/ . Embryol. exp. Morph. Vol. 33, 2, pp. 355-370, 1975
Printed in Great Britain
355
Limb development in mouse embryos:
protection against teratogenic effects of 6-diazo-5
oxo-L-norleucine (DON) in vivo and in vitro
By R. M. GREENE 1 AND D. M. KOCHHAR 1
From the Department of Anatomy, School of Medicine,
University of Virginia
SUMMARY
The glutamine analogue, 6-diazo-5-oxo-L-norleucine (DON), has been shown to inhibit
biosynthesis of purines and glycosaminoglycans, presumably by blocking the glutaminedependent steps in the biosynthetic pathways. The teratogenic potential of DON on the
developing mouse limb-bud in vivo and in vitro was studied in an attempt to discriminate
whether DON is exerting its teratogenic effect by interfering with glycosaminoglycan or
purine metabolism.
A single intramuscular injection of DON (0-5 mg/kg) to ICR/DUB mice on day 10 of
gestation resulted in 76% resorption, while fetuses surviving to day 17 exhibited growth
retardation, median cleft lip, and limb malformations. Concurrent administration of Lglutamine (250 mg/kg) provided no protection against resorption or malformations, while
5-aminoimidazolecarboxamide (AIC, 250 mg/kg) decreased the resorption rate to 34 %
without significantly altering the incidence of malformations. Injection of DON alone on
day 11 resulted in 87 % of fetuses exhibiting limb malformations, with only 2 % resorption.
Concurrent injection of AIC decreased the frequency of limb malformations to 32 %.
L-Glutamine, D-glucosamine, or inosinic acid were without any protective effect in vivo.
DON (5 /tg/ml medium) added in vitro to organ cultures of day 11 mouse limb-buds
caused all limbs to evidence cartilage abnormalities. In this system, either L-glutamine or
D-glucosamine (0-5 mg/ml medium) provided protection against DON effects while AIC
(0-5 mg/ml medium) offered no protection in vitro.
These data suggest that DON exerts its effects in vivo by interfering with purine metabolism
while in vitro its teratogenic action may be interruption of glycosaminoglycan biosynthesis.
This may reflect upon the relative importance of growth and differentiation to limb development in vivo and in vitro. These data infer that limb development in vitro relies more on the
differentiative process (differentiation of cartilage) than on growth, whereas limb development
in vivo is dependent, at this stage, to a greater extent on growth for normal phenotypic
expression.
INTRODUCTION
In the 1960s, the L-glutamine antagonists, azaserine (O-diazoacetyl-L-serine)
and DON (6-diazo-5-oxo-L-norleucine) were undergoing active clinical trials as
anticancer drugs (Duvall, 1960; Karnofsky, Golby & Li, 1967). However, due
to their lack of effectiveness in human malignancies relative to other agents
1
Authors' address: Department of Anatomy, School of Medicine, University of Virginia,
Charlottesville, Virginia 22901, U.S.A.
23
EMB 33
356
R. M. GREENE AND D. M. KOCHHAR
already available, interest in these compounds rapidly declined. Teratogenicity
of DON did not, however, go unnoticed (Dagg, Karnofsky, Lacon & Roddy,
1956; Friedman, 1957; Murphy, Dagg & Karnofsky, 1957; Thiersch, 1957). It
was found that DON produced cleft palate and a variety of skeletal defects.
Recently, this agent has once again gained favor as a tool with which to investigate mechanisms of abnormal development (Aydelotte & Kochhar, 1972;
Pratt, Goggins, Wilk & King, 1973).
Since it is known that DON acts as an antagonist to glutamine (Levenberg,
Melnick & Buchanan, 1957; Anderson & Law, 1960; Chu & Henderson, 1972),
it was proposed in this study to investigate the role of some metabolic pathways
in which glutamine is involved, for discovering possible sites of teratogenic
action by DON.
The teratogenic potential of DON on the developing mouse limb-bud has
been studied in an attempt to discriminate whether DON is exerting its teratogenic effect by interfering with glutamine-dependent steps in glycosaminoglycan
or purine metabolism. This was accomplished by supplementation of DON
administration with metabolic products of glutamine-dependent steps in these
bio synthetic pathways.
MATERIALS AND METHODS
ICR/DUB mice (Flow Laboratories, Dublin, Virginia), fed Purina mouse
chow and tap water ad libitum, were mated between 8.00 a.m. and 12.00 noon in
order to closely estimate the time of fertilization of eggs. The presence of a
vaginal plug immediately afterwards was regarded as evidence of mating. 10.00
a.m. of that morning was regarded as the beginning of day zero of gestation
(first day of development).
Days 10 through 12 of gestation were chosen as times at which to administer
experimental agents, since they include the susceptible periods of organogenesis
of the limb (Kochhar, 1973) in this strain of mice. All experimental agents were
administered via intramuscular injection between 10.00 a.m. and 12 noon.
Pregnant females were injected with a single dose of 1 mg/kg or 0-5 mg/kg
6-diazo-5-oxo-L-norleucine (DON; obtained from National Cancer Institute,
Bethesda, Md.) dissolved in distilled water, on days 10, 11 or 12 of gestation.
Pregnant females were injected with either DON, or DON and a concurrent
intramuscular injection in the opposite thigh of either 125 mg/kg or 250 mg/kg
of one of the following metabolites (all obtained from Sigma Chemical Co.)
dissolved in distilled water; L-glutamine; D ( + ) glucosamine hydrochloride;
5-amino-4-imidazole carboxamide hydrochloride; inosine-5'-monophosphoric
acid (inosinic acid). Control animals were injected with 0-1 ml distilled water.
Animals were killed on day 17 of gestation. Fetuses were examined for external malformations and either fixed in Bouin's fluid for 3 days and stored in
70 % ethyl alcohol or fixed in 95 % alcohol for 3 days, cleared and stained with
alizarin Red S (Matheson. Coleman & Bell Inc.) (Kochhar, 1973).
DON and limb development
357
Another group of untreated pregnant females was used as a source of fetal
material for limb-bud organ cultures. Pregnant females, carrying litters in their
12th gestational day (day 11), were killed by cervical dislocation. Uterine horns
were aseptically removed and placed in a sterile Petri dish containing Tyrode's
solution. Limb-buds were excised and explanted in organ culture by a method
described previously (Kochhar & Aydelotte, 1974). Limb-buds were placed on
ultra-thin Millipore filter, supported on a stainless steel grid in a medium consisting of 75 % BGJ (Difco) and 25 % fetal calf serum (Flow Laboratories). The
medium was supplemented with 150/ig/ml ascorbic acid, 10//g/ml streptomycin, and 5 /<g/ml penicillin G. Somite number of donor embryos ranged
between 42-46 somites.
A total of 63 pairs of limb-buds were organ cultured. DON was added to some
cultures at a concentration of 5/^g/ml media (Aydelotte & Kochhar, 1972).
Other cultures were maintained on media containing both DON (5 /*g/ml) and
either 0-5 mg 5-amino-4-imidazole carboxamide hydrochloride/ml media, 0-5 mg
L-glutamine/ml or 0-5 mg D ( + ) glucosamine hydrochloride/ml. Fresh medium
was supplied on the third day of culture. Cultures were maintained for a total of
6 days at 37 °C and flushed with fresh 5 % CO2/95 % air daily.
At the end of the culture period the limbs were first rinsed in saline. Some
limbs were then fixed in Bouin's fluid overnight and stained with toluidine blue
for cartilage, while the majority were placed in cold acetone for at least 2 h.
This tissue was prepared for DNA and protein determinations using a modification of the method used by Ellison & Lash (1971). Acetone-dried limbs were
transferred to an ice cold 10 % perchloric acid (PCA) solution, thoroughly
homogenized and stored overnight at 4 °C. The homogenate was then centrifuged at 4 °C for 30 min at 2500 rev/min. The resultant supernatant was discarded while the acid-insoluble pellet was washed again with 10 % PCA solution.
The precipitate was then resuspended by vigorous shaking in 2-0 ml of 6 %
PCA and hydrolyzed for 30 min at 90 °C. The hydrolyzed tissue suspension was
centrifuged for 15 min at 2500 rev/min and the supernatant saved as the DNA
fraction. Another 2-0 ml of hot 6 % PCA was added to the precipitate, vortexed
and centrifuged for 15 min at 2500 rev/min. The resulting supernatant was
combined with the previous supernatant, and these were immediately processed
for DNA determination by the method of Burton (1956). Precipitates were
washed once with 1-0 ml of distilled water to wash out any excess PCA, and then
suspended in 1-0 ml of 1 N-NaOH and processed for protein determination by
the method of Lowry, Rosebrough, Farr & Randall (1951).
RESULTS
A single intramuscular injection of DON (0-5 mg/kg) to mice on day 10 of
gestation resulted in 76 % foetal resorption, while fetuses surviving to day 17 of
gestation exhibited growth retardation, median cleft lip (Figs. 1, 2) and limb
23-2
358
R. M. GREENE AND D. M. KOCHHAR
F I G U R E S 1 AND 2
Frontal view of a mouse embryo head on day 17 of gestation after maternal injection
of 0-5 mg DON/kg on day 10 of gestation. Note the median cleft lip. Bouin
Fixation, x 12-5.
9
3
DON, 0-5 + glucosamine, 250
DON, 0-5 + L-glutamine, 250
3
1
2
DON, 1-0 + AIC, 250
DON, 0-5
DON, 10
45
8
DON, 0-5 + inosinic acid.,250
DON, 10
3
DON, 0-5 + AIC, 250
68
100
DON, 0-5 + L-glutamine, 250
DON, 10
2
76
34
DON, 0-5
DON, 0 5 + AIC, 250
DON, 0 5
(%)
Dosage (mg/kg)
Resorption
55
25
32
22
35
28
20
64
45
29
—
33
19
Number of
surviving
fetuses
Day 12
None
None
58
82
80
93
60
32
87
Day 11
20
—
25
26
Day 10
(%)
Total limb
malformations
—
—
24 ectrodactyly
27 syndactyly
37 brachydactyly
8 ectrodactyly
1 syndactyly
20 ectrodactyly
45 syndactyly
47 ectrodactyly
58 syndactyly
42 brachydactyly
27 ectrodactyly
21 syndactyly
32 syndactyly
41 ectrodactyly
28 syndactyly
13 ectrodactyly
—
—
5 polydactyly
—
(%)
Forelimb
malformations
—
—
25 kinky tail
12 ectrodactyly
37 syndactyly
32 syndactyly
5 ectrodactyly
25 syndactyly
None
None
86 kinky tail
9 cleft palate
16 cleft palate
100 kinky tail
—
1 kinky tail
10 ectrodactyly
42 syndactyly
5 ectrodactyly
23 syndactyly
25 micromelia
33 ectrodactyly
49 syndactyly
47 kinky tail
—
—
20 polydactyly
38 median cleft lip
42 cleft palate
31 kinky tail
27 median cleft lip
(%)
Other
malformations
25 polydactyly
10 polydactyly
(%)
Hind limb
malformations
Table 1. Occurrence of resorption and malformation in mouse fetuses after injection of 6-diazo-5-oxo-norleucine
(DON), or DON+metabolite on days 10, 11 or 12 of gestation
?-*
1
^^
o
3'
<^-
Si
o
o
360
R. M. GREENE AND D. M. KOCHHAR
DON and limb development
361
malformations (Table 1). Concurrent administration of L-glutamine (250 mg/kg)
provided no protection against resorption or malformation (Table 1), while
5-amino-4-imidazole carboxamide hydrochloride (AIC, 250 mg/kg) decreased
the resorption rate to 34 % without significantly altering the incidence of malformations (Table 1). Administration of 1 mg DON/kg maternal body weight
proved highly embryolethal, resulting in 100 % frequency of resorption (Table 1).
Intramuscular injection of DON (0-5 mg/kg) alone on day 11 of gestation
resulted in 87 % of the fetuses exhibiting various fore- and hind limb malformations with only a 2 % frequency of resorption (Table 1; Figs. 3, 4, 5, 6). Concurrent injection of AIC (250 mg/kg) decreased the frequency of limb
malformations to 32 %. L-Glutamine and D-glucosamine (250 mg/kg) were
without any protective effect; inosinic acid (250 mg/kg) afforded only partial
protection against limb defects (Table 1; Fig. 7).
A larger dose of DON (1 -0 mg/kg) on day 11 of gestation proved to be more
embryolethal than 0-5 mg/kg, increasing the resorption frequency to 45 %.
Concurrent AIC (250 mg/kg) administration provided protection against resorption (Table 1) but was not as effective in preventing limb defects in surviving
young as it was against a lower dosage (0-5 mg/kg) of DON.
Administered on day 12 of gestation, both 0-5 mg/kg and 1-0 mg/kg DON
failed to elicit any malformations in 80 recovered fetuses examined on day 17
(Table 1). Percentages of resorption was no higher than control levels.
Organ-cultured fetal limb-buds exhibited pronounced growth over a 6-day
in vitro period, as determined by increased protein and DNA content (Fig. 8).
Production of glycosaminoglycans was also evident, as seen by toluidine blue
staining of cartilagenous skeletal rudiments (Fig. 9). Limb-buds cultured, how-
FlGURES 3-6
Fig. 3. The hind limb on the left is from a mouse fetus on day 17 of gestation after
maternal injection of 0-5 mg DON/kg on day 11 of gestation. Note hind limb
syndactyly. The hind limb on the right is from a control animal. Bouin Fixation,
x 12-5.
Fig. 4. Forelimbs of a mouse fetus on day 17 of gestation after maternal injection of
0-5 mg DON/kg on day 11 of gestation. Note forelimb ectrodactyly. The forelimb
on the right is from a control animal. Bouin Fixation, x 12-5.
Fig. 5. Hind limbs from two day 17 mouse fetuses. The limb on the left is a control
from a carrier-injected mother. On the right is a fetal hind limb after maternal
injection of 0-5 mg DON/kg on day 11 of gestation. Note absence of long bone
ossification as well as limb deformation. Fixed in 95 % ethyl alcohol, cleared in 1 %
KOH and stained with alizarin red. x 12-5.
Fig. 6. Forelimbs from two day 17 mouse fetuses. The limb on the left is a control
from a can ier-injected mother. On the right is a fetal forelimb after maternal
injection of 0-5 mg DON/kg on day 11 of gestation. Note fusion of metacarpal
ossification centers indicating a syndactylous defect. Long bone ossification appears
normal. Fixed in 95 % ethyl alcohol, cleared in 1 % KOH and stained with alizarin
red. xl2-5.
362
R. M. GREENE AND D. M. KOCHHAR
lOO—i
80 —
.2
60
~
I
E
40 —
S
j
—'
20 —
DON
(1)
DON
DON
DON
DON
glutamine
glucosamine
I-5-P
AIC
(2)
(3)
(4)
(5)
Treatment on day 11 of gestation
Fig. 7. Percentage limb malformations resulting from DON or DON + metabolite
intramuscular injection on day 11 of gestation. (1) DON, 0-5 mg/kg maternal body
weight. (2) L-glutamine, 250 mg/kg. In opposite thigh, 0-5 mg/kg DON. (3)
D( + )glucosamine hydrochloride, 250 mg/kg. In opposite thigh, 0-5 mg/kg DON.
(4) Inosine-5'-monophosphoric acid, 250 mg/kg. In opposite thigh, 0-5 mg/kg DON.
(5) 5-amino-4-imidazole carboxamide HC1, 250 mg/kg. In opposite thigh, 05 mg/kg
DON.
ever, on medium containing 5 jug DON/ml for a period of 6 days exhibited a
drastic decrease in glycosaminoglycan production, evidenced by virtual absence
of toluidine blue metachromasia (Fig. 10). These DON-treated limbs were
visibly smaller than controls, an observation quantitatively verified by appreciably lower protein and DNA content after 6 days in culture (Fig. 8).
Cultures maintained on media containing both 5 /*g DON/ml and either 0-5
mg L-glutamine/ml or 0-5 mg D ( + ) glucosamine hydrochloride/ml demonstrated marked protection against the inhibitory effects of DON by exhibiting
pronounced cartilagenous skeletal rudiments (Figs. 11, 12). Although the degree
of cartilagenous differentiation was not comparable to controls (Fig. 9), development of skeletal rudiments was far better than in limb cultures exposed to DON
alone. Addition of 0-5 mg/ml of 5-amino-4-imidazole carboxamide hydrochloride to the culture media did not appear to protect against the effects of
DON in vitro.
363
DON and limb development
200 —|
145-3
+ 19-4
180 —
18 —
160 —
16 —
140 —
14 —
91 4
+ 12-9
J 120 —
I ioo —
—
D.
«» 80 -
16-6
+ 2-9
20 —
7-7
+ 1-95
Q
61-53
+ 6-56
_
6
60 —
4 —I
5-94
±1-2
Si
40 —
2Q —
0
Day
Day
0
6
Control
(1)
(2)
Day
0
Day
6
DON
(3)
(1)
Day
6
Day
6
Control
DON
(3)
(2)
Fig. 8. Comparison of amounts (mean±s.E.) of protein and DNA in cultured fetal
mouse limb-buds after 6days in culture. (1) Amount of protein or DNA at beginning
of culture period. (2) Amount of protein or DNA after 6 days on control culture
media. (3) Amount of protein or DNA after 6 days on culture media supplemented
with 5 jug DON/ml media.
DISCUSSION
It was once thought that the L-giutamine antagonists, azaserine (0-diazoacetyl-L-serine) and DON (6-diazo-5-oxo-L-norleucine), would prove highly
successful in the treatment of human malignancies, since they were shown to
inhibit the growth of rodent tumors (Burchenal & Dagg, 1956) and to be effective in the treatment of choriocarcinoma in man (Karnofsky et al. 1967). The
instances of disease regression, however, as indicated by decrease in size of
cancerous lesions, were relatively few and transient (Duvall, 1960). Consequently, interest in the chemotherapeutic value of these agents waned.
The teratogenic potential of these agents, especially DON, has also been well
documented in a wide variety of laboratory animals. DON, 50 times more
teratogenically potent than azaserine (Dagg et al. 1956; Murphy et al. 1957),
has been shown to produce skeletal defects and growth inhibition in developing
chicks (Dagg et al. 1956), mice (Jackson, Robson & Wander, 1959) and rats
(Thiersch, 1957). Cleft palate has also been induced by DON administration in
the rat (Murphy, 1960), dog (Friedman, 1957), and mouse (Pratt et al 1973;
R. M. Greene and D. M. Kochhar, unpublished observations).
The general mechanisms of DON action at the metabolic level are well known.
364
R. M. GREENE AND D. M. KOCHHAR
DON and limb development
365
As an L-glutamine antagonist, DON serves to interrupt and block glutaminedependent steps in several biosynthetic pathways.
Glutamine serves as the amino donor in the conversion of fructoses-phosphate to glucosamine-6-phosphate, an important step in the biosynthetic
production of glycosaminoglycans (Fig. 13). DON is known to prevent formation of glucosamine from glucose (Ghosh, Blumenthal, Davidson & Roseman,
1960) and presumably eventually inhibit the synthesis of glycosaminoglycans
(Fig. 13). Preparations of minced embryonic cartilage have been shown to
catalyze in vitro incorporation of 14C-serine, 14C-acetate and inorganic 35S
into proteoglycans (Telser, Robinson & Dorfman, 1965). DON was shown to
inhibit incorporation of all three components into proteoglycans. This inhibition
was decreased by the addition of glucosamine into the system. Incorporation
of hexosamine into glycosaminoglycans was not affected by DON. These data
are consistent with the postulated site of action of DON being the glutaminedependent transamination schematized in Fig. 13.
DON-induced limb defects were refractory to protection with administration
of 250 mg/kg glucosamine. This would seem to indicate that the biosynthesis of
glycosaminoglycans, albeit of great importance to the developing limb-bud, may
not be the metabolic site of DON-induced teratogenesis in the fetus in vivo.
The metabolic pathway involved in purine biosynthesis also contains several
glutamine-dependent transamination reactions (Hartman & Buchanan, 1959)
(Fig. 14). DON has been shown to inhibit the conversion of formylglycineamide
ribotide to formylglycineamidine ribotide in a number of cell-free systems
(Levenberg et ah 1957; Moore & LePage, 1957; Hartman & Buchanan, 1959;
Anderson & Law, 1960). Chu & Henderson (1972) demonstrated that DON
irreversibly inactivated the phosphoribosylformylglycineamidine synthetase of
Ehrlich ascites tumor cells in the presence of glutamine, a reaction shown to
F I G U R E S 9-12
Fig. 9. Photomicrograph showing two fetal mouse forelimb buds after 6 days in
organ culture grown on control media. Bouin Fixation and toluidine blue staining.
xl2-5.
Fig. 10. Photomicrograph showing two fetal mouse limb-buds after 6 days in organ
culture. On the left is a forelimb grown on control media. On the right is a forelimb
grown on media containing 5 /tg DON/ml media. Note the absence of toluidine
blue staining in the limb grown in presence of DON. Bouin Fixation and
toluidine blue staining, x 12-5.
Fig. 11. Organ cultured fetal mouse limb-buds. On the left is a forelimb grown on
media containing 5 fig DON/ml media. On the right is a forelimb grown on media
containing 5 /<g DON/ml media and 0-5 mg glutamine/ml media. Bouin Fixation
and toluidine blue staining, x 12-5.
Fig. 12. Organ cultured fetal mouse limb-buds. On the left is a forelimb grown on
media containing 5/tg DON/ml media. On the right is a forelimb grown in media
containing 5 /.eg DON/ml media and 0-5 mgD(+)glucosamine hydrochloride/ml
media. Bouin Fixation and toluidine blue staining, x 12-5.
O
ADP
NH 2 CCCH 3
OH
OH
Uridine diphosphate glucosamine (UDPG)
CH,OH
Phosphaglucomutase
Phosplioglucoisomerase
CH,
Fig. 13. Scheme of uridine diphosphate glucosamine (TJDFG) production necessary for normal glycosaminoglycan
(GAG) synthesis. Note how DON, by blocking the conversion of fructose-6-P to gluccsamine-6-P, could obstruct
GAG biosynthesis. The UDPG molecule acts as a. carrier of sugar residues. The glucosamine moiety is transferred, via
a transferase enzyme, to a growing glycosaminoglycan molecule.
Glucosamine-6-P
OH
/V-Acetylglucosamine
OH
Glucose-6-P
o.
.V-Acctylglucosaminc —^ OH
Acetvl CoA
OH
NH2COCH3
M
CH,OP- +glutamic acid
O
DON OH OH
\
COOH
Glutaminc
/
NH,
I
OH
(CH,) 2
c
CH,OH
X
X
o
o
p
d
Z
w
m
m
o
ON
OS
HPO 4 3 -
\
>THFA
NH2
I
/>CH2-NH-CHO
\NH
ATP
+
11
O
OH OH
5-Aminoimidazoie ribotide
CH2OPO
C —N
AIC
OH OH
NH
coo-
DON
—»
Glutamate + ADP-f HPO4
Inosinic acid
Fig. 14. Biosynthesis of inosinic acid en route to purine biosynthesis (after Hartman and Buchanan, 1959).
OH OH
Formylglycineamidine ribotide
ADP +
HPO4
NH-CHO
Glutamine-f ATP+H2O
CH-N
NH, „ n
S
CH
Q C
^ o
y C-CH 2 CH2OPO32"/ \ NH
/
+ /
\
OH OH
5-Phosphoribosylamine
Glutamine-f H2O
Glutamate + HP2O73Glycine + ATP ADP + HPO4
Q
V
J*
CH2OPO3y \NH 2
^
y*
^^-^//^^
>
v
N
<£
_^>—^1
^
M 2+
DON
\
7
S
OH OH
Formylglycineamide ribotide
\NH
O
o
11
/ C-CH 2 NH 3
OH OH
CH2OPO32V
CH,OPO3
o
o
OH OH
5-Phosphoribosylpyrophosphate
AMP
^CH
N5, N10-Anhydroformyl-THFA + H2O
O
/
\
f
OH OH
Ribose-5-P
CH2OPO3 /
ATP
V V
GMP
AMP
ON
I"
2;
o
o
368
R. M. GREENE AND D. M. KOCHHAR
exhibit kinetics characteristic of competitive inhibition. When, however, DON
was incubated with formylglycineamidine synthetase prior to addition of substrate, inhibition with respect to glutamine was non-competitive. Therefore,
DON was not just competing for enzyme binding sites with glutamine, but
eliminating the enzyme from the system.
That this might be the case in our in vivo studies was suggested by the fact
that glutamine (250 mg/kg), when administered concurrently with DON, was
ineffectual in preventing DON-induced limb malformations (Table 1). DONinduced cartilagenous abnormalities in mouse limb-buds grown in vitro were,
however, prevented by the addition of either 0-5 mg L-glutamine or 0-5 mg D( + )
glucosamine hydrochloride per ml of culture media. The reasons for this
apparent discrepancy between in vivo and in vitro results may lie in differences in
L-glutamine availability to fetal tissue in the two systems.
These data suggest that although DON acts to non-competitively eliminate an
enzyme necessary for normal purine biosynthesis, this may not be the case for
DON action in the glycosaminoglycan biosynthetic pathway (Fig. 13). L-Glutamine's protection against DON action on limb development in vitro is suggestive
of competition between DON and glutamine for the transaminase involved in
the synthesis of glucosamine-6-P (Fig. 13).
Since DON-induced embryolethality and malformations have been shown to
be prevented by concurrent administration of adenine and/or guanine (Dagg
et al. 1956; Thiersch, 1957; Murphy, 1959, 1960), it was thought that DONinduced embryolethality and malformation in our mice might be attributable to
an inhibition of purine biosynthesis.
If DON exerted its teratogenic and embryolethal effect by inhibiting glutamine-dependent steps in the purine biosynthetic pathway (Fig. 14), supplemental
administration of the products of these steps should effectively bypass the
metabolic block imposed by DON. Maternal injection of AIC (250 mg/kg) did
indeed reduce DON-induced embryolethality and limb malformations, while
inosinic acid provided a lesser degree of protection against DON-induced limb
defects (Table 1).
These results therefore suggest that DON-induced embryolethality and teratogenicity in vivo may be due to a DON block of glutamine-dependent steps in
purine biosynthesis. Indeed, quantitative measurements of limb-buds grown
in vitro in the presence of DON exhibit pronounced inhibition of DNA synthesis
(Fig. 8).
An alternative explanation for the apparently discrepant results between in
vivo and in vitro data may lie in the relative importance of growth and differentiation to limb development in vitro and in vivo. When a limb-bud is isolated
from a day 11 embryo and explanted in culture, it grows from an initial value of
about 60 fig total protein to only about 140 /tg after 6 days in culture (Fig. 8).
Levels of hydroxyproline, measured as a component of collagen to quantitate
the process of chondrogenesis, rise from an initial value of 0-175 /tg/limb at the
DON and limb development
369
beginning of culture to about 1-0/tg/limb after 6 days of culture (Kochhar &
Vest, 1974). The nucleic acid antimetabolite DON was effective to a similar
extent (about 50 %) in inhibition of growth (total protein) and differentiation
(total collagen). It may therefore be inferred from these data that limb development in vitro relies more on the differentiative process (differentiation of
cartilage) than on growth, whereas limb development in vivo is dependent to a
greater extent on growth for normal phenotypic expression. The fact that
glucosamine counteracted the deleterious effects of DON in vitro while aminoimidazole carboxamide provided protection against DON in vivo seems to
support this hypothesis.
Supported by NIH grant HD06550. The authors wish to acknowledge Dr M. Aydelotte
and Dr R. M. Pratt whose comments and suggestions, in personal discussions, greatly
contributed to this manuscript.
REFERENCES
ANDERSON,
E. P. &
LAW,
L, W. (1960). Biochemistry of Cancer. A. Rev. Biochem. 29,
577-608.
M. B. & KOCHHAR, D. M. (1972). The effects of 6-diazo-5-oxo-L-norleucine
(DON) on the development of mouse limb buds in vitro. Teratology 5, 249.
AYDELOTTE,
BURCHENAL, J. H. & DAGG, M. K. (1956). Effects of 6-diazo-5-oxo-L-norleucine and 2-
ethylaminothiadiazole on strains of transplanted mouse leukemia. Proc. Am. Ass. Cancer
Res. 2, 97-98.
BURTON, K. (1956). A study of the conditions and mechanism of the diphenylamine reaction
for the colorimetric estimation of deoxyribonucleic acid. Biochem. J. 62, 315-322.
CHU, S. Y. & HENDERSON, J. F. (1972). Inhibition of the phosphoribosylformylglycineamide
synthetase of Ehilich ascites tumor cells by glutamine analogues. Biochem. Pharmac. 21,
401-406.
DAGG, C. P., KARNOFSKY, D. A., LACON, C. & RODDY, J. (1956). Comparative effects of
6-diazo-5-oxo-L-norleucine and O-diazoacetyl-L-serine on the chick embryo. Proc. Am. Ass.
Cancer Res. 2, 101.
DUVALL, L. R. (1960). New agent data summaries - Azaserine, 6-diazo-5-oxo-L-norleucine.
Cancer Chemother. Rep. 7, 86-98.
ELLISON, M. & LASH, J. W. (1971). Environmental enhancement of in vitro chondrogenesis.
Devi Biol. 26, 486-496.
FRIEDMAN, M. H. (1957). The effect of O-diazoacetyl-L-serine (azaserine) on the pregnancy of
the d o g - a preliminary report. /. Am. vet. med. Ass. 130, 159-162.
GHOSH, S., BLUMENTHAL, H. J., DAVIDSON, E. A. & ROSEMAN, S. (1960). Glucosamine
metabolism. V. Enzymatic synthesis of glucosamine-6-phosphate. J. biol. Chem. 235,
1265-1272.
HARTMAN, S. C. & BUCHANAN, J. M. (1959). Nucleic acids, purines, pyrimidines (nucleotide
synthesis). A. Rev. Biochem. 28, 365-410.
JACKSON, D., ROBSON, J. M. & WANDER, A. C. E. (1959). The effect of 6-diazo-5-oxo-Lnorleucine (DON) on pregnancy. /. Endocr. 18, 204-207.
KARNOFSKY, D. A., GOLBY, R. B. & Li, M. C. (1967). Remissions induced in patients with
trophoblastic tumors by 6-diazo-5-oxo-L-norleucine (DON). UICC Monogr. Ser. 1,
1-5.
KOCHHAR, D. M. (1973). Limb development in mouse embryos. I. Analysis of teratogenic
effects of retinoic acid. Teratology 7, 289-298.
KOCHHAR, D. M. & AYDELOTTE, M. B. (1974). Susceptible stages and abnormal morphogenesis in the developing mouse limb, analyzed in organ culture after transplacental
exposure to vitamin A (retinoic acid). /. Embryol. exp. Morph. 31, 721-734.
370
R. M. GREENE AND D. M. K O C H H A R
D. M. & VEST, T. K. (1974). Comparative effects of teratogenic agents on collagen
biosynthesis in organ cultures of mouse limb buds. Teratology 10, 313-314.
LEVENBERG, B., MELNICK, I. & BUCHANAN, J. M. (1957). Biosynthesis of the purines. XV.
The effect of aza-L-serine and 6-diazo-5-oxo-L-norleucine on inosinic acid biosynthesis
de novo. J. biol. Chem. 225, 163-176.
LOWRY, O. H., ROSEBROUGH, N. J., FARR, A. L. & RANDALL, R. J. (1951). Protein measurement with the Folin phenol reagent. /. biol. Chem. 193, 265-275.
MOORE, E. C. & LEPAGE, G. A. (1957). In vivo sensitivity of normal and neoplastic mouse
tissues to azaserine. Cancer Res. 17, 804.
MURPHY, M. L. (1959). A comparison of the teratogenic effects of five polyfunctional
alkylating agents on the rat fetus. Pediatrics, Springfield 23, 23.1-244.
MURPHY, M. L. (1960). Teratogenic effects of tumor-inhibiting chemicals in the fetal rat. In
Ciba Foundation Symposium on Congenital Malformations (ed. G. E. W. Wolstenholme
& C. M. O'Connor), pp. 78-114. New York: Little, Brown & Co.
MURPHY, M. L., DAGG, C. P. & KARNOFSKY, D. A. (1957). Comparison of teratogenic
chemicals in the rat and chick embryos. Pediatrics, Springfield 19, 701-714.
PRATT, R. M., GOGGINS, J. F., WILK, A. L. & KING, C. T. G. (1973). Acid mucopolysaccharide synthesis in the secondary palate of the developing rat at the time of rotation
and fusion. Devi Biol. 32, 230-237.
TELSER, A., ROBINSON, H. C. & DORFMAN, A. (1965). The biosynthesis of chondroitin-sulfate
protein complex. Proc. natn. Acad. Sci. U.S.A. 54, 912-919.
THIERSCH, J. B. (1957). Effect of 6-diazo-5-oxo-L-norleucine (DON) on the rat litter in utero.
Proc. Soc. exp. Biol. Med. 94, 33-35.
KOCHHAR,
(Received 3 June 1974)