The Effects of Milk Yield and Stage of Lactation on the Partitioning of Nutrients in Lactating Dairy Cows

J. Dairy Sci. 84:233–240  American Dairy Science Association, 2001.

The Effects of Milk Yield and Stage of Lactation on the Partitioning of Nutrients in Lactating Dairy Cows R. M. Kirkland* and F. J. Gordon† *The Agricultural Research Institute of Northern Ireland, Hillsborough, Co. Down, Northern Ireland BT26 6DR, and The Queen’s University of Belfast, Belfast, Northern Ireland BT9 5PX †Department of Agriculture and Rural Development, Northern Ireland

ABSTRACT The objective of the experiment was to examine, using indirect calorimetry, the effects of milk yield and stage of lactation on the response in milk and body tissue energy, and heat production, to a reduction (decrement) in nutrient intake (assessed as metabolizable energy intake). Eight lactating dairy cows, four representing each of two stages of lactation [either mean initial days in milk (DIM) 158 (SD 6.1) or 414 (SD 51.1)] were used. Each cow underwent four 17-d periods incorporating two physiological states [number of mammary glands milked: either four (periods 1 and 2), or two (periods 3 and 4)], and two levels of metabolizable energy intake within each physiological state [either sufficient to meet requirements for zero tissue balance plus 10 MJ/d (periods 1 and 3)] or these allowances reduced by 20 MJ/d in the subsequent period (periods 2 and 4, respectively). Partitioning was calculated from the changes in metabolizable energy intake, milk energy, tissue energy, and heat production between DIM groups and between four and two gland milking (milk yield) components of the study. Partitioning of the changes in metabolizable energy intake was not influenced by DIM, but milk yield response was greater in the early lactation cows compared with the late group. Cows milked in four glands (higher milk yield) partitioned a significantly greater proportion of decremental changes in metabolizable energy intake to milk energy and less to tissue energy, than when milked in only two glands (lower milk yield). (Key words: dairy cow, nutrient partitioning, stage of lactation, milk yield) Abbreviation key: Eg = energy balance, El = milk energy, F = entire udder milked (four glands), HP = heat production, M = MEI sufficient to meet requirements for maintenance and milk production plus an additional

Received March 10, 2000. Accepted August 16, 2000. Corresponding author: R. M. Kirkland; e-mail: rmkirkland@ yahoo.com.

10 MJ/d, MEI = metabolizable energy intake, R = MEI 20 MJ/d less than M, T = half udder milked (two glands). INTRODUCTION Ration formulation schemes for dairy cows (e.g. 2, 17) mainly use empirical data from calorimetric studies to produce factorial estimates of energy requirements. In such systems, the underlying principles of ration formulation are based on a constant efficiency of conversion of each unit of energy consumed to milk output (14), which implies that a certain amount of energy intake produces a specific output of milk of a given composition. This simplistic approach has limited value in the practical feeding of dairy cows, where it is known that responses in milk yield, or milk energy output, to increments of nutrient intake are curvilinear and obey the law of diminishing returns. This curvilinearity has been explained by the increased partitioning of nutrients to body tissue as intake increases (1, 7, 21). Axiomatically, these responses in milk output and body tissue to incremental changes in feed input are the key factors in ration formulation, enabling the nutritionist to determine optimum feeding regimes in contrasting economic environments. In view of the significance of nutrient partitioning on production responses (milk and tissue), and hence the economic implications of the effect, considerable effort has been directed towards modeling the myriad of complex biochemical pathways that comprise intermediary metabolism at a tissue level (3, 4, 5). Such dynamic, mechanistic modeling approaches have provided an understanding of the nature of, and factors influencing, milk and tissue responses to changes in nutrient intake. However, these also require the support of quantitative data on the partitioning of increments of feed nutrients recorded at the whole animal level. The inadequacies of production studies as a way to quantify responses in milk output and tissue gain are well recognized, reflecting the difficulty of accurately obtaining in vivo measurements of tissue energy changes. Calorimetric studies, however, enable simultaneous measurement

233

234

KIRKLAND AND GORDON

of changes in milk energy output, tissue energy change, and heat production to incremental changes in nutrient intake. This approach has, therefore, been adopted in the present study to quantify some of the factors influencing nutrient partitioning. It is recognized that both dietary and animal factors influence the partitioning of nutrients between milk and tissue (5). Burt (9) used data from production studies to try to disentangle the effects of milk yield and stage of lactation on milk yield response. Through multiple regression analysis, he suggested that milk yield, rather than stage of lactation per se, was the key factor influencing this partitioning response. However, the paucity of clear evidence to enable these two confounding aspects to be disentangled is well recognized (19). The primary objective of the present study was therefore to use calorimetric techniques, coupled with more novel experimental methodology, to enable the confounding effects of milk yield and stage of lactation on the partitioning of unitary changes in nutrient intake between milk energy output (El), body tissue energy (Eg), and heat production (HP) to be separated and quantified. The novel methodology employed to disentangle milk yield and stage of lactation involved altering milk yield at a given stage of lactation by “drying off” two of the four mammary glands within each cow. Over a number of years, many studies of the physiological aspects of the mammary gland have shown that milk yield of the cow is positively correlated with secretory tissue mass (20), and increased concentration of mammary parenchyma DNA (number of cells) (13, 16, 23). Hence, the “drying off” technique used in this study directly reduced secretory tissue mass (and the number of mammary cells), and thereby facilitated a direct method of assessing the impact of these attributes on partitioning responses. In the present study, the ratio of nutrients in the diet was maintained constant throughout. While it is recognized that changes in any nutrient, particularly protein, may influence responses obtained, the present study related the responses obtained to the change in metabolizable energy intake (MEI). MATERIALS AND METHODS Eight high genetic merit [PTA(95) fat + protein = 42 kg (SD 9.8 kg)] Holstein-Friesian dairy cows were subjected to energy metabolism studies at this Institute. PTA(95) is a measure of genetic merit used in the United Kingdom [for further details see Birnie et al. (6)]. Four cows were chosen to represent each of two stages of lactation (DIM), treatment 150 [initial DIM 158 (SD 6.1)], and treatment 400 [initial DIM 414 (SD 51.1)]. Mean cow liveweight, 610 kg (SD 37.0), and condition Journal of Dairy Science Vol. 84, No. 1, 2001

Table 1. Details of the diet offered throughout the study. Composition of TMR (proportion DM basis) Straw Concentrate Composition of the concentrate mixture (kg/t, DM basis) Barley Maize Molassed sugar beet pulp Citrus pulp Soybean Rape meal Megalac (protected fat supplement) White fish meal Mineral supplement Chemical composition of TMR (g/kg of DM unless otherwise stated) DM (g/kg fresh) CP ADF NDF GE (MJ/kg DM)

0.18 0.82 104 99 132 200 198 153 49 49 16

840 213 203 308 18.9

score, 2.5 (SD 0.22), were similar across both groups (condition score was based on a five-point scale, where 1 is a very thin animal and 5 very fat). All animals were less than 60 d pregnant at the commencement of the study and between their second and fourth parities (mean 3.1, SD 0.64). A TMR consisting of straw and concentrate (0.18:0.82 DM basis) was offered throughout the study. The composition of the concentrate portion of the diet, along with the chemical composition of the TMR is presented in Table 1. The experimental design consisted of subjecting each animal to four 17-d periods incorporating two physiological states [proportion of udder milked: entire udder (four glands) (F), half udder (two glands) (T)], and two levels of MEI within each physiological state [either MEI sufficient to meet requirements for maintenance and milk production (2), plus an additional 10 MJ/d (M); or MEI approximately 20 MJ/d less than M (R)]. The sequence of treatments for each cow was as follows: Period (1)—physiological state F; diet M, Period (2)—physiological state F; diet R, Period (3)—physiological state T; diet M, Period (4)—physiological state T; diet R. An additional 7 d was allowed between the end of period 2 and beginning of period 3 to allow each cow to adjust to its new physiological state and its associated dietary regime. For the final 3 d of each period, cows were transferred to indirect, open-circuit respiration calorimeters. Measurement of gaseous exchange over the final 48 h in the chambers was used in the calculation of heat production. The digestibility of the total ration was determined twice for each individual cow. The first was a 6-d study undertaken during period 2 and incorporated the 3 d

235

NUTRIENT PARTITIONING IN DAIRY COWS

prior to, and the 3 d when the animals were in the respiration calorimeters. The second digestibility was undertaken as a 9-d study, which included the 3 d in the calorimeters during period 4 and the subsequent 6 d. In each digestibility study, feces and urine were collected over the total balance period, while ME was determined by assuming that methane as a proportion of GE intake was the same for that part of the balance external to the calorimeters as was recorded for the 3 d in the calorimeters. The mean of the two values recorded per cow was taken as the diet ME concentration for that cow throughout all four periods. All sampling procedures and analytical methods used in the calorimetric studies were as described by Gordon et al. (12), and calibration tests were undertaken in accordance with Yan et al. (24). The calibration tests gave recoveries of 100.89 and 101.35% for CO2 and O2, respectively. The effects of the reductions in MEI (decrements) on HP, El, and Eg were calculated for each cow, in each physiological state, from the changes between periods 1 and 2 for full lactation (F), and between periods 3 and 4 for half lactation (T). From these data, the partitioning of unit decrements in MEI between HP, El, and Eg were calculated for each cow within each physiological state. Data were analyzed by analysis of variance using Genstat 5 (Release 4.1, Rothamsted, England) with lactation stage (150 or 400 DIM) and lactation state (F or T) as main factors. The analysis model fitted effects for DIM and lactation state and had 12 df for error. As the relationship between MEI and El is known to display negative curvilinearity, and the response in El and Eg to be interrelated (1, 7), Eg in periods 1 and 3 were incorporated as a covariate for energy partitioning at the F and T lactation states, respectively. Data on milk yield and milk component responses (per MJ decrement MEI) were similarly analyzed, while DMI, milk yield, and energy utilization parameters were analyzed by analysis of variance without covariate adjustment. Two cows from the 400T group were removed from the study after period 2 due to either ill health or drying off, while three sets of data (one from the 150F group and two from the 150T group) were omitted from the analyses due to equipment irregularities in HP determinations. Cows omitted from the study were included in the data set as missing plots. RESULTS The overall mean values for digestibility of DM and energy of the total diet, along with the determined ME concentration are shown in Table 2. Digestibility of DM was not influenced by stage of lactation (P > 0.05). The mean data for DMI, milk yield, and the range of energy

Table 2. Mean DM and energy digestibility along with metabolizability data for the diet offered throughout the study (n = 15).

DM digestibility Energy digestibility ME1 concentration (MJ/kg of DM) ME/GE2

Mean

SD

0.733 0.753 11.98 0.634

0.0130 0.0150 0.369 0.0170

1

Metabolizable energy. Metabolizable energy/gross energy.

2

input/output data for each treatment group, at each feed level (M and R), are presented in Table 3. Table 4 shows the mean proportions of the reductions in MEI that were partitioned to El, Eg, and HP, as well as the milk yield and individual milk component responses, for each treatment group. Physiological State (Milk Yield) Milk yield, DMI, HP, El, and MEI were significantly higher in the F than the T lactation state (P < 0.001). While the overall mean reductions in MEI (between feed levels) recorded across the whole study approximated to the planned figure of 20 MJ/d, the mean treatment differences achieved were significantly greater in the F compared with the T lactation state (P < 0.001). This disparity reflected both differences in feed DM throughout the study, and the discrepancy between the ME concentrations assumed in the calculation of the original feed levels, and the actual ME concentrations subsequently determined for each animal. Mean Eg values were close to zero and not significantly different between F and T lactation states (P > 0.05), indicating that the original objective of feeding animals to near their requirements (zero Eg) was reasonably well met. Physiological state had a significant effect on partitioning of decrements of MEI to milk (Table 4), being significantly higher (P < 0.01) for cows milked in four glands (150F and 400F) than when the same animals had their milk yield significantly reduced by drying off two glands (150T and 400T) (0.433 vs. 0.205 for F and T lactation states, respectively). In contrast, partitioning to Eg was significantly lower (P < 0.05) in the F compared with the T lactation state (0.216 vs. 0.456 for the F and T lactation states, respectively), while partitioning responses to HP were not affected by whether cows were in the F or T state (P > 0.05). The changes in milk fat and lactose output per MJ decrement in MEI were significantly higher when cows were milked in four (F) compared with two (T) glands (P < 0.05), while the responses in milk yield and milk protein were not influenced by physiological state (P > 0.05). Journal of Dairy Science Vol. 84, No. 1, 2001

236

*** *** NS 1.075 19.84 20.47 14.17 28.89

Cows with mean initial DIM of 158 (treatment 150), or 414 (treatment 400), milked in either four glands (F) or two glands (T). Feed level either M: level of metabolizable energy intake sufficient to meet the requirements for maintenance and milk production, plus an additional 10 MJ/d, or R: metabolizable energy intake 20 MJ/d less than M. 3 Metabolizable energy. 4 Reduction in metabolizable energy intake between periods (1) and (2) for F, and between periods (3) and (4) for T treatment groups. *P < 0.05. **P < 0.01. ***P < 0.001. NS = Not significant. 2

1

NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS * *** NS ** NS * * ** NS NS * * *** *** *** *** NS *** *** *** *** *** NS *** 15.22 18.30 119.3 59.9 6.7 185.9

DMI (kg/d) Milk yield (kg/d) Heat production (MJ/d) Milk energy (MJ/d) Tissue balance (MJ/d) ME3 Intake (MJ/d) Reduction in ME intake (MJ/d)4

21.54 34.28 143.5 103.6 9.1 256.1

19.11 29.23 135.2 90.9 1.1 227.2

12.16 13.55 96.7 43.2 1.4 141.2

10.94 11.45 92.3 39.1 −4.3 127.0

13.54 16.18 111.3 51.9 2.3 165.4

9.79 6.80 86.0 23.4 8.8 118.2

8.15 5.85 78.4 20.5 −0.5 98.4

0.845 1.212 5.51 4.62 3.52 11.00

F vs. T2 R 4 R 2 M 1 R 4 Feed level (Period)

2

M 1

R 2

M 3

n=2 150T1 n=3 150F1

Table 3. Feed intake, milk production, and energy utilization data.

n=4 400F1

M 3

n=2 400T1

SEM

150 vs. 4001

M vs. R3

1×2

Significance

1×3

2×3

1×2×3

KIRKLAND AND GORDON

Journal of Dairy Science Vol. 84, No. 1, 2001

Stage of Lactation Cows in the 150 DIM treatment group had significantly higher DMI, milk yield, HP, El, and MEI compared with cows in the 400 DIM treatment group (P < 0.001). Level of Eg was similar between the 150 and 400 DIM groups (P > 0.05) and, similarly, the reductions in MEI achieved between feed levels (M and R) were not significantly different between the 150 and 400 DIM groups, as was originally intended (P > 0.05). Stage of lactation (150 or 400 DIM) had no influence on the partitioning of decrements of MEI to either El, Eg, or HP (P > 0.05). Milk yield response was higher in the 150-treatment group than in the 400-treatment group (0.160 vs. 0.079 kg of milk/MJ decrement MEI respectively) (P < 0.01). The response in yield of milk fat was not affected by stage of lactation (P > 0.05), while both protein and lactose responses were significantly greater in the 150 DIM group than in the 400 DIM group (P < 0.01). Feeding Levels Both DMI and MEI were significantly reduced during the restricted feed level periods of the study (diet level R) in comparison with diet level M (P < 0.05). Milk yield showed a corresponding reduction when feed level was reduced from M to R (P < 0.01), although the decrease in El resulting from the reduction in feed intake did not reach statistical significance (P > 0.05). Similarly, HP was not significantly influenced by level of feed intake (M or R). Tissue energy balance was, however, significantly lower at the R compared with M feed levels (−0.40 and 6.50 MJ/d for the R and M feed levels, respectively). Interactions There were significant interactions between DIM and physiological state for DMI and MEI (P < 0.05) and for both milk yield (P < 0.001) and El (P < 0.01) components. Similarly, there was a significant interaction between DIM and physiological state in the reduction in MEI achieved between M and R feed levels (P < 0.001). There were no interactions between treatment groups in either the HP or tissue balance components of the study (P > 0.05). No interactions between DIM and physiological state were recorded in either the energy partitioning, or milk yield and milk component response, parts of the study (P > 0.05). DISCUSSION The present study examined the effects of two factors, stage of lactation and milk yield (effected by altering

237

NUTRIENT PARTITIONING IN DAIRY COWS Table 4. The proportion of the reduction (decrement) in metabolizable energy intake (MEI) partitioned to milk energy, energy balance, and heat production, along with responses in milk yield and milk components. Significance 400F1

400T1

SEM

150 vs. 4001

F vs. T2

1×2

partitioned to: 0.230 0.400 0.494 0.200 0.276 0.399

0.179 0.419 0.402

0.0589 0.0855 0.0489

NS NS NS

** * NS

NS NS NS

MJ decrement): 0.143 0.105 0.24 4.41 4.92 3.20 7.96 5.29

0.052 2.36 1.47 2.12

0.0201 0.849 0.834 0.956

** NS ** **

NS * NS *

NS NS NS NS

150F1

150T1

Proportion of decrement of MEI Milk energy 0.466 Tissue energy 0.231 Heat production 0.303 Milk production responses (per Milk yield (kg) 0.177 Milk fat (g) 4.45 Milk protein (g) 5.91 Milk lactose (g) 9.85

1 Cows with mean initial DIM of 158 (treatment 150), or 414 (treatment 400), milked in either four glands (F), or two glands (T). *P < 0.05. **P < 0.01.

the number of glands milked), on the partitioning of decrements of MEI to El, Eg, and HP. However, as there were no significant interactions between the factors in terms of energy partitioning, and in view of the relatively small numbers of replicates per treatment, these factors will be discussed as main effects only. Effect of Stage of Lactation on Partitioning One of the main requirements in the present study was that cows were selected at widely different stages of lactation. This was necessary for two reasons: 1) to ensure that any effects, which might be a result of this factor, would be maximized, and 2), to ensure that there was no overlap of actual DIM between stage of lactation groups throughout the four periods of the trial. Furthermore, it was also decided that the animals chosen for the “early” lactation group should be in the postpeak stage of lactation, to minimize the possible impact of short-term fluctuations in partitioning that might occur during the very early lactation stages. It was also important, however, that cows at the more advanced stage of lactation had a relatively high level of milk production to ensure that any responses recorded were not solely a reflection of very low milk yields per se in this stage of lactation group. All these requirements were met in the study, with mean initial DIM being 158 and 414 for the two groups, and the respective data for initial milk yield being 34.3 and 18.4 kg/d. The target reduction in MEI in the study was 20 MJ/d, but due to differences in feed DM during the study, and the disparity between assumed ME concentrations used in the initial calculation of feed level and actual determined ME concentrations, recorded values deviated around this objective (28.9 MJ/d in the 150F group, compared with 14.2 MJ/d in the 150T group). However,

as all responses in El, Eg, and HP were computed per unit change in MEI, and considering the improbability that these responses would have exhibited nonlinearity over such small changes in MEI, these differences are unlikely to have influenced the partitioning responses obtained. The present study indicated that partitioning of MEI to HP was not significantly influenced by stage of lactation of the cows. However, assuming that ME requirements for maintenance are not influenced by the level of milk production, the mean values recorded would suggest a declining trend in the combined efficiencies (kl + kg) between the 150 and 400 DIM groups. While there is no evidence in the literature to suggest such an effect, this may reflect the fact that previous studies have not examined this relationship. However, recognizing the small number of animals in the present study, and the absence of significant differences in the HP data between stages of lactation, this study would not support a view that stage of lactation influences the efficiency with which energy is used by the lactating cow. The present results indicated no differences in partitioning of decrements of MEI to either El or Eg between the two stage of lactation groups (150 and 400 DIM) examined here. Similarly, when the individual animal responses in El and Eg were regressed against each animals’ DIM, no significant relationship was found. These data therefore provide evidence that partitioning of decrements of MEI to El or Eg is not influenced by DIM over a very wide section of the lactation (150 to 400 DIM). However, there are two important issues to bear in mind: 1) none of the present data relate to the very early stages of lactation, where it remains possible that partitioning responses might be influenced by the rapid changes in milk yield which occur prior to estabJournal of Dairy Science Vol. 84, No. 1, 2001

238

KIRKLAND AND GORDON

lishment of peak milk yield, and 2), the level of milk yield in the 400 DIM group in the present study was reasonably high in comparison to animals at advanced stages of lactation used in previous production studies quoted in the literature. Paradoxically, the absence of significant differences in energy partitioning (to El or Eg) between cows in the 150 or 400 DIM groups in the present study is in contrast to the milk yield data obtained, where the response in yield to decreasing MEI was higher in the 150 compared with the 400 DIM group. Furthermore, when this relationship was tested by regression analysis, the response in milk yield was found to be negatively correlated with DIM (P < 0.01). Other production-orientated studies in the literature have found effects comparable to those in this study. For example, in a review of a large number of published studies (n = 66), Coulon and Re´ mond (10) reported that milk yield response to energy supply was greater for cows in early (56 to 91 DIM) than in midlactation. Wallace (22) similarly noted a greater milk yield response with cows between 63 and 84 DIM compared with the response obtained when these same animals were 203 to 224 DIM. It is important that we try to reconcile the apparently contradictory results obtained between the milk yield and energy partitioning responses presently reported. Milk yield responses purely reflect the volume component, while responses in El reflect the combined responses of individual milk components to alterations in MEI. While milk protein and lactose responses differed between the 150 and 400 DIM groups, the energy-dense fat component was not affected by DIM. It is therefore postulated that the absence of any significant differences in energy partitioning between lactation stages, despite the concomitant effects recorded in milk yield responses, reflects this lack of difference recorded in milk fat responses between stage of lactation groups. In contrast to the above finding, early work by Burt (9), using multiple regression techniques on data from a production study, suggested that responses in milk yield to changes in nutrient intake were not influenced by DIM per se, but that level of current milk yield was the important factor determining such responses. Other production studies, (7, 8) have drawn similar conclusions. While it is recognized that the above trials (7, 8, 9) were carried out with cows of very limited production capacities and abilities in comparison to today’s standards, it is feasible to suggest, circumspectly, that observed or absent responses in this section of the present study might reflect the relative level of milk yield of the cows used, rather than stage of lactation per se. The influence of milk yield on partitioning responses is examined in the following section of the discussion. Journal of Dairy Science Vol. 84, No. 1, 2001

Effect of Milk Yield (Full vs. Half Udder) on Partitioning This component of the present study was designed to alter the level of milk production within each cow with minimum alteration of the DIM component. The innovative approach of ceasing to milk two glands of the cow represents one of the few direct attempts aimed specifically at eliminating the inherently confounding effects of milk yield and stage of lactation on partitioning responses. This alteration in milk production and physiological state resulted in a major reduction in the proportion of the decrement in MEI that was partitioned to milk energy (from 0.433 to 0.205 for cows in F and T states, respectively), while partitioning to Eg was increased from 0.216 to 0.456 between the F and T groups, respectively. The absence of any significant differences in partitioning to HP is largely as expected (assuming that ME requirements for maintenance remain unchanged between physiological states) given the lack of evidence in the literature to support any influence of milk yield on kl or kg. The limitations of developing empirical relationships between animal factors, such as milk yield and partitioning of nutrients, from such a limited data set are fully accepted, but they do assist in the interpretation of the responses obtained. For example, when the relationship between level of milk production (MJ/d) and proportion of decrements of MEI partitioned to El (assuming DIM effects are zero) was tested by regression techniques, significant relationships both linearly and asymptotically were recorded between the variables. The latter, although not a significant improvement on the linear model, must be regarded as the more relevant and applicable model biologically, where it seems unlikely that such a partitioning effect could continue in a linear manner ad infinitum. The relationship developed was described by the following equation: Partitioning to El = 0.509(SE 0.1670) + (−0.600)(SE 0.2770) * 0.976(SE 0.0277)Initial El (R2 = 0.52; P < 0.05), and is displayed graphically in Figure 1. This regression suggests that when a cow is fed to near her requirements (zero Eg), the maximum proportion of an increment of MEI partitioned to El would be 0.51. However, Figure 1 demonstrates that such a maximum would only be reached at high levels of El (>120 MJ/d). Interestingly, if an El of 64 MJ/d is considered, the proportion of decrements of MEI partitioned to El would be 0.38, which is relatively similar to that of 0.30 reported by Agnew et al. (1) in a recent calorimetric study at this Institute using cows with mean El of 64 MJ/d.

NUTRIENT PARTITIONING IN DAIRY COWS

Figure 1. The relationship between proportion of decrements of metabolizable energy intake (MEI) partitioned to milk energy and milk energy output (El).

However, while it is recognized that the current methodology involving drying off two glands of each cow will decrease milk yield through reductions in active mammary tissue (13, 16, 20), and that this procedure alters the partition of MEI in the dairy cow (25), it cannot be assumed that this methodology is totally representative of the effects that might be obtained with naturally lower yielding cows. For example, in the model of the mammary gland proposed by Neal and Thornley (18), and incorporated into the lactating cow model developed by Baldwin et al. (5), udder biosynthetic capacity increases in early lactation and subsequently decreases in later lactation in response to hormone concentration. One may speculate that the normal homeostatic and homeorhetic controls on lactation might have been altered by the sudden drying off of two glands, which could potentially have influenced the partitioning results obtained between lactation states in the present study. Furthermore, it was not possible to determine whether the two functional glands (remaining after the drying off procedure) showed altered activity, or even if they experienced a degree of compensatory growth in response to the drying off procedure. However, the adaptation period was relatively short between F and H states, which might have precluded such mammary adjustment. Nevertheless, it is feasible that physiological changes within the animal resulting from the drying off procedure might have affected the animal response recorded presently. Despite the potential limitations of the present approach, the methodology achieved its objectives. This is clearly evidenced by the reduction in El from a mean of 76.6 MJ/d when cows were milked in all four glands

239

(F), to a mean of 31.6 MJ/d when the same cows were dried off in two glands (T). While the primary interest in this study concerns the partitioning of decrements of MEI, previous studies in the literature on partitioning of increments of feed nutrients have mainly been of a production nature, and hence do not help to determine accurate partitioning responses to El, Eg, and HP. Such production studies are therefore of limited comparative value to the current data set. Nevertheless, work by Burt (9) and Blaxter (7) suggested that responses in milk yield to additional nutrient intake were positively correlated with level of current milk yield, while the converse effect was true of liveweight. However, the inadequacy of liveweight as a measure of Eg is well recognized (15). The production data (milk yield responses), obtained in the current trial are, however, in line with previous production trials. For example, in the present study, although the response in milk yield between the four and two glands only approached significance (P = 0.07), when the overall data were included in a regression analysis, a significant (P < 0.05) relationship between milk yield response (kg of milk/MJ decrement in MEI) and current milk yield (kg/d) was obtained. This relationship was described by the following regression equation: Milk yield response = 0.00369(SE 0.001220) milk yield + 0.0490(SE 0.02700) (R2 = 0.50; P < 0.05) Incorporating the mean milk yield recorded in the present study (17.0 kg/d) into this equation predicts milk yield response to a unitary change in MEI to be 0.112 kg of milk/MJ of MEI. This latter figure is directly in line with the response of 0.113 obtained when the current mean milk yield was applied to the linear regression equation produced by Blaxter (7). These responses are, however, marginally greater than that recorded by Friggens et al. (11), 0.088 kg of milk/MJ of MEI, with cows of relatively similar milk yield to the mean value recorded in the present study. The major effects of El on the partitioning of decrements of MEI to El recorded in the present study reflect the substantial reduction in milk yield concomitant with the reduction in mammary cell numbers resulting from drying cows off in two glands. For example, El was reduced from 76.6 to 31.6 MJ/d when cows were milked in two rather than four mammary glands. However, given this major effect and the fact that there was also a considerable difference in El between stage of lactation groups in the present trial (69.2 vs. 38.9 MJ/d, for 150 and 400 DIM groups, respectively), it appears paradoxical that no differences in partitioning to El were recorded in that part of the study. However, it is possible Journal of Dairy Science Vol. 84, No. 1, 2001

240

KIRKLAND AND GORDON

that the extent of the difference in El recorded between cows at 150 and 400 DIM might not have been sufficiently large for significant differences to become apparent on the asymptotic curve. Nevertheless, it seems feasible to suggest that the differences in production responses discussed previously between the stage of lactation groups might reflect existing differences in milk yield, rather than stage of lactation per se. CONCLUSIONS The results of the present study suggest that partitioning of decrements of MEI to El or Eg is not influenced by stage of lactation per se (when determined between 150 and 400 DIM), but is strongly affected by level of current El. Partitioning of decrements of MEI to El has been shown to increase asymptotically as El increases.

9

10 11

12

13 14

ACKNOWLEDGMENTS

15

The authors wish to thank the staff of the Agricultural Research Institute of Northern Ireland for their assistance throughout the course of this study.

16

REFERENCES

18

1 Agnew, R. E., F. J. Gordon, D. J. Kilpatrick, T. Yan, and M. G. Porter. 1999. The effect of incremental changes in metabolisable energy intake on the partitioning of metabolisable energy in dairy cows. Page 31 in Proc. Br. Soc. An. Sci. Ann. Conf. Scarborough, England. 2 Agricultural and Food Research Council. 1990. Technical Committee on Responses to Nutrients, Report Number 5, Nutritive Requirements of Ruminant Animals, Energy. Nutr. Abstr. Rev. (Series B) 60:729–804. 3 Baldwin, R. L. 1995. Lactation. Pages 469–518 in Modelling Ruminant Digestion and Metabolism. Chapman and Hall, New York, NY. 4 Baldwin, R. L., J. France, and M. Gill. 1987. Metabolism of the lactating cow. I. Animal elements of a mechanistic model. J. Dairy Res. 54:77–105. 5 Baldwin, R. L., J. France, D. E. Beever, M. Gill, and J. H. M. Thornley. 1987. Metabolism of the lactating cow. III. Properties of mechanistic models suitable for evaluation of energetic relationships and factors involved in the partition of nutrients. J. Dairy Res. 54:133–145. 6 Birnie, J. W., R. E. Agnew, and F. J. Gordon. 2000. The influence of body condition on the fasting energy metabolism of nonpregnant, nonlactating dairy cows. J. Dairy Sci. 83:1–7. 7 Blaxter, K. L. 1966. The feeding of dairy cows for optimal production. George Scott Robertson Memorial Lecture, Queen’s University, Belfast, N. Ireland. 8 Broster, W. H., V. J. Broster, and T. Smith. 1969. Experiments on the nutrition of the dairy heifer. VIII. Effect on milk production

Journal of Dairy Science Vol. 84, No. 1, 2001

17

19 20

21

22 23

24

25

of level of feeding at two stages of the lactation. J. Agric. Sci. 72:229–245. Burt, A.W.A. 1957. The effect of variations in nutrient intake upon the yield and composition of milk. 2. Factors affecting rate of eating roughage and responses to an increase in the amount of concentrates fed. J. Dairy Res. 24:296–315. Coulon, J. B., and B. Re´ mond. 1991. Variations in milk output and milk protein content in response to the level of energy supply to the dairy cow: a review. Livest. Prod. Sci. 29:31–47. Friggens, N., G. C. Emmans, S. Robertson, D. G. Chamberlain, C. T. Whittemore, and J. D. Oldham. 1995. The lactational responses of dairy cows to amount of feed and to the source of carbohydrate energy. J. Dairy Sci. 78:1734–1744. Gordon, F. J., M. G. Porter, C. S. Mayne, E. F. Unsworth, and D. J. Kilpatrick. 1995. Effect of forage digestibility and type of concentrate on nutrient utilisation for lactating dairy cattle. J. Dairy Res. 62:15–27. Hackett, A. J., and H. A. Tucker. 1969. Correlation between mammary nucleic acids of rats during immaturity and lactation. J. Dairy Sci. 52:1268–1272. MacRae, J. C., P. J. Buttery, and D. E. Beever. 1988. Nutrient interactions in the dairy cow. Pages 55–75 in Nutrition and Lactation in the Dairy Cow. P. C. Garnsworthy, ed. Butterworths, London, England. Moe, P. W., H. F. Tyrrell, and W. P. Flatt. 1971. Energetics of body tissue mobilisation. J. Dairy Sci. 54:548–553. Nagai, J., and N. K. Sarkar. 1978. Relationship between milk yield and mammary gland development in mice. J. Dairy Sci. 61:733–739. National Research Council (NRC). 1988. Nutrient Requirements for Dairy Cattle. 6th revised ed. National Academy Press, Washington, DC. Neal, H. D. St. C., and J.H.M. Thornley. 1983. The lactation curve in cattle: a mathematical model of the mammary gland. J. Agric. Sci. 101:389–400. Oldham, J. D., and G. C. Emmans. 1989. Prediction of responses to required nutrients of dairy cows. J. Dairy Sci. 72:3212–3229. Sorensen, A., M. Alamer, and C. H. Knight. 1998. Physiological characteristics of high genetic merit and low genetic merit dairy cows: a review. Page 4 in Proc. Br. Soc. An. Sci. Ann. Conf. Scarborough, England. Thomas, P. C., D. G. Chamberlain, P. A. Martin, and S. Robertson. 1987. Dietary energy intake and milk yield and composition in dairy cows. Pages 188–191 in Energy Metabolism of Farm Animals. EAAP Publication 32. P. W. Moe, H. F. Tyrrell, and P. J. Reynolds, ed. Rowman Littlefield, New Jersey. Wallace, L. R. 1959. The nutrients required for milk production at successive stages of lactation: A study with identical twins. Proc. 15th Dairy Grassl. Congr. 1:196–203. Wilde, C. J., A. J. Henderson, C. H. Knight, D. R. Blatchford, A. Faulkner, and R. G. Vernon. 1987. Effects of long-term thricedaily milking on mammary enzyme activity, cell proliferation, and milk yield in the goat. J. Anim. Sci. 64:533–539. Yan, T., F. J. Gordon, R. E. Agnew, M. G. Porter, and D. C. Patterson. 1997. The metabolisable energy requirement for maintenance and the efficiency of utilisation of metabolisable energy for lactation by dairy cows offered grass silage-based diets. Livest. Prod. Sci. 51:141–150. Yan, T., F. J. Gordon, C. P. Ferris, R. E. Agnew, M. G. Porter, and D. C. Patterson. 1997. The fasting heat production and effect of lactation on energy utilisation by dairy cows offered foragebased diets. Livest. Prod. Sci. 52:177–186.