Cambridge Healthtech Institute’s Ann Nguyen recently interviewed Andrea Throop of Arizona State University. Dr. Throop shares her presentation on “Selecting the Optimal Vector for High-Throughput Cloning and Protein Arrays” at the Engineering Genes, Vectors, Constructs and Clones conference taking place January 18-19, 2016, as part of the 15th Annual PepTalk in San Diego, CA.

Q1: How does the Biodesign Institute at ASU support your research on high-throughput cloning? What special resources do you have there?

The Biodesign Institute has been instrumental in providing us the ample space for large equipment and many freezers required for high-throughput cloning. Our cloning team, a part of the Center of Personalized Diagnostics under the direction of Dr. Josh LaBaer, is responsible for the full-length sequence validation of over 150,000 Protein Structure Initiative (PSI) clones in human and other major model organisms. In addition, our team does all of the recombinational cloning and clone validation projects both within the center and numerous collaborators throughout the country. Efficient and accurate management of HT cloning relies on a solid informatics pipeline, which supports all aspects of DNA manipulation, sample handling, sequence validation, and protein and microarrays. Our center tracks and analyzes each step of the cloning and sequencing project using Laboratory Information Management Systems (LIMS) developed in our laboratory. These LIMSs generate bar codes and text files that communicate directly with our robots and parallel every step of the recombinational cloning system from primer design through cloning, and ultimately sequence verification of full-length clones, thus ensuring a seamless flow of information between robots and LIMSs throughout.

In order to handle large cloning workloads, our team has a Biomek FX automated liquid handler that is iatrical in our cloning protocol. Furthermore, we have an automated colony picker that carefully selects individual transformed clones and inoculates them into overnight cultures to be screened. The Biodesign Institute also houses a sequencing core that offers both Sanger and Next Generation Sequencing. The sequencing core provides us fast turnaround time of our clone sequences, allowing us to triage the accepted clones and those that need additional work to capture the genes of interest. To quickly analyze the thousands of DNA sequences produced by the sequence core we utilize our Automated Colony Evaluation (ACE) software. ACE uses artificial intelligence (AI) to analyze the raw sequence data by assembling the actual sequence of an insert into a contig and comparing it to its expected sequence. Each difference between actual and expected sequences is recorded as a discrepancy object including its position, bases, and sequence confidence. Acceptance and rejection are determined by counting the number of each discrepancy type and its confidence and comparing that against what is considered acceptable for a particular project.

All clones created by our team are annotated, stored and made available to the scientific community in an open repository (DNASU).The DNASU Plasmid Repository provides centralized archival and distribution for plasmids created within our laboratory, by consortia (such as the Protein Structure Initiative and the ORFeome Collaboration), and by individual laboratories. DNASU distributes plasmids to all researchers, including companies both domestically and internationally for non-commercial, research purposes, and thus far has distributed over 200,000 plasmids to researchers in over 45 countries. Biodesign has provided us space and security for this large freezer repository to guarantee quality of our clones.

Q2: You’ll elaborate on “Selecting the Optimal Vector for High-Throughput Cloning and Protein Arrays” during your presentation on January 18. What particular inefficiencies are you trying to prevent when choosing the most appropriate vectors and cloning schemes?

The most critical property we consider when selecting a vector for high-throughput cloning is the ability for that vector to function in our recombinational high-throughput cloning schemes. We clone thousands of genes at a time, and the routinely used digestion/ligation cloning schemes do not function well in high-throughput cloning. Digestion/ligation schemes, or any other cloning schemes that you have to consider restriction sites and individual gene and vector properties, is just not feasible. My team utilizes Gateway recombinational cloning to construct our clones. This technology allows us to clone thousands of genes under the same conditions and a single cloning master mix. The vectors we utilize must also contain Gateway-compatible ends. We have also engineered numerous vectors to have these ends in order to work with them in our cloning pipelines.

Another inefficiency we try to avoid in our high-throughput cloning is individual PCR optimization of our clones. Again, for thousands of genes this not possible. To avoid this, we use a Python nearest neighbor script that generates PCR primers for thousands of genes that will amplify the region of interested under relatively similar conditions.

Lastly, in the past a known cloning inefficiency was cloning large (>3kb) using high-throughput methods. Clontech Laboratories, Inc. has developed a technology to efficiently clone large genes (up to 15kb) into any vector. Being aware of the size of your inserts you are trying to clone and then selecting the appropriate cloning scheme is critical in successful high-throughput cloning. Overall, we are trying to avoid the inefficiency of having to optimize our cloning protocols for individual genes and vectors.

Q3: Do you foresee an “optimization ceiling” with cloning the genes into vectors for protein expression? If so, what other techniques or strategies might arise?

We have just come across an example of this question recently in our center. For most of our protein expression work we utilize NAPPA protein microarray platform. This is a patented method, where full-length proteins are produced in situ on the array by human ribosomes and chaperone proteins minutes before testing them, ensuring that proteins stay naturally folded for the study. A key advantage of our method is that we can rapidly produce custom arrays of any proteins from our list of available genes and the robust expression method displays >95% of proteins, including large proteins (>100 kDa and membrane proteins). The amount of protein displayed per feature is approximately 100 fmol. NAPPA technology utilizes a GST-tagged expression vector. When proteins are expressed they are bound to the array by a GST antibody. The GST-tag works well for NAPPA and identifying protein-protein interactions; however, the GST protein is quite large and “heavy”. This became apparent when we wanted to move our NAPPA technology to an SPR (surface plasmon resonance) slide to study the kinetics of the protein-protein interactions. SPR is more sensitive to protein size and background. To work around this issue, we have recently constructed a HaloTag® (Promega) vector compatible with our Gateway cloning system. HALO is based on the formation of a covalent bond between the HaloTag® protein fusion tag and the SPR slide. With this technology we can avoid antibodies, BSA and other large proteins in the NAPPA master mix. In conclusion, we found that modifying tags and Gateway-compatible expression vectors can allow the NAPPA technology to be interchangeable between protein expression platforms.

To learn more about Dr. Throop’s presentation at the 8th Annual Engineering Genes, Vectors, Constructs and Clones conference, visit CHI-PepTalk.com/Genes-Vectors-Clones