Superoxide Dismutases, SOD1 and SOD2, Play a Distinct Role in the Fat Body during Pupation in Silkworm Bombyx mori

One way that aerobic biological systems counteract the generation of reactive oxygen species (ROS) is with superoxide dismutase proteins SOD1 and SOD2 that metabolize superoxide radicals to molecular oxygen and hydrogen peroxide or scavenge oxygen radicals produced by the extensive oxidation-reduction and electron-transport reactions that occur in mitochondria. We characterized SOD1 and SOD2 of Bombyx mori isolated from the fat body of larvae. Immunological analysis demonstrated the presence of BmSOD1 and BmSOD2 in the silk gland, midgut, fat body, Malpighian tubules, testis and ovary from larvae to adults. We found that BmSOD2 had a unique expression pattern in the fat body through the fifth instar larval developmental stage. The anti-oxidative functions of BmSOD1 and BmSOD2 were assessed by exposing larvae to insecticide rotenone or vasodilator isosorbide dinitrate, which is an ROS generator in BmN4 cells; however, exposure to these compounds had no effect on the expression levels of either BmSOD protein. Next, we investigated the physiological role of BmSOD1 and BmSOD2 under environmental oxidative stress, applied through whole-body UV irradiation and assayed using quantitative RT-PCR, immunoblotting and microarray analysis. The mRNA expression level of both BmSOD1 and BmSOD2 was markedly increased but protein expression level was increased only slightly. To examine the differences in mRNA and protein level due to UV irradiation intensity, we performed microarray analysis. Gene set enrichment analysis revealed that genes in the insulin signaling pathway and PPAR signaling pathway were significantly up-regulated after 6 and 12 hours of UV irradiation. Taken together, the activities of BmSOD1 and BmSOD2 may be related to the response to UV irradiation stress in B. mori. These results suggest that BmSOD1 and BmSOD2 modulate environmental oxidative stress in the cell and have a specific role in fat body of B. mori during pupation.


UV irradiation
Day 3 fifth instar larvae were treated with ultraviolet (UV) rays using UVL-56 (1350 μW/cm 2 , UVP) for 1, 2, 6 and 12 hours (4.86, 9.72, 29.2 and 58.32 J/cm 2 ). We prepared a cardboard box with two stands at either side to position two UVL-56 lamps at the top of the box (S1 Fig.). Irradiation distance was adjusted to 7.5cm from the face of the UVL-56 lamps to inside of the plastic box. Silkworm larvae were placed on the inside of the box for UV irradiation, after which they were fed artificial diet for 16 hours until dissection of the fat body.

Microarray analysis
Total RNA was isolated from the fat body of day 3 fifth instar non-irradiated and UV-irradiated (29.2 and 58.32 J/cm 2 ) larvae using RNeasy Mini Kits (Qiagen, Valencia, CA) and quantified on an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA). Microarrays of Cy3-labeled cRNA were prepared for non-irradiated and UV-irradiated larvae (n = 5 each). Briefly, 300 ng of total RNA was processed to Cy3-labeled cRNA using an Agilent Quick Amp Labeling Kit (Agilent Technologies) according to manufacturer instructions followed by purification with an RNeasy Mini Kit (Qiagen) and quantification on an Agilent 2100 Bioanalyzer (Agilent Technologies) and a NanoDrop 1000 spectrophotometer (Thermo Scientific, Waltham, MA). Hybridization was performed using a Gene Expression Hybridization Kit (Agilent Technologies) for which 1.65 μg of Cy3-labeled cRNA was mixed, fragmented and hybridized on a silkworm 4 × 44 K custom oligo-microarray slide containing 43,864 spots of 60-mer oligonucleotides constructed from 17,615 EST silkworm sequences (Agilent Technologies) and incubated at 65°C for 17 hours with shaking at 10 rpm. UV-irradiated and non-irradiated (control) samples were applied to a single array slide. The arrays were washed in Agilent Gene Expression Wash Buffer 1 (Agilent Technologies) for 1 min at room temperature and Agilent Gene Expression Wash Buffer 2 (Agilent Technologies) for 1 min at 37°C. The intensity of hybridized probes was detected with an Agilent G2565BA Microarray Scanner (Agilent Technologies) set to 5-μm scan resolution, and signals were extracted with G2565AA Feature Extraction Software v.9.5 (Agilent Technologies). Raw data collected in text files were normalized 'per spot' and 'per chip' using the GENESPRING GX v.10.3 program (Agilent Technologies). The microarray data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus (GEO) database under GEO accession numbers GSM1346427, GSM1346428, GSM1346429 and GSM1346430 (series GSE55816).

Microarray Data analysis
TIBCO Spotfire (TIBCO Software, Inc., Somerville, MA, USA) was used for hierarchical clustering and identification of gene sets differentially expressed in the UV irradiation treatment. Differential expression analysis for gene sets after UV irradiation was also performed with TIBCO Spotfire. Differences in gene expression greater than 2-fold and with P-values less than 0.05 were considered to be significant. The silkworm-retrieved gene sets that were up-and down-regulated by UV irradiation were converted to corresponding human homologous genes by a systematic BLAST search (tblastx), as described previously [18]. Gene set enrichment analysis (GSEA) was then performed for up-and down-regulated gene sets for the UV-irradiated mRNA expression profiles. A database for annotation, visualization and integrated discovery (DAVID) functional annotation tools [19,20] was used to identity UV-irradiation-specific gene sets in the KEGG pathway and Gene Ontology (GO) biological process category.

Preparation of recombinant BmSOD1 and BmSOD2
Primers for PCR were designed from BmSOD1 and BmSOD2 cDNA sequences obtained from GenBank and RefSeq (BmSOD1, Accession no. AB179561; BmSOD2, AB190802). Total RNA was extracted from the fat body of day 3 fifth instar larvae using an RNeasy mini kit (Qiagen). DNase-treated total RNA was processed for cDNA synthesis using oligo(dT) 12-18 primers and SuperScript II reverse transcriptase (Invitrogen). The ORFs of BmSOD1 and BmSOD2 were amplified by PCR using PfuTurbo DNA polymerase (Agilent Technologies) with the following primers: BmSOD1, 5 0 -CCCGCCAAAGCAGTTTGCGTACTTC-3 0 and 5 0 -TT AAATCTTGGCCAAGCCAATGACT-3 0 ; BmSOD2, 5 0 -TTAATGTCACAAAGGATTGGAT CA-3 0 and 5 0 -TCACTTGAGCGCTTTTTCATATCT-3 0 . Products were cloned into prokaryotic expression vector pTrcHis-TOPO using a TOPO TA cloning kit (Invitrogen) and were expressed in Escherichia coli XL-1 blue strain as fusion proteins with N-terminal Xpress tags. The nucleotide sequence was confirmed by DNA sequencing. Recombinant BmSOD1 and BmSOD2 expressed in E. coli were purified with HIS-Select spin columns (Sigma, St. Louis, MO) according to supplier instructions.

Immunoblotting
First, we examined the specificity of antibodies against BmSOD1 and BmSOD2 using the following samples: 10 μg of HeLa cell lysate, 10 μg of BmN4 cell lysate, 10 μg of larvae testis lysate, 0.5 μl of recombinant BmSOD1 with Xpress-tag, and 0.1 μl of recombinant BmSOD2 with Xpress-tag as a positive control. Also, the tissue distribution of BmSOD1 and BmSOD2 was determined for the silk gland, midgut, fat body, Malpighian tubules, testis and ovary from day 3 fifth instar larvae or the fat body from day 0 to 6 fifth instar larvae. The distribution of BmSOD1 and BmSOD2 in the whole body from stages including first to fifth instar larvae, pupae and adults was also determined. All tissues were homogenized in RIPA lysis buffer composed of 50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% Nonidet P40, 0.5% sodium deoxycholate, 0.1% SDS (Sigma-Aldrich, St. Louis, MO, USA) and protease inhibitor cocktail (Roche Diagnostics, Mannheim, Germany), followed by centrifugation at 10,000 × g for 15 min. Protein concentration was determined by a BCA protein assay kit (Thermo Scientific Co., Ltd., Rockford, IL). To identify the presence of BmSOD1 and BmSOD2 in different tissues, protein samples (10 μg) were separated on SDS-PAGE, transferred to nitrocellulose membranes using the method of Towbin et al. [21], and immunoblotted using rabbit anti-SOD1 antibody 1:2000, for BmSOD1 (ab13498; Abcam Cambridge, UK), rabbit anti-SOD2 antibody 1:2000, for BmSOD2 (ab13534; Abcam) and goat anti-rabbit IgG-conjugated horseradish peroxidase (HRP) 1:2000 (sc-2004; Santa Cruz Biotechnology, Santa Cruz, CA) Membranes were developed using a chemiluminescent substrate (Pierce, Rockford, IL).
After antibodies were stripped by incubating the membranes at 50°C for 30 min in stripping buffer composed of 62.5 mM Tris-HCl, pH 6.8, 2% SDS and 100 mM 2-mercaptoethanol, membranes were processed for relabeling with different antibodies. Protein levels were measured by ImageJ ver. 1.37 c (http://imagej.nih.gov/ij/index.html).

Quantitative RT-PCR
In order to quantify RNA expression levels, total RNA was extracted from pooled fat body tissue dissected from day 3 fifth instar larvae (n = 3-5 each) using an RNeasy Mini Kit (Qiagen). One-step RT-PCR was performed in 20 μl reaction volumes with 1 μg of RNA template and custom-made primers and probes (Table 1) with a TaqMan RNA-to-CT 1-Step Kit (Applied Biosystems, Foster City, CA), in accordance to manufacturer instructions. Quantitative RT-PCR (qRT-PCR) was performed on a 7500 Fast Real-Time PCR System (Applied Biosystems) following the Delta-Delta Ct method. Actin was utilized as an endogenous reference against which RNA expression levels were standardized, and all data were calibrated against universal reference data. Relative quantification (RQ) values represent the relative expression level against a reference sample. All sample sets were assayed in triplicate as technical replications.

Northern blot analysis
Total RNA derived from the fat body of day 3 fifth instar larvae was used. Total RNA (12 μg) was separated on a 1.5% agarose, 6% formaldehyde gel and stained with ethidium bromide. Then, the gel was transferred to a nylon membrane. DIG-labeled probes were synthesized using the PCR DIG probe synthesis kit (Roche Diagnostics, Mannheim, Germany) in accordance with supplier instructions using the following primers: BmSOD1, 5 0 -CACGAATTTGGTGACAAC ACAAATG-3 0 and 5 0 -TTAAATCTTGGCCAAGCCAATGACT-3 0 ; and BmSOD2, 5 0 -ATCAAC TGTCGACAGCTTCTGT-3 0 and 5 0 -TCACTTGAGCGCTTTTTCATA-3 0 . After pre-hybridization, membranes were hybridized with DIG-labeled probes at 50°C overnight. The specific reaction was visualized on Kodak XOMAT AR X-ray film using a DIG chemiluminescence detection kit (Roche Diagnostics). 18S ribosomal RNA (rRNA) was used as a control.
The size of the mRNA for both SODs was calculated using image analysis software CS Analyzer 3.0 (ATTO, Tokyo, Japan). A calibration curve was determined using the mobility of the DIG RNA ladder marker (Roche Diagnostics).

BmN4 cells treated with ROT or ISDN and detection of BmSOD1 and BmSOD2
BmN4 cells (2 × 10 6 ) were grown on 6-well Falcon plates (BD Biosciences, Franklin Lakes, NJ) and washed twice with PBS followed by treatment with TC-100 medium containing rotenone (ROT; 50 μM) dissolved in 0.1% DMSO or 100 μM of isosorbide dinitrate (ISDN; prepared immediately prior to use and kept in the dark) dissolved in 0.1% ethanol for 3 or 6 hours. Control experiments were performed with either 0.1% DMSO or 0.1% ethanol. Total protein extracts were prepared using RIPA buffer (Sigma-Aldrich) for immunoblotting.

Statistical analysis
Student's t-test was carried out using JMP 10.0 software (SAS Institute Japan Ltd., Tokyo, Japan), and P values of <0.05 were considered significant. Table 1. Primers and probes used for quantitative RT-PCR.

cDNA cloning and sequence analysis of BmSOD1 and BmSOD2
Using RNA isolated from the fat body of day 3 fifth instar larvae, the deduced ORF of BmSOD1 was 465 nucleotides coding for a protein with 154 amino acids, a molecular weight of 15,841 Da and a putative isoelectric point (pI) of 5.78, while the deduced ORF of BmSOD2 was 651 nucleotides coding for a protein with 216 amino acids, a molecular weight of 24,226 Da, and a putative isoelectric point (pI) of 9.18. A protein motif search revealed that BmSOD1 contains a copper/zinc superoxide dismutase domain (SOD_Cu, pfam; PF00080) at position 5A-149G and BmSOD2 contains the following domains: iron/manganese superoxide dismutases alpha-hairpin domain (Sod_Fe_N, pfam; PF00081) at position 20R-101N and an iron/manganese superoxide dismutase c-terminal domain (Sod_Fe_C, pfam; PF02777) at 105F-211V. Thus, both SODs were successfully cloned from the fat body of B. mori larvae. We localized the BmSOD1 gene to four blocks between bp 356678 and 360180 on chromosome 23 and the BmSOD2 gene to six blocks between bp 916397 and 921150 on chromosome 3 by linkage mapping with 28 chromosomes using SNP markers [22]. BmSOD1 and BmSOD2 have greater than 60% amino acid sequence identity to vertebrate SOD1 and SOD2 proteins, including those from Rattus norvegicus, Homo sapiens, Mus musculus, Danio rerio and Xenopus tropicalis. BmSOD1 and BmSOD2 also show high homology to insect SOD1 and SOD2 proteins, including those from Anopheles gambiae, Drosophila melanogaster, Plutella xylostella and Manduca sexta (Tables 2 and 3). In particular, BmSOD1 and BmSOD2 showed high similarity to proteins from lepidopteran insects. Alignment of the deduced BmSOD1 and BmSOD2 amino acid sequences and SOD1 and SOD2 orthologs from other species showed that the BmSOD1 protein sequence contains all of the conserved His and Asp residues ( Fig. 1-A, red asterisks). In addition, the BmSOD2 protein sequence contains all of the conserved His, Asp and Glu residues ( Fig. 1-B, red asterisks). These amino acid residues are involved in the coordination of the metal domain.
In the phylogenetic tree, vertebrate SOD1 and SOD2 and insect SOD1 and SOD2 were placed into four distinct clusters ( Fig. 1-C). Northern blot analysis revealed that both transcription products were single products with characteristic sizes: 936 bases for BmSOD1 and 922 bases for BmSOD2 (Fig. 2).

Specificity of antibodies against BmSOD1 and BmSOD2
Based on the conserved evolution of amino acid sequences of SOD proteins, we examined the utility of commercially available antibodies raised against human SOD1 and rat SOD2 to identify BmSOD1 and BmSOD2. Anti-SOD-1 antibody reacted with recombinant BmSOD1 protein as a 22-kDa band and with BmSOD1 in cell and tissue lysates from B. mori as a 16-kDa band ( Fig. 3-A, C). Anti-SOD-2 antibody reacted with recombinant BmSOD2 protein as a 30-kDa band and with BmSOD2 in cell and tissue lysates from B. mori as a 24-kDa band ( Fig. 3-B, D). Further, SOD1 antibodies did not recognize recombinant BmSOD2, and SOD2 antibodies did not recognize recombinant BmSOD1 ( Fig. 3-C, D). The molecular weight of the recombinant BmSOD1 and BmSOD2 proteins ( Fig. 3-C, D lanes 4 and lane 5) was slightly greater than the endogenous BmSOD1 and BmSOD2 proteins, as these proteins included the Xpress tag protein. Thus, we concluded that the commercial antibodies were useful for immunoblotting.

Identification of developmental stage and tissue-specific expression patterns of BmSOD1 and BmSOD2 by immunoblotting
Distribution of BmSOD1 protein and BmSOD2 protein expression by developmental stage and tissue is shown in Fig. 4. Both BmSOD proteins were expressed in the whole body through all developmental stages (Fig. 4 and S2-A Fig.). In addition, BmSOD proteins were expressed in various tissues from day 3 fifth instar larvae, but BmSOD2 protein expression levels were higher in the midgut and Malpighian tubules than in other tissues ( Fig. 4 and S2-B Fig.). To determine the distribution pattern of BmSODs, we studied the fat body from day 0 to 6 fifth instar larvae by immunoblotting. BmSOD proteins were expressed throughout the larval  developmental stages in the fat body, but BmSOD2 protein expression was low in day 1 and 6 fifth instar larvae (Fig. 5-A, B and S3 Fig.).

BmN4 cells treated with ROT, ISDN and detection of BmSODs
Next, we investigated the effect of chemical factors on the expression of BmSODs. We used ROT and ISDN as ROS generators, as these chemicals induced oxidative stress on BmN4 cells. Expression levels of both BmSOD proteins were the same for exposure to ROT for 3 and 6 hours (Fig. 6-A and S4 Fig.). With exposure to ISDN, a nitric oxide (NO) generator [23], expression levels of ISDN-treated BmN4 cells were the same as for the control (Fig. 6-B and  S5 Fig.). Thus, ROT and ISDN did not affect the expression levels of either BmSOD protein (Fig. 6 A, B).
BmSOD response to UV irradiation at mRNA and protein level in B. mori larvae Examination of expression levels of mRNA for both BmSOD by qRT-PCR after UV irradiation at 4.86, 9.72 and 58.32 J/cm 2 showed that BmSOD1 and BmSOD2 mRNA expression was significantly increased in the 9.72 J/cm 2 group (Fig. 7), while protein expression levels of BmSOD1 and BmSOD2 were slightly increased in the UV irradiation groups (Fig. 8 and  S6 Fig.).
To examine the differences in mRNA level due to UV irradiation, we performed microarray analysis. Gene set enrichment analysis by DAVID of genes in larvae without and with UV irradiation showed distinct features in gene functions. Genes up-regulated after 6 and 12 hours of UV irradiation were present at significant level in the insulin signaling (Table 4-A) and PPAR signaling pathways (Table 4-C), respectively.
Moreover, genes up-regulated and down-regulated after 6 hours of UV irradiation contained statistically significant Gene Ontology (GO) terms and included those involved in protein localization (GO:0008104) and oxidative stress response (GO:0006979), respectively (S1-A and B Table).

Discussion
In this study, we characterized the expression and distribution of two BmSOD proteins in B. mori. Both SODs had highly conserved amino acid sequences. Characterization of the B. mori homologs BmSOD1 and BmSOD2 by cDNA cloning from the fat body of fifth instar larvae confirmed the presence of His and Asp, which are key functional residues in both SODs ( Fig. 1-A, B). On a phylogenetic tree of SOD proteins, vertebrate SODs and insect SODs placed in distinct clusters ( Fig. 1-C). These results show that insect SODs may have distinct functions from vertebrate SODs. Each BmSOD was well conserved across the species and also, each BmSOD might have a distinct function. In mammals, SOD1-deficient mice appear normal at birth but show increasing levels of oxidative stress over time [24], while SOD2-knockout mice can only live for a short span (10 days) before showing cardiomyopathy [25], as SOD2 responds to numerous inflammatory and oxidative stimuli [26].
In insects, SOD1-null flies show markedly decreased lifespans and are highly susceptible to oxidation [27]. Diapause-associated protein 3 has a copper/zinc superoxide dismutase (Cu/Zn-SOD) domain and is involved in the regulation of diapause in Antheraea pernyi [28]. Hyphantria cunea SOD1 and SOD2 play a role in resistance to oxidative stress, and SOD2, in particular, plays a role with apo lipophorin III in apoptosis [29,30]. Apis cerana SOD2 transcript levels fluctuate across developmental stages and are involved in development, growth and environmental stress responses [31]. Thus, insect SODs have physiological functions that confer resistance to some types of oxidative stress that are common across various species but that also play species-specific roles.
Both BmSODs exist as single isoforms that map to different chromosomes in B. mori. Accordingly, B. mori SOD1 and SOD2 might have different roles within cells. Previous studies indicated that BmSOD1 and BmSOD2 mRNA are expressed in various tissues in day 3 fifth instar larvae [16,17] with BmSOD2 mRNA being more strongly expressed in the testis. In our current study, both SOD proteins were expressed in the whole body throughout all  developmental stages, and BmSOD1 protein was expressed uniformly in various tissues in day 3 fifth instar larvae. BmSOD2 protein expression levels were higher in the midgut and Malpighian tubules compared to levels in other tissues.
The fat body is an important organ with endocrine and storage functions similar to the vertebrate liver [32]. Thus, we examined developmental distribution of both BmSOD proteins in the fat body of fifth instar larva. BmSOD2 protein expression level was dramatically decreased in the fat body of day 6 fifth instar larvae.
To identify the factors that regulate expression levels of BmSOD1 and BmSOD2 proteins, we used two chemical agents that generate ROS. First, BmN4 cells were treated with ROT, which inhibits the mitochondrial electron transfer chain of mitochondrial complex I. ROT enhances the production of ROS and reduces the production of ATP, resulting in mitochondrial dysfunction [4]. However, expression of BmSOD1 and BmSOD2 proteins did not change following treatment with ROT. Then, BmN4 cells were treated with ISDN, a NO generator, and similarly, ISDN did not affect the protein expression levels of BmSOD1 and BmSOD2 (Fig. 6).
In contrast to the observation for the chemical induction of oxidative stresses, UV-irradiated BmN4 cells showed marked mRNA level changes for both BmSODs (data not shown) as was previously observed for mammalian SODs [33]. UV exposure was found to generate oxidative stress, to produce DNA damage and to rapidly enhance the activity of SOD in Helicoverpa armigera adults [34]. Consequently, examination of the physiological function of BmSOD1 and BmSOD2 proteins in B. mori larvae following UV irradiation using qRT-PCR and immunoblotting showed that the mRNA expression level of both BmSODs was significantly elevated in the 9.72 and 58.32 J/cm 2 UV irradiation groups, but protein expression levels were only    slightly increased in B. mori larvae treated with 58.32 J/cm 2 UV irradiation. Expression levels of BmSOD1 and BmSOD2 mRNA were not matched to protein expression.
Microarray analysis showed that genes up-regulated after 6 hours of UV irradiation predominantly contained insulin signaling pathway related genes (Table 4-A). Moreover, genes down-regulated after 6 hours of UV irradiation predominantly contained GO terms for oxidative stress response-related genes (S1-B Table). In this study, BmSOD1 and BmSOD2 mRNA expression levels were markedly up-regulated but protein expression levels were almost unaltered in the UV irradiated groups. These observations suggest that the translational efficiency of BmSOD1 and BmSOD2 may be decreased, or that the speed of destructive metabolism in BmSOD1 and BmSOD2 protein may be increased by protein phosphorylation. In fact, we found that many genes coding for protein kinase and phosphatase were included in the insulin signaling pathway that was significantly up-regulated after 6 hours of UV irradiation based on microarray analysis (data not shown). Therefore, expressed BmSOD1 and BmSOD2 proteins may be rapidly degraded in the body of B. mori.
B. mori larvae are known to accumulate uric acid (UA) as quite a urate granules, which causes a whitening of integument color [35], and UA plays a powerful role as a physiological antioxidant and in protecting individuals from environmental stress factors [36]. In the case of UV irradiation to the whole body of B. mori larvae, UA in the integument may also have played a role in providing protection against UV irradiation stress. Both BmSODs may play protective roles against UV irradiation stress. Accordingly, we will examine the relation between UA and BmSODs in the future studies.
There is another candidate biological factor that controls the expression level of BmSOD1 and BmSOD2 in B. mori-growth factors. Ecdysteroid and juvenile hormone (JH) titers are up-regulated in day 6 of fifth instar larvae hemolymph in M. sexta and B. mori [37,38]. Furthermore, JH esterase (JHE) plays an important role in the control of hemolymph JH titer and the induction of metamorphosis, and JHE mRNA expression is detected in the fat body from day 4 to 7 fifth instar larvae [39]. Also, insulin receptor substrate (IRS) and single insulin receptor (InR), which operate in the insulin/insulin growth factor signaling (IIS) pathway, are up-regulated in the fatbody of day 6 fifth instar B. mori larvae [39]. In addition, InR and IRS mRNA levels were found to be elevated by treatment with 20-hydroxyecdysone in B. mori [40]. Thus, it is possible that BmSOD2 is down-regulated in the fat body on the last day of the larvae stage by growth factors such as hormones. This suggests that BmSOD2 plays some role in pupation. However, we were unable to clarify whether BmSOD1 and BmSOD2 were related to these hormones due to the application of oxidative stress through the application of ISDN or ROT or UV irradiation in this study. Consequently, the distinct expression patterns in the tissues for these two SODs during the final larval stage suggest that each SOD may play distinct roles in the larval developmental stage.
In summary, we characterized two B. mori SOD proteins and their different roles in the fat body in response to environmental stress and in the metamorphosis from larval to pupal stages in B. mori. These results suggest that BmSOD1 and BmSOD2 modulate environmental oxidative stress in the cell and might serve as metamorphosis-related proteins in B. mori. In future studies, we plan to further investigate the advanced molecular mechanisms in pupation B. mori in vivo using a gene knockout approach.
Supporting Information  Table. (DOC)