On the persistence of memory

Abstract

The circadian clock is considered a central “orchestrator” of gene expression and metabolism. Concomitantly, the circadian clock is considered of negligible influence in the field and beyond leaf levels, where direct physiological responses to environmental cues are considered the main drivers of diel fluctuations. I propose to bridge the gap across scales by examining current evidence on whether circadian rhythmicity in gas exchange is relevant for field settings and at the ecosystem scale. Nocturnal stomatal conductance and water fluxes appear to be influenced by a “hard” clock that may override the direct physiological responses to the environment. Tests on potential clock controls over photosynthetic carbon assimilation and daytime transpiration are scant yet, if present, could have a large impact on our current understanding and modeling of the exchanges of carbon dioxide and water between terrestrial ecosystems and the atmosphere.

Keywords: :

• biosphere-atmosphere interactions
• circadian clock
• diel cycles
• gas exchange
• endogenous regulation
• evapotranspiration
• gross ecosystem exchange
• hysteresis
• up-scaling

This article refers to:

“We propose that if we are to reliably scale measurements up from the leaf level, we must be able to explain the basic patterns of ecosystem CO₂ exchange according to our present understanding of the biochemistry of leaf gas exchange.” – Hollinger et al. 1994,¹ a proposition that has been extensively tested and validated over the past 19 y, and that also underpins current mathematical modeling of gas exchange.² Similar analogous premises underlie the scaling and modeling of transpiration.³ Yet, there is a fundamental physiological process that regulates leaf level fluxes for which we do not know whether or not it exerts a notable influence in the field and, by extension, whether or not it scales up to affect the exchanges of water and carbon between terrestrial ecosystems and the atmosphere: the circadian clock.⁴

Hennessey et al.⁵ observed a 20% oscillation in leaf photosynthesis in Phaseouls vulgaris in the free-running. If the same effect occurs at the ecosystem level, we could expect that the diel oscillation in Gross Ecosystem Exchange (GEE) under no variation in the physical environment (of light, temperature, etc.) would still be a fifth of that observed during a normally oscillating environment. This proportion would likely be higher for water loss because the relative oscillation in g_s is usually higher than in leaf photosynthesis in the free-running.⁵

The key question to understand whether circadian rhythmicity is a significant driver of the daily pattern of GEE and ET (evapotranspiration) is whether it acts as a “soft” clock (its action is overridden by environmental cues) or as a “hard” clock (it remains a driver of importance similar to that of environmental cues). The circadian clock may impact ecosystem fluxes by at least 2 different ways. First, by setting a time-dependent potential value. That is, GEE (or ET) at noon will necessarily be higher than in the afternoon, when “everything else” remains constant, if the clock is entrained to increase GEE (or ET) at noon.⁶ Suboptimal or limiting environmental conditions of temperature, vapor pressure deficit, and the like then reduce this time varying potential to its actual value. Second, the circadian clock may affect the sensitivity of gas exchange to environmental drivers in time. That is, the amplitude in the response to a change in light or temperature, to name a couple of environmental cues, could be different at mid-morning than at noon if the clock affects the responsiveness of gas exchange with time.

Field studies have traditionally assumed that the control of the clock over leaf-to-ecosystem scales is soft, and of negligible importance relative to that of environmental cues. Recent evidence suggests otherwise. Nocturnal water loss often contributes around 10% of the total transpired water (although it can exceed 25% of the total water loss in some desert plants⁷). The temporal pattern of nocturnal stomatal conductance (g_s) under non-limiting water supply is often characterized by an initial decline during the first hours after dusk, and a posterior increase that peaks around dawn. This nocturnal increase in g_s may be attributed to the circadian clock when variation in the environmental drivers of g_s is negligible (a “constant environment”).⁸ However, with the exception of some overcast nights, temperature tends to decrease while relative humidity tends to increase during the night, which leads to a decrease in vapor pressure deficit (VPD). In those species where nocturnal g_s shows a negative response to VPD,⁹ the nocturnal increase in g_s may be attributed to a combination of endogenous controls and of direct physiological responses to the environment. However, nocturnal g_s in many species has been found to respond positively to VPD under a changing environment.^10-12 A nocturnal increase in g_s in species that respond positively to VPD can only be explained by the action of a hard clock, that overrides the effect of environmental variation.

Unraveling the effect of the clock on daytime processes is more complicated than on nocturnal processes, because of the larger number of environmental cues present. However, current evidence points toward the circadian clock being also a general driver of daytime carbon and water fluxes. The diurnal pattern in GEE or ET is often asymmetrical and, “everything else” being equal, significant differences in fluxes arise depending on time of day. This hysteresis has traditionally been attributed to a combination of hydraulic feedbacks after decreased leaf water availability as the day advances, decreased source demand, and increased photorespiration, among others.^13-15 Although scarce, current evidence points toward circadian regulation of g_s as an additional driver of the diurnal hysteresis on water and carbon fluxes.¹⁶ For instance, Mencuccini et al.¹⁷ observed that stomatal opening after root pressurization was lower in the afternoon than in the morning, which could not be attributed to changes in abscisic acid. Indeed, after employing a series of filters to remove environmental variation, Resco de Dios et al.⁶ observed a diel pattern in Net Ecosystem Exchange resembling that under a changing environment.

Even if current evidence may be pointing toward a hard clock action, lack of data still impedes any generalization. In fact, we do not even know how widespread circadian regulation of gas exchange may be across species. However, the potential risk of ignoring whether such a “memory” of the recent past is persistent and leaves an imprint on GEE and ET seems too high under current efforts toward quantifying and predicting the responses of the carbon and water cycles to environmental changes. Because response curves are logistically complex at the ecosystem scale, models are parameterized from the relationship between GEE or ET and the parameter of interest (light, temperature, etc.). These relationships will necessarily be confounded by the clock, because of the uneven distribution of light and temperature over a day. Low temperature and light levels are common early and late in the day, and vice versa, high temperature and light levels only occur around midday. Thus, we could be attributing the temporal variation in fluxes to variation in environmental cues, when they could be driven by a combination of direct responses to environmental cues and endogenous circadian processes. Like Don Quixote, we could be fighting windmills, thinking that they are giants.

Abbreviations:
ET	=	evapotranspiration
GEE	=	gross ecosystem exchange
g s	=	stomatal conductance
VPD	=	leaf-to-air vapor pressure deficit

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.