TERATOLOGY 43:201-215 (1991)

Amelioration by Leucovorin of Methotrexate Developmental Toxicity in Rabbits JOHN M. DESESSO AND GERALD C . GOERINGER Toxic and Hazardous Materials Assessment and Control, Civil Systems Division, MITRE Corporation, McLean, Virginia 22102 (J.M.D.); Department of Anatomy and Cell Biology, Georgetown University Medical School, Washington, D.C. 20007 (J.M.D., G.C.G.)

ABSTRACT Methotrexate (MTX) is lethal or teratogenic to embryos of all species tested. New Zealand white rabbit embryos are relatively resistant to the embryolethal effects of MTX. However, when pregnant does were injected iv with 19.2 mg MTXikg on gestational day 12, virtually all surviving fetuses exhibited multiple malformations of the head, limbs, and trunk. MTX is a structural analogue of folic acid that competitively inhibits dihydrofolate reductase, thereby preventing formation of folinic acid and essentially stopping one carbon metabolism. One carbon metabolism is important in the synthesis of methionine, histidine, glycine, and purine bases that are required for the de novo synthesis of DNA. Presumably these metabolic effects of MTX relate directly to its mechanism of developmental toxicity. An ameliorative treatment has been tested utilizing iv injection of pregnant rabbits with leucovorin (LV), a close structural analogue of folinic acid (the product of the inhibited enzyme), a t various times after MTX exposure. When LV was injected a t times up to 24 hours after MTX, fewer malformed fetuses resulted and the incidence of specific malformations was reduced. When given at times up to 20 hours after MTX administration, LV virtually eliminated the grossly apparent effects of MTX at term. In the forelimb bud, MTX increased the extracellular space surrounding limb bud mesenchymal cells within 8-10 hours; this process continued through 16 hours and remained unabated by 24 hours. Mesenchymal cell nuclei became hyperchromatic and pyknotic during this time period. By 24 hours, a moderate amount of cellular debris was observed in the mesenchymal compartment of limb buds from approximately one-third of the embryos examined. Endothelial cell nuclei of the limb bud vasculature did not exhibit the histopathological alterations observed in the mesenchymal cells. Limb buds from embryos injected with LV at times up to 6 hours after MTX were histologically normal. When LV treatment was delayed until 16 or 20 hours after MTX, mesenchymal nuclei regained normal appearance within 2 hours of treatment; further, the abnormally large intercellular space began to decrease during the next 4 hours. Cellular debris was not a prominent feature of limb buds from LV-treated embryos examined a t any time. Embryos from rabbits injected with LV at 24 hours after MTX exhibited either typical MTX-induced lesions or a sequence of reparative events similar t o those described for the 16 and 20 hour LV-treated embryos. These results demonstrate that LV, administered up to 24 hours after MTX, is able to ameliorate the developmentally toxic effects of MTX seen in rabbits a t term. Histological examination revealed that LV is able to reverse the early sequelae of MTX exposure in limb bud mesenchymal tissue.

'

Folic (pteroylglutamic) acid is the parent substance in -tKe folic acid group vitaacid is neither the mins* In the most Common nor the most important con0 1991 WILEY-LISS, INC.

Received July 11, 1990; accepted October 16, 1990, Address reprint requests to Dr. John M. DeSesso, MITRE Corporation, 7525 Colshire Drive, McLean, VA 22102.

202

J.M. DESESSO AND G.C. GOERINGER

gener; after absorption, folic acid must be reduced to its active form, tetrahydrofolate (folinic acid), in order to participate in cellular metabolism. Folinic acid is an essential component in cellular anabolic processes involving one carbon (methylene) transport, e.g., the synthesis and metabolism of methionine, glycine, and histidine as well as de novo purine biosynthesis. Deficiency of folinic acid, whether induced by inadequate dietary intake of its precursors or by disruption of its formation in cells, is termed "folate deficiency." Folate deficiency due to inadequate dietary intake has been associated with developmental toxicity in animals (reviewed by Kalter and Warkany, '59; Giroud and Tuchmann-Duplessis, '62) and with adverse outcomes of human pregnancy (Milunsky et al., '89). Consistent with both animal dietary deficiency studies and epidemiological findings, abnormal growth of cultured rat embryos in serum from dietinduced folic acid-deficient rats has been reported (Miller et al., '89). Methotrexate (MTX; amethopterin; 4amino-N1O-methylpteroylglutamic acid) is a chemotherapeutic agent that is used in the management of acute lymphocytic leukemia, several neoplasms of the female reproductive tract, carcinoma of the breast, recalcitrant psoriasis, and rheumatoid arthritis (Calabresi and Parks, '80). MTX has been reported as a potent developmental toxicant in animals (Berry, '71; Skalko and Gold, '74; Khera, '76; Wilson, '71; Darab et al., '87) and a known human teratogen (Thiersch, '52; Milunsky et al., '68; Warkany, '71; Schardein, '85). In rabbits, treatment of pregnant females with MTX causes a spectrum of severe craniofacial and limb reduction malformations at term (Jordan et al., '77; DeSesso and Jordan, '77). MTX is a structural analogue of folic acid (structures in Fig. 1)that binds tightly, but reversibly, to dihydrofolate reductase, the enzyme responsible for reduction of folic acid to folinic acid (Goldman, '74). Inhibition of dihydrofolate reductase also causes a depletion of folinic acid and a virtual stoppage of one carbon metabolism. Interruption of one carbon metabolism, regardless of the cause, selectively affects dividing cells. In severe cases, folate deficiency results ultimately in death of the affected cells. MTX not only inhibits dihydrofolate reductase, but also inhibits thymidylate synthetase, which converts deoxyuridylate to

thymidylate, a rate-limiting step in the formation of DNA (Calabresi and Parks, '80). Presumably, these effects contribute to the mechanism underlying the developmental toxicity of MTX. The histological effects in rabbit limb buds after maternal treatment with MTX have been described previously (DeSesso and Jordan, '77; DeSesso, '81a). Briefly, MTX caused enlargement of intercellular spaces among mesenchymal cells beginning by 6-8 hours after treatment. By 16 hours, the mesenchymal cells appeared shrunken and angular with hyperchromatic, pyknotic nuclei. Cellular debris was not a prominent finding at any of the times studied. Leucovorin (LV; 5-formyl-5,6,7,8-tetrahydropteroylglutamic acid) is a structural analogue of folinic acid (structures in Fig. 1) that can act as a surrogate source of methylene groups. Treatment with LV should, therefore, counter the antifolate effects of MTX and allow one carbon metabolism to recommence. The experimental design is depicted graphically in Figure 2. Clinically, LV has been used to mitigate the effects of MTX on normal tissues in cases of high-dose MTX therapy for head and neck cancer and leukemia (see review by Bertino, '77), although the establishment of improved therapeutic efficacy for this regimen in the treatment of cancer remains under investigation (Calabresi and Parks, '80). The purpose of the present investigation was to determine whether LV could ameliorate the developmentally toxic effects of MTX seen at term in rabbits and, if so, to determine 1)the interval of time that could elapse between injections of MTX and LV without reducing the amelioration, and 2) the effect that LV exerts on the histological sequelae of MTX treatment within treated embryos. MATERIALS AND METHODS

A total of 139 virgin female New Zealand white rabbits was used in this study. They were maintained in a climate controlled animal facility at 22 3°C with a 12-hour light-dark cycle. Under the observation of an investigator, each female was mated randomly to two virile bucks of the same strain. The time of mating was designated as hour 0, day 0 of gestation. Experiments were begun on the morning of gestational day 12 (i.e., 288-292 hours post coitum) when pregnant females were assigned randomly to one

*

LV REVERSES MTX DEVELOPMENTAL TOXICITY

203

,COOH \

N H

/CH\

CH2 -CHz -COOH

Folic Acid

Methotrexate (MTX)

H

I

COOH CH2-CH2 -COOH

H I

Tetrahydrofolate (Folinic Acid)

NH2 COOH CH2-CH2 -COOH

Leucovorin (LV) Fig. 1. Structural formulae for folic acid, its derivatives, and structural analogues discussed in this paper. The stippled circles highlight moieties that differ between the natural folates and their administered structural analogues.

of the three control or 11 experimental venous injections of 19.2 mg MTWkg followed at intervals of 30 minutes, 1 , 2 , 3 , 4 , 5 , treatment groups. Control treatment groups included posi- 6, 8, 16, 20, or 24 hours by intravenous intive controls, which received intravenous jection of 75 mg LV/kg. Injection volumes injections of 19.2 mg MTWkg (Methotrex- were 2.5-3.0 ml for MTX, 7.0-7.5 ml for ate-sodium, Lederle); saline controls, which LV, and 7.5 ml for saline. Fifty-five rabbits were allowed to conreceived intravenous injections of sterile physiological saline; and LV controls, which tinue pregnancy until gestational day 29, received 75 mg LV/kg (Leucovorin, Led- when they were anesthetized and sacrificed erle). Experimental groups received intra- by an air embolus. The abdominal wall was

J.M. DESESSO AND G.C. GOERINGER

204

Summary of Proposed Mechanism for Action of Leucovorin

MTX

-1 Leucovorin

Methionine

2

Fig. 2. Schematic representation of normal folk acid metabolism with its contribution to the one carbon pool, the site of action for MTX, and the way that LV is proposed to alleviate the MTX effects.

opened and the positions of all fetuses and resorption sites were recorded graphically. Each fetus was dissected free of the placenta and fetal membranes, weighed, and examined grossly for externally visible signs of malformations. The fetuses were then prepared for examination of osseous structures by the alizarin red S staining method (Staples and Schnell, ’68). Statistical evaluations were performed on both an individual fetus and a litter basis. Statistical methods included chi-square analysis for comparison of frequency data (e.g., incidences of specific anomalies) from three or more treatment groups as prepared in the Statistical Analysis System (SAS; Ray et al., ’82);Fisher’s exact test for analysis of frequency data between a treatment group and the control group as prepared in the Toxicology Risk Assessment Program (Crump et al., ’87); and one-way analysis of variance followed by Duncan’s multiple range test as prepared in SAS was used to analyze fetal weights and numbers of digits. Analyses of mean litter percentages were performed using Student’s “t”test. The level of significance was chosen to be P 5 .05. In a separate experiment, pregnant rabbits (gestational day 12, hour 0) were untreated or injected with either MTX, LV, or saline; experimental rabbits were injected with MTX followed by LV a t either 30 minutes, 4,6, 16, 20, or 24 hours after MTX. In order t o investigate the time course of events after each type of treatment, em-

bryos were harvested a t various times after injection ranging from 1to 36 hours. Due to the complexity of this portion of the experimental design, the numbers of treated females, their schedules of injection, and the times of sacrifice are summarized in Table 1. Each treatment group is listed on the left of the table; the schedule of injection and sacrifice is depicted in the body of the table. A time line beginning with the first maternal injection at hour 0 on gestational day 12 and ending at 36 hours after injection appears at the top of the table. The time and identity of each injection are designated by capital letters; the number of litters examined at each time point is given by the numeral under the appropriate hour. A minimum of three embryos from each litter was examined at each time point. Limb buds from a total of 265 embryos taken from 84 females were analyzed histologically. In all cases the embryos were fixed in Bouin’s fluid, embedded in paraplast, serially sectioned at 5 pm, and stained with haematoxylin and eosin for light microscopic analysis. RESULTS

Teratology The gestational and developmental toxicity data for rabbits injected with saline, LV, MTX, or MTX plus LV are presented in Table 2. The data for the saline-treated litters are consistent with the numbers of implantations and the low background incidence of

205

LV REVERSES MTX DEVELOPMENTAL TOXICITY TABLE 1 . Treatment schedule and litter distribution for New Zealand white rabbits prepared for histological examination'

Maternal Total No. treatment (day 12) litters None 11 Saline 9 LV 3 MTX 17 MTX + 30 m LV 9 MTX + 4 h LV 1

No. litters harvested at indicated hours after maternal treatment 6 8 10 12 16 17 18 20 21 22 24 25 26 28 30 32 36 3 3 2 2 2 3 2 L 1 1 1 M 2 2 3 1 4 3 2 ML 3 3 3 M L 1

0 3 s

?S = time of injection with saline; L h = hours.

=

4

time of injection with leucovorin; M

=

time of injection with methotrexate; m

=

minutes;

TABLE 2 . Gestational and embryotoxicity data for New Zealand white rabbits whose mothers were injected intravenously with methotrexate, andlor leucovorin, or saline on gestational day 12* ImplantaMalformed Number tions Resorptions Viable fetuses Mean fetal weight3 fetuses Treatment (g) f SD No? (%)5 MPL litters No. MPL' No. (%)' MPL No. (%)' MPL Saline 7 64 9.1 3 (5) 0.4 61 (95) 8.7 37.3 f 7.Fd 1 (2) 0.1 Leucovorin 7 52 7.4 7 (13) 1.0 45 (86) 6.4 35.4 f 5.Sd,' 1 (2) 0.1 MTX 7 63 9.0 7 (11) 1.0 56 (89) 8.0 50, (89) 7.1 28.6 -+ 5.2g MTX + 30 m LV 8 65 8.1 (5) 0.4 62 (95) 7.8 39.6 f 5.7' 3 8 (13) 1.0 MTX + 1-8 h LV 7 63 9.0 5 (8) 0.7 58 (92) 8.3 35.7 f 5.1d+ 7b (12) 1.0 MTX + 16 h LV 7 66 9.4 6 (9) 0.9 60 (91) 8.6 31.8 f 4.gf Sb (13) 1.1 35.1 f 4.4dr MTX + 20 h LV 6 47 7.8 l Z b (31) 2.0 8 (17) 1.3 39 (83) 6.5 MTX + 24 h LV 6 49 8.2 5 (10) 0.8 44 (90) 7.3 33.6 f 3.7ezf 34" (77) 5.7 ~~~~~~~~

~

~

~

*"P 5 0.05. bP c 0.0001. "gHomogeneous subsets determined by Duncan's Multiple Range Test (P 5 0.01). 'MPL = Mean per litter. 'Percentage based on implantations. 3Analyzed by Analysis of Variance. 4Analyzed by Fisher's Exact Test verus MTX. 5Percentage based on viable fetuses.

resorptions and malformed fetuses among New Zealand white rabbits that have been published both by ourselves and others (DeSesso and Jordan, '77; DeSesso, '81b; DeSesso and Goeringer, '90; Palmer, '68, '72, '77; Sawin and Crary, '64; Cozens, '65). The single malformed fetus in the salinetreated group exhibited flexed wrists. That LV exerted no adverse effects on the outcome of pregnancy is indicated by a low incidence of resorptions, no significant change in mean fetal body weight compared to saline controls, and only a single malformed fetus. The malformed fetus was a cyclops (exhibiting fused eyeballs in a single orbit) with a proboscis. The cyclopia was deemed not to have been induced by LV treatment because 1)cyclopia did not occur in any of the other 41 litters treated with LV and 2) experimental production of this anomaly requires exposure to teratogenic agents at developmental stages earlier than

gestational day 12 rabbit embryos (Wolfe, '36, '58; Giroud et al., '63; ROUX,'64; Otis and Brent, '54; Shepard, '83). MTX treatment caused pronounced developmental toxicity evidenced by a significant depression in mean fetal weight compared to saline and LV controls and by the occurrence of gross malformations in nearly all surviving fetuses. The malformed fetuses generally exhibited a spectrum of malformations including severe craniofacial anomalies and limb reduction deformities (Fig. 3). The incidences of the most prominent malformations are presented in Table 3. Other malformations that were observed at lower incidences included syndactyly, pedunculated thumb, cleft lip, protruding tongue, and short tail. Alizarin preparations revealed osseous defects consistent with the aforementioned malformations. Prominent osseous defects included reductions in the size and number of the long

206

J.M. DESESSO AND G.C. GOEFUNGER

Fig. 3. A litter of gestational day 29 rabbit fetuses whose mother had been injected intravenously on gestational day 12 with MTX a t 19.2 mglkg. All fetuses exhibit reduction deformities of all limbs, ranging from the absence of one or two digits (fetuses B,C) to absence

of all digits (fetus D). The fetuses exhibited craniofacial defects including micrognathia (fetuses C-E), hydrocephalus (fetus E), and cleft palate (all fetuses, not illustrated).

bones of the distal limbs of ectrodactylous and hemimelic fetuses; delayed ossification and increased size of fontanelles in hydrocephalic fetuses; and delayed development and retarded ossification of the caudal vertebrae in fetuses with short tails. Table 2 shows that injection of MTXtreated rabbits with LV a t times up to 20 hours after MTX resulted in a great reduction in MTX-induced developmental toxicity as evidenced by both increased mean fetal weights and decreased incidences of malformed fetuses. Fetuses from rabbits treated with LV at 24 hours after MTX also demonstrated decreased developmental toxicity in that they were significantly heavier than their MTX-only counterparts. In addition, as demonstrated in Table 3, the incidence of all major malformations was significantly decreased when compared to the MTX-only group. LV treatment not only decreased the incidence of ectrodactylous fetuses, but also increased the mean number of digits on both fore- and hindpaws. These

data clearly demonstrate that LV treatment is capable of ameliorating MTX-induced developmental toxicity up to 24 hours after MTX injection, although LV treatments at times up to 20 hours after MTX were more efficacious than those at 24 hours postMTX.

Histology Because a large percentage of MTXtreated fetuses exhibited reduction deformities of the limb, the microscopic analyses concentrated on changes in the limb buds, which undergo major organogenesis on gestational days 12-13. The forelimb buds of embryos of this age (Fig. 4) were composed of a core of undifferentiated mesenchyme ensheathed within a bilaminar jacket of ectoderm. At the distal tip of the limb bud, the ectoderm was organized as a stratified epithelial placque, the apical ectodermal ridge (AER). The cytoplasm of the mesenchyme cells stained slightly eosinophilic; the nuclei were round to oval and contained one to sev-

Treatment

6

MTX + 24hLV

Hydrocephalus

34

(77)

79

Od

Od

Od

(13)

Oa

0'

Ob

Ob

oc

12

(2)

(2)

7d

ld

2d

75

17d (39)

36

(%)

Id (2)

Od

Od

Od

od

2'

oc

OC

oc

OC

43

(%)

Od

Od

Od

od

Od

Hemimelia

oc

OC

oc

oc

0"

54

(%)

Od

Od

od

od

od

Fore

Ectrodactyly

0'

OC

oc

oc

0'

36

(5)

(5)

(2)

20d (45)

2d

3d

Id

Sd (10)

48 (86)

4 9

4d

4d

2d

lod

80

(%)

lld (25)

Od

od

od

Od

9.9 f 0.4'

Od

28'

8.6

f

5

bP 5

'P

X

l X g 7.5

9.9 2 0.4'

9.9 ? 0.3'

Od

Od

Hind

l.lg

100.

f

8.0'

6.0'

8.0'

8.0'

5 . 0 ? 2.V 3 . 9 f 2.7e

Fore

Mean No. digits -r SE4

Od 9.8 f 0.6'

72

ML%3

Hind 42 (75)

ML@ No.' (%) ML9b3 No.'

Hind 22 (39)

ML%3 No.'

Fore 33 (59)

ML%3 No.'

Cleft palate 23 (41)

ML%3 No.'

lod (25) 27

5d (8)

Id

Id

43 (77)

(%)

Micrognathia

of methotrexate-induced birth defects in fetal rabbits*

0.02. 0.01. dP 5 0.0001. "'Womogeneous subsets determined by Duncan's Multiple Range Test (P5 0.01). 'Analyzed by Fisher's Exact Test verus MTX. 'ML % = Mean of litter percent incidence. The litter percent incidence was calculated as the No. of affected fetuses in a given litter + total No. of fetuses in that litter 3Analyzed by Student's t Test versus MTX. 4Analyzed by Analysis of Variance. 51neludes five cases with protruding tongue as the only defect.

44

28'

(31)

lZb

39

6

*=Pr 0.05.

Ild

(13)

gd

60

Od

lod

7

75,d (12)

Od

7

13d

58

85

7

(89)

Sd (13)

50

62

56

8

7

MTX + 30mLV MTX + 1-8 hLV MTX + 16 h LV MTX + 20 h LV

MTX

Malformed

fetuses lit. ble fe. ters tuses No.' (%) ML%'13 No.' (%) ML%3 No.'

No, Via-

TABLE 3. Amelioration by leucouorin

208

J.M. DESESSO AND G.C. GOERINGER

Fig. 4. Control. Limb bud from a gestational day 12, hour 12 negative control embryo. The limb bud is enveloped in a jacket of ectoderm that is bilaminar except at the thickened apical ectodermal ridge. The mesenchyme is uniformly compact with cells that contain round to oval nuclei. The intercellular space within the mesenchymal compartment is also uniform. Fig. 5. LV + 16 h. Limb bud from an embryo whose mother had been injected 16 hours earlier with LV a t 75 mgkg. The microscopic anatomy is identical to the control limb bud. Compare to Figure 4.

era1 nucleoli. Numerous small blood vessels permeated the mesenchyme. A relatively large, conspicuous blood vessel, the mar-

Fig. 6. MTX + 6 h LV + 10 h. Limb bud from an embryo whose mother had been injected with MTX at hour 0, followed 6 hours later with an injection of LV. The embryo was harvested 10 hours after LV (i.e., 16 hours post-MTX). Note the normal appearance of the mesenchymal cells and the uniformity of the intercellular space. Compare to the control configuration (Fig. 4) and to the MTX + 16 h condition in the absence of an ameliorative dose of LV (Fig. 8).

ginal venous sinus, was located subjacent to the AER. Cell proliferation was common in both mesenchyme and ectoderm as evi-

LV REVERSES MTX DEVELOPMENTAL TOXICITY

denced by the presence of numerous mitotic figures. Injection of pregnant rabbits with LV caused no histological changes from the control configuration a t any post-treatment stage examined (Fig. 5). The histological changes that appeared in limb buds of MTX-treated embryos were consistent with those of our previous reports (DeSesso and Jordan, '77; DeSesso, '81a). Briefly, the earliest change, observed at 810 hours after maternal injection, was an increase in the intercellular space within the mesenchymal compartment (Fig. 7). By 16 hours, the mesenchymal nuclei were angular and darkly stained, and the intercellular space in the mesenchymal compartment had enlarged further (Fig. 8). These histological changes remained prominent features of the limb buds during the next 8 hours (Figs. 9, 10). The nuclei of the endothelial cells lining the vasculature of the limb buds did not change in the dramatic way seen in the mesenchymal cells (Figs. 11, 12). A small amount of cellular debris, an index of cell death, was observed in the mesenchyme of some limb buds at 24 hours (Fig. 10); cell debris was not common prior to that time. For descriptive purposes, the MTX-LV treatment regimes are divided into three categories based on the time elapsed between MTX and LV injections. First, early LV treatment includes embryos from rabbits treated at 30 min to 6 hours after MTX; second, middle LV treatment includes embryos from does injected at 16-20 hours after MTX; and third, late LV treatment is made up of embryos from rabbits treated with LV at 24 hours after MTX. Histological examination of limb buds from MTX-early-LV embryos revealed none of the typical MTX-associated pathology. Limb buds from this group were indistinguishable from control limb buds a t all times studied (Fig. 6). In contrast, MTXmiddle-LV embryos had developed pronounced MTX-induced histological changes, including pyknotic mesenchymal nuclei and increased intercellular space, prior to the time of LV injection. Within 1-2 hours after LV injection, the nuclei of the mesenchymal cells had regained their normal round to ovoid contours, although the intercellular spaces among the mesenchymal cells remained enlarged (Fig. 13). Over the course of the next 4 hours, however, the previously prominent intercellular spaces decreased to

209

the control condition (Fig. 14). Cellular debris was not a prominent feature in the limb buds of MTX-middle-LV embryos. Limb buds from MTX-late-LV-treated embryos appeared to exhibit one of two responses to LV injection. One response involved partial amelioration of the MTX-induced pathology; the other resulted in little change in the limb buds. Embryos that exhibited the ameliorative sequelae possessed limb buds that reacted to LV treatment in a manner similar to that described for the MTX-middle-LV treated embryos. During the first few hours after LV injection, many of the mesenchymal cells regained their normal appearance and the enlarged intercellular spaces decreased (Fig. 15). In contrast to the MTX-middle-LV-treated embryos, a small amount of cell debris was observed in the mesenchyme. MTX-late-LV embryos that did not respond to LV treatment possessed limb bud mesenchymal cells that continued to exhibit angular, basophilic nuclei, greatly enlarged intercellular spaces, and a small-to-moderate amount of cellular debris scattered throughout the mesenchyme (Fig. 16). Approximately onethird of the embryos from the MTX-late-LV groups did not appear to respond to the LV treatment. DISCUSSION

Our results clearly demonstrate that LV treatment, when administered up to 24 hours after MTX, ameliorates the developmental toxicity of MTX in fetal rabbits. Evidence for this amelioration is 1) a highly significant decrease in the incidence of fetuses with malformations, and 2) dramatic reductions in the incidences of the major specific anomalies induced by MTX. Further, not only were the specific anomalies reduced, but also the typical spectrum of MTX-related defects was not observed in LV-treated fetuses. When specific malformations were observed, they usually exhibited diminished severity as documented by the presence of more digits per fetus in the MTX-LV fetuses than in the MTX fetuses. In addition t o the preceding measures of diminished developmental toxicity, all MTXLV groups showed increased mean fetal weights when compared with the MTX-only group. The histological sequelae caused by MTX in limb buds are similar to those published previously by this laboratory (DeSesso and

210

J.M. DESESSO AND G.C. GOERINGER

Figs. 7-10.

LV REVERSES MTX DEVELOPMENTAL TOXICITY

Jordan, '77; DeSesso, '81a) with the exception of the small amount of cellular debris observed at 24 hours after treatment. Although cell debris was not observed in our previous work, embryos from a 24 hour treatment group were not examined in those studies. Furthermore, the amount of cell debris seen in the present study was so small that it could have been cleared from the limb buds prior to the later observations in the previous work. Recent histological observations of MTX-treated chicken limb buds reported by Brewton and MacCabe ('90) agree with our findings of increased intercellular spaces (at 24 hours post treatment in their system) and small amounts of cellular debris appearing thereafter. LV appears to be more effective in reducing the incidence of hemimelia than that of ectrodactyly for late LV-treated embryos (Table 3) because of the delay between MTX and LV injections. Since the limbs develop in a proximo-distal sequence, the forelimb buds of gestational day 12 rabbit embryos, have already laid down the anlagen for the shoulder girdle and arm, and the forearm is undergoing rapid development. This makes the forearm a primary available target when MTX is injected on day 12. Early changes caused by MTX include pyknosis of mesenchymal cell nuclei and accumulation Fig. 7. MTX + 8 h. Limb bud from a n embryo whose mother had been injected 8 hours earlier with methotrexate at 19.2mg/kg. While the ectodermal and mesenchymal cells appear morphologically normal, the intercellular space in the mesenchyme appears enlarged. Compare to Figure 4. Fig. 8. MTX + 16 h. Limb bud from a n embryo whose mother had been injected 16 hours earlier with MTX. The mesenchymal cells appear basophilic and shrunken; the nuclei exhibit angular contours. The intercellular space has enlarged. Note the absence of cellular debris particles in the mesenchymal compartment. Fig. 9. MTX + 20 h. Limb bud from a n embryo whose mother had been injected 20 hours earlier with MTX. The intercellular space has continued to enlarge; the mesenchymal cell nuclei retain their angular appearance. Fig. 10. MTX + 24 h. Limb bud from an embryo whose mother had been injected 24 hours earlier with MTX. The intercellular space in the mesenchymal compartment remains enlarged relative t o the control condition (see Fig. 4); the mesenchymal cells still appear darkly stained and shrunken. Note the small amount of cellular debris (examples at arrows) located within the mesenchymal compartment.

211

of extracellular matrix material. These changes are readily reversed by LV treatment, which allows normal development to recommence after a period of suspended development that varies according to the length of time between MTX and LV injections. However, it should be recalled that during limb development, a finite interval exists during which the extracellular matrix remains undifferentiated (i.e., rich in hyaluronic acid, but poor in sulfated glycosaminoglycans) and the mesenchymal cells are permitted to proliferate rapidly (Toole and Gross, '71; Toole and Trelstad, '71; Toole, '72; Toole et al., '72). As shown in the rat (Sugrue and DeSesso, '821, this finite interval ends on gestational day 15.5 (about 24 hours after the first appearance of digital rays) in the forelimb bud of both control and teratogen-treated embryos. The comparable time in rabbit embryos is day 15 (Edwards, '68). All embryonic repair or compensation must be completed prior to this deadline. This means that the time available for compensation after MTX treatment will be dictated by the elapsed time prior to providing the MTX-treated embryo the LV treatment. For embryos receiving early and middle LV treatments, sufficient time is available for normal limb development to occur. We propose, however, that for embryos receiving late LV treatment, there is not enough time available for normal limb bud development to be completed. Although the proximodistal ontogeny is re-initiated and forearm anlagen are laid down, the deadline is reached before all anlagen for the digital structures of the paws are laid down resulting in the observed reduction in the numbers of digits. The reason that late LV treatment appears to be more effective at reducing hindlimb ectrodactyly compared to the forelimb (Table 3) is also related to both the schedule of administration of agents and the developmental stage of the embryo. The cranial regions of the embryo (including the forelimb buds) are developmentally advanced with respect to the caudal regions (including the hindlimb buds). In rabbit embryos, the forelimb buds appear on gestational day 10 and exhibit digital rays by day 14, whereas the hindlimb buds appear on gestational day 11 and develop digital rays on day 15 (Edwards, '68). This means that relative to the forelimb buds, the hindlimb buds have approximately 24 additional hours in which to

212

J.M. DESESSO AND G.C. GOERINGER

Figs. 11-14

LV REVERSES MTX DEVELOPMENTAL TOXICITY

compensate for damage caused by MTX treatment on gestational day 12. We propose that this longer period for compensation is responsible for the lower incidence of hindpaw anomalies relative to forepaw anomalies in the late LV-treated fetuses. The ability of LV treatment, given as late as 16-24 hours after MTX, to ameliorate the effects seen in fetuses at term appears to be related to the fact that MTX does not cause extensive cell death among mesenchymal cells during that time period. Rather, the embryonic lesions involve reversible alterations in cellular morphology of mesenchymal cells and distortions in the extracellular matrix of target tissues. LV treatment in rabbit embryos is effective at relatively late times after MTX because, in the presence of a tetrahydrofolate analogue, the embryo is able to effectively regulate its developmentalschedule in response to MTX challenge.

Fig. 11. Control blood vessel. A high-power view of the central mesenchyme of a gestational day 12 control embryo. A blood vessel (v)is seen in the right portion of the field. Note the appearance of the endothelial cells (e), which have oval nuclei. The mesenchyme surrounding the vessel is characterized by the presence of many nuclei with smooth, oval contours. The mesenchymal cells give the appearance of being relatively homogeneous and of nearly filling the extracellular space. Fig. 12. MTX blood vessel. A high-power view of a portion of a limb bud from a gestational day 12 embryo whose mother was injected 16 hours previously with MTX. A blood vessel (v) can be seen in the right portion of the field. Note the normal appearance of the endothel i d cells (e) as evidenced by the smooth, oval contours of nuclei and cell membranes in contrast to the darkly staining, angular mesenchymal cell nuclei (m). Note the enlarged appearance of the extracellular space. Compare to Figure 11. Fig. 13. MTX + 16 h LV + 2 h. Limb bud from an embryo whose mother had received an ameliorative dose of LV 16 hours after having been administered a teratogenic dose of MTX. The embryo was harvested 2 hours after the LV injection (i.e., 18 hours post-MTX). Note that while the mesenchymal nuclei have regained their normal smooth, oval contours, the intercellular space remains enlarged. Compare to Figure 8. Fig. 14. MTX + 16 h LV + 6 h. Limb bud from an embryo whose mother had received an ameliorative dose of LV 16 hours after having been administered a teratogenic dose of MTX. The embryo was harvested 6 hours after the LV injection (i.e., 22 hours post-MTX). The anatomy of the limb bud is virutally normal. The intercellular space in the mesenchyme is small and uniform; the mesenchymal cells are uniformly compact. Compare to Figures 10 (MTX + 24h) and 4 (Control).

213

Enlarged intercellular spaces were observed in the mesenchymal compartments of limb buds after MTX injection of pregnant rabbits as well as embryonated chicken eggs (Brewton and MacCabe, ’90). In the living embryo, these spaces are not empty, but rather are filled with interstitial fluid, ions, glycosaminoglycans, and collagen. Brewton and MacCabe relied upon the ultrastructural observations of Darab et al. (’87), who reported vascular disruption in the heads of MTX-treatedmouse embryos in their diagnosis of the lesions in the limb buds of MTX-treated chick embryos as “edema.” This diagnosis implies that the increased spaces resulted from an accumulation of fluid. We note, however, that increased intercellular spaces in the limb bud mesenchymal compartment may be caused by other physiological processes. For instance, Sugrue and DeSesso (‘82) reported that rat embryos treated with hydroxyurea exhibited increased intercellular space in conjunction with the episode of hydroxyurea-induced mesenchymal cell death in affected limb buds. They demonstrated that the enlarged intercellular spaces were caused by an accumulation of hyaluronic acid. Consequently, observations t o date concerning the MTX-induced increases intercellular spaces in both chicken and rabbit embryos are insufficient t o indict any particular physiological process in causing such increases. The identification in rabbit limb buds of two populations of cells that appear to be affected differentially by MTX is a provocative observation. The reasons for this are 1) endothelial cells in the proliferating limb bud are the source of hyaluronic acid (Ausprunk, ’82), and 2) the enzyme that removes hyaluronate from the extracellular compartment is located within the mesenchyma1 cells, which must internalize the hyaluronate in order to degrade it (Toole et al., ’84). In normally developing, undifferentiated tissues, these two activities (synthesis and removal of hyaluronate) are in dynamic equilibrium (see discussion by Toole et al., ’84). Recent preliminary observations from our laboratory have demonstrated increased hyaluronate-positive material in the intercellular spaces of MTX-treated rabbit embryos (DeSesso et al., ’89). We suggest that MTX, through its selective effect on mesenchymal cells, destroys the balance between the synthesis and removal of hyaluronate in

214

J.M. DESESSO AND G.C. GOERINGER

Fig. 15. MTX + 24 h LV + 6 h. Limb bud from a n embryo whose mother had received a n ameliorative dose of LV 24 hours after having been administered a teratogenic dose of MTX. The embryo was harvested 6 hours after the LV injection (i.e., 30 hours post-MTX). Note the nearly complete return to the appearance of a control limb bud, including mesenchymal cell morphology and uniform intercellular space. A small amount of cellular debris can be seen (examples at arrows). Compare to Figure 9 (MTX + 24 h); the sequence of events is similar to that illustrated by Figures 8 (MTX + 16 h) and 14 (MTX + 16 h LV + 6 h).

Fig. 16. MTX + 24 h LV + 6 h. Limb bud from a n embryo whose mother had received an unsuccessful dose of LV 24 hours after MTX injection. The embryo was harvested 6 hours after the LV injection (i.e., 30 hours post-MTX). Note the continued pyknnsis among mesenchymal cells, the enlarged intercellular space (especially evident a t the distal tip), and the small amount of cellular debris (examples at arrows) throughout the mesenchyme. Approximately one third of examined embryos from this time period exhibited this condition.

the affected limb buds, thereby allowing hyaluronate to accumulate and so produce the increased intercellular space. LV appears able to reverse the effect of MTX on mesenchymal cells for up to 24 hours (or until the mesenchymal cells die) thereby re-establishing the equilibrium between the activities of endothelial and mesenchymal cells.

Berry, C.L. (1971) Transient inhibition of DNA synthesis by methotrexate in the rat embryo and foetus. J. Embryol. Exp. Morphol., 26:469-474. Bertino, J.R. (1977) ‘Rescue’ techniques in cancer chemotherapy: use of leucovorin and other rescue agents after methotrexate treatment. Semin. Oncol., 4:203216. Brewton, R.G., and J.A. MacCabe (1990) Studies of methotrexate-induced limb dysplasias utilizing a 51chromium release assay. Teratology, 41t211-221. Calabresi, P., and R.E. Parks, Jr. (1980) Chemotherapy of neoplastic diseases. In: Goodman and Gilman’s The Pharmacological Basis of Therapeutics, 6th ed. A.G. Goodman, L.S. Gilman, and A. Gilman, eds. MacMillan Publishing Company, New York, pp. 12491313. Cozens, D.D. (1965) Abnormalities of the external form and of the skeleton of the New Zealand white rabbit. Fd. Cosmet. Toxicol., 3:695-700. Crump, K.S., R.B. Howe, and C. Van Landingham (1987) Toxicology Risk Assessment Program. Clement Associates, K. S. Crump Division, Ruston, LA. Darab, D.J., R. Minkoff, J. Sciote, andK.K. Sulik (1987)

ACKNOWLEDGMENTS

The authors express their sincere gratitude to Mrs. Zofia Opalka for her outstanding and dedicated technical assistance. This research was funded by MITRE Sponsored Research Project 9587H. LITERATURE CITED Ausprunk, D.H. (1982) Synthesis of glycoproteins by endothelial cells in embryonic blood vessels. Dev. Biol., 9Ot79-90.

LV REVERSES MTX DEVELOPMENTAL TOXICITY Pathogenesis of median facial clefts in mice treated with methotrexate. Teratology, 36:77-86. DeSesso, J.M. (1981a) Ultrastructural analysis of rabbit limb buds subsequent to a teratogenic dose of either hydroxyurea or methotrexate. Teratology, 23:197215. DeSesso, J.M. (1981b) Amelioration of teratogenesis I: modification of hydroxyurea-induced teratogenesis by the antioxidant propyl gallate. Teratology, 24:19-35. DeSesso, J.M., and R.L. Jordan (1977) Drug-induced limb dysplasias in fetal rabbits. Teratology 15:122199. DeSesso, J.M., and G.C. Goeringer (1990) Ethoxyquin and nordihydroguaiaretic acid reduce hydroxyurea developmental toxicity. Reprod. Toxicol., 4:267-275. DeSesso J.M., G.C. Goeringer, and C.B. Underhill (1989) Early events in rabbit limb buds associated with methotrexate developmental toxicity and its amelioration by leucovorin. Teratology 39:450. Edwards, J.A. (1968) The external development of the rabbit and rat embryo. Adv. Teratol., 3t239-263. Giroud, A., and H. Tuchmann-Duplessis (1962) Malformations congenitales r6les des facteurs exogenes. Pathol. Biol. (Paris), IOr119-151. Giroud, A., A. Delmas, and M. Martinet (1963) Cyclocephalie: morphogenese e t mechanisme de sa production. Arch. Anat. Cytol. Pathol., 47:295. Goldman, I.D. (1974) The mechanism of action of methotrexate: I. Interaction with a low-affinity intracellular site required for maximum inhibition of deoxyribonucleic acid synthesis in L-cell mouse fibroblasts. Mol. Pharmacol., lOr257-274. Jordan, R.L., J.G. Wilson, and H.J. Schumacher (1977) Embryotoxicity of the folate antagonist methotrexate in rats and rabbits. Teratology, 15:73-79. Kalter, H., and J . Warkany (1959) Experimental production of congenital malformations in mammals by metabolic procedure. Physiol. Rev., 39~69-115. Khera, K.S. (1976) Teratogenicity studies with methotrexate, aminopterin, and acetylsalicylic acid in domestic cats. Teratology, 14:21-28. Miller, P.N., M.K. Pratten, and F. Beck (1989) Growth of 9.5-day rat embryos in folic-acid-deficient serum. Teratology, 39:375-385. Milunsky, A,, J.W. Graef, and M.F. Gaynor, J r . (1968) Methotrexate-induced congenital malformations. J. Pediatr., 72:790-795. Milunsky, A,, H. Jick, S.S. Jick, C.L. Bruell, D.S. MacLaughlin, K.J. Rothman, and W. Willett (1989) Multivitamidfolic acid sumlementation in earlv oreenancy reduces the pre&ence of neural tube h&ec&. JAMA, 262:2847-2852. Otis, E.M., and R.L. Brent (1954) Equivalent ages in mouse and human embryos. Anat. Rec., 12:33-63. Palmer, A.K. (1968) Spontaneous malformations of the New Zealand white rabbit: the background to safety tests. Lab. Anim., 2:195-206. Palmer, A.K. (1972) Sporadic malformations in laboratory animals and their influence on drug testing. In: Drugs and Fetal Development. M.A. Klingberg, A.

215

Abromvici, and J. Chemke, eds. Plenum Publishing Company, New York, pp. 54-60. Palmer, A.K. (1977) Incidence of sporadic malformations, anomalies, and variations in random bred laboratory animals. In: Methods in Prenatal Toxicology. D. Niebert, H.J. Merker, and T. Kwasigroch, eds. P.S.G. Publishing Company, Boston, pp. 52-71. Ray, A.A., J.P. Sall, M. Saffer, S.P. Joyner, and J.K. Whatley (1982) Statistical Analysis System (SAS) Computer Program. SAS Institute, Cary, NC. RQUX, C. (1964) Action teratogene du triparanol chez l'animal. Arch. Fr. Pediatr., 21:451-464. Sawin, P.B., and D.D. Crary (1964) Genetics of skeletal deformities in the domestic rabbit (Oryctolagus cuniculus). Clin. Orthop., 33:71-90. Schardein, J.L. (1985) Chemically Induced Birth Defects. Marcel Dekker, Inc., New York. Shepard, T.H. (1983)Catalog of Teratogenic Agents, 4th ed. The Johns Hopkins University Press, Baltimore, MD . Skalko, R.G., and M.P. Gold (1974) Teratogenicity of methotrexate in mice. Teratology, 9:159-163. Staples, R.E., and V.L. Schnell (1968) Refinements in rapid clearing technique in the KOH-alizarin red S method for fetal bone. Stain Technol., 43:61-63. Sugrue, S.P., and J.M. DeSesso (1982) Altered glycosaminoglycan composition of r a t forelimb buds during hydroxyurea teratogenesis: a n indication of repair. Teratology, 26:71-83. Thiersch, J.B. (1952) Therapeutic abortions with a folic acid antagonist, 4-aminopteroylglutamic acid (4amino PGA). Am. J . Obstet. Gynecol., 63:1298-1304. Toole, B.P. (1972) Hyaluronate turnover during chondrogenesis in the developing chick limb and axial skeleton. Dev. Biol., 29:321-329. Toole, B.P., and J . Gross (1971) Extracellular matrix of the regenerating newt limb: synthesis and removal of hyaluronate prior to differentiation. Dev. Biol., 25: 57-77. Toole, B.P., and R.L. Trelstad (1971) Hyaluronate production and removal during corneal development in the chick. Dev. Biol., 26:28-35. Toole, B.P., G. Jackson, and J . Gross (1972) Hyaluronate in morphogenesis: inhibition of chondrogenesis in uitro. Proc. Natl. Acad. Sci. U.S.A., 69~1384. Toole, B.P., R.L. Goldberg, G. Chi-Rosso, C.B. Underhill, and R.W. Orkin (1984) Hyaluronate-cell interactions. In: The Role of Extracellular Matrix in Development. R.L. Trelstad, ed. Alan R. Liss, Inc., New York, pp. 43-66. Warkany, J . (1971) Congenital Malformations. Year Book Medical Publishers, Chicago, IL. Wilson, J.G. (1971) Use of rhesus monkeys in teratological studies. Fed. Proc., 30:104-109. Wolfe, E. (1936) Les bases de la t6ratogenese experimentale chez les vert6br6s amniotes d'apr6s les resultats des m6thodes directes. Arch. Anat. Cytol. Pathol., 22:l. Wolfe, E. (1958) Un aperGu de l'embryologie experimentale du systeme nerveux. In: Colloque sur les Malformations Congenitales de YEncephale. College de France, Paris, pp. 17-39.