Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae) is the most widely used parasitoid in biological control of Tephritidae programs around the world. Nevertheless, we have little information about the use of these parasitoids against Anastrepha fraterculus (Wiedemann) (Diptera: Tephritidae). This study was conducted to evaluate the thermal requirements for the development of D. longicaudata in 2 of its hosts, A. fraterculus and Ceratitis capitata (Wiedemann) (Diptera: Tephritidae). Third instars of each fly species were exposed for 1 h to 50 couples of the parasitoid. Subsequently, the larvae were placed individually in glass jars maintained in chambers at temperatures of 15, 18, 21, 25, 28, and 31 °C, at 60 ± 10% RH and a photoperiod of 14:10 h L:D until the emergence of D. longicaudata. The rate of development of D. longicaudata, in both hosts, increased as temperature increased. The development period of egg-to-adult phases was used to estimate, by the hyperbola method, the lower development threshold (Dt) and the thermal constant (K) of D. longicaudata. Dt and K were, respectively, 7.83 °C and 322.58 degree days for individuals that developed in C. capitata, and 12.5 °C and 227.27 degree days for those that developed in A. fraterculus. Adult longevity was inversely proportional to temperature. The results indicated that D. longicaudata may not develop in Rio Grande do Sul in the winter, because it is very common that minimum temperatures in the months of Jun and Jul fall below the Dt. In the 4 fruit-producing regions of the state of Rio Grande do Sul, the average numbers of generations of D. longicaudata per year were estimated in A. fraterculus and C. capitata, respectively, as follows: Porto Alegre (11.16 and 13.12), Pelotas (7.97 and 10.89), Bento Gonçalves (6.99 and 10.05), and Vacaria (4.84 and 8.28).
Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae), which originated in the Indo-Australian region, is a koinobiont endoparasitoid of various tephritid species. It has been introduced into various tropical and subtropical regions for control of Anastrepha species (Diptera: Tephritidae) and Ceratitis capitata (Wiedemann) (Diptera: Tephritidae) (Montoya et al. 2000). This parasitoid is considered the most important biological control agent for inundative releases against diverse species of fruit flies in Latin American countries (López et al. 2009).
In 1994 Embrapa (the Brazilian Corporation of Agricultural Research) introduced D. longicaudata into Brazil from the Florida Department of Agriculture and Consumer Services, Division of Plant Industry, Gainesville, Florida, USA, and released it in northeastern Brazil, where it is now established in the Recôncavo Baiano region and in the Submé dio São Franscisco region (Carvalho & Nascimento 2002). Its establishment has been demonstrated in the states of Minas Gerais (Alvarenga et al. 2005) and Rio de Janeiro (Leal et al. 2008). However, there are no records in regards to its release in Rio Grande do Sul.
Fruit flies (Diptera: Tephritidae) are first-order pests in fruit crops in Brazil (Nava & Botton 2010). In Rio Grande do Sul, C. capitata and Anastrepha fraterculus (Wiedemann) are important pests of exotic fruit crops such as apple and peach (Kovaleski & Ribeiro 2003). Anastrepha fraterculus also damages native fruit crops including several Myrtaceae (Gattelli et al. 2008).
Temperature is an important climatic variable that can directly influence the phenology and distribution of insects (Damos & Savopoulou- Soultani 2012). The optimal temperature for the development of a pest species may not be the same as that of its natural enemy (Horn 1998). Determination of the most appropriate temperature for the development of a parasitoid is important for its laboratory production and for planning field releases, and is a fundamental requirement of any program of biological control of pests. Meirelles et al. (2013) determined the development of D. longicaudata on C. capitata and A. fraterculus at a constant temperature (25 ± 1 °C). This study sought to determine the thermal requirements of D. longicaudata reared in these 2 host species, to estimate the potential number of generations of D. longicaudata per year in 4 fruit-producing regions of the state of Rio Grande do Sul, and to reveal information about the development and possible establishment of this parasitoid in these regions.
Materials and Methods
REARING OF INSECTS
All insects were reared at 25 ± 2 °C, 65 ± 10% RH, and a photoperiod of 14:10 h L:D.
Fruit Flies. Adults of A. fraterculus and C. capitata were kept in separate cages (45 × 30 × 30 cm). Food—adapted from Salles (1995) including wheat germ, beer yeast, the hydrolyzed protein Biorigin®, crystal sugar (at a ratio of 1:1:1:3), and water—was supplied ad libitum. Adults of A. fraterculus received oviposition substrates for 24 h (Salles 1992) composed of agar (5 g), blackberry juice (80 mL), and water (160 mL). The eggs were removed from the substrate and kept in distilled water for 48 h in a climate chamber (25 ± 2 °C, 65 ± 10% RH; in the dark), and then transferred to a larval diet. A 250 mL orange plastic tube was used as a substrate for C. capitata oviposition, following the method of FAO/IAEA/USDA (2003). The eggs were removed daily from the substrate and held 24 h before being placed in a climate chamber on the larval diet. The larval diet for both fly species was composed of raw carrot (500 g), cooked carrot (500 g), crystal sugar (500 g), beer yeast (100 g), corn flour (600 g), nipagin (4.4 g), sodium benzoate (4.4 g), citric acid (14.4 g), and distilled water (400 mL) (Terán 1977). The diet was placed on trays, where the eggs were distributed. The trays with larval diet and eggs were wrapped in newspaper and held for 7 d. Then the trays were unwrapped, placed on sterilized sand, and held for an additional 10 d. During this time the larvae exited the diet and pupated in the sand. The sand was then sifted, and the pupae were placed in plastic boxes (15 ´ 15 ´ 5 cm) with sterilized sand until emergence.
Parasitoid. The first specimens of D. longicaudata were obtained from the laboratory of the Embrapa Mandioca e Fruticultura research institute (located in Cruz das Almas, Bahia, Brazil) in Oct 2008. The adults, approximately 50 couples, were kept in cages (20 × 15 × 15 cm) covered with voile. Water and food (a mixture of water, honey, sugar, ascorbic acid, and nipagin) were supplied ad libitum. Third instars of C. capitata were exposed for 1 h (40 to 50 larvae per parasitism unit) to adults of D. longicaudata. Parasitized larvae were placed in plastic boxes (15 × 15 × 5 cm) with sterilized sand until emergence.
EVALUATION OF THE THERMAL REQUIREMENTS AND LONGEVITY OF D. LONGICAUDATA
Six sets of 50 pairs of D. longicaudata, between 3 and 10 d old, were kept in separate cages (20 × 15 × 15 cm), receiving food and water ad libitum. During several days, each set of D. longicaudata was exposed for 1 h to 40 to 50 third instars of A. fraterculus (~9 d old). About 18 sets of larvae were used. After exposure, the larvae were placed individually in glass tubes (25 mL) with sterilized sand and damp filter paper, and the tubes were sealed with plastic wrap and distributed evenly in chambers maintained at 15, 18, 21, 25, 28, and 31 °C, at 60 ± 10% RH and a photoperiod of 14:10 h L:D. The insects that emerged at each temperature regime (between 109 and 200 individuals) were placed individually in plastic jars (140 mL) with food and water, and maintained at the same temperature regime until death. Adult emergence and mortality were recorded each morning. The same procedure was repeated exposing parasitoids to larvae of C. capitata.
The thermal requirements of the development periods (defined as the period of time from egg to adult emergence) of D. longicaudata at the 6 temperatures were estimated through the hyperbola method (Haddad et al. 1999), which calculates the lower development threshold (Dt), the thermal constant (K), and the optimum temperature range for development. The average development period and the longevity of the parasitoids reared in the same host at different temperatures were analyzed through the Kruskal—Wallis test (P ≤⃒ 0.05) and Dunn's paired comparisons. The development period of parasitoids reared in the 2 different hosts at the same temperature regimes were compared by the Mann—Whitney test (P ≤⃒ 0.05) using the program BioEstat® 5.0 (Ayres et al. 2007). The possible number of generations of D. longicaudata per year was estimated for the counties of Bento Gonçalves, Pelotas, Porto Alegre, and Vacaria, Rio Grande do Sul, based on the parasitoid's thermal requirements and the mean temperatures of these locations (Applied Meteorology Center of FEPAGRO and the 8th Meteorology District, Porto Alegre), according to Cividanes (2000).
No D. longicaudata completed development at either 15 or 31 °C in A. fraterculus, or at 15 °C in C. capitata (Table 1). The viability (completion of the egg-to-adult cycle) of D. longicaudata was greatest at 25 °C when the parasitoid was reared in either host species, and this value was 61.3% in A. fraterculus and 58.3% in C. capitata. The mean development period of D. longicaudata reared in A. fraterculus was longest at 18 °C and shortest at 28 °C, and this period was inversely related to temperature and differed among all temperature regimes (H = 424.54; df = 3; P < 0.0001) (Table 1). When the host was C. capitata, the development period varied inversely with temperature among the temperatures of 18, 21, and 25 °C (H = 603.04; df = 4; P < 0.0001) but was not significantly different between 28 and 31 °C.
The development period of D. longicaudata was longer in A. fraterculus than in C. capitata at all temperatures. Females of D. longicaudata took more time to reach the adult phase than males at all temperatures, both in C. capitata (H = 634.96; df = 9; P < 0.0001) and in A. fraterculus (H = 439.62; df = 7; P < 0.0001).
The lower development threshold (Dt) of males and females of D. longicaudata in A. fraterculus (12.5 °C) was higher than in C. capitata (7.83 °C) (Table 2). Approximately 99% and 93% of the reduction of the development time at increasing temperatures of D. longicaudata reared in A. fraterculus and C. capitata, respectively, was explained by the increase in temperature (Table 2). The optimum temperature range for development of D. longicaudata was 16.37 to 22.48 °C in A. fraterculus and 11.56 to 21.86 °C in C. capitata.
Average (± SE) development periods and their ranges (d) of Diachasmimorpha longicaudata reared in Anastrepha fraterculus [Dl (Af)] and in Ceratitis capitata [Dl (Cc)] larvae at 15, 21, 25, 28, and 31 °C, at 65 ± 10% RH and a photoperiod of 14:10 h L:D.
An inverse relationship between the mean longevity of D. longicaudata and temperature was found. When D. longicaudata was reared in A. fraterculus, the longest development period was recorded at 18 °C (52.7 d) and the shortest at 28 °C (9.2 d). The longevities at 21 °C (29.5 d) and 25 °C (27.0 d) were statistically similar (Table 3). With C. capitata as its host, the parasitoid had statistically similar longevities at 18 °C (53.3 d) and 21 °C (39.4 d), intermediate longevity at 25 °C (29.5 d), and the shortest longevities at 28 °C (11.3 d) and 31 °C (11.5 d). Individuals developing in C. capitata lived longer at 21 °C (U = 4742.0; P < 0.0001) than those developing in A. fraterculus, i.e., 39.4 d vs. 29.5 d. There was no statistically significant difference between the longevities of D. longicaudata reared in the 2 hosts at 18, 25, or 28 °C.
The estimated number of generations of D. longicaudata per year for the 4 counties (Table 4) when the host was A. fraterculus varied from 4.84 to 11.16, and the corresponding values for C. capitata as the host varied from 8.28 to 13.12. Thus D. longicaudata has the potential for more generations per year in C. capitata than in A. fraterculus.
The failure of D. longicaudata to produce adults at 15 °C in both host species may be attributed to the existence of a dormancy phase stimulated by low temperature. Diapause in larvae of D. longicaudata collected in regions of Hawaii, USA, with winter temperatures varying between 2 and 10 °C, was recorded by Clausen et al. (1965). However, the method utilized in our study did not permit us to observe this phenomenon. The failure of the parasitoid to develop at 31 °C in A. fraterculus deserves further studies using individuals more experienced on that host. In C. capitata, at 28 and 31 °C, emergence of D. longicaudata was similar to that recorded in Anastrepha suspensa (Loew) (Diptera: Tephritidae) by Ashley et al. (1976). The results obtained in our study may be associated with the fact that the parasitoid had been reared for many generations in C. capitata. The rearing of this parasitoid in Brazil has been conducted in C. capitata ever since its introduction in 1994. If D. longicaudata had been reared for many generations in A. fraterculus, it seems possible that its Dt value would be different from that reported here.
Lower development threshold (Dt), thermal constant (K) in degree days (DD), linear equation of development speed (1/D), and determination coefficient (R2) of the biological cycle (egg to adult) of Diachasmimorpha longicaudata reared either in Anastrepha fraterculus or in Ceratitis capitata at 15, 18, 21, 25, 28, and 31 °C, at 65 ± 10% RH and a photoperiod of 14:10 h L:D.
Vacaria has an annual average temperature (15.2 °C) that is unfavorable to the parasitoid, regardless of the host, when compared with other regions; thus, fewer generations would be expected per year in this region. Among the 4 counties, Porto Alegre had the largest estimated number of generations per year. We expect that in warmer regions, or in periods with mean temperatures above 25 °C, D. longicaudata would develop more rapidly, as the time necessary to double its population at 25 °C is 3.92 and 4.73 d in A. fraterculus and C. capitata, respectively (Meirelles et al. 2013). Perhaps the success of D. longicaudata in regions of many countries where it has been introduced as a biological control agent can be explained in part by higher mean temperatures in these regions compared with the fruit-growing counties of Rio Grande do Sul.
During the winter in Rio Grande do Sul, during the months of Jun through Aug, temperatures may restrict or impair the establishment of D. longicaudata. Studies of population fluctuations of A. fraterculus in states of southern Brazil point to a reduction in populations during the months of Jun and Jul (Kovaleski et al. 2000), when the mean minimum temperatures are frequently below the development threshold for this parasitoid. The primary hosts of the South American fruit fly in Rio Grande do Sul are species of fruit trees in the Myrtaceae and Rosaceae (Zucchi 2000), which do not have mature fruits in Jun and Jul (Simão 1971; Donadio 2000; Manica 2002; Lira Júnior et al. 2007). Between the maturation of fruit of feijoa (pineapple guava) (Acca sellowiana (O. Berg) Burret; Myrtales: Myrtaceae) in May, and the maturation of fruit of Brazilian cherry (Eugenia species; Myrtales: Myrtaceae) and loquat (Eriobotrya japonica [Thunb.] Lindl.; Rosales: Rosaceae) in Aug, there could be a significant scarcity of hosts for fruit flies and consequently for D. longicaudata. Individuals of D. longicaudata originating from C. capitata had a lower thermal threshold than those reared in A. fraterculus. The development rate of this parasitoid, reared in both hosts, increases with an increase in temperature. It is likely that in some regions of southern Brazil, D. longicaudata would have difficulty surviving through the winter, making necessary the releases of this parasitoid early in spring.
Average (± SE) longevities and their ranges (d) of Diachasmimorpha longicaudata reared either in Anastrepha fraterculus [Dl (Af)] or in Ceratitis capitata [Dl (Cc)] at 18, 21, 25, 28, and 31 °C, at 65 ± 10% RH and a photoperiod of 14:10 h L:D.
Estimated number of generations of Diachasmimorpha longicaudata per year in 4 fruit-producing counties in Rio Grande Do Sul, Brazil, based on annual average temperatures (°C) and annual accumulated degree days (DD), for D. longicaudata developing in Anastrepha fraterculus [Dl (Af)] (Dt of 12.5 °C for the eggto- adult period) or Ceratitis capitata [Dl (Cc)] (Dt of 7.83 °C for the egg-to-adult period).
We thank the National Council for Scientific and Technological Development (CNPq) for fellowships awarded to the first and second authors (proc. n. 305956/2010-8) and for financial support (proc. n. 475287/2010-0). We would like to thank Antonio Nascimento at Embrapa Mandioca e Fruticultura Tropical for providing the first D. longicaudata individuals that we used to establish our laboratory colony.