loxP 5xUAS
P-transposase promoter
18
Results
Generation of RNAi strains against OdsHsim and OdsHmau in Drosophila melanogaster
The mir-6-1-based RNAi constructs, pV22_OdsHsimAB and pV22_OdsHmauAB, were introduced into D. melanogaster y1 sc1 v1 P{nos-phiC31\int.NLS}X;
P{CaryP}attP2 embryos. For the pV22_OdsHsimAB construct, 63.1% of hatched larvae eclosed, 82.9% of the adults were fertile, and 26.5% of the fertile adults were
transformants. For the pV22_OdsHmauAB construct, 60% of hatched larvae eclosed, 70.8% of the adults were fertile, and 29.4% of the fertile adults were transformants (Table 3). Two UAS-OdsHRNAi strains were therefore generated and named
OdsHRNAi_simAB and OdsHRNAi_mauAB.
RNAi knockdown rescues the OdsHmel ectopic expression phenotype in Drosophila eye
A total of six UAS-OdsHRNAi strains, including two newly generated and four obtained form stock centers, were investigated in the GMR-Gal4>UAS-OdsHmel background. Observations on Drosophila eye phenotypic changes demonstrate that all the six RNAi strains yield the knockdown effect. Compared to the wild-type eyes (Figure 5, A-A′), GMR-driven OdsHmel caused severe eye defects: reduction in size, and loss of ommatidia and bristles (Figure 5, B-B′). The two VDRC and two TRiP lines restored the eye defects evidently (Figure 5, C-F′). All of them displayed a
wild-type-like eye of normal size. However, only three of them showed regular
ommatidia and bristles (Figure 5, D-F′), except for the strain GD51289, which produced fused ommatidia and disorganized bristles (Figure 5, C-C′). The newly synthesized
19
OdsHRNAi_simAB and OdsHRNAi_mauAB, whose guide and passenger strands are based on the coding sequences of OdsHsim and OdsHmau, showed minor but palpable knockdown effect. In the GMR-Gal4>UAS-OdsHmel background, OdsHRNAi_simAB and OdsHRNAi_mauAB partially restored not only the size of compound eyes but also the growth of ommatidia and bristles (Figure 5, G-G′ and H-H′). Measurements of eye area and circumference provide quantification results in strong agreement with phenotypic observations that the four strains from stock centers had higher efficiency than the two self-made (Figures 6 and 7).
To further confirm the rescue effect, I used RT-PCR to detect the OdsHmel mRNA level of the above-mentioned strains. The result showed that the VDRC and TRiP stocks obviously reduced OdsHmel expression towards the wild-type (w1118 and GMR-Gal4) level. It is expected that the decrease of OdsHmel mRNA was ambiguous in
OdsHRNAi_simAB and OdsHRNAi_mauAB, whose RNAi constructs were designed according to OdsHsim and OdsHmau coding sequences, because OdsH has experienced rapid evolution and diverged at the sequence level (Figure 8).
20
Table 3. Summary of the introduction of mir-6-1-based RNAi constructs into Drosophila melanogaster via C31 integrase-mediated recombination. *Only a part of the hatched larvae were delivered by Rainbow Transgenic Flies, Inc.
Symbol Genotype Hatching rate
(larvae/injected eggs)
Eclosing rate (P0 adults/larvae)
Fertility rate (fertile P0 adults/P0 adults)
Transformation rate (transformants/fertile P0 adults)
OdsHRNAi_simAB y1 sc1 v1; P{OdsHRNAi_simAB}attP2 * 63.1% (41/65) 82.9% (34/41) 26.5% (9/34)
OdsHRNAi_mauAB y1 sc1 v1; P{OdsHRNAi_mauAB}attP2 * 60% (24/40) 70.8% (17/24) 29.4% (5/17)
21
Figure 5. Drosophila eye images from wild-type and transgenic flies. (A-H) Scanning electron microscope images. (A′-H′) Light microscope images. (A-A′) w1118. (B-B′) GMR>OdsHmel. (C-C′) GMR>OdsHmel; GD51289 (VDRC). (D-D′) GMR>OdsHmel; KK103949 (VDRC). (E-E′) GMR>OdsHmel; JF02198 (TRiP 1st Gen.). (F-F′) GMR>OdsHmel; HMS01554 (TRiP 2nd Gen.). (G-G′) GMR>OdsHmel; OdsHRNAi_simAB. (H-H′) GMR>OdsHmel; OdsHRNAi_mauAB. Scale bars: 100 m.
A B C D E F G H
A′ B′ C′ D′ E′ F′ G′ H′
GMR>OdsHmel GMR>OdsHmel
w1118
GD51289 KK103949 JF02198 HMS01554 OdsHRNAi_simAB OdsHRNAi_mauAB
22
Figure 6. Normalized values of eye area to w1118 control. Genotypes are indicated below the graph. Error bars indicate the standard deviation of the mean. Statistical
significance was determined with the Mann-Whitney U test. *, P < 0.05; ***, P < 0.001;
ns, not significant. n = 10 testes for each data point.
***
ns
ns
* *
***
***
Normalized value of eye area
23
Figure 7. Normalized values of eye circumference to w1118 control. Genotypes are indicated below the graph. Error bars indicate the standard deviation of the mean.
Statistical significance was determined with the Mann-Whitney U test. *, P < 0.05; ***, P < 0.001; ns, not significant. n = 10 testes for each data point.
***
ns
ns * *
* *
Normalized value of eye circumference
24
Figure 8. RT-PCR detection of OdsHmel expression in wild-type and transgenic male flies. Genotypes are indicated above the graph. rp49 was used as a loading control.
25
His-GFP expression in testes of OdsH+ and OdsH0 flies
To determine which testicular cells can be observed in the his-GFP background, it is necessary to clarify the construction of this transgenic line. his-GFP, synonymous with His2AvD-GFP, encodes a variant H2A.F/Z class histone of D. melanogaster fused with the green fluorescence protein of Aequorea victoria. Its localization to
chromosomes makes it a superb reporter gene to visualize nuclear activity during interphase, mitosis and meiosis (CLARKSON and SAINT 1999; WHITE-COOPER 2004). In post-meiotic spermatids, chromatin undergoes reorganization and condensation;
histones are modified, removed and replaced by protamines (JAYARAMAIAH RAJA and RENKAWITZ-POHL 2005; RATHKE et al. 2007). Therefore, in his-GFP testes, where GFP signal appears indicates an aggregate of chromosome in germ cells, including GSCs, spermatogonia, spermatocytes and early spermatids. The only somatic cells that can be easily observed are the dividing somatic stem cells (SSCs) and the newly divided somatic cyst cells. Unlike GSCs that undergo four rounds of mitosis and two rounds of meiosis, SSCs merely divide once during spermatogenesis, which makes them relatively few in number.
I first examined the His-GFP expression in the testes of OdsH+ and OdsH0 flies to uncover how OdsH affects the male germ cells in general. Immediately, I noted that His-GFP displayed various expression patterns in OdsH+ and OdsH0 testes at day 0. The distribution of His-GFP in testes was in the apical region (Figure 9, A and E), somewhat extending (Figure 9, B and F), or further extending towards the basal end (Figure 9, C and G). On the contrary, at day 32, His-GFP distribution was quite similar in OdsH+ and OdsH0 testes (Figure 9, D and H), from the apical tip towards the distal end.
Quantification assay showed a significantly higher GFP signal in OdsH+ testes than in
26
OdsH0 testes at day 0, but not at day 32 (Figure 10).
Sa-GFP expression in testes of OdsH+ and OdsH0 flies
Since I observed a significant difference of GFP expression between OdsH+; his-GFP and OdsH0; his-GFP, I investigated the expression pattern of Sa-GFP, a primary spermatocyte marker (CHEN et al. 2005), in testes of different ages by
immunostaining (Figure 11, A-F). Quantification assay showed at days 1, 15 and 55, no significant difference of GFP signal in sa-GFP, OdsH+ and sa-GFP, OdsH0 testes (Figure 12).
BamP-GFP expression in testes of OdsH+ and OdsH0 flies
Next, I characterized the BamP-GFP expression pattern in testes of different ages.
Immunostaining shows that, in both OdsH+; bamP-GFP and OdsH0; bamP-GFP testes, bam promoter-driven GFP expression either resembled the endogenous bam expression, which starts from the 4-cell cyst stage until the entry into premeiotic G2 phase and peaks in 8-cell cysts (Figure 13, A-C and G-I), or displayed somewhat misimpression pattern that spread extensively from the apical region to the distal end (Figure 13, D-F and J-L).
No quantification was done due to inconsistency in expression pattern.
27
Figure 9. His-GFP expression in testes of 0- and 32-day-old OdsH+ and OdsH0 flies.
Testes from (A-C) 0- and (D) 32-day-old flies of OdsH+; his-GFP and from (E-G) 0- and (H) 32-day-old flies of OdsH0; his-GFP were shown. Testes were stained with anti-GFP (green). Scale bars: 100 m.
OdsH0 ; his-GFP
Day 32
OdsH+ ; his-GFP
Day 0
A B C D
E F G H
28
Figure 10. The average volume of His-GFP signal per testis. The histogram depicts the average volume of His-GFP signal per testis from OdsH+; his-GFP and OdsH0; his-GFP flies at 0 and 32 days of age. Numbers on the bars indicate the sample size for each genotype. Error bars indicate the standard deviation of the mean. Statistical significance was determined with the Mann-Whitney U test. ***, P < 0.001; ns, not significant.
ns
(14) (15) (4) (3)
***
29
Figure 11. Sa-GFP expression in testes of 1-, 15-, and 55-day-old OdsH+ and OdsH0 flies. Testes from (A) 1- (B) 15- and (C) 55-day-old flies of sa-GFP, OdsH+ and from (D) 1- (E) 15- and (F) 55-day-old flies of sa-GFP, OdsH0 were shown. Testes were stained with anti-GFP (green) and anti-Fasciclin III (*, hub cells). Scale bars: 100 m.
A B C
D E F
sa-GFP, OdsH+ sa-GFP, OdsH0
Day 15 Day 55
Day 1
*
* * *
* *
30
Figure 12. The average volume of Sa-GFP signals in spermatocytes. The histogram depicts the average volume of Sa-GFP signal in spermatocytes per testis from sa-GFP, OdsH+ and sa-GFP, OdsH0 flies at 1, 15 and 55 days of age. Numbers on the bars indicate the sample size for each genotype. Error bars indicate the standard deviation of the mean. Statistical significance was determined with the Mann-Whitney U test. ns, not significant.
ns
(17) (16) (12) (14) (14) (10)
ns
ns
31
Figure 13. BamP-GFP expression in testes of 1-, 10-, and 32-day-old OdsH+ and OdsH0 flies. Testes from (AD) 1- (BE) 10- and (CF) 32-day-old flies of OdsH+; bamP-GFP and from (GJ) 1- (HK) 10- and (IL) 32-day-old flies of OdsH0; bamP-GFP were shown. Testes were stained with anti-GFP (green) and anti-Fasciclin III (*, hub cells). Scale bars: 100 m.
A B C
D E F
G H I
J K L
Day 10 Day 32
Day 1
OdsH+ ; bamP-GFPOdsH0 ; bamP-GFP
*
*
*
*
*
*
*
* *
*
* *
32
Discussion
OdsH RNAi strains should facilitate future studies on the functional divergence of duplicate genes and the normal function of OdsHsim
Four OdsH RNAi strains from VDRC and TRiP were tested in GMR>OdsHmel background. All of them rescued the ectopic expression phenotype to different levels, which indicates their efficiency. It seems unlikely to associate their efficiency with their targeting regions; all of the targeting regions fall within the exon 4 of OdsH, with one (JF02198 from TRiP) including the 3′ UTR (Figure 14 and Table 4). The observed less efficiency of GD51289 from VDRC is probably because the RNAi construct is
generated via the P-element transformation, and its expression is subject to the position effect. Ultimately, these strains should be examined in a fertility assay to see whether their phenotype—males produce less offspring—shows consistency with OdsH null mutant. These strains should facilitate the simultaneous knockdown of OdsH and unc-4 for future studies on the functional divergence between the two paralogs.
For the newly generated OdsHRNAi_simAB and OdsHRNAi_mauAB, two target sites for each construct fall within the homeobox (mir-6-1-OdsHsimA and mir-6-1-OdsHmauA) and the exon 4 (mir-6-1-OdsHsimB and mir-6-1-OdsHmauB), respectively (Figure 14).
Considering the sequence divergence of OdsH (Table 5), it is expected to see the incomplete rescue effect. Although the phenotypic rescue effect were observed, these two strains need to be further tested in the GMR>OdsHsim and GMR>OdsHmau
background to determine the knockdown efficiency.
33
Figure 14. OdsH structure and the target regions of UAS-OdsHRNAi strains used in this thesis. Self-made RNAi constructs
(mir-6-1-OdsHsimA, mir-6-1-OdsHsimB, mir-6-1-OdsHmauA and mir-6-1-OdsHmauB) are denoted as simA, simB, mauA and mauB. Note that target regions of GD51289, HMS01554 and simB do not overlap.
OdsH
HMS01554
simA mauB simB
mauA
KK103949 GD51289
JF02198
Intron
Homeobox UTR
Coding sequence Exon 3
Exon 1 Exon 2 Exon 4
34
Table 4. The target sequences of UAS-OdsHRNAi strains obtained from stock centers. OdsH 3′ UTR is underlined.
Strain Length (bp) Sequence (5′ to 3′)
HMS01554 21 CAGCAGCGAACTGGATTCCAA
Table 5. Homology of the target sequences of mir-6-1-based RNAi constructs in Drosophila melanogaster, D. simulans and D. mauritiana.
The designed sequences are shown in bold letters. Mismatched nucleotides are indicated in red. Gaps are denoted as “-.” Percentages in parentheses refer to the level of sequence homology in comparison with the designed sequence (number of identical matches divided by the sequence length).
mir-6-1-OdsHmauA mir-6-1-OdsHsimA mir-6-1-OdsHmauB mir-6-1-OdsHsimB
D. melanogaster GGAGAGAGTCTTCCAGGGAAGT(86.4%) ATAGCGGTTTGGTTCCAGAATC(81.8%) ATATGGGACTCCAGG--GAAGA(63.6%) AATTCGCCCTTCAGCATCGAAT(86.4%) D. simulans GGAGAGAGCCTTCCAGGACAAT(90.9%) ATAGCGGTTTGGTGTCAAAATA ATATGGAAGTCCAGG--GAATA(77.3%) AAATCGTCCTTCAGCATCAAAT
D. mauritiana GGAGAAAGCCTTCCAGGAAAAT ATAAAGATATGGTTCCAAAATA(72.7%) ATATGGAAGTGGAAGTGGAATA AAATCGTCCTTCAGCATCAAAT (100%)
35
His-GFP expression implies GSC loss or disruption of subsequent spermatogenic development in young OdsH0 flies
Spermatogenesis in D. melanogaster is a series of cellular divisions and
metamorphosis events. At the apex of a testis tub lies a group of somatic support cells termed hub cells, which are surrounded by 6-12 germline stem cells (GSCs) (SPRADLING
et al. 2011). Each GSC is flanked by a pair of cyst progenitor cells (i.e. somatic stem cells, SSCs); both GSCs and SSCs are connected to the hub cells. As a GSC divides asymmetrically into two daughter cells, one maintains the GSC identity, and the other becomes a spermatogonium encapsulated by two somatic cyst cells. While these two cyst cells undergo no further division, the spermatogonium enters four rounds of synchronous mitotic division with incomplete cytokinesis, producing a cyst of 16 interconnected spermatogonia. The spermatogonia then embark on premeiotic DNA replication, and switch to an extended G2 phase for cell growth as spermatocytes (LIM et al. 2012). During the G2 phase, each spermatocyte substantially increases 25 times in volume. This premeiotic G2 phase continues for more than 3 days before two meiotic divisions proceed, which generates 64 round spermatids (WHITE-COOPER 2010). Still interconnected by a cytoplasmic bridge (ring canal), these round spermatids
subsequently enter the elongation program and become individual, mature sperms (Figure 15).
At day 0, His-GFP expression patterns were various in OdsH+ and OdsH0 testes.
This is probably due to the individual variation. Because the development of testes starts in the late larval stage (CHENG et al. 2008), it is likely that flies collected on the same day differ in sexual maturation. The quantification difference of GFP signal indicates there are more cells (predominantly germ cells) in OdsH+ testes. One explanation
36
(“explanation A”) is that OdsH0 flies fail to maintain as many GSCs as OdsH+ flies do, and GSC loss in OdsH0 testes causes a decrease in germ cell number. An alternative (“explanation B”) is that the GSCs are intact, but the subsequent development in OdsH0 testes is disrupted. However, at day 32, the quantification of GFP shows no difference between OdsH+ and OdsH0 testes. How does the age factor coordinate with these two explanations? One hypothesis is that the age effect on GSC maintenance, or the subsequent developmental stages, is accumulative and relatively minor than OdsH0 in young flies. In aged flies, OdsH0 effect is masked by the increasing age effect, so the germ cell number decreases in OdsH+ and OdsH0 testes without significant difference.
Both explanations are contradictory to the previous discovery that the GSC number is significantly less in 10-day-old, but not in 3-day-old, OdsH0 testes (CHENG et al. 2012).
37
Spermatogonium Spermatocyte Spermatid
Figure 15. Male germline differentiation in Drosophila melanogaster. The schema is not in scale. Expression periods of three GFP reporter lines used in this study, His2AvD-GFP, sa-GFP and bamP-GFP, are indicated. Note that the various spermatid “stages” are defined by the nuclear shape, not mitochondrial.
GSC Sperm
His2AvD-GFP
bamP-GFP
sa-GFP
38
Sa-GFP expression hints a possible role of OdsH in meiosis or spermatid differentiation
In early Drosophila spermatocytes, two classes of “meiotic arrest” genes have been identified and proved to be essential in meiotic divisions as well as spermatid differentiation (WHITE-COOPER et al. 1998). The meiotic arrest genes, named after the shared morphology of undifferentiated primary spermatocytes, are classified into aly- and can-class genes. The aly-class genes, including always early (aly), achintya/vismay (achi/vis), cookie monster (comr), matotopetli (topi) and tombola (tomb), constitute the testis-specific meiotic arrest complex (tMAC). tMAC is similar to Drosophila
dREAM/MMB complex, an evolutionary conserved complex that exists in human and worm (WHITE-COOPER 2010). This complex functions as a transcriptional repressor and activator on a large number of developmentally regulated and cell-autonomous genes (GEORLETTE et al. 2007; LEWIS et al. 2004). tMAC consists of dREAM/MMB subunits (Caf1/p55 and Mip40), homologs of dREAM/MMB components (Aly [Mip130
homolog], Tomb [Mip120 homolog]), and the previous unknown (Comr and Topi) (LIM
et al. 2012). The can-class genes encode testis-specific TAF (TBP-associated factor) homologs (tTAFs), including cannonball (can, TAF5 homolog), meiosis I arrest (mia, TAF6 homolog), no hitter (nht, TAF4 homolog), ryan express (rye, TAF12 homolog) and sa (TAF8 homolog) (LIM et al. 2012). In all eukaryotes, Transcription factor IID (TFIID) is the prototypical core promoter recognition factor. Comprised of the
TATA-box-binding protein (TBP) and 13-14 TAFs, TFIID recognizes and binds core promoter elements such as the TATA box (through TBP), the initiator element (through TAF1 and TAF2), and the downstream promoter element (through TAF6 and TAF9), thus starting the formation of the RNA polymerase II transcription preinitiation complex.
39
It has been suggested that, in association with the prototypical core promoter recognition complexes, cell-type-specific TAFs play a crucial role in mediating cell-specific
transcription and cell differentiation (GOODRICH and TJIAN 2010).
The roles of tMAC and tTAFs in Drosophila spermatogenesis have yet to be fully elucidated, but previous research works have showed an outline already. In general, tMAC regulates the entry into meiotic division, while tTAFs are associated with spermatid differentiation. Based on the observations on mutants, it is likely that tMAC acts upstream of tTAFs (WHITE-COOPER et al. 1998). tTAFs regulate the spermatid differentiation genes by 1) reducing the Polycomb group (PcG) binding to the target genes and 2) recruiting the Trithorax (TrxG) to the target genes (CHEN et al. 2005). PcG is related to gene silencing; one of its subunit, polycomb repressive complex 2 (PRC2) binds to the PcG target gene, creates trimethylation of histone H3 at lysine 27
(H3K27me3) marks and recruits PRC1 to repress gene expression by forming
heterochromatin. TrxG catalyzes trimethylation of histone H3 at lysine 4 (H3K4me3) marks, which results in gene expression by forming euchromatin (JOBE et al. 2012).
Also, in primary spermatocytes, tMAC is required for the localization of tTAFs to target genes and for the colocalization of tTAFs and PcG to a subcompartment within the nucleolus (CHEN et al. 2011).
sa is one of the can-class meiotic arrest genes. Quantification of Sa-GFP shows a declining trend of signal, but no significant difference between 1-, 15- and 55-day-old OdsH+ and OdsH0 testes. If the GFP intensity represents the cell number, there seems no difference in the total number of primary spermatocytes between OdsH+ and OdsH0 testes. This is, however, inconsistent with the two explanations for the his-GFP result, because in either case it is expected to see a significant difference between, at least,
40
1-day-old OdsH+ and OdsH0 testes. Following explanation A (OdsH0 causes GSC loss in young flies), a wild guess is that knockout of OdsH upregulates sa expression in young OdsH0 flies, and this upregulation effect is compromised as OdsH0 flies age.
Following explanation B (OdsH0 disrupts the subsequent spermatogenesis in young flies), it seems plausible when the disruption occurs after the primary spermatocyte stage (Figure 15), because the difference cannot be detected in sa-GFP background.
Since OdsH expresses in the early stage of spermatogenesis (TING et al. 2004;
VIBRANOVSKI et al. 2009), it is attractive to hypothesize that OdsH plays a role in the tMAC-tTAFs network. Knockout of OdsH causes a minor defect in meiosis and/or spermatid differentiation, and eventually leads to a decrease in germ cell number (Figure 16).
BamP-GFP expression yields no information on OdsH in early spermatogenesis Either explanation A or B contradicts the previous study, which has demonstrated significant GSC loss in aged OdsH0 flies (CHENG et al. 2012). Therefore, it is necessary to examine the germ cells in early stage—GSCs and spermatogonia—in OdsH+ and OdsH0 testes. bamP-GFP is a suitable reporter line to monitor the effect of OdsH0 on spermatogonia from the 4-cell to 16-cell cyst stage (GONCZY et al. 1997) (Figure 10).
Unfortunately, at my hands, the pattern of BamP-GFP expression showed inconsistent.
No matter how old the flies are, BamP-GFP in both OdsH+ and OdsH0 testes either resembles the canonical pattern (a stripe of cells in the apical region of testes), or appears misexpressed throughout testes (Figure 13). This result hindered the further inference about the role of OdsH in early stages of spermatogenesis.
41
Figure 16. tMAC (aly-class genes) and tTAFs (can-class genes) network in Drosophila melanogaster spermatocytes.
Pc binding Trx binding
Spermatid differentiation tTAFs (can-class genes)
Meiotic division
twine mRNA
Boule Cyclin B
Cdk1 kinase
Cdk1-Cyclin B complex (inactive)
Twine (Cdc25 phosphatase)
Cdk1-Clyclin B complex (active)
tMAC (aly-class genes) Mip 40?
42
References
AYALA,F. J., and W. M. FITCH, 1997 Genetics and the origin of species: an introduction.
Proc Natl Acad Sci U S A 94: 7691-7697.
BATESON,W., 1909 Heredity and variation in modern lights, pp. 85-101 in Darwin and Modern Science, edited by A. C. SEWARD. Cambridge University Press,
Cambridge.
BAYES,J. J., and H. S. MALIK, 2009 Altered heterochromatin binding by a hybrid sterility protein in Drosophila sibling species. Science 326: 1538-1541.
BROWNE,J., 2009 Darwin the scientist. Cold Spring Harb Symp Quant Biol 74: 1-7.
BROWNE,J., 2009 Darwin the scientist. Cold Spring Harb Symp Quant Biol 74: 1-7.