Detoxification of Dissolved SO₂ (Bisulfite) by Terricolous Mosses

Abstract

• Background and Aims The widespread calcifuge moss Pleurozium schreberi is moderately tolerant of SO₂, whereas Rhytidiadelphus triquetrus is limited to calcareous soils in regions of the UK that were strongly affected by SO₂ pollution in the 20th century. The proposition that tolerance of SO₂ by these terricolous mosses depends on metabolic detoxification of dissolved bisulfite was investigated.

• Methods The capacities of the two mosses to accelerate loss of bisulfite from aqueous solutions of NaHSO₃ were studied using DTNB [5, 5-dithio-(2-nitrobenzoic acid)] to assay bisulfite, and HPLC to assay sulfate in the incubation solutions. Incubations were performed for different durations, in the presence and absence of light, at a range of solution pH values, in the presence of metabolic inhibitors and with altered moss apoplastic Ca²⁺ and Fe³⁺ levels.

• Key Results Bisulfite disappearance was markedly stimulated in the light and twice as great for R. triquetrus as for P. schreberi. DCMU, an inhibitor of photosynthetic electron chain transport, significantly reduced bisufite loss.

• Conclusions Bisulfite (SO₂) tolerance in these terricolous mosses involves extracellular oxidation using metabolic (photo-oxidative) energy, passive oxidation by adsorbed Fe³⁺ (only available to the calcifuge) and probably also internal metabolic detoxification.

INTRODUCTION

Comparative studies of pollution sensitivity among and within different plant groups have provided many insights into mechanisms of phytotoxicity and tolerance of the major atmospheric pollutants. As a group, bryophytes are highly sensitive to atmospheric pollution by sulfur dioxide (Rao, 1982; Winner et al., 1988; Bates, 2002). This sensitivity, comparable with that of lichens and generally greater than that of vascular plants, is usually attributed to the limited development of a cuticle in most species, and to their large surface areas facilitating pollutant absorption (Winner et al., 1988). Low metabolic activities of bryophyte cells together with modest innate growth rates have also been suggested as reasons for their high sensitivities to SO₂. Resistant bryophyte species are often characterized by relatively high growth rates and they perhaps have greater capacities to detoxify absorbed pollutants than sensitive species (Gilbert, 1970). However, little is known about the metabolic basis, if any, of observed differences in SO₂ tolerance among bryophytes.

In vascular plants, apart from the important pollutant-excluding effect of cuticle and stomata, tolerance towards SO₂ appears to be based on detoxification mechanisms. The pollutant is oxidized to the relatively harmless sulfate ion and/or reduced to sulfide (Legge and Krupa, 2002). Reduction of the pollutant is light-mediated, probably occurring within the chloroplasts, and with eventual loss of S as gaseous H₂S (Rennenberg, 1991; Rennenberg and Herschbach, 1996). However, this process accounted for <1 % of the SO₂-sulfur taken up by a range of tolerant and sensitive lichens studied by Gries et al. (1997). Oxidation of SO₂ probably involves expenditure of metabolic energy. The reaction depends on the formation in the light of the superoxide anion () by the photosynthetic electron transport chain, at least in isolated chloroplasts of spinach and Avena sativa (Asada and Kiso, 1973; Miszalski, 1991a). However, according to Pfanz et al. (1990), the oxidation of dissolved SO₂ (including ) actually occurs in the cell walls through the activity of apoplast peroxidases. It may also occur in the dark to some extent, through mitochondrial activity (Ballantyne, 1977).

Paradoxically, an important cause of SO₂ phytotoxicity is believed to be through intracellular production (Miszalski, 1991b). Plants with high SO₂-tolerance tend also to exhibit elevated activities of superoxide dismutase (SOD) (Miszalski and Niewiadomska, 1993b). However, as SOD catalyses decomposition of , it might be expected to inhibit rather than stimulate photo-oxidation of SO₂ within the cells of tolerant plants (Miszalski and Ziegler, 1992). The inhibitor diethyldithiocarbamate (DETC), which blocks SOD activity (Heikkila et al., 1976; Scarpa et al., 1991), provides a means of assessing the role of this enzyme in a plant's defences against SO₂ (Tanaka and Sugahara, 1980; (Miszalski, 1991b).

The only detailed report of SO₂ detoxification in bryophytes is for the isolated genus Sphagnum. A population of S. cuspidatum, growing in the polluted Southern Pennine Hills (UK) had greater tolerance to bisulfite than plants from a relatively unpolluted site in North Wales (Baxter et al., 1989). Baxter et al. (1991) showed that oxidation of bisulfite to sulfate in the external solution was brought about passively in the Pennine S. cuspidatum and S. recurvum plants by metal cations (Fe³⁺, Mn²⁺ and Cu²⁺) adsorbed onto the cell wall exchange sites and originating from industrial emissions. This effect was reproduced in the sensitive plants by experimental additions of the cations and banished in tolerant plants by removing the metal ions with the chelator EDTA. No metabolic component to the bisulfite detoxification was identified in the Sphagnum study.

The focus of this paper is on the detoxifying potential of terricolous bryophyte shoots incubated in dilute bisulfite solutions. Two robust moss species are compared. The strict calcifuge Pleurozium schreberi is moderately tolerant of SO₂ (Farmer et al., 1992). In contrast, Rhytidiadelphus triquetrus has become restricted to calcareous soils in formerly SO₂-polluted areas, although it is also capable of colonizing relatively acid soils in unpolluted regions (Bates, 1993). Bharali and Bates (2002) observed that photosynthesis in these mosses was strongly inhibited by short (2-h) incubations with bisulfite, but that progressive recovery occurred when the incubation was extended (4–10 h). The eventual recovery was corroborated by growth-chamber experiments in which weekly additions of sulfite to P. schreberi and R. triquetrus failed to inhibit elongation growth. The results suggest that the shoots have a high capacity to detoxify dissolved SO₂. The presence of Ca²⁺ ions in the apoplast of R. triquetrus and of Fe³⁺ in that of P. schreberi, derived from the respective underlying soils of each species or by experimental application, further ameliorated bisulfite toxicity, although Fe³⁺ addition also led to increased membrane leakiness (Bharali and Bates, 2004). Here, several related hypotheses are investigated: (a) tolerance of bisulfite by P. schreberi and R. triquetrus depends primarily on detoxification (oxidation) of the pollutant; (b) bisulfite detoxification involves metabolic energy; (c) Ca²⁺ and Fe³⁺ stimulate the detoxification process; (d) SOD is involved in bisulfite detoxification. The results are interpreted with reference to the contrasted soil preferences of P. schreberi and R. triquetrus.

MATERIALS AND METHODS

Plant material

Two robust, terricolous, pleurocarpous mosses, Pleurozium schreberi (Brid.) Mitt. and Rhytidiaelphus triquetrus (Hedw.) Warnst. were compared as they show contrasted responses to SO₂ pollution and soil reaction (Bharali and Bates, 2002, 2004). Pleurozium schreberi was collected from an acid, sandy loam soil under grassland and scrub at Silwood Park, Berkshire (grid ref. SU9468). Rhytidiaelphus triquetrus was collected from chalk grassland on a rendzina soil in the Chiltern Hills, Oxfordshire (grid ref. SU7093).

Incubation treatments

Samples consisting of ten healthy green 2-cm apices of P. schreberi or R. triquetrus were excised and initially washed (twice, 100 cm³ for 1 min each) with double-distilled water (DDW) to remove any particulate contamination. After discarding the washings, excess water on the tissue surfaces was allowed to soak into a sheet of Whatman No. 15 filter paper to minimize dilution of the bisulfite solution. All subsequent incubations were performed in 100 × 25 mm glass specimen tubes sealed with Parafilm and gently agitated on a mechanical shaker at 20 ± 2 °C. The final concentrations of bisulfite were determined in the incubation solutions immediately upon completion of each treatment (see below). Specific details for individual experiments follow.

Short-term incubation experiment

The shoot samples were incubated with 20 cm³ of freshly prepared aqueous NaHSO₃ solution (0·6 mm) for intervals up to 12 h in the light (72 µmol photons m⁻² s⁻¹) or in the dark.

Long-term incubation experiment

The shoot samples were incubated for periods up to 5 d with 20 cm³ of freshly prepared aqueous NaHSO₃ solution (0·6 mm). All samples were subjected to an 18 : 6 h light : dark regime. A replicated set of solutions lacking moss shoots was included to estimate the background rate of oxidation of bisulfite to sulfate.

Bisulfite disappearance in relation to initial concentration

Shoot samples were incubated (2 h) in 0·3, 0·6 or 1·0 mm NaHSO₃ solutions (20 cm³, 20 °C) in the light or in the dark.

Influence of pH of the incubation solution on bisulfite loss

Solutions (0·3 mm) of NaHSO₃ were buffered at pH 3·0, 4·0, 5·0 or 6·0 employing citric acid buffer (20 mm) and drop-wise additions of 70 % (w/v) aqueous NaOH whilst magnetically stirring (Ferguson and Lee, 1979). Shoot samples were incubated for (2 h, 20 °C) in 20 cm³ NaHSO₃ in the light.

DCMU experiment

DCMU [3-(3′,4′-dichlorophenyl)-1,1-dimethylurea; 0·6 mm] was included in the incubation medium (0·6 mm NaHSO₃) of P. schreberi and R. triquetrus to inhibit photosynthetic electron transport and oxygen evolution. Shoot samples were incubated (2 h, 20 °C) in 20 cm³ of this solution, in the light or in the dark.

DETC experiment

An inhibitor of superoxide dismutase, DETC (diethyldithiocarbamate; 25 mm), was included in the bisulfite solutions (0·6 mm) to determine whether the enzyme may play a role in bisulfite oxidation when moss shoots are incubated in the presence of the pollutant. Shoot samples were incubated (2 h, 20 °C) in 20 cm³ of solution, in the light or in the dark.

Bisulfite oxidation: influence of Ca²⁺ and Fe³⁺

To manipulate the concentrations of the cations Ca²⁺ and Fe³⁺ in the apoplasts of the moss shoots, samples were pre-treated by incubation with 20 cm³ CaCl₂ (0·6 mm), FeCl₃ (0·6 mM) or Na₂-EDTA (5·0 mm), for 2 h prior to incubation in 0·6 mm NaHSO₃ solution. A ‘control’ was provided by pretreatment in 20 cm³ DDW for the same period followed by incubation in 0·6 mm NaHSO₃ solution. At the end of the incubation period the sulfate concentration of the solution was determined as described below.

Bisulfite and sulfate determinations

The bisulfite concentration remaining in the incubation solutions was determined by the spectrophotometric method of Humphrey et al. (1970) in which thiosulfate forms when the sulfite ion reacts with an organic disulfide, DTNB [5, 5-dithio-(2-nitrobenzoic acid)], to displace the thiol anion (also see Bharali and Bates, 2004). Results are expressed in two forms: (1) absolute concentrations (mm), which show the net bisulfite loss but do not take into account the variations in tissue mass involved in any transformations of the ion; and (2) relative concentrations (mmol HSO_{3⁻ g⁻¹} d. wt), which relate the final bisulfite concentration to the oven dry weight of the incubated moss shoots.

In the experiments (see above) in which metal concentrations were altered in the extracellular environment of the moss shoots, there was a specific need to determine whether sulfite in the incubation solution was being oxidized to sulfate. Sulfate concentration was assayed by high performance liquid chromatography (HPLC) using a Waters ion chromatograph (Millipore Ltd, Watford, UK). An IC-PAK anion column and a sodium gluconate/sodium tetraborate decahydrate eluent at pH 8·5 and flow rate of 1·2 cm³ min⁻¹ were used. The HPLC results were also expressed both in absolute terms (mm SO₄²⁻) and corrected for shoot dry weight (mmol SO₄² g⁻¹ oven d. wt).

Statistical analyses

The experimental data were analysed by one-way ANOVA using the GLIM software package (version 3·77, update 1; Royal Statistical Society, London). Where a significant treatment effect was found, differences between treatment means were investigated with Duncan's multiple range test.

RESULTS

Disappearance of bisulfite during short-term incubations with moss shoots

The concentration of bisulfite in the incubation solution (initially 0·6 mm) declined in an approximately logarithmic fashion in the presence of shoots of either Pleurozium schreberi or Rhytidiadelphus triquetrus (Fig. 1). There was a significant effect of light on this process with a greater rate of bisulfite decay observed and lower final remaining concentration in the samples incubated in the light. There was also a significant difference between the two species, with R. triquetrus apparently possessing a greater capacity to catalyse the disappearance of bisulfite. These patterns were strongly evident in terms of absolute concentrations, but also observed when the results were corrected for shoot dry weights. However, only in the case of R. triquetrus incubated in the light for 10 h did bisulfite become completely undetectable.

Fig. 1.

Effects of short-term incubations with moss shoots on the disappearance of bisulfite from 0·6 mm sodium bisulfite solutions: (A) Pleurozium schreberi; (B) Rhytidiadelphus triquetrus. Bar charts show absolute concentration (mm) of bisulfite remaining in solution (open triangles, light; closed triangles, dark; mean ± 1 s.e., n = 3). Line graphs show relative bisulfite content of solution (mmol g⁻¹ d. wt).

Disappearance of bisulfite during long-term incubations

Bisulfite disappearance (presumably by oxidation to sulfate) in the absence of moss shoots was roughly linear, with 28 % remaining at the end of the fifth day (Fig. 2A). In comparison, approx. 95 % of bisulfite had disappeared from the medium after a 3-d incubation period with either moss species present (Fig. 2B). As previously, the relative bisulfite data show that the rate of disappearance of bisulfite was initially significantly higher for R. triquetrus than for P. schreberi.

Fig. 2.

Disappearance of bisulfite from 0·6 mm sodium bisulfite solution over 5 d with an 18 : 6 h (light: dark) regime: (A) in the absence of moss shoots; (B) with moss shoots present. In (B) the bar chart shows absolute concentrations of bisulfite remaining in solution (mean ± 1 s.e., n = 3), whereas line graphs show relative bisulfite content of solution (mmol g⁻¹ d. wt).

Disappearance of bisulfite in relation to initial concentration

The results in Fig. 3 show that disappearance of bisulfite varied in response to both the initial concentration and the presence of light. In P. schreberi the amount of bisulfite remaining in the incubation solution was roughly proportional to the initial bisulfite concentration and there was no significant effect of light. The proportion of the bisulfite that disappeared varied from 74 % (of 0·3 mm) to 57 % (of 1 mm) in the dark, and from 77 % (of 0·3 mM) to 63 % (of 1 mm) in the light. In R. triquetrus the loss of bisulfite increased 3- to 4-fold over the initial concentration range 0·3–1·0 mm. Therefore the proportion of the total bisulfite lost from solution varied little (dark, 72–60 %; light, 71–86 %) over the concentration range used. The stimulatory effect of light on bisulfite loss was very marked at 0·6 and 1·0 mm in R. triquetrus whether considering absolute or relative concentrations.

Fig. 3.

Disappearance of bisulfite from dilute sodium bisulfite solutions as a function of initial concentration in the presence of moss shoots: (A) Pleurozium schrberi; (B) Rhytidiadelphus triquetrus. Bar charts show absolute concentrations (mm) of bisulfite remaining in solution (mean ± 1 s.e., n = 3). Line graphs show relative bisulfite content of solution (mmol g⁻¹ d. wt).

Disappearance of bisulfite in relation to acidity

The quantity of bisulfite remaining in the solution following incubation with shoots of P. schreberi or R. triquetrus was strongly influenced by pH (Fig. 4). Increasing the pH from 3 to 5 units caused a small reduction in the final bisulfite concentration in both species when results were corrected for shoot mass but not in absolute terms. A much greater reduction in bisulfite was observed at pH 6 in both absolute concentrations and when corrected for differences in shoot dry weights.

Fig. 4.

Disappearance of bisulfite from 0·3 mm sodium bisulfite as a function of solution pH. Bar chart shows absolute concentrations of bisulfite remaining in solution (mean ± 1 s.e., n = 3). Line graphs show relative bisulfite content of solution (mmol g⁻¹ d. wt).

Effects of DCMU on disappearance of bisulfite

In the experiment summarized in Table 1, bisulfite persistence was significantly greater in the presence of DCMU under both light and dark. In incubations with P. schreberi, the final bisulfite concentration increased from 61 to 88 % (light) and from 72 to 88 % (dark) of the initial (0·6 mm) concentration, in response to presence of DCMU. The increases for the R. triquetrus solutions were from 26 to 60 % (light) and from 34 to 65 % (dark) upon inclusion of DCMU. When the results are adjusted to take into account the different dry weights of the shoots of the two species (Table 1), it is evident that approximately twice as much bisulfite remained in the incubation solutions of P. schreberi as in those of R. triquetrus.

Table 1.

Effect of the inhibitor DCMU (0·6 mm) on loss of bisulfite from the incubation solution (initial concentration 0·6 mm HSO₃⁻) containing shoots of Pleurozium schreberi or Rhytidiadelphus triquetrus

	Absolute bisulfite (mm)		Relative bisulfite (mmol g−1 d. wt)
Species/inhibitor regime	Light	Dark	Light	Dark
Pleurozium schreberi
Bisulfite alone	0·367a	0·429a	5·08a	8·22b
Bisulfite + DCMU	0·529b	0·530b	8·44b	9·20b
Rhytidiadelphus triquetrus
Bisulfite alone	0·157a	0·203a	2·48a	3·23a
Bisulfite + DCMU	0·361b	0·390b	4·52b	5·10b

Incubations for 2 h, in the light (72 µmol m⁻² s⁻¹), at 20 ± 2 °C.

Data are means of three replicates.

ANOVA: within a species and data type, figures followed by the same superscript letter are not significantly different at P < 0·05.

Effects of DETC on disappearance of bisulphite

Inclusion of DETC (Table 2) significantly inhibited bisulfite loss from the incubation medium in both species. In incubations with shoots of P. schreberi, the presence of DETC caused, approximately, a doubling of the concentration of bisulfite remaining in solution, in both light and dark, after 2 h. For R. triquetrus, the presence of DETC caused even greater proportionate increases (light, 3·4–4·1; dark, 2·6) in the bisulfite remaining in solution, although the absolute concentrations were lower than those remaining after incubation with P. schreberii.

Table 2.

Effect of the inhibitor DETC (25 mm) on loss of bisulfite from the incubation solution (initial concentration 0·6 mM HSO₃⁻) containing shoots of Pleurozium schreberi or Rhytidiadelphus triquetrus

	Absolute bisulfite (µm)		Relative bisulfite (mmol g−1 d. wt)
Species/inhibitor regime	Light	Dark	Light	Dark
Pleurozium schreberi
Bisulfite alone	327a	405a	3·87a	4·37a
Bisulfite + DETC	630b	642b	7·76b	8·69b
Rhytidiadelphus triquetrus
Bisulfite alone	152a	241a	1·34a	2·03a
Bisulfite + DETC	625b	647b	4·53b	5·30b

Incubations for 2 h, in the light (72 µmol m⁻² s⁻¹), at 20 ± 2 °C.

Data are means of three replicates.

ANOVA: within a species and data type, figures followed by the same superscript letter are not significantly different at P < 0·05.

Evidence for extracellular oxidation of bisulfite to sulfate: influence of Ca and Fe

The results in Table 3 show consistently higher sulfate formation during incubation of the bisulfite solutions with Rhytidiadelphus triquetrus than with Pleurozium schreberi, especially after correction for shoot weights. In the calcicolous R. triquetrus, the pretreatment with Fe resulted in a small but significant increase in sulfate above the relatively high concentrations detected in the controls (DDW-pretreated). Pretreatment with CaCl₂ led to a marked reduction in sulfate production in the bathing solution of the R. triquetrus shoots. Na₂-EDTA pretreatment also caused a reduction in sulfate generation (absolute concentration) though not in terms of relative contents. For the calcifuge P. schreberi, there was no difference between the DDW or CaCl₂ pretreatments, but significantly more sulfate was detected in the incubation medium of FeCl₂-treated shoots. The Na₂-EDTA pretreatment of P. schreberi, intended to reduce exchangeably bound Fe³⁺, actually caused no change (absolute concentration) or a small increase (relative concentration) in the sulfate measured in the incubation medium.

Table 3.

Influence of cation addition and removal pretreatments on conversion of bisulfite (0·6 mm) to sulfate during incubation (2 h) of shoots of Pleurozium schreberi and Rhytidiadelphus triquetrus

Species/pretreatment	Absolute sulfate (µm)	Relative sulfate (µmol g−1 d. wt)
Pleurozium schreberi
DDW (control)	454a	117a
CaCl2	455a	112a
Na2EDTA	462a	128b
FeCl3	608b	181c
Rhytidiadelphus triquetrus
DDW (control)	600a	202a
CaCl2	439b	163b
Na2EDTA	502b	213a
FeCl3	633a	220c

Incubations for 2 h, in the light (72 µmol m⁻² s⁻¹), at 20±2 °C.

Data are means of three replicates.

ANOVA: within a species and data type, figures followed by the same superscript letter are not significantly different at P < 0·05.

DISCUSSION

In an earlier paper (Bharali and Bates, 2002) it was shown that addition of bisulfite caused a rapid cessation of photosynthetic CO₂ fixation in the mosses Pleurozium schreberi and Rhytidiadelphus triquetrus. However, with increasing incubation periods (beyond 2 h) photosynthesis was eventually restored. It was hypothesized that this recovery was linked to the progressive oxidation of bisulfite (yielding the considerably less toxic sulfate ion), a process that occurs naturally though rather slowly in solutions in contact with the air (Fig. 2A). The present results show that addition of living shoots of P. schreberi or R. triquetrus greatly accelerates the disappearance of bisulfite ions from the incubation solution. In earlier studies bisulfite ‘scavenging’ by Sphagnum shoots (Baxter et al., 1989, 1991) and lichen thalli (Miszalski and Niewiadomska, 1993a) was attributed to passive oxidation by transition metal cations (Fe³⁺, Mn²⁺ and Cu²⁺) adsorbed in the apoplast. However, there is now little doubt that, in lichens at least, tolerance of SO₂ is also linked to the activities of anti-oxidant enzymes and intracellular compounds such as ascorbate (Calatayud et al., 1999; Deltoro et al., 1999). In the present study, the rate of bisulfite loss from the solution bathing moss shoots varied depending on presence or absence of light, the application of metabolic inhibitors, acidity, the nature and concentrations of adsorbed metal cations and the species of moss. It is proposed that there is evidence for three processes, occurring to varying extents in the two moss species, which account for this loss of bisulfite: (1) external oxidation of bisulfite using metabolic (including photo-oxidative) energy; (2) ‘passive’ external oxidation of bisulfite catalysed by adsorbed Fe³⁺ ions; (3) cellular uptake and metabolic detoxification of bisulfite.

Light significantly stimulated bisulfite loss from the external solution. This supports the hypothesis that harvested light energy is being channelled into oxidizing power which is then available for detoxification of dissolved SO₂ (Miszalski and Niewiadomska, 1993a). The oxidant(s) involved are unlikely to include reactive oxygen species (ROS) such as superoxide as the PPFD used in this work (72 µmol photons m⁻² s⁻¹) was well below the probable saturating intensity for these mosses (see below). This leaves molecular O₂, generated by the splitting of water, the normal endpoint of the photosynthetic electron transport chain, as the likely driving force for bisulfite oxidation. Sérgio et al. (2000) observed concentric deposits of manganese oxides on the cell walls of the aquatic moss Fontinalis antipyretica growing in a stream contaminated by mine effluent. This implies that appreciable oxidative potential is present at ‘hotspots’ of photosynthetic O₂ release around the cell periphery. It is perhaps surprising that stronger evidence of a metabolic component in the oxidation of bisulfite was not observed in the study of Sphagnum by Baxter et al. (1989, 1991) and that of lichens by Miszalski and Niewiadomska (1993b).

In relation to a possible role of ROS in bisulfite detoxification, Marschall and Proctor (2004) pointed out that many bryophytes are essentially shade-adapted species with light saturation levels (PPFD_{95 %}) of 300 µmol photons m⁻² s⁻¹ or below. Light saturations about that level are characteristic of Pleurozium schreberi and Rhytidiadelphus triquetrus (Bates, 1979; J. W. Bates, unpubl. res.). On the basis of chlorophyll fluorescence measurements, they deduced that photoprotection (non-photochemical quenching, NPQ) is relatively undeveloped in mosses from shady sites but extensive in those from sunny habitats. Arguably, the shade plants would be the ones most likely to generate quantities of ROS capable of detoxifying bisulfite given sufficient light. Interestingly, Marschall and Proctor (2004) mentioned Sphagnum species as having high NPQ (though rather low PPFD_{95 %}), indicative of a high degree of photoprotection and a reflection of their unshaded habitats. It follows (but requires confirmation) that any superoxide generated in these plants would be quickly scavenged by SOD and not available for bisulfite oxidation. No comparable chlorophyll fluorescence data are available for the three lichens studied by Miszalski and Niewiadomska (1993b), but these species also live in relatively unshaded habitats and might be expected to have substantial photoprotection. Differences in the degree of photoprotection between the two mosses (arguably, P. schreberi is less shade-adapted than R. triquetrus) might also explain their different abilities to detoxify bisulfite solutions. The issue of whether ROS or O₂ is involved in the light-driven oxidation of bisulfite deserves further investigation, as does the possibility that some bisulfite may also be reduced to hydrogen sulfide.

The experiments with the inhibitor DCMU further support the hypothesis of a metabolically based detoxification. DCMU inhibits photosynthetic electron transport and oxygen evolution. It caused a substantially reduced rate of bisulfite loss from the incubation solution, especially in R. triquetrus. The observation that bisulfite loss was accentuated at pH 6 may also be relevant here. In experiments with isolated oat chloroplasts, Miszalski (1991a) observed that increasing the pH led to increasing generation of ROS. This result may not be transferable to intact plants; however, such an effect could possibly explain the present results. It might also help explain the restriction of R. triquetrus to alkaline calcareous soils under SO₂ pollution.

Bharali and Bates (2002) demonstrated accentuated loss of bisulfite from the incubation solution in P. schreberi and R. triquetrus following pretreatment with Fe³⁺. Analytical data showed that this treatment increased the total Fe concentration of the tissues, with probably most of the metal becoming exchangeably bound or otherwise adsorbed in the cell walls. There is little doubt that the Fe³⁺ catalysed extracellular oxidation of bisulfite as the pretreatment was accompanied by an increase in the sulfate concentration of the external solution (Table 3). In nature, Fe³⁺ is rarely likely to become sufficiently abundant in the apoplast of R. triquetrus to act in this manner because this moss becomes restricted to calcareous soils (with limited iron mobility) in SO₂-polluted regions. On the acid soils inhabited by P. schreberi, however, Fe³⁺ is potentially much more highly mobile and, particularly on soils with significant mineral content, shoot concentrations of the cation may be sufficient to have a significant bisulfite-detoxifying effect (Bharali and Bates, 2002).

So far this discussion has considered detoxification of bisulfite in terms of extracellular reactions. The possibility that a proportion of the supplied bisulfite has been absorbed and metabolized intracellularly, however, cannot be ruled out. However, the evidence that this might be an important process is rather equivocal. The study of sulfate generation indicates that much of the lost bisulfite can be accounted for by the appearance of sulfate in the external solution. However, sulfate generation in the external solution was reduced when R. triquetrus was pre-treated with CaCl₂ solution. As Bharali and Bates (2002, 2004) demonstrated, Ca²⁺ alleviates the inhibitory effects of bisulfite on photosynthetic carbon fixation and potassium leakage in both P. schreberi and R. triquetrus, but, unlike Fe³⁺ and the other transition metals, it presumably does not react directly with the bisulfite ions. It is more likely that it functions by stabilizing cell membranes or embedded protein channels against loss of permeability control (Bates, 1982; Lee, 1999; Plieth, 2005). The reduced extracellular sulfate observed could indicate that Ca²⁺ enhances cellular uptake of bisulfite. The experiment with DETC, an inhibitor of SOD, led to a very significant reduction in the rate of bisulfite loss from the incubation solution with both mosses. This could indicate that intracellular reactions, linked to the activity of an intracellular SOD are involved in bisulfite detoxification. A possibility would be the evolution of oxygen from superoxide dismutation. Another possibility is that the DETC, a chelating agent, is having an unexpected effect on the moss shoots, such as removing metal ions from the cell walls that would otherwise catalyse extracellular oxidation of bisulfite. In the present study EDTA (5·0 mm) pretreatment did not cause lowered bisulphite disappearance. Possibly, EDTA was relatively ineffective in chelating metals such as Fe³⁺ from the moss shoots. Alternatively, the EDTA may have disrupted normal membrane function so that bisulfite uptake or retention rates were modified.

Recent studies of SO₂ effects in lichens indicate that intracellular anti-oxidant defences may be both affected by the pollutant in sensitive species and important in resisting its effects in more-tolerant taxa (Silberstein et al., 1996; Calatayud et al., 1999; Deltoro et al., 1999). Much less is known about the involvement of metabolic defences in mosses. The present work, involving two relatively SO₂-sensitive terricolous species, reveals a complex situation where appreciable capacity exists to detoxify dissolved SO₂ via passive oxidation utilizing adsorbed metal ions and by a metabolically based oxidation, possibly utilizing O₂ evolved by photosynthesis and scavenging of ROS. The extent of intracellular detoxification is uncertain and requires further investigation. Differences in the apparent SO₂-tolerances, and soil preferences under pollution stress, of the two mosses studied here may reflect different degrees of reliance on the various proposed detoxification pathways.