versión impresa ISSN 0187-7151
Act. Bot. Mex no.97 Pátzcuaro oct. 2011
Phenotypic variation and leaf fluctuating asymmetry in natural populations of Parrotia persica (Hamamelidaceae), an endemic species from the Hyrcanian forest (Iran)
Variación fenotípica y fluctuación de la asimetría de la hoja en una población natural de Parrotia persica (Hamamelidaceae), una especie endémica del bosque hircaniano (Iran)
Ali Sattarian1, Mohammad Reza Akbarian2, Mehrdad Zarafshar3,6, Piero Bruschi 4, Payam Fayyaz5
1 GonbadKavoos University, Department of Forestry, Golestan, Iran.
2,3 Tarbiat Modares University, Natural Resources Faculty, Department of Forestry, Noor, Mazandran, Iran.
4 University of Florence, Department of Agriculture Biotechnology, Section of Environmental and Applied Botany, Florence, Italy.
5 University of Yasouj, Faculty of Agriculture and Institute of Natural Research and Environment, Department of Forestry, Yasouj, Iran.
6 Corresponding author: Mehrdadzarafshar@gmail.com
Recibido en octubre de 2010.
Aceptado en agosto de 2011.
Phenotypic variation in four natural populations of Parrotia persica (Hamamelidaceae), a species considered a living fossil endemic to the Hyrcanian forest, was evaluated through leaf morphometric and multivariate analysis. Furthermore, plasticity and leaf fluctuating asymmetry (LFA) were studied. Our findings clearly displayed significant divergence among sites. The smallest lamina size, the highest plasticity, and the lowest LFA values were recorded in the eastern and more xeric sites (Ghorogh and Daland), thus it seems these sites endure a larger environmental stress. Although our findings showed that multivariate and leaf morphometric analyses are suitable instruments to detect levels of phenotypic variability among P. persica natural populations, further study of allozyme and DNA diversity is necessary.
Key words: leaf morphometrics, multivariate analysis, natural populations, plasticity.
Se evaluó la variación fenotípica en cuatro poblaciones naturales de Parrotia persica (Hamamelidaceae), una especie endémica considerada como un fósil viviente del bosque hircaniano, a través de un análisis morfométrico y multivariado de las hojas. Asimismo, se estudiaron la plasticidad y la asimetría foliar fluctuante (AFF). Nuestros hallazgos mostraron claramente una divergencia significativa entre sitios. El tamaño más pequeño de la lámina foliar, la plasticidad más grande y los valores más pequeños de AFF fueron registrados en los sitios más xéricos de la parte oriental (Ghorogh y Daland), por lo que parece que estos sitios están sujetos a un mayor estrés ambiental. Obtuvimos resultados interesantes, pero hace falta realizar nuevos estudios de la diversidad de aloenzimas y ADN.
Palabras clave: análisis multivariado, morfometría foliar, plasticidad, poblaciones naturales.
Parrotia persica C. A. Mey. (Persian ironwood tree) is a biological treasure of the forest ecosystems of Northern Iran, where it is considered as one of the endemic trees to the Hyrcanian forest region. This species is the last representative of a widespread and speciesrich genus that originated in the Tertiary (Mosaddegh, 2006). Unfortunately, forest managers usually overlook P. persica because this species does not produce valuable timber. Conversely, it is widely used by indigenous people for char coal and firewood production. The wood is also used by wood tuners and for weaving telephone poles. Browicz (1982) noted that wood of P. persica was "remarkably hard, heavy, dense and durable", which is why it is called the iron tree. Although P. persica is not considered to be at risk of extinction, presently it faces conservation problems. Certainly, if the destructive process continues, genetic erosion of the species is likely. For this reason, it is important to collect more information about the adaptation of this species and its ability to respond to different types and patterns of disturbance.
The capability of plants to modify their morphology and physiology in response to environmental conditions is called phenotypic plasticity (Bradshaw, 1965). Phenotypic plasticity plays an important role in resource acquisition by plants; variation of the size and placement of resourceacquiring organs such as leaves are critical to a plant's adjustment to resource availability. A further source of variation and developmental stability, is the inability of an organism to grow and mature in a steady manner, according to its phenotypic potential, under a wide variety of environmental conditions (Palmer, 1994), and has been mostly studied as fluctuating asymmetry (FA). The idea is that lowquality individuals, with respect to reproductive fitness, are developmentally less stable and unable to develop traits with precisely one value equal on the left and right sides. According to Møller & Swaddle (1997), increase in the variability of cellular growth rates under suboptimal conditions is a source of developmental instability and it might induce asymmetric growth in plant organs with bilateral symmetry. Hence, any deviation from the genetically programmed phenotype may indicate the existence of stress during development, which in turn may be related to environmental quality (Leary & Allendorf, 1989). For example, increased FA has been observed both in animals and plants in response to pollution, parasitism, grazing, competition, and drought (Freeman et al., 1995; Kozlov et al., 1996; Alados et al., 2003; Hódar, 2002, Graham et al., 2010).
The central objective of this study was to analyze patterns of morphological diversity and trait asymmetry in four wild populations of P. persica, chosen to cover a wide enough rainfall range to admit predictions of shifts in relative growth optima. Topography and climate features have an important influence on plant diversity and richness of Iran. In particular, drought coinciding with the thermal maximum is a general factor that plays a decisive role in the selection of species that make up the natural vegetation in the Hyrcanian forest (Akhani et al., 2010). A first test was per formed to verify whether there were significant morphological differences among the populations and whether these were related to environmental characteristics. Secondly, we examined if there were significant differences in levels of FA among populations. Our expectation was for increasing FA away from an optimum especially at higher drought levels.
MATERIALS AND METHODS
We sampled four natural populations of Parrotia persica in the Hyrcanian Forest (Northern Iran), including the Sisangan and Noor populations in the western Hyrcanian forest, and Ghorogh and Daland populations in its eastern portion. The geographical location of the sampling sites and their characteristics are shown in Figure 1 and Table 1, respectively. Rainfall in the region decreases from west to east (Mohajer, 2007). All four study sites have been impacted by humans, especially by tourists for fuel. On May 27 2010, we selected twenty individual trees from each site. To minimize the possibility of intraspecific crossing, at each site the trees were chosen to be at least 50 m apart from each other (Zarafshar et al., 2010). Similarly, in order to minimize positional variability, 20 leaves were collected within the crown from the external part of the canopy on the aspect facing south and always in the middle part of the first flush shoot. Ten leaves were chosen at random from each subsample after elimination of broken, incomplete, or damaged leaves.
The parameters measured included four morphological characters: lamina length (LL), lamina width (LW), petiole length (PL), and number of veins (V) (count variable). The leaf length/leaf width (LL/LW) and leaf length/petiole length (LL/ PL) ratios were calculated, because ratios relate to shape rather than size variation, and hence may provide additional information (Dickinson et al., 1987; Frampton & Ward, 1990). Fig. 2.
We calculated a measure of leaf overall size as the root square of the total leaf length Χ blade width product to correct for allometric effects (Blue & Jensen, 1988). This new variable was regressed against leaf parameters and the residuals were used as input in the successive analyses. Thus, for each character i and OTU j (Operational Taxonomic Units) sizeadjusted variables Yij were determined according to the formula:
Yij adj. = Yij Yijcap
where Yijcap was the expected value of Yij given the size of OTU j.
Descriptive statistics were computed for the four study populations. Assumptions of normality were checked for all variables with ShapiroWilk's test. Principal Components Analysis (PCA) with standardized varimax rotation was used to separate interrelationships into statistically independent basic components (Bruschi, 2010). In PCA, eigenvectors were calculated to determine the contribution of each variable to the separation of the populations. An Analysis of Variance (Oneway ANOVA) was performed in order to test the main effects of populations on the sco res of the first three factors extracted in the PCA. A Pearson correlation analysis was performed to identify the significant climatic variables (predictors) affecting PC scores extracted.
In addition, we performed a Discriminant Analysis on the total data set to investigate the multivariate relationships among the morphometric traits, and how they change among the populations. The scatter plot of the discriminant scores co rresponding to each case of each population in the multivariate space, as defined by the first two discriminant functions, was obtained to help visualize multivariate phenotypic variations. The statistical program SPSS (version 11.5 and 16) was used for all the analyses.
Total withinpopulation plasticity (Pl) was calculated for each parameter using the smallest (x) and greatest (X) mean values for any given measure, as Pl=1(x/X) (Ashton et al., 1998; Bruschi et al., 2003).
Leaf fluctuating asymmetry
Signed fluctuating asymmetry (RL) was estimated on the maximum width of lamina (MW). As suggested by Palmer (1994), distributions of signed asymmetries in all samples were tested for normality (KolmogorovSmirnov test) and a mean of zero (onesample ttest) to detect the presence of antisymmetry and directional asymmetry. Skewness and Kurtosis were also calculated for all signed differences between sides in each of the leaf traits. Given that there was no evidence of antisymmetry (KolmogorovSmirnov test: P > 0.10) or directional asymmetry (ttest: P = 0.12), the observed variation in leaf size of P. persica, expressed as the difference between right and left side, was considered to be a measure of FA (Palmer & Strobeck, 1986). FA levels were measured as the absolute difference between sides (FA1) (Palmer, 1994). The mean for a trait in each population was calculated by summing the individual FA values and dividing the total by the number of individuals in that population (Palmer & Strobeck, 1986). Linear regressions of |RL|, against the mean of the two sides (R+L)/2, were calculated in all samples to test the linear dependency of the asymmetry on trait character size. Because some relationship between individual trait size and asymmetry was observed, we used (|RL|/0.5 (R+L) as the FA index, controlled for trait size (Palmer & Strobeck, 1986). A Oneway ANOVA was performed to test the main effects of populations on FA values.
Means and standard errors (SE) for studied leaf parameters in P. persica trees of four natural populations are shown in Table 2. The highest mean leaf size was observed at western, wetter and lower sites (Sisangan and Noor), whereas the lowest mean values were measured at the eastern, higher and drier sites (Ghorogh and Daland). Leaf length/leaf width and leaf length/petiole length ratios showed the hig hest values for Ghorogh and Daland. Coefficients of variation and plasticity were generally lower at the western sites (Sisangan and Noor), while higher values were measured at the eastern sites (Table 2). The Sisangan population, with higher mean annual precipitation, seems to have on average a more homogeneous morphology (CV = 17.35 % ± 5.16, and P = 0.56 ± 0.14).
Among all leaf characters measured, the greatest plasticity was found for petiole length and LL/LP (0.76 ± 0.02 and 0.78 ± 0.02, respectively) while the lowest plasticity was observed for LL/LW (0.46 ± 0.05) (Fig 3).
To find the most important traits responsible for differences between the di fferent populations studied, a PCA was performed on all nine leaf traits. The eigenvalues, proportion of variance and cumulative proportion of the principal compo nents are presented in Table 3. Regarding PCA, the first three principal components explained > 80% of the total variance of all traits, whereas the other components comprised a small percentage of total variation (18%). The first component explai ned more than 50% of the total variance and showed high eigenvector values for all leaf size traits (lamina length, lamina width, petiole length, width in right, width in left, and number of veins). PC2 was positively and highly related with leaf length/ petiole length only. The number of the veins (V) trait contributed positively to PC3. Also, the LFA parameter contributed highly and positively to PC3.
An ANOVA showed that variation due to populations was significant in the first and third principal components extracted (Table 4). A posthoc analysis (LSD test) showed that mean factor scores of the first principal component were signifi cantly different between Daland and all other populations (P < 0.05), and between Sisangan and Noor (P < 0.05). Moreover, leaves from Ghorogh trees were morphologically more similar to Sisangan and Noor than to Daland (Table 5). Mean factor scores of the third principal component were significantly different among all popu lations (P < 0.05) (Table 5). A Pearson correlation analysis indicated that all environmental variables considered (i.e. altitude, rainfall and temperature) may explain the morphological variation among studied populations. The population average score values of PC1 were positively related to altitude (R = 0.37, P < 0.05) and temperature (R = 0.59, P < 0.05), but negatively related to rainfall (R = 0.58, P < 0.05). This suggests a significant pattern of environmental variation for the morphological traits in PC1, according to which populations from higher altitudes, growing in wetter stands, have a larger leaf size.
The results of the discriminant analysis showed that first two roots (discri minant functions) had significantly associated eigenvalues (Table 6). Function 1, accounting for more than 50% of total variance, clearly separates the Dalan site from the other three (Fig. 3). Function 2, representing 20% of the total variance, separates the Ghorogh site from the remaining three (Fig. 3). The matrix of cases correctly classified by the discriminant analysis shows that only the Daland site had the highest discriminating power (100% of cases correctly classified). Con trastingly, the Noor site showed the lowest discriminating power (50% of cases correctly classified). Generally, results of the discriminant analysis classification showed a high percentage of correctly classified cases to all actual sites (in total 75%: Table 7).
Table 7 shows the means and SE of |RL|, as well as the relative (RL) and absolute (|RL|) asymmetries for the different populations. Based on ANOVA, there were no betweenpopulation differences in Parrotia persica FA index (g.l. =3; F = 0.93; P = 0.42).
To date, information on the morphological variation among and within P. persica natural populations is scarce. Yosefzadeh et al. (2010) examined morphology of this species along an altitudinal gradient and found that leaf size variation in this species was significantly affected by altitude. However, the authors did not deter mine which environmental factors contributed to influence this trend. In this study, multivariate analyses showed significant differences among sampled populations and that these differences probably reflect morphological adaptations to different climatic conditions. Indeed, both rainfall and temperature appeared to be signifi cantly correlated with PC1, suggesting that climate variables have a significant in fluence on the leaf size of the populations analyzed. On the other hand, discriminant analysis revealed a structure in which the Daland population is well differentiated from more mesic sites (Table 7, Fig. 4) suggesting a morphological adaptation to different moisture conditions. Mean data values showed that leaves from the western sites (Sisangan and Noor) were bigger than those from the drier eastern sites (Ghorogh and Daland). The smallest leaf size was measured in the Daland population where the lowest mean annual precipitation and highest mean annual temperature were observed.
Our results agree with other studies showing that environmental heterogeneity is a major source of variation in leaf morphology (Geeske et al., 1994; VelázquezRosas et al., 2002; Mcpherson et al., 2004; Xu et al., 2009; Bruschi, 2010). However, common garden experiments are needed to determine whether morphological differences among these populations remain when individuals are reared under identical environmental conditions, and to separate genotypic from environmental effects. Differences in plasticity were found among sites. Particularly, the two more xeric sites (Ghorogh and Daland) displayed higher plasticity values than the two mesic ones (Sisangan and Noor). Among all studied characteristics, petiole length and the LL/PL ratio displayed the largest plasticity values while LL/LW showed the smallest one. Zarafshar et al. (2010) and Bruschi et al. (2003), in studying leaf morphology of natural populations of Castanea sativa and Quercus petraea, respectively, reported high plasticity for petiole length. It can be hypothesized that petiole length is strongly associated with environmental conditions and petiole length variation could play an important adaptive role by influencing the spatial distribution of leaves and the refore light interception. Differently, Yousefzadeh et al. (2010) reported the lowest plasticity value for the number of vein pairs and base angle in Parrotia persica along an elevational gradient.
We also evaluated Leaf Fluctuating Asymmetry because this trait has been shown to be positively associated with environmental stresses such as pollution load (Kozlov et al. 1996), high elevation (Wilsey et al. 1998), cold climate (Valkama & Kozlov 2001), and drought (Alados et al., 2001; Hódar, 2002; Fair & Breshears, 2005). The idea that environmental stress can increase FA makes it a potential tool to be used for biomonitoring. As suggested by Palmer (1994), knowledge of baseline levels of FA in natural populations is crucial to application of FA to conservation efforts. FA in Quercus petraea leaves (mean absolute FA= 0.127 cm) was within the reported range for a number of other woody species (Wilsey et al., 1998; Valkama & Kozlov, 2001; Hódar, 2002).
In this study we did not find any significant difference among populations. P. persica seems capable of resisting environmental changes, since no effects of climatic influences were detected. Probably, physiological plasticity buffered the environmental stress conditions, allowing the maintenance of the developmental program that is exhibited under optimum or normal conditions. As shown in Table 8, leaves from wetter sites were more asymmetrical than those of drier sites. Some evidence suggests that FA does not always increase in drier environments, although this may depend on the traits and species being investigated and on the selection history experienced by different populations. For example, in a study conducted during a period of over 8 yr, Valkama & Kozlov (2001) did not find any effect of the amount of rainfall on FA of birch leaves. Llorens et al. (2002) found that drought treatment increased leaf area FA on Vaccinium myrtillus while it did not affect leaf width FA. An analysis of FA in Quercus ilex leaves from areas under different precipitation regimes (Hódar, 2002) showed that plants living in stressful sites were more symmetrical, and that leaves from nonstressful sites were prone to FA in drought years. Moreover, field conditions expose plants to several stresses, the relative roles of which on these populations are not known in this study. Clarke (1993) emphasized that, for developmental stability studies, the 'use of natural populations is fraught with danger unless there is a good understanding of the genetic structure and evolutionary history of the populations under examination'.
In conclusion, our results indicated considerable leaf variation in the sites studied. Moreover, we showed that multivariate and leaf morphometric analyses are suitable instruments to detect levels of phenotypic variability among P. persica natural populations.
It is our duty to thank Professor John Graham (Department of Biology, Berry College, Georgia, USA) for his valuable comments on the manuscript. Also Timothy Kim (Berry College, Georgia, USA) helped us for English grammatical corrections.
Akhani, H., M. Djamali, A. Ghorbanalizadeh & E. Ramezani. 2010. Plant biodiversity of Hyrcanian relict forests, N Iran: an overview of the flora, vegetation, palaeoecology and conservation. Pak. J. Bot., Special Issue (S.I. Ali Festschrift) 42: 231258. [ Links ]
Alados, C. L., T. Navarro, J. Escós, B. Cabezudo & J. M. Emlen. 2001. Translational and fluctuating asymmetry as tools to detect stress in stressadapted and nonadapted plants. Int. J. Plant Sci. 162(3): 607616. [ Links ]
Alados, C. L., Y. Pueyo, M. L. Giner, T. Navarro, J. Escós, F. Barroso, B. Cabezudo & J. M. Emlen. 2003. Quantitative characterization of the regressive ecological succession by fractal analysis of plant spatial patterns. Ecol. Model. 163(12): 117. [ Links ]
Ashton, P. M. S., L. P. Olander, G. P. Berlyn, R. Thadani & I. R. Cameron. 1998. Change in leaf structure in relation to crown position and tree size of Betula papyrifera within fireorigin stands of interior cedarhemlock. Can. J. Bot. 76(7): 11891187. [ Links ]
Blue, M. P. & R. J. Jensen. 1988. Positional and seasonal variation in oak (Quercus: Fagaceae) leaf morphology. Am. J. Bot. 75: 939947. [ Links ]
Bradshaw, A. D. 1965. Evolutionary significance of phenotypic plasticity in plants. Adv. Genet. 13:115155. [ Links ]
Browicz, K. 1982. Chorology of trees and shrubs in south west Asia and adjacent regions. Vol. 1. Polish Scientific Publishers. WarszawaPoznaÅ. 172 pp. [ Links ]
Bruschi, P. 2010. Geographical variation in morphology of Quercus petraea (Matt.) Liebl. as related to drought stress. Plant Biosyst. 144(1): 298307. [ Links ]
Bruschi, P., P. Grossoni & F. Bussotti. 2003. Withinand amongtree variation in leaf morphology of Quercus petraea (Matt.) Liebl. natural populations. Trees 17(2): 164172. [ Links ]
Clarke, G. M. 1993. The genetic basis of developmental stability. I. Relationships between stability, heterozygosity and genomic coadaptation . Genetica 89 : 15 23. [ Links ]
Dickinson, T. A., W. H. Parker & R. E. Strauss. 1987. Another approach to leaf shape comparisons. Taxon 36(1): 120. [ Links ]
Fair, J. M. & D. D. Breshears. 2005. Drought stress and fluctuating asymmetry in Quercus undulata leaves: confounding effects of absolute and relative amounts of stress? J. Arid Environ. 62(2): 235249. [ Links ]
Frampton, C. M. & J. M. Ward. 1990. The use of ratio variables in systematics. Taxon 39(4): 586592. [ Links ]
Freeman D. C., J. H. Graham, D. W. Byrd, E. D. McArthur & W. A. Turner. 1995. Narrow hybrid zone between two subspecies of big sagebrush, Artemisia tridentata (Asteraceae). III. Developmental instability. Am. J. Bot. 82(9): 11441152. [ Links ]
Graham, J. H., S. Raz, H. HelOr & E. Nevo. 2010. Fluctuating asymmetry: methods, theory, and applications. Symmetry 2(2): 466540. [ Links ]
Geeske, J., G. Aplet & M. P. Vitousek. 1994. Leaf morphology along environmental gradients in Hawaiian Metrosideros polymorpha. Biotropica 26(1): 1722. [ Links ]
Hódar, J. A. 2002. Leaf fluctuating asymmetry of Holm oak in response to drought under contrasting climatic conditions. J. Arid Environ. 52(2): 233243. [ Links ]
Kozlov, M. V., B. J. Wilsey, J. Koricheva & E. Haukioja. 1996. Fluctuating asymmetry of birch leaves increases under pollution impact. J. Appl. Ecol. 33(6): 14891495. [ Links ]
Leary, R. F. & F. W. Allendorf. 1989. Fluctuating asymmetry as an indicator of stress: implications for conservation biology. Trends Ecol. Evol. 4(7): 214217. [ Links ]
Llorens, L., J. Peñuelas & B. Emmet. 2002. Developmental instability and gas exchange responses of a heathland shrub to experimental drought and warming. Int. J. Plant Sci. 163(6): 959967. [ Links ]
McPherson, S., D. Eamus & B. R. Murray. 2004. Seasonal impacts on leaf attributes of several tree species growing in three diverse ecosystems of southeastern Australia. Austr. J. Bot. 52: 293301. [ Links ]
Mohajer, M. R. 2007. Silviculture. University of Tehran Press. Tehran. 387 pp. [ Links ]
Møller, A. P. & J. P. Swaddle. 1997. Asymmetry, developmental stability and evolution. Oxford University Press. Oxford. 291 pp. [ Links ]
Mosaddegh, A. 2006. Geography of world forest. University of Tehran Press. Teheran. 404 pp. [ Links ]
Palmer, A. R. & C. Strobeck. 1986. Fluctuating asymmetry: measurement, analysis, patterns. Ann. Rev. Ecol. Syst. 17: 391421. [ Links ]
Palmer, A. R. 1994. Fluctuating asymmetry analyses: a primer. In: Markow, T. A. (ed.). Developmental instability: its origins and evolutionary implications. Kluwer. Dordrecht, Netherlands. pp. 335364. [ Links ]
Palmer, A. R. & C. Strobeck. 1997. Fluctuating asymmetry and developmental stability: heritability of observable variation vs. heritability of inferred cause. J. Evolution. Biol. 10(1): 3949. [ Links ]
VelázquezRosas, N., J. Meave & S. VázquezSantana. 2002. Elevational variation of leaf traits in montane rain forest tree species at La Chinantla, México. Biotropica 34(4): 534546. [ Links ]
Valkama, J. & M. V. Kozlov. 2001. Impact of climatic factors on the developmental stability of mountain birch growing in a contaminated area. J. Appl. Ecol. 38: 665673. [ Links ]
Wilsey, B. J., E. Haukioja, J. Koricheva & M. Sulkinoja. 1998. Leaf fluctuating asymmetry increases with hybridization and elevation in treeline birches. Ecology 79(6): 20922099. [ Links ]
Xu, F., W. Guo, W. Xu, Y. Wei & R. Wang. 2009. Leaf morphology correlates with water and light availability: what consequences for simple and compound leaves? Prog. Nat. Sci. 19(12): 17891798. [ Links ]
Yousefzadeh, H., M. Tabari, M. Akbarinia, M. R. Akbarian & F. Bussoti. 2010. Morphological plasticity of Parrotia persica leaves in eastern of Hyrcanain forests (Iran) is related to altitude. Nord. J. Bot. 28(3): 344349. [ Links ]
Zarafshar, M., M. Akbarinia, P. Bruschi, S. M. Hosseini, H. Yousefzadeh, M. Taieby & A. Sattarian. 2010. Phenotypic variation in chestnut (Castanea sativa Mill.) natural populations in Hyrcanian forest (north of Iran), revealed by leaf morphometrics. Folia Oecol. 37(1): 105113. [ Links ]