Life History Traits and Metabolic Pool Variation in Neotropical Species of Drosophila (Diptera, Drosophilidae)

© 2023 Academia Sinica, Taiwan Open Access Life History Traits and Metabolic Pool Variation in Neotropical Species of Drosophila (Diptera, Drosophilidae) Camila Heloise dos Santos1, Karoline Aparecida Vieira dos Santos1, Luciana Paes de Barros Machado1, and Rogério Pincela Mateus1,* 1UNICENTRO, CEDETEG, Universidade Estadual do Centro-Oeste, LaGEv, Laboratório de Genética e Evolução do Departamento de Ciências Biológicas, Élio Antonio Dalla Vecchia Alley, 838, Vila Carli, Guarapuava-PR 85040-167, Brazil. *Correspondence: E-mail: [email protected] (Mateus). Tel: +55 42 3629-8179. Fax: +55 42 3629-8145 E-mail: [email protected] (CH dos Santos); [email protected] (KAV dos Santos); [email protected] (de Barros Machado) Received 15 December 2020 / Accepted 20 November 2023 / Published 27 December 2023 Communicated by Benny K.K. Chan The differential exploration of natural resources by Drosophila species has effects on fitness, with changes in life history and metabolic traits. There is a lack of research on the variation in these characters in different environments in Neotropical species of Drosophila. The purpose of this study was to evaluate the profile of life history traits, including viability, development time, and dry weight (as a measure of size), as well as the metabolic pools of triglyceride, glycogen, and protein, in populations from the southern and southeastern regions of Brazil of four Neotropical Drosophila species: D. willistoni, of the Sophophora subgenus, and D. mercatorum, D. maculifrons, and D. ornatifrons, which belong to the Drosophila subgenus. Life history and metabolic traits showed interpopulational variation in at least one species. When significant differences in life history parameters occurred, species of the same subgenus presented similar profiles, i.e., southern populations were larger, less viable, and showed longer development time. This was also observed for triglyceride. However, for the other two metabolic pools (glycogen and total proteins), D. maculifrons and D. ornatifrons presented inverse patterns to the other two species, with the highest values in southeastern populations and the lowest in southern populations. These populational variations indicate plasticity of the examined life history traits, which allows distinctive responses to different environmental conditions shared by species of the same subgenus. Nevertheless, interspecific comparisons did not reflect phylogenetic relationships, with the highest viability being found for D. willistoni and D. mercatorum, which is probably correlated to the ability of these species to explore a broader variety of habitats. On the other hand, the storage capability of metabolic pools seems to be species specific, determined by the adaptive history to the quality and availability of resources, with D. mercatorum (low) and D. ornatifrons (high) having opposing capacities to store metabolites from their diets. Key words: Viability, Development time, Dry weight, Adaptive traits, Energy storage molecules Citation: dos Santos CH, dos Santos KAV, Machado LPB, Mateus RP. 2023. Life history traits and metabolic pool variation in Neotropical species of Drosophila (Diptera, Drosophilidae). Zool Stud 62:56. doi:10.6620/ZS.2023.62-56. BACKGROUND Life history traits, such as viability, development time and weight, and the concentration of metabolites in an organism, are all quantitative characteristics resulting from the interaction between the genotype and the environment. These traits are influenced by several environmental variables, as well as internal factors, such as the stress levels, immunological system activity, composition of the intestinal microbiome, and quality Zoological Studies 62:56 (2023) doi:10.6620/ZS.2023.62-56 1 © 2023 Academia Sinica, Taiwan and caloric content of the diet, among others (Rose 1983; Hoffmann and Parsons 1989; Partridge and Sibly 1991; Rose and Bradley 1998; Matzkin et al. 2009 2011; Jumbo-Lucioni et al. 2010; Jehrke et al. 2018; Flatt 2020). The result of the joint action of all these parameters determines the fitness of the individuals. Thus, research on these relationships is essential to understand the adaptive process and evolution of life history traits of a species. Matzkin et al. (2009) studied the variation in the metabolic pools of twelve ecologically divergent Drosophila species with sequenced genomes. Nine of these species belonged to the Sophophora subgenus, and three to the Drosophila subgenus, two of which are cactophilic. Therefore, there is a gap for non-cactophilic Neotropical models of the Drosophila subgenus, including the analysis of interpopulational variation in these characteristics in populations from different types of habitats for both subgenera of this region, where the process of adaptive divergence could be accentuated by the variety of resources and accessible habitats. One of the conclusions of Matzkin et al. (2009) indicated that although the control of the analyzed metabolites may be similar among species, it appears to be evolutionarily plastic, being able to reflect the response to nutritional necessities of populations. In this context, in order to shed some light on this matter and to increase sampling for species in other regions, the current study analyzed metabolism variation and life history traits in four non-cactophilic Drosophila species native to South America, which were recently collected from different natural areas of the Brazilian Atlantic Forest. Three of the analyzed species belong to the Drosophila subgenus: D. mercatorum (repleta group), D. maculifrons (guaramunu group, Robe et al. 2010), and D. ornatifrons (guarani group, Robe et al. 2010). These species are more closely associated with natural areas of the Neotropical region, the first being found in a broader diversity of environments (open areas and forests, frequently containing enclaves of xerophytic vegetation), and the last two in forest fragments. We also analyzed one species from the Sophophora subgenus, Drosophila willistoni, which, in contrast to the other studied species, also inhabits anthropized environments. The Drosophila populations of this work were collected from the Brazilian southeastern and southern regions. The first region is in the tropical zone and the second is in the subtropical zone. These regions have distinctive Atlantic Forest phytophysiognomies and climate conditions. In the Southeast, seasonal semideciduous forest (SSF) phytophysiognomy of Atlantic Forest is the main type of vegetation, characterized most of the year by higher temperatures and two annual distinctive seasons, wet and dry. In the South region, another Atlantic Forest phytophysiognomy can be detected, mixed ombrophilous forest (MOF), with a predominance of Araucaria pine (Araucaria angustifolia). Part of this forest is known for its high altitude, rigorous winters with frequent frost, high rainfall, and high relative humidity rates. However, in the southern region, the SSF can also be found as fragments within the Pampa Biome. In this case, it has intermediate climate conditions from the southeastern SSF and southern MOF (Instituto Brasileiro de Geografia e Estatística 1992; Backes 1999; OliveiraFilho et al. 2015; Instituto Nacional de Meteorologia 2019). Furthermore, the species selected for this study belong to groups of the most generalist Neotropical species, with high ecological versatility for breeding sites in different plant tissues, fungi, and even dung and carrion (Val et al. 1981; Pereira et al. 1983; Medeiros and Klaczko 2004; Mateus et al. 2006 2018; Gottschalk et al. 2007 2009; Döge et al. 2008; Hochmüller et al. 2010; Goñi et al. 2012; Cavasini et al. 2014; CoutinhoSilva et al. 2017; Mendes et al. 2017; Valadão et al. 2019; TaxoDros 2022). The applied approach allows the inference of which factors related to adaptation and historical evolution could be important to determine the extent of resource usage and habitat occupation of a species. Considering the variety of resources, types of habitats and climate conditions available in the different regions of the Brazilian Atlantic Forest, the main purpose of this study was to evaluate intra and interspecific differences in classical life history traits (viability, development time, dry weight) and metabolic response (triglyceride, glycogen, and protein contents) in Neotropical species of Drosophila. Ultimately, all analyzed characteristics demonstrated different populational adaptive responses in at least one species. Drosophila mercatorum, D. maculifrons, and D. ornatifrons, species from the Drosophila subgenus and associated with natural environments, showed a similar populational pattern for most of the studied traits. However, the interspecific comparison did not result in the same pattern, i.e., D. willistoni and D. mercatorum were similar regarding life history traits, and D. willistoni presented metabolic pools similar to D. ornatifrons and D. maculifrons. The possible role of the environment in the adaptive response of the examined characteristics and the capability of resource exploration demonstrated by the species are discussed. page 2 of 15Zoological Studies 62:56 (2023) © 2023 Academia Sinica, Taiwan MATERIALS AND METHODS Collection areas Drosophilids were sampled in five areas of the Atlantic Forest in the southern and southeastern regions of Brazil, with distinctive climatic and phytophysiognomic characteristics. The collections occurred in two sites of the southeastern region, Serrana and Cajuru, in the state of São Paulo, and in three southern sites, Guarapuava, in the state of Paraná, and Santiago and Porto Alegre in the state of Rio Grande do Sul (Fig. 1). The distance between the closest populations from the southeastern and southern regions (Serrana-SP and Guarapuava-PR) is approximately 600 km, in a straight line, and these areas are divided by the Tropic of Capricorn, which determines, besides different vegetation landscapes, distinct climate conditions: southeastern populations are in the Tropical zone and southern populations are in the Subtropical zone. Additional information about each area is described below: 1) Serrana-SP (SER) – fragment of the Seasonal Semideciduous Forest with xerophytic vegetation; average temperature (T°) = 22.7°C (minimum – MIN = 15.6°C; maximum – MAX = 30.4°C); average relative humidity (RH) = 67%. Summer is warm and humid and winter is characterized by long periods without precipitation; 2) Fazenda Santa Cecília, Cajuru-SP (CAJ) – 32 km away in a straight line from SER, fragment of the Seasonal Semideciduous Forest without xerophytic vegetation; T°, RH, and seasonality data are the same as described for SER; 3) Parque Municipal das Araucárias, GuarapuavaPR (PMA) – fragment of the Mixed Ombrophilous Forest (Araucaria Forest), without xerophytic vegetation; T° = 17.6°C (MIN = 13.4°C; MAX = 24.1°C); RH = 82.2%. Mild temperatures characterize summer, and frequent and severe frosts occur in autumn and winter; 4) Santiago-RS (SAN) – fragment of the Seasonal Semideciduous Forest with xerophytic vegetation, inside Pampa biome; T° = 18.9°C (MIN = 14.4°C; MAX = 24.9°C); RH = 74.8%. Region with humid subtropical climate, no defined dry season; 5) Morro Santana, Campus of Universidade Federal do Rio Grande do Sul, Porto Alegre-RS (POA) – 380 km away in a straight line from SAN, fragment of the Seasonal Semideciduous Forest without xerophytic vegetation, inside Pampa Biome; T° = 20.1°C (MIN = 16.1°C; MAX = 25.8°C); RH = 77.1%. Region with Fig. 1. Map of the collection areas of Drosophila populations from which isofemale lines were obtained and analyzed. The locations are described in the Material and Methods section. SER: 21°15'15.23"S, 47°34'34.95"W - Altitude 830 m; CAJ: 21°21'35.45"S, 47°17'32.89"W - Altitude 830 m; PMA: 25°21'3.23"S, 51°28'4.41"W - Altitude: 1,000 m; SAN: 29°23'0.54"S, 54°45'41.58"W - Altitude: 135 m; POA: 30°4'9.94"S, 51°7'36.34"W - Altitude: 115 m. N page 3 of 15Zoological Studies 62:56 (2023) © 2023 Academia Sinica, Taiwan humid subtropical climate, no defined dry season. The data for temperature and relative humidity refer to the average in the period between January 2009 and December 2018 (Instituto Nacional de Meteorologia 2019). In the current study, D. mercatorum isofemale lines (referred to as populations throughout this paper) from all collection areas, except from PMA, were utilized. The analyzed Drosophila willistoni populations were from CAJ, PMA, and POA; and for D. maculifrons and D. ornatifrons, the populations were from CAJ and PMA. The collections were performed between February and April 2018 and the experiments were initiated in the same year, as soon as isofemale lines were established in laboratory conditions, around three generations. Experimental design Between 50 and 80 sexually mature female and male virgins of each population were put in embryonic chambers containing a Petri dish with agar 0.5% and enriched with Saccharomyces cerevisiae and sucrose to induce the larvae to hatch. Daily, dishes with agar were replaced and the removed dishes that contained eggs were stored. At 48h (D. willistoni and D. mercatorum), and 144h (D. maculifrons and D. ornatifrons) after mating, 2nd instar F1 larvae were transferred to vials with a standard banana diet. Ten replicates for each population of the four species were obtained, each containing 40 (D. willistoni and D. mercatorum) and 30 (D. maculifrons and D. ornatifrons) larvae. The experiments and D. willistoni and D. mercatorum isofemale line maintenance were performed at 25°C ± 1°C, and at 20°C ± 1°C for D. maculifrons and D. ornatifrons, all in a natural photoperiod. These experimental incubation temperatures followed previous knowledge about the best adaptation conditions for each species to laboratory conditions, which provides higher fitness for each species. The Drosophila groups of the four species analyzed in this study exhibit a broader amplitude of resource usage among drosophilids captured with fruit baits (Valadão et al. 2019). Therefore, we assumed that the standard banana diet utilized in the experiments should not be a stressful factor for larval development; and the differences in the examined traits would reflect the adaptation to the collection environment, rather than a response to a resource different to those found in nature. Analysis of life history traits From larvae to emerged adults, viability (VI) and development time (DT) were estimated for each of the four populations of the four species. VI was expressed as the proportion of larvae that survived until the adult stage, and DT was measured as the average period elapsed (in hours) between the transference of second instar larvae to the culture medium and the appearance of adults (males and females). The observations were carried out every four hours after the emergence of the first adults. All the emerged adults were separated by sex and stored at -20°C for posterior analysis of the metabolic pools. Dry weight (DW) was determined in 10 groups for each population and each sex of the four species. Each group contained five (for D. mercatorum, D. maculifrons, and D. ornatifrons) or 15 flies (for D. willistoni). The higher number for D. willistoni is due to its smaller size, requiring more individuals to be able to establish the weight. Each group of flies was incubated at 50°C in an oven for three days, and then weighed in a Shimadzu micro scale, model AY 220. The weight obtained for each group was then divided by the number of individuals in the group. Analysis of metabolic pools The metabolic pool analyses of the dried groups of flies were performed according to Matzkin et al. (2009). Each group of flies was homogenized in 1 mL of phosphate buffer (25 mM KHPO4, pH 7.4) and centrifuged for 2 minutes at 12,000 rpm in order to remove particles that could interfere with the colorimetric tests. A total of 800 μL of the supernatant from the homogenized mixture was collected and stored at -20°C for later analysis of the metabolic pools. The colorimetric examinations were performed for the quantification of glycogen (GL), triglyceride (TG), and total soluble protein (PR) contents. GL levels were measured using a Glucose Oxidase and Peroxidase enzyme kit (Sigma-Aldrich P7119), adding 0.1 units of Amyloglucosidase (Sigma-Aldrich) per mL of reaction buffer. The samples (40 μL of the homogenized mixture + 200 μL of the reaction buffer) were incubated at 37°C for three hours, and the absorbance was measured at 445 nm. TG content was determined using a Triglycerides kit (Gold Analisa REF. 459, MS 80022230062). The samples (40 μL of the homogenized mixture + 200 μL of reagent kit) were incubated at 37°C for 30 minutes, and the absorbance was measured at 500 nm. Only the triglyceride analysis was performed, and not the total lipids (which include cuticular lipids), due to the fact that the objective of this study was to evaluate the components involved in storing energy. The PR concentration was determined using the Bicinchoninic acid assay following the instructions of page 4 of 15Zoological Studies 62:56 (2023) © 2023 Academia Sinica, Taiwan the manufacturer (Sigma B9643), and the absorbance for quantifying the proteins was measured at 562 nm. These measurements were performed in the Spectramax 190 spectrophotometer from Molecular Devices. Each metabolic grouping was calculated by the average of triplicates and normalized by DW before the statistical analyses. Statistical analyses The VI, DT, and metabolic pool data were examined by the multifactorial analysis of variance (ANOVA) method, utilizing population and sex as factors. We also used temperature as a factor in the DT analysis because two culture temperatures were applied in different species. The comparison between the species was carried out using nested ANOVA of populations within species. Paired comparisons between fixed factors were carried out by applying Tukey posthoc analysis. All data were transformed before analysis: VI – arc sine of square root; DT and DW – square root; and metabolic pools (TG, GL, and PR) – arc sine. All statistical examinations were performed in Statistica 7 (StatSoft, Inc.) software, using α = 0.05 (Sokal and Rohlf 1995). RESULTS Viability (VI) A significant difference in viability among populations within species was found only for D. ornatifrons (Fig. 2), with PMA being significantly less viable than CAJ (F = 147.00, p ≤ 0.001). The interspecific comparison with nested populations within species showed a significant difference and similar viability for D. willistoni and D. mercatorum, with values significantly higher than the similarly viable D. maculifrons and D. ornatifrons (Table 1). Development time (DT) A significant difference in development time between males and females was found only in the PMA population of D. willistoni, in which the females presented faster DT than the males (F = 5.454, p = 0.002). Among populations within species, except for D. willistoni, the populations in lower latitudes (southeastern populations) tended to have shorter development time. There was significant variance in D. mercatorum (F = 1,641, p = 0): POA with the longest DT, followed by SAN, and SER with the shortest DT (p < 0.001 for all paired comparisons). In D. maculifrons and D. ornatifrons, PMA showed significantly higher DT than CAJ (F = 292.15, p ≤ 0.001; F = 451.53, p ≤ 0.001, respectively), which had similar DT in both species (Fig. 3). The general comparative analysis revealed different DTs among species, populations nested within species, and temperature, between sexes, and in the interaction between sex and nested populations (Table 2). The paired comparison revealed that D. willistoni has a significantly faster development time, followed Fig. 2. Viability of second instar larva to adult for populations of Drosophila willistoni (CAJ, PMA and POA), D. mercatorum (SER, CAJ, SAN and POA), D. maculifrons (CAJ and PMA), and D. ornatifrons (CAJ and PMA). Viability was calculated as the emerged adults/total larvae ratio. Different letters above the bars indicate significant differences at a p value of 0.05 determined by the ANOVA followed by Tukey post-hoc test. Error bars represent standard error of the mean. page 5 of 15Zoological Studies 62:56 (2023) © 2023 Academia Sinica, Taiwan by D. mercatorum, while both D. maculifrons and D. ornatifrons, were similarly slower (Fig. 3, Table 2). Dry weight (DW) of adults Females were larger than males in some comparisons: combining the three populations of D. willistoni (F = 16.96, p < 0.001); SER (Tukey p < 0.001) and CAJ (Tukey p < 0.05) of D. mercatorum; and CAJ of D. maculifrons (Tukey p < 0.05). The population/ sex interaction was significantly different only in D. mercatorum (F = 6.36, p < 0.001). Only D. willistoni did not show a significant difference in DW among populations (D. mercatorum - F = 13.64, p = 0; D. Table 1. Nested ANOVA (1) and post-hoc Tukey test (2) of viability for Drosophila willistoni, D. mercatorum, D. maculifrons, and D. ornatifrons 1SS MS DF F Species 1.8209 0. 6070 3 10.298*** Population (Species) 1.1503 0.1643 7 2.788** Error 5.8349 0.0589 99 2D. willistoni D. mercatorum D. maculifrons D. mercatorum 0.6733 D. maculifrons 4.653** 5.499*** D. ornatifrons 4.866** 5.723*** 0.1943 SS = sum of squares; MS = mean squares; DF = degrees of freedom. *p ≤ 0.05; ***p ≤ 0.01; ***p ≤ 0.001. Fig. 3. Development time from second instar larva to emerged adults for populations of Drosophila willistoni (CAJ, PMA and POA), D. mercatorum (SER, CAJ, SAN and POA), D. maculifrons (CAJ and PMA), and D. ornatifrons (CAJ and PMA). The development times were measured in hours. Lines above bars group statistically similar values. Asterisks above bars and different letters above lines indicate significant differences (sex and populational, respectively) at a p value of 0.05, determined by the ANOVA followed by Tukey post-hoc test. Error bars represent standard error of the mean. page 6 of 15Zoological Studies 62:56 (2023) © 2023 Academia Sinica, Taiwan maculifrons - F = 10.164, p = 0.003; D. ornatifrons - F = 30.256, p < 0.001), with southern populations being larger than the southeastern ones (Fig. 4). Dry weight showed significant differences for species, populations nested in species, sex, the interaction of populations nested in species with sex, and in all paired species comparisons. Drosophila mercatorum was the largest species, followed in order of size by D. ornatifrons, D. maculifrons, and D. willistoni (Table 3, Fig. 4). Table 2. Nested ANOVA (1) and post-hoc Tukey test (2) of development time for Drosophila willistoni, D. mercatorum, D. maculifrons, and D. ornatifrons 1SS MS DF F Species (Temperature) 4963.1 1654.4 3 2061.6*** Population (Species (Temperature)) 3433.1 490.4 7 611.2*** Sex 14.4 14.4 1 18.0*** Population (Species (Temperature)) x Sex 15.3 1.5 10 1.9* Error 2290.2 0.8 2854 2D. willistoni D. mercatorum D. maculifrons D. mercatorum 34.5*** D. maculifrons 54.14*** 31.84*** D. ornatifrons 51.63*** 29.39*** 1.719 SS = sum of squares; MS = mean squares; DF = degrees of freedom. *p ≤ 0.05; ***p ≤ 0.01; ***p ≤ 0.001. Fig. 4. Dry weights in milligrams of adult females and males from populations of Drosophila willistoni (CAJ, PMA and POA), D. mercatorum (SER, CAJ, SAN and POA), D. maculifrons (CAJ and PMA), and D. ornatifrons (CAJ and PMA). Individual dry weights were calculated from samples of 15 recently emerged flies of D. willistoni, and 5 recently emerged flies of D. mercatorum, D. maculifrons, and D. ornatifrons. Lines above bars group statistically similar values. Asterisks above bars and different letters above lines indicate significant differences (sex and populational, respectively) at a p value of 0.05, determined by the ANOVA followed by Tukey post-hoc test. Error bars represent standard error of the mean. page 7 of 15Zoological Studies 62:56 (2023) © 2023 Academia Sinica, Taiwan Metabolic pools Triglyceride (TG) Differences between sexes were detected only in the PMA population of D. maculifrons; females showed significantly higher TG contents than males (Tukey p < 0.01). Differences among populations were observed for D. mercatorum (F = 166.741, p = 0), D. maculifrons (F = 5.943, p < 0.01), and D. ornatifrons (F = 17.6531, p < 0.001), evidently because of southern versus southeastern population comparisons. For these species, populations from the South of Brazil had higher TG content than populations in the southeastern region. Furthermore, in the case of D. mercatorum, no significant differences were detected between populations of the same region, i.e., between SER and CAJ (from the southeastern region), and between SAN and POA populations (from the southern region) (Fig. 5A). There was a significant difference in TG content among species, populations nested in species, and in the interaction of populations nested in species with sex. The paired comparison revealed that D. maculifrons presented, in average, the highest TG content, followed by D. ornatifrons, and by D. willistoni and D. mercatorum, which showed similarly lower values (Fig. 5A, Table 4). Glycogen (GL) In the comparisons between sexes within populations, only D. maculifrons from PMA showed females with significantly higher concentrations of GL than males (Fig. 5B). No other comparisons resulted in a significant difference in this metabolite content between sexes and in the population/sex interaction. There was a significant difference in GL content among populations in D. willistoni (F = 35.636, p = 0), D. mercatorum (F = 27.6538, p = 0), and D. maculifrons (F = 18.9932, p < 0.001). Drosophila willistoni and D. mercatorum presented the same populational pattern, with southern region populations (PMA and POA of D. willistoni, SAN and POA of D. mercatorum) presenting higher GL contents than southeastern region populations, with no difference between populations of the same region. On the other hand, Drosophila maculifrons displayed the opposite pattern compared to the first two species: The southeastern population (CAJ) showed higher GL content than the southern population (PMA). Though D. ornatifrons showed a similar pattern (Fig. 5B), the difference between its populations was not significant. There were significant differences in GL content among species (Table 4); the paired comparison revealed that D. mercatorum has a significantly lower concentration of this metabolite in relation to the other species, mostly because of the southeastern populations, which were statistically different to all others (Fig. 5B). There was also significance in the comparison of populations nested in the species, between sexes, and in the interactions of populations nested in species with sex (Table 4). Total proteins (PR) The PR content was similar between sexes within populations of all species. Regarding populations within species, the same pattern observed for GL was partially detected for PR, i.e., higher concentration in southern when compared to southeastern populations of D. willistoni and D. mercatorum, and the opposite for the other two species (Fig. 5C). For D. willistoni, Table 3. Nested ANOVA (1) and post-hoc Tukey test (2) of dry weight for Drosophila willistoni, D. mercatorum, D. maculifrons, and D. ornatifrons 1SS MS DF F Species 3.50758 1.16919 3 646.15*** Population (Species) 0.18371 0.02624 7 14.50*** Sex 0.15639 0.15639 1 86.43*** Population (Species) x Sex 0.06070 0.00607 10 3.35*** Error 0.35828 0.00181 198 2D. willistoni D. mercatorum D. maculifrons D. mercatorum 43.68*** D. maculifrons 12.67*** 25.17*** D. ornatifrons 19.53*** 17.94*** 6.267*** SS = sum of squares; MS = mean squares; DF = degrees of freedom. *p ≤ 0.05; ***p ≤ 0.01; ***p ≤ 0.001. page 8 of 15Zoological Studies 62:56 (2023) © 2023 Academia Sinica, Taiwan Fig. 5. Triglycerides (A), Glycogen (B), and Total Protein (C) concentrations per mg of dry weight of adult females and males from populations of Drosophila willistoni (CAJ, PMA, and POA), D. mercatorum (SER, CAJ, SAN, and POA), D. maculifrons (CAJ and PMA), and D. ornatifrons (CAJ and PMA). Values are means and SE for homogenates of 15 flies for D. willistoni and 5 flies for D. mercatorum, D. maculifrons, and D. ornatifrons. Lines above bars group statistically similar values. Asterisks above bars and different letters above lines indicate significant differences (sex and populational, respectively) at a p value of 0.05, determined by the ANOVA followed by Tukey post-hoc test. Error bars represent standard error of the mean. page 9 of 15Zoological Studies 62:56 (2023)