Two Novel Volatile Compounds as the Key for Intraspecific Colony Recognition in Macrotermes gilvus (Isoptera: Termitidae)

MATERIALS AND METHODS

Source colonies and maintenance in the laboratory: Three colonies of termites M. gilvus obtained from three different locations: Bandung, Bogor and Bandar Lampung, in July, 2014. Distance Bogor-Bandar Lampung is 156 km, Bandung-Bogor 164 km and Bandung-Bandar Lampung 210 km. Three nests of three colonies were moved and maintained to the laboratory for one week before bioassay. Nest consists some of combs and the royal chambers in which there are both caste workers, soldiers (minor soldier and major soldier) and reproductive (queen, king and alates). Maintenance termites using sealed plastic containers (volume 5.6 L), which filled moistened sand on the bottom and tissue (Tessa TP-01) as, a food source for termites. Maintenance is performed in the incubation chamber at temperature 28±2°C and humidity >85%.

Intra and intercolony agonistic test: Individuals that are used to test agonistic are major workers, minor soldiers and major soldiers from those three colonies. The testing arena, modified by Wong and Lee (2010) and Neoh et al. (2012), using a filter paper disc (diameter 5 cm) moistened with distilled water was placed in a 5-cm-diameter plastic dish. The arena was equally divided into two areas by a plastic card. Termites in the ratio 1: 1 (five individuals from each colony) was tested with a combination of workers versus workers, workers versus minor soldiers, workers versus major soldiers, minor soldiers versus minor soldiers and minor soldiers versus major soldiers. Termites from one colony in a pair of workers versus workers, minor soldiers versus minor soldiers and major soldiers versus major soldiers were marked with a dot of white paint on the head capsule to distinguish them from the same individuals of the other colony. Termites were acclimatized in the test arena for 5 min before the separating plastic card was removed. The behaviors were video recorded during encounters (Ricoh WG-4) for 5 min. The videos were reviewed and analyzed for agonistic reactions. The agression behaviors were leveled from 1-3 based on the definition described by Jmhasly and Leuthold (1999) with slight modifications of Neoh et al. (2012): (1) examination or antennation: contact between antenna, head or body (it means no agonistic behavior), (2) alarm reaction or avoidance, jerking, chasing and/or being chased (it means no agonistic behavior) and (3) agression: mandibles open, seizing or biting (it means no agonistic behavior). The agression behaviors that occur during interactions for each 10 sec intervals were recorded and a total of 30 observations for each test was made to scale and averaged from five replicates were there.

Chemical analysis of volatile compounds: For volatile sampling used live animals, such as; the method of Matsuura et al. (2010) and Himuro et al. (2011). A total of 20 individuals of major workers each colonies from Bogor (code W_B), Bandung (code W_D) and Bandar Lampung (code W_L), then 20 individuals of minor soldiers of Bandung's colony (code MS_D) and Bandar Lampung's colony (code MS_L), also 10 individuals of major soldiers from Bogor (code JS_B) and Bandar Lampung (code JS_L), those seven of the samples were washed for 15 min in 100 mL of n-hexane. Then 1 μL sample leaching results were analyzed using gas chromatography. Gas Chromatography-Mass Spectrophotometer (GC-MS) analyzes of volatiles were performed on Hewlett Packard Agilent 5890, equipped with a capillary column HP-1 column (30 m long×250 μm phase thickness×0.25 μm diameter column). Column temperature was raised from 40°C (5 min hold) at 10°C min^–1 to 280°C. The injection was splitless, with helium as the carrier gas (1.5 mL min^–1). The MS data was obtained under the following conditions: ionization current, 100 μA; ionization energy, 70 eV; accelerating voltage, 1470.6V; scan range of 40-450 m/z. The gas chromatographic and mass spectrophotometer system were both controlled with a data processing system. Compounds were identified from comparisons of mass spectra with a National Institute of Standards and Technology (NIST) data library system.

Data analyses: Quantitative behavioral test with total 45 combination, i.e., 18 intracolony combinations and 27 intercolony combinations (Table 1), was executed in July, 2014. The agression behaviors that occur during interactions, such as; examination and antennation (level 1), alarm reaction and avoidance (level 2) and agression (level 3) were observed. If the behavior on each level occurred ≥80% in any combination then it is expressed as, + means existing and if, this is only ≤20% in any combination then expressed as means no existing.

RESULTS AND DISCUSSION

In general, no agonistic happened intraspecies M. gilvus. Caste differences both inter- and intracolony not indicated agonistic behavior, as well as the distance (ranged from 156-210 km) was not significantly to describe about intraspecific on M. gilvus. Combination treatment between workers versus workers or workers versus minor soldiers, or workers versus major soldiers, or minor soldier versus minor soldiers and minor soldier versus major soldier did not show any agonistic behavior in the species M. gilvus. M. gilvus termite colony from Bandung, Bogor, Bandar Lampung with the closest distance 0 km (intracolony) and the longest 210 km (intercolony) did not show any agonistic behavior. No agonistic behavior in M. gilvus presented in Table 1.

The scale of agonistic behavior only at the level of alarm/avoidance in the different colonies distance and at the level of examination/antennation on intracolony while the level of agression, which is a sign of agonistic never occurred in all treatment combinations both intra- and intercolony in intraspecies M. gilvus. This is consistent with the results of research Neoh et al. (2012) that a combination of worker versus worker intra- and intercolony showed very low intraspecific aggression on M. gilvus, whereas in M. carbonarius showed little agression towards non-nestmates. According Kaib et al. (2004), colony recognition in Macrotermes intraspecies not influenced by morphology, genetics, geography or distance significantly. This is also corroborated by Jmhasly and Leuthold (1999) that aggressive behavior and an examination on major worker M. subhyalinus and M. bellicosus not correlated with spatial distance intercolony, also not correlated with differences in the volume of the nest and colony age involved. Shelton and Grace (1997) also stated, the same thing that the geographic separation of termite colonies was not a factor in predicting agonism on Coptotermes formosanus. Low agonistic response not only happened in intraspecies Macrotermes spp. but also occurs on Reticulitermes spp. (Grace, 1996; Delphia et al., 2003; Harahap et al., 2005), on Microcerotermes crassus (Wong and Lee, 2010) and on Coptotermes spp. (Uchima and Grace, 2009; Shelton and Grace, 1997; Cornelius and Osbrink, 2003).

Table 1:	Behavioral existence of major workers resulting from different colony combination in Macrotermes gilvus

Of 22 volatile components were identified, it can be specified that the six main volatile components derived from termites workers M. gilvus, 11 main volatile components of minor soldier and 15 main volatile components of major soldier. Compounds detected by GC-MS analysis presented in Table 2.

There are two main volatile components are believed to play an important role in the non-agonistic behavior on M. gilvus, namely (Z)-6-octadecenoic acid and (E)-9-octadecenoic acid. Component (Z)-6-octadecenoic acid is contained in the worker caste of termites M. gilvus from all three colonies and also at major soldiers of Bogor.

Table 2:	Compounds detected by GC-MS analysis from worker, minor soldier, major soldier of M. gilvus
WB, JSB: Colony from Bogor, WD, MSD: Colony from Bandung, WL, MSL, JSL: Colony from bandar lampung, +: Identified, -: Not identified

While (E)-9-octadecenoic acid specifically found on the soldier caste both minor- and major soldiers in all colonies.

This is a new novelty and can explain about the components that play a role in nestmate and non-nestmate recognition intraspecies M. gilvus. A similar study of volatile compounds ever done Matsuura et al. (2010) and Himuro et al. (2011) on the species Reticulitermes speratus and Nasutitermes takasagoensis. Volatile compounds produced specifically queen on Reticulitermes speratus and Nasutitermes takasagoensis influence the behavior and physiology of colony members, these compounds have been identified by them i.e., n-butyl-n-butyrate and 2-methyl-1-butanol (in R. speratus) and phenylethanol (in N. takasagoensis), respectively.

Compounds (Z)-6-octadecenoic acid and (E)-9-octadecenoic acid has a formula C₁₈H₃₄O₂ with molecular weight 282.46 (Fig. 1 and 2). Both chemicals are thought to be a component of chemical messages in communication intra- and intercolony on termites M. gilvus. As explained Richard and Hunt (2013), also Trhlin and Rajchard (2011) that the chemical message is the primary mode of communication for the majority of intra-social insect species. Previous research as did by Stuart (1961), Smythe and Coppel (1966), Matsumura et al. (1968) and Tokoro et al. (1994) also began to assert that the pheromone trail markers on the various types of termites have been identified and played a major role in the introduction of the colony and kin recognition that influence the behavior of the colony in general in terms of sourcing food and defend the colony from predators. Pheromone trail marker or recruitment has been discovered in the 1960’s, when three substances has been identified successfully. The first is the (E)-6-cembrene A of Nasutitermes exitiosus (Moore, 1966). The substance is not specific to the species because that is also found in N. walker and N. graveolus (Moore, 1966) and in Trinervitermes bettonianus (McDowell and Oloo, 1984). Then (3Z,6Z,8E)-3,6,8-dodecatrienol first identified as a pheromone trail marker of Reticulitermes virginicus (Matsumura et al., 1968; Tai et al., 1969).

Fig. 1(a-b):	Mass spectra of (Z)-6-Octadecenoic acid from workers and soldiers of M. gilvus (top) and from a commercial sources (bottom)

Fig. 2(a-b):	Mass spectra of (E)-9-Octadecenoic acid from soldiers of M. gilvus (top) and from a commercial sources (bottom)

A novelty of volatile compounds in M. gilvus as an important chemical component in a chemical message intraspecies M. gilvus expected to expand further research on how to reset the definition of a colony in M. gilvus or may be in termite generally. Colony, which is known as a unique organization that does not apply to M. gilvus and perhaps more appropriately categorized as, a separate colony as supracolony, where a great distance but still occurs recognition colony members as well as, in intracolony.

CONCLUSION

Agonistic behavior is closely related to the individual chemical components issued in the colony to recognize members of intra- and intercolony intraspecific in Macrotermes gilvus. The absence of agonistic in M. gilvus both intra- and intercolony associated with volatile compounds, which are owned by the workers and soldiers. Both the main volatile compound is 6-octadecenoic acid, (Z) -and 9-octadecenoic acid, (E), respectively, which are owned by individual workers and soldiers of M. gilvus be key intra- and intercolony recognition in the same species.

ACKNOWLEDGMENT

The authors would like to thanks to the Academy of Analytical Chemistry (Politeknik AKA), Bogor. for facilitating equipments for GC-MS, also thanks to the Ministry of Religious Affairs, Republic of Indonesia for their financial support by providing the scholarship study No. Dj.I/Dt.I.IV/3/HN.01/1182/2010.