Sentinel plants as programmable processing units: insights from a multidisciplinary perspective about stress memory and plant signaling and their relevance at community level

ABSTRACT

Stress memory and an effective signaling among individuals in a given community are recognized to improve plant performance under recurrent stressful conditions. As living beings with memory and signaling abilities, plants can be considered as processing units and then be trained – or programmable from a computational viewpoint – and prepared for facing biotic and abiotic stresses. Here, we propose that sentinel plants could improve the resilience of agricultural and natural communities by reducing the impact of biotic or abiotic stressors on their neighbors. Modeling plants as programmable (or trainable) processing units compels us to think about a multidisciplinary perspective for integrating stress memory, signaling, and resilience of biological systems into executable programs, fostering the creation of applications and technologies that would benefit from the spatiotemporal dynamics related to plant-plant and plant-environment interactions.

Have you already imagined plants taking care of each other in a community? Can plants be trained to face environmental challenges? It is time to deal with plants from a non-naïve perspective. Such organisms are able to learn with stressful events and improve their performance when facing a new constraining condition.^1–3 Regardless the underlying processes leading to stress memory – epigenetic changes, presence of transcription factors or both – plants can acquire stress imprints and then pass through limiting conditions without significant impairment of plant growth and development.^1,2 Such phenomenon is so important in nature that antecedent events have been included in plant and ecosystem models and some species can even present transgenerational memory,^3,4 increasing the ability of the next generation in dealing with environmental constrains.

On one hand we have the plant ability in learning from past events while on the other hand we have plant signaling, a step before stress memory. One facet of signaling is related to plant signals within an organism, as happens when roots produce and send chemical substances to leaves for avoiding excessive water loss under low water availability.⁵ Another facet is the signaling among plants within a community. Such communication among plants may occur through volatile compounds (air signaling – AS, in Figure 1) produced by plants under biotic stress or through substances released in rhizosphere (soil signaling – SS, in Fig. 1).^6–8 Then, an effective communication among plants does not necessarily involve plant-to-plant contact and it occurs in an invisible and hidden way. For instance, plants facing biotic or abiotic stresses at the edge of a forest fragment or in a crop field (Figure 1, top) would warn neighboring plants about insect presence or reduction of resources such as water. Such sentinel plants – those able to respond rapidly against a stressor and inform efficiently other ones about the threat – would induce responses in other plants to alleviate or even avoid severe damage caused by environmental limitations such as drought (e.g., reducing water consumption) or biotic pressure (e.g., increasing the production of secondary metabolites involved in plant defense).

Figure 1.

Biotic and abiotic interactions in plant systems: crop systems have low diversity and are homogenous (top left), whereas natural systems have high diversity and are heterogeneous (top right). Air (AS) and soil (SS) signaling refer to water deficit and herbivory threats. Computer networks representing both systems and plant-plant interactions are shown at bottom. Computers/plants could exchange information by physical contact (wired) or through aerial/soil chemical signals (wireless). Effective signaling would reduce herbivory damage and water consumption (transpiration) under decreasing water availability.

Recently, Elhakeem and colleagues⁹ found that aboveground stimuli modify belowground interactions between neighbors, expanding our vision about biotic and abiotic interactions while revealing our limited understanding about how plants interact with each other in a community. As such complexity involved in plant signaling either at organism or community level has been recognized, the study of biological systems, which are complex by nature, requires non-conventional and dynamic tools. From a computational point of view, plants would be considered as processing units and the sentinel plants would be powerful processing units able to maintain the overall system resilience,¹⁰ avoiding its collapse. A typical computer central processing unit, or processor, is an engine in charge of interpreting and executing instructions stored in main memory.¹¹ A set of instructions defines possible steps to be performed and a processor operates according to an instruction execution model, defined by the computer architecture. ¹¹

In fact, every plant works as a processing unit that interprets (or executes) instructions represented as stress imprints stored in its memory, producing outputs. Under this perspective, instructions are sequence of steps, each one associated with a particular biotic and abiotic pressure. A set of instructions comprises a kind of programming language, which would be used for coding plant functioning. From this perspective, plants from different species are processing units with different architectures, each one associated with a possibly different instruction execution model. Such analogy between plants and computers was proposed initially by Adamatzky and colleagues,¹² suggesting that plant-based computers would not only open new opportunities for multidisciplinary research, but also promote breakthroughs and advances in computer science and plant biology. Starting from the same perspective and analogy, we argue that resilience of plant communities could be improved by training (for biologists) or programming (for computer scientists) plants for dealing with environmental challenges, i.e., our sentinel plants.

If plants memorize past stressful events, in theory, they could be trained – a biological expression for plant programming – for reaching the status of sentinels and protecting the community. At this organizational level, a group of plants working as autonomous processing units that exchange information forms a network (Figure 1, bottom). By understanding how plants acquire memory and which indices are useful to evaluate their biological state, plant functioning programming turns an interesting alternative for complex tasks such as reforestation or improving crop yield. In both cases, young plants would be initially grown under nursery conditions where they can acquire stressful imprints through training. For instance, three cycles of dehydration/rehydration turned the photosynthetic apparatus of sugarcane plants resistant to drought and improved water use efficiency.¹³ While there is no warrant that imprints will improve field performance, experimental data testing this hypothesis are still lacking. Some methods and indices for detecting early signals related to critical transitions in plants were proposed recently.^14–16 By applying metric- or model-based indicators, we might evaluate if stressed plants crossed the tipping point and reached an alternative status, indicating stress memory.^14–16 Then, conceptual frameworks for plant resilience would be useful for evaluating the significance of such stress memory for plant growth and development. After recurrent stresses, reductions in perturbation, impact, and recovery time and increase in recovery rate would indicate positive changes leading to high resilience.¹⁷

Considering that we have (i) indices for evaluating stress memory and (ii) technology for non-destructive and high frequency sampling of plant signaling in a network, we can now train plants – our programmable processing units – and evaluate their importance at community level. Not only are designing appropriate processing unit models and architectures for each plant species mandatory, but also developing methods for quantitatively assessing their efficiency is of paramount importance.¹⁰ Plant programming itself might not be a simple task as well. While devising appropriate languages and tools to support plant coding is still necessary, plant programmers need to develop or improve their computational thinking skills.¹⁷ Another challenge concerns the study of communication network models for properly representing and assessing biotic and abiotic interactions, with the goal of ensuring reliability and then revealing how plant communities avoid failures or reduce their vulnerability to stressors.¹⁸

With relevant and insightful information integrating stress memory, signaling, and resilience of biological systems from a new and multidisciplinary perspective, we would burst the current knowledge about systems dynamics and improve our limited spatial and temporal concepts about the interaction among plants and between plants and their surrounding environment. From a practical perspective, the sensitivity of biological systems to biotic and abiotic stresses would be improved by sentinel plants (trained plants) if these processing units are able to ring the bell and alarm other individuals about a potential threat. An effective signaling among species is likely an important functional component in natural and agricultural systems,¹⁹ and we must be able to understand this talk in a changing world. Whether this talk is effective and can occur among plants of a given species in crop systems or among plant species in a natural ecosystem remains to be revealed.

Funding Statement

This work was supported by the Coordination for Improvement of Higher Level Personnel [88881.145912/2017-01]; the São Paulo Research Foundation (FAPESP, Brazil) [Grant numbers 2015/24494-8 and 2017/12302-2].

Acknowledgments

RVR and RST acknowledge the financial support granted by the São Paulo Research Foundation (FAPESP, Brazil; Grants n. #2015/24494-8 and #2017/12302-2) and the Coordination for Improvement of Higher Level Personnel (CAPES, Brazil; Grant n. #88881.145912/2017-01).

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.