Protocols to Study Aging in Drosophila

The fruit fly Drosophila melanogaster offers a host of advantages for studying the biology of aging: a well-understood biology, a wide range of genetic reagents, well-defined dietary requirements, and a relatively short life span, with a median of ~80 days and maximum ~100 days. Several phenotypes can be used to assess the aging process, but the simplest and most widely used metric is length of life. Here we describe a standard life span assay for Drosophila housed on a simple sugar/yeast diet.


Introduction
As populations around the world age, increasing effort is being devoted to the development of new approaches to improve the health of older people. Remarkably, experimental work on worms, fl ies and mice over the last 20 years has provided a positive outlook on this prospect [ 1 , 2 ] For these model organisms, genetic, environmental, and pharmacological interventions have been described that extend healthy life span [ 3 ]. Even more remarkably given the very different life spans of these model organisms, these interventions often act on common mechanisms to extend life span, implying some degree of evolutionary conservation of mechanisms of aging. Thus there is great promise that studies of aging in laboratory model organisms will yield insights into aging that will ultimately benefi t humans.
The challenges of experimental gerontology are enormous. Experiments require long time-scales, genetic manipulations, large populations, and well-controlled animal stocks and conditions. These factors make the work perfectly suited to the small, shortlived, and well-characterized model organisms such as the fruit fl y Drosophila melanogaster .
Life span experiments have been conducted on Drosophila for the last 100 years [ 4 ] and over time the conditions have been refi ned [ 5 ]. In general, the protocol can be simple, but small and seemingly insignifi cant modifi cations to experimental protocols can have large effects on outcomes. For example, by not controlling for diet quality, genetic background or the interactions between mating frequency and diet, the experiment may report the effects on life span of an uncontrolled, trivial experimental procedure, rather than the focal intervention of the study [ 6 ].
Here we outline the basic procedure for rearing, isolating, and maintaining fl ies for life span experiments, highlighting a number of the known pitfalls that have misled researchers in the past. We provide a basic protocol for wild type fl ies housed under our standard laboratory conditions and then we provide modifi ed protocols for studying the effects on life span of diet , drugs, or genetic interventions .

Materials
All media are prepared using reverse osmosis water. Cooking can be done on a gas hob using a standard saucepan and stirring with a heavy-duty whisk ( see Note 1 ).
1. Egg collection medium (volume suffi cient for ~10 × 15 cm petri dishes): to 250 ml cold water add 12.5 g agar and stir to mix. Bring to boil while stirring and maintain boiling for ~2 min to ensure agar is completely melted. Add 150 ml red grape juice ( see Note 2 ) and stir until the mixture returns to the boil. Remove from heat. Add 25 ml cold water and stir until temperature drops to ~65 °C. Make 10.5 ml 10 % Nipagin (methyl 4-hydroxybenzoate in 95 % ethanol) and pour solution into petri dishes. Allow to cool at room temperature, allowing steam to escape. Ensure to protect the plates from any fl ies at this stage to avoid contamination. Cover and store at 4 °C.
2. Fly food for rearing and maintenance (makes 1 L of 1SY [ 7 ], see Note 3 ): add 15 g agar to 700 ml cold water and stir. Heat until boiling. While continuing to stir, add 50 g table sugar (sucrose) and 100 g yeast (whole yeast autolysate and not water soluble yeast extract). After returned to boil, remove from heat and add cold water to make up to fi nal volume of 1 L. Stir and allow to cool to ~65 °C. Mix in 30 ml 10 % nipagin and 3 ml propionic acid to act as preservatives. This is also the point at which to mix in any small volume additions of drugs/transgene inducers/vehicle control. For larger volume additions, reduce the cold water addition after cooking to ensure fi nal total volume remains at 1 L. Using a peristaltic pump with clean, sterilized tubing, dispense into clean vials or

Media
bottles. Allow to cool at room temperature for several hours ( see Note 4 ). To avoid contamination, ensure to protect cooling food from fl ies ( see Note 5 ). Plug individual vials with cotton balls ( see Notes 6 and 7 ). Store at 4 °C.
3. Live yeast paste for stimulating egg laying: Mix dried baker's yeast granules with cold water at a ratio of approx. 1:1 by weight to make a stiff paste (ice cream consistency). Best when used immediately, but can be stored covered at 4 °C for 2 weeks.
2. Fly "cage" for housing parental fl ies: an ~15 cm long plastic cylinder that fi ts a petri dish snugly at one end, and is covered with mesh at the other.

Methods
Most laboratory stocks are kept in small numbers and under crowded conditions, both of which alter adult life span [ 6 , 8 ].
It is important to implement procedures to control these factors so that they do not confound interpretations of alterations in fl y life span.
1. "wild types": to escape the transgenerational effects of stock crowding on life span, we passage stock-derived fl ies through two generations of our standard density procedure before use in life span experiments.

Preparing Stocks for Egg Collection
2. Genetic crosses: it is extremely important to standardize the genetic background of all mutant lines to be compared in a life span experiment. Failure to do so is common and leads to incorrect conclusions about the effects of genetic interventions to extend life. Most experimental transgenic fl ies are generated by crossing two inbred lines, with one containing the transgene to be activated and the other containing a genetic construct that drives the expression of the fi rst. This cross also produces a hybrid genetic background, and this will generally increase life span when compared with that of the inbred controls, as a consequence of heterosis and irrespective of any effect of the transgenes [ 9 ]. To avoid this problem, all transgenes and mutants should fi rst be backcrossed into a standardized genetic background for at least six generations. To maintain the lines an additional 2-3 backcrosses should be repeated every 6-12 months ( see Note 11 ). Furthermore, each of the transgenic lines used to construct the experimental line should be included as a control in the life span experiment, because transgenes can cause insertional mutagenesis, which can in turn modify longevity.
1. House parental fl ies in "cages." Provide a generous smear of live yeast paste (~1 tsp) at the center of the egg laying plate.
2. After 48 h, replace egg laying surface ( see Note 12 ) with a fresh plate harboring a fresh aliquot of live yeast paste (egg laying peaks ~72 h after introduction to rich food) ( see Note 13 ).
4. Collect embryos for development at standard density. To achieve this, we either use a pipette to allocate a fi xed volume of a dense embryo suspension into new media for development , or use a mounted metal pick to collect and transfer individual larvae to development media ( see Note 15 ): Pipetting method (ideal for robust genotypes , to yield large numbers of experimental fl ies) : 1. Anesthetize fl ies in cage, remove egg laying plate on which fertilized eggs lie and discard any yeast paste not consumed ( see Note 16 ).
2. Using a squeeze bottle containing PBS, cover the plate with a thin layer of buffer.
3. Dislodge eggs by "brushing" the egg laying surface with a fi ne paint brush.
4. Pour egg/PBS suspension into a 15 ml falcon tube and allow eggs to settle.

5.
Pour off most of the PBS and add more fresh PBS to wash the eggs.

To Collect Staged Embryos
6. Allow eggs to settle and pour off most of the PBS, leaving only suffi cient to cover the settled egg mass.
8. Using 100 μl pipette with wide bore tip ( see Note 10 ), set volume to 18-20 μl and insert tip into the solution so the tip is level with the top of settled egg mass; quickly release plunger while dropping tip into the mass of eggs.
9. Inspect tip for a dense, even, mass of eggs ( see Note 17 ).
10. Dispense egg mass on to surface of ~70 ml SY medium in a 250 ml bottle.
Picking method (more labor-intensive than pipetting, but more fragile genotypes tend to fare better using this method ) :  [ 10 -14 ]. Also, genotype and food quality interact with courtship and mating frequency [ 15 , 16 ]. Housing experimental fl ies as a single sex population avoids the confounding effects of sex X treatment interactions that modify life span .   There are relatively few deaths up until day 60, from which point there is rapid loss of life. By contrast, the population illustrated by the red line suffers substantial numbers of deaths beginning at day 20. Thus many fl ies are dying at young and middle ages, rather than predominantly at old age. This is a sign of poor housing conditions or a genetically fragile stock 4. An important recent advance has been the publication of an openly available database for storing life span data, called SurvCurv [ 17 , 19 , 20 ]. Users can upload data for secure storage as well as use an array of statistical tools to analyze the experimental outcomes. Additional tools available on the site allow the life spans to be compared to others in the database and so can be used to aid further biological discoveries.

Notes
1. Automatic cookers with built-in stirrers like the Joni Multimix (Joni Foodline) are useful for standardizing large volume cooks.
2. We use red grape juice that is designed for use in home wine production. Many laboratories use apple juice.
3. Our simple recipe of sugar and whole yeast lysate provides nutrition for optimal development and life span . Many alternatives exist, but not all are optimal (see supplement to [ 5 ]

Data Handling
Lifespan Measurement in Drosophila Melanogaster that contains all necessary nutrients to support long life [ 21 ]. It is important to note that our recipes contain the nutritional complement of whole yeast preparations, which cannot simply be replaced by water soluble yeast extract that does not support long life [ 7 ]. 4. In a relatively cool climate where room temperatures do not exceed 22 °C, this can be overnight. If medium shrinks in vials and pulls away from the edges, this is a sign of over-drying.
5. Housing trays of vials/bottles in pillowslips as they cool is a useful way to protect them from stray fl ies.
6. Alternatives to cotton wool balls exist: for example polyurethane foam plugs (available from www.drosophilacenter.com ) are mite resistant, retain their structure and are reusable after washing.
7. To avoid the need to plug hundreds of vials before storage, it is possible to seal trays with Glad ® Press'n Seal. If doing so, it is extremely important to ensure the seal is sound, there are no holes in the plastic fi lm and all vials are covered to avoid both contamination and food from excessive drying when cooled.

A useful resource for equipment suitable for use in Drosophila
research is the supplier: www.fl ystuff.com (a division of Genesee Scientifi c). 9. In situations where small numbers of parental fl ies are used for egg lays, it is more space and resource effi cient to use small (~5 cm diameter) petri dishes and cages. 10. We use tips from StarLab (Cat Number: E1011-5100), but it is also possible to cut back a standard pipette tip a few mm to make a wide opening.
11. Backcrossing for six generations is, in almost all cases, suffi cient to eliminate the confounding effects of genetic background. It should be noted that this should be performed to each laboratory's own genetic stocks since even inbred lines with the same name will differ between laboratories [ 22 ].
12. If not experienced with fl y handling, replacing the egg laying plate may require fl ies to be lightly anesthetized with CO 2 .
13. It is important not to use too much live yeast for the egg collection plate as it interferes with egg collection. Nor do you want to use too little such that the yeast supply is exhausted. Aiming to have a small amount left at the egg lay is ideal. A cage of ~300 fl ies will consume ~1 tsp of live yeast overnight.
14. While overnight egg lays produce adequate synchronization for most life span experiments, this egg collection window can be reduced. 15. To time the emergence of adults so that it falls on a weekday, transfer embryos to fresh food for development on a Friday. Emerging fl ies will be available on Monday, 9.5 days later.
16. If egg yield is a problem, the same parents can be used for an additional lay on a fresh plate containing live yeast. 17. A dense mass of eggs yields ~300 adult fl ies.
18. Before incubating, it is best to remove any leftover yeast paste from the egg laying plate as emerging larvae will burrow into it, making them hard to collect.
19. This is the smallest of three larval stages.
20. Before transferring larvae into fresh media, make a dent in the food to make it easier to wipe off the larvae against a slope of food.
21. For practical reasons, collecting larvae by picking is more manageable using 30 ml vials containing ~7 ml of SY food.
Overcrowding can be avoided with 30-50 larvae.
22. At the larval densities recommended in this protocol, larvae will migrate to the cotton wool to pupate. If the container is not tightly plugged, the larvae will escape from the bottle. 23. In order to collect virgin fl ies, check bottles at 9 days after egg transfer and clear any fl ies that have emerged. Check the bottle every ~4 h for newly emerged fl ies-these will all be virgins. Transfer virgin fl ies to a cold, clean bottle on ice and sort males from females while they remain in a chill coma. CO 2 should be avoided as the fl y's cuticle is immature, and exposure to the gas can lead to adverse effects on life span. Genders can only be distinguished with the use of a dissecting microscope to examine the genitalia.
24. It is best to transfer newly emerged fl ies without using CO 2 .
25. In order to anesthetize a whole bottle of fl ies rapidly, fi ll a fresh empty (without food), dry bottle with CO 2 and transfer all fl ies into it. When sorting anesthetized fl ies, work on a perforated plate through which a stream of CO 2 is passing. To avoid desiccating the fl ies, it is ideal to bubble the CO 2 through water before it reaches the fl ies. 28. Experimental conditions can be blinded to the experimenter at this stage.
29. Fly life span varies with the size of their housing. Experiments comparing the life spans of fl ies kept in 25 ml versus 500 ml fl asks, but at a standard density per container volume, found that life span was signifi cantly shorter in the larger vol ume fl asks. Shorter life span was associated with higher levels of fl ying activity [ 23 ].
30. There is a range of densities of fl ies per container that is optimum for life span. In a series of 30 ml vials [ 24 ] found the life span optimum to be for 2-15 fl ies per container and above this density saw a decrease in life span for each increase in population density. 31. As fl ies age and become frail, they have an increased risk of falling and becoming stuck in their food. They will also spend more time at the base of the vial. Storin g the vials on their sides during life spans, so the food is a vertical surface at one end of the vial rather than the fl oor, reduces the risk of these accidental causes of death.
32. Depending on experience with handling Drosophila , during the fi rst 2 weeks of a life span experiment the fl ies may be too fast to transfer between vials without light anesthesia. Males are more active and move more quickly than females and so are more likely to escape without anesthesia. With practice and good technique, it should be possible to transfer fl ies without CO 2 .
33. When transferring fl ies, record deaths and censors (accidental deaths or escapees). Remember to note any dead fl ies transferred to new vials so that they can be deducted from the number of deaths recorded during the next transfer.
34. Some fl ies are bang sensitive, and appear to become mores so with age. These can appear dead during the disturbance of transfer. To avoid counting these as dead, fi rst scan vials for deaths, then transfer all vials to new food and after that, check vials for dead fl ies transferred to fresh media.
35. To reduce labor and use of resources, it is possible to reuse the cotton ball that stoppers a vial by transferring it to the fresh vial to which fl ies are transferred. However, over time the cotton balls will deteriorate and so it is best to replace them at least once a month. 40. Some genetically modifi ed lines will have altered (usually longer) development time. In order to synchronize the start of the life span experiment, initiate the parental crosses for the retarded lines so that egg collection is performed before that of the non-delayed lines. To buffer against slight variations in the delay, it is best to rear multiple batches of the experimental generation, derived from consecutive days of egg laying. This way it will be possible to collect fl ies from all lines that have emerged within 24 h of each other.
41. It may not be possible to control for rearing conditions when using different genotypes in the same way as for single genotypes between multiple experimental foods. However, if the genetic scheme allows, it may be possible to use sibling fl ies as controls for experimental fl ies. Alternatively, it may be possible to rear multiple genotypes in a single rearing container. However, it is important to determine fi rst that these larval conditions to not interact with life span outcomes.
42. To control for rearing conditions, use the protocol employed for testing the effects of multiple food types on one genotype ( Note 40 ).
43. Each laboratory has its own method of recording and plotting these data. An Excel sheet used in our laboratories can be found at: http://piperlab.org/resources/ . More sophisticated and automated packages can be found through the Pletcher laboratory ( see ref. 25 and associated URLs).