Characterization of the phosphatase activities of mosses in relation to their environment


  • B. L. Turner,

    1. Department of Biological Sciences, University of Durham, Durham DH1 3LE, UK and
    Search for more papers by this author
    • *Present address: United States Department of Agriculture - Agricultural Research Services, North-west Irrigation and Soils Research Laboratory, 3793 N. 3600 E. Kimberly, ID 83341, USA.

  • R. Baxter,

    1. Department of Biological Sciences, University of Durham, Durham DH1 3LE, UK and
    Search for more papers by this author
  • N. T. W. Ellwood,

    1. Department of Biological Sciences, University of Durham, Durham DH1 3LE, UK and
    2. Department of Chemical and Process Engineering, University of Newcastle, Newcastle-upon-Tyne NE1 7RU, UK
    Search for more papers by this author
  • B. A. Whitton

    1. Department of Biological Sciences, University of Durham, Durham DH1 3LE, UK and
    Search for more papers by this author

Correspondence: B. A. Whitton. Fax: + 44 191 374 2427; e-mail:


Phosphatase activities and environmental features were characterized for 12 terrestrial and aquatic mosses in upland northern England, along with four species sampled from subarctic Sweden. Phosphomonoesterase (PMEase) and phosphodiesterase (PDEase) activities of shoot tips were measured using para-nitrophenyl phosphate (p-NPP) and bis-pNPP. All species showed PMEase activity, but not all showed PDEase activity. The mean pH optimum was 5·0 for PMEase and 5·7 for PDEase. The kinetic parameters Km and Vmax were calculated from three linear transformations of the Michaelis–Menten equation. The mean Km values of the mosses ranged between 77 and 468 µm for PMEase and 26 and 414 µm for PDEase. The corresponding Vmax values were 0·6–205 µmol pNP g−1 DW h−1 for PMEase and 1·4–110 µmol pNP g−1 DW h−1 for PDEase. Mosses from Sweden displayed greater Km and smaller Vmax values than those from England. The aquatics Fontinalis antipyretica and Rhynchostegium riparioides displayed two-phase kinetics for PMEase and PDEase, with Km and Vmax being dependent on substrate concentration. Staining suggested that PMEase activity was located in the cell wall of most mosses. Phosphatase assays provide a rapid method for screening environmental nutrient status and a standard procedure is recommended.


bromo-4-chloro-3-indolyl phosphate


bis-para-nitrophenyl phosphate


3,3-dimethylglutaric acid


ethylenediaminetetra-acetate, Km, Michaelis constant


limit of detection








para-nitrophenyl phosphate


maximum velocity of enzyme reaction


Mosses have been used extensively as indicators of pollution in terrestrial and aquatic environments and, to a lesser extent, as indicators of ambient nutrients. Several species have been shown to be sensitive indicators of metal pollution from both the atmosphere (Brown 1984; Bates 2000) and stream water (Say & Whitton 1983; Wehr & Whitton 1983). Terrestrial species are thought to acquire metals and nutrients through rainfall and aquatic species from the ambient water (Brown 1984): they are thus sensitive indicators of their respective environments, reflecting the levels of external pollutants and nutrients. Additional advantages as biomonitors include the facts that mosses are non-motile, evergreen, relatively tolerant to various pollutants and mostly easy to identify.

Mosses have been used in three main ways to assess environmental nutrient concentrations. For example, rates of atmospheric N deposition have been studied using the internal N content of Hylocomium splendens (Hicks et al. 2000) and the P status of terrestrial and aquatic environments has been determined by P content, rate of P uptake and the activities of phosphatase enzymes (Christmas & Whitton 1998a). These methods are indirect in that the measured values reflect in some way the P status of the moss at a site rather than a direct measure of ambient phosphate. The most widely used approach is to measure phosphatase activity, which itself varies in response to tissue P content (Press & Lee 1983; Christmas & Whitton 1998a). Phosphatase enzymes catalyse the hydrolysis of phosphate esters, releasing orthophosphate and an organic moiety. Some of these enzymes are synthesized and secreted externally to the cytoplasmic membrane and in most cases this has been shown to be a response to a P requirement (e.g. Bieleski 1974; Grainger et al. 1989). Phosphatase activity external to the cytoplasmic membranes can catalyse the release of orthophosphate from organic P compounds and hence increase the available phosphate in the immediate environment. This is important, because organic phosphates are often a major component of soil solution and stream water P in P-limited environments (Livingstone & Whitton 1984; Shand et al. 1994; Turner & Haygarth 2000). Phosphorus deficiency also increases the internal phosphatase activity of plants, presumably to provide orthophosphate to the growing parts by accelerating the turnover of internal organic P (Plaxton & Carswell 1999).

Phosphatase activities of whole or parts of organisms can be measured by standard techniques using analogue substrates such as para-nitrophenol (pNP). A relatively large number of studies have been conducted on algae, because of the ease of obtaining samples dominated by one species (Fitzgerald & Nelson 1966); in some cases, parallel studies have been made on axenic isolates (Whitton 1991). Field populations of terrestrial (Press & Lee 1983) and aquatic (Christmas & Whitton 1998a) mosses have been assayed in a similar way. Mosses where at least some populations show marked phosphatase activity include 11 Sphagnum spp. (Press & Lee 1983) and two widespread submerged aquatics Fontinalis antipyretica and Rhynchostegium riparioides (Christmas & Whitton 1998a, b). This approach is particularly useful for assessing the nutrient status of areas subject to atmospheric N deposition, as such deposition may be expected often to increase the demand for P and hence presumably enhance phosphatase activity.

The aim of this study was to develop a methodology for assessing phosphatase activity that could be applied to a range of moss species and environments. Previous studies have used widely different assay conditions, especially substrate concentration, thus making comparisons difficult. Before broad spatial and temporal surveys can be made, phosphatase enzymes require characterization to optimize and standardize the procedures for their assay. We therefore set out to: (i) characterize the kinetic parameters and pH optima of the enzymes; (ii) standardize the assay procedure for comparative research; (iii) investigate the location of the enzymes, because this has implications for the potential ecological role of phosphatase in P acquisition. Some species were chosen because they occur in a wide range of nutrient environments and are likely to show a wide range of phosphatase activities; others are more restricted and likely to be characteristic of environments where P limitation occurs for much of the time.

Materials and methods

Sampling locations

Mosses were sampled from terrestrial, wetland and aquatic environments in the UK and terrestrial and wetland environments in Sweden, representing a wide range of habitats, climates and levels of nutrient availability (Table 1). The UK terrestrial and some semi-aquatic samples were taken from Widdybank Fell, Upper Teesdale (grid ref. NY 820 290; 54°40′ N; about 520 m a.s.l.; mean annual rainfall 1560 mm). Mean daily temperatures range from an average of 0·1 °C in February to 12·3 °C in July. The area contains relict late-glacial plant assemblages and has three distinct soil types (supporting blanket peat, acid grassland and calcareous grassland) within a geographically small area, which is grazed by sheep. Terrestrial mosses were sampled from each soil type, whereas semi-aquatic mosses were sampled from areas adjacent to two streams, Slapestone Sike and Red Sike, which combine drainage from blanket peat and grassland with subterranean drainage from limestone (Livingstone & Whitton 1984).

Table 1.  Species used and the locations from which they were sampled. All sites are in the UK except Abisko, in northern Sweden
Ref. noSpeciesAuthoritySite and locationEnvironment
 1Hylocomium splendens(Hedw.) Bruch,
 Schimp. & Gümbel
Widdybank FellCalcareous grassland
 2Hylocomium splendens(Hedw.) Bruch,
 Schimp. & Gümbel
AbiskoAcid, peaty soil under birch
 3Hypnum jutlandicumHolmen & E. WarnckeWiddybank FellBlanket peat, growing under
 4Plagiothecium nemorale(Mitt.) A. JaegerWiddybank FellAcid grassland soil
 5Polytrichum communeHedw.Widdybank FellAcid grassland soil
 6Polytrichum communeHedw.AbiskoAcid, peaty soil under birch
 7Racomitrium lanuginosum(Hedw.) Brid.Widdybank FellBlanket peat
 8Racomitrium lanuginosum(Hedw.) Brid.Widdybank FellCalcareous grassland
 9Racomitrium lanuginosum(Hedw.) Brid.AbiskoAcid, peaty soil under birch forest
10Rhytidiadelphus squarrosus(Hedw.) Warnst.Widdybank FellAcid grassland soil
11Palustriella commutata
 var. commutata
(Hedw.) OchyraSlapestone Sike,
 Widdybank Fell
Calcareous/peat spring
12Palustriella commutata
 var. falcata
(Brid.) OchyraRed Sike, Widdybank FellStream draining peat and
 calcareous grassland
13Sphagnum cuspidatumEhrh. ex. Hoffm.Widdybank FellSpring on blanket peat
14Sphagnum cuspidatumEhrh. ex. Hoffm.AbiskoEdge of small peat bog
15Warnstorfia fluitans(Hedw.) LoeskePithouses Stream, BrandonAcid stream
16Fontinalis antipyreticaHedw.Stony Gill Beck, SwaledaleStream draining peat and
 calcareous grassland
17Fontinalis antipyreticaHedw.Cranecleugh Burn,
Stream draining peat and
 calcareous grassland
18Fontinalis antipyreticaHedw.Yet Burn, NorthumberlandStream draining peat and
 calcareous grassland
19Rhynchostegium riparioides(Hedw.) Cardot.Stony Gill Beck, SwaledaleStream draining peat and
 calcareous grassland

Aquatic mosses were sampled from designated 10 m reaches in four relatively fast-flowing streams in northern England. Much of the beds are covered in boulders, cobbles or pebbles. Stony Gill Beck, a tributary of the River Swale in north Yorkshire, has a catchment with year-round grazing for sheep and late summer/early autumn grazing for cattle. The water chemistry is influenced by moorland peat soils and underlying limestone. The sample site (O/S grid ref. SD 964 929; 380 m a.s.l.) has highly variable flow; the unshaded bed typically had about 5% moss cover.

Two streams were sampled in the Kielder Forest, Northumberland, which are less influenced by calcareous material. Cranecleugh Burn (NY 851 657) drains a deforested catchment, with the sample site at 240 m a.s.l. There is no shade and 2–4% moss cover within the reach. Yet Burn (NY 853 613) drains a catchment with plantation Picea abies and Picea sitchensis; the site (300 m a.s.l.) is shaded on both banks and has 2–3% moss cover. Pithouses Stream is a small acid stream degraded in recent years by mining activities (O/S grid ref. NZ 222 405). Unlike most such situations, this has led to a rise in pH from 2·6 (Hargreaves, Lloyd & Whitton 1975) to 3·5–4·0. The acidity probably arises from pyrite oxidation. All these UK upland areas have probably been subject to substantial atmospheric N deposition for the past 200 years. Current rates of N deposition on Great Dun Fell, close to Teesdale, are 20–40 kg N ha−1 year−1, depending on altitude (Hicks et al. 2000), whereas Lewthwaite (1999) reported 15–20 kg N ha−1 year−1 for Upper Teesdale.

An area near the Abisko Scientific Research Station, northern Sweden (68°21′ N, 18°49′ E), was chosen as relatively uncontaminated by atmospheric N deposition. The station is 200 km north of the Arctic circle and approx. 385 m a.s.l. The average annual temperature at low altitudes is approximately −1·0 °C; mean daily temperatures range from an average of −12 °C in February to 11 °C in July. Annual precipitation in Abisko valley is around 300 mm.

Moss species and sampling

Twelve taxa were used (Table 1), though, for simplicity, the two varieties of Palustriella commutata are termed ‘species’. All were sampled monthly from November 1999 to October 2000. The nomenclature is based on Blockeel & Long (1988), which differs from that widely used in the literature for Warnstorfia fluitans (=Drepanocladus fluitans), Palustriella commutata var. falcata (=Cratoneuron commutatum var. falcatum) and Palustriella commutata var. commutata (= C. commutatum var. commutatum).

Whole plants were sampled and returned immediately to the laboratory, where they were stored in the dark at 4 °C until analysis within 24 h after collection. In the case of aquatics, only submerged plants were taken. The mosses were washed in assay medium (see below) to remove soil debris and epiphytes, and 2 cm tips were removed for assays. The use of tips means that the region of active growth is included, which presumably best reflects recent changes in the environment, and which ensures minimum contamination by epiphytes and, in the case of the aquatics, deposits of Mn and Fe ‘oxides’. Mosses were taken from healthy clumps dominated by the particular species. For terrestrial and semi-aquatic mosses, a single 2 cm tip was used for each replicate, except for Sphagnum cuspidatum, where the capitulum (terminal group of leaves) was used. For the three aquatics, several tips were used for each replicate, partly because of shoot variability on a plant and partly to ensure sufficient activity for a short-term assay.

Phosphatase assay

Phosphatase activities were determined using para-nitrophenyl phosphate (pNPP) and bis-para-nitrophenyl phosphate (bis-pNPP) as analogue substrates for phosphomonoesterase (PMEase) and phosphodiesterase (PDEase), respectively. Moss tips were placed in glass vials, with 2·9 mL appropriate buffer made in assay medium (see below). The assay was initiated by adding 0·1 mL substrate (pNPP or bis-pNPP) and the vials were incubated at 20 °C in a shaking water bath (≈ 100 strokes min−1). Irradiance was in the range 20–60 µmol photon m−2 s−1 (although there is no evidence that light has an influence on short-term phosphatase activity in these species). Phosphatase activities were linear with time up to at least 1 h, so the duration of the assay was selected to give levels of product well above the detection limit. This ranged between 15 and 30 min, apart from samples from Sweden, for 1 h was used. To terminate the assay and develop the yellow colour, 2·5 mL assay mixture was transferred to 0·25 mL terminator solution in a test tube. The terminator consisted of 1·1 m NaOH (100 mm final), 27·5 mm EDTA (2·5 mm final) and 0·55 m K2HPO4 (50 mm final). It was designed to inactivate any free enzyme and raise the pH to > 11, but not cause hydrolysis of bis-pNPP, which occurs slowly at pH > 12 (Christmas & Whitton 1998a). Absorbance was measured at 405 nm; values > 0·8 were diluted with deionized water. The activity was determined using calibration curves constructed from pNP standards (0–25 µm) in assay medium. Following analysis, the mosses were dried at 105 °C for 24 h and then weighed to 0·0001 g on a microbalance. Enzyme activity was expressed as µmol pNP released g−1 DW h−1. Four replicates plus one control (no substrate) were used for each moss and a blank vial (substrate and buffer only). The values of the blank and control vials were subtracted from the final measured value. Only small amounts of chemical hydrolysis were detected from the substrates.

The assay medium was a modification of the no. 10 medium of Chu (1942), an important change from the original medium being the inclusion of EDTA as chelator and the absence of N and P. The assay medium was obtained by dilution from stock solutions. The final assay medium contained: 246·6 µm CaCl2, 188·7 µm NaHCO3, 101·4 µm MgSO4, 57·41 µm KCl, 6·73 µm FeCl3, 8·97 µm Na2.EDTA, 11·56 µm H3BO3, 0·229 µm MnCl2, 0·061 µm CuSO4, 0·037 µm CoSO4, 0·028 µm Na2MoO4 and 0·019 µm ZnSO4.

The limits of detection (LOD) were determined for each moss using the formula: LOD = mean blank + 1·96 standard deviations. For the terrestrial mosses, values ranged between 0·45 and 5·23 µmol pNP g−1 DW h−1 for Hypnum jutlandicum and Plagiothecium nemorale, respectively. For the semi-aquatic mosses, the LOD ranged between 2·53 and 10·15 µmol pNP g−1 DW h−1 for Sphagnum cuspidatum and P. commutata var. falcata, respectively. Aquatic mosses had LODs within this range.

The choice of buffer concentration was determined by assaying seven mosses in varying concentrations of 3,3-dimethylglutaric acid (DMG) at the optimum pH selected for kinetic studies (pH 5·0–5·5). Mosses were incubated in buffer concentrations from 0 to 50 mm and then phosphatase activities determined, together with the pH of the solution at the end of the assay.

The phosphatase activity of the mosses was insensitive to orthophosphate at concentrations up to 0·7 mm, which far exceeds that likely in nature or standard assays. Activity was linear with time to at least 1 h and temperatures up to at least 35 °C. We opted for 20 °C.

Mosses were always analysed within 24 h of collection. However, no significant differences in phosphatase activities were detected for three mosses (P. commutata var. commutata, Racomitrium lanuginosum and Hylocomium splendens) in a test of the effects of storage for 4 d in the dark at 4 °C (P > 0·10). Air-drying for 3 d at 30 °C resulted in variable and often substantial changes in PMEase and PDEase activity. However, no significant differences in PMEase activity were observed for Hyl. splendens and R. lanuginosum that had dried naturally in the field, when compared with mosses that had remained moist in adjacent microsites, suggesting that mild drying is a potential means of storing mosses prior to assays.

Characterization of phosphatase pH optima and kinetic parameters

Buffers in the pH range 3·5–10 were tested: DMG for pH 3·5–6·5, N-2-hydroxyethyl piperazine-N′-2-ethanesulphonic acid (Hepes) for pH 7–8 and glycine for pH 9–10. All were prepared in assay medium at 50 mm concentration, except where stated. Drop-wise additions of NaOH were used to adjust the pH. The pH tests were conducted using 500 µm substrate, which appeared to approach saturation on the basis of preliminary tests. The kinetics of the moss phosphatase enzymes were characterized using final pNPP concentrations in the range 40–1000 µm for terrestrial and semi-aquatic mosses and 25–1000 µm for aquatic mosses. For bis-pNPP, final concentrations in the range 25–1000 µm were used for all mosses. All kinetics assays were conducted at pH 5·0 for terrestrial and semi-aquatic mosses and 5·5 for aquatic mosses. Assays were conducted at 0·5 pH unit intervals for UK samples, but only 1·0 pH unit intervals for Swedish samples.

The velocity of enzyme catalysis increases with increasing substrate concentration up to a certain point that approaches a maximum velocity (Vmax). The Michaelis constant (Km) is the substrate concentration at which the velocity is half the maximum velocity. The Michaelis constant is an indication of the affinity of the enzyme for the substrate: the lower the Km, the higher is the affinity. These parameters are described by the Michaelis–Menten equation:


where V is the velocity of the reaction at any time and S is substrate concentration. The kinetic parameters (Vmax and Km) were calculated using the three possible linear transformations of the Michaelis–Menten equation. These were the Lineweaver–Burk plot (1/V versus 1/S, where the y intercept is 1/Vmax and the gradient is Km/Vmax), the Eadie–Hofstee plot (V versus V/S, where the y intercept is Vmax and the gradient is –Km) and the Hanes–Woolf plot (S/V versus S, where the y intercept is Km/Vmax and the gradient is 1/Vmax). Each transformation gives different weighting to the various sources of error and therefore gives different results for the kinetic parameters. It is considered that the Hanes–Woolf plot is the most reliable. The Lineweaver–Burk plot gives the best-looking straight line when fitted by eye, but can give a grossly misleading impression of the experimental error and the points are not distributed homogeneously (too much weight is given to low substrate concentrations, which are greatly affected by slight variations in the blank). The Eadie–Hofstee plot does not fill all the conditions for a least-squares method, as the error-containing variable (velocity, V) is present on both axes. The Hanes–Woolf plot avoids all these problems and is preferred over the other straight-line plots for most purposes (Cornish-Bowden 1996).

Staining of phosphatase activity

Staining with bromo-4-chloro-3-indolyl phosphate (BCIP), which produces a blue/purple stain when hydrolysed (Coston & Holt 1958), was used to establish possible enzyme localization. BCIP solution in alkaline buffer was used as received from Sigma-Aldrich Chemicals, Poole, UK. Moss tips were incubated in 4 mL BCIP solution with the addition of 1 mL 50 mm DMG buffer, pH 5. The reaction was terminated with 1 mL 0·5 m NaOH. Leaves were inspected at × 1000 magnification with standard and epifluorescence microscopy.


All species showed PMEase activity, but not all showed PDEase activity. Those not showing the latter were P. commutata var. commutata (although var. falcata showed marked PDEase activity) and Polytrichum commune from both UK and Sweden. Pl. nemorale showed highly variable PDEase activity, so kinetic parameters and pH optimum were not calculated. PMEase activity was greater than PDEase activity in all other species. The terrestrial mosses from northern English uplands generally displayed much greater phosphatase activity than the Swedish mosses; for instance, Hyl. splendens from England and Sweden displayed PMEase Vmax values of 127 and 21 µmol pNP g−1 DW h−1, respectively.

PH Optima

The PMEase pH optima were quite similar for all 12 species, with a mean of pH 5·0 (Fig. 1; Table 2). All the terrestrial mosses from Upper Teesdale showed pH optima at 5·0. One of the terrestrial mosses from Sweden had an optimum at 5·0, but the other two had optima at 4·0. The semi-aquatic and aquatic mosses had optima at 5·0 and 5·5. Some PMEase activity was detected in the alkaline region of all species. In some species, activity increased sharply in the acid region (pH 3–3·5, e.g. Pl. nemorale). This might represent cell damage to the cells, but W. fluitans also showed this phenomenon (for PDEase) despite growing at pH 3·5 and lower. The similarity of the pH optimum for PMEase contrasted with the range of environmental pH values from which they were sampled. The optimum pH of PMEase activity only coincided with the environmental pH for Po. commune and Hyl. splendens from Sweden.

Figure 1.

pH profiles of the PMEase and PDEase activities of eight mosses.

Table 2.  pH optima of PMEase and PDEase activities of the mosses assayed under conditions used here. Where species displayed a broad high activity range, this range is indicated on the basis of standard errors
Ref. noSpeciesEnvironmental pHPhosphomo-noesterasePhosphodiesterase
  1. Note that soil pH does not necessarily reflect the moss environmental pH (see text). ND, not determined.

 1Hylocomium splendens, UK7·35·04·5–5·05·54·0–7·0
 2Hylocomium splendens, Sweden3·94·0 7·0 
 3Hypnum jutlandicum4·25·0 6·5 
 4Plagiothecium nemorale4·05·0 4·5 
 5Polytrichum commune, UK4·05·0 < LOD 
 6Polytrichum commune, Sweden3·94·04·0–5·0< LOD 
 7Racomitrium lanuginosum, peat, UK4·25·04·5–5·56·05·5–7·0
 8Racomitrium lanuginosum, calc. grass, UK7·35·0 ND 
 9Racomitrium lanuginosum, Sweden3·95·04·0–6·06·05·5–7·0
10Rhytidiadelphus squarrosus4·05·04·5–5·55·55·0–7·0
11Palustriella commutata var. commutata7·65·04·5–5·5Too variable for determination 
12Palustriella commutata var. falcata7·85·55·5–6·56·04·0–8·0
13Sphagnum cuspidatum, UK4·05·5 6·04·5–6·5
14Sphagnum cuspidatum, Sweden3·95·03·0–6·07·03·0–7·0
15Warnstorfia fluitans3·55·0 3·03·0–7·0
16Fontinalis antipyretica, Stony Gill Beck8·04·54·5–5·05·5 
17Fontinalis antipyretica, Cranecleugh Burn6·05·55·5–6·06·05·0–7·0
18Fontinalis antipyretica, Yet Burn7·06·04·5–6·55·54·5–6·0
19Rhynchostegium riparioides8·05·0 6·06·0–7·0
Mean optimum pH  5·0 5·7 
SD  0·5 1·0 

For PDEase, the optimum pH levels were broader and differed more between species (Fig. 1). The mean optimum pH for PDEase across all mosses was 5·7, but the optima ranged from 4·5 to 7·0 for terrestrial, 6·0–7·0 for semi-aquatic and 3·0–6·0 for aquatic mosses (Table 2). For P. commutata var. falcata, considerable activity occurred up to pH 10. The only mosses for which PDEase activity largely coincided with the environmental pH were Fontinalis antipyretica from Cranecleugh Burn (pH 6·0) and W. fluitans (pH 3·0 for PDEase optimum and 3·5–4·0 for environment).

Enzyme kinetics of terrestrial and semi-aquatic mosses

Phosphatase activity in most species obeyed Michaelis–Menten kinetics (Fig. 2). The three linear transformations of the equation gave slightly different results for the kinetic parameters. For PMEase, the mean Km values of all three transformations ranged between 77 and 425 µm for the terrestrial mosses and 168 and 402 µm for the semi-aquatic mosses (Table 3). The corresponding Vmax values were similar for individual mosses across all the transformations and ranged from 0·6 and 149 µmol pNP g−1 DW h−1 for the terrestrial mosses and 3·1 and 205 µmol pNP g−1 DW h−1 for the semi-aquatic mosses. With the exception of R. lanuginosum, comparison between UK and Swedish mosses showed lower Vmax and higher Km for the Swedish mosses. Comparison of R. lanuginosum growing on peat and calcareous grassland at Widdybank Fell showed that the moss growing on the calcareous grassland displayed a lower Km and higher Vmax than when growing on the peat, suggesting greater phosphorus limitation on the peat.

Figure 2.

Michaelis–Menten plots of PMEase and PDEase activities of three selected mosses from terrestrial, semi-aquatic and aquatic environments.

Table 3.  Kinetic parameters of PMEase activity of the mosses studied as determined from the three linear transformations of the Michaelis–Menten equation. The units of Km are µm and of Vmax are µmol pNP g−1 DW h−1. The values of R2 refer to the fit of the straight line of the transformations of the Michaelis–Menten equation and indicate the fit of the data to that transformation (see also Table 4)
Ref. noSpeciesLineweaver–BurkEadie–HofsteeHanes–WoolfMean
 1Hylocomium splendens, UK1921250·981911270·841961280·97193127
 2Hylocomium splendens, Sweden444 21·60·99333 17·80·81483 22·00·93420 20·5
 3Hypnum jutlandicum3691350·983861420·775201690·93425149
 4Plagiothecium nemorale108 73·50·99109 74·00·93107 73·50·99108 73·7
 5Polytrichum commune, UK 49·7 13·20·90 59·4 14·00·77121 16·70·98 76·7 14·6
 6Polytrichum commune, Sweden200  0·70·93110  0·50·55178  0·60·94163 0·6
 7Racomitrium lanuginosum, peat, UK403 61·30·99384 60·10·86460 66·70·98416 62·7
 8Racomitrium lanuginosum, calc. grass, UK310 90·90·99348 99·30·903801040·98346 98·1
 9Racomitrium lanuginosum, Sweden289 14·10·99247 12·80·89304 14·20·98280 13·7
10Rhytidiadelphus squarrosus103 27·40·97118 29·20·88171 33·20·99131 29·9
11Palustriella commutata var. commutata4171080·983641010·724251100·93402106
12Palustriella commutata var. falcata3561181·003931260·974241320·99391125
13Sphagnum cuspidatum, UK1391890·981541990·862132270·97168205
14Sphagnum cuspidatum, Sweden277 2·81·00322 3·10·86432 3·60·92344 3·1
15Warnstorfia fluitans1281010·99111 96·00·843792780·92206158
16Fontinalis antipyretica, Stony Gill Beck 76·3 25·10·83108 32·30·58219 44·40·91135 33·9
17Fontinalis antipyretica, Cranecleugh Burn 86·0 48·80·89129 62·80·67234 80·00·95150 63·9
18Fontinalis antipyretica, Yet Burn1121350·781741890·59104 94·30·99130140
19Rhynchostegium riparioides192 52·10·9831646·90·71314 69·40·95274 56·1

For PDEase, Km values were generally lower than for PMEase and the mean values of all transformations ranged between 26 and 414 µm for the terrestrial mosses and 73 and 102 µm for the semi-aquatic mosses (Table 4). The corresponding Vmax values ranged from 2·7 and 110 µmol pNP g−1 DW h−1 for the terrestrial mosses and 1·4 and 58 µmol pNP g−1 DW h−1 for the semi-aquatic mosses. Although the Km values of PMEase activity of the Swedish mosses were generally higher than those of the same species growing in the UK, the Km values of the PDEase of the Swedish mosses were lower than the respective species growing in the UK.

Table 4.  Kinetic parameters of PDEase activity of the mosses studied as determined from the three linear transformations of the Michaelis–Menten equation. See Table 3 for explanation of units and R2 values
Ref. noSpeciesLineweaver–BurkEadie–HofsteeHanes–WoolfMean
 1Hylocomium splendens, UK16741·50·99177 43·30·91187 44·20·99177 43·0
 2Hylocomium splendens, Sweden 65·2 3·40·89 69·5 3·60·59101 3·80·88 78·4 3·6
 3Hypnum jutlandicum34896·20·984201130·864721220·97414110
 4Plagiothecium nemoraleToo variable to conduct kinetic tests          
 5Polytrichum commune, UK< LOD          
 6Polytrichum commune, Sweden< LOD          
 7Racomitrium lanuginosum, peat, UK 77·422·60·97 94·6 24·80·85155 29·40·98109 25·6
 8Racomitrium lanuginosum, calc.grass., UKKinetics not determined          
 9Racomitrium lanuginosum, Sweden 29·4 2·80·82 24·8  2·70·49 23·3 2·70·95 25·8 2·7
10Rhytidiadelphus squarrosus 92·213·61·00 97·3 14·00·97 96·5 13·91·0 95·3 13·8
11Palustriella commutata var. commutata< LOD          
12Palustriella commutata var. falcata 92·656·20·99 98·8 58·10·95113 60·60·99102 58·3
13Sphagnum cuspidatum, UK 74·143·51·00 74·1 43·60·9672·0 43·10·99 73·4 43·4
14Sphagnum cuspidatum, Sweden 64·0 1·30·97 73·6 1·40·88111 1·50·99 82·9 1·4
15Warnstorfia fluitans 70·590·90·92 91·81050·731701280·94111108
16Fontinalis antipyretica, Stony  Gill Beck 73·4 6·60·91104 8·00·57335 12·70·82171 9·1
17Fontinalis antipyretica, Cranecleugh Burn 78·610·00·86 24·9 7·90·10 9·0 7·20·98 37·5 8·4
18Fontinalis antipyretica, Yet Burn 72·683·30·94 78·9 89·60·54125 95·20·77 92·0 89·4
19Rhynchostegium riparioides 53·312·50·83 74·8 14·80·65167 19·20·96 98·3 15·5

Enzyme kinetics of aquatic mosses

Aquatic mosses did not all obey apparent Michaelis–Menten kinetics. Using the full substrate range, the Km values calculated from the means of the three linear transformations of the Michaelis–Menten equation ranged between 130 and 274 µm for PMEase and 37·5 and 171 µm for PDEase. The corresponding Vmax values ranged between 33·9 and 158 µmol pNP g−1 DW h−1 for PMEase and 8·4 and 108 µmol pNP g−1 DW h−1 for PDEase. The PMEase activity of F. antipyretica was in the sequence Yet Burn > Cranecleugh Burn > Stony Gill Beck, although Km values were similar across all sites. However, some aquatic mosses did not display apparent Michaelis–Menten kinetics. Hanes–Woolf plots indicated two distinct rates of enzyme kinetics operating at ‘high’ and ‘low’ substrate concentrations (e.g. for the PMEase activity of F. antipyretica from Cranecleugh Burn; Fig. 3). By separating the data into ‘low’ and ‘high’ substrate ranges, it was possible to calculate two separate enzyme kinetics depending on substrate concentration (Table 5). For example, for the PMEase activity of F. antipyretica from Cranecleugh Burn, the ‘low’ concentration range (25–100 µm) had a Km of 19·5 µm and a Vmax of 23·1 µmol pNP g−1 DW h−1, whereas the ‘high’ substrate concentration range (100–1000 µm) had a Km of 379 µm and a Vmax of 91·7 µmol pNP g−1 DW h−1. Therefore, at ‘low’ substrate concentrations PMEase appeared to display a much greater affinity for the substrate (lower Km) than at high substrate concentrations. The same phenomenon was evident for the PMEase activity of F. antipyretica from the other two sites and Rh. riparioides and for the PDEase activity of F. antipyretica and Rh. riparioides from Stony Gill Beck. The other aquatic moss W. fluitans did not display the two-phase kinetics for either activity.

Figure 3.

Hanes–Woolf plot of PMEase activity of Fontinalis antipyretica from Cranecleugh Burn, showing two rates of reaction at ‘low’ (0–100 µm) and ‘high’ (100–1000 µm) substrate concentrations [where the Y-axis (S/V) is substrate concentration (µm) divided by the velocity of the reaction (µmol pNP g−1 DW h−1)]. Units of Vmax are µmol pNP g−1 DW h−1.

Table 5.  Kinetic parameters for aquatic mosses calculated using Hanes–Woolf plots for ‘low’ and ‘high’ substrate ranges, showing two-phase kinetics for PMEase and PDEase
Ref. noSpecies and siteSubstrate range
(µmol pNP g−1 DW h−1)
16Fontinalis antipyretica
 Stony Gill Beck100–750 250 46·5
17Fontinalis antipyretica
 25–100 19·5 23·1
 Cranecleugh Burn


18Fontinalis antipyretica
 Yet Burn250–10000

19Rhynchostegium riparioides 25–250 109 34·5

16Fontinalis antipyretica
 28·0 4·2
 Stony Gill Beck200–1000
19Rhynchostegium riparioides 25–250 33·7  9·3



Buffer tests

The effects of DMG concentration on the phosphatase activity were negligible in the mosses studied (F. antipyretica, Hyl. splendens, Hyp. jutlandicum, P. commutata var. falcata, Rh. riparioides, S. cuspidatum, W. fluitans) when assayed at the pH optimum of 5·0 (e.g. for Hyp. jutlandicum;Fig. 4). Small decreases in activity were observed at low buffer concentrations (around 1 mm), but these were largely explained by a decrease in enzyme activity at the greater pH levels in these weakly buffered samples. The desired pH was maintained using a buffer concentration of approximately 25 mm for most species except for S. cuspidatum, for which 50 mm was required. The main factor affecting the pH of the assay medium seemed to be the release of orthophosphate from the substrate during the assay. Therefore, to ensure that the desired pH was maintained throughout the assay, 50 mm buffer was chosen for all subsequent assays.

Figure 4.

Plot of PMEase and PDEase activities and final assay pH (pretermination) against buffer concentration for Hypnum jutlandicum. d, PMEase; s, PDEase; m, final PMEase pH; n, final PDEase pH.

Location of enzyme

Staining revealed PMEase activity in different locations depending on the moss species. In S. cuspidatum, P. commutata var. falcata, Hyl. splendens and R. lanuginosum, staining appeared inside the cell wall. In R. lanuginosum, strong staining was evident in the cells near to the moss tips, with comparatively little staining lower down the leaves. P. commutata var. falcata also appeared to show staining inside the cytoplasmic membrane, although the staining was not clear enough to show this definitively. The moss with one of the lowest PMEase activities, Po. commune, showed highly localized staining in the edges of small sections of the leaves. Staining was poorly visible in the other mosses. No staining on the exterior cell surfaces was observed, which also highlighted the lack of epiphyte contribution towards the measured phosphatase activity.


The study showed the potential for using assays of moss phosphatase activities for assessing environmental nutrient status. For instance, mosses from northern Sweden displayed markedly lower phosphatase activity than the same species in upland England. This probably reflects the large amounts of atmospheric N deposition in the latter and hence enhanced competition for P. Studies were conducted on the youngest parts of the plants, in which phosphatase activities presumably reflect more recent environmental events. However, detailed interpretation of the results in relation to a particular time period is difficult because of the uncertainty of the extent to which different species of moss can store P and transport it elsewhere. Well-differentiated translocation systems in mosses are largely confined to the biggest species, such as the primitive vascular system in Po. commune (Thomas, Schiele & Damburg 1990). However, other species are known to translocate nutrients to the growing tips, for example N and P in Hyl. splendens (Eckstein & Karlsson 1999; Eckstein 2000) and various elements in Rhytidiadelphus squarrosus (Wells & Brown 1996).

Two features of the assay procedure need to be considered in particular. The substrate concentrations used were high compared with those likely in the environment, where organic P concentrations in soil solutions and stream waters seldom exceed 10 µm and are more typically in the region of 1 µm (Livingstone & Whitton 1984; Shand et al. 1994; Turner & Haygarth 2000), at least in upland regions. The pH optimum for PMEase activity was surprisingly similar for all species. One possibility might be that the pH of the micro-environment inside cell walls is quite similar for all moss species, whatever the pH of the external environment. Staining indicated that some or all of the cell walls provide an important site of phosphatase activity responsible for hydrolysis of BCIP. However, P. commutata var. falcata showed apparent intracellular staining, so the possibility cannot be ruled out that organic phosphates may have passed into the cell in all the assays and some hydrolysis taken place there. If this played an important role in the present study, then the observed pH optima might reflect the influence of pH on transport rather than the enzyme.

The differences between enzyme pH optima and the environmental pH raise questions about the ecological significance of the enzyme and are not confined to moss phosphatase (Turner & Haygarth 2001). A possible explanation for this phenomenon is a difference between the pH of the environment and that at the enzyme microsite, whether this occurs in the cell wall or inside the cytoplasmic membrane. It should also be noted that the PMEase pH optima resemble more closely the natural pH of rainfall (around pH 5–6). The variable and broad pH optima of PDEase are also unusual, especially considering the apparent wide pH range for PDEase activity, for example, P. commutata var. falcata. The slightly more alkaline pH optimum of moss PDEase relative to PMEase is similar to these enzymes in soils, where the optimum activities of PMEase and PDEase are typically 6·5 and 8·0, respectively, irrespective of soil pH (Tabatabai 1982).

In spite of the assays being conducted on moss shoots, the observed phosphatase kinetics resembled those of pure enzyme systems. The differences between the three linear transformations of the Michaelis–Menten equation were generally small. The greater Km values for PMEase relative to PDEase are unusual, because they indicate a greater affinity for bis-pNPP than for pNPP, despite the latter being a much smaller molecule (and therefore more likely to be able to enter the cell wall). In general, the differences in Km values appear to reflect the various locations of the enzyme. For example, P. commutata var. falcata displayed high Km values for both PMEase and PDEase (391 and 102 µm, respectively), perhaps because this moss displayed phosphatase activity throughout the cell structure, including within the cell wall and internal to it. In contrast, Po. commune gave amongst the lowest Km values for PMEase (77 and 162 µm for UK and Swedish plants, respectively) and displayed highly localized phosphatase activity on small sections of the leaf edge. However, this pattern was not evident in all mosses.

The presence of both low- and high-affinity systems found in F. antipyretica and Rh. riparioides has previously been reported for the red algae Corallina elongata and various species of Gelidium (Hernández, Fernández & Niell 1995, Hernández, Niell & Fernández 1996). The kinetic parameters at ‘low’ substrate concentrations for F. antipyretica are similar to those determined previously for moss sampled from the River Swale, UK (Christmas & Whitton 1998b). Using substrate concentrations up to 100 µm, they measured Km values ranging between 24 and 186 for PMEase and between 10·5 and 114 for PDEase; the corresponding Vmax values were 30–114 µmol g−1 DW h−1 for PMEase and 3·8–136 µmol g−1 DW h−1 for PDEase. The high affinity for the substrate at low substrate concentrations may represent the more realistic kinetics of the enzyme, because ‘low’ substrate concentrations are typically found in the environment. In contrast, the severe reduction in affinity at high substrate concentrations may result from restricted access to enzymes located in the cell wall, competitive product inhibition by orthophosphate or some form of negative co-operativity. Alternatively, there may be two enzyme systems, with the low affinity system activated in the presence of high substrate concentrations to prevent physiological cell damage or phosphate saturation. In any case, the two-phase kinetics has important implications for phosphatase assays of aquatic mosses, because the lower Km values in the ‘low’ substrate range mean that substrate concentrations in the assay mixture of only 100–200 µm is required for substrate saturation and zero-order kinetics; any greater substrate concentration will enter the kinetic range of greater Km values and render the assay at less than saturation.

Recommended assay conditions

The following standardized and optimized assay conditions are recommended for comparative studies:

  • • Buffer pH 5·0 for both PMEase and PDEase at a concentration of 20 mm, except for Sphagnum, which requires 50 mm.

  • • The final substrate concentration in the assay mixture should be 500 µm for both PMEase and PDEase. However, for aquatic mosses, it may be advisable to test the effect of substrate concentration in assays in order to permit a lower substrate concentration to be used, preferably 100–200 µm.

  • • The assay time can be chosen to give measurable readings above the LOD, but 15–30 min should be sufficient for most mosses.

  • • Temperature of the assay should be 20 °C.

  • • Irradiance of about 50 µmol photon m−2 s−1. (Sufficient studies have not yet been made to be sure that light never has an effect during short-term assays.)

The recommended buffer pH is the optimum for PMEase in most mosses. The PDEase optima vary considerably about this range, but it appears sensible to use the same buffer pH to standardize the assay procedure. The use of assay medium for the buffer is recommended, although there is considerable scope for simplifying the composition of this medium. The buffer concentration is sufficient to hold the pH for the duration of the assay for all mosses with negligible effect on the observed activity.

The final substrate concentrations are recommended on the basis of the measured Km values and is at, or greater than, the Km of these enzymes in mosses. This will ensure zero-order kinetics and avoid substrate depletion during the assay. In studies of soil phosphatase, it has been recommended that substrate concentrations should be five times the Km value (Malcolm 1983), but this is not recommended for mosses because it seems likely that substrate concentrations in the mM range could lead to cell damage. The low affinity kinetics observed in F. antipyretica and Rh. riparioides at substrate concentrations greater than 100–200 µm requires a lower substrate concentration for routine assays. This effect might also be seen in other mosses if concentrations greater than 1000 µm are employed.

The substrate concentrations recommended for routine assays differ markedly from those likely to occur in nature and different procedures will be needed to assess the quantitative role of ‘surface’ phosphatase activities at field sites. Assays can be conducted with fluorogenic substrates at concentrations down to about 0·5 µm, far below the saturating concentrations used in the present study. Furthermore, the pH characteristics of many environments differ considerably from the range found to be optimum in the assays reported here with high substrate concentrations.

If moss phosphatase can be shown to correlate with intracellular phosphorus and/or environmental phosphorus availability, it will provide a rapid, simple and inexpensive screening for P status or atmospheric N pollution, even if the results give little indication of quantitative importance under field conditions. Ideal mosses for assessing nutrient status should be widespread and tolerant to a range of nutrient levels, conditions which Hyl. splendens fits well.


This project was funded by Natural Environment Research Council grant GR9/04458 to R.B. and B.A.W. An Engineering and Physical Sciences Research Council studentship is held by N.E.