Carbon Dioxide Enrichment Inhibits Nitrate Assimilation in Wheat and Arabidopsis

See allHide authors and affiliations

Science  14 May 2010:
Vol. 328, Issue 5980, pp. 899-903
DOI: 10.1126/science.1186440

Nitrate for Me, Ammonium for You

The interdependence of plant nitrogen uptake and plant responses to carbon dioxide is well established, but the influence of inorganic nitrogen form—i.e., whether nitrate or ammonium—has been largely ignored. Bloom et al. (p. 899) present evidence from five independent methods in both a monocot and dicot species that carbon dioxide inhibition of nitrate assimilation is a major determinant of plant responses to rising atmospheric concentrations of carbon dioxide. This finding explains several phenomena, including carbon dioxide acclimation and decline in food quality. The large variation in these phenomena among species, locations, or years derives from the large variation in the relative dependence of plants on nitrate and ammonium as nitrogen sources among species, locations, or years. The relative importance of ammonium and nitrate for plant N nutrition in future cropping systems will be critical for quantity and quality of food.


The concentration of carbon dioxide in Earth’s atmosphere may double by the end of the 21st century. The response of higher plants to a carbon dioxide doubling often includes a decline in their nitrogen status, but the reasons for this decline have been uncertain. We used five independent methods with wheat and Arabidopsis to show that atmospheric carbon dioxide enrichment inhibited the assimilation of nitrate into organic nitrogen compounds. This inhibition may be largely responsible for carbon dioxide acclimation, the decrease in photosynthesis and growth of plants conducting C3 carbon fixation after long exposures (days to years) to carbon dioxide enrichment. These results suggest that the relative availability of soil ammonium and nitrate to most plants will become increasingly important in determining their productivity as well as their quality as food.

The concentration of CO2 in Earth’s atmosphere has increased from about 280 to 390 μmol CO2 per mol of atmosphere (μmol mol–1) since 1800, and predictions are that it will reach between 530 and 970 μmol mol–1 by the end of the 21st century (1). Plants could mitigate these changes through photosynthetic conversion of atmospheric CO2 into carbohydrates and other organic compounds, yet the potential for this mitigation remains uncertain. Photorespiration is the biochemical pathway in which the chloroplast enzyme Rubisco catalyzes the oxidation of the high-energy substrate RuBP rather than catalyzes the carboxylation of RuBP through the C3 carbon-fixation pathway (2). Elevated CO2 (or low O2) atmospheric concentrations decrease rates of photorespiration and initially enhance rates of photosynthesis and growth by as much as 35% in most plants (C3 plants). This enhancement, however, diminishes over time (days to years), a phenomenon known as CO2 acclimation (3, 4). Most studies suggest a strong link between CO2 acclimation and plant nitrogen status [for example, (5)].

Nitrogen is the mineral element that organisms require in greatest quantity (6). The primary source of N for terrestrial plants is soil inorganic N in the forms of nitrate (NO3) and ammonium (NH4+). Root absorption of NO3 and NH4+ from the soil and assimilation of NO3 and NH4+ into organic N compounds within plant tissues have a large influence on primary productivity. The assimilation of NO3 involves the sequential conversion of NO3 into NO2, then into NH4+, then into glutamine, and finally into other organic N compounds. The first step of this process occurs in the cytosol, and the subsequent ones occur within chloroplasts or plastids.

Previously, we reported that atmospheric CO2 enrichment does not stimulate the growth of wheat plants receiving NO3 as a sole N source to the same extent as those receiving NH4+ (7). This result, as well as gas exchange measurements of wheat and Arabidopsis, suggested that elevated CO2 (or low O2) atmospheric concentrations, which are conditions that decrease photorespiration, inhibit NO3 assimilation in the shoots of C3 plants (7, 8).In the present study, we assessed the influence of elevated CO2 and sometimes low O2 atmospheric concentrations on NO3 assimilation in wheat and Arabidopsis using several independent methods.

Our first method, NO3 depletion, involved growing plants under ambient CO2 and O2 conditions and depriving them of NO3 nutrition until their tissue NO3 contents decreased to low steady levels (9). The shoots of these NO3-depleted plants were then subjected to an ambient or elevated CO2 atmospheric concentration and an ambient or low O2 atmospheric concentration. The roots then received a pulse of NO3 in the nutrient medium. The decline of NO3 concentrations in the medium provided an estimate of net plant NO3 absorption, and the difference between this net NO3 absorption and the accumulation of free NO3 within the plants provided an estimate of plant NO3 assimilation (8).

By the NO3-depletion method, wheat assimilated nearly all of the NO3 that its roots absorbed, whereas Arabidopsis assimilated less than half of the NO3 that its roots absorbed (Fig. 1). Estimates of NO3 absorption via this method were slower than estimates via the isotopic methods (15N and 14N labeling) that are described below. The NO3-depletion method deprives plants of NO3 for several days, and this has been shown to down-regulate the expression of several NO3 transporters (10). In contrast, the isotopic methods maintain a constant NO3 concentration in the medium and would not alter the expression of transporters.

Fig. 1

Three methods for assessing nitrate absorption (Absorb) and assimilation (Assim.) in wheat and Arabidopsis plants where the shoots were exposed to atmospheres containing 380 μmol mol–1 CO2 and 21% O2, 720 μmol mol–1 CO2 and 21% O2, or 380 μmol mol–1 CO2 and 2% O2. Shown are means ± SE (n = 6 to 18) in μmol NO3 per gram plant per minute. Within a species and for absorption separately from assimilation, bars labeled with different letters differ significantly (P < 0.05). Data for the NO3 depletion method include those from an earlier study (8).

The NO3-depletion method showed that an elevated CO2 atmospheric concentration around the shoots decreased the rate of NO3 assimilation (Fig. 1). A low O2 atmospheric concentration also decreased the rate of NO3 assimilation, but the decrease in Arabidopsis was not statistically significant. By this method, as in the isotopic methods described below, NO3 absorption varied with elevated CO2 and low O2 in a pattern that was similar to NO3 assimilation.

Our second method for assessing NO3 assimilation, 15N labeling, entailed growing plants under ambient CO2 and O2 conditions in a hydroponic medium containing 0.2 mM NO3 at natural abundance levels of N isotopes (≈0.366% 15N). We then shifted the plants to an ambient or elevated CO2 atmospheric concentration and an ambient or low O2 atmospheric concentration and to a root medium containing 0.2 mM NO3 that was 25%-enriched in 15N-NO3. After a 12-hour labeling period, we analyzed the plant tissues for 15N enrichment of total N and free NO3; the 15N enrichment of total N provided an estimate of net 15N absorption, and the difference between the 15N enrichment of total N and that of free NO3 provided an estimate of 15NO3 assimilation.

According to 15N labeling, wheat and Arabidopsis assimilated about two-thirds of the 15N-NO3 they absorbed (Fig. 1). In wheat, net 15NO3 absorption and assimilation were significantly greater under ambient CO2 and O2 atmospheric concentrations than under an elevated CO2 or low O2 concentration. In Arabidopsis, net 15NO3 assimilation was significantly greater under an ambient CO2 and O2 atmospheric concentration than under an elevated CO2 concentration.

In our third method, 14N-NO3 labeling, we grew plants under ambient CO2 and O2 conditions in a hydroponic medium that contained 99.9% enriched 15N-NO3 as the sole N source. When the wheat and Arabidopsis plants were about 14 and 36 days old, respectively, the shoots were exposed to an ambient or elevated CO2 atmospheric concentration and an ambient or low O2 atmospheric concentration. After a few hours under these atmospheric conditions, the roots received a pulse of NO3 containing the isotopes at their natural abundance levels of nitrogen (99.633% 14N). We estimated 14N-NO3 absorption and assimilation from the decreases in the 15N enrichment of total N and free NO3 in plant tissues after a 12-hour exposure to 14N-NO3.

Differences in atmospheric CO2 or O2 concentration produced distinct patterns of the 14N labeling (Fig. 1). Wheat and Arabidopsis assimilated about two-thirds of the 14N-NO3 they absorbed. In both species, an elevated CO2 or low O2 atmospheric concentration significantly decreased 14NO3 assimilation.

Our fourth method for assessing NO3 assimilation depended on the assimilatory quotient (AQ), the ratio of net CO2 consumption to net O2 evolution from shoots during photosynthesis. Values of AQ decrease as NO3 assimilation increases, because additional electrons generated from the light-dependent reactions of photosynthesis are transferred first to NO3 and then to NO2. This stimulates net O2 evolution but has little effect on CO2 consumption (7, 8, 1113). The ∆AQ—the difference in the AQ between plants receiving NH4+ as the sole N source and those receiving NO3—is strongly correlated with shoot NO3 assimilation. For example, ∆AQ did not deviate significantly from zero in plants with relatively low NO3 reductase activities (that is, Arabidopsis knockout mutants and 48-day-old wild-type Arabidopsis), whereas ∆AQ was positive in plants with significant NO3 reductase activities (that is, wild-type wheat, transgenic Arabidopsis that overexpresses NO3 reductase, and 36-day-old wild-type Arabidopsis) (8). Rates of photorespiration do not influence AQ or ∆AQ, because this process changes neither net CO2 consumption nor net O2 evolution (2).

During these gas-exchange measurements, we exposed the shoots of wheat and Arabidopsis to various atmospheric CO2 concentrations. To account for differences in stomatal conductance, we expressed shoot CO2 and O2 fluxes as a function of apparent shoot internal CO2 concentrations (Ci) that we calculated from water vapor exchange. With increasing Ci, the ∆AQ in both species declined from a positive value to one that was not significantly different from zero (Fig. 2), indicating that NO3 assimilation was significant at subambient and ambient CO2 concentrations but was negligible at elevated CO2 concentrations.

Fig. 2

The ΔAQ, the change in the ratio of shoot CO2 consumption to O2 evolution with a shift from NO3 to NH4+ nutrition, as a function of shoot internal CO2 concentration (Ci) in wheat and Arabidopsis. An instrumental system described previously (12) monitored shoot gas fluxes. A biochemical model of photosynthesis (33, 34), which we fitted to the data, interpolated the values at regular Ci intervals. Shown are the means ± SE (n = 8 for wheat and 4 for Arabidopsis).

Our fifth method relied on the isotopic discrimination of NO3 reductase. When both isotopic forms of NO3 are readily available, this enzyme preferentially converts 14N-NO3 into organic N compounds by about 15 per mil (‰) (14), and so organic N compounds in plant tissues are more depleted in 15N than the N compounds in the growth medium are. In contrast, when NO3 availability limits assimilation, this enzyme discriminates less against 15N-NO3 and assimilates relatively more 15N-NO3, and so organic N compounds in plant tissues become more enriched in 15N.

We grew wheat and Arabidopsis for 14 and 22 days, respectively, in a medium that contained 0.2 or 1.0 mM NO315N = –4‰) as the sole N source and under an ambient or elevated CO2 atmospheric concentration. The difference between total plant N and free NO3 in the tissues provided an estimate of organic N. Wheat shoot growth did not respond to any of the treatments, whereas Arabidopsis shoot growth was greater at the higher NO3 level but did not respond to CO2 treatment (Fig. 3). Shoot NO3 contents were higher in plants that were grown at the higher NO3 level, and shoot organic N contents decreased under CO2 enrichment, although the decrease in wheat grown at 1.0 mM was not statistically significant (Fig. 3).

Fig. 3

Shoot biomass, NO3 content, organic N content, and δ15N of organic N in wheat (upper panels) and Arabidopsis (lower panels) grown at 0.2 or 1.0 mM NO3 and 380 or 720 μmol mol–1 CO2. Shown are means ± SE (n = 6 to 12). For each parameter, bars labeled with different letters differ significantly (P < 0.05).

Shoot δ15N of organic N (Fig. 3) and plant NO3 assimilation rate assessed via the isotopic methods (Fig. 1) were lower in wheat than in Arabidopsis. In both species, δ15N of shoot organic N decreased in plants grown at the higher NO3 level or under CO2 enrichment, although the decrease with CO2 in Arabidopsis that was grown at 1.0 mM was not statistically significant (Fig. 3). All of these results are consistent with NO3 assimilation discriminating more against 15N-NO3 when the availability of NO3 at the sites of NO3 reduction was higher, as a consequence of either slower NO3 assimilation in a species (wheat versus Arabidopsis), increased NO3 supply (1.0 versus 0.2 mM), or decreased NO3 assimilation (elevated versus ambient CO2).

In this study, five independent methods affirm that CO2 enrichment inhibits NO3 assimilation in wheat and Arabidopsis plants. The predominant form of N available to plants in most environments is NO3 (6); therefore, CO2 inhibition of NO3 assimilation would lead to lower organic N production. Indeed, this could be responsible for the 7.4 to 11% decrease in wheat grain protein (15, 16) and the 20% decrease in total protein content of A. thaliana (Columbia) (17) observed under CO2 enrichment in FACE (free-air CO2 enrichment) experiments. Because the influence of CO2 enrichment on leaf organic N contents is highly correlated with its influence on photosynthesis and growth (5), it is reasonable to assume that CO2 inhibition of NO3 assimilation and the resultant decline in plant organic N contents play a major role in the phenomenon of CO2 acclimation, the decline of photosynthesis, and growth of C3 plants after long exposures (days to years) to CO2 enrichment.

The extent to which plants use NO3 versus NH4+ as N sources varies over seasons, years, locations, and species (6). This variation in the relative dependence on NO3 could explain the observed variation in CO2 acclimation. Net primary productivity diminished under CO2 enrichment in an annual California grassland for which NO3 was the predominant N source (18), presumably because NO3 assimilation was inhibited and plant organic N compounds became limiting. In contrast, Scirpus olneyi, the prominent C3 plant in the Chesapeake Bay marsh, which is an NH4+-dominated ecosystem, showed little CO2 acclimation. Even after a decade of treatment, photosynthesis and growth of this species remained about 35% greater under CO2 enrichment (19), with little change in N contents (20).

Root NO3 absorption and plant NO3 assimilation were generally correlated in the N-depletion and isotopic methods, yet differences in the responses to elevated CO2 or low O2 atmospheric concentrations were sometimes significant for assimilation but not for absorption (Fig. 1). Moreover, CO2 enrichment decreased shoot organic N contents, but did not change or even increased shoot NO3 contents (Fig. 3). CO2 enrichment also increased 15N isotope discrimination during NO3 assimilation (Fig. 3), indicating that NO3 availability became less limiting to assimilation. Finally, changes in atmospheric CO2 concentration influenced shoot NO3 assimilation within minutes (time response of Fig. 2 not shown). These CO2 changes also influenced transpiration rapidly, but root NO3 and NH4+ absorption from well-mixed hydroponic solutions is independent of transpiration (21). Together, these results indicate that elevated CO2 or low O2 atmospheric concentrations inhibited NO3 assimilation and that assimilation controlled root absorption, rather than elevated CO2 or low O2 atmospheric concentrations influencing root NO3 absorption directly.

One physiological mechanism that may be responsible for the relationship between elevated CO2 or low O2 atmospheric concentrations and NO3 assimilation involves the first biochemical step of NO3 assimilation, the conversion of NO3 to NO2 in the cytoplasm of leaf mesophyll cells. Photorespiration stimulates the export of malic acid from chloroplasts (22) and increases the availability of the reduced form of nicotinamide adenine dinucleotide (NADH) in the cytoplasm (23) that powers this first step (24, 25). Elevated CO2 or low O2 atmospheric concentrations decrease photorespiration and thereby decrease the amount of reductant available to power NO3 reduction. In contrast, the C4 carbon fixation pathway generates ample amounts of malic acid and NADH in the cytoplasm of mesophyll cells. This may explain why shoot NO3 assimilation is relatively independent of CO2 concentrations in C4 plants (26) and limited to the mesophyll (27).

Another physiological mechanism that may link NO3 assimilation and elevated CO2 is NO2 translocation from the cytosol into the chloroplast. Six transporters of the Narl family are involved in NO2 translocation from the cytosol into the chloroplast in Chlamydomonas, and some of these transport both NO2 and HCO3 (28). We have shown that HCO3 inhibits NO2 influx into isolated wheat and pea chloroplasts (7), indicating that an analogous system is operating in higher plants. Slower NO2 influx into the chloroplast under CO2 enrichment would decrease NO3 assimilation.

A third physiological mechanism linking CO2 enrichment and NO3 assimilation involves competition for reductant in the chloroplast stroma. Several processes within the stroma—C3 carbon fixation, the reduction of NO2 to NH4+, and the incorporation of NH4+ into amino acids—require reduced ferredoxin generated by photosynthetic electron transport. Key enzymes in these processes have different affinities for reduced ferredoxin: FNR (ferredoxin-NADP reductase) has a Michaelis constant Km of 0.1 μM, NiR (nitrite reductase) has a Km of 0.6 μM, and GOGAT (glutamate synthase) has a Km of 60 μM (29). As a result, NO3 assimilation may proceed only if the availability of reduced ferredoxin exceeds that needed for the formation of the reduced form of nicotinamide adenine dinucleotide phosphate (NADPH) (24, 30). For most plants, this occurs when CO2 availability limits C3 carbon fixation (7).

The phenomenon of CO2 acclimation may have several explanations. According to the carbohydrate sink limitation hypothesis, plants under CO2 enrichment initially assimilate more CO2 into carbohydrates than they can incorporate into their growing tissues; in response, they diminish CO2 assimilation by decreasing their levels of Rubisco and other proteins (3). An alternative explanation is the progressive N limitation hypothesis in which shoots accumulate carbohydrates faster than plants can acquire N, making leaf N contents decrease (4, 31, 32). As these leaves senesce and drop to the ground, plant litter quality declines, microbial immobilization of soil N increases because of the high C-to-N ratios in the litter, soil N availability to plants further diminishes because more soil N is tied up in microorganisms, plants become even more N limited, plant protein levels decline, and plant processes including photosynthesis and growth slow down.

Both of these hypotheses about CO2 acclimation fit nicely into the framework of our results. The decline in Rubisco predicted by the carbohydrate sink hypothesis might derive from CO2 inhibition of NO3 assimilation and the subsequent decline in plant organic N compounds (Fig. 3). The decline in leaf N contents predicted by the progressive N limitation hypothesis might derive from CO2 inhibition of NO3 assimilation and the subsequent decline in plant NO3 absorption (Fig. 1).

Our findings have implications for food production. Nitrate is the most abundant form of N in agricultural soils (6). As atmospheric CO2 concentrations rise and NO3 assimilation diminishes, crops will become depleted of organic N compounds (see Fig. 3), including protein, and food quality will suffer. Increasing nitrogen fertilization might compensate for slower NO3 assimilation rates (Fig. 3), but such fertilization rates might not be economically or environmentally feasible. Greater reliance on NH4+ fertilizers and inhibitors of nitrification (microbial conversion of NH4+ to NO3) might avoid the bottleneck of NO3 assimilation, but would require sophisticated fertilizer management to prevent NH4+ toxicity, which occurs when free NH4+ accumulates in plant tissues if they absorb more of this compound than they can assimilate into amino acids. To address these issues, a better understanding of plant NH4+ and NO3 assimilation is critical.

Supporting Online Material

  • Present address: Department of Land, Air and Water Resources, University of California at Davis, Davis, CA 95616, USA.

  • Present address: School of Biological Sciences, Post Office Box 646340, Washington State University, Pullman, WA 99164–6340, USA.

References and Notes

  1. Materials and methods are available as supporting material on Science Online.
  2. We thank C. van Kessel, L. Jackson, and E. Carlisle for their review of the manuscript. This research was supported by NSF grants IBN-03-43127 and IOS-08-18435, by the National Research Initiative Competitive grant number 2008-35100-04459 from the U.S. Department of Agriculture National Institute of Food and Agriculture, and by a postdoctoral fellowship from Agencia Regional de Ciencia y Tecnologia, Region de Murcia, Spain to J.S.R.A. A.J.B. is a paid consultant to the Monsanto Corporation.
View Abstract

Stay Connected to Science

Navigate This Article