Results

Despite the overwhelming large extension of our planned project, we set out in the wet lab to achieve as many results as possible. Although our final purpose was not reached due to time limitations, we firmly believe that our evidence is enough to expect the system to work as predicted and to prove that our project is not only viable but functional.

ACHIEVED RESULTS

---

HEK293T CELL LINE IS SENSIBLE TO PUROMYCIN AND HYGROMYCIN AS INDICATED BY MTT ASSAYS

MTT assay is a colorimetric assay used to measure cellular metabolic activity as an indicator of cell viability, proliferation, and cytotoxicity; thus, widely used to determine the antibiotic resistance of cellular lines. In our case, the assay was used to determine the most optimal antibiotic concentration to select the cells that incorporated the constructs, which included an antibiotic resistance gene. Initially, we prepared the same dilution bank for hygromycin and puromycin, although the second one turned out to be more damaging than expected for the HEK293T cell line and required the use of lower concentrations. Each well of the plate was cultured with the proper antibiotic and concentration of it for 7 days, during which the team performed a qualitative assay concerning the cells' growth inhibition. Moreover, on the last day of the assay, we performed a quantitative assay to obtain the inhibition curved. Results are shown in Figure 1 (Fig. 1.1 and Fig. 1.2) and Figure 2.

The following concentrations were chosen due to their capacity to maintain the cell's viability during the first 48 hours, but to kill them after 7 days:

• Hygromycin: 600 ug/uL
• Puromycin: 0,4 ug/uL

EXOSOMES OF THE HEK293T CELL LINE ARE SUCCESSFULLY PURIFIED THROUGH LOW REVOLUTIONS ULTRACENTRIFUGATION

The first step of our project was focused on assessing the default exosome production of the HEK293T cells, despite the existence of previous scientific data that proved it1, 2; establishing a baseline exosome production of the cells, and validating the exosome purification and collection methods.

It must be pointed out that FBS medium, which is necessary for the survival and growth of the HEK293T cell line, contains exosomes and contaminants for its visualization; thus, needs to be removed and replaced by 0% FBS medium before the analysis of the cells. To assess the impact of the medium removal on the survival and growth of the cells, we studied two different conditions: changing the complete medium for 0% FBS 48 hours or 24 hours before collecting the exosomes. The ultracentrifugation protocol (see Protocols) was followed to isolate the exosomes of the cells.

To visualize and quantify the exosomes we used the Nanoparticle Tracking Analysis (NTA) with the generous help of Marc Nicolau (Dr. Hernando Del Portillo’s group). The results of both conditions (Fig.3.) indicate a significant exosome enrichment in the samples resulting from the ultracentrifugation protocol. The '24h with 0% FBS' condition showed a higher purity outcome in the 100-300 nm sized particle fraction; which, in addition to its simpler procedure, made it elective for further experiments.

GOLDEN GATE TECHNIQUE ALLOWS THE ASSEMBLY OF SEVERAL GENE FRAGMENTS INTO LEVEL 1 pJUMP BACKBONE IN A SINGLE REACTION AND EVEN IN THE REVERSE ORIENTATION

The second step of our project was focused on selecting and applying an assembly strategy for our constructions. Considering the amenability of the method, the existence of commercial sponsors, and the availability of suitable plasmid backbones for it in the Distribution Kit, we chose the Golden Gate technique. Our team divided each of the composite Parts (see Parts webpage) into a different number of fragments, which were adapted to be assembled into the pJUMP29-1A KanR Level 1 pUC backbone (http://parts.igem.org/Part:BBa_J428353) with a single reaction. The economization of the number of reactions allowed us to avoid unnecessary intermediate cloning steps and to obtain the construction through a simple step. Nonetheless, we were misled by the sequence map of the backbone in the Part Registry and we designed our fragments to fit in a BsmBI assembly, which generated a restriction site with reverse orientation in the backbone (see Figure 1 of Engineering success webpage).

Since the recognition sites of the BsmBI were not being removed from the backbone, the enzyme was cutting and ligating them in each cycle, thus greatly reducing the efficiency of the reaction. To obtain a smooth ligation, we should have employed the BsaI restriction site, but this one was not present in the sequence map of the Part Registry (http://parts.igem.org/Part:BBa_J428353). Our team was not aware of this mistake until upon request it was shown the correct sequence map of the pJUMP backbone. Despite this huge setback and trying out dozens of different conditions for the assembly reactions, we were able to successfully clone several of our constructs (Figure 4). Hence, we only had to re-design some of the constructs for a BsaI-based assembly. The BsmBI-based assembly protocol that turned out to be the most successful consisted of cutting the constructs with the BsmBI restriction enzyme, purifying the linearized backbone, and afterward performing several cycles of 37ºC-16ºC.

To visualize and quantify the exosomes we used the Nanoparticle Tracking Analysis (NTA) with the generous help of Marc Nicolau (Dr. Hernando Del Portillo’s group). The results of both conditions (Fig.3.) indicate a significant exosome enrichment in the samples resulting from the ultracentrifugation protocol. The '24h with 0% FBS' condition showed a higher purity outcome in the 100-300 nm sized particle fraction; which, in addition to its simpler procedure, made it elective for further experiments.

FUSION PCR FOR THE LIGATION OF GENE FRAGMENTS IS NOT ACHIEVED

In parallel to the inefficient Golden Gate assemblies, our team decided to try the Fusion PCR technique with the CD63_His-tag construct (BBa_K4501013). We designed primers that incorporated overlapping sequences (consisting of 15-22 bp homology regions with the same melting temperature) to the extremes of the different parts, in order to fuse the consecutive ones in each cycle of the reaction. Two different backbones and restriction enzymes (BsmBI or EcoRI/BamHI) were used to assess this approach. However, despite trying four different conditions, any of them allowed us to obtain the desired construct (data not shown).

THE 3’ MODULE ON THE pJUMP BACKBONE IS NOT SUFFICIENT FOR SUCCESSFUL LENTIVIRUS PRODUCTION

Due to time limitations, our team decided to set up an experiment to verify the ability of our pJUMP-based transfer plasmid to produce a lentiviral vector, in parallel with the development of the constructs. However, considering the presence of BsmBI and BsaI recognition sites in the 5' module, which impeded the assembly of the main site, we had to opt for sequential cloning: 1. Cloning of the 3' module, 2. Cloning of the main assembly, and 3. Cloning of the 5' module.

The previous design hindered the cloning of the 5' module on time. However, we tried to produce lentivirus with the pJUMP containing exclusively the 3' module, which could be assessed by GFP expression. The lentivirus were produced with 10 ug of the transfer plasmid, 7.5 ug of pPAX.2, and 2.5 ug of pMD2.G with the TurboFect reagent, in the HEK293T packaging cell line; and then used to infect wild-type HEK293T cells. Unfortunately, no GFP fluorescence was observed (data not shown).

Taking the previous into account, our team decided to re-design the 5’ and 3’ modules to exclude the BsaI and BsmBi restriction sites; and, thus, allow its pre-assembly on the pJUMP backbone. An SV40 origin of replication was further added to improve the transfection outcome of the backbone in the HEK293T cell line (see Engineering Success webpage).

TRANSIENT TRANSFECTION WITH THE CD63_His-tag-L7Ae ALLOWS LOW-YIELD AFFINITY CHROMATOGRAPHY PURIFICATION OF THE EXOSOMES

After accomplishing the assembly of CD63_His-tag and CD63_His-tag-L7Ae, our team decided to set up an experiment to verify the ability of the HEK293T cell line to produce exosomes that could be purified by simple His-tag affinity chromatography. We transfected HEK293T cells by electroporation, either with 5 ug of CD63_His-tag or 5 ug of CD63_His-tag-L7Ae. After 24 hours, the media was changed to 0% FBS and after 24 hours more the supernatant was collected to purify the exosomes. The expression of the reporter of the construct (RFP) was assessed by FACS and exhibited very low values (Figure 6A).

Our team followed the ultracentrifugation protocol and affinity chromatography protocol as described. For the last one, 7 mL of cleared supernatant was passed through a falcon tube containing 2 mL of HisPurTM Ni-NTA resin (ThermoFisher) and the sample was bound according to the instructions of the manufacturer.

The eluted fractions resulting from the ultracentrifugation protocol and the affinity chromatography protocol were assessed by NTA analysis. The fraction obtained through the affinity chromatography protocol displayed a concentration two orders of magnitude lower than the one obtained through the ultracentrifugation protocol. However, the values of the first one were considered significantly high in comparison with the background signal (approximately 104 particles/mL); and, thus, this protocol proved to be able to retain some of the particles of interest. Moreover, particle size analysis confirmed the presence of fewer contaminants (particles above 200 nm) in the fraction obtained with the affinity chromatography protocol, in comparison to that one of the ultracentrifugation protocol (Figure 6C). Nonetheless, it must be pointed out that the unbound and the eluted fraction of the experiment displayed similar amounts of particles (Figure 6C), leading us to think that the protocol is not optimized. Results of our improved CD63_His-tag-L7Ae are equivalent (data shown in the Part Registry, BBa_K4501014).

Conjointly, these results may suggest that the His-tag-mediated affinity chromatography protocol is viable and, when optimized, could be useful to purify exosomes more efficiently. The observed low efficiency might be due to a low electroporation efficiency of the cells. In the future, the lentiviral transfection with CD63_His-tag and CD63_His-tag-L7Ae will be optimized until the detection of high levels of reporter fluorescence.

Additionally, it must be highlighted that our team also managed to visualize exosomes in a FACS Canto cytometer through exosome binding to DynabeadsTM His-tag isolation (Invitrogen) (data available upon request).

OUR DESIGNED shRNA INDUCES DOWNREGULATION OF MYC WITH A SINGLE TRANSIENT TRANSFECTION

The final step of our project was focused on assessing the capacity of our shRNA to downregulate myc expression. We transiently transfected HEK293T and Ramos cell lines with three different versions of the shRNA (BBa_K4501018, BBa_K4501019, and BBa_K4501020) and, after 72 hours, the cells were collected to extract their total RNA. The total RNA was used to create a total cDNA population and perform a qPCR to asses myc expression. Actin-Beta and GADPH genes were used as housekeeping references. The Ramos condition, in which our team only tested the raw version of the shRNA in expectancy of lower efficiency, showed almost a 2-fold reduction in the expression of myc (Figure 7A); hence, proving its functionality. However, little to no effect was observed on all of the HEK293T conditions, perhaps due to their conditioning in 0% FBS medium for 24 hours, which could have masked the effect of the shRNA.

Interestingly, almost no GFP expression was detected by flow cytometry in any of the conditions (Figure 7B); in line with the RFP expression results of the CD63 constructs. Considering the proper functionality of the parts, this event may be indicative of an issue with the reporter expression (e.g. in the bicistronic construction, in the promoter, or by proximity with the main expression cassette). Further attempts with lentiviral vectors will be needed to elucidate the source of this conflict.