We are continuing to investigate the control of
her-1 expression, which results in a rare larger transcript encoding the HER-1A protein, necessary and sufficient for masculinization, and a moderately abundant smaller transcript of unknown function but with the potential to encode a HER-1B protein representing the C-terminal portion of HER-1A. These transcripts are driven by separate promoters, named P1 and P2, respectively (1), both under negative control by the sdc genes (2), which in turn are negatively regulated by
xol-1 X and also are responsible for turning on dosage compensation in XX animals (see Ref. 3 and others from the Meyer lab). We have recent results consistent with the possibility that a
her-1 gene product may feed back to help regulate dosage compensation. In experiments related to those reported by Perry et al. (WBG 14(4): 60, 1996), we had independently noted that some constructs including the
her-1 P2 promoter region, when carried as transgenic arrays in a
him-8 strain, cause extensive embryonic lethality and a preponderance of males among surviving progeny, suggesting XX-specific lethality. In further experiments to understand this effect, we have obtained the following preliminary results, each in multiple independent transmitting lines. a) A P2-HER-1B construct including 3.4kb of promoter region, the HER-1B coding sequence, an HA tag sequence, and 0.4kb of 3' flanking region, causes the above described apparent XX lethality in
him-8 transgenic lines. Unlike the effect described by Perry et al., we observe the skewed sex ratio primarily in the F2 and subsequent generations following injection. In
xol-1 mutant lines, this construct causes much less embryonic lethality and also substantial masculinization of the
xol-1 XX animals, which is dependent on presence of a functional
her-1 gene. When mated to N2 males, hermaphrodites from these lines produce viable, mating XO male progeny, i.e.,
xol-1 XO lethality is strongly suppressed. b) We have begun similar experiments with a P2-HER-1B construct lacking the HA tag sequence but otherwise identical to the above. It appears to cause more severe XX lethality in
him-8 strains (transmitting lines have not yet been recovered); it causes no apparent masculinization in
xol-1 lines; but these animals produce viable mating
xol-1 XO cross progeny. c) P2 promoter-only constructs (P2::lacZ and P2::GFP) with the same promoter region as above but producing no
her-1 transcripts cause little apparent XX lethality in
xol-1(+) lines and extensive masculinization of
xol-1 lines, including production of mating XX males, in F2 and subsequent generations. Thus the effects of P2 promoter-only arrays resemble those of weak or Tra sdc mutations, which result in low XX lethality and production of masculinized
xol-1 XX animals, while the effects of P2-HER-1B arrays resemble those of some strong sdc mutations such as
sdc-3 nulls, which result in high XX lethality and no masculinization (see Ref. 4 and references therein). These differences suggest that presence of the small
her-1 transcript influences the effects seen. One possible model to explain our results so far is as follows: P2 promoter-only constructs include DNA sequences that in high copy number can titrate a negative regulator of
her-1, such as maternally and embryonically produced SDC-1. P2-HER-1B constructs embryonically express the small
her-1 transcript, which acts somehow to repress activation of dosage compensation, perhaps via an effect on SDC-3 (3). The differences seen between the HA-tag and no-HA-tag constructs could result from less repressing activity when the tag sequence is present. The lack of masculinization by the no-HA-tag construct could result from the stronger interference with dosage compensation, leading to repression of masculinization (presumably via repression of
her-1), as observed earlier in genetic experiments (4). Such a function for the small transcript could possibly serve as a backup mechanism for repressing dosage compensation and for maintaining an appropriate level of the large
her-1 transcript in normal XO animals. One prediction of this model is that production of
her-1 transcripts from the gf allele
her-1(
n695) [which is not well repressed by the sdc genes (2,6)] might allow survival of
xol-1 XO animals by repressing dosage compensation. To test this possibility, we constructed
him-8(
e1489);
her-1(
n695);
xol-1(
y9). Analysis of this strain is still in progress; however, it is viable, producing some dead embryos, arrested larvae, and adults which are almost all partially to fully masculinized, including a high percentage (not yet accurately determined) of mating males. Based on (a still small number of) single-male matings, some of these are XX as expected (5), producing only a few XO male progeny (from nullo-X sperm resulting from
him-8-induced nondisjunction). However, some appear to be XO, producing a much higher proportion (not yet accurately determined) of XO mating males among their cross progeny. If the triple mutant is indeed producing XO males, then
her-1(
n695) suppresses
xol-1 XO lethality. More definitive tests of this suppression are in progress. These results together suggest that one or more
her-1 gene products may feed back on dosage compensation, possibly by regulating functions of SDC proteins. Further experiments are in progress to address some of many remaining questions, including whether HER-1A or its transcript are involved, whether translation of the small
her-1 transcript occurs and is necessary, and whether SDC proteins are being directly regulated. In addition, we are using the masculinization of
xol-1 XX animals as a preliminary assay for segments of the promoter region that interact with negative regulators of
her-1. References: 1) Perry et al. (1993) Genes & Dev. 7:216; 2) Trent et al. (1991) Mech. Dev. 34:43; 3) Davis and Meyer (1997) Devel., in press; 4) DeLong et al. (1993) Genetics 133:875; 5) Miller et al. (1988) Cell 55:167; 6) Perry et al. (1994) Genetics 138:317.