- Email: [email protected]
Behavior of life-shortening genes in genetic mosaics of Drosophila melanogaster
Mechanisms of Ageing and Development, 23 (1983) 1-10 Elsevier Scientific Publishers Ireland Ltd.
t
BEHAVIOR OF LIFE-SHORTENING GENES IN GENETIC MOSAICS OF DROSOPHILA MELANOGASTER A D A I R B. G O U L D and A R N O L D M. C L A R K
School of Life and Health Sciences, University of Delaware, Newark, Delaware 19711 (U.S.A.) (Received March 23rd, 1982) (Revision received February 16th, 1983)
SUMMARY Genetic mosaics of two life-shortening (adult lethal) genes in Drosophila melanogaster were analyzed with respect to the location and extent of external tissues which exhibited the life-shortening genotype. It was hoped by this method to localize the site of action of each gene. For one mutant, A L 2, no correlation was found between the site of the adult lethal genotype and the adult life span of the mosaic fly. However, there was a direct correlation between the total amount of tissue exhibiting the adult lethal genotype and length of adult life. For the other mutant, A L 4, a more direct correlation could be made between location of adult lethal tissue and a shortened adult life span. Its focus was oriented toward the cephalic region. Like A L 2, it exhibited a direct correlation between total adult lethal tissue and length of adult life. The importance of this method of genetic analysis for the understanding of the action of life-shortening genes is discussed.
K e y words: Life-shortening; Mosaic; Adult lethal; Mutation; Mean adult life span
INTRODUCTION The control of adult life span in any organism is a complex developmental function which is a product of the interaction of the genome with the internal and external environment [1]. U n d e r experimental conditions, where the external environment can be kept constant, the influence of the genome on the adult life span of an organism can be investigated. This kind of genetic control has been demonstrated in Drosophila melanogaster [2]. In continuing our study of the 0047-6374/83/$03.00 Printed and Published in Ireland
© 1983 Elsevier Scientific Publishers Ireland Ltd.
genetics of longevity we have employed "adult lethal mutations" whose phenotypic effect is a radical reduction in mean adult life span [2, 3]. Actually, a number of morphological variants in D. melanogaster have been shown to have shortened adult life spans [2, 4, 5], and thus are to some extent "lethal" in the adult stage. The life-shortening mutations which we have recently investigated were induced by either X-rays or ethyl methanesulfonate and were found to have a severe effect on mean adult life span, reducing it from a wild type average of 65-75 days to 12-18 days [6]. The present study of mosaic flies was begun in an effort to determine the site of action of the adult lethal genes. If the life-shortening mutation could be identified in the tissues of the adult fly then a correlation could be made between adult life span and that part of the fly which had the short-lived gene. For instance, if the head of the fly carried the short-lived mutation and the rest of the body was wild type, and the fly lived only 15 days, then it could be assumed that the site of action of the gene with respect to longevity is in the head region. Such a correlation was made by Hotta and Benzer [7] in their construction of a morphogenetic fate map for the "drop dead" mutant of D. melanogaster. Genetic markers would be needed to identify the adult lethal tissues in our flies since our sex-linked short-lived mutants are phenotypically wild type. We chose white eye (w) and yellow body (y) for our markers, and bred two of our adult lethal mutants into such a marker strain. Preliminary testing showed that the yw stock had an adult life span approaching that of wild type (Table I). MATERIALS
AND METHODS
Two of our short-lived, sex-linked mutants, A L 2 and A L 4, were chosen for incorporation into a yellow white stock. Males of each short-lived strain were crossed to y w females and the F~ females backcrossed to A L 2 or A L 4 males, TABLE MEAN
l ADULT
LIFE SPAN OF EXPERIMENTAL
Stra in
Sex
n
L(fe span ( days )
Ring X yw yw AL 2 AL 2 y w AL 2
Females Males Females Males Females Males
47 387 57 250 79 356
31.0 + 1.6 58.4 -+-0.8 57.5 +- 1.9 12.~7-+ 0.2 14.5 -+ 0.3 11.7 +- 0.2
AL4 AL 4 yw AL 4
Males Females Males
116 37 109
15.3 ~ 0 . 3 17.8-+0.7 19.0±0.3
STRAINS
respectively. The recombinant progeny from this backcross should include some y w males which would also be short-lived. In order to identify these males all of the y w males from these backcrosses were mated individually to Basc females. The resulting female progeny, each of which potentially carried the life-shortening gene, were then mated singly to Basc males. The y w male progeny from each of these crosses were tested for mean adult life span by the method previously described [6[, and those which proved to carry the adult lethal genes were used in the mosaic analyses. Genetic mosaics were produced by using a ring X stock which was kindly supplied to us by Dr. Jeffrey Hall. Females of this stock had the genetic constitution R(1)wvC/In (1) dl-492ywlz. Mosaics occur among their progeny because the ring chromosome tends to get lost during early mitotic divisions. This leaves some cells of the developing embryo with only one X chromosome, the one derived from the male parent [8]. One group of ring X females was mated to y w A L 2 males, and another group to y w A L 4 males. Mosaic flies among their progeny were scored with respect to site and extent of y w (male) tissue. Life span determinations were made for these mosaics, housing them singly in 8-dram vials with food [6]. Life spans of ring X females were similarly tested. RESULTS Table I gives the values for mean adult life span of the various genetic strains pertinent to these experiments. There is clearly a significant difference between the mean adult life spans of the A L 2 and A L 4 strains and those without these mutations. The process of incorporating A L 2 and At. 4 into the y w stock yielded information which enabled us to assign a place on the X chromosome to each mutant. Life spans were tested for 62 lines in which the single male progenitor was either y w + or y w A L 2. Of these, 15 lines were identified by adult life span as y w A L 2 and 47 as y w + . This gives a recombination value of 24.2 (16/52). Since w is at 1.5 on the X chromosome, this places A L 2 at approximately 25.7. Similarly, 4 out of 58 lines which were either y w + or y w At. 4 were identified as y w A L 4. The recombination value is 6.9 (4/58); hence A L 4 is located at 8.4 on the X chromosome. We obtained 110 mosaics from crosses of ring X females with y w A L 2 males, and their adult life spans ranged from 1 day to 86 days. We also obtained 121 mosaics, with adult life spans of 4 to 74 days, from crosses of ring X females with y w At. 4 males. The mosaics were classified in several ways. First they were characterized as having a short (1-23 days), intermediate (24-49 days), or long (50-86 days) adult life span. These particular limits were chosen because they coincide with our observations of the range of adult life spans in short- and long-lived lines. Each fly was also examined for the amount and location of y w A L 2 or y w A L 4 tissue in the external cuticle. A correlation was made between
the length of adult life (short, intermediate, or long) and the kind of tissue ( y w A L 2 or y w A L 4 [d], mosaic [M], or wild type [~?]) in the head, thorax, or abdomen. These correlations are found in Tables II and III in which are listed the percentage and location of tissue types with respect to the adult life span of the mosaic flies.
T A B L E I1 D I S T R I B U T I O N O F T I S S U E T Y P E S W I T H R E S P E C T T O A D U L T L I F E SPAN IN F LIES MOSAIC FOR AL 2 Head A d u l t life span
Thorax
Abdomen
Type a
n
%
Type
n
%
Type
n
%
d" M 9
10 12 4
38 46 15
d M 9
9 16 1
35 61 4
d" M (2
5 12 9
19 46 35
(24-49 days) n = 45
d M 9
6 20 19
13 44 42
d M 9
3 31 11
7 69 24
d M ?
4 20 21
9 44 47
Long (50-86 days) n = 39
d M 9
9 17 13
23 44 33
d M 9
5 26 8
13 67 20
d M ?
4 18 t7
10 46 44
Short (1-23 days) n = 26
Intermediate
ad
=
y w A L 2; M = mosaic; 9 = wild type.
T A B L E II1 D I S T R I B U T I O N O F T I S S U E T Y P E S W I T H R E S P E C T TO A D U L T L I F E SPAN IN F LIES MOSAIC FOR AL 4 Head A d u l t life span
Thorax
Abdomen
Type a
n
%
Type
n
%
Type
n
%
d M ~
23 29 7
39 49 12
d M 9
17 37 5
29 63 8
d M 9
16 25 18
27 42 31
(24-M9 days) n = 45
d M 9
3 23 24
6 46 48
d' M 9
7 28 15
14 56 30
d M (2
8 24 18
16 48 36
Long (50-86 days) n = 39
d M 9
0 4 8
0 33 67
d M ~?
0 3 9
0 25 75
d' M 9
1 5 6
8 42 50
Short (1-23 days) n = 26
Intermediate
ac~ = y w A L 4; M = mosaic; 9 = wild type.
For A L 2 (Table II) the percentage of completely male (short-lived) heads is greater in short-lived mosaics (38%) than in either intermediate (13%) or longlived (23%), while the percentage of completely female (long-lived) heads is greatest in intermediate-lived mosaics (42%), least in short-lived (15%) and between these two figures in long-lived (33%). Mosaic heads are found in almost equal proportions in each life span group. Hence, for the head region there is no clear correlation of adult life span with the presence of short-lived tissue. The data for the thoraces and a b d o m e n s show the same lack of correlation. There are few completely male thoraces and a b d o m e n s in any of the life span groups. Completely female a b d o m e n s are in relatively high proportions for all life spans. The data for mosaic a b d o m e n s show the same proportions for each life span as do the mosaic heads, and mosaic thoraces comprise by far the largest percentage in each life span group. From these data we get no clear indication of where the action of the short-lived mutant A L 2 is localized, or if indeed there is such a localized site. For A L 4 (Table Ill) the percentage of completely male (short-lived) heads, thoraces and a b d o m e n s is higher in the short-lived mosaics than in either the intermediate- or long-lived. In fact, one finds relatively few completely male body segments in intermediate-lived mosaics and no completely male heads or thoraces in the long-lived. The data for A L 4 resemble those for A L 2 for the short- and intermediate-lived mosaics in that the percentages of tissue types for each life span group are similar, and in that a majority of the body segments are of mosaic composition. However, for the long-lived mosaics there is a definite difference in the data for the two mutants. A L 4 long-lived mosaics show almost no completely male (short-lived) body segments, and for each segment the percentage of completely female (long-lived) parts is much greater than for A L 2 mosaics. These data suggest that there may be a correlation between life span of mosaics and the location of male (AL) or female (+/AL) tissue. The short-lived A L 4 mosaics tend to have heads which are short-lived to a much greater degree than is indicated for A L 2 mosaics. As adult life span of A L 4 mosaics increases there is a corresponding decrease in the percentage of completely male body segments, markedly so for the head region. A n o t h e r comparison of the data for the two mutants shows that A L 4 produced a greater proportion of short-lived mosaics than A L 2 (49% vs. 24%) and a smaller proportion of long-lived mosaics (10% vs. 35%). The proportion of intermediate-lived mosaics was the same for both mutants (41%). Along with the data in Tables II and III these figures suggest that the two mutants have different developmental pathways in achieving their effect on adult life span. If one combines the male and mosaic areas with respect to the body s e g m ~ t ~ and length of adult life (Tables IV and V), one can see the same tre~d in :~" distribution of adult lethal tissue for all body segments and all life span lengths. with the exception of the long-lived A L 4 mosaics. There is a greater percentage of areas in which there is some expression of the y w A L 2 genome in all segments and for all life spans than there is of areas of completely wild type (female) tissue. Again, this furnishes no correlation of the action of A L 2 with a particular part of
TABLE
IV
DISTRIBUTION
OF AREAS
O F A L 2 T I S S U E IN M O S A I C S
Head
Adult lift" span Short Intermediate Long
Thorax ?
Abdomen
8"+M ~ (%)
(%)
~+M (%)
? (%)
~'+M (%)
(%)
84 57 67
15 42 33
96 76 80
4 24 2(1
65 53 56
35 47 44
~ = y w A L 2: M = m o s a i c ; 9 = wild t y p e ; d + M = c o m b i n e d y w A L 2 a n d m o s a i c a r e a s .
TABLE
V
DISTRIBUTION
OF AREAS
O F A L 4 T I S S U E IN M O S A I C S
Head
Adult life span Short Intermediate Long
Th orax
A bdom en
8'+M"
~2
(%)
(%)
~+M (%)
9 (%)
~+M (%)
? (%)
78 52 33
12 48 67
92 70 25
8 30 75
69 64 50
31 36 50
"~' = y w A L 4; M = m o s a i c ; ? = wild t y p e : 6' + M = c o m b i n e d y w A L 4 a n d m o s a i c a r e a s .
the fly's exterior. There is a conspicuous change in the trend in the expression of the y w A L 4 genome in the long-lived mosaics, most notably for the head and thorax. This trend suggests that the focus of A L 4 is oriented toward the cephalic region. A n o t h e r approach to the characterization of these mosaic flies was made. Each fly was examined and an estimate made of the total amount of y w A L 2 or y w A L 4 tissue in the whole of its external anatomy. Then a correlation was made between the life span and the total area of short-lived genotype. These data are shown in Figs. 1 and 2. In this analysis one can see a trend in the short-lived flies, with very few having only 10% male (short-lived) tissue but they progress to become the largest category in the 90% group. For A L 2 the long-lived flies, however, do not show a complementary trend, except at the 90% level. At the other levels they have essentially the same amount of adult lethal tissue in each. For A L 4 the long-lived flies do complement this trend, and none of them had more than 50% male tissue. A m o n g the progeny of the cross of ring X/y w lz females with y w A L 2 males we obtained 48 males which were y and w but not lz, and which had a mean adult life span of 10.7 _+ 1.2 days. Only two of them could be classified as intermediate in
F7~ long-lived
oo]
[]
•
mutant
intermediate - lived
801 i i
ii /
/
ii
,,'1' //
60
.
O
E
4o.
N
2o-
II
I0
25
50
75
90 patroclinous males
% of mutant tissue Fig. 1. D i s t r i b u t i o n of m u t a n t tissue in A L 2 mosaics.
F
long- lived
7
-
••
[]
intermediate - lived
mutant
I00
(/)
80
i i i i i i
i i i i i i
N.(..) .m
60
(/) 0
E 40
o~ 20
I0
25
50
7'5
90 patroclinous males
% of mutant tissue Fig. 2. D i s t r i b u t i o n of m u t a n t tissue in A L 4 mosaics.
life span, a n d n o n e w e r e l o n g - l i v e d . H e n c e t h e y w e r e c e r t a i n l y of the g e n e t i c constitution y w AL 2 and thus were patroclinous. Similar y w males were found a m o n g t h e p r o g e n y of t h e cross of ring X f e m a l e s with y w A L 4 males. T h e s e 78 m a l e s h a d a m e a n a d u l t life s p a n of 16.9 + 0.4 days, a n d all of t h e m w e r e within
the short-lived category. The probable origin of these patroclinous males is in the loss of the ring X c h r o m o s o m e very early in development, with subsequent external expression of only the X c h r o m o s o m e genes derived from the male parent. There is possibly some internal mosaicism which we could not detect. These males were sterile, since they lack a Y chromosome. When the patroclinous males are c o m p a r e d with those mosaics with the greatest (90%) proportion of male tissue (Figs. 1 and 2), they continue the trend of larger areas of male tissue being correlated with short adult life span. DISCUSSION Genetic mosaics have proven to be a fruitful tool in analyses of insect development [9-11]. They have provided examples of a u t o n o m o u s behavior of one genotype surrounded by tissue of another [12]. They have been used to explore cell lineages [11] and insect behavior [13, 14]. Hence we had expected that we could use this tool to localize the site of action of our adult lethal genes. This did not prove to be the case for the mutant A L 2, and is only suggested for A L 4. We have made the assumption that the external phenotype is a reliable indicator of the genotype of the tissues which lie underneath or are in the same general area of the organism. This assumption is based on studies of gynand r o m o r p h s in Drosophila melanogaster by Morgan [15] and in Habrobraconjuglandis [13, 16] in which the sexual behavior of a sex mosaic was correlated with the sexual genotype of tissue in the head region and not with the sexual genotype of the thorax, abdomen, or genitalia. Further evidence that such an assumption is valid was provided by Egen [17] who showed that in Habrobracon mosaics the sexual phenotype of the a b d o m e n was correlated with the sex of the underlying gonads. Ikeda and Kaplan [18] found a high, but not absolute, correlation between external genotype and the genotype of underlying neural tissue of the thorax of Drosophila melanogaster mosaics. On the other hand, Hall [14] has obtained mosaics with male heads which did not exhibit male courtship behavior and which proved to have brains of predominantly female genotype. Benzer [19] found that the external genotype of the head did not always correspond with that of the brain in his " d r o p d e a d " mutant flies. Some flies with normal heads died early, while some with mutant heads had normal life spans. In these cases the external phenotype was not an indication of the genotype of underlying tissue. Similarly in A L 2 the external phenotype may not furnish positive identification of the internal genotype. An imaginal disc may contain anlagen of both external and internal structures, but the development of these structures is autonomous with respect to the genotype of the cells involved. Thus a mosaic disc could result in external structures of one genotype and internal structures of another. A possible solution to our problem in mosaic analysis would be to employ a genetic m a r k e r for internal mosaicism, such as the enzyme acid phosphatase. If a null allele for acid phosphatase production were bred into a short-lived strain, and the
resulting strain bred into a mosaic producing system, then by suitable staining and sectioning techniques [20] the internal mosaicism could be detected, and correlations with adult life span attempted. For certain life-shortening mutations the decreased life span may be due to a localized effect on a specific organ. Such an effect occurred for the "+drop dead" locus of Drosophila where mosaics with the head as "drop dead" genotype died within 3 days ]7], and where the study had shown deterioration of the adult brain [19]. This is not the pattern observed for our A L 2 mutation whose effect is just as dramatic as "drop dead" but where there appears to be no localized region of degeneration. For this adult lethal the degeneration pattern may indeed not be localized at all but rather related to a biochemical defect that influences all cells and tissues, and hence is correlated with the amount of A L 2 tissue. The A L 4 mutant seems to give a clearer pattern with respect to localization of its effect, although from our data one cannot conclude that a specific organ is involved. The orientation toward the head region and the closer correlation of external phenotype with adult life span suggest that A L 4 is involved in a more limited system than A L 2 and that the short adult life span of AL 4 may very well have a different physiological basis from that of A L 2. We believe that experiments using mosaics for still other life-shortening genes can be important in the study of the genetics of aging.
REFERENCES 1 A.M. Clark, Genetic factors associated with aging, Adv. Gerontol. Res., I (1964) 207-225. 2 A+M. Clark and A.B. Gould, Genetic control of adult life span in Drosophila melanogaster. Exp. Gerontol., 5 (1970) 157. 3 M.B. Baird and A.M. Clark, X-ray induced life-shortening mutations in Habrobracon: A genetic approach to senescence and duration of life. Exp. Gerontol., 6 (1971) 1. 4 R. Pearl and S.L. Parker, Experimental studies on the duration of life. I1. Hereditary differences in duration of life in line-bred strains of Drosophila melanogaster. Am. Nat., 56 (1922) 174. 5 B.M. Gonzales, Experimental studies on the duration of life. VIII. The influence upon duration of life of certain mutant genes of Drosophila melanogaster. Am. Nat., 57 (1923) 289. 6 A.B. Gould and A.M. Clark, X-ray induced mutations causing adult life-shortening in Drosophila melanogaster. Exp. Gerontol., 12 (1977) 107. 7 Y. Hona and S. Benzer, Mapping of behavior in Drosophila mosaics. Nature, 240 (1972) 527. 8 Y. Hotta and S. Benzer, Genetic dissection of the Drosophila nervous system by means of mosaics. Proc. Natl. Acad. Sci. U.S.A., 67 (1970) 1156. 9 A.M., Clark, A.B. Gould and S.F. Graham, Patterns of development among mosaics in Habrobracon juglandis. Dev. Biol., 25 (1971) 133. 10 C. Stern, Gene expression in genetic mosaics. Genetics, 61 (Suppl. 1) (1969) 199. 11 P.J. Bryant and H.A. Schneidermann, cell lineage, growth and determination in the imaginal discs of Drosophila melanogaster. Dev. Biol., 20 (1969) 263. 12 C. Stern, Genetic Mosaics and Other Essays, Harvard University Press, Cambridge, Massachusetts, 1968, pp. 138-141. 13 A.M. Clark and R.C. Egen, Behavior of gynandromorphs of the wasp Habrobracon juglandis. Dev. Biol., 45 (1975) 251. 14 J.C. Hall, Control of male reproductive behavior by the central nervous system of Drosophila: dissection of a courtship pathway by genetic mosaics. Genetics, 92 (1979) 437.
l(J
15 L.V. Morgan, Composites of Drosophila rnelanogaster. Carnegie Inst. Washington Publ., 399 (1929) 225. 16 P.W. Whiting, Reproductive reactions of sex mosaics of a parasitic wasp, Habrobracon ]uglandis. J. Comp. Psychol., 14 (1932) 345. 17 R.C. Egen, A study of mosaicism in the a b d o m e n of Habrobracon juglandis. Ph.D. Thesis, University of Delaware, 1974. 18 K. Ikeda and W.D. Kaplan, Unilaterally patterned neural activity of g y n a n d r o m o r p h s mosaic for a neurological mutant ol Drosophila melanogaster. Proc. Natl. Acad. Sci. U.S.A., 67 (1970) 148(1. 19 S. Benzer, From the gene to behavior. J. Am. IVied. Assoc., 218 (1971) 1015. 20 D.R. Kankel and J.C. Hall, Fate mapping of nervous system and other internal tissues in genetic mosaics of Drosophila melanogaster. Dev. Biol., 48 (1976) 1.
t
BEHAVIOR OF LIFE-SHORTENING GENES IN GENETIC MOSAICS OF DROSOPHILA MELANOGASTER A D A I R B. G O U L D and A R N O L D M. C L A R K
School of Life and Health Sciences, University of Delaware, Newark, Delaware 19711 (U.S.A.) (Received March 23rd, 1982) (Revision received February 16th, 1983)
SUMMARY Genetic mosaics of two life-shortening (adult lethal) genes in Drosophila melanogaster were analyzed with respect to the location and extent of external tissues which exhibited the life-shortening genotype. It was hoped by this method to localize the site of action of each gene. For one mutant, A L 2, no correlation was found between the site of the adult lethal genotype and the adult life span of the mosaic fly. However, there was a direct correlation between the total amount of tissue exhibiting the adult lethal genotype and length of adult life. For the other mutant, A L 4, a more direct correlation could be made between location of adult lethal tissue and a shortened adult life span. Its focus was oriented toward the cephalic region. Like A L 2, it exhibited a direct correlation between total adult lethal tissue and length of adult life. The importance of this method of genetic analysis for the understanding of the action of life-shortening genes is discussed.
K e y words: Life-shortening; Mosaic; Adult lethal; Mutation; Mean adult life span
INTRODUCTION The control of adult life span in any organism is a complex developmental function which is a product of the interaction of the genome with the internal and external environment [1]. U n d e r experimental conditions, where the external environment can be kept constant, the influence of the genome on the adult life span of an organism can be investigated. This kind of genetic control has been demonstrated in Drosophila melanogaster [2]. In continuing our study of the 0047-6374/83/$03.00 Printed and Published in Ireland
© 1983 Elsevier Scientific Publishers Ireland Ltd.
genetics of longevity we have employed "adult lethal mutations" whose phenotypic effect is a radical reduction in mean adult life span [2, 3]. Actually, a number of morphological variants in D. melanogaster have been shown to have shortened adult life spans [2, 4, 5], and thus are to some extent "lethal" in the adult stage. The life-shortening mutations which we have recently investigated were induced by either X-rays or ethyl methanesulfonate and were found to have a severe effect on mean adult life span, reducing it from a wild type average of 65-75 days to 12-18 days [6]. The present study of mosaic flies was begun in an effort to determine the site of action of the adult lethal genes. If the life-shortening mutation could be identified in the tissues of the adult fly then a correlation could be made between adult life span and that part of the fly which had the short-lived gene. For instance, if the head of the fly carried the short-lived mutation and the rest of the body was wild type, and the fly lived only 15 days, then it could be assumed that the site of action of the gene with respect to longevity is in the head region. Such a correlation was made by Hotta and Benzer [7] in their construction of a morphogenetic fate map for the "drop dead" mutant of D. melanogaster. Genetic markers would be needed to identify the adult lethal tissues in our flies since our sex-linked short-lived mutants are phenotypically wild type. We chose white eye (w) and yellow body (y) for our markers, and bred two of our adult lethal mutants into such a marker strain. Preliminary testing showed that the yw stock had an adult life span approaching that of wild type (Table I). MATERIALS
AND METHODS
Two of our short-lived, sex-linked mutants, A L 2 and A L 4, were chosen for incorporation into a yellow white stock. Males of each short-lived strain were crossed to y w females and the F~ females backcrossed to A L 2 or A L 4 males, TABLE MEAN
l ADULT
LIFE SPAN OF EXPERIMENTAL
Stra in
Sex
n
L(fe span ( days )
Ring X yw yw AL 2 AL 2 y w AL 2
Females Males Females Males Females Males
47 387 57 250 79 356
31.0 + 1.6 58.4 -+-0.8 57.5 +- 1.9 12.~7-+ 0.2 14.5 -+ 0.3 11.7 +- 0.2
AL4 AL 4 yw AL 4
Males Females Males
116 37 109
15.3 ~ 0 . 3 17.8-+0.7 19.0±0.3
STRAINS
respectively. The recombinant progeny from this backcross should include some y w males which would also be short-lived. In order to identify these males all of the y w males from these backcrosses were mated individually to Basc females. The resulting female progeny, each of which potentially carried the life-shortening gene, were then mated singly to Basc males. The y w male progeny from each of these crosses were tested for mean adult life span by the method previously described [6[, and those which proved to carry the adult lethal genes were used in the mosaic analyses. Genetic mosaics were produced by using a ring X stock which was kindly supplied to us by Dr. Jeffrey Hall. Females of this stock had the genetic constitution R(1)wvC/In (1) dl-492ywlz. Mosaics occur among their progeny because the ring chromosome tends to get lost during early mitotic divisions. This leaves some cells of the developing embryo with only one X chromosome, the one derived from the male parent [8]. One group of ring X females was mated to y w A L 2 males, and another group to y w A L 4 males. Mosaic flies among their progeny were scored with respect to site and extent of y w (male) tissue. Life span determinations were made for these mosaics, housing them singly in 8-dram vials with food [6]. Life spans of ring X females were similarly tested. RESULTS Table I gives the values for mean adult life span of the various genetic strains pertinent to these experiments. There is clearly a significant difference between the mean adult life spans of the A L 2 and A L 4 strains and those without these mutations. The process of incorporating A L 2 and At. 4 into the y w stock yielded information which enabled us to assign a place on the X chromosome to each mutant. Life spans were tested for 62 lines in which the single male progenitor was either y w + or y w A L 2. Of these, 15 lines were identified by adult life span as y w A L 2 and 47 as y w + . This gives a recombination value of 24.2 (16/52). Since w is at 1.5 on the X chromosome, this places A L 2 at approximately 25.7. Similarly, 4 out of 58 lines which were either y w + or y w At. 4 were identified as y w A L 4. The recombination value is 6.9 (4/58); hence A L 4 is located at 8.4 on the X chromosome. We obtained 110 mosaics from crosses of ring X females with y w A L 2 males, and their adult life spans ranged from 1 day to 86 days. We also obtained 121 mosaics, with adult life spans of 4 to 74 days, from crosses of ring X females with y w At. 4 males. The mosaics were classified in several ways. First they were characterized as having a short (1-23 days), intermediate (24-49 days), or long (50-86 days) adult life span. These particular limits were chosen because they coincide with our observations of the range of adult life spans in short- and long-lived lines. Each fly was also examined for the amount and location of y w A L 2 or y w A L 4 tissue in the external cuticle. A correlation was made between
the length of adult life (short, intermediate, or long) and the kind of tissue ( y w A L 2 or y w A L 4 [d], mosaic [M], or wild type [~?]) in the head, thorax, or abdomen. These correlations are found in Tables II and III in which are listed the percentage and location of tissue types with respect to the adult life span of the mosaic flies.
T A B L E I1 D I S T R I B U T I O N O F T I S S U E T Y P E S W I T H R E S P E C T T O A D U L T L I F E SPAN IN F LIES MOSAIC FOR AL 2 Head A d u l t life span
Thorax
Abdomen
Type a
n
%
Type
n
%
Type
n
%
d" M 9
10 12 4
38 46 15
d M 9
9 16 1
35 61 4
d" M (2
5 12 9
19 46 35
(24-49 days) n = 45
d M 9
6 20 19
13 44 42
d M 9
3 31 11
7 69 24
d M ?
4 20 21
9 44 47
Long (50-86 days) n = 39
d M 9
9 17 13
23 44 33
d M 9
5 26 8
13 67 20
d M ?
4 18 t7
10 46 44
Short (1-23 days) n = 26
Intermediate
ad
=
y w A L 2; M = mosaic; 9 = wild type.
T A B L E II1 D I S T R I B U T I O N O F T I S S U E T Y P E S W I T H R E S P E C T TO A D U L T L I F E SPAN IN F LIES MOSAIC FOR AL 4 Head A d u l t life span
Thorax
Abdomen
Type a
n
%
Type
n
%
Type
n
%
d M ~
23 29 7
39 49 12
d M 9
17 37 5
29 63 8
d M 9
16 25 18
27 42 31
(24-M9 days) n = 45
d M 9
3 23 24
6 46 48
d' M 9
7 28 15
14 56 30
d M (2
8 24 18
16 48 36
Long (50-86 days) n = 39
d M 9
0 4 8
0 33 67
d M ~?
0 3 9
0 25 75
d' M 9
1 5 6
8 42 50
Short (1-23 days) n = 26
Intermediate
ac~ = y w A L 4; M = mosaic; 9 = wild type.
For A L 2 (Table II) the percentage of completely male (short-lived) heads is greater in short-lived mosaics (38%) than in either intermediate (13%) or longlived (23%), while the percentage of completely female (long-lived) heads is greatest in intermediate-lived mosaics (42%), least in short-lived (15%) and between these two figures in long-lived (33%). Mosaic heads are found in almost equal proportions in each life span group. Hence, for the head region there is no clear correlation of adult life span with the presence of short-lived tissue. The data for the thoraces and a b d o m e n s show the same lack of correlation. There are few completely male thoraces and a b d o m e n s in any of the life span groups. Completely female a b d o m e n s are in relatively high proportions for all life spans. The data for mosaic a b d o m e n s show the same proportions for each life span as do the mosaic heads, and mosaic thoraces comprise by far the largest percentage in each life span group. From these data we get no clear indication of where the action of the short-lived mutant A L 2 is localized, or if indeed there is such a localized site. For A L 4 (Table Ill) the percentage of completely male (short-lived) heads, thoraces and a b d o m e n s is higher in the short-lived mosaics than in either the intermediate- or long-lived. In fact, one finds relatively few completely male body segments in intermediate-lived mosaics and no completely male heads or thoraces in the long-lived. The data for A L 4 resemble those for A L 2 for the short- and intermediate-lived mosaics in that the percentages of tissue types for each life span group are similar, and in that a majority of the body segments are of mosaic composition. However, for the long-lived mosaics there is a definite difference in the data for the two mutants. A L 4 long-lived mosaics show almost no completely male (short-lived) body segments, and for each segment the percentage of completely female (long-lived) parts is much greater than for A L 2 mosaics. These data suggest that there may be a correlation between life span of mosaics and the location of male (AL) or female (+/AL) tissue. The short-lived A L 4 mosaics tend to have heads which are short-lived to a much greater degree than is indicated for A L 2 mosaics. As adult life span of A L 4 mosaics increases there is a corresponding decrease in the percentage of completely male body segments, markedly so for the head region. A n o t h e r comparison of the data for the two mutants shows that A L 4 produced a greater proportion of short-lived mosaics than A L 2 (49% vs. 24%) and a smaller proportion of long-lived mosaics (10% vs. 35%). The proportion of intermediate-lived mosaics was the same for both mutants (41%). Along with the data in Tables II and III these figures suggest that the two mutants have different developmental pathways in achieving their effect on adult life span. If one combines the male and mosaic areas with respect to the body s e g m ~ t ~ and length of adult life (Tables IV and V), one can see the same tre~d in :~" distribution of adult lethal tissue for all body segments and all life span lengths. with the exception of the long-lived A L 4 mosaics. There is a greater percentage of areas in which there is some expression of the y w A L 2 genome in all segments and for all life spans than there is of areas of completely wild type (female) tissue. Again, this furnishes no correlation of the action of A L 2 with a particular part of
TABLE
IV
DISTRIBUTION
OF AREAS
O F A L 2 T I S S U E IN M O S A I C S
Head
Adult lift" span Short Intermediate Long
Thorax ?
Abdomen
8"+M ~ (%)
(%)
~+M (%)
? (%)
~'+M (%)
(%)
84 57 67
15 42 33
96 76 80
4 24 2(1
65 53 56
35 47 44
~ = y w A L 2: M = m o s a i c ; 9 = wild t y p e ; d + M = c o m b i n e d y w A L 2 a n d m o s a i c a r e a s .
TABLE
V
DISTRIBUTION
OF AREAS
O F A L 4 T I S S U E IN M O S A I C S
Head
Adult life span Short Intermediate Long
Th orax
A bdom en
8'+M"
~2
(%)
(%)
~+M (%)
9 (%)
~+M (%)
? (%)
78 52 33
12 48 67
92 70 25
8 30 75
69 64 50
31 36 50
"~' = y w A L 4; M = m o s a i c ; ? = wild t y p e : 6' + M = c o m b i n e d y w A L 4 a n d m o s a i c a r e a s .
the fly's exterior. There is a conspicuous change in the trend in the expression of the y w A L 4 genome in the long-lived mosaics, most notably for the head and thorax. This trend suggests that the focus of A L 4 is oriented toward the cephalic region. A n o t h e r approach to the characterization of these mosaic flies was made. Each fly was examined and an estimate made of the total amount of y w A L 2 or y w A L 4 tissue in the whole of its external anatomy. Then a correlation was made between the life span and the total area of short-lived genotype. These data are shown in Figs. 1 and 2. In this analysis one can see a trend in the short-lived flies, with very few having only 10% male (short-lived) tissue but they progress to become the largest category in the 90% group. For A L 2 the long-lived flies, however, do not show a complementary trend, except at the 90% level. At the other levels they have essentially the same amount of adult lethal tissue in each. For A L 4 the long-lived flies do complement this trend, and none of them had more than 50% male tissue. A m o n g the progeny of the cross of ring X/y w lz females with y w A L 2 males we obtained 48 males which were y and w but not lz, and which had a mean adult life span of 10.7 _+ 1.2 days. Only two of them could be classified as intermediate in
F7~ long-lived
oo]
[]
•
mutant
intermediate - lived
801 i i
ii /
/
ii
,,'1' //
60
.
O
E
4o.
N
2o-
II
I0
25
50
75
90 patroclinous males
% of mutant tissue Fig. 1. D i s t r i b u t i o n of m u t a n t tissue in A L 2 mosaics.
F
long- lived
7
-
••
[]
intermediate - lived
mutant
I00
(/)
80
i i i i i i
i i i i i i
N.(..) .m
60
(/) 0
E 40
o~ 20
I0
25
50
7'5
90 patroclinous males
% of mutant tissue Fig. 2. D i s t r i b u t i o n of m u t a n t tissue in A L 4 mosaics.
life span, a n d n o n e w e r e l o n g - l i v e d . H e n c e t h e y w e r e c e r t a i n l y of the g e n e t i c constitution y w AL 2 and thus were patroclinous. Similar y w males were found a m o n g t h e p r o g e n y of t h e cross of ring X f e m a l e s with y w A L 4 males. T h e s e 78 m a l e s h a d a m e a n a d u l t life s p a n of 16.9 + 0.4 days, a n d all of t h e m w e r e within
the short-lived category. The probable origin of these patroclinous males is in the loss of the ring X c h r o m o s o m e very early in development, with subsequent external expression of only the X c h r o m o s o m e genes derived from the male parent. There is possibly some internal mosaicism which we could not detect. These males were sterile, since they lack a Y chromosome. When the patroclinous males are c o m p a r e d with those mosaics with the greatest (90%) proportion of male tissue (Figs. 1 and 2), they continue the trend of larger areas of male tissue being correlated with short adult life span. DISCUSSION Genetic mosaics have proven to be a fruitful tool in analyses of insect development [9-11]. They have provided examples of a u t o n o m o u s behavior of one genotype surrounded by tissue of another [12]. They have been used to explore cell lineages [11] and insect behavior [13, 14]. Hence we had expected that we could use this tool to localize the site of action of our adult lethal genes. This did not prove to be the case for the mutant A L 2, and is only suggested for A L 4. We have made the assumption that the external phenotype is a reliable indicator of the genotype of the tissues which lie underneath or are in the same general area of the organism. This assumption is based on studies of gynand r o m o r p h s in Drosophila melanogaster by Morgan [15] and in Habrobraconjuglandis [13, 16] in which the sexual behavior of a sex mosaic was correlated with the sexual genotype of tissue in the head region and not with the sexual genotype of the thorax, abdomen, or genitalia. Further evidence that such an assumption is valid was provided by Egen [17] who showed that in Habrobracon mosaics the sexual phenotype of the a b d o m e n was correlated with the sex of the underlying gonads. Ikeda and Kaplan [18] found a high, but not absolute, correlation between external genotype and the genotype of underlying neural tissue of the thorax of Drosophila melanogaster mosaics. On the other hand, Hall [14] has obtained mosaics with male heads which did not exhibit male courtship behavior and which proved to have brains of predominantly female genotype. Benzer [19] found that the external genotype of the head did not always correspond with that of the brain in his " d r o p d e a d " mutant flies. Some flies with normal heads died early, while some with mutant heads had normal life spans. In these cases the external phenotype was not an indication of the genotype of underlying tissue. Similarly in A L 2 the external phenotype may not furnish positive identification of the internal genotype. An imaginal disc may contain anlagen of both external and internal structures, but the development of these structures is autonomous with respect to the genotype of the cells involved. Thus a mosaic disc could result in external structures of one genotype and internal structures of another. A possible solution to our problem in mosaic analysis would be to employ a genetic m a r k e r for internal mosaicism, such as the enzyme acid phosphatase. If a null allele for acid phosphatase production were bred into a short-lived strain, and the
resulting strain bred into a mosaic producing system, then by suitable staining and sectioning techniques [20] the internal mosaicism could be detected, and correlations with adult life span attempted. For certain life-shortening mutations the decreased life span may be due to a localized effect on a specific organ. Such an effect occurred for the "+drop dead" locus of Drosophila where mosaics with the head as "drop dead" genotype died within 3 days ]7], and where the study had shown deterioration of the adult brain [19]. This is not the pattern observed for our A L 2 mutation whose effect is just as dramatic as "drop dead" but where there appears to be no localized region of degeneration. For this adult lethal the degeneration pattern may indeed not be localized at all but rather related to a biochemical defect that influences all cells and tissues, and hence is correlated with the amount of A L 2 tissue. The A L 4 mutant seems to give a clearer pattern with respect to localization of its effect, although from our data one cannot conclude that a specific organ is involved. The orientation toward the head region and the closer correlation of external phenotype with adult life span suggest that A L 4 is involved in a more limited system than A L 2 and that the short adult life span of AL 4 may very well have a different physiological basis from that of A L 2. We believe that experiments using mosaics for still other life-shortening genes can be important in the study of the genetics of aging.
REFERENCES 1 A.M. Clark, Genetic factors associated with aging, Adv. Gerontol. Res., I (1964) 207-225. 2 A+M. Clark and A.B. Gould, Genetic control of adult life span in Drosophila melanogaster. Exp. Gerontol., 5 (1970) 157. 3 M.B. Baird and A.M. Clark, X-ray induced life-shortening mutations in Habrobracon: A genetic approach to senescence and duration of life. Exp. Gerontol., 6 (1971) 1. 4 R. Pearl and S.L. Parker, Experimental studies on the duration of life. I1. Hereditary differences in duration of life in line-bred strains of Drosophila melanogaster. Am. Nat., 56 (1922) 174. 5 B.M. Gonzales, Experimental studies on the duration of life. VIII. The influence upon duration of life of certain mutant genes of Drosophila melanogaster. Am. Nat., 57 (1923) 289. 6 A.B. Gould and A.M. Clark, X-ray induced mutations causing adult life-shortening in Drosophila melanogaster. Exp. Gerontol., 12 (1977) 107. 7 Y. Hona and S. Benzer, Mapping of behavior in Drosophila mosaics. Nature, 240 (1972) 527. 8 Y. Hotta and S. Benzer, Genetic dissection of the Drosophila nervous system by means of mosaics. Proc. Natl. Acad. Sci. U.S.A., 67 (1970) 1156. 9 A.M., Clark, A.B. Gould and S.F. Graham, Patterns of development among mosaics in Habrobracon juglandis. Dev. Biol., 25 (1971) 133. 10 C. Stern, Gene expression in genetic mosaics. Genetics, 61 (Suppl. 1) (1969) 199. 11 P.J. Bryant and H.A. Schneidermann, cell lineage, growth and determination in the imaginal discs of Drosophila melanogaster. Dev. Biol., 20 (1969) 263. 12 C. Stern, Genetic Mosaics and Other Essays, Harvard University Press, Cambridge, Massachusetts, 1968, pp. 138-141. 13 A.M. Clark and R.C. Egen, Behavior of gynandromorphs of the wasp Habrobracon juglandis. Dev. Biol., 45 (1975) 251. 14 J.C. Hall, Control of male reproductive behavior by the central nervous system of Drosophila: dissection of a courtship pathway by genetic mosaics. Genetics, 92 (1979) 437.
l(J
15 L.V. Morgan, Composites of Drosophila rnelanogaster. Carnegie Inst. Washington Publ., 399 (1929) 225. 16 P.W. Whiting, Reproductive reactions of sex mosaics of a parasitic wasp, Habrobracon ]uglandis. J. Comp. Psychol., 14 (1932) 345. 17 R.C. Egen, A study of mosaicism in the a b d o m e n of Habrobracon juglandis. Ph.D. Thesis, University of Delaware, 1974. 18 K. Ikeda and W.D. Kaplan, Unilaterally patterned neural activity of g y n a n d r o m o r p h s mosaic for a neurological mutant ol Drosophila melanogaster. Proc. Natl. Acad. Sci. U.S.A., 67 (1970) 148(1. 19 S. Benzer, From the gene to behavior. J. Am. IVied. Assoc., 218 (1971) 1015. 20 D.R. Kankel and J.C. Hall, Fate mapping of nervous system and other internal tissues in genetic mosaics of Drosophila melanogaster. Dev. Biol., 48 (1976) 1.