Images of Science: Making the Secrets of Nature Visible

Mount Desert Island Biological Laboratory (MDI Bio Lab) scientists work collaboratively to understand the secrets of how nature works, seeking to apply that understanding to improve the human condition. Our focus is recognizing novel cellular and molecular mechanisms of disease and translating our basic research discoveries into potential therapies.

Our approach of seeking answers within the natural world, paired with a deep understanding of the scientific method, defines MDI Bio Lab. Images of Science is centered on the belief that both art and science offer legitimate ways to better comprehend our world. Comparing the artistic and scientific processes helps refine research strategies and better communicate our work. What can we learn from nature through the lens of a scientific mind? What do we see in nature through an artistic lens?

The old adage, “a picture is worth a thousand words” rings true within the larger scientific community and here at MDI Bio Lab. Advanced technologies allow us to better understand science and make it visible. The images we produce are an important part of our publications and the basis of our communication with the world around us. As scientists, we are accustomed to using images for their condensed information and rarely reflect on their multiple levels of meaning, the lens they provide to understand the world, and their aesthetic qualities.

Our images contain both scientific information and the beauty of the natural world. This relationship of perception and understanding lies at the heart of making new observations and experiencing nature in a more nuanced way. Teasing out an image’s meaning(s), we are further engaged, seeing additional layers of beauty and interpretation. To enjoy the art of science, we need the time to look at images and truly ‘see’ them; the context of their creation and what they ‘mean’ provide a secondary level of understanding. This is our experience in art museums as well: we are impressed by the objects themselves and thrilled to discover the stories beneath the surface.

We hope to evoke that experience here: the beauty of nature is to look and to understand. We explore this concept mimicking the way we ‘do’ science, employing various tools and scientific methods used by our scientists to unravel the secrets of nature. There are variations in techniques and styles for image generation in both art and science. The way Rembrandt used colors is very different from the approach Andrew Wyeth chose, and the techniques of Picasso are far from the brushstrokes of Renoir. An awareness of the methodologies used and the underlying view of the natural world provides context for appreciating the image and its meanings.

MDI Bio Lab scientists are at the forefront of science, pushing the boundaries of how to see and reveal secrets of nature. We continue the scientific adventure that started more than 500 years ago with novel methods like the anatomical knife and the microscope. Over the last decade, these methods have been greatly refined and are now technologically demanding; sometimes the technology of our science overshadows our questions and findings.

Here we reflect on the biomedical scientific process and demonstrate the tools our researchers use to visually analyze, dissect, and detect. In four sections, we present images from our research projects which can lead us through the process of scientific discovery, revealing the underlying mechanisms of regeneration and aging. This means first identifying the question we want to tackle, finding the structures which are affected, discovering which cells are involved, and finally, pinpointing which molecules interact during these reactions and which genes are turned on and off. Tracing the evolution of the scientific tools we employ, we bring to light the layers of examination and understanding required of basic research.

1. Observation with the Naked Eye

Close observation of the world around us, combined with an open mind to ask relevant questions, is the first step in exploring the wonders of nature. At MDI Bio Lab, our core subjects of metamorphosis and aging begin with straightforward visual analysis.

2. Under the Skin: Anatomy and Dissection

After scientific observation, we move under the skin of our models. To look beyond the surface of our research subject and understand how it ‘works’, we dissect and go deep into the tissue to find structure and principles.

3. Microscopy

MDI Bio Lab’s advanced microscopes allow for a multi-faceted, clear look into the mysteries of nature. Using microscopes as a tool for both science and art, the boundaries of what structures and processes can be made visible is pushed.   

4. Into the Unseeable: Molecules and Genes

A major revolution in biomedical science is the introduction of genetic tools which allow us to tag individual genes and follow them through space and time. This information requires mathematical tools to be accessible and usable; the imaging of these mathematical analyses makes the data understandable and visible. The results are both stunning for scientific purposes and for their aesthetic value.

Art Meets Science beautifully enhances the MDI Bio Lab’s research and education activities, helping us to develop a complete mind and explore how art can communicate and catalyze science. As a community, we will study the science of art; study the art of science. Learn how to see. Realize that everything connects to everything else.

We hope that you enjoy Images of Science and that we Make the Secrets of Nature visible for you.

— Hermann Haller, M.D.

Our 2024 Art Meets Science exhibition is curated by Hermann Haller, M.D., President of MDI Biological Laboratory

Images of Science begins with the perspective of the visual artist on science.  Michael Takeo Magruder, our 2024 MDI Bio Lab artist-in-residence, and Maine-based artist Amanda Lilleston explore biomedical science in their art. They use different media: Lilleston works with collage and woodcuts, a technique first employed in the 16th century to image the body and its anatomical parts, while Takeo Magruder uses AI and video imaging, methods we have introduced into the imaging of science only in the last decade. Both open our eyes to a powerful visual experience of anatomy and biomedical research, and lead us to the underlying questions explored in this exhibition: What do we create when we image science? What are scientific images? What do those images tell us?

 

Michael Takeo Magruder

The Four Horsemen of the Apocalypse are among the best-known characters from the Book of Revelation, even though they appear only briefly in the text. These riders – Conquest, War, Famine, and Death – are seen as the harbingers of humanity’s Last Judgment, and as such, have been the focus of much attention and speculation throughout the Book’s reception history.

The Horse as Technology is a multifaceted new media installation that proposes the ‘horse’ of Revelation is a symbol of technology – an extension of the human body and will – embodying transformation and maintaining the power to either create or destroy. The artwork’s full form is a sculptural environment that is reminiscent of a scientific laboratory filled with modern digital production systems and processes related to 3D scanning, visualization, and printing.

This video component of the installation shows the de/re-construction of the artwork’s central horse skull as a 3D wireframe alongside sequences of its precursory binary data. Viewers are able to see two different manifestations of the same digital entity and reflect on the influence of autonomous technologies to both engender new forms of creation and bring about destructive (perhaps even apocalyptic) changes in ways that are quite similar to the potentialities of the horse in ancient times.

About the Artist:

Michael Takeo Magruder is a visual artist and researcher whose practice utilizes Information Age technologies and systems to examine our networked, media-rich world. In 2020, Takeo was the first ever artist-in-residence at the UK National Archives where he reflected upon the institution’s ongoing digital transformation and what constitutes an archive in the 21st century. He is presently MDI Bio Lab’s Art Meets Science artist-in-residence and is developing a new body of artwork in-dialogue with the Laboratory’s world-class research community.

Amanda Lilleston

Amanda Lilleston’s woodcut collages lead us into the anatomical theatre. These striking  images present  anatomical structures that are clearly corporeal. As we look inside the body, we are confronted with torn pieces and dissected organs. Yet at the same, there is a beauty to be admired as we venture beneath the skin.

Amanda Lilleston is assistant professor of art at INBRE institution Colby College.  Lilleston earned her BA in biology at the Colorado College, Colorado Springs and received her MFA from the University of Michigan. Her artwork focuses on biology, plants and flowers, and the human body, reflecting her life as an educator and her personal experiences.

In her own words, “I am interested in understanding the vulnerability and strength of the human body. In much of my work, I investigated the inner workings of our anatomy. This exploration began by watching surgeries, researching physiology, and assisting in dissections. I used these methods to investigate, first-hand, the components that make us human. My primary medium is woodblock printmaking. I cut the woodblock surface by peeling, digging, and scraping, recording each stage of carving by inking the block and printing. I use multiples, collage, and chine-collé to transform my prints into new forms. I think of these as human parts: susceptible to gravity, discomfort and decay. The form on the paper becomes saturated and heavy, and over time structures develop, details accumulate, and layers fuse. While my imagery is grounded in reality, exaggeration and abstraction of shape, form, and color create new structures and spaces.”

Using woodcuts to image the human body and its anatomy has a long tradition.  The first modern book on human anatomy stems from the famous work of Andreas Vesalius in the 16th century.  He used artistic woodcuts for visualizing human anatomy. Like in Lilleston’s work, science, education and art are intertwined in these early images of the human body.  And they share the precision of the woodcuts, the focus on the anatomical detail.  After the first visual experience of aesthetic beauty, the viewer immediately questions: What is this?  What are the structures? Where do they come from?

Earlier this summer during their first morning at MDI Bio Lab, our 2024 summer undergraduate students sat in front of Lilleston’s work and immediately asked these questions.  The reflective discussion that followed, as the students made sense of what they saw, pivoted between art,science, and education. In Lilleston’s anatomical art her personal experiences as a patient are also represented.  Her thoughts about her own body, as well as her experience with disease, pain and fear are part of her work.  Using powerful, bold colors, and transforming what are usually small hidden pieces of our anatomy into towering, large scale images reminiscent of religious relics Lilleston provokes a visceral, personalized response in her audience.

This relationship between the experience of disease and the art is not uncommon. We have famous examples such as the art of Frida Kahlo relating to a failed pregnancy or the transformation of Francisco Goya’s painting after his loss of hearing. The cathartic creation of art becomes a channel for the pain and challenge of living with the reality of damaged, afflicted bodies. Amanda Lilleston’s artwork forces us to face our own vulnerability, as well as the strength gained through perseverance with disease.

Imaging science has a long tradition. During the Renaissance period in Europe (1400-1600) artist-scientists used drawings and paintings to diligently depict nature and its wonders, to catalogue knowledge and to describe mechanisms and principles of the world as they saw it. Famous examples of images of nature during this period are from Albrecht Dürer and Sybille von Merian.

Murawala Lab

A master of regeneration and with a genome 10 times larger than a human’s, the axolotl carries secrets to understanding how to regrow lost or damaged tissue. Most amphibians begin their lives as aquatic animals that are unable to live on dry land, often being dubbed as tadpoles. To reach adulthood, they go through a process called metamorphosis, in which they lose their gills and start living on land. This Mexican salamander is unusual in that it lacks a thyroid-stimulating hormone, which is necessary for metamorphosis to occur. Therefore, the axolotl keeps its gills and lives in water all its life.

At MDI Bio Lab, Assistant Professor Prayag Murawala, Ph.D., studies the axolotl both in its Neotenic and Metamorphic states in an effort to unlock its ability to heal and regenerate. His long-term goal is to understand the implications of this process on human health and wound-healing.

A quick visual analysis demonstrates a simple differential comparison. On the top left we see a Neotenic animal with its bright pink external gills allowing it to breath and flourish underwater.  On the bottom left are two Metamorphic animals, their fin and gills lost, their skin tightened and their body weight reduced as they thrive on dry land.

The process of Metamorphosis is a wonderful example of how looking closer and digging deeper provides an opportunity for ever more precise understanding of the secrets nature holds.

Haller Lab

In recent years, MDI Bio Lab scientists have developed the African Turquoise Killifish (ATK) as a model for studying aging. Its short lifespan of four to six months offers rapid insight into the effectiveness of anti-aging interventions, while its biological traits, including vertebrate-specific genes and organs and a complex immune system, are directly relevant to the study of human aging.

Here we see two pairs demonstrating opposite ends of the killifish lifespan curve. The young pair is 8 weeks old, which is equivalent to a 20 year old human. In contrast, the older ATK pair is 18 weeks old, a point where half of all fish their age have died, similar to 70 year old humans. The typical ATK reaches very old age around 20 weeks, equivalent to about 90 years of age for humans. Recently, the Haller Lab at MDI Bio Lab has increased the lifespan of ATK to over 50 weeks – equal to more than 200 human years!

With the naked eye, we can see the phenotypical signs of aging in the ATK, similar to how humans age: changes in color, hunched backs, reduced mobility, and memory loss.  In comparing these images the aging process is clearly visible, a benefit to the researchers who can visually determine not just the chronological age, but the biological age of the fish.  This plainly visible decline allows the Haller lab to study how to delay the onset of aging in the ATK and its potential implications for humans.

Dissection of the body has a long tradition reaching back to the Middle Ages. It was during the Renaissance period that scientist and artists began to sketch and paint what they saw under the skin. It can be difficult to tell the difference between art and science during this time.

Leonardo’s drawings are an early example of art and science.  To understand and explain the human body he dissected experimentally, he observed carefully and illustrated masterly.

Medical students and doctors have been trained by anatomical textbooks for 500 years.  Starting with Vesalius in the 1530s  to Gray’s Anatomy, images of the body have structured our understanding of health and disease.

Leonard Worcester Williams

Leonard Worcester Williams, M.D.
Lab notebook drawings, 1904
Colored pencil on paper

Charles Darwin’s On the Origin of Species was published in 1859, inspiring scientists to go into the field to study, compare, and elucidate basic biological processes in a diverse range of organisms. The MDI Biological Laboratory was founded in 1898 as a summer research and education facility to provide this opportunity. Its location on the rocky Maine coastline, in Harpswell (1898-1920) and since 1921 in Salisbury Cove, offered access to countless diverse animal species.

Leonard Worcester Williams (1875-1912) was an Assistant in the Department of Invertebrate Zoology at the American Museum of Natural History in the early 1900s, before joining Harvard University’s School of Medicine as an instructor in Comparative Anatomy. It was while at Harvard that Dr. Williams became a colleague of MDI Bio Lab founder John Kingsley and came to the laboratory in Harpswell as a visiting researcher.  His lab notebook drawings detailing the anatomy of model organisms of the time, the dogfish and common gull, speaks to the practice of close observation of the natural world.

The skill of visual analysis led to careful documentation of what the scientists observed.  We no longer have the written explanations of Dr. William’s work, but his observations via hand drawing remains in his original lab notebook from 1904. Williams diligently peels back the layers of his model organism to simultaneously lay bare the external appearance and internal functioning of his research subject.  The act of drawing not only allowed his findings to be immortalized and shareable with his students and marine biology colleagues, but documents Williams’ perceptions of what he considered to be significant. Drawing, like writing, helps us understand our subjects. With this visual representation of what he sees, Williams shows the shapes and colors of anatomical parts, the relationship of one part to another, to the whole organism and illustrates how these parts interact.  Beyond simple documentation, what he chooses to draw and how he chooses to represent them hints at his own understanding of what he sees.

Evelyn Olsen Kok

Evelyn Olsen Kok’s artistic career spanned 70 years, beginning with her education at the School of Practical Arts (now a part of Lesley University) in Cambridge, MA. After marrying Jan Kok, a professor of music, the couple settled in Presque Isle and began teaching at the University of Maine. Kok’s art spanned watercolor, sculpture, paper and fiber arts, and handcrafted various musical instruments. In 1970 she discovered Stonington Harbor on Deer Isle and purchased an old building on the waterfront that evolved into a gallery and studio for the couple. Today, her niece Christina Shipps continues her legacy, managing The Art of Evelyn Kok Gallery in Stonington, Maine.

It was in 1949 that Kok began to work with Dr. Richard Overholt, head of an internationally respected thoracic clinic in Brookline, Massachusetts, and launched a successful career in medical illustration. During this time, artists would accompany surgeons into the operating theater to construct a visual record of the details of the procedure being performed.

Kok called herself “an interior decorator,” and her wonderful sense of humor gave her the edge to observe and document scenes not for the faint of heart. Peering over the shoulder of a surgeon, Kok had a front row seat to the procedure, the technique, the tools, and the moments of education found within a teaching hospital focused on pulmonary disease.

The images she made in the operation theatre are a wonderful example of how scientific illustrations are more than a photographic image. Kok draws what she sees and documents what the surgeons describe to her. While a photograph can capture a scene, the hand-drawn nature of these illustrations directs the focus and offers space for annotation, labeling, and a stunning scientific exactitude. In fact, the goal of the drawing was not the objective representation of what the eye saw, rather it was documentation of a successful operational procedure created for the purpose of educating young doctors and students. These drawings provided generations of young surgeons with a map of the thorax and surgical guidance.

Through her artwork, Kok contributed to decades of medical knowledge combining science and art to communicate the anatomy and methodologies critical to thoracic surgery. These are more than clinical representations; Kok’s artistic style and skill allowed her to utilize color and line to simultaneously draw the eye to an instructive point of focus while maintaining a pleasing aesthetic quality.

Goro Eguchi, Ph.D.

Goro Eguchi, Ph.D.
Study of the Lens of the Newt, 1997
Pen and ink.
On loan from James Godwin, Ph.D.

Dr. Goro Eguchi’s artistic talents led to his most fruitful scientific relationship.  As a 3rd-year student in Nagoya, Japan, he was asked by his future mentor, Professor Sato, to draw an illustration of a fellow Scientist and Nobel Laureate’s house for the cover of his biography. Impressed by his skills, Prof. Sato offered Eguchi a graduate student fellowship in his lab. Eguchi would go on to dedicate 50 years of research as a developmental biologist studying the regeneration of the lens in the eyes of newts.

Sketching with pen and ink, Eguchi translates electron microscopy data of the late 1980s-1990s. Eguchi’s drawings are a master class in using fine art to describe a regeneration process without words. Here he clearly identifies macrophages as a key cell type in this process in 1988/1992.  His prescience of the importance of macrophages, a type of white blood cell that surrounds and kills microorganisms, removes dead cells, and stimulates the action of other immune system cells, inspired many of the young researchers who followed his work, including MDI Bio Lab Assistant Professor, James Godwin, Ph.D.  Today, Godwin performs modern immunostaining of the eye structures in the axolotl (a Mexican salamander) with modern tools, building upon the work of Eguchi’s own examination of regeneration.

The invention of the microscope was a major step for the imaging of science; its introduction added a new dimension to looking at the world. Suddenly the world had become much larger. More objects, more details and a better understanding of body functions became possible. Gaining insight into nature became more technical, while technical innovations triggered imaging and imagination. Under the microscope the imaging of objects now required measurements and scaling.

The first microscopes were made in the Netherlands around 1600 by Antonie van Leeuwenhoek. These instruments could be carried around in your pocket and used to look at the hidden wonders of the world.

Advanced custom-built light sheet microscope at MDI Bio Lab in which comparatively large animal samples can be moved efficiently through sheets of light produced by lasers to build 3-dimensional images of entire organisms, zoomable down to scales smaller than a single cell.

Madelaine Lab

Contrary to humans, some animals have the powerful capacity to fully regenerate damaged tissue and organs, including the nervous system. Dr. Romain Madelaine’s research focuses on the function of the stem cell population during tissue regeneration, using the zebrafish as the model organism.

The Madelaine Lab studies the regeneration of retinal ganglion cells and the optic nerve in zebrafish. In humans, retinal ganglion cells and optic nerve degeneration are associated with the neurodegenerative disease glaucoma. Contrary to humans, zebrafish can regenerate retinal neurons and the optic nerve.

Dr. Madelaine aims to understand the molecular and cellular mechanisms underlying this process. The Madelaine lab’s long-term objective is to apply their discoveries in promoting retinal neuron regeneration in humans.

The image above shows the dorsal view of a zebrafish larva brain using fluorescent confocal microscopy. Retinal neurons (retinal ganglion cells) form the optic nerve which is labeled in green, axons from neurons are labeled in magenta and synapses from axons in cyan.

In the image above, we see the microscopic dorsal view of a zebrafish larva brain generated to better understand the function of a multipotent cell population (cells that can self-renew and differentiate into multiple specialized cell types in a specific tissue or organ) called neural crest cells. These cells have the capacity to differentiate into multiple lineages, including melanocytes, neurons, oligodendrocytes and peri-vascular pericytes. Melanocytes express melanin: a black pigment. In the first image, you can clearly see the retina of the zebrafish eye is highly pigmented, indicating a concentration of melanocytes.

Zooming further into the dorsal view of a zebrafish eye, the third image shows a retina imaged with fluorescent confocal microscopy. Neural crest cells are labeled in green and peri-vascular pericytes are labeled in magenta. The Madelaine Lab explores the function of a peri-vascular (tissues surrounding a blood vessel) cell population called pericytes during retinal neurons and optic nerve regeneration. This peri-vascular cell population potentially shares the multipotency (or ability to differentiate into multiple specialized cell types) of the neural crest cell population. The Madelaine Lab aims to test the hypothesis that pericytes underly a new mode of neuronal regeneration in the zebrafish retina. This discovery would open new avenues to promote the formation of new retinal neurons and treat glaucoma in humans.

Murawala Lab

Further exploring the variations between Neotenic and Metamorphic axolotl, the Murawala lab characterizes the differences in the skin of these model organisms under the microscope.

Here, we see the upper arm limb skin from (A) Neotenic and (B) Metamorphosed sibling axolotl, sectioned, and stained with H&E. H&E is the combination of two histological stains: hematoxylin and eosin, that illuminates the skin architecture and the underlying Extracellular Matrix (a complex scaffold of biologically active molecules critical for determining the action of the cells it surrounds).

The hematoxylin stains cell nuclei a purplish blue and eosin stains the extracellular matrix and cytoplasm pink, with other structures taking on different shades, hues, and combinations of these colors.

This staining makes evident that the epidermis (outer most layer of skin) is pseudostratified (closely packed cells arranged in layers), whereas the dermis (the middle layer of skin) is thinner and denser in the metamorphic axolotl. Seeing the rigid skin of metamorphic animals acting as a barrier to limb regeneration prompts us to ask how does loose skin of neotenic animals promote limb regeneration?

Updike Lab

These three images above depict the head regions of living Caenorhabditis elegans (C. elegans) nematodes and represent two microscopy techniques used in the Updike lab at the MDI Bio Lab. How do stem cells maintain pluripotency — the capacity to become other cell types in the body? Can our ability to regenerate after disease or injury be improved with the ability to switch on and off cellular pluripotency? To answer these and other questions, the Updike Lab derives cues from quintessential stem cells found in the germline. Germline stem cells (GSCs) are the precursors to oocytes and sperm, and these cells must retain their stem cell attributes so that fertilized embryos can give rise to all of the cell types of each subsequent generation.

C. elegans are commonly used in biomedical research to study genetic pathways relevant to human health and disease. These images feature a newly engineered strain, created using CRISPR (a technology used to selectively modify the DNA of living organisms), which expresses a fluorescent protein (mScarlet, purple) in two pairs of mechanosensory neurons (neurons that can sense mechanical stimuli as pressure or vibration)known as the FLP neurons. These neurons have extensive dendritic, or immune, processes throughout the worm’s head.

The first two images were captured using a widefield epifluorescence microscope, typically used for routine work in the lab. The yellow channel outlines the nematode heads with transmitted light, while the cyan channel highlights autofluorescent granules in the intestine. The third image was taken with a higher-end confocal microscope, offering enhanced resolution, particularly in capturing the smallest dendritic processes and commissures.

This strain of C. elegans, developed by Dr. Dustin Updike and Dr. Emily Spaulding and created by Hugo Schodet and Catherine Sharp, will aid in tracking the movement of biomolecular condensates within these neurons in living worms, comparing “wild-type” and “mutant” genetic backgrounds where abnormal condensation affects the function of these mechanoreceptor neurons.

Drummond Lab

The Drummond Lab established the zebrafish as a model for studying kidney development, human kidney disease pathology, and kidney injury/regeneration. They study how the kidney develops, identify zebrafish genetic mutants, and determine key points of equivalence between fish and mammalian kidneys that make zebrafish a relevant model of human disease. Fish make new nephrons after kidney injury to replace lost organ function.

These images show newly formed kidney nephrons in the process of fusing to the existing kidney system. The images reveal the cellular structure of the tubules (green), cell nuclei (blue), and nuclei that have divided recently (red).

Also visualized are individual mRNA molecules encoding proteins associated with the fusion event (lef1 brown dots; wnt4a magenta dots) and localized right at the junction of two tubules. Identifying genes expressed right at the fusion junction reveals the cellular process of fusion and gives us tools to engineer tubule fusion and engraft new nephrons.

These images point the way to engrafting stem cell derived kidney tissue into injured or diseased human kidneys to help replace lost kidney function.

The better and larger our microscopes become, the more difficult it is to directly ‘see’ an object and generate what a researcher wants to see. Already with ‘old-fashioned’ electron microscopes (EM) the scientist must make choices throughout the creation of the image that determine its final size, scope, and perspective. During the last decade, a revolution occurred in the imaging of the world: we went from the microscope into the unseeable. It is like in astronomy where we went from Galileo’s telescope to the Hubble telescope and beyond. Our technology allows us to ‘see’ so much that we need mathematics and AI to image it. But we still need the images to capture, comprehend, and make sense of the data.

A Revolution in Gene Identification

Thirty years ago, we were proud to rapidly sequence a couple of genes in our labs. Now, new sequencing methods allow the measurement of all expressed genes in a cell at once, and in thousands of cells simultaneously. It has been almost impossible to analyze this amount of data by reading through millions of data points. We have to create new images to understand the molecular universe. Using the algorithms of artificial intelligence to generate colored dots from individual cells, which then form colorful clusters of similar genetic patterns, we can interpret meaningful patterns of cell and gene behavior.

The first multiplex PCR machine (left) to measure expression of a single gene was introduced at MDIBL in the 1990s.

Today we analyze thousands of genes in thousands of cells simultaneously (single cell sequencing) and use AI and AI-generated images to understand and make sense of it.

 

Murawala Lab

Building upon what they learned through visual analysis, dissection, and microscopy imaging, the Murawala lab next utilizes single cell analysis, spatial transcriptomics, and chord plots as tools to understand what is thyroid hormone induced metamorphosis.

At first glance, the generated images appear to be alternately quilt squares, abstract paintings, or color wheels. Instead, they are AI generated visualizations of thousands of genes in thousands of cells, the communication that happens between them, and the molecules that make these cell-to-cell interactions possible – single cell analysis.

Uniform Manifold Approximation and Projection (UMAP) is a method to image the huge amount of data generated from single cell sequencing. This algorithm takes a high-dimensional dataset (single-cell gene expression) and reduces it to a low-dimensional plot that retains much of the original information. Single-cell RNA sequencing datasets are so vast and multidimensional that it is virtually impossible to find a biologically meaningful pattern by reading through the data points. UMAP is designed to help visualize the vast array of information so that it can be easily studied, making clear patterns within the data. The goal of UMAP in single-cell studies is to place similar cells together and different cells further apart to capture the similarities between cells in a high-dimensional space. This way, dots that form clusters on a UMAP plot can be potentially interpreted as separate cell (sub)types.

Each point in the image at left represents thousands of measurements of RNA, such that each point has an ‘expression profile’ a set of numbers telling us which genes are expressed and at what level. The simple interpretation is that the closer two points are, the more similar their expression profiles are. As such, we infer that they represent similar or even the same cell type. Then, mathematical clustering algorithms are used to design which groups are similar enough to each other (and different enough from others) to be called a distinct cluster.

What is gene expression profiling?

As the building blocks of life, every cell has the same blueprint (DNA that is manifested in our genes). Our cell’s form, there are hundreds of cell types in a human, and function are determined by how that blueprint is interpreted or activated. Genes are the active part of the blueprint, and modern biomedical technology now lets us measure which of the 20,000 or more genes are activated and at what abundance on a cell-by-cell basis. This results in an ‘expression profile’ which is a picture of the complex mixture of genes activated or deactivated in each cell.

We can compare these profiles between thousands of cells in an experimental sample, and mathematics allows us to group them into clusters of cells with similar profiles (as such we infer that they are likely of the same type). We visualize these cells and clusters by generating an image in which each cell is represented by a single point, and the relative distance between any pair of points is a representation of their similarity (cells of a common cluster will usually plot very near each other). We can map the cell profiles of these clusters onto decades of biomedical knowledge to get a likely identification of what type of cell each dot represents.

Chord Plots and Spatial Transcriptomics: AI-generated images to see the world

Cells respond to their environment by changing which genes are activated. Cells communicate via messengers, which bind to specific receptors on neighboring cells. When the messenger and receptor bind to each other, the receptor will start a process inside the cell which alters its gene expression. Since humans have thousands of receptors and messengers (and genes) analysis is virtually impossible. Instead, AI is used to screen all databases of published data from decades of biomedical studies that have identified messengers and receptors, and their effect on genes. Single cell data is then mapped onto these databases to identify which pairs of messengers and receptors are being activated in our samples, which further suggests which processes (or pathways) are being activated in response to the environment. The chord plot is a visual representation of this connectivity, with the different types of cells represented around the outer surface, and colored lines connecting pairs of cells that are activating matching ligands and receptors. A chord plot images the communication and connections between different cell groups, making visible the unseeable. Each cell group is represented by a fragment on the outer part of the circular layout. Arcs are drawn between each the different groups. The size of the arc is proportional to the importance of the communication.

Spatial transcriptomic data is gathered by extracting the genetic material (RNA) from sectioned tissue (here a section of axolotl limb) in a spatially regular grid, and subsequently sequence and assess gene expression. Then the microscopy and the expression data are aligned such that we know what the expression profile is for each region. The dots in the images represent the overall expression profile of each region and dots that are colored with the same color are inferred to have the same cell type. The data represented in the two images on page 44 are both cross sections of an amputated axolotl limb, with the center region (shown in purple) representing the bone, blue and green outward from that muscle.

Using spatial transcriptomics we use the technology to bring the information of profiling the expression of thousands of genes back to the tissue they come from. We try to localize the AI-generated information in a more familiar setting, in a tissue with muscle, nerves, skin and blood vessels. Spatial transcriptomics combines a fascination of the unknown with the sound foundation of a real limb.

 

Prayag Murawala, Ph.D., Joel Graber, Ph.D., Comparative Genomics & Data Science Core
Understanding the molecular communication between cells; spatial transcriptomics, 2024
Digital print

In this particular analysis, data from Neotenic (left) and Metamorphic (right) samples were merged into a single clustering analysis, so that commonalities and differences between the cell activity in both animal types could be characterized. As such, the coloring of dots across all samples is the same, making clear that the biggest differences in cellular activity following metamorphosis are in the outermost skin layer.

What are we doing with these images?

Building upon what they have learned from visual analysis, dissection, and microscopy imaging, the Murawala lab utilizes both single cell analysis and chord plots as tools to try to understand what is thyroid hormone induced metamorphosis. After microscopically characterizing the differences between neotenic and metamorphic axolotl skin, the Murawala Lab seeks to understand what are the underlying molecular changes in key cell types (Keratinocytes and Fibroblasts) of these two presentations of the same animal model. Keratinocytes are the main constituent cells in the epidermis and fibroblasts are a type of cell that contributes to the formation of connective tissue, a fibrous cellular material that supports and connects other tissues or organs in the body.

In examining the ‘cross-talk’ (who produces what ligands, which cell type receives those ligands, and how do they alter skin architecture) between keratinocytes and fibroblasts is altered in neotenic versus metamorphic axolotl. The outcome of rigid skin observed in metamorphic animals, is suspected to be caused by the altered cross-talk in these two cell types. It is only through digging into the unseeable that the secrets of nature can be examined and understood.