isobutyric acid, which decreased in caecal digesta in the LZn group (p < 0.05). The change of the diet from HZn to LZn and feeding it for consecutive 2 weeks resulted in decrease of caecal pH and increase of native water concentrations of propionic and butyric acid (p < 0.01), and tendencies for increased acetic acid (p = 0.076), valeric acid (p = 0.061) and decreased A: P ratio (p = 0.093). In faeces, after the change from HZn to LZn, an increase of concentrations of iso- and butyric acid, and valeric acid was recorded (p < 0.05), as well as a tendency for an increase of ammonia concentration (p = 0.095). The detailed results of the tissue Zn and Cu measurements are provided in Table 5. ZnO treatment had an effect on the Zn concentration of liver, kidneys and bone (rib) (p < 0.05), but no effect on the concentration in muscle tissue was measured. The pharmacological level of ZnO (HZn) fed for 4 weeks after weaning resulted in a fourfold increase of the Zn concentration in liver, a threefold increase of Zn in kidneys and an almost twofold increase of Zn in bones in comparison with LZn (p < 0.05). After reduction of the ZnO in the feed from HZn to LZn level and consecutive feeding for 2 weeks, almost half the Zn concentrations could be observed in liver and kidneys compared to the animals of the HZn group after
16
Journal of Animal Physiology and Animal Nutrition © 2015 Blackwell Verlag GmbH
Table 2 Body weight, average daily gain (ADG), average daily feed intake (ADFI) based on dry matter, and feed conversion ratio (FCR) of pigs fed different zinc oxide levels in the diet for 6 weeks Trait* Body weight [kg] At weaning After week 1 After week 2 After week 3 After week 4 After week 5 After week 6 ADG [g] Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 ADFI [g] Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 FCR Week 1 Week 2 Week 3 Week 4 Week 5 Week 6
LZn†
8.48 10.15 11.81 14.82 18.44 22.12 28.04
HZn†
8.51 10.19 12.17 16.02 19.46 23.44 29.36
SEM
0.21 0.27 0.35 0.47 0.57 1.07 1.36
p-value††
0.952 0.944 0.625 0.210 0.385 0.570 0.654
200 237 430 517 648 833
171 282 550 491 657 919
11 11 28 28 23 86
0.224 0.033 0.022 0.676 0.859 0.670
240 360 579 787 926 1184
270 449 635 811 1025 1332
9 18 19 21 44 81
0.080 0.005 0.152 0.600 0.315 0.419
1.21 1.52 1.35 1.54 1.43 1.42
1.62 1.59 1.17 1.68 1.56 1.49
0.10 0.04 0.04 0.06 0.03 0.05
0.031 0.385 0.012 0.289 0.013 0.556
P. Janczyk et al.
Dietary zinc and intestinal microbial activity in pigs
Table 3 Dry matter of digesta and pH, lactic and short-chain fatty acids in native water of caecal digesta and faeces of piglets fed different levels of zinc oxide in the diet from weaning for 4 weeks Caecum Trait DM pH LA AcetA PropA isoButA ButA iValerA ValerA nCapronA A:P NH3
[%] mmol/l
mmol/l
Faeces
LZn*
HZn
pSEM
p value†
LZn
HZn
pSEM
p value
15.4 5.48b 76.27 258.65 156.83 11.61a 80.24a 1.55 14.47a u.d.l. 1.65b 45.87
15.6 6.62a 87.54 255.33 121.08 3.99b 31.72b 1.78 4.00b 0.63 2.11a 36.08
0.68 0.20 21.65 14.72 10.20 1.86 10.44 0.21 2.18 – 0.10 4.15
0.914 0.000 0.822 0.918 0.076 0.029 0.009 0.610 0.005 – 0.006 0.261
25.7 6.89 u.d.l. 418.9a 195.0 19.5a 100.9a 23.2a 28.9a 5.3 2.22a 202.1a
25.0 7.08 19.0 297.1b 162.4 12.0b 60.6b 15.1b 15.5b 1.7 1.85b 136.2b
0.67 0.09 4.96 22.79 11.52 1.38 7.56 1.70 2.73 1.57 0.08 14.46
0.604 0.310 – 0.004 0.163 0.003 0.004 0.012 0.010 0.468 0.025 0.018
DM, dry matter; LA, lactic acid; AcetA, acetic acid; PropA, propionic acid; isoButA, iso-butyric acid; ButA, butyric acid; iValerA, iso-valeric acid; ValerA, valeric acid; nCapronA, n-capron acid; A:P, acetic to propionic acid ratio; P:A, propionic to acetic acid ratio; NH3, ammonia; pSEM, pooled standard error of the mean; u.d.l., under detection limit of 0.1 mmol/l. Significant differences within row (superscript lower case letters) (p < 0.05). *Pigs from LZn group received diet containing 100 mg ZnO/kg diet for 6 weeks; HZn group received diet with 3100 mg ZnO/kg for 4 weeks, then the same diet as LZn for another 2 weeks. †statistically significant differences at P < 0.05 presented in bold. Table 4 Dry matter of digesta and pH, lactic and short-chain fatty acids in native water of caecal digesta and faeces of piglets fed different levels of zinc oxide in the diet from weaning for 6 weeks Caecum Trait DM pH LA AcetA PropA isoButA ButA iValerA ValerA nCapronA A:P NH3
% mmol/l
mmol/l
Faeces
LZn*
HZn
pSEM
p value†
LZn
HZn
pSEM
p value
16.3 5.52 110.1 291.6 188.7 4.6 77.0 1.3 11.4 u.d.l. 1.57 42.5
16.0 5.71 31.9 299.1 182.2 2.2 68.5 1.1 10.7 0.8 1.70 31.0
0.71 0.11 21.80 8.77 9.54 0.75 6.11 0.18 1.74 – 0.11 5.27
0.857 0.390 0.077 0.695 0.757 0.119 0.518 0.588 0.862 – 0.562 0.319
28.1 7.11 u.d.l. 404.3a 176.4 21.0 92.4 25.6 22.2 3.3 2.30a 193.5
31.2 7.05 u.d.l. 333.9b 198.9 18.0 90.6 20.1 25.5 2.1 1.74b 209.8
1. 84 0.06 – 14.75 12.48 1.75 4.39 2.97 2.26 0.66 0.12 28.04
0.435 0.596 – 0.006 0.398 0.427 0.852 0.380 0.487 0.490 0.004 0.790
DM, dry matter; LA, lactic acid; AcetA, acetic acid; PropA, propionic acid; isoButA, iso-butyric acid; ButA, butyric acid; iValerA, iso-valeric acid; ValerA, valeric acid; nCapronA, n-capron acid; A:P, acetic to propionic acid ratio; P:A, propionic to acetic acid ratio; NH3, ammonia; pSEM, pooled standard error of the mean; u.d.l., under detection limit of 0.1 mmol/l. Significant differences within row (superscript lower case letters) (p < 0.05). *Pigs from LZn group received diet containing 100 mg ZnO/kg diet for 6 weeks; HZn group received diet with 3100 mg ZnO/kg for 4 weeks, then the same diet as LZn for another 2 weeks. †statistically significant differences at P < 0.05 presented in bold.
4 weeks of feeding (p ≤ 0.01). Despite this reduction in the Zn tissue concentration in the HZn group, these concentrations remained greater than in the LZn (p < 0.01). The reduction of Zn in the diet did not affected concentrations in bones.
An effect was observed of the ZnO amount in the diet on the Cu concentration in kidneys but not in muscle, liver and bones (Table 5). After 4 weeks of feeding, the HZn diet an almost fourfold increase of the Cu concentration in the kidneys was recorded
Journal of Animal Physiology and Animal Nutrition © 2015 Blackwell Verlag GmbH
17
P. Janczyk et al.
Dietary zinc and intestinal microbial activity in pigs
Table 5 Zn and Cu concentrations in liver, kidney, muscle (ham) and bone (rib) of pigs fed different levels of Zn (as ZnO) in the diet 4 weeks LZn* Zn [mg/kg DM] Liver 234.0 Kidney 122.3 Muscle 69.6 Bone 154.9 Cu [mg/kg DM] Liver 38.14A Kidney 38.82A Muscle 3.42 Bone 3.56
6 weeks HZn
SEM
p value† (LZn vs. HZn, 4 weeks)
LZn*
HZn
SEM
p value† (LZn vs. HZn, 6 weeks)
855.1A 372.7A 66.7 273.4
119.7 49.8 1.9 26.4
0.001 0.003 0.472 0.013
213.3 124.4 66.3 159.8
462.1B 162.0B 61.3 241.2
48.6 7.2 1.9 14.5
0.002 0.001 0.208
ORIGINAL ARTICLE
Effect of high dietary zinc oxide on the caecal and faecal shortchain fatty acids and tissue zinc and copper concentration in pigs is reversible after withdrawal of the high zinc oxide from the diet € ckler1 and A. Zeyner4 €sing2, B. Dobenecker3, K. No P. Janczyk1, K. Bu 1 Unit for Molecular Diagnostics, Genetics and Pathogen Characterisation, Department of Biological Safety, Federal Institute for Risk Assessment Berlin, Germany 2 Chair of Nutrition Physiology and Animal Nutrition, Faculty of Agricultural and Environmental Sciences, University of Rostock, Rostock, Germany 3 Animal Nutrition and Dietetics, Department of Veterinary Science, Ludwig-Maximilians-University Munich, Oberschleißheim, Germany, and 4 Group Animal Nutrition, Institute of Agricultural and Nutritional Sciences, Martin-Luther-University Halle-Wittenberg, Halle (Saale), Germany
Summary Zinc oxide (ZnO) used in high (‘pharmacological’) levels to prevent diarrhoea in pigs is assumed to reduce copper (Cu) in tissues and inhibits large intestinal microbial fermentation. To test it, German Landrace pigs were weaned on d28 of age and fed diets containing either 100 (LowZinc, LZn, n = 10) or 3100 mg ZnO/kg (HighZinc, HZn, n = 10). The mixed feed (13.0 MJ ME, 18.5% crude protein) was based on wheat, barley, soya bean meal and maize. After 4 weeks, the HZn group was further fed 100 mg ZnO/kg for another 2 weeks. Caecal contents, faeces and tissues were collected after 4 weeks (n = 5 and n = 10 respectively) and 6 weeks (n = 5 and n = 5 respectively). Faeces and caecal content were analysed for dry matter (DM), pH, ammonia, lactic acid (LA) and shortchain fatty acids (SCFA) on native water basis. ANOVA was performed to elucidate significant differences at p < 0.05. No diarrhoea occurred. After 4 weeks, the caecal contents’ pH increased (p < 0.001) and butyric (p < 0.05) and valeric acid (p < 0.01) decreased in the HZn group in comparison with LZn. In faeces, a decrease of acetic (p = 0.009), butyric (p = 0.007) and valeric acid (p = 0.046), as well as reduced acetic:propionic acid (A: P) ratio (p = 0.025) was observed in the HZn group in comparison with LZn. Faecal ammonia decreased in HZn (p = 0.018). No differences (p > 0.05) were recorded in caecal contents after 6 weeks. In faeces, acetic acid remained lower in the HZn group in comparison with LZn (p = 0.006), as did the A:P ratio (p = 0.004). Zn concentration in liver, kidneys and ribs, and Cu concentrations in kidneys increased in HZn. Withdrawal of ZnO resulted in reversibility of the changes. The effect on butyric acid should be discussed critically regarding the energetic support for the enterocytes. High Zn and Cu tissue concentrations should be considered by pet food producers. Keywords intestinal microbial activity, pig, short-chain fatty acids, zinc oxide, copper Correspondence Pawel Janczyk, Karl-Heinrich-Ulrichs-Strasse 8A, 10787 Berlin, Germany. Tel: +49-1708095390; Fax: +49-32221007198; E-mail: [email protected] Received: 30 January 2014; accepted: 27 June 2014
Introduction Zinc (Zn) is a component of over 300 enzymes in the mammalian cells and thus an essential micronutrient (Poulsen and Larsen, 1995; Terr es et al., 2001). Growing pigs (up to 20 kg body weight) require approximately 100 mg Zn/kg dry matter (DM) of feed, and later approximately 80 mg Zn/kg DM of feed (Gesellschaft f€ ur Ern€ ahrungsphysiologie, 2006). Because the availability of Zn occurring naturally in feedstuffs may often be reduced by many factors (Smith et al., 1962; Pond et al., 1985;
Poulsen and Carlson, 2001), supplementation of diets with Zn by way of both organic and inorganic sources has become a standard (Richards et al., 2010). Furthermore, supplementation of dietary zinc oxide (ZnO) at high levels (2000–3000 mg/kg) has been used as prophylactic measure against postweaning diarrhoea and to improve growth performance of weaning pigs (Poulsen, 1995; Wang et al., 2010). Next to its direct effect on the Zn homeostasis in the host, it is believed that ZnO acts locally in the intestinal lumen affecting the microbiome (Pieper et al., 2011), but its exact growth-promoting
Journal of Animal Physiology and Animal Nutrition 99 (Suppl. 1) (2015) 13-22 © 2015 Blackwell Verlag GmbH
13
P. Janczyk et al.
Dietary zinc and intestinal microbial activity in pigs
and antidiarrhoea mode of action still remains unclear. Enteric bacteria can be inhibited by increased Zn concentrations (Surjawidjaja et al., 2004), and effects on the composition of ileal core microbiome have been reported (Vahjen et al., 2011). High concentrations of Zn were shown to affect the functionality and the microbial community of soils (Bewley and Stotzky, 1983). So far there are little data on functional changes of the large intestinal microbiome caused by increased ZnO levels in the diet of pigs (Højberg et al., 2005), but no data were found on their reversibility after removal of the excess of ZnO. Few studies investigated the concentrations of Zn in the tissues after feeding high amounts of ZnO (Hahn and Baker, 1993; Poulsen, 1995; Shell and Kornegay, 1996; Jensen-Waeren et al., 1998; Case and Carlson, 2002; Rincker et al., 2005). However, no data could be found for tissue concentrations of Zn when ZnO was fed in high amounts for a period of time with the following reduction of ZnO level in the diet. Excessive Zn in the diet is made responsible for a reduction of the copper (Cu) bioavailability (Hill et al., 1983). Increased dietary ZnO could therefore result in a decrease of Cu in tissues of weaned pigs. Tissue (especially edible as muscles) concentration of Zn and Cu is important for the calculations of dietary intake of these micronutrients, both for humans and especially for pets, as most of the animal by-products such as liver and kidney are taken for pet food production. This study aimed to evaluate the effect of a withdrawal period for 2 weeks following the 4 weeks high-Zn nursery period with large intestinal pH, short-chain fatty acids, lactate and ammonia, as well as tissue Zn and Cu concentrations as the response criteria. Material and methods Animals and treatment
Twenty German Landrace piglets of both sexes were weaned at 28 days of age (8.5 1.0 kg body weight), transported to the experimental facility and allocated to pens, two pigs each (male and female). The piglets received a commercial creep feed from 2 weeks of age (Turbostart, Trede & von Pein, Itzehoe, Germany) that contained 19.0% of crude protein, 7.5% of crude fat, 3.0% of crude fibre, 5.5% of crude ash, 0.7% Ca, 0.6% P, 0.25% Na, 1.5% Lys, 0,5% Met and 14.8 MJ ME/kg. The feed containing the following supplements (calculated values): 22 500 I.U. Vitamin A (E672), 2000 I.U. Vitamin D (E671), 150 mg Fe (FeII-sulphate), 1.0 mg I (potassium iodide), 0.3 mg Co 14
(alcalic Co-II-carbonate), 150 mg Cu (Cu- II-sulphate), 70 mg Mn (Mn-III-oxide), 0.3 mg Se (sodium selenite), 100 mg Zn (Zn sulphate). The study was planned and performed before the ban of the use of Co as supplement for pigs in the EU (European Community, 2013). Five pens formed one group. Wheat–barley–soya– maize diet (Table 1) was prepared to contain 150 mg Zn/kg diet that is the maximal allowed level of Zn according to the EU legislation at that time (European Community, 2003), by adding 100 mg of ZnO per kg of diet (CAS: 1314-13-2; Sigma-Aldrich Chemie GmbH, Germany). This feed was fed to the low-zinc (LZn) group (control). The experimental group received the same diet, but the Zn level was increased by adding 3100 mg ZnO/kg – high zinc group (HZn).
Table 1 Composition of basal and experimental diets used in the study Item [g as fed]
LZn
HZn
Wheat Barley Soya bean meal Maize meal Calcium carbonate Monocalcium phosphate Mineral mix* Salt Lysine-HCl Methionine Soybean oil ZnO Total Nutrients (as calculated) Dry matter [%] Metabolisable energy [MJ] Crude protein [%] Starch [%] Fibre [%] Crude fat [%] Crude ash [%] Lysine [%] Methionine [%] Methionine + Cysteine [%] [mg/kg DM] as analysed Zn Cu Fe
384 303 234 10 18 18 13 2 2 1 15 0.1 1000
384 303 234 7 18 18 13 2 2 1 15 3.1 1000
87.9 13.0 18.5 37.6 3.5 3.4 8.1 1.15 0.35 0.7 179 33 308
87.9 13.0 18.5 37.6 3.5 3.4 8.1 1.15 0.35 0.7 2353 31 308
*Mineral mix contained per kg: semolina bran (36%), NaCl (33.6%), MgO (10.5%), vit. A (600 000 IU), vit. D3 (120 000 IU), vit. E as alpha-tocopherol acetate (8000 mg), vit. K3 as menadione-NaHSO3 (300 mg), vit. B1 as thamin-HCl (250 mg), vit. B2 as riboflavine (250 mg), vit. B6 as pyridoxol-HCl (400 mg), vit. B12 (2000 lg), nicotin acid (2500 mg), folic acid (100 mg), biotine (25 000 lg), Ca-D-panthotenate (1000 mg), choline-Cl (80 000 mg), MnO (6000 mg), FeCO3 (5000 mg), CuSO4 9 5H2O (1000 mg), CoSO4 9 7H2O (30 mg), Ca(IO3)2 (45 mg), Na2SeO3 (35 mg).
Journal of Animal Physiology and Animal Nutrition © 2015 Blackwell Verlag GmbH
P. Janczyk et al.
To investigate the reversibility of potential changes caused by the high content of ZnO, after 4 weeks of treatment, the HZn group received the same feed as LZn for another 2 weeks. Feed was fed in meal form (88% DM) and offered mixed with some water twice daily for an hour as semi ad libitum, to avoid refusals. Water was provided ad libitum via nipple drinkers. Ambient temperature was kept at 25 1 °C for the first 4 weeks, and then reduced to 22 1 °C, with humidity 30–55% and light regime of 12 h light and 12 h darkness. Pigs were weighed once a week before the morning feeding. Feed intake was recorded daily on dry matter basis for each pen. Average daily gain (ADG), average daily feed intake (ADFI) and feed conversion ratio (FCR) were calculated. Faeces were collected after 4 and 6 weeks of treatment, directly from the rectum, and kept cooled on ice until further processing. At these time-points, five pigs from each group were euthanized by overdose of pentobarbital under general azaperon (Stresnil; Janssen Animal Health, Neuss, Germany) and ketamine (Ketamin 10%; Bremer Pharma GmbH, Warburg, Germany) anaesthesia; the feed was withdrawn 2 h before the euthanasia. Total digesta from caecum was collected. The whole liver, both kidneys, left ham and X-XIIth ribs were dissected for analysis of Zn and Cu concentrations. The animal experiment was approved by the local authority (Landesamt f€ ur Gesundheit und Soziales, LAGeSo, Berlin) under the accession number G 0349/ 09. Analysis of digesta and faeces
Faeces and caecal digesta were analysed for DM, pH, ammonia, lactic acid (LA) and short-chain fatty acids (SCFA), namely acetic, butyric, propionic, valeric, and capronic acid and their isoforms. DM was determined by freezing fresh digesta or faeces at 20 °C with final lyophilisation. For other analyses, fresh samples were diluted 1:3 with sterile distilled water (20 g sample, 60 ml water) and homogenised in BagMixer (Interscience, Saint Nom, France) at maximum speed for 30 s. Forty-five millilitre of the homogenate was transferred into 50 ml-Falcon tube and pH was measured using Digital-pH-Meter 646 (Knick, Berlin, Germany). Further, the homogenates were centrifuged at 4000 g for 10 min, at 20 °C. Supernatant was collected and stored at 20 °C before further analyses. Within the thawed samples, the contents of LA and SCFA were analysed by HPLC and gas chromatography, respectively, as described (Hackl et al., 2010). The content of ammonia was determined by the Journal of Animal Physiology and Animal Nutrition © 2015 Blackwell Verlag GmbH
Dietary zinc and intestinal microbial activity in pigs
modified microdiffusion method (Voigt and Steger, 1967). All concentrations were calculated as mmol of substance per litre of native water content in caecum digesta/faeces according to Zeyner et al. (2004). The ratio of acetic to propionic acid (A:P) was calculated. Zinc and copper analysis
Dissected organs, ham and ribs were weighed and homogenised using a kitchen mill (Moulinex, SEB S.A., Ecully Cedex, France). The homogenised tissues were put into aluminium bowls and frozen at 20 °C. They were subsequently lyophilised, packed hermetically and sent to the analytic laboratory. The determination of Zn and Cu in the homogenised and lyophilised tissues was performed in an acetylene flame using atomic absorption spectrometry (PerkinElmer Inc., Waltham, Massachusetts, USA) after wet hydrolysis with HNO3 in a microwave (Milestone Inc., Shelton, Connecticut, USA). Statistical analysis
Levene test was applied for testing the homogeneity of variance of all traits. Effect of different ZnO levels on the body weight of the pigs was analysed as analysis of variance (ANOVA) with repeated measures with pig as experimental unit. The effect of ZnO on DM, pH, ammonia, LA and SCFA, as well as on tissue Zn and Cu concentrations, were tested by t-test to elucidate which differences were responsible for the observed effects. The calculations were performed using SPSS for Windows version 12.0.2 (The Appache Software Foundation; IBM Corp., Armonk, NY, USA). Mean values with pooled standard error of the mean (pSEM) are provided in the tables. Differences for the traits within groups and between both time-points were analysed performing also a t-test. When t-test was performed, the Bonferroni correction was applied. Differences between mean trait values were considered significant at p < 0.05. Tendency for differences was considered at 0.05 < p ≤ 0.1. Results The mean analysed Zn concentration in the feed was 180 mg/kg DM (160 mg/kg feed) in LZn and 2400 mg/kg DM (2250 mg/kg feed) in HZn (Table 1). The Cu concentration in the feed was analysed to be 33 and 31 mg/kg DM in LZn and HZn respectively (Table 1). All piglets were in good condition throughout the time of the treatment; no diarrhoea occurred. 15
P. Janczyk et al.
Dietary zinc and intestinal microbial activity in pigs
There was no effect of ZnO on the body weight. However, the pigs from HZn weighed numerically more than pigs from LZn from d53 onwards (Table 2). During the 2nd and 3rd week, the HZn gained more than LZn (p < 0.05). ADFI was higher in the HZn during the 2nd week (p < 0.05). The FCR was higher in HZn during the 1st week and lower in the HZn during the 3rd and 5th week (Table 2). After 4 weeks of the experimental period, pH of the caecal digesta was greater in the HZn in comparison with LZn (p < 0.001). Whereas (iso-)butyric and valeric acid decreased in the HZn (p < 0.05), propionic acid tended to decrease (0.05 < p < 0.1) but lactic and acetic acid were not affected there (p > 0.1). The con-
SEM, standard error of the mean. *ADG was calculated using pig as experimental unit. ADFI and FCR were calculated using pen as experimental unit. †LZn received diet containing 100 mg ZnO/kg diet for 6 weeks; HZn group received diet with 3100 mg ZnO/kg diet for 4 weeks, then the same diet as LZn group for another 2 weeks. ††statistically significant differences at P < 0.05 presented in bold.
tents of DM and ammonia were neither affected (p > 0.1). A weak negative correlation was calculated between the pH and the total SCFA in caecum (r2 = 0.461), and the strongest correlation was observed between the butyric acid concentration and pH (r2 = 0.572). In faeces, acetic and (iso-)butyric, (iso-) valeric acid decreased in the HZn (p ≤ 0.01). Faecal propionic acid concentration was not affected (p > 0.1). The A:P ratio in caecum was greater in HZn in comparison with LZn (p < 0.05). In faeces, an opposite was observed – A:P ratio was lower in HZn (p < 0.05). Faecal ammonia concentration decreased in HZn (p < 0.05). No correlation between the total SCFA and pH was detected (r2 = 0.147). All traits are summarised in Table 3. After the change of diet in the HZn group to the feed administered to the LZn group and feeding the animals for two consecutive weeks, no differences were recorded in the caecal digesta between the groups. In faeces, again, acetic acid was lower in the HZn in comparison with LZn (p < 0.01). Similar to the results obtained after 4 weeks, in faeces, the A:P ratio was lower in HZn in comparison with LZn (p < 0.01). No further differences were recorded (Table 4). Comparison of the data obtained after 4 and 6 weeks of feeding within the groups revealed no differences for LZn except for isobutyric acid, which decreased in caecal digesta in the LZn group (p < 0.05). The change of the diet from HZn to LZn and feeding it for consecutive 2 weeks resulted in decrease of caecal pH and increase of native water concentrations of propionic and butyric acid (p < 0.01), and tendencies for increased acetic acid (p = 0.076), valeric acid (p = 0.061) and decreased A: P ratio (p = 0.093). In faeces, after the change from HZn to LZn, an increase of concentrations of iso- and butyric acid, and valeric acid was recorded (p < 0.05), as well as a tendency for an increase of ammonia concentration (p = 0.095). The detailed results of the tissue Zn and Cu measurements are provided in Table 5. ZnO treatment had an effect on the Zn concentration of liver, kidneys and bone (rib) (p < 0.05), but no effect on the concentration in muscle tissue was measured. The pharmacological level of ZnO (HZn) fed for 4 weeks after weaning resulted in a fourfold increase of the Zn concentration in liver, a threefold increase of Zn in kidneys and an almost twofold increase of Zn in bones in comparison with LZn (p < 0.05). After reduction of the ZnO in the feed from HZn to LZn level and consecutive feeding for 2 weeks, almost half the Zn concentrations could be observed in liver and kidneys compared to the animals of the HZn group after
16
Journal of Animal Physiology and Animal Nutrition © 2015 Blackwell Verlag GmbH
Table 2 Body weight, average daily gain (ADG), average daily feed intake (ADFI) based on dry matter, and feed conversion ratio (FCR) of pigs fed different zinc oxide levels in the diet for 6 weeks Trait* Body weight [kg] At weaning After week 1 After week 2 After week 3 After week 4 After week 5 After week 6 ADG [g] Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 ADFI [g] Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 FCR Week 1 Week 2 Week 3 Week 4 Week 5 Week 6
LZn†
8.48 10.15 11.81 14.82 18.44 22.12 28.04
HZn†
8.51 10.19 12.17 16.02 19.46 23.44 29.36
SEM
0.21 0.27 0.35 0.47 0.57 1.07 1.36
p-value††
0.952 0.944 0.625 0.210 0.385 0.570 0.654
200 237 430 517 648 833
171 282 550 491 657 919
11 11 28 28 23 86
0.224 0.033 0.022 0.676 0.859 0.670
240 360 579 787 926 1184
270 449 635 811 1025 1332
9 18 19 21 44 81
0.080 0.005 0.152 0.600 0.315 0.419
1.21 1.52 1.35 1.54 1.43 1.42
1.62 1.59 1.17 1.68 1.56 1.49
0.10 0.04 0.04 0.06 0.03 0.05
0.031 0.385 0.012 0.289 0.013 0.556
P. Janczyk et al.
Dietary zinc and intestinal microbial activity in pigs
Table 3 Dry matter of digesta and pH, lactic and short-chain fatty acids in native water of caecal digesta and faeces of piglets fed different levels of zinc oxide in the diet from weaning for 4 weeks Caecum Trait DM pH LA AcetA PropA isoButA ButA iValerA ValerA nCapronA A:P NH3
[%] mmol/l
mmol/l
Faeces
LZn*
HZn
pSEM
p value†
LZn
HZn
pSEM
p value
15.4 5.48b 76.27 258.65 156.83 11.61a 80.24a 1.55 14.47a u.d.l. 1.65b 45.87
15.6 6.62a 87.54 255.33 121.08 3.99b 31.72b 1.78 4.00b 0.63 2.11a 36.08
0.68 0.20 21.65 14.72 10.20 1.86 10.44 0.21 2.18 – 0.10 4.15
0.914 0.000 0.822 0.918 0.076 0.029 0.009 0.610 0.005 – 0.006 0.261
25.7 6.89 u.d.l. 418.9a 195.0 19.5a 100.9a 23.2a 28.9a 5.3 2.22a 202.1a
25.0 7.08 19.0 297.1b 162.4 12.0b 60.6b 15.1b 15.5b 1.7 1.85b 136.2b
0.67 0.09 4.96 22.79 11.52 1.38 7.56 1.70 2.73 1.57 0.08 14.46
0.604 0.310 – 0.004 0.163 0.003 0.004 0.012 0.010 0.468 0.025 0.018
DM, dry matter; LA, lactic acid; AcetA, acetic acid; PropA, propionic acid; isoButA, iso-butyric acid; ButA, butyric acid; iValerA, iso-valeric acid; ValerA, valeric acid; nCapronA, n-capron acid; A:P, acetic to propionic acid ratio; P:A, propionic to acetic acid ratio; NH3, ammonia; pSEM, pooled standard error of the mean; u.d.l., under detection limit of 0.1 mmol/l. Significant differences within row (superscript lower case letters) (p < 0.05). *Pigs from LZn group received diet containing 100 mg ZnO/kg diet for 6 weeks; HZn group received diet with 3100 mg ZnO/kg for 4 weeks, then the same diet as LZn for another 2 weeks. †statistically significant differences at P < 0.05 presented in bold. Table 4 Dry matter of digesta and pH, lactic and short-chain fatty acids in native water of caecal digesta and faeces of piglets fed different levels of zinc oxide in the diet from weaning for 6 weeks Caecum Trait DM pH LA AcetA PropA isoButA ButA iValerA ValerA nCapronA A:P NH3
% mmol/l
mmol/l
Faeces
LZn*
HZn
pSEM
p value†
LZn
HZn
pSEM
p value
16.3 5.52 110.1 291.6 188.7 4.6 77.0 1.3 11.4 u.d.l. 1.57 42.5
16.0 5.71 31.9 299.1 182.2 2.2 68.5 1.1 10.7 0.8 1.70 31.0
0.71 0.11 21.80 8.77 9.54 0.75 6.11 0.18 1.74 – 0.11 5.27
0.857 0.390 0.077 0.695 0.757 0.119 0.518 0.588 0.862 – 0.562 0.319
28.1 7.11 u.d.l. 404.3a 176.4 21.0 92.4 25.6 22.2 3.3 2.30a 193.5
31.2 7.05 u.d.l. 333.9b 198.9 18.0 90.6 20.1 25.5 2.1 1.74b 209.8
1. 84 0.06 – 14.75 12.48 1.75 4.39 2.97 2.26 0.66 0.12 28.04
0.435 0.596 – 0.006 0.398 0.427 0.852 0.380 0.487 0.490 0.004 0.790
DM, dry matter; LA, lactic acid; AcetA, acetic acid; PropA, propionic acid; isoButA, iso-butyric acid; ButA, butyric acid; iValerA, iso-valeric acid; ValerA, valeric acid; nCapronA, n-capron acid; A:P, acetic to propionic acid ratio; P:A, propionic to acetic acid ratio; NH3, ammonia; pSEM, pooled standard error of the mean; u.d.l., under detection limit of 0.1 mmol/l. Significant differences within row (superscript lower case letters) (p < 0.05). *Pigs from LZn group received diet containing 100 mg ZnO/kg diet for 6 weeks; HZn group received diet with 3100 mg ZnO/kg for 4 weeks, then the same diet as LZn for another 2 weeks. †statistically significant differences at P < 0.05 presented in bold.
4 weeks of feeding (p ≤ 0.01). Despite this reduction in the Zn tissue concentration in the HZn group, these concentrations remained greater than in the LZn (p < 0.01). The reduction of Zn in the diet did not affected concentrations in bones.
An effect was observed of the ZnO amount in the diet on the Cu concentration in kidneys but not in muscle, liver and bones (Table 5). After 4 weeks of feeding, the HZn diet an almost fourfold increase of the Cu concentration in the kidneys was recorded
Journal of Animal Physiology and Animal Nutrition © 2015 Blackwell Verlag GmbH
17
P. Janczyk et al.
Dietary zinc and intestinal microbial activity in pigs
Table 5 Zn and Cu concentrations in liver, kidney, muscle (ham) and bone (rib) of pigs fed different levels of Zn (as ZnO) in the diet 4 weeks LZn* Zn [mg/kg DM] Liver 234.0 Kidney 122.3 Muscle 69.6 Bone 154.9 Cu [mg/kg DM] Liver 38.14A Kidney 38.82A Muscle 3.42 Bone 3.56
6 weeks HZn
SEM
p value† (LZn vs. HZn, 4 weeks)
LZn*
HZn
SEM
p value† (LZn vs. HZn, 6 weeks)
855.1A 372.7A 66.7 273.4
119.7 49.8 1.9 26.4
0.001 0.003 0.472 0.013
213.3 124.4 66.3 159.8
462.1B 162.0B 61.3 241.2
48.6 7.2 1.9 14.5
0.002 0.001 0.208
