Open access journal: http://periodicos.uefs.br/ojs/index.php/sociobiology ISSN: 0361-6525 Sociobiology 60(1): 65-68 (2013) The Unusual Neotenic System of the Asian Dry Wood Termite, Neotermes koshunensis (Isoptera: Kalotermitidae) Y Miyaguni1, K Sugio2, K Tsuji1 Introduction Termites are highly eusocial insects that show re- productive division of labor among morphologically distin- guishable castes (Weesner, 1969). The reproductive castes of termites are classified broadly into primary reproduc- tives, namely the queen and king, which are the dealates that found new colonies after a swarming, and the second- ary reproductives, which are recruited within an established colony (Roisin, 2000). Depending on circumstances of the emergence of secondary reproductives, they are classified into either replacement reproductives, which emerge after the death (or disappearance) of the queen or the king, or both, and supplementary reproductives, which emerge even in the presence of the primary reproductives (Roisin, 2000). On the basis of their ontogenetic origin, secondary reproduc- tives are classified into neotenics, which are reproductives raised from immature individuals (they are more specifically called nymphoids, ergatoids, or pseudergatoids, depending on their origin from nymphs, workers or pseudergates), and adultoids, which are dealate imagoes and are therefore indis- Abstract In most lower termites, colonies are headed by neotenic reproductives of both sexes after the primary reproductives (i.e., the queen and king) are lost. The pro- duction of a neotenic sexual is inhibited by the presence of a primary reproductive of the same sex. We found an exception in the caste system of the dry wood termite Neotermes koshunensis (Kalotermitidae). The neotenic caste is exclusively male. Moreover, production of male neotenics is completely inhibited not only by the presence of a king but also by the presence of a queen. Therefore, it is likely to be difficult for N. koshunensis colonies to replace their reproductive pairs. Sociobiology An international journal on social insects 1 - Kagoshima University, Kagoshima, Japan 2 - University of the Ryukyus, Japan ReSeaRch aRTIcle - TeRMITeS Article History Edited by: Og DeSousa, UFV - Brazil Received 12 October 2012 Initial acceptance 09 December 2012 Final acceptance 03 January 2013 Keywords neotenic reproductives, orphaned colony, caste differentiation, Neotermes koshunensis Corresponding author Yasushi Miyaguni Department of environmental Science and conservation Biology, United Graduate School of agricultural Sciences, Kagoshima University, Kagoshima 890-8580, Japan e-Mail: neotenic_of_termite@yahoo.com.jp tinguishable from the primary reproductives (Roisin, 2000). The neotenic castes are widespread in various termite taxa, whereas adultoids are rare (Myles, 1999). Termite society is in principle bisexual, and males and females in most species perform the same social tasks (Roisin, 2000). The neotenic castes in termites are also often bisexual, although a biased numerical sex ratio is reported in neotenics of some species (Myles, 1999). The termite genus Neotermes (Kalotermitidae) shows an interesting diversity in the neotenic caste system. In Neot- ermes connexus and Neotermes papua (Kalotermitidae) neo- tenics are found only in males (Myles & Chang, 1984; Roi- sin & Pasteels, 1991), whereas in Neotermes tectonae and Neotermes jouteli the common pattern of bisexual neotenic production is observed (Kalshoven, 1930 cited by Myles, 1999; Nagin, 1972). To understand the biological reasons for this variation in the neotenic system more species should be studied in this genus. Lüscher (1961) proposed the currently well-acknowl- edged model of sex-specific inhibition of neotenic production in termites, i.e., the emergence of male and female neoten- Y Miyaguni, K Sugio, K Tsuji - The Unusual Neotenic System of the Asian Dry Wood Termite66 ics is inhibited by the presence of the primary reproductive of the same sex. In Kalotermes flavicollis (Kalotermitidae), these inhibitory effects are lost when the transmission of anal excretions from royal reproductives to pseudergates is blocked, suggesting that there is a pheromone in the anal excretions (Lüscher, 1961; Miller, 1969). Here, we report an unusual neotenic system in the Asian dry wood termite, Neotermes koshunensis (Kaloter- mitidae); the production of male neotenics is totally inhib- ited by the presence of a queen. Although partial inhibition of male neotenic production by the presence of a queen was reported in a related species, N. jouteli (Nagin, 1972), the system found in N. koshunensis is exceptional and differs markedly from the model proposed by Lüscher (1961). Materials and Methods Termite N. koshunensis is distributed from Taiwan to Okina- wa (Ikehara, 1966). Colonies nest in dead branches of living trees and in dead trees that serve as sources of both food and shelter (i.e., the termites are of the “one-piece” type, sensu Abe, 1987); this enabled us to collect entire colonies. This species has a linear caste development pathway: all castes, including neotenics (pseudergatoids) differentiate from pseudergates (older larvae of the functional worker caste in termite species with a linear caste development pathway) af- ter molting (Roisin, 2000; Katoh et al., 2007). Effect of the presence of primary reproductives on the emer- gence of neotenics We collected 22 colonies of N. koshunensis on the main island of Okinawa during 2007–2010. Collected indi- viduals were classified into the following castes according to Katoh et al., (2007): small larvae, pseudergates, nymphs, pre-alate nymphs, alates, pre-soldiers, soldiers, the queen (we found a maximum of one individual in each colony) and the king (also a maximum of one individual in each colony). From each colony, we established subcolonies receiving two or three of the following four treatments (note that it was impossible to apply treatments i and ii, or i and iii, simulta- neously in the same colony). (i) Normal colony: 50 haphazardly sampled pseuder- gates + a queen and a king (n = 7), (ii) Queen colony: 50 haphazardly sampled pseuder- gates + a queen (n = 9), (iii) King colony: 50 haphazardly sampled pseuder- gates + a king (n = 13), (iv) Orphaned colony: 50 haphazardly sampled pseu- dergates (n = 22). Each subcolony was put into a petri dish (90 mm di- ameter × 16 mm height) with a piece of damp filter paper, which served as a source of both food and moisture. The subcolonies were then kept in the laboratory (25 ± 1 °C) in the dark for 6 weeks. The caste composition of each sub- colony was investigated every 7 days, and the filter paper was replaced at the same time. All statistical tests were per- formed using R version 2.12.2 (R Development Core Team, 2011). Results Neotenics were produced only in the orphaned sub- colonies: the presence either a queen or a king, or both, led to total inhibition of neotenic production (Fig. 1). The pro- portion of subcolonies that produced neotenics differed sig- nificantly between the orphaned and any other treatments (all pairs P < 0.001, Fisher’s exact test followed by Bonfer- roni correction). This result holds true after controlling for the possible confounding effect of colonies: in the likelihood ratio test after applying the generalized linear model (GLM) with binomial error structure (dependent variable = the pro- portion of subcolonies producing male neotenic(s), explana- tory variable = treatment and colony, link function = logit), a significant effect of treatment was detected (χ2 = 58.958, P < 0.0001), but the effect of colony was non-significant (χ2 = 17.526, P = 0.6788). Moreover, all of the neotenics that emerged in these orphaned subcolonies were males (Fig. 1). At the end of the experiment, we carefully checked the contents of all petri dishes. We found eggs in the normal and queen subcolonies, Figure 1. Effects of the presence of either a king or a queen, or both, on the production of neotenics in Neotermes koshunensis. Normal colony: subcolonies with both a queen and a king; Queen colony: subcolonies with a queen but no king; King colony: subcolonies with a king but no queen; Orphaned colony: subcolonies without reproductives. Numbers in parentheses denote sample sizes, i.e., the number of subcolonies of each type formed. Sociobiology 60(1): 65-68 (2013) 67 whereas eggs were absent in the orphaned and king sub- colonies. Immediately after the experiment, we haphazardly chose 4 of the 19 orphaned subcolonies that had produced at least one neotenic and dissected all individuals to observe their gonads. In these colonies, the numerical sex ratio in females was 25% (10 of 40), 26% (10 of 39), 36% (16 of 45) and 40% (18 of 45), but no individuals except male neoten- ics had clearly developed gonads. In 16 of the 19 orphaned subcolonies, each subcolony produced a maximum of one male neotenic during the entire experimental period. In 3 subcolonies, a maximum of two male neotenics were found at the same time, however only one individual survived after 1-2 weeks. Emergence of a neotenic was apparent by day 7 in 1 of the 19 orphaned subcolonies, by day 14 in 15 sub- colonies, and by day 21 in 3 subcolonies. Discussion The neotenic system of N. koshunensis is unusual for two reasons. First, the neotenic caste is male-specific. Second, the inhibitory effect is not sex-specific, i.e., both the king and the queen, whether present individually or to- gether, inhibit the production of male neotenics. The apparent absence of female neotenics is not ex- plained by the presence of cryptic female neotenics that had no body color change and therefore went unrecognized, be- cause no eggs were found in the orphaned and king sub- colonies, which lacked a queen. Furthermore, none of the females in the orphaned subcolonies that produced male neotenics had developed gonads. The experimental period seemed to be long enough at least for males, because the testes size (the width of a testis) of neotenics increased twice that of the testes of pseudergates during the experimental period of 6 weeks (Miyaguni, Y. unpublished data). There is a rather well-circulated description on the biology of this species by Myles (1999) asserting the exis- tence female neotenics in this species. There is a historical reason why this incorrect information was circulated in the scientific literature. Myles (1999) referred to Takuya Abe’s personal communication in which Abe wrote that in late au- tumn and winter female reproductives (neotenics) were pro- duced in this species. It was also written that female neoten- ics were not as darkly colored as males but they have mature ovaries. However, Abe’s descriptions were all based on the results of examination of gonad size of pseudergates by our coauthor Koji Sugio who was a student of Abe in those days. What Sugio’s data showed was only that female pseuder- gates in late autumn and winter had on average larger ovary size than in summer. This phenomenon is a simple reflection of higher fat storage of individuals in winter (Sugio, K. un- published data). Unlike in caste differentiation, no specific individual showed more prominent ovarian development than the others. We conclude that Abe simply misinterpreted Sugio’s data, which has since been communicated to Myles (personal communication). Indeed, in addition to the cur- rent experimental data, we have much evidence in the field and laboratory strongly indicating the true absence of female neotenics in this species (Miyaguni, Y. and Sugio, K. unpub- lished data). Male-specific neotenic systems also occur in N. connexus and N. papua (Myles & Chang, 1984; Roisin & Pasteels, 1991). However, in these two species whether both the queen and the king inhibit male neotenic production remains to be determined. Our finding that the presence of not only the king but also the queen inhibited the production of male neotenics is not explained by the model of Lüscher (1961), which as- sumes the sex specificity of pheromonal inhibition (see also Matsuura et al. 2010). However, there are a few known ex- amples that the Lüscher sex-specific inhibition does not ful- ly account for. The possible inhibition of neotenic produc- tion by the presence of a reproductive caste of the opposite sex has also been suggested in Mastotermes darwiniensis (Mastotermitidae), which has both male and female neoten- ics (Watson et al., 1975). In N. jouteli, the functional queen (female neotenic) and the functional king (male neotenic) inhibit independently the emergence of new neotenics of both sexes, although the inhibition is not as complete as in N. koshunensis (Nagin, 1972). These previous data together with our current results suggest that the production of neo- tenics in some termites is controlled by a non-sex-specific factor. Our results suggest that it is difficult for the reproduc- tive pairs of N. koshunensis colonies to be recovered when colonies lose one or both primary reproductives. One pos- sibility, which occurs in N. papua, is the recovery of repro- ductive castes through the production of both male neotenics and female adultoids in orphaned laboratory colonies (Roi- sin & Pasteels, 1991). At this stage, it is difficult to understand the adaptive significance of the production of male-only neotenics that are inhibited by the presence of a queen or king, or both. To answer this question we need more information, especially on the life history of this species and on the reproductive ability of its neotenics. Acknowledgments We thank S. Dobata, M. K. Hojo and Ed Vargo who kindly gave us invaluable advice during the course of the study. Two anonimous reviewrs provid valuable coments that improved this article. References Abe, T. (1987). Evolution of life types in termites. In: Kawano, S., Connell, J. H. & Hidaka, T.(eds.). Evolution and Coadaptation in Biotic Communities (pp. 125–148). Univer- sity of Tokyo Press. Tokyo. Y Miyaguni, K Sugio, K Tsuji - The Unusual Neotenic System of the Asian Dry Wood Termite68 Ikehara, S. (1966). Distribution of termites in Ryukyu Archi- pelago. Bulletin of Arts & Science Division, University of the Ryukyus, (Mathematics & Natural Sciences). 9: 49–178. Kalshoven, L. G. E. (1930). Bionomics of Kalotermes tec- tonae Damm. as a base for its control. Mededeel. Inst. Plan- tenziekten. 76: 1-154. Katoh, H., Matsumoto, T. & Miura, T. (2007). Alate diffe- rentiation and compound-eye development in the dry-wood termite Neotermes koshunensis (Isoptera, Kalotermitidae). Insectes Soc., 54: 11–19. doi: http://dx.doi.org /10.1007/ s00040-006-0900-y Lüscher, M. (1961). Social control of polymorphism in ter- mites. R. Entomol. Soc. London, 1: 57–67. Matsuura, K., Himuro, C., Yokoi, T., Yamamoto, Y., Vargo, E. L. & Keller, L. (2010). Identification of a pheromone regu- lating caste differentiation in termites. Proc. Natl. Acad. Sci., USA 107: 12963-12968. doi: 10.1073/pnas.1004675107 Miller, E. M. (1969). Caste differentiation in the lower ter- mites. In: Krishna, K. & Weesner, F. M. (eds.). Biology of Termites Vol. 1 (pp. 283–310). Academic Press. New York. Myles, T. G. (1999). Review of secondary reproduction in termites (Insecta: Isoptera) with comments on its role in ter- mite ecology and social evolution. Sociobiology, 33: 1–87 Myles, T. G. & Chang, F. (1984). The caste system and caste mechanisms of Neotermes connexus (Isoptera: Kalotermiti- dae). Sociobiology, 9: 163–319. Nagin, R. (1972). Caste determination in Neotermes jouteli (Banks). Insectes Soc., 19: 39–61. R Development Core Team (2011). R: A language and envi- ronment for statistical computing. R Foundation for Statisti- cal Computing, Vienna, Austria. URL http://www.R-project. org. Roisin, Y. (2000). Diversity and evolution of caste patterns. In: Abe, T., Bignell, D. E. & Higashi, M. (eds.). Termites: Evolution, Sociality, Symbioses, Ecology (pp. 95–119). Kluwer Academic Publishers. Dordrecht. Roisin, Y. & Pasteels, J. M. (1991). Sex ratio and asymmetry between the sexes in the production of replacement repro- ductives in the termite, Neotermes papua (Desneux). Ethol. Ecol. Evol., 3: 327–335. Watson, J. A. L., Metcalf, E. C. & Sewell, J. J. (1975). Preli- minary studies on the control of neotenic formation in Mas- totermes darwiniensis Froggatt (Isoptera). Insectes Soc., 22: 415–426. Weesner, F. M. (1969). External anatomy. In: Krishna, K. & Weesner, F. M. (eds.). Biology of Termites Vol. 1 (pp. 19–47). Academic Press. New York.