Summary We examined the effects of three foliar potassium concentrations (high, intermediate and low) on the morphol-ogy, ultrastructure and polyamine concentrations of current-year and 1- and 2-current-year-old needles of 30-current-year-old Scots pine (Pinus sylvestris L.) trees. Foliar K concentration had only a slight effect on needle morphology. The sclerenchyma cell walls were thinner, the xylem area was larger, and the resin ducts were smaller in needles with a low K concentration than in needles with a high or intermediate K concentration. In addition, the bundle sheath cells were collapsed in needles having a low K concentration. The secondary growth of phloem tissue and the mesophyll area were greater in needles with a high or intermediate K concentration than in needles with a low K concentration, possibly indicating greater production of photoassimilates in these trees. At the ultrastructural level, mesophyll cells with enlarged central vacuoles and small vacu-oles containing electron-dense material were common in need-les having a low K concentration. Enlargement of the central vacuole coincided with an exponential increase in putrescine concentration in needles with a low K concentration, suggest-ing that the central vacuole may function as a storage site for putrescine.
Keywords: needle morphology, Pinus sylvestris, potassium de-ficiency, putrescine.
Introduction
Potassium (K) is characterized by high mobility in plants; however, K uptake by plants is highly selective and closely coupled with metabolic activity (Marschner 1995). Potassium has important roles in enzyme activation, osmoregulation and carbohydrate translocation in plant cells (Lüttge and Clarkson 1989). There is evidence that cell extension is related to the K content of leaves. Thus, leaf area, cell size and turgor are lower in expanding leaves of bean plants suffering from K deficiency than in expanding leaves of bean plants well-supplied with K (Mengel and Arneke 1982). Although there have been no studies of the morphological responses of conifer needles to K deficiency, Holopainen and Nygren (1989) reported that K deficiency results in specific ultrastructural changes in Scots pine seedlings including extension of the vacuolar system,
injuries to the tonoplast structure and increased deposition of cytoplasmic lipids. However, it is not known whether similar ultrastructural changes also occur in K-deficient needles of different age classes in mature Scots pine trees.
A foliar K concentration of 3.5--4.0 mg gDW−1 is indicative of severe K deficiency in Scots pine (Pinus sylvestris L.) dur-ing the nongrowdur-ing period (Paarlahti et al. 1971). Throughout the year, the characteristic biochemical response of Scots pine to K deficiency is the accumulation of putrescine (Sarjala and Kaunisto 1993). Although putrescine accumulation is consid-ered indicative of K deficiency, it has also been shown to occur in response to various other stresses (Flores 1991).
The present study was undertaken to determine whether K deficiency induces morphological and ultrastructural changes in needles of different age classes in mature Scots pine trees. Specifically, we tested the hypotheses that (1) foliar K concen-trations alter both the ultrastructure and morphology of Scots pine needles and (2) there is a close relationship among needle microscopic structure and foliar K and putrescine concentra-tions.
Materials and methods
Site description and foliar sampling
Scots pine needles were collected in August 1992 from a fertilization experiment located at Kuru (61°55′ N, 23°44′ E) in western Finland. The site is an ombrotrophic, low-sedge open bog with a deep peat layer. It was fertilized in 1967 with 800 kg ha−1 of rock phosphate (P 115 kg ha−1) and 200 kg ha−1 of KCl (K 100 kg ha−1). Severe potassium deficiency symp-toms were observed in the late 1980s and a potassium refertili-zation trial was established in 1989 with seven treatments: unfertilized and phosphorus fertilized controls, four potassium sources of different solubility (KCl, K2CO3, KPO3, biotite) and a mixture of KCl and biotite (Sarjala and Kaunisto 1993). Plot mean height ranged from 4 to 6 m.
According to earlier observations (Sarjala and Kaunisto 1993), the needles of trees in the different plots differ widely in K and polyamine concentrations. On the basis of these results, we selected a subset of 25 trees for foliar polyamine and nutrient analyses, of which three sets of five trees
repre-Effects of foliar potassium concentration on morphology,
ultrastructure and polyamine concentrations of Scots pine needles
ANNE JOKELA,
1TYTTI SARJALA,
2SEPPO KAUNISTO
2and SATU HUTTUNEN
11
University of Oulu, Department of Biology, Botany, P.O. Box 333, FIN-90571 Oulu, Finland 2 The Finnish Forest Research Institute, Parkano Research Station, FIN-39700 Parkano, Finland
Received November 27, 1996
senting high (> 5.3 mg gDW−1), intermediate (3.5--5.0 mg gDW−1) and low (< 3.0 mg gDW−1) needle K concentrations were selected for microscopy studies of needle morphology and ultrastructure in August 1992. Needles of the selected trees sampled for the microscopy studies were generally green, but some older needles with low K concentrations often exhibited yellow tips. The polyamine, K, N and P concentrations of current-year (c), 1-year-old (c + 1) and 2-year-old (c + 2) needles were analyzed. Because needles for nutrient analyses are usually collected during the nongrowing period, foliar sampling for polyamine and nutrient analyses was performed in December 1992 from the same 15 trees. The data obtained from the nutrient analyses were used to interpret the results of the microscopy studies. Additionally, needles were sampled in September 1995 for measurements of needle length, thickness and width and for potassium and putrescine analyses.
Nutrient analyses and foliar free polyamines
Needles for nutrient analyses were taken to the laboratory in plastic bags and stored at −20 °C until analyzed. Nutrients were analyzed by methods routinely used at the Forest Re-search Institute, Parkano, Finland as described by Halonen et al. (1983). Total N was measured in oven-dried material by the Kjeldahl method. Dry-ashed material was used for the determination of K by flame atomic spectrophotometry (Varian AA-30) and for the spectrophotometric analysis of P.
Needle samples for polyamine analysis were kept in ice until taken to the laboratory and stored at −80 °C. Free polyamines (putrescine, spermidine and spermine) were extracted from the needle samples with 5% HClO4, dansylated and analyzed by HPLC as described by Sarjala and Kaunisto (1993).
Sample preparation for microscopy studies
In August 1992, needles (c = current-year, c + 1 = 1-year-old and c + 2 = 2-year-old needles) for the microscopy studies were collected in test tubes containing 0.05 M sodium cacodylate buffer (pH 7) and 1.5% glutaraldehyde + paraformaldehyde prefixative. After prefixation, 0.5-mm-thick cross sections were cut from the middle portion of each needle, postfixed with OsO4 and embedded in Ladd’s Epon as described by Reinikainen and Huttunen (1989). An ultramicrotome (Reichert Jung ULTRACUT E, Vienna, Austria) was used to cut semi-thin sections (1--3 µm) for the light microscopy studies and ultra-thin sections for electron microscopy (50--70 nm). The semi-thin sections were stained with toluidine blue and the ultra-thin sections were stained with lead citrate and uranyl acetate.
Needle morphology
Cross sections for light microscopy were taken from the mid-dle of neemid-dles that were green and had no visible symptoms of nutrient deficiency. The total number of needles examined was 111, i.e., two to three needles per needle year and eight to nine needles per tree were observed. The samples were examined with a Nikon OPTIPHOT-2 light microscope, and morphologi-cal measurements were performed with a digital image ana-lyzer (Microscale TM/TC, Digithurst Ltd., Royston, England)
and a video camera (Hitachi KP-C571 CCD color camera). The needle morphological variables measured were: needle thickness, needle width, needle area, mesophyll area per nee-dle area (%), epidermis + hypodermis area per neenee-dle area (%), central cylinder area per needle area (%), sclerenchyma cell wall thickness, phloem area per needle area (%), xylem area per needle area (%), bundle sheath cell index (radial width of bundle sheath cell/tangential width of bundle sheath cell meas-ured on the adaxial side of needle cross sections), resin duct area per needle area (%, measured on the abaxial side of needle cross sections) and resin duct number per needle area (mm−2) (Figure 1). Phloem and xylem were observed in one of the two vascular bundles of the needle. The resin duct area was based on measurements of two resin ducts from the abaxial side of the needle. The compression and shrinkage of tissues during embedding in plastic and sectioning were assumed to be mini-mal and similar in all samples (Toth 1982). Injured phloem cells in the vascular bundle were also determined. The swelling of parenchyma cells and collapse of sieve cells were regarded as phloem injuries (Fink 1991, Jokela et al. 1995).
In September 1995, current-year needles were collected from the same trees that were sampled in August and Decem-ber 1992, and the length, thickness and width of the needles were measured with a digital caliper. Needles were collected from three trees from an unfertilized control plot and from three trees from a plot fertilized with KPO3. Measurements were made on 50 needles (one needle per fascicle) from cur-rent-year branches. Thickness and width were measured in the middle region of the needle and were, therefore, comparable with the image analysis measurements of needle thickness and width made on needle cross sections in August 1992 (Fig-ure 1).
Observations at the ultrastructural level
Ultra-thin sections were examined by the scanning transmis-sion electron microscope (JEM 100CX II, JEOL, Tokyo, Ja-pan). Cell organelles and cytoplasm in mesophyll were examined in a total of 234 cells of 37 needles. The central vacuole and cytoplasm in transfusion parenchyma were exam-ined in 34 cells of 12 needles. Altogether, one to two needles per needle year, and two to six needles per tree were studied.
Statistical analysis
Figure 1. Morphological measurements made by image analysis on cross sections of Scots pine needles.
Table 1. Effects of needle age and classification by foliar K concentration based on an earlier study (Sarjala and Kaunisto 1993) on concentrations of K (mg gDW−1), P (mg gDW−1) and N (%DW) and on N/P, N/K and K/P ratios in Scots pine needles sampled from five trees per treatment in
August and December 1992. Different letters within a row indicate significant differences between foliar K concentrations (P = 0.05). Needle age: c = current-year, c + 1 = 1-year-old, and c + 2 = 2-year-old needles.
Variable Needle High K Intermediate K Low K P
age Mean ± SD Mean ± SD Mean ± SD
August
K (mg gDW−1) c 5.95 ± 0.79 a 4.12 ± 0.47 b 2.55 ± 0.30 c 0.000
c + 1 4.73 ± 0.80 a 3.33 ± 0.68 b 2.41 ± 0.10 c 0.000
c + 2 4.35 ± 0.80 a 2.95 ± 0.60 b 1.86 ± 0.15 c 0.000
P (mg gDW−1) c 1.74 ± 0.35 a 1.35 ± 0.10 b 1.72 ± 0.19 a 0.037
c + 1 1.39 ± 0.26 1.17 ± 0.19 1.43 ± 0.16 0.145
c + 2 1.28 ± 0.23 1.17 ± 0.16 1.06 ± 0.24 0.305
N (%DW) c 1.17 ± 0.08 a 1.12 ± 0.05 a 1.35 ± 0.05 b 0.000
c + 1 1.08 ± 0.06 a 1.05 ± 0.09 a 1.38 ± 0.13 b 0.000
c + 2 1.08 ± 0.05 1.07 ± 0.12 1.24 ± 0.27 0.299
N/P c 6.9 ± 1.2 8.3 ± 0.5 7.9 ± 1.0 0.071
c + 1 8.0 ± 1.3 9.1 ± 1.1 9.7 ± 1.0 0.100
c + 2 8.4 ± 1.4 9.2 ± 0.9 12.3 ± 4.6 0.118
N/K c 2.0 ± 0.3 a 2.7 ± 0.3 b 5.3 ± 0.9 c 0.000
c + 1 2.3 ± 0.3 a 3.2 ± 0.5 b 5.7 ± 0.6 c 0.000
c + 2 2.7 ± 0.3 a 3.7 ± 0.4 b 6.8 ± 1.9 c 0.003
K/P c 3.5 ± 0.8 a 3.1 ± 0.4 a 1.5 ± 0.2 b 0.000
c + 1 3.5 ± 0.9 a 2.9 ± 0.6 a 1.7 ± 0.2 b 0.002
c + 2 3.5 ± 0.8 a 2.5 ± 0.4 b 1.8 ± 0.4 c 0.002
December
K (mg gDW−1) c 5.99 ± 0.85 a 4.44 ± 0.35 b 3.04 ± 0.36 c 0.000
c + 1 5.55 ± 1.02 a 3.95 ± 0.40 b 3.09 ± 0.23 c 0.000
P (mg gDW−1) c 1.94 ± 0.49 a 1.44 ± 0.17 b 1.91 ± 0.12 a 0.047
c + 1 1.86 ± 0.43 1.36 ± 0.24 1.70 ± 0.16 0.056
N (%DW) c 1.25 ± 0.12 a 1.18 ± 0.05 a 1.40 ± 0.07 b 0.004
c + 1 1.24 ± 0.07 1.35 ± 0.47 1.42 ± 0.07 0.593
N/P c 6.8 ± 1.6 8.2 ± 0.8 7.4 ± 0.8 0.406
c + 1 6.9 ± 1.5 10.5 ± 5.3 8.4 ± 1.1 0.422
N/K c 2.1 ± 0.3 a 2.7 ± 0.2 b 4.6 ± 0.5 c 0.000
c + 1 2.3 ± 0.4 a 3.5 ± 1.7 ab 4.6 ± 0.4 b 0.014
K/P c 3.3 ± 0.9 a 3.1 ± 0.4 a 1.6 ± 0.3 b 0.001
Results
Nutrient concentrations
Trees classified on the basis of an earlier study (Sarjala and Kaunisto 1993) as having high, intermediate or low foliar K concentrations fell into the same categories on the basis of measurements made in the present study (Table 1). In trees in the intermediate foliar K class, most macronutrients were present in optimal concentrations, with the exception that P concentration and the N/K ratio were well below optimum values (Table 1). Optimum P concentrations were present in needles in trees of the high and low foliar K classes. Foliar N concentrations were below the deficiency limit in trees of the intermediate and high foliar K classes (N, N/P and N/K, Table 1), whereas the K/P ratio was in balance. The N/K and
K/P ratios indicated a K deficiency in needles of trees in the low foliar K class.
For both the August and December samplings, K and P concentrations decreased with increasing needle age. Both K and P concentrations were higher in December than in August for all needle age classes and for all foliar K classes.
Needle morphology
Foliar K concentration had only a slight effect on needle morphology. Needle thickness, width (Tables 2 and 3) and length (Table 3) were not significantly affected by foliar K concentration. Needles with a low K concentration had the largest needle area, whereas needles with a high K concentra-tion had the largest relative area of mesophyll and the smallest central cylinder area (Table 2). The sclerenchyma cell walls were thinnest in needles with a low K concentration
(Fig-Table 2. Light microscopic image analysis measurements of needle morphological variables. Different letters within a row indicate significant differences between the foliar K concentration classes (P = 0.05). Needles were collected in August 1992. Needle age: c = current-year, c + 1 = 1-year-old, and c + 2 = 2-year-old needles; and n = number of trees observed.
Variable Needle High K Intermediate K Low K P
age n Mean ± SD n Mean ± SD n Mean ± SD
Needle thickness c 5 0.781 ± 0.033 4 0.811 ± 0.064 5 0.812 ± 0.066 0.622
(mm) c + 1 5 0.801 ± 0.069 5 0.772 ± 0.017 5 0.802 ± 0.093 0.878
c + 2 5 0.780 ± 0.066 5 0.774 ± 0.026 5 0.808 ± 0.104 0.613
Needle width (mm) c 5 1.671 ± 0.069 5 1.705 ± 0.161 5 1.766 ± 0.189 0.600
c + 1 5 1.690 ± 0.104 5 1.540 ± 0.420 5 1.824 ± 0.238 0.651
c + 2 5 1.647 ± 0.182 5 1.681 ± 0.149 5 1.658 ± 0.357 0.977
Needle area (mm2) c 5 1.071 ± 0.870 4 1.110 ± 0.153 5 1.210 ± 0.268 0.491
c + 1 5 1.105 ± 0.139 5 0.976 ± 0.248 5 1.203 ± 0.262 0.445
c + 2 5 1.122 ± 0.205 5 1.082 ± 0.086 5 1.149 ± 0.368 0.913
Mesophyll area (%) c 5 57.59 ± 2.59 4 56.63 ± 1.27 5 57.12 ± 2.11 0.677
c + 1 5 59.65 ± 2.03 5 57.57 ± 2.71 5 56.68 ± 1.58 0.075
c + 2 5 58.20 ± 2.29 5 57.96 ± 3.39 5 57.63 ± 1.29 0.763
Epidermis + hypo- c 5 12.68 ± 1.59 4 11.41 ± 0.53 5 11.95 ± 1.58 0.432
dermis area (%) c + 1 5 10.95 ± 1.06 5 10.79 ± 1.44 5 11.78 ± 1.29 0.326
c + 2 5 12.11 ± 1.25 5 11.33 ± 0.57 5 11.83 ± 1.32 0.566
Central cylinder c 5 29.72 ± 1.85 4 31.96 ± 1.43 5 30.93 ± 3.28 0.298
area (%) c + 1 5 29.40 ± 2.16 5 31.64 ± 2.00 5 31.58 ± 2.56 0.310
c + 2 5 29.69 ± 2.02 5 30.71 ± 2.98 5 30.55 ± 1.43 0.733
Sclerenchyma cell c 4 6.02 ± 0.58 4 5.66 ± 0.91 5 6.44 ± 1.22 0.359
wall thickness (µm) c + 1 5 5.73 ± 0.66 5 5.12 ± 0.49 5 5.61 ± 0.88 0.364
c + 2 5 5.73 ± 0.71 a 5 4.68 ± 0.42 b 5 4.63 ± 0.67 b 0.026
Phloem area (%) c 5 0.54 ± 0.09 5 0.60 ± 0.06 5 0.62 ± 0.01 0.566
c + 1 5 0.76 ± 0.03 5 0.73 ± 0.02 5 0.70 ± 0.01 0.878
c + 2 5 0.90 ± 0.15 5 0.90 ± 0.18 5 0.80 ± 0.11 0.468
Xylem area (%) c 5 0.60 ± 0.01 5 0.58 ± 0.05 5 0.67 ± 0.01 0.275
c + 1 5 0.60 ± 0.11 5 0.69 ± 0.31 5 0.66 ± 0.16 0.878
c + 2 5 0.62 ± 0.09 5 0.60 ± 0.07 5 0.66 ± 0.13 0.733
Bundle sheath c 5 0.57 ± 0.10 4 0.58 ± 0.10 5 0.56 ± 0.08 0.650
cell index c + 1 5 0.62 ± 0.03 5 0.66 ± 0.08 5 0.60 ± 0.04 0.185
c + 2 5 0.73 ± 0.04 5 0.76 ± 0.24 5 0.72 ± 0.08 0.651
Resin duct area (%) c 5 0.91 ± 0.17 5 0.99 ± 0.17 5 0.87 ± 0.11 0.185
c + 1 5 0.76 ± 0.12 5 1.05 ± 0.33 5 0.73 ± 0.18 0.114
c + 2 5 0.72 ± 0.13 5 0.76 ± 0.12 5 0.84 ± 0.28 0.810
Resin duct c 5 9.69 ± 1.26 5 9.04 ± 2.08 5 8.19 ± 1.15 0.249
number mm−2 c + 1 5 8.86 ± 0.83 5 8.94 ± 4.24 5 7.92 ± 0.56 0.196
ure 2a) and thickest in 2-year-old needles with a high K con-centration (Figure 2b; P = 0.026, Table 2).
The phloem area was greater in 1- and 2-year-old needles than in current-year needles, and the difference was significant in needles with a high K (P = 0.025) or an intermediate K concentration (P = 0.022) but not in needles with a low K
concentration (P = 0.098). In contrast, needle age and foliar K concentration had few significant effects on xylem area, al-though xylem area was larger in needles with a low K concen-tration than in needles with an intermediate or a high K concentration. When calculated in relation to the central cylin-der area, the phloem and xylem areas showed the same rela-tionship as for whole needle area (data not shown). There was no relationship between phloem cell injury and foliar K con-centration.
The bundle sheath cell index was smallest (radial to tangen-tial width of cell was smaller) in needles with a low K concen-tration and was independent of needle age (Table 2). In needles with a low K concentration, the bundle sheath cells were collapsed (Figure 2c), whereas the bundle sheath cells were normal in shape in needles with a high K concentration (Fig-ure 2d). The resin duct area was smallest in current-year and 1-year-old needles with a low K concentration, and the number of resin ducts was smallest in needles with a low K concentra-tion and was independent of needle age.
Observations at ultrastructural level
Ultrastructural observations of the mesophyll (Table 4) re-vealed an enlarged central vacuole in current-year and 1- and
Table 3. Leaf morphological variables (measured with a digital cali-per) and potassium (K) and putrescine concentrations of current-year needles collected in September 1995. Fifty needles per tree were measured on three trees per plot.
Variable High K Low K P
Mean ± SD Mean ± SD (t-test)
Needle length (mm) 37.03 ± 9.23 42.78 ± 4.64 0.389
Needle thickness 0.67 ± 0.02 0.65 ± 0.04 0.426 (mm)
Needle width (mm) 1.46 ± 0.02 1.50 ± 0.10 0.562
K mg gDW−1 5.58 ± 0.10 2.37 ± 0.33 0.002
Putrescine 150.26 ± 52.77 2392.74 ± 1568.10 0.131 nmol gFW−1
Table 4. Effect of foliar K concentration class on ultrastructure of needles collected in August 1992. Needle age: c = current-year, c + 1 = 1-year-old, and c + 2 = 2-year-old needles.
Variable High K Intermediate K Low K
c c + 1 c + 2 c c + 1 c + 2 c c + 1 c + 2
MESOPHYLL
Central vacuole
Enlarged central vacuole --1 -- +2 -- -- + + + +
Proliferation of tonoplast + + + + ++3 + -- + +
Chloroplast
Swelling of thylakoids + + -- + + + ++ + +
Cytoplasm
Lipid accumulations + ++ ++ + + + + ++ ++
Small vacuoles with -- + + -- -- -- -- ++ +
electron-dense material
Extensive vesiculation -- -- + -- -- -- -- + +
Mitochondria
Swelling of mitochondria -- -- + -- -- -- -- -- +
TRANSFUSION PARENCHYMA
Central vacuole
Enlarged central vacuole -- + -- -- -- + -- +
--Cytoplasm
Lipid accumulations -- + -- -- -- -- -- + +
1
-- = Not observed.
2
+ = Occasionally observed.
3 ++ = Frequently observed.
2-year-old needles with a low K concentration. Proliferation of the tonoplast was slightly less frequent in needles with a low K concentration than in needles with an intermediate or high K concentration. The formation of small vacuoles with elec-tron-dense material in mesophyll cells (Figure 3a) was ob-served in needles with a low or high K concentration, but not in needles with an intermediate K concentration. Sometimes, these small vacuoles gave the cell a vesiculated appearance (extensive vesiculation in Table 4) (Figure 3b).
The swelling of chloroplast thylakoids and occurrence of lipid accumulations in the cytoplasm were abundant and inde-pendent of foliar K concentration. Lipid accumulations were often adjacent to the tonoplast (Figure 3c). Slight mitochon-drial swelling was observed in the 2-year-old needles with a low or high K concentration. Cytoplasmic lipid accumulations in the transfusion parenchyma occurred slightly more fre-quently in needles with a low K concentration than in needles with an intermediate or high K concentration (Figure 3d), whereas an enlarged central vacuole in transfusion paren-chyma cells was observed at all foliar K concentrations.
Free polyamine concentrations
Polyamine analysis of current-year and 1- and 2-year-old needles of the 25 selected trees indicated a negative correlation between putrescine and K concentrations that declined with needle age (regression analysis: current-year needles y = 22191x−3.20 and r2 = 0.622, 1-year-old needles y = 138019x−5.33 and r2 = 0.554, 2-year-old needles y=9154x−2.97 and r2 = 0.313; x = K concentration and y = pu-trescine concentration). In 1992, pupu-trescine concentrations in needles with a low K concentration were significantly higher
than putrescine concentrations in needles with an intermediate or high K concentration (Figure 4). Spermidine and spermine concentrations were lower in needles with a low K tion than in needles with an intermediate or high K concentra-tion (Figure 4). The changes in polyamine concentraconcentra-tion in response to foliar K concentration were independent of needle age and were observed in both August and December, except for spermine concentration, which was lower in December than in August for all needle age classes. In September 1995, putrescine concentration in needles with a low K concentration was higher than in needles with a high K concentration (Ta-ble 3), but the difference was not statistically significant.
Discussion
Foliar K concentration had only slight effects on needle length, thickness and width. The higher proportion of the mesophyll area in needles with a high K concentration may indicate that these needles had a high photosynthetic potential, which in turn may have affected tree growth. Potassium fertilization has been shown to increase the volume growth and basal area of Scots pine growing on peatlands (Kaunisto 1989). Moreover, Baillon et al. (1988) reported that photosynthetic rates were lower in K-deficient Norway spruce seedlings than in seed-lings receiving an adequate supply of K; however, in Scots pine seedlings, photosynthetic rate was not inhibited until foliar K concentration was reduced to 2.4 mg gDW−1 (Nygren and Hari 1992). Neither photosynthesis nor tree growth was measured in our study.
The thin cell walls of the sclerenchyma tissue in 2-year-old needles with a low K concentration may indicate injury similar to that found in needles deficient in boron, potassium or phos-phorus (Raitio 1979, 1981). The development of thin scler-enchyma cell walls may be a response to nutrient imbalance caused by an excess N supply (Jokela et al. 1995).
The enhanced formation of secondary phloem in needles with an intermediate or high K concentration compared to needles with a low K concentration may have been a result of enhanced photosynthesis caused by K fertilization (Dünisch and Bauch 1994). Furthermore, Hartt (1969) showed that K has an important role in photoassimilate transport in the phloem of sugar cane. The formation of secondary phloem but not secondary xylem in Scots pine needles is in agreement with the findings of Ewers (1982), who reported that cambium in the vascular bundle produces secondary phloem but not secon-dary xylem in several conifer species including Scots pine.
The finding that the xylem area was slightly larger in cur-rent-year needles with a low K concentration than in current-year needles with an intermediate or a high K concentration is of interest because polyamine transport takes place in the xylem (Bagni and Pistocchi 1991). Putrescine enhances xylo-genesis of xylem cells in Helianthus tuber (Phillips et al. 1988) and it may influence cambial activity in gymnosperms. For example, in the stem cambial zone of Norway spruce, putre-scine is the most abundant polyamine during the period of greatest cambial activity in spring and summer (Königshofer 1991).
Although phloem injuries have often been attributed to nu-trient imbalance (Fink 1991, 1993, Jokela et al. 1995, 1996), we found no effect of foliar K concentration on phloem cell integrity. The slightly collapsed bundle sheath cells of needles with a low K concentration may be related to nutrient stress in general or K deficiency in particular. Collapsed bundle sheath cells have been associated with the accumulation of secondary substances, and modifications in the bundle sheath structure have been found in response to fumigation with SO2 and O3 (Maier-Maercker and Koch 1992). The decrease in size and number of resin ducts with decreasing foliar K concentration may indicate that the terpene-based defensive mechanism is compromised in trees suffering from K deficiency (Burr and Clancy 1992).
In our study, foliar K concentration influenced the size of the central vacuole of mesophyll cells but had no effect on the occurrence of tonoplast injuries or lipid accumulation. Lipid accumulations were frequently observed but they were inde-pendent of foliar K concentration. They were often located adjacent to the tonoplast, and may indicate tonoplast disinte-gration (Zwiazek and Shay 1987) or cold hardening during late summer (Holopainen et al. 1992). Holopainen and Nygren (1989) observed that K deficiency causes an extension of the vacuolar system in mesophyll cells, fragmentation of the tonoplast and an increase in cytoplasmic lipids in needles of Scots pine seedlings. According to Holopainen and Nygren (1989), small vacuoles with tannin deposits are associated with K deficiency. We observed many small vacuoles with electron-dense material that gave the cytoplasm a vesiculated appear-ance in needles with a low K concentration. We conclude that the vacuole system may be particularly sensitive to K defi-ciency, because K is important in osmoregulation (Lüttge and Clarkson 1989).
Changes in the vacuolar system could also be explained by changes in the polyamine concentration of K-deficient need-les. We observed increased concentrations of putrescine and decreased concentrations of spermidine and spermine in need-les with a low K concentration (cf. Sarjala and Kaunisto 1993). Flores (1991) suggested that the accumulation of putrescine, possibly in the vacuole (Pistocchi et al. 1988), compensates for the decrease in pH of cells of K-deficient plants.
Changes in Scots pine needle structure caused by K defi-ciency differed from those caused by P (injuries to mitochon-drial structure) and N deficiencies (changes in the cytoplasm and chloroplast structure) (Palomäki and Holopainen 1994, 1995). Interpretation of the symptoms of K deficiency is com-plicated because of the effects of foliar K concentration on the balance of the other macronutrients. For example, needle N was at an adequate concentration for growth in needles with a low K concentration, whereas needles with an intermediate or high K concentration were deficient in N (cf. Paarlahti et al. 1971, Kaunisto 1987). Differences between nutrient interac-tions in Scots pine trees growing in the field and those that occur in seedlings growing under controlled environment con-ditions may account for the differences between our observa-tions and those reported by Holopainen and Nygren (1989).
We conclude that foliar K concentration has a small effect on needle morphology. Needles with a low K concentration were characterized by thinner sclerenchyma cell walls, a slightly larger xylem area, less secondary growth of phloem, smaller resin ducts and more collapsed bundle sheath cells than needles classified as containing an intermediate or high K concentration. A large accumulation of putrescine occurred in needles with a low K concentration and was accompanied by the development of a vacuolar system in mesophyll cells.
Acknowledgments
This research was supported by the Kone Foundation and the Finnish Research Program on Climate Change. Dr. Jaana Bäck and Dr. Sagar V. Krupa are thanked for their valuable comments on the manuscript as are Mrs. Tellervo Siltakoski, Mrs. Mervi Saaranen and M.Sc. Eija Kukkola for technical assistance. The language of the manuscript was revised by Mrs. Leena Kaunisto.
References
Bagni, N. and R. Pistocchi. 1991. Uptake and transport of polyamines and inhibitors of polyamine metabolism in plants. In Biochemistry and Physiology of Polyamines in Plants. Eds. R.D. Slocum and H.E. Flores. CRC Press, Boca Raton, FL, pp 105--120.
Baillon, F., X. Dalschaert, S. Grassi and F. Gaiss. 1988. Spruce photosynthesis: possibility of early damage diagnosis due to expo-sure to magnesium or potassium deficiency. Trees 2:173--179. Burr, K.E. and K.M. Clancy. 1992. Douglas-fir needle anatomy in
relation to western spruce budworm (Lepidoptera: Tortricidae). J. Econ. Entomol. 86:93--99.
Dünisch, O. and J. Bauch. 1994. Influence of mineral elements on wood formation of old growth spruce (Picea abies (L.) Karst.). Holzforschung 48 Suppl:5--14.
Ewers, F.W. 1982. Secondary growth in needle leaves of Pinus lon-gaeva (bristlecone pine) and other conifers: quantitative data. Am. J. Bot. 69:1552--1559.
Fink, S. 1991. Structural changes in conifer needles due to Mg and K deficiency. Fert. Res. 27:23--27.
Fink, S. 1993. Microscopic criteria for the diagnosis of abiotic injuries to conifer needles. In Forest Decline in Atlantic and Pacific Re-gions. Eds. R.F. Hüttl and D. Müller-Dombois. Springer-Verlag, Berlin, pp 175--188.
Flores, H.E. 1991. Changes in polyamine metabolism in response to abiotic stress. In Biochemistry and Physiology of Polyamines in Plants. Eds. R.D. Slocum and H.E. Flores. CRC Press, Boca Raton, FL, pp 213--228.
Halonen, O., H. Tulkki and J. Derome. 1983. Nutrient analysis meth-ods. Metsäntutkimuslaitoksen tiedonantoja 121. Valtion painatus-keskus, Helsinki, 28 p.
Hartt, C.E. 1969. Effects of potassium deficiency upon translocation of 14C in attached blades and entire plants of sugarcane. Plant Physiol. 44:1461--1469.
Holopainen, T. and P. Nygren. 1989. Effects of potassium deficiency and simulated acid rain, alone and in combination, on the ultrastruc-ture of Scots pine needles. Can. J. For. Res. 19:1402--1411. Holopainen, T., S. Anttonen, A. Wulff, V. Palomäki and L. Kärenlampi.
Jokela, A., J. Bäck, S. Huttunen and R. Jalkanen. 1995. Excess nitro-gen fertilization and the structure of Scots pine needles. Eur. J. For. Pathol. 25:109--124.
Jokela, A., V. Palomäki, S. Huttunen and R. Jalkanen. 1996. Effects of root damage on the nutritional status and structure of Scots pine needles. J. Plant Physiol. 148:317--323.
Kaunisto, S. 1987. Effect of refertilization on the development and foliar nutrient contents of young Scots pine stands on drained mires of different nitrogen status. Commun. Inst. For. Fenn. 140, 58 p. Kaunisto, S. 1989. Jatkolannoituksen vaikutus puuston kasvuun
vanhalla ojitusalueella. Summary: Effect of refertilization on tree growth in an old drainage area. Folia For. 724:1--15.
Königshofer, H. 1991. Distribution and seasonal variation of poly-amines in shoot-axes of spruce (Picea abies (L.) Karst.). J. Plant Physiol. 137:607--612.
Lüttge, U. and D.T. Clarkson. 1989. Mineral nutrition: Potassium. Prog. Bot. 50:51--73.
Maier-Maercker, U. and W. Koch. 1992. Histological examination of spruce needles from a long-term gas exchange experiment in pure and polluted air in the field. Trees 6:186--194.
Marschner, H. 1995. Mineral nutrition of higher plants. Academic Press, New York, 889 p.
Mengel, K. and W.W. Arneke. 1982. Effect of potassium on the water potential, the osmotic potential, and cell elongation in leaves of
Phaseolus vulgaris. Physiol. Plant. 54:402--408.
Nygren, P. and P. Hari. 1992. Effect of foliar application with acid mist on the photosynthesis of potassium-deficient Scots pine seedlings. Silva Fenn. 26:133--144.
Paarlahti, K., A. Reinikainen and H. Veijalainen. 1971. Nutritional diagnosis of Scots pine by needle and peat analysis. Commun. Inst. For. Fenn. 74, 58 p.
Palomäki, V. and T. Holopainen. 1994. Effects of phosphorus defi-ciency and recovery fertilization on growth, mineral concentration, and ultrastructure of Scots pine needles. Can. J. For. Res. 24:2459--2468.
Palomäki, V. and T. Holopainen. 1995. Effects of nitrogen deficiency and recovery fertilization on ultrastructure, growth and mineral concentrations of Scots pine needles. Can. J. For. Res. 25:198--207. Phillips, R., M.C. Press, L. Bingham and G. Grimmer. 1988. Poly-amines in cultured artichoke explants: effects are primarily on xylogenesis rather than cell division. J. Exp. Bot. 39:473--480. Pistocchi, R., F. Keller, N. Bagni and P. Matile. 1988. Transport and
subcellular localization of polyamines in carrot protoplasts and vacuoles. Plant Physiol. 87:514--518.
Raitio, H. 1979. Growth disturbances of Scots pine caused by boron deficiency on an afforested abandoned peatland field. Folia For. 412:1--16.
Raitio, H. 1981. Effects of macronutrient fertilization on the structure and nutrient content of pine needles on a drained short sedge bog. Folia For. 465:1--9.
Reinikainen, J. and S. Huttunen. 1989. The level of injury and needle ultrastructure of acid rain-irrigated pine and spruce seedlings after low temperature treatment. New Phytol. 112:29--39.
Sarjala, T. and S. Kaunisto. 1993. Needle polyamine concentrations and potassium nutrition in Scots pine. Tree Physiol. 13:87--96. Toth, R. 1982. An introduction to morphometric cytology and its
application to botanical research. Am. J. Bot. 69:1694--1706. Zwiazek, J. and J. Shay. 1987. Fluoride- and drought-induced
