Effects of Dietary Gossypol Consumption on ...

Effects of dietary gossypol consumption on metabolic homeostasis and reproductive endocrine function in beef heifers and cows M. L. Gray, L. W. Greene and G. L. Williams J ANIM SCI 1993, 71:3052-3059.

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://jas.fass.org/content/71/11/3052

www.asas.org

Downloaded from jas.fass.org by guest on July 12, 2011

Effects of Dietary Gossypol Consumption on Metabolic Homeostasis and Reproductive Endocrine Function in Beef Heifers and Cows1f2 M. L. Gray*, L. W. Greenet, and G. L. Williams*J *Animal Reproduction Laboratory, Texas Agricultural Experiment Station, The Texas A&M University System, Beeville 78102 and +Department of Animal Science, Texas A&M University, College Station 77843

ABSTRACT: Our objectives were to determine the effects of incremental increases in dietary gossypol on metabolic homeostasis and reproductive endocrine function in postpubertal beef heifers and the long-term effects of elevated dietary gossypol on various metabolic and reproductive endocrine characteristics in mature cows. In Exp. 1, heifers ( n = 6/group) were fed either 0, .5, 2.5, 5, 10, or 20 g.anima1-l.d-l of dietary free gossypol for 62 d. Erythrocyte membrane osmotic fragility was increased ( P < .0001) in both the 10- and 20-g groups. Slight alterations in plasma concentrations of sorbitol dehydrogenase and K+ were also detected in the latter group. Treatment did not affect ADG, body condition scores, or concentrations of progesterone during the estrous cycle; however, mean concentrations of LH were higher ( P< .001) in heifers

fed 20 g/d of gossypol than in heifers in all other groups. In Exp. 2, lactating cows ( n = 17) exhibiting regular estrous cycles were fed a control (no gossypol, n = 8) or high-gossypol (20 mg-kg BW-I-d-l free gossypol, n = 9 ) diet for 33 wk. Mean BW and body condition scores did not differ during the feeding period. Erythrocyte membrane fragility was greater ( P < .05) in the high-gossypol than in the control group. Magnitude of the preovulatory LH surge, luteal phase concentrations of progesterone, follicular fluid concentrations of estradiol and progesterone, in vitro granulosa cell estradiol production, and 60-d pregnancy rates were similar between groups. The amounts of gossypol fed in these experiments are not likely to affect reproductive performance adversely in beef heifers or cows.

Key Words: Gossypol, Toxicity, Beef Cows, Reproduction, Endocrine System J . h i m . Sci. 1993. 71:3052-3059

Introduction Whole cottonseed and cottonseed byproducts have been common supplemental feedstuffs for livestock for > 100 yr, especially in the southern and southwestern regions of the United States (Ensminger and Olentine, 1978). Moreover, our recent work indicates that whole cottonseed can be used as a practical source of supplemental dietary fat for modifying metabolism, modulating ovarian follicular recruitment and luteal

ITexas Agric. Exp. Sta. journal article no. 30753. ?Supported in part by the National Cottonseed Products Assoc. and the TAES-ERA program. We acknowledge the NIDDK and USDA for pituitary hormones and Jerry Reeves for anti-ovine LH. The expert technical assistance of C. Ball, J. Cutshaw, K. Hitchcock, E. Lugo, J. Perez, L. Perez, F. Sablan, R. Schnitz, M. Smith, and R. Spoon is gratefully acknowledged. 3T0 whom correspondence should be addressed. Received October 20, 1992. Accepted July 14, 1993.

activity, and enhancing reproductive potential in cattle (Williams, 1989a; Wehrman et al., 1991; Ryan et al., 1992). Whole cottonseed and other cotton products also contain gossypol, a yellow, polyphenolic binapthylaldehyde that is toxic to prefunctional ruminants and nonruminants (Kerr, 1989). Gossypol in excessive quantities may escape detoxification and produce toxicosis in functional ruminants as well (Kerr, 1989). Recent reports have documented histopathological changes in testes of bulls and rams (Arshami and Ruttle, 1988, 1989). As a result of these reports, gossypol as a potential feed toxin has received renewed attention, both regionally (Williams, 198913) and nationally (Jones et al., 1991). The objectives of the experiments reported herein were to quantify the effects of dose and duration of ingestion of dietary gossypol on several measures of growth, metabolic homeostasis, and reproductive endocrine function in young, sexually mature beef heifers and mature, estrual beef cows.

3052 Downloaded from jas.fass.org by guest on July 12, 2011

DIETARY GOSSYPOL EFFECTS IN BEEF COWS

3053

Table 1. Composition of diets 1% DM) fed to heifers receiving various amounts of dietary free gossypol [Experiment 1) Dietary group (grams of free gossypol)

0

Item

Whole cottonseed Cottonseed meal Milo Soybean meal Coastal bermudagrass haya Cottonseed hulls Chemical analysis Protein, % ME, Mcal/kgb

.5

2.5

5

-

-

32.0 68.0 -

2.5

.2 1.2 10.8 53.2

20.1 ,990

-

34.5

2.4 8.8 14.4 26.8 47.5

20.4 ,992

21.2 1.04

21.4 1.05

-

30.4 67.1

Materials and Methods Experiment 1 Animals and Diets. Sexually mature beef heifers weighing 383 k 3.8 kg and exhibiting regular estrous cycles were allotted randomly t o receive one of six complete-mixed diets containing differing amounts of dietary free gossypol ( n = 6 heifers/group). Diets were formulated to contain 0, .5, 2.5, 5, 10, or 20 g of free gossypol-animal-l&l (Table 1). The dose range was intended to encompass quantities of gossypol previously shown to produce some evidence of toxicity in bulls and dairy cows (Linsey et al., 1980; Arshami and Ruttle, 1988). Cottonseed meal was the primary source of gossypol; however, the 20-g group required both cottonseed meal and whole cottonseed to achieve the desired dose of gossypol. Therefore, the total lipid content in the 20-g group (2.9%) was unavoidably higher than that for all other groups (2.1%). Total and free gossypol contents of the whole cottonseed and cottonseed meal used in this study were estimated with standard AOAC (1990) procedures (Pope Testing Laboratories, Dallas, TX). All diets were formulated to be isoenergetic and isonitrogenous and to promote .27 kg of gain-animal-l.d-l. Heifers were fed individually and received the diets for 62 d. Sampling and Analysis. All heifers were weighed on d 0,30, and 60 of the feeding period and were assigned body condition scores using a 1 to 9 scale (Herd and Sprott, 1986). After 6 wk of feeding, heifers were observed for estrus twice daily, and all heifers not exhibiting estrus by d 5 were administered 25 mg of prostaglandin F2a ( PGF2,; Lutalyse, Upjohn, Kalamazoo, MI). Observations continued until all heifers were in estrus. Blood samples were collected at 10-min intervals for 30 min during the mid-luteal phase ( d 10) to determine mean basal LH concentrations. Single blood samples were collected, via jugular venipuncture, into heparinized tubes every other day to determine plasma progesterone concentrations dur-

-

10

20

2.6 45.6 .5

30.6 26.8 3.3

51.2

39.3

21.1 1.04

22.0 1.13

-

-

ing the estrous cycle. All blood samples were centrifuged for 30 min at 640 x g . Data for samples collected between d 0 to 8 and 9 to 16 were combined for treatment comparisons within early and late luteal phases, respectively, of the estrous cycle. In this experiment, plasma samples collected every other day were examined subjectively by visual appraisal for hemolysis to gain preliminary insight into the effects of diet on erythrocyte osmotic fragdity. Hemolysis was considered to have occurred if plasma exhibited a reddish discoloration as opposed to a transparent, straw-color. We recognize that this method is not quantitative and may be subject to significant technical variability. Therefore, a quantitative measure of osmotic fragility (Nelson, 1979) was used in Exp. 2. Concentrations of potassium ions [K+l and sorbitol dehydrogenase ( SDH) were determined in plasma samples collected on d 42 and 62. These two plasma constituents change as sequelae of gossypol toxicosis in various species of animals (Tam, 1985; Holmberg et al., 1988). After 62 d, dietary treatments ceased, and two fertile bulls were placed with the heifers for 30 d. Thirty-day pregnancy rates were assessed with transrectal ultrasonography (LS-1000; Tokyo-Keiki, Japan) using a 5-MHz probe. Concentrations of plasma progesterone and LH were determined with RIA using methods validated by Williams (1989a) and McVey and Williams (19891, respectively. The sensitivity of the progesterone assay averaged .02 ng/mL, with intra- ( n = 2 ) and interassay ( n = 8) CV averaging 5.7 and 2.2%, respectively. The LH assay had a sensitivity of .12 ng/ mL, with intra- and interassay CV of 1.4 and 9.2%, respectively. Concentrations of SDH were determined using enzymatic procedures (Sigma Chemical, St. Louis, MO). Concentrations of K+ were measured using atomic absorption spectrophotometry (ThermoJarrel Ash S-11 unit). Hemolysis data were combined by week for analyses. Body weights, body condition scores, hormone concentrations, and metabolic data as influenced by


3054

GRAY ET AL.

dose of gossypol were first analyzed with regression procedures (SAS, 1985). In the absence of significant linear, quadratic, or cubic effects of dose, analyses of variance for repeated measures were used. Sources of variation included dose of gossypol, cow(dose), time, and dose x time. Main effects of dose were tested using cow(dose) as the error term and the effects of time were tested using dose x time as the error term. When a significant F-value was obtained, pairwise comparisons of each gossypol dose were conducted using Scheffe’s conservative test. Frequency of hemolysis was analyzed with chi-square contingency tables. The microcomputer version of SAS (1985) was used for these analyses.

Table 2. Composition (W DM) of high-gossypol and control supplements fed to mature beef cows (Experiment 2) Diet Item Whole cottonseed Cottonseed meal Grain sorghum Soybean meal Vegetable oil Chemical analysis Protein, % Ether extract, % Free gossypol, g k g

GOSSVDO~ Control 10.9 78.7 10.4 -

41.0 2.96 2.5

-

.5 96.7 2.8 41.6 2.93 -

Experiment 2 Animals and Diets. Multiparous, nonpregnant, lactating cows were allotted randomly to one of two dietary groups: 1) control (no gossypol, n = 8 ) and 2 ) high gossypol (20 mgkg BW-l&l, n = 9). Diets were formulated t o meet the nutritional requirements of lactating beef cows (NRC, 1984) and were designed to be both isoenergetic and isonitrogenous. Cows were fed individually their control or high-gossypol supplement ( . 8 kg.100 kg BW-l.d-l) for 33 wk (Table 2). Each group was allowed continuous access to a single lot of Coastal bermudagrass hay. Free gossypol content of cottonseed meal and whole cottonseed was analyzed with AOAC (1990) procedures (Pope Testing Laboratory, Dallas, TX). Composite samples of Coastal bermudagrass hay were also analyzed chemically and yielded crude protein, crude fiber, and TDN estimates of 11.3, 31.2, and 59.5%, respectively.

Body Weights, Body Condition Scorzs, Sampling, and Analysis. Cows were weighed and scored for body condition according to the method described by Herd and Sprott (1986) at the beginning of the study and every 2 wk thereafter. Beginning at wk 4 of the trial, jugular blood samples were obtained once weekly t o estimate osmotic fragility of erythrocytes as described by Nelson (1979). Briefly, test solutions containing 0, .4, .45, -5, .55, .6, .65, .7, .75, .8, and .85% NaCl were prepared. One milliliter of each fresh, heparinized whole-blood sample was diluted in 3 mL of .85% saline. One hundred microliters of each diluted sample was added to 5 mL of each test solution in duplicate. Contents of the tubes were mixed gently by inversion, incubated at room temperature for 20 min, vortexed gently, and centrifuged at 650 x g for 5 min at room temperature. Absorbance was measured in 1mL of the supernatant at 540 nm in a Spectronic 1001 spectrophotometer (Milton Roy, Rochester, NY). Percent hemolysis of each sample in each test solution was calculated by dividing its absorbance by the absorbance obtained in the 0% saline solution (100% hemolysis) and multiplying by 100. After 90 d of feeding, estrous cycles were synchronized with a two-injection sequence of PGFzo,

(Lutalyse). A jugular vein was cannulated at the time of the second injection of PGFzo,. After the second injection, 15-mL blood samples were collected every 4 h until the onset of estrus. Observations for signs of estrus were conducted every 2 h for 30 min before each 4-h bleeding period with the aid of androgenized cows. Hourly blood sampling commenced at the first signs of estrus (i.e., mounting by other cows, attempts to mount) for LH analysis. Upon initiation of hourly sampling, cows were kept in pens containing fresh hay and water; they were sampled in an adjacent handling chute. Ovarian morphology was evaluated per rectum using linear array ultrasonography (7.5-MHz probe, LS-1000; Tokyo-Keiki, Japan) before the onset of hourly blood sampling to determine the size of the dominant follicle. Ovaries were evaluated every 6 h after the 12-h bleeding period to determine the time of ovulation. Plasma concentrations of progesterone were monitored in daily blood samples collected during the first 11 d of the estrous cycle. Ultrasonography and Ovariectomy. After 150 d, ultrasonography was used to follow follicular waves and development of a dominant follicle during the early luteal phase of an estrous cycle. A PGF2, injection ( 2 5 mg of Lutalyse) was given when the dominant follicle reached 10 mm in diameter. Ovaries with the dominant follicle were removed 18 h after PGF2, injection by supravaginal colpotomy (Williams and Forrest, 1989). Ovaries were immediately placed in ice-cold Medium-199 (Sigma Chemical) and transferred to the laboratory for further processing. After unilateral ovariectomy, cows remained on their respective diets during a 2-wk recovery period and a 60-d breeding period. Cows were maintained with a single fertile bull. Pregnancy rates were determined 30 d after the end of the breeding period using ultrasonography. Collection of Follicular Fluid and Culture o f Granulosa Cells. Sterile techniques were used for in vitro culture of granulosa cells, similar t o that previously reported (Wehrman et al., 1991; Ryan et al., 1992). The dominant follicle was dissected from


3055


the surrounding ovarian tissue, washed, and placed in a 15-mm x 100-mm Petri dish (Fisher Scientific, Fairlawn, NJ). Follicular fluid was aspirated and frozen at -20°C for later analysis of hormone concentrations. Granulosa cells were plated in 3 mL of Medium-199 containing 10% fetal bovine serum, 2% penicillinhtreptomycin (5,000 IU/5,000 pg/mL; Sigma M 25-hydroxycholesterol (Sigma Chemical), M andros-4-ene-3, 17-dione Chemical), and (Sigma Chemical). Cell viability was determined using .4% trypan blue (Sigma Chemical). Cells were incubated at 100,000 cells per well in multiwell tissue culture plates (24 welldplate; Fisher Scientific), and in a humid incubator (Precision, Chicago, IL) containing 95% air and 5% CO2. Gonadotropin treatments included 50 ng/mL of LH (USDA-bLH-b5; AFP-5500) and 50 ng/mL of FSH (USDA-FSH-BP 3). Each treatment was replicated four times for each follicle. After a 48-h incubation period, media were collected and stored at -20°C for later analyses of progesterone and estradiol-17P concentrations. Culture wells were filled with distilled water and frozen for subsequent protein analyses. Culture of Theca Interna. After the collection of granulosa cells, a section of the theca interna was peeled from the theca externa and scraped free of granulosa cells (Kotwica et al., 1982). This section of theca was weighed and dissected into nine equal pieces (approximately 1 mm2 each). Pieces were placed on dry, sterile sections of lens paper for support in multiwell tissue culture plates (24 wells/plate; Fisher Scientific). A volume of 950 1 L of Medium-199 with 2% penicillinhtreptomycin (5,000 IU/5,000 pg/ M 25-hydroxycholesterol, and 100 ng/mL of mL), LH (USDA-bLH-b5) was added t o each well. Cells from each follicle were incubated in triplicate in a humid incubator at 39°C for 48 h. The atmosphere was 95% air and 5% (302. Media were collected and frozen at -20°C for later analyses of androstenedione concentrations. Hormone Analyses. Concentrations of plasma progesterone and LH were determined with methods reported by Williams (1989a) and McVey and Williams (19891, respectively. The sensitivity of the progesterone assay averaged .02 ng/mL, with intraand interassay ( n = 3 ) CV averaging 6.2 and 5.0%, respectively. The LH assay had a sensitivity of .13 ng/ mL, with intra- and interassay ( n = 3) CV averaging 4.4 and 10.8%, respectively. Follicular fluid was analyzed for progesterone, estradiol- 170, and androstenedione according to methods described by Wehrman et al. (1991). All hormones in follicular fluid were assayed in single assays, and their sensitivities were 2.5 pg/mL for androstenedione, .02 ng/mL for progesterone, and .51 pg/mL for estradiol. Intraassay CV averaged 4.7, 5.9, and 4.7%, respectively. Granulosa cell culture medium was assayed directly for progesterone according to the method described by Wehrman et al. (1991). Sensitivity of the assay was

.02 ng/mL with intra- and interassay ( n = 4) CV averaging 7.8 and 3.7%, respectively. Estradiol-17P was measured directly in cell incubation medium after assay validation. Displacement curves for standards and dilutions of culture medium containing increasing quantities of estradiol were parallel. Sensitivity for estradiol assays averaged .51 pg/mL, and intra- and interassay ( n = 3) CV were 9.3 and 15.8%, respectively. Androstenedione was measured in theca tissue incubation medium using direct assay methods described by Wehrman et al. (1991) after validation in medium. Sensitivity of the androstenedione assay was 1.5 pg/mL, and intra- and interassay ( n = 2 ) CV averaged 7.7 and 17.2%, respectively. Protein analysis was performed by the Lowry method (Sigma Chemical). Data were analyzed using one-way analyses of variance or split-plot analyses for repeated measures where appropriate. For the latter, treatment served as the main plot and time as the subplot as previously described. Areas under the curve for preovulatory LH surges were determined by the trapezoid rule. The microcomputer version of SAS (1985) was used for all statistical analyses.

Results Experiment 1 Body weight increased ( P < .001) gradually over time for all groups (ADG = .17 k .10 kg/dj, but it was not affected by dose of gossypol. Average initial and final BW were 383.1 It 3.8 and 391 & 11.3 kg, respectively. Body condition scores did not change throughout the experiment and averaged 5.93 k .12. Evidence of increased ( P < .0001) erythrocyte fragility was subjectively observed in heifers consuming the two highest doses of gossypol after 8 wk of dietary treatment (Figure 1). Two weeks after treatment withdrawal, heifers in the highest dose group began t o recover, but frequency of hemolysis was still increased ( P < . O O O l j compared with frequency of hemolysis in heifers in all other groups (Figure 2). Table 3 summarizes the concentrations of SDH, K+, progesterone, and LH in plasma, as well as final pregnancy rates. Plasma [K+l was lower ( P< .05) in heifers consuming the highest dose of gossypol for 62 d compared to the 2.5, 5-, and 10-g groups, but it was not lower than that for the control group. Sorbitol dehydrogenase concentrations had a tendency ( P = .14) to be elevated after 42 d in the two groups receiving the highest doses of gossypol; however, SDH concentrations were not elevated on d 62 in any group. Progesterone concentrations were measured throughout an estrous cycle. Days 0 to 8 and d 9 to 16 were designated as early or late luteal phase, respectively. Progesterone concentrations did not differ between


3056

GRAY ET AL b b

E

7

8

b b

1

0

WEEKS AFTER TREATMENT INITIATION

Figure 1. Erythrocyte membrane fragility observed at 6, 7, 8, and 9 wk after initiation of diets containing various amounts of free gossypol (gos).Values are based on subjective evaluations of frequency of occurrence of hemolysis and represent differences based on chisquare analysis. Means with different superscripts (a,b) differ (P < .001).

groups in either the early or late luteal phases. However) mean LH concentrations were greater ( P < .001) during the mid-luteal phase in heifers fed the highest dose of dietary gossypol than in heifers in all other groups. Thirty-day pregnancy rates were not affected by gossypol treatment.

Experiment 2 One control cow died during the experiment due to causes unrelated to the study. A treated cow died due to hemorrhagic shock after ovariectomy. Therefore, data were not available throughout all phases of the study for these cows. Body weights and body condition scores were similar for cows consuming control and gossypolcontaining diets throughout the study, averaging

b

h

2 3 WEEKS AFTER TREATMENT WITHDRAWAL

Figure 2. Erythrocyte membrane fragility observed at 1, 2, and 3 wk after withdrawal of diets containing various amounts of free gossypol (gos). Values are based on subjective evaluations of frequency of occurrence of hemolysis and represent differences based on chisquare analysis. Means with different superscripts (a,b) differ (P < .001).

554.8 k 68 and 5.03 .78 kg, respectively. After 8 wk of consumption) the high-gossypol diet increased ( P < .05) erythrocyte membrane fragility compared with the control diets when cells were stressed in .5 and .55% NaCl (Figure 3 ) . Table 4 summarizes the effects of diet on several endocrine variables during the estrous cycle. Mean areas under the curve for the preovulatory surge of LH were determined in six control cows and in seven cows receiving gossypol that responded to PGF2, estrus synchronization. Similar patterns of LH release were seen for cows fed control and gossypol-containing diets. We were unable to detect differences in mean progesterone concentrations during the early- to midluteal phase. Follicular fluid obtained from preovulatory follicles did not differ in concentrations of androstenedione) estradiol) or progesterone between _+

Table 3. Mean plasma concentrations of sorbitol dehydrogenase (SDH), potassium, progesterone, and LH and pregnancy rates in heifers fed increasing amounts of dietary gossypol for 62 d (Experiment 1) Daily gossypol, g Variable Potassium, mgldL Day 42 Day 62

SDH, mIU Day 42 Day 62 Progesterone, ng1mL Early luteal phase Late luteal phase LH, ng/mL Pregnancy rate

0

.5

2.5

5

10

20

SE

25.3 20.3ab

22.9 20.gab

21.1 22.4a

22.5 23.1a

21.1 23.6a

23.6 18.7b

.7 1.1

5.0 5.1

5.6 5.0

5.0 3.8

5.3 4.2

7.6 3.5

8.0 5.3

1.1 .9

2.6 6.7

2.8 8.5

1.9 7.9

2.9 7.1

4.1 7.6

4.1 7.8

.5

.4c 516

.5c

.5c

516

516

a,bMeans within a row with different superscripts differ ( P < .05). C,dMeans within a row with different superscripts differ ( P < ,001). Downloaded from jas.fass.org by guest on July 12, 2011

516

.4c 616

.9d

616

.9

.1 -


.7b

.E5 .55 NmCI CONCENTRATION (X)

.45

Figure 3. Erythrocyte membrane osmotic fragility at 4, 8, and 12 wk in cows fed control or high-gossypol diets. Hemolysis values are expressed as a percentage of that obtained in 0% saline (100% hemolysis] at each test concentration shown. Beginning at wk 8, percentage of hemolysis differed (P < .05) between control and highgossypol (Gos) groups at NaCl concentrations of .55 and .50% (n = 8/group].

control or gossypol-fed groups. Estradiol production by granulosa cells also did not differ. Progesterone production by granulosa cells was higher ( P e .06) in cows fed gossypol-containing diets than in control cows. Thecal production of androstenedione also tended ( P = .11)to be greater in tissue from gossypolfed cows than in that from control cows. Sixty-day pregnancy rates of unilaterally ovariectomized cows consuming control and gossypol-containing diets for > 5 mo did not differ. Although the number of cows in the current experiment was small, and fertility endpoints were not a major objective, chronic consumption of large amounts of gossypol for 33 wk had no obvious effects on fertility. Final 60-d pregnancy rates were 87.5 (7/8) and 85.7% (6/7) for control and high-gossypol groups, respectively.

Discussion Feed intake, BW, and body condition scores were not affected by the amount of gossypol fed in Exp. 1 and 2. Although anorexia is an early sign of gossypol toxicity in nonruminants (Abou-Donia, 1976; Hahn et al., 1981) and preruminants (Hollon et al., 1958; Rogers et al., 1975), dairy cows consuming up to 30% whole cottonseed exhibited no signs of anorexia (Coppock et al., 1985). In the present study, heifers fed the highest dose of gossypol also received 30% of their DM intake as whole cottonseed, but there was no evidence of anorexia. Increased erythrocyte fragility was detected in heifers and cows consuming high doses of gossypol for 6 to 8 wk. Increased red blood cell fragility is a good indicator of gossypol intoxication in goats (Menges, 19911, bulls (Wyse et al., 19911, and dairy cows (Linsey et al., 1980). Various other measures of

3057

hematologic dysfunction have been studied. In the male bonnet monkey, Kalla et al. (1985) found no effects of gossypol on total erythrocyte or leukocyte count, hematocrit, or hemoglobin content. However, Linsey et al. (1980) found that cows consuming solvent-extracted cottonseed meal (24 mgkgl.d-l) had increased erythrocyte fragility after 7 wk of treatment, which is similar to our data from Exp. 1 and 2. Although all values were within the normal range (Kincaid, 1988), our findings in Exp. 1 indicate that gossypol can decrease plasma [K+l. Tam (1985) reported decreases in plasma [K+l in men after ingestion of gossypol. No differences were detected in urinary [K+l for these men before or during gossypol treatment. This observation supports the idea that gossypol causes a shift of K+ from extracellular fluid into cells rather than an increase in net excretion. Because gossypol toxicosis causes erythrocyte membrane fragility, an accurate assessment of CK+l may not be possible after centrifugation of whole blood, because [K+l is released into plasma due to hemolysis. The tendency for SDH concentrations t o be elevated at 6 wk in groups of heifers receiving the highest doses (10 and 20 g ) of gossypol, but not on d 62, is not readily explained. Accumulation of gossypol in the liver has been speculated to cause hepatic damage. Holmberg et al. (1988) reported elevated concentrations of SDH due to hepatic damage in calves consuming cottonseed meal. Both plasma K+ and SDH concentrations in the current experiment were within the normal ranges (Cornelius, 1989). Diets did not affect progesterone concentrations during the early, middle, and late luteal phases of the estrous cycle of heifers or during the early to midluteal phase of cows. Progesterone concentrations have been reported to be lower in rats treated orally with gossypol acetic acid (Lin et al., 1985; Yang and Wu, 1987). In contrast, Wyse et al. (1991) reported that progesterone concentrations in luteal tissue obtained from FSH-treated, superovulated heifers increased with increasing consumption of dietary gossypol. Follicular fluid steroid hormone concentrations and in vitro granulosa cell estradiol production were not affected by diet in the current study. However, progesterone production by granulosa cells was higher in cows fed gossypol-containing diets than in control cows. Female rats fed toxic amounts of gossypol had lower serum estradiol and progesterone (Lin et al., 1985), but Wu et al. (1981) reported higher serum and ovarian concentrations of estrone and estradiol in gossypol-treated female hamsters. Wyse et al. ( 199 1) reported that luteal tissue concentrations of progesterone increased in heifers as dietary free gossypol increased. No information concerning luteal or granulosa cell production of progesterone is available in nonruminant animals fed diets containing gossypol.


GRAY ET AL.

3058

Table 4. Mean areas under the curve (AUC] for the preovulatory LH surge, concentrations of progesterone during the early- to mid-luteal phase, preovulatory follicular fluid hormone concentrations, and in vitro production of hormonal steroids by granulosa cells and theca interna (Experiment 2) Group Control

Item

LH surge AUCa Early- to mid-luteal phase plasma progesterone, ng/mL Follicular fluid steroids Androstenedione, ng/mL Estradiol, ng/mL Progesterone, ng/mL Granulosa cell production in vitro Progesterone, ng/mg protein Estradiol-17& ngimg protein Theca interna production in vitro Androstenedione, pg/mg tissue

320

High gossypol 340

SE 16

2.5

2.9

.2

12.9 1,070.3 50

10.6 791.5 50.8

4.2 236 7.1

125 175

185 161

20 23

2,000

1,700

254

aDetermined by the trapezoid rule in arbitrary units.

Thecal production of androstenedione also tended to be greater in tissue from gossypol-fed cows. Reports of serum androgen concentrations in gossypol-treated animals are limited to the male. Testosterone concentrations were lower in male rats fed 30 mg of gossypol/ kg of BW for 30 d (Chang et al., 1982). However, others have reported that gossypol does not affect serum testosterone concentrations in male rats (Hoffer, 19851, bonnet monkeys (Kalla et al., 19851, or bulls (Chase et al., 1990). The increase in mean concentrations of LH observed during the mid-luteal phase of heifers receiving 20 g of gossypol daily is in contrast to all previously reported studies in females of both ruminant and nonruminant species (Chang et al., 1982; Frick and Danner, 1985; Hoffer 1985). We have no explanation for this observation, and none of the mean values could be considered physiologically abnormal (i.e., < 1 ng/mL). Wu et al. ( 1 9 8 1 ) reported increases in serum concentrations of FSH and decreases in pituitary FSH of female hamsters treated with gossypol. Male rats had decreased concentrations of LH after ingesting 30 mg of gossypovkg BW for 30 d (Chang et al., 1982). Finally, lower conception rates have been reported for rodents consuming gossypol acetic acid ( G u and Anderson, 1985; Lagerlof and Tone, 1985). Although numbers were small, extremely high doses of dietary free gossypol did not reduce pregnancy rates in heifers previously exposed to gossypol for 62 d. We interpret this to mean that, at these levels of consumption, gossypol has no effect on fertility. Similarly, 60-d pregnancy rates were not affected in mature cows fed large amounts of gossypol for 33 wk.

Implications Heifers consuming 2 10 g.anima1-l.d-' of dietary free gossypol for 1 to 2 mo exhibit clinical signs of

toxicity manifested as increased red blood cell fragility. However, no detrimental effects on growth rates, body condition scores, or reproduction should be expected. Similarly, consumption of large amounts ( 2 0 mgkg BW-l.d-l) of dietary free gossypol by mature beef cows results in increased red blood cell fragility, but body weights, body condition scores, and reproduction should not be altered. Amounts of dietary free gossypol producing evidence of erythrocyte fragility are generally in excess of those used in typical production schemes.

Literature Cited Abou-Donia, M. B. 1976. Physiological effects of metabolism of gossypol. Residue Rev. 61:125. AOAC. 1990. Official Methods of Analysis (15th Ed.). Association of Official Analytical Chemists, Arlington, VA. Arshami, J., and J . L. Ruttle. 1988. Effects of diets containing gossypol on spermatogenic tissue of young bulls. Theriogenology 30:507. Arshami, J., and J . L. Ruttle. 1989. Effects of diets containing gossypol on semen quality and testicular tissue in fine-wool rams. Proc. West. Sect. Am. SOC.Anim. Sci. 40:277. Chang, C.-C., Z. Gu, and Y. Tsong. 1982. Studies on gossypol. I. Toxicity, antifertility and endocrine analysis in male rats. Int. J. Fertil. 27:213. Chase, C. C., Jr., P. Bastidas, J. L. Ruttle, C. R. Long, and R. D. Randel. 1990. Serum concentrations of testosterone in Brahman bulls fed diets containing gossypol. J. Anim. Sci. 68(Suppl. 1):15 (Abstr.). Coppock, C. E., J. W. West, J. R. Moy, D. H. Nave, J. M. LaBore, K. G. Thompson, L. D. Rowe, Jr., and C. E. Gates. 1985. Effects of amount of whole cottonseed on intake, digestibility, and physiological responses in dairy cows. J. Dairy Sci. 68:2248. Cornelius, C. E. 1989. Liver function: VI. Serum Activities and other markers for detecting hepatic necrosis, cholestasis or carcinoma. In: J. J. Kaneko ( E d . ) Clinical Biochemistry of Domestic Animals. p 383. Academic Press, New York. Ensminger, M. E., and C. G. Olentine, J r . 1978. Feed and Nutrition (2nd Ed.). Ensminger Publishing Co., Clovis, CA.


DIETARY GOSSYPOL EFFECTS IN BEEF COWS Frick, J., and C. Danner. 1985. Effects of gossypol on the human testicular function: Evaluation of seminal and hormonal parameters. In: S. J. Segal ( E d . ) Gossypol: A Potential Contraceptive for Men. p 17. Plenum Press, New York. Gu, J., and N. 0. Anderson. 1985. Effects of gossypol on the estrous cycle and ovarian weight of the rat. Contraception 32:491. Hahn, D. W., C. Rusticus, A. Probst, R. Homm, and A. N. Johnson. 1981. Antifertility and endocrine activities of gossypol in rodents. Contraception 23:97. Herd, D. B., and L. R. Sprott. 1986. Body condition, Nutrition and Reproduction of Beef Cows. Texas Agric. Ext. Bull. B-1526. Hoffer, A. P. 1985. Ultrastructural, biochemical and endocrine studies on effects of gossypol and its isomeric derivatives on the male reproductive tract. In: S. J. Segal ( E d . ) Gossypol: A Potential Contraceptive for Men. p 17. Plenum Press, New York. Hollon, B. F., R. K. Waugh, G. H. Wise, and F. H. Smith. 1958. Cottonseed meal a s the primary protein supplement in concentrate feeds for young calves. J . Dairy Sci. 41:286. Holmberg, C. A., L. D. Weaver, W. M. Guterbock, J . Genes, and P. Montgomery. 1988. Pathological and toxological studies of calves fed a high concentration cottonseed meal diet. Vet. Pathol. 25:147. Jones, L. A,, D. H. Kinard, and J. S. Mills. 1991. Cattle research with gossypol-containing feeds. Natl. Cottonseed Prod. Assoc. Rep., Memphis, TN. Kalla, N. R., J . Foo, T. W. Kalpana, S. Hurkadli, and A. R. Sheth. 1985. Effects of gossypol on the fertility of the male bonnet monkey. In: S. J. Segal ( E d . ) Gossypol: A Potential Contraceptive for Men. p 33. Plenum Press, New York. Kerr, L. A. 1989. Gossypol toxicosis in cattle. Compend. Cont. Ed. Prac. Vet. 11:1139. Kincaid, R. 1988. Macro Elements for Ruminants. In: D. C. Church ( E d . ) The Ruminant Animal: Digestive Physiology and Nutrition. p 326. Prentice-Hall, Engelwood Cliffs, NJ. Kotwica, J., G. L. Williams, and M. J . Marchello. 1982. Countercurrent transfer of testosterone by the ovarian vascular pedicle of the cow: Evidence for a relationship to follicular steroidogenesis. Biol. Reprod. 27:778. Lagerlof, R. K., and J. N. Tone. 1985. The effects of gossypol acetic acid on the female reproduction. Drug Chem. Toxicol. 8(6):469. Lin, Y. C., T. Fukaya, Y. Rikihisa, and A. Walton. 1985. Gossypol in female fertility control: Ovum implantation and early pregnancy inhibited in rats. Life Sci. 37:39. Linsey, T. O., G. E. Hawkins, and L. D. Guthrie. 1980. Physiological responses of lactating cows to gossypol from cottonseed meal rations. J . Dairy Sci. 63:562.

3059

McVey, W. R., Jr., and G. L. Williams. 1989. Effects of temporary calf removal and osmotic pump delivery of gonadotropin-releasing hormone on synchronized estrus, conception to a timed artificial insemination and gonadotropin secretion in norgestomet-estradiol valerate-treated cattle. Theriogenology 32:969. Menges, W. L. 1991. Effects of feeding cottonseed on mohair growth and performance of angora goats. M.S. Thesis. Angelo State Univ., San Angelo, TX. Nelson, D. A. 1979. Erythrocytic disorders. In: J. B. Henry ( E d . ) Clinical Diagnosis and Management. p 117. W. B. Saunders, Philadelphia, PA. NRC. 1984. Nutrient Requirements of Beef Cattle (6th Ed.). National Academy Press, Washington, DC. Rogers, P.A.M., T. P. Henaghan, and B. Wheeler. 1975. Gossypol poisoning in young calves. Ir. Vet. J. 29:9. Ryan, D. P., R. A. Spoon, and G. L. Williams. 1992. Ovarian follicular characteristics, embryo recovery, and embryo viability in heifers fed high-fat diets and treated with follicle-stimulating hormone. J. Anim. Sci. 70:3505. SAS. 1985. SAS User’s Guide: Statistics (5th Ed.). SAS Inst. Inc., Cary, NC. Tam, S.-C. 1985. Acute hypokalemia effect of gossypol. Can. J. Physiol. Pharmacol. 642381. Wehrman, M. E., T. H. Welsh, Jr., and G. L. Williams. 1991. Dietinduced hyperlipidemia in cattle modifies the intrafollicular cholesterol environment, modulates ovarian follicle dynamics and hastens the onset of postpartum luteal activity. Biol. Reprod. 45:514. Williams, G. L. 1989a. Modulation of luteal activity in postpartum beef cows through changes in dietary lipid. J . Anim. Sci. 67:785. Williams, G. L. 1989b. Use and abuse of cottonseed and cottonseed products in cattle diets: A 1989 perspective. Texas Agric. Exp. Sta. Rep. No. 89-1, Beeville. Williams, G. L., and D. W. Forrest. 1989. A note on fertility of cyclic beef females following acute unilateral ovariectomy. h i m . Reprod. Sci. 18:155. Wu, Y. M., S. C. Chappel, and G. L. Flickinger. 1981. Effects of gossypol on the pituitary and ovarian endocrine function, ovulation and fertility in female hamsters. Contraception 29:259. Wyse, S. J., J . S. Velez, R. C. Stahringer, L. W. Greene, and R. D. Randel. 1991. Effect of diets containing free gossypol or erythrocyte fragility and packed cell volume in cattle. J. Anim. Sci. 69 (Suppl. 1):43 (Abstr.). Yang, Y. Q., and X. Y. Wu. 1987. Antifertility mechanisms of gossypol acetic acid in female rats. J. Reprod. Fertil. 80:425.


Citations

This article has been cited by 1 HighWire-hosted articles: http://jas.fass.org/content/71/11/3052#otherarticles
