(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Germline protein, Cup, non-cell autonomously limits migratory cell fate in Drosophila oogenesis [1]

['Banhisikha Saha', 'Department Of Biological Sciences Indian Institute Of Science Education', 'Research- Kolkata Mohanpur Campus Mohanpur', 'Nadia', 'West Bengal', 'Laboratory Of Malaria', 'Vector Research', 'National Institute Of Allergy', 'Infectious Diseases', 'Nih']

Date: 2023-02

Specification of migratory cell fate from a stationary population is complex and indispensable both for metazoan development as well for the progression of the pathological condition like tumor metastasis. Though this cell fate transformation is widely prevalent, the molecular understanding of this phenomenon remains largely elusive. We have employed the model of border cells (BC) in Drosophila oogenesis and identified germline activity of an RNA binding protein, Cup that limits acquisition of migratory cell fate from the neighbouring follicle epithelial cells. As activation of JAK-STAT in the follicle cells is critical for BC specification, our data suggest that Cup, non-cell autonomously restricts the domain of JAK-STAT by activating Notch in the follicle cells. Employing genetics and Delta endocytosis assay, we demonstrate that Cup regulates Delta recycling in the nurse cells through Rab11GTPase thus facilitating Notch activation in the adjacent follicle cells. Since Notch and JAK-STAT are antagonistic, we propose that germline Cup functions through Notch and JAK-STAT to modulate BC fate specification from their static epithelial progenitors.

Transformation of stationary epithelial cells into a migratory fate is critical for development, tissue repair and progression of diseases like tumor metastasis. We propose a novel role for the germline soma communication in modulating migratory border cell fate specification from the somatic follicle cells using the Drosophila oogenesis model. By coupling fly genetics, immunohistochemistry and live endocytosis, we propose that germline protein, Cup, modulates Delta recycling thus potentiating Notch activation and fine tuning STAT stimulation in the adjacent somatic follicle cells. A fine balance between the Notch and STAT signalling aids in specifying an optimum number of follicle cells to acquire migratory border cell fate.

Funding: B.S. and G.G. received student fellowship support from the University Grants Commission and Council of Scientific & Industrial Research respectively. S.A. and P.D. were supported by the Innovation in Science Pursuit for Inspired Research, Department of Science and Technology student fellowship. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

In this study, we report a novel role of nurse cells in BC fate specification. Specifically, our data suggest that germline-specific activity of Cup, non-cell autonomously modulates Notch signaling in the AFCs. As Notch and JAK-STAT signaling work in an antagonistic fashion, Cup mutants exhibit an excess number of BCs likely due to elevated STAT in the AFCs. Further, we demonstrate that Cup mutants exhibit aberrant actin cytoskeleton and enrichment of Delta puncta in the nurse cell cytoplasm. Employing classical genetics and tissue immunohistochemistry in various genetic backgrounds, we show that Cup maintains the integrity of the germline cytoskeleton and modulates Delta trafficking in the nurse cells. Also, overexpression of constitutively active Rab11GTPase in the germline in cup mutants can rescue the reduction in the BC numbers. Together these data argue that recycling Delta in the germline nurse cells is critical for Notch activation in the AFCs of vitellogenic egg chambers. Activation of Notch in the AFCs modulates STAT activity, thus controlling the total number of AFCs that acquire BC fate.

The JAK-STAT signaling in the AFCs is carefully modulated at multiple levels to recruit an optimum number of FCs to BCs fate (generally 4–6 cells). First, both the production and the distribution of Upd ligand are regulated to form a gradient across the anterior follicle cells. Second, Yorkie, a component of the Hippo signaling, negatively regulates Upd production from the polar cells [ 27 ]. Furthermore, the Glypicans, Dally, and Dally-like shape the distribution of Upd ligand, thus calibrating STAT activation and BC fate specification [ 28 ]. Within the AFCs, various intracellular components modulate the STAT activity. The suppressor of Cytokine Signaling (SOCS36E) regulates the ubiquitination of several components of the JAK-STAT pathway to limit STAT activation [ 29 , 30 ]. In addition, other checkpoints operate at the level of transcription. In the follicle cells (FCs), antagonistic interactions between STAT and transcriptional repressor Apontic, restrict the domain of STAT activation, thereby limiting BC fate [ 31 ]. A recent study shows that Insulin signaling also constrains BC fate by stabilizing the negative regulator SOCS36E in the AFCs [ 32 ]. Thus, the JAK-STAT pathway is regulated at multiple levels in the somatic FCs to give rise to a fixed number of BCs during Drosophila oogenesis. Since interaction between germline nurse cells and somatic FCs is critical for oogenesis progression and polar cell fate specification, we were curious to examine if the germline cells also have any direct role in BC fate specification [ 33 ].

Border cells (BCs) in Drosophila oogenesis have emerged as an excellent genetic model system for studying how stationary epithelial cells transition into motile cells [ 10 ]. Drosophila oogenesis is a synchronized developmental process consisting of 14 stages of interconnected oval egg chambers [ 11 , 12 ]. Each egg chamber consists of 16 central germline cells, of which only a single cell acquires the oocyte identity, while the remaining 15 cells become nurse cells involved in nourishing the growing oocyte [ 13 – 15 ]. Enveloping the germline cells is a single layer of approximately 750 follicular epithelial cells. A pair of specialized follicle cells, called the polar cells, mark both the anterior and posterior ends of the egg chamber [ 16 ] During the germarium stage, the activity of Delta ligand emanating from the nurse cells activates Notch signaling which allows specification of the anterior polar cells [ 17 ]. Subsequently, the second round of Delta mediated Notch activation inhibits the proliferation of the follicle cells in stage 6–7 egg chambers and assists their differentiation into distinct cell fates [ 18 , 19 ]. The polar cells secrete cytokine, Unpaired (Upd) which activates the JAK-STAT pathway to confer migratory fate onto a select group of 4–6 anterior follicle cells (AFCs) [ 20 , 21 ] also termed border cells (BCs). BCs undergo partial epithelial to mesenchymal fate transition and initiate posterior movement towards the oocyte [ 14 ]. The BCs are marked by the STAT-mediated activation of CEBP transcription factor, Slow border cells (Slbo) [ 21 – 23 ]. After the BCs are specified their posterior movement is guided under the influence of combined yet graded action of growth factors (PVF1-Platelet Derived Growth Factor and Vascular Endothelial Growth Factor-related Factor 1 and EGF-Epidermal growth factor) secreted from the oocyte [ 24 – 26 ]. After the BC cluster reaches the oocyte, it aids in the formation of a channel in the micropyle. This channel permits sperm entry during fertilization [ 23 ]. Any defect in BC specification or their efficient movement, impedes micropyle function, rendering eggs sterile.

Transformation of a stationary epithelial cell population into a migratory one is critical not simply for normal metazoan development but is also linked to various pathological conditions including tumor cell metastasis. [ 1 – 3 ]. In fact, the inappropriate acquisition of migratory capabilities by the cells from solid tumors underlies the high degree of fatality associated with metastasis [ 4 – 7 ]. Cells acquire migratory potential via diverse mechanisms which can be broadly classified into two categories: autonomous and regulative or nonautonomous [ 8 , 9 ]. Regulative or nonautonomous communication is more prevalent during such type of cell fate transformation as metazoans employ diverse modes of cell-cell communication.

Results

The germline specific function of cup affects the size of the somatic BC cluster In Drosophila melanogaster both proper specification and migration of BCs are critical determinants of female fertility. Several autocrine, and paracrine factors associated with AFCs mediate the specification of BCs. Though signals from the nurse cells regulate the polar and stalk cell fate in previtellogenic egg chambers (stages 1–2), it is not clear if the germline can directly impact the specification of BCs [34]. Hence, we sought to assess if the germline nurse cells directly participate in the specification of somatic BCs during early vitellogenesis. To address this question, we narrowed down 14 candidate genes known to be expressed in the nurse cells. We also ensured that mutations in these genes result in female sterility (S1 Table). Among these 14 genes, we examined the status of BC fate specification in 3 mutant lines that are homozygous viable. We reasoned that the size of BC clusters in the homozygous mutant egg chambers will be altered if the gene product directly controls BC specification. We measured the size of the BC cluster for each of the three homozygous mutant lines and found that mutation in the Cup gene (cup01355) resulted in the largest BC cluster among the three mutants (3073.40±127.59μm3 SEM, n = 32 clusters) compared to the WT (1373.33±54.86 μm3 SEM, n = 31 clusters) (Fig 1A–1C). The BC clusters of Gprk06923 and cdc27L7123 mutants were also larger than the WT (Gprk06923: 2342±98.6 μm3 SEM n = 31; cdc27L7123: 1926±121.4 μm3 SEM, n = 29) (S1A–S1D Fig). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 1. Cup functions in the germline cells to limit the size of the BC cluster. (A-C) cup01355 egg chambers exhibit larger border cell cluster, Armadillo (red), DAPI (blue), White dotted line marks BC cluster. (D-F) cup01355 egg chambers exhibit increased cell count, DAPI (grey), compared to wild type. (G-H) cup01355 egg chambers lack Cup Expression, Cup (red), DAPI (blue). (I-L) Increased BC number (white arrow) is rescued by UASp-CupCDS, driven by nos. GAL4-VP16 in cup01355 egg chambers F-actin (magenta), DAPI (cyan, grey in inset). Student t-test, p value between WT and cup01355 is <0.0001 indicated by ****, p value between cup01355 and UASp-CupCDS in cup01355 background is <0.0001 indicated by ****, p value between WT and UASp-CupCDS in cup01355 background is 0.0698 indicated by ns (not significant). (M-O) Down regulation of Cup in germline by overexpressing cupRNAi with germline mat alpha4-tub-GAL4 exhibit increased border cell count compared with control, F-actin (magenta), DAPI (cyan, inset in grey). n represents the number of egg chambers evaluated. SEM represent the error bars. https://doi.org/10.1371/journal.pgen.1010631.g001 Cup protein has been shown to regulate the translation and stability of several maternal mRNAs including oskar and nanos during Drosophila oogenesis [35–37]. cup01355, a hypomorphic allele, has a P-lacZ insertion in the untranslated region of the first exon of the cup gene. It belongs to the least severe class of alleles where the phenotype manifests only during post-vitellogenic stages of Drosophila oogenesis [38]. To establish that the larger clusters observed in cup01355 homozygotes were indeed due to the altered number of BCs, we stained the egg chambers with DAPI to quantify the number of BCs. Consistent with our expectation, we observed that the number of BCs in cup01355 mutant egg chambers was higher (8.13±0.4 SEM, n = 22) compared to WT (6.41±0.13 SEM, n = 29) (Fig 1D–1F). To exclude the possibility that a second-site mutation induced the phenotype, we analyzed BCs numbers in the heterozygous genetic background of cup01355 and cup8 background. cup8 is an ethyl methyl sulfonate induced allele that was independently isolated and thus has a different genetic background. It exhibits morphological defects in stage 8–9 egg chambers [39]. Satisfyingly, we also observed an increase in the number of BCs compared to the control in cup8/ cup01355 trans-heterozygous background (cup8/cup01355-8.37±0.11 SEM, wild type-5.34±0.07 SEM, n≥50 egg chambers) (S1E–S1G Fig). Altogether these data suggest that Cup modulates the number of BCs in the migrating clusters from the developing egg chambers. Next, we sought to examine if Cup activity is required in the germline cells or somatic follicular cells to influence BC fate. We stained the egg chambers with anti-Cup antibodies and observed that Cup is highly expressed in the cytoplasm of the germline nurse cells both in the early and late stages of oogenesis (Fig 1G). Consistent with previously published reports, we failed to detect any Cup protein in the somatic FCs (Fig 1G) [38]. Since cup01355 is a hypomorphic allele, we examined the levels of cup transcript and protein in cup01355 ovaries. We observed reduced levels of cup transcript (1/10th of WT), and also failed to detect any Cup protein in the homozygous cup01355 egg chambers (Fig 1G–1H, S1H–S1I Fig). The expression analysis of the Cup gene product in WT and cup01355 egg chambers suggested that Cup is primarily expressed in the germline and likely works in a non-cell autonomous manner to specify BC fate. To confirm that Cup activity is not required in the FCs, we employed Mosaic Analysis with a Repressible Cell Marker (MARCM) technique to generate homozygous mutant Cup FCs using a stronger allele of Cup (cup15). cup15 is an EMS allele, and mutant ovaries are known to exhibit a negligible amount of Cup protein as compared to WT [38]. We examined the status of BC fate specification when follicle cells were nearly devoid of Cup protein [40]. As expected, we didn’t observe any significant difference in BC numbers specified in cup15 mutant AFCs (5.68±0.06 SEM, n = 97) compared to WT AFCs (5.76±0.07 SEM, n = 50) (S1J and S1K Fig). To validate that the increased BC number is indeed due to the absence of Cup in the nurse cells, we restored the cup activity by expressing the Cup-coding region (Cup-CDS) in cup01355 nurse cells using mat alpha4-tub-GAL4. Upon reconstitution of Cup-CDS in cup01355 nurse cells, the BC number was significantly restored close to that of the WT (cup01355-8.32±0.12SEM, rescue-5.62±0.05 SEM, wild type-5.45±0.76 SEM, n≥49 egg chambers (Fig 1I–1L). However, overexpression of Cup-CDS in the anterior follicle cells failed to rescue the elevated number of BCs observed in the cup mutant. (cup01355-8.15±0.12SEM, c306; cup01355; Cup-CDS-8.02±0.08 SEM, wild type-5.46±0.07 SEM, n≥35 egg chambers) (S1L–S1O Fig) To further support the non-autonomous role of Cup in border cell fate specification, we downregulated cup function in germline nurse cells employing mat alpha-tub-GAL4 and cup RNAi. Since the mat alpha4-tub-GAL4 driver has weak expression in the nurse cells of early-stage chambers, it enabled us to evaluate cup function during mid-oogenesis. Consistent with our expectation, we observed a higher number of BCs in the migratory cluster compared to the control supporting our conclusion that germline Cup non-cell autonomously affects the BC fate in the AFCs (CupRNAi-9.375±0.4 SEM n = 16 egg chambers, control-5.59±0.07 SEM, n = 39 egg chambers) (Fig 1M–1O) Altogether our results suggest that nurse cell specific activity of Cup protein non-cell autonomously modulates the size of the BC cluster which is specified from the overlying somatic AFCs.

Cup controls BC fate by negatively regulating the JAK-STAT pathway Since cup01355 mutant egg chambers exhibit more nuclei in the migrating cluster, we investigated if the extra cells were indeed BCs. To check this, we stained the egg chambers for the Slbo protein which marks the BCs. We observed a significantly higher number of Slbo-positive cells in the cluster (7.04±0.19 SEM, n = 23) of cup01355 egg compared to the WT (5.27±0.11SEM, n = 22) (Fig 2A–2C). This suggests that Cup depletion results in the aberrant specification of BCs from the follicular epithelium. As Notch signaling modulates the mitotic to endocycle switch in developing eggs, we curious to know if an increase in BC numbers in cup01355 background is due to a prolonged mitotic phase [18]. To check this, we compared the expression pattern of two markers associated with proliferating follicle cells, Cut and phospho-Histone 3 (pH3) [18,34]. We examined 170 egg chambers each of WT, and cup01355 and observed no difference in the staining pattern for Cut (S2A–S2H Fig); nor did we observe any pH3-positive cells in stage 8, or higher stage egg chambers in 168 samples analyzed each for WT and cup01355 (S2I–S2N” Fig). As the expression pattern of both Cut and pH3 was similar in both the WT and the cup01355 egg chambers, we ruled out the possibility that extra rounds of mitotic division are responsible for the excess BC numbers observed in the cup mutants. In addition, we assessed the total number of follicle cell nuclei in stage 8 egg chambers for both wild-type and cup01355 mutants. We counted the number of follicle cell nuclei plane by plane of a confocal z stack to ensure that each nucleus (DAPI) was counted only once. Satisfyingly we didn’t observe any significant difference in the number of follicle cell nuclei between WT and cup mutant egg chambers (cup01355-968.6±6.97 SEM n = 10, wild-type-963.6±5.63 SEM n = 10) (S2O–S2Q Fig). This further supported our claim that inappropriate FC proliferation is not responsible for the elevated number of BCs observed in the cup01355 mutant egg chambers. Since JAK-STAT signaling activates Slbo expression in the AFCs, we next examined if the increase in the number of BCs in cup01355 mutant egg chambers were linked to enhanced STAT function [21,22]. Nuclear STAT is used as a molecular reporter for assessing the status of JAK-STAT signaling [41]. We quantified nuclear STAT and observed higher levels (1.64-fold) of STAT in cup01355 mutant FCs (82.91±7.76 SEM, n = 17 egg chambers) compared to the WT FCs (50.32±3.36 SEM, n = 17) (Fig 2D–2F). We also observed that the number of AFCs exhibiting distinct nuclear STAT in cup01355 egg chambers (15.12±0.67 SEM, n = 16) was higher compared to WT (9.41±0.47 SEM, n = 17) (Fig 2G). In addition, we observed an elevated number of nuclear STAT positive cells which extend as far as 6th FC from the polar cell in cup01355 egg chambers as opposed to the 3 cells observed in the control (Fig 2H and 2H’). Taken together these results suggest that both the levels and spread of STAT activation are enhanced in cup01355 egg chambers. To assess if the elevated STAT was indeed responsible for excess BCs observed in the cup01355 egg chambers, we compared BC clusters in cup01355 egg chambers in WT and STAT heterozygous background (statP1681/+). BC number in the cup01355 cluster was reduced in STAT heterozygous background as compared to the cup01355 mutants itself (cup01355-8.31±0.07 SEM, cup01355; statP1681 /+-6.0±0.1 SEM, wild type-5.50±0.07 SEM, n≥40 egg chambers). Expectedly, the observed rescue of BC fate was partial likely due to heterozygosity for STAT. Collectively these data suggest that elevated STAT levels are responsible for the enhancement in BC fate specification observed in the cup01355 egg chambers (Fig 2I–2M). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. Cup modulates STAT in the follicle cells to limit BC fate. (A-C) cup01355 egg chambers exhibit increased Slbo positive cells (yellow arrow head) compared to wild type, Slbo (red), DAPI (blue). (D-G) cup01355 egg chambers exhibit higher STAT levels and more STAT positive cells in anterior end of egg chamber as compared to wild type (dotted area), STAT (red). (H-H’) STAT level is higher in 6th FC from polar cell in cup01355 egg chambers as compared to control. (I-M) Higher BC numbers is rescued when statp1681/+ background is introduced in cup01355 homozygous background (yellow arrow heads mark BC) F-actin (magenta), DAPI (blue, grey in inset). Student t-test, p value between WT and cup01355 is <0.0001 indicated by ****, p value between cup01355 and cup01355; statp1681/+ is <0.0001 indicated by ****, p value between WT and cup01355; statp1681/+ is <0.0001 indicated by ****. n represents the number of egg chambers evaluated. SEM represent the error bars. https://doi.org/10.1371/journal.pgen.1010631.g002

Loss of cup reduces Notch signaling and increases STAT activity As higher STAT levels can lead to the specification of an elevated number of BCs, we next decided to analyze why STAT function is higher in cup01355 FCs. JAK-STAT signaling in the AFCs is positively regulated by the Upd ligand produced by the anterior polar cells [20,21]. Since we detected increased JAK-STAT signaling in the Cup mutant FCs, we wondered if this was due to an increase in the number of polar cells. To this end, we examined the pattern of Fasciclin III (FasIII), the lateral membrane protein that marks the junction between the two polar cells [16]. Like the WT, we observed a single distinct junction labeled by FasIII at the anterior and posterior ends of the polar cells in the cup01355 egg chambers. This observation suggested that the number of polar cells is unaffected in cup01355 egg chambers (S3A and S3B Fig). We also observed similar FasIII expression in the early stages of oogenesis in both WT and cup01355 egg chambers indicating that polar cell fate is unaffected in the cup01355 hypomorphic background (S3C–S3J’ Fig). As the polar cell number is unaffected, we tested if enhanced JAK-STAT signaling was due to transcriptional upregulation of the upd gene. To examine this, we measured the expression of the upd reporter construct, upd-lacZ, and observed no significant difference in the intensity of β-gal antibody staining from upd-lacZ between the WT (118.69±7.02 SEM, n = 20) and the cup01355 stage 8 egg chambers (124.73±11.03 SEM, n = 20) (S3K–S3M Fig). Similarly, no change in the β-gal antibody staining intensity was observed in the younger egg chambers between the WT and the cup01355 (S3N–S3V Fig). This ruled out the possibility that elevated upd transcription could be responsible for increased STAT signaling in the cup01355 egg chambers (S3K–S3M Fig). This prompted us to explore the possibility of Cup modulating the function of other JAK-STAT regulators which, in turn, may affect BC specification. For instance, the upregulation of STAT activity could be explained by the downregulation of one (or more) of the negative regulators of the JAK-STAT signaling pathway in the cup01355 egg chambers. The known negative regulators of the JAK-STAT signaling include Protein tyrosine phosphatase 61F (Ptp61f), Brahma (Brm), Suppressor of Cytokine Signaling 36E (SOCS36E), and Notch [33,42,43]. Among these, we decided to focus on Notch primarily for two reasons. First, communication between nurse cells and FCs depends on Notch signaling during egg chamber development. Secondly, Notch signaling inhibits JAK-STAT signaling in a context-specific manner in the FCs [33,34]. To assess the level of Notch signaling in the FCs we employed a Notch reporter construct, where the Notch Response Element (NRE) is tagged upstream of eGFP (NRE-eGFP) [44]. NRE consists of binding sites for the transcription factor, Suppressor of Hairless, and the transcriptional activator Grainy head. Activation of Notch signaling leads to the binding of these transcriptional activators to the NRE sequence leading to GFP expression. We checked Notch activity in AFCs by measuring eGFP reporter expression (under NRE) in both control and cup01355 stage 8 egg chambers. Interestingly, we observed a fivefold decrease in the levels of eGFP in the AFCs of cup01355 egg chambers (cup01355-0.2±0.008 SEM, n≥30 egg chambers) compared to the control (1.0±0.06 SEM, n≥30 egg chambers) (Fig 3A–3C). We observed a rescue in the levels of NRE-eGFP when Cup CDS was overexpressed in the nurse cells of cup01355 mutant egg chambers (cup01355-0.4±0.06 SEM n = 10;cup01355; CupCDS/ nos GAL4-VP16–1.1±0.3 SEM n = 6, wild type-1.0±0.2 SEM, n = 5 egg chambers) (S4D–S4G Fig). However, over-expression of Cup-CDS in the AFCs failed to rescue the low levels of NRE-eGFP observed in the Cup mutants suggesting that germline-specific function of Cup modulates Notch signaling in the AFCs (cup01355-0.28±0.05 SEM n = 5, c306 GAL4; cup01355; CupCDS -0.26±0.01 SEM n = 12, WT-1.0±0.05 SEM, n = 7 egg chambers) (S4H–S4K Fig). Altogether these data suggest that germline Cup modulates the strength of Notch signaling in the AFCs of developing egg chambers. To support this observation further, we examined the levels and distribution of Notch in the FC. It is known that ligand binding stimulates two sequential proteolytic cleavages in the Notch receptor generating a fragment with the extracellular domain (NECD) and the other with the intracellular domain (NICD) [45]. The distribution of NICD and NECD is routinely used to evaluate the status of Notch signaling. Ligand stimulation promotes NECD and NICD internalization in the ligand-producing cell and signal-receiving cell respectively [46–48]. We observed numerous NICD and NECD puncta in the WT FC and nurse cells respectively (Fig 3D–3J). The presence of a large number of NICD and NECD puncta suggests that Notch signaling is active in WT FCs. On the contrary, we observed very few internalized puncta of both NICD and NECD in the FC and the nurse cell of the cup mutant egg chambers respectively supporting the fact the Notch signaling is downregulated (For NICD; WT-101±8.32 SEM n = 10, cup01355-64.10±3.345 SEM n = 10 (Fig 3G) and for NECD; WT-271.9±24.18 SEM n = 11, cup01355-103.4±15.30 SEM n = 11. (Fig 3J). We observed a similar trend in the distribution of NICD puncta in the cup8/cup01355 background (WT-138.3± 18.9 SEM, n = 10, cup8/cup01355-44.6±7.8 SEM n = 10) (S4A–S4C Fig). However, we didn’t observe any significant difference in the total NICD intensity between the AFCs of WT and the Cup mutant egg chambers (WT-39.8±2.24 SEM n = 16, cup01355-37.5±2.4 SEM n = 10) (Fig 3D–3F) suggesting against the transcriptional downregulation of Notch receptor. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. Lower Notch activation in the follicle cells is responsible for the excessive border cell fate observed in the Cup mutants. (A-C) Intensity of NRE-eGFP (in green) is significantly decreased in the anterior follicle cells of stage 8 cup01355 egg chambers. (A-B) Dotted line marks the anterior follicle cells expressing NRE-eGFP in the indicated genotypes. Depleted FASII (red) staining indicates stage 8 egg chamber. DAPI (blue). (C) Quantification of the NRE-eGFP. p value of <0.0001 is indicated by ****, (D-G) Stage 8 egg chambers stained with NICD (green). (D-E) Reduced number of NICD puncta (yellow arrow head) observed in the anterior follicle cells of cup01355 mutant egg chamber compared to wild type. Inset is the representative egg chambers (F-G) Quantification of NICD intensity (F) and puncta (G). Though mean intensity of NICD is not significantly different between cup01355 mutant and wild type follicle cells (F), the number of NICD puncta is significantly reduced (G). (H-J) cup01355 exhibits lesser number of cytoplasmic puncta of NECD (green) (yellow arrow head) in the nurse cell compared to the wild type. (K-N) Increased BC number is rescued by UAS-NICD, driven by slbo-GAL4 in cup01355 egg chambers, yellow arrow heads mark BC cluster, F-actin (magenta), DAPI (cyan, grey in inset). Student t-test, p value between WT and cup01355 is <0.0001 indicated by ****, p value between cup01355 and UAS NICD in cup01355 is <0.0001 indicated by ****, p value between WT and UAS NICD in cup01355 is 0.0004 indicated by ***. n represents the number of egg chambers evaluated. SEM represent the error bars. https://doi.org/10.1371/journal.pgen.1010631.g003 Next we decided to test if simply reducing levels of Notch signaling in cup01355 egg chambers alone can enhance the BC fate specification. Thus, we upregulated Notch signaling in the AFCs of cup01355 egg chambers and examined if this can mitigate the increase in total number of BCs. Notch signaling is required during early as well as late stages of oogenesis. To activate Notch signaling specifically during mid to late stages of oogenesis, we overexpressed the Notch intracellular Domain (UAS-NICD) using the late driver slbo-GAL4 [49]. We observed a lower number of BCs in the cup01355 egg chambers overexpressing the NICD (cup01355, UAS-NICD; slbo-GAL4- 6.2±0.11 SEM n = 22) compared to the cup01355 mutant egg chambers (cup01355-8.3±0.15 SEM n = 22, WT-5.5±0.1 SEM, n = 18 egg chambers) (Fig 3K–3N) although the rescue was partial, possibly due to the late expression of the driver. Taken together these results suggest that the increase in the BC fate observed due to loss of function of cup is mediated by the suppression of Notch signaling in the follicle cells. Thus, Cup functions via Notch to modulate the number of AFC that eventually acquire BC fate in the developing egg chambers.

Cup regulates the nurse cell organization and Delta trafficking Decrease in the levels of NICD and NECD is indicative of inefficient Notch proteolysis which results in compromised signaling [50,51]. To analyze underpinnings of these alterations, we wondered if these changes can be correlated with the nurse cell morphology. Consistently, unlike the normal round shaped nurse cell nuclei in the WT, we observed elongated, mispositioned nurse cell nuclei in cup01355 egg chambers (Fig 4A and 4B). Since mispositioned nurse cell nuclei have been reported when the cytoskeleton is disorganized, we also examined the status of the actin cytoskeleton in cup01355 mutant egg chambers [52]. We stained the egg chambers with rhodamine-phalloidin and observed reduced levels of F-actin fibers in the nurse cells of cup01355 egg chambers unlike the control (Fig 4C and 4C’). In addition, we observed very sparse Tubulin fibers in the germline cells of cup mutant egg chambers compared to the control (Fig 4D–4D’). Together these changes suggested that aberrant nurse cell cytoskeleton could be one of the reasons for the disorganized germline and mispositioned nuclei seen in the cup mutant egg chambers. Delta ligand is known to activate Notch in the follicle cells of developing egg chambers. Since cytoskeleton is critical for Delta trafficking and Notch activation, we next focussed our attention on the distribution Delta ligand in the nurse cells of cup mutant egg chambers [53]. First, we compared the levels of total Delta protein in the nurse cells of WT and cup01355 egg chambers. We did not observe any significant difference in the mean Delta intensity in nurse cells between WT (1810.22±97.63 SEM, n = 9 egg chambers) and cup01355 homozygous egg chambers (1614.43±105.29 SEM, n = 9 egg chambers) (Fig 4E–4F’ and 4G). Interestingly, however, the asymmetric posterior localization of Delta protein in WT oocytes was absent in the cup01355 mutant egg chambers (Fig 4E–4F’). Strikingly, we observed a large number of Delta positive puncta in the cytoplasm of nurse cells of cup01355 mutant egg chambers (767± 20.51 SEM, n = 9 egg chambers) unlike the control (328.6±10.35 SEM, n = 9 egg chambers) (Fig 4F–4G’ and 4H). cup8/ cup01355egg chambers also showed a similar increase (WT-341.4 ±30.3 SEM n = 5, cup8/cup01355- 747.5±47.2 SEM, n = 4) (S5A–S5C Fig). In addition, we observed very few Delta positive puncta at the apical interface of AFC and germline nurse cells (control—150.4± 4.68 SEM, n = 5; cup01355- 45.69±3.2225 SEM, n = 5) (Fig 4I–4J and 4K). Incidentally, it is the anterior-most FCs that acquire the migratory BC fate as oogenesis progresses. Delta being a transmembrane protein, its enrichment in the cytoplasmic fraction of the cup mutant nurse cells and its absence from the apical interface of AFCs suggested that Delta trafficking is probably perturbed in the cup mutants. As proper trafficking of Delta ligand in the nurse cells is critical for Notch activation, we decided to analyze it further [34]. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. Cup modulates nurse cell cytoskeleton and Delta trafficking. (A-B) Nurse cell nuclear morphology is disrupted in cup01355 egg chambers. Nuclei (yellow arrow heads) are elongated and mispositioned in cup01355 egg chambers compared to round nuclei observed in wild type, F-actin (Magenta), DAPI (cyan). (C-C’) Phalloidin staining of cup01355 egg chambers exhibit reduced actin fibers (yellow arrow heads) compared wild type egg, F-actin (grey, green in inset), DAPI (red). (D-D’) Tubulin stained cup01355 egg chambers show smaller, randomly distributed tubulin fibres (yellow arrow heads) in nurse cell cytoplasm compared to distinct radially arranged fibers observed in the wild type, tubulin (black, green in inset), DAPI (red). (E-H) Delta stained cup01355 egg chambers exhibit more cytoplasmic puncta in nurse cells as compared to wild type (yellow arrows). Oocyte Delta localisation is absent in cup01355 egg chambers as observed in the wild type (white arrow head). Delta (red), F-actin (green), DAPI (blue). Mean Delta intensity of wild type and cup01355 egg chambers is similar (G). However, the number of cytoplasmic puncta in the cup01355 nurse cells is higher than that observed in the wild type (H). (I-J) Stage 8 egg chamber of indicated genotypes stained for Delta (black) (yellow arrowhead). Inset is the representative egg chamber. (K) Quantification of Delta puncta at the junction of nurse cell and anterior follicle cell suggest that Delta trafficking is perturbed in the cup mutants. (L-N) Delta endocytosis assay has shown that cup01355 egg chambers exhibit more cytoplasmic puncta (yellow arrow heads) in the nurse cells as compared to wild type (yellow arrow heads), Delta (black, red in inset), DAPI (white). n represents the number of egg chambers evaluated. SEM represent the error bars. https://doi.org/10.1371/journal.pgen.1010631.g004

Delta endocytosis is impaired in cup mutants Delta internalization by endocytosis in the signal sending i.e. ligand-producing cells is important for the activation of Notch signaling in the receiving i.e. receptor-expressing cells [53–55]. As the accumulation of Delta puncta in the nurse cells can be an outcome of either defective endocytosis or exocytosis, we first examined the status of these two processes in the cup mutant nurse cells. To this end, we carried out an Delta endocytosis assay on egg chambers with an antibody that recognizes the extracellular domain of Delta ligand (c594.9B). In live samples, the c594.9B antibody can bind only the extracellular fraction of total Delta ligand while the intracellular Delta fraction remains unbound. During the chase, Delta is internalized and moves through the endocytotic vesicles, so does the labelled fragment. Thus, in the endocytosis assay, the enriched Delta puncta observed in the cup01355 mutant nurse cells will be labelled if there are defects in endocytosis. While any aberrations in exocytosis will not result in accumulation of labelled fragment in the puncta [56,57]. When we conducted this experiment, we observed a conspicuous apical enrichment and a few randomly distributed cytoplasmic puncta of Delta in the follicle cells of both the WT and cup mutant egg chambers. Strikingly however, unlike the WT, we observed a significantly higher number of cytoplasmic Delta in the nurse cells of cup01355 mutant egg chambers similar to what was observed in fixed sample analysis (Delta particle count: WT- 468.3±27.84 SEM n = 11, cup01355- 1104±60.88 SEM n = 11) (Fig 4L–4N). As the cytoplasmic Delta in the cup01355 mutant nurse cells was significantly labelled in the endocytosis assay, it suggested that defects in Delta trafficking observed in the cup01355 mutants were predominantly due to impaired endocytosis. Given that the endocytic pathway of Delta trafficking is perturbed, we sought to identify which specific component of endocytosis may be responsible for this behaviour.

Elevating Rab11 activity limits border cell fate Endocytosis is a multi-step process wherein the fate of the internalized cargo is decided in the early endosomes between either recycling or degradation [58–61]. Rab5GTPase plays a crucial role in the biogenesis of endosomes and aids in the maturation of early endosomes to late endosomes. While Rab11GTPase facilitates the recycling of the cargo from the early endosomes to the plasma membrane [62–64]. The cargo marked for degradation moves from the late endosome to the lysosome with the help of the activity of Rab7GTPase [65]. We were curious if endocytosis of Delta ligand was impaired at a specific point during endocytosis in cup mutants. We thus wondered if overexpression of a specific constitutively active RabGTPase (CA) in the nurse cells of the cup01355 egg chamber could reduce the number of BCs cup mutant egg chambers. We observed that overexpression of Rab11GTPaseCA in the cup01355 nurse cells rescued the number of BCs to the control levels (cup01355-8.36±0.081SEM, rescue-5.889±0.0.12 SEM, wild type-5.757±0.082SEM, n≥80 egg chambers) (Fig 5A–5E). However, we didn’t observe any significant difference in the BC numbers when Rab5GTPaseCA (8.260±0.1479) and Rab7GTPaseCA (8.913±0.1632) were over-expressed in the cup01355 nurse cells (Figs 5E and S6D–S6G). Neither did we observe any restriction of BC fate specification, when Rab11GTPaseCA was overexpressed in the WT germline nurse cells. In addition, unlike Rab11GTPaseCA, we observed a modest rescue in the border cell fate in cup mutants overexpressing Rab11WT(cup01355-8.36±0.081SEM; cup01355; UASp Rab11YFP-WT-6.53±0.08 SEM n≥100 egg chambers) (Figs 5E and S6A–S6C). As specifically activating the recycling component of the endocytosis can restore the BC fate to near WT numbers in the cup01355 mutant egg chambers, it suggested that Rab11GTPase activity downstream of Cup modulates BC fate specification in the AFCs. Next, we tested if the rescue is indeed due to the restoration of Notch signaling in the AFCs of cup01355 egg chambers upon overactivation of Rab11GTPase. We measured Notch reporter activity by quantifying NRE-eGFP levels, and observed a 0.5-fold upregulation of Notch activity when Rab11GTPaseCA was overexpressed in the nurse cells of cup01355 egg chambers as compared to that of control (cup01355 mutant egg chambers) (WT- 1.00±0.05 SEM, n = 13, cup01355- 0.16±0.028SEM, n = 10, rescue- 0.52±0.040 SEM, n = 16) (Fig 5F–5I). We also observed rescue in the number of Delta puncta at the apical interface of AFC and germline cells of Cup depleted egg chambers that were over-expressing Rab11CA-YFP(wild type- 164.3±3.125 SEM, n = 10, cup01355- 41.39±2.16SEM, n = 10, rescue- 138.3±3.396 SEM, n = 10) (Fig 5J–5M). Irrespective of partial rescue in the levels of NRE-eGFP and Delta puncta count, we noticed a complete reversion of border cell numbers when Rab11GTPaseCA was overexpressed in the cup mutant germline. This may suggest that BC fate specification is quite robust beyond a certain level of signaling or that Cup may affect other aspects of egg chamber development independent of Rab11 function. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 5. Overexpression of Rab11CA limits the border cell fate in the Cup mutant. (A-E) Stage 10 egg chambers of indicated genotypes stained with EYA (magenta), DAPI (blue, grey in inset) and YFP (green). yellow arrowheads mark the border cell cluster. Number of border cells is rescued when Rab11CA is overexpressed in the nurse cells of cup01355 egg chambers. **** indicates a p value <0.0001 (Student t-test). ** indicates a p value <0.01 (Student t-test). “ns” indicates statistically not significant. (F-I) Stage 8 egg chambers of indicated genotypes. White dotted line outlines the Anterior follicle cells. NRE-eGFP (green) and DAPI (blue, white in inset). (I) NRE-eGFP intensity fold change with respect to the control is partially rescued when Rab11CA is overexpressed in nurse cells of cup01355 egg chambers as compared to cup01355 mutant egg chambers alone. Red asterisk in nurse cells indicates tagged YFP expression when Rab11CA is over expressed with germline specific nos GAL4- VP16. **** indicates a p value <0.0001 (Student t-test) (J-M) Stage 8 egg chambers of indicated genotypes. Yellow arrowheads mark the Delta puncta at the junction of anterior follicle cell and nurse cells, Delta (black, red in inset), and DAPI (white). (M) Quantification indicates that Number of Delta particles at the junction of anterior follicle cell and nurse cells is partially rescued when Rab11CA is overexpressed in the nurse cells of cup01355 egg chambers as compared to that observed in cup01355 egg chambers alone. **** indicates a p value <0.0001 (Student t-test). n represents the number of egg chambers evaluated and SEM represent the error bars in each panel. https://doi.org/10.1371/journal.pgen.1010631.g005 The rescue of border cell numbers and Delta puncta in cup mutant upon over-expression of Rab11GTPaseCA was clearcut. We were thus curious whether such a rescue was possible despite the disruption of cytoskeleton. When we examined the status of actin and tubulin upon overexpression of Rab11GTPaseCA in nurse cells, we observed restoration of actin filaments in the cup mutant nurse cells like WT (S6H–S6K Fig). However, we didn’t see any rescue of aberrations in tubulin distribution. Altogether, these results suggest that Rab11GTPaseCA overexpression rescues the actin filament structure which may help Delta trafficking and appropriate activation of Notch in the AFCs. However, over-expression of either actin or tubulin or both in cup mutant nurse cells failed to restrict the increase in border cell count observed in the cup mutant egg chambers (S5D Fig). Altogether these data underscore the importance of Rab11 downstream of cup during BC specification and also argue that there may be other additional targets downstream of Rab11. Next, we were curious to examine the role of Rab11 in border cell fate specification with a specific focus on Delta trafficking.

Germline Rab11 affects Delta trafficking and border cell fate in the anterior follicle cells Given that increasing Rab11 activity rescues the border cell fate in cup01355 egg chambers, we were curious if manipulating Rab11 activity can influence Delta trafficking and border cell fate specification on its own. To test this, we downregulated Rab11 function in nurse cells by expressing a dominant negative construct of Rab11 that inhibits GTP binding [66]. We employed mat alpha4-tub-GAL4 to bypass the early requirement of Rab11 and targeted its downregulation around mid-oogenesis. Satisfyingly, we observed that functional depletion of Rab11 in the nurse cells resulted in a higher number of AFCs acquiring BC fate as compared to the controls (Rab11DN- 6.3± 0.08 SEM, n = 55; cnn-GFP- 5.3± 0.07 SEM, n = 57) (Fig 6A–6C). The modest increase in the number of cells that acquire BC fate could be attributed to the corresponding level of overexpression of Rab11DN construct. Nevertheless our results suggest that Rab11 activity in the nurse cells modulates BC fate in the AFCs in a non-cell autonomous manner. Next, we examined the status of Delta in Rab11 depleted nurse cells and observed that the number of cytoplasmic Delta puncta was much higher than the control (Rab11DN- 496.5± 12.05 SEM, n = 8; control- 185.2± 4.51 SEM, n = 8) (Fig 6D–6F) In addition, we observed less number of Delta puncta at the apical interface of AFC and germline nurse cells similar to what we observed in cup mutants (Rab11DN- 47.1± 3.1 SEM, n = 5; cnn-GFP- 92.91± 4.6 SEM, n = 5) (Fig 6G–6I). Altogether these results reinforce our claim that Rab11 activity downstream of Cup non-cell autonomously limits the border cell fate in the AFCs. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 6. Rab11 function in the germline modulates Delta Trafficking and Border cell specification from the follicle cells. (A-C) Downregulation of Rab11 activity by over-expressing Rab11DN in the nurse cells results in increased Border cell count compared to control (mat alpha4-tubulin-GAL4; UASp-cnnGFP). Yellow arrow heads marked border cell cluster, F-actin (magenta), DAPI (cyan, grey in inset). (D-F) Overexpression of Rab11DN in nurse cells using mat alpha4-tubulin-GAL4 exhibits cytoplasmic Delta puncta (yellow arrows), Delta (red) in D and E, Delta (black) in D’ and E’. DAPI (blue), GFP (green). (G-I) Overexpression of Rab11DN in the nurse cells with mat alpha4-tubulin-GAL4 exhibits reduced Delta puncta at the junction of anterior follicle cell and nurse cells, Delta (black, red in inset), DAPI (grey in inset). **** indicates a p value <0.0001 (Student t-test). n represents the number of egg chambers evaluated in each panel. https://doi.org/10.1371/journal.pgen.1010631.g006 Next, we were curious to examine how Cup may regulate the Rab11 activity in the nurse cells to influence the BC fate specification from the AFCs.

[END]
---
[1] Url: https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1010631

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.