A principle of organization which facilitates broad Lamarckian-like adaptations by improvisation
As a fundamental principle of organization, adaptive improvisation by drive-reduction may be implemented in complex systems beyond life sciences (e.g. in physics, engineering and economy), thus improving the ability of these systems to overcome arbitrary types of novel malfunctions.
The article by Soen, Knafo and Elgart discusses “adaptive improvisation” which the authors portray as a novel principle of evolution driven by unfamiliar (i.e. not encountered previously) forms of stress. In my view the paper does not quite live up to such dramatic claims. The term of adaptive improvisation is new but the specific examples of such “exploratory response” presented by the authors, namely stress-induced mobilization of MGE, DNA transfer by exosome, and changes in the microbiome, are well known. I do agree that these mechanisms (and other similar ones, e.g. stress-induced mutagenesis) can be legitimately described as exploratory response and do represent a distinct evolutionary modality (in Ref. 59, Yuri Wolf and I defined this class of phenomena as quasi-Lamarckian). If this is all that is meant, the proposition is not novel. If there something more to “adaptive improvisation”, the present text is too vague to discern these potential new aspects. To my regret, the proposed tests of the hypothesis are also presented vaguely, and at least from the current version of the paper, I could not figure out what if any specific experiments or analyses are proposed. As it stands, the article is, in my opinion, not a hypothesis but an essay but even in that capacity, is unnecessarily opaque and not sufficiently specific.
I cannot rule out that I missed the more novel and concrete aspects of the hypothesis. If that is the case, however, I think the authors should make an effort to present their ideas simply and clearly.
Quality of written English: Needs some language corrections before being published.
Authors’ response: We revised the manuscript (and its title) to provide a clearer description of the hypothesis. Below we emphasize the novel elements of this hypothesis. To avoid potential confusion, it is important to keep in mind that stress-mediated induction of variation (heritable or not) is not the subject of this hypothesis. The provided mechanistic examples (e.g. MGEs, exosomes and symbiotic microorganisms) are also not the subject. We include these mechanisms only as a way of demonstrating the feasibility and generality of the hypothesis and to provide clear (though non-exclusive) examples for the kind of variations which satisfy its underlying assumptions. The hypothesis is first and foremost about a fundamental principle of organization and its ability to address an adaptation problem that has not been previously discussed, namely: Howevery individualcan acquirediverse physiologicaltolerance tohigh loadof novel combinations ofinternalperturbations that areindividual-specific(distinguishing features of this problem are highlighted in boldface).
While addressing this new problem, the hypothesis indeed use key elements that were previously suggested to contribute to Lamarckian evolution (Koonin & Wolf, [59]). It is therefore important to explain the distinguishing features of the current hypothesis. The aforementioned previous publication discussed a framework of co-existing Lamarckian-like and Darwinian mechanisms. It suggested that Lamarckian-like acquisition of heritable adaptation to novel environments can be achieved by stress-induced mutagenesis. It also proposed that the deleterious potential of random mutagenesis is addressed by regulation which involves DNA repair and particular structure of genome architecture (in prokaryotes). The latter are thought to bias mutations in a way which allows coordinated evolvability of functionally linked genes in rare cells where beneficial mutations emerge. Although we agree with this view, it does not address the conceptual problem which we propose to solve, namely: how any regulation could use mostly non-beneficial variation to provide every individual with a beneficial response to arbitrary types of novel conditions. This becomes far more challenging when each multicellular individual has to cope with many (individual-specific) combinations of internal perturbations (or ‘failures’) in different cells, tissues and organs. As we explain in this paper, every individual encounters many of these failures during its lifetime even under regular environments. Thus, the difficulty is not in explaining how variation can be induced or inherited. The problem is to explain how regulation within an individual can organize variation into emergent adaptations to novel scenarios of diverse kinds.
A solution which is based on random variation must also explain how deleterious effects of non-beneficial variations are avoided and how emergent adaptations in different cells, tissues and organs are coordinated into a coherent benefit for the whole organism. Without such ability to organize variation in a beneficial way, the benefit of inheriting this variation is also questionable. While genome architecture and DNA repair likely participate in the solution (as are other mechanisms), the conceptual problem remains unanswered. Moreover, substantial part of the variation which can participate in emergent physiological adaptations is not at all genetic. Since our hypothesis focuses on providing a conceptual solution to this yet unaddressed organizational/regulation problem, its core message does not overlap with prior suggestions.
The proposed solution (Adaptive Improvisation) is achieved by incorporating a principle of organization which is not part of current evolutionary thinking and is not exclusively dependent on any specific mechanism. It explains how large volume of random variation of any kind (including, but not limited to genetic variation) and every scale (from intracellular to the whole organism) can be safely organized de novo into many beneficial physiological adaptations during development and homeostasis of every single individual. This should not be confused with previously reported examples of Lamarckian adaptations by pre-evolved mechanisms which address a specific type of novelty. Hallmark examples are the CRISPR system in prokaryotes [69–71] and the small RNA-based mechanism of viral immunity in C. elegans [72]. Both mechanisms enable rapid acquisition of heritable resistance to new viruses, but the tolerance is always acquired using the same mechanistic pathway. While the preexistence of such pathway is very instrumental for coping with new viruses, it does not provide solutions for all other types of novel conditions (including those which can arise as outcomes of new pathogenic viruses that have managed to evade rapid elimination). This highlights the need for an adaptive approach which can use arbitrary mechanisms in conjunction with any kind of new variation at any scale.
Another major and critical distinction with respect to prior discussions is the perception of stress. Specifically, the stress in this hypothesis is not necessarily regarded as the activation of a defined stress pathway. Rather, it is viewed as deviations from steady states (or disruption of ongoing processes) which are subsequently capable of driving more deviations (with or without activation of specific stress pathways). In this new view, stress is a collection of driving forces which operate on diverse processes across multiple scales (from the scale of intracellular processes, through tissues and organs, and up to the whole organism level and above). Regarding stress as a mere driver of variation (analogously to physical stress) is required to: 1) account for a tendency to re-achieve stable states and processes of any kind (drive reduction), and 2) enable coordination between adaptive changes which take place across different scales of a multi-cellular organism. It is important to note that de novo emergence of adaptations in cells, tissues and organs are not enough for coping with diverse internal failures. Such coping also requires that the coordination between adaptive outcomes in different scales will also emerge de novo. This cannot be achieved by pre-specified mechanisms of coordination. Yet, by regarding stress as a driver of variation in every scale, coordination can arise as an outcome of stress-reductions which occur in parallel over multiple scales.
The necessity of ‘real time-like’adaptive capabilities beyond natural selection leads us to propose a new picture of gene regulation as (an inseparable) synergy between outcomes of natural selection and adaptation by improvisation. To prevent confusion, we do not just mean to suggest that adaptive improvisation is merely assisting natural selection. Rather, we argue that the dependence of every individual on its ability to address high load of novel internal failures, means that natural selection and adaptive improvisation are inter-dependent. Simply put, one is insufficient without the other. Similarly to natural selection, adaptation by improvisation can involve any kind of mechanism. To further establish this view, the hypothesis explains how the workload can be divided between pre-evolved responses and adaptive improvisation without any need for a ‘decision-making’ process. Here again, stress is the key factor; the less it is reduced by outcomes of natural selection, the more it drives improvisation-based processes. In this new picture, stress is a major physiological organizer of variation, whereas fitness is the combined outcome of natural selection and adaptive improvisation. Note that the stress in this picture cannot be reduced to zero, because it is constantly influenced by two opposing factors: it is increased by the 'entropic' effect of the higher abundance of non-beneficial (vs. beneficial) variations and it is reduced by (less frequent) beneficial variations.
As a fundamental principle, this hypothesis has broad implications which can be translated into many predictions (and applications). However, unlike hypotheses which are based on one or a few deterministic mechanisms, our proposal is inherently based on largely random organizations which involve a wide range of processes and mechanisms. This limits the ability of verifying or falsifying the hypothesis using a single experiment. Of course, some evaluations are potentially more informative, especially those which are designed to test if coping with similar conditions of novel stress (e.g. in replicated experiments) can result in substantially different adaptive outcomes. Notwithstanding, we believe that sufficient level of confidence would ultimately require different types of analyses, each focusing on one or few predictions of the kind provided in thismanuscript.
For completeness, the paper also discusses the potential relevance of adaptive improvisation beyond life-sciences (e.g. in man-made systems).
I appreciate the extended response given by the authors to my comments. I must admit that the level of abstraction in this discussion remains largely above my head. Hopefully, other, theoretically better versed readers receive inspiration from this treatise.
Soen, Knafo and Elgart touch upon an important topic of the nature of adaptation to novel challenges. To me, however, this discussion seems to be a bit too vague and general.
Authors’ response: As indicated above, we revised the manuscript (and its title) to provide a clearer description.
The authors improved the presentation, but the main problem remains - in my opinion it is impossible to criticize or embrace ideas, formulated in such a general terms. Even the very existence of a problem in need of solution is not entirely obvious - yes, none of the generic plasticity mechanisms can fully eliminate all the problems (p.4), but at the same time not all and every individual of the population must survive for the population to survive. Without a careful quantitative analysis it is difficult to understand which of the proposed novel entities (principle of Adaptive Improvisation, self-stabilization of stress-reduction, etc.) are driven by necessity.
We would like to acknowledge the inspiring and pioneering work of Prof. Erez Braun (Technion, Israel) and Prof. Shimon Marom (Technion, Israel), and the insightful research of Dr. Shay Stern and members of our lab. These studies were instrumental for developing a perspective which motivated and helped shaping these ideas. We thank Prof. Erez Braun (Technion, Israel), Prof. Naama Brenner (Technion, Israel), Prof. Eva Jablonka (Tel-Aviv University) and Prof. Shimon Marom (Technion, Israel) for ongoing discussions of this subject and related ideas. We would also like to thank Prof. Michael Walker (Weizmann Institute of Science, Israel), Prof. Yitzhak Rabin (Bar-Ilan Univ.), Prof. Denis Noble (University of Oxford, UK), Prof. Zvi Livneh (Weizmann Institute of Science, Israel), Prof. Eytan Domany (Weizmann Institute of Science, Israel), Prof. Nir Friedman (Weizmann Institute of Science, Israel) and Dr. Yael Heifetz (Hebrew University, Israel) for helpful discussions. We thank Prof. Moshe Szyf (McGill University, Canada) for helpful discussion of potential implications to cancer treatment. We also thank Profs. Michael Walker, Eva Jablonka, Denis Noble, Yosef Shaul, Eytan Domany and Dr. Idan Efroni (NYU, USA) for comments on the manuscript, and we thank Shira Holand for contribution to the preparation of the graphic presentations.
This work was supported by the Sir John Templeton Foundation (grant ID: #40663) and the ISRAEL SCIENCE FOUNDATION (grant No.1860/13). Yoav Soen is Incumbent of the Daniel E. Koshland Sr. Career Development Chair at the Weizmann Institute.