Effects of Supplemental Methionine onAntiserum-Induced Dysmorphology in RatEmbryos Cultured In VitroLYNDA B. FAWCETT,1,2* JOAN E. PUGARELLI,2 AND ROBERT L. BRENT1,2

1Department of Pediatrics, Jefferson Medical College, Philadelphia, Pennsylvania 191072Department of Research, Alfred I. duPont Hospital for Children, Wilmington, Delaware 19899

ABSTRACT

Background: Heterologous antiserum to the visceralyolk sac (AVYS) is teratogenic, inducing a spectrum ofmalformations in vivo and producing similar effects invitro. Numerous studies support the concept thatAVYS-induced malformations result from embryonic nu-tritional deficiency, without affecting the maternal nu-tritional status. This has provided a useful model withwhich to investigate the nutritional requirements of theearly embryo, as well as the role of various nutrients inthe etiology of congenital defects.Methods: In the current investigation, we examinedthe effects of methionine and other nutrients on AVYS-induced embryotoxicity in vitro. For these experiments,we cultured rat embryos (9.5 p.c) for 48 hr with AVYSand/or methionine at several concentration levels.Results: The addition of L-methionine to AVYS-exposedcultures reduced dysmorphology and open neuraltube; this effect was concentration dependent. AVYS-induced dysmorphology was completely prevented at aconcentration of L-methionine corresponding to 50-foldthe basal serum concentration. Utilization of D-methio-nine, L-leucine, or folic acid (5-methyltetrahydrofolate,MTHF) instead of L-methionine had no protective ef-fects.Conclusions: These results suggest that, althoughAVYS limits the supply of all amino acids to the embryo,embryopathy largely results from a deficiency of me-thionine. Furthermore, although endocytosis and deg-radation of proteins by the VYS supplies most aminoacids to the embryo, free amino acids may be com-pensatory when this source is reduced. These resultssupport those of previous investigations that suggestmethionine is required for normal NT closure and thatmethionine is a limiting nutrient for embryonicdevelopment.Teratology 61:332–341, 2000. © 2000 Wiley-Liss, Inc.

INTRODUCTION

Although the etiology of many human birth defectsremains unknown (Brent, ’86; Beckman et al., ’97a), ithas become increasingly evident that a significant pro-portion may result from effects relating to maternal

and/or embryonic nutritional status. It is now wellknown that periconceptional supplementation with fo-late, or vitamin supplements containing folate, reducesboth the overall incidence and the recurrence of neuraltube defects (NTDs) (MRC, ’91; Czeizel and Dudas, ’92).Unfortunately not all NTDs are prevented by folatesupplementation, and the mechanism by which folateprevents NTDs has not been determined. While re-duced levels of folate have been reported in some stud-ies for women with at least one prior NTD birth (Bun-duki et al., ’95), this is not a universal observation(Scott et al., ’90; Economides et al., ’92), and attentionhas turned to the hypothesis that folate supplementa-tion prevents NTDs by correcting subtle alterations infolate metabolism, rather than overt deficiency (Kirkeet al., ’96). This hypothesis is supported by observa-tions that alterations in key components of folate me-tabolism, such as homocysteine (Steegers-Theunissenet al., ’91, ’94) and vitamin B12 (Kirke et al., ’96), arealso associated with an increased risk of NTD.

There is growing evidence that supplementationwith methionine may also be able to prevent someNTDs, and that a methionine deficiency may increaserisk of NTD. In vitro culture of rat embryos usingserum-deficient methionine results in abnormalitiesincluding open NT (Flynne et al., ’87; Coelho et al., ’89),while supplementation of the culture serum with freemethionine promoted normal development and NT clo-sure (Flynne et al., ’87; Coelho et al., ’89). It has furtherbeen shown that methionine supplementation of wholerat embryo cultures reduces the dysmorphology thatresults from culture of embryos with anti-laminin an-tibodies (Chambers et al., ’95b), valproic acid (Noseland Klein, ’92), and nitrous oxide (Fujinaga and Baden,’94). Methionine supplementation in vivo reduced thespontaneous incidence of spina bifida in Axd mice by41–47% (Essien, ’92; Essien and Wannberg, ’93) andreduced the incidence of NTDs resulting from admin-istration of valproic acid to pregnant dams (Ehlers etal., ’96).

*Correspondence to: Lynda B. Fawcett, Department of Clinical Sci-ence, Alfred I. duPont Hospital for Children, P.O. Box 269, Wilming-ton, DE 19899. E-mail: lfawcett@nemours.org

Received 10 June 1999; Accepted 12 November 1999

TERATOLOGY 61:332–341 (2000)

© 2000 WILEY-LISS, INC.

The majority of amino acids used during organogen-esis by rat embryos are derived from digestion of ma-ternal proteins within the visceral yolk sac (VYS)(Beckman et al., ’90, ’97b; Lloyd et al., ’96). Greaterthan 90% of methionine, leucine, and serine used forprotein synthesis by the embryo have been shown to bederived from this source, rather than from free aminoacids (Rowe and Kalaizis, ’85; Lloyd et al., ’96; Beck-man et al., ’97b). This is perhaps not surprising, as freeamino acids account for only 1% of the total aminoacids present in rat serum (Koszalka et al., ’94). De-spite the requirement for methionine for both proteinsynthesis and for synthesis of the methyl group donorS-adenosylmethionine (SAM), methionine constitutesonly 1% of all serum amino acids, both free and incor-porated into protein (Milakofsky et al., ’84; Koszalka etal., ’94). Thus, it is of great interest that rat embryosgrown in cultures deficient in free methionine developabnormally, and supplementation of these cultureswith free methionine prevents the dysmorphology (Fly-nne et al., ’87; Coelho et al., ’89). This implies that freemethionine, although apparently present in insignifi-cant amounts in the serum as compared with methio-nine derived from maternal proteins, may assumegreater importance under circumstances in which theprimary methionine supply is limited, both in vitro andin vivo.

Investigations from our laboratory have long focusedon the mechanisms by which the early embryo obtainsnutrients and on the deleterious effects of embryonicmalnutrition (Brent and Fawcett, ’98). Much of ourcurrent understanding of early embryonic nutrition isderived from studies that use various teratogens toinhibit the transfer of nutrients to the embryo. In par-ticular, the discovery of teratogenic antibodies raised tocertain rat tissues, especially antiserum to the VYS(Brent et al., ’71), has provided a model that hasgreatly facilitated both the study of the nutritionalrequirements of the organogenesis stage rat embryoand the mechanisms by which the embryo derivesthese nutrients (for review, see Brent and Fawcett,’98). Anti-VYS serum (AVYS) inhibits VYS endocytosis(Freeman et al., ’82; Lerman et al., ’86). Because mostamino acids required by the rat embryo during orga-nogenesis are derived from endocytosed protein, expo-sure to AVYS significantly decreases the availability ofamino acids to the embryo, inducing a nutritional de-ficiency in the embryo without altering the nutritionalstatus of the mother. Exposure of yolk sacs to AVYSduring organogenesis results in aberrant in vitro de-velopment and in congenital malformations in vivo,including a high incidence of exencephaly. Direct expo-sure of the embryo the AVYS does not induce embry-opathy (New and Brent, ’72). Although AVYS inhibitsthe uptake of macromolecules such as protein, evidenceto date suggests that the transport processes of smallmolecules, such as free amino acids, are not signifi-cantly affected by AVYS (Lerman et al., ’86; Beckmanet al., ’90).

For the present investigation, we have addressed thepossibility that AVYS-mediated teratogenesis may re-sult in part from a methionine deficiency. This hypoth-esis was based on the current understanding of aminoacid supply to the embryo, the relatively low concen-tration of methionine in serum as compared with otheramino acids, and the previous reports that methioninedeficiency resulted in NTDs in vitro. Because mostamino acids are supplied to the embryo by endocytosisand degradation of protein in the VYS endoderm, inhi-bition of VYS endocytosis would theoretically decreaseavailability of all amino acids. However, because of itslow concentration and potential developmental impor-tance, methionine may become limiting before othermore abundant amino acids in AVYS-treated embryos.For these experiments we first measured the effects ofvarious doses of AVYS on embryonic development invitro and compared these results with development invivo after a teratogenic in vivo dose of AVYS was ad-ministered. We then supplemented 9.5-day rat embryoculture with various concentrations of methionine todetermine if methionine could ameliorate the embryo-pathic effects of AVYS in vitro. Finally we conductedexperiments to determine if supplemental folic acidreduced AVYS-induced embryopathy.

MATERIALS AND METHODS

Animals and timed pregnancies

Wistar rats weighing 200–250 g (Charles River Lab-oratories, Wilmington, DE) were used for all experi-ments. The rats were mated overnight, and the pres-ence of sperm in the vaginal lavage the next morningwas used to indicate 0.5 days postconception (p.c.).Food and water were available ad libitum.

Whole embryo culture

Mid-head-fold stage embryos from pregnant ratswere obtained at 9.5 days p.c. Embryos were culturedaccording to the method of New (’78). Pregnant femaleswere euthanized and gravid uteri were removed. Em-bryos were isolated for culture by dissecting uterine,decidual, and Reichart’s membrane tissues away fromthe embryo in phosphate buffered saline containing0.1% glucose at 37°C. Embryos with attached yolk sacsand ectoplacental cones were cultured at 37°C underconstant rotation (35–40 rpm) in whole rat serum thatwas prepared by immediate centrifugation, heat-inac-tivated at 56°C for 30 min, and stored at 270°C beforeuse. Culture serum was supplemented with 50 mg/mlstreptomycin and 50 U/ml penicillin (Sigma ChemicalCo., St. Louis, MO). At 3 hr after culture initiation,embryos were evaluated for normal VYS expansion andabsence of damage resulting from the isolation proce-dure. Defective embryos were excluded from furtheranalysis. Up to four embryos were cultured together,using 1 ml of rat serum per embryo. Cultures weregassed with 5%O2:5%CO2:90%N2 on the first day ofculture, with 20%O2:5%CO2:75%N2 on the second dayof culture, and with 40%O2:5%CO2:55%N2 for the final

EFFECTS OF METHIONINE SUPPLEMENT 333

6 hr. Embryos were harvested from culture 48 hr afterinitiation of the treatments and evaluated.

Evaluation of embryos

At the termination of culture, embryos were trans-ferred to sterile phosphate-buffered saline (PBS) con-taining 1% glucose at 37°C, and immediately examinedusing a stereomicroscope. The presence of a heartbeatwas used to indicate viability, and only viable embryoswere evaluated. The maximum diameter of the visceralyolk sac was measured using an ocular micrometer,and the relative degree of axial rotation was assessed.Embryos were then dissected free of the VYS, ectopla-cental cone, and amnion; growth parameters and mor-phology were assessed. Crown-rump length, defined asthe maximum diameter of the embryo in its naturalposition (Brown and Fabro, ’81), was measured usingan ocular micrometer before removal of the amnion.After evaluation, embryos and visceral yolk sacs werefrozen at 270°C. Total protein was determined by dis-solving yolk sacs and embryos in 0.5 M NaOH for 1 hrat 37°C. The protein content was determined using themethod of Lowry et al. (’51).

Teratogenic antiserum and culturesupplements

Cultures were supplemented with L- or D-methionine(Sigma) dissolved in sterile deionized water in a vol-ume no greater than 50 ml to achieve a final concen-tration of 100 or 500 mg methionine/ml culture me-dium, which corresponded to 10-fold or 50-fold basalserum concentrations respectively. Alternatively, cul-ture medium was supplemented with L-leucine toachieve a final concentration of 328 or 656 mg/ml (cor-responding to 10-fold or 20-fold basal concentrationsrespectively), or with 5MTHF (Sigma) to achieve a finalconcentration of 25 mg/ml or 500-fold the basal serumconcentration. Control cultures received an equal vol-ume of sterile deionized water. After the initial 3 hr ofculture, one-half of the cultures were treated withAVYS. The AVYS used for this study was a sheepanti-whole visceral yolk sac that has been stored lyoph-ilized at 220°C since its initial preparation (Brent etal., ’71), and has been used in previous investigations(Brent et al., ’71; New and Brent, ’72; Fawcett et al.,’95; Beckman et al., ’97b). Antiserum was reconstitutedwith sterile deionized water and was added to embryocultures in a volume no greater than 15 ml, to achievea final concentration of either 50, 75, or 100 mg anti-serum protein per ml culture serum. An equal volumeof sterile deionized water or normal (pre-immune)sheep serum (NSS) was added to control cultures.

Embryopathy of AVYS in vivo

Pregnant rats (9.5 p.c.) were administered a singleintraperitoneal injection of AVYS (70 mg protein/ml) ata dose of 100 mg/kg body weight (LD50). On day 11.5days p.c., 48 hr postinjection, dams were euthanized,using CO2. Gravid uteri were removed, the total sitescounted, and then 5 sites were randomly chosen and

removed for analysis. Embryos were dissected free ofthe uterine wall and decidua; the diameter of the VYSand vitelline circulation was assessed. Embryos werethen removed from the VYS and assessed for morphol-ogy and size, using the same criteria used for the eval-uation of cultured embryos.

Statistical analysis

All statistical analysis was performed with the aid ofSigma Stat statistical software (SPSS). Embryo andVYS size and protein measurements were comparedusing one-way analysis of variance (ANOVA). Pairwisecomparisons were achieved using Student-Newman-Keuls method. Developmental parameters and inci-dence of malformations were analyzed using the chi-square test with the Yates correction, or the Fisher’sexact test where appropriate. The level of significancewere set at P # 0.05 for all tests.

RESULTS

Comparison of AVYS in vitro dose responsewith in vivo teratogenic effects

In the first series of experiments, we ascertained theeffect of several doses of AVYS on embryonic develop-ment in vitro using 48-hr embryo culture. We thencompared the outcome of embryos with embryos ex-posed for 48 hr in vivo by injecting pregnant dams witha teratogenic dose of AVYS on 9.5 days p.c. The resultsfrom these experiments are presented for comparisonin Table 1.

When pregnant dams were administered a dose ofAVYS on day 9.5 p.c. and embryos were examined onday 11.5 p.c., most embryos were found to be dysmor-phic. In particular, these embryos had open NTs andabsent or abnormal development of optic structures.Although approximately one-fourth of embryos ap-peared morphologically normal, all were severelygrowth retarded as evidenced by the protein content ofembryos and VYSs and the number of somites, both ofwhich were significantly reduced. Administration of anequal dose of preimmune sheep serum (NSS), ratherthan AVYS, did not significantly affect embryonic de-velopment as previously reported (Brent et al., ’71;New and Brent, ’72); however, two embryos from onelitter had at least one dysmorphic feature involving theNT and/or eye which resulted in an incidence of 8%dysmorphology overall.

Embryos cultured for 48 hr in vitro in whole ratserum were smaller than same stage embryos but nor-mal in morphology and consistent with previously pub-lished data with regard to size and protein content.(Table 1). Embryos had an average of 26 somite pairs,three brachial arches, full dorsal flexion, and the vastmajority had completed NT closure. Embryos andVYSs averaged approximately 170–180 mg and 100 mgof protein respectively. The addition of NSS to culturesdid not adversely affect development; these data havebeen included with the data for embryos cultured with-out additional supplements. The addition of AVYS to

334 FAWCETT ET AL.

rat embryo cultures significantly inhibited embryonicgrowth and development and significantly increasedmorphologic abnormalities, especially with regard tooptic structures and NT development. Consistent withthe in vivo effects of AVYS, the effects observed withAVYS in vitro were dose dependent. A concentration of50 mg AVYS/ml culture medium significantly inhibitedaxial rotation, somite number, and hindlimb bud de-velopment of the embryos. In addition, only 33% ofembryos had normal eye development, and only 41%had completed NT closure. These abnormalities usu-ally manifested in the same embryos. Open NTs in thisgroup were restricted to the anterior regions of theneural groove. Surprisingly, exposure of embryos toAVYS did not measurably affect the protein content ofeither the VYS or embryo at this dose. Increasing theconcentration of AVYS to 75 mg/ml resulted in a de-creased proportion of embryos completing axial flexionand an increased proportion of embryos with abnormalor absent eye development and open NTs (Table 1).Most embryos treated with 75 mg/ml AVYS had openNTs (95%). The severity of the open NT was also in-creased; 45% of embryos treated with 75 mg/ml AVYShad an open NT encompassing the entire neuralgroove. Finally, addition of AVYS to cultures at 100mg/ml resulted in the most severe effects on culturedembryos. Protein accretion was only one-third of con-trol embryos, and embryos had significantly reducedsomite number, reduced or absent limb bud develop-ment and lacked axial flexion. Additionally, the vastmajority of these embryos had open NTs encompassingthe entire neural groove. Because the embryos failed toachieve appropriate rotation, the cranial and caudalaspects of the neural tubes that were in oppositionbecame fused together in a large proportion of theseembryos. A significant number of embryos treated with100 mg/ml AVYS were also absent or had dysmorphicotic vesicles and the majority (52%) of yolk sacs lackeda vitelline circulation. Despite these severe effects on

growth and development, most embryos (.98%)treated with this dose of AVYS were viable at exami-nation. It should be noted that although the embryoswere significantly retarded with regard to growth anddevelopment, the failure of the embryos to initiate NTclosure cannot be explained entirely by developmentaldelay, because the morphology of the NT at all treat-ment doses of AVYS was frankly abnormal. Open neu-ral tubes were consistently flared ventrally. Further-more, in the more severely affected embryos, theanterior and posterior NT were fused, a malformationwhich has been previously described as dorsifusion byother investigators (Chambers et al., ’95a).

Effect of supplemental free L-methionineon AVYS-induced dysmorphology

Results of morphologic assessment of embryos cul-tured for 48 hr in the presence of AVYS serum at either50 or 100 mg/ml were entirely consistent with thosedescribed for the dose response analysis presented inTable 1; these results are presented in Table 2. Addi-tion of 100 mg/ml L-methionine (100 mg or 500 mg/ml)alone had no effect on the growth or development ofembryos. The addition of D-methionine alone at 100mg/ml significantly (P , 0.05) affected embryonic de-velopment with regard to optic morphology as com-pared with embryos cultured without supplements, orembryos cultured with L-methionine or NSS. Eye ves-icles were present but abnormal in 35% of these em-bryos. Some embryos cultured with 100 mg/ml D-methi-onine also had an open anterior NT; however, thisvalue was not significantly different from control val-ues, perhaps because of the small sample size.

The addition of free L-methionine to cultures signif-icantly improved the outcome of embryos treated withAVYS, particularly at a dose of 50 mg/ml AVYS. Whencultures were supplemented with free L-methionine at10 times the reported basal serum concentration, (100mg/ml; Milakofsky et al., ’84), the percentage of em-

TABLE 1. Comparison of rat embryos from dams treated with a teratogenic dose of AVYS (LD50) in vivo andrat embryos grown with several concentrations of AVYS in vitro 48 hr postexposure (11.5 days p.c)†

AVYS

In vivo In vitro

0 mg/kg 100 mg/kg 0 mg/ml 50 mg/ml 75 mg/ml 100 mg/ml

No. of embryos 25 43 62 39 20 33Crown-rump (mm) 4.2 6 0.3 2.4 6 0.3* 3.2 6 0.4 3.1 6 0.4 3.0 6 0.3 1.8 6 0.4*VYS (mm) 4.8 6 0.3 3.1 6 0.3* 3.7 6 0.5 3.6 6 0.6 3.7 6 0.3 2.8 6 0.4*Embryo protein (mg) 368.5 6 38.2 92.9 6 15.8* 176.7 6 49.0 172.9 6 48.1 172.2 6 29.4 50.6 6 23.8*VYS protein (mg) 151.4 6 14.5 62.6 6 11.1* 100.7 6 27.4 105.2 6 27.0 106.3 6 20.6 48.9 6 17.7*No. of somites 30.0 6 1.9 24.2 6 6.8* 26.5 6 4.1 25.5 6 4.9 24.3 6 6.1 5.8 6 8*No. of arches 3.1 6 0.3 2.7 6 0.5 3.0 6 0.4 2.8 6 0.6 3.0 6 0.5 1.3 6 1.4*Rotation (%) 100 97.7 95.2 71.8* 45.0* 3*VYS circulation (%) 100 100 98.4 92.3 100 52*Forelimb bud (%) 100 95.3 100 100 100 36*Hindlimb bud (%) 100 88.4 56 25.6* 10* 0*Otic (% normal) 100 97.7 100 97.4 95 58*Optic (% normal) 92 25.6* 93.5 33.3* 5* 0*Neural tube (% closed) 96 30.2* 93.5 41.0* 5* 0*Dysmorphic (%) 8 74.4* 6.4 66.7* 95* 100*

*Significantly different from control values P , 0.05.†Values are mean 6 SD, or percentage of embryos where indicated.

EFFECTS OF METHIONINE SUPPLEMENT 335

TA

BL

E2.

Eff

ect

ofsu

pp

lem

enta

lL-M

eth

ion

ine

onA

VY

S-i

nd

uce

dd

ysm

orp

hol

ogy

inra

tem

bry

oscu

ltu

red

invi

tro

from

9.5

to11

.5d

ays†

Dos

eof

AV

YS

supp

lem

ent

0m

gA

VY

S/m

l50

mg

AV

YS

/ml

100

mg

AV

YS

/ml

Non

eD-M

eth

ion

ine

100

mg/

ml

L-M

eth

ion

ine

100

mg/

ml

L-M

eth

ion

ine

500

mg/

ml

Non

eL-M

eth

ion

ine

100

mg/

ml

L-M

eth

ion

ine

500

mg/

ml

Non

eD-M

eth

ion

ine

100

mg/

ml

L-M

eth

ion

ine

100

mg/

ml

L-M

eth

ion

ine

500

mg/

ml

No.

ofem

bryo

s88

1726

2255

6032

4822

4432

Cro

wn

-ru

mp

(mm

)3.

26

0.4

3.1

60.

33.

26

0.4

3.3

60.

33.

26

0.4

3.1

60.

53.

36

0.3

2.0

60.

6*1.

96

0.5*

2.2

60.

4**

2.4

60.

4**

VY

S(m

m)

3.9

60.

63.

76

0.4

3.7

60.

34.

56

0.5

3.7

60.

63.

66

0.6

4.0

60.

23.

06

0.4*

2.8

60.

3*2.

96

0.4*

3.1

60.

3*E

mbr

yopr

otei

n(m

g)18

7.1

647

.515

1.6

642

.118

0.0

634

.518

6.9

633

.118

0.0

655

.317

5.2

653

.217

6.8

625

.463

.76

36.3

*55

.36

17.8

*66

.46

11.9

*70

.56

26.9

*V

YS

prot

ein

(mg)

110.

76

30.1

83.3

614

.391

.56

14.8

133.

86

25.2

104.

96

27.9

105.

96

23.4

117.

86

11.1

58.4

622

.6*

58.1

69.

8*54

.26

10.1

*64

.76

13.5

*N

o.of

som

ites

27.2

63.

926

.56

5.0

24.1

67.

729

.06

3.4

24.0

66.

1*21

.46

5.3*

29.6

62.

57.

46

7.8*

8.4

66.

4*12

.86

10.0

**13

.36

9.1*

*N

o.of

arch

es3.

06

0.4

2.9

60.

32.

96

0.5

3.0

60.

62.

86

0.6*

2.8

60.

5*3.

36

0.5

1.6

61.

4*1.

76

1.2*

2.5

61.

1**

2.1

61.

2**

Rot

atio

n(%

)96

.682

.496

.190

.170

.9*

80*

100

2.1*

0*2.

3*12

.5V

YS

circ

ula

tion

(%)

98.9

94.1

100

95.4

92.7

100

100

66.7

*54

.5*

87.5

*93

.8F

orel

imb

bud

(%)

100

94.1

96.1

100

100

100

100

50.0

*50

.0*

77.3

**81

.2**

Hin

dlim

bbu

d(%

)58

.052

.946

.168

.221

.8*

48.3

53.1

0*0*

0*6.

2*O

tic

(%n

orm

al)

100

94.1

96.1

100

96.4

98.3

100

56.2

*59

.1*

97.7

87.5

Opt

ic(%

nor

mal

)98

.964

.7*

8890

.936

.4*

56.7

*96

.90*

0*2.

3*3.

1*N

eura

ltu

be(%

clos

ed)

96.6

82.4

100

100

45.4

*75

.0**

100

0*0*

0*6.

2*D

ysm

orph

ic(%

)2.

435

.3*

11.5

9.1

65.4

*46

.7*

3.1

100*

100*

100*

96.9

*

*Sig

nifi

can

tly

diff

eren

tfr

omco

ntr

olva

lues

P,

0.05

.**

Sig

nifi

can

tly

diff

eren

tfr

omco

ntr

olva

lues

and

AV

YS

alon

eP

,0.

05.

† Val

ues

are

mea

n6

SD

orpe

rcen

tage

ofem

bryo

sw

her

ein

dica

ted.

336 FAWCETT ET AL.

bryos treated with 50 mg/ml AVYS completing normalNT closure increased significantly from 45% to 75%(Table 2). This dose also restored hindlimb bud devel-opment to control values, and modestly increased thepercentage of embryos with normal optic developmentand axial rotation. However, a significant number ofembryos in this group were still dysmorphic (47%; Ta-ble 2).

Supplementation of cultures with L-methionine at 50times the basal serum concentration significantly im-proved embryonic development such that embryostreated with 50 mg/ml AVYS in the presence of 50-foldL-methionine were indistinguishable from control em-bryos. Specifically, all (100%) of these embryos com-pleted NT closure and appropriate axial rotation. Inaddition, only one embryo from this group had an opticabnormality, an incidence that does not differ fromcontrol values.

Results obtained with L-methionine supplementa-tion of embryo cultures at higher doses of AVYS werenot as dramatic as those obtained with 50 mg/ml AVYS.Supplementation of embryo cultures treated with 100mg/ml AVYS with L-methionine at 10-fold the normalrat serum concentration (770 mM) led to significantimprovement for some developmental parameters ascompared with embryos treated with AVYS alone. Inparticular, supplementation with L-methionine signif-icantly improved the number of somite pairs as well ascrown-rump length. Embryos cultured with AVYS and10-fold L-methionine also had significantly fewer oticmalformations compared with embryos cultured withAVYS alone. A significant improvement was also notedin the development of the vitelline circulation. Supple-mentation of cultures with 50 times the basal concen-tration of free L-methionine in the rat serum, followedby treatment of the cultures with 100 mg/ml AVYS,achieved essentially the same results obtained with the10-fold concentration of L-methionine. The percentageof embryos achieving partial closure of the NT wassignificantly increased in AVYS-treated cultures sup-plemented with L-methionine as compared with cul-tures treated with 100 mg/ml AVYS alone. This re-sulted in a concurrent decrease in the percentage ofembryos treated with antiserum that had open NTsencompassing the entire neural groove.

Despite these differences, all embryos exposed to thehigher dose of AVYS (100 mg/ml) and supplementedwith L-methionine were still abnormal, and proteinaccretion in either the VYS or embryo was not affected.Addition of D-methionine instead of L-methionine inthese experiments yielded results virtually identical tothose obtained with AVYS alone, indicating that theeffects of L-methionine are stereo specific. Further-more, although D-methionine did not increase the tox-icity of AVYS in culture, in experiments in which D-methionine was added to cultures alone the number ofembryos with abnormal optic development was signif-icantly increased, indicating that D-methionine itselfmay have adverse effects on embryonic development.

Effects of 5MTHF and L-leucine on AVYS-induceddysmorphology in vitro

Embryos cultured for 48 hr in whole rat serum withsupplemental folic acid (5MTHF) or L-leucine grew nor-mally and did not differ from embryos grown withoutsupplements. The addition of 50 mg/ml AVYS to thesecultures resulted in similar morphology as that re-ported for the previous experiments described abovewith the exception that a significant decrease in totalprotein accretion by both the embryo and VYS werenoted. The addition of either 5MTHF or L-leucine toAVYS-treated cultures for 48 hr did not affect embry-onic growth or morphology, compared to embryosgrown with AVYS alone. Culture of embryos with 50mg/ml AVYS and either leucine or 5MTHF produceddysmorphology in more than two-thirds of embryos anddid not differ significantly from embryos cultured withAVYS alone in which 50% had NTD and almost 60%were dysmorphic (Table 3).

DISCUSSION

The degradation of endocytosed proteins within lyso-somes of the VYS endoderm provides most (85–97%) ofthe amino acids used by the organogenesis stage ratembryo (Rowe and Kalaizis, ’85; Lloyd et al., ’96). It hasbeen well documented that the primary effect of AVYSis an inhibition of VYS endocytosis that produces anembryonic nutritional deficiency (Freeman et al., ’82;Lerman et al., ’86; Brent et al., ’90; Beckman et al., ’91).Although AVYS disrupts the supply of all amino acidsto the embryo, amino acids such as methionine are insuch short supply that the embryo may become defi-cient in these before other, more abundant, amino ac-ids. In the present study, we have addressed the pos-sibility that the embryopathy induced by AVYS results,at least in part, from a deficiency in methionine, ratherthan a generalized reduction of amino acids. Experi-ments performed to assess the ability of methionine toovercome AVYS-induced teratogenicity in vitro supportthe hypothesis that AVYS-induced embryopathy re-sults in large part from reduced availability of methi-onine.

The effects of AVYS, like other teratogens, are dosedependent. Thus, for our initial experiments, weneeded to determine the relationship between the invivo versus the in vitro effects of AVYS. The predomi-nant malformations observed with a teratogenic (LD50;100-mg/kg) dose of AVYS administered during organo-genesis in vivo are micro/anophthalmia and exen-cephaly (Brent et al., ’71, ’90; Fawcett et al., ’95; Brentand Fawcett, ’98). NTD encompassing other regions ofthe NT, such as spina bifida, are infrequent. Thepresent study confirmed that 48 hr after administra-tion of a teratogenic dose of AVYS to 9.5-day pregnantrats in vivo, the most common defects observed in theembryos are anophthalmia and a failure of anterior NTclosure. On the basis of these results, we were able todetermine that the concentration of AVYS that re-sulted in a similar morphology in vitro was 50 mg/ml

EFFECTS OF METHIONINE SUPPLEMENT 337

AVYS. At this dose, a significant majority of embryosfailed to close the anterior NT and had a high incidenceof optic dysmorphology, but were of normal size andhad completed axial flexion. Higher doses of AVYS invitro led to a failure to close the entire NT, lack of axialflexion, and significant inhibition of other growth pa-rameters. This morphology is more consistent withthat expected for a dose that would not permit survivalif administered in vivo; thus, none of these malforma-tions would be observed at term.

It is noteworthy that the plasma concentration ofAVYS resulting from an in vivo teratogenic dose isapproximately 50-fold higher than the concentration ofAVYS found to exert a similar morphologic effect on theembryos in vitro. This difference is best explained bythe decreased accessibility of the VYS to the maternalcirculation in vivo, and the length of time the VYS isexposed. Previous investigations have demonstratedthat the VYS-reactive IgG is rapidly removed from thematernal circulation (Fawcett et al., ’95). This is due, atleast in part, to the presence of cross-reactive proteinsin other rat tissues. In culture the VYS is directlyexposed to the entire dose administered over a moreprolonged period of time. Thus while maternal plasmalevels of AVYS may be higher with an in vivo admin-istration, the amount of AVYS that the conceptus isexposed to is likely to be only a fraction of the maternalplasma levels. Based on the morphological criteriaused for the present study, it is likely that the overallexposure achieved in vivo is similar to that obtained bythe lower in vitro dose.

Once we identified an in vitro concentration of AVYSthat produced a similar embryopathy as that resultingfrom a teratogenic in vivo dose, and that would permitsurvival to term, we were able to assess the effects ofmethionine on AVYS induced embryopathy in vitrousing both a “teratogenic” (50 mg/ml) and an “embryo-lethal” (100 mg/ml) dose of AVYS. At the higher of thetwo concentrations of AVYS (100 mg/ml), supplementalL-methionine significantly improved the degree of NT

closure, although the defect was still present. Supple-mental L-methionine also reduced the incidence ofother defects observed, especially with regard to oticdevelopment, and somite number. When L-methioninewas supplied at 50-fold the basal concentration and thelower “teratogenic” dose of AVYS was used (50 mg/ml),the embryopathy was completely prevented and theAVYS-exposed embryos were indistinguishable fromlittermate control embryos. No effects were observedwhen D-methionine was used instead of L-methionineat the same concentration. Furthermore, supplementa-tion of cultures with L-leucine or 5MTHF did not alterthe outcome of embryos treated with AVYS. These datastrongly support the concept that the effects of L-me-thionine are specific to this amino acid, although theefficacy of other compounds for amelioration of AVYS-induced embryopathy cannot be ruled out at this time.

Previous reports have established that adequate me-thionine is essential for proper NT development in therat. Culture of embryos in serum deficient in methio-nine produces NTDs in vitro (Flynne et al., ’87; Coelhoet al., ’89). Furthermore, supplemental methionineameliorates the embryopathy induced by a variety ofsubstances both in vivo (Essien, ’92; Essien and Wann-berg, ’93; Ehlers et al., ’96) and in vitro (Flynne et al.,’87; Coelho and Klein, ’90; Van Aerts et al., ’93, ’94;Chambers et al., ’95b). Of particular interest to thepresent investigation are studies demonstrating thatsupplemental free methionine ameliorates embryopa-thy induced by anti-laminin antibodies in vitro (Cham-bers et al., ’95b). As in the present study, D-methionineproved ineffective, as were other amino acids or vita-mins (Chambers et al., ’95). The in vitro mechanism bywhich anti-laminin induces embryopathy may be sim-ilar to AVYS in that it may involve inhibition of nutri-ent transport to the embryo. Anti-laminin has beenreported to inhibit both endocytosis, as measured byuptake of sucrose, and the transport of free aminoacids. (Chambers et al., ’95a), and could therefore leadto nutritional deprivation of the embryo. However, al-

TABLE 3. Effect of leucine and folate on incidence of dysmorphology in cultured rat embryos resultingfrom in vitro exposure to AVYS from 9.5 to 11.5 days

AVYS supplement

0 mg/ml culture medium 50 mg/ml culture medium

None L-Leucine MTHF† None L-Leucine 103 L-Leucine 203 MTHF (mg/ml)

No. of embryos 51 11 12 26 28 21 32Crown-rump (mm) 3.5 6 0.4 3.3 6 0.3 3.5 6 0.5 2.99 6 0.7* 2.7 6 0.5* 2.94 6 0.4* 3.0 6 0.5*VYS (mm) 4.4 6 0.5 4.2 6 0.56 4.0 6 0.4 3.57 6 0.8* 3.46 6 0.42* 3.8 6 0.4* 3.6 6 0.5*Embryo protein (mg) 206.6 6 38.1 195.8 6 34.5 209.4 6 48.03 147.0 6 52.2* 135.9 6 36.4* 143.2 6 39.1* 153.2 6 38.6*VYS protein (mg) 118.7 6 15.4 114.4 6 18.8 104.7 6 30.9 82.1 6 26.3* 75.8 6 24.0* 106.4 6 24.6 85.7 6 20.6*No. of somites 28.84 6 2.9 28.45 6 3.9 28.6 6 1.5 24.7 6 7.7 22.65 6 7.9* 22.4 6 7.4* 22.6 6 6.3*No. of arches 2.9 6 0.3 3.0 6 0 3.0 6 0 2.60 6 0.85 2.7 6 0.5 2.6 6 0.6 2.6 6 0.8Rotation (%) 96.1 91 91.7 73.1* 64.3* 76.2* 62.5*VYS circulation (%) 98 100 100 80.8* 89.3 95.2 93.8Forelimb bud (%) 100 100 100 92.3 100 100 96.9Hindlimb bud (%) 62.7 72.7 91.7 23.1* 28.6* 23.8* 15.6*Otic vesicles (%) 100 91 100 92.3 96.4 100 90.6Optic vesicles (%) 98 91 100 50* 42.9* 19* 28.1*Neural tube (% closed) 98 91 100 50* 50* 19* 34.4*Dysmorphic (%) 3.9 9.1 0 57.7* 57.1* 80.9* 71.9*

*Significantly different from control value (P , 0.05).†MTHF; 5-methyltetrahydrofolate.

338 FAWCETT ET AL.

though anti-laminin may induce NTD by a similarmechanism as AVYS in vitro, it is likely that the in vivomechanism of anti-laminin embryotoxicity differs fromAVYS. While AVYS induces malformations in vivo in-cluding NTD, anti-laminin antibodies have not beenassociated with increased incidence of NTD or othermalformations in vivo, but rather are embryolethal(Foidart et al., ’83) and have been associated with re-productive failure and/or spontaneous abortion (Szarf-man et al., ’82; Foidart et al., ’86; Weeks et al., ’89;Chambers et al., ’95a,b).

Previous studies have demonstrated that AVYS in-hibits endocytosis by the VYS but does not affect thetransport of small molecules, such as leucine, a-aminoisobuteric acid, and deoxyglucose (Lerman et al., ’86;Beckman et al., ’90). Transport of free methioninewould not be expected to be affected by exposure toAVYS, as confirmed by recent data from our laboratory(Fawcett et al., ’99). Thus, it is likely that transport ofsupplemental free methionine can compensate for theloss of methionine normally provided by endocytosisand protein degradation, even though greater than90% of the methionine used by the embryo is normallyderived from endocytosis (Lloyd et al, ’96). This mayexplain why free methionine was able to overcomeAVYS-induced embryopathy in vitro in the present in-vestigation. It should be noted that although transportof free amino acids is not affected, even at the higher ofthe two concentrations used in the present study (100mg/ml; Lerman et al., ’86; Beckman et al., ’90), theeffects of this concentration on the VYS may be sosevere that deficiencies in other amino acids/nutrientsmay also occur, resulting in growth retardation andreduced protein accretion. Under these circumstances,additional methionine cannot completely prevent theembryopathic effects of AVYS.

Attempts to purify the antigenic target of AVYS haveshown that the antigen responsible for teratogenesismay be a high-molecular-weight glycoprotein on theapical membrane of the visceral endoderm (Leung, ’82;Leung et al., ’85). Combined evidence from adsorptionusing several rat tissues, immunodiffusion, electro-phoresis, and Western blot has demonstrated that theantigen has immunologic homology with a protein onthe brush border of kidney proximal tubules. Althoughadsorption of AVYS with kidney proximal tubules de-creases teratogenicity, adsorption with basement mem-brane material, including collagen, laminin, or base-ment membrane tissues, did not reduce AVYSteratogenicity, indicating that the antigen has no ho-mology to these proteins (Jensen et al., ’75; Brent et al.,’90). Previously isolated proposed targets had molecu-lar weights of 30 and 60 kD, respectively (Leung et al.,’85). Recently, Moestrup et al. (’98) reported that thetarget of teratogenic monoclonals to the VYS used intheir studies is a 420-kD intrinsic factor-vitamin B12receptor also found on the brush border of kidney prox-imal tubules. This finding is of great interest in light ofthe fact that vitamin B12, methionine, and folate path-ways are intimately linked. The putative antigens pu-

rified thus far in our laboratory are not of high molec-ular weight; however, the possibility remains thatthese proteins are subunits of a larger protein (Brentand Fawcett, ’98). The similarity of the target of ourantiserum to other potential teratogenic targets canonly be fully addressed with final purification and se-quencing of the proteins currently under study.

The mechanism by which methionine prevents NTDshas not been definitively established. However, evi-dence from several studies, including the present in-vestigation, indicate that methionine supplementationovercomes a direct deficiency of methionine or SAM,rather than metabolic alterations of other componentsof one carbon metabolism (Coelho et al., ’89; Coelho andKlein, ’90; Moephuli et al., ’97). In the present study, upto 500 times the basal concentration of folic acid, pro-vided as 5MTHF, failed to improve the outcome ofembryos exposed to AVYS. Furthermore, because sup-plemental methionine did not appear to affect proteincontent in either normal embryos, or in those exposedto AVYS, it is doubtful that the mechanism involvesincreased protein synthesis. Klein and colleagues haveproposed that insufficient methionine, and conse-quently reduced SAM, results in reduced methylationof key proteins involved in NT closure (Coelho et al.,’89; Coelho and Klein, ’90; Moephuli et al., ’97). Thishypothesis is supported by the demonstration that ac-tin, myosin, and several other proteins in the NT arehypomethylated when embryos are cultured in serumthat is deficient in methionine (bovine serum) in vitro(Coelho and Klein, ’90; Moephuli et al., ’97). In a morerecent study, NTDs in rodent embryos grown in methi-onine-deficient serum could be prevented using homo-cysteine, as well as methionine, but not folinic acid(Van Aerts et al., ’94). At high concentrations, L-homo-cysteine is embryotoxic in vitro but does not induceNTD. The embryotoxicity resulting from elevated ho-mocysteine was attenuated using serine (a methylgroup donor), 5MTHF, or vitamin B12 (Van Aerts et al.,’94). These results suggest that methionine was pro-vided to the embryo by the methylation of homocys-teine and the demethylation of 5MTHF via methioninesynthase, and further demonstrates the importance ofmethylation for normal embryonic development.

The importance of methionine in human pregnancyoutcome remains unknown; few studies have ad-dressed this question. In one early report, dietary sup-plements containing methionine were found to increasefavorable pregnancy outcome for women with a historyof recurrent spontaneous abortion (Farrari et al., ’94).A recent epidemiological study reported an associationbetween dietary intake of methionine and reduced riskof NTD independent of folate status, although the datacould easily support an association with other nutri-ents (Shaw et al., ’97). Unfortunately, it is difficult tostudy the in vivo effects of methionine deficiency usinganimals models. Several early investigations (Sims,’51; Tagliamonte et al., ’76) demonstrated that al-though pregnant rats could be made methionine defi-cient, litters were normal with respect to morphology

EFFECTS OF METHIONINE SUPPLEMENT 339

and levels of methionine. This was largely due to theefficient transfer of methionine from maternal stores tothe fetus. Deficiency of methionine resulted in majorliver changes, failure to lactate and low serum andprotein methionine levels in the dams. Thus, defectsthat may result as a result of inefficient transport ofmethionine (from fetal membranes) to the embryo,whether genetic or teratogen-induced, cannot be accu-rately modeled by maternal dietary deficiency alone.The model used in this laboratory may circumventthese problems since AVYS induces nutritional depri-vation of the embryo in vivo as well as in vitro withsimilar outcome (i.e., NTDs) without affecting mater-nal nutrition or general health. It will be of consider-able interest to determine whether methionine canovercome the malformations induced by AVYS in vivo.

In conclusion, these studies confirm those of previousinvestigations demonstrating the importance of methi-onine in embryonic development, especially with re-gard to NT closure. Supplementation of embryo cul-tures with methionine significantly improved theoutcome of AVYS-treated embryos and NT closure.This effect occurred without significantly altering pro-tein accretion by the embryo, indicating that the actionof methionine may occur through other metabolic roles.Further studies are necessary to determine the effect ofmethionine on AVYS-induced defects in vivo, the levelsof methionine required for maximum beneficial effectson development, and whether these levels are attain-able, and can be maintained, in utero.

LITERATURE CITED

Beckman DA, Pugarelli JE, Jensen M, Koszalka TR, Brent RL, LloydJB. 1990. Sources of amino acids for protein synthesis during orga-nogenesis in the rat I. Relative contributions of free amino acids andof proteins. Placenta 11:109–121.

Beckman DA, Ornoy A, Jensen M, Arnon J, Brent RL. 1991. Ultra-structure and function of the rat yolk sac: damage caused by tera-togenic anti-VYS serum and recovery. Teratology 44:181–192.

Beckman DA, Fawcett LB, Brent RL. 1997a. Developmental toxicity.In: Schardien JL, editor. Handbook of human toxicology. Boca Ra-ton, FL: CRC Press. p 1007–1084.

Beckman DA, Lloyd JB, Brent RL. 1997b. Investigations into mech-anisms of amino acid supply to the rat embryo using whole-embryoculture. Int J Dev Biol 41:315–318.

Brent RL. 1986. The complexities of solving the problem of humanmalformations. Clin Perinatol 13:491–503.

Brent RL, Beckman DA, Jensen M, Koszalka TR. 1990. Experimentalyolk sac dysfunction as a model for studying nutritional distur-bances in the embryo during early organogenesis. Teratology 41:405–414.

Brent RL, Fawcett LB. 1998. Nutritional studies of the embryo duringearly organogenesis with normal embryos and embryos exhibitingyolk sac dysfunction. J Pediatr 132:S6–S16.

Brent RL, Johnson AJ, Jensen M. 1971. The production of malforma-tions using tissue antisera VII. Yolk-sac antiserum. Teratology4:255–275.

Brown NA, Fabro S. 1981. Quantitation of rat embryonic developmentin vitro: a morphological scoring system. Teratology 24:65–78.

Bunduki V, Dommergues M, Zitton J. 1995. Maternal-folate statusand neural tube defects: a case control study. Biol Neonate 67:154–159.

Chambers BJ, Klein NW, Conrad SH, Ruppenthal GC, Sackett GP,Weeks BS, Kleinman H. 1995a. Reproduction and sera embryotox-icity after immunization of monkeys with the laminin peptidesYIGSR, RGD, and IKVAV. Proc Natl Acad Sci USA 92:6818–6822.

Chambers BJ, Klein NW, Nosel PG, Khairallah LH, Romanow JS.1995b. Methionine overcomes neural tube defects in rat embryoscultured on sera from laminin-immunized monkeys. J Nutr 125:1587–1599.

Coelho CND, Klein NW. 1990. Methionine and neural tube closure incultured rat embryos: morphological and biochemical analysis. Ter-atology 42:437–451.

Coelho CND, Weber JA, Klein NW, Daniels WG, Hoagland TA. 1989.Whole rat embryos require methionine for neural tube closure whencultured on cow serum. J Nut 119:1716–1725.

Czeizel AE, Dudas I. 1992. Prevention of the first occurrence of neuraltube defects by periconceptional vitamin supplementation. N Engl JMed 327:1832–1835.

Economides DL, Ferguson J, Mackenzie IZ. 1992. Folate and vitaminB12 concentrations in maternal and fetal blood, and amniotic fluidin second trimester pregnancies complicated by neural tube defects.Br J Obstet Gynaecol 99:23–25.

Ehlers K, Elmazar MMA, Nau H. 1996. Methionine reduces the val-proic acid-induced spina bifida in mice without altering valproicacid kinetics. J Nutr 126:67–75.

Essien FB. 1992. Maternal methionine supplementation promotes theremediation of axial defects in Axd mouse neural tube mutants.Teratology 45:205–212.

Essien FB, Wannberg SL. 1993. Methionine but not folinic acid orvitamin B-12 alters the frequency of neural tube defects in Axdmutant mice. J Nutr 123:27–34.

Farrari DA, Gilles PA, Klein NW, Nadler D, Weeks BS, Lammi-KeefeCJ, Hillman RE, Carey SW, Ying YK, Maier D, Olsen P, WempleDW, Greenstein R, Muechler EK, Miller RK, Mariona FG. 1994. Ratembryo development on human sera is related to numbers of pre-vious spontaneous abortions and nutritional factors. Am J ObstetGynecol 170:228–236.

Fawcett LB, Beckman DA, Fawcett PT, Brent RL, Lloyd JB. 1995.Pharmacokinetics of teratogenic antibodies administered to rats.Teratology 52:260–266.

Fawcett LB, Pugarelli JE, Brent RL. 1999. Prevention of neural tubedefects: effects of teratogenic antiserum and supplemental methio-nine on methionine uptake and utilization by rat embryos culturedin vitro. Teratology 59:403.

Flynne TJ, Friedman L, Black TN, Klein NW. 1987. Methionine andiron as growth factors for rat embryos cultured on canine serum. JExp Zool 244:319–324.

Foidart JM, Yaar M, Figueroa A, Wilk A, Brown KS, Liotta LA. 1983.Abortion in mice induced by intravenous injections of antibodies totype IV collagen or laminin. Am J Pathol 110:346–357.

Foidart JM, Hunt J, Lapiere CM, Nusgens B, de Rycker C, Bruweir M,Lambotte R, Bernard A, Mahieu P. 1986. Antibodies to laminin inpreeclampsia. Kidney Int 29:1050–1057.

Freeman SJ, Brent RL, Lloyd JB. 1982. The effect of teratogenicantiserum on yolk sac function in rat embryos cultured in vitro. JEmbryol Exp Morphol 71:63–74.

Fujinaga M, Baden JM. 1994. Methionine prevents nitrous oxide N2Oinduced teratogenicity in rat embryos grown in culture. Anesthesi-ology 81:184–189.

Jensen MJ, Koszalka TR, Brent RL. 1975. Production of congenitalmalformations using tissue antisera. XV. Reicharts membrane andvisceral yolk sac antisera. Dev Biol 42:1–12.

Kirke PN, Mills JL, Whitehead AS, Molloy A, Scott JM. 1996. Meth-ylene tetrahydrofolate reductase mutation and neural tube defects.Lancet 348:1037–1038.

Koszalka TR, Andrew CL, Brent RL, Beckman DA, Lloyd JB. 1994.Amino acid requirements in the early post-implantation rat concep-tus. Placenta 15:311–320.

Lerman S, Koszalka TR, Jensen M, Andrew CL, Beckman DA, BrentRL. 1986. In vitro studies on the effect of yolk sac antisera onfunctions of the visceral yolk sac. I. Pinocytosis and transport ofsmall molecules. Teratology 34:335–341.

Leung CCW. 1982. Isolation, partial characterization, and localiza-tion of a rat renal tubular glycoprotein antigen. Antibody-inducedbirth defects. J Exp Med 156:372–384.

Leung CCK, Lee C, Cheewatrakoolpong B, Hilton D. 1985. Abnormalembryonic development induced by antibodies to rat visceral yolk-

340 FAWCETT ET AL.

sac endoderm: isolation of the antigen and localization to microvil-lar membrane. Dev Biol 107:432–441.

Lloyd JB, Brent RL, Beckman DA. 1996. Sources of amino acids forprotein synthesis during early organogenesis in the rat 3. Methio-nine incorporation. Placenta 17:629–634.

Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. 1951. Protein mea-surements with the Folin phenol reagent. J Biol Chem 193:265–275.

Medical Research Council (MRC) Vitamin Study Research Group.1991. Prevention of neural tube defects: results of the MedicalResearch Council vitamin study. Lancet 338:131–137.

Milakofsky L, Hare TA, Miller JM, Vogel WH. 1984. Comparison ofamino acid levels in rat blood obtained by catheterization anddecapitation. Life Sci 34:1333–1340.

Moephuli SR, Klein NW, Baldwin MT, Krider HM. 1997. Effects ofmethionine on the cytoplasmic distribution of actin and tubulinduring neural tube closure. Proc Natl Acad Sci USA 94:543–548.

Moestrup SK, Kozyraki R, Kristiansen M, Kaysen JH, RasmussenHH, Brault D, Pontillon F, Goda FO, Christensen EI, HammondTG, Verroust PJ. 1998. The intrinsic factor-vitamin B12 receptorand target of teratogenic antibodies is a megalin-binding peripheralmembrane protein with homology to developmental proteins. J BiolChem 273:5235–5242.

New DAT. 1978. Whole embryo culture of mammalian embryos dur-ing organogenesis. Biol Rev 53:81–122.

New DAT, Brent RL. 1972. Effect of yolk sac antibody on rat embryosgrown in culture. J Embryol Exp Morphol 27:543–553.

Nosel PG, Klein NW. 1992. Methionine decreases the embryotoxicityof sodium valproate in the rat: in vivo and in vitro observations.Teratology 46:499–507.

Rowe PB, Kalaizis A. 1985. Serine metabolism in rat embryos under-going organogenesis. J Embryol Exp Morphol 87:137–144.

Scott JM, Kirke PN, Weir DG. 1990. The role of nutrition in neuraltube defects. Annu Rev Nutr 10:277–295.

Shaw GM, Velie EM, Schaffer DM. 1997. Is dietary intake of methi-onine associated with a reduction in risk for neural tube defect-affected pregnancies? Teratology 56:295–299.

Sims FH. 1951. Methionine and choline deficiency in the rat withspecial reference to the pregnant state. J Biol Chem 189:481–492.

Steegers-Theunissen RPM, Boers GHJ, Trijbels FJM, Eskes TKAB.1991. Neural tube defects and derangement of homocysteine me-tabolism. N Engl J Med 324:199–200.

Steegers-Theunissen RPM, Boers GHJ, Trijbels JMF, Finkelstein JD,Blom HJ, Thomas CMG, Borm GF, Wouters GAJ, Eskes TKAB.1994. Maternal hyperhomocyteinemia: a risk factor for neural tubedefects? Metabolism 43:1475–1480.

Szarfman A, Terranova VP, Rennard SI, Foidart JM, Lima MF, Schei-nman JI, Martin GR. 1982. Antibodies to laminin in Chagas’ dis-ease. J Exp Med 155:1161–1171.

Tagliamonte B, Benedetti PC, Semprinin ME, Gentili V. 1976. Effectof different protein and methionine level on the reproduction of therats. I. Pregnancy and perinatal survival of the progeny. NutritionReports International 14:633–638.

Van Aerts LAGJM, Klaasboer HH, Postma NS, Pertijs JCLM, Peere-boom JHJC, Eskes TKAB, Noordhook J. 1993. Stereospecific invitro embryotoxicity of L-homocyteine in pre- and post-implantationrodent embryos. Toxicol In Vitro 7:743–749.

Van Aerts LAGJM, Blom HJ, Deabreu RA, Trijbels FJM, EskesTKAB, Peereboom-Stegeman JHJC, Noordhoek J. 1994. Preventionof neural tube defects by and toxicity of L-homocysteine in culturedpreimplantation rat embryos. Teratology 50:348–360.

Weeks BS, Klein NW, Kleinman H, Fredrickson T, Sackett GP. 1989.Laminin immunized monkeys develop sera toxic to cultured ratembryos and fail to reproduce. Teratology 40:47–57.

EFFECTS OF METHIONINE SUPPLEMENT 341