Introduction

The carbon (C) and nitrogen (N) contents of plant tissues change during the growing season in relation to resource uptake, storage and mobilization according to seasonal demand, combined with variability of resource availability in the environment (Millard and Grelet 2010). Carbon is stored throughout a tree (Millard and Grelet 2010) but N is stored in specific tissues and organs. Deciduous trees store N in stems, roots or wood, but evergreen trees store N mostly in the youngest needles (Dickson 1989; Millard et al. 2001; Pallardy 2008; Teitel et al. 2016). Storage of N in the leaves of evergreen trees is carried out after the completion of leaf growth. These resources are then mobilized from the start of the next growing season (Dickson 1989).

Resources (including C and N) are divided between three main functions: maintenance (including defence), growth and reproduction. These functions are often carried out at the same time and, because an organism has a limited supply of resources, allocation to one function leads to a decreased allocation to others (Levins 1968; Obeso 2002). The proportion of investment to each of these three basic functions in woody plants depends on many factors including plant age (young plants invest mostly in fast growth and defence, older plants in reproduction) and habitat richness. Most organisms on reaching maturity stop allocating resources to growth and start to allocate all the surplus energy (i.e. that not used for maintenance) to reproduction (Kozłowski and Wiegert 1986, 1987). Trees are an exception to this principle since after reaching sexual maturity, resources are still shared between primary and secondary growth and reproduction (Koenig and Knops 1998). In the case of dioecious trees with sexual dimorphism there may, therefore, be different strategies of resources allocation in males and females. This is a consequence of differences in life-history, morphological or functional traits (Agren et al. 1999; Dawson and Geber 1999; Delph 1999). Females and males have different strategies of resource investment to maximize reproductive success when resources are limited (Obeso 2002). Females usually support higher reproductive effort than males (e.g. Obeso 2002; Montesinos et al. 2006) and consequently allocate more resources to reproduction resulting in slower radial growth after reaching sexual maturity (Obeso 2002; Banuelos and Obeso 2004; Montesinos et al. 2006; Chen et al. 2014). Females are also often more demanding of habitat requirements (Obeso 2002; Chen et al. 2014) which may be a reason behind the spatial segregation of the sexes in some species (Bierzychudek and Eckhart 1988; Dawson and Ehleringer 1993; Matsushita et al. 2016). Females can, however, compensate for the higher reproductive investment (Dawson and Ehleringer 1993; Obeso 2002; Tozawa et al. 2009; Jiang et al. 2016). They can show greater gas exchange per unit area and mass in comparison to males (Dawson and Ehleringer 1993; Obeso 2002; Montesinos et al. 2012) and increase photosynthesis by, for example, increasing leaf area (Wallace and Rundel 1979; Meagher 1992; Kohorn 1994).

Carbon demand increases at the time of formation and maturation of reproductive structures (McDowell et al. 2000). However, so far, no significant differences between C concentration between males and females have been found (Chen et al. 2014; Jiang et al. 2016). For example, no seasonal difference was found in non-structural carbohydrate concentration between males and females in Juniperus thurifera (DeSoto et al. 2016).

Variable differences in N concentration have, however, been found within dioecious species. The N concentration of Rubus chamaemorus leaves was found to be higher in males, particularly within 2–3 weeks after flowering (Agren 1988). Nyssa sylvatica stems were analysed four times a year and these studies demonstrated a significant effect of gender on N concentration. At fruit maturation (September) the concentration of N was significantly higher in males than females but only in one of two sites investigated (Cipollini and Stiles 1991). Lower N concentration was also demonstrated in females of Populus cathayana (Chen et al. 2014) and Acer negundo (Dawson and Ehleringer 1993). In other studies where the level of N was examined once in the season, no differences were usually found between the sexes (Agren et al. 1999; Jiang et al. 2016). However, the C and N concentration depends on the stage of plant growth and development; thus, the lack of differences seen in these studies may be due to analyses being performed just once a year. This is confirmed by a study of Honckenya peploides by Vilas and Retuerto (2016) who found a decrease of nitrogen concentration in females during growing season, whereas in males, the decrease was noted only from April to June.

In J. thurifera, Montesinos et al. (2012) claimed that females had a long-term strategy to increase N storage to compensate for massive reproductive masting events (synchronous production of many seeds at periodic intervals) because they found that an increase in shoot N was associated with increased photosynthesis in males but not in females. This suggests that females store surplus N while males are more reactive to current nutrient availability.

Although there have been a number of studies on J. thurifera, other gymnosperms and evergreen species have been little studied. European yew (Taxus baccata L.) and common juniper (Juniperus communis L.) are appropriate models to study dioecious species because males and females have been found to respond differently to their environments. Males of T. baccata grow taller (Iszkuło et al. 2009) and both T. baccata and J. communis males show greater radial growth than females after the beginning of sexual maturity (Cedro and Iszkuło 2011; Iszkuło and Boratyński 2011). Females of T. baccata have compensatory mechanisms allowing higher reproductive investment in the form of a greater area of needles and stomatal density, and these traits are highly correlated with gas exchange intensity (Iszkuło et al. 2009; Iszkulo et al. 2013). T. baccata and J. communis are evergreen gymnosperm species. Since evergreen species are known to store N in the youngest needles, the level of this element may be particularly sensitive to differences in phenology and different male and female reproductive functions. Analyses carried out throughout the growing season are still rare and, therefore, it is not known how the C and N concentration changes during the year in dioecious plants and whether indeed females have a different strategy of N-storage than males. If so, is it a general strategy of dioecious plants or is it limited to species growing under conditions of N deficiency? Therefore, the stated hypotheses are (1) phenology influences the sex-specific pattern of yearly foliar nitrogen concentration; (2) T. baccata and J. communis show the same sex-specific pattern of foliar nitrogen concentration.

Materials and methods

Study species

Juniper and yew have common features (gymnosperms, dioecy, evergreen, anemophily, zoochory), and yet have significant reproductive differences (Table 1). For example, yew strobili open in March, while juniper much later in May. Female cones of yew (seeds with arils) develop within one growing season, while juniper female cones (“juniper berries”) mature over 2 or 3 seasons (Thomas and Polwart 2003; Thomas et al. 2007). These species also occur in different habitats; juniper is a pioneer species, occurring in dry and nutrient poor habitat while yew persists into late-successional stages where it grows in fertile habitats. Yew is an endangered species and certain woodland communities containing T. baccata have received priority habitat status EU Habitat Directive 92/43/EEC (Thomas and Polwart 2003; Thomas and Garcia-Marti 2015) although, as recently shown, juniper populations are also reducing (Thomas et al. 2007; Verheyen et al. 2009).

Table 1 Characteristics of yew and juniper
Full size table

Study area and sampling

The study was conducted using T. baccata in the Kórnik Arboretum (52°14′27.93″N; 17°5′28.21″E) Poland and J. communis in the Torzym Forestry Commission land (52°18′36.93″N; 15°7′43.07″E), Poland. Juniper is a natural species in the latter area with a density of about 553 individuals per ha (Mandryk et al. unpublished data). They grow on poor, iron-rich soils (an outwash plain with sand and gravel) in a mixed oak–pine forest Querco–Pinetum. The yew population arose from spontaneous natural regeneration and forms the understorey in the Kórnik Arboretum on brown earth soils; there are about 2500 individuals >1 m tall (Giertych 2000). The potential vegetation is a rich, eutrophic variant of Central European lowland oak–hornbeam wood—Galio sylvatici–Carpinetum (Matuszkiewicz et al. 1995). Further details about this population are given in Iszkulo and Boratynski (2004, 2005).

For analysis, 20 individuals of yew (10 male and 10 female) and 20 individuals of juniper (10 male and 10 female) of similar height and diameter with visible reproductive structures (male or female strobili buds and/or female cones) were randomly chosen. Minimum distance between individuals was 3 m.

Sampling was repeated nine times in 2012, from March to November at intervals of 1 month. Phenological observations (beginning and end of flowering, growth, strobili maturation) were also made.

Twigs with last year’s growth (determined using bud scale scars) were collected intact from the middle part of crown. A total of 30 needles from one tree were collected from last year’s growth (2011 increment). Needles were removed from the twig; the area, length and width of fresh needles were measured using WinSeedle software with accuracy 0.0001 mm (Regent Inc.). Needles were then frozen before analysis. After thawing, needles were dried at 65 °C for 3 days and weighed to allow calculation of the specific leaf area (SLA: leaf area/leaf dry mass, cm2 g−1). Dried needles were finely ground and nitrogen (N) and carbon (C) concentrations in the needles were measured using an Elemental Combustion System CHNS-O (Costech Analytical Technologies Inc., Valentia, USA).

A multivariate approach (ANOVA, Obrien and Kaiser 1985) for repeated measures was used to analyse the morphological traits, and for N and C concentrations. All data were analysed using JMP 8.0 (SAS Institute Inc.).

Results

Morphological analysis

A significant effect of sex on the length of yew and juniper needles, and on the area of yew needles were found (Tables 2, 3). Needles of females were longer and had greater area in both species (Figs. 1, 2). These traits did not change significantly in each of analysed sampling dates (Tables 2, 3).

Table 2 ANOVA repeated measures analysis of J. communis needle area, length and specific leaf area (SLA), nitrogen (N) and carbon (C) content against time effect and sex
Full size table
Table 3 ANOVA-repeated measures analysis of T. baccata needle area, length and specific leaf area (SLA), nitrogen (N) and carbon (C) content against time effect and sex
Full size table
Fig. 1
figure 1

Mean area (±SE) for the season (between March and November) of male and female Taxus baccata and Juniperus communis needles. White bars male individuals, grey bars female individuals. Different letters significant differences between male and female individuals (Student’s t test)

Full size image
Fig. 2
figure 2

Mean length (±SE) for the season (between March and November) of male and female Taxus baccata and Juniperus communis needles. White bars male individuals, grey bars female individuals. Different letters significant differences between male and female individuals (Student’s t test)

Full size image

No significant influence was found of sex and time on specific leaf area (SLA) (Tables 2, 3). Much higher average SLA was found in yew (31.03 cm2 g−1 ± 0.43 SE) compared with juniper (9.16 cm2 g−1 ± 0.14 SE).

Phenology

Both species began and ended vegetative growth at similar times (from mid-April to the end of July), but strobili opening of yew was observed before the development of vegetative growth (between 15 March and 6 April; Fig. 3) and in junipers the opening began in a similar period as the growth phase (between 6 April and 30 April; Fig. 4).

Fig. 3
figure 3

Mean (±SE) annual nitrogen concentration of needles by gender in Taxus baccata. Black line male individuals, grey line female individuals

Full size image
Fig. 4
figure 4

Mean (±SE) annual nitrogen concentration of needles by gender in Juniperus communis. Black line male individuals, grey line female individuals

Full size image

Nitrogen concentration

Yews were characterized by higher average levels of N (a mean of 1.92% ± 0.02 SE across all sampling dates) compared to the juniper (1.33% ± 0.01 SE). A significant effect of sex and time on the average of N concentration was found in yews (Table 3). Across all sampling dates the N concentration of yew needles was higher in males, with the largest differences shown in May, June and July (Fig. 3). In the needles of juniper, there was no overall effect of sex on the N concentration, but concentration varied significantly with time and with interaction of time and sex (Table 2). This was attributed to N concentration being higher in females in March and April just prior to flowering but not in the following months, and indeed after July there was a non-significant tendency for N to be higher in males (Fig. 4). In both species the N concentration of leaves decreased in early spring during the flowering period and then generally increased from May onwards as radial and elongation growth slowed and stopped (Figs. 3, 4).

Carbon concentration

The average C concentration in the needle was slightly higher in yew (49.4% ± 0.15 SE) compared with juniper (48.6% ± 0.11 SE). In neither species there was a significant difference between the sexes but in both juniper and yew C concentration varied significantly over time (Tables 2, 3). The lowest C concentration of needles was observed in May, June and July in both species, and in March in the case of juniper (Fig. 5). The C concentration was considerably higher in yew than in juniper in March and then from July to November. In the intervening months of April, May and June the trend was reversed with N being higher in Juniper (Fig. 5).

Fig. 5
figure 5

Mean (±SE) annual carbon concentration of needles by species. Data for male and female have been aggregated in each species, solid line represents Juniperus, dotted line represents Taxus

Full size image

Discussion

N concentration of leaves varied significantly through the year in both species but only in T. baccata there was a significant difference between male and female trees; males exhibited higher N concentration than females throughout the year. In J. communis differences between the sexes were apparent only in early spring and at this time females had higher level of N concentration. These differences can be linked to phenology. There was, however, no effect of sex on the carbon concentration.

Nitrogen concentration

The changes of N concentration observed in our study are consistent with annual fluctuations in other evergreen plants (Matyssek 1986; Silla and Escudero 2003) where the N concentration drops during early spring and then storage starts after the vegetative growth and increases until the end of the season (Dickson 1989). The decrease of N concentration during the growth of new shoots in both species is caused by the translocation of resources from the last year’s needles to emerging vegetative and reproductive tissues (Bryant et al. 1983; Grelet et al. 2001; Millard et al. 2001; Teitel et al. 2016). However, yew and juniper showed different patterns for N concentration fluctuations.

Taxus baccata

In the case of yew, males had a higher concentration of N throughout the entire period of measurement (Fig. 3). Nitrogen levels reduced in both sexes during flowering. Thereafter, N began to increase after flowering in the male, but in females it carried on decreasing, leading to the greatest differences between sexes from April to August. This accords with previous studies on Rubus chamaemorus and Populus cathayana where N concentration was frequently higher in male than in female plants (Agren et al. 1999; Chen et al. 2014), and partly with observations on Honkenya peploides where the N concentration of females decreased from April to August, while in males a decrease was only found from April to June (Vilas and Retuerto 2016). The data in our study, however, demonstrate these changes for the first time in an evergreen dioecious woody species, and that the greatest differences in N concentration are during intensive plant growth.

This difference between male and female trees corresponds with the period of most intensive vegetative growth, when females allocate resources not only for the vegetative growth but also for the formation and growth of seeds and arils. Yew strobili open before the new leaves, but development of arils and seeds takes place simultaneously with vegetative growth. This is the probable reason for the greater decrease of the N level in females as they invest N simultaneously in the growth of new shoots and maturing cones. This is compounded by the seeds and arils of T. baccata completing their development in one growing season which thus requires more resources in a short period of time in comparison with juniper that takes longer (Table 1). This greater N demand in yew may be related to the greater reproductive effort of females found in many species (e.g. Obeso 2002).

Since N level in female yews declines during periods of high demand, there was, therefore, no evidence of a compensatory mechanism connected with the storage of N in yew. This may result from the fact that yew typically grows in rich soils (see “Materials and methods” section) with less pressure of N deficiency than juniper. It should also be taken into consideration that flowering and vegetative growths do not overlap (Fig. 3) which probably helps to reduce the temporal demand for N (Delph 1990; Milla et al. 2006).

Juniperus communis

Changes in the N concentration of juniper followed a different pattern. Nitrogen concentration was higher in female junipers before the start of strobili opening in March and April (Fig. 4) which suggests the existence of a long-term strategy for increasing N storage before the growing season (Dawson and Geber 1999; Montesinos et al. 2012). Juniper had a lower overall N concentration than yew (Figs. 3, 4) and grows on soils poorer in N, thus it appears that female junipers are more deficient in N than male junipers and the females put greater emphasis in accumulating and storing N for the start of the growing season. A similar strategy has been demonstrated in another species of the genus, J. thurifera, which also grows on poor N habitats in Mediterranean area (Montesinos et al. 2012). The rapid decrease of N at the beginning of growth in the female juniper may also be exacerbated by vegetative growth and strobili opening overlapping in this species (Fig. 3) (Delph 1990; Milla et al. 2006). The fact that the female cones mature over 2–3 years may be a compensatory mechanism for the slow accumulation of N from growing in poor sites. In a similar way, lower N level in male needles, compared to female needles from the beginning of sampling until after the start of strobili opening could also be connected with production of pollen, which is very costly in terms of N (Ashman 1994; Vilas and Pannell 2011).

Carbon concentration

Lack of differences in C concentration between the sexes in both species should be considered in the context of morphological and physiological adaptations. Carbon demand is very high during the formation and maturation of reproductive structures (McDowell et al. 2000) and yet no difference was found in the C concentration of the leaves between the sexes. A similar result was observed, however, in Populus cathayana (Chen et al. 2014) suggesting that this is a more general relationship in dioecious plants. Greater reproductive effort in females is probably compensated for by a higher rate of gas exchange and photosynthesis. A second reason for the lack of differences in the level of C in the leaves of different sexes may be due to storage of C in all parts of the plant including roots and shoots, not just the needles. This is especially pertinent since remobilization is closely linked to the intensity of photosynthesis and growth (Millard and Grelet 2010). Thus, C concentration of the needles is not strictly linked to the maturation of the reproductive structures because the transport of C may be largely from the roots and shoots at the beginning of the growing season. Indeed, in the case of J. thurifera, a similar lack of differences was found between sexes in total non-structural carbohydrates, but females stored more non-soluble sugars in the wood than did males (DeSoto et al. 2016). On the other hand, it has to be taken into consideration that developing seeds and associated structures (cone berries and arils) also carry out photosynthesis and are partly independent in terms of the C economy (Dickson 1989; Ashman 1994).

Morphological adaptations

Greater area and length of female T. baccata needles may be a morphological adaptation for supporting greater reproductive effort of females. This confirms the trend shown in many studies in different dioecious plant species (Wallace and Rundel 1979; Meagher 1992; Kohorn 1994) as well as T. baccata (Iszkuło et al. 2009, 2013). There is a close correlation between the assimilation apparatus area and gas exchange such that larger leaves are associated most often with a higher rate of gas exchange in females (Dawson and Bliss 1989; Dawson and Ehleringer 1993). This phenomenon is probably a consequence of a greater reproductive effort of females connected with the production of seed and accompanying structures. Females are selected to have higher rates of leaf gas exchange (Dawson and Ehleringer 1993; Montesinos et al. 2006). Larger size of needles in the female vs. male is more distinct in the yew—which may be related to the differences in habitat. Junipers often grow in full light exposure but yews grow in shade conditions and light is a limiting factor.

Conclusions

So far, to our knowledge, this is the first demonstration of changes in N concentration in dioecious evergreen woody plants in short time sequences. We confirmed the hypothesis that phenology influences the sex-specific pattern of yearly foliar nitrogen concentration. Our other hypothesis was rejected as these two species had different sex-specific pattern of foliar nitrogen concentration. Juniper and yew are ecologically similar in many aspects: they both are gymnosperms, dioecious, anemophilous and zoochorous (Table 1). However, both species have opposite traits that are likely to significantly affect the species-specific N management. Juniper growing in N-poor pioneer habitats makes it necessary to have a long-term strategy for N storage by females. This is compounded by vegetative and reproductive growth overlapping in juniper. Yew grows in richer, late-successional habitats in mainland Europe and does not show such adaptation. The longer maturation time of juniper strobili over 2–3 years may be a compensatory mechanism to reduce the demand for N within a growing season.

Author contribution statement

Study conception and design: KND, MJG, GI; acquisition of data: KND, GI; analysis and interpretation of data: KND, MJG, PT, GI; drafting of manuscript: KND, MJG, GI; critical revision and final version: KND, MJG, PT, GI.