We have been doing lab work to experimentally prove our idea ‘biomissile’ for 7 months. During these 7 months, we got heaps of data from our experiments. However, most of the time the experiments didn’t work out as we expected. By continuous trying, we got some preliminary data, which can support our idea. We include all our data here because both the failed and successful data are from our hard work, and they are the footprints towards the realization of our idea.

293T cells were transfected with 308/467/516 plasmids by lipo (lipofectamine 2000) respectively and the 293T cells of control groups were all treated with empty lipo. The setting of experimental groups and control groups is shown in Table 0513.24 hours and 36 hours after transfection，both 293T cells and the culture medium were harvested for RNA extraction. And relative siRNA expression levels were detected by RT-qPCR.
The data revealed whether the expected siRNAs may express normally in 293t cells, and if the siRNAs express perfectly in the cells, whether they could be enfolded into exosomes secreted by 293T cells. We found that the expression levels of the 308/467/516 siRNA gene in the experimental group cells are up-regulated remarkably compared to those in control group cells, especially 467 siRNA gene, more than 1000 times up-regulated.
It also shows that the siRNA expression levels of 308 and 467 genes siRNA in the experimental group cells and exosomes was higher under the condition of extracting RNA 36 hours after transfection. In conclusion, to get exosomes containing more target siRNA, it is more appropriate to harvest cell culture medium and collect exosomes 36 hours after transfection.

Silencing abilities of these three siRNA towards relevant genes were tested. The setting of experimental groups and control groups is shown in Table 0518. RNA was extracted from 293T cells 36 hours after transfection.
A. The expression levels of three siRNA were all up-regulated apparently.
B. 293T cells were transfected with HBX over-expression plasmids and HBsAg over-expression plasmids to test siRNA silencing abilities. The HBX gene expression level in 293T cells was up regulated around 30% and that of HBsAg gene was around 140% compared with control groups. Theoretically, 308 siRNA silences HBX gene ， 467 and 516 siRNA silence HBsAg gene. However, the results were reverse. HBX gene expression level in 293T cells transfected with both HBX over-expression plasmids and 308 siRNA plasmids was higher than that in 293T cells transfected with HBX over-expression plasmids alone. The results of 467 and 516 siRNA were the same. It confused us.

We tested silencing abilities of these three siRNA towards relevant genes again. The setting of experimental groups and control groups is shown in Table 0529. RNA was extracted from 293T cells 36 hours after transfection.
A. The expression levels of three siRNA were all up-regulated apparently. Transfected 293T cells were able to secrete exosomes containing relevant siRNA ideally.
B. There was nearly no difference of HBX and HBsAg gene expression levels between experimental groups and control groups. This time, we didn’t get good results again. HBX gene expression level in 293T cells transfected with HBX over-expression plasmids and 308 siRNA plasmids together was higher than that in 293T cells transfected with HBX overexpression plasmids alone. 467 and 516 siRNA seemed didn’t work at all.

We repeated siRNA silencing abilities screen again. The setting of experimental groups and control groups is shown in Table 0624. RNA was extracted from 293T cells 36 hours after transfection.
A. The expression levels of three siRNA were all up-regulated apparently, especially 467 siRNA, up to more than 700 times.
B. When transfected with HBX overexpression plasmids alone into 293T cells, HBX gene expression level was up regulated only by 80%, compared to control group. The expression level of HBsAg gene was up regulated to one order of magnitude. 308 siRNA inhibited HBX gene expression slightly. The silencing ability of 467 siRNA towards HBsAg gene expression was not apparent, while 516 siRNA obviously inhibited HBsAg gene expression. In conclusion, 308 siRNA and HBX gene were eliminated, highly expressed 467 siRNA and effective 516 siRNA were selected for further experiments.

Silencing abilities of these three siRNA towards relevant genes were tested. The setting of experimental groups and control groups is shown in Table 0704. RNA was extracted from 293T cells 24 hours and 48hours after transfection.
A. Compared with control group, the expression levels of both 467 siRNA and 516 siRNA were up-regulated apparently, and the effect is obvious no matter extracting RNA from cells 24 hours or 48hours after transfection.
B. 293T cells were transfected with HBSAG over-expression plasmids and siRNA plasmids to test siRNA silencing abilities. Compared with the expression level of HBSAG of control group, the expression level of that was up regulated about 50 times. And it is obviously down regulated after transfected with both HBSAG over-expression plasmids and siRNA plasmids compared with only transfected with HBSAG over-expression plasmids. The results shows that 46g7 siRNA and 516 siRNA gene were not only expressed normally in the cells ,but also had good silence effects towards the expression of HBSAG gene.

The figures show the standard curves of 467 siRNA and 516 siRNA, and the curve tells the linearity between circulation times(ct value) of QPCR and the density of siRNA, we can get the density of 467 or 516 siRNA in different MVs according to the standard curves and the ct value.

To test the down-regulating effect of si RNA, we set experiment groups as Table 0729A. Figure 0729A shows that both the siRNA were transfected into the cells effectively, cause all their siRNA expression level are higher than control group, But the level of cells transfected with siRNA plasmids was lower than that of transfected with siRNA. However, from Figure 0729B, we can figure out that both siRNA and siRNA plasmids were efficient in silencing HBSAG gene, it was obvious that compared with HBSAG+irrelevant plasmid group, the HBSAG expression is lower of that in group that transfected with siRNA or siRNA plasmids. In these effective groups, the effect of siRNA is better than plasmid, and the 467siRNA plasmid is better than 516 siRNA plasmid. In view of the results, we choose 467siRNA as our final targeting RNA.

We transfected 293t cells with RVG plasmids ,467 siRNA plasmids and GFP plasmids, expecting that the products of three gene can be expressed on the surface of exosomes or be enfolded in exosomes , with which we inject into mice. The RVG protein is supposed to target to the brain. We conducted two control groups, one is esxosmes that contain nothing, empty exosomes. The other is esxosmes harvested from cells transfected with RVG plasmids ,308 siRNA plasmids and GFP plasmids, and here 308 siRNA plasmids function as irrelevant plasmids.
When compared with the empty control group, the figure shows that the 467 siRNA expression level is higher injecting mice with RVG+308+GFP exosomes or RVG+467+GFP exosomes both in serum and brain , but the background of 467siRNA is a little high in the brain of mice. Furthermore, comparing RVG+308+GFP group with RVG+467+GFP group, we find that the 467 siRNA expression level is higher in RVG+308+GFP group, which does not show the effect of RVG targeting. Regarding these results, we will conduct experiments to select some kind of siRNA of which the background is low in the brain.

To selecting some kind of RNA of which expression was not so high in mice’s brain, we selected several kinds of plasmids to test. From Figure 0809 and Table 0809, we find that the expression of Her2 and anti214 in the brain had little difference with that of the cells transfected with relevant plasmids, which meant their expression was high in normal station. While the background of 308 siRNA or 516 siRNA was lower than that of the cells transfected with relevant plasmids, especially obvious for 516 siRNA, it was more than 3,000 times less than that of the cells. In view of the results, we decide to use 516 siRNA to conduct the brain-targeting experiment.

We conducted the background RNA screening experiment with 308 siRNA plasmids and 516 siRNA plasmids one more time. The results are the same as last time, the expression level of brain are apparently lower than cells transfected with 516 siRNA plasmids. That is to say, 516 siRNA is what we need for the subsequent experiments.

From Figure 0812B, we can see that the HBSAG expression of cells from 467exosomes group was inhibited compared with cells only transfected with HBASG plasmids or co-cultured with empty exosomes, but from Figure 0812C, we can find out that the 467 siRNA level of cells from 467exosomes group are surprisingly lower than that from HBSAG group and empty exosomes group, Which reveals that the down-regulating of HBSAG of 467exosomes group may not because of 467siRNA, and the results may be not so persuasive.

It is another failed experiment on RNA screening of brain. We can see that both the expression level of 516 siRNA and 308 siRNA of DEPC water is much high, and on the contrary, the level of that in cells transfected with relevant plasmids is almost zero. We must have do something wrong in the process of the experiment, like polluting the DEPC water or tagging sample with mistake, cause, the level in cells transfected with relevant siRNA is supposed to be high.

The data and diagrams shows the standard curves of 467 siRNA and 516 siRNA, the linearity between circulation times(ct value) of QPCR and the density of siRNA is supposed to be told. However, we can see that both the two standard curves from from our results are not so good, for the R2 has not reached 0.99, so the linearity is not persuasive. We have found that the standard siRNA we used was incorrect, and we will conduct the experiment for another time with the right standard siRNA.

From the two standard curves, we can get the linearity between the ct value and the siRNA density, the function and R2 have been show on the figure. And according to the two standard curves, we quantified the siRNA of exosomes we collected, the density is showed as Table 0816.

We collected exosomes from 293T cells transfected with 467 siRNA overexpression plasmids and expected that these exosomes could target HBsAg mRNA and degraded them. In figure 1 we can see that the 467 siRNA level in group 1b is higher than 1a, the same as group 2b and 2a. However, group 3b and 3a reflect an opposite trend. Furthermore, figure 2 shows that the HBsAg mRNA level in group 2b is lower than 2a. Both group 1 and 3 reflect an opposite trend. Therefore, we conclude that the content of HBsAg overexpression plasmids and the content of exosomes in group 2 are the better relative condition to do the exosome-cell coculture experiment.

Relative siRNA level and absolute siRNA level were showed in Figure 0824A and Figure 0824B. We can see from the Figure 0824C that the level of siRNA in the mice cerebral cortex injected with RVG+siRNA exosomes is higher than the level in the mice cerebral cortex injected with empty exosomes. RVG+siRNA exosomes were able to cross the blood-brain barrier. What’s more, when injected with RVG+siRNA exosomes, the siRNA level of cerebral cortex was higher than the level of cerebral medulla. It showed that RVG targeted to cerebral cortex.

In order to construct a standardized plasmids that can express siRNA 467, we choose a kind of pSB1C3 plasmids from kit and amplified the plasmids by transfection them into E.coli DH5α Competent Cells. We extracted amplified pSB1C3 plasmids and did agarose gel electrophoriesis with them. Sample 3 and 4 were of high purity, so we choose pSB1C3 plasmids in sample 3 and 4 for digestion.

pSB1C3 were digested by XbaⅠ enzyme and SpeⅠenzyme to get 2000bp carrier segments(CS).

We connected CS with siRNA 467 part by T4 DNA ligase and got siRNA 467 recombinant plasmids.

We transfected 293t cells with 467 siRNA plasmids and pre-s1 plasmids and collected exosomes which we will inject into mice. The pre-s1 protein will supposed to target to the liver. We conducted two control groups, one is esxosmes that contain nothing, empty exosomes. The other is esxosmes harvested from 293T cells transfected with only 467 plasmids. Before injection the absolute 467 siRNA level in exosomes must be examined.

In order to examine the absolute concentration of 467 siRNA in exosomes, we used single strand 467 siRNA (20 pmol/μl) to do RT-PCR and then diluted these cDNA into five gradient concentration(pmol/μl): 10-2, 10-3, 10-4, 10-5, 10-6. We did Q-PCR with the five cDNAs and got five Ct value spectively. After ploting with Ct value and the log of concentration, we got 467 siRNA standard curve as shown in Figure 0925A. The data are shown in Table 0925B.

We examined the protein and 467 siRNA concentration through BCA method and the standard curve spectively. We founded the relationship between the content of 467 siRNA and the volume of exosomes(marked by the content of protein). The data are shown in Table 0925C.

In liver, the relative siRNA 467 level of control groups was much higher than two experiment groups. And the siRNA 467 level of experiment group treated with pre-s1+467 exosomes was lower than groups treated with 467 exosomes (Figure 0925C). The relative HBsAg mRNA level of experimental group treated with pre-s1+467 exosomes was highest, experimental group treated with 467 exosomes was a little lower and that of control group was the lowest (Figure 0925D). All these results were opposite to expectation.

As Figure 1019 shows, the cells surrounded with red line are CD4+ T cell, and the percentage is 15.96%.

As for Figure A , we cocultured T cells with exosomes containing siRNA, and for figure B, we cocultured T cells with exosomes with gp120 on the surface (gp120 can specifically bind to CD4+ cell), and contained siRNA. Comparing the two figures, it showed that exosomes with gp120 on the surface can bring more siRNA into T cells, which proved the targeting ability of our modified exosomes.

We added empty exosomes and our modified exosomes into HepG2 cells, and tested the proliferation of cells in 12/24/36/48 hours, and as figure 1026Cshows, there is no difference between exosomes group and control group. It means that exosome has no toxicity and was safe to cells.

We collected the medium of cells that cocultured with empty exosomesand modified exosomes, and did elisa to test the level of IL-6,IL-8,TNF-α . As shown in figure 1026D-1/2/3, there were almost no difference in the expression level of the three kinds of cytokines. It means that exosome has little immunogenicity.

We injected GFP mice with empty exosomes(Figure1026E-1), exosomes containing GFP-inhibiting siRNA (Figure1026E-2), exosomes containing GFP-inhibiting siRNA with surface protein pres1 respectively (Figure1026E-3), and separated the liver of the GFP mice, and observed the GFP expression level. As shown in figures above, Compared with the mice injected with empty exosomes and exosomes containing GFP-inhibiting siRNA, the GFP expression level in liver of the mice injected with exosomes containing GFP-inhibiting siRNA with surface protein pres1 was significantly down regulated. The result indicated that our modified exosomes with pres1-lamp2b fusion protein on the surface can effectively target to liver.