|
|
||||||||
1 Department of Animal Sciences, Ohio Agricultural Research and Development Center, The Ohio State University, Wooster 44691
2 Zinpro Corp., Eden Prairie, MN 55344
Corresponding author: W. P. Weiss; e-mail: weiss.6{at}osu.edu.
| ABSTRACT |
|---|
|
|
|---|
Key Words: manganese digestion dairy cow
Abbreviation key: AC = absorption coefficient.
| INTRODUCTION |
|---|
|
|
|---|
The NRC (2001) model uses absorption coefficients (AC) to convert dietary Mn into absorbed Mn. The AC (g of absorbed Mn/g of total Mn) for Mn in feedstuffs other than Mn supplements is 0.0075 and between 0.0015 and 0.012 for Mn supplements (NRC, 2001). The NRC does not provide AC for organic sources of supplemental Mn, but, based on tissue concentrations in sheep, Mn from Mn-Met might have a higher relative bioavailability than Mn sulfate (Henry et al., 1992). Data comparing organic and inorganic sources of supplemental Mn are lacking for dairy cows.
The first objective of this experiment was to determine whether source (inorganic or organic) of Mn influenced apparent absorption and retention of Mn in late gestation cows. The second objective was to use data from digestibility studies to estimate the maintenance requirement for Mn of lactating and dry dairy cows.
| MATERIALS AND METHODS |
|---|
|
|
|---|
|
|
Sampling and Analyses
Blood samples were taken via the tail vein from all cows at dry-off, the day cows left the metabolism stalls (approximately 25 d before anticipated calving), and within 15 h after parturition. Blood from calves (jugular vein) was sampled within 15 h of birth. A sample of first milking colostrum was taken from each cow. Blood samples were collected into heparinized, trace metal-free vacutainers, and colostrum samples were placed in acid-washed containers.
Over the experiment (approximately 4 mo were required for all cows to finish the experiment), forages and concentrates were sampled weekly and composited into monthly samples. Weekly forage samples were analyzed for DM to adjust TMR for changes in forage moisture. During the metabolism study, feeds, orts (if any), urine, and feces were sampled daily (samples of orts, urine, and feces represented 10, 2, and 1%, respectively, of daily amounts) and were composited within cow. Feeds, orts, and fecal samples were lyophilized and ground (1-mm screen; Wiley mill; Arthur Thomas, Philadelphia, PA). Water samples were taken on approximately d 30 and 60 of the experiment during the second and third collection period (samples were taken from the water stream immediately before water entered the drinking cups).
Organic nutrients, energy, and N in feed, orts, and fecal samples were analyzed as described by Weiss and Wyatt (2004). For calculating N digestibility and retention, samples of undried feces and acidified urine were assayed for N. Samples of dried, ground feed; orts; feces; non-acidified urine; whole blood; and colostrum were digested in nitric and perchloric acids (Timmons et al., 2001), and Mn was assayed using flame atomic absorption spectroscopy (Varian Spectra AA 200; Varian Inc., Palo Alto, CA). Urine, whole blood, and colostrum were concentrated before analysis using the method described for milk by Timmons et al. (2001). The starting mass of whole blood and milk was 20 g and was 30 g for urine.
Calculations and Statistical Analyses
Apparent digestibility or absorption was calculated as follows: (intake fecal excretion)/intake. Apparent retention was calculated as follows: intake fecal excretion urinary excretion. The Mn consumed via the bolus was included in the Mn intake. Because water intake was not measured, consumption of Mn via water was not included. The water consumed by these cows contained 0.03 mg of Mn/L and, based on an estimated (Holter and Urban, 1992) water intake of 35 L/d, consumption of water would have increased intake of Mn by about 1 mg/d.
Data were analyzed using PROC MIXED (SAS, 1999). The model included block (cows were blocked by anticipated calving date) as a random effect (5 df), treatment as a fixed effect (2 df), and error (10 df). The treatment effect was partitioned into 2 contrasts: effect of Mn supplementation (control vs. MnSO4 + Mn-Met) and the effect of type of supplement (MnSO4 vs. Mn-Met). Retention of Mn was compared to 0 using least squares means and a t-test (SAS, 1999). One calf was born dead (MnSO4 group); therefore, error degrees of freedom for calf blood data was 9. A second analysis was conducted to compare the concentration of Mn in whole blood collected from the cow and calf at parturition using PROC MIXED. The model included treatment (fixed, 2 df), animal type (cow or calf, fixed, 1 df), treatment x animal type interaction (2 df), block (random, 5 df), and error (23 df).
Estimating Mn Requirements (Experiment 2)
Apparent Mn digestibility data from experiment 1 were combined with data collected from lactating dairy cows from experiments conducted at the Ohio Agricultural Research and Development Center. The protocol for total collection was the same as described previously, except that cows were milked twice daily in the stalls, and milk was measured and sampled daily. Data from lactating cows (160 cows or cow-periods, if the experiment was a Latin square) came from 8 different experiments with 39 dietary treatments. Details concerning the lactating cow data set can be found in Weiss and Wyatt (2004). The combined data set had 178 observations. Dietary Mn was not a treatment in any experiment, except for the dry cow study (experiment 1). All diets exceeded the 1989 NRC recommendation for Mn (40 mg/kg), but diets differed in Mn concentration. With the exception of the dry cow experiment (experiment 1), MnO was the only form of supplemental Mn fed. Excretion of Mn via milk and urine are trivial in comparison with fecal losses. The concentration of Mn in urine and milk was measured in 2 experiments (43 observations), and urinary and milk losses of Mn averaged 0.4 and 0.6 mg/d, respectively. The sum of urine and milk Mn was <0.1% of average Mn intake. The relationship between intake of Mn and intake of apparently digestible Mn was quantified using PROC MIXED (SAS, 1999) with experiment (trial) included as a random class effect (St-Pierre, 2001). Analyses were conducted using individual observations (n = 178). Trial-adjusted data were calculated as described by St-Pierre (2001). The intercept and slope obtained when intake of digestible Mn was regressed on Mn intake (i.e., Lucas test) are estimates of metabolic fecal Mn and true digestibility of Mn, respectively (Van Soest, 1982).
| RESULTS AND DISCUSSION |
|---|
|
|
|---|
|
|
The marginal digestibilities of Mn from the supplements were calculated by first determining intake of Mn provided by basal diets (i.e., Mn intake 200 mg Mn provided by bolus). Then, fecal Mn that originated from basal ingredients was estimated by multiplying intake of Mn from basal ingredients by 0.977 (i.e., the mean indigestibility coefficient for the control treatment). That value was subtracted from total fecal Mn to estimate fecal Mn provided by the supplement. Marginal digestibility = [intake of supplemental Mn (i.e., 200 mg) fecal Mn provided by supplement] ÷ 200. This calculation assumes that digestibility of basal Mn was not influenced by Mn supplementation. The resulting marginal digestibilities were 16 and 17% for MnSO4 and Mn-Met, respectively.
Urinary excretion of Mn was not affected by treatment and represented <0.1% of Mn intake, which is consistent with other data (Watson et al., 1973; Gustafson and Olsson, 2004). Apparent retention of Mn was 12 mg/d for control cows and was increased (P < 0.01) to 44 mg/d for cows fed supplemental Mn (no effect of source). Sheep fed diets with Mn-Met had greater Mn concentrations in bone, but not liver or kidney, than sheep fed MnSO4 (Henry et al., 1992). Averaged across all 3 tissues (but not weighted by relative mass of tissues), Mn-Met resulted in 21% higher Mn concentrations than MnSO4. Diets in that experiment contained 900 to 2700 mg/kg of supplemental Mn.
Apparent retention of Mn was not different from zero for the control treatment (P > 0.33), but was greater than zero (P < 0.05) for cows fed MnSO4 and for cows fed Mn-Met. Apparent retention of Mn by cows fed supplemental Mn was higher than what would likely be needed for growth of fetal and associated maternal tissues. The accretion of Mn by the conceptus of late gestation dairy cows averaged about 1 mg/d when cows were fed diets with 50 to 60 mg/kg Mn (House and Bell, 1993). Howes and Dyer (1971) reported that the concentration of liver Mn in newborn calves increased when gestating cows were fed diets with higher concentrations of Mn (20 vs. 44 mg/kg), suggesting that feeding supplemental Mn might have caused increased fetal retention of Mn. The concentration of Mn in whole blood of newborn calves from cows fed supplemental Mn was numerically higher (P = 0.20) than that from calves from control cows, but variation among calves was extremely high (Table 3
). Concentrations of Mn in colostrum and whole blood from cows were not affected by treatment. The concentration of Mn in whole blood from newborn calves was higher (P < 0.01) than the concentration in whole blood from cows at parturition, but no treatment x animal type interaction (P > 0.38) was observed.
Estimated Mn Requirement (Experiment 2)
Mean intake of Mn was 1005 mg/d (SD = 308), mean apparent digestibility of Mn was 10.3% (SD = 12.6), and average Mn retention was 109 mg/d (SD = 123) for the lactating cows in the data set. With trial included as a random effect, the relationship (Figure 1
) between intake of digestible Mn (mg/d) and Mn intake (mg/d) was
|
![]() | ([1]) |
The intercept and slope in that equation were different from zero (P < 0.01). The estimates of metabolic fecal Mn and true digestibility from equation [1] are 125 and 43 times greater than the corresponding maintenance requirement and AC from NRC (2001). However, comparing the individual components is not appropriate because, within a system (NRC or equation [1]), the components are dependent upon each other. The NRC (2001) estimated the maintenance requirement by assuming that an intake of approximately 0.3 mg of dietary Mn/kg of BW was adequate to prevent deficiency signs. That value was multiplied by an assumed AC (0.0075) to yield the estimated requirement of 0.002 mg of available Mn/kg of BW. Therefore, the estimated maintenance requirement is a function of the AC that was used. Similarly, the intercept and slope in equation [1] are not statistically independent; therefore, the maintenance requirement for Mn (i.e., metabolic fecal Mn) estimated using equation [1] is a function of the estimated true digestibility of Mn. The amount of dietary Mn needed to meet the maintenance requirement (NRC, 2001) and the amount of dietary Mn needed for metabolic fecal Mn (equation [1]) can be compared.
Based on NRC (2001) equations, the mean maintenance requirement (mean BW of lactating cows was 605 kg) for available Mn was 1.21, the lactation requirement (mean milk production was 30.5 kg/d) was 0.92, and the mean total requirement was 2.13 mg/d for lactating cows in our data set. The average AC for diets fed to lactating cows in this data set was 0.006 (approximately 25% of dietary Mn was from MnO) based on values in NRC (2001). Therefore, the maintenance and total requirement for dietary Mn was 201 and 355 mg/ d. For the dry cows in the data set (experiment 1), the available Mn requirement (NRC, 2001) for maintenance and the total requirement averaged 1.5 and 1.8 mg/d. The average AC for the dry cow diets (assuming Mn-Met and MnSO4 were equal) was 0.0083. Therefore, the maintenance and total dietary requirement for Mn of dry cows (NRC, 2001) were 181 and 217 mg/d.
Based on equation [1], metabolic fecal Mn (i.e., an estimate of the maintenance requirement) is 151 mg/ d. Using the true digestibility (0.26) in equation [1], 580 mg/d of dietary Mn is needed to meet the maintenance requirement of dairy cows and an additional 1 to 3 mg of dietary Mn were needed to meet the lactation or gestation requirement. Therefore, the total dietary requirement for Mn of dry and lactating dairy cows based on equation [1] was approximately 582 mg/d. The requirements for dietary Mn based on equation [1] were approximately 1.6 and 2.7 times higher than NRC (2001) requirements for lactating and dry cows, respectively. Average DMI for lactating and dry cows in this data set were 20.9 and 11.8 kg/d; therefore, dietary concentrations of 28 and 49 mg of Mn/kg of DMI would be needed to meet the requirements calculated with equation [1] compared with 17 and 18 mg/kg calculated from NRC (2001) equations.
Both the NRC (2001) method and equation [1] have limitations. The maintenance requirement for Mn of dairy cows has not been quantified directly (NRC, 2001), and the value used by NRC is dependent on the AC that was used. The AC values in NRC are based on very limited data. Because of substantial fecal excretion of endogenous Mn (Van Bruwaene et al., 1984), measuring true absorption of Mn is extremely difficult. Perhaps the most direct measurements were obtained by Sansom et al. (1978). They calculated that 0.5 to 0.75% of dietary Mn (mostly from MnCl2) was absorbed from the gut based on differences in blood Mn concentrations between the portal vein and systemic circulation, but those data were collected from only 2 cows. The coefficients in equation [1], although statistically significant (P < 0.01), have high standard errors. Based on those standard errors, estimates of dietary Mn requirements could easily vary by 30%. The 580 mg/d dietary Mn calculated from equation [1] assumes absorption efficiency is constant; however, in ruminants, absorption of Mn appears to become more efficient when dietary supply is low (Watson et al., 1973). Therefore, the amount of dietary Mn needed to meet the maintenance requirement as estimated using equation [1] may be < 582 mg/d.
The final criticism of equation [1] is that retention data in general are not appropriate for determining requirements for trace minerals (Mertz, 1987). The basis of this criticism is that excretion of a trace mineral is proportional to the current pool size, whereas uptake of the mineral depends on supply of available mineral (among other factors). Therefore, assuming intake of available mineral is greater than obligatory losses and less than toxic levels, a balance study does not determine the requirement for a mineral element, but rather the intake required to maintain the existing pool size (Mertz, 1987). Because fecal loss of Mn represents >99% of the total loss of Mn from the cow, the data in Figure 1
are quantitatively the same as if Mn intake was plotted against Mn retention (Y-axis). Conceptually, however, Figure 1
differs from a plot of Mn intake and Mn retention. The Y-intercept in Figure 1
is the extrapolated estimate of fecal excretion when no Mn is consumed and is an estimate of obligatory losses of Mn. In theory, the Y-intercept does not include endogenous fecal Mn that is excreted in an attempt to maintain Mn homeostasis.
| CONCLUSIONS |
|---|
|
|
|---|
Received for publication November 15, 2004. Accepted for publication March 10, 2005.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. L. Hansen, J. W. Spears, K. E. Lloyd, and C. S. Whisnant Feeding a low manganese diet to heifers during gestation impairs fetal growth and development. J Dairy Sci, November 1, 2006; 89(11): 4305 - 4311. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. E. Nocek, M. T. Socha, and D. J. Tomlinson The effect of trace mineral fortification level and source on performance of dairy cattle. J Dairy Sci, July 1, 2006; 89(7): 2679 - 2693. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |