Invasion of Patchiella reaumuri (Kaltenbach, 1843) on taro roots in Asia: a biosecurity concern and new threat to indigenous taro landraces

The invasive taro root aphid, Patchiella reaumuri (Kaltenbach) (Hemiptera: Aphididae) of Western Palearctic origin, is a highly devastating pest of taro, Colocasia esculenta (L.) Schott (Family: Araceae) in the Pacific region and Europe. Consequently, P. reaumuri is considered as an important quarantine pest for taro growing regions. We detected for the first time the infestation of P. reaumuri causing 34.3% yield loss in the field and 62.4% loss in storage. This is the first taxonomically confirmed record of P. reaumuri from Asia and a global first record of its damage in storage. The northeast region of India, being a part of primary centre of origin of taro, is very rich in diversity of taro, where several traditional landraces are important dietary components of the residents, therefore, the invasion of destructive P. reaumuri may wreak havoc, if not timely diagnosed and managed. This study focussed on the detection of P. reaumuri in a new invaded region, on reporting losses caused by P. reaumuri in its new habitat, and the presentation of an illustrative morphological diagnostic and DNA barcode for its diagnosis at quarantine ports .


Introduction
The taro root aphid, Patchiella reaumuri (Kaltenbach, 1843) (Hemiptera: Aphididae) of Western Palearctic origin, is a highly destructive pest of taro, Colocasia esculenta (L.) Schott (Family: Araceae), known to cause 75-100% yield losses in Hawaii and O'ahu (Sato and Hara 1997).This species got accidentally introduced in Hawaii before 1971 (Sato and Hara 1997;Cook et al. 2010) and now has become invasive, feeding on Araceae plants in the Hawaiian Islands (Mondor et al. 2007).In Europe, this species forms leafnest galls on its primary host, Tilia spp., as lime leaf-nest aphid; but anholocyclic populations of P. reaumuri are mainly found on the roots of its secondary host "taro" and related aroid species, where apterous form multiplies parthenogenetically (Cook et al. 2010).
Patchiella reaumuri is highly host-specific, apparently infesting only taro and Araceae plants (Sato and Hara 1997).It feeds on the roots and the affected plant leaves turn small and yellow, and plant remains stunted.Due to its subterranean habits, the infestation cannot be detected before the appearance of aboveground symptoms (Cook et al. 2010).Feeding of P. reaumuri causes root rot, and infestation appears as a white mould on the fibrous taro roots.In the absence of above ground symptoms, the presence of ants near the base of the taro plants can be an indication of aphid infestation (Carmichael et al. 2008).Owing to its destructive nature, subterranean habit and difficulties in management, P. reaumuri has been listed as a pest of quarantine importance for taro growing regions (Biosecurity Australia 2011;Carmichael et al. 2008;Anonymous 2010).In India, we found an infestation of P. reaumuri for the first time on roots/corms of taro (C.esculenta cv.Muktakeshi) in the field and storage at Umiam, Meghalaya.The known distribution of P. reaumuri was restricted to Europe, Hawaii and the Solomon Islands, until the report of Qiaoand Zhang (1999) who described the new species of aphid Gharesia kolokasia infesting taro roots in China, which is believed to be a synonym of P. reaumuri (Sano and Akimoto 2011).
Taro or Cocoyam is an economical, medicinal and traditionally important vegetable crop, primarily grown for its edible starchy corm and is considered a staple in African, Oceanic and Asian cultures (Prajapati et al. 2011;Li and Siddique 2018).Taro is the fifth most consumed root crop in the world, and the global taro production reached 10.53 million tons harvested from 1.63 million hectares in 2017 (FAO 2020a).Nigeria, China, Cameroon, Ghana and Papua New Guinea are the top five taro producing countries in the world (FAO 2020b).Taro, considered as "potato" of the humid tropics, is also cultivated throughout India.Northeast region of India, being a part of primary centre of origin of taro, is very rich in diversity of taro, where several traditional landraces are important dietary components of the residents (Yen and Wheeler 1968).A recent report revealed that many landraces in northeast India have become extinct due to several reasons, including biotic stresses (Thirugnanavel et al. 2015), thus urgent steps are warranted for the conservation of these valuable landraces of the region.Under such circumstances, the invasion of destructive P. reaumuri may wreak havoc, if not timely diagnosed and managed.Besides China, northeast India shares international borders with Bhutan, Myanmar, Bangladesh and Nepal and therefore appropriate management strategies would be certainly essential to stop further spread of this quarantine pest to nearby countries and the rest of India, where it has not been recorded so far.In this framework, our study focussed on detection and feeding damage of P. reaumuri on taro in northeast India.Here, we have also provided an illustrative morphological diagnosis and DNA barcode for the first time for diagnosis of P. reaumuri at quarantine ports.

Location of the study
The field and laboratory studies were conducted at ICAR Research Complex for NEH region, Umiam, Meghalaya (India).The institute and its farms are located at Umiam (25°41′21″N and 91°55′25″E having an elevation of 1010 m above msl).
Occurrence, feeding damage and losses due to P. reaumuri in field and storage The field infestation of P. reaumuri was observed for the first time during the corm development stage (November).Subsequently, the taro roots were monitored for the presence of root aphids till harvest time.Number of infested plants (corms) were recorded during harvesting, and % infestation was determined.Patchiella reaumuri incidence was also noticed in storage.Later, the aphids infesting taro corms were confirmed to be the apterous form of P. reaumuri (see results section).The number of infested corms, the aphid population per corm, rotted corms (%) and weight loss (%) were recorded in storage at monthly interval till May.Number of aphids/corm was counted from 20 corms randomly selected from the stored materials.
In Meghalaya, the taro crop is planted during April-May and harvested during December-January.To understand the genesis of apterous aphids (which appears to be field borne), the infested and healthy (aphid free) corms of taro (variety: Muktakeshi) were planted separately each in five different plots (Size: 3 m × 4 m; Plant spacing: 45 × 45 cm) during June 2019.Plots planted with infested and healthy corms were separated with 25 m distance to avoid aphid movements towards healthy plants.Observations on aphid incidence and damage symptoms (if any) were recorded at monthly interval.The roots of five randomly selected plants (each from infested and healthy plots) were examined to understand the seasonal dynamics of P. reaumuri.For that purpose, the number of aphids per plant (root zone) was recorded at monthly intervals till January 2020 (harvesting time).Since the infestation of P. reaumuri was not observed throughout the season in plants developed from healthy seeds (free from P. reaumuri), such plants were used for comparison of yield with infested plants.In order to know the pattern of yield loss with respect to aphid infestation, the yield of corms was recorded from five plants (each from infested and healthy plots) during different stages of corm development, i.e. from November (24 weeks old plants) till January (33 weeks old plants.The % yield loss was determined by the following formula: Yield loss (%) = 100 − Yield (kg) of infested plants x 100 Yield (kg)of healthy plants

Taxonomic characterization of P. reaumuri infesting taro roots in Meghalaya
Aphid samples were preserved in 70% ethanol and specimens were mounted in Canada balsam (Blackman and Eastop 2000).Digital camera (Nikon Digital Sight DSVI-1) attached to a Nikon Eclipse 80 I microscope was used to illustrate slide-mounted specimens of aphids.Morphological measurements were made using an eyepiece reticle in an Olympus BX 51 microscope, which was calibrated using a stage micrometer.Measurements are maximum dimensions, and setal lengths include the setal base.All plates were generated using Photoshop CS2.The aphid species was identified using keys prepared by Blackman and Eastop (2000) and description by Stroyan (1979).All specimens have been deposited in the National Insect Museum, ICAR-National Bureau of Agricultural Insect Resources (NBAIR), Bengaluru, Karnataka, India.

Development of species-specific DNA barcode of P. reaumuri
DNA extraction, PCR amplification and sequencing were carried out as per the protocols described in Firake et al. (2013) and Behere et al. (2015).
We sequenced two individual specimens and after trimming ambiguous ends (both 5' and 3'), a final 600 bp good quality sequence was obtained from both the specimens.Sequence identity was compared in BLASTN search (Altschul et al. 1997) against the nr DNA database deposited in NCBI GenBank.

Statistical analysis
Data on infestation, yield loss (in field and storage) at different growth stages was analysed by using one way ANOVA at the 5% level of significance.Before conducting "F" test, the homogeneity of variances among treatments was tested (Levene 1960).Data on yields of infested and undamaged plants were checked for normality, and Mann-Whitney U test (non-parametric test) was used to determine the variation (if any) in the yield of the two groups, i.e. infested and undamaged plants.Data analysis was conducted by using SPSS 21.0 software for windows.

Occurrence, feeding damage of P. reaumuri and losses in taro yield
The field infestation of P. reaumuri was observed for the first time in November 2018 and 38.67% plants were found to be infested by P. reaumuri at harvesting time (December 2018).Interestingly, P. reaumuri infestation was also detected during storage, wherein aphids were found colonising on the young buds/sprouts of stored corms (Figure 1A).In February 2019, a total of 23.19% of corms was found to be infested with aphids (81.85 ± 5.52 aphids/corm, n = 20) with 21.98% losses in stored corms (Figure 1B).Patchiella reaumuri infestation was recorded to be increasing over the storage  time (Figure 2A) and significantly higher infestation (F 8,11 = 64.13,p < 0.001) was observed in May (75.31%).Rotted corms (%) were also found to increase with storage duration (F 8, 11 = 73.33,p < 0.001),where a total of 69.69% of harvested product was lost till May (Figure 2A).Globally, this is the first report of infestation of P. reaumuri to taro corms in storage.
In order to understand their life cycle, we examined root aphid incidence in taro field during June 2019 to January 2020.The number of P. reaumuri declined from planting till December (Figure 2B) but increased enormously during January (329 ± 73.73 aphids/plant) (Figure 1C, D).An unidentified ant species of subfamily myrmicinae was found accompanying P. reaumuri in the field.The yield of infested plants was significantly reduced (Mann-Whitney test, U = 0, p = 0.009) during December and January (Figure 2C).Infested plants produced 34.29% less yield than undamaged plants (Figure 2D).The experimental location received high (293 to 436 mm) rainfall during June to October (Figure 2E).Interestingly, when we examined few unharvested corms in the field, a very severe infestation of P. reamuri was observed in May 2020 (Figure 1E).It indicates that, P. reaumuri reproduces parthenogenetically over the year on the roots and young buds of taro without producing winged (alate) form in Meghalaya.

Morphological characterization of P. reaumuri infesting taro roots in Meghalaya
Live characters of P. reaumuri infesting taro in field and storage First instar elongate translucent yellow (Figure 1F) with transparent appendages and dark brown eyes.Nymphs greenish yellow (Figure 1G), completely grown apterous viviparous females elliptical, broader at the abdomen; appears white due to wax coating (Figure 1H); posterior part of abdomen with thick flocculent wax.All appendages are dark brown.
Taxonomic description of slide mounted specimens of P. reaumuri Taxonomically, the collected root aphid specimens were found matching to the key characters of Patchiella reaumuri described by Stroyan (1979) and Blackman and Eastop (2000).The key characters include, head dark except for marginal areas which are paler; three faceted eyes on dark areas; three apical segments of rostrum dark, proximal segments pale; all antennal segment and all segments of legs dark; dorsum and venter membranous, pale except for slightly darker wax plates; genital plate, anal plate and cauda dark.
Biometric data of adult female of P. reaumuri can be seen in Supplementary material Table S1.Brief description: Body length 1.9-2.2mm (Figure 3A).Antennae short, 5-jointed (Figure 3B), together 0.48-0.49mm long.Primary rhinaria small, oval, with a well-developed ciliate fringe and last segment of antenna with 4 to 5 short terminal processes (Figure 3C).Secondary rhinaria absent.Eyes three faceted (Figure 3D).Rostrum (Figure 3E) about 1.10-1.11mm long from base of retractor apodeme of segment II to apex; apical segment (IV+V) 0.185-0.1907mm long which is 0.64 times as long as hind tarsus II, and bearing 3-6 subsidiary hairs in addition to the 3 apical pairs and the small pair at extreme base of segment; there is a distinct pale zone between the dark main part of the apical segment and the strongly sclerotic apex.Legs dark sclerotic; tibia maximally 42-43 µm long; 3-4 hairs at apex of each tibia modified into stouter, spur-like spines 25-27 µm (Figure 2F).First tarsal joints each with 2 acute hairs at apex, 20 µm long but slender than apical spurs of tibiae (Figure 3F).Anal plate rectangular (Figure 3G) with 18-20 hairs.Subgenital plate narrow (Figure 3H), transverse, with 7-12 hairs across its anterior half and 9-12 in the posterior marginal series.Siphunculi absent.Cauda rounded (Figure 3I) and bearing 4-6 hairs.Dorsal wax gland plates present on all segments from head to abdominal tergite 7. The gland plates are of two different types: type I (Figure 3J), in which the individual facets are irregularly angular or polygonal, usually well separated from each other and without any apparent differentiation of their central area; and the type II (Figure 3K) in which the facets are rather closely grouped, so that they influence one another's shape towards a more regular sub-hexagonal honeycomb arrangement, and in which each facet is seen under phase contrast to contain a small darker central area compared to the peripheral part.Tergite 8 with two pairs of long stiff hairs (Figure 3L).Ventral side of coxa of all the legs with reticulate sculpturing on anterior side, such network was also seen on the sclerites that joins coxa to the ventral derm (Figure 3M).

DNA barcoding and molecular characterization of P. reaumuri
From molecular data, the BLASTN search of 600bp sequence derived from our P. reaumuri specimens shown 91.51 to 95.50% identity with top five hits, namely; Eriosonatinae sp.(MF837565) 95.5%, Pemphigini sp.(JX536292) 95.5%, Patchiella reaumuri (KF639584) 93.5%, Aphis sanguisorbicola (GQ904111) 92.3% and Aphis fabae (KY323029) 91.5%.Interestingly, the percentage identity of COI sequence of our P. reaumuri was only 93.5% with the sequence of Italian P. reaumuri (KF639584) deposited by Coeur d'acier et al. (2014).Pair wise sequence alignment of partial COI gene (600 bp) of Indian (MT796073) and Italian P. reamuri (KF639584) detected 39 variable sites.Of the 39 substitutions, 34 were synonymous and 5 were nonsynonymous substitutions.Three non-synonymous substitutions resulted in change of amino acid from L-leucine (Leu) to L-methionine (Met) and others were L-isoleucine (Ile) to L-valine (Val) and L-Asparagine (Asn) to L-tyrosine (Tyr).Overall findings indicate that, the COI sequence of Italian specimen of aphid (KF639584, referred as P. reaumuri) is substantially different from our P. reaumuri.The representative DNA barcode (COI sequence) of correctly identified specimens of our P. reaumuri has been deposited in the NCBI GenBank vide accession number "MT796073".

Discussion
We detected anapterous viviparous form of P. reaumuri infesting taro corms in India.Patchiella reaumuri is a dioecious aphid species, which host alternates between lime or linden (Tilia spp.) and the roots of Araceae plants (Blackman and Eastop 2020).In Europe, P. reaumuri feed on the primary host (Tilia spp.) during winter as lime leaf-nest aphid; however, they are known to be the most destructive pest of secondary hosts (Araceae) as taro root aphid in Hawaii and Oahu (Sato 2000;Sato and Hara 1997).In order to complete the life cycle, P. reaumuri has to alter their host during winter, where they exclusively feed on the lime or linden plants (Tilia spp.).In Europe, the mature alate leave the lime plants during May-June and colonise the roots of Arum maculatum as a wingless form (Stroyan 1979).Lime plants and Tilia spp.are native to Europe and are not naturally occurring in India, although a few species of Tilia trees have been introduced in Himachal Pradesh for cultivation as ornamental plants (Anonymous 2020a, b, c).The absence of lime trees (Tilia spp.) in northeast India further confirms that the P. reaumuri population on taro roots in Meghalaya is an anholocyclic, developing parthenogenetically on taro roots throughout the year.Similar to our conditions, P. reaumuri has also lost their primary host in Hawaii and Solomon Islands, where the wingless forms reproduce without fertilization round the year and cause heavy losses (up to 100%) to the dryland taro (Sato and Hara 1997;Sato 2000;Biosecurity Australia 2011).In the last decade, India has witnessed the introduction of two species of aphids viz., Wahlgreniella nervata (Gillette) and Liosomaphis ornata Miyazaki, which have become pests on ornamental roses and the medicinal plant Berberis lyceum Royle, respectively (Joshi et al. 2014;Nadda and Joshi 2015).
Patchiella reaumuri can reproduce without fertilization by males; thereby increases the probabilities of a population upsurge in short time (Sato and Hara 1997).Early growth stages of plants are highly susceptible to aphid attack, particularly in drought condition (Cook et al. 2010).In the Hawaiian Islands, the newly planted taro plants face severe attack of P. reaumuri, where yellow-green aphid colonies are sometimes found on both the roots and the basal portions of the leaf sheaths, ultimately causing death of the taro plant (Sato and Hara 1997;Sato 2000).Although severe plant mortality (young stage) due to P. reaumuri was not observed in Meghalaya, nevertheless we experienced a total of 34.29% yield loss in the field.Since our experimental location received very high rainfall (293 to 436 mm) during June to November, the soil moisture level was very high, which might have hindered the development of P. reaumuri and their associated ants.Conversely, relatively low moisture during critical stages of corm development (December-January) might have favoured aphid multiplication in the field leading to yield loss.Furthermore, significant correlation of aphid population and yield loss was also supported by our findings in the field, as we observed an increase in the trend of an aphid population with decrease in the corm yield from November to January (low moisture period).Very high level of soil moisture could be a probable reason for its absence in wetland taro cultivation across the globe (Sato and Hara 1997;see Cook et al. 2010).Therefore, our findings indicate that, the prolonged drought conditions in the future may favour aphid development in such areas of the country.
Besides field infestation, we found significant damage of P. reaumuri during storage, leading to corm rot.Faster development of aphids in storage attributed to the accessibility of sprouting buds for feeding triggered with favourable climatic conditions (21-24 °C temperature and 75-85% relative humidity) during storage (March to May).In order to survive over the year, the wingless form needs to feed on taro roots/corms in the absence of its primary host.Patchiella reaumuri infested corms were often found in rotted conditions in storage.Patchiella reaumuri is known to cause root/corm rot in field conditions (Carmichael et al. 2008); however, since worldwide this is a first report of P. reaumuri infestation in storage, the information is hitherto not available in association of root aphid damage and corm rot.Although a number of pathogens are reported to cause corm rot in storage (Jackson and Gollifer 1975;Ugwuanyi and Obeta 1996), we found rotting only in aphid infested corms and aphid free corms were totally free from rot.Thus, there are strong possibilities of direct role of P. reaumuri in corm rot.Based on our observations, we believe that the corm rotting in storage was possibly due to root aphids, P. reaumuri, as they are known to cause root rot or corm rot in field conditions.Nevertheless, further detailed studies would provide more insights on the role of P. reaumuri in corm rot or their association (if any) with pathogens responsible for rotting.
Patchiella reaumuri has been recorded so far from Hawaiian and Solomon islands of the Pacific (Sato 2000;Foottit et al. 2012) and from 19 countries/regions in Europe (de Jong et al. 2014;Tsitsipis et al. 2007;Basky 2013).An aphid species Gharesia kolokasia has been reported in China infesting Colocasia sp.(Qiaoand Zhang 1999), which certainly is a synonym of P. reaumuri (Sano and Akimoto 2011).We have also found substantial variation in COI sequences in between our P. reaumuri and Italian P. reaumuri sequence deposited in NCBI by Coeur d'acier et al. (2014).The genus Patchiella is monotypic and hence it has become difficult to establish the relationship of P. reaumuri with other aphids at molecular level and therefore further detailed studies are required to understand the genetic variations in P. reaumuri from different localities.
Overall facts indicate that, the simplified and illustrative diagnostic characters are urgently needed for the reliable identification of P. reaumuri.Therefore, in this report, we have provided the key diagnostic characters of P. reaumuri along with a newly recognised character "possession of reticulate ornamentation of coxae of legs", which was not indicated in earlier detailed description by Stroyan (1979).We have also supported the morphological identity of the species with its molecular data.This study is the first to taxonomically confirm the record of P. reaumuri from Asia.
Invasion routes and time of P. reaumuri in India are yet unclear.Short distance spread or dispersal of root aphids generally occurs through ants, but long distance spread occurs mainly by movement of planting material, leaves and corms (Carmichael et al. 2008).Many ants pecies are known to be associated with taro root aphids, which indirectly contribute in plant damage by increasing the aphid population (Sato and Hara 1997).Ants usually farm aphids for honeydew; in turn they provide protection to aphids from predators.Human transmission of planting materials is supposed to be the most common pathway to disperse the taro root aphid in new areas (see Cook et al. 2010).Two decades back, Gharesia kolokasia (which is certainly a synonym of P. reaumuri) has been reported in China, albeit no further status of this species is available hitherto from China.Since the northeast India is geographically near to China, the chances of spread of P. reaumuri through planting materials can't be ignored.

Conclusions
The present study concludes that P. reaumuri develops parthenogenetically on taro roots/corms all over the year and cause significant yield losses in India.This study provides morphological illustrations supported by molecular data for its identification.Considering the biosecurity significance, the key diagnostic characters of P. reaumuri mentioned in this study such as black antennae and legs, 3-facetted eyes, 4 to 5 short terminal processes on the last segment of antenna, absence of siphunculi and newly identified character "possession of reticulate ornamentation of coxae of legs" would be helpful to quarantine personnel and other stakeholders for reliable identification of P. reaumuri.

Figure 1 .
Figure 1.Infestation of Patchiella reaumuri to taro corms in field and storage: A. Apterous viviparous females colonizing on buds; B. Corm rot due to heavy infestation of aphids; C. Aphid infestation on roots in the fields during January; D. Close up showing severity of the infestation on roots; E. Increase in aphid infestation to unharvested corms in moisture stress condition.F. First instar nymphs; G. Grown up nymphs without wax coating on dorsum; H. Completely grown up females with thick flocculent wax on posterior parts of abdomen.Photographs by Dr. D.M. Firake, ICAR RC NEHR, Umiam, India.

Figure 2 .
Figure 2. Development and losses caused by P. reaumuri to taro in field and storage.Error bar represents standard error of means: A. Infestation and losses due to P. reaumuri in storage; B. Seasonal incidence of P. reaumuri on taro roots under field conditions; C. Effect of P. reaumuri damage on yield at different stages of corm development; D. Yield loss caused by P. reaumuri in field; E. Rainfall received during the study period and RD stands for number of rainy days in a month.

Figure 3 .
Figure 3. Diagnostic characters of P. reaumuri.A. Apterous viviparous female; B. Five segmented antenna; C. Primary rhinaria with ciliate fringe and last segment of antenna with 4 to 5 short terminal processes; D. Three faceted eyes; E. Ultimate rostral segment; F. Apex of tibia with spur like spine and first tarsal segment with acute hairs at apex; G. Anal plate; H. Subgenital plate; I. Cauda; J. Type I dorsal wax gland plate; K. Type II dorsal wax gland plate; L. Tergite 8 with long stiff hairs; M. Reticulate ornamentation on coxae of legs.Photographs by Dr Sunil Joshi, ICAR-NBAIR, Bengaluru, India.