Inclusion of wheat bran in barley-soybean meal diets with different phosphorus levels for growing-finishing pigs I . Effects on nutrient digestibility and mineral balances in finishing pigs

Helander, E„ Näsi, M. & Partanen, K. 1994. Inclusion of wheat bran in barley-soybean meal diets with different phosphorus levels for growing-finishing pigs. I. Effects on nutrient digestibility and mineral balances in finishing pigs. Agricultural Science in Finland 3: 27-39. (Department of Animal Science, P.O. Box 28, FIN-00014 University of Helsinki, Finland. Present address: Eija Helander, Cultor Ltd, P.O. Box 105,FIN-00241 Helsinki, Finland.)


Introduction
More than half of the phosphorus (P) in feedstuffs of plant origin occurs in the form of phytate, which is poorly available to non-ruminants (CROMWELL 1992).Phytic acid readily forms complexes with several minerals, like calcium (Ca) and iron (Fe) (Harris 1955),zinc (Zn) and manganese (Mn) (ref. Erdman 1979) as well as with proteins (Scheuer- mann et al. 1988 a), thus reducing their availability to animals.High levels of phytate in the diet can also cause magnesium (Mg) deficiency (Pomer- ANZ 1978).Pig diets are generally supplemented with inorganic P to ensure that the P requirement is satisfied.If the phytate P of plants could be made more available, a reduction in inorganic P supplementation of the feeds would be possible, making P-emissions through manure decrease consider- ably.
Phytase catalyzes the hydrolysis ofphytic acid to inositol and orthophosphate.The intrinsic phytase activity of different plants varies considerably.
Wheat is one of the feedstuffs which has a high phytase activity ranging from 300 to 2000 lU/kg (1 IU is 1 pmol P liberated from sodium phytate/min at 37°C) (Pointillart 1988).Most of its phytase appears in the surface layers (Pomeranz 1978).In an experiment by Scheuermann et al. (1988 a), wheat phytase was found to hydrolyze maize phytate to the same extent as wheat phytate.Bagheri and Gueguen (1985) have established that the utilization of P improved when an oats-soybean meal diet was supplemented with 200 g of wheat bran (WB) per kg.WB addition also improved the utilization of Mg but decreased that of Ca and Zn.
The present work was undertaken to evaluate whether the intrinsic phytase ofWB could improve the digestibility of phytic P in commercial barleysoybean meal diets enough to serve as a partial or complete replacement for the added inorganic P in the diets of growing-finishing pigs.

Material and methods
The digestibility and balance assay was conducted with six Large White x Landrace barrows using a 2x3 factorial arrangement in a 6x5 cyclic change-over design.The factors were WB inclusion (0 or 100 g per kg, later referred to as WB-and WB+, respectively) and three P levels, high (HP), medium (MP) and low (LP) corresponding to 4.33 g(3/3), 2.99 g (2/3) and 1.64 g (1/3) digestible P per feed unit (FT) = 0.7 kg starch equivalent), respectively.Thus, the experiment consisted of six different diets: HPWB-, HPWB+, MPWB-, MPWB+, LPWB-and LPWB+.The nutrient digestibility, nitrogen (N) balance and the balance of P, calcium (Ca), magnesium (Mg), potassium (K) and zinc (Zn) were determined.The pigs were kept in meta- bolism cages made of galvanized iron, which al- lowed separate quantitative collection offaeces and urine.Each period lasted 10 days; five days of adjustment and five days of collection.The initial weight of the pigs was 70.6 kg (SE 3.57) and the final weight 122.5 kg (SE 5.52).
The feeds were pelleted (65°C, 4 mm diameter) barley-soybean meal complete mixtures.The com- position of the experimental diets is shown in Table I.HP diets have been found adequate for pigs with respect to all nutrients (Salo et al. 1990), and the others with respect to all other nutrients except P. The inorganic source of P in the present experiment was dicalciumphosphate.The energy content was 0.99 FU/kg in WB-diets and 0.96 FU/kg in WB+ diets.The intention was to give the same    Lysine 111111 HP = high P, MP = medium P, LP = low P, WB-= no wheat bran, WB+ = 100 g wheat bran/kg diet.
amount of FUs/d to each pig (ranging from 2.6 FU to 3.0 FU/pig/d in periods 1 to 5, respectively) but, in practice, the pigs on WB-diets got 0.16 FU/d more than the other pigs.Just before feeding, the feed was mixed with water (one litre water per kg feed) and afterfeeding the pigs were given water ad libitum.
The feeds and faeces were analyzed by standard methods (AOAC 1984).Amino acids were assayed with a Beckman 6300 amino acid analyzer.ICP- AES equipment was used in phytic acid determina- tion (Flaami and Kumpulainen 1991).The sample was first burned by inductively coupled plasma and then the P content of the sample was indirectly measured by atomic emission spectrophotometry.P from feeds, water and faeces was analyzed col- orimetrically after dry ashing by the vanadomolybdate procedure of Tayssky and Shorr (1953).The other minerals of the feeds, drinking water, faeces and urine were measured with a Perkin-Elmer 5100 PC atomic-absorption spectrophotometer, except K, which was assayed with a Corning 435 flame photometer.Phytase activity was measured as free phosphate from phytate after incubating the sample in a 0.1 M sodium acetate buffer, pH 5.0, at 35°C for 30 min.A phytase unit (U) is defined as the amount of enzyme that liberates 1 pmol of inor- ganic P from sodium-phytate in one minute.Phytase activity was measured from the main raw ma- terials and from the complete feeds.
The data were subjected to an analysis of vari- ance using the following model (Snedecor and Cochran 1989): Yijk H + Ai + Pj + Tk + eijk where p = overall mean, Ai = the effect ofanimal i, Pj = the effect of period j, Tk = the effect of treat- ment k and eijk = residual error.
In analyzing the crude protein digestibility, N intake was used as a covariate.All other results were first corrected by using FU intake as a covari- ate, but because it did not have any significant effect on the digestibility and retention of the nutri- ents, this covariate was omitted.The degrees of freedom for treatment effects were further parti-tioned into single degrees offreedom by the follow- ing orthogonal contrasts: Cl = WB-vs.WB+, C 2 = linear effect of P level, C 3 = quadratic effect of P level, C 4 = interaction Cl x C 2, C 5 = interaction Cl x C 3.

Chemical analyses
The analyzed chemical composition of the experimental diets is presented in Table 2.Although the experimental feeds were produced in the normal production line of a big feed factory, not designed for mixing small feed batches, the chemical composition of the diets was relatively close to the target.Only the protein and amino acid contents of the diets were sligtly higher than targeted and, there- fore, the HPWB+ and MPWB+ diets contained more protein per FU than the other diets.The hemi- cellulose content varied from 101 g/kg DM to 130 g/kg DM, naturally being higher on WB+ diets.
Table 3 shows the P, Ca and phytase content of the diets and feed ingredients.The P content of barley was 3.5 g/kg DM and two thirds of it was of phytic origin, which is in accordance with the figures reported earlier by PoiNTILLART (1988) and JONGBLOED et al. (1991).In WB, 0.89 of total P was of phytic origin.PoiNTILLART (1988) has re- ported the corresponding figure to be 0.75 for wheat bran and Jongbloed et al. (1991) 0.80 for wheat middlings.According to a review by House- man and de Bruyne (1989), phytate P is located in the aleurone and pericarp layers in monocotyledons.Thus, the content of phytate P in wheat bran is highly dependent on the technical process it comes from.
The P contents in HPWB-and HPWB+ diets were 8.2 and 8.0 g/kg DM, in MPWB-and MPWB+ diets 6.1 and 6.2 g/kg DM, and in LPWBand LPWB+ diets 4.4 and 4.9 g/kg DM, respec- tively.The phytic P content of the experimental diets varied from 0.305 to 0.621 in total P, being always higher on WB+ diets.The digestible P con- tent of the diets was calculated by using the deter- mined digestibility coefficients (Table 6).Digest-Table 2. Chemical composition (g/kg DM) and calculated energy content of the experimental diets.12.0 HP = high phosphorus (P), MP = medium P, LP = low P, WB-= no wheat bran, WB+ = 100 g wheat bran/kg diet. 2 FU = 0.7 kg starch equivalent.Table 3. Phosphorus, calcium and phytase contents of the experimental diets and feed ingredients (g/kg DM) HP = high phosphorus (P), MP = medium P, LP = low P, WB-=no wheat bran, WB + = 100 g wheat bran/kg diet.
ible P ranged from 4.2 to 1.4 g/kg DM.On a LPWB+ diet, the measured digestible P content was 0.4 g/kg DM (28%) higher than on a LPWBdiet.This difference cannot be explained by the differences in the total P content of the diets, al- though theLPWB+ diet contained 11% more P than the LPWB-diet.The difference may rather be at- tributed to WB phytase.The phytase activity of WB was 2800 U/kg.Pointillart (1988) has measured activities between 600 and 1700 lU/kg for wheat bran.The highest enzyme activities in wheat are found from its surface layers (Pomeranz 1978); thus, the processing technology affects the phytase activity of the end product.The phytase activity of barley was the same as in another experiment by Helander (1993, preliminary results).Soybean meal did not show any phytase activity.COSGROVE and IRVING (1980) have reported phytase activity in soybeans, and Bagheri and Gueguen (1982) have also indicated SBM to have some phytase activity.It is likely that the phytase activity in SBM depends on the temperature of the process SBM comes from.
There is some inaccuracy in the figures concern- ing the phytase activitity of the diets; e.g., 100 g of WB per kg should have increased the phytase activ- ity of WB+ diets by 280 U/kg.No such increase, however, could be found, which lends support to activity loss during pelleting.Steam pelleting at 80°C reduced the phytase activity in an experiment by Jongbloed and Kemme (1990), and Schwarz and Schöner (1991) reported 15-25 % losses in phytase activity when the temperature of the pellets was 70°C.Since the pelleting temperature in the present trial was lower, this does not explain the smaller-than-expected differences between the treatments.It is probably more a question of an analytical problem: when the phytase activity of the feeds is low, the sensitivity of the analytical method may not be sufficient (Puhakka 1993, personal communication).In any case, the phytase activity was always higher on WB+ diets than on other diets.

Digestibilities and balances
The average dry matter intake of the pigs was 2526 g (82 g/VT 15 ) on WB-diets and 2435 g (79 g/W 0 ' 75 ) on WB+ diets.The difference in daily energy intake was 0.16 FU.The pigs on WB+ diets ate on average 38 g more hemicellulose and 8.6 g more cellulose per day in spite of their lower total feed intake.WB inclusion was found to impair the dry matter digestibility (pcO.OOl) (Table 4).There tended to be an interaction between WB inclusion and P level in organic matter digestibility: the di- gestibility decreased linearly on WB+ diets (p<0.05) when the P level of the diets decreased and tended (p<o.l) to decrease quadratically on WBdiets.No effect on ash digestibility could be found due to treatments.Diggs et al. (1965) have reported WB to decrease the metabolizable energy content of a diet.Also, other fibre sources are known to decrease digestibility: the faecal digestibility of or- ganic matter was reduced by the inclusion of 50 g cellulose or 50 g of straw meal per kg diet in an experiment conducted by Den Hartog et al.The amounts of excreted N and digested N were affected by the daily N intake.Correcting the di- gested N by N intake decreased the digestibility coefficient on all WB-diets and increased it on WB+ diets (Table 5).Since phytic acid readily forms complexes with protein, it was interesting to note that the crude protein digestibility improved significantly (p<0.01) by WB inclusion.This may be due to the intrinsic phytase of WB as in the experiment by Mroz et al. (1991), where a significant improvement in ileal amino acid digestibility by microbial phytase addition was found.Neither N excretion in urine nor N retention was significantly affected by daily N intake; however, N excretion in urine was always lower (p<0.05) on WB+ diets.N retention tended to be lower on HPWB+ and MPWB+ diets than on corresponding diets without HP = high phosphorus (P), MP = medium P, LP = low P, WB-= no wheat bran, WB+ = 100 g wheat bran/kg diet.
The average daily intake of P ranged from 20.5 g/d on a HPWB+ diet to 12.5 g/d on a LPWB-diet (Table 6).The dicalciumphosphate content of the diets varied from 16 g/kg in a HPWB-diet to 1 g/kg on a LPWB-diet.On LPWB+ diets, all P was of plant origin.Phytic P ranged from 0.305 to 0.560 and from 0.388 to 0.621 of total P in WB-and WB+ diets, respectively.Although the excretion of P in faeces decreased linearly (p<0.01) with a decreas- ing P content of the diet, the differences in excre- tion between treatments were not significant.P ex- cretion ranged from 9.88 g/d on a HPWB-diet to 8.22 g/d on a LPWB+ diet.Also the apparent di- gestibility of P (% of intake), decreased linearly (p<0.001) with a decreasing P content of the diet.In the previously mentioned digestibility and balance study conducted by Näsi and Helander (1993), the faecal excretion of P was about twice as high in diets with inorganic P supplementation as com- pared to unsupplemented diets.The absorption of P was then on an equal level (3.8-3.9 g/d) with P absorption on the LPWB-diet (4.0 g/d) in this study, although the daily intake of P in the present experiment was one-third higher.A possible reason for the low absorption rate of P in the present experiment may be related to the age of the pigs.In Table 6.Excretion, apparent digestibility and retention of phosphorus in pigs fed on the experimental diets.NS NS HP = high phosphorus (P), MP = medium P, LP = low P, WB-= no wheat bran, WB+ = 100 g wheat bran/kg diet. 2 (P < 0.10) o, (P < 0.05) *, (P < 0.01) **, (P < 0.001) ***, (P > 0.10) NS Cl = WB-vs.WB + , C 2 = P lin, C 3 = P quadr, C 4 = Cl x C2, C 5 = Cl x C3 the earlier study, the pigs were younger and, thus, weighed less (weight 36-88 kg) than the pigs in this study (71-122 kg).Jongbloed (1987) reviewed the adaptation of young animals to widely differing amounts of P and Ca in the diet by varying their intestinal absorption, and found that a decrease in the adaptation of the intestine to dietary P and Ca restriction occurs with increasing age.This is supported by the findings of NÄSI (1990): in older pigs the retention of P remained low in spite of a high P supply on a maize-soybean meal diet.Another ex- planation for the lower absorption rate of P on a LPWB-diet as compared to the earlier results may be the less favourable Ca:P ratio in this experiments (2.3 vs 1.8).A wide Ca:P ratio has an important negative influence on a low-phosphorus diet (De Wilde and Jourquin 1992).Jongbloed (1987) concludedin his trials that the higher the Ca content of the diet, the more the absorption percentage ofP and Ca decreased at increasing live weights.
Phosphorus homeostasis appears to be achieved mainly by regulating excretion via urine (Linder  1991).When the P supply in the diet is low, the animal tries to retain the plasma P level constant by reabsorbing P from the kidneys.In the present study, the urinary P excretion decreased linearly (pcO.001) with a decreasing P content of the diet.The proportion of absorbed P that was retained increased linearly (p<0.001) when the P content of the diet decreased, ranging from 0.620 on a HPWB+ diet to 0.946 on a LPWB+ diet.This is in agreement with earlier observations (NÄSI 1990,  NÄSI and Helander 1993).
The inclusion of WB in the diet did not improve the digestibility of P.However, the amount of P excreted in urine was non-significantly lower on WB+ diets.The absorption ofP appeared to be 0.75 g/d (18.8%) higher on the LPWB+ diet compared to the LPWB-diet.The proportion of absorbed P that was retained was 7.1 percentage units, i.e. almost 1 g/d higher on the LPWB+ diet than on the LPWBdiet.The P intake was 0.5 g/d higher on a LPWBdiet, but this hardly explains the differences in ab- sorption and retention.Thus, it can be assumed that WB (phytase) had improved the utilization of phytic P on the low-P diet.The apparent P absorption was found to improve by 21% in a study by Newton et al. (1983), when a corn-soybean meal diet with a normal P level was supplemented with 100 g WB.Those results are, however, not fully HP = high phosphorus (P), MP = medium P, LP = low P, WB-= no wheat bran, WB + = 100 g wheat bran/kg diet. 2 (P < 0.10) o, (P < 0.05) *, (P < 0.01) **, (P < 0.001) ***, (P > 0.10) NS Cl = WBvs.WB + , C 2 = P lin, C 3 = P quadr, C 4 = Cl x C2, C 5 = Cl x C3 comparable with ours, because WB inclusion also enhanced the total P intake.Pointillart (1991) reported that P from a diet to which no inorganic P was added and which was supplemented by 20% of rye bran, was better absorbed (55 vs 36%) and retained (50 vs 36%) by pigs of 12-43 kg live weight than P from the control diet.
The Ca intake of the pigs ranged between 22-23 g/d (Table 7).Ca digestibility was not affected by WB inclusion or by decreasing the P content of the diet, whereas urinary Ca excretion increased line- arly (p<0.01) and quadratically (p<0.05) with a decreasing dietary P content.The results of the present study show that Ca balance is mainly regulated through the intestine as was also concluded by Fernandez (1992).In an experiment by Den Har- tog et al. (1988) apparent faecal digestibility of Ca decreased, when the diet was supplemented with 50 g pectin, cellulose or straw meal per kg.No significant differences were found in Ca retention due to treatments.However, the retention tended to be lower on LP than on other diets.Jongbloed(1987) reported the daily Ca retention to vary from 6 to 8 g after 55 kg live weight, which is well in accordance with our results.The proportion of digested Ca that was retained decreased linearly (p<0.01) when the P content of the diet decreased.In our previous digestibility and balance study (NÄSI and Helander 1993), Ca absorption was lower in a P-supplemented diet compared to an unsupplemented diet, but no significant differences were found in Ca retention because the urinary Ca excre- tion was also lower.POINTILLART (1991) re- ported no change in Ca absorption (53 vs. 50%) but higher Ca retention (52 vs. 45%) on a diet with 200 g inclusion of rye bran per kg feed than on a diet without any inorganic P supplement.
Mg intake was higher (p<0.001) on WB+ diets and decreased when the dietary P level decreased (pcO.001)(Table 8).This indicates that both WB and dicalciumphosphate have contributed some Mg to the diet.Daily Mg absorption did not differ be- tween treatments.According to Linder (1991), dietary Ca and P do not affect the absorption ofMg

NS NS
• HP = high phosphorus (P), MP = medium P, LP = low P, WB-= no wheat bran, WB + = 100 g wheat bran/kg diet. 2 (P < 0.10) o, (P < 0.05) *, (P < 0.01) »*, (P < 0.001) ***, (P > 0.10) NS Cl = WB-vs.WB + , C 2 = P lin, C 3 = P quadr, C 4 = Cl x C2, C 5 = Cl x C3 in humans (the studies have, however, usually been done with animals).Mg homeostasis occurs simi- larly as with P, and thus, mainly via adjustments of urinary excretion in humans (Linder 1991).Ac- cording to Crenshaw (1991) the regulation of body Mg levels in pigs is not very well understood.In a study by Newton et al. (1983) WB inclusion was not found to affect the apparent digestibility of Mg in pigs in a study by Newton et al. (1983).In the present experiment, Mg excretion in urine increased (p<0.001) with a lower dietary P content.An interaction between WB inclusion and P level was recorded in Mg retention: on WB+ diets Mg retention decreased linearly (p<0.05) and on WBdiets both linearly (p<0.05) and curvilinearly (p<o.()s) with a decreasing P content.These results contradict those ofa study by NÄSI (1990), in which Mg absorption improved and retention remained unchanged on a unsupplemented maize-soybean meal diet compared with a diet supplemented by inorganic P. The results are, however, well in ac- cordance with those of an earlier study by NÄSI and Helander (1993), where barley-soybean meal diets were used.
K digestibility was not affected by WB inclusion (Table 9).This result is in conflict with the results of Newton et al. (1983), who reported that the apparent absorption of K decreased significantly when the diet was supplemented by 100 g of WB/kg and non-significantly when supplemented by 200 g of WB/kg.Urinary K excretion was increased slightly when WB was included in the diet.In K retention, an interaction was observed between WB and P level: on WB-diets the retention de- creased curvilinearly (p<0.05) while on WB+ diets the retention remained constant when the P level of the diet decreased.
Due to unknown reasons there was a large varia- tion in zinc (Zn) intake between treatments (Table 9).A high Zn intake led to a higher Zn excretion in faeces.The daily Zn absorption and retention were the highest on MPWB-and MPWB+ diets and were unaffected by WB inclusion.In retention, again, an interaction was found between Table 9. Apparent digestibility and retention of potassium and zinc in pigs fed on the experimental diets.HP = high phosphorus (P), MP = medium P, LP = low P, WB-= no wheat bran, WB+ = 100 g wheat bran/kg diet. 2 (P < 0.10) o, (P < 0.05) *, (PcO.Ol) **, (P < 0.001) ***, (P > 0.10) NS Cl = WB-vs.WB +, C 2 = P lin, C 3 = P quadr, C 4 = Cl x C2, C 5 = Cl x C3 WB inclusion and P level: on WB+ diets Zn reten- tion increased linearly (p<0.05) and quadratically (p<0.001) and on WB-diets it decreased quadratically (p<0.001) with a decreasing dietary P level.In a study by NÄSI (1990) Zn absorption and retention were higher in diets withoutphosphate addition, but remained unaffected in a later study by NÄSI and Helander (1993).The proportion of digested Zn that was retained was higher on all WB+ diets than on WB-diets.In a trial made by Newton et al. (1983) Zn apparent absorption decreased, when the WB level of the diet increased.The indigestible, fibrous bran fraction was found to accumulate some minerals such as Zn.Unfortunately, the retention of minerals was not determined in that study.
The rather low effect of WB phytase on the digestibility of P and other minerals may be due to several reasons.The optimum temperature for wheat phytase, 55°C, is higher than the animal body temperature and the optimum pH for its activity is 5.15 (Pomeranz 1978).For pigs, the stomach seems to be the principal place for the hydrolysis of phytate to utilizable P (Jongbloed 1987).The low pH of the stomach can cause the inactivation of plant phytase.The solubility of wheat phytate has been reported to be lowest within the pH range from 1.5 to 2.5 (Scheuermann et al. 1988 a).According to Scheuermann et al. (1988b), pepsin also inhib- its the activity of wheat phytase.JONGBLOED (1987) found that an increase in dietary protein level improved the digestibility of P. In the present study, the pigs on WB+ diets ate less protein than the other pigs.This might also explain the low effect of WB on P utilization.
The age of the pigs may also partly explain the results.Older pigs may need less P in tissue meta- bolism and/or they are able to utilize phytate P better (Reinhart and Mahan 1986).
One reason for the poor effect of WB could be that the inclusion of 100 g WB used in this experiment per kg diet may not have been a sufficient amount.The calculated increase in phytase activity in the diet was only 280 U/kg.The addition level of microbiologically produced phytase for growingfinishing pigs recommended by BASF (1993) is 500 U per kg feed, having a total P content of 4.0-5.0g/kg.
The possible decreasing effect of pelleting on enzyme activity was discussed previously.Pellet- ing may, on the other hand, have a positive effect on the digestibility of P. Bayley et al. (1975) have reported improved P absorption due to steam pellet- ing.It is interesting to note that such an improvement was found only in diets not supplemented with inorganic P. The pelleting temperature, how- ever, was not mentioned.JONGBLOED (1987) has also reported that the absorption and retention of P improved with pelleted diets.
In conclusion, the supplementation of a barleysoybean meal diet with 100 g WB per kg was not found to improve the dietary P utilization significantly.However, the digestion and retention of P appeared to be slightly improved by WB inclusion in the LP diet.This improvement may be due to WB phytase.The effects of WB on the digestibility and balance of other minerals remained relatively small.The P level had a greater effect on mineral balances.The dry matter and organic matter digestibilities were impaired when 100 g of WB was included in the diet.N absorption was higher on WB+ diets.The retention ofN appeared to decrease on HP and MP diets due to WB, but no decrease was found on LP diets.Ash digestibility was not affected by the treatments.The LPWB+ diet ap- peared to be the most favourable regarding the polluting P-emission" since the faecal excretion of P on that diet was the lowest and the overall reten- tion coefficient of P was the highest.This conclu- sion can, however, be confirmed only after a growth trial.On the basis of the present results, an addition of 100 g of WB to the diet in order to improve the P utilization of finishing pigs, does not seem very beneficial.

Table
Composition of the experimental diets, g/kg.

Table 4 .
Nutrient digestibilities of the experimental diets.
* o NS

Table 5 .
Nitrogen balance and protein utilization in pigs fed on the experimental diets.

Table 7 .
Calcium excretion, apparent digestibility and retention in pigs fed on the experimental diets.

Table 8 ,
Magnesium excretion, apparent digestibility and retention in pigs fed on the experimental diets.