The oriental fruit fly, Bactrocera dorsalis Hendel (Diptera: Tephritidae), is a generalist feeder that is known to successfully feed and breed on a variety of fruits and vegetables in the tropical and subtropical zones of Asia. Recently, the northward extension of B. dorsalis has been reported, and this geographic range increase relates to both climate changes and the rapid development of domestic and international trade and travel. However, it has not been possible to determine the origin of this pest, which is trapped in north China. In this study, traps baited with the sex attractant methyl eugenol were used to collect B. dorsalis in Beijing, China. With these traps B. dorsalis populations were sampled at the same time in 5 different latitudes of China and used to determine the distribution of δ2H stable isotope values. The relationship between rainwater and the B. dorsalis δ2H stable isotope values at these sites was modelled, which then could be used to hypothesize the population origin of B. dorsalis trapped in Beijing, China. The results showed that δ2H stable isotope values for B. dorsalis from Beijing were not consistent with that of the rainwater in Beijing, but rather were consistent with the results obtained from Fuzhou in southeast China. Bactrocera dorsalis trapped in Beijing was not a resident population, and may have come from South China. The fruit and vegetable trade may have vectored the fly northward in China. Our results also showed that δ2H stable isotope technology is a promising strategy for tracing the population origins of invasive fruit flies.
The oriental fruit fly, Bactrocera dorsalis Hendel (Diptera: Tephritidae), is a generalist feeder that has been known to successfully feed and breed on a variety of fruits and vegetables, including citrus, guava, litchi, sugar apple, mango, pepper, and papaya (Clarke et al. 2005; Vargas et al. 2012). Bactrocera dorsalis is recorded mainly from the tropical and subtropical zones of Asia, but as of 2016 it had spread to 5 continents, including more than 70 countries and 120 geographical regions (Stephens et al. 2007; Manrakhan et al. 2015). The distribution of B. dorsalis is still increasing through invasion of climatically suitable regions throughout the world (De Villiers et al. 2016). Larval feeding of B. dorsalis causes the abscission of immature fruits and vegetables, leading to major economic losses to many such crops. Adults of B. dorsalis live more than 3 mo in tropical regions, feed primarily on nectar and pollen, and actively fly to seek oviposition sites in fresh fruits or vegetables (Mwatawala et al. 2015).
Bactrocera dorsalis historically had only a narrow distribution in south China (south of 25° NL) (Fan 1998). A climate-matching model supported the assumption that B. dorsalis could not survive in northern China (Fan 1998; Zhan et al. 2006), and that the area north of the Yangtze River was not suitable for the population survival and overwintering of B. dorsalis (Li et al. 2011). Until recently, researchers believed that B. dorsalis could not permanently establish in northern China.
In 2008, however, B. dorsalis was reported in the Wuxi district of Jiangsu province (31° north latitude), where it caused serious economic damage to citrus fruits due to a high population density (Qi et al. 2008). The domestic trade of fruits and vegetables from south China to north China likely was an important factor enabling the spread of B. dorsalis to new areas (Qi et al. 2008). Prior research has shown that the larvae of B. dorsalis can be transported in fruits and vegetables to new regions (Goergen et al. 2011; De Villiers et al. 2016). Movement of infested fruits or vegetables in the vectoring process is further supported by the collection of many adults of B. dorsalis found in Jun 2012 in apple and peach orchards located close to a large fruit and vegetable wholesale market in Fangshan District of Beijing (Qu & Sun 2013; Wang et al. 2016). In 2014, adult B. dorsalis were trapped during Jun in a vineyard (1.2633°E, 39.72°N) in Beijing. In 2015 and 2016, no larval damage or pupae were found in this orchard.
Although the population origin of B. dorsalis trapped in northern China has been surmised to be southern China, and based on movement of infested product, this has been surmised only, not experimentally demonstrated. However, the natural variation in a hydrogen stable isotope ratio (δ2H) is a useful tool to investigate such questions and determine the temporal-spatial dynamics of ecological pathways (Bortolotti et al. 2013; Voigt et al. 2015). Recently, stable isotope technology has been applied successfully to track the dispersal pathways of birds and insects (Rubenstein & Hobson 2004; Forbes & Gratton 2011). The principle of the technique is that the stable isotope composition found in the tissues of an organism is set by the introduction of H isotopes through its diet, which in turn reflects the signature of the micro-environment in which the organism has grown (Solomon et al. 2009). The stable isotope ratio of δ2H precipitation varies geographically and this ratio is determined by local geophysical and chemical cycles (Wang et al. 2009; Wu et al. 2016). The δ2H stable isotope ratio forms a continuous gradient in China from south to north and from the coast inland (Voigt et al. 2015; Deng et al. 2016). In addition, the δ2H stable isotope has a reliable relationship within ecosystems from low to high trophic levels. The δ2H stable isotope composition of an organism therefore provides a signature of the organism's natal environment through its diet (Holder et al. 2014; Susilawati et al. 2016; Peng et al. 2016).
Stable isotope (δ2H) relationships have been used to determine the origin of various insect populations, including Danaus plexippus Linnaeus (Lepidoptera: Nymphalidae), Episyrphus balteatus (De Geer) (Diptera: Syrphidae), Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae), and Arhopalus ferus (Mulsant) (Coleoptera: Cerambycidae), and the results have demonstrated that a stable isotope ratio can be used as a biogeographical location marker (Dawadi & Lugtenburg 2013; Holder et al. 2014, 2015).
We hypothesized that B. dorsalis populations were repeated entries in the early season each year, and did not colonize in the area of Beijing because there were no detections in winter. In the present study, δ2H isotope analyses were used to determine the population origin of the first generation of trapped B. dorsalis adults collected in Jun 2014 and 2015 in northern China (Beijing).
Materials and Methods
SAMPLE COLLECTION IN NORTHERN CHINA
Bactrocera dorsalis collections were carried out in the Fangshan district of Beijing, North China, and the surrounding rural areas in Jun 2014 and 2015. Adults were attracted to traps with an attractant (methyl eugenol) in an organic Bolongbao grape vineyard (1.2663°E, 39.72°N) using the random 5-point collection method of Zhao et al. (2015) with at least 10 m between sampling points. Bactrocera dorsalis was trapped in early Jun to collect the first generation of B. dorsalis populations in that year (the first arrivals into the district). Every 2 wk, the traps were examined, and the sex attractant renewed. All collected B. dorsalis individuals were transferred to small vials with 100% alcohol. Although adults were collected in traps, field surveys of fruit did not result in any collections of eggs or larvae in fruit from the field investigation.
Also, we collected samples of B. dorsalis adults from a nearby fruit and vegetable market of Xinfadi (1.2723°E, 39.82°N) using the same trapping method. All collected adult samples were taken back to the laboratory and stored at -20 °C for further δ2H determination (see below for method).
COLLECTION AND HANDLING OF SAMPLES FOR δ2H STABLE ISOTOPE ANALYSIS
Bactrocera dorsalis adults, collected in a variety of locations in southern and central China, were placed individually into Eppendorf (EP) tubes and transferred to a drying closet at 60 °C for 48 h. After flies had been allowed to dry for 48 h, samples were transferred to a mortar for grinding. To guarantee the fineness of samples, each individual B. dorsalis was ground for more than 10 min. The entire body of the B. dorsalis adult was ground, which represents the isotope ratios derived from the fruit consumed by its larva, marking its geographic origin, through its match with hydrogen isotopes in rainwater (Bortolotti et al. 2013). Ground samples were weighed using a microbalance (0.0001 g) and wrapped in silver paper for further δ2H analysis. Samples were stored at room temperature for 2 to 3 d to equilibrate before examination by spectrometer. Finally, an isotope ratio mass spectrometer (Thermo Scientific MAT 253, Thermo Fisher Scientific, Inc., Waltham, Massachusetts, USA) was used to examine the δ2H stable isotope of B. dorsalis samples through the differences of neutron number in hydrogen.
SAMPLE COLLECTION IN SOUTHERN AND CENTRAL CHINA
We collected B. dorsalis in southern and central China, in Haikou, Guangzhou, Fuzhou, Wuhan, and Yixing, using the same trapping methods as described above for northern China. These location samples represent an anticipated latitudinal gradient of δ2H stable isotope values. All sampling sites were located with a Geographical Position System (GPS) and their elevation recorded at the same time.
We obtained the precipitation δ2H stable isotope values for each sample location from Waterisotopes.org of the Online Isotopes in Precipitation Calculator (OIPC), which is provided by the International Atomic Energy Association (IAEA) and the World Meteorological Organization (WMO). This database includes the global net precipitation station and provides the water δ2H values of stable isotopes (Yann et al. 2013).
Using the laboratory method described above, we determined the δ2H stable isotope values of the flies collected from each location in southern and central China. Those data were then used to construct a standard curve equation of δ2H stable isotopes relating values in B. dorsalis to those of geographical locations.
STATISTICAL ANALYSIS
The relative abundance of the δ2H stable isotope (heavy vs light H) in each B. dorsalis fly relative to the international calibration standard, Vienna Standard Mean Ocean Water (VSMOW) (Tanaka & Nakamura 2013), was determined using the equation below:
where δ2H is the ratio of heavy hydrogen element isotopic (deuterium) to the light hydrogen element stable isotope (protium), which is an international standard of isotope measurement. Rsample is the ratio of heavier to lighter hydrogen element isotopes in the B. dorsalis sample, and Rstandard is the hydrogen element isotopic ratio of VSMOW.Stable isotope values of δ2H stable isotopes (i.e., precipitation and B. dorsalis) were examined using a Gaussian distribution. ANOVA was used to compare the differences of the δ2H stable isotope for B. dorsalis among different geographical locations (Duncan's method). Then, the relationship between the δ2H stable isotopes from trapped B. dorsalis in the 5 aforementioned geographical locations and the local water was then established using a simple linear regression model, which was a standard curve with a linear equation for calculating the theoretical δ2H stable isotopes. The ANOVA also was used to examine the differences of the δ2H stable isotope values for B. dorsalis between Xinfadi market and Bolongbao grape vineyard (Duncan's multiple range test). The theoretical values of the B. dorsalis δ2H stable isotope then could be obtained through the equation of standard curve at a given water δ2H stable isotope value. We also calculated the diet δ2H stable isotope values of B. dorsalis trapped in Beijing by using the equation. All statistical analysis was performed using R 3.4.1. (R Development Core Team 2016).
Results
The δ2H stable isotope values of B. dorsalis samples decreased from south to north, which was consistent with the relationship between water and the δ2H stable isotope values. The δ2H stable isotope value of B. dorsalis was the highest (-80.3 ± 3.78) in Haikou and lowest (-93.5 ± 4.34) in Yixing (Table 1). Additionally, the δ2H stable isotope values of B. dorsalis among 5 geographical locations had significant differences (Table 1).
Table 1.
The δ2H stable isotope values of Bactrocera dorsalis populations for five geographical locations in China.
The δ2H value of water in Guangzhou for 2015 to 2016 was -38, based on the monthly average precipitation), which corresponded to a δ2H value for flies of -82.3 ± 3.86, empirically derived for B. dorsalis trapped in that location. Based on the 5 geographical locations analyzed, the relationship between the δ2H stable isotope values for B. dorsalis and water can be described well by a simple linear regression (y = 2.8268x + 23.745; r = 0.8169; P1,24 < 0.001; Fig. 1). This indicates a strong relationship between δ2H in the precipitation and in the local fruit flies, with the flies being less enriched with the heavy isotope of deuterium.
Based on this standard curve and given precipitation in this northern area, the theoretical values of δ2H stable isotope in the B. dorsalis from the Fangshan district of Beijing, north China, should be -130.85. However, the δ2H stable isotope of B. dorsalis in the Xinfadi market in Beijing actually ranged from -81.6 to -93.9, which is significantly higher, more enriched for the heavy isotope of deuterium, than that of the theoretical fly values (F1,9 = 8.98; P < 0.001; Table 2). Similarly, the δ2H stable isotope values for B. dorsalis in Bolongbao grape vineyard in Beijing ranged from -81.6 to -90.3, which also were significantly higher than the theoretical values of -130.85 for that latitude (F1,9 = 9.74; P < 0.001; Table 2). Furthermore, there were no differences of the δ2H stable isotope values for B. dorsalis between Xinfadi market and Bolongbao grape vineyard (F1,9 = 0.23; P = 0.64; Table 2).
Conversely, the δ2H stable isotope value of B. dorsalis from the Xinfadi market in Beijing (-88.14 ± 4.71) was not statistically different from the water δ2H stable isotope of Fuzhou (-87.8 ± 4.85), in southern China (F1,9 = 0.46; P = 0.51; Table 1). In the Xinfadi market (the origin of flies from imported fruit), the δ2H stable isotope value of B. dorsalis was -87.17 ± 3.66, which also was consistent with some locations (Fuzhou of Fujian province) in southern China, according to the standard curve equation (F1,9 = 0.34; P = 0.57; Table 2; Fig. 1). This suggests that the B. dorsalis trapped in the Xinfadi market and Bolongbao grape vineyard in Beijing likely originated from further south, within the known geographic range for this species.
Discussion
Our research showed that δ2H stable isotope technology could be used to determine the likely population origin of B. dorsalis. Previous research also has found that some migratory insects could be traced to their population origins using stable isotope technology (Brattstrom et al. 2010). In plant quarantine especially, population tracing of the emergent species and invasive species could be achieved by determining the δ2H stable isotope composition during commodity trading (Holder et al. 2015).
The stable isotopes were used to determine the population origins of B. dorsalis, and such information could be valuable for detecting alternative hosts of invasive species or invasive pathways on populations outside their original environment (Voigt et al. 2015; Adams et al. 2016). The survival of larval B. dorsalis on hosts such as mango, sugar apple, and lychee, may lead to individuals being transferred to new environments. The stable isotopes or elemental markers could be used to determine the host species of B. dorsalis when the emergent generation of B. dorsalis was trapped in the field (Rubenstein & Hobson 2004; Holder et al. 2014).
Geologically driven δ2H stable isotopes reflect the geochemical cycle characteristics of the origin point and type of climate (precipitation) as intrinsic markers (Holder et al. 2014). Because δ2H stable isotopes could be applied to determine the population origin of emergent species and invasive species, stable isotope technology has great potential for use in plant quarantine (Holder et al. 2015).
In this study, the δ2H stable isotope ratio of sampled B. dorsalis trapped in Beijing was higher than the theoretical values. This inconsistency revealed that the B. dorsalis population came from south China or another country at a similar latitude to southern China where this species is widespread. The movement of infected fruits and vegetables may be the most important avenue for causing the population spread of B. dorsalis in China (Wang et al. 2015).
The δ2H stable isotope also is transmitted within the food web from low trophic level species to high trophic level species at a constant fractional distillation (Dawadi & Lugtenburg 2013; Górka et al. 2017). However, the mechanism of fractionation of the δ2H stable isotope in the precipitation-host-fruit fly relationship is not clear (Bortolotti et al. 2013), and the transmitting mechanism of the δ2H stable isotope in the food web of the ecosystem is an unexplored field for future work (Weber et al. 2017). Other stable isotopes (S, P, and N) should also be evaluated as potential markers for population tracing because the availability of several additional markers would enhance the accuracy and validity of stable isotope technology (Murray et al. 2016).
Table 2.
The measured values and theoretical values of Bactrocera dorsalis δ2H stable isotope from the Xinfadi market and the Bolongbao grape vineyard (the capital letters indicated the differences of Bactrocera dorsalis δ2H stable isotope between measured values and theoretical values, the lower case letters indicated the differences of Bactrocera dorsalis δ2H stable isotope between Xinfadi market and Bolongbao grape vineyard).
In a terrestrial ecosystem, the δ2H stable isotopic is one of the most important elements for tracing diets and origins due to the stable fractionation associated with plant photosynthetic pathways. Some experts have suggested that flight wings may be more suitable markers because they are largely metabolically inert after adult emergence (Holder et al. 2014). One of the major advantages of using such a technique for invasive species such as B. dorsalis is that this pest only feeds on one fruit throughout the larval stage on plant hosts, while the adult stage of B. dorsalis is relatively non-feeding, which mean that the adult signatures of H isotopes only would be derived from larval feeding, and would not be altered or masked due to adult feeding (Wang et al. 2009). Thus, the application of δ2H stable isotopes may be a reliable technology to track population origins or original hosts if conspicuous differences exist among population isotopic signatures. Such δ2H stable isotope technology also could be used to determine the population origins of invasive species.
Based on the research reported herein, we conclude that the first generation of B. dorsalis in 2014 and 2015 in Beijing is not a resident population, and may come from southern China. The fruit and vegetable trade may have vectored the fly northward in China. Further research is needed to evaluate whether combining δ2H element stable isotope data with other element isotopic and trace element concentration profiles would be useful for determining accurate insect provenance (Nagoshi et al. 2007; Ziegler et al. 2016). Stable isotopes, including δ2H and many other isotope elements, may be an important technology for precise population tracing in the field of biosecurity (Hobson et al. 1999; Simard et al. 2008; Hood-Nowotny et al. 2011).
Acknowledgments
We would like to thank Yongjiang Huai and Zhenxia Zhao for assisting with the field sample collection. We also would like to thank Dr. Fan Jiang for assisting with fruit fly species identification, Prof. Zhiyong Pang for sample analysis, and the Stable Isotope Geosciences Facility at Tsinghua University. This work was funded by 13th National Key Scientific Research Projects (No. 2016YFC1200605). Zi-Hua Zhao and Zhenglong Lu contributed equally to this research.