Sean M. Gleason‎ > ‎


Making plants and cropping systems more productive in dry environments

It is well understood that subjecting plants to drought reduces their productivity.  Although this statement is intuitive, the sequence of physiological failures that accompany drought are poorly understood.  Do plants “starve”, i.e., net CO2 assimilation falls below rates necessary for survival?  Does their water transport system “break”?  Do they down-regulate their photochemistry?  Although responses are likely to be species- and context-specific, are there general responses to water stress that occur across most angiosperm species?

Our first-order question should be, if we look across the entire world, “what plant traits confer drought resistance in naturally occurring species?”  A more specific question might be, “how has the heterogeneous distribution of water across the earth’s terrestrial ecosystems affected the distribution of plant traits present in those ecosystems?”  Then, we can choose the most likely traits to confer drought resistance, but not result in reduced productivity when water is plentiful.  Impossible?  Not if we carefully consider the outcomes of natural selection. 



Gas exchange traits 

Many plant traits are more common in arid environments than in wet environments.  Unfortunately, many of these plant traits also result in reduced productivity.  A good example of this is the “leafiness” of branches.  If we consider the amount of leaf area on a plant, relative to the cross-sectional area of living wood (xylem) that supports it (“leafiness”), we notice that arid-land plants are less leafy than plants from wet places, like rainforests.  In fact, the power of leafiness to predict where species occur is pretty good – about 20% of the total variation in leafiness across the world’s angiosperms is aligned with variation in mean annual precipitation.  However, less leafy plants have less photosynthetic tissue than plants in wetter habitats, and are thus likely to be less productive. 

The ratio of internal leaf CO2 (ci) to external leaf CO2 (ca) is another good example.  Plants with low ci/ca values are often found in dry habitats.  This is because plants in dry habitats tend have leaves with high resistance to gas conductance across the cuticle.  This is a direct response to reduce water loss in hot and dry habitats, i.e., increased resistance results in slower rates of water loss.  Unfortunately, increasing the resistance to gas diffusion, including CO2, results in lower concentrations of CO2 at the chloroplasts and lower rates of photosynthesis.  This is obviously a “bad” trait for crop plants because less net CO2 assimilation would lead to less plant biomass when water is plentiful, all else being equal.  On the other hand, plants from dry habitats tend to have higher activities of photosynthetic enzymes.  This also reduces ci/ca values and results in a direct increase in net CO2 assimilation, without adversely affecting water loss, but also requires better N nutrition (Wright et al. 2003).  This contributor to lower ci/ca might be a good trait to look for in dryland crop species, particularly if N fertility is not an issue.  Interestingly, evolution has resulted in the tight optimization of photosynthetic efficiency and water use, depending on the relative “costs” of N and water in different habitats (Prentice et al. 2013). 


Xylem traits

The plant tissue that transports water and nutrients to the canopy is known as “xylem”.  Xylem has other functions as well, but for now, let’s consider only its role in water transport.  Xylem is a network of conduits that transport water from the roots, through the stems and leaves, to the atmosphere.  Under water stress, these columns of water come under tension, break, and then fill with gas – becoming useless.  Although it is well-understood that xylem conduits do fill with gas (become “embolized”) when tensions increase, there is much variation in the susceptibility of species to embolization (ca two orders of magnitude) (Maherali et al. 2004). 

It is not completely understood what anatomical traits cause this variation across species.  Not knowing the direct causes of embolism resistance/susceptibility has thwarted our efforts to enhance embolism resistant traits in crop plants.  Nevertheless, there exists good correlative evidence linking pit properties, mainly the size and number of pits, as well as pit and membrane ultrastructure, with embolism resistance (Hargrave et al. 1994; Wheeler et al. 2005; Jansen et al. 2009, Lens et al. 2011).  Unfortunately, these traits are difficult to measure across many species and, thus, the comparative power of these analyses are presently limited. 

Embolism resistance itself can be measured (even though the causes are unknown).  Embolism resistance is most often expressed as the xylem water potential (i.e., tension) at which 50% of the maximal conductive capacity is lost (i.e., P50).  P50 values have been measured on over 1000 angiosperm species, from nearly every biome on earth (Choat et al. 2012).  It is a fairly good predictor of climate with P50 of branch xylem explaining ca 31% of the variation in habitat soil water potential (unpublished data) and the P50 of leaf xylem explaining somewhere between 20% and 68% of the variation in habitat precipitation (Blackman et al. 2014, Nardini & Luglio 2014).  Furthermore, the simple measure of xylem density (i.e., “wood” density) explains about 25% of the variation in soil water potential across species and habitats (unpublished data).  Clearly, 170 million years of natural selection has resulted in strong modification of angiosperm xylem in arid habitats.  Should we desire any of these traits in crop plants?


The embolism resistance vs efficiency tradeoff

It has long been thought that a strong tradeoff should exist between xylem embolism resistance and the capacity of xylem to transport water (hereafter “efficiency”).  This is a very intuitive and logical theory and reaches back at least as far as Dixon (1914), who realized the role of bordered pits in conferring both efficiency and embolism resistance.  But now, with the aid of large global trait databases, we can assess the validity of this idea, at least to a large extent.  If there existed a strong tradeoff between embolism resistance and efficiency throughout the evolutionary history of angiosperms I would expect a negative (not necessarily linear) correlation between embolism resistance and efficiency.  Across the world’s angiosperm species the r2 (% variation explained) of this relationship falls somewhere between 0.05 and 0.10 (Maherali et al. 2004; Gleason et al. 2016).   

I would suggest that for crop scientists this is very good news.  It suggests that the relationship between embolism resistance and transport efficiency has not been an important factor driving the divergence of angiosperm xylem.  Furthermore, it is likely that efforts to increase the transport efficiency of crop plants would not necessarily result in reduced embolism safety.   


The way forward

Selection for embolism resistance in crop plants appears possible in maize (genotype used in Li et al. 2009).  Perhaps more importantly, careful breeding efforts coupled with anatomical, proteomic, and physiological measurements, may help elucidate the structural and physiological causes underpinning embolism resistance. Considering the rapid pace of molecular techniques to directly modify genes and gene expression, understanding the underpinnings of embolism resistance and other traits of drought tolerance should be a high priority for plant physiologists, geneticists, and plant breeders.      


Some recent research outputs...