Transcriptive and replicative activity of the X chromosome in an autosomal segmental hyperploid in Drosophila and its significance

It has been demonstrated, with substantial supporting evidence, that dosage compensation in Drosophila operates by hyperactivity of the X chromosome in the male (Mukherjee & Beermann, 1965; Mukherjee, 1966; Lakhotia & Mukherjee, 1969; Chatterjee & Mukherjee, 1973; Korge, l970a,b; Lucchesi, Rawls & Maroni, 1974; Maroni & Plaut, 1973a,ò). The hyperactivity of the X chromosome remains unaltered when either an X segment is translocated to an autosome (Tobler, Bowman & Simmons, 1971), or an autosomal segment interrupts the contiguity of the X as an insertion (as evidenced by early completion of DNA replication of the X in such cases; Lakhotia, 1970).

Studies on transcription and enzyme activity in normal females and males, and in metafemales and metamales (Lucchesi et al. 1974; Lucchesi, Belote & Maroni, 1977), have provided evidence in favour of a role for some autosomal factors in the determination of hyperactivity of the X chromosome in male Drosophila. The discovery of the action of autosomal sex-specific lethal mutants has provided strong evidence in support of autosomal control (Belote & Lucchesi, l980a,ò; Cline, 1983).

Yet, evidence against exclusive autosomal control has come from results on the X-chromosome activity in different Sxl alleles (Lucchesi & Skripsky, 1981; Cline, 1983) and also from those on transcription in segmental hyperploids (Prasad, Duttagupta & Mukherjee, 1981), and from studies on sex-mosaics in hyperploids (Prasad-Sinha & Mukherjee, 1985). More direct evidence in favour of X-autosomal dual control has been claimed by the transcription pattern in the In(1) BM² (reinverted) strain where the X chromosome is nearly four times as hyperactive and it is suppressible by a duplication of 16A–20D (Ghosh & Mukherjee, 1983; Mukherjee & Ghosh, unpublished data).

We have examined the transcription and replication patterns in different segments of the X chromosome and that of an autosomal segmental hyperploid in which a piece of the autosomal segment is translocated to the X chromosome in addition to the normal two doses. Results of this investigation should help us to determine whether or not a change of position and dosage in a given chromosome region leads to a concomitant change in the transcriptive and replicative behaviour of the X-chromosomal hyperactivity in male Drosophila, and whether chromosomal pairing of the autosomal hyperploid has any bearing on the activity of the X chromosome.

The translocation stock used in these experiments was T(1,3) O5 D/+; yf := of Drosophila melanogaster. The duplication-bearing males Dp(3;1)05 obtained from this stock carry a duplicated segment of autosome (88B—92A, 3R), one extra dose being inserted between 4F and 5A of Bridges’ map of the X chromosome while the other two doses (88B—92A) are normally present on the 3R segment (Fig. 1A,B). These males are maintained with attached X females (Fig. 2A). TO obtain segmental autosomal hyperploid female, Oregon R⁺ virgins were crossed to Dp (3;1)05 D/+ males and the cross yielded daughters having three doses of the 88B–92A fragment (Fig. 2B).

Flies were reared on standard Drosophila food at 25 °C. Larvae for cytological observations were grown on food supplemented with yeast. Salivary glands were dissected out from third instar larvae (either male or female) of the duplication strain in Ringer (pH 7.2). To measure the degree of RNA synthesis on the X chromosome and on the autosome, glands were labelled with [³H]uridine (500µCiml⁻¹, sp. act. 9700 mCi mmol⁻¹; obtained from Bhabha Atomic Research Centre, Bombay) for 10 min, and, for replication, glands were incubated in [³H]thymidine (400µCiml⁻¹, sp. act. l8800mCimmol⁻¹, obtained from BARC, Bombay) for 2Omin. After labelling (with either [³H]uridine or [³H]thymidine) glands were fixed in 1:3 (v/v) acetic acid/methanol and squashed on 50% (v/v) acetic acid; coverslips were removed and autoradiography was performed by the usual technique with Kodak AR 10 stripping film (Lakhotia & Mukherjee, 1969). The autoradiograms were exposed for 15–16 days and developed in Kodak Dl9b. Preparations were stained with 1% Toluidene Blue. Oregon R⁺ males and females and T(1,3) 0.5 males were used as controls for the duplication males and females, respectively, labelled with either [³H]uridine or [³H]thymidine at the same concentrations, and processed for autoradiography using the same procedure. A minimum of three silver grains over a given chromosomal segment was deemed necessary to indicate significant labelling. Photographs were taken under oil-immersion objective (×100) in a Zeiss Photomicroscope III with bright-field transmitted or phase-contrast illumination.

In the present investigation RNA synthesis on larval salivary gland chromosomes has been examined autoradiographically in an autosomal segmental hyperploid (88B—92A) present as a duplication inserted in the X chromosome of D. melanogaster, and compared with two controls, Oregon R⁺ and X–3 translocation T(1,3)0.5 with two doses of the segment 88B—92A. The intensity of silver grains due to [³H]uridine was computed from actual grain counts on different parts of the X chromosome and the autosomal segment. The data (Table 1) reveal that the relative intensity of grains on 1A–4F of the X chromosome, i.e. the means of ratios of the number of silver grains on the X-chromosomal part and that on the part of the 2R (49E–56EF) of the corresponding nuclei is 0.60 ± 0.04 on Oregon R⁺ male (Fig. 3A), 0.66 ± 0.02 in the T(1,3)0.5 X-3rd translocation male with two doses of 88B—92A (Fig. 3B) and 0.80±0.04, in the hyperploid male nuclei in which the two normal segments were not paired with the inserted 3R segment (Fig. 3c); 0.95 ±0.07, in those hyperploid male nuclei in which one inserted and one normal 88B-92A segment were paired while the third dose was free (Fig. 3D); and 0.92 + 0.08, in those in which all three doses of 88B-92A were paired together (Fig. 3E). The differences between the means of ratios of grains (1A–4F/49E-56EF) in the hyperploid males under all three conditions of pairing and that in the Oregon R⁺ male are significant (P<0.0l, by/⁷ test), while the ratios in the –3 translocationbearing T(1,3)0.5 male with two doses of the segment and the Oregon R⁺ male are not significantly different. Comparison of the X/2R ratios for the segment 5A–2OF also reveals the same picture (Table 1). In the female, the corresponding ratios of grain numbers for 1A–4F and 5A–20F among hyperploids and Oregon R⁺ are not significantly different. The [3H]uridine labelling of the Oregon R⁺ female X chromosome and the X chromosome with autosomal insertion under different conditions of pairing is shown in Fig. 4. On the other hand, the mean 88B–92A/49E-56EF grain ratios in Oregon R⁺ male (0.86 ± 0.04) and female (0.75 ± 0.03) are significantly less than in the duplicated strain. The increased ratios in the hyperploid male and female are nearly 1.5-1.7 and T5-T9 times (depending on the condition of synapsis) that in the Oregon R⁺ male and female, respectively. Thus, these data clearly support a dosage effect for the autosomal segment (expected increase for dosage effect being T5 times) regardless of whether the segment is present in the X chromosome or the autosome. The relative incorporation of [³H]uridine on the non-duplicated segment 96EF-100DE of the 3R has been analysed as the 3R/2R grain ratio and this serves as an internal control. The results show that there is no significant difference in the labelling intensity in the 96EF-100DE region in Oregon R⁺ and hyperploid males and females (Table 1).

These findings imply that the addition of an autosomal piece to a diploid male (1×2A) tends to enhance the hyperactivity of the X chromosome in the male, but has no effect in the female. Furthermore, it indicates that the pairing condition of the autosomal segment has apparently no effect on the gene activity of the X chromosome.

The DNA synthesis of Dp(3;1)05 male and female larvae has been compared with that of Oregon R⁺ male and female larvae. The representative autoradiograms of [³H]thymidine labelling in the Dp(3;1) female bearing the duplication is shown in Fig. 5A,B. For a comparison of the replicative behaviour, the labelling frequencies of the segments 1A-8ABC on X and 56EF-6OD on 2R were scored. Results presented in Table 2 clearly establish that the labelling frequencies of most of the sites on the X chromosome of the Dp(3;1) males are less than those on that of the Oregon R⁺ males and even much less than those on that of the Oregon R⁺ females. In contrast, the labelling frequencies on 2R are highly comparable in both Oregon R⁺ and Dp(3;1) male and female. A comparison of the intranuclear variation in [3H]thymidine-labelling frequencies is presented in Fig. 6. Clearly both Oregon R⁺male X and Dp(3;1) male X show a polynomial regression fitted to the equation: Y = a+bX+cX², in contrast to linear regression for the Oregon R⁺ female X and the duplication-bearing female X chromosome. The deviation from linearity for the two types of males is significant and, as resolved by the F values, the difference between the two males is also significant. Such a comparison strongly attests to the fact that the increased transcriptional competence of the X chromosome in the autosomal segmental hyperploid is decided at the template level and that the X chromosome of such a duplication-bearing complement replicates even earlier. Forty five different sites (see Lakhotia & Mukherjee, 1970) on the X chromosome (1A–12DE), 20 sites (delineated according to Nash & Bell, 1968) on the autosome 2R (56F—6OF) and 18 sites on the duplication-bearing segment (88B—92A) were chosen for grain counting. To minimize the well-known difficulties of autoradiography and cytology (overlapping grains, dense packing of bands within the duplication loop of Dp(3;1)05, D/+ chromosomes etc.), only the nuclei at the terminal phase of replication (2D-ID pattern) were taken into account for a replication study of both the control (Oregon R⁺) and the duplication strain. The labelling-intensity profile based on the ratio of silver grains on the specific replication units (1A, IB, IC etc…l2DE) of the X chromosome to that on 56F of the autosome 2R (of the same nuclei) reveals that while more than 50% of the 45 sites of the X chromosome have a ratio greater than 0.4 in the female, over 80% of the sites have a ratio of less than 0.4 in the male in both Oregon R⁺ and Dp(3;1) (Fig. 7A,B). NO such differences could be detected for the autosomes.

However, no difference in the intensity profile is detectable for the X chromosome within the sexes. Thus, these data on the frequency of labelling and intensity profile taken together suggest that as the intensity is the same for the duplication-bearing male and the control male, and the frequency is considerably lower in the duplication male, the autosomal segmental hyperploidy does influence the level of activity of the X chromosome in the male. And, therefore, although the template form of X in the presence of autosomal duplication is indeed augmented, the replicative competence (as resolved by intensity profile) remains unaltered.

The results on the transcription and replication of the autosomal hyperploid strain presented here reveal that the addition of a piece of the third chromosome to an 1×2A genome enhances the hyperactivity and the template state of the male X chromosome.

On the other hand, the duplicated segment of the autosome (3R) shows a dosage effect as expected, while a non-duplicated segment of the autosome (96EF—100DE) of the duplication strain reveals no change from the normal. The transcriptive activity of the X chromosome and of specific segments of an autosome (88B-92A) has been analysed under various conditions of pairing between the duplicated segment of the autosome and its homologue in the genome. The result indicates that the transcriptional activity is independent of pairing.

Our results regarding transcription in a general way are in agreement with the earlier findings by Maroni & Plaut (l973a,ò) and Lucchesi et al. (1974, 1977), which suggest the presence of regulatory factors of autosomal origin that might control the hyperactivity of the male X chromosome in Drosophila. However, since there is considerable evidence now that suggests that transcription of the X chromosome may be under dual regulatory control (Cline, 1983), it is conceivable that dislocation of the contiguity of the regulatory signals that may be interspersed on the X chromosome (Prasad-Sinha & Mukherjee, 1985; M. Ghosh & A. S. Mukherjee, unpublished data) may interfere with the realization of hyperactivity of the male X chromosome. In contrast, the present results reveal that the hyperactivity is not lowered due to the transposed segment of the autosome, but rather the autosomal hyperploid condition enhances it, as expected. In the case of diplo-autosomal translocation (T(X-3)05D) male the hyperactivity is maintained as in Oregon R⁺. These facts rule out the possible position effect at the site of insertion on the X chromosome. Furthermore, as shown earlier by Lakhotia (1970) and as evident from the results, the X-chromosomal sections and the autosomal inserted segment manifest their own levels of activity, respectively, right from the site of insertion and there is no spreading effect. It may also be pointed out that there is no sex-related or sex-specific gene known to be located at the site of insertion, namely 4F-5A. The well-known dominant sex-specific lethal (Sxl⁺) is at 6E1-7B7 (Belote & Lucchesi, 1980a). It may therefore be surmised that the X-chromosomal organization is predisposed by certain interacting regulatory genes.

Analysis of replication as done by the sitewise examination of the [³H] thymidine labelling of the X chromosome and the duplicated part of the autosome reveals, as reported earlier by Lakhotia (1970), that while no change in the characteristic replication of the autosome is evident, the autosomal hyperploidy tends to enhance the template state of the X chromosome by an even earlier replication than euploidy.

In all earlier papers from this laboratory (Chatterjee & Mukherjee, 1973), it was reported that in whatever genetic situation, when the hyperactivity was altered, replication of DNA on the X chromosome was altered accordingly. For example, in different species of Drosophila, hyperactivity of the X chromosome in the male was always concurrent with the early completion of DNA replication of the X chromosome (Mukherjee & Chatterjee, 1975, 1976; Lakhotia & Mukherjee, 1970; Das, Mutsuddi, Duttagupta & Mukherjee, 1982).

No such observation has been made in metamales, but studies on replicative behaviour in metafemales have also confirmed this finding. In segmental hyperploids this parallelism has always been reinstated (Prasad-Sinha & Mukherjee, 1985). In hyperploids in which long segments from proximal ends of the X chromosome have been added* early completion and hyperactivity appear to be maintained up to the addition of 0.62 length of the X (Mutsuddi, Mutsuddi, Duttagupta & Mukherjee, 1983), but beyond that hyperactivity is abolished and the X-chromosomal replication becomes synchronous. Recently, Goldman et al. (1984) convincingly established that genes that can be transcribed must replicate early, and the inactive genes must replicate and be transformed into a transcriptionally competent template form in order to be in the active state. The data presented here corroborate this relation between higher transcriptional competence and early replication.

It is conceivable on the basis of these observations that there is a parallel between enhanced transcriptional competence of the X chromosome in the male and its early completion of replication. However, the distinctly different regression (Fig. 6) for the Oregon R⁺ and Dp(3;1) males, in contrast to the females, strongly suggests that the template forms resulting from the regulatory influence might be set differently in the two sexes.

What we therefore propose is that the autosomal signals do not interfere with the inherent primary modulation in the X chromosome. The autosomal signals act on the modulated organization of the X chromosome, guided by a modulator gene complex located on the X chromosome (Ghosh & Mukherjee, 1983; and unpublished) at a second step, evoking the higher activity of the male X chromosome. So an autosomal hyperploidy makes the male X more hyperactive. The findings of the present study lead us to conclude that a certain autonomously acting control system is inherent in the X chromosome, which may be influenced by the dosage of the autosome only at a second step preceding the hyperactivity. The results further indicate that the gene activity of the X chromosome is not influenced by the pairing of the autosome.

This work was supported by University Grants Commission Special Assistance to the Zoology Department and S.G. was a Junior Research Fellow in a pre-doctoral fellowship sponsored by the University Grants Commission, which we acknowledge with gratitude.