Hormonal signaling cascades required for phototaxis switch in wandering Leptinotarsa decemlineata larvae

Many animals exploit several niches sequentially during their life cycles, a fitness referred to as ontogenetic niche shift (ONS). To successfully accomplish ONS, transition between development stages is often coupled with changes in one or more primitive, instinctive behaviors. Yet, the underlining molecular mechanisms remain elusive. We show here that Leptinotarsa decemlineata larvae finish their ONS at the wandering stage by leaving the plant and pupating in soil. At middle wandering phase, larvae also switch their phototactic behavior, from photophilic at foraging period to photophobic. We find that enhancement of juvenile hormone (JH) signal delays the phototactic switch, and vise verse. Moreover, RNA interference (RNAi)-aided knockdown of LdPTTH (prothoracicotropic hormone gene) or LdTorso (PTTH receptor gene) impairs avoidance response to light, a phenotype nonrescuable by 20-hydroxyecdysone. Consequently, the RNAi beetles pupate at the soil surface or in shallow layer of soil, with most of them failing to construct pupation chambers. Furthermore, a combination of depletion of LdPTTH/LdTorso and disturbance of JH signal causes no additive effects on light avoidance response and pupation site selection. Finally, we establish that TrpA1 (transient receptor potential (TRP) cation channel) is necessary for light avoidance behavior, acting downstream of PTTH. We conclude that JH/PTTH cascade concomitantly regulates metamorphosis and the phototaxis switch, to drive ONS of the wandering beetles from plant into soil to start the immobile pupal stage.

and pupating in soil. At middle wandering phase, larvae also switch their phototactic behavior, 23 from photophilic at foraging period to photophobic. We find that enhancement of juvenile

50
Movements to stage-dependent resources, i.e., ontogenetic niche shifts (hereafter ONS), 51 occur in nearly 80% of animal taxa. The shifts enable animals to exploit several niches 52 sequentially during their life cycles to meet stage-dependent nutritional requirements, to 53 overcome stage-specific physiological limitations, and to reduce intraspecific competition 54 between juveniles and adults. ONS is thus widely accepted as an evolutionary adaptation 4 55 [1][2][3][4]. To successfully finish ONS, transition between development stages is often 56 accompanied with changes in one or more primitive, instinctive behaviors, allowing 57 inexperienced novices to obtain novel abilities to detect new environmental cues [5]. To date, 58 however, the underlining mechanisms driving these behavior switches are still largely 59 unexplored. 60 Insects are a suitable animal group to explore the molecular modes of these instinctive [5]. Accordingly, it can be reasonably hypothesized that the change from foraging to 77 wandering stages should be coupled with a switch for phototactic behavior from photophilic 78 to photophobic in most herbivorous Holometabolans. 79 For a soil-pupated insect species, a wandering larva usually shows a sequence of three 80 primary behavioral components before pupation: a) crawling to the ground and searching for 81 a suitable location, b) mining into soil, and c) building a pupation chamber in soil [18][19][20]. The first aim of the current study was to determine whether Leptinotarsa larvae changed 108 their light preference from photophilic to photophobic during wandering stage. We then 109 uncovered that JH/PTTH cascade concomitantly regulated metamorphosis and the phototaxis 110 switch. Finally, we provided clear evidence that PTTH-induced light avoidance drove ONS 111 from plant to pupation refuge in an insect herbivore.

113
Phototaxis of L. decemlineata larvae 114 We first observed phototaxis of Leptinotarsa beetles on the natural potato field. The

125
When given a choice in the laboratory, second-instar, third-instar and foraging 126 fourth-instar larvae prefer light-exposed over shaded areas (Fig 1J). For wandering 127 Leptinotarsa larvae, however, light avoidance index significantly increased at 5 days post 128 ecdysis, and reached nearly 1.0 at 6 days ( Fig 1K). These basic findings demonstrate that an 129 obvious phototactic switch occurs at the middle wandering period.  Wandering behavior occurred at 4.5-4.7 days after ecdysis (S2 Table), which was earlier 142 than the occurrence of the phototaxis switch. Moreover, ingestion of 20E accelerated the 143 onset of wandering behavior, whereas interruption of 20E signal retarded the onset (S2 Table).  Table).  Table); whereas knockdown of a JH biosynthesis gene (JH acid methyl transferase, 9 159 wandering (S2 Table). It is clear that JH signal inhibits the premature switch of light 160 preference and the early onset of wandering. LdTorso) in the larvae whose JH signal had been disturbed. As expected, the mRNA levels of 165 the two genes were reduced in the larval specimens whose JH signal had been enhanced  It is suggested that PTTH promotes larval light avoidance in L. decemlineata. We next 169 knocked down PTTH gene (Fig 3A), and verified the knockdown by determination of the  Table). Moreover, silencing of PTTH reduced avoidance response 174 to light (Fig 3D), rate of larvae that had buried in soil per day (RLB) ( Fig 3E) and  Table). 183 We then examined the relative mRNA levels of a JH biosynthesis gene LdJHAMT and 184 two JH signaling pathway genes (LdMet and LdKr-h1), and found that these genes varied  Table). We found no additional effect on avoidance response to light (Figs 4A-D),  Lastly, no additional effects on light avoidance response and pupation were observed by 230 simultaneous inhibition of PTTH pathway and disturbance of JH signal (Fig 4).   Table). The latency between appearance of 20E and the onset of wandering ]. Moreover, we found here that Leptinotarsa wandering larvae kept crawling at the soil 243 surface; they did not burrow into soil (Fig 3 and S3 Fig) until the phototaxis switch had 244 finished on day 5.5-6.0 post ecdysis (Fig 1). Therefore, PTTH-induced phototaxis switch 245 drives Leptinotarsa larvae to burrow into soil.  Table). However, the reduced PTTH signal cannot trigger the light and construct chambers before pupation. JH/PTTH signaling is thus at the core of a hormonal 296 network that coordinates developmental progression and appropriate phototactic behavior to 297 maximize insect fitness. We propose a model summarizing these findings (Fig 5E).  Specific primers used to clone the fragments of dsRNAs were listed in S1 Table. These

391
The same method as previously reported [60] was used to introduce dsRNA into larvae.

392
Potato leaves were immersed with a bacterial suspension containing a dsRNA for 5 s, and LdHR3, the newly-ecdysed fourth-instar larvae were used. For other dsRNA feeding 397 bioassays, the newly-ecdysed third-instar larvae were used. The larvae were starved for at 398 least 4 h prior to the experiment. Then, ten larvae were transferred to each dish as a repeat.

399
For each treatment, three repeats were set. The larvae were allowed to feed treated foliage for 400 3 days (replaced with freshly treated ones each day).  LdRP18, LdARF1 and LdARF4, see S1 Table)