In this post, we will discuss three cases of how evolutionary processes (two explicitly intra-generational, one extensively inter-generational) affect the operation of physiological systems. In the case of our inter-generational examples (I and III), physiological processes exhibit their own adaptive dynamics. In the case of our inter-generational example, the macro-evolutionary distribution of gene variants provides a basis for intra-generational adaptation.
I. Role of Intra-generational Selection in tRNA Availability and Translation
Mahlab, S. and Linial, M. Speed Controls in Translating Secretory Proteins in Eukaryotes: an Evolutionary Perspective. PLoS Computational Biology, 10(1), e1003294 (2014).
This paper deals with transcription and the production of secretory molecules (production of the secretome) as an intra-generational evolutionary process. The secretome involves a vast array of peptides produced via a special ribosomal structure located at the cell membrane (Figure 1). The resulting peptides (manufactured on the outside of this membrane) are then involved in cell-cell communication.
Figure 1. The specialized translational pathway that leads to the production of signaling peptides.
The authors focus on role of tRNA (or transfer RNA) adaptation and the N' terminal of secretory proteins. tRNAs are internal to the translational process, and serve to translate open reading frames of DNA into amino acids. As adaptors, tRNAs are specialized by codon and function. Each type of specialization results in a population (or pool) which has a certain degree of diversity. This results in something called codon usage bias, a concept to which we will return. The N' terminal regions of the signaling peptide are associated with the so-called "fast" tRNAs. Secretory proteins made in this fashion are also found to contain segmental information that allow for various signaling functions. The signaling functionality is in turn a product of adaptation by natural selection within a single human generation (and perhaps even within a single cellular generation). "Fast" tRNAs are just one type of specialized tRNA molecule that exist in different proportions depending on various factors. The consequences of selection on these ratios is to affect the production of some codons (and thus peptides) over others.
Changes in the proportion of tRNA types requires an adaptive mechanism. While the specifics of this mechanism are unknown (but see Figure 2), the amount of diversity is governed by the number of tRNA molecules of a certain specialized type. To illustrate this, I use a conceptual model of translation called the "Hungry, Hungry Hippos" model, named after the popular children's board game (see Figure 3). The game begins with four hippos and a game board full of freely-moving marbles. Then, each hippo eats as many marbles as they can before the marbles are all eaten. This race exemplifies how tRNAs are utilized in the process of translation: mRNA is moved through the ribosome at different speeds, which tRNA molecules compete to bind to the incoming sequence and replicate their information in a new generation of peptides.
Figure 2. The site of translation and the trade-offs inherent in tRNA selection. COURTESY: Figure 1 (a and b) from [2].
Figure 3. The game "Hungry, Hungry Hippos", a model for transcription?
While one might think of this as a stochastic process, tRNA pools adapt to the needs of a given cell, including the speed of translation and amino acid bias. One measure of how these pools evolve is the tRNA adaptation index [1], which is based on the concept of codon usage bias (Figure 4).
Figure 4. the tRNA (codon) adaptation index, a intra-generational natural selection index. CAI is a weighted geometric mean for all categories of codon.
The premise of tRNA adaptation and the potential role of natural selection is that gene expression is correlated with codon bias [3]. Depending on the needs of the cell, the production of proteins can be biased towards certain amino acids through controlling both the speed of translation and the (perhaps more importantly) the composition of tRNA pools. The consequences of this codon bias can be observed when plotted against gene expression (e.g. the production of mRNAs -- see Figure 5). In general, when gene expression (or transcriptional noise) is more active, the greater the bias in codon-specific tRNA activity (in the form of codon frequency).
Figure 5. Changes in codon frequency with respect to gene expression. Figure 2 from [4].
II. Inter-generational Selection for Antigens
Forni, D., Cagliani, R., Tresoldi, C., Pozzoli, U., De Gioia, L., Filippi, G., Riva, S., Menozzi, G., Colleoni, M., Biasin, M., Lo Caputo, S., Mazzotta, F., Comi, G.P., Bresolin, N., Clerici, M., and Sironi, M. An Evolutionary Analysis of Antigen Processing and Presentation across Different Timescales Reveals Pervasive Selection. PLoS Genetics, 10(3), e1004189 (2014).
The human immune system is a complex system that consists of recognition and defense mechanisms (Figure 6). These mechanisms operate both intracellularly and extracellularly. In addition (see Figure 7), there is both an innate system (which is evolutionarily conserved) and an adaptive system (which is derived but shared among vertebrates). Given this complexity, it is often hard to find the inter-generational underpinnings of intra-generational adaptation. One form of intra-generational adaptation in the immune system involves antigen processing and presentation. This is determined by both the inter-generational evolutionary history of antigen-specific genes and the role of selection within and between generations.
Figure 6. A quick refresher on the human immune system architecture. COURTESY: [5].
In this study, the authors examined the evolutionary history of 45 antigen-specific genes in Homo sapiens. In doing so, they and looked at both the intra-specific variation and inter-specific diversity of genes related to antigen-related processes. This study also used a comparative genomic approach to better understand the evolutionary history of antigen-specific genes in humans. This was done in two different ways. The first was to use several different statistical tests to identify the target of selection. Then, the targets were characterized using low-coverage, whole-genome Sanger sequencing (e.g. high-throughput analysis using next-gen sequencing). In the end, it was found that 9 genes in the antigen processing and presentation (APP) pathway have undergone adaptation within Homo sapiens. Taken collectively, this study gives us a structural view of diversity in the immune system that may predict variation in immune-related physiological responses.
Figure 7. Evolution of adaptive immunity in the Tree of Life. COURTESY: [6]
III. Intra-generational Selection in Tumor Survival (Cancer Evolution)
Ostrow, S.L., Barshir, R., DeGregori, J., Yeger-Lotem, E., and Hershberg, R. Cancer Evolution Is Associated with Pervasive Positive Selection on Globally Expressed Genes. PLoS Genetics, 10(3), e1004239 (2014).
Much like Evolutionary Psychology, evolutionary views of cancer has become increasingly popular as conceptual models. Unlike Evolutionary Psychology, however, evolutionary views of cancer are not based on attempts to broadly characterize human behavior. The evolutionary view of cancer is similar to the population dynamics of organismal evolution by natural selection. Except that in this case, population processes are intra-generational and occur within specific tissues. What makes them "evolutionary"? For one, cancer can be characterized as a genealogical (branching) process, with many cancer cells originating from a single deleterious mutant (Figure 8).
Figure 8. Oncogenesis as a branching bush (evolution from common descent). COURTESY: [8].
In fact, one could think of evolutionary models of cancer as an instance of evolutionary dynamics rather than the outcome of reproductive fitness. Nevertheless, the usual suspects still participate in the process. For example, genetic variation in the form of standing variation or somatic mutations is selected upon through the process of tumorigenesis [7]. Mutations that are robust to positive selection contribute to the proliferation and microenvironmental maintenance of tumors (Figure 9). This is distinct from the natural selection that acts on germ line cells. Nevertheless, reproductive fitness is still the criterion for selection.
Figure 9. LEFT: Schematic showing the role of selection on cell populations and their microenvironmental ecosystem. RIGHT: comparison of clonal populations and their evolution with organismal species and their evolution. COURTESY: [9].
One important but often overlooked aspect of treating cancer as an intra-generational evolutionary process is that the constituent cells of a tumor can be viewed as replicators. Figure 10 demonstrates how lineages bud from single mitotically-dividing cells given various environmental and microenvironmental triggers. Yet single- cell replicators are also theoretical units upon which selection acts. In the case of Eukaryotic somatic and stem cells, variants can compete to determine the intensity or metastatic ability or a given type of cancer. These replicators also operate in an environmental context that often acts as a source of selection.
Figure 10. Evolutionary process in a single body (e.g. intra-generational cell population). COURTESY: Figure 2 in [10].
Despite some conceptual difficulties, these three studies give us a window into intra-generational adaptive and evolutionary processes. Far from being a black box, these processes are often distinct from but are influenced by inter-generational evolution. While these studies ignore the role of currently hyped adaptive mechanisms such as epigenetics and the microbiome, there is a lesson for interpreting the true contribution of these types of mechanisms on the long-term evolutionary process.
NOTES:
[1] dos Reis, M., Savva, R., and Wernisch, L. Solving the riddle of codon usage preferences: a test for translational selection. Nucleic Acids Research, 32(17), 5036–5044 (2004).
[2] Pechmann, S. and Frydman, J. Evolutionary conservation of codon optimality reveals hidden signatures of co-translational folding. Nature Structural and Molecular Biology, 20, 237–243 (2013).
[3] Neame, E. Structure vs. Codon Bias. Nature Reviews Microbiology, 7, 406 (2009).
[4] Shah, P. and Gilchrist, M.A. Explaining complex codon usage patterns with selection for translational efficiency, mutation bias, and genetic drift. PNAS, 108(25), 10231-10236 (2011).
[5] The Human Immune System. The Molecules of HIV website (2006).
[6] Danilova, N. Evolution of the Immune System. MIT OpenCourseWare, Spring (2005).
[7] Anderson, A.R.A., Weaver, A.M., Cummings, P.T., and Quaranta, V. Tumor Morphology and Phenotypic Evolution Driven by Selective Pressure from the Microenvironment. Cell, 127, 905-915 (2006) AND Magiliocco, A.M. Tumor Heterogeneity in Breast Cancer, Concepts, and Tools. Figshare.
[8] Looi, M-K. Cancer, genomes, evolution, and personalized medicine - it's complicated. Wellcome Trust blog, March 7 (2012).
[9] Greaves, M. and Maley, C.C. Clonal Evolution and Cancer. Nature, 481, 306-313 (2012).
[10] Yates, L.R. and Campbell, P.J. Evolution of the Cancer Genome. Nature Review Genetics, 13, 795-806 (2012).
