Watercress is a semi-aquatic plant that grows in f lowing shallow freshwater and is found across Europe, Asia, the Americas, the Caribbean, New Zealand and Australia. Watercress is placed within the Brassicaceae family together with several other important food crops including broccoli, kale, cabbage, and mustard. A significant amount of commercial aquatic watercress cultivation is centred in a few locations including Florida in the USA, southern Spain and Portugal, France and the south of England, with 90% of production occurring in Dorset, Hampshire and Wiltshire. These chalky areas provide nutrient-rich spring water and boreholes that directly supply the watercress beds. Phosphate-rich fertiliser is used to boost crop yield; however, this presents a major challenge in watercress production since it results in direct leakage of phosphate into the waterways which have high conservation value. Excess phosphate results in eutrophication of aquatic ecosystems, a process where nutrient enrichment of water sources results in excessive algal and plant growth, and subsequent disruption of ecosystem community dynamic. Approximately 90% of watercress farms in the UK are on, or upstream of, a Site of Special Scientific Interest , increasing the pressure to minimise phosphate release.Phosphate is vital for plant survival; it forms the phosphodiester bonds that link nucleotides in nucleic acids and is critical for the structure of proteins and carbohydrate polymers, for powering cells through the release of phosphate from ATP and for regulating several metabolic pathways. Symptoms of P deficiency are retarded growth, increased root:shoot biomass,vertical farming racks decreased leaf area and often dark green or purplered colouration in severely deficient plants due to anthocyanin production.
Ninety percent of the global demand for phosphorus is used for food production, however, rock phosphate is a limited resource with estimates that reserves could be exhausted in the next 50–100 years. In addition, most of the remaining rock phosphate reserves areunder the control of only a few countries, with Morocco and the Western Sahara holding over 70% of the total reserves, making it sensitive to political instability. This, combined with increasing costs of extraction and issues of eutrophication,make reducing fertiliser use an important global driver. The high reactivity and low solubility of phosphates make them commonly the growth-limiting nutrient for plants. Accounting for fertiliser application, approximately 30% of global cropland area exhibits soil P deficits although global P imbalances in water sources have not been investigated to any significant extent.Phosphates in soils are classified into inorganic P and organic P and fractionised based on their solubility. Po typically comprises 30–70% of total soil P, however this is highly variable with values up to 90% reported. Pi derived from rocks is only present in poorly soluble forms and any soluble phosphates released in the soil are rapidly immobilised into insoluble forms by interacting with adsorbing agents such as Al and Fe oxides or Ca and Mg carbonates, such that only a small amount of P in soils is accessible to plants. The availability of P is dynamic, changing over time depending on various factors such as temperature, water content and soil pH.Like soil, P in aquatic systems is also divided into different fractions based on solubility and reactivity in aquatic systems, with dissolved orthophosphate the most bio-available. P in water adsorbs to oxides and tightly binds with carbonates in the same manner as when in soil. However, the P inputs to natural water systems and the interaction with P in bed sediments is altered.
This creates a dynamic source of phosphorus that transfers between particulate and dissolved forms, between bed sediments and the water column, and between dead and living material . In a watercress bed, the sediment is shallow gravel and thus P uptake from water likely represents the major P source. This is reflected in a study by Cumbus and Robinson who found that a greater proportion of P was absorbed by the adventitious roots of watercress , compared to basal roots. However, some organic detritus held within the sediment should still be considered. Phosphate dynamics in hydroponic agricultural systems such as watercress beds have not been studied, representing a knowledge gap, but P is likely uniformly distributed due to f lowing water and regular maintenance of P concentrations. Since P retention in sediments is high, P delivery into freshwater systems is largely governed by release from point sources such as sewage treatment works , leaking septic tanks, and from excess fertiliser application. Globally, domestic sources contribute 54% to total P inputs into freshwater systems, 38% from agriculture and 8% from industry . Although substantial steps towards P reduction in fresh waters have been made over the last 50 years there is still much to be done, with only 40% of European surface waters currently in good ecological status. Eutrophication of watercourses is also prevalent across the UK: in the most recent analysis, 55% of river water bodies in England failed to meet the revised P standards for good ecological status. Eutrophication is both an economic as well as environmental issue. In the US, the economic damage of eutrophication equates to $2.2 billion annually, due to losses in recreational water use, waterfront real estate, recovery of endangered species and drinking water. Naturally, phosphate levels in chalk aquifers are less than 20 μg/l, however, inputs of phosphate rapidly increase these concentrations above P targets downstream of watercress farms. In the river Itchen where several watercress farms are located, total SRP load comes predominantly from sewage treatment works but watercress beds can be responsible for up to 62% of the total reactive phosphate in some chalk streams, suggesting room for improvement in P management. Casey and Smith found watercress beds increased mean P concentrations which may cause undesirable growth of algae and disruption of community dynamics.
One important strategy to tackle this problem of eutrophication is through plant breeding. By breeding watercress varieties with improved phosphorus use efficiency , the impact of watercress farming on eutrophication could be minimised. To date, no breeding for nutrient use has been conducted in watercress even though P release represents a clear issue in watercress productionHere PUE is used as a broader term that also encompasses phosphate acquisition efficiency, defined as the ability to take up P, as has been used in several studies. Plant traits underpinning PUE can be observed at the macroscopic, microscopic, and molecular levels and we consider their relevance to future breeding for enhanced PUE. To date, knowledge on P acquisition by aquatic plants only covers the effectiveness of plants for phytoremediation , rather than breeding for PUE in aquatic crops such as water chestnut , water spinach , lotus and watercress. Present information does not cover morphological or genetic components to improve PUE in aquatic species, and with new plant species emerging as suggested model organisms, watercress is offered as a model crop for aquatic systems. The need for aquatic modelcrops is only exacerbated by increasing market value in indoor hydroponic cultivation systems .Root structural architecture defines the spatial configuration of the root system and variation in RSA can ref lect efficient phosphate uptake by plants. Much of the current literature focuses on RSA traits in maize , common bean and Arabidopsis thaliana with RSA shown to be a highly plastic trait, changing in response to the availability of water,vertical racking system nutrients and hormone signalling. For example, ethylene is involved in the promotion of lateral root growth, root hair growth and inhibition of primary root growth under low P conditions. Reduced root growth angle : is one of the most important RSA traits for improving P acquisition in many soil-grown species. In maize and common bean, lower RGA is associated with increased P accumulation and improved growth in P deficient soil where P is concentrated in the topsoil.However, in aquatic systems when P is likely homogenously distributed, a lower RGA is unlikely to be advantageous. Root growth angle therefore is not considered important as a trait for enhance PUE in aquatic plants. Increased lateral root density: also enhances P acquisition by allowing plants to explore outside P-depleted zones. Using maize recombinant inbred lines with contrasting lateral rooting phenotypes, significant differences in phosphorus acquisition, biomass accumulation and relative growth rate were observed under low phosphorus availability. Increased investment in the production of lateral roots was shown to be cost-effective under low P. For aquatic crops, enhanced lateral root density is an important trait for enhanced PUE since it increases root surface area for P uptake. Adventitious roots: from above ground structures can enhance topsoil foraging by up to 10% in stratified soil.
These require a lower metabolic investment than basal roots, however, in uniform soil they can limit P acquisition by hindering the growth of basal roots . In aquatic and f looding-tolerant plants, adventitious roots are important for P uptake within the water column. There is positive correlation between the size of the adventitious root system and P uptake in bittersweet plants under long-term submergence, and as mentioned, adventitious roots are responsible for a higher proportion of P acquisition in watercress than basal roots. Thus, in an aquatic crop such as watercress, increased adventitious root production is a key trait for enhanced PUE.Root cortical aerenchyma: are an adaptation to waterlogged soils and reduce the risk of asphyxiation and their formation has been suggested to increase P uptake in deficient soils. Several studies have reported increase in RCA formation in maize under these conditions. Modelling of root architecture in maize and bean has shown RCA may increase the growth of plants up to 70% in maize and 14% in bean under low P availability largely due to remobilisation of P from dying cells. However, formation of aerenchyma is also associated with reduced root hydraulic conductivity in maize which may impede transport of water, but is unlikely to be relevant to hydroponically grown crops . Most aquatic plants, including watercress, form aerenchyma constitutively in roots and stems to aid internal gas exchange and maintain strength under water pressure. For aquatic crops such as watercress, formation of root aerenchyma is an important trait for selection for enhanced PUE. Root hair density and root hair length: root hair density increases up to 5 times in low P conditions. Using Arabidopsis mutants, Bates & Lynch found hairless plants had lower biomass and produced less seed than wild-type plants at low phosphorus availability. Root hairs increased root surface area by 2.5 to 3.5-fold in barley and wheat , respectively, and there was an almost perfect correlation between P uptake and root hair surface area [74]. Root hair traits vary substantially between genotypes and the genetic control underlying their formation is well understood, thus making them an excellent target for plant breeding programs. Root hair length and density are likely to be important for PUE in aquatic crops as they significantly increase root surface area for P uptake. Synergism of root phenotypes should also be considered. A modelling approach in Arabidopsis showed that the combined effects of root hair length, root hair density, tip to first root hair distance and number of trichoblast files on P acquisition was 3.7-fold greater than their additive effects. For aquatic species such as watercress, the root ideotype for determining optimal P acquisition remains unknown. Although, absorption of P through the shoots is still debated, root uptake is generally regarded as the mode of P uptake in aquatic plants.Watercress beds have a fine gravel substrate which contains negligible amounts of P. The substrate within watercress beds is likely too shallow to allow for significant stratification of P, thus a shallower basal root angle would be unlikely to provide much adaptive benefit. In groundwater sources, P may be distributed more homogenously due to turbulent f lowing water, so RGA will not assist within the water column. Nevertheless, even with homogenous P distribution, plants with shallower root systems have been shown to encounter less inter-root competition with roots on the same plant so RGA could provide an adaptive value in this sense. Cumbus & Robinson studied P absorption by the adventitious and basal roots of watercress and found that the adventitious roots absorbed a higher proportion of P at low P concentrations, despite having a lower biomass compared to the basal root tissue. Thus, adventitious roots are also a key trait for analysing watercress PUE.