Survey of Fruit Flies ( Diptera : Tephritidae ) from 23 Species of Fruits Collected in Sleman , Yogyakarta

Fruit flies (Diptera: Tephritidae) are a major pest of fruits and vegetables in many countries, including Indonesia. The females lay their eggs in the fruit, and the larvae that hatch feed on the flesh, resulting in decay, discoloration, and a significant decrease in the economic value of harvested fruit. In addition, fruit flies are a major obstacle to international trade of fruits. Quarantine restrictions have been implemented by countries around the world to prevent the introduction of exotic fruit fly species. Globally, there are many fruit fly species of economic importance, and these fall into 6 different genera, namely Anastrepha, Ceratitis, Rhagoletis, Dacus, Zeugodacus, and Bactrocera (Van Houdt et al., 2010; Virgilio et al., 2015). Pestiferous Anastrepha is native to tropical and subtropical America (CABI, 2019a; 2019b; 2019c; 2019d), Rhagoletis is native to North America (Bush, 1966), Ceratitis capitata is native to sub-Saharan Africa (CABI, 2019e), while Bactrocera, Zeugodacus, and Dacus are native to Asia, Oceania and Afrotropical region (White, 2000). The Oriental fruit fly, Bactrocera dorsalis is known as the world’s worst horticultural pest, highly invasive, and widely distributed in tropical Asia. It was introduced to Africa, Oceania, and parts of America and was recently reported in Europe (Vargas et al., 2015; Nugnes et al., 2018; CABI, 2019f). In addition to the Oriental fruit fly, there are several other species of fruit flies in the genus Bactrocera that are of economic importance in Indonesia. Accurate knowledge of the larval host range and distribution of the various fruit fly species is essential for pest management programs and quarantine authorities. Indonesia is an archipelago located in the tropical area of South-East Asia where fruits and vegetables are available throughout the year. Various ecosystems or habitats can be found on each island in Indonesia, ABSTRACT


INTRODUCTION
Fruit flies (Diptera: Tephritidae) are a major pest of fruits and vegetables in many countries, including Indonesia. The females lay their eggs in the fruit, and the larvae that hatch feed on the flesh, resulting in decay, discoloration, and a significant decrease in the economic value of harvested fruit. In addition, fruit flies are a major obstacle to international trade of fruits. Quarantine restrictions have been implemented by countries around the world to prevent the introduction of exotic fruit fly species. Globally, there are many fruit fly species of economic importance, and these fall into 6 different genera, namely Anastrepha, Ceratitis, Rhagoletis, Dacus, Zeugodacus, and Bactrocera (Van Houdt et al., 2010;Virgilio et al., 2015). Pestiferous Anastrepha is native to tropical and subtropical America (CABI, 2019a;2019b;2019c;2019d), Rhagoletis is native to North America (Bush, 1966), Ceratitis capitata is native to sub-Saharan Africa (CABI, 2019e), while Bactrocera, Zeugodacus, and Dacus are native to Asia, Oceania and Afrotropical region (White, 2000). The Oriental fruit fly, Bactrocera dorsalis is known as the world's worst horticultural pest, highly invasive, and widely distributed in tropical Asia. It was introduced to Africa, Oceania, and parts of America and was recently reported in Europe (Vargas et al., 2015;Nugnes et al., 2018;CABI, 2019f).
In addition to the Oriental fruit fly, there are several other species of fruit flies in the genus Bactrocera that are of economic importance in Indonesia. Accurate knowledge of the larval host range and distribution of the various fruit fly species is essential for pest management programs and quarantine authorities.
Indonesia is an archipelago located in the tropical area of South-East Asia where fruits and vegetables are available throughout the year. Various ecosystems or habitats can be found on each island in Indonesia, ranging from densely to sparsely populated areas, highland to lowland forests, and monoculture to polyculture. Most of the islands in Indonesia provide suitable habitats for fruit flies, especially B. dorsalis, due to the favourable climate and the availability of various types of tropical fruits that are good larval hosts for the species. The various types of larval host plants identified through surveys in some parts of Indonesia have been reported by Allwood et al. (1999) and Suputa et al. (2010).
These previous surveys employed two common methods: trapping, collection, and rearing from host fruits. The trapping method involved the use of plastic traps baited with a strong chemical lure that attracts male flies and a toxicant to kill them. The rearing method involved collecting fleshy fruits that have the potential to be infested by fruit flies and holding them in rearing cages in the laboratory until adult flies emerged (Allwood et al., 1999;Suputa et al., 2007). Rearing of flies from host is a very useful and sensitive tool in any survey of fruit flies, particularly for fly species such as B. latifrons that do not respond to the commonly used male attractants. Host rearing also provides information on the overlap in the host range of different species of fruit flies (Harris et al., 2003).
Host records for fruit flies from a number of regions in Indonesia have been documented by Suputa et al. (2010). Our study aims to extend and update the previous data by surveying the District of Sleman, Special Province of Yogyakarta, a region that has different types of agro-ecosystems. Sleman is a district consisting of dryland and wetland zones, urban and suburban areas, and forest and grasslands on the slopes of Mt. Merapi. The agricultural and forest areas in the dry and wetland zones occupy almost 70% of this district. The crops grown in Sleman include rice, various vegetables, sugarcane, tobacco, coconut, salak (snake fruit), mango, and banana (BPS, 2018;2019). Sleman is also known as one of the centers for salak or snake fruit (Salacca zalacca) production (Ministry of Agriculture, 2019). In 2018, fruit flies had been reported to infest salak plantations in Indonesia. Salak fruit exports to China now require quarantine treatments to ensure the absence of fruit flies (IAQA, 2014;Astuti et al., 2019;DHCP, 2020). Recently, the European and Mediterranean Plant Protection Organization (EPPO) issued a non-compliance notification that listed Bactrocera found in salak exported from Indonesia to the Netherlands (EPPO, 2019). The host status and level of infestation in salak by Bactrocera fruit flies in Indonesia remains unclear and needs further studies to provide a better foundation for managing this pest. Considering the diversity of the ecosystem and the economic importance of salak, we selected the District of Sleman for extending and updating the information of fruit fly host. The findings of our study would be an essential element in fruit fly management deploying an area-wide approach.

Fruit Fly Collection from Host
The study was conducted over a six month period from August 2019 to February 2020. Fruits were collected from four different sub-districts in the District of Sleman, Yogyakarta ( Figure 1) with 15 species of fruits from the sub-district of Berbah, 5 from Depok, 3 from Turi, and 2 from Gamping (Table 1). Berbah sub-district consists of rural and semi-rural areas. In this sub-district, the fruit sampling was concentrated in the Agrotechnology Innovation Center of Universitas Gadjah Mada. This Center is a 35-hectare block of land with various trees, horticultural crops, rice fields, small farms, and an arboretum, which has more than 50 species of trees. Depok sub-district is a densely populated urban and suburban area with a mixture of trees growing in home gardens and parks. Turi sub-district is the center of salak production in Yogyakarta, mostly of the cultivar Pondoh. Thus, the area is dominated by salak plantations and clustered settlements. Gamping sub-district also consists of rural and semi-rural areas where the central fruit market in Yogyakarta Province is located. Ripe and senescing fruits from attached or detached fruits were collected from the various species of plants in each location. Ripe and senescing fruits were selected because they have a higher probability of being infested by fruit flies compared to immature or green fruits. However, for papaya (Carica papaya), it was not possible to obtain fully ripe fruits from the trees, thus we sampled that was between 40-50% yellow. Not all fruit types were available in equal numbers because of seasonality, thus the sample size of collected fruits varied widely between 1 to 62 with a total of 305 fruit samples collected in total.
Altogether 23 species of fruits belonging to 14 different plant families were sampled in this study. The collected fruits were then transferred to the laboratory for fruit fly rearing. The rearing method used followed Suputa et al. (2007)  depending on the size of the fruit. Large fruits were placed in 5 L plastic containers covered by gauze at the top of the container, with sterilized sawdust that served as a pupating medium. Smaller fruits were placed on top of gauze covered 90 mm plastic petri dish, and this was then placed above sterilized sawdust in a 750 mL plastic container. A petri dish was placed under the smaller fruits to contain liquids that oozed from the decaying fruit. Rectangular ventilation holes, (7 cm×3 cm) covered by gauze were made on each side of the plastic container. The number of pupae recovered from each species of fruit was counted and recorded. All pupae recovered from the containers were transferred into separate containers (a maximum of 5 pupae/ container) for adult rearing. This separate container was a 500 mL clear plastic bowl with 2 gauze-covered rectangular holes (5 cm×3 cm) on opposite sides of the wall for ventilation. When adult flies emerged, they were fed sugar, yeast, and water for seven days at room temperature to mature and fully develop their body colours and be suitable for accurate identification. The number of flies and parasitoids that emerged were counted and recorded. Fruit flies were kept in the container and transferred to a freezer for a minimum of one hour to kill them before identification under a dissecting microscope.

Host Plant Identification
All host plants collected in the study were commonly known species.

Collected Fruit Flies from Host Rearing
Fruit fly pupae were recovered from all of the fruit species except papaya with adult emergence rates that varied between 0 to 96.3% (Table 2).
Parasitoids were recorded from pupae obtained from mango (Mangifera indica), salak, Indian almond (Terminalia catappa), melinjo (Gnetum gnemon), starfruit (Averrhoa carambola), and God's crown (Phaleria macrocarpa). Pupae collected from mango and Indian almond from an urban area in Depok had parasitism rates of 3.5 and 1.4%, respectively, while salak from the salak plantation in Turi had a higher level of parasitism (24.8%). The pupae collected from melinjo, starfruit, and God's crown in the rural area in Berbah showed 28.6, 4.9, and 12.5% parasitism, respectively. Pupal mortality was high, at more than 50% from pupae reared from ridge gourd (Luffa acutangula), breadfruit (Artocarpus altilis), Java plum (Syzygium cumini), and God's crown. No pupae were recovered from papaya which could be due to the immature state of the collected fruits. Only immature papaya was available for collection. It has been established that green unripe papaya is not attractive for fruit fly infestation (Seo et al., 1983) and ovipositing flies have been reported to be attracted to ripe papaya only (Jang & Light, 1991;Cugala et al., 2017).
The variation in pupal mortality and adult emergence might be due to the quality and variation in nutrients, and the growth environment within the host fruit. Nutrients in the host fruit affected the performance of immature flies including growth, development, survival, and adult fecundity as well as longevity (Bateman, 1972;Tsitsipis, 1989). The nutrient content within the same or different host(s) varies, which could affect insect development including pupal survival and adult emergence (Christenson & Foote, 1960;Tsitsipis, 1989). Newell & Haramoto (1986), also suggests that the natural mortality of B. dorsalis pupae from field-collected fruits might be caused by unfavourable conditions during larval development that in turn lead to decreased larval fitness which affects pupal survival. The pupae in different species of fruit flies also required different moisture conditions for survival, with some fly species preferring a humid environment while other species were not easily affected by relative humidity (Bateman, 1972).
Parasitism was observed in pupae recovered from mango, salak, Indian almond, melinjo, starfruit, and God's crown. These fruits are widely cultivated and some are native to Indonesia. According to Clarke (2019), infested native fruits are more likely 125 Jurnal Perlindungan Tanaman Indonesia Vol. 24 No. 2 to have a level of higher parasitism than exotic fruits because the searching mechanism of the parasitoids is more adapted to native host fruit. Regarding B. carambolae and B. dorsalis found infesting salak, both species of flies and salak are native to Indonesia. We are uncertain as to whether the infestation in salak has simply not been noted in the past or whether it is the result of a relatively recent expansion of the host utilization by these two fruit fly species. Based on the information from the leaders of farmer groups, many salak farmers do not use insecticide and some do so minimally. Thus, a relatively high rate of parasitism in salak could be attributed to the very low rate of insecticide use in salak cultivation.

Host Utilization by Various Species of Fruit Flies
Six species of fruit flies belong to the genus Bactrocera and Zeugodacus i.e. B. dorsalis, B. carambolae, B. umbrosa, B. albistrigata, B. mcgregori, and Z. cucurbitae were identified in this study (Table 3) (Allwood et al., 1999). However, Java plum and God's crown have not been reported to be infested by B. carambolae in other countries. This study also provides further data to support the reports by DHCP (2020) and Astuti (2019) that B. dorsalis and B. carambolae were found to infest salak. Bactrocera dorsalis infested the widest range of hosts, followed by B. carambolae, Z. cucurbitae, B. albistrigata, B. umbrosa, and B. mcgregori. In contrast, B. mcgregori was the most host-specific fruit fly species as it was only associated with melinjo. This record concurs with previous publications by White and Elson-Harris (1992), Allwood et al. (1999), Ranganath & Veenakumari (1999), Suputa et al. (2010), Drew & Romig (2013), and Larasati et al. (2013). However, the infestation of B. mcgregori in melinjo did not appear to be a serious threat and this species was still categorized as a non-pest (Doorenweerd et al., 2018). Melinjo, as fresh fruit was not exported from Indonesia and therefore not a quarantine issue (Cadiz & Florido, 2001

Artocarpus altilis
Syzygium aqueum Watery Rose Apple Cucurbitaceae, respectively. Bactrocera albistrigata was found in three different fruit species belong to two different plant families. Bactrocera dorsalis and B. carambolae infested the widest range of fruits. Indian almond and guava (Psidium guajava) were found to be hosts to three different species of Bactrocera i.e. B. dorsalis, B. carambolae, and B. albistrigata, whilst 8 fruits were the hosts of two species of flies, and 12 fruits were the host of one fly species. In this study, we were not able to justify whether there was more than one species per fruit because more than one fruit was placed in one container. The high frequency of overlapping hosts is evidence that the overlapping of host utilization by fruit flies is common particularly for B. dorsalis and B. carambolae as reported by Harris et al. (2003) and Danjuma et al. (2013). Bactrocera dorsalis and B. carambolae are known as sympatric sibling species, native to South-East Asia (Wee & Tan, 2005;Vargas et al., 2015), and frequently found in the same area (Clarke et al., 2001;Wee & Tan, 2005;Suputa et al., 2010;Larasati et al., 2013;Linda et al., 2018). Most of B. carambolae hosts are also the hosts of B. dorsalis (Allwood et al., 1999;CABI, 2019f, CABI, 2020. The infestation of B. carambolae on P. macrocarpa, M. emarginata, and S. cumini is a new report. Previous works reported associated flies with P. macrocarpa were B. papayae (now B. dorsalis), B. bullata, and B. trilobata (Suputa et al., 2010;Drew & Hancock, 2016). Outside Indonesia, several species of flies were reported to be associated with M. emarginata such as B. caryeae, B. correcta, B. dorsalis, B. tryoni, B. jarvisi, B. neohumeralis, and B. zonata (Allwood et al.,1999;CABI, 2019h) while associated flies with S. cumini were B. correcta, B. tryoni, and B. dorsalis (Allwood et al.,1999;CABI, 2019). This new report in Indonesia extends the host range of this species.
Salak is not commonly thought to be a host of B. dorsalis and B. carambolae in Indonesia although the infestation of fruit flies in salak was reported previously by IAQA (2013), EPPO (2019), and DHCP (2020). Salak was not originally thought to be a host of fruit flies since there have been no reports of fruit fly infestation in salak before 2013. We collected only detached salak, and it remains unclear how fruit fly oviposits through the hard outer skin and utilizes salak as a host. It is possible that the flies oviposit only in ripe or overripe fruits that tend to detach easily from the plant when they are ripe (Haryoto & Priyanto, 2018). We collected such detached fruits for our study and some being cracked open and increasing the possibility for the females to lay eggs. Thus, further research is required to determine the real host status of this fruit. On the other hand, Barbados cherry and Surinam cherry are non-native fruits that are mainly planted as a hedge or as ornamentals (Hanelt et al., 20011), and are not extensively cultivated and rarely sold in the markets in Indonesia. Barbados cherry, however, is a crop of major economic importance in the Mekong delta in South Vietnam. It was grown, processed, and exported to Japan (Vijaysegaran, 2016). Thus, Barbados cherry has the potential to develop into a crop of economic importance in the future in Indonesia. The infestation of fruit flies in these exotic or ornamental fruiting trees showed the possibility of utilization in non-commercial hosts that might affect the orchards.
Bactrocera dorsalis and B. carambolae are known as highly polyphagous fruit flies, B. dorsalis has been reported to utilize more than 300 species of plants (CABI, 2019f) and B. carambolae up to 75 species of plants (Allwood et al., 1999;CABI, 2020). Many tephritid fruit flies are polyphagous which is uncommon for herbivorous insects although the polyphagous trait of some fruit fly species still indicates the presence of the host preference between available fruits (Clarke et al., 2001;Clarke, 2017). Bactrocera is also classified as an opportunistic and broad-range exploiter of pulpy fruit (Aluja & Mangan, 2008). Bactrocera is native to Asia while the fruits of Asia are largely non-toxic, therefore the polyphagous Bactrocera is easier to switch and expand the host range as no or low fitness cost for this behaviour (Clarke, 2017).
Determining the host status of fruit fly is not a simple mechanism since the level of host utilization pattern also needs to be considered (Aluja et al., 1987). Ours was a preliminary study that recorded the host utilization by different species of fruit flies in Sleman, but with no information on infestation levels or regularity of host utilization. Further research is thus required to provide such data. Understanding the host use of Bactrocera is important since fruit conditions such as maturity levels or skin damage may influence the host utilization by Bactrocera (Clarke, 2019). Fruit could be classified as a non-host or a conditional or potential host due to skin thickness of different varieties or when the fruit has a disease, physiological or mechanical damage (Clarke, 2019).
The international trading of fresh fruits and vegetables among countries is often severely restricted when these commodities were infested by fruit flies. Quarantine treatments have to be negotiated and applied for trade to commence. Tropical fruits such as banana (Musa domestica), guava, mango, melon (Cucumis melo), papaya, and Citrus spp. are hosts of fruit flies (Allwood et al., 1999;Suputa et al., 2010;Larasati et al., 2013;Leblanc et al., 2013) and are widely cultivated in Indonesia (BPS & Directorate General of Horticulture, 2020). These commodities are equally promoted in local and international trading (FAO, 2020;BPS, 2020). Therefore, Indonesia is obliged to follow the quarantine protocols to avoid the risk of introducing exotic fruit fly species into importing countries. The management of non-commercial/alternate hosts in the production area by removing or replacing non-economic host plants with non-host plants is recommended as a component of a systems approach for pest risk management of fruit flies (FAO/IPPC, 2012). Furthermore, the diversity of hosts for the economically important species of fruit flies supports the idea that the management of these insects should be based on the ecosystem approach e.g. area-wide pest management, rather than relying on managing a specific commodity in a particular area. The infestation levels of B. dorsalis and B. carambolae in host plants that have no economic importance should be monitored because of the possibility that these plants could serve as a breeding ground for pest fruit flies.

CONCLUSION
Six species of Bactrocera were associated with 23 different species of host plants collected from Sleman, Yogyakarta. The six fruit fly species are B. dorsalis, B. carambolae, B. albistrigata, B. umbrosa, and B. mcgregori. Two species of fruit flies, B. dorsalis and B. carambolae, utilized most of the collected fruits. Java plum and God's crown were first reported to be infested by B. carambolae in the world while Barbados cherry was the first in Indonesia. Further studies to determine the susceptibility of this host should be determined. Rearing of B. dorsalis and B. carambolae from collected salak provided strong evidence that these two fruit fly species are extending their utilization of salak as a host and are likely to become a major pest problem in salak cultivation. A follow-up study involving periodic sampling of the various fruit types and detailed observations on host infestation by the different fly species is suggested to obtain additional information on fruit fly infestation levels and host preference in Sleman. Ours was a preliminary study and detailed information such as damage to the skin, cultivar, etc. was not collected. Further studies should include these categories to determine whether the hosts are good or poor.

ACKNOWLEDGEMENT
This study was supported by the funding from the Australian Centre for International Agricultural Research under Project HORT/2015/042 'Development of area-wide management approaches for fruit flies in mango for Indonesia, Philippines, Australia, and the Asia-Pacific region.' This article is a part of the first author master thesis required as partial fulfillment for the completion of a Master degree.