The physiological effects of feeding warfarin poultry

The Physiological Effects of Feeding Warfarin to Poultry1,

J. R. VELTMANN, JR.,3 ERNEST ROSS,4 and STEVEN E. OLBRICH5

Departmentof AnimalSciences, University of Hawaii, 1800 East-West Road,Honolulu, Hawaii 96822

(Received for publication April 3, 1981)

ABSTRACT A 20 week study using layer and broiler strain chicks of both sexes was undertakento determine whether poultry were susceptible to warfarin-induced granulomatous endocardiallesions.

Birds were fed a corn-soybean meal basal diet with no added vitamin K, supplemented witheither 0, 25, 50, or 100 ppm of warfarin or vitamin K at .6 mg/kg of diet.

Broiler chicks showed a higher incidence of hemorrhages, more mortality, and longer prothrombin times than did the layer strain fed the same diets. Regardless of the breed, female chicks fedthe highest warfarin level had significantly longer prothrombin times than the male chicks. However,there were no sex differences associated with mortality or incidence of hemorrhages among birdsfed the experimental diets. Growth was most significantly reduced for chicks fed the highestwarfarin level and to a lesser degree for birds fed 50 ppm of warfarin. In contrast to the first 10weeks of the study, there was a sharp decline in mortality, incidence of hemorrhages, and prothrombin times during the last 10 weeks of the study.

Layer and broiler strains of chickens fed warfarin for 20 weeks showed no evidence of granulomatous endocardial lesions as was reported for swine (Oshiro and Brooks, 1975).(Key words: warfarin, vitamin K, heart lesions, layer, broiler chicks, prothrombin times)

INTRODUCTION

Blood coagulation in mammals requires fourvitamin K dependent blood clotting proteins(prothrombin, Factor VII, Factor IX, andFactor X) synthesized in the liver (Olson,1974). Impairment in the synthesis of theseproteins due to either a vitamin K deficiency ora vitamin K antagonist results in subcutaneousand intramuscular hemorrhages and hypo-prothrombinemia.

In association with a hemorrhagic conditionin swine fed a 63% sucrose diet, Brooks et al.(1971, 1972, 1973) found irregular shapedlesions in the heart and aorta. Although Gre-sham and Howard (1963) and Sanda (1968)reported similar pathological changes in the

'Journal Series No. 2586 of the Hawaii Agricultural Experiment Station.

2Submitted by the Senior author to the Graduate School in partial fulfillment of the requirementsfor the degree of Master of Science at the Universityof Hawaii.

3Present address: Department of Poultry Science,University of Georgia, Athens, GA 30602.

4Direct reprint requests to University of Hawaii.'Present address: P.O. Box 718, Waianae, HI

96792.

ot r

L_ynj'

1981 Poultry Science 60:2603-2611

aorta of other mammals fed sucrose diets,Brooks et al. (1973) prevented the heartlesions by supplementing the diet with avitamin K derivative, menadione dimethy-lpyrimidinol bisulfite (MSB). Subsequently,Oshiro and Brooks (1975) demonstrated thatgranulomatous lesions also developed in swinefed a practical diet containing a vitamin Kantagonist, 3-(acetonylbenzyl)-4-hydroxycou-marin (warfarin) at .5 mg/kg of diet. Supplementing this diet with MSB (1 mg/kg ofdiet) prevented the warfarin-induced heartlesions and hemorrhagic condition.

Warfarin has been used extensively since1952 as an anticoagulant rodenticide and inrecent years as a vitamin K antagonist both inclinical and animal research studies, yet themode of action of the drug is unknown. Moreover, there are incomplete data on the long-term effects of warfarin feeding in animalsother than rodents. Warfarin feeding studiesin the domestic fowl conducted during the1950's described only the clinical symptoms ofacute warfarin toxicity (Papworth 1958;McGirr 1953; Stableforth 1953). Moreover,these findings are difficult to interpret due tothe few animals used per treatment and theshort duration of each study. This paper reportsthe effects of long term feeding of various levels

2603

t

2604 VELTMANN, JR., ET AL.

of warfarin to poultry and the apparent absenceof warfarin-induced heart lesions in two breedsof chickens.

MATERIALS AND METHODS

Chicks. Single Comb White Leghorn andbroiler strain chicks were obtained from acommercial hatchery6 at 1 day of age andplaced in thermostatically-controlled starterbatteries with raised wire floors. At 40 days ofage all chicks were transferred to growerbatteries and at 68 days of age birds weremoved to wire floor developer pens measuring75 X 75 cm until the experiment was terminated at 152 days of age.

Dietary Treatments. For the first 12 days all

'Asagi Hatchery, Honolulu, HI 96817.'Menadione (2-Methyl-l,4-naphthoquinone) ob

tained from ICN Life Science Group, Cleveland,OH 44128.

"Dethmore is the registered trade name by J. C.Peneck Conpany, New York, NY for a commercialwarfarin product. Lot number 20-PRF-3 used forthis study was obtained from Hawaii Chemical Company, Ltd., Honolulu, HI 96814.

'WARF Institute, Incorporated, Madison, WI53701.

chicks received a modified vitamin K-deficientbasal diet (Table 1, Diet A), originally designedby Charles and Huston (1972) except that 2%of the corn was replaced with an equal amountof corn starch. The experiment started at 13days of age when 75 layer and 75 broiler chickswere randomly assigned to the followingdietary treatments: negative control (basaldiet), positive control (basal diet + .6 mgvitamin K7 per kg of diet), or three warfarindiets (basal + either 25, 50, or 100 mg warfarin/kg of diet). Warfarin doses were preparedby replacing either 25, 50, or 100% of the cornstarch in the basal diet with Dethmore, acommercial rodenticide, which contained .5%of warfarin in a corn starch base. A warfarinassay9 revealed that Dethmore was within 2%of the stated amount of warfarin. Because ofthe very high mortality in broilers fed 100 ppmof warfarin, the highest warfarin dose wasreduced to 75 ppm for both strains after 10weeks on treatment. The protein contentof the experimental diets was reduced from23% to 18% (Diet B), to 16% (Diet C), and to12% (Diet D) when chicks were 40, 70, and 88days of age, respectively (Table 1). All dietswere calculated to meet the National ResearchCouncil (1971) minimum nutrient requirementsof poultry except for vitamin K. Feed andwater were supplied ad libitum. Individual Hart

TABLE 1. Composition of the starter andgrower diets (g)

Diet

Ingredient A B C D

Yellow corn

Soybean meal (48.4% protein)56.40

35.60 &©67.20

24.80

72.20

19.80

77.30

14.70

Cottonseed oil1 3.00 3.00 3.00 3.00

Corn starch 2.00 2.00 2.00 2.00

Defluorinated rock phosphate 1.85 1.85 1.85 1.85

Ground limestone .45 .45 .45 .45

Iodized salt .38 .38 .38 .38

dl-methionine .19 .19 .19 .19

Micro-ingredient mix2 .13 .13 .13 .13

Calculated analysis:Crude protein, % 23.12 18.00 16.00 12.00

ME (kcal/kg) 2979 3107 3166 3226

Energy-to-protein ratio 128.9 172.6 197.9 268.8

Phosphorous, % .72 .68 .65 .63

Calcium, % 1.00 .97 .97 .95

'Wesson Oil Sales Company, Fullerton, California.

2Supplied the following per kg of diet: vitamin A, 7500 IU; vitamin D3, 2010 ICU; vitamin E, 2.2 IU;vitamin B,2, .01 mg; riboflavin, 4.4 mg; Ca-pantothenate, 8.8 mg; niacin, 39.8 mg; choline chloride, 375 mg;manganese, 60mg; iron, 16 mg; copper, 2.0 mg; zinc, 20mg; cobalt, .4mg; and folic acid, .2mg.

-

PHYSIOLOGICAL EFFECTS OF WARFARIN 2605

TABLE 2. Experimental design showing the number of chicks assigned to the respective dietary treatmentfor the determination ofprothrombin times and main effects

Dietary treatments

Vitamin

K

Warfarin (mg/kg)

0 25 50 100'

Prothrombin time

determinations2

Layer breed 15 15 15 15 15

Broiler breed 15 15 15 15 15

Main treatment effects3

Layer breed 604 60 60 60 60

Broiler breed 60 60 60 60 60

1Warfarin changed from 100 to 75 mg/kg after the first 10 weeks of treatment.

2These birds were only used to obtain blood for periodic determinations of prothrombin time.

3Body weight, feed consumption, mortality and post mortem data obtained only from these animals.

4Each treatment consisted of six groups of ten chicks each. Each replicate of ten birds consisted of fiverandomly selected male and five randomly selected female chicks.

water cups10 were installed in each pen of thebatteries to minimize bacterial production ofvitamin K.

Experimental Design. As shown in Table 2,15 of 75 birds per dietary treatment wereestablished as "bleeder" groups: at 12 days ofage and every 2 weeks thereafter for a full 20weeks, 4 of the 15 birds per treatment wererandomly selected and bled by cardiac puncture, and prothrombin times were determinedaccording to the method of Doerr et al. (1975).Duplicate prothrombin determinations usingthe BBL Fibrometer1 were performed on eachplasma sample. The remaining 60 birds pertreatment, consisting of six replicate groupswith 5 males and 5 females each, were used to

monitor growth and physiological effects.Initial group body weights and feed con

sumption were recorded at 12 days of age andagain at 19, 26, 33, 40, 68, 82, and 152 days ofage. Pertinent mortality data were recorded forall deaths. A thorough examination of the heartchambers and the vascular network leadingfrom the heart, visceral organs, and ectodermal

"H. W. Hart Manufacturing Company, Glendale,CA 91201.

'' Baltimore Biological Laboratories, Cockeysville,MD 21030.

12Taylor Pharmaceutical Co., Decatur, IL 62525.

tissue was conducted. Efforts were made to

examine birds immediately after death; however, if that was not possible, birds were refrigerated at 5 C until examined.

One-third of the surviving broiler and layerchicks from each dietary treatment wererandomly selected at 82 and 152 days of ageand killed with an euthanasia solution, T-61.12Each bird was carefully examined for heartlesions. Cervical dislocation was not used

because our preliminary studies disclosedtissue artifacts in heart tissue, possibly maskingthe presence of heart lesions.

Statistical analysis. The sources of variationin the 2x2x5 factorial used in the studywere sex, strain, and the five dietary treatments. Feed consumption, body weights, andprothrombin times were analyzed by analysis ofvariance (Snedecor and Cochran, 1973) withmultiple range tests (Duncan, 1955; Kramer,1956) used to locate significant differencesbetween means. Interactions between the main

effects were tested using Tukey's test (Cic-chetti, 1972). Mortality data were analyzedusing a modified chi-square (Cochran and Cox,1957).

RESULTS

Post-Mortem Examinations. No gross ormicroscopic lesions were found in the heart

2606 VELTMANN, JR., ET AL.

vascular system or visceral organs in the birdsthat died during the course of the experimentor in those birds that were killed at 82 and 152

days of age.Incidence of Hemorrhages. Incidence and

severity of hemorrhages were directly proportional to the warfarin level. However,compared to the layer breed, broiler chickstended to have a higher incidence of hemorrhages across the three warfarin treatments.Within 1 week of the start of feeding, massivesubcutaneous hemorrhages were observed in thewings of both breeds fed the highest warfarinlevel. Upon post-mortem examination of theseanimals, intramuscular hemorrhages were located in the major and minor pectoralis, bicepsfemoris, and sartorius muscles; some intraperitoneal bleeding was also observed. Similarpathological lesions were found in both thelayer and broiler birds fed 50 ppm of warfarin.However, the number and severity of thelesions were markedly less than was observedin the high warfarin group. The frequency ofhemorrhages was less for broiler and layerchicks fed 25 ppm of warfarin and usually seenonly upon post-mortem examination. Nohemorrhages were found in birds fed either thepositive or negative control diets. During thelast 10 weeks of the study, hemorrhages inwarfarin-fed birds occurred with less frequencyand severity.

Mortality Data. Significantly more deaths

(P<.05) occurred in the broiler breed than inthe layer chicks (Table 3). There were moredeaths among birds fed 100 ppm of warfarincompared with birds fed lowerwarfarin levels orthe control diets, but the death rate for Leghornsfed either 25 or 50 ppm of warfarin or broilersfed 25 ppm of warfarin was not significantlydifferent from the respective negative controls.The death rate for broiler chicks fed 50 ppm ofwarfarin was significantly higher than that forthe lowest warfarin dose or the control diets.There was no difference in the death ratebetween male and female birds within the samedietary treatment for either the layer or broilerbreed. A higher proportion of the warfarin-fedbirds died during the first 10 weeks of thestudy than during the last 10 weeks.

Growth Data. A significant reduction in feedconsumption and body weight gain occurred inboth strains fed the highest warfarin level fornearly all the weeks sampled (Table 4). Likewise, significant declines in feed consumptionand body weight gain were found for bothstrains fed the intermediate warfarin level, butonly during the first 3 weeks (Leghorns) andfirst 7 weeks (broilers). Growth of the birds fed25 ppm of warfarin was not significantlydifferent from either the negative or positivecontrol birds.

Warfarin Consumption. Warfarin intake(ug/g of body weight) of the birds fed thewarfarin diets generally followed the dietary

TABLE 3. Mortality (numberdead) by 10-weekperiodsfor layer and broiler chicks

Dietary treatments

Vitamin

K

Warfarin (mg/kg)

0 25 50 100'

Layer breed0-10 weeks

10-20 weeks

Total no. dead

Total mortality, %

0

2

2

3.3^

2

2

4

6. 7ab

3

1

4

6. 7ab

6

3

9

15.0b

22

12

34

56.7C

Broiler breed

0-10 weeks

10-20 weeksTotal no. dead

Total mortality, %

4

5

9

15.0a

3

8

11

18 3ab

11

8

19

31. 7b

37

8

45

75.0C

602

60

100d

a,b,c,dValues on thesame !ine witn different superscripts differ significantly (P<.05).

1Dietchanged from 100 to 75 mg/kg of warfarin after 10 weeks of treatment.2No broiler chicks fed 100 mg/kg survived past the 10th week of treatment.

6

o-O

13Sa

•2 SS. £

nSt <ij

5 5

? •"

° i1 *

hft

yl

*

Efl

<

PHYSIOLOGICAL EFFECTS OF WARFARIN

<J « U U rjo !-«. rrt Is* Ov 00o hO^VIO,

tN IN f« <N

X) T3 u y u .Os ro 00 »n O0CN00NO

1-H •*

X) y jD iD j2 rt rt00 *h r*» C** I** m© rom O wnh 00s0 00<N ro ro •<♦• so mo\ •*•

^h fNOO Th

X! O X) X> _o rt rt

i-HSOsomOOuiOsO

rt -Q rt rt « rt rt

O <0 N 0\ M hO ">X O Os O fcft m^- roN •+ -t vO ^ l**^" ro

fN CslQv SO

rt -D rt rt rt rt rts0s000»AfN00O»A>ftrf.tO\HQ0m>oH(Srs|fNO\NrnO\

rt rt rt rt rt rt rtN N fi m ^o MO O00 N (s) N m t^oo m(N rj* m so Tf t>.ro •*•

fsl fNOs so

rt rt rt rt rt rt rtO\rniftfnift00"nsommOOOrOrtnNNmCsONNO

> M

rt rt rt rt rt cd rtINrtOOt^rtOOOt"'*'CO ^ h ffiffi Ovrt ^fN •#• ia \o m 00"> O

fN CslOs t>.

rt rt rt rt rt rt rtOONGNOOOOCOOroso^O^ONfomoH<N(s)NO\OONO

rt X) U X) X) X) rt(N 00 m O ^O fNO rort so t> rt in Osi-h <N^h »-* r-i fN fs. O-* Os

^HTJ- SO

XI <j X> X» X) rt rtsO^-0\sOrtrotnm

t< rt £ if it al elos rj so (N rsi ooo osi-H co os »a oo Hfr» t*-rt rt rt (N 00 fNOO SO

S Xi rt rt rt rt rt

vONO\0\HtNO\^ro ro in m

.. rt rtN sO ^- ON ^ OO soh ON (s] W ^- sOsO w»rt r-i csi rsi os rs?tN. t^

rt^ r-*

rt rt rt rt

rtCOWH^'lONVOtsCsONHHTj-ON

ro ro so in

rt ti rt rt rt rt rtSO SO Tj- OS ITl rtO ^hrM a fN <n ^ oom oort rt r>j rsi o\ (Sin in

rt-rj- r>.

rt rt rt rt rt rt rtrs|sosot-»r^mfNinsOOOOsOsOrslroO

ro ro so so

§rt

> rt

rt rt rt rt rt rt rtOs so Th fN Tf OsO ^h On N w fM irtO mrt rt <S Cs» Os (SOs 00

r-nt r>.

rt rt rt rt rt rt rtfslOfN^sOOOOfNsOOOOsOsOSfsiOs^"

fsj ro in m

si ,s

w a00

c —

3 •&4-* .2ra. V

E3ft

GO(J

•t)

a; T3

o•*- O

on "«

ert 6

^

<N 4 l«6 o

*> o

(N ro ^

fN ro

o O u. S2,2

o

V

>- I

.1 J2

i I<*- cO o

nn to

fi 2oo

E3

T3

60C

T3

I |8 2a s

2607

2608 VELTMANN, JR., ET AL.

levels (Table 5). Chicks that received theintermediate and highest warfarin levels consumed almost two and four times as much

warfarin per unit of body weight, respectively,than the birds fed the lowest warfarin level.The layer strain consumed a slightly higheramount of warfarin per unit of body weightthan the broiler strain birds over the twentyweek study, yet for both strains the rate ofwarfarin consumption per unit of body weightdeclined as body size increased.

Prothrombin Time. Mean biweekly prothrombin times for the Leghorn and broilerchickens are presented in Figures 1 and 2,respectively; samples collected during the 15thweek of treatment were lost because of a

refrigeration failure. Birds fed the three warfarin diets had significantly higher prothrombintimes than chicks given either the positive orthe negative control diets, regardless of strain.However, compared to the layer strain birds fedthe same warfarin diets, mean prothrombintimes for the broiler chicks were significantlyhigher at each warfarin level. Mean prothrombin times for birds fed the negative and positivecontrol diets were not significantly differentfrom each other.

The highest prothrombin times for nearly all

TABLE 5. Calculated mean weekly Warfarin intake(mg/g of body weight)

Weeks

on

trial

Warfarin (mg/kg)

25 50 100'

Layer breed0-1 16.3 33.0 68.8

1-2 19.9 38.3 74.3

2-3 15.9 29.4 58.8

3-4 14.7 34.4 57.1

4-7 10.6 19.8 42.3

7-10 9.9 19.1 38.1

10-20 7.0 14.6 21.8

Broiler breed

0-1 18.6 42.4 70.0

1-2 16.6 30.4 53.1

2-3 14.5 27.0 61.5

3-4 13.3 24.1 44.7

4-7 9.5 17.1 9.6

7-10 8.1 17.72

10-20 6.0 12.3

'Diet changed from 100 to 75 mg/kg after tenweeks on treatment.

2Insufficient data due to early high mortality.

birds fed warfarin diets were observed 1 week

after treatment was initiated; however, broilerchicks fed 100 ppm of warfarin recorded theirlongest prothrombin times after the 3rd weekof treatment. Prothrombin times for layer andbroiler chicks fed either 25 or 50 ppm ofwarfarin declined gradually until the 7th weekof treatment at which time they remainedrelatively constant for the next 13 weeks.Birds fed the highest warfarin level succumbedeither to the dietary treatment or cardiacpuncture prior to the 7th week for the broilersand the 15th week for the layer breed.

There were no significant sex differences inmean prothrombin times among chickens fedeither the control diets or the lowest and

intermediate warfarin levels; however, at thehighest warfarin level mean prothrombin timesfor female chickens (54 sec) were significantlyhigher than those for male chickens (43.7 sec).

DISCUSSION

The fact that vitamin K prevented thedevelopment of granulomatous endocardiallesions in swine made vitamin K deficient byeither a high sucrose diet or by feeding thevitamin K antagonist, warfarin, strongly suggested that the vitamin had a more universalmetabolic function in animals than just inblood clotting.

so

45

- 40

E~ 30c

| 25o

i Me

* 15

IO

1 3 5 7 9 11 13 15 17 19 IO

W..k

Figure 1. Biweekly mean prothrombin times forlayer strain chickens fed different levels of warfarinor vitamin K during the 20 week experimental period.

PHYSIOLOGICAL EFFECTS OF WARFARIN 2609

Stenflo et al. (1974), Nelsestuen etal. (1974), and Magnusson et al. 1974)established that the vitamin K-dependentmodification of bovine blood clotting precursors involved the carboxylation of specificglutamyl residues to form 7-carboxyglutamylresidues in the completed blood clottingproteins. The 7-carboxyglutamic acid residueswere shown to act as functional calcium bind

ing sites for the normal calcium phospholi-pid-dependent activation of blood coagulation(Esmon et al, 1975). Since then, numerous7-carboxyglutamic acid-containing proteins,some of which are vitamin K-dependent, havebeen isolated in animal tissues predominantlynoted for calcium homeostasis. A vitamin

K-dependent 7-glutamyl carboxylase activitywas found in chick bone (Hauschka and Reid,1978) and chick chorioallantoic membrane(Tuan, 1979). In other species 7-carboxyglutamic acid has been detected in bovine bone

(Price et al, 1976), urine (Fernlund, 1976),human kidney (Hauschka et al, 1976), renalcalculi (Lian et al, 1977), and sites of ectopiccalcification, including hardened artheroscle-rotic plagues (Lian et al, 1976). Interestingly

111

103

95

87

79

71

V

\ "E 47

Eo 39

X

S 31c

23

15

7 9 II 13 15 17 19 20

W..k

Figure 2. Biweekly mean prothrombin timesfor broiler strain chickens fed different levels of

warfarin or vitamin K during 20 week experimentalperiods.

enough, Brooks et al. (1973) described theheart lesions in swine as the degeneration ofheart elastic tissue followed by calcification.Whether7-carboxyglutamicacid-containing proteins are present at the site of warfarin-inducedendocardial lesions in swine has not been

reported.The reason for the apparent absence of heart

lesions in poultry fed 50, 100, or 200 timesmore dietary warfarin than swine and for 10 to12 weeks longer is not clear. However, amongdifferent animal species, vitamin K-dependentreactions respond differently to warfarinadministration. Esmon et al. (1975) reportedthat rats treated with warfarin secreted in

significant amounts of abnormal prothrombinin the plasma whereas the endoplasmic reticulum of liver cells accumulated abnormal

prothrombin. In warfarin-treated chicks abnormal prothrombin has been detected in bothliver microsomes and blood plasma (Olson,1974; Carlisle et al, 1975), whereas the distribution of abnormal prothrombin in porcinecellular components has not been reported.Thus, the chick, like the cow and man, secretean abnormal prothrombin into the plasmarather than accumulating it in the liver (Stenfloand Ganrot, 1972; Morrissey et al, 1972).Whether the abnormal prothrombin from thechick plasma retains affinity for calciumbinding is not certain. Olson (1974) reportedthat because chick abnormal prothrombin wasbarium adsorbable, it retained some affinity forcalcium binding. However, Carlisle et al (1975)determined that only about 30% of abnormalprothrombin from adult chickens given warfarin was barium sulfate adsorbable. In addition, the adsorbable abnormal prothrombinwas detected by Echis carinatus venom but notby a two-stage prothrombin assay, suggestingthat the adsorbable abnormal prothrombincould not be activated in vivo. Garvey andOlson (1978) were unable to duplicate thisphenomenon with growing chicks but showedlow maximum values for abnormal prothrombin with viper venom as was expected in atwo-stage assay. The reduction of hemorrhages,prothrombin times (Figs. 1 and 2), and mortality (Table 3) over time in warfarin-fed chicksand absence of heart lesions in chicken necropsies may be related to the physiological activation of abnormal prothrombin in the chick.During this same period of time warfarin intakeper unit of body weight (Table 5) declined witheach successive week. Thus, the pharmaco-

2610 VELTMANN, JR., ET AL.

logical effect of the drug was probably reduced,which may have contributed to the diminutionof vitamin K deficiency symptoms. Becausethere were no significant differences betweenthe mean prothrombin times for birds fed thenegative and positive control diets (Figs. 1 and2), there was apparently sufficient vitamin K inthe basal diet to prevent hemorrhages andhypoprothrombinemia. Therefore, an endogenous source of vitamin K may have minimized the physiological response of Leghornand broiler chicks to dietary warfarin. Inaddition, the establishment of a microflorapopulation in the gut capable of synthesizingvitamin K may have significantly affected thevitamin K status of the warfarin-fed birdsand the presence of warfarin-induced heartlesions. Almquist and Stokstad (1936) demonstrated that hemorrhages in vitamin K-deficient chicks were prevented when chickexcreta was added to the diet. Even though thewarfarin-fed chicks were reared on raised wirefloors in this experiment, as the birds grewolder, excreta would often become trappedin the wire mesh. Coprophagy would then notonly establish and perpetuate bacterial synthesis of vitamin K but could also provide thechicks with a source of vitamin K for immediate use. It is conceivable that both exogenous and endogenous sources of vitamin Kmay have modified the pharmacological effectsof dietary warfarin in poultry.

Another significant piece of physiologicalinformation uncovered in this experiment wasthe differential response to warfarin feedingbetween the layer and broiler breeds. Althoughthere are genetic differences between the twobreeds with respect to growth rate, feed consumption, and body weight gains, the differences in prothrombin times, mortality,and incidence of hemorrhages cannot beexplained merely on the basis of increasedintake of the drug by the broiler chicks, sincewarfarin consumption calculated on the basis ofbody weight did not vary significantly betweenbreeds (Table 5). The differential response towarfarin feeding by the layerand broiler chicksmay be due to different rates of absorption andexcretion of the drug, different rates of warfarin uptake by the plasma, or the degree ofvitamin K inhibition as well as differences incalcium binding induced by warfarin feeding.

ACKNOWLEDGMENTS

The senior author gratefully acknowledges

the generous donation of T-61 by NationalLaboratories Corporation, Somerville, NJ, andto Jean Murai for her secretarial expertise in thepreparation of this manuscript. This work wassupported in part by HEW Grant R01HL15840(principal investigator, Coy C. Brooks).

REFERENCES

Almquist, H. J., and E.L.R. Stokstad, 1936. Factorsinfluencing the incidence of dietary hemorrhagicdisease in chicks. J. Nutr. 12:329-335.

Brooks, C. C, A. Y. Miyahara, and D. W. Huck, 1971.Effect of slaughter weight and certain dietaryadditives on sucrose induced heart lesions in thepig. Proc. West. Sec. Amer. Soc. Anim. Sci.22:233-237.

Brooks, C. C, A. Y. Miyahara, D. W. Huck, and S. M.Ishizaki, 1972. Relationship of sugar-inducedlesions in the heart of the pig to live weight,serum cholesterol and diet. J. Anim. Sci. 35:31-37.

Brooks, C. C, R. M. Nakamura, and A. Y. Miyahara,1973. Effect of menadione and other factors onsugar-induced heart lesions and hemorrhagicsyndrome in the pig. J. Anim. Sci. 37:1344-1350.

Carlisle, T. L., D. V. Shah, R. Schlegel, and J. W.Suttie, 1975. Plasma abnormal prothrombin andmicrosomal prothrombin precursor in variousspecies. Proc. Soc. Exp. Biol. 148:140-144.

Charles, O. W., and T. M. Huston, 1972. The biologicalactivity of vitamin K materials following storageand pelleting. Poultry Sci. 51:1421-1427.

Cicchetti, D. V., 1972. Extension of multiple-rangetests to interaction tables in the analysis ofvariance: a rapid approximate solution. Psychol.Bull. 77:405-408.

Cochran, W. G., and G. M. Cox, 1957. Experimentaldesigns. 2nd ed. John Wiley and Sons, Inc., NewYork, NY.

Doerr, J. A., R. D. Wyatt, and P. B. Hamilton, 1975.Investigation and standardization of prothrombintimes in chickens. Poultry Sci. 54:969-980.

Duncan, D. B., 1955. Multiple range and multiple Ftests. Biometrics 11:1—42.

Esmon, C. T., J. W. Suttie, and C. M. Jackson, 1975.The functional significance of vitamin K action.Differences in phospholipid binding betweennormal and abnormal prothrombin. J. Biol.Chem. 250:4095-4099.

Fernlund, P., 1976. 7-carboxyglutamic acid in humanurine. Clin. Chem. Acta 72:147-155.

Garvey, W. T., and R. E. Olson, 1978. In vitro vitaminK dependent conversion of precursor to prothrombin in chick liver. J. Nutr. 108:1078-1086.

Gresham, G., and A. N. Howard, 1963. Comparativehistopathology of the atherosclerotic lesion. J.Atheroscler. Res. 3:161-177.

Hauschka, P. V., P. A. Friedman, H. P. Traverso, andP. M. Gallop, 1976. Vitamin K-dependent 7-carboxyglutamic acid formation by kidneymicrosomes in vitro. Biochem. Biophys. Res.Commun. 71:1207-1213.

Hauschka, P. V., and M. L. Reid, 1978. Vitamin K

C

PHYSIOLOGICAL EFFECTS OF WARFARIN 2611

dependence of a calcium-binding protein containing 7-carboxyglutamic acid in chicken bone.J. Biol. Chem. 253:9063-9068.

Kramer, C. Y., 1956. Extension of multiple range teststo group means with unequal number of replications. Biometrics 12:307-310.

Lian. J. B., E. L. Prien, M. J. Glimcher, and P. M.Gallop, 1977. The presence of protein-bound7-carboxyglutamic acid in calcium-containingrenal calculi. J. Clin. Invest. 59:1151-1157.

Lian, J. B., M. Skinner, M. J. Glimcher, and P. Gallop,1976. The presence of 7-carboxyglutamic acid inthe proteins associated with ectopic calcification.Biochem. Biophys. Res. Comm. 73:349-355.

Magnusson, S., L. Slottrup-Jensen, T. F. Petersen, H.R. Morris, and A. Dell, 1974. Primary structureof the vitamin K-dependent part of prothrombin.FEBS Lett. 44:189-193.

McGirr, J. L., 1953. Poisoning of livestock by newerrodenticides, insecticides and weed killers. Proc.15th Int. Vet. Congr. Stockholm 1:479-484.

Morrissey, J., R. K. Kipfer, and R. E. Olson, 1972.Non-existence of a circulating preprothrombin inrats. Fed. Proc. 31:218.

National Research Council, 1971. Nutrient requirements of poultry. 6th rev. ed. Nat. Acad. Sci.,Washington, DC.

Nelsestuen, G. L., T. H. Zytkovicz, and J. B. Howard,1974. The mode of action of vitamin K: identification of 7-carboxyglutamic acid as a component of prothrombin. J. Biol. Chem. 249:6347-6350.

Olson, R. E., 1974. New concepts relating to the modeof Vitamin K. Vit. Horm. 32:483-511.

Oshiro, D. K., and C. C. Brooks, 1975. Effect ofCoumadin and antibiotics on heart lesions. J.Anim. Sci. 41:324, (Abstr.)

Papworth, D. S., 1958. A review of the dangers ofwarfarin poisoning to animals other than rodents.Royal Soc. Health 78:52-60.

Price, P. S., A. S. Otsuka, J. W. Poser, J. Kristaponis,and N. Raman, 1976. Characterization of a7-carboxyglutamic acid-containing protein frombone. Proc. Nat. Acad. Sci. 73:1447-1451.

Sanda, I., 1968. Effect of dietary sucrose on vascularlesions in the rabbit. Japanese Circ. J. 32:1613—1620.

Snedecor, G. W., and W. G. Cochran, 1973. Statisticalmethods. 6th ed. The Iowa State UniversityPress, Ames, IA.

Stableforth, A. W., 1953. Recent advances at Wey-bridge. Vet. Rec. 65:709.

Stenflo, J., P. Fernlund, W. Egan, and P. Roepstroff,1974. Vitamin K-dependent modifications ofglutamic acid residues in prothrombin. Proc. Nat.Acad. Sci. 71:2730-2733.

Stenflo, J., and P.-O. Ganrot, 1972. Vitamin K andthe biosynthesis of prothrombin identificationand purification of a dicumarol-induced prothrombin. J. Biol. Chem. 247:8160-8166.

Tuan, R. S., 1979. Vitamin K-dependent 7-glutamylcarboxylase activity in the chick embryonicchorioallantoic membrane. J. Biol. Chem. 254:1356-1364.