Fig 1.
Mutations in the bithorax complex used in this study.
The top line shows a scale map of the region of the Drosophila bithorax complex from abd-A to Abd-B with the parasegment specific cis-regulatory domains bracketed above the line. The exonic structure of the abd-A and Abd-B primary transcripts are shown as broken arrows below the line, connected with lines that indicate their splice patterns. Shown as a thinner broken arrow under the bithorax complex DNA is the most prominent splice version of the iab-8 ncRNA. The expected msa transcript is shown beneath the iab-8 transcript. Under the primary transcripts, are some of the mutations used in this work. Deletions are indicated by () separating two horizontal lines. The msa inversion is also shown and labeled accordingly. The arrows above the bithorax map indicate the positions of chromosomal rearrangement breaks used in this work. The chromosomal breaks are label and are color-coded (red for breaks that do not complement Df(P9) in the accessory gland, and blue for breaks that complement the Df(P9) in the accessory gland).
Fig 2.
A lncRNA is plays a role in the development of the accessory gland.
Accessory gland phenotypes associated with BX-C mutations. In each case, a chromosome (here called AGFP) containing a BAC transgene containing the Abd-B regulatory sequence driving Gal-4 expression specifically in the secondary cells and a UAS-nGFP transgene was used to visualize the secondary cell cytoplasm and nuclei (more concentrated GFP). Vacuoles can be seen as black spaces within the cell. Genotypes are labeled above each panel. A. A wild-type accessory gland. B. iab-6cocuD1/ iab-6cocuD1. The rest of the panels are hemizygous for the BX-C mutants described in Fig 1: C. iab-611/Df(3R)P9, D. iab-4186A/Df(3R)P9, E. Fab-864/Df(3R)P9 and F. iab-386A/Df(3R)P9. Note the smaller vacuoles, similar to B. (iab-6cocuD1/ iab-6cocuD1) in C. (iab-611/Df(3R)P9) and D. (iab-4186A/Df(3R)P9). Scale bar equals 50μm. A’.-F’. are enlargements of individual secondary cells from the A.-F. White arrow heads mark the cells that were enlarged.
Fig 3.
Detection of miR iab-8 using a GFP sensor.
Dissected accessory glands from A. a fly carrying an mCherry reporter in the msa second exon (exon2-mCherry), B. a wild type fly carrying a mir iab-8 GFP sensor and C. an iab-6cocuD1 homozygous fly in the presence of a mir iab-8 GFP sensor (22). C’. is an enlargement of the box shown in C. In A., mCherry is visualized directly, while in B., C. and C’. GFP is visualized. B. In wild-type AGs, secondary cells appear as non-GFP expressing holes in the plane of GFP-expressing main cells (examples shown with white arrows). C. In the iab-6cocuD1 homozygous mutants, the secondary cells express GFP (examples shown with orange arrows). B’ is an enlargement of the orange box in B. The enlargement of the secondary cells allows for the visualization of the abnormal morphological phenotype (lack of large vacuoles) of iab-6cocuD1 secondary cells in the sensor line. Scale bar equals 50μm.
Fig 4.
Loss of the iab-8 miRNA and/or Abd-B leads to defects in vacuole formation.
Accessory glands are shown from three to four-day old virgin males that express the Abd-Gal4, UAS-GFP reporter and carry the following mutations: A. iab-6cocuD1/+ (control), B. iab-6cocuD1/ iab-6cocuD1(MSA-, Abd-B-), C. iab-6cocuD1/ΔmiR-iab-8 (miRNAiab-8-) and D. iab-6cocuD1/Abd-BD16 (Abd-B-). Below panels A.-D. are the median and mean vacuole sizes for each genotype along with the standard deviation for the mean vacuole size and the number of vacuoles measured (n). Based on Kruskal-Wallis one-way analysis of variance, followed by post-hoc Dunn’s tests, the differences in values is significant between all groups (p < .0001 for all pairwise comparisons except iab-6cocuD1 homozygotes vs iab-6cocuD1/miR-iab-8 mutants, where p = .0085. Figures E. and F. show AGs stained with antibodies to ABD-B (green), imaged using the same settings. In F., the AG is from iab-6cocuD1/Abd-BD16 males (where Abd-B should not be expressed), and in E. the AG is from a sibling control of the cross, iab-6cocuD1/Dp(P5) (where two wild-type copies of Abd-B should be expressed). An enlargement of a single secondary cell is present in each panel to show the difference in Abd-B expression level in the two lines. Scale bars = 25μm for A.-D., and = 50μm for E. and F.
Fig 5.
CrispR-mediated deletions and inversions of msa confirm the role of ABD-B in the accessory glands.
A. Accessory glands from iab-3,6CR6YO1M20M2 (msa deletion) / iab-6cocuD1 in the presence of the Abd-B-Gal4, UAS-GFP reporter show a phenotype like that of iab-6cocuD1 homozygotes. B. Accessory glands from iab-3,6CR6YO1M20M1 (msa inversion) / iab-6cocuD1 in the presence of the Abd-B-Gal4, UAS-GFP reporter show a mixed phenotype with some cells resembling iab-6cocuD1 homozygotes and other cells being less affected. This can be seen more easily in B’., which is an enlargement of the rectangle indicated in B. C. Accessory glands from iab-3,6CR6YO1M20M2 (msa deletion) / iab-6cocuD1 in the presence of the iab-8 miRNA sensor shows that the iab-8 miRNA is not present. D. Accessory glands from iab-3,6CR6YO1M20M1 (msa inversion) / iab-6cocuD1 in the presence of the iab-8 miRNA sensor show some cells expressing the miRNA while others do not. D’. shows an enlargement of the rectangle present in D. with two secondary cells traced with a dashed blue line). F. and F’. are phase contrast images the same glands shown in D. and D’. From D’. and F’. one can see how cells expressing the miRNA retain vacuoles visible by phase contrast (as a rougher appearance), while the cells not expressing the miRNA do not (resembling iab-6cocuD1 homozygotes). E. ABD-B (green) and DE-Cadherin (red) immunostaining of glands from iab-3,6CR6YO1M20M1 (msa inversion) / iab-6cocuD1. ABD-B is absent in this mutant line. Scale = 50μm.
Fig 6.
miR iab-8 loss-of-function causes defects in the LTR.
Two different fertility fecundity assays were used to assess the LTR of mutants lacking the iab-8 miRNA: (A. and B.) egg-laying over the 10 days post-mating and (C. and D.) receptivity after four days post-mating. The ΔmiR-iab-8 mutation was tested over two different iab-6cocu alleles to overcome possible effects of genetic background iab-6cocuD1(D1) or iab-6cocuD5 (D5). For the 10-day egg laying assay, in A., the D1 curve (in blue, n = 15) refers to iab-6cocuD1homozygous males, the D1Resc curve (in red, n = 18) refers to iab-6cocuD1/ iab-5,6rescue males, where iab-5,6rescue is a chromosome made with the same InSiRT platform [70] used to make the iab-6cocu mutations but with wild-type sequence added in its place. The mirResc curve (in green, n = 22) refers to ΔmiR-iab-8/ iab-5,6rescue males and the mirD1 curve (in purple, n = 16) refers to iab-6cocuD1/ ΔmiR-iab-8 males. In B., the D5 curve (in blue, n = 18) refers to iab-6cocuD5 homozygous males, the D5Resc curve (in red, n = 22) refers to iab-6cocuD5/ iab-5,6rescue males, the mirResc curve (in green, same as in A., n = 22) refers to ΔmiR-iab-8/ iab-5,6rescue males and the mirD5 curve (in purple, n = 21) refers to iab-6cocuD5/ ΔmiR-iab-8 males. In both cases there is a significant drop in egg laying in mates of either iab-6cocu mutant or in mates of males transheterozygous for either iab-6cocu mutant and the ΔmiR-iab-8 chromosome (rmANOVA p>0.001 relative to controls). The receptivity assay was performed on the same genotypes. In C. and D., sample sizes are as follows, D1 (n = 19), D1 Resc (n = 18), mir Resc (n = 20) mirD1 (n = 20), D5 (n = 19), D5 Resc (n = 20), mir Resc (same as in C., n = 20) and mirD5 (n = 20). In both cases, there is a significant increase in remating (p < .05) by mates of either iab-6cocu mutants or in mates of males transheterozygous for either iab-6cocu mutant and the ΔmiR-iab-8 chromosome (* indicates p < .007 vs D1 res, p≤.02 mir Res and p < .001 D1 relative to controls, ** indicates p < .001 relative to controls, as accessed by the Wilcoxin Ranked Sums Test). E. Shows extracts from single AGs run on Western blots and probed with antibodies against CG1656 or CG1652. Below these blots are images of the same blots stripped and reprobed with a loading control antibody against the main cell protein, Acp36DE. The genotype of the flies from which the AGs were dissected are indicated above each lane and the antibodies used indicated on the left. F. Shows western blots of extracts from wild type or mutant AGs probed with antibodies against CG1656. The extracts on the left were treated with PGNase F, while those on the right were not. Genotypes of the flies used for the extracts are indicated above each lane. Below the blot is an image of the same blot stripped and reprobed with a loading control antibody against the main cell protein, Acp36DE.
Fig 7.
Known miR-iab-8 target abd-A is not overexpressed in iab-6cocuD1secondary cells.
ABD-A staining in wild-type or mutant secondary cells. Accessory glands from A.-C. CantonS (wild-type), or D.-F. iab-6cocuD1/ iab-6cocuD1 males immunostained for ABD-B (in green A. and D.), ABD-A (in red, B. and E.) and DAPI (in the merged image in blue C. and F.). Scale = 50μm.
Fig 8.
UBX and EXD level do not change by detectable levels in iab-6cocuD1 mutant accessory glands.
UBX and EXD staining in wild-type or mutant secondary cells. Genotypes and detected proteins are labeled on the image. Accessory glands from CantonS (wild-type) (A.-D. and I.-L.), or iab-6cocuD1/ iab-6cocuD1 (E.-H. and M.-P.). Panels A.-H. are maximum intensity projections of glands stained for UBX (in green (A.,D. (merge of A.-C.),E. and H. (merge of E.-G.))), DE-Cadherin, an apical junction marker to show cell outlines (in red (B., D. (merge of A.-C.), F. and H. (merge of E.-G.))) and DAPI (in blue (C., D. (merge of A.-C.), G. and H. (merge of E.-G.))). Panels I.-P. are single slices from a Z-stack of glands stained for EXD (in red (J.,L. (merge of I.-K.), N. and P. (merge of M.-O.))), Disc-Large, a basal lateral cell junction marker to show cell outlines (in green (I., L. (merge of I.-K.),), M. and P. (merge of M.-O.))) and DAPI (in blue (K., L. (merge of I.-K.),), O. and P. (merge of M.-O.))). Arrows point to representative secondary cells (identified based on their characteristic pattern of DE-Cadherin staining) in each merged image. Scale = 50μm.