Phoretic mites associated with Rhynchophorus phoenicis Fabricius (1880) (Coleoptera: Curculionidae) in the Kisangani region, D.R. Congo

We sampled the mites present on individuals of Rhynchophorus phoenicis collected at different sites in the Kisangani region of the Democratic Republic of the Congo. Two species belonging to two families (Uroactiniidae and Urodinychidae) were identified: Centrouropoda almerodai and Uroobovella phoenicola. This is the first time that C. almerodai has been reported on R. phoenicis. The number of mites per adult may reach 39 and differs between sexes and regions.


Introduction
Many species of mites are closely associated with invertebrates, colonizing any part of their bodies and sometimes in large numbers Błoszyk 2004, Bajerlein andPrzewozny 2005). All insect orders, including many families of beetles (Atakan et al. 2009, Pernek et al. 2012, take part in these associations that may correspond to different degrees of interactions such as commensalism, phoresy (Athias-Binche et al. 1993), ectoparasitoidism, parasitism or predation (El-Sharabasy 2010. In many cases, mites reduce the fitness potential of their host, even in some cases of phoresy, when the mites are particularly abundant.
Phoresy is a temporary association during which an organism, the phoront, uses its host as a means of transport between available resources (Farish and Axtell 1971). Phoresy was also defined by Athias-Binche et al. (1993) as a dispersal strategy in which one animal actively seeks out and attaches to another animal for transport (Athias-Binche et al. 1993, Szymkowiak et al. 2007). In the same way, Peter (1989) considered phoresy as a symbiotic interaction that leads to dispersal. The fitness benefit received by the phoront is thus linked to its dispersal and to the possibility of access to a new habitat (White et al. 2017). This mechanism of dispersal is usually caused by the existence of poor conditions in the initial habitat, overcrowding or lack of food and can be facultative or obligatory (Athias-Binche et al. 1993, Salmane andTelnov 2009). Normally, the phoront does not feed on its host, and in mites, it may correspond to a particular developmental stage, usually the deutonymph, which exhibits particular morphological adaptations for attaching to its host (Binns 1982, Lorber 2017.
The larvae of palm tree weevils of the Rhynchophorus genus feed on many species of palms, boring into the trunk while the adults feed on the terminal buds. These weevils are thus considered the main pests of Arecaceae but also of banana, cacao, sugarcane, and papaya (Beaudoin-Ollivier et al. 2017). On the other hand, local human populations consume the larvae of Rhynchophorus phoenicis because of their nutritional quality. This happens in several parts of central Africa, including the Democratic Republic of the Congo (DRC), where they are considered a delicacy and may contain special nutrients (Bahuchet 1975, Dounias 1993, Thies 1995, Malaisse 1997, Van Huis 2003, Monzenga Lokela et al. 2017. Adults of Rhynchophorus sp. are large insects that disperse from trunk to trunk in search of a mate or to lay eggs. They are often colonized by mites belonging to different families and may contain up to several tens of individuals under their elytra or attached to various parts of their body (Atakan et al. 2009, Mazza et al. 2011, Kontschán et al. 2012. In particular, R. phoenicis usually houses a significant number of mites at all stages of their development (Monzenga Lokela 2015).
There are very few studies on the relationship between mites and R. phoenicis. The study of Kontschan et al. (2012) is the only study to identify the mites present on R. phoenicis from samples collected in Cameroon. As R. phoenicis may provoke damage to many economic Arecaceae, a better knowledge of its associated fauna in Africa in general and in the DRC in particular is particularly worthwhile. Moreover, in the context of mass rearing of this species on artificial substrates for human nutrition, associated mites are a potential problem that must be more deeply understood (Monzenga Lokela et al. 2017).
The purpose of this study was thus to identify the mites associated with R. phoenicis recovered under the elytra of males and females and cocoons in the Kisangani area to better understand the nature of their association.

Sampling of Mites
Rynchophorus phoenicis individuals were trapped using standing-type traps (Monzenga Lokela 2015). These standing traps consist of a young wild oil palm that is pruned to keep only ten internal leaves that form the arrow. The base of the arrow receives a large V-shaped wound that results in folding of the arrow, thus creating a shelter for the insects. The rest of the arrow is cut 1.3 m from its base. After the first 5 days, the wounds were refreshed with a machete to continue the release of volatile substances that attract insects.
During 2012, trapping was carried out at three different sites: Tshopo (00°34 ′ N 25°08 ′ E and at an altitude of 450 m), Lubunga (00°41 ′ N 25°15 ′ E and at an altitude of 438 m), and Bangboka (00°50 ′ N 25°29 ′ E at an altitude of 413 m). At each site, 10 traps were visited every day to check for the presence of adult Rhynchophorus. Collecting was performed by unfolding the arrow to recover the adults. Adults were then stored in aerated plastic boxes and brought back to the laboratory for identification.
One hundred beetles per site were chosen at random among sampled individuals (50 of each sex). Mites were then collected from each beetle using a moistened fine brush and transferred to an Eppendorf tube containing alcohol (86%). Drs. Durmus Ali Bal (Turkey) and Hans Klompen (U.S.A.) kindly performed specimen identification. Mites were slide-mounted in Hoyer's solution and voucher specimens were deposited in the collection of the Ohio State University Acarology Laboratory (OSAL), Columbus, Ohio by Dr. Hans Klompen.

Statistical analysis
A generalized linear model (Poisson distribution) was performed to test the potential influence of the site and of the host sex on the number of mites found on an individual. Statistical analyses were performed using R version 2.14.1 Copyright (C) 2011 The R Foundation for Statistical Computing for Mac. All tests were applied under two-tailed hypotheses, and the significance level, P, was set at 0.05.

Discussion
A summary of the available information of the mite species present on Rhynchophorus spp. is presented in Table 1. In total, twelve species of mites have been identified on Rhynchophorus spp., ten on R. ferrugineus and only two on R. phoenicis. This illustrates that there are few studies on these species. Among the mite species, only Hypoaspis sardoa, Berlese is recognized as a true parasite.
Centrouropoda almerodai was described in 1992 by Wisniewski et al. in the Philippines on Rhynchophorus spp. This mite is generally regarded as a phoretic mite and has accidentally been introduced into the island of Malta on R. ferrugineus (Ragusa et al. 2009). Moreover, Mazza et al. (2011) demonstrated that, in Italy, the individuals of R. ferrugineus associated with this species have life spans shortened by a third compared to those exempts of mites, suggesting that the interaction between these two species may not be limited to phoresy.

R. ferrugineus
Uroobovella phoenicola was described in 2012 by Kontschan et al. (2012) from a sample of mites collected in Cameroon. Kontschan et al. (2012) suggested that it might be a predator of weevil eggs and early instar larvae and proposed its use for the biological management of the beetle. Eggs, larvae and adults of these mites heve been found under the elytra and on the abdomen of the weevil (Monzenga Lokela, 2015). This distribution of stages is hardly compatible with only phoresy. Indeed, after emergence, the larvae must grow and therefore feed. It is possible that they feed directly on the insect; in this case, the mites would use Rhynchophorus individuals as a means of transport, as a shelter and as a food source. The mites could then be considered ectoparasites of the beetles. These observations could confirm the survival reduction of R. ferrugineus in the presence of parasitic mites (Mazza et al. 2011). Furthermore, as observed by El-Sharabasy (2010), mites can be phoretic for a given stage of the insect and become predators or parasites for other stages of the same insect. Therefore, phoretic mites may represent a cost for their hosts, as they may increase the energy balance of the host, disrupt its aerodynamics in fixing their colonies under the elytra, or cause an extra weight that diminishes its velocity (Athias-Binche et al., 1993). Other studies have also identified C. almerodai on R. ferrugineus, without being able to formally determine whether these mites are phoretic or not (Wattanapongsiri 1966, Atakan et al. 2009). This point has to be investigated in the future as well as the life cycle of the mite outside of its host.
We also observed a higher number of mites on females than on males, which confirms the results of Atakan et al. (2009). As there is no difference in size between males and females of R. phoenicis (Monzenga Lokela 2015), it is probably not their morphology but rather a difference in the behaviour or pheromone emissions between the two sexes that make the female a preferential host for the mite. Indeed, although the way a dispersing mite finds a new host is still unknown, the female could represent a good opportunity for such a mite to find a new host. The fact that R. phoenicis females remain slightly longer on palm trees than males to lay their eggs makes them easier to find than males. Moreover, these mites being considered as egg predators of the weevils (Kontshan et al., 2012), infecting a female could then increase their opportunities to find a prey. However, these hypotheses remain to be verified.

Conclusions
The objective of this work was to identify the mites present on R. phoenicis in the Kisangani region and to understand the nature of their association. Centrouropoda almerodai (Wisniewski et al. 1992) and Uroobovella phoenicola (Kontschan et al. 2012) belonging to families Uropodidae and Urodinychidae respectively were identified. To our knowledge, C. almerodai has never been reported on R. phoenicis before our study. The role of this species remains to be determined. It is possible that in addition to pure phoresy, C. almerodai also has a negative impact on the host.
Finally, in the case of mass rearing of R. phoenicis to use the larvae as a food source, the presence of mites on the adult stage poses many difficulties for the production of these insects and could affect the final quality of the proposed product.