The papaya mealybug, Paracoccus marginatus Williams and Granara de Willink (Hemiptera: Pseudococcidae), is a polyphagous pest that damages many tropical crops. It is a native of Central America and spread to the Caribbean region and South America in the 1990s; since then it has accidentally been introduced to some islands in the Pacific region and some countries in Africa and Asia. We recorded its presence in China for the first time in 2013 from Guangdong Province and Yunnan Province in southern and southwestern China, respectively. Our genetic analysis revealed that only 1 haplotype of P. marginatus has been recorded in all of Asia, including China, reflecting the very recent invasion of P. marginatus in Asia. This study also includes a summary of the global geographical distribution of P. marginatus.
The papaya mealybug, Paracoccus marginatus Williams and Granara de Willink (Hemiptera: Pseudococcidae), is a small, yellowish, polyphagous sucking insect and is considered a significant pest of many tropical and subtropical fruits, vegetables, and ornamental plants (Miller et al. 2002). It feeds on the sap of plants by inserting its stylets into the epidermis of the leaf and into the fruit and stem. The leaves become crinkled, yellowish, and withered (Tanwar et al. 2010; Suganthy et al. 2012; Kirsur et al. 2014; Ben-Dov 2015). Sooty mold develops on honeydew excreted by this mealybug; this mold eventually covers the leaves, fruits, and stems, impeding photosynthesis and gaseous exchange. The result is chlorosis, plant stunting, leaf deformation, early dropping of the leaves and fruits, a heavy accumulation of honeydew, and death of the host plants. The mealybug also deposits a thick, white, waxy material that can make plants inedible (Muniappan 2011). Although the papaya mealybug originated in Mexico, it was not a serious pest there or in Central America as its natural enemies kept it under control (Miller et al. 1999). However, when it spread to the Caribbean and to Florida, USA, the mealybug caused significant yield losses not only in papaya but in more than 60 other crops, particularly horticultural species (Miller et al. 2002; Meyerdirk et al. 2004; Walker et al. 2006; Heu et al. 2007).
Paracoccus marginatus was first collected in Mexico in 1955 but first described by Williams and Granara de Willink in 1992 from Mexican specimens collected on cassava (Manihot esculenta Crantz; Malpighiales: Euphorbiaceae); it was thus thought to be native to Mexico and South/Central America (Williams & de Willink 1992; Miller et al. 1999, 2001; Miller & Miller 2002). It was first reported as a pest of papaya (Carica papaya L.; Capparales: Caricaceae) in the Caribbean during the 1990s, and by 2000 it had spread to many Caribbean countries as well as Florida in the United States (Pollard 1999; Muniappan 2009). Over the next decade, there were additional reports from many countries in Oceania, Asia, and Africa (Meyerdirk et al. 2004; Muniappan et al. 2006, 2008, 2009; Muniappan 2009).
Paracoccus marginatus was first reported in India in 2008, and in subsequent years it has invaded 9 additional Asian countries (Muniappan 2009, 2011; Chen et al. 2011; Myrick et al. 2014). In this study, we provide the first report of its presence in southern and eastern parts of China and identify its genetic haplotype in China and other Asian and African countries. These results provide the only genetic survey of P. marginatus from Asia and enhance our understanding of the invasion of this detrimental insect pest.
Materials and Methods
We designated in total 12 sites in mainland China, Cambodia, Malaysia, and Mozambique (Table 1) and used 3 individuals from each. The samples from Cambodia, Malaysia, and Mozambique were intercepted at the entry port of China. However, the Chinese samples were taken randomly from those sites wherever the pest was reported in China. Specimens of P. marginatus were collected with forceps and preserved in 95% alcohol. Species identifications were confirmed using both morphological characters (Miller & Miller 2002) and DNA barcodes from BOLD System V3 (Ratnasingham & Hebert 2007). For the haplotype analysis, we also incorporated genetic data from specimens collected at an additional 9 sites in India, Indonesia, and Thailand; sequences were downloaded from GenBank, BOLD, and EMBLEBI (Table 1).
The total genomic DNA was extracted from individual specimens of P. marginatus using a Promega Wizard® Genomic DNA Purification Kit (Madison, Wisconsin, USA). We used Wizard® Genomic DNA Purification Kit animal tissue protocols as described in their technical manual, #TM050 ( http://www.promega.com/resources/protocols/technical-manuals/0/wizard-genomic-dna-purification-kit-protocol/).
Identification and distribution for the haplotype of Paracoccus marginatus samples along with its genetic identities.
Polymerase chain reaction (PCR) amplification was performed in 25 µL volume reactions using final concentrations of the following reagents: 2.5 µL of 1× PCR buffer, 1.5 µL of 25 mM MgCl2, 2.5 µL of 2 mM dNTP mix with 0.2 mM of each nucleotide, 1.25 µL of each 0.5 µM primer, 2.0 µL of 50–125 ng DNA template, 0.5 µL of 0.5 U of DNA polymerase containing 5 U/µL (Takara), and 14.75 µL of nuclease-free water. We amplified partial sequences (~800 bp) of the cytochrome oxidase subunit I (COI) gene using the universal primer pairs CJ-J-2183 (alias Jerry, 5′-CAACATTTATTTTGATT TTTTGG-3′) and TL2-N-3014 (alias Pat, 5′-TCCAATGCACTAATCTGCCATATTA-3′) (Simon et al. 1994). The COI gene was amplified under the following conditions: 3 min at 95 °C, followed by 30 cycles at 94 °C for 1 min, 57 °C for 30 s, and 72 °C for 1min, and a post-cycle incubation at 72 °C for 5 min. After that, 5 µL of the PCR product underwent electrophoresis on 1.2% agarose gels. When bands of the expected size, ~800 bp, were visible in the gels, the other 20 µL of the PCR product was used for sequencing using an ABI 3730xl sequencer (Invitrogen Life Tech Co. Ltd., Shanghai, China) with the same primers for both directions.
Sequences were aligned using multiple sequence alignment with the “GINSI” algorithm in MAFFT (Tamura et al. 2011). The ends with incomplete codons were then trimmed to produce 712 bp alignments. In order to exclude possible nuclear mitochondrial pseudogenes (numts) commonly found in invertebrates (Buhay 2009), we screened translated COI sequences for stop codons.
Analyzing closely related populations is challenging because the small genetic distances between individuals can obscure evolutionary pathways. In these cases, tree-based phylogenetic methods are poor at representing intraspecific gene evolution (De Barro & Ahmed 2011). Network analysis is a more appropriate means of exploring relationships within species, and many different networking approaches have been developed to estimate intraspecific genealogies (Fitch 1997; Bandelt et al. 1999).
The haplotype identifier refers to haplotypes belonging to various locations; the haplotype identifiers are listed in Table 1. The sequences were analyzed using statistical parsimony (Templeton et al. 1992; Posada & Crandall 2001) with the program TCS v.1.21 (Clement et al. 2000) with a cut-off set at 95% (Hart & Sunday 2007; Chen et al. 2010). Each sequence was assigned a haplotype identifier (Table 1), and the resulting network provided both the relationship between the different haplotypes and the significant number of substitutions connecting haplotypes (Templeton et al. 1992).
Twenty-one COI sequences were analyzed for genetic identity, 12 of which were sequenced specifically for this study. Most sequences were from China (2 from Yunnan and 4 from Guangzhou) and India (7), and the rest from Cambodia (3), Indonesia (1), Malaysia (1), Mozambique (2), and Thailand (1).
Pair-wise genetic distance values were calculated in MEGA5 (Tamura et al. 2011). The use of the genetic species concept and these values are not meant to be a strict delimitation, but are a guide to help estimate the number of possible genetic lineages in our samples.
FIRST RECORD OF P. MARGINATUS IN CHINA AND ITS DISTRIBUTION IN THE WORLD
Our study recorded the first presence of P. marginatus from 2 provinces of China: the southwestern Yunnan Province and the southern Guangzhou Province. Chinese specimens were found on 4 host plants: C. papaya, Jatropha integerrima Jacquard, Jatropha cuarcas L. (Malpighiales: Euphorbiaceae), and Plumeria rubra L. (Gentianales: Apocynaceae). Our study also provides the first Southeast African record of P. marginatus, in Mozambique. In addition, we surveyed the invasion history of P. marginatus and found that it has invaded more than 54 countries and regions in the last 22 yr (Table 2).
HAPLOTYPE DIVERSITY AND DISTRIBUTION OF P. MARGINATUS
We analyzed the genetic identities of Chinese populations of P. marginatus and compared them with genetic identities of P. marginatus populations from 6 other countries in Asia and Africa. The results indicated that only 1 haplotype is present across all sampled countries (Table 1).
Shortly after P. marginatus was first reported to have invaded India in 2008, it was found to have invaded a number of other Asian countries Our study adds the first Chinese records, from 2 noncontiguous provinces, and the first record from southeastern Africa. This further demonstrates the potential for P. marginatus to rapidly invade new regions.
The presence of the same of P. marginatus haplotype across all sampled Asian countries suggests that all specimens stem from the same population resource that initially invaded Asia in 2008, and perhaps from the same population that entered Mozambique. This is another typical case of rapid invasion given that the reported rate of COI evolution for some model insects is ~1.5 to 2.3% per million years ago (Nakamine & Takeda 2008). Similar cases have been found with other invasive Sternorrhyncha. A single haplotype of the cotton mealybug Phenacoccus solenopsis Tinsley (Hemiptera: Pseudococcidae) has undergone a very recent rapid invasion in China (Ahmed et al. 2015), and a single haplotype of an invasive whitefly has established a global distribution (De Barro & Ahmed 2011).
The low levels of variation between geographically distant populations of Asia support the hypothesis that this was a recent invasion that likely occurred during the last 4 to 6 yr. However, the route that P. marginatus used as it spread across regions is still unclear. More sampling and investigating of genetic identities of P. marginatus from its native countries in South and Central America and from other invaded countries, especially those in the Caribbean, will help to understand the precise route of its invasion.
One interesting topic we were unable to address with our data is whether certain external factors contributed to the rapid spread of P. marginatus populations across Asia. One report suggests trade of ornamental flowers could have facilitated the invasion (Qian et al. 2007). However, the mealybug itself was reported to have high biological adaptability, ecological adjustability, temperature tolerance, and a wide range of host plants that may help it to establish quickly in newly invaded places (Arif et al. 2009; Hodgson 2009; Vennila et al. 2011, 2013; Xin et al. 2011). This rapid spread suggests that current quarantine measures have failed to stop the invasion of P. marginatus into China and other neighboring countries. Therefore, officially listing P. marginatus as an invasive species, along with revised quarantine regulations and protocols, may help stop further invasion by this devastating pest.
The recent establishment of P. marginatus is of serious concern because it is a major pest of many species of tropical fruits, vegetable crops, and ornamental plants. Multiple control strategies should be developed such as early detection and eradication in the areas with new records, and biological control for areas where P. marginatus has been established. Within 3 mo of the first discovery of P. marginatus in Indonesia, the magnitude of the mealybug problem caused the government to consider development of a classical biological control program against it (Muniappan et al. 2008). Successful results with classical biological control were achieved in India, Guam, Palau, Florida, and Hawaii in recent years (Meyerdirk et al. 2004; Muniappan et al. 2006; Walker et al. 2006; Heu et al. 2007). In India, an encyrtid parasitoid of papaya mealybug, namely, Acerophagus papayae Noyes & Schauff (Hymenoptera: Encyrtidae), was imported from Puerto Rico and released after multiplication. The introduction of A. papayae provided excellent control of the papaya mealybug within 5 mo along with reduced pesticide use, increased production, and increased income (Myrick et al. 2014). This success suggests that classical biological control may also be a viable option against P. marginatus in China.
This study was supported by the National “Twelfth Five-Year” Plan program for Science & Technology Support (2012BAK11B01) and Guangdong CIQ project (2015GDK14). Many thanks go to Li Hua Lv and Fu Zhong Wu for their help in sample collections. Special thanks to David Plotkin (Florida Museum of Natural History, University of Florida, USA) and Asma Zeiri (Department of Biology, Faculty of Sciences of Bizerte, Bizerte, Tunisia) for their useful comments on our initial analysis.