Some DNA Sequences Associated With Mitotic Heterochromatin Are Underrepresented in Polytene Chromosomes

The notion that the DNA of both the Y and pericentric regions of mitotic chromosomes is underrepresented in polytene chromosomes dates back to Heitz's pioneering cytological studies of heterochromatin around 1930 (see ref. 19). However, direct evidence for such underrepresentation in D. melanogaster was first provided by Rudkin in the 1960s (68,165). Using Feulgen-DNA cytophotometry, Rudkin found that the successive replicative classes of salivary gland polytene nuclei of young (0-72 h) larvae contained significantly less DNA than 2n-tuples of 2C expected for complete genome replication. Moreover, the deviations he observed were greater for males than females, consistent with males having a heterochromatic Y chromosome. It should be noted that two later Feulgen cytophotometry studies yielded results at odds with Rudkin's data. Dennhofer (168) reported that the DNA levels in small polytene nuclei (8C-128C range) of salivary glands were exact doublings of the diploid values she measured in brain nuclei, and Lamb (173) found no evidence of sequence underrepresentation in polytene nuclei of midgut and Malpighian tubule cells of adults (see Table 3). However, the balance of experiments addressing this question favor Rudkin's observations and conclusions, as discussed next.

Compelling evidence for underrepresentaion of satellite DNA in Drosophila polytene nuclei was first provided by Gall and colleagues (58). Using the technique of CsCl equilibrium centrifugation, they found that the 1.688 satellite band (III) present in DNA isolated from diploid tissues (imaginal discs and brains) was "almost undetectable" in salivary gland DNA. Consistent with this result, they also found that in situ hybridization to this satellite produced signals of nearly equivalent intensities in mitotic and polytene nuclei, suggesting that both types of chromosome contain similar amounts of 1.688 satellite despite having vastly different overall ploidies. It has since been shown that both dodeca satellite (48) and the AAGAC repeat of satellite II (160) are also underrepresented in salivary gland polytene chromosomes, and sequences corresponding to the 1.705 satellite are underrepresented in nurse cell polytenes (181).

Further evidence for underrepresentation of heterochromatic sequences of polytene chromosomes includes the following: (1) some transposable elements in heterochromatin were found not to be polytenized in salivary gland chromosomes (see refs. 60, 62,141); and (2) a gradient of polytenization of >50-fold was detected at the euchromatin-heterochromatin (E-H) junction of a minichromosome (Dp1187) in salivary nuclei (182).

P element constructs inserted into mitotic heterochromatin represent unique sequence DNAs whose ploidy in polytene chromosomes can be estimated by quantitative Southern blotting. Zhang and Spradling (183) found that 15 of 15 PZ elements (P[ry+, lacZ]) distributed over much of YL were underrepresented by at least 20-fold in salivary gland DNA compared to DNA of adult males (the latter have a mix of diploid and low-level polytene nuclei). Furthermore, 13 of those inserts were undetectable by FISH analysis on salivary gland polytene chromosomes. These results argue that much of the Y chromosome is not polytenized in salivary gland nuclei.

However, not all DNA in heterochromatin is underrepresented. For example, at least three unique-sequence genes located in P-heterochromatin—light, rolled, and suppressor of forked—are extensively polytenized in salivary gland chromosomes (141,184,185), and the pericentric AAGAC satellite repeat is polytenized in pseudo-nurse-cell polytene chromosomes (160).

Zhang and Spradling (183) found that 16 of 16 PZ inserts in pericentric het-erochromatin of chromosomes 2 and 3 were fully polytenized and produced FISH signals in the chromocenter. Some FISH signals were like bands, similar to those in euchromatin, whereas others appeared as dots or covered relatively large areas of the chromocenter while remaining intensely fluorescent. Did the PZ elements integrate into heterochromatin domains that normally undergo polytenization or did the very presence of the insert, euchromatic as it is, stimulate polytenization where it does not normally occur? The answer to this question remains unknown. The mere presence of a PZ insert in heterochromatin does not guarantee its polytenization, as shown by the Y chromosome inserts mentioned earlier. Furthermore, polytenization in the 16 autosomal lines tested was not restricted to just the PZ inserts because middle repetitive DNAs flanking the inserts were also fully polytenized, although, in such cases, a "coattail" effect of the PZ insert causing adjacent sequences to polytenize cannot be excluded.

Taken together, these results reveal a complex state of polyteny in the salivary gland chromocenter, with some heterochromatin-associated DNAs under-represented (e.g., most satellite DNAs and much or all of the Y chromosome) and some others more or less fully represented in dispersed polytenized domains.

It is generally assumed that the nonpolytenized sequences of the chromocenter give rise to a-heterochromatin and the polytenized regions to P-heterochromatin. Looping out of DNAs from interspersed polytenized domains and ectopic contacts between them are proposed to generate the reticular morphology of P-heterochromatin (see refs. 74,160,184,186). Such ectopic contacts may be stimulated by DNA breaks in heterochromatin (187,188). Koryakov et al. (160) reported finding no obvious cytological differences in the a-heterochromatin between X0, XY, XYY, XX, and XXY animals, which led them to suggest that the Y chromosome does not contribute substantially to a-heterochromatin formation. Given that the Y chromosome is heavily laden with satellite DNAs, this unexpected observation would challenge current models of chromocenter organization, so that further critical studies are required to address this issue.

0 0

Post a comment