Journal of Dairy Science
Volume 91, Issue 9 , Pages 3311-3322, September 2008

Acute Metabolic Responses of Postpartal Dairy Cows to Subcutaneous Glucagon Injections, Oral Glycerol, or Both1

  • M.A. Osman

      Affiliations

    • Department of Animal Science, and Molecular Biology, Iowa State University, Ames 50011
    • ,
  • P.S. Allen

      Affiliations

    • Department of Animal Science, and Molecular Biology, Iowa State University, Ames 50011
    • ,
  • N.A. Mehyar

      Affiliations

    • Department of Biochemistry, Biophysics, and Molecular Biology, Iowa State University, Ames 50011
    • ,
  • G. Bobe

      Affiliations

    • Department of Animal Science, and Molecular Biology, Iowa State University, Ames 50011
    • G. Bobe (e-mail: bobeg@mail.nih.gov) is currently a fellow in the Cancer Prevention Fellowship Program, Office of Preventive Oncology, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892.
    • ,
  • J.F. Coetzee

      Affiliations

    • Department of Veterinary Clinical Sciences, Kansas State University, Manhattan 66506
    • ,
  • K.J. Koehler

      Affiliations

    • Department of Statistics, Iowa State University, Ames 50011
    • ,
  • D.C. Beitz

      Affiliations

    • Department of Animal Science, and Molecular Biology, Iowa State University, Ames 50011
    • Department of Biochemistry, Biophysics, and Molecular Biology, Iowa State University, Ames 50011
    • Corresponding Author InformationCorresponding author.

Received 6 January 2008; accepted 19 May 2008.

Article Outline

Abstract 

This study examined the effects of multiple subcutaneous glucagon injections with or without co-administration of oral glycerol on energy status-related blood metabolites and hormones of Holstein dairy cows in the first 2 wk postpartum. Twenty multiparous cows were fed a dry cow ration supplemented with 6kg of cracked corn during the dry period to increase the likelihood of developing postpartal fatty liver syndrome. Cows with a body condition score of ≥3.5 points (1- to 5-point scale) were assigned randomly to 1 of 4 treatment groups: saline, glucagon, glycerol, or glucagon plus glycerol. Following treatment, serial blood samples were collected over an 8-h period to determine the effects of glucagon and glycerol on blood metabolites and hormones. Treatment effects were determined by comparing the concentrations of metabolites and hormones during the first 4-h period and the entire 8-h period after treatment administration (time 0) with the concentration of the same compounds at time 0 on d 1, 7, and 13 postpartum. Administration of glucagon alone increased concentrations of plasma glucagon and insulin on d 1, 7, and 13 and increased plasma glucose and decreased plasma nonesterified fatty acids (NEFA) on d 7 and 13 postpartum relative to the saline group. Administration of glycerol alone increased plasma glucose on d 7 and plasma triacylglycerols on d 1 postpartum. Glycerol administration also decreased plasma glucagon and NEFA on d 1, 7, and 13 and plasma β-hydroxybutyrate (BHBA) on d 1 postpartum relative to the saline group. Administration of glucagon plus glycerol increased and sustained concentrations of plasma glucagon, glucose, and insulin on d 1, 7, and 13 and decreased plasma NEFA on d 1, 7, and 13 and BHBA on d 1 and 7. Early postpartal treatment of dairy cows with glucagon plus glycerol increased plasma glucose and insulin, decreased plasma NEFA and BHBA, and increased secretion of liver NEFA as plasma triacylglycerols. This suggests that glucagon and glycerol, when co-administered, act to decrease the likelihood of metabolism-related syndrome development in dairy cows.

Keywords: glucagon, glycerol, metabolism

 

Back to Article Outline

Introduction 

Fatty liver syndrome (FLS) affects dairy cows in the late prepartal and early postpartal periods up to 4 wk postpartum (Grummer, 1993). The severity of FLS is often classified based on the percentage of accumulated triacylglycerols (TAG) in the liver (Bobe et al., 2004) to assess the potential for deleterious effects on the health, productivity, and reproductivity of dairy cows.

During the peripartal period, DMI by dairy cows spontaneously declines. Regardless of the energy density and protein content of the peripartal diet, DMI decreases by about 30% during the final 3 wk prepartum, and about 90% of that DMI decrease happens during the last week (Doepel et al., 2002; Hayirli et al., 2002). Because the sharp decrease in DMI is accompanied with an increased peripartal energy and nutrient requirement by the gravid uterus and the mammary gland (Goff and Horst, 1997), the dairy cow develops a negative energy balance (NEB) that reaches -16 Mcal/d (Doepel et al., 2002). To compensate for the NEB, the dairy cow mobilizes body fat reserves in the form of NEFA and glycerol. Mobilized NEFA are taken up mainly by liver and are either oxidized in the mitochondria to produce energy or exported in the form of TAG-rich very low density lipoproteins. When uptake of NEFA by the liver far exceeds their disposal through oxidation or export as TAG-rich very low density lipoprotein, FLS develops in different degrees (Grummer, 1993).

Development of FLS is considered the number-one risk factor for ketosis induction in dairy cows (Veenhuizen et al., 1991). Furthermore, FLS impairs the immune system during the peripartal period making the dairy cow especially vulnerable to develop mastitis, endometritis (Zerbe et al., 2000), and retained placenta (LeBlanc et al., 2005). In addition, FLS is associated intimately with left displacement of the abomasum and milk fever (Katoh and Nakagawa-Ueta, 2001) and is implicated with decreased reproductive health of the dairy cow (Jorritsma et al., 2003).

Attempts to use nutritional means to alleviate the peripartal NEB and prevent the development of FLS during the postpartal period, by use of a low- or high-energy prepartal diet (Grum et al., 1996; Douglas et al., 2004) or by depression of milk fat to decrease the energy expenditure (Castaneda-Gutierrez et al., 2005), were not as successful. Similarly, the use of the glucogenic precursors such as propylene glycol, propionate salts, and amino acids gave inconsistent results (Hoedemaker et al., 2004). In addition, the use of slow-release insulin on d 3 postpartum decreased plasma glucose more precipitously than plasma NEFA, resulting in hypoglycemic shock in some experimental cows (Hayirli et al., 2002).

Subcutaneous injection of Holstein dairy cows with experimental FLS with 15mg/d of glucagon for 14 d during the postpartal period increased plasma glucose and insulin, decreased plasma NEFA and hepatic TAG, and prevented and cured FLS (Nafikov et al., 2006). One metabolic pathway by which glucagon cures FLS involves upregulating gluconeogenesis and increasing plasma glucose concentrations (Hippen et al., 1999). Because DMI decreases during the peripartal period, gluconeogenic substrates are in short supply. Oral glycerol is metabolized in the rumen and liver to intermediates of gluconeogenesis and glycolysis (Remond et al., 1993; Goff and Horst, 2001) with the potential to alleviate peripartal NEB. Because of the absence of glycerol kinase in adipocytes, adipose tissue cannot use glycerol released resulting from lipolysis in adipose tissue. This glycerol is transferred to the liver for metabolism.

Therefore, we hypothesized that glucagon and glycerol, when co-administered postpartally, may act to alleviate the peripartal NEB and treat FLS in dairy cows. Thus, the objectives of this study were to examine how glucagon and glycerol might act to treat FLS and to understand the biochemical mechanisms whereby they positively influence the metabolism of postpartal dairy cows. The ultimate goal of this research is to provide farmers with a slow-release subcutaneous implant of glucagon or a glycerol feeding protocol to use as an effective tool for treating fatty liver syndrome in dairy cows.

Back to Article Outline

Materials and Methods 

Experimental Design 

Twenty multiparous Holstein dairy cows from the Iowa State University dairy herd were selected during the dry period to participate in the study, based on parity and a BCS of 3.5 or greater (1 to 5 scale). Cows were assigned randomly to 1 of 4 treatment groups: saline (control), glucagon, glycerol, or glucagon plus glycerol. Starting at 6 wk prepartum, cows were housed in straw-bedded free-stalls and offered ad libitum access to a dry-cow TMR diet based on NRC recommendations (NRC, 2001) that was supplemented with 6kg of cracked corn to induce FLS development postpartum (Bobe et al., 2003b). Additionally, cows were offered ad libitum access to grass hay. Experimental cows received a physical examination weekly during the prepartal period and daily during the postpartal period. Details of physical examinations included measuring the heart rate, respiratory rate, ruminal movements, urinary ketones, fecal consistency, rectal temperature, BCS, and signs of lameness. A test for detection of abomasum displacement was included during the postpartal period.

Immediately after parturition, experimental cows were housed in a sand-bedded tie-stall and offered an NRC (2001)-recommended TMR for high-producing dairy cows twice daily at 0900 and 1500h. Treatment administration began 4h after parturition. Control group cows (n=4) received 60mL of 0.9% sodium chloride (pH 10.25) under the skin of the neck each 8h daily for 14 d. Cows in the glucagon group (n=4) received 5mg of glucagon (donated by Eli Lilly and Co., Indianapolis, IN) dissolved in 60mL of 0.9% sodium chloride (pH 10.25) and injected under the skin of the neck each 8h for 13 d, 16h apart on d 14, and a single injection on d 15 postpartum. Cows in the glycerol group (n=6) received 400mL of pure glycerol (West Central Inc., Ralston, IA) diluted with 100mL of water orally once a day at 0600h for 14 d. The dosage was selected on basis of results of Remond et al. (1993), Goff and Horst (2001), and De Frain et al. (2004). Glucagon plus glycerol group cows (n=6) received both treatments as described above. Glucagon solutions were prepared in autoclaved glassware that was prerinsed with 1% BSA (cat. number A8327, Sigma, St. Louis, MO) and kept at 4°C for no more than 36h. All experimental and surgical procedures were done in accordance with the guidelines of the Iowa State University Institutional Animal Care and Use Committee.

Sampling and Analysis 

Blood 

To monitor the acute effects of glucagon and glycerol on blood hormones and metabolites, blood samples were collected on d 1, 7, and 13 postpartum by using a jugular catheter (cat. no. GAS65, Abbott, Redwood City, CA) at -15, -10, -5, 15, 30, 45, 60, 80, 100, 120, 150, 180, 210, 240, 300, 360, 420, and 480min relative to each treatment administration. Twenty milliliters of blood was collected into 2 Vacutainer tubes (Tyco Health Care Group LP, Mansfield, MA) containing Na2-EDTA and placed on ice until plasma was separated by centrifugation at 500×g for 20min at 4°C. Harvested plasma was stored at -20°C until analyzed for concentrations of glucose (kit number G7519-500, Pointe Scientific Inc., Canton, MI), NEFA (NEFA C kit number 994-75409, Wako, Richmond, VA), BHBA (kit number H7587-58, Pointe Scientific Inc., Canton, MI), BUN (kit number B7551-120, Pointe Scientific Inc.), glycerol (kit number F6428, Sigma), and TAG (kit number T7532-120, Pointe Scientific Inc.) adapted for the microplate reader (SPECTRA Max PLUS, Sunnyvale, CA). For plasma hormones determination, 1mL of plasma was mixed with 500 kIU of aprotinin (Boehringer-Mannheim, Indianapolis, IN) and analyzed for glucagon and insulin concentrations via RIA (kit numbers GL-32K and PI-12K, Linco Research Inc., St. Louis, MO).

Statistical Analysis 

The dependent variable was the change of the response variable (plasma glucagon, glucose, insulin, NEFA, BHBA, BUN, and TAG) from baseline concentrations (-15, -10, and -5min before glucagon or glycerol administration) to the concentrations after glucagon or glycerol administration (15, 30, 45, 60, 80, 100, 120, 150, 180, 210, 240, 300, 360, 420, and 480min after glucagon or glycerol administration). For all plasma metabolites and hormones responses, the change was adjusted for differences in baseline and calculated from log-transformed concentrations. The fixed effects were treatment (saline, glucagon, glycerol, glucagon plus glycerol), day of administration (d 1, 7, or 13), time after treatment administration (15, 30, 45, 60, 80, 100, 120, 150, 180, 210, 240, 300, 360, 420, or 480min), and their 2-way interactions. The Kronecker product of a completely unrestricted variance-covariance matrix (for day of administration) and a first-order autoregressive variance-covariance matrix (for time after administration) was used to account for double repeated measures taken on individual cows across time.

The overall effects of glucagon, glycerol, and glucagon plus glycerol administration were evaluated by comparing their estimated changes from baseline concentrations averaged across time after injection (15 to 480min) with the corresponding estimated change of the saline group using Student's t-test. The average effects of glucagon, glycerol, and glucagon plus glycerol administration for the first 4-h period and the entire 8-h posttreatment period were evaluated by calculating the area under the curve minus the area below the baseline concentration by using the trapezoidal rule divided through the first 240 or the entire 480min. The estimated average areas of glucagon, glycerol, and glucagon plus glycerol administration were compared with the corresponding average area of the saline group using a Student t-test. To obtain the correct degrees of freedom the Kenward-Roger option was invoked that uses the Satterthwaite adjustment for degrees of freedom with a Kenward-Roger adjustment on standard errors. Values presented in the figures are raw means and SEM. All statistical tests were 2-sided. Significance was declared at P≤0.05 and tendency toward significance was declared at P≤0.09. We examined additivity of effect of glucagon and glycerol by determining whether the effect of the combined treatment is greater or smaller than the sum of the effect of the individual glucagon and glycerol treatments. Result of significant additivity only will be reported.

Back to Article Outline

Results 

Plasma Constituents 

Plasma Glucagon 

To verify that injected glucagon treatment increased plasma glucagon concentrations, the glucagon content of blood sampled during the first 4-h and entire 8-h periods was evaluated on d 1, 7, and 13 postpartum. Overall, we detected a significant effect (P=0.0001) of treatment on plasma glucagon concentrations during the treatment period. Effects of day and interaction of treatment×day were not significant (P ≥0.12; Figures 1A, B, and C). Glucagon administration increased (all P=0.0001) plasma glucagon concentrations during the first 4h and the entire 8h on d 1, 7, and 13 postpartum. Similarly, administration of glucagon plus glycerol increased (all P=0.0001; Figures 1A, B, and C) plasma glucagon concentrations during the first 4-h and the entire 8-h periods. Conversely, administration of glycerol decreased (P≥0.04; Figures 1A, B, and C) plasma glucagon concentration during the first 4h on d 1, 7, and 13. This effect did not persist (P≥0.12) during the entire 8-h periods on the same days.

  • View full-size image.
  • Figure 1. 

    Plasma glucagon concentrations of cows treated with saline, glycerol, glucagon, or glucagon plus glycerol (GlucGlyc) on d 1, 7, and 13 postpartum. Subcutaneous glucagon injections, given either alone or in combination with glycerol, increased plasma concentrations of glucagon during the first 4 and entire 8h after injections on d 1, 7, and 13 postpartum, respectively (all P<0.001, SEM = 0.13 to 0.23). Precisely, glucagon effect on plasma glucagon concentration was significant between 15 and 300min (P=0.04 and P=0.05), between 15min and 210min (P=0.0001 and P=0.02), and between 15 and 210min (P=0.003 and P=0.02) on d 1, 7, and 13 postpartum, respectively. Glucagon plus glycerol effect was significant between 15 and 360min (P=0.003 and P=0.04), between 15 and 210min (P=0.0003 and P=0.02), and between 15 and 180min (P=0.008 and P=0.02) on d 1, 7, and 13 postpartum respectively. Glucagon plus glycerol also tended to increase plasma glucagon at 240 and 300min (P=0.05 and P=0.06) on d 7 postpartum. Oral glycerol decreased plasma glucagon concentrations during the first 4h postadministration on d 1, 7, and 13 (P=0.05, P=0.04, and P=0.05, respectively; SEM = 0.10 to 0.21) with the effect being nonsignificant for the entire 8-h period (P=0.15, P=0.12, and P=0.14, respectively). No significant effects of day (P=0.12) and treatment×day (P=0.90) on plasma glucagon concentrations were observed.

Plasma Glucose 

Overall, the effects of treatment on plasma glucose concentrations were significant (P=0.0001). There were no effects (P≥0.15) of day or interaction of the treatment×day on plasma glucose concentration. Glucagon administration alone increased glucose concentrations during the first 4 and entire 8h on d 7 and 13 postpartum (P≤0.03; Figures 2B and C), but did not affect plasma glucose concentration on d 1 postpartum (P≥0.34; Figure 2A). Oral glycerol increased plasma glucose concentration on d 7 postpartum during the first 4h (P=0.04) and tended (P=0.06; Figure 2B) to increase it during the entire 8-h period postinjection. The combined glucagon and glycerol treatment increased glucose concentrations during the first 4h and entire 8h on d 1, 7, and 13 postpartum, respectively (P ≤0.002; Figures 2A, B, and C).

  • View full-size image.
  • Figure 2. 

    Plasma glucose (panels A, B, and C) and insulin (panels D, E, and F) concentrations on d 1, 7, and 13 postpartum in cows treated with saline (n=4), glucagon (n=4), glycerol (n=6), or glucagon plus glycerol (GlucGlyc; n=6). For plasma glucose, effects of glucagon were not significant (P≥0.34) during the first 4h or the entire 8-h period on d 1, although a significant effect was detected between 60 and 120min (P=0.02 and P=0.05, respectively). Glucagon effect became significant (P ≤0.03; SEM=0.06 to 0.10) on d 7 and 13 postpartum. In particular, glucagon effect was significant between 15 and 240min (P=0.0001 and P=0.02) and between 30 and 240min (P=0.03 and P=0.02) on d 7 and 13 postpartum, respectively. Overall, effect of treatment and the interaction of treatment×day were significant (P≤0.03) and effect of day was not (P=0.15). Effect of glycerol was not significant (P≥0.22; SEM=0.06 to 0.09) on d 1 and 13 postpartum, although it was significant at 100 and 120min (P=0.04 and P=0.04, respectively) with a tendency at 80min (P=0.07). However, it became significant (P=0.04) during the first 4h and tended (P=0.06) to be significant over the 8-h periods on d 7 postpartum. Overall, effects of treatment, day, and the interaction of treatment×day were not significant (P≥0.11). Glucagon plus glycerol effects were significant (P≤0.008; SEM=0.06 to 0.10) during the first 4h and entire 8h on d 1, 7, and 13 postpartum. Glucagon plus glycerol effects were significant between 15 and 210min (P=0.003 and P=0.01), between 15 and 300min (P=0.002 and P=0.003), and between 30 and 240min (P=0.004 and P=0.03) on d 1, 7, and 13, respectively. Effect of glucagon plus glycerol tended (P=0.07) to be synergistic on d 1. Overall, the effect of treatment was significant (P=0.0001) and those of day and interaction of treatment×day were not significant (P≥0.31). For insulin, effect of glucagon during the first 4h was not significant (P=0.32) on d 1, but was significant (P≤0.004) on d 7 and 13 postpartum. During the entire 8-h period, glucagon effects were significant (P≤0.04) on d 1 and 7 and not (P=0.22; SEM 0.16 to 0.27) on d 13 postpartum. Glucagon effect was significant between 15 and 210min (P=0.01 and P=0.04) on d 7 and between 60 and 210min (P=0.03 and P=0.04, respectively) on d 13 postpartum. Overall, effects of treatment, day, and the interaction of treatment×day were significant (P≤0.01). Glycerol did not influence (P≥0.33; SEM = 0.15 to 0.25) plasma insulin on any day. Overall, effects of treatment and the interaction of treatment×day were not significant (P≥0.25). Effect of day was significant (P=0.01). Effects of glucagon plus glycerol were significant (P≤0.01; SEM = 0.15 to 0.24) at all times on d 1, 7, and 13 postpartum. In particular, the effect of glucagon plus glycerol was significant between 60 and 300min (P=0.0009 and P=0.01), between 15 and 240min (P=0.002 and P=0.01), and between 60 and 240min (P=0.01 and P=0.02). Overall, effects of the treatment were significant (P<0.0001) and synergistic (P=0.02). Effect of day was significant (P=0.01), but the interaction of treatment×day was not significant (P=0.47).

Plasma Insulin 

Significant effects of treatment, day, and treatment×day interaction (P ≤0.01) on plasma insulin concentrations were detected. Glucagon did not change (P=0.32) plasma insulin concentration during the first 4h on d 1 but its effect became significant (P=0.04; Figure 2D) over the entire 8-h period. On d 7 and 13, glucagon increased (P ≤0.02; Figures 2E and F) plasma insulin concentrations during the first 4h and over the entire 8-h period only on d 7 postpartum. This effect was attenuated (P=0.22; Figure 2F) over the 8-h period on d 13 postpartum. No effect (P≥0.33; Figure 2D, E, and F) of glycerol treatment on plasma insulin concentrations was detected during the first 4h or entire 8h on d 1, 7, or 13 postpartum. Glucagon plus glycerol treatment increased plasma insulin concentrations at all times on d 1, 7, and 13 postpartum (P ≤0.01; Figures 2D, E, and F).

Plasma NEFA 

Overall, we observed a significant effect (P ≤0.01) of treatment and the interaction of treatment×day on plasma NEFA concentrations; effects of the day were not significant (P=0.13). Plasma NEFA concentration was steeply decreased (P≤0.005; Figures 3A, B, and C) by the glucagon plus glycerol treatment during the first 4h of d 1, 7 and 13 postpartum. Over the 8-h period, the effect of glucagon plus glycerol was blunted (P=0.25; Figure 3A) on d 1, but remained significant (P ≤0.03; Figures 3B and C) on d 7 and 13 postpartum. During the first 4h postadministration, glucagon alone caused plasma NEFA concentrations to decrease (P≤0.04; Figures 3B and C) on d 7 and 13 postpartum. This effect did not persist (P ≥0.13; Figures 3B and C) for the entire 8h postadministration on the same days. Glucagon alone did not alter (P≥0.76; Figure 2C) plasma NEFA concentration at any time on d 1 postpartum. As illustrated in Figures 3A, B, and C, oral glycerol significantly decreased (P≤0.01) plasma NEFA concentrations during the first 4- and entire 8-h periods on d 1, 7, and 13 postpartum.

  • View full-size image.
  • Figure 3. 

    Plasma NEFA (panels A, B, and C) and BHBA (panels D, E, and F) concentrations on d 1, d 7, and d 13 postpartum in cows treated with saline (n=4), glucagon (n=4), glycerol (n=6), or glucagon plus glycerol (GlucGlyc; n=6). For NEFA, glucagon effects did not differ (P=0.76) at any time on d 1 and differed (P≤0.04; SEM = 0.16 to 0.25) during the first 4h, but were not different (P≥0.13) during the 8-h period on d 7 and 13. Precisely, glucagon effect was significant between 120 and 240min (P=0.02 and P=0.03, respectively) on d 7 and between 100 and 240min (P=0.04 and P=0.04, respectively) overall, whereas effects of treatment and the interaction of treatment×day were significant (P≥0.003), effect of the day was not (P=0.13). Glycerol effects were highly significant (P≤0.01; SEM = 0.15 to 0.24) at all times on d 1, 7, and 13. Glycerol effect was significant between 60 and 360min (P=0.02 and P=0.002) on d 1, between 80 and 240min (P=0.03 and P=0.04) on d 7, and between 45 and 300min (P=0.004 and P=0.05, respectively) on d 13 postpartum. Overall, treatment and the interaction the treatment×day effects were significant (P ≤0.05) and day effect was not (P=0.13). Effects of glucagon plus glycerol were significant (P≤0.005; SEM = 0.15 to 0.23) during the first 4h on d 1, 7, and 13 postpartum. Over the entire 8-h period, these effects were significant (P≤0.03) only on d 7 and 13, but not (P=0.25) on d 1 postpartum. Precisely, glucagon plus glycerol treatment effect was significant between 80 and 300min (P=0.04 and P=0.03) on d 1, between 30 and 210min (P=0.02 and P=0.04) on d 7, and between 45 and 300min (P=0.004 and P=0.05, respectively) on d 13 postpartum. Overall, effect of treatment was significant (P=0.0001), but effects of day and the interaction of treatment×day were not significant (P≥0.30). For plasma BHBA, glucagon effect was not significant (P≥0.21; SEM = 0.11 to 0.16) during the first 4h or entire 8h on d 1, 7, and 13. However, glucagon tended to increase BHBA at 360 and 420min (P=0.09 and P=0.09, respectively) and increased BHBA at 480min (P=0.03). Overall, effects of treatment and the interaction of treatment×day tended to be significant (P≤0.06). Effect of day was not significant (P=0.25). For BHBA, glycerol effects were significant (all P=0.01; SEM = 0.10 to 0.16) during the first 4h and entire 8h on d 1 postpartum. Effect of glycerol was significant between 120 and 300min (P=0.04 and P=0.02, respectively). On d 7 postpartum, glycerol did not affect (P=0.13) BHBA concentration during the first 4h. Over the entire 8-h period, glycerol effect became significant (P=0.04). This effect was most significant at 240min (P=0.03). No effect (P≥0.21) of glycerol was detected on d 13 postpartum. Overall, effect of treatment was significant (P=0.003) and effects of day and the interaction of treatment×day were not (P≥0.25). Effects of glucagon plus glycerol were significant (P≤0.04) during the first 4h on d 1 and 7 postpartum and continued to be significant (P≤0.04; SEM=0.10 to 0.15) over the entire 8-h period on the same days. Precisely, the glucagon effect was significant between 120 and 240min (P=0.05 and P=0.008) with a tendency at 300 and 420min (P=0.06 and P=0.05, respectively). No effect (P ≥0.35) was detected on d 13 postpartum. Overall, effect of treatment was significant (P=0.002) and synergistic (P=0.05), but effects of day and the interaction of treatment×day were not significant (P≥0.25).

Plasma BHBA 

Overall effects of treatment and day on plasma BHBA were not significant (P≥0.25). Effects of the interaction of the treatment×day tended to be significant (P=0.08). Glucagon alone did not affect plasma BHBA concentrations (P ≥0.18; Figures 3D, E and F) during the first 4h and entire 8h on the 3 d of measurement postpartum. Conversely, glucagon plus glycerol treatment decreased (P≤0.04; Figures 3D and E) plasma BHBA concentration during the first 4- and the whole 8-h periods on d 1 and 7 postpartum. Plasma BHBA was not responsive (P≥0.35; Figure 3F) to glucagon plus glycerol treatment on d 13 postpartum. Glycerol treatment decreased (all P=0.01; Figure 3D) plasma BHBA concentration at all times on d 1 post-partum. Although was not significant (P=0.13; Figure 3E) during the first 4h, the effect of glycerol on plasma BHBA became significant (P=0.04) on d 7 after 8h. Glycerol administered on d 13 postpartum did not alter (P≥0.21; Figure 3F) BHBA concentration at any time on that day.

BUN 

The overall effects of treatment on BUN were significant (P=0.02). Nevertheless, effects of day and interaction of treatment×day were not significant (P ≥0.16). Administration of glycerol alone decreased (P ≤0.03; Figure 4A) BUN concentrations during the first 4-h period and the entire 8-h period on d 1 postpartum. A tendency (P=0.06 and P=0.07; Figures 4B and C) of glycerol to decrease BUN was detected during the first 4h on d 7 and 13, respectively, but not (P ≥0.10; Figures 4B and C) over the 8-h period. Similarly, glucagon plus glycerol administration decreased (P ≤0.05; Figure 4B) BUN concentration during the first 4- and the entire 8-h periods on d 7. Glucagon alone did not seem to alter (P=0.58) BUN concentrations on either of d 1, 7, or 13 postpartum.

  • View full-size image.
  • Figure 4. 

    Blood urea nitrogen (panels A, B, and C) and plasma triacylglycerol (TAG, panels D, E, and F) concentrations on d 1, 7, and 13 postpartum in cows treated with saline (n=4), glucagon (n=4), glycerol (n=6), or glucagon plus glycerol (GlucGlyc; n=6). For BUN, glucagon effect was not significant during the first 4 or entire 8h on d 1, 7, and 13 (P≥0.58). The glycerol effect was not significant (P≥0.03) at all times on d 1 and tended to be significant (P=0.07) during the first 4h on d 7 and 13, but was not significant (P≥0.10; SEM = 0.26 to 0.11) over the 8-h periods. Overall, treatment effect tended to be significant (P=0.08) and effects of day and the interaction of treatment×day were not (P≥0.21). For BUN, the glucagon plus glycerol effect was significant (P≤0.05) at the first 4h and entire 8-h period on d 1. A tendency for glucagon plus glycerol started at 30min (P=0.08) and a significant effect started at 80min (P=0.04). Glucagon plus glycerol was not significant (P≥0.19; SEM = 0.26 to 0.10) at any time on d 7 and 13 postpartum. Overall, effects of treatment and day were not significant (P≥0.20) and effect of the interaction of treatment×day was (P=0.05). For TAG, glucagon effect was not significant (P=0.16). Effect of glycerol was significant (P=0.04; SEM = 0.35 to 0.55) during the first 4h on d 1 postpartum only. Specifically, the glycerol effect was significant at 45, 800, and 180min (P=0.03, P=0.04, and P=0.05, respectively). Glycerol also tended to increase plasma TAG at 30, 100, and 150min (P=0.07, P=0.08, and P=0.06) on d 1 postpartum. No effect (P≥0.43; SEM = 0.35 to 0.55) of glycerol was detected on d 7 and 13 postpartum. Overall, effects of treatment, day, and the interaction of treatment×day were not significant (P≥0.10). Glucagon plus glycerol effects were significant between 80 and 120min (P=0.04 and P=0.04; SEM = 0.33 to 0.43) on d 1. Glucagon plus glycerol effect also tended to be significant at 15 and 150min on d 7 (P=0.08 and P=0.07) and on d 1 postpartum, respectively. No effect (P≥0.17 and P≥0.22) of glucagon plus glycerol was detected on d 7 and 13 postpartum, respectively. Overall, the effects of treatment, day, and the interaction of treatment×day were not significant (P≥0.19).

Plasma TAG 

Effect of treatment, day, and the interaction of the treatment×day were not significant (P ≥0.19). Plasma TAG concentration increased (P=0.04; Figure 4D) in cows given the glycerol treatment on d 1 during the first 4h and tended (P=0.08; Figure 4D) to be significant after 8h. Glycerol treatment did not alter plasma TAG concentrations (P≥0.45; Figures 4E and F) at any time on d 7 and 13. Similarly, glucagon and glucagon plus glycerol treatments did not significantly affect (P≥0.24; Figures 4D, E, and F) plasma TAG concentrations during the first 4- or the entire 8-h periods on d 1, 7, and 13. To verify whether administered glycerol facilitates the rapid removal of liver NEFA, the Pearson coefficient for the correlation between plasma NEFA and TAG concentrations was calculated. In the glucagon plus glycerol-treated cows, the Pearson coefficient for the correlation was -0.28 on d 1 postpartum and tended (P=0.08) to be significant. In the glucagon-treated group, the Pearson coefficients for the correlation between plasma NEFA and TAG concentrations were 0.71 and 0.78 (All P=0.0001) on d 1 and 7 postpartum, respectively.

Back to Article Outline

Discussion 

The objective of the present study was to examine whether glucagon and glycerol, when co-administered, act to improve the peripartal metabolic parameters of Holstein dairy cows induced with experimental FLS. Precisely, we wanted to understand and elucidate the biochemical mechanisms whereby glucagon and glycerol may improve the peripartal metabolism efficiency and treat FLS.

Because DMI decreases (Doepel et al., 2002) during the peripartal period, dietary glucogenic amino acids and propionate from rumen fermentation are in short supply. Furthermore, the need for amino acids to support postpartal colostrum and milk protein may lessen their contribution to net hepatic glucose synthesis (Overton and Waldron, 2004). Therefore, gluconeo-genesis is deprived of a suitable amount of substrates. Subsequently, plasma glucose concentrations would be inadequate and the dairy cow will develop NEB and FLS. Therefore, the rationale for using glycerol as a glucogenic precursor in this study was based on the assumption that, whereas other gluconeogenic substrates are in a short supply during the peripartal period, a more versatile intermediate such as glyceraldehyde-3-phosphate (G-3-P) may be needed to keep both gluconeogenesis and glycolysis accelerated.

In this study, administration of glucagon alone or glucagon plus glycerol starting immediately after parturition increased plasma glucagon concentrations to about 700 pg/mL within 15min on d 1, 7, and 13 postpartum (Figure 1). These findings are consistent with previously published results (Bobe et al., 2003a). These results further corroborate earlier findings that the subcutaneous route is reliable for glucagon administration and produces reproducible results (Nafikov et al., 2006).

Plasma glucose concentration was increased by administration of glucagon plus glycerol but not by administration of glucagon alone on d 1 postpartum (Figure 2A). Inadequate postpartal liver glycogen likely impeded the ability of glucagon to increase plasma glucose. It is possible that most of the glycogen was converted to glucose to support milk lactose synthesis (unpublished data). Additionally, the inadequate peripartal DMI coupled with the need to partition amino acids to support colostrum and milk protein synthesis probably deprived gluconeogenesis of a suitable supply of gluconeogenic substrates. The ability of glucagon plus glycerol to increase plasma glucose could be attributed to the ability of the glucagon-upregulated gluconeogenesis to recruit the G-3-P derived from glycerol to provide the needed carbon atoms for glucose synthesis.

In this study, glycerol treatment increased plasma glucose, which is consistent with other results (Goff and Horst, 2001). It was noticeable that plasma glucose responses increased with greater glycerol dosages for a longer time (Goff and Horst, 2001).

On d 7 postpartum, both glucagon and glucagon plus glycerol treatments increased plasma glucose concentrations (Figure 2B). The ability of glucagon alone to increase plasma glucose concentrations on d 7 postpartum may be facilitated by the improved DMI, which provided more gluconeogenic substrates to fuel gluconeogenesis. Similarly, on d 13 postpartum, glucagon and glucagon plus glycerol treatments increased plasma glucose (Figure 2C) above that of the control. The increase in concentrations of plasma glucose caused by glucagon observed in this study is consistent with that of Bobe et al. (2003b) and Nafikov et al. (2006).

It is reasonable to speculate that when glycerol is administered alone or in an amount <1L, it is mostly converted to 1,3-bisphosphoglycerate (1,3-BPG), which supplies ATP and pyruvate in glycolysis. Increased availability of ATP would alleviate the need for more glucose synthesis and spare glucose for lactose synthesis. Consistent with this concept is the acute decrease in plasma glucagon concentrations caused by glycerol in this study. When co-administered with glucagon (this study) or in a larger amount (Goff and Horst, 2001), glycerol is likely converted to fructose-1,6-bisphosphate (F-1,6-P2), which then supplies glucose through gluconeogenesis (Figure 2A, B, C, and Figure 5).

  • View full-size image.
  • Figure 5. 

    Metabolic use of glycerol. α-GP = α-glycerolphosphate; G-3-P = glyceraldehyde-3-phosphate; F-1,6-P2 = fructose-1,6-bisphos-phate; F-6-P = fructose-6-phosphate; G-6-P = glucose-6-phosphate; 1,3-BPG = 1,3-bisphosphoglycerate; 3-PG = 3-phosphoglycerate; 2-PG = 2-phosphoglycerate; PEP = phosphoenol pyruvate; TAG = triacylglycerol; VLDL = very low density lipoprotein.

The increase in postpartal plasma insulin concentrations caused by the glucagon plus glycerol treatment on d 1, 7, and 13 postpartum is consistent with the concurrent increase in plasma glucose concentrations caused by the same treatments (Figures 2D, E, and F). The glycerol treatment did not increase plasma insulin concentrations during the postpartal period because its effect on plasma glucose concentrations was small (Figures 2D, E, and F) during the same period. Thus, our data support previous findings that glucagon indirectly increases plasma insulin concentrations by increasing plasma glucose via signals through the Gas class glucagon receptor (Jiang and Zhang, 2003). Furthermore, the signaling of glucagon through the Gq class glucagon receptor to directly increase plasma insulin concentrations (Gromada et al., 1998) is probably only effective when plasma glucose is increased simultaneously.

On d 7 postpartum, both glucagon and glucagon plus glycerol treatments increased plasma insulin concentrations by 27.4 and 33.1μU/mL, respectively (Figures 2E and F), which are significantly greater than that of the control. The increase in plasma insulin concentrations caused by glucagon on d 7 and 13 postpartum (Figures 2B and C) is consistent with previous published findings (Bobe et al., 2003a) and may be attributed to the concurrent increase in plasma glucose caused by glucagon treatment (Figures 2B and C).

Glucagon administration significantly decreased plasma NEFA concentrations (Figures 3B and C), which lends support to other results (Bobe et al., 2003a). Oral administration of glycerol steeply decreased plasma NEFA concentrations (Figures 3A, B, and C), which is also similar to results of other studies (Bodarski et al., 2005). Because glycerol only slightly increased plasma glucose and did not increase plasma insulin concentrations, the mechanism whereby glycerol decreases plasma NEFA concentrations might be through increasing its esterification into TAG in the liver, not the adipose tissue. Furthermore, glycerol can be absorbed through the ruminal wall and phosphorylated in the liver by glycerol kinase to α-glycerolphosphate (α-GP). The liver could use α-GP to esterify NEFA into TAG before it is exported to extrahepatic tissues (Figures 3A, B, and C and Figures 4D, E, and F). Therefore, availability of α-GP probably enhances liver uptake of NEFA from the blood decreasing plasma NEFA concentrations.

We speculate that one purpose of the accelerated peripartal lipolysis is to supply glycerol that can be converted to G-3-P in the liver and utilized in gluconeogenesis to synthesize glucose to support milk lactose or in glycolysis to directly produce the needed energy in the form of ATP and pyruvate. In addition, because G-3-P enters gluconeogenesis after the reactions catalyzed by the rate-limiting enzymes pyruvate carboxy-lase and phosphoenolpyruvate carboxykinase (Goff and Horst, 2001) are completed, glycerol can be used more efficiently than most other gluconeogenic substrates. Moreover, the overwhelming of the liver by NEFA during the peripartal period could be explained by the fact that for each demanded molecule of glycerol, the liver has to cope with 3 molecules of excess NEFA. Therefore, because of the short supply of α-GP that is needed to esterify NEFA into TAG, NEFA can accumulate in the liver. It could also be speculated that accumulated NEFA may disturb the charge balance in hepatocytes because of their negatively charged carbonyl tail, which could play an active role in the pathogenesis of FLS.

The plausible mechanism by which the glucagon plus glycerol treatment decreased plasma NEFA concentrations (Figures 3A, B, and C) is likely through increasing the postpartal plasma glucose and insulin concentrations. Whereas increased plasma glucose improves the metabolism and energy status of the cow and lessens the need for lipolysis, increased plasma insulin increases uptake and esterification of NEFA by the adipose tissues (Brockman et al., 1975). This further explains why glucagon plus glycerol, unlike glycerol alone, did not significantly increase plasma TAG concentration (Figures 4D, E, and F).

In general, plasma BHBA concentrations increased in all treatment groups during wk 1 and decreased during wk 2 postpartum in accordance with the decrease of plasma NEFA concentrations. Glucagon, however, did not affect plasma BHBA concentrations on d 1, 7, and 13 postpartum. The lack of effects of glucagon on plasma BHBA is consistent with other published results (Bobe et al., 2003a). Glycerol administration decreased plasma BHBA concentrations during the postpartal period, which agrees with the results of DeFrain et al. (2004) who fed 430 g/d of glycerol. Additionally, Goff and Horst (2001) drenched dairy cows with glycerol and observed a decrease in urinary ketones and symptoms of ketosis complicated with FLS. It is possible that glycerol decreased plasma BHBA concentrations postpartum by increasing NEFA esterification and accordingly the disposal of NEFA from the liver as TAG. It is known that glycerol is fermented in the rumen to propionate, acetate, and butyrate (Remond et al., 1993). Furthermore, the omasal and ruminal epithelium convert butyrate to BHBA to provide energy and lessen the toxic effect of butyrate on digestive mucosa (Schröder and Südekum, 1999). Thus, feeding glycerol at 500 g/d (Bodarski et al., 2005) or greater (DeFrain et al., 2004) may increase plasma BHBA. Because these increased plasma BHBA concentrations, when a high amount of glycerol is fed, are not the result of down-regulation of the tricarboxylic acid cycle, it is not likely related to ketosis development. Furthermore, glucagon plus glycerol decreased postpartal plasma BHBA concentrations (Figures 3D, E, and F) possibly because this treatment substantially decreased concentrations of plasma NEFA (Figures 3A, B, and C), the precursor of BHBA.

Although direct stimulation of the uptake of amino acids by the liver (Brockman et al., 1975) and an increase in protein degradation are attributes of glucagon, glucagon alone did not increase postpartal BUN concentrations in our study. In contrast, the glucagon plus glycerol treatment caused BUN to decrease (Figures 4A, B, and C) possibly because the glucagon-stimulated gluconeogenesis recruited the glycerol component as a glucogenic precursor and spared the use of glucogenic amino acids. In agreement with this speculation is the significant decrease in BUN concentrations caused by the glycerol treatment in this study.

Glucagon administration did not affect plasma TAG concentrations (Figure 2E). However, glycerol and the glucagon plus glycerol treatment increased plasma TAG concentrations in the first 4h posttreatment (Figures 4D, E, and F). It is possible that glycerol, after conversion to α-GP, increased NEFA disposal from the liver in the form of TAG. Although the correlation between plasma NEFA and TAG concentrations was small (-0.28) in the cows given the glucagon plus glycerol treatment, it suggests that glycerol possibly facilitated the removal of liver NEFA in the form of TAG-rich very low density lipoprotein. Furthermore, the increase in plasma TAG concentration immediately after the administration of the glycerol and the glucagon plus glycerol treatments indicates possible shortages in α-GP availability in the liver. Limited availability of α-GP might cause NEFA to overwhelm the liver, inducing FLS disease. This possible concept is further supported by the significant, positive Pearson coefficients (0.71 to 0.78) for the correlation between plasma NEFA and TAG concentration observed in cows treated with the glucagon treatment alone on d 1 and 7 postpartum, respectively.

Back to Article Outline

Conclusions 

The results of this study suggest that glucagon and glycerol, when co-administered during early lactation, act to increase plasma glucose and insulin concentrations and decrease plasma NEFA and BHBA concentrations. Therefore, co-administration of glucagon and glycerol may have the potential to positively influence the postpartal metabolism of dairy cows and subsequently treat other metabolism-related peripartal syndromes and infectious health disorders than either glycerol or glucagon alone.

Back to Article Outline

Acknowledgments 

The authors thank Eli Lilly (Indianapolis, IN) for donation of glucagon and the US Department of Agriculture for financing the study (grant number 2002-02060). Our appreciation is extended to the management of Iowa State University's Teaching Dairy Farm (Ankeny, IA) for providing the research cows, to Derek Widman (Dept. Anim. Sci., Iowa State University) for assisting with sample collection, and to Gayle Johnson (Dept. Chemistry, Iowa State University) and Almass Abuzaid (Dept. Biochem., Biophys., Mol. Biol., Iowa State University) for assisting with laboratory analysis.

Back to Article Outline

References 

  1. Bobe G, Ametaj BN, Young JW, Beitz DC. Effects ofexogenous glucagon on lipids in lipoprotein and liver of lactating dairy cows. J. Dairy Sci. 2003;86:2895–2903
  2. Bobe G, Sonon RN, Ametaj BN, Young JW, Beitz DC. Metabolic responses of lactating dairy cows to single and multiple injections of glucagon. J. Dairy Sci. 2003;86:2072–2081
  3. Bobe G, Young JW, Beitz DC. Invited review: Pathology, etiology, prevention and treatment of fatty liver syndrome in dairy cows. J. Dairy Sci. 2004;87:3105–3124
  4. Bodarski R, Wertelecki T, Bommer F, Gosiewski S. The changes of metabolic status and lactation performance in dairy cows under feeding TMR with glycerin (glycerol) at periparturient period. Electronic J. Polish Agric. Univ. 2005;8:22
  5. Brockman RP, Bergman EN, Joo PK, Manns JG. Effects of glucagon and insulin on net hepatic metabolism of glucose precursors in sheep. Am. J. Physiol. 1975;229:1344–1349
  6. Castaneda-Gutierrez E, Overton TR, Butler WR, Bauman DE. Dietary supplements of two doses of calcium salts of conjugated linoleic acid during the transition period and early lactation. J. Dairy Sci. 2005;88:1078–1089
  7. DeFrain JM, Hippen AR, Kalscheur KF, Jardon PW. Feeding glycerol to transition dairy cows: Effects on blood metabolites and lactation performance. J. Dairy Sci. 2004;87:4195–4206
  8. Doepel L, Lapierre H, Kennelly JJ. Peripartum performance and metabolism of dairy cows in response to prepartum energy and protein intake. J. Dairy Sci. 2002;85:2315–2334
  9. Douglas GN, Overton TR, Bateman HG, Drackley JK. Peripartal metabolism and production of Holstein cows fed diets supplemented with fat during the dry period. J. Dairy Sci. 2004;87:4210–4220
  10. Goff JP, Horst L. Physiological changes at parturition and their relationship to metabolic disorders. J. Dairy Sci. 1997;80:1260–1268
  11. Goff JP, Horst RL. Oral glycerol as an aid in the treatment of ketosis/fatty liver complex. J. Dairy Sci. 2001;84(Suppl. 1):153–154(Abstr.)
  12. Gromada J, Holst JJ, Rorsman P. Cellular regulation of islet hormone secretion by the incertin hormone glucagon like peptide 1. Eur. J. Phys. 1998;435:583–594
  13. Grum DE, Drackley JK, Younker RS, LaCount DW, Veenhuizen JJ. Nutrition during the dry period and hepatic lipid metabolism of periparturient dairy cows. J. Dairy Sci. 1996;79:1850–1864
  14. Grummer RR. Etiology of lipid-related metabolic disorders in periparturient dairy cows. J. Dairy Sci. 1993;76:3882–3896
  15. Hayirli A, Bertics SJ, Grummer RR. Effects of slow- release insulin on production, liver triglyceride, and metabolites profiles of Holsteins in early lactation. J. Dairy Sci. 2002;85:2180–2191
  16. Hippen AR, She P, Young JW, Beitz DC, Lindberg GL, Richardson LF, et al. Metabolic responses of dairy cows and heifers to various intravenous dosages of glucagon. J. Dairy Sci. 1999;82:1128–1138
  17. Hoedemaker M, Prange D, Zerbe H, Frank J, Daxenberger A, Meyer HHD. Peripartal propylene glycol supplementation and metabolism, animal health, fertility and production in dairy cows. J. Dairy Sci. 2004;87:2136–2145
  18. Jiang G, Zhang BB. Glucagon and regulation of glucose metabolism. Am. J. Physiol. Endocrinol. Metab. 2003;284:E671–E678
  19. Jorritsma R, Wensing T, Kruip TAM, Vos PLAM, Noordhuizen JPTM. Metabolic changes in early lactation and impaired reproductive performance in dairy cows. Vet. Res. 2003;34:11–26
  20. Katoh N, Nakagawa-Ueta H. Concentrations of apolipoprotein C-III in healthy cows during the peripartum period and cows with milk fever. J. Vet. Med. Sci. 2001;63:597–601
  21. LeBlanc SJ, Leslie KE, Duffield TF. Metabolic predictors of displaced abomasum in dairy cattle. J. Dairy Sci. 2005;88:159–170
  22. Nafikov RA, Ametaj BN, Bobe G, Koehler KJ, Young JW, Beitz DC. Prevention of fatty liver in transition dairy cows by subcutaneous injections of glucagon. J. Dairy Sci. 2006;89:1533–1545
  23. NRC. Nutrient Requirements of Dairy Cattle. 7th rev. ed.. Washington, DC: Natl. Acad. Press; 2001;
  24. Overton TR, Waldron MR. Nutritional management of transition dairy cows: Strategies to optimize metabolic health. J. Dairy Sci. 2004;87(E Suppl.):E105–E119
  25. Remond B, Souday E, Jouany JP. In vitro and in vivo fermentation of glycerol by rumen microbes. Anim. Feed Sci. Technol. 1993;41:121–132
  26. Schröder A, Südekum K-H. Glycerol as a by-product of biodiesel production in diets for ruminants. New Horizon for an Old Crop. In:  Wartten N,  Salisbury PA editor. Proc. 10th Int. Rapeseed Congr., Canberra, Australia. Gosford, Australia: The Regional Institute Ltd; 1999;Paper No. 241
  27. Veenhuizen JJ, Drackley JK, Richard MJ, Sanderson TP, Miller LD, Young JW. Metabolic changes in blood and liver during development and early treatment of experimental fatty liver and ketosis in cows. J. Dairy Sci. 1991;74:4238–4253
  28. Zerbe H, Schneider N, Leibold W, Wensing T, Kruip TAM, Schuberth HJ. Altered functional and immunophenotypical properties of neutrophilic granulocytes in postpartum cows associated with fatty liver. Theriogenology. 2000;54:771–786

PII: S0022-0302(08)71046-4

doi:10.3168/jds.2008-0997

Journal of Dairy Science
Volume 91, Issue 9 , Pages 3311-3322, September 2008

Article Tools

* Image Usage & Resolution