- Email: [email protected]
Analysis on the soil factor and physiological response of the plants in the process of sandy desertification on grassland
ACTA ECOLOGICA SINICA Volume 27, Issue 1, January 2007 Online English edition of the Chinese language journal Cite this article as: Acta Ecologica Sinica, 2007, 27(1), 48−57.
RESEARCH PAPER
Analysis on the soil factor and physiological response of the plants in the process of sandy desertification on grassland Zhu Zhimei1,2, Yang Chi1,*, Cao Mingming2, Liu Kang2, Yang Lianan2 1 Ecology and Environment Science Department, Life Science College, Inner Mongolia University, Huhhot 010021, China 2 Environment Science Department, Northwest University, Xi’an 710069, China
Abstract: The aim was to find the most practical measure to father desertification on grassland. Three-year research was carried out from 2001 to 2003 in Duolun County, Xilin Gol, Inner Mongolia, China. A series of degradation gradients or stages were established by clustering analysis, and corresponded to 5 community types. Five community types were selected as the sampling sites. Four common plant populations (i.e., Leymus chinensis, Cleistogenes squarrosa, Artemisia frigida and Melilotoides ruthenica) were chosen because of their different physiological responses to sandy desertification. The correlation analysis was made between the soil factor (5 soil indices) and the physiological response of per plant population (7 physiological indices). The results showed that in the course of the sandy desertification on grassland, the physiological response of L. chinensis, an impressible type, had more significant correlations with the soil moisture and C/N in the soil than others (P < 0.01) relatively. The soil moisture and C/N in the soil are likely the key factors for the damage on a physiological level. Its malondialdehyde (MDA) and abscisic acid (ABA) had more significant correlations than others with 5 soil indices as a whole (P < 0.01). C. squarrosa and A. frigida are of resistant types. Only the correlations of C/N among 5 soil indices with both MDAs were consistent and very significant (P < 0.01). Both ABA relations to 5 soil indices were consistent, and, similarly, were very significant (P < 0.01). M. ruthenica, a retarded type, was more sensitive to soil physical character (the soil moisture and the content of clay) than soil chemical character (the content of C, the total N and C/N in the soil), suggesting that the degradation of nutritious elements in the soil is not the leading factor in holding back its growth. Its MDA showed a more significant correlation (P < 0.01) than others as a whole, but its ABA did not show a significant correlation with each of the soil indices (P > 0.05). The synthesis result showed that MDA and ABA in the plants responded intensively to desertification stress. For each of the stress resistant types, there were different soil response mechanisms under different stages. The impressible type responded intensively to the soil moisture and C/N. The response of the resistant type to the soil factor did not appear to be a dominant factor. Altogether, the physiological response of plants mostly had a significant correlation with the C/N in the soil. Key Words: sandy desertification; physiological response; stress resistant type; soil factor; grassland steppe
Duolun County is situated at the farming-pastoral zone of the semi-arid region in North China. Expansion of desertification land area has caused serious harm to people's production life and social economy development. In recent years, the farming-pastoral zone has received universal attention from various aspects. There are two reasons from the viewpoints of the national social economy development and the ecological environment construction. The first reason is the most impoverished areas in China; the second reason is the most serious desertification areas in China. It has typical ecological vul-
nerability[1]. Duolun County is the nearest county of Inner Mongolia apart from Beijing, and is also one of the sand storm sources in North China. Therefore, it has the vital significance fathering desertification to safeguard the ecological environment of the capital circle. The essence of the fathering desertification is to restore and reconstruct the damaged vegetation and soil, and it is also the basic safeguard of the sustainable development in the farming-pastoral zone. The efficiency of restoration and reconstruction of the desertification land relates directly to the choice and the disposition way of the plant
Received date: 20006-02-28; Accepted date: 2006-12-10 *Corresponding author. E-mail: [email protected] Copyright © 2007, Ecological Society of China. Published by Elsevier BV. All rights reserved.
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
species and the edaphic factors[2]. The former study on relationship of the soil and plant has concentrated to the community, the population and the individual level in the desertifica－ tion process[3 7], the study on damaged mechanism in view of － the grassland desertification is limited[8 11], and little involves relation change of the edaphic factor and the plant physiological biochemistry under the natural state. The plant can make the response to environmental factors on a series of different scales. Intuitive and obvious changes on the community, population, individual and shape level can occur through complex physiological processes in the plant. Under environmental stress, the plant adjusts physiological reaction to compensate its lack in nutrition for growth. In this way, the plant can grow continually. Perhaps plants’ physiological reaction and the adaptability are not only short-term and irritable, but also a more in-depth response mechanism. In the process of sandy desertification on grassland, soil is an important environmental factor. This article has studied the mechanism that the plant adjusts in the physiological reaction to adapt to the soil factor. The aim is to find the most practical measure fitting for different typical plants to father desertification on grassland, and to provide scientific proof restoring and reconstructing the damaged ecosystem.
1 Survey on research area and research technique 1. 1 Survey on research area The research area is established in Duolun County, Inner Mongolia, China. This county is located southeast Xilin Gol, Inner Mongolia: east longitude is 115º51′–116º 54′, north latitude is 41º46′–42º36′, and elevation is 1150–1800m. In the area the climate is of the monsoon climate, the wettability is 0.54, the year average temperature is 1.6℃, the annual temperature variation is 36.8℃, the diurnal temperature variation is 14.1℃, and the average frost-free period is 100d. The yearly transpiration rate is 4.5 times the yearly precipitation. The year average wind speed is 3.6m/s, and the number of gale days is 67.3d, many in spring. This county is in the south of Inner Mongolia Plateau, the northern slope of Yin Mountain, and the southern edge of Otindag Sandland. Its eastern part links up to Daxinganling Mountains. Because of the influence of the terrain, the landform is complicated with 5 types: the
low mountain knap, the knap, the river valley or the ditch valley, the piedmont clinoplain or the high platform, and the piling sand dune. The zonal soil in the county is the chestnut soil; the main soil type occupying the area with the percentage from high to low is in turn the chestnut soil (70.1%), the wind sandy soil (16.6%), the meadow soil (6.95%), and the black calcium earth (3.38%). The vegetation type is of the medium warm prairie belt in semi-drought and semi-moist zones of transition. The main vegetation types include the typical prairie, the forest prairie, the meadow prairie, the sand dune and so on. The research is carried out on the chosen desertification gradient. 1.2 Research technique 1.2.1 Choice of type places On the basis of disposal spots of the prairie at large observed in Duolun County in 1985, in August 2001 therein 35 types of places were sought with GPS, and each type place took three quadrats (1×1m2) to do the ecology investigation stochastically. In August 2002 and 2003, respectively, on the natural sandy prairie region in the county, 12 obvious desertification gradient type places were selected and each type place took three quadrats (1×1m2) to do the ecology investigation stochastically. On the basis of the basic clustering analysis on the data of community, population and soil obtained in three years, the natural sandy desertification grassland in Duolun County was classified into 5 community types:Ⅰ(OV), Ⅱ(PD), Ⅲ(LD), Ⅳ(MD) and Ⅴ(HD) (Table 1). 1.2.2 Method of sampling Plant: in 5 kinds of type places, such common populations as Leymus chinensis, Cleistogenes squarrosa, Artemisia frigida and Melilotoides ruthenica were selected. Fresh plant leaves (middle) were stochastically gathered and mixed up to every species. Some part of them was fixed by fluid nitrogen, and then transferred over to a refrigerator of –80℃ when returning back to the laboratory to be used for testing the enzyme and ABA. Some other part was used for testing the fresh weight, the chlorophyll and the relative membrane penetrability (EC). Still some other samples were heated at 105℃ for 1 h, and then dried at 65℃ for 24 h to be used for testing the leaf water content and the proline. The plant quantity of Ⅴgradient was limited, so for some indices only 4 gradients could be tested.
Table 1 Community types of different desertification stages of sandy grassland during the desertification process Desertification stage
Community type
Constructive species
Original vegetation (OV)
L. chinensis + Stipa kryovii+ bunch grass
Leymus chinensis
Potential desertification (PD)
C. squarrosa + Agrostis cristatum
Cleistogenes squarrosa
Light desertification (LD)
A. frigida + A.cristatum + C. squarrosa
Artemisia frigida
Moderate desertification (MD)
A. intramongolica + weed
Artemisia intramongolica
Heavy desertification (HD)
Annual plant
Chenopodium album
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
Soil: in August 2001, a soil profile was taken in each type place, respectively, divided into 5 layers: 0–5 cm, 5–10 cm, 10–20 cm, 20–30 cm and 30–50 cm. Soil was taken with the soil box and the ring sampler to be used for testing the soil moisture content and carrying out the granule analysis. Soil samples in layers of 0–10 cm and 10–20 cm were taken thrice at each type place with an earth drill, respectively, to be used for testing the total C and the total N contents in the soil. In August 2002 and 2003 a soil profile of 0–20 cm was taken in each type place, respectively (below 20 cm the soil change was not obvious in 2001, so the soil profile of 0–20 cm was focused in judging the change in the soil); other operations were the same as those in 2001. 1.2.3 Determination method The determinations of relative leaf water content, EC, proline (PRO) content, MDA content, protection enzyme (superoxide dismutase SOD, peroxidase POD, catalase CAT) activity and ABA content can be seen in another article of this research[12]. The content of soil moisture (SM): drying and weighing method. The content of organic C in the soil (S.C): after being processed by using 0.3 mm sieves, the soil sample was determined
using the law of K2Cr2O7 capacity[13]. The content of the total N in the soil (S.N): after being processed by using 0.3 mm sieves, the soil sample was determined using the law of Kjeldahl[13]. The granulometric analysis: soil grains with different grain sizes were separated and its percentage was calculated using pipette analysis (combined with the sieving and the water sedimentation)[14], and applying the Stokes’ Law, where the clay (S. Clay) size < 0.01mm. The above indices were determined thrice. 1.2.4 Data processing By synthesizing the tested data in 2001–2003, the Bivariate correlation analysis and the ANOVA variance analysis were carried out.
2
Result and analysis
2.1 Change in the soil factors and plant's physiological response under different desertification stages With the desertification progress of grassland, physiological response of plants and change in the soil factors are shown in Tables 2 and 3. 2.2 Correlation analysis between plant’s EC and soil factors
Table 2 EC, MDA, proline content, protection enzyme and ABA content of the plants under different desertification stages (mean ± SD) Proline content (PRO) (µg·g–1DW)
Species
Desertification stage
L．chinensis
Ⅰ
4.94±0.25
14.71±2.62
679.06±38.96
123.42±11.10
1280.49±117.16
961.98±73.89
2150.00±343.17
Ⅱ
6.08±0.40
26.96±4.08
802.26±46.36
156.36±8.70
1234.13±137.71
1111.54±97.43
1142.21±126.36
Ⅲ
5.11±0.16
25.32±3.10
405.21±41.97
98.91±6.86
881.46±53.47
856.97±54.20
1628.89±292.01
Ⅳ
6.61±1.03
55.05±12.38
564.51±106.81
162.11±25.38
961.79±56.70
643.41±77.99
2654.92±136.37
Ⅴ
9.48
Ⅰ
10.73±1.52
23.35±2.34
275.96±21.61
92.66±13.21
309.61±46.98
151.45±19.00
1330.71±145.65
Ⅱ
14.41±2.69
35.60±6.10
286.37±28.16
109.49±8.40
434.45±79.04
209.96±16.70
Ⅲ
14.51±1.31
26.21±3.36
190.85±23.05
84.93±5.75
287.47±37.28
344.41±23.77
Ⅳ
10.76±0.52
79.00±27.00
343.75±46.88
132.61±11.02
202.22±36.21
281.82±34.79
Ⅴ
15.17±1.46
80.44±8.00
70.00±13.50
225.88±28.30
Ⅰ
2.43±0.41
53.82±18.85
961.79±41.02
195.09±21.87
83.33±12.60
89.71±7.34
2552.66±202.97
Ⅱ
4.26±0.94
132.01±20.79
1246.41±56.85
234.83±9.21
101.39±14.54
133.98±29.16
4077.20±262.85
Ⅲ
5.79±0.39
91.10±21.12
536.14±61.28
155.48±18.14
111.76±32.68
402.17±45.07
2197.94±118.93
Ⅳ
4.58±0.32
119.72±23.88
950.55±101.08
218.03±18.14
84.89±12.97
353.50±46.63
2438.20±291.40
Ⅴ
6.44±0.35
173.42±17.06
85.41±22.10
504.17±82.04
Ⅰ
3.78±0.43
15.80±4.19
1484.73±98.74
219.50±20.27
481.57±75.32
617.49±83.19
3094.77±278.21
Ⅱ
4.28±0.51
66.81±26.44
1687.89±124.80
277.13±19.36
563.26±76.45
1118.98±73.70
3833.46±583.21
Ⅲ
5.11±0.35
61.53±20.03
1172.36±102.76
289.53±18.99
293.72±35.00
675.59±43.81
3886.04±516.68
Ⅳ
5.70±0.34
51.66±21.03
1272.72±88.84
417.81±19.99
428.26±55.67
654.14±41.35
3658.82±463.61
Ⅴ
5.45±0.40
318.30±19.88
105.25±20.30
896.28±24.12
C．squarrosa
A．frigida
M．ruthenica
EC (%)
EC: Relative membrane permeability
MDA content (nmol·g–1DW)
SOD activity (U.g–1DW)
POD activity (U·g–1DW·min-1)
CAT activity (µmol·g–1DW·min-1)
ABA content (ng·g–1DW)
718.54±102.57 762.57±58.75 861.07±123.00
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
L. chinensis's leaf EC has a significant negative correlation with SM (P < 0.05) (Table 4), and has very significant negative correlations with S.C and S.C/N (P < 0.01). It also has negative correlations with S.Clay and S.N, but the correlations are not close (P > 0.05). It is explained that S.C/N and the decline of S.C are essential factors influencing the damage of cell membrane of L. chinensis. C. squarrosa's leaf EC has significant positive correlations with S.Clay and S.C (P < 0.05), and also has positive correlations with SM, S.C/N and S.N, but the correlations are not close (P > 0.05). It is explained that S.Clay and S.C influence its EC mainly. With the decline of S.Clay and S.C, C. squarrosa's EC reduces, and the situation of its damaged cell membrane is different from that of L. chinensis entirely. A. frigida's leaf EC has a very significant positive correlation with S.C/N (P < 0.01), has a significant negative correlation with SM (P < 0.05), and has a very significant negative correlation with S.N (P < 0.01), but the negative correlations with S.C and S.Clay do not reach a significant level (P > 0.05). It is explained that S.C/N and S.N affect its EC greatly. With the decline of S.N and the increase of S.C/N, A. frigida's EC increases. The correlation of M. ru-
thenica's leaf EC with the soil factor is the same as that of A. frigida’s leaf EC with differences that M. ruthenica's leaf EC has very significant correlations with S.Clay and S.C (P < 0.01), and A. frigida’s leaf EC does not have significant correlations with S.Clay and S.C (P > 0.05). 2.3 Correlation analysis between plant’s proline content and soil factors L. chinensis's leaf PRO has very significant negative correlations with SM, S.C, S.N and S.Clay (P < 0.01) (Table 5), and has a negative correlation with S.C/N, but the correlation is not close (P > 0.05). The decline of SM, S.C, S.N and S.Clay arouses a mass of accumulation of L. chinensis's leaf PRO. The correlation of C. squarrosa's leaf PRO with the soil factor is the same as that of L. chinensis’s leaf PRO, which also reflects that the decline of most soil factors (except S.C/N) arouses a mass of accumulation of C. squarrosa's leaf PRO. The correlations of A. frigida's leaf PRO with S.N and S. C/N are opposite to those of A. frigida's leaf EC, and has a very significant positive correlation with S.Clay. It is explained that S.C/N and S.Clay greatly affect A. frigida's leaf PRO. The correlation of M. ruthenica’s endogenesis PRO with the soil
Table 3 The soil moisture and the content of clay, C , N, C/N in the soil under different desertification stages (mean ± SD) Desertification stage
Soil moisture (SM)
The content of clay (S.Clay) (<0.01mm)
The content of organic carbon in the soil (S.C)
The content of total N in the soil (S.N)
C/N in the soil (S.C/N)
Ⅱ
24.84±3.96
12.53±6.76
1.33±0.82
0.16±0.06
8.11
Ⅲ
15.13±2.68
4.86±3.82
1.07±0.58
0.08±0.03
13.96
Ⅳ
10.86±1.83
2.49±2.35
0.82±0.48
0.07±0.04
12.36
Ⅴ
4.28±0.01
2.19±1.32
0.29±0.14
0.03±0.02
28.17
Table 4 Correlations between EC of the plants and the soil moisture, the content of clay, C , N and C/N in the soil Correlation coefficient
Species SM
S.C/N
L．chinensis
− 0.361*
− 0.492**
C．squarrosa
0.285
0.294
S.Clay
S.C
− 0.298
S.N
− 0.517**
0.348*
− 0.184
0.350*
0.242
A．frigida
− 0.329*
0.462**
− 0.297
− 0.263
− 0.545**
M．ruthenica
− 0.45**
0.329*
− 0.478**
− 0.461**
− 0.441**
SM: soil moisture; S.Clay: the content of clay in the soil; S.C: the content of organic carbon in the soil; S.N: the content of the total N in the soil * indicates significant correlations (P<0.05); ** indicates very significant correlations (P<0.01); the same below
Table 5 Correlations between PRO of the plants and the soil moisture, the content of clay, C , N and C/N in the soil Correlation coefficient
Species SM
S.C/N
S.Clay
S.C
S.N
L．chinensis
− 0.781**
− 0.200
− 0.716**
− 0.926**
− 0.587**
C．squarrosa
− 0.696**
− 0.277
− 0.628**
− 0.849**
− 0.498**
A．frigida M．ruthenica
0.318
− 0.674**
0.372**
0.395**
− 0.256
0.384**
− 0.185
0.416**
0.464** 0.358*
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
factor is the same as that of A. frigida’s endogenesis PRO, but is opposite to that of L. chinensis’s endogenesis PRO and C. squarrosa’s endogenesis PRO. 2.4 Correlation analysis between plant’s MDA content and soil factors L. chinensis's leaf MDA content has a very significant negative correlation with S.C/N (P < 0.01) (Table 6), and has very significant positive correlations with SM, S.C, S.N and S.Clay (P < 0.01). It is explained that the soil factor is very close to L. chinensis's leaf MDA content. C. squarrosa's leaf MDA content has very significant negative correlations with SM, S.C/N and S.C (P < 0.01), and the negative correlations
of C. squarrosa's leaf MDA content with S.N and S.Clay are not significant (P > 0.05). It is explained that S.C/N and the decline of S.C and SM are essential factors arousing C. squarrosa's MDA accumulation. A. frigida's leaf MDA content has very significant positive correlations with SM, S.N and S.Clay (P < 0.01), has a very significant negative correlation with S.C/N (P < 0.01), and does not have significant correlations with S.C (P > 0.05). It is explained that with the increase of S.C/N and the decline of S.Clay and S.N, A. frigida's MDA content reduces. The correlation of M. ruthenica’s MDA with the soil factor is similar to that of A. frigida’s MDA, but the correlation of M. ruthenica’s MDA with S.C is very signifi-
Table 6 Correlations between MDA contents of the plants and the soil moisture, the S.Clay, the contents of C and N, and C/N in soil Correlation coefficient
Species SM
S.C/N
S.Clay
0.655**
− 0.997**
0.719**
− 0.376**
− 0.486**
A．frigida
0.485**
− 0.947**
0.561**
0.275
0.684**
M．ruthenica
0.646**
− 0.772**
0.680**
0.535**
0.728**
L．chinensis C．squarrosa
− 0.312
S.C 0.464** − 0.533**
S.N 0.820** − 0.197
Table 7 Correlations between the SOD activity of the plants and the soil moisture, the S.Clay, the content of C and N, and C/N in soil Correlation coefficient
Species SM
S.C/N
L．chinensis
− 0.224
− 0.551**
S.Clay
S.C
S.N
− 0.187
− 0.370**
0.287
− 0.328*
C．squarrosa
− 0.231
− 0.398**
− 0.283
A．frigida
− 0.245
− 0.378**
0.260
M．ruthenica
− 0.382**
0.192
− 0.419**
− 0.178 − 0.441**
− 0.196
0.312 − 0.334*
Table 8 Correlations between the POD activity of the plants and the soil moisture, the S.Clay, the content of C and N, and C/N in soil Correlation coefficient
Species SM
S.C/N
S.Clay
S.C
S.N
L．chinensis
0.474**
− 0.547**
0.495**
0.406**
0.524**
C．squarrosa
0.661**
− 0.576**
0.611**
0.654**
0.637**
A．frigida
0.245
− 0.192
0.230
0.264
0.221
M．ruthenica
0.596**
− 0.766**
0.523**
0.494**
0.654**
Table 9 Correlations between the CAT activity of the plants and the soil moisture, the S.Clay, the content of C and N, and C/N in soil Correlation coefficient
Species SM L．chinensis
0.648**
C．squarrosa
0.205
A．frigida M．ruthenica
− 0.491**
0.451**
S.C/N
S.Clay
− 0.440**
0.636**
S.C 0.658**
S.N 0.603**
0.259
− 0.222
0.518**
− 0.487**
− 0.443**
− 0.525**
0.593**
0.429**
0.511**
− 0.331*
0.271
− 0.234
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
cant (P < 0.01). 2.5 Correlation analysis between plant’s protection enzyme activity and soil factors L. chinensis's leaf SOD activity has very significant negative correlations with S.C and S.C/N (P < 0.01) (Table 7), has negative correlations with SM and S.Clay, and has a positive correlation with S.N, but the three correlation coefficients do not reach significant levels (P > 0.05). It is explained that S.C/N and the decline of S.C are essential factors to arouse L. chinensis's SOD activity. But with the decline of S.N, its SOD activity reduces. L. chinensis's leaf POD and CAT activities have a very significant negative correlation with S.C/N (P < 0.01) (Tables 8, 9), and have very significant positive correlations with SM, S.C, S.N and S.Clay (P < 0.01). With the increase of S.C/N and the decline of SM, S.C, S.Clay and S.N, L. chinensis's leaf POD and CAT activities reduce greatly. Among them, relation between its POD activity and S.C/N are closest, so is the relation between its CAT activity and S.C. The correlation of C. squarrosa's leaf SOD activity with the soil factor accords with that of L. chinensis’s leaf SOD activity, and thus S.C/N and the decline of S.C are essential factors to arouse C. squarrosa's SOD activity with the difference from L. chinensis that the inapparent correlation of C. squarrosa's SOD activity with S.N is negative. It is explained that with the decline of S.N, C. squarrosa's SOD activity increases. The correlation direction between A. frigida's leaf SOD activity and the soil factor accords with that of L. chinensis and C. squarrosa, but the correlation degree is inferior to that of the latter two. None but S.C/N has a very significant negative correlation with its SOD activity (P<0.01), and the rest do not have significant correlations (P>0.05). The correlation of A. frigida's with the soil factor is opposite to that of L. chinensis’s CAT activity. The correlation between M. ruthenica's leaf protection enzyme activity and the soil factor is stronger, and with the decline of SM, nutrient and fine matter in the soil, M. ruthenica's leaf SOD activity increases, and M. ruthenica's leaf POD and CAT activities reduce. Among the 3 protection enzymes, M. ruthenica's leaf POD and CAT activities respond more intensively to environmental changes of soil. 2.6 Correlation analysis between plant’s ABA content and soil factors L. chinensis's leaf endogenesis ABA content has a very significant positive correlation with S.C/N (P < 0.01) (Table 10),
has very significant negative correlations with SM, S.Clay, S.N and S.C (P < 0.01), among witch the correlation with S.C is closest. C. squarrosa's leaf endogenesis ABA content has a significant negative correlation merely with S.C (P < 0.05). The negative correlations of C. squarrosa's leaf endogenesis ABA content with SM, S.N and S.Clay, and the positive correlation of C. squarrosa's leaf endogenesis ABA content with S.C/N do not reach significant levels (P > 0.05). It is explained that in the process of desertification, C. squarrosa's ABA content accumulates largely with the decline of S.C. The correlation direction of A. frigida's leaf endogenesis ABA content with the soil factor is opposite to that of L. chinensis’s leaf endogenesis ABA content, and the correlation degree is all very significant (P < 0.01). The correlation between M. ruthe- nica's leaf endogenesis ABA content and the soil factor is insignificant (P > 0.05).
3
Discussion
In the process of grassland desertification, as an impressible kind[12], L. chinensis’ physiological indices (EC, PRO, MDA, protection enzyme and ABA) have higher correlations with SM and S.C/N than others; therefore, SM and S.C/N are the key factors to result in the damage on the physiological level of L. chinensis. Among these physiological indices its MDA and ABA have a higher correlation with the soil factor. Considerable analysis on the environmental stress and the plant physiological indices has indicated that MDA is a more active － factor[15 20]. Organism's damage mechanism of produced ROS (active oxygen) is mainly related to DNA oxidation damage[21]. In recent years, more and more careful research on ABA and resistibility has discovered that ABA as the important cell adversity signal matter has conducted a series of gene expres－ sion[22 24]. Moreover, the route from the perception of cells to the primary signal of water stress to the coding of the key enzyme gene expression in ABA biosynthesis is a most essential pass of cell adversity information transmission[25]. In addition, under the drought stress ABA as the signal conduction pass causes the gene expression of responses, and is a kind of required mediator when the plant responds to adversity[26]. So in this experiment it is guessed that ABA is possibly the key starter, which causes the injury that L. chinensis suffers on its physiological level. ABA takes the roles of conduction and
Table 10 Correlations between ABA contents of the plants and soil moisture, the content of clay, C , N, C/N in soil Species
Correlation coefficient SM
L．chinensis
− 0.780**
C．squarrosa
− 0.321
A．frigida
0.723**
M．ruthenica
0.211
S.C/N
S.Clay
0.380**
− 0.745**
− 0.842**
− 0.672**
0.224
− 0.313
− 0.337*
− 0.295
− 0.816**
0.186
S.C
S.N
0.755**
0.613**
0.798**
0.206
0.222
0.197
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
response linkage, thus aggravating the peroxidation of membrane lipid until responding to DNA damage through information transmission and gene expression[27]. In the process of grassland sandy desertification, C. squarrosa and A. frigida as resistant types[12] have different response mechanism to the soil from that of L. chinensis. The leading factor is not prominent, and comparatively S.C is more important to C.squarrosa’s physiological metabolism. The phy- siological influence of S.C/N to A .frigida is more obvious, the one of S.N is the next, and even those of SM and S.Clay are also very important. Both ABA relations to 5 soil indices are consistent. Similarly, their correlations were very significant (P<0.01). Only the relation of S.C/N among 5 soil indices to both MDA was consistent. And their correlations were very significant (P<0.01). On the physiological level, on the whole, A. frigida community has more obvious correlation with the soil factor than C. squarrosa community. Changing the soil factor would be more influential to A. frigida's physiology. So it is very important to protect A. frigida's soil under LD stage of the sandy desertification. At the same time among these physiological indices, A. frigida’s ABA is affected by soil greatly, which is the same as L. chinensis and also relates to ABA inducing resistive gene[28] besides the characteristics acting on L. chinensis. This suggests that the ecosystem is likely not only to strengthen its resistivity and thus develop to the direction of restoration under this stage, but also to continue to be damaged and thus develop to heavy desertification. The correlation direction of A. frigida's ABA content (influenced heaviest by soil) with the soil factor is opposite to that of L. chinensis’s ABA content, which reflects their difference in the desertification resistance and adaptability, and finally causes the difference in their shape and growth[29]. M. ruthenica (a retarded type)[12] is more sensitive to the soil’s physical characteristics (SM and S.Clay) than to the soil’s chemistry characteristics (S.C, S.N and S.C/N). It is suggested that the decline of nutritious elements in the soil was not the leading factor of holding back its growth. But in the process of desertification evolvement the most important characteristic of the soil environment is the decline of element C and N in the soil. This point has also confirmed that it is not sensitive to the desertification. Moreover, the relation between M. ruthenica’s ABA and the soil factor is not close (P > 0.05), which also tallies with it. Under the environmental stress, the change in the plant’s physiological metabolism is a traditional focus of research on the mechanism of resistibility. Synthesizing of the correlation analysis showed that the plant’s MDA and ABA respond intensively to the environment stress of desertification. MDA is an important index to reflect the peroxidation of membrane － lipid[15 20]. According to Gayler et al.’s report[30], under the drought stress the plant’s ABA may remarkably increase osmoregulation, and induct PRO accumulation. In the process of
desertfication, ABA as a "stress hormone" causes the peroxidation of membrane lipid through signal conduction[26], increases EC, and thus leads to ABA accumulation; furthermore it may induce the increase of PRO[31], which has been firmly confirmed by this research. Moreover, from the correlation analysis between the above 4 common populations and the soil factor, by comparing the physiological indices not related to all the 5 soil indices, it can be found that no such index exists in L. chinensis; only one such index exists in C. squarrosa, that is, the CAT activity; only one such index exists in A. frigida', that is, the POD activity; however, this index is ABA in M. ruthenica. The response of ABA as a "stress hormone" is usually the most sensitive and the most essential under numerous environmental stresses[32,33]. And some experiments have already proved that ABA is a starter for such physiological reactions as damage and adaptation[34,35]. So analysis of the correlation between the common population and the soil factor further explains and shows that M. ruthenica is the most insensitive in the desertification environment, A. frigida and C. squarrosa are the next, but L. chinensis is the most sensitive to the desertification environment. The results prove that for each of the stress-resistant types, there are differences in the response mechanism to the soil factor under different desertification stages on sandy grassland. The sensitive type (L. chinensis) responds intensively to SM and S.C/N. The response of the resistant type (A. frigida and C. squarrosa) to the soil factor does not show dominant factors. Altogether, the physiological response of the plants among the 4 common populations has an almost significant correlation with S.C/N. So for preventing the sandy desertification of grassland, the most effective measure is to control the key soil factor, that is, S.C/N. This point can fully apply to our actual fathering work. Certain measure can be taken to the soil so that the soil can be restored gradually to benefit the better development of grassland. In brief, some physiological changes in plants are contradictory themselves. It is necessary to explore the mechanism, and solve contradictions using considerable experiments. This experiment has obtained some new understanding; however, it needs to be further confirmed.
Acknowledgements The authors would like to express their gratitude to Duolun restoration ecology experimental demonstration station, Institute of Botany, CAS, China, and grassland workstation in Duolun County, Inner Mongolia, China. This research was supported by the National 973 Project, China (No.G2000048704), Natural Science Foundation of Shaanxi Province, China (No. 2006Z05) and Scientific Rese arch Foundation of Northwest University, China (No. 04NW60).
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
References [ 1 ] Cheng X. Frontier issue of modern ecology in the study of ecotone between agriculture and animal husbandry. Resources Science, 1999, 21(5): 2–8. [ 2 ] Wang B F, Wang Z G, Jiang Z P, et al. Impact of shelter-forest patterns on rehabilitation of desertified land in arid Area. Journal of Desert Research, 2003, 23(3): 236–241. [ 3 ] Kang M Y, Dong S K, Huang X X, et al. Ecological regionalization of suitable tree shrubs and herbages for vegetation restoration in the farming—pastoral zone of North China. Acta Botanica Sinica, 2003, 45(10): 1157–1165. [ 4 ] Yang X H, Zhang K B, Hou R P, et al. Vegetation variations under different exclusion measures and their correlation to soil factors. Acta Ecologica Sinica, 2005, 25(12): 3212–3219. [ 5 ] Feng W, Zhang W J, Feng X Z. Character research of plantation and soil factors in restoration process in sandlot near altiplano bashing. Agricultural Research in the Arid Areas, 2006, 24(3): 130–224. [ 6 ] Zhou Z Y, Li F R, Chen Y M, et al. Growth and reproduction and their relationships with soil moisture in artificially established Artemisia sphaerocephala populations of different densities in the Alxa Desert. Acta Ecologica Sinica, 2004, 24(5): 895–899. [ 7 ] Hao Z Q, Guo S L. Canonical correspondence analysis on relationship of herbs with their environments on northern slope of Changbai Mountain. Acta Ecologica Sinica, 2003, 23(10): 2000– 2008. [ 8 ] Cui S X, Wang W, Chen G C, et al. Seasonal changes of endogenous phytohormones, fatty acid composition of chloroplast membrane and enzymes of membrane lipid antioxidation system in two desert plants. Acta Phytoecologica Sinica, 2000, 24(1): 96–101. [ 9 ] Zhou R L, Zhao H L, Wang H O. Characters of resisting adverse environment in vegetation evolution in Horqin Sandland. Journal of Desert Research, 1999, 19(Supp.1): 1–6. [10] Zhou R L, Wang H O, Zhao H L. Physiological changes and high-temperature resistance of Artemisia halodendron and A. frigida. Journal of Desert Research, 1999, 19(Supp.1): 59–64. [11] Zhou R L, Sun G J, Wang H O. Osmoregulation changes in desert plants under drought and high temperature stresses, related to their resistance. Journal of Desert Research, 1999, 19(Supp.1): 18–22. [12] Zhu Z M, Yang C. Researches on stress resistant types of the plants during the grassland sandy desertification process. Acta Ecologica Sinica, 2004, 24(6): 1093–1100. [13] Nanjing Agriculture Institute ed. Soil Farming and Chemistry Analysis. Bejing, China: Agricultural Press, 1985. 31–52. [14] Lao P F. Soil Physical Chemistry. Beijing, China: Agricultural Press, 1985. 23–27. [15] Foyer C H, Harbinson J. Oxygen metabolism and the regulation of photosynthetic electron transport. In: Foyer C H,
Mullineaux P M, eds. Causes of Photooxidative Stress and Amelioration of Defense Systems in Plants. Boca Raton: CRC Press, 1994. 1–42. [16] Bowler C, van Montagu M, Inze D. Superoxide dismutase and stress tolerance. Ann Rev Plant Physiol Plant Mol Biol, 1992, 43: 83–116. [17] Moran J F, Becana M, Iturbe-Ormaetxe I, et al. Drought induces oxidative stress in pea plants. Planta, 1994, 94: 346– 352. [18] Cakmak I, Marschner H. Magnesium deficiency and high light intensity enhance activity of superoxide dismutase ascobate peroxidase, and glutathione reductase in bean leaves. Plant Physiol, 1992, 98: 1222–1227. [19] Baisak R, Rana D, Acharga P B, et al. Alterrations in the activities of active oxygen scavenging enzymes of wheat leaves subjected to water stress. Plant and Cell Physiol, 1994, 35: 489–495. [20] Mehlhorn H, Wenzel A. Manganese deficiency enhances ozone toxicity in bush beans (Phaseolus vulgaris L. cv. Saxa). Plant Physiol, 1996, 148: 155–159. [21] Halliwell B, Aruoma O I. DNA and Free Radical. New York: Ellis Horwood Limited Press, 1993. [22] Davies W J, Zhang J. Root to shoot signals and regulation of growth and development of plants in drying soil. Annu Rev Plant Physiol Plant Mol Biol, 1991, 42: 55–76. [23] Giraudat J, Parey F, Bertauche N, et al. Current advances in abscisic acid action and signaling. Plant Mol Biol, 1994, 26: 1557–1577. [24] Ingram J, Bartels D. The molecular basis of dehydration tolerance in plants. Annu Rev Plant Physiol Plant Mol Biol, 1996, 47: 377–403. [25] Jia W S, Xing Y, Lu C M, et al. Signal transduction from water stress perception to ABA accumulation. Acta Botanica Sinica, 2002, 44(10): 1135–1141. [26] Merlot S, Giraudat J. Genetic analysis of abscisic acid signal transduction. Plant Physiol, 1997, 114: 751–757. [27] Zeng Z H, Zhang Z Y. The oxidative damage of mitochondrial DNA by free radicals and aging. Progress in Biochemistry and Biophysics, 1995, 22: 429–432. [28] Smirnoff N, Cumbes Q J. Hydroxyl radical scavenging activity of compatible solutes. Phytochemistry, 1989, 28: 1057–1060. [29] Zhu Z M, Yang C. Changes of four common plant populations growth and their anti-oxidative enzymatic system in desertification process. Chinese Journal of Applied Ecology, 2004, 15(12): 2261–2266. [30] Gayler K R, Glaziou K T. Plant enzymes synthesis: hormonal regulation of invertase and peroxidase synthesis in sugar cane. Planta, 1969, 84: 185–194. [31] Zeevaart J A, Creelman R A. Metabolism and physiology of abscisic acid. Ann Rev Plant Physiol, 1988, 39: 439–473. [32] Lopez-Carbonell M, Alegre L, van Dnckelen H A. Changes in cell ultrastructure and endogenous abscisic acid and indole-3-
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
acetic acid concentrations in Fatsia japonica leaves under
[34] Pastor A, Lopez-Carbonell M, Alegre L. Abscisic acid immu-
polyethylene glycol-induced water stress, Plant Growth Regul,
nolocalization and ultrastructural changes in water-stressed
1994, 15: 165–174.
lamender (Lavandula stoechas L.) plants. Physiol Plant, 1999,
[33] Xiong L, Schumaker K S, Zhu K J. Cell signaling during cold, drought and salt stress. Plant Cell, 2004 (Supp.1): 165– 183.
105: 272–279. [35] Shinozaki K, Yamaguchi-Shinozaki K. Gene expression and signal transduction in water-stress response. Plant Physiol, 1997, 117: 263–271.
RESEARCH PAPER
Analysis on the soil factor and physiological response of the plants in the process of sandy desertification on grassland Zhu Zhimei1,2, Yang Chi1,*, Cao Mingming2, Liu Kang2, Yang Lianan2 1 Ecology and Environment Science Department, Life Science College, Inner Mongolia University, Huhhot 010021, China 2 Environment Science Department, Northwest University, Xi’an 710069, China
Abstract: The aim was to find the most practical measure to father desertification on grassland. Three-year research was carried out from 2001 to 2003 in Duolun County, Xilin Gol, Inner Mongolia, China. A series of degradation gradients or stages were established by clustering analysis, and corresponded to 5 community types. Five community types were selected as the sampling sites. Four common plant populations (i.e., Leymus chinensis, Cleistogenes squarrosa, Artemisia frigida and Melilotoides ruthenica) were chosen because of their different physiological responses to sandy desertification. The correlation analysis was made between the soil factor (5 soil indices) and the physiological response of per plant population (7 physiological indices). The results showed that in the course of the sandy desertification on grassland, the physiological response of L. chinensis, an impressible type, had more significant correlations with the soil moisture and C/N in the soil than others (P < 0.01) relatively. The soil moisture and C/N in the soil are likely the key factors for the damage on a physiological level. Its malondialdehyde (MDA) and abscisic acid (ABA) had more significant correlations than others with 5 soil indices as a whole (P < 0.01). C. squarrosa and A. frigida are of resistant types. Only the correlations of C/N among 5 soil indices with both MDAs were consistent and very significant (P < 0.01). Both ABA relations to 5 soil indices were consistent, and, similarly, were very significant (P < 0.01). M. ruthenica, a retarded type, was more sensitive to soil physical character (the soil moisture and the content of clay) than soil chemical character (the content of C, the total N and C/N in the soil), suggesting that the degradation of nutritious elements in the soil is not the leading factor in holding back its growth. Its MDA showed a more significant correlation (P < 0.01) than others as a whole, but its ABA did not show a significant correlation with each of the soil indices (P > 0.05). The synthesis result showed that MDA and ABA in the plants responded intensively to desertification stress. For each of the stress resistant types, there were different soil response mechanisms under different stages. The impressible type responded intensively to the soil moisture and C/N. The response of the resistant type to the soil factor did not appear to be a dominant factor. Altogether, the physiological response of plants mostly had a significant correlation with the C/N in the soil. Key Words: sandy desertification; physiological response; stress resistant type; soil factor; grassland steppe
Duolun County is situated at the farming-pastoral zone of the semi-arid region in North China. Expansion of desertification land area has caused serious harm to people's production life and social economy development. In recent years, the farming-pastoral zone has received universal attention from various aspects. There are two reasons from the viewpoints of the national social economy development and the ecological environment construction. The first reason is the most impoverished areas in China; the second reason is the most serious desertification areas in China. It has typical ecological vul-
nerability[1]. Duolun County is the nearest county of Inner Mongolia apart from Beijing, and is also one of the sand storm sources in North China. Therefore, it has the vital significance fathering desertification to safeguard the ecological environment of the capital circle. The essence of the fathering desertification is to restore and reconstruct the damaged vegetation and soil, and it is also the basic safeguard of the sustainable development in the farming-pastoral zone. The efficiency of restoration and reconstruction of the desertification land relates directly to the choice and the disposition way of the plant
Received date: 20006-02-28; Accepted date: 2006-12-10 *Corresponding author. E-mail: [email protected] Copyright © 2007, Ecological Society of China. Published by Elsevier BV. All rights reserved.
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
species and the edaphic factors[2]. The former study on relationship of the soil and plant has concentrated to the community, the population and the individual level in the desertifica－ tion process[3 7], the study on damaged mechanism in view of － the grassland desertification is limited[8 11], and little involves relation change of the edaphic factor and the plant physiological biochemistry under the natural state. The plant can make the response to environmental factors on a series of different scales. Intuitive and obvious changes on the community, population, individual and shape level can occur through complex physiological processes in the plant. Under environmental stress, the plant adjusts physiological reaction to compensate its lack in nutrition for growth. In this way, the plant can grow continually. Perhaps plants’ physiological reaction and the adaptability are not only short-term and irritable, but also a more in-depth response mechanism. In the process of sandy desertification on grassland, soil is an important environmental factor. This article has studied the mechanism that the plant adjusts in the physiological reaction to adapt to the soil factor. The aim is to find the most practical measure fitting for different typical plants to father desertification on grassland, and to provide scientific proof restoring and reconstructing the damaged ecosystem.
1 Survey on research area and research technique 1. 1 Survey on research area The research area is established in Duolun County, Inner Mongolia, China. This county is located southeast Xilin Gol, Inner Mongolia: east longitude is 115º51′–116º 54′, north latitude is 41º46′–42º36′, and elevation is 1150–1800m. In the area the climate is of the monsoon climate, the wettability is 0.54, the year average temperature is 1.6℃, the annual temperature variation is 36.8℃, the diurnal temperature variation is 14.1℃, and the average frost-free period is 100d. The yearly transpiration rate is 4.5 times the yearly precipitation. The year average wind speed is 3.6m/s, and the number of gale days is 67.3d, many in spring. This county is in the south of Inner Mongolia Plateau, the northern slope of Yin Mountain, and the southern edge of Otindag Sandland. Its eastern part links up to Daxinganling Mountains. Because of the influence of the terrain, the landform is complicated with 5 types: the
low mountain knap, the knap, the river valley or the ditch valley, the piedmont clinoplain or the high platform, and the piling sand dune. The zonal soil in the county is the chestnut soil; the main soil type occupying the area with the percentage from high to low is in turn the chestnut soil (70.1%), the wind sandy soil (16.6%), the meadow soil (6.95%), and the black calcium earth (3.38%). The vegetation type is of the medium warm prairie belt in semi-drought and semi-moist zones of transition. The main vegetation types include the typical prairie, the forest prairie, the meadow prairie, the sand dune and so on. The research is carried out on the chosen desertification gradient. 1.2 Research technique 1.2.1 Choice of type places On the basis of disposal spots of the prairie at large observed in Duolun County in 1985, in August 2001 therein 35 types of places were sought with GPS, and each type place took three quadrats (1×1m2) to do the ecology investigation stochastically. In August 2002 and 2003, respectively, on the natural sandy prairie region in the county, 12 obvious desertification gradient type places were selected and each type place took three quadrats (1×1m2) to do the ecology investigation stochastically. On the basis of the basic clustering analysis on the data of community, population and soil obtained in three years, the natural sandy desertification grassland in Duolun County was classified into 5 community types:Ⅰ(OV), Ⅱ(PD), Ⅲ(LD), Ⅳ(MD) and Ⅴ(HD) (Table 1). 1.2.2 Method of sampling Plant: in 5 kinds of type places, such common populations as Leymus chinensis, Cleistogenes squarrosa, Artemisia frigida and Melilotoides ruthenica were selected. Fresh plant leaves (middle) were stochastically gathered and mixed up to every species. Some part of them was fixed by fluid nitrogen, and then transferred over to a refrigerator of –80℃ when returning back to the laboratory to be used for testing the enzyme and ABA. Some other part was used for testing the fresh weight, the chlorophyll and the relative membrane penetrability (EC). Still some other samples were heated at 105℃ for 1 h, and then dried at 65℃ for 24 h to be used for testing the leaf water content and the proline. The plant quantity of Ⅴgradient was limited, so for some indices only 4 gradients could be tested.
Table 1 Community types of different desertification stages of sandy grassland during the desertification process Desertification stage
Community type
Constructive species
Original vegetation (OV)
L. chinensis + Stipa kryovii+ bunch grass
Leymus chinensis
Potential desertification (PD)
C. squarrosa + Agrostis cristatum
Cleistogenes squarrosa
Light desertification (LD)
A. frigida + A.cristatum + C. squarrosa
Artemisia frigida
Moderate desertification (MD)
A. intramongolica + weed
Artemisia intramongolica
Heavy desertification (HD)
Annual plant
Chenopodium album
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
Soil: in August 2001, a soil profile was taken in each type place, respectively, divided into 5 layers: 0–5 cm, 5–10 cm, 10–20 cm, 20–30 cm and 30–50 cm. Soil was taken with the soil box and the ring sampler to be used for testing the soil moisture content and carrying out the granule analysis. Soil samples in layers of 0–10 cm and 10–20 cm were taken thrice at each type place with an earth drill, respectively, to be used for testing the total C and the total N contents in the soil. In August 2002 and 2003 a soil profile of 0–20 cm was taken in each type place, respectively (below 20 cm the soil change was not obvious in 2001, so the soil profile of 0–20 cm was focused in judging the change in the soil); other operations were the same as those in 2001. 1.2.3 Determination method The determinations of relative leaf water content, EC, proline (PRO) content, MDA content, protection enzyme (superoxide dismutase SOD, peroxidase POD, catalase CAT) activity and ABA content can be seen in another article of this research[12]. The content of soil moisture (SM): drying and weighing method. The content of organic C in the soil (S.C): after being processed by using 0.3 mm sieves, the soil sample was determined
using the law of K2Cr2O7 capacity[13]. The content of the total N in the soil (S.N): after being processed by using 0.3 mm sieves, the soil sample was determined using the law of Kjeldahl[13]. The granulometric analysis: soil grains with different grain sizes were separated and its percentage was calculated using pipette analysis (combined with the sieving and the water sedimentation)[14], and applying the Stokes’ Law, where the clay (S. Clay) size < 0.01mm. The above indices were determined thrice. 1.2.4 Data processing By synthesizing the tested data in 2001–2003, the Bivariate correlation analysis and the ANOVA variance analysis were carried out.
2
Result and analysis
2.1 Change in the soil factors and plant's physiological response under different desertification stages With the desertification progress of grassland, physiological response of plants and change in the soil factors are shown in Tables 2 and 3. 2.2 Correlation analysis between plant’s EC and soil factors
Table 2 EC, MDA, proline content, protection enzyme and ABA content of the plants under different desertification stages (mean ± SD) Proline content (PRO) (µg·g–1DW)
Species
Desertification stage
L．chinensis
Ⅰ
4.94±0.25
14.71±2.62
679.06±38.96
123.42±11.10
1280.49±117.16
961.98±73.89
2150.00±343.17
Ⅱ
6.08±0.40
26.96±4.08
802.26±46.36
156.36±8.70
1234.13±137.71
1111.54±97.43
1142.21±126.36
Ⅲ
5.11±0.16
25.32±3.10
405.21±41.97
98.91±6.86
881.46±53.47
856.97±54.20
1628.89±292.01
Ⅳ
6.61±1.03
55.05±12.38
564.51±106.81
162.11±25.38
961.79±56.70
643.41±77.99
2654.92±136.37
Ⅴ
9.48
Ⅰ
10.73±1.52
23.35±2.34
275.96±21.61
92.66±13.21
309.61±46.98
151.45±19.00
1330.71±145.65
Ⅱ
14.41±2.69
35.60±6.10
286.37±28.16
109.49±8.40
434.45±79.04
209.96±16.70
Ⅲ
14.51±1.31
26.21±3.36
190.85±23.05
84.93±5.75
287.47±37.28
344.41±23.77
Ⅳ
10.76±0.52
79.00±27.00
343.75±46.88
132.61±11.02
202.22±36.21
281.82±34.79
Ⅴ
15.17±1.46
80.44±8.00
70.00±13.50
225.88±28.30
Ⅰ
2.43±0.41
53.82±18.85
961.79±41.02
195.09±21.87
83.33±12.60
89.71±7.34
2552.66±202.97
Ⅱ
4.26±0.94
132.01±20.79
1246.41±56.85
234.83±9.21
101.39±14.54
133.98±29.16
4077.20±262.85
Ⅲ
5.79±0.39
91.10±21.12
536.14±61.28
155.48±18.14
111.76±32.68
402.17±45.07
2197.94±118.93
Ⅳ
4.58±0.32
119.72±23.88
950.55±101.08
218.03±18.14
84.89±12.97
353.50±46.63
2438.20±291.40
Ⅴ
6.44±0.35
173.42±17.06
85.41±22.10
504.17±82.04
Ⅰ
3.78±0.43
15.80±4.19
1484.73±98.74
219.50±20.27
481.57±75.32
617.49±83.19
3094.77±278.21
Ⅱ
4.28±0.51
66.81±26.44
1687.89±124.80
277.13±19.36
563.26±76.45
1118.98±73.70
3833.46±583.21
Ⅲ
5.11±0.35
61.53±20.03
1172.36±102.76
289.53±18.99
293.72±35.00
675.59±43.81
3886.04±516.68
Ⅳ
5.70±0.34
51.66±21.03
1272.72±88.84
417.81±19.99
428.26±55.67
654.14±41.35
3658.82±463.61
Ⅴ
5.45±0.40
318.30±19.88
105.25±20.30
896.28±24.12
C．squarrosa
A．frigida
M．ruthenica
EC (%)
EC: Relative membrane permeability
MDA content (nmol·g–1DW)
SOD activity (U.g–1DW)
POD activity (U·g–1DW·min-1)
CAT activity (µmol·g–1DW·min-1)
ABA content (ng·g–1DW)
718.54±102.57 762.57±58.75 861.07±123.00
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
L. chinensis's leaf EC has a significant negative correlation with SM (P < 0.05) (Table 4), and has very significant negative correlations with S.C and S.C/N (P < 0.01). It also has negative correlations with S.Clay and S.N, but the correlations are not close (P > 0.05). It is explained that S.C/N and the decline of S.C are essential factors influencing the damage of cell membrane of L. chinensis. C. squarrosa's leaf EC has significant positive correlations with S.Clay and S.C (P < 0.05), and also has positive correlations with SM, S.C/N and S.N, but the correlations are not close (P > 0.05). It is explained that S.Clay and S.C influence its EC mainly. With the decline of S.Clay and S.C, C. squarrosa's EC reduces, and the situation of its damaged cell membrane is different from that of L. chinensis entirely. A. frigida's leaf EC has a very significant positive correlation with S.C/N (P < 0.01), has a significant negative correlation with SM (P < 0.05), and has a very significant negative correlation with S.N (P < 0.01), but the negative correlations with S.C and S.Clay do not reach a significant level (P > 0.05). It is explained that S.C/N and S.N affect its EC greatly. With the decline of S.N and the increase of S.C/N, A. frigida's EC increases. The correlation of M. ru-
thenica's leaf EC with the soil factor is the same as that of A. frigida’s leaf EC with differences that M. ruthenica's leaf EC has very significant correlations with S.Clay and S.C (P < 0.01), and A. frigida’s leaf EC does not have significant correlations with S.Clay and S.C (P > 0.05). 2.3 Correlation analysis between plant’s proline content and soil factors L. chinensis's leaf PRO has very significant negative correlations with SM, S.C, S.N and S.Clay (P < 0.01) (Table 5), and has a negative correlation with S.C/N, but the correlation is not close (P > 0.05). The decline of SM, S.C, S.N and S.Clay arouses a mass of accumulation of L. chinensis's leaf PRO. The correlation of C. squarrosa's leaf PRO with the soil factor is the same as that of L. chinensis’s leaf PRO, which also reflects that the decline of most soil factors (except S.C/N) arouses a mass of accumulation of C. squarrosa's leaf PRO. The correlations of A. frigida's leaf PRO with S.N and S. C/N are opposite to those of A. frigida's leaf EC, and has a very significant positive correlation with S.Clay. It is explained that S.C/N and S.Clay greatly affect A. frigida's leaf PRO. The correlation of M. ruthenica’s endogenesis PRO with the soil
Table 3 The soil moisture and the content of clay, C , N, C/N in the soil under different desertification stages (mean ± SD) Desertification stage
Soil moisture (SM)
The content of clay (S.Clay) (<0.01mm)
The content of organic carbon in the soil (S.C)
The content of total N in the soil (S.N)
C/N in the soil (S.C/N)
Ⅱ
24.84±3.96
12.53±6.76
1.33±0.82
0.16±0.06
8.11
Ⅲ
15.13±2.68
4.86±3.82
1.07±0.58
0.08±0.03
13.96
Ⅳ
10.86±1.83
2.49±2.35
0.82±0.48
0.07±0.04
12.36
Ⅴ
4.28±0.01
2.19±1.32
0.29±0.14
0.03±0.02
28.17
Table 4 Correlations between EC of the plants and the soil moisture, the content of clay, C , N and C/N in the soil Correlation coefficient
Species SM
S.C/N
L．chinensis
− 0.361*
− 0.492**
C．squarrosa
0.285
0.294
S.Clay
S.C
− 0.298
S.N
− 0.517**
0.348*
− 0.184
0.350*
0.242
A．frigida
− 0.329*
0.462**
− 0.297
− 0.263
− 0.545**
M．ruthenica
− 0.45**
0.329*
− 0.478**
− 0.461**
− 0.441**
SM: soil moisture; S.Clay: the content of clay in the soil; S.C: the content of organic carbon in the soil; S.N: the content of the total N in the soil * indicates significant correlations (P<0.05); ** indicates very significant correlations (P<0.01); the same below
Table 5 Correlations between PRO of the plants and the soil moisture, the content of clay, C , N and C/N in the soil Correlation coefficient
Species SM
S.C/N
S.Clay
S.C
S.N
L．chinensis
− 0.781**
− 0.200
− 0.716**
− 0.926**
− 0.587**
C．squarrosa
− 0.696**
− 0.277
− 0.628**
− 0.849**
− 0.498**
A．frigida M．ruthenica
0.318
− 0.674**
0.372**
0.395**
− 0.256
0.384**
− 0.185
0.416**
0.464** 0.358*
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
factor is the same as that of A. frigida’s endogenesis PRO, but is opposite to that of L. chinensis’s endogenesis PRO and C. squarrosa’s endogenesis PRO. 2.4 Correlation analysis between plant’s MDA content and soil factors L. chinensis's leaf MDA content has a very significant negative correlation with S.C/N (P < 0.01) (Table 6), and has very significant positive correlations with SM, S.C, S.N and S.Clay (P < 0.01). It is explained that the soil factor is very close to L. chinensis's leaf MDA content. C. squarrosa's leaf MDA content has very significant negative correlations with SM, S.C/N and S.C (P < 0.01), and the negative correlations
of C. squarrosa's leaf MDA content with S.N and S.Clay are not significant (P > 0.05). It is explained that S.C/N and the decline of S.C and SM are essential factors arousing C. squarrosa's MDA accumulation. A. frigida's leaf MDA content has very significant positive correlations with SM, S.N and S.Clay (P < 0.01), has a very significant negative correlation with S.C/N (P < 0.01), and does not have significant correlations with S.C (P > 0.05). It is explained that with the increase of S.C/N and the decline of S.Clay and S.N, A. frigida's MDA content reduces. The correlation of M. ruthenica’s MDA with the soil factor is similar to that of A. frigida’s MDA, but the correlation of M. ruthenica’s MDA with S.C is very signifi-
Table 6 Correlations between MDA contents of the plants and the soil moisture, the S.Clay, the contents of C and N, and C/N in soil Correlation coefficient
Species SM
S.C/N
S.Clay
0.655**
− 0.997**
0.719**
− 0.376**
− 0.486**
A．frigida
0.485**
− 0.947**
0.561**
0.275
0.684**
M．ruthenica
0.646**
− 0.772**
0.680**
0.535**
0.728**
L．chinensis C．squarrosa
− 0.312
S.C 0.464** − 0.533**
S.N 0.820** − 0.197
Table 7 Correlations between the SOD activity of the plants and the soil moisture, the S.Clay, the content of C and N, and C/N in soil Correlation coefficient
Species SM
S.C/N
L．chinensis
− 0.224
− 0.551**
S.Clay
S.C
S.N
− 0.187
− 0.370**
0.287
− 0.328*
C．squarrosa
− 0.231
− 0.398**
− 0.283
A．frigida
− 0.245
− 0.378**
0.260
M．ruthenica
− 0.382**
0.192
− 0.419**
− 0.178 − 0.441**
− 0.196
0.312 − 0.334*
Table 8 Correlations between the POD activity of the plants and the soil moisture, the S.Clay, the content of C and N, and C/N in soil Correlation coefficient
Species SM
S.C/N
S.Clay
S.C
S.N
L．chinensis
0.474**
− 0.547**
0.495**
0.406**
0.524**
C．squarrosa
0.661**
− 0.576**
0.611**
0.654**
0.637**
A．frigida
0.245
− 0.192
0.230
0.264
0.221
M．ruthenica
0.596**
− 0.766**
0.523**
0.494**
0.654**
Table 9 Correlations between the CAT activity of the plants and the soil moisture, the S.Clay, the content of C and N, and C/N in soil Correlation coefficient
Species SM L．chinensis
0.648**
C．squarrosa
0.205
A．frigida M．ruthenica
− 0.491**
0.451**
S.C/N
S.Clay
− 0.440**
0.636**
S.C 0.658**
S.N 0.603**
0.259
− 0.222
0.518**
− 0.487**
− 0.443**
− 0.525**
0.593**
0.429**
0.511**
− 0.331*
0.271
− 0.234
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
cant (P < 0.01). 2.5 Correlation analysis between plant’s protection enzyme activity and soil factors L. chinensis's leaf SOD activity has very significant negative correlations with S.C and S.C/N (P < 0.01) (Table 7), has negative correlations with SM and S.Clay, and has a positive correlation with S.N, but the three correlation coefficients do not reach significant levels (P > 0.05). It is explained that S.C/N and the decline of S.C are essential factors to arouse L. chinensis's SOD activity. But with the decline of S.N, its SOD activity reduces. L. chinensis's leaf POD and CAT activities have a very significant negative correlation with S.C/N (P < 0.01) (Tables 8, 9), and have very significant positive correlations with SM, S.C, S.N and S.Clay (P < 0.01). With the increase of S.C/N and the decline of SM, S.C, S.Clay and S.N, L. chinensis's leaf POD and CAT activities reduce greatly. Among them, relation between its POD activity and S.C/N are closest, so is the relation between its CAT activity and S.C. The correlation of C. squarrosa's leaf SOD activity with the soil factor accords with that of L. chinensis’s leaf SOD activity, and thus S.C/N and the decline of S.C are essential factors to arouse C. squarrosa's SOD activity with the difference from L. chinensis that the inapparent correlation of C. squarrosa's SOD activity with S.N is negative. It is explained that with the decline of S.N, C. squarrosa's SOD activity increases. The correlation direction between A. frigida's leaf SOD activity and the soil factor accords with that of L. chinensis and C. squarrosa, but the correlation degree is inferior to that of the latter two. None but S.C/N has a very significant negative correlation with its SOD activity (P<0.01), and the rest do not have significant correlations (P>0.05). The correlation of A. frigida's with the soil factor is opposite to that of L. chinensis’s CAT activity. The correlation between M. ruthenica's leaf protection enzyme activity and the soil factor is stronger, and with the decline of SM, nutrient and fine matter in the soil, M. ruthenica's leaf SOD activity increases, and M. ruthenica's leaf POD and CAT activities reduce. Among the 3 protection enzymes, M. ruthenica's leaf POD and CAT activities respond more intensively to environmental changes of soil. 2.6 Correlation analysis between plant’s ABA content and soil factors L. chinensis's leaf endogenesis ABA content has a very significant positive correlation with S.C/N (P < 0.01) (Table 10),
has very significant negative correlations with SM, S.Clay, S.N and S.C (P < 0.01), among witch the correlation with S.C is closest. C. squarrosa's leaf endogenesis ABA content has a significant negative correlation merely with S.C (P < 0.05). The negative correlations of C. squarrosa's leaf endogenesis ABA content with SM, S.N and S.Clay, and the positive correlation of C. squarrosa's leaf endogenesis ABA content with S.C/N do not reach significant levels (P > 0.05). It is explained that in the process of desertification, C. squarrosa's ABA content accumulates largely with the decline of S.C. The correlation direction of A. frigida's leaf endogenesis ABA content with the soil factor is opposite to that of L. chinensis’s leaf endogenesis ABA content, and the correlation degree is all very significant (P < 0.01). The correlation between M. ruthe- nica's leaf endogenesis ABA content and the soil factor is insignificant (P > 0.05).
3
Discussion
In the process of grassland desertification, as an impressible kind[12], L. chinensis’ physiological indices (EC, PRO, MDA, protection enzyme and ABA) have higher correlations with SM and S.C/N than others; therefore, SM and S.C/N are the key factors to result in the damage on the physiological level of L. chinensis. Among these physiological indices its MDA and ABA have a higher correlation with the soil factor. Considerable analysis on the environmental stress and the plant physiological indices has indicated that MDA is a more active － factor[15 20]. Organism's damage mechanism of produced ROS (active oxygen) is mainly related to DNA oxidation damage[21]. In recent years, more and more careful research on ABA and resistibility has discovered that ABA as the important cell adversity signal matter has conducted a series of gene expres－ sion[22 24]. Moreover, the route from the perception of cells to the primary signal of water stress to the coding of the key enzyme gene expression in ABA biosynthesis is a most essential pass of cell adversity information transmission[25]. In addition, under the drought stress ABA as the signal conduction pass causes the gene expression of responses, and is a kind of required mediator when the plant responds to adversity[26]. So in this experiment it is guessed that ABA is possibly the key starter, which causes the injury that L. chinensis suffers on its physiological level. ABA takes the roles of conduction and
Table 10 Correlations between ABA contents of the plants and soil moisture, the content of clay, C , N, C/N in soil Species
Correlation coefficient SM
L．chinensis
− 0.780**
C．squarrosa
− 0.321
A．frigida
0.723**
M．ruthenica
0.211
S.C/N
S.Clay
0.380**
− 0.745**
− 0.842**
− 0.672**
0.224
− 0.313
− 0.337*
− 0.295
− 0.816**
0.186
S.C
S.N
0.755**
0.613**
0.798**
0.206
0.222
0.197
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
response linkage, thus aggravating the peroxidation of membrane lipid until responding to DNA damage through information transmission and gene expression[27]. In the process of grassland sandy desertification, C. squarrosa and A. frigida as resistant types[12] have different response mechanism to the soil from that of L. chinensis. The leading factor is not prominent, and comparatively S.C is more important to C.squarrosa’s physiological metabolism. The phy- siological influence of S.C/N to A .frigida is more obvious, the one of S.N is the next, and even those of SM and S.Clay are also very important. Both ABA relations to 5 soil indices are consistent. Similarly, their correlations were very significant (P<0.01). Only the relation of S.C/N among 5 soil indices to both MDA was consistent. And their correlations were very significant (P<0.01). On the physiological level, on the whole, A. frigida community has more obvious correlation with the soil factor than C. squarrosa community. Changing the soil factor would be more influential to A. frigida's physiology. So it is very important to protect A. frigida's soil under LD stage of the sandy desertification. At the same time among these physiological indices, A. frigida’s ABA is affected by soil greatly, which is the same as L. chinensis and also relates to ABA inducing resistive gene[28] besides the characteristics acting on L. chinensis. This suggests that the ecosystem is likely not only to strengthen its resistivity and thus develop to the direction of restoration under this stage, but also to continue to be damaged and thus develop to heavy desertification. The correlation direction of A. frigida's ABA content (influenced heaviest by soil) with the soil factor is opposite to that of L. chinensis’s ABA content, which reflects their difference in the desertification resistance and adaptability, and finally causes the difference in their shape and growth[29]. M. ruthenica (a retarded type)[12] is more sensitive to the soil’s physical characteristics (SM and S.Clay) than to the soil’s chemistry characteristics (S.C, S.N and S.C/N). It is suggested that the decline of nutritious elements in the soil was not the leading factor of holding back its growth. But in the process of desertification evolvement the most important characteristic of the soil environment is the decline of element C and N in the soil. This point has also confirmed that it is not sensitive to the desertification. Moreover, the relation between M. ruthenica’s ABA and the soil factor is not close (P > 0.05), which also tallies with it. Under the environmental stress, the change in the plant’s physiological metabolism is a traditional focus of research on the mechanism of resistibility. Synthesizing of the correlation analysis showed that the plant’s MDA and ABA respond intensively to the environment stress of desertification. MDA is an important index to reflect the peroxidation of membrane － lipid[15 20]. According to Gayler et al.’s report[30], under the drought stress the plant’s ABA may remarkably increase osmoregulation, and induct PRO accumulation. In the process of
desertfication, ABA as a "stress hormone" causes the peroxidation of membrane lipid through signal conduction[26], increases EC, and thus leads to ABA accumulation; furthermore it may induce the increase of PRO[31], which has been firmly confirmed by this research. Moreover, from the correlation analysis between the above 4 common populations and the soil factor, by comparing the physiological indices not related to all the 5 soil indices, it can be found that no such index exists in L. chinensis; only one such index exists in C. squarrosa, that is, the CAT activity; only one such index exists in A. frigida', that is, the POD activity; however, this index is ABA in M. ruthenica. The response of ABA as a "stress hormone" is usually the most sensitive and the most essential under numerous environmental stresses[32,33]. And some experiments have already proved that ABA is a starter for such physiological reactions as damage and adaptation[34,35]. So analysis of the correlation between the common population and the soil factor further explains and shows that M. ruthenica is the most insensitive in the desertification environment, A. frigida and C. squarrosa are the next, but L. chinensis is the most sensitive to the desertification environment. The results prove that for each of the stress-resistant types, there are differences in the response mechanism to the soil factor under different desertification stages on sandy grassland. The sensitive type (L. chinensis) responds intensively to SM and S.C/N. The response of the resistant type (A. frigida and C. squarrosa) to the soil factor does not show dominant factors. Altogether, the physiological response of the plants among the 4 common populations has an almost significant correlation with S.C/N. So for preventing the sandy desertification of grassland, the most effective measure is to control the key soil factor, that is, S.C/N. This point can fully apply to our actual fathering work. Certain measure can be taken to the soil so that the soil can be restored gradually to benefit the better development of grassland. In brief, some physiological changes in plants are contradictory themselves. It is necessary to explore the mechanism, and solve contradictions using considerable experiments. This experiment has obtained some new understanding; however, it needs to be further confirmed.
Acknowledgements The authors would like to express their gratitude to Duolun restoration ecology experimental demonstration station, Institute of Botany, CAS, China, and grassland workstation in Duolun County, Inner Mongolia, China. This research was supported by the National 973 Project, China (No.G2000048704), Natural Science Foundation of Shaanxi Province, China (No. 2006Z05) and Scientific Rese arch Foundation of Northwest University, China (No. 04NW60).
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
References [ 1 ] Cheng X. Frontier issue of modern ecology in the study of ecotone between agriculture and animal husbandry. Resources Science, 1999, 21(5): 2–8. [ 2 ] Wang B F, Wang Z G, Jiang Z P, et al. Impact of shelter-forest patterns on rehabilitation of desertified land in arid Area. Journal of Desert Research, 2003, 23(3): 236–241. [ 3 ] Kang M Y, Dong S K, Huang X X, et al. Ecological regionalization of suitable tree shrubs and herbages for vegetation restoration in the farming—pastoral zone of North China. Acta Botanica Sinica, 2003, 45(10): 1157–1165. [ 4 ] Yang X H, Zhang K B, Hou R P, et al. Vegetation variations under different exclusion measures and their correlation to soil factors. Acta Ecologica Sinica, 2005, 25(12): 3212–3219. [ 5 ] Feng W, Zhang W J, Feng X Z. Character research of plantation and soil factors in restoration process in sandlot near altiplano bashing. Agricultural Research in the Arid Areas, 2006, 24(3): 130–224. [ 6 ] Zhou Z Y, Li F R, Chen Y M, et al. Growth and reproduction and their relationships with soil moisture in artificially established Artemisia sphaerocephala populations of different densities in the Alxa Desert. Acta Ecologica Sinica, 2004, 24(5): 895–899. [ 7 ] Hao Z Q, Guo S L. Canonical correspondence analysis on relationship of herbs with their environments on northern slope of Changbai Mountain. Acta Ecologica Sinica, 2003, 23(10): 2000– 2008. [ 8 ] Cui S X, Wang W, Chen G C, et al. Seasonal changes of endogenous phytohormones, fatty acid composition of chloroplast membrane and enzymes of membrane lipid antioxidation system in two desert plants. Acta Phytoecologica Sinica, 2000, 24(1): 96–101. [ 9 ] Zhou R L, Zhao H L, Wang H O. Characters of resisting adverse environment in vegetation evolution in Horqin Sandland. Journal of Desert Research, 1999, 19(Supp.1): 1–6. [10] Zhou R L, Wang H O, Zhao H L. Physiological changes and high-temperature resistance of Artemisia halodendron and A. frigida. Journal of Desert Research, 1999, 19(Supp.1): 59–64. [11] Zhou R L, Sun G J, Wang H O. Osmoregulation changes in desert plants under drought and high temperature stresses, related to their resistance. Journal of Desert Research, 1999, 19(Supp.1): 18–22. [12] Zhu Z M, Yang C. Researches on stress resistant types of the plants during the grassland sandy desertification process. Acta Ecologica Sinica, 2004, 24(6): 1093–1100. [13] Nanjing Agriculture Institute ed. Soil Farming and Chemistry Analysis. Bejing, China: Agricultural Press, 1985. 31–52. [14] Lao P F. Soil Physical Chemistry. Beijing, China: Agricultural Press, 1985. 23–27. [15] Foyer C H, Harbinson J. Oxygen metabolism and the regulation of photosynthetic electron transport. In: Foyer C H,
Mullineaux P M, eds. Causes of Photooxidative Stress and Amelioration of Defense Systems in Plants. Boca Raton: CRC Press, 1994. 1–42. [16] Bowler C, van Montagu M, Inze D. Superoxide dismutase and stress tolerance. Ann Rev Plant Physiol Plant Mol Biol, 1992, 43: 83–116. [17] Moran J F, Becana M, Iturbe-Ormaetxe I, et al. Drought induces oxidative stress in pea plants. Planta, 1994, 94: 346– 352. [18] Cakmak I, Marschner H. Magnesium deficiency and high light intensity enhance activity of superoxide dismutase ascobate peroxidase, and glutathione reductase in bean leaves. Plant Physiol, 1992, 98: 1222–1227. [19] Baisak R, Rana D, Acharga P B, et al. Alterrations in the activities of active oxygen scavenging enzymes of wheat leaves subjected to water stress. Plant and Cell Physiol, 1994, 35: 489–495. [20] Mehlhorn H, Wenzel A. Manganese deficiency enhances ozone toxicity in bush beans (Phaseolus vulgaris L. cv. Saxa). Plant Physiol, 1996, 148: 155–159. [21] Halliwell B, Aruoma O I. DNA and Free Radical. New York: Ellis Horwood Limited Press, 1993. [22] Davies W J, Zhang J. Root to shoot signals and regulation of growth and development of plants in drying soil. Annu Rev Plant Physiol Plant Mol Biol, 1991, 42: 55–76. [23] Giraudat J, Parey F, Bertauche N, et al. Current advances in abscisic acid action and signaling. Plant Mol Biol, 1994, 26: 1557–1577. [24] Ingram J, Bartels D. The molecular basis of dehydration tolerance in plants. Annu Rev Plant Physiol Plant Mol Biol, 1996, 47: 377–403. [25] Jia W S, Xing Y, Lu C M, et al. Signal transduction from water stress perception to ABA accumulation. Acta Botanica Sinica, 2002, 44(10): 1135–1141. [26] Merlot S, Giraudat J. Genetic analysis of abscisic acid signal transduction. Plant Physiol, 1997, 114: 751–757. [27] Zeng Z H, Zhang Z Y. The oxidative damage of mitochondrial DNA by free radicals and aging. Progress in Biochemistry and Biophysics, 1995, 22: 429–432. [28] Smirnoff N, Cumbes Q J. Hydroxyl radical scavenging activity of compatible solutes. Phytochemistry, 1989, 28: 1057–1060. [29] Zhu Z M, Yang C. Changes of four common plant populations growth and their anti-oxidative enzymatic system in desertification process. Chinese Journal of Applied Ecology, 2004, 15(12): 2261–2266. [30] Gayler K R, Glaziou K T. Plant enzymes synthesis: hormonal regulation of invertase and peroxidase synthesis in sugar cane. Planta, 1969, 84: 185–194. [31] Zeevaart J A, Creelman R A. Metabolism and physiology of abscisic acid. Ann Rev Plant Physiol, 1988, 39: 439–473. [32] Lopez-Carbonell M, Alegre L, van Dnckelen H A. Changes in cell ultrastructure and endogenous abscisic acid and indole-3-
ZHU Zhimei et al. / Acta Ecologica Sinica, 2007, 27(1): 48–57
acetic acid concentrations in Fatsia japonica leaves under
[34] Pastor A, Lopez-Carbonell M, Alegre L. Abscisic acid immu-
polyethylene glycol-induced water stress, Plant Growth Regul,
nolocalization and ultrastructural changes in water-stressed
1994, 15: 165–174.
lamender (Lavandula stoechas L.) plants. Physiol Plant, 1999,
[33] Xiong L, Schumaker K S, Zhu K J. Cell signaling during cold, drought and salt stress. Plant Cell, 2004 (Supp.1): 165– 183.
105: 272–279. [35] Shinozaki K, Yamaguchi-Shinozaki K. Gene expression and signal transduction in water-stress response. Plant Physiol, 1997, 117: 263–271.