**3. Discussion**

For research of small molecular drugs, it is of grea<sup>t</sup> importance to elucidate the target protein in special cells or tissues. In this study, we conducted the experiments of BLI, CETSA, qPCR, RNAi, and enzyme activity, as well as modeling and docking analysis. As results, we carefully validated that BmArgRS and BmLamin-C are the DA-binding proteins and elucidated their interaction modes. Obviously, it is significant for further developments of DA-like drugs and researches of target proteins. However, BmPRP1 was not DA-binding protein. In fact, similar to other RNA helicases, BmPRP1 was an important component of constitution of stress granules (SGs) under stimuli [16], while SGs, composed of mRNA and ribosomal subunit, expression initial factor and RNA-binding proteins, appear in cellular stress conditions such as hypoxia, oxidative stress, and virus infection so as to stall translation and expression for energy saving [17]. Interestingly, in our previous study, we found a DA-binding protein (BmTudor-sn) was also component of SGs [8]. In addition, there is a report showed that destruxin can induce oxidative stress in insect [18]. Therefore, it is more likely that DA injures cells in sorts of unknown mechanism to trigger mRNA translation and protein synthesis of related SGs components such as BmPRP1. It probably leads to a false-positivity of BmPRP1 in DARTS experiments.

Aminoacyl tRNA synthetases play an important role in protein synthesis and are usually considered as medicine targets [19–21]; among them, ArgRS is involved in the formation of arginyl-tRNAArg complex in peptide chain extension [22]. Sequence analysis indicates that BmArgRS contains a catalytic core domain of arginyl-tRNA synthetases in the region of 216–481. Interestingly, in this study, we proved that DA binds to a pocket of BmArgRS in the conserved active catalysis sites. It might be speculated that DA suppresses the synthesis of some proteins containing arginine, especially the arginine rich basic proteins (histone), subsequently causes and leads to cells chaos in insects. Our research results give some clues to development of ArgRS inhibitors. Although DA interacts with BmArgRS at lower level as 10−<sup>5</sup> M, more other chemicals with higher bioactivity can be found on the basis of the quantitative structure-activity relationship (QSAR) of molecules and BmArgRS.

Lamins-C has many functions which usually acts as intermediate filament proteins providing stability and strength to cells, as supporting (sca ffolding) components of the nuclear envelope regulating the movement of molecules into and out of the nucleus, and as a role in regulating the activity (expression) of certain genes [12]. Previously, researchers found out that DA showed no a ffinity to nucleus, but was abundant in cytosol [23]. As above indicated, nonpolar DA might access into nucleus at inner membrane to bind BmLamin-C but does not result in mechanical stress. In cytoplasm, lamins mainly function as response to eliminate heat shock and oxidative stress [7]. Interestingly, DA induced heat shock proteins upregulation and oxidative stress in cells. It is suggested that DA binds to BmLamin-C leading to maintain unsuitable environment in cells which constant to previous study that DA acts as immunosuppressor [2]. Intriguingly, C-terminal lamin tail domain is highly conserved and contains immunoglobulin fold. However, researches of Lamin-C were insu fficient in *B. mori*, so it is hard to in-depth study in function with DA and make an assumption. But it provides new evidences and insight to further study owing to several nuclear related candidate proteins we have.

Reviewed in previous research on molecular mechanism of DA, the widely accepted hypothesis is that DA acts as immunosuppressor to inhibit innate immune system when *M. anisopliae* infected host. However, traditional molecular mechanism of DA target sites such as hemocyte and hemolymph are obscure. Here, we can speculate that DA stresses and leads to cell in uncomfortable environment and the meantime DA binds to those stress induced proteins to repress host defense as an immunosuppressor. Because we previously found that DA bond to heat shock protein [10], stress granule protein BmTudor-sn [8] and immunophilin peptidyl-prolylcis-transisomerase (BmPPI) [9], and interacted with BmArgRS to inhibit protein synthesis and bond to stress support protein BmLamin-C as well in this study.

In conclusion, we investigated the interaction between DA and three candidate proteins, and found that BmArgRS and BmLamin-C are DA-binding proteins but not BmPRP1. In addition, structurally, DA interacts with BmArgRS at Lys228, His231, Gln437 and Lys475 in catalysis active domain and with BmLamin-C at Lys528 and His524 in lamin tail domain. The above findings would o ffer new insight and evidence to DA target protein discovery.

#### **4. Materials and Methods**

#### *4.1. Cell Culture and Destruxin A*

The *Bombyx mori* Bm12 cell line was donated by our colleague Cao Yang (College of Animal Science, South China Agricultural University) and cultured in Grace's culture medium (Hyclone, Pittsburgh, MA, USA) and 10% fetal bovine serum (Gibco, Waltham, MA, USA). Cells were cultured at 27 ◦C and maintained at over a period of 2–4 days.

Destruxin A (DA) was isolated and purified from the *Metarhizium anisopliae* var. *anisopliae* strain MaQ10 in our laboratory [24]. Stock solution is 10,000 μg/mL in dimethyl sulfoxide (DMSO, Sigma-Aldrich, Darmstadt, Germany).

#### *4.2. Bio-Layer Interferometry (BLI)*

All proteins were prepared by expression in *E. coli*, and were tagged with His-tag and purified by nickel a ffinity chromatography. Protein accession number of BmArgRS, BmLamin-C, and BmPRP1 are XP\_004931696.1, XP\_004930078.1, and XP\_004926349.1, respectively.

BLI analysis was performed on a ForteBio OctetQK System (K2, Pall Fortebio Corp, Menlo Park, CA, USA) [25]. Generally, the protein samples were coupled with a biosensor for immobilization. Dilutions of DA were used for treatment. PBST buffer (0.05% Tween20, 5% DMSO) was used for the running and dilution buffers. The working procedure was baseline for 60 s, association for 60 s, and dissociation for 60 s. Finally, the raw data were processed with Data Analysis Software (9.0, Pall ForteBio Corp, Menlo Park, CA, USA).

#### *4.3. Cellular Thermal Shift Assay (CETSA)*

The Bm12 cell line was used to conduct the CETSA experiments [26]. Firstly, Bm12 cells were treated with 100 μg/mL DA and divided into 8–10 aliquots. After heated at 37–58 ◦C at 3 min, cells were lysed by a freeze-thawcycle. After centrifugated, the supernatants of the lysed cells were used for the western blot analysis.

#### *4.4. RNAi and Toxicity Assessment RNAi*

SiRNAs were prepared by synthesis in vitro. The sequence of BmLamin-C siRNA: 5-GCUGAUACCCGUAAGACUUTT-3 and 5-AAGUCUUACGGGUAUCAGCTT-3. The sequence of BmArgRS: 5-GCGAUCAAGAAGGAAGCUATT-3 and 5-UAGCUUCCUUCUUGAUCGCTT-3. SiRNA and FuGENE (Promega, Beijing, China) transfection reagen<sup>t</sup> were each diluted in serum-free medium and then mixed. The mixture was added to Bm12 cells and DA treatments. Toxicity assessment was performed by LDH-Glo™ Cytotoxicity Assay (Promega, Beijing, China) and detected in Synergy™ H1 (BioTek, Winooski, VT, USA).

#### *4.5. Survey of Gene Expression*

Expression of target gene was measured by RT-qPCR. Primer of BmLamin-C: 5-CTTCACCACGGCTCTGCTCAAC-3 and 5- TGCGGCAATTCTCTTCACCTTCG-3. Primer of BmArgRS: 5-GAGGTTAGAAGAGCGAACCACCAG-3 and 5-CGCCGCTGAAGATTCGGTCTC-3. Primer of BmPRP1: 5-GGTGCTGTGATGGCGGAGTTC-3 and 5-GCATGGCAGTGATGGAGAGGATC-3. The qPCR reactive program was subjected to 39 cycles at 95 ◦C for 10 s, 60 ◦C for 10 s, 72 ◦C for 30 s, then 95 ◦C for 10 s, and 65–95 ◦C for 5 s. The experiment was repeated three times. The silkworm GAPDH (glyceraldehyde-3-phosphate dehydrogenase) gene was taken as the reference gene. The qPCR data were analyzed by using the 2−ΔΔCt method. The means and DMRT (Duncan multiple range test) were evaluated by employing SPSS software (IBM, Armonk, NY, USA).

#### *4.6. Homology Modeling and Molecular Docking*

The target sequence of BmArgRS and BmLamin-C were acquired from Uniprot, with the Uniprot ID H9JHD3 and H9J2B5. Template crystal structures of RARS and LAMC were identified through BLAST and downloaded from RCSB Protein Data Bank (PDB ID were 4R3Z1 and 6GHD2). Homology modeling and docking were conducted in MOE v2014.09014 (Chemical Computing Group, Montreal, Canada).

#### *4.7. Circular Dichroism Spectrum*

Circular dichroism (CD) experiments were conducted by Chirascan Plus V100 (Leatherhead, Surrey, United Kingdom) provided by Applied Photophysics Ltd. Analysis processing in 1.0 nm band width. Measurement range include 190~260 nm (far-UV region scan) and 250–340 nm (near-UV region scan). The test was repeated three times. All raw scanning data were processed in Pro-Data Viewer (Applied Photophysics, Leatherhead, Surrey, United Kingdom) with subtract baseline and smoothing analysis.

**Author Contributions:** Conceived the experiments and revised the manuscript, Q.H.; Performed the experiments, analyzed the data, and wrote the paper, J.W.; Analyzed the data, Q.W. and F.Y. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research is supported by National Natural Science Foundation of China (U1901205).

**Conflicts of Interest:** The authors declare no conflict of interest.
