Journal of Experimental Botany

This Article Abstract FREE Full Text (PDF) E-letters: Submit a response Alert me when this article is cited Alert me when E-letters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (12) Request Permissions Disclaimer Google Scholar Articles by Repo, T. Articles by Rikala, R. Search for Related Content PubMed PubMed Citation Articles by Repo, T. Articles by Rikala, R. Agricola Articles by Repo, T. Articles by Rikala, R. Social Bookmarking          What's this?

Journal of Experimental Botany, Vol. 51, No. 353, pp. 2095-2107, December 2000
© 2000 Oxford University Press

Original Papers

The electrical impedance spectroscopy of Scots pine (Pinus sylvestris L.) shoots in relation to cold acclimation

T. Repo^1,4, G. Zhang^1,3, A. Ryyppö¹ and R. Rikala²

¹ Faculty of Forestry, University of Joensuu, PO Box 111, FIN-80101 Joensuu, Finland
² Finnish Forest Research Institute, Suonenjoki Research Station, Juntintie 40, FIN-77600 Suonenjoki, Finland

Received 15 May 2000; Accepted 13 July 2000

	Abstract

Top Abstract Introduction Materials and methods Results Discussion Appendix 1 Appendix 2 References

 
Electrical impedance spectroscopy (EIS) was applied to stems of Scots pine (Pinus sylvestris L.) in a provenance field trial during frost hardening to find an EIS parameter for assessing frost hardiness (FH) without a controlled freezing test. The FH of stems and needles assessed by controlled freezing tests was compared with the equivalent circuit EIS parameters of a distributed model of stems (not exposed to controlled freezing treatment) and with dry matter (DM) content of stems. Significant differences in the equivalent circuit parameters, FH and DM content were found between provenances. The relaxation time (₁), describing the peak of the high frequency arc of the impedance spectrum, and the intracellular resistance (r_i) of stems increased with increasing FH. According to the linear regression, the coefficient of determination (R²) between the FH of stems and needles with ₁ of the stem was 0.87 and 0.89, and with r_i of the stem 0.74 and 0.85, respectively. The relation between FH and ₁ changed with the degree of hardiness. The highest coefficient of determination was 0.95 in September when the FH of needles, ranging from -10 °C to -25 °C, was predicted with an accuracy of ±2.0 °C. The resistance parameter r₂, describing the width of the low frequency arc of the impedance spectrum, decreased prior to and during the initial hardening: significant differences were found between provenances. This indicates that r₂ was not related to frost hardening per se. It is concluded that it is possible to distinguish the hardening patterns of different provenances by ₁ in the rapid phase of hardening without controlled freezing tests.

Key words: Dry matter content, frost hardiness, impedance spectroscopy, relaxation time, Scots pine.

	Introduction

Top Abstract Introduction Materials and methods Results Discussion Appendix 1 Appendix 2 References

 
An easy and fast method for the determination of frost hardiness (FH) in plants is still lacking. Several techniques are currently used that necessitate a controlled freezing test with several freezing temperatures. These techniques include the electrolyte leakage test, the visual scoring of damage, the measurement of chlorophyll fluorescence, differential thermal analysis, plasma membrane H⁺-ATPase activity, and electrical impedance analysis (Timmis, 1976; Palta et al., 1978; Glerum, 1985; Ryyppö et al., 1998). Thus the FH assessment is time-consuming, requires expensive equipment, and also a considerable amount of material.

Electrical impedance spectroscopy is a method for studying the structure of organic and inorganic materials (Ackmann and Seitz, 1984; Foster and Schwan, 1989; Macdonald, 1992; Repo and Zhang, 1993; Repo et al., 1997; Schwan, 1999). In this method, alternating current (AC) is applied to a piece of plant tissue. Alternating current causes polarization and relaxation in the sample leading to changes in amplitude and phase of the applied AC signal. According to those changes an impedance of the sample can be determined which is formed of a real and an imaginary part in a complex plane. When the real and imaginary part is measured at different frequencies an impedance spectrum is obtained.

The proportion of the current passing through the apoplastic and symplastic space of the tissue sample depends on the AC frequency and the tissue properties. Cell membranes have high impedance at low frequencies. Accordingly, the current flows in the apoplastic space which determines the total impedance. In the apoplastic space ions are the main current carriers. The impedance of the cell membranes decreases with increasing frequency. Accordingly the symplastic space become conductive and at high frequencies the symplastic and apoplastic resistance form a parallel circuitry.

Changes in cellular features appear in their EIS-characteristics. The tissue features can be quantified in EIS by the equivalent electrical circuit analysis (Zhang and Willison, 1992; Repo and Zhang, 1993; Repo et al., 1994; Zhang et al., 1995). With a proper equivalent electrical model it is possible to study the effects of different stress factors on the tissue properties according to the changes in the parameters of the model (Zhang and Willison, 1992; Zhang et al., 1993; Repo et al., 1994; Repo and Pulli, 1996; Ryyppö et al., 1998).

It has been found that several equivalent circuit parameters of the distributed model for the stems of Scots pine undergo seasonal variation (Repo et al., 1995, 1997). Some changes in the parameters coincided with FH, and the highest correlation was previously found between intracellular resistance and FH. It has been hypothesized that impedance analysis might be used for assessing FH without a controlled freezing test. In order to develop practical applications in nurseries and tree breeding, this hypothesis had to be tested in an experiment with genotypes, each with its own different hardening pattern. If there is a common relationship between FH measured by a conventional method and the selected parameter of a non-frost exposed sample, then the method could be used for FH assessment without using a controlled freezing test.

The aim of this study was to examine the relationship between the frost hardiness of stems and needles, assessed by controlled freezing tests, with equivalent circuit EIS parameters of stems not exposed to frost. The equivalent circuit EIS parameters and FH were compared with the dry matter content of the samples. The study took place by using a provenance field trial of Scots pine.

	Materials and methods

Top Abstract Introduction Materials and methods Results Discussion Appendix 1 Appendix 2 References

 
Plant material
The study was based on a field provenance trial with six Scots pine (Pinus sylvestris L.) origins. The seeds for the trial were collected from six open-pollinated natural stands over an area ranging from Estonia to northern Finland. The bareroot 2+0 seedlings (see Nerg et al., 1994, for more details) were planted in an old nursery field at the Research Nursery of the Suonenjoki Research Station, Finnish Forest Research Institute (62°39'N, 27°03'E, 130 m.a.s.l.) in 1993 (Table 1). The soil in the nursery bed was a mixture of fine sand and peat (<3% w/w). Two hundred seedlings from each origin were planted with a spacing of 1.0x0.5 m between the seedlings. The saplings were 8-year-old during the experimental growing season in 1998.

View this table: [in this window] [in a new window]   Table 1. Latitude, longitude and mean temperature sum (T+5 °C as degree days) of the seed origins used in the study (mean of 30 years) (Nerg et al., 1994)

 
Samples from 16 saplings from each provenance were taken at 2–3 week intervals between 3 August and 30 November 1998. On each sampling occasion, shoots of lateral branches grown during the current year were sampled for impedance analysis of stems (termed ‘non-frost exposed sample’), for determination of dry matter content, and for assessment of FH by means of controlled freezing tests (‘frost exposed sample’). After sampling, the needles were immediately separated from the shoots in the laboratory. In November, owing to the below zero outside air temperatures, the shoots were first kept at 5 °C for 2 h prior to sample preparation in the laboratory.

Frost hardiness by controlled freezing test
The FH of the stems and needles was determined by controlled freezing tests. For the tests, 8 stems (16 in the last two tests) and 32 needles from each origin and each test temperature were prepared in plastic bags. Distilled water was sprayed into the bags to avoid excessive supercooling. In each test 6–7 freezing temperatures and a control temperature of +3 °C were used to determine the critical temperature range for FH. The freezing temperatures used were between 0 °C and -130 °C according to the predicted hardiness level. Both the initial and end temperature of the exposure was 3 °C and the rate of cooling and warming was 5 °C h^-1. The samples were kept at the target temperature for approximately 4 h. Immediately after the exposure the degree of frost damage in the stems and needles was quantified by the electrical impedance analysis and the electrolyte leakage method, respectively (Flint et al., 1967; Repo et al., 1994; Ryyppö et al., 1998). The total numbers of stems and needles used for FH determination by controlled freezing tests were 2 800 and 11 500, respectively.

The FH of the frost exposed stems was determined as the extracellular resistance of each specimen obtained by means of electrical impedance analysis (Repo et al., 1994). Immediately after the frost exposure and thawing, a 15 mm section was cut from the central portion of the stem. The section was placed directly in contact with the electrode pastes of the Ag/AgCl-cell (the type of electrodes RC1, WPI Ltd., Sarasota, U.S.A.) to measure an impedance spectrum at 42 frequencies between 80 Hz and 1 MHz (HP 4284 A) (Repo, 1994). The input voltage level of the sine signal was 100 mV (rms). The measurement of each sample took 30 s.

The distributed circuit element model (see below) was used to calculate the extracellular resistance according to the impedance spectra (Repo et al., 1994). The EIS parameters of the equivalent circuit were estimated with an automated Complex Non-linear Least Squares (CNLS) fitting program (T Repo) which uses LEVM v6.0 (JR Macdonald, Department of Physics and Astronomy, University of North Carolina, Chapel Hill, NC). The FH was estimated as the inflection point of the logistic sigmoid function (equation 1).

(1)

where y and x refer to the specific extracellular resistance (unit m) and the exposure temperature, respectively, A and D define asymptotes of the function, and B is the slope at the inflection point C. The specific resistance was obtained by normalization of the resistance in respect of the cross-sectional area and the length of the sample.

The FH of needles was assessed by electrolyte leakage method (Flint et al., 1967; Sutinen et al., 1992). After freezing, 32 needles from each provenance and each testing temperature were randomly selected. Ten millimetre pieces were cut from the middle of the needles, rinsed with distilled water and placed in test tubes (eight samples per tube) as four replicates. Six millilitres of distilled water was added to each test tube. The tubes were shaken at room temperature for 24 h before the measurement of the first conductivity (L₁). Then the samples were heat-killed at 92 °C for 20 min and shaken for another 24 h before the measurement of the second conductivity (L₂). The relative electrolyte leakage (REL) was calculated as:

(2)

The FH was estimated as the inflection point (parameter C) of equation 1, where y refers to the relative electrolyte leakage (REL).

Impedance analysis of non-frost exposed stems
The electrical impedance spectra of the 16 stems (one stem per sapling) from each origin at each sampling time were measured in the laboratory in the way described above immediately after sampling (a total of 8 times). The total number of non-frost exposed samples used for the EIS-analysis was 768.

The impedance spectra of stems were modelled by an equivalent circuit with two distributed circuit elements (DCE) in series with a resistor (double-DCE model) (Repo et al., 1994). Both DCE-elements are composed of a constant phase element in parallel with a resistor (Fig. 1; Appendix 1) (Macdonald, 1987). The total complex impedance (Z) of the double-DCE is as shown (for derivation of this equation see Appendix 1):

(3)

where the angular velocity =2f (f=frequency). In the double- DCE model, there are three resistances (R, R₁ and R₂), two relaxation times (₁ and ₂) and two distribution coefficients (₁ and ₂) of the relaxation times (for mathematical interpretation of each parameter see Fig. 1). The letter i refers to the imaginary unit. The parameters were estimated with the automated CNLS program (see above).

View larger version (19K): [in this window] [in a new window]   Fig. 1. Determination of the distributed model parameters (double-DCE) of an impedance spectrum of stem () of Scots pine schematically. Real part of impedance on x-axis and imaginary part of impedance on y-axis (upper figure). Frequency increases from right (80 Hz) to left (1 MHz). The two circles represent the two arcs of the spectrum. DCE1 and DCE2 are distributed circuit elements composed of constant phase elements (CPE) in parallel with resistors (lower figure). Resistances (R, R1 and R2) of the equivalent model are obtained according to the intersections of the circles with x-axis. The centre of the circles is below x-axis‘depressed centre’ defined by parameters 1 and 2. Relaxation times 1 and 2 are obtained from the apex of the circles. ZCPE1 and ZCPE2 are impedances of Constant Phase Elements. For more explanation see Appendix 1 and 2.

 
Since the low frequency current may not pass the cell membranes but flows in the apoplastic space the extracellular resistance (R_e) is obtained as:

(4)

At high frequencies the current may pass the cell membranes and accordingly flows both in the apoplastic and symplastic space, the intracellular resistance (R_i) is obtained as:

(5)

The resistance parameters were normalized with respect to the cross-sectional area (A_s=d²/4, d=diameter) and the length (l) of the sample in order to obtain the specific resistances (equation 6)

(6)

Lower case letters have been used to indicate the normalized values.

Determination of the dry matter (DM) content
The same 16 shoots from each provenance as for the ‘non-frost exposed’ impedance measurements were used for determination of the dry matter content at each of the eight times. After fresh weight measurements, the samples were oven-dried at 80 °C for 48 h before weighing of their dry weight. The DM content was calculated as the percentage of dry weight in relation to fresh weight.

Analysis of the data
In order to compare the differences between the provenances at different times, the original data was pooled into three groups: southern, intermediate and northern, with two provenances in each (Table 1). The significance of the monthly mean differences in FH, equivalent circuit EIS parameters and DM content between the groups was tested by means of a paired t-test.

The relation of the FH of stems and needles to the equivalent circuit EIS parameters of stems and to the DM content of stems was studied by linear and exponential models, respectively. The original data (the means of each origin at each given time) from all the six provenances over the whole study period were pooled and the linear and exponential regression models were applied (SPSS 8.0 for Windows, SPSS Inc.). Then the pooled data was split according to the assessment dates, and the correlation of FH with the parameters of the non-frost exposed samples was calculated separately for each time. For the evaluation of the reliability and accuracy of the regression models, the coefficient of determination, the confidence intervals and the residuals were examined (SPSS Inc.).

	Results

Top Abstract Introduction Materials and methods Results Discussion Appendix 1 Appendix 2 References

 
Frost hardiness by controlled freezing tests
There were no significant differences in the FH between the three provenance groups in August (Table 2). Differences developed in September, when the frost hardening commenced (Fig. 2). No difference was found in the FH between the provenances in November, except between the stems of the southern and northern groups. The needles reached a higher level of hardiness (around -90 °C) than the stems (from -40 to -50 °C) in November.

View this table: [in this window] [in a new window]   Table 2. Comparison of frost hardiness in stems and needles between groups of provenances with time (t-test) For the test the data gathered twice per month were pooled. S–I, southern versus intermediate provenances; I–N, intermediate versus northern provenances; S–N, southern versus northern provenances. There are two origins in each provenance group.

 

View larger version (14K): [in this window] [in a new window]   Fig. 2. The frost hardiness of stems (A) and needles (B) of Scots pine of different origins in the provenance field trial at Suonenjoki, Finland, in 1998. Each point is the mean of two origins: () two southernmost, () two intermediate, () two northernmost provenances (see Table 1).

 

Impedance analysis and dry matter content of non-frost exposed stems
The form of the impedance spectra of the non-frost exposed stems of Scots pine changed during the study (Fig. 3). In early August, the spectra were clearly characterized by two arcs. With frost hardening, the proportion of the low and high frequency arcs changed as the high frequency arc became more dominant.

View larger version (20K): [in this window] [in a new window]   Fig. 3. Typical impedance spectra of non-frost-exposed stems of Scots pine on August 3, September 7, October 19, and November 30. The spectra are composed of 42 different frequencies ranging from 80 Hz to 1 MHz (from right to left, respectively).

 
The trend in the equivalent circuit parameters was similar for all provenances (Fig. 4). All of the resistance parameters, except r₂, increased during the study, especially from September onwards. The intracellular resistance (r_i) was 7 m at the beginning of September and rose to 15 m until the end of November (Fig. 4E). In September there were significant differences between the provenance groups with regard to r_i.

View larger version (27K): [in this window] [in a new window]   Fig. 4. The parameters for the double-DCE model of the non-frost-exposed stems of Scots pine by impedance analysis (A–I) and the dry matter content (DM) of stems (J) of southern (), intermediate () and northern () provenances (see Table 1) in a field trial with Scots pine. The parameters r, r1, r2, re and ri are resistances, and 1 and 2 are the distribution coefficients of the relaxation times 1 and 2, respectively. Each point is the mean of two origins (n=32, 16 per origin). The bars indicate standard errors.

 
The resistance r₂ decreased from 11 m in August to 3 m in October (Fig. 4C). The southern and northern provenances differed significantly in r₂ from the beginning of August until the end of September (Table 3).

View this table: [in this window] [in a new window]   Table 3. The dates when the differences between the provenance groups in the electrical impedance spectroscopic (EIS) parameters and the dry matter content (DM) of stems of Scots pine were statistically significant Otherwise the differences were not significant. The paired provenance groups were: (A) Southern versus intermediate. (B) Intermediate versus northern. (C) Southern versus northern.

 
The relaxation time ₁ increased from 3.5 µs to 8 µs between the end of August and the end of November (i.e. the characteristic frequency f_c1 decreased from 286 to 125 kHz). The change occurred first with the northern and last with the southern provenances (Fig. 4F). There was a time lag of approximately 1 month in ₁ between the provenances; the level of 6 µs was reached on 29 September and 24 October in the southern and the northern groups, respectively. For most of the time, the differences between the provenance groups were significant (Table 3).

The relaxation time ₂ behaved more irregularly than ₁ (Fig. 4G). At the end of August all of the groups differed significantly with respect to ₂ (Table 3). In mid-September, ₂ increased from 400 µs to 550–600 µs (i.e. the characteristic frequency f_c2 decreased from 2.5 to 1.7 kHz) concomitantly with the increase in the coefficient ₂ (Fig. 4I). All the provenance groups differed significantly from each other at the end of September with respect to ₂ (Table 3).

The DM content of the stems increased during the experimental period for all groups (Fig. 4J). There was a temporary drop in DM content on 19 October. On that date the frost hardening of the stems temporarily ceased (Fig. 2). For most of the study period the DM content of the northern and intermediate groups was significantly higher than that of the southern one (Table 3, Fig. 4J).

Comparison of the EIS parameters and the DM content of non-frost exposed samples with frost hardiness
The relaxation time ₁ and the intracellular resistance r_i of the stems displayed the highest correlation with the FH of the stems and needles (Table 4). The linear regression model described the relation between the FH and ₁ well (Fig. 5A, B). Some model error existed for the lowest r_i values in the linear model as the r_i increased without any clear change in FH (Fig. 5C, D). The relation between FH and DM content was non-linear (Fig. 5E, F). The DM content increased at first without hardening, which was initiated when the DM content reached a level between 32% and 35%.

View this table: [in this window] [in a new window]   Table 4. The coefficient of determination (R2) of the linear regression model for the comparison of the electrical impedance spectroscopic (EIS) parameter and dry matter (DM) content of stems (non-frost exposed samples) with frost hardiness of stems and needles (FHstem and FHneedle, respectively) (frost exposed samples) at dates when the correlation was significant (A) EIS parameter of stems versus FHstem. (B) EIS parameter of stems versus FHneedle. (C) DM content of stem versus FHstem and FHneedle. The EIS parameters were obtained by the double-DCE model. The linear regression was calculated using the pooled data of all provenances and test dates (n=48) and at different dates separately (n=6).

 

View larger version (32K): [in this window] [in a new window]   Fig. 5. The relationship of the frost hardiness of needles and stems with the relaxation time 1 (A, B), intracellular resistance ri (C, D), and the dry matter (DM) content (E, F) of non-frost-exposed stems of Scots pine. The data from six provenances and different times from August to November have been pooled. Each point represents the mean of one origin at one time: 1, ri and the DM are the means of 16 stems, and the respective frost hardiness is obtained by means of controlled freezing tests (impedance analysis and electrolyte leakage for stems and needles, respectively). The best fit function, the coefficient of determination (R2) and the 90% confidence intervals are indicated.

 
In the pooled data the deviation between measured and predicted FH was occasionally high for ₁. According to the linear model the 90% confidence intervals of FH for needles and stems were ±15 °C and ±9 °C for ₁ and ±20 °C and ±14 °C for r_i, respectively. The predictive power of the DM content was variable; an increase of DM at low values had no effect on the FH. The FH increased with increasing DM by over 32–35%, but concomitantly the confidence intervals expanded remarkably.

At selected times in the hardening phase the correlation between the FH and the ₁ was higher and the residuals between the predicted and measured FH lower than for the pooled data (Table 4; Fig. 6). The coefficient of determination was highest at 0.95 between the FH of the needles and the ₁ of the stems on 21 September (Fig. 6B). Then the residuals were within ±2.0 °C. The slope of the regression line gradually decreased from 24 August to 5 October with increasing hardiness (Fig. 6A). The variation of the residuals typically increased with the increase in FH. The correlation of FH with other parameters than ₁ was temporarily high and significant, e.g. for r₂-FH_stem and r₂-FH_needle on 21 September, r_i-FH_needle on 7 September, ₂-FH_needle on 21 September and DM-FH_stem on 30 November (Table 4).

View larger version (14K): [in this window] [in a new window]   Fig. 6. (A) The relationship between the frost hardiness of needles and the relaxation time 1 of stems of Scots pine at different times: 24 August (y1, ), 7 September (y2, ), 21 September (y3, ), 5 October (y4, •), 19 October (y5, ) and 9 November (y6, ). Each point of the 1 is the mean of 16 stems. The FH of the needles is assessed by means of electrolyte leakage method. The best fit regression lines are indicated. (B) The residuals between the measured and predicted FH for the linear regressions as a function of predicted FH at different times (for symbols see (A)).

 

Comparison of the EIS parameters and the dry matter content
According to the non-linear regression model for stems, the dry matter content had the highest coefficient of determination with the ₁ and r₁ which both increased with an increase in the DM content (Fig. 7). There was a minor change in the parameters when the DM content increased from 25% to 35%. Thereafter ₁ and r₁ increased faster with a further increase in the DM content above 40%.

View larger version (27K): [in this window] [in a new window]   Fig. 7. The relationship between the dry matter content (DM) and the equivalent circuit parameters of the double-DCE model of the non-frost-exposed stems of Scots pine. The data from all six provenances have been pooled. Each point is the mean of 16 stems. The best fit exponential function with the coefficient of determination (R2) is indicated.

 

	Discussion

Top Abstract Introduction Materials and methods Results Discussion Appendix 1 Appendix 2 References

 
The electrical impedance spectra of Scots pine saplings changed during the study period in autumn. The changes were mainly similar to those found previously (Repo et al., 1994, 1995, 1997), i.e. the magnitude of the real and the imaginary levels as well as the proportion of the high and low frequency arc changed. The CNLS analysis of the spectra modelled with the double-DCE showed that, all of the impedance parameters of the non-frost-exposed stems changed during the study period between early August and late November. Typically, there was first a latent period in August and early September when the parameters remained fairly constant and no frost hardening was found. Then the parameters increased concomitantly with frost hardening. The resistance r₂ formed an exception to that trend, since it already decreased in August without any indication of frost hardening or increase in DM content, and continued to do so until early October.

The relaxation time ₁ had the highest correlation with FH compared with the other equivalent circuit parameters or the DM content. Thus the ₁ seemed to predict the FH without a controlled freezing test. On some occasions the deviation between the predicted and measured FH was high, however, e.g. ranging up to 25 °C for ₁ between 4 µs and 6 µs (Fig. 5A). Due to the large deviation between the measured and predicted FH of the samples covering the temperature range from -5 °C to -90 °C, none of the parameters were accurate enough for applications.

At selected times during the hardening period the coefficient of determination between the FH and the relaxation time ₁ was high, and the deviation between the predicted and measured FH was small. These assessments were located in the initial phase of hardening which is the most interesting phase in several applications. During that time the differences between the provenances were the most obvious, i.e. the FH of needles of different provenances were between -10 °C and -25 °C. Thus, in that phase of hardening it was possible to grade the provenances according to their hardening pattern without a controlled freezing test.

Interpretation of the EIS parameters
In mathematical terms, the ₁ is obtained from the apex of the high frequency arc of the impedance spectrum (Fig. 1). Using the normalized values for r₁ and c₁ we get for the relaxation time , with the capacitance c₁ representing some interface in the tissue. Then the c₁ remained fairly constant during the study period, ranging from 10–40 nF m^-1 (data not shown), whereas the parameters ₁ and r₁ were linearly correlated. This suggests that the change in the ₁ is due to the r₁ rather than the c₁, and thus the capacitance c₁ would not change during frost hardening. The increase in ₁ is then caused by the change in the ion mobility in a cellular compartment represented by r₁.

A biological interpretation of the ₁ and the derived parameters r₁ and c₁ is unknown. It has been proposed that altered cell membrane properties affect the relaxation time of the leaves and stems of olive trees (Olea europaea L.) (Mancuso and Rinaldelli, 1996) and accordingly r₁ and/or c₁. If a general number of 1 µF cm^-2 is used for the cell membrane capacitance and a cell membrane thickness of 10 nm is assumed, the specific capacitance of 0.1 nF m^-1 is obtained for the cell membrane. This number is about 100 times smaller than that calculated from this study's data.

The relaxation time ₁ and the resistance r₁ increased with an increase in the dry matter content. Thus, the water content may partially explain the behaviour of ₁ and r₁. Change or more specifically redistribution of water between symplast and apoplast may partially explain the decrease in the ₁ in case of freezing tests too (Repo et al., 1994). The net water content of the tissues does not change there, but there is a drift of water towards apoplastic ice by exosmosis. This together with damage in cell membranes and consequent ion leakage from the symplast to the apoplast (Palta and Weiss, 1993; Ryyppö et al., 1998) leads to decrease in ₁. This may, in turn, contribute to a decrease in the r₁ due to an increase in the charge-carrying ion content inside or outside the cell.

The dry matter content of the stems ranged from 25% to 45%, i.e. the moisture content (MC) from 75% to 55%, respectively. Thus the MC was well above the fibre saturation point (FSP) of 30% in woody plants. Above the FSP, the low frequency impedance should not depend much on the moisture content (Tattar et al., 1972; Glerum, 1980; Pukacki, 1982; Kucera, 1986), but this effect may not totally absent, as these data reveal. This study's data also include the acclimation period and the concomitant changes in the cellular properties, for example, cell wall thickness and rigidity, the size of the apoplastic space and membrane fluidity and symplastic compartmentalization. Those changes may partly explain the increase in certain EIS parameters which took place coincidentally with the decrease in the water content.

Prior to the hardening phase in August, when the diameter growth of the stems was ceasing, the relaxation time ₂ of the southern and intermediate provenances was higher than that of the northern ones. The resistance r₂ decreased from August to September and the decrease occurred later in the southern than in the northern origins. The parameters ₂ and r₂ were not correlated (data not shown). According to the relation , the capacitance c₂ decreased exponentially from 400 to 10 µF m^-1 with an increase of r₂ from 2 to 15 m, i.e. the c₂ increased prior to the frost hardening. The capacitance values of the c₂ were 1000 to 10000 times higher than the values of the c₁ and more than 1x10⁵ times higher than the cell membrane capacitance. Judging from the timing of the change in the r₂ and ₂ it can be assumed that r₂ and ₂ are connected with cellular differentiation and lignification.

The intracellular resistance r_i increased with frost hardening as was found earlier in Scots pine (Repo et al., 1995), alfalfa (Medicago stativa L.), birdsfoot trefoil (Lotus corniculatus L.) (Stout 1988a, b) and willow (Salix viminalis) (Repo et al., 1997). In the pooled data the coefficient of determination of FH for r_i was less than that for ₁, however, but it was occasionally as high as 0.83 (Table 4B). The increased concentration of the intracellular sap and the impaired intracellular ion mobility due to ‘frictional effect’ (Pauly and Schwan, 1966) may explain the increase in the r_i with frost hardening.

The DM content of stem started to increase before frost hardening. After the DM content of stems had increased to over 30%, the FH increased steeply in accordance with the previous studies (Junttila and Kaurin, 1990; Toivonen et al., 1991; Sutinen, 1992; Repo et al., 1997). The decrease in the water content is probably one of the first reactions at the moment of growth cessation and in reaction to physiological and structural changes in cells preceding cold acclimation. Significant differences were found in the DM content between provenances during frost hardening, but the DM content as such was not a reliable method for the determination of FH without a controlled freezing test. The DM content and FH of stems decreased temporarily on 19 October. This decrease can be explained by a fairly warm and rainy period just before that sampling date.

In conclusion, impedance spectroscopic analysis is a useful method for studying cold acclimation. The method is fast, and with proper planning over 300 samples can be measured in 1 d which is an advantage with regard to most other methods. The relaxation time of the non-frost exposed stems of Scots pine showed a high correlation with their frost hardiness. With proper timing of the measurement, the accuracy of impedance analysis may be high enough for forestry applications in nurseries, in seed orchards and in provenance trials for grading genotypes for the timing of frost hardening.

	Appendix 1

Top Abstract Introduction Materials and methods Results Discussion Appendix 1 Appendix 2 References

 

(1APPX)

	Appendix 2

Top Abstract Introduction Materials and methods Results Discussion Appendix 1 Appendix 2 References

 
List of symbols and abbreviations:

Symbol Explanation Unit A coefficient to define the asymptote % or m    of the sigmoid function B slope of the sigmoid function at % °C-1    the inflection point    or m °C-1 C inflection point of the sigmoid °C    function D coefficient to define the % or m    asymptote of the    sigmoid function L1 conductivity µS cm-1 L2 conductivity µS cm-1 REL relative electrolyte leakage % FH frost hardiness °C DM dry matter content: %    (dry/fresh)x100 CNLS Complex Non-linear    Least Squares DCE Distributed Circuit    Element model Z complex impedance ZCPE1 impedance of the    Constant Phase Element 1 ZCPE2 impedance of the    Constant Phase Element 2 i imaginary unit =2f angular velocity rads-1 f frequency Hz Rx resistance rx specific resistance m R, R1, R2 resistances in the    double-DCE model Re extracellular resistance Ri intracellular resistance r, r1, r2 specific resistances in the m    double-DCE re specific extracellular resistance m ri specific intracellular resistance m 1 relaxation time s 2 relaxation time s fc1 characteristic frequency=1/1 Hz fc2 characteristic frequency=1/2 Hz 1 distribution coefficient of 1 2 distribution coefficient of 2 c1 specific capacitance Fm-1 c2 specific capacitance Fm-1 d diameter m As cross-sectional area m2 l length of the sample m R2 coefficient of determination

	Acknowledgments

 
We thank Mr Martti Vuorinen, The Finnish Forest Research Institute, Suonenjoki Research Station, for providing us with the opportunity to make use of the provenance field trial. Professor Pere Riu and Dr Ilkka Leinonen also deserve to be thanked for their comments on this manuscript, Dr John A Stotesbury for its English revision, and Mr Jaakko Heinonen for his advice on the statistical analyses. Ms Anna-Maija Väänänen and Ms Eija Lappalainen are thanked for their technical assistance. This study was funded by the University of Joensuu, the China Scholarship Council (CSC), the Academy of Finland Research Council for Environmental and Natural Resources (projects No. 38012 and No. 41277), and the Finnish Forest Research Institute.

	Notes

 
³ Current address: Department of Horticulture, Agricultural University of Hebei, Baoding, Hebei. 071001, P.R.China.

⁴ To whom correspondence should be addressed. Fax: +358 13 2514444. E-mail: tapani.repo{at}forest.joensuu.fi

	References

Top Abstract Introduction Materials and methods Results Discussion Appendix 1 Appendix 2 References

 
Ackmann JJ, Seitz MA.1984. Methods of complex impedance measurements in biological tissues. CRC Critical Review in Biomedical Engineering 11, 281–311.

Flint HI, Bouse BR, Beattie DJ.1967. Index of injury–a useful expression of freezing injury to plant tissues as determined by the electrolytic method. Canadian Journal of Plant Sciences 47, 229–230.

Foster KR, Schwan HP.1989. Dielectric properties of tissues and biological materials: A critical review. In: Bourne JR, ed. Critical reviews in biomedical engineering. Boca Raton, FL: CRC Press, 25–104.

Glerum C.1980. Electrical impedance techniques in physiological studies. New Zealand Journal of Forestry Science 10, 196–207.[Web of Science]

Glerum C.1985. Frost hardiness of coniferous seedlings: principles and applications: In: Duryea ML, ed. Evaluating seedling quality: principles, procedures, and predictive abilities of major tests. Corvallis: Oregon State University, 107–123.

Junttila O, Kaurin Å.1990. Environmental control of cold acclimation in Salix pentandra. Scandinavian Journal of Forest Research 5, 195–204.

Kuera LJ.1986. Kernspintomographie und elektrische Wider-standsmessung als Diagnosemethoden der Vitalität erkrankter Bäume. Schweizerische Zeitschrift für Forstwesen 137, 673–690.

Macdonald JR.1987. Impedance spectroscopy: emphasizing solid materials and systems. New York: John Wiley & Sons, Inc.

Macdonald JR.1992. Impedance spectroscopy. Annals of Biomedical Engineering 20, 289–305.[Web of Science][Medline]

Mancuso S, Rinaldelli E.1996. Response of young mycorrhizal and non-mycorrhizal plants of olive tree (Olea europaea L.) to saline conditions. II. Dynamics of electrical impedance parameters of shoots and leaves. Advances in Horticultural Science 10, 135–145.

Nerg A, Kainulainen P, Vuorinen M, Hanso M, Holopainen JK, Kurkela T.1994. Seasonal and geographical variation of terpenes, resin acids and total phenolics in nursery grown seedlings of Scots pine (Pinus sylvestris L.). New Phytologist 128, 703–713.[Web of Science]

Palta JP, Levitt J, Stadelmann EJ.1978. Plant viability assay. Cryobiology 15, 249–255.[Medline]

Palta JP, Weiss LS.1993. Ice nucleation and freezing injury: an overview on the survival mechanisms and molecular aspects of injury and cold acclimation in herbaceous plants. In: Li PH, Christersson L, eds. Advances in plant cold hardiness. Boca Raton, Fla: CRC Press, 143–176.

Pauly H, Schwan H.1966. Dielectric properties and ion mobility in erythrocytes. Biophysical Journal 6, 621–639.

Pukacki P.1982. Influence of freezing damage on impedance parameters in Magnolia shoots. Arboretum Kornickie XXVII, 219–233.

Repo T.1994. Influence of different electrodes and tissues on the impedance spectra of Scots pine shoots. Electro- and Magnetobiology 13, 1–14.

Repo T, Hiekkala P, Hietala T, Tahvanainen L.1997. Intracellular resistance correlates with initial stage of frost hardening in willow (Salix viminalis). Physiologia Plantarum 101, 627–634.

Repo T, Leinonen M, Pääkkönen T.1995. The application of electrical impedance spectroscopy in assessing the frost hardiness of Scots pine. In: Gersing E, Schäfer M, eds. Proceedings of the IX International Conference on Electrical Bio-Impedance. Heidelberg: ICPRBI, 370–373.

Repo T, Pulli S.1996. Application of impedance spectroscopy for selecting frost hardy varieties of English ryegrass. Annals of Botany 78, 605–609.[Abstract/Free Full Text]

Repo T, Zhang MIN.1993. Modeling woody plant tissues using a distributed electrical circuit. Journal of Experimental Botany 44, 977–992.[Abstract/Free Full Text]

Repo T, Zhang MIN, Ryyppö A, Vapaavuori E, Sutinen S.1994. Effects of freeze–thaw injury on parameters of distributed electrical circuits of stems and needles of Scots pine seedlings at different stages of acclimation. Journal of Experimental Botany 45, 823–833.[Abstract/Free Full Text]

Ryyppö A, Repo T, Vapaavuori E.1998. Development of freezing tolerance in roots and shoots of Scots pine seedlings at non-freezing temperatures. Canadian Journal of Forest Research 28, 557–565.

Schwan H.1999. The practical success of impedance techniques from an historical perspective. In: Riu PJ, Rosell J, Bragós R, Casas Ó, eds. Electrical bioimpedance methods: applications to medicine and biotechnology. Annals of New York Academy of Sciences 873, 1–12.

Stout DG.1988a. Effect of cold acclimation on bulk tissue electrical impedance. I. Measurements with birdsfoot trefoil at subfreezing temperatures. Plant Physiology 86, 275–282.[Abstract/Free Full Text]

Stout DG.1988b. Effect of cold acclimation on bulk tissue electrical impedance. II. Measurements with alfalfa and birdsfoot trefoil at nonfreezing temperatures. Plant Physiology 86, 283–287.[Abstract/Free Full Text]

Sutinen M-L.1992. Physiological changes in the needles of Pinus nigra and Pinus resinosa with seasonal changes in freezing stress resistance. Acta Universitas Ouluensis A240.

Sutinen M-L, Palta JP, Reich PB.1992. Seasonal differences in the freezing stress resistance of needles of Pinus nigra and Pinus resinosa: evaluation of the electrolyte leakage method. Tree Physiology11, 241–254.[Abstract]

Tattar TA, Shigo AL, Chase T.1972. Relationship between the degree of resistance to a pulsed electric current and wood in progressive stages of discoloration and decay in living trees. Canadian Journal of Forest Research 2, 236–243.

Timmis R.1976. Methods of screening tree seedlings for frost hardiness. In: Cannel MGR, Last FH, eds. Tree physiology and yield improvement. London: Academic Press, 421–435.

Toivonen A, Rikala R, Repo T, Smolander H.1991. Autumn coloration of first year Pinus sylvestris seedlings during frost hardening. Scandinavian Journal of Forest Research 6, 31–39.

Zhang MIN, Repo T, Willison JHM, Sutinen S.1995. Electrical impedance analysis in plant tissues: on the biological meaning of Cole-Cole in Scots pine needles. European Biophysics Journal 24, 99–106.

Zhang MIN, Willison JHM.1992. Electrical impedance analysis in plant tissues: the effect of freeze—thaw injury on the electrical properties of potato tuber and carrot root tissues. Canadian Journal of Plant Science 72, 545–553.[Web of Science]

Zhang MIN, Willison JHM, Cox MA, Hall SA.1993. Measurement of heat injury in plant tissue by using electrical impedance analysis. Canadian Journal of Botany 71, 1605–1611.

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				G. Zhang, Y.-q. Li, and S.-h. Dong Assessing frost hardiness of Pinus bungeana shoots and needles by electrical impedance spectroscopy with and without freezing tests J Plant Ecol, November 23, 2009; (2009) rtp023v1. [Abstract] [Full Text] [PDF]

				E. Marsili, J. B. Rollefson, D. B. Baron, R. M. Hozalski, and D. R. Bond Microbial Biofilm Voltammetry: Direct Electrochemical Characterization of Catalytic Electrode-Attached Biofilms Appl. Envir. Microbiol., December 1, 2008; 74(23): 7329 - 7337. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) E-letters: Submit a response Alert me when this article is cited Alert me when E-letters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (12) Request Permissions Disclaimer Google Scholar Articles by Repo, T. Articles by Rikala, R. Search for Related Content PubMed PubMed Citation Articles by Repo, T. Articles by Rikala, R. Agricola Articles by Repo, T. Articles by Rikala, R. Social Bookmarking          What's this?

Online ISSN 1460-2431 - Print ISSN 0022-0957

Copyright © 2005 Society for Experimental Biology

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics