Vertically- and horizontally-transmitted memories - the fading boundaries between regeneration and inheritance in planaria

The main open questions concern what changes (transcriptional, chromatin-level, or bioelectrical) distinguish a trunk fragment from a wild-type worm (destined to make one head) and an anatomically-normal trunk fragment from a two-headed worm (which will make two heads). Examination of bioelectric state (using fluorescent reporters of voltage distributions) (Adams and Levin, 2012; Oviedo et al., 2008), transcriptional profiling, and chromatin state analysis must be used to understand what is different about these fragments. Quantitative models must be developed to explain how stable states can be stored, and edited, in physiological circuits (Cervera et al., 2015; Law and Levin, 2015; Levin, 2014a).

To determine how and where memory may be stored outside the brain during head regeneration, it would be necessary to first optimize training protocols (Abbott and Wong, 2008; Blackiston et al., 2010; Inoue et al., 2015; Nicolas et al., 2008; Pagán et al., 2012), capitalizing on more ecologically-salient stimuli and learning paradigms to achieve high-throughput induction of robust learning. The key experiments would be to assess the persistence of memory in fragments of different sizes, anatomical locations and body compositions. A variety of molecular and biophysical tools now exist to establish suppression screens targeting various pathways, to begin to probe the mechanisms necessary for imprinting of the memory upon a newly-regenerating brain (Aoki et al., 2009; Gentile et al., 2011; Sheı˘man and Kreshchenko, 2015).

In planaria, and other organisms that reproduce by fission, producing and maintaining variation between fragments after asymmetric division may be adaptive (much like the beneficial increase in variation following sexual reproduction and recombination). Therefore, the theoretical ability of asymmetric division to create variability in an otherwise isogenic population could be considered as a tool for producing evolutionary progress. Thus, asymmetric fission is a mechanism that challenges our current view of what defines the temporal axis of evolution, since epigenetic processes, environmental cues, biochemical gradients and generation of a complete individual from a community of cells can generate natural variation, without requiring so called ‘distinct’ generations. It is likely that we have only begun to glimpse the prevalence and variety of long-term memory in somatic tissues during lifespan and across reproduction throughout phyla. The continued future analysis of such instructive interactions is likely to have profound implications for understanding evolution. Moreover, a mature understanding of these fascinating processes will drive numerous applications in regenerative medicine and bioengineering that exploit the rich informational plasticity of tissues for the rational control of form and function.

We are grateful to Fallon Durant, Maya Emmons-Bell, and Jennifer Hammelman for helpful comments on the draft. We especially thank Keith Harris for his great contribution to the writing of the paper and for helping with the development of the ideas that constitute this work. We thank all members of the Rechavi and Levin lab for fruitful discussions and advice. We are grateful to Eva Jablonka, Yehu Moran, and Nir Ohad for reading the manuscript and for their helpful comments.