Conception and development of the Second Life® Embryo Physics Course

Abstract

The study of embryos with the tools and mindset of physics, started by Wilhelm His in the 1880s, has resumed after a hiatus of a century. The Embryo Physics Course convenes online allowing interested researchers and students, who are scattered around the world, to gather weekly in one place, the virtual world of Second Life®. It attracts people from a wide variety of disciplines and walks of life: applied mathematics, artificial life, bioengineering, biophysics, cancer biology, cellular automata, civil engineering, computer science, embryology, electrical engineering, evolution, finite element methods, history of biology, human genetics, mathematics, molecular developmental biology, molecular biology, nanotechnology, philosophy of biology, phycology, physics, self-reproducing systems, stem cells, tensegrity structures, theoretical biology, and tissue engineering. Now in its fifth year, the Embryo Physics Course provides a focus for research on the central question of how an embryo builds itself.

Keywords:

• embryo physics
• online teaching
• Second Life®
• virtual worlds

Introduction

Embryo physics had an excellent start in the late 1800s when Wilhelm His investigated the buckling (‘foldings’) of laminates as a model for neural tube closure [His 1874; 1888; 1894]. The idea was unfortunately squelched by Wilhelm Roux [Roux 1888, translated 1964], who showed that embryos with slits cut lateral to the neural plate, which he thought should prevent lateral pushing, nevertheless completed neural tube formation. This apparent contradiction between theory and experiment [Gordon 1999], combined with the rise of biochemistry and its molecular view of life [Gordon 1985], postponed the further development of embryo physics for a century (except for limited forays, reviewed in [Gordon 1985; §1.12 of Gordon 1999; Davidson 2011; Obladen 2011]). In retrospect, both Wilhelms were right: axial elongation of the notochord and attached notoplate [Gordon and Jacobson 1978] (now called ‘extension-convergence’ [Keller 1984; Yin et al. 2009]) generates lateral compression, whether or not tissue lateral to the neural plate is cut [Jacobson and Gordon 1976]. As a result of this misunderstanding and the consequent widespread premature abandonment of a physics approach to embryology, we have a lot of catching up to do.

Approaching Embryo Physics

Embryogenesis has been a lifelong passion of mine, launched while I was a graduate student in Chemical Physics at the University of Oregon, by chance reading of a paper by Magorah Maruyama in which he proposed a simple lattice model for cell interactions in a growing embryo [Maruyama 1963]. He told me that he was inspired by a lecture by Stanislaw Ulam [Ulam 1962], with whom I later undertook postdoctoral studies [Gordon 2011a]. At the time I was learning statistical mechanics from Terrell L. Hill, who then was working out the theory of small systems thermodynamics [Hill 1963; 1964]. I reasoned that cells are small systems, and thus, unlike Maruyama's deterministic rules of cell interaction, behave stochastically. Thus, just as we try to derive the macroscopic parameters of temperature, pressure, and volume from the partly random activity of molecules, I sought the macroscopic form of an organism in the myriad, partly random interactions of its cells [Gordon 1966]. Later work on the self-sorting of reaggregated embryonic cells took me away from the rigid lattice cellular automata [Codd 1968] approach to continuous space, where cells can generate and transmit long range physical forces to one another [Gordon et al. 1972; 1975].

I started at the University of Manitoba in 1978 in Pathology, with a minor appointment in Radiology, and an administrative position as the Director of the Computer Department for Health Sciences, one of the last vestiges of the days of the mainframe computers, before personal computers. This did not offer any opportunities for teaching, so I created and taught noncredit courses on the now defunct computer language Pascal, and on how to use a computer controlled digitized video microscope in the Quantitative Morphology Unit, an effort that proved ahead of its time.

In 1984 my main appointment shifted to Botany, and then a decade later I became full-time Professor of Radiology. Along the way I had adjunct professorships in Electrical and Computer Engineering, Physics, Zoology, Biosystems Engineering, Obstetrics and Gynecology, and Computer Science. The University of Manitoba was a good place to work (retired end of 2011), where I had full academic freedom to pursue any and all kinds of research that I felt important, leading to my unusual career as a theoretical biologist in which I contributed to computed tomography algorithms [Gordon 2011c], diatom nanotechnology [Gordon 2010], HIV/AIDS prevention [Smith and Gordon 2009], and vertebrate embryogenesis [Beloussov and Gordon 2006]. Some of my appointments permitted me to supervise graduate students and to teach graduate courses in image processing plus my favorite, Reverse Engineering the Embryo [Gordon and Melvin 2003]. However, all of my adjunct positions ended after 26 years, perhaps due to tighter competition for grants (cf. [Gordon and Poulin 2009a; 2009b]). As Radiology was not part of the Faculty of Graduate Studies, I no longer had an opportunity to teach at that level. I was depressed over this matter until William R. Buckley encouraged me to start what we called the Embryo Physics Course in the virtual world Second Life® (http://secondlife.com), because of his interest in applying John von Neumann's ideas on self-reproducing automata [Buckley 2008; Buckley and Mukherjee 2005; von Neumann 1966] to real embryos.

The theme ‘embryo physics’ comes from my observation that the so-called genetic program of an embryo is a bifurcating alternation of changes in gene regulation and physics [Gordon 1999]. The central physics component of the genetic program appears to be easily observable [Gordon and Björklund 1996] waves of contraction or expansion of epithelial tissues [Brodland et al. 1994; Gordon 2001; 2011b]. The first found, and perhaps most important example is the splitting of the ectoderm in urodele amphibians into the neuroepithelium and epidermis by observable contraction [Brodland, et al. 1994] and expansion [Gordon et al. 1994] waves, respectively. These observations provide the physics of the classical neural induction problem [Spemann and Mangold 2001], for which Hans Spemann received the 1933 Nobel Prize (discussed in [Gordon 1999]).

Embryo Physics: the Course

We started the course in January, 2009, and except for occasional breaks of a few months it has run once a week since then. It has attracted 7 to 17 people each week, with people coming in and out according to their interests and time. At this writing we have scheduled 129 lectures (see Supplement).

The basic theme of the Embryo Physics Course has been the fundamental question: how does an embryo build itself? I regard this as one of the three big unsolved problems of biology, the other two being the origin of life [Damer et al. 2012] and the explanation of consciousness [Tuszynski 2006]. Anything that bears on this central question is fair game in the Embryo Physics Course, which is wide ranging in its scope. As explanations in embryogenesis sometime revert to the old question of Aristotle: “Which came first, the chicken or the egg?” [Wikipedia contributors 2011], we occasionally have talks on origin of life. Similarly, insofar as embryogenesis is the self-construction of the observer, we sometimes delve into philosophy of science and the nature and origin of consciousness. Indeed, one of my ‘morphogenesis heroes’, Edmund Sinnott [Sinnott 1960], saw morphogenesis as possibly the key to understanding consciousness [Sinnott 1950; 1962; 1966]. A reviewer of this Invited Review wrote:

“Embryology, consciousness, and life's origin are the three big questions precisely because they are systems-level problems demanding unfamiliar ways of thinking, so highlighting an interdisciplinary and radical approach to tackling one of these questions is appropriate for this journal…”.

The ramifications of a proper, working explanation of embryogenesis are many. First and foremost would be the lifting of the awe of the ‘miracle of birth’ to uncover a deep understanding of how it comes about. An analogy may be made with the religious concept of ‘the breath of life’ which has been completely replaced by the science of respiration [Sober 1993]. Are we better off for seeing through the magic? I would think so, since understanding respiration allows us to explore the depths of the oceans and outer space in person, and we are beginning to get a handle on respiratory diseases.

A real understanding of embryogenesis should have significant impact on the study and perhaps amelioration of birth defects and cancer. It should expedite research on stem cells [Gordon 2006], tissue engineering, cloning, the relationship between the genome and the phenotype [Gordon 2010], genetic engineering, and mechanisms of evolution [Gordon 1999]. The budding field of embryonics [Tempesti et al. 2007], the attempt to build and grow robots and computers using principles of embryogenesis, might succeed much better if we understood the latter [Gordon 2001; 2006]. The Embryo Physics Course has therefore attracted speakers in a wide variety of fields, showing the tremendously interdisciplinary nature of the problem:

• Anthropology (Terrence W. Deacon)

• Applied mathematics (José J. Muñoz)

• Artificial life (Bruce Damer, Steve Grand, Rudolf Nico Penninkhof, Andrew J. Wade)

• Bioengineering (Gerald H. Pollack, Glauco R. Souza, Michelangelo von Dassow, Vito Conte)

• Biophysics (Ille C. Gebeshuber, Timothy J. Newman)

• Biosemiotics (Alexei A. Sharov)

• Cancer biology (Vasily V. Ogryzko)

• Cellular automata (Genaro Juárez Martínez)

• Civil Engineering (G. Wayne Brodland)

• Computer science (Peter Newman)

• Embryology (Michael Danilchik, Antony J. Durston, Albert K. Harris, Ryan Kerney, Daniel Mietchen)

• Electrical engineering (Susan Crawford-Young)

• Evolution (Marta Linde)

• Finite element methods (Evgenii Rudnyi)

• History of biology (Ute Deichmann)

• Human genetics (Natalie K. Björklund-Gordon)

• Marine Biology (Shin Kubota)

• Mathematics (David M. Holloway)

• Molecular developmental biology (Ramray Bhat, Stephen A. Krawetz, Spyros Papageorgiou)

• Molecular biology (Eluem R. Blyden, Eva E. Deinum)

• Nanotechnology (Akhlesh Lakhtakia)

• Perinatology (Barbara Tzschentke)

• Philosophy of biology (Chris Chetland, Melinda Bonnie Fagan, Michael Ruse, Liz Stillwaggon Swan, Clément Vidal)

• Phycology (Lisa Willis)

• Physics (Steve P. McGrew, Vincent Fleury, M. Lisa Manning, Vasily Ogryzko, Alexis M. Pietak, Jack A. Tuszynski)

• Self-reproducing systems (William R. Buckley)

• Stem cells (Bradly Alicea, Lu Kai)

• Tensegrity structures (Gerald de Jong, Stephen M. Levin)

• Theoretical biology (Eva E. Deinum, Richard Gordon, Stuart A. Newman)

• Tissue engineering (Glauco R. Souza)

Names and talk titles by our speakers are available on course web pages maintained by volunteers William R. Buckley [Buckley and Gordon 2010] at first and now Evgenii Rudnyi [Rudnyi 2011] (see Supplement). At first I gave approximately every other lecture, and now only lecture on occasion. Our lecturers have spoken via their computers at home or work in: Argentina, Australia, Belgium, Canada, Denmark, France, Germany, Greece, India, Israel, Japan, Malaysia, Mexico, Netherlands, New Zealand, Singapore, Spain, United Kingdom, and the United States.

Meeting in a Virtual World

Meeting in the virtual world Second Life® is both an unusual and comforting experience. Each person is represented as an avatar that is controlled much like a puppet. However, unlike in a staged play, the personalities of the avatars tend to correspond to that of the people running them. We have therefore designated the (human, avatar) pair by the coined word ‘havatar’ [Gordon et al. 2009; cf. Friedman et al. 2009; Wieser et al. 2010; Yee et al. 2007]. The first book on avatars was written by one of our participants [Damer 1997].

Second Life® allows us to build our own world, and the landscape and buildings used for the Embryo Physics Course were designed and constructed by Natalie K. Björklund-Gordon, also a lecturer in the course. We tried a rooftop, open air forum for a while (Fig. 1), but found that having the avatars sitting in a semicircle in a natural setting was more comfortable (Fig. 2). Entering Second Life® is uncannily realistic, close to settling in and interacting with other people meeting in the same room. So in a sense we come together from all over the world to meet in one room, with no travel expenses or jet lag. We have considered commercial Internet meeting services, but they all cost from $20 to $278 per person per year, which might lead to near zero participation. Real costs, for uploading slides and rental of Second Life® virtual land, are under $30 per month, total. Second Life® is available on a number of platforms [Linden Research 2011]. But the Internet continues to improve, so the course may not continue in Second Life® forever.

Figure 1.  The original rooftop setting in Second Life® for the Embryo Physics Course.

Figure 2.  The present setting for the Embryo Physics Course in Second Life®.

One problem is that the real world is not flat, and so each person participates at a different time of their day. We meet at 2PM Pacific Time, which keeps our European participants up late at night and gets our Asian participants up early in the morning. Fortunately a good proportion of our participants are young graduate students, who are used to surviving on little sleep. We ask all of our participants to come to as many lectures as they can. Some of course just come once to say their piece, others have stuck with us from the beginning, and most drift in and out as their availability and interests dictate. Our cumulative mailing list is over 700 people, of which about 75 are on our active list in 2013.

One of the delights for me is that the Embryo Physics Course has spun off a number of collaborations that might not have occurred otherwise or were enhanced by mutual participation in the course [Damer et al. 2013; Damer et al. 2012; Fleury and Gordon 2012; Gebeshuber et al. 2012; Gebeshuber and Gordon 2011; Gordon et al. 2011, 2014; Igamberdiev et al. 2012; Lu et al. 2013; Lu et al. 2012; Nouri et al. 2012; Stillwaggon Swan et al. 2013; Tuszynski and Gordon 2012]. The Embryo Physics Course has been a forum to test ideas and get comments on a draft of a new book [Gordon and Gordon 2013]. This use of a virtual world for feedback as one writes has been touted as a new way of writing books [Anderson 2011].

There are a few technical problems that we have had to overcome or work around. Slides are uploaded as JPG files, and so animation in them is not possible. While movies can be shown within Second Life®, the computational burden is high, which can cause lag or ‘res’ (resolution) problems. We have instead asked speakers to create YouTube movies and paste their YouTube web page addresses into the Second Life® chat window. One click opens them in a separate browser. (The Second Life® Viewer is essentially a special purpose browser.) Furthermore, if two or more people have their computer microphones turned on simultaneously, screeching feedbacks or reverberations can occur. Thus during a talk the speaker's microphone is kept on and everyone else asks questions via text chat. On rare occasions the server for our region of this virtual world requires restarting, and once we had to postpone a lecture.

One drawback of Second Life® is lack of eye contact. This may improve when software becomes available for reading one's own expression from a web camera and transferring the gaze and expression to one's avatar [Murray et al. 2007; cf. Friedman et al. 2009; Schreer et al. 2008].

The number of people actually engaged in physics of whole embryos is still rather small. However, with the discovery of mechanical effects on gene expression [Adamo and Garcia-Cardena 2011; Avvisato et al. 2007; Ben-Ze'ev et al. 1980; Brouzés and Farge 2004; Chien et al. 1998; Dahl et al. 2008; Fu et al. 2011; Gieni and Hendzel 2008; Komuro and Yazaki 1993] slowly taking its place alongside a mostly symbolic and cartoon oriented molecular biology, we can expect ever more people will graduate from the cell culture to the whole embryo level of inquiry. The Embryo Physics Course may have a niche in accelerating this process, by bringing people together from across the world, simultaneously, in a single, albeit virtual space, where they can meet, talk, and plan collaborations.

One curious development is that many of the participants are young people, graduate and undergraduate students, who appear to be reaching beyond the limitations of their current milieu and thesis projects. As a consequence, the Editor of this journal and I are planning to offer the Embryo Physics Course for credit, which would hopefully be honored everywhere. Here is our course description:

The purpose of Readings and Lectures in Embryo Construction in Second Life® is to expose each participant to the emergence of a new field that brings together many disciplines that span from basic biology to the computational sciences and engineering. Readings and lectures will examine the current literature and research addressing the question of how an embryo builds itself. All model systems of development will be considered. Special emphasis will be placed on mechanics and other physical phenomena involved in the determination of cell fate, and their interactions with the molecular biology of development. This will include nuclear, cell, tissue, and whole organism structure and function as delineated from microscopy, molecular, and bioinformatic strategies. The course will review the current literature, with attention paid to hypothesis, experimental design, computer simulation, image data collection, and data interpretation. New readings will be assigned for each session and literature that is ‘just off the press’ will be included. Alternate sessions will be by students and guest speakers, all of whom participate by meeting at the same time and place in the virtual world Second Life®. Student sessions will consider one or two papers. One student will be assigned to present each paper. At the beginning of each session each student participant, with the exception of the student presenter, will be required to provide a one-page computer-written summary of each paper that is up for discussion that day. This will consist of a clear statement of the hypothesis tested, and a brief literature review as part of the synopsis of the work presented. When guest speakers present, each student will be expected to provide a one-page computer-written summary within the following two days. Questions during each presentation will be encouraged. A formal discussion will follow after each presentation, reflecting its contribution to the field while considering the logical next steps. In this course, critical literature review, presentation, and scientific writing skills will be emphasized. Students will be evaluated on presentations and weekly reviews in the following manner: presentation(s) 30%, participation 20%, weekly reviews 20%, and a maximum 1500 word (excluding references) final review paper 30%. The final review paper will be assigned during the second month and will be due the last day of the course. Enrollment is limited and is by permission of the lead instructor.

This plan may evolve with experience.

Acknowledgments

I would like to thank Ille C. Gebeshuber and Stephen A. Krawetz for their helpful comments, and Natalie K. Björklund-Gordon, William R. Buckley, Evgenii Rudnyi, and Lu Kai for being steadfast partners in pulling this off.

Declaration of interests: The author has no financial interest in Second Life®.

The bibliography for this paper includes references or a web page related to talks given or planned in the Embryo Physics Course as listed in the Supplement. Most of these are not cited in the body of this paper.