Mechanism of Enhanced Bombyx mori Nucleopolyhedrovirus-Resistance by Titanium Dioxide Nanoparticles in Silkworm

The infection of Bombyx mori nucleopolyhedrovirus (BmNPV) in silkworms is often lethal. It is difficult to prevent, and its lethality is correlated with both viral particle characteristics and silkworm strains. Low doses of titanium dioxide nanoparticles (TiO2 NPs) can promote silkworm growth and improve its resistance to organophosphate pesticides. In this study, TiO2 NPs’ effect on BmNPV resistance was investigated by analyzing the characteristics of BmNPV proliferation and transcriptional differences in silkworm midgut and the transcriptional changes of immunity related genes after feeding with TiO2 NPs. We found that low doses of TiO2 NPs improved the resistance of silkworm against BmNPV by 14.88-fold, with the mortalities of the experimental group and control group being 0.56% and 8.33% at 144 h, respectively. The proliferation of BmNPV in the midgut was significantly increased 72 h after infection in both experimental and control groups; the control group reached the peak at 120 h, while the experimental group took 24 more hours to reach the maximal value that was 12.63 times lower than the control, indicating that TiO2 NPs can inhibit BmNPV proliferation in the midgut. Consistently, the expression of the BmNPV-resistant gene Bmlipase-1 had the same increase pattern as the proliferation changes. Immune signaling pathway analysis revealed that TiO2 NPs inhibited the proliferation of silkworm BmNPV to reduce the activation levels of janus kinase/signal transducer and activator of transcription (JAK/STAT) and phosphatidylinositol 3-kinase (PI3K)-Akt signaling pathway, while promoting the expression of Bmakt to improve the immunity. Overall, our results demonstrate that TiO2 NPs increase silkworm resistance against BmNPV by inhibiting virus proliferation and improving immunity in silkworms.

NPs' effect on BmNPV proliferation focusing on JAK/STAT and PI3K-Akt signaling pathways have important significance.

Insects and Chemicals
The silkworm strain was Jingsong × Haoyue, and the BmNPV strain was T3 (GenBank:L3318), both of which were preserved in our laboratory.
The preparation of anatase TiO 2 NPs was through controlling the hydrolysis oftitanium tetrabutoxide. The synthesis and characterization of TiO 2 NPs were following the method described by Yang et al. [32,33]. The average particle sizes of powders suspended in 0.5% w/v hydroxypropylmethylcellulose (HPMC) K 4 M solvent ranged from 5 to 6 nm after 12 h and 24 h incubation. As measured by DLS, themean hydrodynamic diameter of TiO 2 NPs in HPMC solvent ranged from 208 to 330 nm (mostly 294 nm), and the zeta potentials after 12 h and 24 h incubation were 7.57 mV and 9.28 mV, respectively, and more detailed characterization of TiO 2 NPs has been described by our team previously [33]. The morphology of the obtained TiO 2 NPs was characterized by a transmission electron microscope (TEM) (Hitachi H-600, Japan). The detection of BmVPV in the hemolymph was carried by a scanning electron microscope (SEM).
TRIzol, chloroform, isopropanol, RNasin Inhibitor, dNTP, SYBR Premix, and other routine chemical reagents were all purchased from TAKARA Biotechnology (Dalian) Co., Ltd. Primers were synthesized by Shanghai Sangon Biological Technology and Services Co., Ltd.

Silkworm Treatments
First to third instar larvae were reared with fresh mulberry leaves. From fourth instar, silkworms were reared in control or experimental zones, and each zone had 3 groups with 60 larvae in each group for the determination of morbidity and cocoon quality. The larvae of the experimental zones were continuously fed with TiO 2 NPs at 5 mg/L [34] until mounting. The larvae of the control zones were fed with mulberry leaves which treated with sterile water. All treated leaves were dried before feeding for three times each day. From fifth instar, silkworms were fed with leaves with BmNPV (titer: 5.6 × 10 6 polyhedral/mL). 100 g fresh mulberry leaves were dipped in BmNPV solution for 1 min and dried at room temperature before feeding. The rearing condition was long-day photoperiod (16: 8 h light/dark) at 25°C and approximately 70% relative humidity. After feeding with BmNPV, silkworms in both control zone and experimental zone were dissected to isolate midgut and fat body once every 24 h.

Investigation of Biological Characteristics
Seven days after mounting, cocoon quality was surveyed by analyzing number of cocooning and non-cocooning, number of dead worm cocoons, whole cocoon mass, and cocoon shell mass. The cocooning rate, rate of death worm cocoons, and ratio of cocoon shell were calculated: cocooning rate (%) = number of cocooning / (number of cocooning + number of noncocooning + number of dead worm cocoons) × 100; rate of death worm cocoons (%) = (number of dead worm cocoons / number of cocooning) × 100; ratio of cocoon shell (%) = (cocoon shell mass / whole cocoon mass) × 100.

Detection of BmNPV Proliferation in Silkworm
For real-time detection of BmNPV proliferation in silkworms, mixed genomic DNA was extracted from midgut and BmNPV which in the midgut. Genomic DNA extraction was following the method described by Hughes et al. [35]. Quantitative real-time PCR (qPCR) primers were designed based on the sequences of polyhedrin genes lef-1 and gp64. Bmactin3 was used as the internal reference gene. To avoid the interference from RNAs, the primers were designed to target introns (Table 1). qPCR analysis was performed using the ViiA 7 Realtime PCR System [36] with SYBR Premix Ex Taq (Takara) following previously described method by our laboratory [37,38].

Detection of Expression of Silkworm Anti-BmNPV Genes and Related Important Immune Signaling Genes
To explore the effects of TiO 2 NPs on silkworm anti-BmNPV innate immune system, the endogenous BmNPV-resistance gene Bmlipase-1 and the key genes of JAK/STAT and PI3K-Akt signaling pathways were selected for qPCR analysis. Total RNA was extracted from the midgut and fat body samples using Trizol reagent (Takara, Japan) and then treated with DNase to remove potential contamination from genomic DNA. RNA quality was assessed by formaldehyde agarose gel electrophoresis and was quantitated spectrophotometrically, and primer sequences showed in Table 1.

Western Blot Analysis
Fat body samples of the control and TiO 2 NPs treated groups were homogenized in lysis buffer supplemented with 1 mM of PMSF. The samples were centrifuged at 10,000 g for 10 min, and the supernatants were collected. The following procedure was carried out following Gu et al. [39]. A rabbit polyclonal phospho-Akt (Ser 505)-specific antibody or a rabbit polyclonal total Akt-specific antibody (Cell signaling, USA; 1:2000) was used as the primary antibody, and the HRP-conjugated goat anti-rabbit IgG (Santa Cruz Biotechnology, USA; 1: 5000) was used as the secondary antibody.

Data Processing
Statistical analyses were conducted using the SPSS 19 software. Data are presented as mean ± standard error (SE). One-way analysis of variance (ANOVA) was carried out to compare the differences of means among multi-group data. Dunnett's test was performed when each data set was compared with the solvent control data. Statistical significance for all tests was judged at a probability level of 0.05 (P<0.05).

Characterization of TiO 2 NPs
The size of the TiO 2 NPs was distributed from 5 to 6 nm as shown in the images of TEM (Fig. 1).

TiO 2 NPs Improves Silkworm Resistance to BmNPV
After feeding with BmNPV for 120 h, some silkworms in the control group showed BmNPV disease symptoms, manifested as white body color, bulged intersegmental membrane, and manic crawling ( Fig. 2A). As a contrast, silkworms in the experimental group grew well without disease symptoms (Fig. 2B). In order to detect the proliferation of BmNPV in vivo, 100 μL hemolymph was taken at 120 and 144 h for SEM analysis, respectively (Fig. 2C, E and D, F). BmNPV particles were observed in the control group, which were arranged on monolayer at 120 h but became aggregated at 144 h with apparently increased number of particles. In the experimental group, no BmNPV particles were observed at either 120 h or 144 h. These results indicated that TiO 2 NPs significantly inhibited the proliferation of BmNPV in silkworm larvae. As shown in Fig. 3, the average mortalities of larvae in the control group and the experimental group were 7.22% and 0%, respectively, at 120 h. Death of larvae was only observed at 144 h at a rate of 0.56%, while that of the control group reached 8.33%; at 168 h, the experimental group's mortality was 0.56%, compared with the control group's 10.56%. These results indicated that TiO 2 NPs not only delayed the onset of BmNPV disease in silkworm larvae but also significantly reduced larvae mortality (14.88-fold increased in silkworm resistance against BmNPV). As shown in Table 2, the larva survival rate of the experimental group at 99.44% ± 0.01% was significantly higher than that of the control group at 89.44% ± 0.02% (P<0.01); the cocooning rate of the experimental group was 49.13% ± 0.05%, significantly higher than the control group's 40.94±0.04% (P<0.01). The rate of death worm cocoons of the experimental group was 38.51±0.05%, significantly lower than the control group's 54.13±0.05% (P<0.05). In view of the survival rate and cocooning rate in H 2 O group was 100%, and there was no death worm   cocoons, we mainly focused on the differences after NPV infection in the latter experiment. As shown in Table 3, the whole cocoon mass of the control group was 1.67±0.21 g, similar to the experimental group's 1.73±0.19 g. The cocoon shell mass of the control group was 0.38 ± 0.009 g, significantly lower than that of the experimental group (0.41 ± 0.013 g) (P<0.05). The two groups' ratio of cocoon shell were 22.94±2.80% and 23.82±2.66%, respectively, not significantly different from each other. These results indicated that feeding with TiO 2 NPs significantly improved silkworm survival and cocoon and reduced the rate of death worm cocoons after BmNPV infection. Although TiO 2 NPs did not significantly improve whole cocoon mass and ratio of cocoon shell, cocoon production was still significantly increased due to significantly improved silkworm cocooning rates that increased the total number of cocoons.

TiO 2 NPs Affects BmNPV Proliferation in Silkworm Midgut
In order to detect BmNPV proliferation levels in silkworms accurately, genomic DNA was extracted from the mixture of silkworm midgut and BmNPV, and the relative copys of two essential genes for BmNPV amplification, lef-1 and gp64, were selected as the detection indicators for qPCR analysis with BmactinA3 as the internal reference gene. As shown in Fig. 4, the control group's relative copys of lef-1 was significantly higher than that of the experimental group. At 24 h and 48 h, lef-1's relative copys were relatively low in both control and experimental groups, the control group's were 12.69-and 13.02-fold of those of the experimental group, respectively. From 72 h, the relative copys of lef-1 were apparently increased in both groups, and the control group entered a rapid increase period from 96 h to 120 h and reached the maximum at 120 h, and then maintained at a stable level; the experimental group's rapid increase period was from 120 h to 144 h, and it reached the maximum at 144 h while reducing to 8.48% of the maximal level at 168 h. The peak value of the control group was 12.5-fold higher than that of the experimental group. BmNPV envelope protein gene gp64 showed similar amplification pattern as that of lef-1 (Fig. 5). The relative copys of gp64 of the control group were all higher than those of the experimental group at all periods after BmNPV infection. 24 h and 48 h after the BmNPV infection, the relative copys of gp64 were 2.18-and 1.13-fold of those of the experimental group, respectively. At 72 h, significant differences started to be observed, with the relative copys of gp64 of control group showing 6.98-fold of the experimental group. Similar to the amplification of lef-1, gp64 entered the rapid growth period also from 96 h to 120 h and reached the maximum at 120 h and maintained the level until mounting. In the experimental group, the rapid growth period was from 120 h to 144 h and the maximum value was reached at 144 h; it was reduced at 168 h to only 17.78% of the maximal level; the peak value of the control group was 12.76fold higher than that of the experimental group.
Therefore, BmNPV proliferation in the control group experienced the classic latency period, rapid growth period, and plateau period; in each period, its amplification level was significantly higher than that of the experimental group; the control group entered the rapid growth period and reached the peak value both much more earlier than the experimental group. It indicated that TiO 2 NPs inhibited the proliferation of BmNPV, delayed the emergence of the peak of virus proliferation, consistent with the results of larva morbidity. We also discovered that the amplifications of lef-1 and gp64 did not enter the plateau period after the peaks but were significantly decreased to 8.48% and 17.78% of the peak values, indicating that the inhibition of TiO 2 NPs changed the proliferation trend of BmNPV in silkworm midgut.

Transcriptional Characteristics of BmNPV-Resistance Relate Gene
Bmlipase-1 The transcription levels of Bmlipase-1 in both experimental and control groups were measured in this study to investigate the effects of different titers of BmNPV on the induction of Bmlipase-1 expression. As shown in Fig. 6, BmNPV infection led to mRNA levels of Bmlipase-1 first increasing then decreasing in both groups. In the control group, its transcription level reached the peak at 96 h, while the experimental group had the maximum level at 120 h with only 18.7% of the control level. In addition, the control group's peak value was maintained at about 3-fold of the experimental group's since 96 h. The experimental group's Bmlipase-1 level reached the maximum at 120 h but decreased to the level similar to the initial infection period at 168 h. These results indicated that the transcription of Bmlipase-1 was induced by BmNPV infection, and TiO 2 NPs decreased the induction of Bmlipase-1 by reducing the titer of BmNPV in silkworms. As a result, the occurrence of peak values was delayed by TiO 2 NPs, which consistent with the changes in larva mortality. In addition, the peak values of Bmlipase-1 transcription both occurred 24 h before the death of larvae. In the TiO 2 NPs-treated group, the

Expression Characteristics of Key Genes in Immune Pathway
The resistance of silkworms against BmNPV is associated with not only resistance genes but also immune signaling pathways. In this study, the transcription levels of some key genes in the JAK/STAT and PI3K-Akt pathways were measured. The transcription levels of JAK/STAT pathway marker gene Bmstat were already upregulated at 24 h after BmNPV infection (Fig. 7), with the control group's level being higher than the experimental group's. The control group's Bmstat transcription peaked at 120 h with 9.17 times of the level at 0 h. The transcription levels of Bmstat in the control group remained high level after reach the peak until mounting, which consistent with the trend of Bmlipase-1 expression. In the experimental group, Bmstat's relative transcription level remained low before 120 h, followed by increases until the peak at 144 h with 4.15-fold of the 0 h level. However, the peak value of Bmstat's transcription level was only 44.04% of that of the control group. At 168 h, the relative transcription level of Bmstat was Besides JAK/STAT immune signaling pathway, PI3K-Akt pathway is also correlated with BmNPV infection in insects. It has been reported that PI3K-Akt pathway is required for the replication of baculoviruses, and Bmpi3k activation increases AcMNPV production [14]. In order to confirm the effects of TiO 2 NPs on the PI3K-Akt signaling pathway response to BmNPV infection, the expression characteristics of Bmpi3k and Bmakt were examined in this study. As shown in Fig. 8, no Bmpi3k expression was detected 48 h after BmNPV infection; at 72 h, the control group showed upregulaton in Bmpi3k expression and achieved the maximum at 120 h, which was 15.99-fold of the level at 72 h; at 144 h and 168 h, its expression was downregualted to 8.77-fold and 5.66-fold of the level at 72 h, respectively. The relative expression Akt is the effector of PI3K, and PI3K activation leads to Akt phosphorylation. However, Akt phosphorylation can be mediated through either PI3K-dependent or-independent mechanism. In order to verify whether Bmakt is activated, we examined the mRNA transcription level of Bmakt in silkworm midgut. As shown in Fig. 9, the experimental and TiO 2 NPs groups' Bmakt transcription levels were all higher than the control group's at different time points. Without BmNPV infection at 0 h, the relative transcription level of Bmakt of the experimental group was higher than that of the control group and reached the maximum at 144 h after infection, which was 2.99-fold of the value at 0 h. In the TiO 2 NPs-treated group, the transcription level of Bmakt showed significant increase and reached the maximum at 144 h; as shown in Fig. 8, Bmpi3k level started to be upregulated at 96 h and reached the highest value at 144 h, indicating that the upregulatoin of Bmakt was induced by TiO 2 NPs, not by Bmpi3k. At 144 h, both Bmakt and Bmpi3k's transcription levels reached the maximum, speculating that the upregulation of Bmakt was a joint effect of TiO 2 NPs treatments and Bmpi3k activation.

Effects of BmNPV and TiO 2 NPs on Akt Phosphorylation
Western blot was performed for fat body tissues. The total Akt was measured by an antibody recognizing total Akt, which demonstrated that the amount of total Akt protein remained stable throughout the infection (Fig. 10, upper panel). In contrast, an increased amount of total Akt protein was detected from 120 h after infection in TiO 2 NPs treated group (Fig. 10, third  panel). The phosphorylation of Akt was measured by an antibody that only recognizes Akt phosphorylation on Ser 505. As showed in Fig. 10 (second panel), the level of phosphorylated Akt in control group increased from 72 h after infection, clearing indicating that BmNPV infection induces Akt phosphorylation to resist the virus's infection in silkworm fat body. In contrast, a high level of phosphorylated Akt was detected throughout the infection and without infection at 0 h in TiO 2 NPs treated group (Fig. 10, fourth panel), especially at the time 96 to 144 h after infection, which consistent with the BmNPV proliferation characteristics of gene lef-1 and gp64 in silkworms. The bottom two panels represent the total Akt protein and phosphorylated Akt in TiO 2 NPs-treated without BmNPV infection group, respectively, the result showed that both remained stable throughout the fifth instar. These results indicate that the upregulation of Akt phosphorylation was due to an upregulation of total Akt levels caused by the activation of upstream PI3K and TiO 2 NPs treatments.

Discussion
China's annual production of cocoons is about 6.61×10 8 kg; its sericulture farmers' income is about 3.65 billion dollars and the annual total silk exports is about 33.22 billion dollars [1]. Each year, silkworm diseases may lead to about 20% economic losses [40], 80% of which was contributed by BmNPV disease. BmNPV has also become a serious threat to global sericulture [41]. Therefore, an effective method is urgently needed to improve BmNPV resistance. In recent years, nanoparticles have opened up new ways to treat viral diseases, and there had been reported that TiO 2 NPs could ease silkworm injury caused by BmNPV [31,42]. This study is the first one to reveal the inhibition of BmNPV proliferation in silkworms by TiO 2 NPs and the activation of Bmakt in PI3K-Akt pathway that lead to improved resistance of silkworms to BmNPV. This provides a new method for BmNPV disease treatment.
Polyhedra occurred since 72 h after BmNPV infection, and traditional identification of polyhedra is through light microscopy with apparent lag and poor accuracy [43,44]. In this study, qPCR was performed to detect BmNPV proliferation dynamics using genomic DNA extracted from silkworm midgut and midgut-BmNPV mixture for the first time. The differences in BmNPV copies could be detected as early as 24 h after infection, indicating the high sensitivity of qPCR that should improve early detection efficiency of BmNPV and provide a new idea for the research on the amplification of other viruses.
The proliferation characteristics of BmNPV in silkworm midgut indicated that TiO 2 NPs can inhibit BmNPV proliferation in the organ (Fig. 4-5). Immune signaling pathway analysis revealed that BmNPV infection in silkworms led to upreguated transcription of both Bmstat and Bmpi3k (Fig. 6-8), indicating the activation of JAK/STAT and PI3K-Akt signaling pathways. Because the activation of pi3k may increase AcMNPV yield, and inhibits the transcription of it may reduce NPV production [14]. In the present study, TiO 2 NPs effectively suppressed the upregulation of Bmpi3k transcription after BmNPV infection, which was probably the main reason for the reduction in BmNPV production. The relative transcription level of the downstream effector Bmakt was also detected, its transcription in the experimental group was not reduced along with the reduction of Bmpi3k levels and higher than that of the control group, indicating that Bmakt activation after BmNPV infection with TiO 2 NPs was PI3K-independent. In addition, Bmakt expression in the experimental group was higher than that of the control group at 0 h ( Fig. 9), confirming that TiO 2 NPs can upregulate Bmakt expression. Studies have shown that Akt protein is a serine-threonine kinase that mediates the activities of other kinases, signaling proteins, and cell growth-, cell cycle-, and cell survivalassociated transcription factors to improve immune response [45,46], which is probably one of the reasons why TiO 2 NPs improve silkworm production.

Conclusions
In summary, the increased BmNPV-resistance in silkworms caused by TiO 2 NPs was probably through the inhibition of BmNPV proliferation and the improvement in immune response. The inhibition of BmNPV proliferation was probably by decreasing the expression of Bmpi3k, and the enhanced immunity was through promoting Bmakt expression. However, the mechanisms behind these effects need further investigations.