Flight muscles degenerate by programmed cell death after migration in the wheat aphid, Sitobion avenae

Objective Previous studies showed that flight muscles degenerate after migration in some aphid species; however, the underlying molecular mechanism remains virtually unknown. In this study, using the wheat aphid, Sitobion avenae, we aim to investigate aphid flight muscle degeneration and the underlying molecular mechanism. Results Sitobion avenae started to differentiate winged or wingless morphs at the second instar, the winged aphids were fully determined at the third instar, and their wings were fully developed at the fourth instar. After migration, the aphid flight muscles degenerated via programmed cell death, which is evidenced by a Terminal deoxynucleotidyl transferase dUTP-biotin nick-end labeling assay. Then, we identified a list of differentially expressed genes before and after tethered flights using differential-display reverse transcription-PCR. One of the differentially expressed genes, ubiquitin-ribosomal S27a, was confirmed using qPCR. Ubiquitin-ribosomal S27a is drastically up regulated following the aphids’ migration and before the flight muscle degeneration. Our data suggested that aphid flight muscles degenerate after migration. During flight muscle degeneration, endogenous proteins may be degraded to reallocate energy for reproduction.

IFM degeneration is previously depicted as a regulated physiological process (7,8) in insects including the vetch aphid, Megoura viciae (4) and the pea aphid, Acyrthosiphon pisum (5). The regulation of IFM degeneration involves multiple factors: (i) juvenile hormone (JH), JH treatment induces IFM degeneration (8)(9)(10). (ii) neural factor, denervated muscles in the presence of JH initiated degeneration earlier than the innervated muscles in cricket (11). (iii) specific protein, several specific proteins were induced despite that the overall protein synthesis decreases during A. pisum IFM degeneration (8). For example, ubiquitin, a marker for programmed cell death (PCD), accumulates when aphids undergone extensive IFM degeneration (7,8). Although IFM degeneration has been suggested as an active PCD, limited is the direct evidence and unclear is the underlying molecular mechanism.
The wheat aphid, Sitobion avenae, is a major grain crop pest. In China, S. avenae migrates from southern to northern China (12) and causes ~10% wheat yield losses every year (13).
Here using S. avenae, we investigated aphid wing development and PCD during IFM degeneration. Then we identified genes involved in IFM degeneration by comparing gene expression pre-and post-migration. One differentially expressed gene, ubiquitin-ribosomal S27a (RPS27a) was analyzed for its expression during IFM degeneration.

Aphids
We generated a S. avenae population with identical genetic background using a single wingless aphid. Aphids were raised on wheat seedlings >30 generations of parthenogenesis at 22°C with a 16h light/8h dark photoperiod.
We obtained the winged aphids by manipulating aphid densities. Under low-density, one wingless aphid was reared on a joint-stage wheat to maintain the wingless morph. Under high-density, 80 wingless adult aphids were reared on a single ripe wheat to induce the winged morph (6).
To investigate flight muscle development and degeneration, we collected the winged aphids every 24h from eclosion (0 day) to migration (5 th day), to reproduction (8 th day) until death observed. For each timepoint, half of the aphids were kept in petri dishes with moist filter paper for morphological, histological, and apoptosis examinations. The other half of aphids were treated following a freeze-drying procedure (14), then dissected for qPCR analyses.

Morphological and histological examination
The external morphology of aphid thorax was observed using a scanning electron microscope (SEM). Aphids were fixed in 3% glutaraldehyde for 24 hrs and transferred to 1% osmic acid. Then aphids were saturated with ethanol, exchanged using isopentyl acetate, and dried in a Hitachi CO 2 Critical Point Dryer system. Aphids were then coated with gold (Au) in a sputter coater (Hitachi, IB-5) and imaged under a Hitachi S-570 SEM.
The inner morphology of aphid thorax using histological staining. Aphids were fixed in 4% paraformaldehyde for 4 hrs. The specimens were dehydrated in a serial of ethanol solution (70%, 80%, 90%, 100%, 10 mins/each), cleared in xylene, and embedded in paraffin.
Serial sections (5μm) were cut and stained with hematoxylin and eosin for imaging ( Fig   S1).
Terminal deoxynucleotidyl transferase dUTP-biotin nick end labeling (TUNEL) assay To examine apoptosis, we performed a TUNEL assay using an in situ apoptosis detection kit (Boster, China). Briefly, paraffin-embedded aphids were sliced into 5μm serial sections, which were rehydrated in xylene 20 mins and a serial of ethanol solutions (100%, 90%, 80%, 70%, 10 mins/each). Then all specimens were permeabilized using Proteinase K Using TUNEL, we also examined the IFM degeneration on fasted aphid. For fasting, newly eclosed winged aphids were transferred to water-soaked sponges. Every 6 hrs, aphids were transferred back to rearing plants for 2 hrs to avoid death. In parallel, aphids reared on plants were collected as a control.

PCR)
To identify genes differentially expressed pre-and post-migration, we performed tethered flight using 3-4 days post-eclosion winged aphids. For pre-migration, aphids were tethered (not flighted) and flash-frozen. For post-migration, aphids were tethered and flighted for 24 hrs. Aphid flight were monitored using a flight-mill program (15) and flighted aphids were collected for analyses.
For DDRT-PCR, we designed three one-base anchored oligo-dT 3' primers and eight arbitrary 5' primers (13-mers) designed according to the RNAimage DD Kits (GenHunter) (Table S1). We performed a PCR with each anchor and arbitrary primer pair (3 x 8 pairs) using cDNAs generated from RNAs that extracted from single aphid. The PCR products were visualized using 6% SDS polyacrylamide gel electrophoresis ( Fig S2). Differentially expressed genes were excised from the gel and purified for sequencing. Obtained sequences were blast annotated on NCBI.
The dynamic expression of RPS27a The sequence of RPS27a was obtained using a rapid amplification of cDNA ends method (3'-RACE). Partial RPS27a sequence containing the start codon was amplified using UbS/UbA primers (Table S2) designed based on A. pisum RPS27a. The complete C-terminal of RPS27a was obtained using a RACE kit (Takara, China) with gene specific primers (GSP1/GSP2) and the kit-provided outer/inner primers (Table S2). To check the conservation of RPS27a, the nucleotide and deduced amino acid sequences were aligned with homologous from other insects using DNAMAN (Fig S3).
We quantified the RPS27a expression during flight muscle development and degeneration using qPCR in tissues including head, thorax, and abdomen. To calculate gene expression, a credible standard curve was constructed using a series of 10x dilution of a standard sample. The experiment was repeated three times and each experiment includes 3-4 technical replicates for each sample.

Flight muscle development in S. avenae
The external and internal structure in the winged and wingless aphids at the 1 st instar was similar (Fig S1A, a). Consistent to A. pisum (6), wing primordia were observed in the internal structures of nymphs in both morphs (Fig S1a). From the 2 nd instar, wind primordia developed and enlarged in the winged morph (Fig S1b'), whereas disappeared in the wingless morph (Fig S1b). At the 3 rd instar, swollen structures appeared in the winged morph ( Fig S1C, C'), which later developed into wing bud. Flight muscles fibers also differentiated (Fig S1c'); the corresponding area was occupied by fat bodies in the wingless morph hereafter (Fig S1c). To the 4 th instar, wing buds enlarged into a plate shape; the wing epithelia were folded in a complicated structure, differentiated between the forewings and hindwings (Fig S1d'). Wing hair sensilla were also seen in the winged morph (Fig S1E, F). To adults, the wings were fully developed (Fig S1E, F).

IFM degeneration after S. avenae migration
Histological examination showed that aphid flight muscle degenerate after migration. The flight muscle of an alatae adult was plump at the 1 st day after eclosion till the 5 th day ( Fig  1A, 1B). From the 9 th day, the flight muscle started to degenerate. The myofibrils appear to be thin. The diameter of the myofibrils was reduced, and the volume of the interfibrillar sarcoplasmic region increased ( Fig 1C); and the degradation continues through the 11 th day (Fig 1D). To the 13 th day, the flight muscle was degenerated completely (Fig 1E). The contractile fiber vanished, and the intact flight muscle was not visible.
IFM degeneration is a PCD process IFM degeneration after migration can be attributed to PCD. In sections of aphid pterothorax stained in the TUNEL experiment, no apoptotic signals were observed in alatae aphid before 6 th day after eclosion (Fig 2A, B). The first apoptotic signal appeared in aphids at the 7 th day as brownish yellow grains (Fig 2C), which is similar to the apoptosis signals in the positive controls. In parallel, we did not find any apoptotic signals in the muscles of the 1 st , 3 rd , and 7 th day fasted aphids (Fig 2D, E, F).
Differentially expressed genes in aphid thorax preand post-migration We identified 36 differentially expressed genes that were annotated and classified into three groups: 1) genes related to apoptosis, including the apoptosis marker, RPS27a; 2) genes related to metabolism, including energy re-allocation genes; and 3) genes with unknown functions (Table S3).

S. avenae RPS27a and its dynamic expression
The S. avenae RPS27a encodes a 150 aa protein (76 aa ubiquitin monomer + 74 aa ribosomal protein). The RPS27a are highly conserved across different insect species (84-93.3%) with the highest identity to A. pisum ( i.e. 93.3%) (Fig S3). The divergence of RPS27a was mainly from the N-terminal ribosomal protein, whereas the ubiquitin monomers were almost identical (Fig S3).
The expression of S. avenae RPS27a change dynamically pre-and post-IFM degeneration.
Initially after eclosion, RPS27a remains a constant low expression in all head, thorax and abdomen tissues (Fig 3). The expression started raising from the 3 rd day post-eclosion.
RPS27a expression significantly increased and is significant higher in abdomen than head and thorax in the 6 th day, which is one day post-migration and one day before we detected the apoptotic signals (Fig 2C). Right after migration, a sharp decrease occurred in the 7 th day and the apoptosis signals were detected (Fig 3).

Discussion
IFM degeneration has been mostly studied physiologically in insects such as fruit fly (16), crickets (17)(18)(19)(20), bugs (21,22), and some aphid species (7). Here, we depicted the flight muscle development and degeneration in S. avenae and identified some genes that may participate in the degeneration process.
IFM degeneration for energy re-allocation A trade-off has been proposed between IFM degeneration and reproduction with regard to energy allocation (23). IFM degeneration is regulated throughout development in many insects (21)(22)(23)(24)(25), and the products of IFM degeneration are considered to be involved in insect reproduction (26,27). Here, we identified two groups of genes that differentially expressed pre-and post-aphid migration, including genes relative to cell apoptosis and genes involved in metabolism (Table S3). We hypothesize that aphid IFM degeneration is an active process that first degrades flight muscles by triggering cell apoptosis after migration, which provides energy resources for oogenesis and reproduction by accelerating metabolism (5,26,28,29).
Ubiquitin function as genetic marker for aphid IFM degeneration Ubiquitin degrades proteins in eukaryotic cells. During eukaryotic cell apoptosis, many short-lived proteins are subjected to ubiquitination and triggers different degenerative processes (30,31). For example, RPS27a can regulate cell apoptosis by mediating P53 (an apoptosis gene) via inhibiting E3 ubiquitin-protein ligase Mdm2 (32). Kobayashi found that ubiquitin is one of the regulators of flight muscles degeneration in A. pisum (8). The increased activity of the ubiquitin indicates a preferential role for apoptosis in turn decreases muscle function.
Here, we found the S. avenae RPS27a was significantly differentially expressed pre-and post-migration. RPS27a showed a sharp increase before migration and a sudden decrease after migration, which indicated that aphids have been prepared for IFM degeneration.
During IFM degeneration, RPS27a may be a trigger or recognized as a signal for other genes in the cell apoptosis pathway to regulate IFM degeneration (30)(31)(32). RPS27 may also be used to degenerate the waste proteins to facilitate energy translocation and reproduction (33).
Flight muscle is crucial to migratory insects. The timing of migration and IFM degeneration determines whether these insects can locate preferable plant hosts or better habitats (34). To illustrate the mechanism of IFM degeneration will not only advance our understanding of insects' migratory behavior, but also shed light on the regulation of energy allocation from migration to reproduction.

Limitations
(1) Aphids flighted less distance in the tethered flight. In nature, the timing of aphid migration and IFM degeneration may be different.
(2) The differentially expressed genes pre-/post aphid migration can be further explored using next generation sequencing.

Availability of data and materials
All data generated or analyzed during this study are included in this published article and its supplementary information files.  The dynamic expression of S. avenae ubiquitin-ribosomal S27a gene. The aphids we collected are from eclosion (0 day) to breeding finished (8th day). The expression in all three tissues always show stable and relatively lower, except a sharp increase in 5th day, with higher in abdomen than in head and thorax.

Supplementary Files
This is a list of supplementary files associated with the primary manuscript. Click to download. TableSvFinal.pdf