Single Cell RNA Sequencing Allows Comprehensive Study of the Genetic Programs of All Cell Types in a Diseased Human Lung
by Frank McCormack, MD, The LAM Foundation Scientific Director
Studying LAM biological behavior in the laboratory is difficult because the model-based approaches that have been so vital to progress in other diseases have simply not been very good in LAM. There are many reasons for the inadequacy of LAM cell models, including that we don’t know the origin of the LAM cell, and that LAM cells grow slowly, change their signature functions with time in a dish, are often outcompeted by other cells in culture, and do not survive well through multiple passages. Immortalizing LAM cells with viruses can keep them alive but alters them in unpredictable and inauthentic ways. LAM animal models are also problematic; there is no known naturally occurring model of LAM other than in humans, and because we don’t know where LAM starts in the body, we don’t know what cell to target with our genetic manipulations. Tumor xenograft models, in which LAM cells are injected or implanted into mice, are also suboptimal because they are typically done in mice with compromised immune function (so they won’t reject the tumors), colonize and metastasize to the lung in an un-LAM-like manner, and induce reactions that are not typical in humans. Because the LAM models we have are imperfect, it has been difficult to be confident about how findings made in them can be translated to humans.
Said another way, although we know a lot about the genetic basis of LAM, the laboratory limitations above have left us with a gap in our understanding of what those mutations do to LAM cell functions. Enter single cell RNA sequencing (ssRNAseq), a technique that allows us to comprehensively study the genetic programs of all cell types in a diseased human lung. We are fortunate to have a strong core facility that offers this sophisticated technique in Cincinnati, through the efforts of Drs. Steve Potter and Jeffrey Whitsett. The LAM Foundation has provided funds to perform approximately 10-12 ssRNAseq experiments in LAM, angiomyolipoma (AML) and normal human tissue to map the ‘transcriptome’ of the LAM cell; that is, to produce a catalog of their most highly expressed genes. In this technique, tissues that are harvested at lung transplant, biopsies or resections are immediately placed on ice, rapidly dissociated into single cells and labeled in a manner that allows them to be easily identified. Using the 10x Chromium ssRNAseq method, about 2,000-4,000 genes in each of up to 10,000 of the dissociated cells are sequenced and identified. The genetic profile of each cell sequenced is displayed as a dot on a two-dimensional plot in which distances between dots reflects differences between gene programs (see figure). The dots segregate into ‘island like’ clusters representing different cell types, such as lymphocytes, macrophages, fibroblasts, endothelial cells, airway epithelial cells, and alveolar epithelial cells. When this process was followed for the two LAM lungs that have been completed, small islands of cells that were not present in normal human lung, and which expressed several of the ‘usual suspects’ known to be made in LAM cells (such as VEGF-D, gp100, cathepsin K, osteoglycin, etc.) were identified (see pink circle in the figure).
By studying these LAM clusters, we can validate the few dozen or genes that we strongly suspected were being expressed in LAM cells, and identify hundreds of new ones with may become future drug targets or useful biomarkers. We might also learn something about where the LAM cell is coming from—for instance if the LAM cluster is found to express many of the genes that are commonly found in the uterus or the ovary, perhaps our attention will be focused to those potential sources. We can also determine what effects LAM is having on other cells in the lung, such as the fibroblasts they are recruiting, the alveolar epithelial cells they are inducing to proliferate, and the immune cells that are trying to kill them. ssRNAseq allows us to study these processes in LAM cells ‘where they live’ and in their natural state, and does not suffer from the many of pitfalls that occur with isolation and manipulation, and other limitations that plague the study of cultured cells.
This is an incredibly exciting time for LAM research. Many of us feel as if we are entering a new era, and that have been given a precious new window into the LAM cell playbook. Based on the advice of the Scientific Advisory Board, the LAM Foundation devoted $80,000 to ssRNAseq, on the condition that the raw data would be immediately available to all interested researchers. Three donors have already generously provided tissues, from two LAM lung transplants and one from an angiomyolipoma (AML) resection. As it has always been, the success of this effort depends on the patients. If you are approaching a procedure where LAM or AML tissue will be obtained, please contact the Foundation and the NDRI by emailing Anne McKenna at firstname.lastname@example.org and sign all necessary paperwork early. It is vitally important to make your wishes regarding research known to your surgical team. The logistics of timely acquisition, processing and shipping of the samples are complicated and easily overlooked on transplant day. The time from resection to ssRNAseq processing is key to success.
Hats off to our first three donors, to our future donors and to The LAM Foundation Board of Directors for immediately recognizing the promise of this cutting edge technology and for being nimble with generous support.