BEETLE EVOLUTION

Paleontological evidence presents the first insects as a group of small, scurrying hexapods in the earliest Devonian period. DNA studies have estimated that insects originated 434 MYA (in the Early Silurian) and winged insects (pterygotes) originated 387 MYA ago (in the mid-Devonian). It is during the Early Permian (280 MYA) that beetles arose diverging from their common ancestors the Neuropterida and possibly the Glosselytrodea. These protocoleoptera were among the first holometabolous insects and by the Late Triassic, 240-220 MYA, twenty families had evolved including what are considered true beetles, these exhibiting hardened, veinless wing cases (elytra) (Ponomorenko, 2002).

Four major suborders of beetles exist today, the relationship between which is still contested. The Archostemata comprise the specialised wood borers, the Myxophaga are a small group of just 65 aquatic or semi-aquatic species, the Adephaga include approximately 10% of all beetles and include predatory ground and aquatic beetles and the Polyphaga that comprise 90% of all beetle species and as the name suggests have extensively varied diets.
All of the six bioluminescent beetle families belong to the suborder Polyphaga.

The suborder is composed of a number of superfamilies and the superfamily Cantharoidea was revised by Crowson to contain all bioluminescent beetle families, with the exception of Elateridae and Staphilinidae. However the Cantharoidea were later combined into the Elateroidea when Elateriformia was redefined. With the exception of the dubious Staphilinidae, the series Elateriformia contain all bioluminescent Coleoptera. However, despite this classification Crowson concluded that nothing indicated a particular link between the Cantharoidea and the bioluminescent members of the Elateridae and that it was unlikely that the luminescence of elaterid members and of various cantharoids derived from a common ancestor.

The oldest fossil records of cantharoids are from the Eocene (Cantharidae and Lycidae). Much younger are the lampyrids where definitive specimens have been found preserved in amber from the Dominican Republic, 20 MYA (Lawrence & Newton, 1982). The evolution of bioluminescence in cantharids is constantly debated. Several studies have presented hypotheses concerning the evolutionary relationships within and around the Cantharoidea. These different phylogenies, when aligned with luminescent members, provide a range of evolutionary scenarios ranging from three origins of luminescence through to one origin and three losses.

Evolutionary relationships within and around the Cantharoidea. Plotting luminescence onto different trees supports interpretations ranging from three origins A to one origin and three loses D. Non-bioluminescent lineages are shown in grey and bioluminescent lineages in black. Dark boxes denote an emergence and white boxes a loss of bioluminescence. A striped box and branch indicates the emergence of a non-cantharoid bioluminescent lineage. Cantharoidea families are shown in upper case and bioluminescent Cantharoidea families in bold. A and B are derived from a dendrogram of nine cantharoid families by Crowson (1972). C is a dendrogram derived from a condensed strict consensus tree based upon seventy-four morphological states in eighty-five exemplary taxa by Bramham and Wenzel (2001). D is based upon the majority rule consensus tree of Beutel (1995).

A number of adaptive functions have been proposed for luminosity in cantharoids, the most substantiated are aposematism in larvae and mate attraction in the adults. This leads to the question as to whether bioluminescence arose first in adults or in larvae as each seem to present an adaptive function. Based upon the fact the Omalisidae and some Phengodidae are luminous as larvae but not as adults and that conversely there were no established instances of luminous adult cantharoids lacking luminous larvae, Crowson first tentatively proposed that bioluminescence arose in the larvae (Crowson, 1972). It is fair to assume that if bioluminescence first arose in larvae then the Omalisidae, which contains no members with luminous adults, should be the most basal family of the Cantharoidea. According to Branham and Wenzel the next basal family after Plastoceridae and Drilidae, both of which are no bioluminescent families, is the Omalisidae which they consider the origin or ancestral lineage that has resulted in all the bioluminescent cantharoids (Branham & Wenzel, 2001).

Evolution of the bioluminescent mechanism

The emergence of bioluminescence in beetles is dependent upon two fundamental components being present within the beetle, the enzyme luciferase and the substrate luciferin. The presence of these in conjunction with other common components within living organisms would have provided the first light in the proto-bioluminescent beetle. It is possible to envisage a low light emission which is diffuse and not localised within the particular cells or organs. Evidence for such diffuse light is seen in the images of Lampyris noctiluca larvae where light leaking out between segments. It has been proposed by a number of authors that bioluminescence would have arisen in larvae first and later sequestered by adults for communication purposes.

Evolution of luciferin

Luciferin appears to be conserved in structure between bioluminescent beetle species and even families irrespective of metamorphic stage or lantern location. No evidence has been found for beetle luciferin being present in any organism other than bioluminescent beetles. Furthermore, the levels of beetle luciferin in luminous and non-luminous beetles were recently investigated and no luciferin was detected in the non-luminous cantharoids and elaterids (Oba et al., 2008). The luciferin biosynthetic pathway for is still not established although based upon the structure of beetle luciferin and its chemical synthesis, it has been proposed that the origin of the thiazoline ring is likely to be a cysteine.

Only D-luciferin contributes to beetle bioluminescence and several researchers have noted that no light is produced from L-luciferin. However, Lembert reported that L-luciferin produced a weak emission but extremely slowly. Consequently Lembert proposed that L-luciferin was racemized to give d-luciferin. It has been recently presented that luciferase could be responsible for the stereoisomeric inversion of L-luciferin to D-luciferin thereby explaining the weak bioluminescence observed by Lembert [82]. In Luciola lateralis both D- and L-luciferins were detectable in all firefly life stages, including the egg [83]. The enantiomeric excess of D-luciferin was highest at the adult stage, while it was lower during larval and pupal stages suggesting L-luciferin is converted to D-luciferin as the beetle matures.

It is therefore plausible that L-luciferin was the first compound to be synthesised in biolumnescent beetles and either through the origination of a luciferase enzyme or a yet to be described mechanism was converted into the more efficient D-luciferin form.

Evolution of luciferase

Beetle luciferase belongs to a large family of adenylate-forming enzymes (PFAM00501). The adenylate forming proteins catalyze a two-step reaction converting an organic acid to a CoA thioester. This mode of substrate activation is commonly used by adenylate-forming enzymes such as acyl-CoA ligases, acetyl-CoA synthetases, non-ribosomal peptide synthetases (NRPSs) and aminoacyl-tRNA synthetases, as well as beetleluciferase. These enzymes are relatively large, ranging in size from 500 to 700 residues. Structurally they are composed of two domains, an N-terminal domain of 400-550 residues and a smaller C-terminal domain of 100-140 residues. An active site is situated at their interface. Members share limited sequence homology of 20-30%, however, several well-conserved sequence motifs have been identified between members and three principle motifs have been attributed with an adenylation function.

These enzymes activate a variety of different substrates, including aromatic acids, acetic acid and long-chain fatty acids, to the corresponding enzyme-bound acyl-adenylates, which are then transferred to the thiol group of CoA. It has recently been speculated that beetle luciferase may have evolved from an ancestral fatty acyl-CoA synthetase as firefly luciferase retains this activity in vitro. As such beetle luciferin may not itself have originally been the substrate for the ancestral luciferase, but rather a ‘luciferin-like’ molecule, with beetle luciferin appearing as a substrate later in evolution In support of this, dehydroluciferin, differing from luciferin by only two hydrogen atoms and inactive for chemiluminescence, can be efficiently ligated to CoA by firefly luciferase. Luciferase may still function as a fatty acyl-CoA synthetase involved in the oxidation of fatty acids in the peroxisome of beetles. Interestingly, studies have shown that firefly luciferase has a marked preference for fatty acids such as arachidonic acid. This may be unsurprising as arachidonic acid, although typically occurring in very small amounts in the phospholipids of terrestrial insects, has been found in very high levels in the tissue lipids of the adult firefly.