BY SAM WU, BS and KEVIN T. FITZGERALD, SJ, PhD

Genome editing, or the ability to manipulate the DNA of an organism, has been facilitated by gene function studies made possible by the progress and affordability of genome sequencing (Gaj T, et al 2013; Ding Y, et al 2016).  To make ethical decisions regarding evaluation and regulation of new genome editing technologies, it is important to gain an understanding of their mechanisms of action and potential applications.

The scientific community has developed several commonly used genome-editing techniques: zinc finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs), and clustered regularly interspaced short palindromic repeat (CRISPR). ZFNs, TALENs, and CRISPR all use nucleases (proteins or enzymes that can cleave DNA), which can be customized to recognize a particular sequence of base pairs in DNA. More recently, Gao et al (2016) demonstrated the utility of yet another nuclease, Argonaute, for genome editing. In general, once recognition of the target DNA sequence occurs, the system’s nuclease binds to the DNA and creates breaks in both strands of the DNA, also known as a double-strand break (DSB).

The DSB can then be repaired within the cell through either of two DNA repair mechanisms: error-prone nonhomologous end joining (NHEJ) or homology-directed repair (HDR) (Wyman and Kanaar 2006). NHEJ is usually the default repair mechanism for DSBs in cells and is useful in research, but is prone to producing errors in the repaired DNA. When DSBs need to be repaired with greater precision (e.g. for therapeutic purposes), use of the more accurate repair mechanism, HDR, is recommended (Ding Y, et al 2016; Cortez 2015). HDR requires the addition of a fragment of DNA that is identical to the original, unbroken DNA sequence. The proteins involved in HDR use the DNA fragment as a template to ultimately repair the DSB, restoring the original DNA sequence. In genome editing applications, the DNA template can be manipulated such that HDR will introduce a new mutation, or change in the base pair sequence, into a particular gene (Cortez 2015). Depending upon the purpose of the experiment, the outcome of such gene editing could be to inactivate or activate a gene, or to induce or repress expression of a gene. This technology has potential applications in areas such as disease modeling, disease treatment and prevention, and agriculture.

Zinc Finger Nucleases

Zinc finger nucleases (ZFNs) were the first genome editing technology to be popularized in the scientific community (Pabo CO, et al 2001). Zinc finger proteins, major components of ZFNs, were first identified in a species of African aquatic frog, X. laevis (Miller J, McLachlan AD, and Klug A 1985). They are approximately 30 amino acids long and are characterized by the spacing of two particular types of amino acids: cysteine and histidine. In a ZFN, each zinc finger binds to three base pairs on the DNA strand (Pabo CO, et al 2001).

After determining the structure of the zinc finger-binding domain, researchers recognized the utility of the zinc finger framework for the design and selection of new DNA-binding proteins.   Kim, Cha, and Chandrasegaran (1996) were the first to fuse zinc finger proteins with the cleavage domain of another protein, Fok I. This fusion created a hybrid protein that could be customized to cleave DNA at any particular site. Still, zinc finger interactions with DNA were complex and unpredictable. To improve target specificity, more zinc fingers were added to the hybrid. Each zinc finger protein is capable of recognizing three DNA base pairs on the target DNA site; adding more zinc fingers increases the length of the recognition sequence, and thus, target specificity (Bartsevich VV and Juliano RL 2000; Laity JH, Dyson HJ, and Wright PE 2000).

Zinc finger nucleases (ZFNs) are composed of two parts: zinc finger proteins and a DNA cleavage domain (derived from Fok I). An active ZFN (see Figure 1 below) requires two different ZFN complexes—one for each strand of the double-stranded target DNA. This requirement expands the length of recognition sites, further increasing target specificity.   Researchers can alter the DNA binding domain of zinc finger proteins to customize them to recognize a genomic target of choice (Desjarlais and Berg 1992).  Additionally, ZFNs are relatively small in size, allowing for easy delivery into cells compared to other genome editing techniques (Lee J et al 2015). Despite such advantages and improvements to the performance of ZFNs, challenges with efficiency, target availability, off-target effects, and specificity remain (Bae K, et al 2003; Kim et al 2009; Ramirez et al 2008; Cornu et al 2008).

Transcription Activator-Like Effector Nucleases (TALENs)

Transcription activator-like effector nucleases (TALENs) are similar to ZFNs, in that they consist of a DNA-binding domain and a DNA cleavage domain (also derived from Fok I) (Miller JC, et al 2011). The DNA-binding domains of TALENs are proteins called transcriptor-like effectors (TALEs), derived from plant bacteria. TALEs are comprised of 33-35 amino acid repeats, each of which recognizes a single base pair on the DNA (Deng D, et al 2012).   Usually, TALENs are used in pairs. See Figure 2 below for the structure of TALENs.

TALENs are relatively easy to design and construct. Large-scale, systematic studies indicate that TALE repeats can be combined to recognize any user-defined sequence (Reyon et al 2012). The assembly of these relatively large protein complexes has been facilitated by the development of systems such as FLASH (fast ligation-based automatable solid-phase high-throughput). FLASH is both a rapid and cost-effective way to facilitate large-scale assembly of TALENs for use in genome editing. Cermak et al (2015) used another system, the Golden Gate method, demonstrating that it could be used to construct TALENs within five days. On the other hand, delivery and optimization of these constructs can be complicated by their large size and repetitive nature (Holkers, et al 2013). Targeting multiple sites on DNA with TALENs may also be hindered by the large size of the nuclease, although this issue could be ameliorated through diversification of the TALEs (Yang, et al 2013).

Clustered Regularly Interspaced Short Palindromic Repeat (CRISPR)

Ishino et al (1987) were the first to observe CRISPR clustered repeats in E. coli. More research followed, with Francisco Mojica characterizing CRISPR as a microbial immune system designed to adapt to and eliminate foreign genetic material (Mojica FJ, et al. 2005; Reiss A, et al 2014). The novel Cas9 (CRISPR-associated protein-9) gene, found to code for a nuclease, was discovered next by Bolotin, et al (2005) in the bacteria S. pyogenes. The team also found that a particular sequence—PAM (protospacer adjacent motif)—of approximately two to five nucleotides is required for target recognition. The PAM sequence itself may differ depending upon the type of Cas9 being used, but it must always be present in proximity to the DNA target site.

Over the next decade, researchers sought to understand the role of CRISPR and how it interferes with invading genetic material (Makarova et al 2006), focusing mostly on the CRISPR type II system that requires only one Cas protein to introduce DSBs (Barrangou et al 2007). Brouns et al (2008) found that small RNAs (CRISPR RNAs or crRNAs) guide Cas proteins to target DNA sites. Further research (Marraffini and Sontheimer 2008; Garneau et al 2010) informed Emmanuelle Charpentier’s work that identified yet another guide RNA: trans-activating CRISPR RNA (tracrRNA) (Deltcheva et al 2011), which forms a duplex with crRNA to guide the Cas9 nuclease to its DNA targets. In June 2012, Charpentier and Doudna published findings confirming the role of Cas9 as an endonuclease (enzyme that cleaves DNA) and demonstrating the fusion of crRNA and tracrRNA to generate a single guide RNA (sgRNA) (Jinek et al 2012), significantly increasing the simplicity and specificity of the gene-editing system. In 2013, a team led by Feng Zhang successfully demonstrated targeted genome cleavage using Cas9 in both human and mouse cells, in addition to showing the system’s utility in targeting multiple DNA sites and driving HDR DNA repair (Cong et al 2013). See Figure 3 below for the Cas9 mechanism of action.

The targeting efficiency of CRISPR/Cas9 relative to other techniques such as ZFNs or TALENs is high, in part due to the guidance of the sgRNA and the PAM requirement. Still, there has been much work to improve the system’s performance, including increasing the efficiency of HDR DNA repair for use with CRISPR—specifically by suppressing factors associated with the NHEJ pathway (Chu et al 2015). Furthermore, other variants of Cas9 (i.e. Cas9s from other species or synthetically-developed mutant Cas9s) have been identified and can be used in different contexts to better serve the needs of a particular experiment (Jiang et al 2013). The process of targeting multiple DNA sites with CRISPR has been facilitated by the development of a single plasmid (fragment of double-stranded DNA separate from a cell’s chromosomal DNA; plasmids are replicable and are frequently used to deliver genes into a cell) that can contain multiple sgRNAs that can guide Cas9 to different DNA target sites (Sakuma et al 2014; Ma et al 2014; Guo et al 2015). Polstein et al (2015) have also developed a method to activate the CRISPR-Cas9 system at particular points in time via light stimulation, giving researchers more control over when genes are expressed.

While the CRISPR system has its benefits, it also has limitations. There is concern over off-target effects, although efforts have been made to reduce such effects through improved design of guide RNAs and the development of Cas9 variants (Cong et al 2013). Additionally, mismatches of guide RNAs to the wrong target sites remain as obstacles to the effectiveness of CRISPR/Cas9-based targeted genome editing. Lastly, delivery of the necessary components of the system may be difficult due to relative size.

Despite such setbacks, CRISPR remains the most affordable and versatile genome editing technique available. Thus, the CRISPR/Cas9 system has seen huge investment and wide publicity, owing to the system’s relative ease of application and versatility for targeted genome editing in multiple species and for many research, therapeutic, diagnostic, public health, and agricultural applications. Some of these applications have proved more controversial than others. Ousterout, et al. (2015) used CRISPR/Cas9 to correct multiple mutations associated with Duchenne muscular dystrophy, a genetic disorder that leads to muscle degeneration over time (Muscular Dystrophy Association 2016). CRISPR has also been used to facilitate the development of a cheaper Zika virus diagnostic test (Gregory 2016).

On the more controversial side, a total of four proof-of-concept studies –in yeast, fruit flies, and two species of mosquitoes—have been published, demonstrating successful development of gene drives in the lab in at least three organisms (Committee on Gene Drive Research 2016). Gene drives are a method in which the odds of passing on a particular gene to the next generation of offspring are significantly increased beyond the means of natural reproduction (Ledford and Callaway 2015). Developments in agriculture include CRISPR-Cas9-edited mushrooms that can be cultivated and sold, without further oversight from the US Department of Agriculture (USDA) (Waltz 2016).   Adding to the list of controversial developments with CRISPR, UK studies involving editing human embryos with CRISPR technology were approved by the Human Fertilisation and Embryology Authority in early 2016 (Callaway 2016). More recently, proposals have been made for human trials that would use CRISPR/Cas9 genome editing in cancer treatment for patients (Begley 2016).

ARGONAUTE

Argonautes are another family of endonucleases that are also important in defending the cell against foreign genetic material (Gao et al 2016). These endonucleases use single-stranded nucleic acids as guides, and exist in nearly all organisms. Some Argonautes have demonstrated that they bind to single-stranded DNAs and can cleave target DNAs (Swarts et al 2014). Gao, et al (2016) demonstrate that NgAgo (Natronobacterium gregoryi Argonaute) is programmable with single-strand DNA guides to be a “precise and efficient tool for genome editing in mammalian cells.”

In response to the Gao et al (2016) study, some have speculated that NgAgo may pose a challenge to proponents of the CRISPR/Cas9 genome editing system. In the CRISPR/Cas9 system, Cas9 requires that the guide RNA exhibit a specific structure for correct binding to occur. In the Argonaute system, however, the guides need not have any specific structure for binding. The CRISPR system is also limited by the PAM sequence—such that the DNA can only be cleaved if it is in proximity to the PAM sequence (Garneau et al 2010). Argonautes, on the other hand, do not require the presence of any specific sequence on targets, and researchers suggest that these endonucleases have a broad targeting range and low tolerance for mismatches (Gao et al 2016).

Looking Forward

Our previous posts discussed recent breakthroughs in genomics, and provided some recommendations on how to address the difficult questions that these scientific advancements raise. Here, we have attempted to provide our readers with an overview of the technologies in question to emphasize the importance of context as the scientific and political communities grapple with regulating these gene-editing technologies.

ABOUT THE AUTHORS

Sam Wu, BS is a research associate at the Pellegrino Center for Clinical Bioethics at Georgetown University Medical Center.

Kevin T. FitzGerald, SJ, PhD is a research associate professor at the Pellegrino Center for Clinical Bioethics, GUMC.

Recommendations for improved public engagement in discussions on emerging biotechnology: Article three in the series on emerging biotechnology.

BY SAM WU, BS and KEVIN T. FITZGERALD, SJ, PhD

“To move ahead on important national issues without public support is to invite being undermined in the long run.”

– Daniel Yankelovich, Coming to Public Judgment: Making Democracy Work in a Complex World

Scientists perceive the public as both critical of and influential on the progress of science. Yet at the same time, scientists fear or distrust public input into the scientific research arena because they perceive the public as more likely to impede or misdirect the research process due to their lack of understanding of science and research.

Recent advances in science demonstrate the field’s far-reaching societal implications, from industry to medicine to what it means to be human. As Yankelovich stated, “to move ahead on important national issues without public support is to invite being undermined in the long run.” This statement makes the supposed “public mistrust crisis” all the more disturbing, and the need to do something about it all the more imperative. We are at a point where many experts would agree that there is a need to build public trust, but recommendations on how to do so vary considerably.

Some call for more inclusive discussions to help develop alternative solutions. Others blame the media and point to the need for the scientific community to take responsibility for communicating effectively with the public about their research, which can mean determining risks, benefits, and goals without public input. Still others blame educational shortcomings, calling for changes to science curriculum that will foster a greater understanding of the research process and its inherent degree of uncertainty. Various models of science communication and public engagement have been described, attempting to bring some order to the growing list of possible options (for example, see Lewenstein and Rowe and Frewer). Despite the abundance of recommended solutions to the “public mistrust crisis,” there remain significant obstacles in choosing, designing, and implementing good public engagement mechanisms.

To move forward and improve these efforts, we need first to address the prevalent and recurring opinion that most of the public is anti-science and uninformed, and therefore, incapable of participating well in science policy-making. Such narrow thinking has encouraged the development of public engagement events that focus primarily on informing the public of relevant technical information and/or benefits of specific technologies in question, with the goal of fostering public acceptance or understanding (i.e. the deficit model of science communication). Allum, et al. (2008), however, find that scientific knowledge is only one factor, and a weak one at that, among many that determine public attitudes and policy preferences on particular issues. Knowledge, whether technical or lay, is filtered by way of an individual’s social and political identities. Hence, while aiming at increasing public scientific understanding may be an important element of public engagement, it is by no means sufficient and should not be the only aim.

The next step forward would be to increase our understanding of the complex network of factors that precipitate controversy and public mistrust in particular contexts – in this case, in discussions regarding emerging biotechnologies. Such insight may, in turn, inform the selection and design of public engagement strategies that aim to address disagreements specific to that particular scientific debate. Notably, not all public engagement events should seek to achieve the same goals or to address the same issues. Different technologies can raise different questions of ethical and practical importance, and can be subject to a variety of regulations depending on the nature of the technology. These differences provide the context within which the dilemmas inherent to the design of any engagement event must be resolved: upstream v. downstream, deliberation v. decisiveness, inform v. elicit, top-down v. bottom-up, and commissioning research v. involving civil society groups (Nuffield Bioethics, Emerging Biotechnology 2014). The purpose and design of a public engagement event will, therefore, depend upon the particular technology or scientific advance in question and the factors, stakeholders, and communities involved.

Additionally, public engagement mechanisms can be designed, not merely with the goal of fostering public understanding or acceptance, but also with the goal of making diverse interests and values explicit and creating room for disagreement and consensus to inform policy-making. Rather than highlighting consensus and glossing over disagreements, both points of agreement and disagreement should be emphasized as valuable outcomes of engagement. Decision-makers can, for instance, use persistent points of disagreement as jumping-off points to refine proposals or to develop new alternatives altogether (O’Doherty, et al. 2009).

Incorporating diverse community interests throughout the decision-making process will encourage more innovative and ethical policy-making that leaves room for deliberation. One could even frame the design of public engagement methods in a public discourse ethics, which would enhance the values of equity, solidarity, and sustainability – helping to prevent biased decision-making that favors the goals of science.

Emerging biotechnologies pose new and challenging ethical questions, fraught with uncertainty and ambiguity as to how they should be answered. As stated earlier, a complex network of factors influences public attitudes and preferences, and to suggest that addressing only some public knowledge “deficit” is the best way forward ignores the dynamic, interconnected nature of our society. It is, therefore, crucial that we re-evaluate our policy-making processes and extend these processes to include stakeholders beyond science, policy, and industry. Employing this approach, we may better integrate our rapidly developing technologies into a future envisioned by our entire society, and not merely the science and technology community.

ABOUT THE AUTHORS

Sam Wu, BS is a research associate at the Pellegrino Center for Clinical Bioethics at Georgetown University Medical Center.

Kevin T. FitzGerald, SJ, PhD is a research associate professor at the Pellegrino Center for Clinical Bioethics, GUMC.

BY KEVIN T. FITZGERALD, SJ, PhD, SAMANTHA (SAM) WU, BS, and Fr. CHARLES BOUCHARD, OP

In February 2016, it was reported that the Human Fertilization and Embryology Authority (HFEA) granted limited permission for researchers in the UK to genetically modify human embryos, with the hope of elucidating which genes are necessary for successful embryological development. Although Dr. Kathy Niakan and her team at the Francis Crick Institute are only allowed to use the embryos for 14 days, and may not implant a modified embryo in the womb, this permission crossed a frontier in genetic research. It was the first time human embryonic genetic modification had been authorized. This followed the publication of the controversial paper by Liang, et al. (2015) that detailed the researchers’ attempt to modify genes that cause β-thalassaemia in non-viable human embryos using the gene-editing technique, CRISPR. The paper, published in April 2015, kicked off a heated ethical debate.

Now, Frederik Lanner at the Karolinska Institute in Sweden, who got the go-ahead on a project that will also involve gene editing in human embryos, is making preparations to begin those experiments. Earlier this month, it was also reported that another team in China, led by Yong Fan, attempted to use CRISPR to generate HIV-resistant human embryos via the introduction of precise genetic modifications. While this project involved non-viable embryos, much like the research conducted by Liang, et al., “the purpose of this study was to evaluate the technology and establish principles for the introduction of precise genetic modifications in early human embryos.” The ethics committee of Guangzhou Medical University in China approved Fan’s work, and has reported that it has since approved two more similar projects.

The rapid rate of investment of both time and money in new projects involving gene editing and CRISPR makes it clear why the novel gene-editing technique was named Science’s “2015 Breakthrough of the Year.” Indeed, the technique and the research it facilitates have the potential to lead not only to treatments, but also to the elimination of some genetic mutations from the human genome altogether. Other novel biotechnologies, such as next-generation sequencing (NGS), have contributed to the revolution in gene-editing, making sequencing of the genome faster and cheaper than ever before. Clearly, these new technologies are altering, and will alter, medicine in ways that were science fiction only a few years ago.

Scientists have hailed the advancement of these projects with enthusiasm, convinced that the recent approval of the human embryo gene-editing research by funding agencies and IRBs is indicative of wider societal approval. Lanner, for instance, is hopeful that his work will be received with more optimism and less heated debate than the paper by Liang, et al. published just a year ago. At the same time, it is worth nothing that many of the ethical and practical questions, which made and continue to make the genome-editing debate controversial, remain unanswered:

Who should guide the potential impacts of these developments in clinical practice and in broader society?  What applications are ethically permissible?  Who will own these new technologies and the information resultant from their use? How should we regulate and oversee the technology in a way that such advances in science are not prioritized at the expense of public health or to the disadvantage of the poor or marginalized? 

We argue that experts in research, medicine, industry, and policy currently dominate and guide the conversation about R&D and regulation of gene-editing technology – leaving it up to the “experts” to answer the aforementioned questions of ethical and practical importance. But one might also ask, do the “experts” have the appropriate knowledge to know what is in the public’s best interest? What is the “appropriate” or relevant knowledge to make such a judgment? Are we moving forward with these projects because we have answered those questions?

CRISPR, NGS, and many other biotechnologies are all pieces of the broader discussion regarding what the future of science, medicine, and society will look like.  The outcomes of such a discussion may affect our conceptualizations of “disease” and what it means to be a “normal” human being – things that impact every human in society, not just a gathering of experts from a particular social stratum. By restricting the debate to only the “experts” or those who have a vested interest in the technology, society risks maintaining the status quo, or worse, exacerbating existing socioeconomic and health inequalities that disproportionately affect marginalized communities.

Recent Attempts at “Public Engagement”

Many scientists and policymakers have recognized that we are at a pivotal moment in research and health care.  They acknowledge the benefits of collaboration and communication in the effort to achieve improved health.  The FDA, for instance, launched precisionFDA, an online portal that enables “scientists from industry, academia, government and other partners to come together to foster innovation and develop the science behind…[NGS].”

With CRISPR, the international research community was prompted by growing ethical concerns to host the International Summit on Human Gene Editing in Washington, DC in December 2015.  The summit brought together experts from across disciplines and continents charged with the task of discussing the “scientific, ethical, and governance issues” central to the debate on human gene-editing research. That the scientific community organized the conference suggests that it recognizes the need to incorporate a variety of perspectives in the debate. But, recognition alone is insufficient to ensure that such diverse interests are represented in the resulting policy recommendations.

At the conclusion of the summit, Chair of the Organizing Committee, David Baltimore, proposed guidelines for regulating human gene-editing research. The guidelines, however, were crafted to reflect the perspectives of scientists and academics, along with a few members of the general public who were invited and present. The guidelines lacked input from other public groups for whom human genome editing could have many implications. Unfortunately, the inadequate model of “public engagement” employed at the Summit is actually quite common across science policy issues.

Dr. Ruha Benjamin, Princeton University professor and author of People’s Science: Bodies & Rights on the Stem Cell Frontier, notes that the processes currently used to gauge public opinion on new scientific developments often create only an “illusion of opening up the science.” Institutions acknowledge, to some degree, the importance of gauging public interests and goals when it comes to scientific progress. The list of terms and mechanisms used in an attempt to determine such interests and goals is long: “public conferences,” “public meeting,” “public comment period, “public forum,” “public engagement,” and so on. Closer inspection of these mechanisms and their outcomes reveal several issues with the scientific community’s current “public engagement” efforts: 1) there is a lack of evidence and consensus to suggest which mechanism is most appropriate and when, 2) the public’s diversity of values and goals are underrepresented by a handful of public representatives, often selected by the host organization themselves, 3) differences are often resolved with more “expert” or “technical” information, rather than deeper discussion of differences in values or goals (expert knowledge is prioritized over lay knowledge), 4) the mechanism’s design provides the public with insufficient leverage to effect change in research or policy development, and 5) more often than not, there are insufficient evaluation measures in place to reinforce accountability. While these mechanisms have the potential to inform and engage the public in policy discussions that bear directly on the common good, these obstacles often lead them to fall short and leave voices unheard.

Our ethical tradition of fostering the common good says we need broader discussions that go beyond the one-way discussion that focuses on informing the public to foster “public acceptance” — beyond this “illusion of opening up the science.” We need deliberative, two-way discussions that make values and goals explicit, and that actively involve the public as community members and participants in research, clinical practice and decision-making. With genome editing, it is imperative that these discussions happen now before boundaries are crossed, even inadvertently, that cause harm to many people which cannot be readily remedied.

Moving forward, we suggest that academic institutions – increasingly home to interdisciplinary efforts and collaborations – make a much greater effort to research and design processes that engage public stakeholders in the discussion around genome editing and, more broadly, emerging biotechnologies. Several organizations have already implemented biomedical research tools and platforms created with participant preferences, values, and communities in mind that can serve as a good starting point. We also call for greater collaboration between natural science and social science researchers, which will increase our understanding of the assumptions and factors that influence policy-making and inform the design of mechanisms intended to generate a more open, deliberative public dialogue.

Sustained public engagement is essential to ensuring that science and medicine’s advancements continue to address broad public needs, and not only those of the few. Indeed, there have been calls for greater public involvement in the genome-editing debate. Despite these calls, however, the field continues to proceed with controversial experiments while ethical and practical questions remain unanswered, with scientists presuming that IRB approval of their controversial projects means that the public is becoming more accepting with “the passage of time.” It is, thus, time to increase our efforts to “open up” the science, not just to the experts, but also to those whose lives will be impacted and whose voices have yet to be heard.

ABOUT THE AUTHORS

Fr. Kevin FitzGerald, S.J., is research associate professor and Samantha (Sam) Wu, B.S., is a research associate at the Pellegrino Center for Clinical Bioethics at Georgetown University Medical Center; Fr. Charles Bouchard, O.P., is Senior Director, Theology and Ethics, at the Catholic Health Association of the United States.