Drosophila larval epidermal cells only exhibit epidermal aging when they persist to the adult stage

ABSTRACT Holometabolous insects undergo a complete transformation of the body plan from the larval to the adult stage. In Drosophila, this transformation includes replacement of larval epidermal cells (LECs) by adult epidermal cells (AECs). AECs in Drosophila undergo a rapid and stereotyped aging program where they lose both cell membranes and nuclei. Whether LECs are capable of undergoing aging in a manner similar to AECs remains unknown. Here, we addressed this question in two ways. First, we looked for hallmarks of epidermal aging in larvae that have a greatly extended third instar and/or carry mutations that would alter the pace of epidermal aging at the adult stage. Such larvae, irrespective of genotype, did not show any of the signs of epidermal aging observed in the adult. Second, we developed a procedure to effect a heterochronic persistence of LECs into the adult epidermal sheet. Lineage tracing verified that presumptive LECs in the adult epidermis are not derived from imaginal epidermal histoblasts. LECs embedded within the adult epidermal sheet undergo clear signs of epidermal aging; they form multinucleate cells with each other and with the surrounding AECs. The incidence of adult cells with mixed AEC nuclei (small) and persistent LEC nuclei (large) increased with age. Our data reveal that epidermal aging in holometabolous Drosophila is a stage-specific phenomenon and that the capacity of LECs to respond to aging signals does exist.

. The UV irradiation was 15-16 mJ/cm 2 for Figure 3C and Figure 4. The larvae after treatment were recovered on fly food. The number of pupae and eclosed adults were counted and the epidermal membranes of adults eclosed after UV treatment were analyzed.

Dissections, immunostaining and microscopy
Larvae were dissected and stained as previously described (Burra et al., 2013). Adult flies were anaesthetized by briefly exposing to CO 2 and dissected as described (Scherfer et al., 2013).
Briefly, flies were placed on a Sylgard plate (Dow Corning). The head and other appendices were removed using forceps and the thorax and abdomen were pinned dorsal side up using 0.1 mm diameter dissection needles (Fine Science Tools). After placing the first set of pins, 1X phosphate buffered saline (PBS) was added to the Sygalrd plate and any bubbles trapped beneath the ventral abdomen were gently flushed. Using dissecting scissors (Fine Science Tools) an incision was made and the flaps were pinned to the side. The viscera and other organs were removed and the pins were repositioned to flatten the epidermis. The dissected samples were fixed in 3.7% formaldehyde prepared in 1X PBS for 1 hour and then washed quickly with 1X PBS. After rinsing out the formaldehyde, samples were unpinned and transferred to 0.5 ml microtubes containing PHT buffer ( phosphate-buffered saline containing 1% heat-inactivated normal goat serum and 0.3% Triton X100) for 1 hour before immunostaining. The samples were incubated overnight in solution containing 1: 50 dilution of mouse anti-Fasciclin III (Drosophila Hybridoma Bank). After primary antibody incubation, the samples were washed with PHT. Next, the samples were incubated in PHT solution containing goat anti-mouse Alexa 647 (1:500, Abcam). The samples were again washed with 1x PBS containing 0.3% Triton X100 and mounted in Vectashield mounting medium (Vector Laboratories). Larval histoblast images were captured at room temperature with a Leica MZ16 FA fluorescent stereomicroscope equipped with a PLAN APO 1.6x stereo-objective and a JENOPTIK ProgRes C14 Plus digital camera.
Image Pro Plus 7.0 softwawre (Media Cybernetics) was used. Other epidermal whole mount images were captured with an Olympus FV 1000 laser scanning confocal microscope and a UPLAPO (10x/0.40 NA or 20x/0.70 NA) objective using FLUOVIEW version FV10-ASW 3.1 software at room temperature and processed with ImageJ 1.52n.

Introduction
Holometabolous insects undergo a complete restructuring of the body plan during metamorphosis. Typically, this involves changing the morphology of a larva (usually a wormlike transitional stage) into an adult that possesses adult appendages (legs, wings, antennae) and functional reproductive organs. In some insects, including Drosophilid flies, the larval stage can enter a diapause which temporarily halts further development (Enomoto, 1981) until the local environment is conducive to further growth. In a phenomenon distinct from diapause, certain dietary restrictions that prevent synthesis of the molting hormone (Parkin and Burnet, 1986) can block the onset of pupariation and result in long-lived Drosophila larvae. A variety of genetic mutations also result in a greatly prolonged larval stage (Belinski-Deutsch et al., 1983, Sandoval et al., 2014, often with no pupariation or metamorphosis. As in most insects, the Drosophila larval epidermis is a monolayer of large polarized polygonal epithelial cells that are adherent to an apical cuticle (Gangishetti et al., 2012) and that synthesize a basal lamina (Fessler and Fessler, 1989) separating them from the hemolymph in the open body cavity. Little is known about whether or how prolonged larval stages affect barrier tissue architecture. During metamorphosis, epidermal histoblast cells (Madhavan and Madhavan, 1980) proliferate and migrate to replace the larval epidermal cells (LECs) undergoing apoptosis (Ninov et al., 2007), thus forming a new epidermal sheet and cuticle. After eclosion, the adult epidermal cells (AECs) are substantially smaller than their larval counterparts and more rounded in shape (Scherfer et al., 2013). Like their larval counterparts (Galko and Krasnow, 2004), they are capable of undertaking physiological responses such as wound healing (Losick et al., 2013, Ramet et al., 2002. Adult flies undergo a genetically-programmed aging process in which the total average lifespan can be shortened by certain mutations (Juhasz et al., 2007, Munoz-Alarcon et al., 2007 and lengthened by others (Clancy et al., 2001, Libert et al., 2007. In a striking example of a tissue-specific aging program, many of the AECs grow thinner, lose the membranes intervening between nuclei, and eventually lose nuclei as well as adult flies age (Scherfer et al., 2013).
The question of whether larval stages can undergo a normal "aging" process, either during the normal window of development or during a prolonged version of this window, has not, to our knowledge, been addressed in the era of molecular/genetic aging research. We approached this question experimentally in two distinct ways. First, we manipulated the larval diet to create "longer-lived" larvae. We did this in control larvae, and in mutants that would normally accelerate or decelerate aging at the adult stage (Juhasz et al., 2007, Munoz-Alarcon et al., 2007. We then examined the larval barrier epidermis for the normal morphological hallmarks of adult aging (Scherfer et al., 2013)-primarily loss of membranes between intervening nuclei. Second, we developed a protocol that effects a "heterochronic" persistence of LECs into the adult epidermal sheet. We then examined whether these hybrid epidermal sheets containing both LECs and AECs, and the different cell types within them, underwent a normal process of adult skin aging. The results are presented and discussed below.

The larval epidermis does not exhibit signs of epidermal aging
Previously, we observed that the Drosophila adult epidermis undergoes an age-dependent loss of cell membranes (See schematic Fig. 1A) and nuclei (Scherfer et al., 2013). To help determine whether skin-aging signals are specific to the adult stage, we asked whether larvae, a transitional juvenile form that precedes metamorphosis, also exhibit a similar age-dependent changes in the epidermis. The Drosophila third instar larval stage (L3) normally lasts for 2 days before the puparial molt. The time spent in this stage can be substantially prolonged through nutrient deprivation that precludes synthesis of the molting hormone (Parkin and Burnet, 1986). Larvae grown on media containing yeast mutant for the ERG2 gene cannot synthesize molting hormone and do not pupariate (Katsuyama and Paro, 2013). We developed a scheme to grow control larvae (w 1118 ) and larvae mutant for aging genes on either normal media (NM) or media made with Erg2 mutant yeast (E2M) (Fig. 1B). Drosophila grown on normal media (NM) typically reach L3 after five days and have large polygonal epidermal cells with distinct cell membranes ( Fig. 1C). On NM, both lam G262 mutants, which exhibit a short lifespan (Munoz-Alarcon et al., 2007) and premature/accelerated skin aging as adults (Scherfer et al., 2013) and ATG 7d77 mutants (Juhasz et al., 2007), which exhibit decelerated skin aging as adults (Scherfer et al., 2013), exhibited a morphologically normal larval epidermis at the middle of the L3 stage ( Fig. 1 D-E). When grown on E2M, ten day old L3 larvae of all genotypes tested also showed an epidermal morphology that was indistinguishable from the younger larvae grown on NM ( Fig. 1 F-H). These results suggest that there is no equivalent, in larvae, to the progressive deterioration of epidermal cell membranes that is observed in the adult. This is true regardless of whether the larval genotype would or would not exhibit an aging phenotype at the adult stage.

UV treatment of larval histoblasts to create adults with persistent larval epidermal cells
The adult epidermis is formed through proliferation and migration of larval epidermal histoblasts after the puparial molt (Madhavan andMadhavan, 1980, Ninov et al., 2007). During normal development these histoblast cells, which are diploid precursors embedded within the polyploidy larval epidermis ( Fig. 2A), expand and migrate to replace dying larval epidermal cells (LECs).
To interfere with this replacement process, and hopefully create Drosophila adults that contained persistent LECs, we developed a protocol where the lateral aspect of larvae was irradiated with UV (see methods and Fig. 2B). We first defined a UV dose that is not lethal. We observed full survival to the pupal stage up to 20 mJ/cm 2 of UVC (Fig. 2C). Survival to the adult stage was complete up to 15 mJ/cm 2 but dropped by over half when the dose was increased to 20 mJ/cm 2 .
Do LECs in irradiated larvae persist through metamorphosis and into the adult stage? To assess this, we irradiated larvae that carried an epidermal Gal4 driver (A29-Gal4) and a nuclear-

LECs that persist until the adult stage do undergo epidermal aging
We next asked what happens to these persistent LECs and the resident AECs as the adults with a hybrid epidermal sheet age. If LECs are immune to adult epidermal aging signals, as might be suggested by their behavior in longer-lived larvae, the expectation is that these cells would stay primarily mononucleate as the adult ages. On the contrary, if LECs can respond to epidermal aging signals when embedded in the adult epidermis we would expect them to lose membranes both between themselves and the surrounding AECs. The latter is what we observed.
In the week-old unirradiated adult epidermis ( The same diversity of cell types is present as the adult ages to two weeks. In the absence of irradiation the loss of epidermal membranes progresses with age (compare Figure 4A Several possibilities might explain the inability of LECs to show morphological signs of aging, even when the larval stage is prolonged. One is that larvae simply do not produce the systemic signal(s) that might accompany normal adult aging. A second is that these signals exist, but that LECs are not responsive to them at this stage. Our heterochronic transplantation experiment tested this latter possibility. By developing a protocol to create adult flies that harbor persistent LECs we were able to examine the cellular morphology of LECs in a tissue that normally undergoes an age-related morphological progression. The resulting epidermis, a hybrid of LECs and AECs, underwent a normal adult epidermal aging program in which all cells participated. This suggests that LECs do have the capacity to respond to systemic aging signals present in the adult.
Most adult insect tissues are epithelial in nature and are not, as in vertebrates, replenished during adult life by resident stem cells. Each of these tissues has its own organ-specific aging program that can likely be monitored at the cellular level. For those tissues, like the barrier epidermis that have a contribution from nests of imaginal cells that are set aside at the larval stage, one can imagine using a similar irradiation strategy to interfere with replacement of the larval cells during metamorphosis. Another such tissue is the Drosophila tracheal system (Weaver and Krasnow, 2008). Although the organ-specific aging program of this tissue has not been examined in cellular detail, some details of the replacement process are known, including its dependence on FGF signaling (Chen and Krasnow, 2014), suggesting that either irradiationbased or genetic strategies (interfering with FGF signaling) might be viable strategies for creating heterochronic animals with a mixture of larval and adult tracheal cells. We hope that the experimental strategies outlined here will prove adaptable to other tissues, allowing an examination of how generalizable the result obtained here, with barrier epidermal cells, proves to be.          T  A  T  A  (  D  I  P  T  E  R  A  :  D  R  O  S  O  P  H  I  L  I  D  A  E  ) .