**5. A case study of genetic and behavioral evaluation of Thai firefly species,** *Sclerotia aquatilis*

## **5.1 Background**

*Sclerotia aquatilis* (*L. aquatilis*) [70] is an aquatic firefly species. Individuals are commonly found in freshwater habitats throughout Thailand, i.e., ponds, ditches, wetlands inhabited by an abundance of aquatic snails and aquatic vegetation such as duck weed, water lettuce, water hyacinth, *Typha* spp., water lily, and Indian lotus. It is a multivotine species appearing all year round with the life cycle duration of 3–5 months [71], **Figure 1**. The larvae live in the water by respiring mainly through a pair of caudal spiracles to receive the air from water surface. They are frequently found back swimming at the surface of water.

The species has high potential for reintroduction programs because of the successful rearing technique developed [15, 16] and their several adaptive characteristics that support recovery of the new populations in old/new habitats. Since *S. aquatilis* occurs throughout Thailand, the reintroduction programs are probably

**Figure 1.** *Life cycle of S. aquatilis.*

*Firefly Translocation: A Case Study of Genetic and Behavioral Evaluation in Thailand DOI: http://dx.doi.org/10.5772/intechopen.97455*

applied widely in the country. The firefly translocation has not previously been reported for this species.

There are many reasons suggesting genetic differentiation among *S. aquatilis* populations could lead to negative impact on translocation programs. Although geographic isolation frequently results in reproductive isolation by limiting gene flow between populations, it still remains unknown for firefly populations in Thailand. The expansion of cities and associated infrastructures not only destroy firefly habitats, but also creates habitat fragmentation. *S. aquatilis* populations are restricted to freshwater habitats, i.e., ponds, wetlands, and ditches. Adult female fireflies lack strong flight ability; therefore, habitat fragmentation seriously limits the range of their dispersal efforts, resulting in little immigration and even local extinctions. These limiting dispersal factors cause an increased the level of inbreeding and minimize interbreeding among spatially isolated populations. Thus, the probability of inbreeding and low genetic variability in nature is high in fragmented habitats. There is evidence of loss of genetic variation and the extinction of populations from habitat fragmentation in a butterfly metapopulation [72]. In addition, most *S. aquatilis* habitats overlap with human-used areas such as residential and agricultural areas, fireflies are subjected to many negative impacts from human urbanization, especially light pollution that can interfere with the sexual communication signals. Moreover, light pollution can be an effective dispersal barrier of fireflies. All these factors might result in both decreasing numbers and promoting inbreeding effects in populations.

### **5.2 Materials and methods**

### *5.2.1 Study areas*

During the process of urbanization, habitat loss and fragmentation have subsequently expanded particularly in Bangkok (BKK) area, where is the focus area for firefly reintroduction in this study. Historically, *S. aquatilis* inhabited in high abundance in the agricultural diches and ponds in the Chao Phraya delta area. However, the recent populations of the species have been decreased and become rare. The sources of translocated populations were from four nearby provinces, Samut Prakarn (SPK), Pathum Thani (PTE), Nakhon Pathom (NPT), and Suphan Buri (SPB) (**Figures 2** and **3**). Seven populations of fireflies from five locations were collected. One population from each province but two subpopulations from Pathum Thani (PTE2) and Nakhon Pathom (NPT2).

#### *5.2.2 Firefly collection and maintenance*

The collection of *S. aquatilis* specimens was conducted in all five locations during firefly season from August to November in 2012–2013, which was during the end of the raining season and the beginning of winter. The adult fireflies were collected at nighttime using a sweep net over freshwater areas. Adults were maintained in insect rearing cages supplied with a 10% honey solution on balls of moist cotton. In case of small populations, aquatic firefly larvae were also collected for molecular work. After observing the flashing behavior, the firefly specimens were placed in vials containing 100% ethanol, and stored in a − 80°C freezer prior the molecular study.

#### *5.2.3 Genetic analysis*

Genomic DNA from the hind legs of the adult specimens was extracted following the manufacturer's protocol using the DNeasy Blood & Tissue

**Figure 2.** *Map of Thailand the S. aquatilis study sites. The map illustration was modified from Vemaps.com.*

Kit (Qiagen). A region encoding mitochondrial cytochrome c oxidase subunit II (COII) was amplified by the polymerase chain reaction (PCR) using the primers 5′-ATGGCAGATTAGTGCAATGG-3′ (TL2-J-3037) and 5′-GTTTAAGAGACCAGTACTTG-3′ (TK-N-3785) [69]. The PCR amplifications were performed as follows: an initial denaturing step at 94°C for 1 min, followed by 35 cycles beginning with a denaturation step at 94°C for 30 sec, an annealing step at 50°C for 30 sec, an extension step at 72°C for 1 minute, and a final step at 72°C for 10 min. The PCR product was verified by running through a 1% TAE agarose gel, stained with ethidium bromide and observed under UV light. The PCR product was treated with ExoSAP-IT PCR clean up reagent (Thermo Fisher Scientific, Massachusetts, USA) and sequenced by the 3130xl Genetic Analyzer (Thermo Fisher Scientific) with the BigDye Terminator v3.1 Cycle Sequencing kit (Thermo Fisher Scientific). The nucleotide sequences were assembled and edited individually using DNASIS Pro (Hitachi Software Engineering, Tokyo, Japan).

The numbers of base differences per site among sequences (p-distance) were calculated and constructed Unweighted Pair Group Method using arithmetic Average (UPGMA) tree using the p-distance by Molecular Evolutionary Genetics Analysis software (MEGA X) [73].

Median-joining networks among firefly haplotypes were constructed and post-processed under maximum parsimony in Network Version 4.6.1.1 (available at http://fluxus-engineering.com/sharenet.htm) to describe phylogeographic and genetic relationships between haplotypes.

#### *5.2.4 Flashing behavior analysis*

The live adult fireflies from each population were brought to the laboratory (26°C) for recording flash patterns within 1–2 days after collection to decrease the error from weakness and death. They were paired 1: 1 for mating in a mating arena that was prepared from a 7.1 × 11.0 × 6.5 cm of transparent plastic box with small moist cotton. They were allowed to have an adaptation period for 15–30 min before *Firefly Translocation: A Case Study of Genetic and Behavioral Evaluation in Thailand DOI: http://dx.doi.org/10.5772/intechopen.97455*

**Figure 3.** *Habitat characteristics of the firefly collection sites, a) SPK, b) PTE, c) PTE2, d) NPT, e) NPT2 and f) SPB.*

starting the experiment. The experiment was carried out under dark conditions (0 lux) for 30 min to 2 hr. after sunset.

The flashing communication was recorded using a Sony Handycam™ digital camera recorder (HDR-SR11E) at nightshot mode. All experimental mating boxes were separated from one another by placing black partitions between each arena to prevent flash interference from other mating pairs. Ten to 15 mating pairs from each population were randomly selected for video recording. Two flash types, courtship and warning flash types (**Figure 4**), which appeared at different periods of mating sequences, were recorded. The "courtship flashes" produced during courtship in responding to females, perhaps displayed during dorsal mounting. On the other hand, the brighter flashes displayed mostly during copulation called were defines as "warning flashes." At least 15 sec intervals or 30–50 flashes were recorded from each male. In case of small populations that had low numbers of females, the males were allowed to mate with virgin captive females to stimulate courtship behavior.

**Figure 4.**

*Flashing behavior of male fireflies, courtship flash type (upper) and warning flash type (lower).*

The video files were converted to audio video interleave or. AVI format files to analyze the flash parameters using time-lapse image analysis (TiLIA), a free software package for signal and flight pattern analyses of fireflies (available at Google Drive: https://drive.google.com/open?id=0B2o7FRVs2VohMmx2QzBVX3ZD eDA) [74] following the technique used by Thancharoen and Masoh [75]. The flash analysis was classified into three parameters, pulse duration, interpulse duration and flash interval, following previous study [76].

#### *5.2.5 Statistical analysis*

At least 30 flashes of courtship and warning flashes from each male were statistically analyzed. The pulse duration, interpulse duration, and flash interval among study sites were compared using One-way ANOVA and Tukey's multiple comparison tests. A value of p < 0.05 was considered statistically significant. The relationship between pulse and interpulse durations was tested using Pearson's correlation. All statistical analysis was performed using SPSS program version 24.

#### **5.3 Results**

#### *5.3.1 Flashing behavior analysis*

During mating behavior of *S. aquatilis*, the pulse durations of both courtship and warning flash types were quite similar, whereas the interpulse duration of warning flashes were twice longer than courtship flashes (**Table 2**). The correlation analysis of interpulse duration and pulse duration in each population showed that both flash parameters were negatively correlated (r in the range of −0.767 to −0.329, *P* < 0.05, n = 13). In case of short pulse duration, the interpulse duration was observed to be prolonged, stabilizing the flash interval.

The comparison of courtship flash parameters of all seven populations from five provinces showed that the fireflies from Suphan Buri province displayed different


*Firefly Translocation: A Case Study of Genetic and Behavioral Evaluation in Thailand DOI: http://dx.doi.org/10.5772/intechopen.97455*

**Table 2.**

*Flash parameters of courtship and warning flash types of* S. aquatilis *(from overall populations).*

#### **Figure 5.**

*The comparison of courtship flash parameters among seven populations of S. aquatilis; different letters indicate significant differences among different populations. Samut Prakarn (SPK), Bangkok (BKK), Pathum Thani (PTE), Nakhon Pathom (NPT), and Suphan Buri (SPB).*

courtship flashes from the other sites located in the Bangkok Metropolitan Region (Samut Prakarn, Pathum Thani, Nakhon Pathom and Bangkok) (One-way ANOVA, P < 0.05; **Figure 5**). Results indicated that the Suphan Buri population had significantly longer pulse duration and flash interval resulting in slow flashing.

The flash parameters of the warning flash type could not be analyzed in all populations because not all experimental mating pairs displayed warning flashes. Therefore, only three populations from Pathum Thani, Nakhon Pathom and Suphan Buri province were analyzed. Perhaps because the mating happened under controlled environments without interference from mate competition and predation. Again, the Suphan Buri population flashed significantly differed when compared with other populations (**Figure 6**). It had a significantly long interpulse duration that resulted in having a long flash interval and a low flash frequency.

#### **Figure 6.**

*The comparison of warning flash parameters among three populations of S. aquatilis; different letters indicate significant differences among different populations. Samut Prakarn (SPK), Bangkok (BKK), Pathum Thani (PTE), Nakhon Pathom (NPT), and Suphan Buri (SPB).*

#### *5.3.2 Genetic diversity of S. aquatilis*

The genetic diversity of COII gene in *S. aquatilis* populations were examined from 132 individuals from seven locations in five provinces in the central part of Thailand. The sequences were registered in GenBank accession nos. MW800771 to MW800823 and MW814512 to MW814587. The p-distances among individuals ranged from 0 to 0.0122. The UPGMA tree revealed that regional cohesion of sequence types was not observed due to short p-distances (data not shown). The median-joining haplotype network was needed to confirm the low genetic diversity. The network revealed 37 haplotypes but not any phylogeographic sub-structuring of the firefly populations (**Figure 7**). Thus, no genetic differentiation was shown among the *S. aquatilis* populations examined.

#### **5.4 Discussion**

The study revealed flash signal variation among populations of *S. aquatilis* in the central part of Thailand. However, a distant population in Suphan Buri province apparently displayed longer pulse duration in the courtship flashes and longer interpulse in the warning flashes. As sexual communication, the pulse duration of the courtship signals is generally quite similar, preserving constant species-specific flash patterns. Most researchers studied "interflash interval" to define flash type from frequency, for instance, slow-flash, fast-flash, intermediate-flash and quick flash types [65–68]. However, the negative correlation between interpulse duration and pulse duration might help to balance the flash interval and flash frequency.

*Firefly Translocation: A Case Study of Genetic and Behavioral Evaluation in Thailand DOI: http://dx.doi.org/10.5772/intechopen.97455*

#### **Figure 7.**

*Median-joining haplotype network generated from COII data from S. aquatilis collected from four locations in Central Thailand, different colors represent different collecting locations, sizes of nodes and pie segments are proportional to haplotype frequency, and length of branches is proportional to number of mutational changes between haplotypes.*

Our finding was that there is intraspecific variation in flash communication of *S. aquatilis*. The fireflies in the Bangkok Metropolitan Region were fast-flash populations whereas the Suphan Buri population was slow-flashing although they did not show genetic differences among populations. This result is similar to the case of *L. lateralis* that *L. lateralis* populations distributed throughout the Korean Peninsula, Northeast China, Sakhalin, and Japan, the two flashing behavioral types could not be separated phylogenetically [69]. However, among populations with different flash types of *L. cruciata* in Japan, the genetic variation associated with flashing behavior was investigated [65, 67, 68]. The geographical differences caused by a great rupture zone of Japanese Islands might have had a strong

effect on this species. Similarly, as the most geographically distant location of our studied populations, the Suphan Buri population (109 kilometers from Bangkok), is probably isolated from the others. Although there are no geographical barriers influencing allopatric populations like in the Japanese case, habitat fragmentation including light pollution barriers probably significantly affect the firefly populations. *S. aquatilis* fireflies normally inhabit in or near freshwater areas, the active males can fly fast and travel a long distance, the inactive females remain near a water area. The reduced female mobility behavior might limit the dispersal ability of the species and result in population isolation. In addition, artificial night lighting could also interfere with flashes of *S. aquatilis* resulting in adaptive behavior to adjust their flashes.

The fireflies inhabiting the area of the Bangkok Metropolitan Region might face a habitat flooded with artificial light that causes reduced ability to communicate with their mates. Selection pressure favors adaptations of their flash pattern to minimize light competition or to increase the clarity of flash signals to improve their mating success. It might be possible that the environmental selection pressure happened in the fireflies. The plasticity of the flashing behavior depending on situation and environmental conditions were examined in many firefly species [75, 77, 78]. The fireflies in light polluted areas will modify their flash patterns to be faster to mitigate steady light from artificial night lighting. Similar adaptations occur in acoustically communicating animals, where ambient noise, especially anthropogenic low-frequency noise, affected acoustic communication in blackbirds [79], tree frogs [80], tree swallows [81], fish [82] and tree crickets [83]. The birds sing louder with higher frequencies to mitigate low frequency traffic noise, while the males of the tree crickets shortened their calls (echemes) and paused singing with a higher probability with increasing noise level without modification of song frequency or interecheme interval. Unfortunately, no work has been done on their genetic differences between the normal and noise polluted populations.

#### **5.5 Recommendations**

Generally, genetic differentiation among populations would happen in a heterogenous or mosaic environment by reduction of population size, genetic drift, gene flow and natural selection and accumulated by geographic isolation. Although there is no geographical isolation in the central region of Thailand, in case of *S. aquatilis*, gene flow is limited by the dispersal ability of adult females and aquatic larvae that are restricted to the aquatic ecosystems. In addition, the light pollution is likely an important barrier limiting the adult dispersal whereas habitat fragmentation reduces population sizes, reduces habitat size of firefly larvae and increases isolation of small subpopulations. The wild populations of the fireflies are at risk of extinction due to the effect of inbreeding depression.

The recommendation for *S. aquatilis* translocation is to consider: (i) no genetic differentiation between the local and the released populations, (ii) no divergence in flash signals to prevent pre-mating isolation between recipient and donor populations, (iii) the distance between populations might promote variation among populations; thus, closer populations are properly used for translocation, (iv) the sources of translocated populations come from a large population or several subpopulations to acquire proper numbers of source populations and decrease the effect of inbreeding depression. In addition, other factors, for example, habitat quality, source of translocated fireflies (from wild or captivity), released stage, frequency of releasing, released area and other environmental conditions during releasing, can relate to the success of program. This information is probably species specific; therefore, the biological and ecological characteristics of the focus species

#### *Firefly Translocation: A Case Study of Genetic and Behavioral Evaluation in Thailand DOI: http://dx.doi.org/10.5772/intechopen.97455*

are needed for translocation application. Significantly, the long-term monitoring of establish populations also is necessary.

In the case study, although the *S. aquatilis* populations in the central part of Thailand have no genetic divergence among populations, the variation of flash signals was found in a location of Suphan Buri province. The translocation of the species could happen if the donor and recipient populations come from Bangkok Metropolitan Region where the fireflies displayed similar flash signals and no genetic divergence among populations.
