COVID-19 breakthroughs: How pioneering scientists are advancing antibody therapies

In a popular SelectScience webinar, Dr. Naveen Suryadevara, scientist at Vanderbilt Vaccine Center, and Dr. Alon Wellner, researcher at the University of California, Irvine, share their insights on how monoclonal antibodies and nanobodies are offering promising new ways to combat SARS-CoV-2 and bring hope to patients with COVID-19. The pair outline their differing approaches and discuss some of the novel technologies being developed to create better, more targeted therapeutics against coronavirus and pave the way to scientific breakthroughs.

The therapeutic power of antibodies

With the promise of better specificity and reliability, antibody therapeutics are becoming a powerful tool to help physicians treat a range of diseases, from cancer to HIV and COVID-19. Many scientists have chosen to pursue antibody drug discovery processes with the aim of identifying new therapeutic agents. Due to the outstanding sensitivity and specificity of antibodies, these immunoglobulins have already revolutionized therapeutics and could be the crucial missing link in the fight against the COVID-19 pandemic. This approach involves researchers isolating antibodies from recovering patients to help identify and analyze antibodies that have the potential to ‘neutralize’ an infection. Monoclonal antibodies (mAb) development is acknowledged as a complex and expensive procedure, which has encouraged researchers to seek and establish new, more effective ways of producing them.

Although traditional hybridoma technology was developed back in the 1970s, it remains one of the most widely used methods for mAb generation. Following the coronavirus outbreak in March 2020, lead investigator, Dr, James E. Crowe and his team at Vanderbilt Vaccine Center, VUMC, responded rapidly and began to identify, isolate memory B cells, and sequence antibodies from the blood of COVID-19 survivors. Taking the blood from convalescence COVID patients is important, because these individuals have already produced the antibodies that can potentially neutralize SARS-CoV-2. “There are many approaches used within antibody discovery processes, but many labs, including the lab at Vanderbilt Vaccine Center continue to use many traditional approaches,” says Suryadevara.

Another approach James E. Crowe and his team have explored is a strategy known as antibody gene rescue. Antibody gene rescue involves enriching certain cell populations, including plasmablasts, as well as antigen-specific B cells and memory B cells to collect the heavy and light chains of an antibody. To help identify and enrich B cells from PBMCs, the team turned to the Sony SH800 Cell Sorter platform. The cell sorter permits sorting of a wide range of cell sizes using the 70-µm, 100-µm, and 130-µm microfluidic sorting chips, whilst being fully integrated with fluidics controls. The information gathered from this process can then be analyzed via bioinformatic platforms to identify antibody lineages, heavy and light chain pairings, and antibody duplications. "With this antibody gene rescue approach, we can select a panel of antibodies that are functional and capable of neutralizing a virus at a much quicker pace than other existing approaches,’ explains Suryadevara.

Before setting out to directly tackle COVID-19 with monoclonal antibodies, the team at Vanderbilt began their investigations using Zika virus. "When the outbreak began, our site was selected as one of the labs in the P3 program, known as the pandemic prevention platform with the goal to identify donors that have already been exposed to an infection, isolate B cells, rescue antibody sequences, express, purify, run functional assays and do protection studies within 90 days,’’ says Suryadevara. The P3 program is a five-year agreement with the university which functions to rapidly develop antibody treatments within 60-90 days after a viral outbreak. The program involves the delivery of antibodies in the form of nucleic acids, termed nucleic acid delivered therapeutic antibodies, which can be used to mount a medical countermeasure within 90 days. Applying nucleic acid delivered therapeutic antibodies can scale up and amplify antibody production compared to other conventional methods. After successfully achieving this goal with Zika virus, the Crowe lab researchers are now using this novel strategy to combat COVID-19.

Webinar attendee question: How many target specific B cells are sorted for sequencing analysis using the 10x Chromium System?

Dr. Suryadevara: From the two donors we used, we had 20 million PBMCs, where we were able to sort approximately 9,000 antigen-specific B cells. Out of the total 9,000 antigen-specific B cells, we split these into two different platforms, where 4,500 were loaded onto a platform and the remaining 4,500 cells were expanded to generate around 40,000 expanded B cells.

Neutralizing antibodies: COVID-19

Spike is the viral glycoprotein of SARS-CoV-2 that binds angiotensin-converting enzyme 2 (ACE2) to host cells, while being recognized as one of the major proteins targeted for antibody therapeutics and COVID-19 vaccine development. It is of paramount importance for researchers to be able to produce neutralizing anti-spike mAbs to help protect against coronavirus. "Due to the complexity of this dynamic glycoprotein, spike can be very difficult to work with as it can come in many conformational states, including both open and closed states,’’ says Suryadevara. The many heterogeneous conformational states of spike can make antibody isolation problematic, particularly when producing antibodies that can neutralize and fully block the receptor-binding domain of the spike protein.

Suryadevara outlines how Dr. Crowe’s lab was able to analyze a large panel of human monoclonal antibodies that specifically targeted the spike glycoprotein, as well as successfully identify antibodies that exhibited potent neutralizing activity to block the receptor-binding domain of the spike protein from interacting with ACE2. The group was able to categorize the resulting monoclonal antibodies into specific unique categories that could target distinct epitopes, as well as conformational states. "We were able to identify close to 900 monoclonal antibodies with 400 exhibiting reactivity to various segments of spike protein. We were able to identify a panel of 50 lead candidates to neutralize the virus from the 400,’’ explains Suryadevara. These findings could potentially be used to support vaccine design and help in the selection of robust therapeutic agents to inhibit the virus from causing disease.

From B cells to yeast surface display 

Over at the University of California, Dr. Alon Wellner, at the lab of Dr. Chang Liu is also exploring new ways to better create antibodies to treat patients suffering from COVID-19. This team has developed a novel technique known as orthogonal DNA replication which is capable of the continuous hypermutation of genes in engineered yeast cells. The team is now using this system to detect and neutralize the interactions that occur between the spike protein and the cellular receptor, ACE2, which is known to act as the main location for the virus to enter into the cells.

Wellner and his group began their investigations targeting G protein-coupled receptors (GPCR). The team is now implementing the same strategy to target coronavirus. “We managed to generate a panel of neutralizing antibodies in 30 days. We handed over the lead antibodies to another team who confirmed that our antibodies demonstrated a high level of protection in mice models.”

Webinar attendee question: What are the advantages and disadvantages of using yeast surface display over B cells for antibody screening?

Dr. Wellner: B cells do have a large set of advantages over synthetic systems like ours. Antibodies originating from patient-derived B cells are known to be less immunogenic and target the correct epitope for neutralizing a virus, since they were selected to do so by our body’s immune system. However, many of the patient-derived antibodies tend to be on the lower end of affinity and often require further optimization, so without rapid and scalable affinity maturation, we tend to get stronger subnanomolar affinities. We are now setting up new selections against a large panel of RBD (the receptor-binding domain of CoV2’s S glycoprotein) mutants. In the case of B cells, you would need to prove a patient is suffering from a new mutant virus, whereas we can generate multiple selections and produce numerous nanobodies against those mutants.

"When isolating antibodies, we need to apply additional affinity maturation processes to increase antibody affinity to a given target,” explains Wellner. Antibody affinity maturation occurs naturally in our own immune system and can be mimicked using various in vitro and in vivo systems via directed evolution using PCR techniques. Directed evolution is a method used in protein engineering that mimics natural selection by steering proteins toward a goal through mutagenesis and selection processes. "One of the most popular platforms for antibody affinity maturation is with yeast surface display,’’ says Wellner. One of the biggest advantages of yeast surface display is the measurement of the expression of each variant using a HA tag or other epitope tags. "Unfortunately, yeast surface detection can be very slow, laborious and low throughput. It is hard to evolve more than a single antibody at a time,” Wellner notes.

In a bid to solve these issues, Wellner decided to incorporate yeast surface display into orthogonal replication technology, a technology termed OrthoRep. OrthoRep is a continuous system for yeast surface display, which functions to accelerate evolution and allow for the simultaneous running of many duplicate selections. The webinar details how Wellner and his team were able to couple clonal selection with affinity maturation by integrating a small computationally designed nanobody library into yeast to mimic the immune system and allow for direct evolution to be achieved. "Combining yeast surface display into orthogonal replication technology is a simple principle. This process often begins with applying a p1 plasmid, which is located in the cytoplasm of yeast. We place our gene of interest onto the p1 plasmid, which begins plasmid replication,” Wellner explains. The team was able to achieve mutational acceleration of about 105 higher than normal genomic mutation rates. "All you need to do is grow your cells and the gene will get automatically mutated for you,” he adds.

Rather than using monoclonal antibodies, the team at the University of California chose to focus on nanobodies, which are unique antibody fragments that originate in llamas and their relatives. “IgG has both heavy and light chains in the variable domain, whereas nanobodies are much easier to engineer as they are smaller proteins and retain high binding affinities like monoclonal antibodies,” explains Wellner. “We use the OrthoRep system to evolve high-quality nanobodies and ACE2 variants to neutralize the SARS-CoV-2 through binding of its spike protein.” Wellner notes that even though it is now possible to streamline affinity maturation, what they wanted to create was a system comparable to a synthetic immune system including clonal selection.

The team needed to generate small libraries that they could easily transfer onto theOrthoRep. "We collaborated with Debbie Marks from Harvard Medical School who used machine learning approaches to design a small library of millions of variants that mimicked the germline repertoire of a llama,” says Wellner. “Using this library, we were able to isolate low-affinity clones to any antigen that then automatically evolve into high-affinity binders. We managed to generate 11 different clones that bound to the receptor-binding domain of spike with different affinities.” This unique synthetic procedure could help combat current and future outbreaks of the coronavirus pandemic by generating therapeutic candidates before a vaccine is made available.

Find out about the latest COVID news in our COVID-19 Special Feature>>