Just phoresy? Reduced lifespan in red palm weevils Rhynchophorus ferrugineus (Coleoptera: Curculionidae) infested by the mite Centrouropoda almerodai (Uroactiniinae: Uropodina)

Abstract

Invasive species usually act as carriers of their associated organisms like parasites and symbionts. This phenomenon has also occurred with the recent colonization of the red palm weevil, Rhynchophorus ferrugineus, in the Mediterranean area: this introduced pest is strictly associated with several species of mites (mainly belonging to the suborder Uropodina). In this paper, we document the high rate of infestation of Central and Southern Italian red palm weevil populations by the mite Centrouropoda almerodai. This mite species was found in all five investigated regions and infested the large majority of individuals (from 57 to 95%) by settling preferentially under the first pair of wings. Although this mite–weevil association is usually considered as a phoretic relationship, i.e. without impact on hosts, our study revealed that infested individuals have a significantly reduced lifespan (by one-third) in comparison with those not infested. Our study provides evidences that C. almerodai imposes a cost on its carrier host, at least under laboratory conditions, leading us to believe that the R. ferrugineus–C. almerodai relationship is also not phoretic in the wild.

Keywords:

• Rhynchophorus ferrugineus
• Uropodina
• mite
• Centrouropoda almerodai
• phoresy

Introduction

One of the risks linked to translocations of living organisms is the accidental spread of their parasites or pathogens. Literature provides examples about the diffusion of parasitic species or endosymbionts carried by their host (Gherardi et al. 2008) as occurred with the worldwide spread of Varroa mites from Asia due to the import/export of Apis colonies (Sammataro et al. 2000). This seems to have occurred for different genera of mites associated with several families of Coleoptera (Atakan et al. 2009). Curculionid beetles of palms are an outstanding example, being highly affected by different species belonging to several families of mites that can be found simultaneously on the same species (Wattanapongsiri 1966), as in the red palm weevil Rhynchophorus ferrugineus (Olivier, 1790) (Coleoptera: Curculionidae) (Longo & Ragusa 2006; Atakan et al. 2009; Ragusa et al. 2009). Three species of the suborder Uropodina can be found on R. ferrugineus: their diffusion outside distribution area (Southern Asia and Melanesia) is probably due to their host (Longo & Ragusa 2006).

Rhynchophorus ferrugineus, a large, polyphagous insect, is an important pest of several species of palms (Murphy & Briscoe 1999; Faleiro 2006); its recent colonization of the Mediterranean area is causing serious economical damages (Faleiro 2006). In Italy, after its first report in Tuscany in 2004, in a few years the species has spread in 11 regions, especially in the Centre and in the South (Sacchetti et al. 2005, 2006; Longo personal observation). Individuals are mainly attracted by dying or damaged palms: inside the palm trunks, females lay hundreds of eggs hatching in a few days. The larvae feed on the tree tissue causing the tree to collapse (Murphy & Briscoe 1999; Faleiro 2006).

The most common mite found on Italian weevils is the Centrouropoda almerodai Wisniewski & Hirschmann, 1992 (Uroactiniinae: Uropodina) (Ragusa et al. 2009): this alien species was found in the large majority of the examined weevil in Sicily (Longo & Ragusa 2006). Deutonimphs of C. almerodai attach themselves with pedicels to the weevils and shelter in clusters especially under the elytra. Centrouropoda almeorodai seems to invade R. ferrugineus in the pupal stage: 86% of the examined pupae carried mites, which were almost absent in the larvae (Longo et al. 2009b).

C. almerodai is considered to establish a phoretic relationship (Longo & Ragusa 2006; Ragusa et al. 2009), i.e. an ‘interaction that enhances dispersal, benefiting the disperser without impacting the phoretic host’ (Holte et al. 2001). However, the life history traits of this mite species are poorly known and the relationship with the R. ferrugineus has never been deeply investigated. The presence of up to hundreds of mites on the same individual and their localization in the ‘softer’ part of the cuticle (under the elitrae and in the intersegmental membranes) lead us to believe that it could represent a cost for the host. Several authors hypothesized that the presence of the mite could in some ways affect the red palm weevil's performance, e.g. by hampering flight behaviour (Atakan et al. 2009). Indeed, even though the target of phoresy is simply dispersal, the relationships between the involved organisms can be complex (Krombein 1962; Fain & Ide 1976; Crawford 1984; Houck & O'Connor 1991). Using a tritium radiolabel, for example, Houck and Cohen (1995) and Holte et al. (2001) found that some astigmatid hypopodidae specimens previously considered as phoretic (Gerson 1967; Gerson & Schneider 1982) extract materials (at least water) from their beetle hosts, revealing a more complex relationship than previously thought.

Here, we investigate the occurrence and the prevalence of the mite in several R. ferrugineus populations of Central and Southern Italy. In laboratory, we compared the lifespan of infested and not infested individuals, hypothesizing that the presence of the mites on the weevil's body could affect and reduce it.

Materials and methods

Sampling sites, collection and rearing of Rhynchophorus ferrugineus

Adult coleopteran females and males (respectively, N = 5909 and N = 5175) have been collected from 2005 to 2008 in five regions of Central and Southern Italy (N = 79 Latium, N = 112 Campania, N = 118 Apulia, N = 10,743 Sicily and N = 32 Sardinia) throughout the whole year. In Sicily, the region most affected by the red palm weevil, individuals were caught by hand-capture on infested Phoenix canariensis Hort. ex Chabaud palms (N = 7574) and by trapping using the aggregation pheromone Rhyfer 220 (Intrachem Bio Italia S.p.A.) (Caldarella et al. 2009) (N = 3169): these two different methods are usually used to monitor and control this species. For the other regions, only hand-capture was used (N = 341).

All collected individuals were immediately examined under the stereomicroscope, in order to check for mite presence. A sample of deutonymphs (100 individuals for site) was also collected and sent to a specialist for the identification. All belonged to C. almerodai. In 2005, a sample of 236 infested and not infested individuals (sex ratio 1:1) from Catania (Sicily) was used for lifespan evaluation. The adults were kept, in Catania (Dipartimento di Scienze e Tecnologie Fitosanitarie), under laboratory conditions (T: 25 ± 5°C, 80–100% UR, natural daylight) in plastic cages (15 × 15 × 15 cm) and fed with apple slices until they died. The animals were monitored daily and we did not find any animals which had visibly died of parasitoids or entomopathogens.

Statistical analyses

The proportions of infested male and female individuals have been compared using the G test after Williams' correction (statistic: G). Differences in survival depending on infestation status, gender and capture method as fixed factors were assessed with a General Linear Model (GLM), with days of survival as the dependent variable. Variances were not homogeneous (Levene test, P < 0.05), but the experimental design was balanced, the sample size was quite large and data were closed to the normal distribution. Under these circumstances GLM is quite robust to deviation from the assumption of omoscedasticity (Hill & Lewicki 2005). In order to be quite confident that the results of this analysis were not inflated by errors due to the violation of the assumption of variances' homogeneity we (a) performed the same analysis with square-root transformed data, obtaining a data set with more equal variances and (b) conducted direct pairwise comparisons with t-test (statistic: t) under the assumption of ‘unequal variances’ of all the relevant subgroups. In both cases the results confirmed the previous analysis. In the text we report mean ± S.D.

Results

Infestation of Rhynchophorus ferrungineus adults by mites of the species Centrouropoda almerodai was very common in our samples. In the investigation carried out in the province of Catania, 91% out of the 10,173 adults examined showed mites under their elitrae. This high prevalence is confirmed both at the regional and subregional scale (5 sampling sites each). Infested adults occurred in all investigated sites, representing a large proportion of the sampled populations (from 57 to 95%, Figure 1). The ratio of infested individuals did not differ between the genders for the plant-captured weevils (Catania population, trap: 847 males infested on 890 total, 2147 females infested on 2275 total, G = 0.81, df = 1, P > 0.1; plant: 3464 males infested on 3846 total, 2815 females infested on 3162 total G = 2.01, df = 1, P 0.1). Both methods collected a high ratio of infested individuals (Catania population, trap: 95% of total sampled individuals, N = 3165; plant: 90% of total sampled individuals, N = 7008) with the highest proportion of infested individuals captured with traps (G = 73.37, df = 1, P < 0.001).

Figure 1. Prevalence of Centrouropoda almerodai in different Italian populations of the red palm weevil. Regional scale on the left, subregional scale on the right (focus on Sicilia). P = proportion of infested individuals, N= total number of collected and examined individuals of R. ferrugineus. Results refer to both females and males collected with the plant-method.

Lifespan is significantly affected by the presence of the mite (F(1,228) = 27.63, P < 0.001): not infested individuals living longer than affected ones (Figure 2). Also ‘gender’ and ‘capture method’ factors have a significant effect. Males live longer than females (F(1,228) = 8.64, P = 0.004; males: 92.63 ± 3.99 days; females: 76.01 ± 3.99 days) and plant-captured individuals live longer than trap-captured ones (F(1,228) = 10.42, P = 0.001; plant-captured: 93.43 ± 3.67 days; trap-captured: 75.22 ± 4.28 days). There is no significant interaction between ‘gender’ and ‘infestation status’ nor between ‘gender’ and ‘capture method’ (gender × capture method, F(1,228) = 0.20, P = 0.654; gender × infection status, F(1,228) = 0.30, P = 0.586).

Figure 2. Adult lifespan (days) of infested (N = 118) and not infested (N = 118) individuals (average and 95% confidence interval); (F(1,228) = 27.63, P< 0.001).

A significant interaction was found between the factor ‘capture method’ and ‘infestation status’ (infestation × capture method, F(1,228) = 4.80, P = 0.030). This means that the way infestation status affects the survival is different depending on the capture method; i.e. in both cases there is a significant reduction in the lifespan of infested individuals, but this decreases more abruptly in trap-captured weevils than in plant-captured ones (Figure 3). Lifespan was similar between the two capture methods in not infested individuals (N plant = 68, N trap = 50, 102.07 ± 39.10 vs. 96.22 ± 51.51 days, t = 0.71, df = 116, P = 0.484) even though a trend is still present, with plant-derived individuals living longer than trap-derived ones, but not in infested individuals (N plant = 68 vs. N trap = 50: 84.78 ± 48.42 vs. 54.22 ± 31.07 days, t = 3.91, df = 116, P< 0.001).

Figure 3. Interactive effect of infestation status and capture method on weevils' lifespan (days) (average and 95% confidence interval are shown). Individual infested: in plant N = 68, in trap N = 50; individuals not infested: in plant N = 68, in trap N = 50; F(1,228) = 4.80, P = 0.030.

Discussion

Our study clearly shows the large association between Rhynchophorus ferrugineus and Centrouropoda almerodai in Central and Southern Italy. Mites were found in all the sampled sites with a great prevalence rate: from more than half to almost all individuals in each population carry C. almerodai deutonymphs. The prevalence rate was quite similar both at the regional and subregional scale. This shows how the diffusion of an invasive pest (R. ferrugineus) could be tracked by the spread of one another alien species (C. almerodai  ) having a close relationship with it. The presence of mites reduced the lifespan in red palm weevils: infested individuals showed a shorter lifespan (1.4-times shorter) compared to not infested ones, suggesting that the presence of mites can impose a cost on the weevils. Mites under the elytra could impede the flight (Atakan et al. 2009), but in our study we also found a high rate of infestation in trap-captured individuals, which should have flown to reach the trap. Moreover, this harm does not seem to be sufficient to explain the extent of the reduced lifespan observed under laboratory conditions.

Mites could directly damage the weevils, as suggested by Ragusa et al. (2009): C. almerodai deutonymphs moult to the adult stage if fed with weevil body parts, leading to the hypothesis that mites do not only use the weevil as a carrier but also exploit it as a protein source. Similarly, for a long time, astigmatid hypopodes (their heteromorphic deutonymphs) have been considered as phoretic (Gerson 1967; Gerson & Schneider 1982), but Houck and Cohen (1995) discovered that hypopodes of an astigmatid mite extract materials from their beetle hosts, suggesting a parasitic relationship. Later, a bi-directional material flow was established in this beetle–mite relationship, which is now interpreted as a mutualistic one.

A significant difference in lifespan was related to the gender, with males surviving more than females, as already reported for some populations of this species (Murphy & Briscoe 1999; Longo et al. 2009a; but see also Avand Faghih 1996), and to the capture method, with plant-captured weevils living longer than trap-captured ones. This could be due to the different collected cohorts of individuals: the plant capture method collects both adults and just-emerged individuals, while the trap capture method based on the aggregation pheromone attracts mostly fertile and ready to mate individuals (Faleiro et al. 2003).

Finally, we found an interaction between the infestation status and the trap method: in both methods, there is a significant reduction in the lifespan of infected individuals, but it decreases more abruptly in trap-captured weevils than in plant-captured ones (Figure 3). A possible explanation could be that a larger number of mites are present in trap-captured individuals than in plant-captured ones. However, research on other Uropodina mite species–R. ferrugineus relationship in Turkey revealed the opposite results (Atakan et al. 2009), while no data are available at present for the C. almerodai–R. ferrugineus association: further studies are necessary to investigate this issue.

Our study does not rule out the possibility that the highlighted difference could be due to the mites' preference for old or unhealthy individuals. Only ad-hoc infestation of randomly chosen healthy individuals of the same age will answer the question. Nevertheless, we reckon that our results are unlikely to be due to an imbalance between healthy/unhealthy or young/old individuals. In fact, since C. almerodai is present in the large majority of the weevil population, it seems improbable that up to 90% of individuals are old or unhealthy. Moreover, it seems that R. ferrugineus–mite association can begin even before the emergence: more than two-thirds of the examined pupae have mites on (Longo et al. 2009b), thus weakening the hypothesis that mites prefer to attack old individuals.

This paper shows the widespread presence in Italy of the phoretic mite C. almerodai, an alien species that has likely tracked the R. ferrugineus diffusion. Our survival essays suggest a possible effect of the mite presence on the weevil's health status. So far Centrouropoda mites have been considered only as phoretics. However, if our results were confirmed by laboratory ad-hoc infestations, the life history of these Centrouropoda should be revisited, as has already happened in other beetle–mite associations (Houck & Cohen 1995).

Acknowledgements

We would like to thank Carlotta Cini for the English revision, Laura Aquiloni, David Baracchi, Leonardo Dapporto and Elena Tricarico for the critical reading of the manuscript. Thanks to Giacomo Santini and Filippo Frizzi for their precious support in statistical analyses. Thanks to Dilara Samancioglu for the translation of the article of Atakan et al. 2009. Thanks to Peter Mašán of the Institute of Zoology, Slovak Academy of Sciences of Bratislava (Slovakia) for identification of the mites.

This research was supported by funds of the Ministero italiano dell'Agricoltura, Project 684/7303/2008 ‘Difesa nei confronti del punteruolo rosso delle palme Rhynchophorus ferrugineus (DIPROPALM)’ and of Regione Siciliana-Assessorato Agricoltura e Foreste (FITOPALMINTRO).