BACKGROUND
Drug delivery encompasses a great array of issues concerning the metabolism, toxicity, specificity, stability, and bioavailability of the therapeutical agent in the recipient organism. Regarding that, lipid-based nanocarriers offer a versatile platform for targeted drug delivery, which has already led to some clinical translation approaches1. Extracellular vesicle-based carrier systems, and in particular, exosomes, hold great promise to provide a solid approach that solves those major issues2.
Notwithstanding, exosomes have been found to promote immunomodulation3, regeneration4, and recovery5 when administered for several diseases. However, its main source in the therapeutics field is mesenchymal and adipose-derived stem cells1, which limit its obtention in high amounts and its range of application. Furthermore, its current production, isolation, and obtention techniques entail a significant set of issues. Hence, further bioengineering refinement is required to address the previous clinical and commercial limitations2. Nonetheless, it must be pointed out that engineered exosomes remain in preclinical stages2, whereas stem cell-derived exosomes are being tested in current clinical trials.
OVERVIEW
In this context, we propose a cell-based system for the production of engineered exosomes. Given the facility of transfection, cell culture availability, and default exosome secretion, we chose the HEK293T cell line to incorporate the machinery. The device includes a system that improves and facilitates exosome purification through affinity chromatography of Histidine-tag labeled exosomes. Besides, it incorporates a circuit for the loading of shRNA into these vesicles at the cellular level, avoiding the inefficient step of artificial loading of the therapeutical agent and simplifying the process.
Concretely, we sought to apply the system to produce therapeutical exosomes against Burkitt lymphoma, a prevalent and aggressive B-cell lymphoma whose main current treatment is chemotherapy. For that, we planned to introduce an shRNA against myc, one of the main oncogenes implicated in this type of tumor, while endowing our exosomes with a CD19 ligand to provide B cell tropism.
Boosting of exosome biogenesis
To increase the default production of exosomes, STEAP3 (BBa_K261900), NadB (BBa_K2796002), SDC4 (BBa_K2619001), nSMase (BBa_1633008) genes were integrated into our cell line with expression cassettes corresponding to the composite parts BBa_K4501011 and BBa_K4501012. These genes were previously proven to boost exosome biogenesis6
Enhancement of exosome purification: His-tag
A Histidine-tag incorporated into the extracellular domain of CD63 (BBa_K3113051 part, taken from Münich 2019) was used to facilitate exosome purification by chromatography affinity.
shRNA cargo loading circuit
In addition to the His-tag, the archaeal domain L7Ae (BBa_K3113009 part, taken from Münich 2019) was incorporated into the cytosolic domain of CD63; and, up to three consecutive L7Ae (BBa_K4501014, BBa_K4501015, and BBa_K4501016) were introduced to increase shRNA binding.
Furthermore, an expression vector containing the shRNA sequence marked with C/D-box, the RNA ligand of L7Ae (BBa_K4501018, BBa_K4501019, and BBa_K4501020), was transfected into the cell line to allow inside-cell loading of the cargo.
Cell type tropism by surface fusion protein expression
Engineered exosomes also express the Lamp2b-CD19-L fusion protein transgene (BBa_K4501017) on their surface, which provides them with tropism for B cells.
CLONING STRATEGY FOR PLASMID CONSTRUCTION
Given the sponsored free fragment gene synthesis from IDT and Twist Bioscience, and the availability of pJUMP plasmids in the Distribution Kit, we chose Golden Gate as the first-line technique for the assembly of our constructions. To do so, we divided the different desired constructions (see Parts, Composite Parts) into different parts (from 2 to 6 parts for each construction) and added flanking adaptors with recognition sites for BsmBI, BbsI, and AarI, resulting in sticky ends (see Figure below). All the parts were inserted on the “pJUMP28-1A KanR Type IIS level 1 vector. Origin pUC” (BBa_J428353) on a single reaction (see Protocols). The strategy for the transfection of the vector was either by using the CRISPR-Cas9 system or lentiviral infection.
a. Integration by CRISPR-Cas9 system
Employing the AAVS1 homology arms located at the 5’ and 3’ terminus of the Exosome booster 1 construct, we aimed to integrate the transgene into that safe harbor to achieve a stable expression. For that, we designed a gRNA with IDT wizard7 that recognizes the AAVS1 region, in which the transgene should be inserted. The HygR selector added to the transgene assured to retain only the cells that incorporated and expressed the construction.
b. pJUMP-based lentiviral transfer plasmid construction
For the rest of the constructions, we opted for a lentiviral vector that ensured time optimization. Usually, the use of a lentiviral vector requires a preliminary step consisting of the cloning of the desired transgene into a preformed transfer plasmid that contains MSCV modules, which is time-consuming and sometimes challenging. Hence, to solve those limitations, our team designed a pJUMP-based lentiviral transfer plasmid that eliminates the extra cloning step and allows the resultant plasmid from the assembly to be used right away.
We designed two parts (BBa_K4501005 and BBa_K4501005) that handle the 5’ and 3’ MSCV modules, which are required on the lentiviral transfer plasmids. Taking advantage of the secondary restriction sites of the pJUMP backbones (BbsI and AarI), the MSCV modules were incorporated into the pJUMP backbone through a one-fragment Golden Gate reaction, before the assembly of the main construction. Consequently, we were able to upgrade the default pJUMP backbone to be used as a lentiviral transfer plasmid itself, while eliminating a whole classic cloning step.
c. Fusion PCR cloning strategy
Given the difficulties in the assembly of some of our constructs with a big number of parts, we also developed them by Fusion PCR techniques. We designed specific primers with overlapping sequences between the consecutive parts, creating 18-25 bp homology sequences that shared the same Tm among the parts of the same construction. A first reaction to amplify each of the parts individually was followed by the Fusion PCR, in which all the fragments of a construct were put together in the same reaction with a forward and a reverse primer of the first and last part respectively. Afterward, the resulting construct required classical cloning to be inserted into the backbone.
DESIGN OF THE shRNA
As a proof of concept of our device, we aimed to produce an shRNA-mediated silencing of the oncogene myc in Burkitt lymphoma. For this purpose, we used the VectorBuilder software8 to design an shRNA directed against the 1741-1761 position of the MYC gene (NM_002467.6) transcript. Moreover, we developed three different versions of the shRNA to prove that the L7Ae-C/D-box loading system worked on our device.
- anti-myc shRNA (BBa_K4501000)
- anti-myc shRNA labeled with the C/D-box (BBa_K4501001)
- anti-myc shRNA in-loop labeled with the C/D-box (BBa_K4501002). Since the C/D-box is a loop structure, we optimized the design by substituting the traditional loop of the shRNA with the C/D-box.
LOADING OF THE shRNA INTO THE EXOSOMES AND PURIFICATION OF THE SYSTEM
Given the presence of L7Ae on the cytosolic domain of the exosomes and the labeling of the shRNA with the C/D-box, we were able to load the exosomes with the shRNA in a specific and effective manner. Through this process, the labeled shRNA is captured by L7Ae and enriched in the exosomes, which then are secreted from the cells (see Animation below).
Regarding its purification, simple affinity chromatography with Niquel resin can be performed to attach the His-tagged exosomes. Notwithstanding, taking into consideration that CD63 synergically contains both the loading system (L7Ae) and the purification tag (His-tag), the device also ensures the loading of the exosomes with the desired shRNA. The previous approach increases the efficiency and specificity of the production process while reducing the undesired heterogeneity of the collected exosomes.