Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Dev Biol. Author manuscript; available in PMC 2010 Feb 1.
Published in final edited form as:
Dev Biol. 2009 Feb 1; 326(1): 60–67.
Published online 2008 Oct 30. doi:  10.1016/j.ydbio.2008.10.022
PMCID: PMC2762819

Larval leg integrity is maintained by Distal-less and is required for proper timing of metamorphosis in the flour beetle, Tribolium castaneum


The dramatic transformation from a larva to an adult must be accompanied by a coordinated activity of genes and hormones that enable an orchestrated transformation from larval to pupal/adult tissues. The maintenance of larval appendages and their subsequent transformation to appendages in holometabolous insects remains elusive at the developmental genetic level. Here the role of a key appendage patterning gene Distal-less (Dll) was examined in mid- to late- larval stages of the flour beetle, Tribolium castaneum. During late larval development, Dll was expressed in appendages in a similar manner as previously reported for the tobacco hornworm, Manduca sexta. Removal of this late Dll expression resulted in disruption of adult appendage patterning. Intriguingly, earlier removal resulted in dramatic loss of structural integrity and identity of larval appendages. A large amount of variability in appendage morphology was observed following Dll dsRNA injection, unlike larvae injected with dachshund dsRNA. These Dll dsRNA-injected larvae underwent numerous supernumerary molts, which could be terminated with injection of either JH methyltransferase or Methoprene-tolerant dsRNA. Apparently, the partial dedifferentiation of the appendages in these larvae acts to maintain high JH and, hence, prevents metamorphosis.

Keywords: developmental stability, Distal-less, wound healing, regeneration, transdetermination, juvenile hormone, metamorphosis, Tribolium castaneum


Holometabolous insects are characterized by three distinct life history stages – the larva, the pupa and the adult. The distinct morphology associated with each life history stage is thought to permit the insect to inhabit distinct habitats as well as allocate resources to different functions, such as growth in the larva, tissue reorganization in the pupa and reproduction in the adult. The larval stage is characterized by dramatic growth that spans over a variable number of instars. Its growth is characterized by isometric growth with relatively few changes in morphology between instars (interval between molts). At the end of the larval stage, a dramatic change in morphology occurs during an event called metamorphosis. The pupa that results from this event has little resemblance to the larva. During this relatively quiescent stage, dramatic tissue reorganization occurs and adult structures develop. The adult again has little resemblance to both the larva and the pupa.

The hormonal mechanisms underlying the transition between the life history stages are well understood. The sesquiterpenoid hormone, juvenile hormone (JH), acts as a status quo hormone that maintains larval molting, and prevents pupae from forming (Williams, 1961; Riddiford, 1996). Upon reaching a certain body size called threshold weight, the larva commits to molt to the final larval instar (Nijhout and Williams, 1974a). The molecular mechanism by which the threshold weight is determined is not known. During the final instar, a second size check point called the critical weight (Nijhout and Williams, 1974b) determines the time at which JH is cleared from the hemolymph (the blood of the insect) and allows the larva to molt to a pupa.

Relatively little is known about the developmental genetic mechanisms underlying the transformation of larval structures to adult structures. In particular, we know very little about the genetic mechanism underlying the maintenance of larval appendages and the subsequent transformation of larval appendages to adult appendages, in part due to the fact that the model insect, Drosophila melanogaster, lacks larval appendages. Presumably, genes involved in patterning the larval structures during embryogenesis are reactivated during the final instar to initiate adult patterning (Truman and Riddiford, 2007). Unlike in Drosophila where the imaginal discs are produced during embryonic development, many holometabolous insects do not form imaginal discs until the final larval instar (Truman and Riddiford, 2002).

There are two ways that adult appendages develop from larval appendages. In the more derived holometabolous insects, such as the tobacco hornworm, Manduca sexta, only a small field of cells called the imaginal primordium in the larval leg contributes to the adult leg (Tanaka and Truman, 2005). Expression analysis in Manduca showed that the embryonic appendage patterning genes are re-expressed during the final instar in these imaginal primordia cells (Tanaka and Truman, 2007). In the more basal holometabolous species, the cells that form the adult leg appear to derive from the entire larval leg (Huet and Lenoir-Rousseaux, 1976; Truman and Riddiford, 2002). How these larval legs transform into adult legs is not well established.

In Drosophila, Dll is required during the embryonic phase when it specifies the fields of cells that will form imaginal discs (Kojima et al., 2004). In fact, the expression of Dll in several species suggests that this gene is expressed in most appendages throughout the Bilateria, suggesting that the specification of appendages is an ancient conserved function of Dll (Panganiban et al., 1997). Thus, Dll is likely to play an important role in the transformation of a larval leg to an adult leg. In the ladybird beetle Harmonia axyridis, Dll dsRNA given during the final instar caused truncation of the distal portions of the legs and antennae (Niimi et al., 2005). In the present study, we have investigated the function of Dll in Tribolium castaneum during the penultimate or earlier larval instar before the onset of transformation of the larval to the pupal leg.

While a number of Tribolium Dll mutants have been identified, most of the embryos with strong mutant alleles are embryonic lethal (Beerman et al., 2001). In addition, because Dll affects embryonic leg development, it is difficult to distinguish between the roles of Dll in larval and embryonic stages in such mutants. Therefore, we used time-specific RNA interference (RNAi) during mid- to late- larval life to study the role of Dll in Tribolium larvae. When Dll dsRNA was injected into the final instar, the larvae pupated and subsequently formed adults with appendage patterning defects. Injection of Dll dsRNA during earlier instars resulted in loss of larval appendage integrity and identity, as well as a large amount of variability in the appendage morphology within each larva. Furthermore, these larvae underwent numerous supernumerary larval molts. We provide evidence that this supernumerary phenotype is due to the maintenance of JH signaling.

Materials and Methods


Wildtype Tribolium castaneum was obtained from the USDA ARS Biological Research Unit, Grain Marketing & Production Research Center, Manhattan, Kansas (gift of Dr. Richard Beeman). The beetles were reared in small plastic containers containing organic wheat flour with 5% nutritional yeast. Stocks were maintained in a 26.5°C walk-in incubator under a long-day photoperiod regimen (17 hours light: 7 hours dark) until use.

Amplification of cDNA and cloning

The sequences for Dll, dac and Methoprene-tolerant (TcMet) were obtained from GenBank. Primers for Dll were designed outside the homeodomain to avoid the non-specific knockdown of other homeodomain proteins. A 425 bp fragment on the 5′ end of the mRNA transcript was amplified using the using the forward primer GGATAACAAACCCTTCACGAC and the reverse primer GCCTCTCCAACGATAAACAC. We also amplified a 301 bp on the 3′ end of the homeodomain using the forward primer GCAATAACAATAATGGGACACC and reverse primer TTGTGGGAGGTAGTTGTGTG. Either pair gave similar results with RNAi (the results reported here are from those injected with the former pair), but for in situ hybridization, we found the latter to work better. To ensure that the RNAi effects were not due to off-target RNAi effects, we have blasted the cloned sequences in Beetlebase and have found no stretch of DNA that had exact matches larger than 18bp. Additionally, none of these short exact matching sequences overlap between the two Dll sequences. For dac, a forward primer CAGCATCGCATCTTCAAC and a reverse primer CTCCTCCCTCAGCCTTTCT were used to amplify a 386 bp region of the dac gene. For TcMet, GAAGCTTCAAGAGAGGAATATG and TTTCAACAGTTCCCTGGTCG were used for forward and reverse primers, respectively, as reported by Jindra and Konopova (2007). We inferred JH acid O-methyltransferase (TcMT3) sequence from BeetleBase and isolated a 266 bp fragment using the forward primer CCGAAAATCCCCAAACA and reverse primer GCAAGGAAAGTCAGGAGCA. The identity of our isolated sequence was verified using the recently reported sequence of TcMT3 (Minakuchi et al., 2008). cDNA was amplified with the primers listed above and cloned using the Topo TA vector (Invitrogen). After confirmation of the insert via sequencing, the plasmids were cut with either Spe1 or Not1 restriction enzymes (NE Biolabs, Ipswich, MA).

Probe preparation

Sense and anti-sense Digoxigenin (DIG) labeled probes were synthesized from the linearized plasmids using the MaxiScript kit (Ambion, Austin, TX) with DIG RNA labeling mix (Roche Molecular Biochemicals, Indianapolis, IN). Alkaline hydrolysis was used to hydrolyze the dac probes. Dll probes were not hydrolyzed. The probes were suspended in hybe buffer (65% deionized formamide, 6.5× saline sodium citrate, 500 ug/ml heparin, 0.1% Tween-20, 500 μg/ml boiled sperm, 500 μg/ml torula RNA) and stored at -80°C until use. Prior to use, the probe was thawed and boiled for 3-5 min.

In situ hybridization

Prepupae at the stationary crooked posture stage were dissected in phosphate-buffered saline [PBS: 0.0038M NaH2PO4, 0.0162 M Na2HPO4, 0.15 M NaCl (pH 7.4)]. The ventral thorax with the head region attached was dissected and fixed in 3.7% formaldehyde (FA) overnight at 4°C. The legs and antennae were then dissected out of the larval cuticle and after several washes in PBS with 1% Tween-20 (PBT), processed using standard in situ procedures (protocol obtained from Drs. David Angelini and Yoshinori Tomoyasu). Briefly, the fixed tissues were dehydrated through 25%, 50, 80% and 100% methanol series in PBT. After at least 1 hour of storage at -20°C, the tissues were rehydrated through 80%, 50% and 25% methanol series in PBT and then rinsed several times in PBT. A 30 minute proteinase K (10 μg/μl) digest was followed by rinses in PBT with 2 mg/ml glycine. After several rinses in PBT, the tissues were postfixed in FA for 20 min. The tissues were then rinsed several times in PBT and subsequently transferred to hybe buffer. After at least 1 hour of incubation in hybe buffer at 55°C, the probe was added. After incubation with the probe for 14-48 hrs, the probe was removed and 5 washes with the following wash buffers: 50% formamide:5X SSC, 50% formamide:2X SSC, 25% formamide:2X SSC, 2X SSC and 0.2X SSC. These were performed at 55°C for 30 min each except the final wash, which was 40 min long. After an additional rinse in 0.2X SSC, the tissues were washed in maleic acid buffer (0.1 M maleic acid, 0.15 M NaCl; pH 7.5) with 3% Triton-X several times and blocked for 1 hr at room temperature in blocking buffer (Tris pH 7.4 with 2 mg/ml bovine serum albumin and 5% normal goat serum). Tissues were incubated overnight at 4°C with anti-DIG-alkaline phosphatase (AP) conjugate (Roche Applied Science, Indianapolis, IN) at a concentration of 1:2000. After washing with maleic acid buffer several times and then with AP buffer, the color reaction was performed using NBT/BCIP as the substrate. The color reaction was stopped by rinses in PBT and the tissues were cleared using a methanol series. The tissues were mounted on slides in glycerol.

dsRNA synthesis and injection

Each of the strands of the dsRNA was synthesized using T3 or T7 promoter sites on the plasmid with the MEGAscript Kit (Ambion, Austin, TX). The two complementary RNA strands were annealed as described by Hughes and Kaufman (2000). Final concentrations were 2 μg/μl for Dll, 4 μg/μl for dac, 4 μg/μl for TcMT3 and 5 μg/μl for TcMet.

Our Tribolium colony typically has a total of 7 or 8 instars. dsRNA was injected into various instars of Tribolium larvae using a 10 μl glass capillary needle connected to a syringe. For larvae older than 7th instar, 0.5 μl was injected. For younger instars, dsRNA was injected until the abdomen started to stretch due to a buildup of pressure. The same volume of bacterial ampicillin-resistance (ampr) dsRNA (plasmid obtained from Dr. Takashi Koyama, our laboratory) was injected for controls. Injected larvae were kept at 30°C, and the resulting phenotypes were analyzed.

Leg ablation

Three legs on one side of the final instar larvae were cut either just proximally to the coxa-trochanter boundary or near the tibia-femur boundary on day 0, 1 or 2 following ecdysis to the final larval instar. A typical larva molts every 3 to 4 days. The larvae were either maintained in flour at 30°C without further treatments or the untreated three legs on the other side of the larvae were ablated on either day 0, 1 or 2 of the subsequent supernumerary instar. The number of larval molts before pupation was recorded.


Expression of Dll

In the antennae, Dll expression was detected throughout the appendage although it was particularly prominent in the more distal segments (Fig. 1A). Dll was also expressed throughout the leg although the expression intensity differed along the proximal-distal axis. The expression was particularly strong at the boundary of the trochanter and femur. In the tibia and tarsus, strong expression was also observed in the more distal ends of the segments. No expression was observed when the control sense probe was used.

Figure 1
Expression of Dll and dac as determined by in situ hybridization. (A) Dll is expressed throughout the legs with variable intensities along the proximo-distal axis (see text). (B) dac is expressed in the medial portions of appendages. No expression is ...

Effect of Dll RNAi during the 7th larval instar

Injection of Dll dsRNA into day 0 7th instar larvae resulted in the formation of either larvae that kept molting as larvae and never pupated (6 out of 16) or pupae (10 out of 16) (Table 1). The pupae had shortened, stout appendages that instead of folding, stuck out laterally (Fig. 2A, black arrow). Each of the leg segments was shortened but broader. In addition, the urogomphi were dramatically shortened with only small stumps (Fig. 2A, arrowhead). The adults were unable to shed their pupal cuticle properly, especially around the wings, but the adult cuticle tanned normally. The legs were shorter, and the tarsi and claws were fused to form a single mass that lacked the normal segmentation (Fig. 2B). The pupal antennae were much reduced in size (not shown), and the adult antennae were severely distorted with very few identifiable segments (Fig. 2B). The supernumerary phenotypes were characterized more carefully in larvae injected at an earlier time point. Most of the larvae injected with ampr dsRNA as a control pupated either at the end of the 7th instar or the 8th instar. One formed a 9th instar larva, then pupated (Table 1A).

Figure 2
Effect of 7th instar Dll and dac dsRNA injection on the pupae and adults. (A) Pupae formed after injection of ampr, Dll and dac dsRNA. Dll dsRNA injected pupae had stout legs that extended laterally (black arrow) and reduced urogomphi (arrowhead). dac ...
Table 1
The effect of Dll dsRNA injection on the timing of pupation. (A) Effect of Dll dsRNA injection at the onset of 7th and 6th instar larvae. (B) Fate of supernumerary larvae undergoing larval-larval molt at the end of the 10th instar. Italicized number in ...

Effect of Dll RNAi during the 6th larval instar

Injection of Dll dsRNA into day 0 6th instar resulted in the formation of supernumerary larvae (Table 1A). Knockdown of the transcript in these animals was confirmed by RT-PCR (data not shown). Typically, Tribolium larvae pupate at the end of the 7th or 8th instar. However, most of the larvae injected with Dll dsRNA on day 0 of the 6th instar molted to the 9th instar and continued to undergo supernumerary larval molts before some started to pupate at the end of the 10th instar (15 out of 67). During these earlier molts, the larvae molted every three or four days, which is the same interval seen in those injected with ampr dsRNA. The larvae that pupated formed adults with defects as described above (not shown). About two-thirds of the larvae continued to molt as larvae (43 out of 67). We tracked the fate of 27 larvae and found that only 3 out of 27 larvae pupated after the 10th instar (Table 1B). The rest underwent subsequent supernumerary molts, with a few reaching the 15th instar (8 out of 27), and eventually died as larvae (Table 1B). These Dll dsRNA-injected larvae grew to a large size (Fig. 3A) and were feeding during the 8th instar (data not shown), so the supernumerary molts were not induced by starvation. In contrast, most larvae injected with ampr dsRNA pupated at the end of the 8th instar with a few stragglers pupating at the end of the 9th instar. None went beyond the 9th instar.

Figure 3
Effect of Dll dsRNA injection on the larval phenotype. (A) A 10th instar Dll dsRNA injected larva (right) and a wild type larva that has entered the prepupal period (left). (B) Diversity of leg and antenna morphologies obtained after injection of Dll ...

Interestingly, the appendages of these Dll dsRNA-injected larvae showed deformities that worsened after every molt. We tracked the number of affected appendages and found that the number of affected appendages increased (Fig. 4). The nature of deformities varied between individuals as well as within an individual (Figs. 3B and and4)4) but typically involved a loss of distinct distal segments and broadening of the remaining segments, presumably resulting from proliferation (Fig. 3B). In a number of the antennae, bilobed structures formed (Fig. 3B, arrowhead). In some larvae, only antennae were affected, while in others, only legs were affected. However, by the time the larvae molted to the 12th instar, most had deformities in all the appendages with their identity mostly indeterminable (data not shown).

Figure 4
The change in the percentage of larvae with affected appendages through development in larvae injected with Dll dsRNA at the onset of the 6th instar. X-axis represents the instar. (left) analysis for all larvae; (center) analysis for larvae undergoing ...

Because a number of larvae pupated at the end of the 10th instar, we separately analyzed those larvae undergoing larval-pupal molts at the end of the 10th instar and those undergoing larval-larval molts (Fig. 4, center and right). Interestingly, those pupating at the end of the 10th instar had fewer legs with noticeable deformities compared to those undergoing larval-larval molts (Fig. 4B, center and right). Thus, the number of deformed legs and the number of supernumerary molts appeared to be correlated. The number of affected antennae was not noticeably different between the two groups (Fig. 4A).

Effect of leg ablation

The above experiments suggested that the larval appendage deformities may play a role in preventing the larvae from pupating. We therefore cut off legs on one side of the final instar larvae to see if they would molt to supernumerary larvae. When the legs were removed on day 0 or day 1 of the final instar, approximately 70% of the larvae underwent a supernumerary molt at the end of that instar (n=9 and 13, respectively; Table 2). When only the tips of the three legs were removed on day 1, all larvae pupated at the end of the instar (n=5; Table 2).

Table 2
Effect of leg ablation. (A) Effect of the timing of leg ablation during the presumptive final instar. Three legs on one side of the larva were removed. L-P molt refers to larvae pupating at the end of the instar. L-L molt refers to larvae undergoing supernumerary ...

When the legs were removed on day 2, most larvae underwent a supernumerary molt (20 out of 22 or 91%; Table 2). When the other three unablated legs on each of the supernumerary larvae were ablated, the larvae typically underwent another supernumerary molt (7 out of 10; Table 2B). Two out of 10 larvae formed larval-pupal intermediates with tiny winglets (not shown). Six of the 7 supernumerary larvae subsequently pupated, and the remaining one underwent another supernumerary molt before pupating (not shown). One of the larval-pupal intermediate animals also underwent another intermediate molt (not shown). Interestingly, even in the larvae that only had one set of legs ablated, a number of larvae underwent another supernumerary molt (3 out of 10; Table 2B).

Effect of removal of JH signaling following Dll dsRNA injection

The production of supernumerary larvae is a hallmark of elevated JH titers. We therefore hypothesized that removal of JH synthesis or reception should cause the supernumerary larvae to pupate. TcMet has been implicated in the transduction of JH signal in the target tissues whereas TcMT3 is a necessary component of JH synthesis in the corpora allata. Both TcMet and TcMT3 dsRNA injections have been shown to induce precocious pupation in Tribolium (Konopova and Jindra, 2007; Parthasarathy et al., 2008; Minakuchi et al., 2008; Supplementary Figures 1A and B).

In Dll dsRNA-injected larvae, injection of TcMT3 or TcMet dsRNA resulted in the formation of prepupae at the end of either the instar of injection or the following instar (Table 3). When we injected Dll dsRNA during the 6th instar and injected TcMet or TcMT3 dsRNA either simultaneously or subsequently during either the 9th or the 11th instar, the larvae formed prepupae either at the end of the same instar that TcMT3/TcMet dsRNA was injected or at the end of the following instar (Table 2). Comparable results were obtained when similar experiments were conducted on earlier instar larvae (see Supplementary materials). Thus, injection of TcMets or TcMT3 dsRNA can rescue the Dll dsRNA supernumerary phenotype if injected during the 6th instar and can terminate the supernumerary molts if injected later. The adults that formed from TcMT3 dsRNA injection were larger than those injected with ampr dsRNA alone that molted earlier as can be seen clearly by comparing the size of the pronotum (Fig. 5A, arrows). When ampr dsRNA was injected as control, the Dll dsRNA-injected larvae never pupated (Table 3).

Figure 5
The adult morphology of Dll dsRNA-injected larvae given TcMT3 or TcMet dsRNA. A) Adults formed after TcMT3 or TcMet dsRNA injection into 11th instar larvae given Dll dsRNA at the onset of the 6th instar. The TcMet dsRNA-injected animal is like a precocious ...
Table 3
The effect of TcMT3 or TcMet dsRNA injection on the timing of pupation in Dll dsRNA-injected larvae. Dll dsRNA was injected at the onset of the 6th instar. TcMT3 or TcMet dsRNA was injected either at the onset of the 6th instar, 9th instar or 11th instar. ...

In these Dll dsRNA-treated animals that were induced to pupate, a large variability in the legs was seen both between individuals and within individuals (Fig. 5B; supplementary materials; compare with those of animals injected with Dll dsRNA during the 7th instar shown in Fig. 2C). Many of the appendages had morphologies that are never seen in normal adults and therefore, the identities of these appendages were not clear. In a few legs, the adult leg was merely a femur with a small segment attached at the end. Interestingly, when larvae were injected with TcMT3 dsRNA at the end of the 11th instar, we found a few adults with enlarged antennal structures that assumed a leg-like identity (Fig. 5B). While the identity of the proximal portion of the appendage was unclear, a forked claw developed at the distal end (Fig. 5B, arrow), indicating that the antenna had partially transformed into a leg. In fact, we have observed a few larvae with claw-like structure at the end of the antennae after numerous molts following Dll dsRNA injection and then ampr dsRNA (not shown).

Appendage expression of dac and the effect of dac dsRNA injection

Next we wished to determine whether this supernumerary phenotype was specific to Dll or a general response to the removal of any appendage patterning gene. dac encodes a nuclear protein that is required for patterning the medial segments of Drosophila, Manduca and larval Tribolium legs (Mardon et al., 1994; Prpic et al., 2001; Tanaka and Truman, 2007). We first determined the expression of dac in the appendages of prepupae. In all appendages, we found that dac was expressed in the medial regions of the appendages (Fig. 1B). In the legs, expression was detected in the femur and tibia, with some variability in the intensity of the signal, suggesting that dac may play a role in defining the segmental boundaries. No signal was observed in the distal portions of the tarsi. In the antennae, we found dac expression confined to the more proximal domain as well.

We next injected dac dsRNA into larvae and observed the phenotypes. Unlike Dll dsRNA injection, injection of dac dsRNA did not result in the dramatic alteration of larval appendage morphology (not shown). Furthermore, when dac dsRNA was injected at the onset of 6th instar, the larvae typically pupated at the end of either the 7th or 8th instar. The pupae of the dac dsRNA injected larvae had appendage deformities with the legs extending laterally (Fig. 2A, open arrow), and the adult that eclosed had truncated appendages that were missing the medial regions (Figs. 2B and C). In the legs, the distal portion of the femur as well as the entire trochanter and tibia appeared to be missing. In addition, only one tarsal segment was present although the claws developed normally (Fig. 2C). Unlike in Dll dsRNA injected animals, we observed very little variability in the legs of dac dsRNA-injected adults. Furthermore, all legs were more or less identical: unlike ampr-injected adults where the fore-, mid- and hind-legs have distinct sizes, in dac dsRNA-injected adults, all legs had similar morphology and size (Fig. 2C, black lines). In addition, the middle segments of the antennae were fused, thus shortening the overall length of the antennae. The proximal and distal-most segments appeared normal in morphology (Fig. 2C).


In the present study, expression patterns of Dll and dac were observed in prepupal appendages of Tribolium. We found that both late Dll and mid- and late- dac dsRNA injections resulted in appendage patterning defects in the adults. However, mid-larval Dll dsRNA injections produced supernumerary larval instars. We also found that the supernumerary molts could be interrupted at any time with the injection of TcMet or TcMT3 dsRNA. The appendages of Dll dsRNA-injected larvae lost their tissue patterning and developed into malformed structures that lacked clear segmental identity. Furthermore, we demonstrated that ablation of legs resulted in the formation of supernumerary larvae.

Expression patterns of Dll and dac in Tribolium prepupal legs resemble those of Manduca sexta

The expression of Dll and dac in the legs of Tribolium is similar to that observed in the legs of Manduca. In both species, Dll is expressed throughout most of the legs except for a small patch near the femur/tibia boundary (Tanaka and Truman, 2007; this study). Because of the small size of Tribolium, expression could only be analyzed in prepupal legs. However, the broad expression throughout the legs is in agreement with the effect of Dll dsRNA injection where adult leg patterning is affected in each leg segment. Similarly, dac is expressed in the medial regions of the appendages in both species (Tanaka and Truman, 2007; this study).

Role of dac in appendage development

The appendages of dac dsRNA-injected adults provide some interesting insights. The adult legs of dac dsRNA-injected animals are of interest as they lack all the allometric differences between the fore-, mid- and hind-legs. Given that Ultrabithorax (Ubx) regulates allometric differences between the serially homologous legs (Mahfooz et al., 2004), it may be that Ubx acts through dac to generate these differences.

We found that removal of dac results in fusion of the medial segments of the adult antennae. This is in stark contrast to Drosophila and Oncopeltus where dac plays minimal roles in the development of antennae. In Oncopeltus, no defects were observed when dac was removed using RNA interference (Angelini and Kaufman, 2004). Similarly, dac mutants of Drosophila only have a minor defect in the a5-arista joint, and very little aberration in the overall morphology of the antennae (Dong et al., 2003). It is noteworthy that Drosophila has only four antennal segments and Oncopeltus has five, whereas Tribolium has 11. It will be of interest to determine whether dac may somehow be linked to the increased number of antennal segments.

Late larval Dll dsRNA injection results in adult patterning defects

Removal of Dll during the final instar resulted in the disruption of proper distal leg patterning. Dll appears to play a major role in defining the tarsal segments. Other leg segments are shorter but broader. The antennae were also severely malformed with most of the antennal segments being indistinguishable, and the antennae overall were greatly reduced in size. These patterning defects are consistent with the defects observed in the Short antennae mutant of Tribolium which has been shown to be a mutant allele of Dll (Beerman et al. 2001). Similar phenotypic effects were observed when Dll dsRNA was injected into the last larval instar of the ladybird beetle, Harmonia axyridis (Niimi et al., 2005).

Role of Dll in maintaining appendage identity, patterned growth and developmental stability

We found that in animals given Dll dsRNA during earlier instars, the larval legs and antennae became distorted greatly with progressive broadening and gross morphological alterations that are at least partially due to the loss of original appendage identity. In contrast, dac dsRNA-injected larvae maintained the general larval leg morphology. Thus, it appears that Dll may have a unique role in maintaining the integrity of larval appendages. In Drosophila, Dll has been implicated in specification of ventral appendage identities (Gorfinkel et al., 1997). In the antennae of Drosophila, it has been observed that Dll is required for specifying antennal fate: removal of Dll results in the formation of legs in the place of antennae (Dong et al., 2000). We have found that a few Dll dsRNA-injected larvae that underwent multiple supernumerary molts before induced to pupate developed claws at the tip of the antenna. This observation suggests that when deprived of Dll for a sufficiently long time, the larval antenna begins to transform to a leg.

This formation of a different appendage type from the original Dll-treated appendage is similar to Drosophila imaginal discs during disc regeneration (reviewed in McClure and Schubiger, 2007). During regeneration, the compartment-specific identity of the disc cells are lost, and the discs become much more flexible in their developmental fate. For example, cells in irradiated antennal discs that are cultured in vivo can change their cellular identity to that of a wing by a process called transdetermination (Gehring, 1967). Such a transformation of tissue identity during regeneration has also been observed in the walking stick, Carausius, as well as a number of crustacean species (Brecher, 1924; Needham, 1965). Thus, appendages of various insects behave similarly in response to disruption of tissue integrity, and this suggests that transdetermination may be a widely conserved feature of insects and crustaceans.

The ability of the larval tissues to transform suggests that in response to Dll removal, the tissues become sufficiently plastic to change their fates. Furthermore, we observed that in larval legs, removal of Dll function results in tissue de-differentiation with 1) a large amount of variability in its qualitative nature; and 2) asymmetry in the malformation of appendages. Thus, removal of Dll appears to unleash a large amount of developmental instability (Fig. 3). Interestingly, Dll has been shown to play a key role in stable development of several traits of the adult Drosophila leg (Dworkin, 2005).

Removal of Dll during mid-larval stages results in the formation of supernumerary larval instars

We have found that when Dll is removed early during the penultimate instar, the larvae will keep molting, and most will not pupate. The few that pupate after a number of supernumerary molts do so probably after the effect of Dll RNAi has waned over two to three weeks. In fact, examination of the larval leg morphology showed that those that pupated after a number of supernumerary molts tended to have fewer legs with deformities compared to those that never pupated. The supernumerary larvae continued to feed and grow, indicating that the effect cannot be explained by starvation.

The supernumerary molt indicated to us that JH might remain high in these larvae. Typically, JH titer declines during the final instar and this allows tissues to commit and metamorphose (Riddiford, 1976). However, when a JH analog is applied topically to Tribolium larvae, they undergo larval molts and do not pupate (Konopova and Jindra, 2007; Suzuki et al., 2008). When dsRNA against TcMT3 or TcMet, which are required for JH synthesis (Bellés et al., 2005; Minakuchi et al., 2008) and action (Konopova and Jindra, 2007; Riddiford, 2008; Parthasarathy et al., 2008) respectively, was injected into the Dll dsRNA-treated larvae, their supernumerary molts were terminated. These data show that JH signaling remains active in these treated larvae.

One possible mechanism by which the JH remains high is that the deformed appendage tissue elicits a wound healing response that is known to elevate JH in some insects (Needham, 1965; Krishnakumaran, 1972). In many insects, including Tribolium reported here, wounding results in the formation of supernumerary instars, presumably to allow larvae to heal the injury before pupating (reviewed in Esperk et al., 2007). We suggest that the removal of Dll results in de-differentiated appendages that either elevates JH titers or sensitizes the animal to JH.

It is important to recognize that unlike in Drosophila where tissue-specific knockdown of genes is possible using tissue-specific transgene expression systems, injection of dsRNA affects the whole animal. Thus, we cannot exclude the possibility of other tissues being involved. However, we have not been able to detect any obvious Dll mRNA expression in the Tribolium brain (unpublished results). Thus, we find the appendage deformity to be the most parsimonious explanation for the supernumerary phenotype.

The current study demonstrates that Dll plays an important role in maintaining the structural integrity and identity of appendage tissues in Tribolium. Furthermore, we find that larval legs have extensive plasticity as well as capacity to undergo transdetermination. This loss of tissue identity may be akin to a reversion to a more stem cell-like identity. Therefore Dll may play an important role in commitment and reprogramming of stem cells.

Supplementary Material



We would like to thank H. Cui, Dr. K. Hiruma, H. Kelstrup, Dr. T. Koyama, Dr. L. Marin, Dr. C. Mirth, Dr. S. McNabb, M. Rynerson and Dr. J. Truman (Janelia Farm, HHMI Institute) for discussions and assistance. We also thank Dr. David Angelini (U. of Connecticut), Dr. Julia Bowsher (U. of Arizona), James Cooley (U. of Arizona), Dr. Kohtaro Tanaka (OHSU) and Dr. Yoshinori Tomoyasu (Kansas State U.) for their help with in situ hybridization and helpful discussions. We thank Dr. Richard Beeman and Dr. Yoshinori Tomoyasu for the Tribolium beetles. This research was supported by NIH 2R01 GM60122.


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


  • Angelini DR, Kaufman TC. Functional analyses in the hemipteran Oncopeltus fasciatus reveal conserved and derived aspects of appendage patterning in insects. Dev Biol. 2004;271:306–321. [PubMed]
  • Beermann A, Jay DG, Beerman RW, Huelskamp M, Tautz D, Juergens G. The Short antenna gene of Tribolium is required for limb development and encodes the orthologue of the Drosophila Distal-less protein. Development. 2001;128:287–297. [PubMed]
  • Bellés X, Martin D, Piulachs MD. The mevalonate pathway and the synthesis of juvenile hormone in insects. Annu Rev Entomol. 2005;50:181–199. [PubMed]
  • Brecher L. Die Bedingungen fur Fühlerfiisse bei Dixippus morosus. Arch Mikr Anat u Entw Mech. 1924;102:549–572.
  • Dong PDS, Chu J, Panganiban G. Coexpression of the homeobox genes Distal-less and homothorax determines Drosophila antennal identity. Development. 2000;127:209–216. [PubMed]
  • Dong PDS, Dicks JS, Panganiban G. Distal-less and homothorax regulate multiple targets to pattern the Drosophila antenna. Development. 2003;129:1967–1974. [PubMed]
  • Dworkin I. Evidence for canalization of Distal-less function in the leg of Drosophila melanogaster. Evol Devel. 2005;7:89–100. [PubMed]
  • Esperk T, Tammaru T, Nylin S. Intraspecific variability in number of larval instars in insects. J Econ Entomol. 2007;100:627–645. [PubMed]
  • Gehring W. Clonal analysis of determination dynamics in cultures of imaginal disks in Drosophila melanogaster. Dev Biol. 1967;16:438–456. [PubMed]
  • Huet C, Lenoir-Rousseaux JJ. Etude de la mise en place de la patte imaginable de Tenebrio molitor. 1. Analyse expérimentale des processus de restauration au cours de la morphogenése. J Embryol Exp Morphol. 1976;35:303–321. [PubMed]
  • Hughes CL, Kaufman TC. RNAi analysis of Deformed, proboscipedia, and Sex combs reduced in the milkweed bug Oncopeltus fasciatus: novel roles for Hox genes in the hemipteran head. Development. 2000;127:3683–3694. [PubMed]
  • Kojima T. The mechanism of Drosophila leg development along the proximodistal axis. Dev Growth Diff. 2004;46:115–129. [PubMed]
  • Konopova B, Jindra M. Juvenile hormone resistance gene Methoprene-tolerant controls entry into metamorphosis in the beetle Tribolium castaneum. Proc Natl Acad Sci USA. 2007;104:10488–10493. [PMC free article] [PubMed]
  • Krishnakumaran A. Injury induced molting in Galleria mellonella larvae. Biol Bull. 1972;142:281–292. [PubMed]
  • Mahfooz NS, Li H, Popadic A. Differential expression patterns of the hox gene are associated with differential growth of insect hind legs. Proc Natl Acad Sci USA. 2004;101:4877–4882. [PMC free article] [PubMed]
  • Mardon G, Solomon NM, Rubin GM. dachshund encodes a nuclear protein required for normal eye and leg development in Drosophila. Development. 1994;120:3473–3486. [PubMed]
  • McClure K, Schubiger G. Transdetermination: Drosophila imaginal disc cells exhibit stem cell-like potency. Int J Biochem Cell Biol. 2007;39:1105–1118. [PMC free article] [PubMed]
  • Minakuchi C, Namiki T, Yoshiyama M, Shinoda T. RNAi-mediated knockdown of juvenile hormone acid O-methyltransferase gene causes precocious metamorphosis in the red flour beetle Tribolium castaneum. FEBS J. 2008;275:2919–2931. [PubMed]
  • Needham AE. Regeneration in the arthropoda and its endocrine control. In: Kiortsis V, Trampusch H, editors. Regeneration in Animals and Related Problems. North Holland Publishing; Amsterdam: 1965. pp. 283–323.
  • Niimi T, Kuwayama H, Yaginuma T. Larval RNAi applied to the analysis of postembryonic development in the ladybird beetle, Harmonia axyridis. J Insect Biotechnol Sericol. 2005;74:95–102.
  • Nijhout HF, Williams CM. Control of moulting and metamorphosis in the tobacco hornworm, Manduca sexta (L.): cessation of juvenile hormone secretion as a trigger for pupation. J Exp Biol. 1974a;61:493–501. [PubMed]
  • Nijhout HF, Williams CM. Control of moulting and metamorphosis in the tobacco hornworm, Manduca sexta (L.): growth of the last-instar larva and the decision to pupate. J Exp Biol. 1974b;61:481–491. [PubMed]
  • Panganiban G, et al. The origin and evolution of animal appendages. Proc Natl Acad Sci USA. 1997;94:5162–5166. [PMC free article] [PubMed]
  • Parthasarathy R, Tan A, Palli SR. bHLH-PAS family transcription factor methoprene-tolerant plays a key role in JH action in preventing the premature development of adult structures during larval-pupal metamorphosis. Mech Dev. 2008;125:601–616. [PMC free article] [PubMed]
  • Prpic NM, Wigand B, Damen WG, Klingler M. Expression of dachshund in wild-type and Distal-less mutant Tribolium corroborates serial homologies in insect appendages. Dev Genes Evol. 2001;211:467–477. [PubMed]
  • Riddiford LM. Hormonal control of insect epidermal cell commitment in vitro. Nature. 1976;259:115–117. [PubMed]
  • Riddiford LM. Juvenile hormone: the status of its “status quo” action. Arch Insect Biochem Physiol. 1996;32:271–286. [PubMed]
  • Riddiford LM. Juvenile hormone action: a 2007 perspective. J Insect Physiol. 2008;54:895–901. [PubMed]
  • Suzuki Y, Truman JW, Riddiford LM. The role of the transcription factor broad in the development and evolution of the holometabolous insect pupa, Tribolium castaneum. Development. 2008;135:569–577. [PubMed]
  • Tanaka K, Truman JW. Development of the adult leg epidermis in Manduca sexta: contribution of different larval cell populations. Dev Genes Evol. 2005;215:78–89. [PubMed]
  • Tanaka K, Truman JW. Molecular patterning mechanism underlying metamorphosis of the thoracic leg in Manduca sexta. Dev Biol. 2007;305:539–550. [PubMed]
  • Truman JW, Riddiford LM. Endocrine insights into the evolution of metamorphosis in insects. Annu Rev Entomol. 2002;47:467–500. [PubMed]
  • Truman JW, Riddiford LM. The morphostatic actions of juvenile hormone. Insect Biochem Mol Biol. 2007;37:761–770. [PubMed]
  • Williams CM. The juvenile hormone. II. Its role in the endocrine control of molting, pupation, and adult development in the cecropia silkworm. Biol Bull. 1961;121:572–585.
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem Compound links
  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...