The yellow-necked drywood termite, Kalotermes flavicollis (F.), so called due to the yellow pronotum of its alates, is the only species in the genus Kalotermes known for Europe. In some Italian localities, K. flavicollis swarms can contain a small proportion of alates with a dark pronotum, but otherwise not different from the normally colored K. flavicollis. This color variation was described by G. Becker in 1955 as Kalotermes flavicollis var. fuscicollis. During collecting trips in central Italy, we found in Grosseto Marina (Tuscany) a Kalotermes population whose alates all have a very dark pronotum. Compared with K. flavicollis alates, Grosseto alates also have paler wings and smaller arolia. Grosseto soldiers have eyes smaller than those of K. flavicollis soldiers. Kalotermes sp. form Grosseto also differs morphologically from the other Kalotermes species known for the circum-Mediterranean lands. The population from Grosseto has mitochondrial DNA sequences (a partial sequence of the control region and a fragment including a portion of COI, tRNA-Leu and a portion of COII) quite different from K flavicollis (p-distance: 5.6–7.3%). Some Kalotermes populations from Tuscany and Marche, morphologically classifiable as K flavicollis, have sequences similar to those of Grosseto Kalotermes (p-distance: 0.1–1.0%). These populations are possibly hybrids between K. flavicollis and Kalotermes sp. from Grosseto. Because of its morphologic and genetic distinctive features, we describe Kalotermes sp. from Grosseto as Kalotermes italicus sp. nov.
Kalotermes flavicollis (F.) is a drywood termite distributed along most of the Mediterranean coastal areas. It is the only species in the genus Kalotermes known to occur in Europe, and was so called (flavus = yellow, collum = neck) due to the yellowish color of the pronotum of its alates (Fabricius 1793), contrasting starkly with the dark brown/black color of most of their body.
Alates with a pronotum of a darker shade of yellow, with a dark posterior margin, or even entirely dark, of the same color as the head, have been reported for some Italian localities, in the regions Veneto, Tuscany, Marche, Apulia, Campania, and Sicily (Becker 1955; Giordani Soika 1962; Springhetti 1964, 1968). Alates with a dark pronotum were found, usually in small proportions, in colonies that contained a majority of normally colored alates, and a population where all the alates had a dark pronotum has never been found (Becker 1955). Outside Italy, the occurrence of K. flavicollis alates with a dark pronotum has never been reported.
This color variation was described by Becker (1955) as Kalotermes flavicollis var. fuscicollis. The same author carried out experimental crosses, whose results were compatible with a monofactorial inheritance of the color of the pronotum, with an incomplete dominance of the “yellow” allele over the “dark” allele. In fact, most heterozygous alates had a pronotum of a somewhat darker shade of yellow compared with the homozygous “yellow” alates, or with a dark posterior margin. However, this difference was sometimes very slight or apparently nonexistent (Becker 1955). No morphological difference other than the color of the pronotum has been reported between the variant fuscicollis and the normally colored K. flavicollis.
From the genetic point of view (mitochondrial DNA, microsatellites, and Inter-SINE markers), K. flavicollis populations can be separated in 4 main groups, distributed in 4 corresponding geographical regions: i) continental France, ii) Corse and Sardinia, except the Sardinian locality Portoscuso, iii) Portoscuso, iv) central and east Mediterranean area, including peninsular Italy, Croatia, Greece, and Crete (Velonà et al. 2011) (Fig. 1). Within this latter group, genetic variability is very low (Luchetti et al. 2004; Velonà et al. 2011).
Kalotermes colonies that differ from K. flavicollis from the genetic point of view (mitochondrial DNA and non-LTR retrotransposon R2), but are morphologically classifiable as K. flavicollis, were recently found in Feniglia wildlife reserve (Grosseto province) (Ghesini et al. 2011), where colonies with typical K. flavicollis mtDNA haplotypes also exist (Velonà et al 2011).
In this study, we report the finding in Grosseto Marina (Tuscany) of a Kalotermes population whose alates all have a dark pronotum. We characterize this population from the morphological point of view, comparing it with the available morphologic data on the Kalotermes species known to occur in the circum-Mediterranean lands. In addition, we compare mitochondrial DNA sequences of Kalotermes from Grosseto with K. flavicollis sequences reported in literature and with sequences of samples collected in Tuscany and Marche, obtained in this study. Because Grosseto Kalotermes is different from K. flavicollis, K. flavicollis var. fuscicollis, and from the other species known to occur in the circum-Mediterranean lands, we describe it as Kalotermes italicus sp. nov.
MATERIALS AND METHODS
Three Kalotermes colonies with black-necked alates were collected near Grosseto (Tuscany), in Jun 2010. Colonies with yellow-necked alates, morphologically classifiable as K. flavicollis based on the description in Springhetti (1964), were collected in Portonovo and Ancona (Marche, 3 and 2 colonies respectively) and in Feniglia wildlife reserve (Tuscany, 2 colonies) in Jun-Jul 2010. Kalotermes colonies are quite small, rarely exceeding a few thousand individuals. A colony is usually confined to a single wood log, and very rarely extends over more than a few connected logs or branches. We collected samples in distant fallen tree trunks—at least 50 m from each other-so we are quite sure that we did not sample the same colony twice.
Alates were collected when they spontaneously flew from their mother colonies. Soldiers and juveniles were collected by breaking the wood they were in. The color of the pronotum and of the wings was checked in all the alates, most of which were subsequently either reintroduced in their mother colonies, kept alive in the laboratory, or used for experimental crosses that are now in progress. Samples for genetic analyses were kept in 100% ethanol. Samples for microscopy analysis were kept in 70% ethanol.
The individuals of K. flavicollis shown for comparison with Grosseto Kalotermes (Figs. 2, 3, 5, and 7) were obtained from a colony collected in Areopolis (Greece), already analyzed from the genetic point of view in previous studies (Luchetti et al. 2004; Velonà et al. 2011).
The color of the pronotum and of the wings was checked by eye on more than 150 alates from Grosseto. The observation of other morphologic characters was carried out either with a stereomicroscope or with a scanning electron microscope (Jeol JSM-5200).
Ten alates and 7 soldiers of Kalotermes from Grosseto were used for SEM observations. In order to obtain pictures of K. flavicollis to show for comparison, 8 alates and 4 soldiers from Areopolis were also observed. Samples were dehydrated in increasing concentrations of ethanol (80%, 90%, 95%, 100%), then kept in a 1:1 solution of hexamethyldisilazane/ethanol for 2–3 h, then immersed in hexamethyldisilazane 100% and air dried. Samples were mounted on aluminium supports and metalized with gold in a metalizer BIORAD SC 502 (45 s at 15 mA). The number of ommatidia forming the compound eye was counted in each of the 8 alates (an eye of each individual). Because in winged termites arolia become atrophied some days after the dispersal flight (Springhetti 1964), arolia were observed on alates that had just flown from their mother nests.
Measurements of Grosseto alates (17 females and 18 males) and soldiers (15 individuals) were taken with an ocular micrometer mounted in a stereomicroscope. For alates, the following measurements were taken: head length (from the distal margin of the clypeus to the posterior end of the head), head width, maximum eye diam, pronotum height along the median line, pronotum width, tibia length of the third right leg, forewing length (from basal suture to wing tip), maximum forewing width, hind wing length (from basal suture to wing tip), maximum hind wing width. For soldiers, the following measurements were taken: head length (from the line joining the 2 mandible insertions to the posterior end of the head), head width, pronotum height along the median line, pronotum width, tibia length in the third right leg. On the same individuals, the number of antennal segments was also counted.
Differences between measurements from female and male alates were tested with MannWhitney U test, performed with the software PAST 1.90 (Hammer et al. 2001).
Genetic analysis was performed on 3 colonies from Grosseto, 3 colonies from Portonovo (Marche), 2 colonies from Ancona (Marche), and 2 colonies from Feniglia (Tuscany). Sampling localities are shown in Fig. 1. Two individual termites were analyzed from each colony.
Total DNA was extracted from the head of each termite using the CTAB protocol (Doyle & Doyle 1987). A 910-bp fragment (including the 3′ end of the cytochrome oxidase subunit I, the tRNA-Leu, and a 5′ portion of the cytochrome oxidase subunit II) and a 297–298-bp fragment of the control region were amplified in PCR with the primers C1-J-2797 (5′-CCT CGA CGT TAT TCA GAT TAC C-3′)/TK-N-3785 (5′-GTT TAA GAG ACC AGT ACT TG-3″) and AT-KR (5′-GTG GCT ATA CCC ACT ATA AA-3′)/TM-N-193 (5′-TGG GGT ATG AAC CAG TAG C-3′), respectively. PCR was performed in a 50 µL reaction using GoTaq® Flexi DNA Polymerase kit (Promega, USA). Reaction conditions were set as follows: initial denaturation at 95 °C for 5 min; 30 cycles composed each by denaturation at 95 °C for 30 s, annealing at 48 °C for 30 s, and extension at 72 °C for 30 s; a final extension at 72 °C for 7 min. Purification and sequencing were performed by Macrogen Europe (Amsterdam, The Netherlands).
DNA sequences were aligned and analyzed with MEGA version 5 (Tamura et al. 2011). Closely related sequences were identified from GenBank using the BLAST network service (Altschul et al. 1990) at NCBI. For comparison with the sequences of Kalotermes sp. from Grosseto, sequences of Kalotermes flavicollis obtained in previous studies (Luchetti et al. 2004; Velonà et al. 2011) were drawn from GenBank and added to the alignments. Sequences of Kalotermes colonies from Portonovo and Ancona (Marche) and Feniglia (Tuscany) obtained in this study were also added to the alignments.
Phylogenetic trees were built based on a 665bp fragment of the COII gene. This fragment was chosen because it was also available for many K. flavicollis populations (Luchetti et al. 2004; Velonà et al. 2011), as well as for some other Kalotermes species from Australia (Thompson et al. 2000) and Madagascar (Monaghan et al. 2009). A COII sequence from Mastotermes darwiniensis Froggatt (Cameron et al. 2012), was used as an outgroup. A sequence from the drywood termite Neotermes insularis (Walker) was also added to the alignment as a possible outgroup. Substitution models were tested using the software jModelTest 2.2.1 (Darriba et al. 2012). According to Bayesian information criterion, the best substitution model resulted Tim2 + I + G, I = 0.4490, G = 1.3740. The maximum likelihood tree was obtained with PAUP* 4.0b10 (Swofford 2003). Bootstrap values were calculated with 100 replicates. The Bayesian tree was obtained with MrBayes 3.1.2 (Huelsenbeck et al. 2001; Ronquist & Huelsenbeck 2003). Convergence was reached after 1 million generations (average standard deviation of split frequencies < 0.01). Trees were sampled every 100 generations, and the first 2,500 trees were discarded as burnin, after graphic visualization.
The alates of Kalotermes from Grosseto had a number of antennal segments varying from 14 to 16, scape included. Asymmetry was very common: 27 out of 35 individuals (77.1%) were found that had a different number of segments in the 2 antennae (14–15, 15–16 or even 14–16). The number of ommatidia forming the compound eye ranged from 216 to 249, with a mean of 237.4 ± 10.43. No evident differences existed in the shape of the head between Kalotermes from Grosseto and K. flavicollis (Fig. 2).
The pronotum was a dark shade of brown, almost black (Fig. 3). When seen from above, the pronotum seemed to be variable in shape across individuals, being more or less narrowed at its anterior margin, but most of this variability was only apparent, depending on the degree to which the sides of the pronotum bent downwards.
Wings were unpigmented, and looked whitish when folded upon the body (Fig. 3). Only 2 individuals out of more than 150 were found that had wings nearly as dark as those of K. flavicollis. Wings of males were slightly, but significantly, longer and wider than those of females (Table 1) (Mann-Whitney test, P < 0.05). Wing venation was variable among individuals. In many individuals, wing venation was markedly different in the 2 wings of the same pair. As an example, in Fig. 4 are shown the 2 pairs of wings of the same individual. Arolia were present, but smaller than those of Kalotermes flavicollis (Fig. 5).
MEASURES (IN MM) OF ALATES AND SOLDIERS OF KALOTERMES FROM GROSSETO.
Measurements of alates are shown in Table 2. When measurements of female and male alates did not differ significantly, the 2 datasets were pooled.
Grosseto alates were not able to climb up the walls of the plastic containers they were kept in, such as boxes, jars and Petri dishes, whereas K. flavicollis alates climbed quite easily and even walked upside down under the lids (unpublished data). The flying ability of Grosseto alates seemed to be lower compared with K. flavicollis alates, but no tests were conducted to verify this impression.
The shape of the head and the shape of the mandibles of Grosseto Kalotermes soldiers did not show any evident differences compared with K. flavicollis soldiers (Fig. 6). Basal tubercles were present on the outer sides of mandibles. Most of the soldiers from Grosseto had mutilated antennae. Out of the 6 complete antennae we observed, 2 had 13, and 4 had 14 segments. Soldiers had eyes, smaller than those of K. flavicollis soldiers (Fig. 6). Measurements of soldiers are presented in Table 1.
GENEBANK ACCESSION NUMBERS OF THE SEQUENCES OBTAINED IN THIS STUDY AND SPECIES WITH CORRESPONDING MTDNA.
GenBank accession numbers of the sequences obtained in this study are shown in Table 2. Among the comparable Kalotermes spp. sequences available in the literature, obtained for European, African, and Australian species (Thompson et al. 2000; Luchetti et al. 2004; Monaghan et al. 2009; Velonà et al. 2011), those from Europe predictably showed the highest affinity with those obtained in this study. In particular, the 2 colonies from Ancona had typical K. flavicollis haplotypes, previously identified in some other Italian localities, as well as in some Greek, Croatian and Cretan localities (Luchetti et al. 2004; Velonà et al. 2011). In contrast, the colonies from Grosseto, Portonovo, and Feniglia have quite different sequences from those of K. flavicollis (see below). For the fragment COI/tRNA-Leu/COII, the 3 colonies from Grosseto all had the same haplotype (C2), whereas, for the control region partial sequence, 2 different haplotypes (R2 and R3), differing by a substitution and a 1-bp deletion, were found, the first one in 2 colonies and the second one in the remaining colony.
The 3 colonies from Portonovo and the 2 colonies from Feniglia shared the same COI/tRNALeu/COII haplotype (C3). For the control region portion, 2 haplotypes were found in Portonovo (R4 and R5), of which the first one is also found in the 2 colonies from Feniglia.
For both mtDNA fragments, the populations with the minimum p-distances from the Grosseto population were those from Tuscany and Marche (Table 3), which were morphologically classifiable as K. flavicollis, but genetically different (Ghesini et al. 2011). When considered together, the Grosseto population and those from Tuscany and Marche differed from K. flavicollis in the 5.4–7.1% of sites of the combined dataset, whereas their within-group p-distances were only 0.1–1.0%, and the within-group distances of K. flavicollis were 0.0–3.1%. Kalotermes flavicollis populations with the minimum p-distances from Grosseto popula- tion were found among those from Central and Eastern Mediterranean lands, whereas those with the maximum p-distances were those from Corsica and Sardinia.
Maximum likelihood and Bayesian phylogenetic trees were found to share the same topology (Fig. 7). Populations from Grosseto, Portonovo, and Feniglia form a well supported clade, sister to K. flavicollis. Inside this clade, populations from Portonovo and Feniglia form a well supported subclade. A sequence from Neotermes insularis was added in the alignment as a possible outgroup, but it unexpectedly turned out to be included in a clade of Kalotermes species from Madagascar.
The Grosseto population is the first European Kalotermes population ever found whose alates all have a dark pronotum. Grosseto alates have another color peculiarity, i.e., unpigmented wings, evidently paler than those of K. flavicollis alates. The light color of wings was quite a striking feature, because dark wings have been indicated as a diagnostic character for the genus Kalotermes (Krishna 1961). The wings of Grosseto alates seemed to be less effective in flying than those of K. flavicollis alates. Wing venation was very variable among individuals, as usually happens in termites (Grassé 1949), and can vary sensibly between the 2 wings of the same pair. Such asymmetries are not unusual in termites, and can be very pronounced, as effectively stated by Tillyard (1931): “We have even noted an extreme case in which right and left wings from the same individual were classified into 2 distinct genera based on venational differences!”.
Grosseto alates had 14- to 16-segmented antennae, and most individuals had a different number of segments in the 2 antennae. Kalotermes flavicollis alates usually have antennae with 17 or 18 segments, even though some authors have found a wider range of 12–20 segments (Springhetti 1964, and references therein).
Arolia are less developed in Grosseto alates than in the normally colored K. flavicollis alates. Because arolia are adhesive organs (Crosland et al. 2005), this afore mentioned difference probably explains the scant ability of Grosseto alates to climb up smooth surfaces, such as the walls of plastic or glass containers.
Grosseto soldiers did not show any important difference from those of K. flavicollis soldiers, except for the size of the eye, that in Grosseto soldiers was smaller. In K. flavicollis, soldier measurements can vary widely even among colonies of the same population (Springhetti 1964). The measurements of Grosseto soldiers tended to be smaller than those obtained for Italian K. flavicollis populations (Rossi & Springhetti 1983), even though these data are not perfectly comparable, because Rossi & Springhetti (1983) reported only mean values, and not the ranges of variation.
PER CENT P-DISTANCES AMONG THE GROSSETO POPULATION, POPULATIONS FROM TUSCANY AND MARCHE, AND THE MAIN EUROPEAN KALOTERMES GROUPS IDENTIFIED IN VELONà ET AL. (2011)1.
Kalotermes flavicollis var. fuscicollis differed from the normally colored K. flavicollis only by its dark pronotum (Becker 1955), whereas the alates from Grosseto also had paler wings and smaller arolia. Thus, we believe that Kalotermes from Grosseto is not K. flavicollis var. fuscicollis.
Besides K. flavicollis, the other Kalotermes species known for the circum-Mediterranean area are K. sinaicus Kemner, found in Israel and Egypt, K. monticola (Sjöstedt), found in Algeria, and K. maroccoensis (Sjöstedt), found in Morocco. Kalotermes maroccoensis and K. monticola have been suspected not to be true species, but instead races of K. flavicollis (Harris 1970). As far as we know, no descriptions of the alates exist in the literature for these 3 taxa and no DNA sequences are available for a comparison with Grosseto sequences. Moreover, the descriptions of K. sinaicus and K. maroccoensis are based on few soldiers (5 and 1, respectively) (Kemner 1932; Sjöstedt 1925), so the interspecific variability is inevitably not well represented in the data. The description of K. monticola is extremely concise, and the number of soldiers examined is not stated. However, based on the available data, we think it can be reasonably excluded that Grosseto Kalotermes belongs to any of these species.
In particular, K. sinaicus soldiers are described as having a peculiarly long head, about twice as long as wide, even though a soldier was found that had a shorter head (Kemner 1932), whereas the head of Grosseto soldiers is about 1.4–1.5 times as long as wide. Unlike Grosseto soldiers, K. sinaicus soldiers have no basal tubercles on the outer side of mandibles and no eyes (Kemner 1932). Kalotermes sinaicus alates have a brown pronotum and no arolia (J. Kugler & D. Simon, personal commun.). The color of the pronotum is similar to the dark brown/black color of Grosseto alates, but the latter character is different in the 2 taxa.
The soldiers of K. monticola have no basal tubercles on the outer sides of the mandibles, and shorter and thicker antennae than those of K. flavicollis soldiers (Sjöstedt 1925), whereas the Grosseto soldiers have tubercles on the outer sides of the mandibles and their antennae are not dissimilar to those of K. flavicollis soldiers.
The soldier of K. maroccoensis has a headwidth of 1.3 mm, its mandibles are as long as its head is wide, and its antennae are 15-segmented (Sjöstedt 1904), whereas Grosseto soldiers have a wider head, their head width is larger than mandible length, and their antennae are 13- or 14-segmented.
The genetic distance between Grosseto population and Kalotermes flavicollis populations is quite high (5.2–6.5% for the COI/tRNA-Leu/ COII fragment, 5.4–6.6% for the control region portion). Some Kalotermes colonies found in Tuscany and Marche, despite having alates with a yellow pronotum and being morphologically classifiable as K. flavicollis, have haplotypes similar to those found in Grosseto (p-distance: 0.1–1.0%). Together with the population from Grosseto, these populations form a well-supported clade in the COII-based phylogeny. This discordance between morphology and mtDNA could have arisen from hybridization events between Kalotermes sp. from Grosseto and K. flavicollis. Individuals of the variant fuscicollis are possibly hybrids homozygous for the “dark” allele, and their presence in variable (usually small) proportions in colonies with a majority of yellow-necked alates could be the result of different levels of introgression characterizing different populations. Further studies are needed to test these hypotheses.
Syntypes: sample IT-GRO1 (20 alates). Paratypes: samples IT-GR02 (4 soldiers), IT-GR03 (3 soldiers), IT-GR04 (10 juveniles), IT-GR05 (10 alates), IT-GR06 (5 alates). All samples from the Grosseto Marina (Tuscany, Italy), 8-VT-2010, M. Cruscanti, S. Ghesini and M. Marini. All these series are in the M. Marini collection - Dipartimento di Scienze Biologiche, Geologiche e Ambientali, University of Bologna, Italy.
Head, thorax and abdominal sclerites dark brown, femura brown, antennae, clypeus, labrum and tibiae pale brown-yellowish; wings unpigmented. Antennae of 14–16 segments, often differing on either side of head. Ommatidia: 216–249. Wing venation variable among individuals, often different in the 2 wings of the same pair. Small arolium present.
Measurements: see Table 1.
Head light brown behind, darkening towards the clypeus. Mandibles dark brown, almost black. Basal tubercles on the outer side of mandibles. Antennae of 13–14 segments. Eye spots present.
Measurements: see Table 1.
The Kalotermes italicus alate differs from the K. flavicollis alate by having a dark brown/black pronotum, unpigmented wings, and smaller arolia. It differs from K. sinaicus by having arolia. The Kalotermes italicus soldier has eyes smaller than those of the K. flavicollis soldier, while the K. sinaicus soldier has no eyes. The head of K. italicus is wider than that of K. maroccoensis, about 1.4–1.5 times as long as wide, while the K. sinaicus head is about twice as long as wide. Kalotermes italicus soldier has antennae of 13–14 segments, while K. maroccoensis soldier has 15-segmented antennae. The Kalotermes italicus soldier and the K. flavicollis soldier have basal tubercles on the outer side of mandibles that are absent in K. sinaicus and K. monticola soldiers.
From the mitochondrial DNA point of view, the sequences of the fragment COI/tRNA-Leu/ COII and of a portion of control region of K. italicus (A.N. JN 029900–02) differ from those of K. flavicollis (p-distance 5.6–6.7%). Populations with mitochondrial DNA sequences similar to those of K. italicus, but morphologically classifiable as K. flavicollis, are found in Feniglia (Tuscany) and Portonovo (Marche), and are possibly hybrids between the 2 species.
Until now, K. italicus has been found only in Grosseto Marina (southern Tuscany, Italy). The sampled area is about 200 × 100 m.
The name refers to the currently known distribution of the species, limited to an Italian locality.
We wish to thank: Marco Cruscanti for indicating the presence of “anomalous” Kalotermes colonies in Grosseto Marina and for help in collecting samples; Eugenio Sereni (Ispettore Capo of Corpo Forestale dello Stato) for help in collecting samples in Feniglia wildlife reserve; Marco Bondini for indicating the presence of Kalotermes alates in Ancona; Ariel-Leib-Leonid (Laibale) Friedman and Dany Simon of Tel Aviv University for providing information about Kalotermes sinaicus morphology. This study was funded by a Canziani donation.