Analyzing Starvation-Induced Autophagy in the Drosophila melanogaster Larval Fat Body

Summary

The present protocol describes the induction of autophagy in the Drosophila melanogaster larval fat body via nutrient depletion and analyzes changes in autophagy using transgenic fly strains.

Abstract

Autophagy is a cellular self-digestion process. It delivers cargo to the lysosomes for degradation in response to various stresses, including starvation. The malfunction of autophagy is associated with aging and multiple human diseases. The autophagy machinery is highly conserved-from yeast to humans. The larval fat body of Drosophila melanogaster, an analog for vertebrate liver and adipose tissue, provides a unique model for monitoring autophagy in vivo. Autophagy can be easily induced by nutrient starvation in the larval fat body. Most autophagy-related genes are conserved in Drosophila. Many transgenic fly strains expressing tagged autophagy markers have been developed, which facilitates the monitoring of different steps in the autophagy process. The clonal analysis enables a close comparison of autophagy markers in cells with different genotypes in the same piece of tissue. The current protocol details procedures for (1) generating somatic clones in the larval fat body, (2) inducing autophagy via amino acid starvation, and (3) dissecting the larval fat body, aiming to create a model for analyzing differences in autophagy using an autophagosome marker (GFP-Atg8a) and clonal analysis.

Introduction

Autophagy is a "self-eating" process induced by various stresses, including amino acid starvation¹. Macroautophagy (hereafter referred to as autophagy) is the most well-studied type of autophagy and plays an irreplaceable role in maintaining cellular homeostasis². The malfunction of autophagy is associated with several human diseases³. In addition, some autophagy-related genes are potential targets for treating various diseases⁴.

Autophagy is regulated in a highly sophisticated manner⁵. Upon starvation, the isolation membranes sequester cytoplasmic materials to form double-membraned autophagosomes⁶. Autophagosomes then fuse with endosomes and lysosomes to form amphisomes and autolysosomes. With the help of lysosomal hydrolytic enzymes, the engulfed cytoplasmic contents are degraded, and the nutrients are recycled⁷.

Autophagy is an evolutionarily conserved process⁸. Drosophila melanogaster is a great model for studying the autophagy process in vivo. Amino acid starvation easily induces autophagy in fly fat body tissue, an analog of human liver and adipose tissue⁹. Defects in autophagy disrupt the distinct puncta patterns of several autophagy-related proteins, such as Atg8, Atg9, Atg18, Syx17, Rab7, LAMP1, and p62, among others¹⁰. Therefore, analyzing the patterns of these autophagy markers will help discern the occurrence of autophagy defects and the defective autophagy step. For example, the ubiquitin-like protein Atg8 is the most commonly used autophagy marker¹¹. In Drosophila melanogaster, transgenic strains with a green fluorescent protein (GFP)-tagged Atg8a have been successfully developed¹². GFP-Atg8a is diffused in the cytosol and nuclei in the fed cells. Upon starvation, GFP-Atg8a is processed and modified by phosphatidylethanolamine (PE) and forms puncta, which label the isolation membranes and fully developed autophagosomes^13,14. Through direct fluorescence microscopy, the autophagy induction can be easily observed as an increase in GFP-Atg8 puncta formation¹⁵. Atg8a puncta would not form in response to starvation in the presence of an autophagy initiation defect. As GFP-Atg8a can be quenched and digested by the low pH in autolysosomes, GFP-Atg8a puncta may increase in numbers if autophagy is blocked at late stages¹⁶.

As autophagy is highly sensitive to nutrition availability¹⁷, slight differences in culture conditions often lead to variations in phenotypes. Therefore, clonal analysis, a method that analyzes mutant cells versus wild-type control cells in the same tissue, has a major advantage in dissecting autophagy defects¹⁸. Taking advantage of flippase/flippase recognition target (FLP/FRT)-mediated site-specific recombination between homologous chromosomes, flies carrying mosaic tissues are readily made^19,20. The wild-type cells surrounding the mutant cells form a perfect internal control to avoid individual differences²¹.

The present study describes how to induce autophagy by amino acid starvation and generate GFP-Atg8a-expressing mosaic fat body tissues. These protocols can be used for analyzing differences in autophagy among mutant clones.

Protocol

1. Drosophila crossing and egg laying

1. Introduce 3 male (genotype hsFLP ubiRFP FRT19A; cgGal4 UAS-GFP-Atg8a) and 15 female (genotype y' w* Mu FRT19A/ FM7, Kr GFP) adult flies (see Table of Materials) into a culture vial (with standard cornmeal/molasses/agar Drosophila media at 25 °C) for mating.
   NOTE: Multiple culture vials of the same cross must be set up to ensure enough larvae for further experiments. The male fly strain with genotype hsFLP ubiRFP FRT19A; cgGal4 UAS-GFP-Atg8a carries an FRT site at the X chromosome near the centromere (FRT19A). It also expresses FLP upon heat shock at 37 °C. A transgene with ubiquitous RFP expression is inserted on the X chromosome. GFP-Atg8a is expressed under the control of cgGal4 in the larval fat body²². The female fly strain with genotype y' w* Mu FRT19A/ FM7, Kr GFP carries a lethal mutation (Mu) and FRT19A on the X chromosome. The detailed crossing scheme is shown in Figure 1.
2. For egg laying, transfer the flies to a new culture vial. At 48 h post introduction, tap the "mating" vial until the flies are stunned and drop down on the media at the base of the vial.
   1. Unplug the "mating" vial and cover its mouth with an unplugged inverted vial with fresh media (fresh culture vial). Then, transfer the flies to the fresh culture vial by flipping and tapping the vials.
   2. Plug the fresh culture vial (with the transferred flies) and discard the old culture vial. Place this fresh culture vial in a 25 °C incubator for egg laying on the fresh media.
      NOTE: Prewarming the fresh culture vials to 25 °C for 15 min before the fly transfer process will help the flies adapt to the new environment quickly and hasten egg laying.
3. Remove the flies after 6 h of egg laying. If the experiments need to be repeated, transfer these flies into another culture vial (as described in step 1.2). Otherwise, discard the flies by dumping them into a flask containing 75% ethanol).
   1. Incubate the vial with embryos in a 37 °C water bath for 1 h to induce FLP expression. Subsequently, place them in a 25 °C incubator and allow the embryos to continue to develop.
      ​NOTE: The successful formation of mutant clones can be confirmed subsequently through imaging. The absence of RFP signals marks the mutant clones.

2. Amino acid starvation induces autophagy

1. Using a laboratory spatula, scoop out the media containing the developing larvae into a Petri dish 75 h post egg-laying. Add 3 mL of 1x PBS to the dish and gently separate out the culture media and the larvae using long forceps. Select 10 to 15 early third instar larvae.
   NOTE: Early third instar larvae need to be chosen carefully. The larvae's developmental stage is critical for this protocol's success. Due to the restricted egg laying duration, most larvae in the culture vial are expected to be in the early third instar stage by 75 h of incubation. However, different culture media recipes may lead to variations in the developmental timing of the larvae. The criteria for distinguishing early third instar larvae are the body length, the presence of anterior and posterior spiracles, as well as the mandibular hooks of the mouth apparatus²³.
2. Fill the wells of a 9-well glass depression spot plate (see Table of Materials) with 1x PBS. Place the separated third instar larvae in the wells using long forceps and wash the larvae thoroughly to remove all the media residues.
3. Take 5 mL of 20% sucrose (in 1x PBS) solution in an empty vial and place the clean third instar larvae in this solution using long forceps. Incubate this vial in a 25 °C incubator for 6 h before harvesting them for dissection.
   ​NOTE: The 20% sucrose (in 1x PBS) solution serves as the amino acid-deficient starvation medium.

3. Third instar larvae dissection and sample tissue processing

1. Sharpen two pairs of #5 forceps (see Table of Materials) evenly on both sides with a sharpening stone.
2. Add 400 µL of 1x PBS into each well of the 9-well glass depression spot plate and transfer the larvae into the wells with long forceps. Place one larva in a well with the dorsal side (the side with the trachea) facing up.
   1. Grip the cuticle of the larva with two #5 forceps in the middle of the larval trunk and gently tear open the cuticles. The exposed fat bodies, along with other larval internal tissues, will still be attached to the carcass of the larva. Pull enough to expose the internal organs as much as possible. Repeat this step for all the larvae.
      NOTE: Each larva has two large pieces of fat body along the body trunk. Fat body tissue is a white, opaque, and flat monolayer that can be easily distinguished under the dissection microscope²⁴.
3. Transfer the larval carcass into a 1.5 mL microcentrifuge tube containing 500 µL of 4% paraformaldehyde (PFA). Incubate for 30 min at 25 °C without shaking the tubes.
   CAUTION: 4% PFA is toxic. Wear gloves and masks for protection.
   NOTE: Preparing 4% PFA in 1x PBS buffer is recommended. PFA powder does not dissolve in 1x PBS instantly. Hence, the mixture must be incubated at 65 °C in an incubator overnight or shaken intermittently for 45 min at 25 °C.
4. After 30 min incubation of the carcass with 4% PFA, pipette out the PFA solution, add 500 µL of 1x PBS into the tube, and gently shake the tube on a flat rotator for 10 min before discarding the 1x PBS solution (repeat 3x).
5. Using long forceps, transfer the fixed and washed larval carcass to a well in the 9-depression spot plate filled with 1x PBS. Using #5 forceps, remove all the non-fat body tissues.
6. Use #5 forceps to mount the pieces of the fat body on a microscope slide using 80% glycerol as the mounting medium and lay a coverslip on top.
   NOTE: Before mounting, check the pH of the 80% glycerol (using pH papers, pH = 7 is optimal) to protect the GFP signal. Mounting media (see Table of Materials) with DAPI in 80% glycerol will help in imaging the nuclei of fat body cells. These slides are stable for 1 week when stored in the dark at 4 °C.

Representative Results

Under fed conditions, the GFP-tagged ubiquitin-like protein, GFP-Atg8a, is diffused inside the cells. Upon starvation, it forms green puncta and labels autophagosomes. Once autophagosomes fuse with lysosomes, GFP is quenched in the acidic autolysosomes, and the green puncta disappear. If autophagy is not induced or the autophagosome maturation is accelerated, the number of GFP puncta is expected to be low. However, if the fusion between autophagosomes and lysosomes is blocked or the pH of the autolysosome becomes basic, the number and/or size of GFP puncta is expected to be high.

In the protocol presented here, the FLP/FRT system induces mitotic recombination and generates tissues with both wild-type and mutant cells. While the wild-type cells express RFP, the mutant cells lack RFP expression. The expression of RFP in all fat body cells would indicate that FLP/FRT-mediated mitotic recombination was not successfully induced or that the mutation is cell-lethal. In the latter case (i.e., if the mutation is cell-lethal), the larval fat body is expected to have a few cells with higher level RFP signals than the surrounding cells.

In Figure 2, GFP-Atg8a (green) formed puncta in wild-type cells (red), implying that autophagy was successfully induced. In the mutant clones (RFP negative), the pattern of GFP-Atg8a puncta was different from that in the surrounding wild-type cells, suggesting autophagy defects. Very few GFP-ATG8a puncta were detected in Mu1 mutant clones (Figure 2B), indicating that autophagy was blocked at the initiation steps or accelerated autolysosome maturation. The numbers and size of GFP-Atg8a puncta were greatly increased in Mu2 mutant clones (Figure 2C), suggesting autophagosome-lysosome fusion or autolysosome acidification defects. Further experiments are required to distinguish these possibilities. Moreover, the sizes and number of puncta need to be quantified to determine whether the observed differences are statistically significant.

Figure 1: Schematic representation of the experimental procedures for monitoring the autophagy marker GFP-Atg8a in a mosaic larval fat body carrying mutant clones. The crossing scheme for generating flies that express GFP-Atg8a in the larval fat body tissues and carry mutant clones is shown (a strain with a lethal point mutation [Mu] on the X-chromosome serves as an example). After heat shock at 37 °C for 1 h to activate FLP expression, the homozygous mutant cells with GFP-Atg8a expression can be generated in the y' w* Mu FRT19A / hsFLP ubiRFP FRT19A; cgGal4 UAS-GFP-Atg8a / + larval fat body. Autophagy is induced in the fat body by incubating the larvae in 20% sucrose solution for 6 h. Please click here to view a larger version of this figure.

Figure 2: The patterns of GFP-Atg8a puncta in Mu1 and Mu2 clones differed from those in the control clones. Mu1 and Mu2 are two independent lethal mutants on the X chromosome. Isogenized y' w*, FRT19A flies serve as a control. GFP-Atg8a (green) patterns were analyzed in the control, Mu1, or Mu2 mosaic larval fat bodies. The mutant clones (or control clones) were negatively marked by RFP (red). (A) In the control clones (RFP negative), the patterns of GFP-Atg8a puncta were similar to those in the surrounding RFP positive cells. (B) In Mu1 mutant clones (RFP negative), the GFP-Atg8a puncta were greatly reduced. (C) In Mu2 mutant clones (RFP negative), the numbers and size of GFP-Atg8a puncta were increased. Scale bar: 10 µm. Please click here to view a larger version of this figure.

Discussion

The present protocol describes the methods to (1) generate flies carrying mutant clones in the larval fat bodies, (2) induce autophagy through amino acid starvation, and (3) dissect the larval fat bodies. In order to generate clones successfully in the larval fat bodies, the following critical steps need to be carried out diligently. (1) Timing the heat shock accurately is crucial because mitotic recombination only happens when the tissue is undergoing mitosis, and (2) both heat shock temperature and duration are critical for inducing FLP expression. A standard 37 °C water bath for 1 h heat shock is recommended. Considering the increased possibility of embryonic/larval death during heat shock and starvation, multiple crossings must be set up for sufficient tissue sample generation for imaging.

In addition, some steps may need to be modified considering the actual culturing conditions. For example, the differences in environment and Drosophila culture media between laboratories may affect the growth rates of larvae. Therefore, the developmental time required for embryos to reach the early third instar stage and the duration of larval starvation (culturing in 20% sucrose to induce autophagy) may need to be standardized in each lab separately.

GFP-Atg8a is the most commonly used marker for determining autophagy. However, the changes in the GFP-Atg8a puncta pattern fail to pinpoint the exact autophagy defect or the affected step directly. Therefore, multiple markers must be analyzed to interpret such data accurately. The protocol presented here can be adapted for other autophagy-related proteins with their respective transgenic strains²⁵. Multiple markers labeled with different fluorescent proteins (such as blue fluorescence protein [BFP]) can be analyzed simultaneously to elucidate the initiation or fusion processes. Controlled autophagy modification, using chemicals²⁶ or RNA interference of critical genes²⁷, can be combined with the present approach to elucidate autophagy mechanisms further. In addition, analyzing endogenous protein markers of autophagy (by using immunohistochemistry) in mosaic tissues is important to confirm the phenotypes²⁸.

Although autophagy was originally recognized as a random, unselective process, increasing evidence indicates that it can be selective and selectively degrades cytoplasmic organelles such as mitochondria and the endoplasmic reticulum, among others²⁹. Further, the procedure described here can be applied to explore selective autophagy. For example, mitophagy can be monitored in fly fat bodies by attaching fluorescent proteins (such as Keima or GFP-RFP tandem tags) to a mitochondrial targeting sequence³⁰.

Materials

Name	Company	Catalog Number	Comments
1.5 mL microcentrifuge tube	Axygen	MCT-150-C
#5 Forceps	Dumont	RS-5015
9 Dressions Spot plate	PYREX	7220-85
Fluorescence Microscope	Nikon	SMZ1500
Glycerol	Sangon Biotech	A100854-0100
KCl	Sangon Biotech	A610440-0500	Composition of 1x PBS solution
KH2PO4	Sangon Biotech	A600445-0500	Composition of 1x PBS solution
Laboratory spatula	Fisher	14-375-10
Long forceps	R' DEER	RST-14
Microscope cover glass	CITOTEST	80340-1130
Microscope slides	CITOTEST	80302-2104
Na2HPO4	Sangon Biotech	A501727-0500	Composition of 1x PBS solution
NaCl	Sangon Biotech	A610476-0005	Composition of 1x PBS solution
Paraformaldehyde	Sigma-Aldrich	158127
Petri dish	Corning	430166
Standard cornmeal/molasses/agar fly food			Tong Lab-made
Stereo microscope	Nikon	SMZ745
Sucrose	Sinopharm Chemical Reagent Co.,Ltd.	10021418
Vectashield antifade mounting medium with DAPI	Vectorlabratory	H-1200-10	Recommended mounting medium
Fly stocks
y'w* Iso FRT19A			Tong Lab's fly stocks
y'w* Mu1FRT19A/ FM7,Kr GFP			Tong Lab's fly stocks
y'w* Mu2 FRT19A/ FM7,Kr GFP			Tong Lab's fly stocks
hsFLP ubiRFP FRT19A; cgGal4 UAS-GFP-Atg8a			Tong Lab's fly stocks