5

Cardiovascular and Lung Tissues

Chronic cardiac and lung disease are among the leading causes of morbidity and mortality worldwide, and currently there are very few, if any, treatments available to alleviate the damage from heart and lung conditions and prevent mortality. Experimental approaches involving in

vivo cellular reprogramming, the therapeutic administration of exosomes, and bioengineered three-dimensional lung models have emerged as exciting new ways for understanding and treating these conditions. However, several challenges need to be overcome before these technologies can move forward in the research and development process or be or applied clinically. This chapter explores new advances in cell and exosome-based approaches to treating heart and lung disease and includes a discussion of remaining questions and challenges that need to be addressed in order for clinical translation to take place.

REPROGRAMMING APPROACHES TO CARDIOVASCULAR DISEASE

In the United States, heart disease is the number one cause of death in men and women and is the leading noninfectious cause of death in children, according to Deepak Srivastava, the Younger Family Director of the Gladstone Institute of Cardiovascular Disease, the director of the Rodenberry Center for Stem Cell Biology and Medicine, and a professor at the University of California, San Francisco. There are 6 million people in the United States who are living with heart failure, and about half of heart failure patients die within 5 years of diagnosis (CDC, 2016). Currently, there is no disease-modifying therapy available for heart failure, so there is tremendous excitement for the potential of cell-based approaches, Srivastava said.

There are several cell replacement strategies to treat heart failure that are currently under investigation. One approach involves the injection of cardiomyocytes or multipotent cardiac progenitor cells directly into the heart. A recent study demonstrated that fibroblast-derived induced pluripotent stem (iPS) cells that were differentiated into cardiomyocytes were able to improve cardiac contractile function in the damaged hearts of nonhuman primates, although there were concerns about post-transplant arrhythmias (Shiba et al., 2016). Furthermore, there is still uncertainty about the mechanism underlying the functional benefits observed in the hearts of the nonhuman primates. Although the approach generated a lot of excitement and has a great deal of potential, it will be important to address the issue of cell maturity in cardiomyocytes that are derived from iPS cells, Srivastava said. Another approach to treating heart failure involves introducing inductive signals into the heart to stimulate resident progenitor cells to regenerate the damaged tissue. A third

strategy involves stimulating resident cardiomyocytes to re-enter the cell cycle and divide in order to regenerate the heart. A fourth strategy, the focus of Srivastava’s presentation, involves reprogramming endogenous and resident fibroblasts into new cardiomyocytes.

Cardiac fibroblasts are in abundance in the adult human heart, comprising approximately half of the heart’s cells, Srivastava said. The cell-based approach that Srivastava described supposes that converting the resident fibroblasts in situ directly into cardiomyocyte-like cells may induce a regenerative effect. The first step for Srivastava’s team involved determining how fibroblasts could be converted into cardiomyocyte-like cells by cellular reprogramming. To do so, they leveraged the vast knowledge amassed over the past two decades about cardiac cell fate decisions during embryogenesis, and they found that a cocktail of three transcription factors (Gata4, Mef2c, and Tbx5, referred to as GMT) was sufficient for generating functional cardiomyocytes from mouse postnatal cardiac or dermal fibroblasts (Ieda et al., 2010). Interestingly, the cell fate conversion did not involve transition through a progenitor or stem cell stage, but rather the cells converted directly from one adult somatic cell type to another, he said.

Srivastava and his team went on to demonstrate that it was possible to do in vivo reprogramming of murine cardiac fibroblasts into cardimyocytelike cells by delivering GMT via myocardial injection (Qian et al., 2012). The reprogrammed cells exhibited a binucleate structure, assembled sarcomeres, ventricular action potentials, beating upon electrical stimulation, and evidence of electrical coupling (Qian et al., 2012). The cardiomyocytes derived from this conversion are electrically most similar to adult ventricular cardiomyocytes, whereas stem cell–derived cardiomyocytes are less mature, he said.

Even in vivo the cell reprogramming process was somewhat inefficient, Srivastava said, so his team used a high-throughput screening process to identify chemical modulators that could improve the process. It was discovered that Wnt signaling and TGF-β signaling inhibitors could independently improve the efficiency, quality, and speed of direct cardiac reprogramming in vitro, and when used together they had the remarkable effect of producing beating cardiomyocytes in a cell culture dish within 1 week. When researchers used an approach that introduced the genes for the three transcription factors (GMT) in combination with Wnt and TGF-β signaling inhibitors, they observed a greater rate of cardiac reprogramming than achieved when using the gene therapy approach alone (Mohamed et al., 2017).

Srivastava and his team hit a roadblock when they discovered that the cell reprogramming protocol that had worked well in mouse cells and models did not translate well into human cells. The researchers re-screened for other factors and found that the addition of MESP1 and ESRRG to the three original factors was sufficient to convert human cardiac fibroblasts to human induced cardiomyocytes (Fu et al., 2013). To further investigate this treatment in a larger heart more akin to the size of a human heart, they decided to use pigs as a model system, Srivastava said. The experimental protocol in porcine models involved inducing myocardial infarction at day 0, performing magnetic resonance imaging (MRI) to measure baseline damage at 3 days post infarction, delivering reprogramming factors via retrovirus at 5 days post infarction, and measuring cardiac function at 56 days post damage. Compared to the control group, the pigs that received the cardiac reprogramming genes had improved cardiac function at day 56 as measured by magnetic resonance imaging. The team is in the process of repeating the experiment using the chemical factors in addition to the gene therapy, Srivastava said.

Further refinement of the technique has reduced the number of factors necessary to reprogram human fibroblasts to three: MEF2C, TBX5, and Myocardin. One important remaining challenge is the issue of delivery, Srivastava said. The use of a retrovirus for gene delivery is not optimal because it may integrate in vivo, while adeno-associated viruses (AAVs) work well for targeting cardiomyocytes but generally have low rates of infectivity in cardiac fibroblasts. A positive-negative screen of AAV variants resulted in the identification of one AAV variant (A2) that has greater tropism for human cardiac fibroblasts and lower efficiency of infection in cardiomyocytes. Although approaches to direct cardiac reprogramming have drastically improved over the last several years, there are important challenges that remain, Srivastava said, including delivery, safety, and regulatory issues.

During the panel discussion a workshop participant asked Srivastava if converting fibroblasts into cardiomyocytes results in a depletion of the fibroblast pool. “Fortunately, [fibroblasts] can reenter the cell cycle and proliferate,” Srivastava said, and his team has not to date observed an issue with depleted fibroblast pools.

EXOSOMES AS NEXT-GENERATION THERAPEUTIC CANDIDATES

The process of culturing cardiosphere-derived cells (CDCs) was first described in 2007 and has since been reproduced by at least 26 labs worldwide, said Eduardo Marban, the director of the heart institute and a professor of medicine at Cedars-Sinai. To create CDCs, small amounts of biopsied cardiac tissue are placed in dishes coated with fibronectin. Stromal-like cells arise from adherent cardiac explants, and they are then re-plated onto non-adherent plastic, poly-lysine-coated dishes where they self-assemble into three-dimensional organoids called cardiospheres. The cardiospheres are then transferred to fresh growth medium and their numbers divided multiple times to yield the therapeutic candidate CDCs. CDCs are uniformly positive for the TGF-β receptor accessory subunit endoglin (also known as CD105) and negative for CD45 and all other hematogenous markers, Marban said. CDCs secrete stromal cell-derived factor (SDF-1), he said, and can induce the secondary secretion of SDF-1 via exosomes that contain a distinctive panel of microRNAs and other non-coding RNAs.

CDCs have been tested in the clinic several times, and the results of the first prospective, randomized trial, titled Cardiosphere-Derived Autologous Stem Cells to Reverse Ventricular Dysfunction (referred to as CADUCEUS), were published in 2012 (Makkar et al., 2012). In the CADUCEUS trial, autologous CDCs were administered to 17 eligible patients, with 8 patients receiving the standard of care. In this Phase I, proof-of-concept clinical trial, patients who received CDCs had an increase in viable myocardial tissue and regional contractility, along with a reduction in scar mass and regional systolic wall thickening (Makkar et al., 2012). The CADUCEUS trial demonstrated that CDCs are safe, and researchers have gone on to create an allogeneic version that is viable even in non-immunosuppressed patients, Marban said. Enrollment was just completed for a Phase II multi-center, randomized, placebo-controlled, double-blinded study of allogeneic CDCs in patients with mild heart failure after myocardial infarction. Other trials are under way to study allogeneic CDCs for advanced heart failure and for Duchenne muscular dystrophy–related cardiomyopathy.

Even though CDCs are cardiac progenitor cells, their mechanism of action is paracrine in nature, Marban said. CDCs have regenerative effects on the heart, including the promotion of cardiomyogenesis, the prevention of cardiomyocyte apoptosis, and increased anti-fibrotic and

anti-inflammatory effects. Transplanted CDCs do not proliferate, differentiate, or produce new tissue of donor origin, Marban said, but rather they survive for several weeks while secreting factors that lead to new healthy tissue of host origin. Discovering this led researchers to wonder if there was a single entity that could mimic all the salient benefits of CDCs which could be developed into a cell-free therapeutic approach. One possible solution was exosomes, bioactive nanoparticles that are secreted by all eukaryotic cells and present in all body fluids. Exosomes are 30 to 150 nanometers in diameter and contain a unique complement of microRNAs and other bioactive contents that vary depending on the cell type and culture conditions. A recent study demonstrated that CDC-secreted exosomes reproduce the therapeutic regeneration associated with the administration of CDCs and that inhibiting the production of exosomes in CDCs negates their positive therapeutic effects (Ibrahim et al., 2014). In response to a workshop participant’s question about reproducibility, Marban said that his team has sent exosomes to various other labs to see if their results can be replicated and if there is bioactivity in other model systems. Their results have been verified in a few other models, he said, noting that CDC-derived exosomes inhibited human T cell degranulation in antibody-dependent cell-mediated cytotoxicity, and induced regenerative effects on skeletal muscle in the mdx mouse model of Duchenne muscular dystrophy. Although CDCs are initially derived from heart tissue, their bioactivity may be applied elsewhere in the body, Marban said.

Several studies have indicated that CDC-derived exosomes have the same regenerative, anti-inflammatory, anti-fibrotic, anti-apoptotic, and immunomodulatory effects as CDCs themselves (Aminzadeh et al., 2015; Chimenti et al., 2010; Ibrahim et al., 2014; Li et al., 2012; Makkar et al., 2012; Smith et al., 2007; Tseliou et al., 2014a,b). In an attempt to understand if one specific component of exosomes was responsible for the regenerative effects, researchers compared the microRNA profiles of exosomes from CDCs to those from normal human dermal fibroblasts. One particular microRNA, miR146a, was heavily enriched in CDC exosomes, but no single RNA species can account for all the benefits of CDC-derived exosomes, Marban said. Instead, “it is the totality of the contents that are required for full manifestation of bioactivity,” he said.

Recently, Marban’s team has turned their attention to newts, amphibians with an exceptional capacity for regeneration. Although newts separated from the mammalian lineage approximately 300 million years ago, there may be important lessons to learn from newt cells, Marban said.

Early experimental results indicate that A1 cells, a naturally immortal type of cell found in newts, have the ability to generate exosomes that contain eight times as much RNA per particle as mammalian exosomes and are bioactive in mammalian injury, Marban said.

Bringing therapeutic exosomes to the clinic required that the technology be transferred to a company that could take it to the next step, Marban said. Therefore, in 2005 he co-founded Capricor, Inc., to further the research on CDCs and exosomes, among other technologies. Marban emphasized that the company is managed exclusively by scientists and clinicians and is supported by grants from the California Institute for Regenerative Medicine, the Department of Defense, and the National Institutes of Health. Capricor’s business model includes a wide range of activities including discovery through development, manufacturing, regulation, and clinical trials design and management, he said.

REGENERATIVE THERAPIES FOR LUNG DISEASE

There is a great unmet clinical need for lung regenerative therapies, but unfortunately the field lags behind many other tissues and organ systems, said Brigitte Gomperts, an associate professor of pulmonary and pediatric medicine at the University of California, Los Angeles. This may be due, in part, to the structural and functional complexities of the lung, she said. In the upper airways, the proximal cartilaginous airways are in direct contact with the environment, and they produce mucus to trap bacteria and viruses and other particles from, for example, pollution. There are also ciliated cells which beat unidirectionally to move the mucus up and out of the body. In the lower airways, gas exchange occurs at the level of the alveolar sacs, which provide a very large surface area where the epithelial and endothelial cells come together to allow the diffusion of oxygen into the capillaries and of carbon dioxide back into the alveoli spaces to be breathed out. The lung contains more than 42 different cell types, Gomperts said, and there does not seem to be one specific stem cell that makes all lung cell types. Lung diseases are complex because all of the anatomical areas of the lung and multiple cell types are affected.

The field of lung disease research tends to be divided into two main areas, Gomperts said. The first area is focused on monogenic lung diseases, such as cystic fibrosis (CF), and researchers in this area have made great therapeutic advances, she said. The second focus is on

complex lung diseases, where unfortunately progress has been much slower for a number of reasons, she said (see Table 5-1).

CF results from an inherited mutation in the cystic fibrosis transmembrane conductance regulator (CFTR) gene. CFTR functions as an ion channel, and although CF is a systemic disease, it has major implications in the lungs, Gomperts said. Currently there are two Food and Drug Administration–approved drugs available for CF; however, there are subsets of patients with specific mutations who will not respond to these drugs, Gomperts said, and stem cell therapies may hold a great deal of therapeutic potential for those individuals. The strategy underlying potential cell-based therapies for CF involves generating iPS cells from patients, correcting the genetic defect in the stem cells, differentiating them into lung stem or progenitor cells, and transplanting them back into patients. A few of those steps have been completed, including the creation of iPS cells from CF patients and the ex vivo gene correction of the most common CF variant, Gomperts said. However, there are many challenges remaining, including determining the correct stem or progenitor cell to use, finding novel ways to expand the cells, and identifying the optimal delivery and engraftment approaches.

TABLE 5-1 Successes and Challenges in Regenerative Therapies for Lung Diseases

NOTE: √ indicates that the specific challenge or issue in the left-hand column has been overcome and × indicates that further research is required.

SOURCE: Brigitte Gomperts, National Academies of Sciences, Engineering, and Medicine workshop presentation, October 13, 2016.

The progress made toward CF therapies is in stark contrast to research on complex lung diseases, such as complex obstructive pulmonary disease or idiopathic pulmonary fibrosis (IPF), for which there are few therapies currently available. This may be due in part to a lack of understanding about the interactions between the genetic and environmental factors that contribute to these diseases, Gomperts said. To address the therapeutic gap for complex lung diseases, Gomperts and her team investigated the possibility of using iPS cells as a model to study IPF. Their protocol involved removing fibroblasts from the damaged lungs of IPF patients when they were undergoing transplantation. Although the IPF lung fibroblasts came from extensively damaged lungs, the cells were phenotypically and genotypically almost identical to normal fibroblasts, Gomperts said. Next her team generated iPS cells from the IPF fibroblasts, allowed them to spontaneously differentiate, and placed them onto 12-kilopascal hydrogels, which mimic the stiffness of the IPF lung. After about 2 weeks on the hydrogels, the cells exhibited a progressive phenotype of aggregation. Immunostaining was used to test the cell aggregates for markers of fibrotic foci, which are the hallmark of IPF. The researchers found alpha smooth muscle actin staining, a marker of activated fibroblasts, as well as collagen production and evidence of proliferation. Further examination revealed that the patient-derived iPS cells had elevated levels of cytokines and chemokines, increased levels of TGF-β activity, damage associated molecular patterns, and increased cellular stiffness—all of which are similar to features of the lung tissue of IPF patients. These cells may be a very useful model of IPF in vitro and could be used for disease modeling as well as drug screening, Gomperts said.

Recently researchers have made exciting advances in bioengineering which will greatly aid lung disease research, Gomperts said. One such advance is the development of a human lung “small airway-on-a-chip,” which consists of primary cells seeded in a two-chambered microfluidics device, allowing for the analysis of organ-level lung pathophysiology (Benam et al., 2016). In her lab, Gomperts and other researchers are attempting to mimic the lung’s cellular architecture in a three-dimensional model system. To do this, researchers allow cells to adhere to alginate beads coated with collagen and dopamine. A rotating bioreactor encourages the cell-coated beads to come together to form organoid-like structures. Sectioning through the organoids reveals that the tissue very closely mimics the three-dimensional structure of normal

human alveolar lung tissue, Gomperts said. This bioengineered “lung” could be very useful for disease modeling and drug discovery.

Looking toward the future of lung disease research, Gomperts commented on possible next steps. One goal should be to generate three-dimensional lung models complete with respiratory membranes that mimic gas exchange, she said. Ideally, regenerative approaches for lung diseases in the future would involve introducing a scaffold or a large amount of functional tissue back into patients, she said.

PANEL DISCUSSION

Partial Reprogramming

A workshop participant queried the panelists about the issue of cells that only get partially reprogrammed during in vivo reprogramming. The participant hypothesized that partially reprogrammed cells may have some deleterious effects. Srivastava responded that he is concerned about partially reprogrammed heart cells acting as a nidus for rhythm disorders because they are not as electrically mature. While they have seen evidence that some of the cells are indeed partially reprogrammed, they have not yet seen evidence of arrhythmias in animal models. There is evidence of a percentage of fibroblasts expressing the reprogramming genes, but not of them undergoing conversion to cardiomyocytes, Srivastava said. These cells remain in a fibroblast-like state, he said, but they do not function like normal fibroblasts and may not lay down as much collagen. These partially reprogrammed, altered fibroblasts may contribute to the improvements observed after attempted cell reprogramming.

Unique Challenges

“What are the unique challenges for this area of medicine besides the fact that this is a new field?” asked Jiwen Zhang, the senior director of regulatory affairs in the Cell Therapy and Regenerative Medicine Division at GE Healthcare and the session moderator. Unlike the case for a single drug or single therapy, these new therapies are extremely complicated, Srivastava said, and likewise, the development and regulatory processes are also complicated. However, he noted, there is a huge unmet medical need that has no other readily available solution, so researchers should not be deterred by the complexities. Marban added

that one of the challenges is having a thorough understanding of the complexities in order to be comfortable with safety and mechanism of action of a new therapy. “In the future it may be possible to come up with defined cocktails of the active factors that may reproduce many of the desired effects of cell therapy as next-generation products,” he said. Cell-based therapies for lung diseases are likely not going to be solutions that use just one type of cell, Gomperts said, but instead they may be organoids or scaffolds. She said that these types of therapies are likely still very far away, but that cell-based disease modeling and drug screening are “low hanging fruit” that may be achievable in the near future.

Physician Involvement

Next, Zhang queried the panelists about how clinicians can be engaged in the research process, specifically in terms of managing patients and helping them navigate these new technologies. All three panelists noted that in addition to their research roles, they are physicians who still work in the clinic and see patients. The training that physician scientists receive is very important, Marban said, because it introduces a certain level of “healthy skepticism,” whereas clinical training alone may not provide that. It was extremely challenging to get physicians to administer new cholesterol-lowering medications, Srivastava said, indicating that the adoption of cell-based therapies by physicians might also be a huge challenge. Even though she works closely with a pulmonologist, Gomperts said, it has been very difficult to collect patient samples, and she added that it “is going to take a lot of work and a lot of interactions and collaborations to really get the physicians on board.”

Scaling Up

Organs such as the lungs and heart are much bigger than the macula, noted a workshop participant, who went on to ask the panelists whether it will be a challenge to scale up cell-based therapies for larger organs. The challenges of such a scale-up will be significant but not impossible, Marban said, noting that bioreactors and other amplifying mechanisms may be useful. Exosomes may be another part of the answer, he continued, because a single eukaryotic cell can make several thousand exosomes per day, which explains why “a few cells seem to make a difference.” The scale-up of production to generate a billion or more

pluripotent stem cell–derived cardiomyocytes is less of a problem today, Srivastava said. The major challenge is getting those cells to engraft and survive upon transplantation, he said. That is one reason why his research is increasingly focusing on harnessing the regenerative power of the resident cells.