-
MATERNAL-ZYGOTIC LETHAL INTERACTIONS IN DROSOPHILA MELANOGASTER:
THE EFFECTS OF DEFICIENCIES IN THE
ZESTE-WHITE REGION OF THE X CHROMOSOME1
LEONARD G. ROBBINS
Genetics Program and Department of Zoology, Colleges of Natural
ScLnces and Osteopathic Medicine, Michigan State University, East
Lansing, Michigan 48824
Manuscript received April 12, 1980 Revised copy received June
16, 1980
ABSTRACT
The possibility that essential loci in the zeste-white region of
the Dro- sophila mlanogaster X chromosome are expressed both
maternally and zy- gotically has been tested. Maternal gene
activity was varied by altering gene dose, and zygotic gene
activity was manipulated by use of position-effect variegation of a
duplication. Viability is affected when both maternal and zygotic
gene activity are reduced, but not when either maternal or zygotic
gene activity is normal. Tests of a set of overlapping deficiencies
demonstrate that at least three sections of the zeste-white region
yield maternal zygotic lethal interactions. Single-cistron
mutations at two loci in one of these seg- ments have been tested,
and maternal heterozygosity for mutations at both loci give lethal
responses of mutant-duplication zygotes. Thus, at least four of the
13 essential functions coded in the zeste-white region are active
both ma- ternally and zygotically, suggesting that a substantial
fraction of the genome may function at both stages. The normal
survival of zygotes when either maternal gene expression or zygotic
gene expression is normal, and their in- viability when both are
depressed, suggests that a developmental stage exists when
maternally determined functions and zygotically coded functions are
both in use.
HERE is substantial genetic and molecular evidence for the
developmental importance of maternal gene action (for reviews see
DAVIDSON 1976; KING
and MOHLER 1975). Are maternally active genes a restricted
subset of genes? Do genes that determine maternal functions also
determine zygotic functions? What fraction of the genome determines
functions required both maternally and zygotically? There is some
evidence that genes with both maternal and zygotic function are not
uncommon. GALAU et al. (1976) have presented evidence that, in the
sea urchin, the subsets of polysomal RNA information found at later
embryonic stages are all included within the oocyte RNA sequences.
RIPOLL (1977) and RIPOLL and GARCIA-BELLIDO (1979) have shown that
a majority of Drosophila lethal mutants that are both
cell-autonomous and cell-lethal never- theless survive till
hatching of first-instar larvae. Their interpretation is that these
embryos survive because maternally specified equivalent functions
are
1 Research supported by National Science Foundation Grant PCM
79-01824.
Genetics 96: 187-200 September, 1980
-
188 L. G . ROBBINS
active during embryogenesis. In addition, GARCIA-BELLIDO and
Moscoso DEL PRADO ( 1979) have suggested that maternal-zygotic
functions are widespread since offspring of deficiency-heterozygous
females often have reduced viability, while genetically identical
embryos fathered by deficiency-heterozygous males do not. Whether
the defect that kills off spring of deficiency-heterozygous mothers
relatively early in embryogenesis is the same as that which kills
deficiency- homozygous embryos at a later stage was not
demonstrated, nor was it demon- strated that reduced maternal gene
dose, rather than some property of the pater- nally transmitted
normal homolog, causes the nonreciprocal lethality. That pa-
ternally and maternally transmitted chromosomes differ has been
demonstrated by the recovery of mutants that differentially affect
the behavior of paternally and maternally derived chromosomes
(BAKER 1975; HALL, GELBART and KAN- KEL 1976) and has been raised
in various reports of parental effects on variega- tion ( SPOFFORD
1976).
A number of Drosophila genes are known that have both maternal
and zy- gotic functions. In most instances these mutants were
detected as the result of chance observations, and the conditions
that expose the maternal-zygotic inter- actions are not of general
utility for detecting maternal-zygotic interactions at other loci.
Lethals, which are common and easily detected, would be the mu-
tants of choice for asking about genes that act both maternally and
zygotically but for the fact that an unconditional zygotic lethal
cannot be tested for maternal effects, at least not as a
homozygote. The recovery of temperature-sensitive ma- ternal and
zygotic lethal mutations ( SHEARN, HERSPERGER and HERSPERGER 1978)
offers one approach to uncovering these functions. Another
approach, and one that is well suited to asking the general
questions posed above, is to select a wel!-defined region of the
genome in which all essential loci are known, where a large
collection of lethal mutants already exists, and to devise a set of
condi- tions that allows testing for maternal effects.
This approach is similar to that used by GARCIA-BELLIDO and
Moscoso DEL PRADO (1979) in that maternal effects of
deficiency-heterozygous females were initially examined. However,
use of duplications corresponding to the deficien- cies, and
modulation of gene activity in those duplications by position-eff
ect variegation, allows demonstration that the maternal-effect
lethality is, in fact, a consequence of reduced gene dose and is
restricted to embryos that do not have fully active complements for
the region. Demonstration of similar maternal- zygotic interactions
for several single-cistron mutants confirms that individual gene
functions are active during both oogenesis and zygotic
development.
CROSSES A N D RESULTS
The deficiencies, duplications and zeste-white regioE mutations
used in these experiments were supplied by B. JUDD. Descriptions
may be found in LINDSLEY and GRELL (1968), JUDD, SHEN and KAUFMAN
(1972), KAUFMAN et al. (1975) or SHANNON et al. (1972). Except as
noted, descriptions of other markers and
-
MATERNAL-ZYGOTIC LETHAL INTERACTIONS 189
chromosomes used may be found in LINDSLEY and GRELL (1968). All
crosses were done on cornmeal, molasses, brewers' yeast medium at
25", except as noted.
Exposing maternal-zygotic gene interactions: Homozygous females
cannot be used to test lethal mutants for maternal effects. Use of
heterozygous females re- quires exposing effects of reduced, rather
than absent, maternal gene activity. For a gene that acts both
maternally and zygotically, reduced maternal gene activity may be
more apparent if the same gene function is depressed in the zygote
as well. Thus, in order to determine whether the zeste-white
interval contains essential functions that are expressed either
maternally or zygotically, or both maternally and zygotically, the
interaction between reduced levels of maternal and zygotic gene
activity was examined. The results of these experi- ments are shown
in Table 1.
Df(1) K95 and Df (1) ~ ~ ~ ~ - 4 ~ are two small, nonoverlapping
deficiencies that between them delete most of the zeste-white
interval. The level of maternal gene product was reduced by using
females heterozygous for either of these deficien- cies. To reduce
the level of zygotic gene activity, the crosses were done so as to
generate a class of male zygotes that carried either one of the
deficiencies and a small duplication. That duplication, Dp(j ;4)mg,
is a segment of the X chro- mosome that includes the entire
zeste-white region translocated to the fourth chromosome (ROBBINS
1977). Thus, Df/O ; Dp(1;4)mg/+ males are euploid but, as is
demonstrated below, position-eff ect variegation of the duplication
re- sults in less than normal gene activity. In the crosses shown
in Table 1, defi- ciency males are generated in two ways: by
crossing deficiency-heterozygous
TABLE 1
Maternal-zygotic lethal interactions of two zeste-white region
deficiencies ~ ~~ ~
Maternal deficiency - Daughters Sons 1.0s~ of Recovery of
Cross Non-Df Df,Dp Non-D\ Df,Dp Y"X.YL Df,Dp sons
Father: X Y / O ; Dp/4 1. Mother: y p n 973 - 1329 - 0.27 - 2.
Mother: K95/y 484 448 626 19' 0.26 0.06 3. Mother: w258/y 482 477
620 216 0.23 0.70
Paternal deficiency Mother: C ( l ) R M / O ; Dp/4
- - 4. Father: y p n 155 - 185 - 5. Father: K95/w+Y 483 - - 196
- 0.81 6. Father: w258/w+Y 309 - - 1 43 - 0.93
* May include some nondeficienry crossovers between y and the
deficiency. == YSX.YL, y B. Dp = Dp(l;4)mg. 4,= spapol. K95 = Df( l
)K95, yL. wz58 = Df(l)w258-45, y z w-. C(1)RM = C(I)RM,yS S U ( W ~
) w5.
The deficiency males produced in these crosses are Df/O;
Dp(l;#)mg/+. In the maternal transmission crosses, the frequency of
females is depressed because of loss of the compound-XY in the male
parent. That loss is calculated as 1 - (daughters/sons) for the
control and as 1 - (daughters/2~non-Df sons) for the two deficiency
crosses. Recovery of Df;Dp sons is cal- culated as ZXDf;Dp/non-Df
sons to account for lethality of Df;non-Dp sons. In the paternal
transmission crosses, recovery is similarly calculated as 2 x Df;
Dp sons/daughters.
-
190 L. G. ROBBINS
females to attached-XY (= YsX.YL) males and by crossing
deficiency-bearing males to attached-X [= C(Z)RM] females.
Attention is first drawn to the Df (1 j K95 results and to the
following observations: (1 ) Df;Dp males survive despite
variegation of the duplication if their mothers are euploid (cross
5 ; re- covery = 0.81). Thus, there is only a slight effect of this
level of reduced zy- gotic gene activity 011 survival; (2) the
recovery of deficiency females, which also carry a normal X
chromosome, and in half the zygotes the duplication as well, is
unaffected by maternal deficiency heterozygosity. They are
recovered as often as nondeficiency females (cross 2; 448 us. 484);
and ( 3 ) there is a striking absence of Df(Z)K95 ; Dp(Z;4)mg males
when the female parent is heterozygous for the deficiency (cross 2;
recovery = 0.06). Thus, there appears to be a synergistic
interaction between reduced maternal and reduced zygotic gene
activity.
The D f ( 1 j ~ 9 ~ ~ - ~ ~ crosses also show a maternal effect
that is exposed in Df;Dp sons, but not in deficiency-heterozygous
daughters. Though less extreme, this effect is statistically
significant (the difference from a 2: 1 expectation of normal :
deficiency males yields x2 = 21.6, with 1 d.f.) and is
replicable.
An objection might be raised to the notion that maternal-zygotic
interaction lethality is more extreme than expected from separate
maternal and zygotic ef- fects. Were it the case that maternal
deficiency heterozygosity caused substan- tial inviability of all
off spring classes. a weak zygotic effect coupled with a strong
maternal effect could yield the apparent interaction. In the
foregoing experi- ments, a maternal lethal effect on all offspring
genotypes would not be detected except as a reduction in
fecundity.
This matter has been addressed in the following way. Df(l)K95, y
z / y females were again crossed to YsX.YL, y B/O ;
Dp(l;4)mg/spaPoz males. Eggs were col- lected over a 24 hr period
on yeasted, grape-juice-darkened medium. After count- ing, the eggs
were collected and transferred to normal food and allowed to
develop. Of the 2.094 eggs sampled, 804 survived as females and 541
as nonde- ficiency males. Recovery of 541 nondeficiency male
zygotes implies that 541 deficiency male zygotes would have been
expected. Thus, 1,886 out of 2,094, or 90%, of the eggs are
accounted for as survivors, deficiency males, or as victims of the
maternal-zygotic interaction. There is no room here for a maternal
effect sufficient to account, in a multiplicative fashion, for the
lethality of Df;Dp sons of deficiency-heterozygous mothers. I t may
be noted that, though no attempt was made to define the particular
stage at which the lethal zygotes succumb, very few survive to o r
beyond the white-pupa stage.
These observations suggest that, for at least two segments of
the zeste-white region: ( 1 ) a necessary oocyte function is
impaired when gene dose is reduced in the female; (2) the
zeste-white genes carried in Dp(l ;$ jmg have less than normal
activity; and (3) combination of the maternal and zygotic defects,
but neither defect alone, results in lethality.
Several experiments have provided substantial confirmation of
the two major elements of this hypothesis: ( 1 ) the maternal
effect is a consequence of reduced gene dose and not a consequence
of a peculiar sex-specific response of the dupli-
-
MATERNAL-ZYGOTIC LETHAL INTERACTIONS 191
cation, and (2) the inability of Dp(1;4)mg to ensure survival of
Df;Dp males is a consequence of reduced activity resulting from
position-eff ect variegation of the duplication.
Gene dose and the maternal e&ct: Two experiments demonstrate
that the maternal effect is a consequence of reduced gene dose.
Since the effect of Df(l)K95 is more readily followed, that
deficiency was used in these experi- ments. First, as shown in
Table 2. paternal transmission of the duplication in the absence of
a maternal deficiency does not result in reduced recovery of Df;Dp
males. The recovery of Df;Dp males is not significantly different,
when both the duplication and deGciency are paternally transmitted,
from that ob- served in Table 1 when only the deficiency came from
the father. The second cross of Table 2 provides an additional
control where, as discussed in more de- tail later, variegation of
Dp(l ;4)mg is suppressed. The difference in recovery is not
statistically significant. Thus, the parental effect on survival is
a maternal and not a paternal effect.
The second experiment tested whether the maternal effect was a
consequence of reduced gene dose, and not a concequence of maternal
transmission per se. To examine this, maternal gene dose was
increased by including both the de- ficiency and duplication in the
female parent’s gefiotype. Two such crosses were done: one where
only the deficiency female parent carried the duplication and the
other where both parents carried the duplication. For the latter
cross, chro- mosome 4 markers allowed identification of progeny
carrying a paternally de- rived duplication. Deficiency;
nonduplication females were tested as a control. The results of
these crosses are shown in Table 3.
It is apparent in these data that the presence of a duplication
in the deficiency- heterozygous mothers substantially improves the
viability of Df; Dp sons. As the second cross indicates, all of the
expected Df;Dp males are recovered when their mothers carry a
duplication as well as a deficiency; whereas, genotypically iden-
tical sons of females lacking the duplication are recovered only 7%
of the time. Furthermore, as seen in the third cross, rescue
afforded by a maternal duplica- tion occurs even when off spring do
not receive the maternal copy of the duplica- tion-47% of the Df;Dp
males survive if they receive the duplication from their fathers.
Even though recovery of sons bearing a paternal duplication is
lower than that of sons bearing a maternal duplication, the
calculated values cannot be taken too literally. The
paternal-duplication-bearing males are ciD in pheno-
TABLE 2
Effect of paternal fransmissicn of Dp (1;4) m g on recovery of
Df ;Dp males
C(1)RM daughters Dj;Dp sons Recovery __- Cross Df/Y; Dp/+ x
C(I)RM/O; spapo~/spapo~ 724 254 0.7 Df /Y; D p / f x C ( l ) R M /
Y ; spapo1/spapo1 200 93 0.9
Df = Df( l )K95, ys. Dp = Dp(l ;4)mg. Recoveries are calculated
as 2 x Df;Dp males/females.
-
1 92 >. G . ROBBINS
-
MATERNAL-ZYGOTIC LETHAL INTERACTIONS 193
type and their viability is depressed for that reason alone. It
may be noted that an equivalent disparity in the recovery of ciD
and 4- is seen for both the deficiency and nondeficiency females,
though ciD and 4- recovery is closer to equality among the
nondeficiency males. In addition, though Dp(l;#)mg is homozygous
lethal in otherwise normal flies, Df ( I ) K 9 5 / 0 ; Dp(l
;4)mg/Dp( l ;4)mg males have been found (in other crosses)
occasionally to survive. Thus, the number of males apparently
carrying only maternally transmitted duplications is inflated by
survival of an unknown number of males that carry both maternally
and pa- ternally derived duplications, while the number of males
carrying a paternally derived duplication is deflated by the effect
of ciD on viability. Even if some of the apparent numerical
discrepancy reflects an effect of parental origin on variegation
(SPOFFORD 1976), it is neverthless clear that normal gene dose in
the female, though probably still providing less than normal gene
activity, sub- stantially restores the viability of Df;Dp sons.
Together, these data establish that the maternal effect observed
for deficiency heterozygotes is an effect of reduced gene dose and
not some peculiarity of maternal transmission itself.
Variegation and reduced zygotic g e m activity: The second major
element of the hypothesis is that Dp(l;#)mg is less than fully
active. Position-effect variega- tion (reviewed by SPOFFORD 1976)
is a common consequence of chromosome rearrangements and was tested
for by inquiring whether the viability of Df;Dp males responds to
agents known to affect variegation. Two responses were tested:
relaxation of variegation by addition of a Y chromosome and
enhancement by reduced temperature. The effect of a Y chromosome on
survival of Df(l)K95 and D f ( l ) ~ d ~ ~ - ~ ~ deficiency;
duplication sons of deficiency-heterozygous females was tested at
25", survival of Df ( l ) K 9 5 / 0 ; Dp(1;4)mg/spdo2 males was
tested at 19", 25" and 29", and both the maternal effect and Y
chromosome effect were checked for at 19". In addition, the
behavior of a different dupli- cation, visibly variegating for an
included white locus, was examined. The ex- pected enhancement of
viability by addition of a Y chromosome is clearly shown in Table
4.
The effect of temperature on the survival of Df( l )K95 /0 j
Dp(l;4)mg/sp.aPo2 males is shown in Table 5. The most common
response of variegating systems to
TABLE 4
Suppression of lethality of D f ;Dp sons b y a Y chromosome
Non-Df D f Non-Df Dp;Df sons Recovery Mother Father daughters
daughters sons
- XY/O; Dp/4 1551 1523 1927 62 0.12 - of (lY{959/Y X Y / Y ;
Dp/4 805 763 822 225 0.55
Df(i)w"8-45/y XY/O; Dp/4 2262 2245 281 1 1068 0.76 - - X Y / Y ;
Dp/4 548 542 515 234 0.19
xy=YSX.YL, y B, Dp=Dp(Z;4)mg. Df(l)K95=Df(l)K95, y2.
Df(l)w258-45=Df(l)wzS8-4s, yg w-, Recovery of Df;Dp sons is
calculated as 2 ~ D f ; D p sons/non-Df sons.
-
1941 L. G . ROBBINS
TABLE 5
The effect of temperature on gene expression in Dp (1 ;4)mg
Kon-DI; of; Non-Df; Non-Dp Non-Df, Dp Non-Dp DJ;Dp Non-Dp
Kon-Df,Dp DI,DP
sons sons -
lemperatuie daughters daughters daughtas daughters sons
29" 385 261 (0.68) 411 437 510 297 (0.58) 19 (0.04) 2 5 O 959
656 (0.68) 833 808 1287 958 (0.74) 43 (0.03) 19" 1079 934 (0.87)
912 958 1650 1343 (0.91) 7 (0.004)
Homogeneity xz, 2 d.f. 38.7 17.0 14.8
Df(l)Ii95,. y2/y; spap*l/spal.lol females were crossed to Y S X
YL, y B/O; Dp(l ;4)mg/~pa~ '~~
Non-Df; Dp daughters: non-Df; Dp daughters/non-Df; non-Dp
daughters. Non-Df;Dp sons: non-Df;Dp sons/non-Df; non-Dp sons.
Df;Dp sons: Df;Dp sons/non-Df; non-Dp sons. Each group of females
tested was also subdivided by whether they were themselves reared
at
29", 25" or 19". However, there was no grandparental temperature
effect and those homogeneous results have been combined.
males at the indicated temperatures. Recoveries (in parentheses)
were calculated as follows:
temperature shifts is reduced gene expression at lower
temperatures. Survival of Df;Dp males is therefore expected to
decrease, and does decrease, as temperature is reduced. In
addition, hyperploidy for Dp(I;4)mg also causes some loss of
viability. One therefore expects, and finds, increased survival of
hyperploids at lower temperatures. These observations are all in
accord with reduced activity of Dp(I;4)mg at reduced temperature
and are therefore in accord with the idea that Dp(I;4)mg variegates
for the functions it encodes.
Df ( I )wP58-45 yields only a weak maternal-zygotic interaction
at 25". If its effect is analogous to the more striking effect of D
f ( l ) K 9 5 , it should be possible to enhance it by increasing
variegatioE of the duplication. Reduced temperature accomplishes
this, as indicated by the data in Table 6. While survival of Df ( I
) ; Dp(I;4)mg/spaPo1 sons of deficiency-heterozygous females is
about 70% at 25" (e . g., Table 7, cross 3 ) , it is reduced to
only 10% at 19" (Table 7, cross 2). As at 25", survival is markedly
improved by addition of a Y chromosome (Table 7, cross 2 us. Table
7, cross 4).
TABLE 6
Suruiual of Df (1) wz58--45; Dp(l;4)mg males and effects of a Y
chromosome at 29" ~
Daughters Sons Recovery of Cross Father Mother Non-Df Df Non-Df
D f Df;Dp sans
1. Df/wfY C(I )RM/O; Dp/4 1228 - - 6 74 1.10 2. m/O; Dp/4 Df/Y
1192 1172 1768 85 0.10 3. Df/w+Y C(I)DX/Y; Dp/4 1588 - - 948 1.19
4. B / Y ; Dp/4 Df/Y 559 635 710 203 0.57
Df =Df(2)~"8-4~, y2 W-. X Y =YsX.YL, y B . C(I )RM=C( l )RM, y2
S U ( W ~ ) @. C(Z)DX=
Calculation of recovery of Df sons is as indicated in Table 1.
C(I)DX, y f . 4 ~ = s p a p o ~ .
-
MATERNAL-ZYGOTIC L E T H A L I N T E R A C T I O N S 195
While the foregoing experiments demonstrate that the inability
of Dp(1;4) mg to rescue deficiency zygotes is a consequence of
variegation, it remains possible that variegation is a necessary,
but not sufficient, explanation for the behavior of Dp(1;4)mg. The
entire phenomenon could be peculiar to this particular dupli-
cation. Whether other variegating duplications could also expose
maternal- zygotic lethal interactions has been examined by using D
p ( i ' ; 4 ) ~ " ~ ~ g in lieu of Dp(l;4)mg. Dp(1;4) wms5g
includes only part of the zeste-white region, but does extend
beyond the segment missing in D f ( l ) ~ ~ ~ * - ~ ~ . Unlike the
situation with Dp(l;4)mg, variegation of D ~ ( 1 ; 4 ) w " ~ ~ g is
quite apparent; w/O ; Dp- (l;4)wnTfi5g/+ males have eyes that are
almost entirely white. Addition of a Y chromosome produces an
obvious increase in the amount of red tissue, but still leaves most
of the eye white. Since Z I ~ ( I ; ~ ) W ~ ~ ~ ~ ~ ~ shows little
gene activity in normal circumstances, tests of its ability to
rescue Df ( I ) w " * - ~ ~ zygotes were done at 29" to reduce
variegation. Survival was examined both in the presence and absence
of a Y chromosome. The results of these crosses are shown in Table
7.
In the absence of a Y chromosome, Df ( I ) U ? J * - ~ ~ ;
Dp(2;4) wmfi5g zygotes do not survive whether the deficiency is
maternally or paternally transmitted (Table 8, crosses 1 and 2).
However, when Df/Y ; Dp/+ males are generated, a maternal effect on
survival is evident. Only 20% of Df/Y;Dp sons of deficiency females
are recovered (Table 8, cross 4), although recovery of identical
males from normal mothers is near normal (Table 8, cross 3) .
Furthermore, survival is much reduced here as compared with the
approximately 70% survival observed when Dp(1;4)mg is used. Thus, a
weak maternal-zygotic lethal interaction is even more clearly
exposed by Dp(l;)wnTfi5g than by Dp(1;4)mg.
The generality of maternul-zygotic lethal interactions: The
preceding results indicate that genes that function both maternally
and zygotically can be de- tected by combination of reduced
maternal and reduced zygotic gene activities. How common are genes
that function at both times? As a first step in examining this, a
set of overlapping deficiencies that dissect the zeste-white region
have been tested for maternal effects on survival of Df;Dp zygotes.
In each case, Df;Dp zygotes were generated by transmission of the
deficiency from the mother in one cross and from the father in a
second cross. As a consequence of previous
TABLE 7
Maternal-zygotic lethaal interactions of Df (1) w258-45 and Dp
(1 ;4) wm65g
Daughters Sons Recovery of Cross Mother Father Non-Df D f Non-Dj
Df;Dp Df ;Dp sons
1. C(I )RM/O; Dp/4 Df/w+Y 2776 - - 0 0.0 2. Df/Y X Y/O; Dp/4
1514 1528 2083 0 0.0
4. Df/Y X Y / Y ; Dp/4 1212 1214 1429 141 0.20
__
3. C(Z)DX/Y; Dp/4 Df/w+Y 1824 - - 883 0.97 __
Df=Df(i)wZ-45, y2 W-. Dp=Dp(l;4)wm659. C(I )RM=C(I )RM, yz
su(wa) @. C ( I ) D X =
Crosses were done at 29". Calculation of recofvery of deficiency
males is as indicated in Table 1. C ( I ) D X , y f . X Y YSX.YL, y
B.
-
196 L. G . ROBBINS
TABLE 8
Maternal-zygotic lethal interactions of zeste-white region
deficiencies
Female transmission- Non-Df D f Noil-Df Df:Dp
Deficiency tested daughters daughters sons sons Recovery
Male transmission _ - _ _ Non-Df DkDu
daughters ions Recovery
Df (1) wrJ1, y2 Df(1)64c4; y+ DI ( 1 ) w258-11 ,Y Df (1)64j4, y
Df(l)wX12, y Df (1)65j26, y + Df (1 ) 62gl8, yf
Df(1)64fl, yf Df (1)K95, y2
Df ( 1 ) W'JZ, y+
Df (1) ~ 2 5 8 - 4 5 , YZ
209 1116 75 1 92 1 1141 1260 1186 1111 1171 1262 968
219 1102 765 947 1137 1427 1228 1144 1148 1216 973
347 1399 1024 1102 1579 1671 1470 1419 1388 1466 1273
1 2 2 11 3
813 696 7
384 25 467
0.006 0.003 0.004 0.019 0.004 0.973 0.947 0.010 0.553 0.034
0.734
302 629 466 278 1431 820 512 740 469 214 158
146 0.97 370 1.18 55 0.24 211 1.52 388 0.54
315 1.23 343 0.93 306 1.30 126 1.18 85 1.08
4344 1.06
Ratio __ 0.01 0.00 0.02 0.01 0.01 0.92 0.77 0.01 0.42 0.03
0.68
Recovery of Df/Dp sons following maternal transmission of the
deficiency was measured in crosses of either: (1) Df/y females,
where the deficiency chromosome was marked by yf or y2; or (2)
Df/y+ females, where the deficiency chromosome was marked by y . In
each case, the females were crossed to YsX.YL, y B/O;
Dp(i;4)mg/spaPoz males and recovery is calculated as 2 x Df;Dp
sons/non-Df sons. Male transmission crosses were Df/w+Y males X C(I
)RM, y2 su(wa) &/O; Dp(l;4)mg/spapol females and recovery is
calculated at 2 x Df sons/non-Df daughters. The last column gives
the ratio of recovery in the female transmission cross to that in
the male transmission cross.
work (ROBBINS 1980), most of the X chromosome proximal to
Df(l)K95 and and most autosomal material had been replaced with a
common
wild-type background. The other deficiencies do not share a
common background and some extraneous viability differences might
be expected. Nevertheless, any nonmaternally determined viability
differences would be exposed in both re-
gt tko z z w l zw8 zw4 zwl0 zw13 zw2 ZW3 zw6 zw12 zw7 zw5 zwll
zw9 w
b D f ( I ) 6 2 g l 8 3 &f(1)64j4-+ L D f ( l ) w 2 5 8 - 4
5 - 68
- ---------------- '.42Df W64f I+
D f ( l )wrJ2 +Of (Il65j26- , .01 I I Df(l)wrJ1 .01
k Df(1)64c4 .oo
.02
I
I
4 I Df ( l)w258-1 I
Df ( l )wXl2 I
.03 - Df (I) K95- FIGURE 1 .-Deficiency map of maternal-zygotic
lethal interactions. The relative recoveries
of Df; Dp males following maternal .versus paternal transmission
of the deficiency are indicated, together with the arrangement of
the deficiencies with respect to the lethal and visible loci in the
zeste-white region. This figure is based on the data in Table
8.
-
MATERNAL-ZYGOTIC LETHAL INTERACTIONS 197
ciprocal crosses. Thus, attention may be focused on maternal
effects by exami- nation of the ratio of recovery in the two
crosses. All of these crosses were done at 25". The results are
shown in Table 8 and are summarized diagrammatically in Figure
1.
There is no single segment of the zeste-white region that
entirely accounts for the maternal effect, though there are
substantial regional differences. The gt to z w l segment, defined
by Df(l)62g18 and Df(l)65j26, evinces very little, and possibly no,
maternal-effect lethality. On the other hand, the zwl3-w seg- ment,
defined by Df(l)64j4, Df ( l )64f l and Df ( 1 ) ~ ~ ~ ~ - ~ ~ must
contain at least three dosage-sensitive, maternally and zygotically
acting sites. Heterozygosity for D f ( l ) ~ $ ~ ~ + ~ produces a
relative recovery of 0.68; deficiency for the zw3- zw6 segment,
included in the segment of Df( l )64f l that is not overlapped by
Df further reduces recovery to 0.42; and maternal deficiency for
the zw13-zw2 segment nearly eliminates survival of Df;Dp sons.
Whether any sites within the zw8-zwlO segment have separate
maternal effects cannot be deter- mined, since the effect of
zw13-zw2 deficiency heterozygosity is so severe that any additional
effects of the longer deficiencies would not be resolvable.
While exhaustive testing of zeste-white region single-gene
mutations for maternal-zygotic lethal interactions is presently
underway, initial observations indicate that the maternal effect
described in this report is not restricted to de- letions, but can
be ascribed to individual gene functions. Futhermore, these tests
indicate that the deficiency map underestimates the number of
maternally and zygotically active functions. Data for two zw2
mutations and three zw13 muta- tions are shown in Table 9. The
maternal effect is quite as striking here as it is for
deficiencies. The zeste-white region, therefore, encodes at least
four, and conceivably more, functions that are both maternally and
zygotically active.
TABLE 9
Maternal-zygotic interactions of zw2 and zw13 mutmts ~~ ~~ ~ _ _
_ _ ~ ~ ~ ~ ~ ~
Recovery of Mutant Daughters sons mutant; D p
Locus Allele parent Non-mutant Mutant Nor-mutant Mutant sons
Ratio
6 2 ~ 2 1 Female 1300 1310 1798 71 0.08 0.08
65123 Female 948 1031 1462 57 0.08 0.07 zwz Male 1296 - - 6 73 I
.04
Male 705 - - 403 1.14
64bl l Female 1814 1667 21 79 119 0.11 0.11
zw13 e50 Female 464 487 756 187 0.49 0.44
13 Female 631 585 838 42 0.10 0.10
Male 95 0 - - 472 0.99
Male 760 - - 420 1.11 Male 492 - - 248 1.01
Lethal/y females were crossed to YSX.YL, y B/O; Dp(l;4)mg/spaPOz
males and lethal/w+Y males were crossed to C ( I ) R M , ye su(wU)
Wa/O; Dp(l;4)mg/spaPo1 females. Calculations were done as indicated
in Table 8.
-
198 L. G . ROBBINS
SHANNON et al. (1972) have reported on the lethality patterns of
I2 of the 13 known essential zeste-white loci, including that of
zw2. Flies hemizygous for zw2 alleles do not die before the
embryonic-larval transition. Thus, an individual gene function
that, when mutant results in lethality at the end of embryogenesis,
is nevertheless active during oogenesis.
DISCUSSION
The combination of matercal mutant heterozygosity with zygotic
gene varie- gation provides a system for the detection and analysis
of genes that act both maternally and zygotically. It has the
advantage, over de nouo searches for con- ditional mutations, of
allowing use of mutations in segments of the genome that have
already been well characterized. It has the disadvantage of being
restricted to segments for which appropriate duplications are
available. Nevertheless, that this circumstance provides
conditional expression of a lethal phenotype suggests that the
effects of maternal and zygotic gene expression must temporally
over- lap. If an early nonlethal effect of maternal insufficiency
were combined with a later nonlethal effect of zygotic
insufficiency, survival would be expected. How- ever, the
combination of maternal and zygotic defects can be lethal. While it
is conceivable that this temporal overlap reflects the existence of
relatively long- lived maternal messages (GERASIMOVA and SMIRNOVA
1979), much less direct functional equivalence is also possible.
These results imply only that the ulti- mate effects of maternal
and zygotic gene activity can at some point substitute for one
another. The Df;Dp or mutant;Dp condition should be amenable to
clonal analysis to define the timing of this interaction. For
example, for a cell- autollomous deficiency, Df ;Dp clones in off
spring of deficiency heterozygous fe- males should not survive
during those stages where maternally determined func- tion is still
needed, but would survive after that time.
There are not many essential genes in the Drosophila genome that
act only maternally (ZALOKAR, AUDIT and ERK 1975). There may,
however, be quite a few that act both maternally and zygotically.
The deficiencies used in the present analysis divide the
zeste-white region into 10 segments, many just one chro- momere in
length. Several divisions, however, are defined only by relatively
long deficiencies that cause such severe maternal-zygotic lethality
that distinc- tions among them are impossible. Nevertheless, three
distinct sites have been identified by the deficiencies that
contain functions that are active both ma- ternally and
zygotically. In addition, tests of the first two single-gene
functions examined have shown that both cause maternal-zygotic
lethality. If the zeste- white region is not atypical, and there is
no a priori reason to suppose that it is, a substantial fraction,
and conceivably most, of the genes of Drosophila are ac- tive both
during oogenesis and during zygotic development. It is therefore
pos- sible that most maternally determined embryonic functions are
not different from the ordinary housekeeping functions that are
required in any cell. If there are maternal messages that are of
special significance in early development, they may be but a small
fraction of all maternally coded information. That the
-
MATERNAL-ZYGOTIC LETHAL INTERACTIONS 199
observation of a larval lethal effective phase cannot be taken
to imply restriction of gene activity to that particular stage has
already been noted by SHEARN, HER- SPERGER and HERSPERGER (1978)
and by GARCIA-BELLIDO and Moscoso DEL PRADO (1979) ; it is
explicitly demonstrated here for two single-gene functions. The
relationship between lethal effective phase and time of gene
expression need not be straightforward. The temptation to conclude
that genes that act ma- ternally are of specific importance during
embryogenesis must be avoided as well; zygotic effects may not be
apparent unless the maternal genome is also defective. These
results, along with those of GARCIA-BELLIDO and Moscoso DEL PRADO (
1 979), RIPOLL (1977) and RIPOLL and GARCIA-BELLIDO (1 979),
suggest that caution should be exercised in the interpretation of
phenogenetic observa- tions, whether the interpretntion be
developmental or in terms of gene organiza- tion. This note of
caution may be of particular importance when duplications are used
to define cytological location or time of gene expression. It is
possible that, without reciprocal crosses and controls for
variegation, the results may be misleading.
appreciated.
KHESIN, R. B., 1947
NANCY VEENSTRA assisted the author in doing these experiments
and her help is gratefully
Dr. A. GARC~A-BELLIDO has recently brought the following
references to my attention: Maternal effect in Drosophila
melanogaster. Spreading of maternal effect.
Doklady Acad. Nauk SSSR 58: 667-671. -- 1948a Maternal effect in
DrosophiZa melanogaster. Influence of maternal genotype upon the
rate of progeny’s development. Doklady Acad. Nauk SSSR 59:
561-564.. - 1948b Duration of the influence of maternal genotype
upon the development of Drosophila mehogaster progeny. Doklady
Acad. Nauk SSSR 59: 751-754.
I t is with pleasure that I note this earlier work that in part,
parallels my work, as well as that of the Madrid group.
LITERATURE CITED
BAKER, B. S., 1973 The maternal and zygotic control of
development by cinnamon. A new mutant in Drosophila melanogastm.
Develop. Biol. 33: 4 2 9 4 0 . -, 1975 Paternal loss (pal) : a
meiotic mutant in Drosophila melanogaster causing loss of paternal
chromo- somes. Genetics 80: 267-296.
DAVIDSON, E. H., 1976 Gene Activity in Early Development, 2nd
edition. Academic Press, New York.
GALAU, C. A., W. H. KLEIN, M. A. DAVIS, B. J. WOLD, B. J.
BRITTEN and E. H. DAVIDSON, 1976 Structural gene sets active in
embryos and adult tissues of the sea urchin. Cell 7: 487-505.
GARCIA-BELLIDO, A. and J. Moscoso DEL PRADO, 1979 Genetic
analysis of maternal information in Drosophila. Nature 278:
346-348.
GAREN, A. and W. GEHRING, 1972 Repair of the lethal
developmental defect in deep orange embryos of Drosophila by
injection of normal egg cytoplasm. Proc. Natl. Acad. Sci. U.S. 69:
2982-2985.
Maternal effect for genes encoding 6-phospho- gluconate
dehydrogenase and glucose-6-phosphate dehydrogenase in Drosophila
melano- gaster. Develop. Genet. 1 : 97-107.
Mosaic systems. pp. 265-314. In: The Genetics and Biology of
Drosophila, vol. la. Edited by M. ASHBURNER and E. NOVITSKI.
Academic Press, New York.
GERASIMOVA, T. I. and S. G. SMIRNOVA, 1979
~ L L , J. C., W. M. GELBART and D. R. KANKEL, 1976
-
200 L. G . ROBBINS
JUDD, B. H., M. W. SHEN and T. C. KAUFMAN, 1972
KAUFMAN, T. C., M. P. SHANNON, M. W. SHEN and B. H. JUDD,
1975
The anatomy and function of a segment of the X chromosome of
Drosophila nzelanogaster. Genetics 71 : 139-156.
A revision of the cy- tology and ontogeny of several
deficiencies in the 3A1-3C6 region of the X chromosome of
Drosophila melanogaster. Genetics 79 : 265-282.
The genetic analys’s of oogenesis in Drosophila melano- gaster.
pp. 757-791. In: Handbook of Genet cs, V01. 3. Edited by R. C.
KING, Plenum, New York.
LINDSLEY, D. L. and E. H. G R E L ~ 1968 Genetic variations of
Drosophila melanogaster. Car- negie Inst. Wash. Publ. No. 627.
Behavior of somatic cells homozygous for zyg3tic lethals in
Drosophila melano- gaster. Genetics 86: 357-376.
Viability of homozygous deficiencies in somat’c cells of
Drosophila melanogaster. Genetics 91 : 443453.
The meiotic effect of a deficiency in Drosophila melanogaster
with a model for the effects of enzyme deficiency on recombination.
Genetics 87: 655-684. -, 1980 The meiotic effects of a deficiency
in Drosophila melanogaster: identification of two dosage sensitive
sites. Genetics 94: 361-381.
Lethality patterns and morphology of selected lethal and
semi-lethal mutations in the zeste-white region of Dro- sophila
melanogaster. Genetics 72 : 615-638.
SHEARN, A , G. HERSPERGER and E. HERSPERGER, 1978 Genetics
analysis of two allelic tempera- ture-sensitive mutants of
Drosophila melanogaster, both of which are zygotic and maternal
effect lethals. Genetics 89: 341-353.
KING, R. C. and J. D. MOHLER, 1975
RIPOLL, P., 1977
RIPOLL, P. and A. GARCI.*-BELLIDO, 1979
ROBBINS, L. G., 1977
SHANNON, M. P., T. C. KAUFMAN, M. W. SHEN and B. H. JUDD,
1972
SPOFFORD, J. B., 1976 Position-effect variegation in Drosophila.
pp. 955-1019. In: T h e Genetics and Biology of Drosophila, vol.
1C. Edited by M. ASHBURNER and E. NOVITSKI. Academic Press, New
York.
Developmental defects of female-sterile mutants of Drosophila
melanogaster. Develop. Biol. 47: 419-432.
Corresponding editor: T. C. KAUFMAN
ZALOKAR, M., C. AUDIT and J. ERK, 1975