Discussion

Nutritional responses on milk production are related to the animal physiological state, which is dependent on stage of lactation and nutrient balances. Stage of lactation may have an effect on nutrient partitioning between milk and body tissues. However, it can be questioned whether the physiological status is more related to nutrient supply than stage of lactation per se (Kirkland and Gordon, Reference Kirkland and Gordon2001). In the present experiment, similar energy balances in early and late lactation make it possible to compare the nutritional effects on the animal performance. The response to absorbed protein is dependent on the energy status of the animal (Hanigan et al., Reference Hanigan, Cant, Weakley and Beckett1998), where glucogenic AA might be oxidised or used as glucose precursors. For cows in negative energy balance higher AA availability will increase gluconeogenesis and thus lactose synthesis and thereby milk and protein yield, but have little effect on protein content as found in a Norwegian study (Schei et al., Reference Schei, Volden and Bævre2005). The negative energy balance in the present study might explain the general low milk protein content, since there is a positive correlation between energy intake and milk protein content (Coulon and Rémond, Reference Coulon and Rémond1991; Rigout et al., Reference Rigout, Hurtaud, Lemosquet, Back and Rulquin2003). However, for cows in positive energy balance, higher glucose availability has been shown to increase tissue energy balance and glucose oxidation, but have little effect on milk or protein yield (Reynolds et al., Reference Reynolds, Harmon and Cecava1994). Total tract digestibility of OM and CP were 2.8 and 1.3% units higher in late lactation than in early lactation, respectively, which corresponds well to observations that increased feeding level decreases the digestibility of the same diet (Volden, Reference Volden1999). Moreover, the percentage of urinary N excretion was 7.3 units higher in late lactation than in early lactation in spite of a lower feeding level, which also corresponds well to Volden (Reference Volden1999). Responses in milk yield and milk protein production were similar in early and late lactation. The higher AA availability due to AA infusions increased urinary N excretion in early lactation and plasma urea in late lactation. Similar ME, and reversed urinary N and plasma urea in early and late lactation might indicate a different fate of the available AA with only minor effect on milk production. Lower fat concentration in the GB treatment in late lactation could be a dilution effect of increased yield resulting from an increase in glucose availability for lactose synthesis without concomitant increases in milk fat synthesis (Gaynor et al., Reference Gaynor, Waldo, Capuco, Erdman, Douglass and Teter1995) which has been observed in other studies with glucose (Fisher et al., Reference Fisher and Elliot1966; Hurtaud et al., Reference Hurtaud, Rulquin and Verite1998) or starch infusions (Reynolds et al., Reference Reynolds, Cammell, Humphries, Beever, Sutton and Newbold2001). The responses in milk yield expressed in percentage were marginally higher in late than in early lactation (4.6 v. 6.6%). This is interesting since the infusion amounts in late lactation were lower than in early lactation. The N balance showed that the recovery of N in milk, urine and faeces were higher in late (92.1%) than in early lactation (88.3%), and therefore, the body protein retention was also higher in early lactation. This could be a result of a higher metabolic activity in the gastrointestinal tissues in early lactation as discussed in our corresponding paper (Schei et al., Reference Schei, Danfær, Boman and Volden2007). Using the same assumption as calculated by Raggio et al. Reference Raggio, Pacheco, Berthiaume, Lobley, Pellerin, Allard, Dubreuil and Lapierre(2004) as we did in the early lactation experiment (Schei et al., Reference Schei, Danfær, Boman and Volden2007) the loss of metabolic faecal nitrogen (MFN) with starch infusion was calculated to 5.6 g/kg DMI. The corresponding figures with intravenous glucose infusion were 4.8 g/kg DMI. Comparing these calculations with the corresponding values for early lactation (Schei et al., Reference Schei, Danfær, Boman and Volden2007), demonstrate that there is a higher loss of MFN in early than in late lactation. This indicates an increased protein turn-over and tissue growth in the gut. This could explain the higher retention in early lactation.

The AB infusion increased plasma concentration of both glucose and insulin above that found for GB infusion. These observations were not found in early lactation (Schei et al., Reference Schei, Danfær, Boman and Volden2007). This is difficult to explain, but Kim et al. (Reference Kim, Choung and Chamberlain2000) also reported higher insulin concentration by intravenous NEAA compared with glucose infusion. Data on blood metabolite concentrations are often difficult to interpret in terms of nutrient fluxes, but the observed glucose concentrations on treatments GB and AB indicate that gluconeogenesis was reduced by glucose infusion more strongly in late than in early lactation. In both lactational stages, the plasma level of EAA was higher on treatment AB than on treatment GB, whereas the opposite was the case for the concentration of NEAA. This suggest that gluconeogenesis from AA was higher with AB than with GB infusion as NEAA is preferred to EAA as glucogenic substrates in cows (Black et al., Reference Black, Egan, Anand and Chapman1968). On the other hand, the insulin concentration in late lactation was higher on treatment AB than on treatment GB, and insulin is known to inhibit gluconeogenesis and to increase the uptake of glucose and AA in muscle cells (Sjaastad et al., Reference Sjaastad, Hove and Sand2003). Moreover, these results suggest that insulin concentration is hardly affected by nutrient supply in early lactation. The higher insulin level for AB infusion in late lactation can be explained by the higher glucose concentration since insulin is more sensitive to glucose than to AA (Lemosquet et al., Reference Lemosquet, Rideau and Rulquin1997). Individual AA like Arg and Leu may stimulate insulin more than other AA, and there is a synergistic effect on insulin release between glucose and AA (Lemosquet et al., Reference Lemosquet, Rideau and Rulquin1997), which may have been important in our experiment. Abomasal AA infusion did not have the same effect on insulin as intravenous AA infusion, probably as a result of lower blood glucose level due to a lower gluconeogenesis. Higher arterial glucose concentrations with jugular compared with abomasal AA infusion was also observed by Thivierge et al. (Reference Thivierge, Bernier, Dubreuil and Lapierre2002) in cows 84 days in milk, but the arterial insulin concentrations did not differ. However, a higher insulin response may be expected with blood infusion of AA because the pancreas in this case would be exposed to the infused AA before these are taken up by the liver. Infusion of glucose in the blood increased IGF-1 concentrations in both early (Schei et al., Reference Schei, Danfær, Boman and Volden2007) and late lactation, which is in line with other studies (Rigout et al., Reference Rigout, Lemosquet, Van Eys and Blum2002; Lemosquet et al., Reference Lemosquet, Rigout, Bach, Rulquin and Blum2004). This is related to the parallel increase in milk yield and milk protein yield (Lemosquet et al., Reference Lemosquet, Rigout, Bach, Rulquin and Blum2004). Low circulating concentrations of IGF-1 in early lactation have been associated with negative energy balance (Spicer et al., Reference Spicer, Tucker and Adams1990; Vicini et al., Reference Vicini, Buonomo, Veenhuizen, Miller, Clemmons and Collier1991) and with negative protein balance when energy balance is not already limiting (Ronge et al., Reference Ronge, Blum, Clement, Jans, Leuenberger and Binder1988).