By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Please be advised that we experienced an unexpected issue that occurred on Saturday and Sunday January 20th and 21st that caused the site to be down for an extended period of time and affected the ability of users to access content on Wiley Online Library. This issue has now been fully resolved. We apologize for any inconvenience this may have caused and are working to ensure that we can alert you immediately of any unplanned periods of downtime or disruption in the future.
•Relationships between nitrogen deposition in the UK and phosphomonoesterase (PME) activity and nitrogen (N) and phosphorus (P) concentrations in Cladonia portentosa were quantified to understand factors limiting lichen growth and to further develop biomarkers for N pollution.
•Lichen was collected from sites differing either in rates of wet N (NH4+ + NO3−) deposition or in annual mean N concentration in rainfall based on both measured and modelled data sets. The PME activity, and total N and P concentrations were measured in specific horizontal strata in lichen mats and PME activity in the thallus was located using an enzyme-labelled fluorescent phosphatase substrate.
•With an increase in modelled N deposition from 4.1 to 32.8 kg N ha−1 yr−1, PME activity, thallus N and N : P ratio increased by factors of 2.3, 1.4 and 1.8, respectively. Correlations with modelled data were generally stronger than with measured data and those with N deposition were stronger than those with N concentration in rainfall. The PME activity was located solely in the lichen fungus in outer regions of the thallus.
•Nitrogen enrichment changes lichen N : P ratios from values typical of N limitation (for example, 10) to those indicative of P limitation (for example, 26) driving upregulation of PME activity.
Mat-forming terricolous lichens in such genera as Cladonia (subgenus Cladina) have been shown to provide coherent biomarkers for N enrichment, that is, biological attributes that respond to N deposition in a consistent and predictable manner. Cladina spp. are widespread and locally abundant on well-drained terrain in subarctic heathlands and boreal forests as well as in temperate heathlands and uplands. Hyvärinen & Crittenden (1998a) demonstrated that c. 60% of variation in tissue N concentration in Cladonia portentosa at different sites across the UK could be explained by variation in wet deposition of inorganic N, and Walker et al. (2003) found that N concentration in Cladonia stellaris decreased northwards along a 240-km transect across the taiga–tundra ecotone in north European Russia which, according to Ryaboshapko et al. (1998), lies along a gradient of decreasing N deposition. Coupling between the chemical composition of mat-forming lichens and that of atmospheric deposits might be particularly close because: (1) lichen mats typically develop in open situations where rainfall is intercepted directly, and hence relationships between lichen chemistry and atmospheric inputs are not confounded by nutrient exchanges between rainfall (or snow meltwater) and overlying vascular plant canopies (cf. Farmer et al., 1991; Søchting, 1995); (2) accumulation, below well-developed lichen mats, of copious quantities of structurally intact dead thallus or necromass partially isolates living thalli from the chemical influence of underlying soil (Ellis et al., 2003, 2004); (3) lichen mats are efficient scavengers of key ions in precipitation such as NO3−, NH4+ and PO43− (Crittenden, 1983, 1989; Hyvärinen & Crittenden, 1998b).
In their studies on C. portentosa, Hyvärinen & Crittenden (1998a) found that concentrations of thallus N and P ([N]lichen and [P]lichen, respectively) covaried and that both were positively correlated with at least some components of N deposition. The reason for covariation between [N]lichen and [P]lichen is not known but two suggestions were advanced. First, N and P deposition might covary spatially because of man-made P contamination of the atmosphere deriving from, for example, mechanized agriculture (cf. Anderson & Downing, 2006). Unfortunately, data on P deposition are scarce and so it is difficult to evaluate this possibility. Second, lower lichen growth rates in N-polluted regions might result in lower growth dilution of tissue P concentrations but no evidence of lower growth rates in C. portentosa at N polluted sites was found (Hyvärinen & Crittenden, 1998c). A further possibility is that N enrichment might promote P capture possibly involving upregulation of phosphatase enzymes catalysing the release of orthophosphate from organic P sources. This suggestion is partially consistent with studies on plant root, mycorrhizal and soil systems in which N enrichment is known to promote phosphomonoesterase (PME) activity (Johnson et al., 1999; Treseder & Vitousek, 2001; Phoenix et al., 2003; Pilkington et al., 2005). However, such upregulation of PME activity in plant root systems is usually associated with low plant tissue P concentration and an increase in N : P values. Lane & Puckett (1979), LeSueur & Puckett (1980) and more recently Stevenson (1994) showed that surface PME activity, as detected by hydrolysis of p-nitrophenyl phosphate (pNPP), is readily measurable in a range of lichen species. Crittenden (1998) speculated that such activity might account for higher detectable phosphate concentrations in snow meltwater that had passed through canopies of the Antarctic lichen Usnea sphacelata compared with unmodified meltwater.
In this paper we examine PME activity and thallus N : P relationships in the common heathland lichen C. portentosa at sites in the UK mainland with contrasting N deposition characteristics. We also seek to distinguish between the effects on lichen chemistry and physiology of N-deposition rate, N concentration in precipitation, and rainfall depth. Sites at which the lichen was collected were selected using both modelled and measured N deposition data facilitating a partial evaluation of the likely accuracy of modelled deposition values.
Materials and Methods
Cladonia portentosa (Dufour) Coem. is a common terricolous lichen in Calluna vulgaris-dominated lowland heathland and upland moorland throughout the UK. It was collected from sites subject to widely differing N deposition rates and mean N concentrations in rainfall. Site selection used two data sets: modelled data and measured data. Modelled data were derived using an atmospheric deposition model paramaterized for moorland terrain (Smith & Fowler, 2001). The model utilizes a simulated rainfall field for the UK generated by the UK Meteorological Office and interpolated maps derived using measurements from the UK Acid Deposition Monitoring Network operated by AEA Technology plc (Didcot, UK). Modelled data were provided for annual mean total wet deposited inorganic N (NO3− + NH4+) (Ns) and annual mean volume weighted concentration of dissolved inorganic N in rainfall ([N]s) as 5 × 5 km gridded data sets for the UK. Each grid square was assigned a site code and, using a plot of Ns vs [N]s (Fig. 1a), individual grid squares were then selected to represent two gradients: one in Ns but in which grid squares had broadly similar [N]s values and the other in [N]s but in which grid squares had similar Ns values (Fig. 1b; Table 1).
Table 1. Details of sites at which Cladonia portentosa was collected (sites selected from those in Fig. 1a,c) including values of modelled and measured* annual mean total inorganic N deposition (N), volume weighted inorganic N concentration in precipitation ([N]ppt), and precipitation for each study site (ss)
Name of site
National grid reference
Wet N deposition (N) (kg N ha−1 yr−1)
[N]ppt (mg l−1)
Annual precipitation (mm yr−1)
Discrepancy in distance between ss and rg (km)†
Discrepancy in altitudes between ss and rg (m)†
Date of lichen collection
*Calculated from 4 yr (2002–2005) average data with wet inorganic N deposition calculated from annual mean volume weighted [N]ppt multiplied by rainfall depth. Data taken from Hayman et al. (2004a,b, 2005, 2007).
†Discrepancies in distance and altitude between study sites and rain gauges (rg) are provided for measured data (ss deviation from rg), along with collection dates for all sites. Sites varying in measured inorganic N deposition and N concentration were selected using 3 yr average data (2002–2004, Fig. 1c,d), while relationships reported in the text and in Table 2 between lichen chemistry and Nm or [N]m are based on 4 yr average values (2002–2005) given.
24 Apr 2006
Tarn at Leaves Reservoir
24 Apr 2006
5 May 2006
24 Apr 2006
21 Mar 2006
8 May 2006
11 May 2006
3 Apr 2006
13 May 2006
21 Mar 2006
24 Mar 2006
25 July 2006
27 Mar 2006
28 Mar 2006
3 Apr 2006
24 Apr 2006
24 Apr 2006
21 Apr 2006
21 Apr 2006
8 Jun 2006
5 May 2006
13 May 2006
12 May 2006
9 May 2006
10 May 2006
4 Apr 2006
5 May 2006
Measured data were obtained from the UK Acid Deposition Monitoring Network, which comprises 38 rainfall collection stations distributed in rural areas at which rainfall chemistry is measured at weekly intervals (Hayman et al., 2004a). Annual rainfall and annual mean volume-weighted concentration of inorganic N (NO3− + NH4+) ([N]m) were available for these sites from which annual mean total wet inorganic N deposition (Nm) could be calculated by summation. This procedure is likely to underestimate N deposition because this type of collector does not catch all rainfall, and the catch efficiency will vary between sites but, unlike the modelled data, it does not rely on model assumptions. Using three-yearly mean values for 2002–2004, subsets of sites were selected representing gradients in Nm and [N]m (Fig. 1c,d; Table 1) as described earlier for modelled data.
We selected sites using data for wet deposition and concentration in rainfall of inorganic N, rather than data for total N deposition, for two reasons. First, wet deposition is measured whereas dry deposition is derived from transfer models and, hence, modelled interpolated values of dry N deposition contain large uncertainties (Magnani et al., 2007). Second, Hyvärinen & Crittenden (1998a) found that the effect of different components of total N deposition on [N]lichen were not all additive: NO3− and NH4+ wet depositions were strong covariables and both strong correlates with [N]lichen but were interchangeable rather than complementary; modelled NH3 deposition did not explain [N]lichen while overall, wet NO3− deposition, with or without measured NO2 concentrations in air, was the variable yielding the highest explanatory power. Note that Marner & Harrison (2004) report that in the West Midlands of England wet deposited NO3− + NH4+ represented 50–70% of total N deposition.
Heathlands and moorlands supporting C. portentosa were sought within each of the 5 × 5 km grid squares selected from the modelled data set (13 sites), and as close as possible to each of the selected rainfall collection stations (9 sites) (Figs 1b,d, 2, Table 1).
Collection and pretreatment of C. portentosa
At each site 10 replicate samples of C. portentosa were collected, each one typically a clump of podetia 50–150 mm in diameter. The samples were collected from open areas to minimize tree canopy effects and from locations > 10 m apart. Powder-free latex gloves were worn at all times when handling lichens both in the field and in the laboratory to minimize contamination. Samples were transported to the laboratory in either seed trays or polythene bags, and air-dried for 24–48 h at room temperature. Material collected for preliminary studies was stored in seed trays in a growth room at 10°C and under a 12 h light (photosynthetic photon fluence rate over the waveband 400–700 nm of 50–200 nmols m−2 s−1)/12 h dark cycle and assayed for phosphatase activity within 3 wk of collection. Collections made during subsequent field surveys were air-dried at room temperature, sealed in polyethylene bags and stored at either 4°C (for chemical analysis) or −15°C (for phosphatase assays).
Thalli were rehydrated overnight in water-saturated air (over water in desiccators) at 10°C, then saturated by spraying lightly with deionized water and cleaned of extraneous debris. Thalli from each replicate sample were then dissected into horizontal strata by cutting with a razor blade at some or all of the following distances downwards from the apices: 5, 10, 15, 25, 35, 40, 50, 65, 80 mm.
Phosphomonoesterase and phosphodiesterase (PDE) activities were determined using pNPP and bis-p-nitrophenyl phosphate (bis-pNPP), respectively, as substrates, as described by Turner et al. (2001). Samples of C. portentosa were added to 2.9 ml citric acid-trisodium citrate buffered assay medium made up in simulated rainfall containing major ions representative of UK precipitation (20 mmol l−1 MgSO4.7H2O, 8 mmol l−1 CaCl2.2H2O, 150 mmol l−1 NaCl, 15 mmol l−1 NH4NO3, 5 mmol l−1 KNO3) (Hayman et al., 2004a). Assays were initiated by the addition of 0.1 ml analogous substrate, to yield a final concentration of either 0.5 mM, 3 mM or other values where stated. Samples were then placed in a shaking water-bath at 15°C for 20 min in the dark after which the reaction was terminated by transferring 2.5 ml assay medium into 0.25 ml terminator solution (1.1 M NaOH, 27.5 mM EDTA, 0.55 M K2HPO4) and the absorbance measured at 405 nm using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). Thalli were then blotted dry, oven dried for 24 h at 80°C and weighed. Enzyme activity was expressed as mmol substrate hydrolysed g−1 dry mass h−1 using p-nitrophenol to calibrate the assay. No-lichen and no-substrate controls were included. Values of assay medium pH, incubation temperature, duration of assay and substrate concentration were all varied in a series of preliminary experiments and the data used to select conditions for subsequent assays.
Enzyme labelling and preparation of material for fluorescence microscopy
Thalli of C. portentosa were rehydrated as described earlier. Individual branches were removed and shaken gently in enzyme-labelled fluorescent phosphatase substrate ELF 97 phosphate (ELF 97 Endogenous Phosphatase Detection Kit; Thermo Fisher Scientific) for 20 min at 15°C, the solution being prepared by adding 100 μl of ELF 97 phosphate stock solution to 1.9 ml 0.02 M citric acid-trisodium citrate buffer (pH 2.5). Like pNPP, ELF 97 is an organic phosphate that is readily hydrolysed by PMEs; it yields an insoluble fluorescent product that precipitates at the site of enzyme action (Paragas et al., 2002). After exposure, the samples were washed twice in water for 5 min. Hand-cut sections were prepared at this stage while material for thin sectioning was fixed in 2% (v : v) glutaraldehyde in 0.1 M sodium phosphate buffer (pH 7.0) at 4°C and processed as described by Davey et al. (1993). Sites of PME activity were visualized by fluorescence microscopy using a Zeiss LSM uv META Kombi confocal on a Zeiss Axiovert 100M microscope fitted with 350 nm excitation and 505 nm emission filters.
Thallus surface pH
Dry thalli were hydrated as already described and the apical 15 mm stratum selected for measurement. Each thallus was pressed against the bottom of a Petri dish using two microscope slides positioned 15 mm apart and 0.025 M KCl was added to the exposed thallus between the slides (to enhance measurement stability) until a visible water film was formed on the thallus surface. A flat tip pH electrode (Gelplas double junction flat tip electrode; VWR International Ltd, Lutterworth, UK) was placed firmly against the wet thallus and the pH was noted at 15 s intervals for 5 min. The electrode output was stable after 1 min at which time the true value was considered to have been recorded.
Determination of total N and P concentration
Samples of the apical 5 mm and the 40–50 mm stratum of lichen thalli were oven dried at 80°C, weighed and digested using the sulphuric acid–hydrogen peroxide procedure of Allen (1989). Ammonium-N in the digest was determined using the fluorometric method of Holmes et al. (1999) and a Wallac 1420 VICTOR2 multilabel counter (PerkinElmer LAS (UK) Ltd, Beaconsfield, UK). Phosphorus was assayed by the malachite green variant of the methylene blue method after van Veldhoven & Mannaerts (1987) using a Pye Unicam SP6-350 visible spectrophotometer. Note that the apical 5 mm and the 40–50 mm stratum were those parts of the thallus selected for analysis by Hyvärinen & Crittenden (1998a) in an earlier survey of the same species and thus their use here facilitated comparisons between the results of the two studies.
SPSS (SPSS (UK) Ltd, Woking, UK) was used to perform standard statistical analyses and all data were checked for normality and homogeneity of variances. Relationships between lichen chemistry and variables describing atmospheric N inputs were subjected to one-way ANOVA, correlation analyses or linear regression. Where test assumptions were not met, data were either log-transformed or nonparametric correlation analysis was applied.
Characteristics of phosphatase activity in C. portentosa
Cladonia portentosa produces both acid PME and PDE activity. Phosphomonoesterase activity was significantly (P =0.007) higher than PDE activity by a factor of c. 4 at each pH interval investigated (Fig. 3). For both PME and PDE, activity was readily measurable across the entire pH range but with a significant trend (P <0.05, as revealed by anova) for increasing activity with decreasing pH. Maximum values for both enzyme systems were recorded at pH c. 2.5, an optimum broadly consistent with the measured thallus surface pH of 3.09 ± 0.07 (mean ± 1 SE, n =4) recorded after 1 min in 0.025M KCl.
Phosphomonoesterase activity was readily measurable to a depth of 80 mm below the apices, which was the lowermost stratum assayed. However, activity was maximal in the apices with rates below 40 mm being lower by c. 40% (Fig. 4). Initial experiments confirmed that the rate of the phosphatase-mediated reaction responded linearly to increasing temperature between 5 and 50°C, and that product generation was linear with time up to 60 min. PME activity followed Michaelis–Menten kinetics in its response to substrate concentration and the value of 0.5 mM pNPP used in these preliminary tests was not enzyme saturating. The PME in C. portentosa collected from a low Ns value site (4.2 kg N ha−1 yr−1, The Halsary, Table 1) had Km and Vmax values, calculated from a Hanes–Woolf plot, of 0.878 mM and 1.111 mmol substrate hydrolysed g−1 dry mass h−1, respectively, whereas in samples from a high Ns value site (19.3 kg N ha−1 yr−1, Ridgewalk Moor, Derbyshire, GR SK 147952, UK) PME had Km and Vmax values of 2.221 mM and 3.333 mmol substrate hydrolysed g−1 dry mass h−1, respectively. On the basis of these initial results we selected the following assay conditions for use in the regional survey: an ecologically relevant temperature of 15°C, an incubation period of 20 min (during which the rate of hydrolysis was likely to be linear in all samples), a pH value of 2.5 and a substrate concentration of 3 mM. These conditions produced a readily detectable quantity of product using a relatively low ratio of thallus mass : assay medium volume (typically 15–30 mg : 3 ml). The pNPP concentration selected was either saturating or near-saturating for all samples and avoided excessive interference from high blank values associated with higher substrate concentrations.
The results were obtained using freshly collected lichen material stored air-dry in a growth room (see the Materials and Methods section). However, the initial enzyme activity at the time of collection could be maintained by storage at −15°C in the air dry state for up to 2 months and hence this method was used to store subsequent lichen collections used in the following subsections.
Location of PME activity in C. portentosa
Phosphomonoesterase activity was concentrated over the outer and inner surfaces of the hollow tube-like branches of C. portentosa thalli, and was most clearly visualized in hand-cut transverse sections (Fig. 5b). Detailed observations on thin sections under high magnification suggested that PME activity was exclusively associated with the fungal symbiont (Fig. 5c) and concentrated at distinct spots, or centres of activity, along the length of hyphae. These centres of activity were associated with hyphal lumina (Fig. 5d) consistent with a location in the inner region of the cell wall and/or on the cell membrane, or in the protoplast. The use of ELF 97 phosphate did not reveal PME activity on or in photobiont cells.
Relationships between PME activity, lichen chemistry and N enrichment
Phosphomonoesterase activity assayed in the apical 10 mm of C. portentosa was positively related to N income (Fig. 6a,c; Table 2). Activity varied roughly by a factor of 2 between sites with the lowest and highest Ns values. Note that the goodness of fit of the regression line in Fig. 6a can be improved by using ln PME activity as the independent variable (r2=0.93). The PME activity was also weakly negatively related to [N]ppt (Fig. 6b,d; Table 2). However, the selected gradients in increasing N and decreasing [N]ppt are both also gradients in increasing rainfall, and there are strong positive relationships between PME activity and precipitation among the sites selected to represent gradients in Ns, Nm and [N]m (Table 2).
Table 2. Bivariate Pearson correlation coefficients (r) between phosphomonoesterase (PME) activity, nitrogen (N) and phosphorus (P) concentrations in Cladonia portentosa and both modelled (s) and measured (2002–2005) (m) values of mean inorganic wet N deposition (Ns, Nm), volume weighted inorganic N concentration in precipitation ([N]s, [N]m) and precipitation
*Correlation is significant at the 0.05 level; **correlation is significant at the 0.01 level.
†Spearman’s correlation coefficient (nonparametric equivalent of the Pearson’s correlation).
Thallus N concentration in both the apical 5 mm ([N]apex) and the basal stratum 40–50 mm downwards from the apices ([N]base) were strongly positively correlated with both Ns and Nm, and with rainfall (Fig. 7a,c; Table 2), but of the two lichen measurements [N]apex was the stronger correlate. Relationships between [N]lichen and [N]s or [N]m were generally weaker: neither [N]apex nor [N]base were significantly related to [N]m (Fig. 7d) while only [N]base was significantly related to [N]s (Fig. 7b, Table 2).
Relationships between [P]lichen and N enrichment were less clear than those for [N]lichen. There were no obvious trends between [P]lichen and Nm (Fig. 7g) but both [P]apex and [P]base were negatively related to Ns (Fig. 7e, Table 2). Despite the lack of consistent relationships between [P]lichen and N deposition, the N : P mass ratio in both lichen apices and bases was strongly positively correlated with Ns (but not significantly so with Nm) (Fig. 8), increasing by a factor of 2.5 over the range of modelled deposition values. In turn, PME activity was also broadly related to ([N] : [P])lichen which varied by a factor of 3 in lichen bases and by a factor of 2 in the apices (Fig. 9).
Correlations between PME activity, [N]lichen and [P]lichen and N deposition parameters were frequently stronger for the modelled data sets than for measured data. In five of the 10 regressions that utilize measured data (Figs 6, 7) the relationships improved when they were recalculated using the modelled values pertaining to the lichen collection sites. However, this procedure also frequently changed the order of sites on the N and [N]ppt gradients such that some of these sites might not have been selected if the original choice of sites had been on the basis of modelled (rather than measured) data. Modelled values spanned a larger range (Ns = 4.1–32.8 kg N ha−1 yr−1; [N]s=0.3–1.9 mg l−1) than measured values (Nm = 2.3–9.3 kg N ha−1; [N]m = 0.4–1.3 mg l−1, respectively (2002–2005 3-yr mean values)).
Characteristics of phosphatase activity in C. portentosa
The present results for C. portentosa compare favourably with data of Lane & Puckett (1979) for C. rangiferina which had acid phosphatase (optimum pH = 2.2) with a Km value of 8.9 mM. These authors also showed the epiphytic lichen Lobaria pulmonaria to have an acid phosphatase with a similar pH optimum. However, it should be noted that in some lichen species (e.g. Xanthoria spp. and Peltigera spp.) the optimum pH for PME activity is in the range 5–6 (Stevenson, 1994; E. J. Hogan & P. D. Crittenden, unpublished) and thus approach an alkaline class of PME (Whitton et al., 2005). The apices in Cladonia (subspecies Cladina) are the most metabolically active region of the thallus with maximum rates of photosynthesis and respiration, and contain maximum concentrations of N and P (Hyvärinen & Crittenden, 1998a; Ellis et al., 2003). Therefore, it is consistent that this region of the thallus has the maximum phosphatase activity on a dry mass basis. Note also that because PME activity is concentrated in hyphae on or near the thallus surfaces (Fig. 5b), the higher surface area to mass ratio in the finer apical branches might also explain the maximal activity on a dry mass basis in the apices.
Phosphomonoesterase activity in the apical 10 mm of C. portentosa ranged between 0.721 and 3.372 mmols pNPP hydrolysed g−1 dry mass h−1. These values can be compared with documented activities associated with plant roots and fungal mycelia ranging from ≤ 1 μmol substrate hydrolysed g−1 dry mass h−1 (Helal, 1990; Jayakumar & Tan, 2005) to 7.5 mmol g−1 dry mass h−1 in the mycorrhizal fungus Hymenoscyphus ericae (Gibson & Mitchell, 2005) (note that the value of 1.5 mmol g−1 fresh mass h−1 for Agrostis capillaris roots reported by Johnson et al. (1999) might be higher) and with those associated with bryophytes in the range 0.02–4.0 mmol substrate hydrolysed g−1 dry mass h−1 (Press & Lee, 1983; Phuyal et al., 2008). However, the large majority of published values are < 1 mmol g−1 dry mass h−1 (Hogan, 2009) and thus the rates of PME activity in C. portentosa appear to be among the higher values recorded in the literature. Further, as we have found rates of PME activity in other lichens species (e.g. Stereocaulon alpinum, Teloschistes capensis and U. sphacelata; E. J. Hogan & P. D. Crittenden, unpublished) broadly similar to those in C. portentosa, it seems probable that lichens as a group have an exceptional capacity for PME activity.
Use of the fluorescent substrate ELF 97 phosphate suggested that PME activity was predominantly, if not entirely, a property of the mycobiont and showed staining to be concentrated in hyphae closest to the thallus surfaces (Fig. 5b,c). High activity on the outer surfaces of the thallus could be ecologically advantageous if it resulted in hydrolysis of organic P-containing compounds in solutes and particles deposited onto the lichen. However, it is perhaps surprising that PME activity is also concentrated on the surfaces of the inner cavity of the thallus which is not exposed to the atmosphere. One explanation for the distribution of enzyme activity over all surfaces is that perforations that frequently occur in the axils of branches might allow significant quantities of rainfall and particulates into the interior of the thallus. However, it has also been suggested (J. R. Leake, pers. comm.) that a primary function of extracellular phosphatases might be conservation of cellular P resources, a function that would be consistent with the distribution of activity over all surfaces of the thallus, the upregulation of activity in response to increasing N : P ratios and the longevity of the lichen thallus (i.e. long ‘leaf’ lifespan). It is widely believed that the substrates pNPP and bis-pNPP are predominantly hydrolysed by surface (i.e. cell-wall)-bound enzymes. Evidence to support this view is that activity is sensitive to the pH of the assay solution and the hydrolysis product, p-nitrophenol, accumulates in the bathing solution (Bartlett & Lewis, 1973; Whitton et al., 2005). However, the results of staining with ELF 97 phosphate are ambiguous on this question. Fluorescence was concentrated at spots or ‘centres’ along hyphae (Fig. 5c) in a manner similar to that reported by Alvarez et al. (2006) for several ectomycorrhizal fungi. However, these centres were consistently located at hyphal lumina perhaps indicating that PME activity is located on the cell membrane and/or innermost region of the hyphal wall. Location on the cell membrane might facilitate recycling of this (and other) enzymes by endocytosis (cf. Read & Kalkman, 2003; Steinberg, 2007) which could be an advantageous trait in the N-scarce habitats occupied by lichen-forming fungi. However, the visualization of PME activity at the cell lumen also raises the possibility that a significant proportion of the pNPP hydrolysis recorded in C. portentosa might be intracellular in which case PME activity might perhaps function more broadly in the optimization of P-use efficiency; for example, by promoting turnover of cellular P.
PME activity and N enrichment
Spatial variation in N-enrichment in the natural environment is mirrored by spatial variation in PME activity in C. portentosa. Activity increased by a factor of c. 2 between sites with the lowest and highest N deposition rates, and in the case of both modelled and measured N deposition gradients. In a companion paper (Hogan et al., 2010) we demonstrate that this relationship is causal since experimental addition of N as either NH4+ or NO3− to peatland plant communities containing C. portentosa resulted in upregulation of PME activity in this lichen species. We do not know whether this upregulation is caused by an increased number of enzyme units or whether it results from a change in the relative proportions of specific PMEs being synthesized. However, values of both Km and Vmax for PME activity were higher at a high, compared with a low, N site pointing to a change in the characteristics of PME capacity and providing some support for a change in the relative abundancies of specific enzymes. Our results are in line with data for soil/plant root-associated PME activity in rainforests (Treseder & Vitousek, 2001), both acid and calcareous grasslands (Johnson et al., 1998, 1999; Phoenix et al., 2003), and in heathlands (Johnson et al., 1998; Pilkington et al., 2005) where N fertilization typically increased PME activity by factors of between 2 and 3. As the habitats for C. portentosa, a species frequently associated with Calluna vulgaris, are typically N-limited, the probable explanation for this response is that increased N availability and capture from atmospheric deposits shifts deficiency from N to P, an element that frequently becomes the limiting factor when N-limited tundra-like ecosystems are N-enriched (Aerts et al., 1992; Bowman et al., 1993; Arens et al., 2008). Moreover, Kirkham (2001) considers that N pollution has changed a substantial proportion of Calluna-dominated uplands in England and Wales from N-limited to P-limited ecosystems. This explanation is consistent with the positive coupling between PME activity and thallus N : P mass ratio (cf. Turner et al., 2003). Values of N : P mass ratio in the apices of C. portentosa generally ranged between 10 and 26 (Figs 8, 9). According to Güsewell (2004) N : P mass ratios in vascular plants > 20 are indicative of P limitation, a situation that obtained in C. portentosa at sites with Ns > 20 kg ha−1 yr−1. In this context it is interesting to note that in N2-fixing lichens, which can be considered to be relatively well-supplied with N, growth is markedly stimulated by the addition of P (Benner & Vitousek, 2007; Benner et al., 2007; McCune & Caldwell, 2009).
The capacity for PME activity was not promoted by higher concentrations of N in rainfall per se when this was not associated with increased N deposition. The trend for negative relationships between PME activity and [N]ppt among sites with similar N values (Fig. 6b,d) might be explained by the underlying rainfall differences between the sites potentially resulting in higher annual lichen growth rates, and hence greater growth-led demand for nutrients, at high precipitation sites with low [N]ppt values. There is abundant evidence that lichen growth rates are higher at sites, or in years, with higher rainfall (Kärenlampi, 1971; Fisher & Proctor, 1978; Boucher & Nash, 1990; Armstrong, 1993a,b), although it should be noted that Hyvärinen & Crittenden (1998c) found that growth in C. portentosa at physiographically distinct sites in the UK was not correlated with annual rainfall. We do not know the extent of seasonal variation in PME activity in lichens. There is some evidence of seasonal changes in PME activity in soil/root systems (Turner et al., 2002; Sardans et al., 2007) and in mosses (Turner et al., 2003) but in C. portentosa one would expect nutrient capture to be maximized throughout the year and such seasonal variation to be small.
Thallus N and P concentrations and N enrichment
It is instructive to compare the results of the present work with those of Hyvärinen & Crittenden (1998a) for C. portentosa collected between 1993 and 1994 as close as possible to 31 rainfall collections stations in the UK Acid Deposition Monitoring Network. In the present study, [N]apices was a superior predictor of Nm than [N]base, whereas Hyvärinen & Crittenden (1998a) found [N]base to be the stronger correlate. Their explanation for the accentuation of N enrichment in lichen bases was that increased N capture in the apices interferes with internal N cycling thereby reducing relocation of N from basal regions of the thallus to the apices. The anomaly between the results of this earlier survey and the present findings might reflect changes in the pollution climate during the 12-yr period between the surveys which include decreasing inorganic N and S deposition rates (Fowler et al., 2004). However, there was no significant difference between 1994 and 2006 Nm values at eight sites common to both studies. It is noteworthy that Walker et al. (2003) found that spatial variation in [N]apices, but not [N]base, in C. stellaris, reflected a putative gradient in N deposition along a 240 north–south transect in northern Russia. Thus it is possible that the usefulness of [N]lichen as a biomarker for N deposition shifts from [N]base to [N]apices as N deposition decreases.
Hyvärinen & Crittenden (1998a) found that [N] and [P] values in both thallus apices and bases of C. portentosa covaried among sites in the UK (Fig. 10a), an observation that partly prompted the present study (see the Introduction section). This trend was not confirmed in the present work (Fig. 10b). In both studies, the ranges of [N]apices, [N]base, [P]apices and [P]base values are broadly similar but the distribution of values differ (Fig. 10), perhaps reflecting the different site locations and different criteria used in each study to select collection sites. For example, sites with both high N and high [N]ppt values were omitted from the present work.
Evaluation of modelled N deposition data
In making comparisons between the efficacy of modelled and measured N deposition data in explaining variation in lichen ecophysiological variables, it should be remembered that the two series of analyses utilized C. portentosa collections from different sets of locations (Fig. 1b,d). Notwithstanding this caveat, outcomes of the regression analyses using modelled and measured data were in good general agreement. Nonetheless, and perhaps surprisingly, on some occasions modelled values proved superior explanatory variables. This might partly be because of the much larger range of Ns compared with Nm values. Two factors underlie the generally large values of Ns. First, the moorland deposition model, the output of which was used in the present study, compensates for variable collection efficiencies of rain gauges by taking into account wind-driven rain, occult precipitation, and orographic seeder–feeder effects which enhance [N]ppt and N values at high altitude. Second, among the c. 11 200 5 × 5 km grid squares for which Ns values were modelled, there is inevitably a greater range of values available than among the 38 measurement sites (Fig. 1); that is, it was possible to collect C. portentosa from sites representing almost the full range of modelled deposition values (4.1–32.8 kg N ha−1 yr−1) which might have contributed to the strong relationships observed with modelled data. Thus, for a combination of these two reasons, values of Ns always exceeded Nm.
We thank K. J. Vincent (AEA Technology plc) for placing at our disposal data on precipitation chemistry and air pollution, D. Smith (UK Meteorological Office) for provision of modelled rainfall data and P. A. Wolseley (Natural History Museum) for the loan of a flat tip pH electrode. We are especially grateful to E. C. Cocking for his guidance with the sectioning of lichen samples and subsequent microscopy, and to K. Vere and T. Self for their help with confocal microscopy. The following individuals are thanked for their help in locating heathland sites: B. J. Coppins, T. Donnelly, I. Evans, A. F. Fletcher, D. Isaac and T. F. Preece. Access to heathland and moorland sites was kindly granted by the following private land owners, individuals and organizations: D. Bruce, Countryside Council for Wales, J. Daniels, Forest Enterprise, Leicestershire County Council, E. Martin, I. Mitchell, Natural England, C. Nixon, L. Robbins, Royal Society for the Protection of Birds, Scottish Natural Heritage and Suffolk Wildlife Trust. G.M. acknowledges the receipt of a Russian Government Scholarship. This research was funded by The School of Biology, The University of Nottingham. We thank four referees for their comments on an earlier draft of this paper.