Documente Academic
Documente Profesional
Documente Cultură
9[10, 1976
INTRODUCTION
Duplicate gene systems are quite common in poikilotherms, where their
fitness value is thought to be related to the ability of their possessors to
acclimate and maintain near-optimum levels of metabolic function in spite of
changes in physical parameters of the environment. For example, there appear
to be two distinct gene-enzymes with acetylcholinesterase activity in the steel-
t Department of Zoology, University of Florida, Gainesville, Florida.
823
© I976 Plenum Publishing Corporation, 227 West 17th Street, New York, N.Y. 1001 I. N o part o f this publica-
tion may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic,
mechanical, photocopying, microfilming, recording, or otherwise, without written permission o f the publisher°
824 Giesel
MATERIALS AND M E T H O D S
Genetic Inferences
Fitness Analysis
To collect the data presented here, individual flies were homogenized in 1.5/~1
of a 20% solution of sucrose in 0.1 M TBEDTA (pH 8.6). The samples were
centrifuged at high speed in a Misco refrigerated centrifuge, after which
supernatants were loaded into the pockets of vertical 5 ~ polyacrylamide gels
made with 0.3 M TBEDTA (pH 8.6) to which TPN had been added. The gels
were electrophoresed in the same bridge at 300 V for 5 hr. Phenotypes were
resolved by the staining method of Brewer (1970) for G6PD in Drosophila.
G6PDs migrated under these conditions at about half the rate of G6PDf (see
Fig. 1), but both forms stained with the same time constant. Phenotypes were
read after 45 rain of staining at room temperature (about 25 C). Both forms
were polymorphic.
The data presented here are from three sets of experiments, only the last
of which was designed specifically to investigate differences between genotypes
of the Zw locus in the components of fitness as related to levels of temperature
and food.
In the first experiment, three large cage populations were kept at constant
food and temperature and censused at monthly intervals and simultaneously
assayed for genotypes of several gene-enzymes, including the two G6PD loci.
Using these data, relationships between population density and the allele
and genotype frequencies of the Zw locus could be statistically investigated.
In the second set of experiments, again done for another purpose, flies
were collected from a large, long-standing cage population as eggs by allowing
the population's females to oviposit on fresh cups of medium for 1 hr.
Separate samples were collected sequentially on Carolina Biological Supply
instant Drosophila medium (CBSIDM), CBSIDM diluted 50~ by weight with
Fuller's earth (which Throckmorton, personal communication, believes is
ingested but not metabolized), and CBSIDM fortified with glucose--5% by
weight. The flies were allowed to develop and later were analyzed for Zw
genotype. Each set of data was investigated for deviations from Hardy-
826 Giesel
Food 28 C 25 C 19 C
Low" X X X
Medium" X X X
High b X X X
" CBSIDM.
CBSIDM plus 5% (wt) sucrose.
RESULTS
Genetic Inference
In the test for genetic independence, 104 flies taken from a cage popula-
tion were scored for genotypes of the loci. The data were organized into a
3 x 3 matrix of genotypes. This revealed that all nine possible genotypes were
present. Moreover, when gamete frequencies were taken it was found that
coupling gametes were significantly in excess of repulsion gametes (X] af =
17.58), which suggests that the loci may be linked and were in this case in a
state of linkage disequilibrium. Specific data on the exact linkage relationships
of the two loci will be published elsewhere.
Thus it becomes necessary to redefine the function of the Z w locus. This
locus acts to determine the in vitro, and probably in vivo, relative activities of i
Components-of-Fitness Analysis
Data from the first experiment revealed, upon inspection, that the hetero-
zygote genotype of the Z w locus was always in excess of Hardy-Weinberg
expectation in the cage populations--even though allele frequencies were
relatively constant at a b o u t p Z w ~ = 0.45-0.52. The data were then examined
graphically to detect evidence of association between extent of heterozy.gote
excess and population rate of increase for the corresponding period. A positive
correlation was revealed between extent of excess and population density for
the eight cases in which sample sizes were large and the nonrandom fre-
quencies of genotypes could be confirmed by X2 (Fig. 2). This suggests that
(some component of) heterozygote relative fitness must increase with popula-
tion density, or, since in these experiments food level was constant while
Biology of Duplicate Gene System 829
40 I I I i I
,m 3o 0 0
¢.~
X D
Lid
LU
o~
2o
t.O
l--
Od
I
I0-
oo ; ~; io ~o .'o
POPULATION 51ZE (THOUSANDS}
Fig. 2, Associations between population density and
Z w heterozygote excess (observed frequency minus
expected frequency).
Phenotype
Development Time
The data were first analyzed by three-way analysis of variance to determine
the relative importance of the independent variables--food level, temperature,
and genotype--and their interactions in explaining the variance of the depen-
dent variable--egg to adult development time. Results of this analysis are
presented in Table IV. Here we see that all three independent variables are
important in the explanation of sample variance in development time and that
significant interactions exist between the two environmental variables--
temperature and genotype--and the environmental variables and genotype
of the Z w locus. As expected for a poikilotherm, food level is relatively
unimportant by itself but interacts strongly with temperature. Since it is of
major interest to this study to define the reactions of the heterozygotes vs.
those of the homozygotes to the environment variable levels, mean develop-
ment times of the two broad genotypic classes---heterozygotes and homozy-
gores--were calculated for the three different temperatures and for the food
levels. In Table V we see that at low temperature, which, for a poikilotherm,
Genotype
Homozygotes significance Heterozygotes
At 19 C
LF 301 P < 0.01 286
MF 289 N.S. 283
HF 264 N.S. 269
At 25 C
HF 253 N.S. 247
MF 270 N.S. 230
LF 269 N.S. 271
At 28 C
HF 210 P<0.01 185
MF 216 P < 0.05 205
LF 225 P<0.01 193.5
Fecundity
Because o f generally low viabilities, there were insufficient d a t a for a three-
w a y analysis o f variance. Therefore, m e a n p e r d a y p e r female fecundities a n d
Table VI. Comparison of Mean and Variance of Fecundity per Day and Age at
Peak Fecundity of Z w Homozygotes and Heterozygotesa
"Means and variances are computed over all environmental treatments for which
there were complete data.
bp (variance ratio F) < 0.01.
K
832 Giesel
average ages at reproduction were calculated and then collected over the
treatments in which all gentotypes were represented, to form genotypic means
and variances over all experimental treatments. Results of this treatment of
the data are presented in Table VI. They are remarkably clear. There are no
genotypic differences in either average egg production per day or average
age of reproduction. However, genotypic differences in variance of these mean
values engendered by experimental conditions are striking and highly signi-
ficant (F ratio, P<0.01). These data suggest, more clearly that any other set
presented here, that the heterozygotes are buffered against variation in
trophic conditions.
DISCUSSION
This set of experiments, designed to evaluate fitness differentials among
genotypes of Zw, demonstrated that Zw heterozygotes are more "fit" under
conditions of trophic deprivation than either Z w homozygote. This fact is
particularly well demonstrated by the third set of experiments, in which the
development times and fecundity distributions of genotypes could be unam-
biguously compared. The first two sets of data, which showed large excesses
of the Z w heterozygote phenotype, share a common ambiguity; Z w homo-
zygotes can, depending on genetic background and trophic conditions,
exhibit phenotypes approaching and sometimes indistinguishable from those
of Z w heterozygotes (Giesel, in preparation). Thus the heterozygote excess
which was noted to increase with degree of trophic deprivation may be due
to two factors. Higher viability of heterozygotes is one, and a set of regulatory
phenomena producing heterozygotelike homozygote phenotypes is the other.
Nevertheless, the fitness polymorphism shown here is inescapable.
The fitness differentials are most likely to be due to the heterozygote's
greater ability to utilize the products of both structural forms of G6PD,
which have quite different substrate and cofactor kinetics, "on demand."
G6PD S is the more efficient utilizer of low G6P levels, and G6PDf is less
adversely affected by low levels of cofactor (see Steel et aL, 1968; Komma,
1968). Thus the Zw heterozygotes, which run both structural forms of G6PD
simultaneously (at least more so than do Z w homozygotes), should more
readily and more efficiently metabolize low levels of dietary lipid and low
levels of dietary glucose than either Z w homozygote.
REFERENCES
Brewer, G. J. (1970). An Introduction to Isozyme Technique, AcademicPress, New York,
186 pp.
Komma,D. J. (1968). Glucose6-phosphate dehydrogenasein Drosophila: A sex influenced
electrophoretic variant. Biochem. Genet. 1:229.
Biology of Duplicate Gene System 833