An evaluation of the evidence for, and implications of, cytoplasmic nitrate homeostasis


  • M. Y. Siddiqi,

    1. Department of Botany, University of British Columbia, 6270 University Boulevard, Vancouver, Canada V6T 1Z4
    Search for more papers by this author
  • A. D. M. Glass

    Corresponding author
    1. Department of Botany, University of British Columbia, 6270 University Boulevard, Vancouver, Canada V6T 1Z4
      Dr A. D. M. Glass. Fax: +1 604 822 6089;
    Search for more papers by this author

Dr A. D. M. Glass. Fax: +1 604 822 6089;


A review of literature, reporting values of cytoplasmic/cytosolic [NO3] in plant cells, identified two major areas of disagreement: (1) disparity in the absolute values within the same system, and (2) constancy versus variability in cytoplasmic/cytosolic [NO3] with varying [NO3]o. These differences are related to the techniques used by the different authors. Estimates of cytoplasmic [NO3] by compartmental analysis and by cell fractionation were consistently higher than the estimates by NO3selective microelectrodes and by techniques based upon in vivo and in vitro nitrate reductase activity (NRA). A model recognizing more than one cytoplasmic ionic pool would satisfactorily reconcile the differences in both aspects, i.e. absolute values and constancy. Compartmental analysis and cell fractionation techniques may measure the amount of NO3 in the cytoplasm as a whole (including organelles); by contrast, NO3 selective microelectrodes and NRA estimate only the cytosolic NO3 and, hence, may result in lower estimates. Thus, variable organellar pool(s) may maintain a constant cytosolic pool as estimated by microelectrodes. However, certain observations remain at odds with the notion of a constant cytosolic [NO3].


Cytoplasmic/cytosolic nitrate concentration has been estimated in several systems (species, organs, and tissues), using different techniques. Reported values of cytoplasmic/cytosolic [NO3] range from 0·01 to 0·1 mm at the lower end to 50 mm at the higher end, depending upon the plant systems and techniques used (Table 1). Table 1 lists five discrete techniques that have contributed estimates of cytoplasmic/cytosolic [NO3]. These fall into two categories, namely methods based upon the use of nitrate-specific electrodes, assays of nitrate reductase activity (NRA) and nuclear magnetic resonance (NMR) studies which determine the activity or concentration of NO3, and those that estimate the total amount of cytoplasmic NO3 and deduce concentrations from estimates of cytoplasmic volume. The latter include compartmental analysis by efflux (CAE) and cell fractionation. Interestingly, these two groups, i.e. one reporting lower values (Zhen et al. 1991; King, Siddiqi & Glass 1992; van der Leij, Smith & Miller 1998), and the other reporting the higher values (Siddiqi, Glass & Ruth 1991) differ in a more fundamental way. Among the former group, Miller and his coworkers (see Table 1 for references) have provided evidence that cytosolic [NO3] is held constant within narrow limits over a wide range of external NO3 concentrations. By contrast, Siddiqi et al. (1991) ­demonstrated that cytoplasmic [NO3] was not rigorously controlled, and that it increased by a factor ∼ 2 with every 10-fold increase of external NO3 concentration ([NO3]o) in the range from 0·01 to 1·0 mm. This lack of constancy has also been demonstrated in several other systems (e.g. ­Devienne, Mary & Lamaze 1994b; Kronzucker, Siddiqi & Glass 1995a; Min et al. 1999). In an attempt to resolve this disparity, Siddiqi et al. (1991) suggested that the different methodologies (NRA and CAE, resulting in micromolar and millimolar values, respectively) might be detecting spatially distinct NO3 pools, such as those of the epidermal and cortical cells, respectively. This suggestion arose from two observations (1) MacKown, Jackson & Volk (1983) reported that NO3 reduction and translocation appeared to proceed from a different pool than that responsible for NO3 efflux, and (2) Rufty et al. (1986) reported that NRA was mainly localized in epidermal cells of corn roots. Subsequent investigations using NO3-specific microelectrodes have not substantiated this hypothesis, indicating that different cell types of the root differ only modestly in cytosolic [NO3] (Zhen et al. 1991; van der Leij et al. 1998).

Table 1.  Measurements or estimates of cytoplasmic/cytosolic [NO3] by different techniques. The techniques are separated into two groups. Group 1 techniques measured or estimated NO3 concentration directly whereas group 2 techniques measured the amount of NO3 in the cytoplasm and then cytoplasmic [NO3] was deduced from measurements or estimates of cytoplasmic volume (see text)
[NO3] (mm)
Group 1
 <1 mmNRA-basedCultured tobacco cellsFerrari, Yoder & Filner (1973)
 0·01–0·1 mmNRA-basedLeaves of barley, maize, pea,
 soybean, rice
Robin et al. (1983)
 0·065–0·14NRA-basedWhole plantsBelton et al. (1985)
 0·065–0·14NRA-basedCultured rose cellsBelton et al. (1985)
 0·66–3·9–17NRA-basedBarley (Steptoe) rootsKing et al. (1992)
 4–8NRA-basedspinach leafSteingröver, Ratering& Siesling (1986).
[NO3]o = 4 mm
 1·6MicroelectrodeChara cellsMiller & Zhen (1991)
 3·2–5·4MicroelectrodeBarley rootsZhen et al. (1991)
 ∼ 4MicroelectrodeBarley rootsMiller et al. (1995).
[NO3]o = 0·1–10 mm
 2·9–3·7MicroelectrodeBarley root cortical cellsvan der Leij et al. (1998)
 3·9–4·6MicroelectrodeBarley root epidermal cellsvan der Leij et al. (1998)
 3·2–5·4MicroelectrodeBarley root cortex,
 epidermis, respectively,
Zhen et al. (1991) (See also Miller &
 Smith 1992
for recalculated values.)
 0·63MicroelectrodeLiverwortTrebacz, Simonis & Schönknecht (1994)
 undetectable14N - NMRBarley, maize, pea, rootsBelton et al. (1985)
Group 2
 4·1Cell fractionationBarley leafMartinoia et al. (1986)
 6·8Cell fractionationMesophyll cellMartinoia et al. (1987)
 ≤ 30Cell fractionationBarley leafWinter et al. (1993). (See also Lohaus et al. 1995.)
[NO3]o = 14 mm
 ≤ 8Cell fractionationSpinach leafWinter et al. (1994). (See also Lohaus et al. 1995.)
[NO3]o = 14 mm
 50–100CAE - 13NO3Maize rootsPresland & McNaughton (1984).
[NO3]o = 1·7–70 mm
 26CAE - 13NO3Barley rootsLee & Clarkson (1986).
[NO3]o = 1·5 mm
 40–50CAE - 15NO3Onion root epidermisMacklon, Ron & Sim (1990).
[NO3]o = 2 mm
 12–37CAE - 13NO3Barley rootsSiddiqi et al. (1991).
[NO3]o = 0·01–1 mm
 0·3–4CAE - 13NO3SpruceKronzucker et al. (1995a).
[NO3]o = 0·01–1·5 mm
 4–8CAE - 15NO3Soybean rootMuller, Tillard & Touraine (1995) [NO3]o = 0·5 mm
 4·4–5·8CAE - 13NO3Trembling aspenMin et al. (1999).
[NO3]O = 0·1–1·5 mm
 0·41–7·37CAE - 13NO3Lodgepole pineMin et al. (1999).
[NO3]O = 0·1–1·5 mm
 0·92–2·37CAE - 13NO3Douglas firMin et al. (1999).
[NO3]O = 0·1–1·5 mm
 18–46CAE - 15NO3Wheat rootsDevienne et al. (1994b).
[NO3]O = 0·1–5 (mm)

Miller & Smith (1996) have suggested that ‘either cytosolic nitrate is very variable or there are errors associated with each of the methods that is used to measure it’. By contrast, we suggest the following:

  • 1The two groups of techniques actually measure quite different parameters, namely the cytosolic activity or concentration of NO3 (nitrate-specific electrodes or NRA) or the cytoplasmic NO3 quantity (CAE and cell fractionation), from which an average value of cytoplasmic [NO3] is calculated.
  • 2Regardless of these distinctions, a number of fundamental problems arise from the concept of a strict homeostat for cytosolic NO3, including variable values of NRA, NO3 efflux, NO3 influx due to high-affinity transport systems (IHATS), NO3 translocation rates and transcript abundance of genes encoding NR and IHATS. These observations appear to be incompatible with a constant cytosolic [NO3].

Literature review and discussion

Table 1 lists NO3 concentrations estimated by different techniques. As outlined above, these techniques may broadly be classified into two categories:

These techniques directly measure the activity (i.e. concentration) of [NO3] in the cytosol, and no measurement or estimation of cytosolic volume is involved in its calculation. One potential source of error associated with the NRA-based technique is the assumption that the NR enzyme behaves similarly in vitro and in vivo. In the case of microelectrodes, one possible source of error may be the calibration of the NO3-selective microelectrode in simple aqueous solution and its use in a complex cytoplasmic milieu. However, until experimentally shown to the contrary, we assume that these aspects have been adequately considered by the respective workers, and accept the reported values as valid measurements.

The most extensively used technique among these is compartmental analysis. It measures the amount of NO3 in the cytoplasm, including NO3 contained in any subcompartments such as the endoplasmic reticulum (ER) or associated with putative NO3 binding proteins (Siddiqi et al. 1991), that is readily exchangeable with the cytosol. In all reported studies, NO3 concentration of the cytoplasm was then calculated by measuring (or assuming) the cytoplasmic volume and presuming that there is no subcompartmentation of cytoplasmic NO3 pools. If the total quantity of cytoplasmic NO3 consisted of a homeostatically maintained cytosolic pool and a variable ER or plastidic pool (Winter et al. 1993), the calculated value of cytoplasmic NO3 would be the average of these two pools. In the compartmental analysis studies, cited in Table 1, cytoplasmic volumes were not actually measured; rather, a figure of 5 or 10% of the total cell volume was used. Obviously, as Miller & Smith (1996) pointed out, the cytoplasmic concentration would be doubled if a figure of 5% were used instead of 10%, as is illustrated in the findings of Siddiqi et al. (1991) and Devienne et al. (1994b).

There are two possible sources of error in determinations of cytoplasmic [NO3] by compartmental analysis: (1) the measurement is based on tracer efflux. If the effluxing N isotope includes any other labelled N species than NO3, the measurements will be confounded. However, Siddiqi et al. (1991) and Kronzucker et al. (1995a) have demonstrated that the effluxing species was indeed 13NO3 (>98% purity). (2) Any significant contribution of 13NO3 effluxing from the vacuole would introduce a source of error in the calculation of cytoplasmic [NO3]. However, the vacuolar half-life for tracer exchange in cereal roots, determined by use of 15NO3, was 7 h compared with the cytoplasmic half-life of 4·6 min (e.g. Devienne, Mary & Lamaze 1994a). In studies using the 13N isotope, roots were loaded with isotope for ∼ 35–40 min (Lee & Clarkson 1986; Siddiqi et al. 1991; Kronzucker et al. 1995a; Min et al. 1999). Therefore, the specific activity of 13NO3 in the vacuole was extremely low, and its contribution to 13NO3 efflux from the cytoplasm correspondingly small.

Miller & Smith (1996) raised ‘another problem with compartmental efflux analysis is that the labelled nitrate present in the cytosol is continuously being assimilated and also being transported to the xylem and so is no longer available for efflux’. In the same vein they state that ‘removal of nitrate from the cell can result in continuous changes in specific activity of nitrate pools’. Neither of these criticisms is valid. It is acknowledged that tracer efflux and cytoplasmic-specific activity decline with time during the washout period. However, tracer efflux during washout is plotted against time in a semi-log form and the regression line for this flux is extrapolated to the beginning of the washout procedure (t = 0). Thus, efflux and other parameters are obtained at t = 0 when the specific activity of the cytoplasm was that of the external solution, by virtue of labelling with 13NO3 for more than five half-lives of the cytoplasmic tracer exchange. In addition, tracer fluxes to shoot and assimilation plus vacuolar storage are separately determined/calculated (Lee & Clarkson 1986; Siddiqi et al. 1991; Kronzucker et al. 1995a; Min et al. 1999).

Cell fractionation techniques also measure the amount of NO3 rather than concentration in various subcellular compartments (Martinoia et al. 1986, 1987; Winter et al. 1993, 1994). Volumes of these compartments were also measured in the epidermal and mesophyll cells of barley (Winter et al. 1993) and spinach (Winter et al. 1994) leaves. These studies clearly demonstrate that relative volumes of the subcellular compartments differ from one species to the other and from one tissue to the other. For example, relative cytoplasmic (cytosol + nuclei) volumes in the epidermal cells of barley and spinach leaves were 1·2 and 8·7%, respectively, and in the mesophyll cells, the cytosolic volumes were 9·7 and 5·4%, respectively. The most important take-home message therefore is that one must be careful in comparing the reported absolute NO3 concentrations across different systems and/or different organs and/or different tissues, if the values of subcellular volumes are not available. Notwithstanding this caveat, relative changes in estimated values of cytoplasmic/cytosolic [NO3] within the same plant system maintained at different external NO3 concentrations, as in a number of studies cited in Table 1, remain valid, irrespective of the absolute values.

A possible explanation for the disparity in literature values for cytoplasmic/cytosolic [NO3]

Table 1 shows that estimates of cytoplasmic/cytosolic cytoplasmic/cytosolic [NO3] vary from 0·01 to 100 mm. Even in the same system (e.g. barley roots), reported estimates of cytoplasmic/cytosolic [NO3] show ∼ 12- to 50-fold differences (compare, e.g. the data of King et al. 1992; Zhen et al. 1991 and van der Leij et al. 1998; with those of Lee & Clarkson 1986 and Siddiqi et al. 1991). This indicates that methodological differences (including growth conditions), more than plant differences, are responsible for this lack of consensus. Indeed, a close examination of Table 1 shows that, within the same plant system, lower estimates of [NO3] are consistently associated with the group 1 techniques (NRA-based and microelectrode) whereas higher estimates are equally consistently associated with the Group 2 techniques (CAE and cell fractionation). The ­latter will include cytosolic NO3 as well as NO3 pools associated with organelles such as ER, plastids and vesicles; these are rapidly exchangeable with the cytosol (with t0·5 of exchange ≤ cytoplasmic t0·5). Models based on more than one distinct cytoplasmic nutrient pool, i.e. cytosol and ER (MacRobbie 1970) and/or plastids (Walker & Pitman 1976) have been proposed earlier (for a discussion of these models, see Lüttge & Higinbotham 1979). We hypothesize that the higher estimates of cytoplasmic [NO3] generated by group 2 techniques might result from the inclusion of these organellar pools; the latter would be invisible to estimates based upon group 1 techniques. We offer the following observations and arguments in favour of our hypothesis regarding the existence of organellar pools of nutrients in general and of NO3 in particular:

  • 1Winter et al. (1993, 1994) showed that the amount of NO3 associated with chloroplasts in the mesophyll cells of spinach and barley leaves (0·59 and 3·4 µmoles g−1, respectively; see also the data of Schröppel-Meier & ­Kaiser 1988a,b for spinach chloroplasts) was even greater than the amount in the cytosol (0·43 and 2·9 µmoles g−1, respectively). To illustrate the impact of inclusion of ­plastidic NO3 in the cytosolic pool, let us consider the data for spinach leaf. In this example, cytosol (5·4% of the cell volume) contained 8 mm NO3 (i.e. 0·43 µmoles NO3 g−1 of fresh weight). If we add together the amount of NO3 in cytosol and chloroplasts (0·43 + 0·59 = 1·02 µmoles) and consider it as all coming from the cytosol, then the [NO3] becomes 19 mm. This example is meant to show that the contribution from the organelles might result in significant over-estimation of cytosolic ion concentrations generated by CAE.
  • 2Given the proposed role of ER in symplastic ion transport, it is reasonable to expect that this organelle should contain ions destined for transport to the stele. The presence of several ATPases associated with ion transport and ion channels in the ER and other organelles (e.g. Bush & Sze 1986; Kawata & Yoshida 1988; White 1995) are indicative of ion transport activity across the membranes of these organelles. In the case of Ca2+, it is known that ER, plastids and mitochondria all serve as storage pools within the cytoplasm. Although the major storage pool is the vacuole, these rapidly exchanging storage pools within the cytoplasm (particularly the ER) appear to be essential to maintain [Ca2+] in the cytosol at such low concentrations as 200–300 nm (Klusener et al. 1995; Persson et al. 2001). Interestingly, Macklon (1975) used compartmental analysis to estimate cytosolic Ca2+ in onion roots and obtained an anomalously high value of 10·2 mm. Such an overestimate would readily be obtained if subcompartments for Ca2+ were exchanging rapidly with cytosolic Ca2+. We recognize that compartmentation of NO3 may differ in detail from that of Ca2+. However, the crucial issue to be derived from this comparison is that compartmental analysis failed to differentiate cytosolic Ca2+ from organellar Ca2+ (Macklon 1975; Siddiqi & Glass 1984) resulting in a substantial over-estimation of cytosolic Ca2+ (actually nM) to mM values.
  • 3Chemical localization studies have shown the presence of thallium and iodide (used as analogs for potassium and chloride, respectively) in the ER (van Iren & van der Spiegel 1975). Interestingly, in barley roots, measurements of cytosolic [K+] (∼75 mm) by microelectrodes (Walker, Leigh & Miller 1996) were also substantially lower than estimates of cytoplasmic [K+] by CAE (Memon, Saccomani & Glass 1985). This may indicate the presence of an organellar K+ pool within the cytoplasm, which CAE failed to detect. Indeed, high chloroplastic K+ (∼ 200 mm) has been demonstrated in spinach (Robinson & Downton 1984; Schröppel-Meier & Kaiser 1988a, b).
  • 4Cytosolic [NO3] measured directly by group 1 techniques (NRA, microelectrodes) is always substantially lower than the estimates by the group 2 techniques (CAE, cell fractionation) in the same system. For example in barley roots, cytoplasmic [NO3] reported by the same laboratory using two different methods differed by a factor ∼10 (Table 1); estimates by CAE (Siddiqi et al. 1991) were ∼10-fold higher than measured by NRA (King et al. 1992). We emphasize that high mM estimate of cytoplasmic [NO3] by CAE is not an artifact of the technique per se, for in coniferous species, white spruce, lodgepole pine and Douglas-fir, <,0·4–1 mm cytoplasmic [NO3] were obtained using CAE (Kronzucker et al. 1995a; Min et al. 1999). Similarly, in spinach and barley leaves, cytosolic [NO3] estimated by cell fractionation (group 2 technique) was 8–30 mm (Winter et al. 1993, 1994), whereas that measured by NRA in the leaves of several species was ≤ 0·1 mm (Robin et al. 1983).
  • 5In the same system (barley roots), reported cytosolic [NO3] values, measured by two different laboratories using two different group 1 techniques, NRA (King et al. 1992) and microelectrodes (Zhen et al. 1991), were similar (Table 1).

Evaluation of a putative cytosolic [NO3] homeostat

Unlike most studies, cytosolic [NO3] measured by NO3-specific microelectrodes (Zhen et al. 1991; Miller, Walker & Smith 1995; Miller & Smith 1996; van der Leij et al. 1998) appears to be constant, independent of external NO3 supply (Table 1). Miller & Smith (1996) have argued that the constancy of cytosolic [NO3] is required particularly in connection with xylem loading in the roots, but also perhaps to avoid the potential toxic effects of high nitrate concentration.

Our hypothesis regarding the existence of separate cytosolic and organellar pools within the cytoplasm is adequate to reconcile the conflicting reports in the literature regarding the variability of cytoplasmic [NO3] estimated by CAE versus the constancy of cytosolic [NO3] measured by microelectrodes (Miller & Smith 1996). It is plausible that a constant cytosolic [NO3] is maintained by a variable organellar NO3 pool (see above). Notwithstanding this hypothesis, and notwithstanding our acceptance of the data resulting from the use of nitrate-specific microelectrodes, a number of observations and implications are difficult to reconcile with the notion of a constant cytosolic [NO3] without a significant revision of currently accepted concepts. These are outlined as follows:

  • 1In contrast to the situation for ions such as K+, H+ and Ca2+, whose essential roles can not be substituted within the cytoplasm, a priori, there appears to be no specific requirement to maintain cytosolic NO3 at a prescribed concentration. Most plants are typically opportunistic, able to absorb and assimilate NO3, NH4+, or even amino acids (e.g. Kronzucker, Siddiqi & Glass 1997 and references therein). Under field conditions most plants typically obtain their required N in the forms of NO3 or NH4+. Because the proportions of these N forms differ substantially seasonally and geographically, a variable uptake and delivery of NO3 to the shoot would appear, a priori, to be the norm. Thus we question the argument that the importance of a constant cytoplasmic NO3 is to sustain NO3 delivery to the shoot (Miller & Smith 1996). Furthermore, should cytosolic [NO3] rise to some value equal to or greater than the proposed 2–5 mm there are no obvious detrimental implications for NO3 reduction; indeed even at 5 mm NO3 most NR enzymes are already saturated with respect to [NO3] (Guerrero, Vega & Losada 1981).
  • 2Measurements by CAE showed that in response to altered external ion concentration, whereas cytoplasmic [NO3] varied from more than three-fold in the roots of barley (Siddiqi et al. 1991) to more than 18-fold in lodgepole pine (Min et al. 1999), cytoplasmic [K+] was maintained within narrow limits in the roots of barley (e.g. Memon et al. 1985). For both K+ and NO3, influx, efflux, flux to xylem, and flux to ‘vacuole + metabolism’ of these ions increased with increasing external concentrations (Siddiqi & Glass 1983a, b; Memon et al. 1985; Siddiqi et al. 1991; Kronzucker et al. 1995a; Min et al. 1999). These flux changes must result in changes in the rates of turnover (t0·5 of exchange) and/or cytoplasmic concentrations (Britto & Kronzucker 2001). Interestingly, K+ and NO3 responded differently. In barley roots, depending on the variety, cytoplasmic [K+] was either maintained or increased by 20–30% as [K+]o increased from 10 to 100 µm. Under the same conditions the t0·5 for cytoplasmic K+ exchange declined (signifying an increased turnover) from an average value of 74 min for the three varieties to 35 min (Memon et al. 1985). It is considered that maintenance of cytosolic [K+] within a narrow limit is an obligate requirement for protein synthesis (Wyn Jones, Brady & Speirs 1979). It will be intuitively obvious that if component fluxes increase the turnover must also increase if pool size is to be held constant. Conversely if turnover rate remains constant then pool size must increase. In the case of NO3, t0·5 for cytoplasmic NO3 was constant over a wide range of [NO3]o, whereas cytoplasmic [NO3] increased ∼ 2-fold with every 10-fold increase in [NO3]o (Siddiqi et al. 1991; Britto & Kronzucker 2001). First, these results demonstrate that the method of compartmental analysis is able to recognize the constancy of cytoplasmic K+, a case that has been confirmed by different methodologies. Second the interactions between fluxes, t0·5-values and pool sizes make it difficult to reconcile constancy of both pool size and t0·5 when component fluxes increase to the extent reported. Using CAE, the same pattern (i.e. variation in cytoplasmic [] with varying []o) has been observed with other ions which are metabolized and are structural components of the plants, e.g. NH4+ (Wang et al. 1993; Kronzucker, Siddiqi & Glass 1995b; Min et al. 1999) and SO42– (Cram 1983). A consequence of the observed constancy of t0·5 of exchange in these instances (e.g. NO3, NH4+) is a variable cytoplasmic [NO3] or [NH4+]. Interestingly, when MSX was used to block NH4+ assimilation in maize roots (Lee & Ratcliffe 1991) there was no indication of an NH4+ homeostat and cytosolic [NH4+] rose from 8 to 80 mm, notwithstanding the potential toxicity of such high values of cytoplasmic [NH4+]. Clearly the constancy of t0·5, a parameter representing the interactions of a large number of fluxes and pool sizes, must arise through proportional changes in these parameters.
  • 3Estimation of cytosolic [NO3] in barley roots based upon in vivo NRA and in vitro measurements of Michaelis–Menten constants demonstrated that cytosolic [NO3] increased from <1 mm at 0–2 mm[NO3]o to ∼ 1·7 mm at 5 mm[NO3]o, and to ∼ 4 mm at 20 mm[NO3]o (Fig. 5 in King et al. 1992).
  • 4King et al. (1992) showed that both in vivo and in vitro NRA of barley roots responded hyperbolically to the [NO3] of the incubation medium, saturating at ∼ 10 mm (Fig. 3 in King et al. 1992) and ∼ 20 mm[NO3]o (Fig. 2 in King et al. 1992), implying that changes in NRA were driven by changes of cytosolic [NO3].
  • 5There is clear evidence that the rapidity and/or extent of induction of NRA and high-affinity NO3 influx and the corresponding gene transcripts are increased with increasing [NO3]o and root [NO3] in barley roots (Siddiqi et al. 1989; Öhlen et al. 1995; Vidmar et al. 2000). Likewise there are many reports demonstrating that fluxes of NO3 from cytosol to cell wall, to the vacuole and to the xylem increase with increasing [NO3]o and concomitant increases of tissue [NO3] (Andrews 1986; Teyker et al. 1988; Siddiqi et al. 1991; Peuke, Hartung & Jeshke 1994; Devienne et al. 1994b; Kronzucker et al. 1995a; Sivasankar, Rothstein & Oaks 1997; Min et al. 1999). Although it can be argued that changes of flux may be achieved by either changing the driving force and/or the conductance of the pathway, the same argument can not be advanced to explain changes of transcript abundances.
  • 6The argument offered by Miller & Smith (1996) regarding the maintenance of cytosolic [NO3] at ∼ 4 mm for passive xylem loading in the roots is unconvincing. Calculations according to the Nernst equation at membrane potentials ∼ −75 mV (Miller & Smith 1996) show that the electrochemical potential gradient will be downhill from cytosol to the xylem even when cytosolic [NO3] is as low as ∼ 0·6 mm. If xylem loading were determined only by the electrochemical potential (Δµcx) for NO3, increasing cytosolic [NO3] and/or more negative membrane potentials would facilitate xylem loading.

In conclusion, we suggest the following:

  • • There may be more than one pool of NO3 (and other ions) within the cytoplasm, e.g. cytosolic, plastidic, ER, and/or putative NO3 binding proteins.
  • • Disparity in cytoplasmic/cytosolic [NO3] in the literature appears to be related to the techniques used, because some techniques may measure the total cytoplasmic NO3 pool, which may include organellar pools, whereas others specifically measure the cytosolic pool.
  • • The terms, cytoplasm and cytosol, represent two different entities and should not be used interchangeably, as sometimes is the case. In particular, the symbol ‘[NO3]cyt’ is commonly used for both, cytoplasmic and cytosolic [NO3].    We    suggest    the    adoption    of    the    symbols ‘[NO3]cp’ and ‘[NO3]cs’ for cytoplasmic [NO3] and ­cytosolic [NO3], respectively. Indeed, the same general convention must be extended to distinguish between the cytoplasmic and cytosolic concentrations of other solutes as well.
  • • If cytosolic [NO3] is held constant under conditions wherein [NO3]o and tissue [NO3] is varied, the reported increases of transcript abundance for NR and NRT2 transporters is particularly problematic.
  • • Although valid arguments can be made for homeostatic regulation of [K+], [Ca2+] and [H+], given the diverse availability of various forms of N, there appears to be no such equivalent rationale for the case of cytosolic [NO3].

Received 2 May 2002;accepted for publication 6 May 2002