Laccase 2 is the phenoloxidase gene required for beetle cuticle tanning

Results

Characterization of Laccase and Tyrosinase Genes. Before the recent release of Version 1 of the Tribolium genome sequence assembly (www.hgsc.bcm.tmc.edu/projects/tribolium) and its incorporation into the Tribolium genome database (BeetleBase; www.bioinformatics.ksu.edu/BeetleBase), we had cloned and characterized full-length cDNAs for laccases (TcLac2A and TcLac2B) and tyrosinases (TcTyr1 and TcTyr2) from this species. Initially, partial cDNA fragments were obtained by PCR amplification using pupal cDNA as template and degenerate primers designed from highly conserved amino acid sequences of laccases and tyrosinases derived from several other insect species (12, 20–22). Complete sequences of cDNAs were assembled by 5′- and 3′-RACE, as described in Materials and Methods. After the release of the Tribolium genome assembly, we used the tblastn program to identify genes corresponding to these four previously characterized phenoloxidase cDNAs and another putative laccase gene, TcLac1. A partial cDNA for TcLac1 was amplified from the pupal cDNA by using gene-specific primers, and then 5′- and 3′-RACE was performed to obtain the full-length cDNA. The TcLac2 gene gave rise to two cDNA clones, TcLac2A and TcLac2B, as a result of alternative splicing (see below). No additional phenoloxidase genes were identified in the genome sequence assembly. The amino acid sequences of the three laccases and the two tyrosinases were deduced from the PCR-amplified cDNA sequences. Conceptual translations of the three laccase and two tyrosinase genes exhibited good alignments and sequence similarities with other insect laccases and tyrosinases, respectively (Figs. 6 and 7 and Tables 2 and 3, which are published as supporting information on the PNAS web site).

All three of the laccase isoforms encoded by the two Tribolium laccase genes have putative signal peptides and are presumed to be secreted proteins with multiple glycosylation sites. A comparison of the sequences of TcLac2A and TcLac2B cDNAs and of the TcLac2 locus revealed the intron–exon organization of this gene (Fig. 1). The genomic sequence encoding the C terminus of TcLac2 consists of two sets of alternative exons with three exons in each set, which results in the production of two alternatively spliced transcripts corresponding to the two laccase 2 cDNAs that we had isolated. These transcripts encode proteins of 717 and 712 aa (Figs. 1 and 6). The amino acid sequence identity is 74% in the variable C-terminal regions of these two isoforms (266 and 261 aa for TcLac2A and TcLac2B, respectively). The gene for the other laccase, TcLac1, has no alternative exons and the encoded protein shares only ≈35% amino acid sequence identity with either of the laccases encoded by TcLac2. A search of the genomic databases of Drosophila melanogaster and A. gambiae identified TcLac2 orthologs in both species. These orthologs have, respectively, three and two sets of alternative exons encoded in C-terminal regions corresponding to the alternatively spliced region of TcLac2 [see also Dittmer et al. (12)]. However, the sizes of the Drosophila and Anopheles alternative exons do not correspond well with those of TcLac2, and the arrangement of exons is more complex in the two dipteran species (data not shown).

The two tyrosinase genes of Tribolium identified by tblastn analysis of the Tribolium genome sequence assembly correspond to the two cDNA clones that we had amplified by RT-PCR. These genes encode proteins of 684 and 683 aa and have an amino acid sequence identity of 71% (Fig. 7 and Table 3). There are no alternative exons in these genes. The amino acid sequence identity of TcTyr1 and TcTyr2 to other insect tyrosinase-like proteins is ≈50%.

Expression of Laccase and Tyrosinase Transcripts During Development. To determine the proper timing of dsRNA injections for effective disruption of the target gene function, the developmental patterns of expression of the two tyrosinase genes and TcLac1 and the two alternatively spliced transcripts of the laccase 2 gene, namely TcLac2A and TcLac2B, were determined by RT-PCR analysis of RNA prepared from prepupal, pupal, and adult stages (Fig. 2A). The highest levels of TcLac2 transcripts were detected in prepupae and pharate adults just before the pupal and adult molts, respectively, whereas TcLac1 transcript levels were the highest in young pupae and pharate adults and the lowest in the interval between these two stages. TcLac2A transcripts were more abundant than TcLac2B transcripts at almost all time points examined, but the relative amount of each alternatively spliced transcript varied during development. In contrast, transcripts of the tyrosinases were abundant during early pupal development and were essentially undetectable in pharate adults and in adults. These differences in temporal patterns of expression during development suggest that the two types of phenoloxidases have different, specialized functions. The timing of expression of TcLac2 just before the initiation of the cuticle tanning process and the strong expression of the orthologous laccase 2 in M. sexta pharate pupal epidermis and its reduced expression in young pupal epidermis (12, 23) are observations consistent with the presumed role of laccases 2A and/or 2B in cuticle tanning.

dsRNA-Mediated Depletion of Laccase and Tyrosinase Transcripts. To obtain direct evidence for the involvement of specific phenoloxidases in the tanning process, we used dsRNA-mediated transcript depletion, which has been demonstrated to be very efficient for other genes in Tribolium (13, 14). dsRNAs were injected into dorsal abdomens at least 3 d before the time of maximal accumulation of the targeted transcript. Administration of dsRNA for TcTyr1 at the late larval stage knocked down transcripts for both tyrosinase genes, TcTyr1 and TcTyr2, at the early pupal stage, as shown by RT-PCR analysis (Fig. 2B). The knockdown of both tyrosinase transcripts by dsRNA for TcTyr1 may be attributed to short regions of high sequence similarity between these two genes. In contrast, injection of dsRNA for TcLac1 reduced the levels of its transcripts substantially without affecting those of TcLac2 (Fig. 2B). dsRNA for TcLac2 was similarly selective in its action, and the levels of TcLac1 transcripts were unaffected after injection of dsTcLac2. Both of the alternatively spliced forms of TcLac2 were down-regulated by injection of this exon-nonspecific dsRNA, because of shared identical sequences in these transcripts. In no case did dsRNA injections result in detectable changes in levels of polyubiquitin transcripts that were used as a control for equal sample loading and for monitoring of nonspecific effects of dsRNA.

RNAi Phenotypes. The phenotypes of the animals obtained from these experiments are shown in Fig. 3A. In beetles injected with buffer alone, tanning was evident by day 5 in the pupal and pharate adult cuticles and in <1-day-old adult cuticle. The adult cuticle tanning process was completed after 3 d, not only in buffer-injected controls but also in animals injected with dsRNA (≈0.2 μg per insect) for TcTyr, demonstrating that down-regulation of the TcTyr1 and TcTyr2 transcripts had no effect on cuticle tanning. Similarly, insects treated with dsRNA for TcLac1 also showed normal pupal and adult cuticle tanning (Fig. 3A). We did not observe any lethality from injection of dsTcTyr,dsTcLac1, or buffer. In contrast, prepupae injected with dsRNA for TcLac2 exhibited little or no tanning on the last day of the pupal stage and relatively little tanning of the pharate adult or adult cuticle. These adults had soft colorless cuticles and abnormally enlarged bodies with elytra that were deformed and leg sockets that were oversized. All of the beetles treated with dsRNA for TcLac2 (n = 40) died within several days after adult eclosion.

In addition to monitoring adult cuticle tanning, we wished to test whether TcLac2 is also critical for tanning of the larval and pupal cuticles of Tribolium. dsRNA for TcLac2 (≈0.2 μg per insect) was injected into larvae that were presumed to be a mixture of penultimate and last instars. As shown in Fig. 4, two different phenotypes were observed. One day after the larval–larval molt, animals treated with buffer showed normal cuticle tanning, whereas dsLac2-treated larvae exhibited no cuticle tanning (Fig. 4A). The larvae became shorter in length and died before any subsequent molting. One-day-old pupae that were treated as larvae with dsLac2 also exhibited no pupal cuticle tanning (Fig. 4B). No tanning was observed in pupa-specific cuticular structures, including setae, urogomphi, or gin traps. Gin traps were markedly distended, probably because the softness associated with the absence of sclerotization rendered them more sensitive to the swelling effect of internal hemolymph pressure (Fig. 4C). These results clearly demonstrated that TcLac2 plays a critical role in the sclerotization and pigmentation of larval and pupal cuticles and adult cuticle.

To evaluate the function(s) of the two alternatively spliced isoforms of laccase 2, TcLac2A and TcLac2B, in pupal and adult cuticle tanning, we performed exon-specific RNAi by using dsRNAs for the 3′ UTRs and extreme 3′ ends of C-terminal coding regions of the two alternatively spliced transcripts. The length of dsRNAs for TcLac2A and TcLac2B were 596 and 424 bp, respectively. The knockdown levels of the transcripts of TcLac2A and TcLac2B were examined by RT-PCR (Fig. 8A, which is published as supporting information on the PNAS web site). Injection of dsLac2A reduced the TcLac2A transcript level without reducing that of TcLac2B. Conversely, the introduction of dsLac2B reduced the TcLac2B transcript level without affecting that of TcLac2A. The resulting phenotypes are shown in Figs. 3B and 8B. Insects injected with dsRNA for TcLac2A, which is the major transcript of this gene, exhibited little tanning of the pupal and pharate adult cuticles. Injection of dsRNA for TcLac2B, which is the minor transcript, caused only a delay in adult cuticle tanning. Insects treated with a mixture of both dsRNAs showed phenotypes similar to those of insects injected with exon-nonspecific TcLac2 dsRNA. Little or no cuticle tanning was observed during the pupal, pharate adult, or adult stages. All of the beetles treated with exon-specific dsRNAs, either alone or in combination, died within a week after adult eclosion. These results indicated that both of the laccase 2 isoforms, TcLac2A and TcLac2B, play different but indispensable roles in pupal and/or adult cuticle tanning, with TcLac2A making the major contribution to tanning at all of the stages. Although injection of dsTcLac2B only delayed the rate of adult tanning, the animals did not survive for more than 1 week after adult emergence.

The effect of administration of exon-nonspecific dsRNA for TcLac2 on the phenotype was dose- and time-dependent. Administration of 200 ng of dsRNA per prepupa resulted in nearly complete inhibition of tanning in adults on day 1, and only a small degree of tanning was observed by day 2 (Fig. 5). All of these insects died on day 2 or 3 and exhibited severe developmental abnormalities. Reducing the dose of dsRNA 10-fold to 20 ng per prepupa resulted in fewer developmental abnormalities but did not result in substantial improvement in tanning or survival. At a dose of 2 ng per prepupa, the treated insects developed into more normal-looking adults, but they still had dented pronota, malformed wings, and incompletely tanned elytra, and the tanning process was considerably slowed, taking several additional days to complete, suggestive of a progressive loss of RNA interference after administration of low levels of dsRNA. In contrast to insects treated with higher doses of dsRNA, these insects were more viable, indicating substantial recovery from the effects of the lowest dosage of dsRNA administered.

Discussion

Insect cuticles vary considerably in stiffness, hardness, and pigmentation, depending on the evolutionary dictates of their specific anatomical and physiological roles. Tanning agents, namely oxidized catechols and their derivatives, could be generated by any one of several phenoloxidases or combinations thereof. The identification of the particular phenoloxidase(s) involved in cuticle tanning from among several candidate isoforms of laccases and tyrosinases had been a matter of debate until now. Our data unambiguously demonstrate that only two of the several phenoloxidases of Tribolium, namely laccases 2A and 2B, are required for larval, pupal, and adult cuticle sclerotization and pigmentation. Both of these isoforms, which are generated from a single gene, TcLac2, as a result of alternative splicing, are indispensable because knock-down of their transcripts at the late larval stage results in incomplete tanning of the larval, pupal, and adult cuticles and, subsequently, death. In contrast, neither laccase 1 nor the two isoforms of tyrosinase are required for tanning. Presumably, these three phenoloxidases have other functions unrelated to cuticle tanning, although a minor, dispensable role in cuticle sclerotization and pigmentation cannot be completely ruled out. Consistent with this interpretation, laccases 2A and 2B are expressed maximally in the epidermis just before the periods of pupal and adult cuticle tanning, when they catalyze the oxidation of endogenous catechols that serve as precursors for cuticle tanning agents (4, 24). Although the dsRNA-mediated down-regulation of TcLac2 was incomplete, beetles injected with TcLac2 dsRNA failed to tan normally, were soft-bodied, enlarged, deformed, and unable to walk, and they subsequently died prematurely in a dsRNA dose-dependent fashion. The morphological abnormalities and mortality are probably secondary consequences of loss of structural integrity of the cuticle in the absence of tanning. By catalyzing the formation of cross-linking agents from catechols for cuticle sclerotization, the two laccases encoded by the TcLac2 gene help to generate the supramolecular architecture of the exoskeleton, which provides not only the requisite mechanical properties for muscle attachment and locomotion but also stability against cuticle-degrading enzymes widely distributed in entomopathogenic microorganisms.

The data presented here clearly demonstrate that laccase 2 is involved in the tanning of not only pupal and adult cuticles and associated structures but also larval cuticle. The laccase 2A isoform appears to play the major role in the tanning process, affecting pupal and adult cuticle tanning and probably larval pigmentation as well. The laccase 2B isoform apparently makes a much smaller contribution compared with laccase 2A because it affects only the rate of adult cuticle tanning and has no effect on pupal tanning. However, this contribution is nonetheless indispensable, as indicated by structural abnormalities and premature death, which occur in young adult beetles when the transcripts of this particular isoform are depleted. It is possible that the requirement for the laccase 2B isoform may be related to morphogenesis and tanning of tracheae, foregut, hindgut, and/or other cuticle-containing tissue types exclusive of the exoskeleton during development of the pharate adult.

Progress in deciphering the supramolecular structure of tanned cuticle and the mechanisms of its assembly has been relatively slow because of the intractable nature of the “finished” product and the irreversibility of the tanning process. However, with the results reported here and elsewhere about structural cuticle proteins (4, 8, 25), catechols (26, 27), oxidative enzymes (2, 4, 12), and chitin (14, 17, 28), as well as their interactions (3, 24, 29, 30), researchers should in the future be able to determine more precisely how insects use these components to stabilize their exoskeletons and other tanned structures. A more complete understanding of the biochemistry of cuticle tanning, a process specific to arthropods, may reveal other targets besides laccases 2A and 2B for biorational agents that could be used for controlling agricultural pests and vectors of animal and human diseases. In addition, when the biochemical mechanisms responsible for cuticle tanning are better understood, new types of cross-linked biopolymers based on that chemistry could be produced, some of which may have medical or industrial applications (31).