Effects of FA and HA on root growth
In the absence of Al addition, root growth in the E-FA40 treatment was greater than that in the comparable Nil-OA, HA40 or FA40HA40 (P<0.05). With the addition of Al to 30 µM there was an apparent, but not significant, enhancement of root growth in the FA40. However, root growth in the HA40 increased substantially with the addition of Al up to 30 µM to the point where it was similar to that in the FA40, whilst root growth in the FA40HA40 was intermediary between the FA40 and HA40 responses. In each of these treatments, monomeric Al concentrations were low, in the range of 0-6.5 µM, and OC concentrations were stable and consistent with the concentrations imposed for each of the OA treatments. The stability in monomeric Al and OC concentrations suggests that the enhanced beneficial effect of the E-HA40 with increasing Al was related to the complexation of inhibitory components in the E-HA. The 6.5 µM monomeric Al in the FA40 at 30 µM treatment Al may have provided the apparent enhanced root growth.
Stimulation of plant growth by FA and HA is well documented (Linehan 1976; Rauthan and Schnitzer 1981; Schnitzer and Poapst 1967), although the mechanism by which stimulation occurs is not well understood. However, a positive impact of OA is not universal. Brunner et al. (1996) found that the high molecular weight fractions (>10,000 Daltons) of an aqueous chestnut (Castanea sativa) leaf litter extract inhibited barley (Hordeum vulgare) growth at concentrations of only 5 mg C L-1 yet the fraction below 1000 Daltons did not. The co-addition of the H-FA with the H-HA (H-FA40HA40) was less inhibitory than was either component alone, suggesting the two components interacted in affecting root growth. This occurred despite the fact the OC concentration in the H-FA40HA40 was twice that of either the H-FA40 or H-HA40 treatments. There was no interaction between the E-FA and E-HA in their effects on root growth. Only limited research has been conducted on interactions of FA and HA, however, Ernst et al. (1987) found that combinations of FA and HA stimulated plant growth to a greater extent than did FA or HA alone.
Confounding effects of monomeric Al and organic acids on root growth
In the Nil-OA treatments, for both the Eucalyptus and hay experiments, the fitted Gaussian functions for the relationship between root growth and monomeric Al mirrored each other validating experimental comparisons on a relative root length basis. At monomeric Al concentrations up to 10 µM root growth was enhanced compared with the 0 µM whilst at higher Al concentrations root growth decreased due to Al toxicity. Note that the Eucalyptus relationship is displaced upward relative to the Nil-OA due to the stimulatory effect of Eucalyptus OA on root growth, while the hay relationship is displaced downward relative to the Nil-OA due to the inhibitory effect of hay OA. Interestingly, in the E-HA40 treatment at 90 µM treatment Al, the monomeric Al concentration was 11.9 µM and OC was 0 mg C L-1 and relative root length sat neatly on the fitted function of RRL to monomeric Al in the absence of OA. In this treatment it appeared that the stimulatory effect observed was a result of the small concentration of monomeric Al rather than the E-HA.
Where greater than about 10 mg C L‑1 remained in solution in the E-FA treatments, the stimulatory effect of the FA was expressed. The E-FA stimulated root growth despite the presence of an otherwise toxic concentration of monomeric Al. However, care should be taken in considering that the FA was able to override the effect of Al toxicity as the accuracy of method of measuring monomeric Al concentration requires further consideration. It appeared the stimulatory effect of the FA had over-ridden the toxic effect of monomeric Al since the relative root growth response. In the presence of OA greater than ≈10 mg C l-1, monomeric Al was not a good index for plant root growth response to Al toxicity. The result of which is a displacement of the OA curves in Figure 6 to the right relative to the Nil-OA curve. However, the presence of only a small amount of un-precipitated FA or HA (<about 5 mg C L‑1) did not provide a stimulatory effect on root growth and the inhibitory effect of monomeric Al on root growth prevailed.
With the addition of Al at concentrations greater than 30 µM, root growth in the presence of E-FA and E-HA treatments decreased substantially since precipitation of the FA and HA occurred (Figure 2), reducing the OA stimulation of root growth, and there was an overriding inhibitory effect of monomeric Al (Figure 6a). Similarly in the hay OA treatments, comparable precipitation of OA (Figure 2) resulted in the removal of the inhibitory effect of the OA and the remaining inhibitory effect is attributed to Al toxicity. Thus at higher Al concentrations the effect of OA on root growth (either stimulatory or inhibitory) is removed, and the remaining effect is attributable to Al toxicity alone, with both experiments producing comparable responses (Figure 6a). The effect of H-FA and H-HA shifted from an inhibition of root growth to stimulation of root growth with increasing treatment Al concentration.
In both OA treatments, the effect of monomeric Al on root growth (initial stimulation then toxicity) is confounded with the direct effect of the OA on root growth (stimulation for Eucalyptus and inhibition for hay). As a means of resolving these two confounded effects, the RRLs were expressed relative to the relevant OA treatment at 0 µM Al (RRLOA) (Figure 6b) as opposed to the Nil-OA at 0 µM Al (Figure 6a). For the hay treatments addition of Al initially reduces the inhibitory effect of the OA, producing an initial increase in RRL compared to the Nil-OA treatment. As the OC concentration of the solutions remains constant up to 30 µM Al, this effect is attributed to complexation of the OA with Al denaturing the OA and detoxifying it, rather than coagulating and precipitating it as occurs at higher Al concentrations. At Al concentrations of 30 µM and above, any positive growth effect of denaturing the inhibitory OA is outweighed by the toxic effect of increasing monomeric Al concentrations. For the Eucalyptus treatment, low rates of Al appear to cause similar denaturation of the OA, but in this instance reducing its stimulatory effect on root growth, with the effect of lowering the response curve below that of the Nil-OA at low Al concentrations.
Though distinct differences can be observed in the organic components of the Eucalyptus and hay FA and HA (Harper et al. 2000) it is not clear which components would be responsible for the stimulatory or inhibitory effects. The 13C NMR spectra for the hay FA and HA (Harper et al. 2000) shows the presence of a strong peak at 147 PPM in the FA and HA spectra and this equates to aromatic diphenols, which are likely to cause phytotoxic effects. These groups would also be likely to complex with Al, and thus their inhibition of root growth would be expected to be altered as Al is added to the solution. Various authors have demonstrated the co-addition of toxic concentrations of soluble cations with toxic organic ligands results in no inhibition to plant growth (Suthipradit et al. 1990; Tadano et al. 1992). Presumably in the present study, non-toxic metal ion (Al) and organic ligand complexes had formed.
Mechanisms for reducing monomeric Al
In both experiments, at treatment Al concentrations of 10 and 30 µM, the concentration of OC and total Al were equivalent to that nominally imposed yet the monomeric Al concentrations in the presence of FA and HA (Eucalyptus or hay) were only 0-2.3 µM in the 10 µM Al treatment and 0-9.7 µM in the 30 µM Al treatment. Irrespective of the OA type, at treatment Al concentrations up to 30 µM complexation of Al was the principal mechanism by which monomeric Al concentration was reduced. By virtue of their greater OC concentration the FA40HA40 treatments (Eucalyptus and hay) in the presence of 30 µM treatment Al had lower monomeric Al concentrations than the comparable FA40 and HA40 treatments (Fig 3a).
As Al addition increased (90 and 270 µM treatments) Al complexation ceased to be the sole mechanism for reducing monomeric Al concentration. In the E-FA40 treatment, in the presence of 90 µM Al, both total Al and OC concentration were substantially lower than that nominally imposed and there was a considerable discrepancy between total and monomeric Al concentrations (Figs 1 and 3). The HA in the E-HA40 and H-HA40 treatments, precipitated from solution at a treatment Al concentration of only 90 µM. This data in conjunction with the observation of a precipitate lining the pots, indicated co-precipitation of Al with HA had removed all HA from solution.
In contrast to the HA, FA in the FA40 and FA40HA40 treatments for both the Eucalyptus and hay was less prone to precipitation and higher total and monomeric Al concentrations were observed. At 90 µM treatment Al concentration, the total Al concentration in the E-FA40 and H-FA40 (45.1 and 77.6 µM respectively) and H-FA40HA40 (102.0 and 87.1 µM respectively) treatments were greater than that in the E-HA40 (19.2 µM) and H-HA40 (31.7 µM) indicating a greater capacity for the FA to hold Al in solution. Consistent with this observation, the OC concentrations in the E-FA40 (13.7 mg l-1) and H-FA40 (37.4 mg l-1) were relatively high representing about 30% of that nominally imposed in the Eucalyptus and 90% in the hay. In contrast, no OC was present in the E-HA40 and only 30% was present in solution in the H-HA40 (13.6 mg l-1). The hay FA was largely not affected by precipitation reactions at this 90 µM treatment Al concentration whilst the Eucalyptus FA was moderately affected. At 90 µM treatment Al, in the FA40 and HA40 treatments, for both the Eucalyptus and hay, complexation and precipitation reactions were simultaneously reducing monomeric Al concentration to varying degrees. However, in the H-FA40 in the presence of 90 µM Al, both total Al and OC concentrations (Figs 2 and 4), though reduced, were similar to that nominally imposed but there was a considerable discrepancy between total and monomeric Al concentrations (Fig 6) indicating complexation was the dominant mechanism for reducing monomeric Al. This result is consistent with the strong 13C NMR spectra peaks for the H-FA40 at 172 ppm representing carboxylic functional groups and 55 PPM for hydroxyl functional groups (Harper et al. 2000) the combination of which confers strong potential for metal ion complexation (dos Santos et al. 2020). A similar result was recorded by Hiradate et al. (2006) that showed a strong complexation reaction between FA and Al was observed where 13C NMR spectra showed strong peaks at 172 ppm (carboxylic functional groups) and 55 ppm (hydroxyl functional groups).
In the E-FA40, E-HA40 and E-FA40HA40 and H-HA40 treatments in the presence of 270 µM Al, almost all of the OA had precipitated from solution and in the H-FA40 and H-FA40HA40 the OC was substantially reduced. This indicated that in the presence of a still higher Al concentration, precipitation supersedes complexation as the most important mechanism in reducing monomeric Al concentration, which is demonstrated with other metal ions with FA and HA when the metal ion activity is high (Chirenje et al. 2002). Hiradate et al. (2006) identify the mechanism for formation of FA-Al precipitates includes a complexation reaction between Al and carboxylic functional groups on the FA.
In both the E-FA40HA40 and H-FA40HA40 treatments, in the presence of 90 µM Al, the total Al and OC concentration remained equivalent to that nominally imposed and a large discrepancy existed between the monomeric and total Al concentrations. Under the higher OC concentration in these treatments, but the same treatment Al concentration (viz. 90 µM ), complexation remained the dominant mechanism for reducing monomeric Al concentration and precipitation was not an important mechanism. With a threefold increase in treatment Al concentration from 90 to 270 µM OC concentration in the E-FA40 and H-HA40 declined to a third of that recorded at 90 µM. In the E-FA40HA40 a 94% reduction in OC concentration occurred from 90 µM Al (OC=91.0 mg l-1) to 270 µM Al (OC=5.4 mg l-1). The comparable reduction in OC concentration for the H-FA40HA40 was only 69%.
In both the E-HA40 and H-HA40 treatments the HA had totally precipitated at 90 µM Al so further changes in solution composition with increasing treatment Al concentration for these treatments are comparable to those in the Nil-OA. However, the subsequent increases in total solution Al concentration for the E-HA40 (44 µM increase) and H-HA40 (15.3 µM increase), with a further increase in treatment Al concentration to 270 µM, were substantially less than that in the comparable Nil-OA treatments with a mean increase in the two experiments of 144.8 µM. This result indicates that further complexation or precipitation of Al to the existing Al-HA precipitate occurred under the subsequent increase in Al concentration and indicates a potential significant role of humin material in the amelioration of Al toxicity. This effect was also observed for the FA40 treatments with an increase in Al from 90 to 270 µM despite not all the OC being precipitated at 90 µM Al.
The solubility of the H-FA at high Al concentrations was greater than that of the H-HA and the E-FA and E-HA. Importantly, the 13C NMR spectra for the H-FA had strongly developed carboxyl and hydroxyl peaks (Harper et al. 2000) and this arrangement is likely to favour complexation of metal ions (Saar and Weber 1982). The findings accord with those of Schnitzer and Skinner (1963) who noted that, at a solution ratio of 1M Al to 1M of a high molecular weight organic extract (viz. FA or HA) (about 115 µM Al to 40 mg C L-1), complexation was the most important reaction and the complex was water soluble (viz. it did not precipitate). In the present experiments, however, a ratio of only 90 µM Al to 40 mg C L-1 resulted in substantial precipitation of organic C supplied as HA and to a lesser extent FA. Schnitzer and Skinner (1963) also showed that, at higher ratios of Al to C, a precipitate of Al with OA formed. Suthipradit et al. (1990) showed that at a ratio of about 31 µM Al to 40 mg C L-1 (as FA), complexation was the only apparent mechanism for reducing monomeric Al.
The capacity of OAs to complex and precipitate Al is dependent not only on the broad acid type (viz. FA or HA), but also on the source from which these are extracted. The differences in complexation and precipitation reactions between the two sources of FA and HA relate to the considerable differences in their structure as shown by 13C NMR and SEC spectra for these acids (Harper et al. 2000). These findings demonstrate a fundamental broad difference in the reactivity of FA and HA with solution Al and further highlight a differential in solubility of the Al-OA complex dependent on the source from which they were extracted. At low Al concentrations (10 and 30 µM) both sources of HA more effectively complexed Al than did FA, while the hay FA and HA were poorer complexers of Al than the comparable Eucalyptus FA and HA. However, at higher treatment Al concentrations the hay FA and HA possessed a greater capacity for complexing Al than the Eucalyptus FA and HA for which precipitation was favoured. The FA of both sources were less affected by precipitation than HA and the hay FA was less prone to precipitation than the Eucalyptus FA.
Implications for measurement of toxic monomeric Al in soil studies
This solution culture study highlights the complex role of high molecular weight OAs in the expression of Al toxicity on plant root growth. In acidic soil studies, plant root elongation is often related to various soil properties, in particular toxic monomeric Al, so as to develop criteria for assessing Al toxicity (e.g. Bruce et al. 1988; Close and Powell 1989; Menzies et al. 1994; Wright et al. 1989). Studies such as these often show poor coefficients of determination. Menzies et al. (1994) in their study discriminated against the effects of non-toxic Al-organic complexes, which gave a better correlation. In our study it was shown that in the presence of non-precipitated high molecular weight OAs (i.e. FA and HA) monomeric Al was not a good index of plant response to Al toxicity.
In some treatments the presence of toxic monomeric Al concentrations (>20 µM) did not inhibit root growth to the extent anticipated. As the effects of Al and the organic acids are both confounded and interactive, it is not possible to fully isolate the respective effects. Thus, our discussion of the results obtained in this experiment considers several possible explanations, including the potential for both positive and negative direct effects of OA on root growth, interaction with Al causing detoxification of the Al and reduction of the direct effects of the OA, and the potential for OA complexation of Al to produce inaccurate estimation of monomeric Al concentration by the kinetic colorimetric approach used.
Firstly considering the direct effects of OA and their interaction with Al, in the absence of added Al the effect of the FA and HA was either negative as for the hay FA and HA or positive as for Eucalyptus FA and HA). Increasing the solution Al concentration altered the direct effect of FA and HA on root growth and the magnitude of this effect varied depending on the source of FA and HA. For the Eucalyptus FA and HA enhanced root growth occurred in the absence of Al and the beneficial effect increased slightly with an increase in Al concentration. However, the hay FA and HA inhibited root growth in the absence of Al. Indeed, the inhibitory effect on root growth of the hay FA and HA per se was equivalent to that of Al toxicity at a monomeric Al concentration of 20-25 µM. Further addition of Al to the hay FA and HA may have resulted in a shift in the direct effect of OA on root growth from inhibition to enhancement of growth (relative elevation of hay FA and HA treatments in Figure 6b).
In acidic soils the direct impact of soluble organic ligands on root growth and complexation of Al complicates the understanding of acidic soil infertility and particularly Al toxicity effects. In degraded acidic soils with low levels of organic matter and high soil solution Al concentrations, inorganic soil solution chemistry principles will prevail and most Al in soil solution will be present as monomeric Al. Various authors (e.g. Adams and Hathcock 1984; Evans and Kamprath 1970) have demonstrated that for soils low in soluble OC, plant growth is well correlated with measures of solution Al concentration, but for soils with higher solution OA concentrations the effect of Al complexation and detoxification must be considered (Hue et al. 1986; Menzies et al 1994; Zhao et al. 2020). In acidic soils amended with organic materials, soluble OAs may be present in soil solution at extremely high concentrations, up to 500 mg C L-1 (Kerven et al. 1995). In this study, monomeric Al was estimated by the rate of reaction of a colorimetric reagent with Al in the treatment solutions; Al bound to organic ligands reacts more slowly than free Al, permitting discrimination. In the method used, Al-malate is the Al-organic complex used for calibration. If an organic ligand bound Al less strongly than malate, and hence reacted with the colorimetric reagent more rapidly, but nevertheless reduced the toxicity of the Al, the method would overestimate the concentration of toxic monomeric Al. This would result in a displacement of points to the right in Figure 6. Thus, the apparent high tolerance of plants in the hay FA and HA treatments may be attributable to a failure of the analytical method to accurately measure monomeric Al concentration.
In the present study concentrations of FA or HA at only about 10 mg C L-1 had direct effects on plant growth. Therefore, it is likely that their direct effect on plant growth will be important in studies where bioassays are used as a means of determining the relative toxicity of Al. Further research is required to refine the diagnostic criteria for Al toxicity in acidic soils, taking into consideration the effect of the organic component per se on plant growth. This direct effect could be either inhibitory or stimulatory depending on the type of OAs present and the crop species being grown. Additionally, the question of the accuracy of kinetic colorimetric methods of monomeric Al measurement should be considered.