|
|
||||||||
ANIMAL GROWTH, PHYSIOLOGY, AND REPRODUCTION |
Division of Animal and Veterinary Sciences, Davis College, West Virginia University, Morgantown 26506
| Abstract |
|---|
|
|
|---|
Key Words: glucagon hepatocyte insulin progesterone progesterone clearance
| INTRODUCTION |
|---|
|
|
|---|
Serum progesterone is elevated within the first 3 wk of pregnancy when ewes are fed below their maintenance energy requirement (Cumming et al., 1971
; Parr et al., 1982
). When underfed, ovariectomized, pregnant ewes were given exogenous progesterone, peripheral progesterone concentrations were elevated compared with ewes not underfed, indicating an impact on metabolic clearance, rather than synthesis, of progesterone (Parr et al., 1982
). Studies designed to investigate the relationship between nutrition and progesterone clearance unfortunately have confounded DMI with ME (Parr, 1992
).
In ruminants, carbohydrates in the feed are largely converted to VFA as a result of microbial fermentation so that concentrations of blood glucose are less than monogastrics and vary little with feeding (De Jong, 1982
). Intravenous administration of VFA uniquely stimulates insulin secretion from the pancreas in ruminants (Horino et al., 1968
; De Jong, 1982
; Harmon, 1992
).
It was hypothesized that changing the form of energy, while balancing for DMI and ME, would alter pancreatic insulin and glucagon secretion, as well as the metabolic clearance of progesterone. Moreover, it was hypothesized that challenging hepatocytes directly with insulin or glucagon would alter progesterone clearance in vitro.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Preliminary Study
A preliminary study was conducted to characterize the magnitude and duration of the increase in hepatic portal vein concentrations of VFA after an oral gavage of 0.146 Mcal of VFA. Four crossbred, yearling ewe lambs (BW = 45.5 ± 2.5 kg) were housed in individual pens (2.1 m x 2.1 m). Lambs were fed grass hay (once daily at 0800) at 2% of BW (89.2% DM, 9.2% CP, 53.9% TDN) to meet NRC (1985)
maintenance requirements. The lambs were assigned randomly to 1 of 2 treatments for an 11-d acclimation period: a once-daily oral gavage of 0.7 mol of sodium acetate in 200 mL of H2O (lot 81K0206, Sigma Chemical Co., St. Louis, MO) or 0.4 mol of sodium propionate in 200 mL of H2O (lot 021K0158, Sigma Chemical Co.). The energy content of the gavage (0.146 Mcal) was equal between the sodium acetate and sodium propionate treatments and represented approximately 10% of the daily energy requirement of the ewe lambs (NRC, 1985
).
Before the acclimation period, all lambs were determined to be anestrus, defined by serum progesterone concentrations <0.7 ng/mL for 2 blood samplings taken 6 d apart (7 and 1 d; Knights et al., 2001
). A portal vein catheter was inserted in each lamb according to the method of Ferrell et al. (1992)
. After induction and maintenance of anesthesia, a paracostal incision (~34 cm) was made parallel to and 5 cm posterior to the last rib. The portal vein was cannulated approximately 8 cm from the liver utilizing a 14-ga needle, and a heparinized (TDMAC heparin complex, lot 73733, Polyscience Inc., Warrington, PA) catheter (i.d. = 1.0 mm, o.d. = 1.8 mm; lot 50466, Cole-Parmer Instrument Co., Vernon Hills, IL) was inserted (~6 cm) until the tip was at the liver. A purse-string suture was tied around the catheter to secure it inside the portal vein. The catheter was sutured to the portal lymph node to further stabilize its placement. The free end of the catheter was tunneled under the skin using a trocar and was exteriorized near the middle of the back. After catheter patency was confirmed, the catheter was filled with heparinized saline (100 IU of heparin/mL of 150 mM NaCl) and tied off. One animal assigned to acetate treatment lost portal vein catheter patency before blood sampling and was removed from the experiment.
On d 12 after the beginning of the acclimation period, 3-mL samples of both portal and jugular venous blood were simultaneously collected at frequent intervals (0.5, 0.5, 1, 2, 3, 4, 5, 6, and 7 h) with respect to feeding and the oral gavage of VFA. Blood was combined with 100 µL of heparinized saline and was stored at 4°C for <1 h until being centrifuged at 3,000 x g for 15 min. After centrifugation, the plasma was aspirated and then frozen until assayed for VFA. To determine the plasma acetate and propionate concentrations from the feed-alone treatment, d-13 blood samples were collected in a similar manner as on d 12, except that sodium acetate and sodium propionate were not gavaged orally.
To determine plasma VFA concentrations, plasma (1 mL) was extracted with 5 mL of 100% ethanol, the precipitate was removed by centrifugation (3,000 x g for 15 min), and the supernatant was mixed with 100 µL of sodium hydroxide (0.2 M) and air-dried. The dry residue was reconstituted in 20 µL of 30 mM oxalic acid, and 1 µL of the reconstituted sample was injected onto a 2-m long x 2-mm i.d. glass column (80/120 Carbopack B-DA/4% Carbowax 20M, Supelco Inc., Bellefonte, PA) and subjected to GLC (Varian 3300 Gas Chromatograph and Varian 4290 Integrator, Varian Inc., Walnut Creek, CA; Remesy and Demigne, 1974
).
Exp. 1
Thirty crossbred, yearling, anestrous ewe lambs (16 in replicate 1 and 14 in replicate 2; BW 45.2 ± 1.9 kg) were blocked by BW and housed (3-m x 3-m pens) with 2 ewes per pen. Within each pen, each lamb was assigned randomly to 1 of 2 treatments: an oral gavage of either sodium acetate or sodium propionate, as in the preliminary study. During an 11-d acclimation period, lambs were fed grass hay once daily at 0700 at 2% of BW on an as-fed basis (89.2% DM, 9.2% CP, 53.9% TDN) to meet NRC (1985)
maintenance requirements. All lambs were anestrus, which was defined as serum progesterone concentrations <0.7 ng/mL at 2 samplings 6 d apart (Knights et al., 2001
).
On d 12, after the acclimation period, animals were injected i.m. with 20 mg of progesterone (lot 100K0204, Sigma Chemical Co.) in corn oil (20 mg/mL) and gavaged orally with either sodium acetate (0.7 mol) or sodium propionate (0.4 mol) as in the preliminary study. Blood samples were collected via jugular venipuncture at 0.5, 0.5, 1, 2, 3, 4, 5, 6, and 8 h relative to feeding, VFA gavage, and progesterone treatment. Blood was stored at 4°C for 24 h and centrifuged at 3,000 x g for 15 min; the serum was aspirated and frozen until assayed. Jugular serum progesterone concentrations were determined by RIA (Sheffel et al., 1982
) with a sensitivity of 100 pg/mL and intra- and interassay CV of 6.8 and 7.1%, respectively.
Exp. 2
Five crossbred, yearling, anestrous ewe lambs (55.6 ± 1.37 kg) were housed in 3-m x 3-m pens (2 ewe lambs in 1 pen, and 3 ewe lambs in an adjacent pen). During an 11-d acclimation period, lambs were fed grass hay once daily at 0700 at 2% of BW on an as-fed basis (90.6% DM, 14.7% CP, 64.8% TDN) to meet NRC (1985)
maintenance requirements.
On d 12, after the acclimation period, each pen was assigned to 1 of 2 treatments, and animals were gavaged orally with either sodium acetate (0.7 mol) or sodium propionate (0.4 mol) as in the preliminary study and Exp. 1. On d 17, each pen was treated alternatively with sodium acetate or sodium propionate in a crossover design. Blood samples were collected via jugular venipuncture at 0.5, 0.5, 1, 2, 3, 4, 5, 6, and 8 h relative to feeding and VFA gavage. Blood was combined with 100 µL of heparinized saline and stored at 4°C for <1 h until centrifuged at 3,000 x g for 15 min. The plasma was aspirated and frozen until assayed for concentrations of glucagon (double-antibody glucagon RIA; lot 0021, Diagnostic Products Corp., Los Angeles, CA) and insulin (insulin ELISA; lot 05264, Diagnostic Systems Laboratories, Inc., Webster, TX) according to the manufacturers instructions. Both assays exhibited parallelism between the respective standard curves and a serial dilution of plasma samples. The glucagon assay had a sensitivity of 3.73 pM and intra- and interassay CV of 5.5 and 9.1%, respectively. The insulin assay had a sensitivity of 174 pM and intra- and interassay CV of 4.5 and 6.5%, respectively.
Exp. 3
A search of the American Type Culture Collection revealed no characterized hepatocyte cell line derived from a ruminant animal; therefore, a hepatocyte cell line derived from a nonruminant was employed. Hepatocytes (cell line FL83B, catalogue number CRL-2390) were obtained from American Type Culture Collection (Manassas, VA). Fetal bovine serum (lot 1125143), penicillin-streptomycin (10,000 U and 10,000 µg/mL, respectively; lot 15140122), and Hanks balanced salt solution (lot 14170112) were obtained from Invitrogen (Carlsbad, CA). Culture medium (F-12K; lot 3000357) was purchased from American Type Culture Collection. Insulin (from bovine pancreas; lot 064K8403), glucagon (from porcine pancreas; lot 123K8928), progesterone (lot 100K0204), and all other reagents were purchased from Sigma Chemical Co.
Hepatocytes (105 per well) were plated in five 12-well plates in 1.0 mL of F-12K medium with 10% fetal bovine serum and 1% penicillin-streptomyocin. To allow the cells to adhere to the wells and to grow to approximately 40% confluency, the plates were incubated for 12 h at 37°C and 5% CO2. After the initial incubation period, the medium was aspirated and replaced with 1.0 mL of fresh medium with the addition of 5 ng of progesterone/mL and 1 of 4 treatments (0, 0.1, 1.0, or 10 nM insulin) in triplicate (i.e., 3 wells per treatment).
Five times (0, 1, 2, 3, and 4 h) were used to calculate the fractional rate constant of progesterone decay [(A) = (A)0ekt, where A is the reactant (progesterone), k is the first-order fractional rate constant, and t is time]. The result of a preliminary experiment demonstrated that from 0 to 4 h, the cells continued to grow in the presence of 5 ng of progesterone/mL, reaching confluency at approximately 4 h. At each time, one plate was removed from the incubator, and the medium was aspirated from each well into a microcentrifuge vial. For 0 h, 1 mL of medium containing treatment was placed in each well and immediately aspirated and placed into a microcentrifuge vial. The other 4 plates were placed in the incubator at 37°C and 5% CO2 and remained undisturbed until the appropriate sampling time.
The microcentrifuge vials were centrifuged at 3,000 x g and were decanted into new vials, which were frozen until assayed for progesterone. The experiment was replicated 3 times (i.e., 3 wells per treatment at each time with the experiment replicated 3 times or 45 wells per treatment in all replicates). Progesterone concentrations were determined in conditioned media by RIA (Sheffel et al., 1982
) with a sensitivity of 100 pg/mL and intra-and interassay CV of 6.5 and 7.6%, respectively.
Cells were also cultured in the presence of glucagon exactly as with insulin, except treatments included 0, 0.01, 0.1, or 1.0 nM glucagon in place of insulin. Finally, cells were cultured in the presence of both insulin and glucagons, as with insulin or glucagon alone, except the treatments contained insulin and glucagon including 0 and 0, 1.0 and 1.0, 1.0 and 0.1, or 0.1 and 1.0 nM insulin and glucagon, respectively.
Statistics
Exp. 1.
Statistical models included treatment effect (sodium acetate or sodium propionate), time, replicate, and a treatment x time interaction in a randomized block design. Preliminary analysis indicated no differences between replicates of the experiment; therefore, the data were combined and analyzed together. Concentrations of progesterone were analyzed using Proc Mixed for repeated measures [ewe (treatment)] utilizing the autoregressive covariance structure, and means separation was performed using the PDiff option of the LSMeans procedure of SAS (SAS Inst., Inc., Cary, NC).
Exp. 2.
Statistical models included treatment effect (acetate or propionate), time, and a treatment x time interaction in a crossover design. Data were transformed to meet the assumptions of normality. Plasma concentrations of insulin (cube root transformed) and glucagon (log transformed) were analyzed using Proc Mixed for repeated measures utilizing the autoregressive covariance structure, and means separation was performed using the PDiff option of the LSMeans procedure of SAS.
Exp. 3.
Fractional rate constants were calculated for each treatment and were then expressed as a percentage of the rate constant calculated for the control treatment. Statistical models included treatment effect (concentration of insulin or glucagons or both) in a complete block design. Preliminary analysis showed no differences among the 3 replicates of each experiment; therefore, the data were combined and analyzed together. The adjusted fractional rate constants were analyzed using Proc GLM, and means separation was performed using the Duncans procedure of SAS.
| RESULTS |
|---|
|
|
|---|
|
|
Exp. 3
The fractional rate constant for progesterone clearance in this system was 14.6 ± 0.5%/h in the absence of exogenous insulin or glucagon. When hepatocytes were challenged with insulin, there was a dose-dependent decrease in progesterone clearance (Figure 3A
). There was a reduction (P < 0.05) in the rate of decay with the addition of 0.1 nM insulin when compared with the control (Figure 3A
). Further, there was a greater reduction in the rate of decay in response to 1.0 and 10 nM insulin (P < 0.05) than the control and 0.1 nM insulin. There was no observable change in progesterone clearance with either physiological (0.01 and 0.1 nM) or supraphysiological (1.0 nM) treatment of hepatocytes with glucagon (Figure 3B
). Supraphysiological concentrations of glucagon (1.0 nM) negated (Figure 3C
) the effects of either 0.1 or 1.0 nM insulin on the progesterone clearance observed when hepatocytes were challenged with insulin alone (Figure 3A
). However, when challenged with physiological concentrations of glucagon (0.1 nM) and 1.0 nM insulin, glucagon was not able to negate the reduction in progesterone clearance caused by insulin (Figure 3C
).
|
| DISCUSSION |
|---|
|
|
|---|
It has been shown that at least 2 ng of serum progesterone/mL is necessary for satisfactory conception rates (Parr et al., 1987
). Ewes fed a below-maintenance diet had greater peripheral progesterone concentrations, but on d 5 of the estrous cycle, had reduced endometrial progesterone concentrations when compared with ewes fed in excess of maintenance (Lozano et al., 1998
). Progesterone supplementation had no effect on ewes that were fed at or below maintenance. However, in ewes fed in excess of maintenance, progesterone supplementation increased the pregnancy rate from 48 to 76% (Parr et al., 1987
). The researchers concluded that exogenous progesterone will increase pregnancy rate only in ewes fed in excess of maintenance. Beyond the effect observed with drastic alterations in DMI, the present data show that at equal energy and DMI, there are alterations in progesterone clearance as a result of the form of energy that could impact the ability of an animal to maintain pregnancy.
Less progesterone was cleared from plasma after sodium propionate gavage than after sodium acetate gavage. These differences could be mediated by alterations in hepatic blood flow. Bensadoun and Reid (1962)
hypothesized that the increased clearance of progesterone from ewes fed high-energy diets may be due to increased blood flow to the liver. Others have demonstrated an association between liver blood flow and progesterone clearance in dairy cows (Sangsritavong et al., 2002
). In this experiment, hepatic portal blood flow was not determined. Bensadoun and Reid (1962)
demonstrated that blood flow to the liver of ewes increases between 3 and 7 h after feeding. Similarly, mean hepatic blood flow increased by about 45% at 3 h after feeding (Katz and Bergman, 1969
) and was influenced by the amount of energy in the diet (Parr et al., 1993b
). Ruminal infusion of VFA has been shown to increase hepatic portal blood flow (Sellers et al., 1964
). In sheep, ruminal infusion of propionate resulted in a 5-fold greater increase in ruminal arterial blood flow than an isoenergetic infusion of acetate (Sellers et al., 1964
). If changes in blood flow were involved in the observed alteration in progesterone clearance, one would expect that a greater increase in hepatic portal blood flow after the sodium propionate gavage would have increased progesterone clearance. Instead, the opposite occurred, i.e., progesterone clearance was reduced after the sodium propionate gavage as compared with the sodium acetate treatment.
Alterations in the type of feedstuffs or the physiological status of the ewe may play a greater role in embryo survival than previously thought. Further, progesterone clearance was altered without alterations in DMI or energy and was opposite to that which would be expected based on likely changes in hepatic portal blood flow. In this experiment, balancing DMI and energy, while altering the VFA profile leaving the rumen, resulted in alterations in progesterone clearance. Therefore, alterations in feedstuffs may influence clearance of progesterone and potentially affect pregnancy rate.
In the present experiment, the amount of sodium acetate or sodium propionate gavaged orally into the rumen was physiologically relevant, representing approximately 10% of the daily energy requirement of the ewe lamb. However, the hepatic portal vein VFA concentration in the 4 h after the oral gavage was supraphysiological. In the ruminant, acetate is the major form of circulating energy in the fed state (De Jong, 1982
). Acetate use is dependent on insulin (Baile and Mayer, 1967
; Skarda and Bartos, 1969
; Schwalm and Schultz, 1976
). However, this experiment showed no change in plasma insulin concentrations after an oral gavage of acetate. This result is in agreement with other researchers who reported similar results in sheep (Manns and Boda, 1967
; Trenkle, 1970
), goats (De Jong, 1982
), and cattle (McAtee and Trenkle, 1971
). Similar to De Jong (1982)
, in the current experiment, acetate also did not effect glucagon concentrations.
In Exp. 2, an oral gavage of sodium propionate did stimulate increased plasma concentrations of both insulin and glucagon in sheep. This result is in agreement with De Jong (1982)
, who reported similar findings in goats. It has been hypothesized that VFA act directly on the
and ß cells of pancreatic islets of Langerhans. There is the possibility that glucagon is an intermediary between VFA and insulin secretion because glucagon is a potent stimulator of insulin secretion (Bassett, 1971
; Samols et al., 1972
). In this study, glucagon concentrations rose coincident with insulin, which is in agreement with the findings of Bassett (1972)
, who reported similar results in sheep. However, it should be noted that glucagon is probably only responsible for a portion of the increase in plasma concentrations of insulin, as plasma concentrations of insulin were reduced before plasma concentrations of glucagon returned to baseline.
From the results of the preliminary study, the concentration of propionate in the hepatic portal vein was increased for approximately 4 h. It is surprising that the increased insulin and glucagon both returned to baseline before propionate leaving the rumen should have decreased. Therefore, we hypothesize that the initial increase in VFA signals insulin and glucagon secretion, but ultimately concentrations of blood glucose would become the predominant regulator of hormone secretion, as observed in the monogastric (Grodsky, 1972
). Rations that stimulate the production of propionate, increasing the portal vein propionate concentration, should alter insulin and glucagon secretion.
The results from the current work show that less progesterone is cleared by the hepatocyte in the presence of insulin or insulin and physiological concentrations of glucagon. Further, there was a dose-dependent relationship between insulin treatment and progesterone clearance with the exception that the greatest dosage of insulin (10 nM) was not different than 1.0 nM. This was expected, as the insulin receptor saturation in the mouse hepatocyte is approximately 1 nM (Valverde et al., 1997
); therefore, the addition of 10 nM insulin should not further decrease progesterone clearance beyond that observed with 1.0 nM insulin. Glucagon had no apparent effect on progesterone clearance; however, a supraphysiological dosage was able to negate the effects of insulin.
Consistent with De Jong (1982)
, the oral gavage of sodium propionate appeared to increase glucagon secretion. However, it should be noted that the increased glucagon concentration was similar to the physiological combination (1.0 nM insulin and 0.1 nM glucagon), which did not prevent insulin from decreasing progesterone clearance. Similarly, Selvaraju et al. (2002)
found that in cattle treated with insulin, circulating progesterone concentrations significantly increased. Diskin et al. (2003)
concluded that reductions in the concentration of insulin could reduce androgen and estradiol production or clearance. One could further extrapolate from this and earlier research with under- vs. overfed ewes, that there would be differences in insulin secretion.
Bidstrup et al. (2003)
confirmed that the in vitro metabolism of progesterone was a reduced nicotinamide adenine dinucleotide phosphate-dependent process and that the activity of CYP3A4 (6-hydroxysteroid dehydrogenase) and CYP2C19 (20-hydroxysteroid dehydrogenase) results in the formation of the major metabolites including 6-hydroxyprogesterone and 20-hydroxyprogesterone, respectively. Both CYP3A4 and CYP2C19 catabolize many different substrates, and competition among these substrates for binding sites or limitations in enzyme cofactors may alter the ability of the cytochrome P450 enzyme to metabolize a given substance (Bidstrup et al., 2003
).
Dietary restriction and negative energy balance reduce circulating insulin concentrations (Mackey et al., 1999
; Sinclair et al., 2000
). Further, Pell et al. (2000)
demonstrated that plasma insulin concentrations are reduced in lactating vs. nonlactating dairy cattle. To further emphasize lactational interactions, dairy cattle selected for milk yield exhibit reduced circulating insulin concentrations (Armstrong et al., 2003
). Selvaraju et al. (2002)
found that in cattle treated with insulin, circulating progesterone concentrations are significantly increased.
Sidhu and Omiecinski (1999)
cultured hepatocytes with either 0 or 1 nM insulin, and when exposed to Phenobarbital, the hepatocytes incubated in the absence of insulin showed a 1.5- to 2.0-fold increase in the messenger RNA for the cytochrome P450 enzymes. In addition, the time required to attain maximal gene expression was reduced in the noninsulin-treated cultures (Sidhu and Omiecinski, 1999
). Similarly, Sidhu et al. (2001)
demonstrated again that suppression of messenger RNA for cytochrome P450 enzymes was reduced by insulin but was not the result of an alteration in phosphatidylinositol 3-kinase. Finally, there was an 80 to 90% reduction in the messenger RNA for hepatocyte-derived cytochrome P450 enzymes when incubated in either 1.0 or 10 nM insulin (Woodcroft and Novak, 1999
). Those researchers also incubated cells with a supraphysiological dose of 100 nM glucagons, resulting in an approximately 7-fold increase in messenger RNA for the cytochrome P450 enzymes. However, when 1 nM insulin was added to the supraphysiological dose of glucagon, expression of the messenger RNA was similar to controls, indicating a mutual antagonistic signaling pathway where insulin and glucagon negate the effect of each other on the expression of messenger RNA for the cytochrome P450 enzymes (Woodcroft and Novak, 1999
). These results demonstrate that the cytochrome P450 enzymes, particularly CYP3A4 and CYP2C19, which contribute substantially to the hepatic progesterone catabolism, are influenced by insulin concentrations. These data are consistent with a paradigm in which elevated propionate increases insulin, which decreases progesterone clearance, thus increasing progesterone concentrations.
| IMPLICATIONS |
|---|
|
|
|---|
| Footnotes |
|---|
2 Current address: Department of Family and Consumer Science and Agriculture, Station 11, Eastern New Mexico University, Portales 88130. ![]()
3 Corresponding author: mwilso25{at}wvu.edu
Received for publication October 11, 2005. Accepted for publication December 30, 2005.
| LITERATURE CITED |
|---|
|
|
|---|
) of radioactive progesterone in plasma and red cells of bovine blood. J. Dairy Sci. 46:14021404.This article has been cited by other articles:
![]() |
C. O. Lemley, S. T. Butler, W. R. Butler, and M. E. Wilson Short Communication: Insulin Alters Hepatic Progesterone Catabolic Enzymes Cytochrome P450 2C and 3A in Dairy Cows J Dairy Sci, February 1, 2008; 91(2): 641 - 645. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |