Stage-specific isolation of mid-to-late Drosophila follicles is useful for a variety of purposes. Such follicles develop in culture, which allows for genetic and/or pharmacologic manipulations to be coupled with in vitro development assays and live imaging. Additionally, follicles can be used for molecular studies, such as isolating mRNA and protein.
Drosophila oogenesis or follicle development has been widely used to advance the understanding of complex developmental and cell biologic processes. This methods paper describes how to isolate mid-to-late stage follicles (Stage 10B-14) and utilize them to provide new insights into the molecular and morphologic events occurring during tight windows of developmental time. Isolated follicles can be used for a variety of experimental techniques, including in vitro development assays, live imaging, mRNA expression analysis and western blot analysis of proteins. Follicles at Stage 10B (S10B) or later will complete development in culture; this allows one to combine genetic or pharmacologic perturbations with in vitro development to define the effects of such manipulations on the processes occurring during specific periods of development. Additionally, because these follicles develop in culture, they are ideally suited for live imaging studies, which often reveal new mechanisms that mediate morphological events. Isolated follicles can also be used for molecular analyses. For example, changes in gene expression that result from genetic perturbations can be defined for specific developmental windows. Additionally, protein level, stability, and/or posttranslational modification state during a particular stage of follicle development can be examined through western blot analyses. Thus, stage-specific isolation of Drosophila follicles provides a rich source of information into widely conserved processes of development and morphogenesis.
Each Drosophila ovary is composed of ~16 ovarioles, or chains of sequentially maturing egg chambers or follicles. Each follicle is composed of a single oocyte, 15 germ line derived nurse or support cells, and ~650 somatic cells termed follicle cells (Figure 1A). Drosophila oogenesis is divided into 14 morphologically defined stages of development1. Each stage of follicle development is observed many times within a single fly, making it relatively easy to isolate a substantial number of stage-specific follicles.
The mid-to-late stages of oogenesis (Stages 10B-14) are particularly well suited for stage isolation (Figure 1). At Stage 10B (S10B), the follicle is fully elongated (i.e. its length is equal to that of a Stage 14 (S14) follicle, see Figure 1 and Figure 2H) and half the length of the follicle is composed of nurse cells while the other half is the oocyte (Figure 1C). At this stage the nurse cells undergo dramatic actin remodeling, strengthening the cortical actin and generating parallel bundles of actin filaments2. At the same time, a population of follicle cells, termed centripetal cells, migrate in between the nurse cells and the oocyte, and two dorsal groups of follicle cells become specified to undergo migration to form the dorsal appendages, tubular respiratory apparatuses for the embryo3. The nurse cells then contract (S11), squeezing their cytoplasmic contents into the oocyte in a process called nurse cell dumping, which provides the oocyte with the factors necessary for it to complete embryogenesis (Figure 1D). The nurse cells then undergo cell death (S12-S13)4, and the follicle cells secrete and pattern the eggshell5 (Figures 1E-G). Thus, the end of oogenesis is rich with important developmental and morphogenetic processes.
Isolated mid-to-late stage follicles (S10B-S14) can be used for a variety of purposes, including molecular analyses. For example, mRNA from staged follicles can be isolated for RT-PCR, microarray, or RNA-seq analyses. This allows one to look at gene expression within a short developmental window, with only a few cell types present, and determine how gene expression is changed by either pharmacologic or genetic perturbations. Stage isolation can also be used to look at proteins by western blotting. Such analysis is important because it allows one to quantify the level of protein expression in wild-type versus mutants at specific stages. While one could use immunofluorescent analyses to achieve similar results, quantification of fluorescence is less robust due to the strict requirements that all of the pixels be within the linear range of detection6. Additionally, western blot analysis may provide other information, such as if the protein is posttranslationally modified or is expressed from a specific splice isoform. Isolated stages can also be used for further protein purification, including subcellular fractionation or coimmunoprecipitation.
Stage-specific follicle isolation can also be used for in vitro development assays7 and live-imaging8. Isolated S10B-S13 follicles will continue to develop to S14 in simple culture media (see below). It is important to note that S10A follicles will not progress through nurse cell dumping using the culture conditions discussed in this manuscript. We have used S10B in vitro development assays to define the role of prostaglandins, both pharmacologically and genetically, in regulating actin remodeling by using nurse cell dumping and development as read-outs7,9. Similarly, the later stages of development can also be isolated to determine the effects of pharmacologic treatments or genetic manipulations on particular processes such as centripetal cell migration, dorsal appendage migration/formation10, and nurse cell death. Such assays can be used to perform dominant interaction screens or assays; for example, while heterozygosity for mutations in pxt or fascin alone have no effect on S10B in vitro development, follicles from double heterozygotes exhibit nurse cell dumping defects and a block in development9.
Additionally, because S10B-13 can develop in culture, all of the processes that occur during this time can be observed by live-imaging. Such imaging can be performed simply using transmitted light (if one is only interested in gross changes in morphology) or with confocal microscopy using transgenic flies expressing fluorescent probes or follicles stained with live imaging dyes. Live imaging is being used to substantially advance our understanding of developmental processes. Indeed, live imaging of late stage follicles has expanded the knowledge of dorsal appendage migration, an example of tubulogenesis10. We expect that live imaging of additional late stage processes, including actin dynamics during nurse cell dumping, will provide novel insights into these developmental events. It is important to note that while S10A and early stages of follicle development will not continue to develop into a S14 in culture, live-imaging of events occurring during those stages of development is possible using alternative culture conditions11-14 (see Discussion for more information).
Here we provide detailed protocols for isolating late stage follicles for either in vitro development and live-imaging, or molecular analyses (mRNA and protein isolation).
1. Preparing Drosophila Prior to Stage Isolation
2. Isolating Mid-to-late Staged Drosophila Follicles
3. In vitro Development of S10B Follicles
4. Stage Isolation for Live Imaging
5. Stage Isolation for mRNA Preparation
6. Stage Isolation for Western Blotting
When isolating specific stages of Drosophila follicle development it is essential to be able to accurately distinguish the different morphological stages. This is somewhat challenging for S10A and S10B, as the nurse cells and the oocyte each take up half the length of the follicle at these stages (Figure 1B compared to 1C). However, S10A follicles are shorter in length than S10B follicles, as the S10B follicles are fully elongated and thus equal in length to a S14 follicle (Figure 1C compared to 1G). Additionally, a subset of the follicle or somatic cells over the oocyte called centripetal follicle cells begin to migrate in between the nurse cells and the oocyte at S10B. During S10B, which takes ~5 hr, the nurse cells undergo dynamic actin remodeling so that at S11 the nurse cells contract, squeezing their cytoplasmic contents into the oocyte. Thus at S11, the nurse cell region has significantly decreased, while the oocyte has expanded (Figure 1D). At S12, the nurse cells begin to undergo death and take on a more opaque appearance (Figure 1E); interestingly, in culture, nurse cell remnants curl dorsally in S12 follicles (see S12s in Figures 3C'-D'). During S13, the nurse cell region is transparent as cell death is being completed, and dorsal appendage formation is occurring (Figure 1F). By S14 only the oocyte and the follicle cells remain, and dorsal appendage formation is complete (Figure 1G).
The in vitro development assay can be used to assess the consequences of various pharmacological treatments and genetic mutations on S10B development and nurse cell dumping (Figure 3). The development of S10Bs in culture is robust, however a number of factors can lead to poor experimental results. In Figure 3A, experimental data from both a successful and a failed experiment are provided. A successful experiment is one in which 80-100% of the wild-type follicles in control media complete nurse cell dumping and progress to S12-14 (Figure 3A, wt 1). Occasionally wild-type controls will fail, meaning that fewer than 80% of the follicles develop (Figure 3A, wt 2). This failure can be caused by a number of issues including: 1) inability to distinguish S10A from S10B follicles (refer to Figures 1 and 2), 2) media problems (temperature, age, etc.), and/or 3) follicles were exposed to debris for too long.
In vitro development of S10Bs can be used in combination with pharmacologic treatment. For such experiments, it is important that drug controls, i.e. treatment of wild-type follicles with the drug, are performed with every experiment because drug effectiveness and/or concentration can change with time. For pharmacologic experiments, it is convenient to analyze the ratio of the percentage of follicles developing in the drug treatment to that in control media; this controls for slight genetic background differences. It is often useful to treat with the IC50 concentration of the drug; this is the concentration that blocks 50% of wild-type (yw) follicles from undergoing nurse cell dumping and further development. Thus, the ratio for wild-type follicles is expected to be 0.5 (Figure 3B, drug 1). Figure 3B is an example of how a drug (aspirin) can become too concentrated (drug 2; likely due to solvent evaporation) and block greater than 50% of the wild-type follicles from developing. To further illustrate the in vitro development assay, images of two wells of developing S10B follicles are provided. Figures 3C and D illustrate the S10B follicles at the beginning of the assay, in control or aspirin treated (~2 mM) media. Figures 3C' and D' illustrate the end of the assay, revealing that the majority of the control treated follicles developed to S14, while the majority of the aspirin treated follicles have not completed nurse cell dumping.
The in vitro development assay can also be used to screen for genetic backgrounds that are more sensitive to a particular drug. Figure 3E contains two examples of this. In the first example, the genetic background (expt1, blue bar) fails to interact with aspirin as the ratio remains ~0.5; conversely, in the second example, the genetic background (expt2, red bar) enhances the effect of aspirin. Thus, the in vitro development assay can be used to define the consequences of mutations and pharmacologic reagents on the developmental and morphological events occurring during late oogenesis; additionally, the assay can be used to assess both pharmacologic and genetic interactions (see9).
Stage isolation can also be used for live imaging of developmental processes. Figure 4 is an example of a time-lapse movie of a S10B follicle expressing Utrophin-GFP, the actin binding domain from human Utrophin fused to green fluorescent protein. Using this imaging tool, actin bundle formation and condensation can be visualized. Many tools are available for live imaging, including protein trap transgenic lines16,17, UAS driven fluorescently tagged organelle markers (mitochondria, golgi, endoplasmic reticulum, etc.), UAS driven fluorescently tagged cytoskeletal markers (actin and actin binding proteins, microtubule and microtubule binding proteins), transgenic lines expressing any fluorescently tagged protein of interest, and vital dyes (Nile Red labels neutral lipids, Edu labels replicating DNA, FM4-64 labels membranes).
Molecular analyses can be performed using isolated stages of Drosophila follicles. For protein analysis by western blotting, it is necessary to empirically determine how many follicles, of a particular stage of development, are required to observe the protein of interest. Figure 5 is an example of how to sequentially dilute a concentrated protein lysate to determine the number of S10B follicles needed to observe a particular protein, Fascin. In this case one S10B follicle is sufficient to observe this protein by western blotting.
Figure 1. Diagram describing the cellular composition of Drosophila follicles and images illustrating the morphological difference of mid-to-late stage Drosophila follicles. A. Diagram depicting the cellular composition of a S10A follicle. B-G. Representative images of S10A-S14 Drosophila follicles taken using a stereo dissecting scope. The Drosophila follicle consists of 16 germline-derived cells: 1 oocyte (red, A) and 15 nurse or support cells (yellow, A), which are surrounded by ~650 somatically-derived epithelial cells (white, magenta, and cyan, A). At S10A, one half of the length of the follicle is composed of the nurse cells (yellow bracket, B), which are covered by the stretch follicle cells (magenta in A), and the other half is composed of the oocyte (red asterisk, B), which is covered by the main body follicle cells (white in A). At S10B, the nurse cells (yellow bracket, C) and oocyte (red asterisk, C) each compose half of the length of the follicle, however the overall length of the follicle is now equal to that of a S14 follicle (compare C 세스 G). During S11, the nurse cells (yellow bracket, D) rapidly squeeze their cytoplasmic contents into the elongating oocyte (red asterisk, D) in an actin/myosin-dependent process termed nurse cell dumping. By S12, the oocyte (red asterisk, E) has fully elongated as nurse cell dumping is complete and only nurse cell remnants remain (yellow bracket, E). The nurse cell remnants (yellow bracket, F) complete cell death at S13. S14 represents the fully mature follicle, which is composed of the oocyte (red asterisk, G), somatic cells, and eggshell, including the dorsal appendages (white arrows, G). B. Scale bar = 0.1 mm.
Figure 2. Images providing an overview of ovary dissection and follicle isolation. A. Submerge the fly in the dissecting media and orient it so that it is held, by forceps, in the nondominant hand. B. Using the dominant hand, grab the cuticle at the posterior of the abdomen. C. Pull the cuticle off, exposing the ovaries (yellow arrow). D. Working from the anterior of the abdomen toward the posterior, gently squeeze the abdomen releasing the ovaries (yellow arrow). E. Separate the ovaries (yellow arrow) from any remaining cuticle (red arrowhead) and/or internal organs (white arrowheads). F. Transfer the isolated ovaries to a fresh well containing dissecting media. G. Tease apart the ovaries, using dissecting needles, to expose individual follicles (compare intact ovary (top) to separated ovary (bottom). White arrowhead points to a stabbed S10B follicle, an example of a follicle that should not be used for subsequent experiments. H. Select the morphological stages of interest (Stages 10A-14 (S10A-S14) shown).
Figure 3. Examples of in vitro development assays. A. Chart of the percentage of S10B follicles that complete development in culture. wt 1 is an example of expected development of wild-type S10B follicles (96% developing), while wt 2 is an example of poor development in culture (68% developing). We would typically discard the whole experiment if the development of wild-type S10B follicles in control media is below 80%. B. Chart of the ratio of the percentage of S10B follicles developing (dev.) in media containing the IC50 for aspirin to the percentage developing in control media. The IC50 is the concentration that blocks 50% of the S10B follicles from developing; this means the wild-type value should be ~0.5. Drug 1 is an example of the expected wild-type ratio (0.52), while drug 2 is an example of what happens when the drug concentration is too strong (0.37). C-D'. Images of in vitro development wells. C-C'. Control treated. D-D'. Aspirin treated (~2 mM). C-D. Image of the S10Bs at the start of the assay. C'-D'. Image of the follicles at the end point of the assay. Control treated S10B follicles develop to S14s in culture (C'), while the majority of aspirin treated follicles fail to complete nurse cell dumping (D'). E. Chart of the ratio of the percentage of S10B follicles developing (dev.) in media containing the IC50 for aspirin to the percentage developing in control media. This is an example of two experiments looking for pharmaco-interactions. The expt1 (blue bar) mutant does not alter the effect of the IC50 for aspirin (0.57), while the expt2 (red bar) mutant enhances the effect of aspirin (0.16).
Figure 4. Example showing the use of isolated S10B follicles for live imaging. A-F. Maximum projections of 3 slices from time-lapse z-stacks of a S10B follicle isolated from Utrophin::GFP expressing transgenic flies (sqh-Utrophin::GFP). A'-F'. Magnified insets highlight a single nurse cell from A-F. A-F'. F-actin (Utrophin::GFP), white. A, A'. time (t) = 0 min. B, B'. t = 20 min. C, C'. t = 40 min. D, D'. t = 60 min. E, E'. t = 80 min. F, F'. t = 100 min. At the initial time point, short actin filaments can be observed extending inwards from the nurse cell membranes (A, A'). These actin filaments elongate throughout the early time points (B-C' compared to A, A') until they are fully elongated (D, D'). In later time points, these fully elongated actin filaments then shorten and condense throughout nurse cell dumping (E-F') A. Scale bar = 50 μm. A'. Scale bar = 10 μm.
Figure 5. Illustration of how to determine the number of follicles needed to examine a particular protein by western blot analysis. This is a western blot looking at Fascin expression in S10B follicles (1:20, sn 7C, Cooley, L.; Developmental Studies Hybridoma Bank (DSHB), developed under the auspices of the NICHD and maintained by the University of Iowa, Department of Biology, Iowa City, IA, 52242). The numbers across the top represent the approximate number of S10B follicles loaded per lane.
The Drosophila follicle is composed of only a small number of cell types, making it ideal for both morphologic and molecular analyses. Furthermore, due to the structure of the ovary, it is relatively easy to obtain large numbers of specific stages of follicle development with a common dissecting scope and minimal training. As each stage represents a short temporal window, stage isolation can provide significant molecular insights into the developmental processes occurring during that stage. For example, we have used mid-to-late stage follicle isolation to characterize the gene expression changes occurring during S10B, S12, and S1418. This analysis suggests a number of previously uncharacterized genes are likely to contribute to follicle development and, in particular, eggshell formation.
Stage isolation can provide dramatic insights into morphological events through live imaging. Such imaging can be performed on all stages of Drosophila follicle development. While the focus of this work is S10B-14, readers are referred to the following articles for live-imaging of earlier stages: germarium13, follicle elongation19, and stage 911,12,14. The culture conditions utilized in these studies are distinct from those discussed in this work. Specifically, these studies used Schneider's Insect Media with variable levels of FBS (2.5-15%) as well as the addition of insulin11-13,19 and, in some cases, the further addition of trehalose, methoprene, 20-hydroxyecdysone, and adenosine deamidase14. It is important to point out that these live imaging studies have significantly advanced our understanding of the events occurring during these earlier stages of development. However, these early stage follicles cannot progress all the way to the final stage of follicle development, S14. Conversely, S10B-13 will develop to S14 in culture.
During S10B-S14 many morphological processes occur that can be studied by live imaging. For example, live imaging can be used to examine the process of nurse cell dumping, the process by which the nurse cells squeeze their cytoplasmic contents into the oocyte to provide it with everything it needs to complete embryogenesis. Nurse cell dumping can be observed by time-lapse imaging using transmitted light 7 or by confocal imaging using transgenic lines expressing fluorescent markers (see Figure 4). Continued use of live imaging is expected to provide novel insights into the cytoskeletal dynamics necessary for nurse cell dumping. Live-imaging of later stages has also advanced our understanding of dorsal appendage primordia migration and tubulogenesis10.
In vitro development of S10B-S13 follicles can be used to define the effects of both pharmacologic reagents and genetic manipulations on the processes occurring during this time. We have used in vitro development of S10B follicles to establish the role of prostaglandins in regulating nurse cell dumping7. Subsequently we have been using this assay to perform a pharmacologic-interaction screen; specifically, we have been testing if loss of one copy of an actin regulator enhances or suppresses the nurse cell dumping defects due to the loss of prostaglandins. This screen has revealed a number of putative downstream targets (9 and Spracklen, Meyer, and Tootle, unpublished data). The assay can also be used to examine genetic interactions by assessing developmental defects, such as a block in nurse cell dumping, due to heterozygosity for two different factors (for an example of this see9).
While isolating mid-to-late stage Drosophila follicles is a fairly simple process, there are a number of critical factors for optimum success. First, the preparation of the flies is very important. It is best to maintain different genotypes of flies in as similar a condition as possible, i.e. the number of flies per vial, the ratio of females to males (2:1 ratio is ideal), and provide fresh, wet yeast consistently. The feeding (wet yeast) and dissection time will alter the prevalence of the different stages of development. We have found that when the flies are consistently fed in the morning, more S10Bs can be isolated in the morning, while more S12s are present in the afternoon. A second critical factor is the media. For in vitro development and live imaging it is essential to prepare fresh IVEM media. Additionally, the media must come to room temperature as cold media will alter the cytoskeletons, both actin and microtubules, and therefore, disrupt further development. It is also important to keep the follicles away from debris. Sometimes, during the dissection, the gut will come out with the ovaries. We have found that if the gut is ruptured and the follicles are kept in the contaminated media, the follicles are unlikely to develop in culture. As the follicles continue to develop in culture, it is essential to reverify the staging after dissection before proceeding with either in vitro development or molecular analyses. Lastly, it is important to have proper controls for the experiments. For in vitro development, it is necessary to use wild-type follicles to test that the media allows for normal development (80-100% complete nurse cell dumping), and to test that pharmacologic reagents act as expected (see Figure 3).
In conclusion, isolation of mid-to-late stage Drosophila follicles can provide significant insight into developmental processes through a variety of experimental techniques.
Table of Specific Reagents and Equipment:
Name of the reagent | Company | Catalogue number | Comments (optional) |
---|---|---|---|
Active Dry Yeast | Genesee Scientific | 62-103 | Any source of Active Dry Yeast is fine |
Grace's Insect Media | Lonza | 04-457F | |
Heat Inactivated Fetal Bovine Serum | Atlanta Biologicals | S11050H | Any Heat Inactivated FBS should work |
10x Pen/Strep | Gibco/Invitrogen | 15140-122 | |
Pin Vises and Needles | Ted Pella, Inc. | 13561-10 | |
Spot Plate, Nine Well | Corning | 7220-85 | |
#5 Dumont forceps | Fine Science Tools | 11252-20 | |
24 multi-well plates | Becton Dickinson | 35 3226 | Any 24-well tissue culture dish should work |
Coverslip Bottom Dishes (35 mm) | MatTek Corporation | P35G-1.0-14-C | Coverslip thickness will depend on the microscope/objective being used |
Glass pipettes | Corning | 7095B-5x (for transferring follicles) 7095B-9 (for producing pulled pipettes) |
|
Sample pestle (1.5 μl; RNase/DNase free) | Research Products International | 199228 | Any plastic pestle that fits 1.5 μl microfuge tubes can be used |
Trizol | Invitrogen | 15596-018 |
The authors have nothing to disclose.
We would like to thank Thomas Lecuit (sqh-Utrophin::GFP line), the Bloomington Stock Center, and the Developmental Studies Hybridoma Bank for reagents. We further thank all members of the Tootle Lab for helpful discussions and critiques of the manuscript. Funding from the National Science Foundation MCB-1158527, and start-up funds from the Anatomy and Cell Biology Department, University of Iowa supported this work. National Institutes of Health Predoctoral Training Grant in Pharmacological Sciences T32GM067795 supported AJS. Data storage support was provided by the ICTS, which is funded through the CTSA supported by the National Center for Research Resources and the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant UL1RR024979.
Active Dry Yeast | Genesee Scientific | 62-103 | Any source of Active Dry Yeast is fine |
Grace’s Insect Media | Lonza | 04-457F | |
Heat Inactivated Fetal Bovine Serum | Atlanta Biologicals | S11050H | Any Heat Inactivated FBS should work |
10x Pen/Strep | Gibco/Invitrogen | 15140-122 | |
Pin Vises and Needles | Ted Pella, Inc. | 13561-10 | |
Spot Plate, Nine Well | Corning | 7220-85 | |
#5 Dumont forceps | Fine Science Tools | 11252-20 | |
24 multi-well plates | Becton Dickinson | 35 3226 | Any 24-well tissue culture dish should work |
Coverslip Bottom Dishes (35mm) | MatTek Corporation | P35G-1.0-14-C | Coverslip thickness will depend on the microscope/objective being used |
Glass pipettes | Corning | 7095B-5x (for transferring follicles) | |
Glass pipettes – long | Corning | 7095B-9 (for producing pulled pipettes) | |
Sample pestle (1.5 μl; RNase/DNase free) | Research Products International | 199228 | Any plastic pestle that fits 1.5 μl microfuge tubes can be used |
Trizol | Invitrogen | 15596-018 |