Originally published as:
Kahmen, A., Dawson, T. E., Vieth, A., Sachse, D. (2011): Leaf wax n‐alkane δ,{delta} D values are determined early in the ontogeny of Populus trichocarpa leaves when grown under controlled environmental conditions. ‐ Plant, Cell & Environment, 34, 10, 1639‐1651
DOI: 10.1111/j.1365‐3040.2011.02360.x 1
2 Leaf wax n-alkane δD values are determined early in the ontogeny of
3 Populus trichocarpa leaves when grown under controlled environmental
4 conditions
5
6
7 Ansgar Kahmen1,*, Todd E. Dawson1, Andrea Vieth2 and Dirk Sachse3
8
9
10 1 Center for Stable Isotope Biogeochemistry, Department of Integrative Biology,
11 University of California – Berkeley, USA
12 2 Helmholtz Centre Potsdam GFZ German Research Centre for Geosciences, Potsdam,
13 Germany
14 3 DFG-Leibniz Center for Surface Process and Climate Studies, Institute of Earth and
15 Environmental Sciences, University of Potsdam, Germany
16
17 *Current address and correspondence:
18 Ansgar Kahmen, ETH Zürich – Institute of Agricultural Sciences
19 Universitätsstrasse 2, LFW C55.2
20 CH-8092 Zürich
21 email: [email protected]
22 Tel: +41-44-6328515
23
1 23 Abstract
24 The stable hydrogen isotope ratios (δD) of leaf-wax n-alkanes record valuable
25 information on plant and ecosystem water relations. It remains, however, unknown if leaf
26 wax n-alkane δD values record only environmental variation during the brief period of
27 time of leaf growth or if leaf wax n-alkane δD values are affected by environmental
28 variability throughout the entire life of a leaf. To resolve these uncertainties, we irrigated
29 Populus trichocarpa trees with a pulse of deuterium-enriched water and used compound
30 specific stable hydrogen isotope analyses to test, if the applied tracer can be recovered
31 from leaf wax n-alkanes of leaves that were at different stages of their development
32 during the tracer application. Our experiment revealed that only leaf wax n-alkanes from
33 leaves that had developed during the time of the tracer application were affected, while
34 fully mature leaves were not. We conclude from our study that under controlled
35 environmental conditions leaf wax n-alkanes are synthesized only early in the ontogeny
36 of a leaf. Our experiment has important implications for the interpretation of leaf wax n-
37 alkane δD values in an environmental context as it suggests that these compounds record
38 only a brief period of the environmental variability that a leaf experiences throughout its
39 life.
40
2 40 Introduction
41 The oxygen and hydrogen isotope composition (δ18O and δD, respectively) of organic
42 plant materials can be used as valuable indicators of environmental and physiological
43 processes (Dawson & Siegwolf, 2007). Over the last two decades plant physiological
44 research has provided a detailed understanding of the mechanisms that determine the
45 δ18O and δD values in plant cellulose and it is now becoming well established as an
46 integrative recorder of environmental and physiological processes (Yakir, 1992, Roden et
47 al., 2000, Brooks & Coulombe, 2009, Sternberg, 2009, Kahmen et al., 2011). The
48 development of new analytical instrumentation and methodologies over the last decade
49 offers now the opportunity to use the δ18O and δD values obtained from other plant
50 compounds such as lignin or leaf waxes as additional recorders of environmental signals
51 that are complementary in the sort of information they can provide to plant cellulose
52 (Burgoyne & Hayes, 1998, Hilkert et al., 1999a, Keppler et al., 2007, Greule et al.,
53 2008). In particular compound-specific stable hydrogen isotope analysis of leaf wax n-
54 alkanes has now been shown to be a powerful tool for investigating present and past
55 hydrological processes as well as an indicator of plant and ecosystem water relations (Xie
56 et al., 2000, Sauer et al., 2001, Huang et al., 2004, Sachse et al., 2004, Schefuss et al.,
57 2005, Tierney et al., 2008).
58 Leaf wax n-alkanes are long-chained alkyl lipids with 25 to 33 carbon atoms that are
59 vital components of higher plant cuticles (Jetter et al., 2006). Several chemical and
60 biological characteristics make leaf wax n-alkanes ideal biomarkers for the investigation
61 of modern or past environments. This is because n-alkanes are relatively easy to extract
62 from plant leaves or from sediment samples where these waxes have accumulated in.
3 63 Also, the analytical tools for determining δD values of individual n-alkanes are now well
64 developed (Burgoyne & Hayes, 1998, Hilkert et al., 1999a). Additionally, n-alkanes can
65 persist in the sedimentary record over geological time scales, which is an important
66 prerequisite for paleoclimatic reconstructions (Radke et al., 2005). Finally, n-alkanes are
67 composed of only carbon and hydrogen atoms and the hydrogen atoms are covalently
68 bound to the carbon atoms in the molecule. Therefore the original hydrogen isotope
69 composition can be preserved in the molecules over geologic time scales (Schimmelmann
70 et al., 2006).
71 Much of the variability in leaf wax n-alkane δD values originates from precipitation,
72 which serves as the plants’ principal source of hydrogen when the plant takes this water
73 up from the soil (Dawson et al., 2002). Several studies have now shown that δD values of
74 leaf wax n-alkanes from terrestrial plants, sediments and soils record the isotope
75 composition of precipitation along environmental gradients (Sauer et al., 2001, Huang et
76 al., 2004, Sachse et al., 2004, Sachse et al., 2006, Smith & Freeman, 2006, Hou et al.,
77 2008, Feakins & Sessions, 2010, McInerney et al., 2011). Since the isotope composition
78 of precipitation is influenced by a number of hydrological processes (Craig & Gordon,
79 1965, Gat, 1996), leaf wax n-alkane δD values in sediments have been suggested to
80 indicate, for example, the intensity or origin of precipitation (Schefuss et al., 2005,
81 Tierney et al., 2008).
82 In addition to the δD values of precipitation leaf wax n-alkanes δD values can also be
83 influenced by soil water and/or leaf water, which is typically enriched in deuterium when
84 compared to precipitation (Sachse et al., 2006, Smith & Freeman, 2006, Sachse et al.,
85 2009, Feakins & Sessions, 2010, McInerney et al., 2011). Further, substantial seasonal or
4 86 cross-species variability has been observed for δD values of leaf wax n-alkanes from
87 temperate, tropical or boreal ecosystems (Liu et al., 2006, Sachse et al., 2006, Smith &
88 Freeman, 2006, Hou et al., 2007, Liu & Yang, 2008, Pederitchouk et al., 2008, Sachse et
89 al., 2009, Feakins & Sessions, 2010). Interestingly, this high interspecific and seasonal
90 variability cannot be fully explained by the influence of precipitation δD values or by leaf
91 water evaporative enrichment in deuterium. This suggests that additional and perhaps
92 fundamentally important plant physiological, biochemical and/or plant ecological
93 processes influence the δD values of leaf wax n-alkanes. These processes are not yet
94 understood and can thus complicate the interpretation of δD values of leaf wax n-alkanes
95 (Zhou et al., 2011).
96 One particularly important plant ecophysiological characteristic that is critical for the
97 robust interpretation of leaf wax n-alkane δD values is the temporal integration with
98 which environmental or physiological signals are recorded in leaf wax n-alkane δD
99 values. The temporal integration of leaf wax n-alkane δD values depends of course on the
100 duration of time over which the leaf wax n-alkanes are synthesized for a particular leaf.
101 The cuticle of plant leaves is typically synthesized early in the ontogeny of a leaf
102 (Kolattukudy, 1970, Jenks et al., 1996, Riederer & Markstaedter, 1996, Hauke &
103 Schreiber, 1998). If n-alkanes are made at the same time then we would expect the δD
104 values of leaf wax n-alkanes to be “locked-in” early in the development of a leaf and to
105 record therefore only a brief period of the environmental or physiological variability that
106 a leaf experiences. The abundance and chemical composition of leaf waxes has, however,
107 been shown to undergo substantial changes in mature and fully expanded leaves (Jetter et
108 al., 2006, Shepherd & Griffiths, 2006). These changes can occur either during the natural
5 109 course of leaf ontogeny (Hauke & Schreiber, 1998, van Maarseveen et al., 2009) or as a
110 response of the leaf to environmental stressors (Baker, 1974, Bengtson et al., 1978, Jetter
111 & Schaffer, 2001, Cameron et al., 2006). Given these observed post-maturation changes
112 in leaf wax abundance and composition it remains unclear if environmental and
113 physiological information in n-alkane δD values is solely recorded and “locked-in” early
114 in the life of a leaf or if the continuous de-novo synthesis of leaf waxes integrates
115 environmental or physiological information in leaf wax n-alkane δD values over the
116 entire lifespan of a leaf.
117 Very few observational studies have investigated the variability of leaf wax n-alkane
118 δD values during the live of a leaf and published studies report contrasting results. Sachse
119 and co-workers (2010) for example have observed that leaf wax n-alkane δD values of
120 barley leaves (Horduem vulgare, Poaceae) are established early in the life of a leaf and
121 show little seasonal variation thereafter. In contrast, Pedentchouk et al. (2008) and Sachse
122 et al. (2009) have shown large seasonal variations in leaf wax n-alkane δD signals in the
123 foliage of deciduous tree species that can reach up to 40‰. In summary, no general
124 pattern for the temporal integration of environmental signals in leaf wax δD values has
125 yet emerged.
126 Here, we present the results of a greenhouse-based experiment where we specifically
127 tested over what timeframe the leaf wax n-alkane δD values are being established. We
128 used the deciduous tree Populus trichocarpa (Salicaceae) as a model species to test if
129 under controlled environmental conditions leaf wax n-alkanes δD values are established
130 and "locked-in" only early in the ontogeny of a leaf or if leaf wax n-alkane δD values can
131 be continuously affected by environmental or physiological drivers throughout the entire
6 132 lifespan of a leaf. For our study, we designed a pulse-chase experiment where P.
133 trichocarpa plants were irrigated with a pulse of deuterium-enriched water. This
134 treatment provided a distinct isotope-based marker that we could then follow into the n-
135 alkanes of leaves using compound-specific stable hydrogen isotope analyses. The
136 purpose of this experiment was to test if the deuterium enrichment could be detected in
137 leaf wax n-alkanes of leaves that were at different stages in their ontogeny: a) young
138 leaves that emerged and developed during the time when the tracer was applied and b)
139 old leaves that were already fully developed and had matured before the application of
140 the tracer. This analysis of different leaf types allowed us to determine if leaf wax n-
141 alkanes are synthesized de-novo only in the early developmental stages of a leaf or if leaf
142 wax n-alkanes are synthesized de-novo continuously throughout the life of a leaf. We
143 purposely performed our experiment under controlled environmental conditions to test
144 the de-novo synthesis of leaf wax n-alkanes in the absence of environmental stressors. As
145 such, this experiment will provide a basic understanding of the timeframe during which
146 environmental or physiological signals are recorded in the δD values of leaf wax n-
147 alkanes.
148
149 Materials and Methods
150 For the experiment we grew 60 individuals of the common deciduous tree Populus
151 trichocapra (Salicaceae) from 20 cm cuttings in pots in a greenhouse under controlled
152 environmental conditions. Cuttings were initially without leaves but developed the first
153 leaves shortly after planting. Diurnal temperature and humidity minima and maxima in
154 the greenhouse were held constant over the entire experiment. Nighttime temperatures
7 155 reached minima of 15°C and daytime maxima temperatures of 30°C. Relative humidity
156 varied between minima of 50% during the day and maxima of 80% during the night.
157 After planting, plants were watered twice a week with Berkeley tap water that had a δD
158 value of -90‰. The soil in pots was covered with 2-3 cm coarse gravel to prevent
159 evaporative enrichment of soil water in deuterium. Growth rates of the plants were high,
160 and the saplings grew about one new leaf every five days. By the end of the experiment,
161 the plants were two months old, had grown on average 15 leaves, and were about 2.5 m
162 high.
163 After plants had developed four mature leaves and were about 40 cm high, we
164 divided the plants into two groups, a treatment group and a control group. The treatment
165 plants were watered with water that was enriched in deuterium (δD = +99‰) for 7 days.
166 Each treatment plant received 250 ml of deuterium-enriched water every second day at
167 9:00 in the morning. The deuterium tracer applications started on May 29th 2007 and
168 ended on June 4th 2007. The control plants were watered with 250 ml Berkeley tap water
169 (δD = -90‰) also every second day at 9:00 in the morning. 250 ml irrigation water was
170 sufficient to reach full field capacity of the soils in the pots of treatment and control
171 plants. After 7 days the labeling treatment was stopped and both treatment and control
172 plants were watered with Berkeley tap water every other day until the end of the
173 experiment. During the entire experiment, treatment and control plants were randomly
174 arranged in the greenhouse to ensure equal environmental conditions for treatment and
175 control plants.
176 On day 1, 6, 8, 13, 21, 28, 38 and 51 after the first tracer application we collected
177 leaves from four replicate treatment and four replicate control plants to determine their
8 178 δD values of their leaf water (n=4). For the same plants, we also determined the
179 corresponding leaf wax n-alkane δD values but we used only three treatment and three
180 control plants for these analyses (n=3). At sampling day 8, we only collected leaves from
181 the treatment plants for leaf water isotope and leaf wax n-alkane analysis. All leaves were
182 collected at midday between 13:00 and 14:00 hours.
183 The specific goal of our experiment was to determine if leaves that were at different
184 stages in their development were differently affected by the tracer application. At each
185 sampling date, we therefore sampled from each plant three different leaf types that were
186 at different developmental stages at the time of tracer application (Fig. 1):
187 1) We collected leaves that had emerged from buds at the beginning of the first
188 tracer application and had developed to fully expanded leaves during the time of
189 tracer application. We call these leaves “Developed During Tracer-addition
190 Leaves" (DDT-Leaves) hereafter. Since these leaves had just started to develop at
191 the time of the first tracer application, their age corresponds to the time since the
192 initial tracer application. For example, 18 days after the start of the tracer
193 application the leaves matured during tracer application were 18 days old.
194 2) We collected leaves that were at least 21 days old and were thus fully developed
195 and matured at the time when the tracer application started. We refer to these
196 leaves as “Developed Before Tracer-addition Leaves" (DBT-leaves) hereafter.
197 3) In addition we always sampled the youngest fully matured leaf of a plant at each
198 sampling date. These leaves were sampled to quantify the turnover of the
199 deuterium tracer in plants and to test if leaves that developed and matured after
200 the tracer application had stopped still showed an impact of the tracer in the δD
9 201 values of their leaf wax n-alkanes as a result of a "tracer memory effect" within
202 the biosynthetic hydrogen pool. We refer to these leaves as “Youngest Mature
203 Leaves" (YM-leaves) in the following.
204 All leaf samples had their mid-vein removed after sampling and were stored in 5 ml PVC
205 vials until processing for leaf water and leaf wax n-alkane extractions.
206
207 Leaf water extractions and isotope analyses
208 Bulk leaf lamina water was extracted from the leaves using cryogenic vacuum distillation
209 at the Center for Stable Isotope Biogeochemistry (CSIB), UC Berkeley (West et al.,
210 2006, Kahmen et al., 2009).
211 Leaf water was analyzed for δD using a Thermo Finnigan (Bremen, Germany)
212 H/Device interfaced with a Delta Plus XL isotope ratio mass spectrometer (IRMS) run in
213 the dual inlet configuration. Water samples were reduced to H2 by injection onto
214 chromium at 900°C then automatically measured after the gas was admitted into the
215 IRMS. Calibration was performed with two different isotope ratio standards to drift
216 correct and normalize the analysis with long-term external precision recorded with a third
217 standard. Long-term external precision is +/- 0.7‰.
218
219 Leaf wax n-alkane extractions and isotope analyses
220 For the extraction of n-alkanes we used the dried leaf samples from the leaf water
221 extractions of three replicate treatment plants and three replicate control plants. The
222 leaves were ground to a fine powder. 100 to 400 mg of the powder was extracted for
223 lipids using an accelerated solvent extractor (ASE200, Dionex Corp., Sunnyvale, U.S.A.)
10 224 with dichloromethane/methanol mixture (9:1) at 100°C and 103 bar (=1500 psi) for 5 min
225 in 3 cycles. The total lipid extracts were separated into three fractions with a medium
226 pressure liquid chromatography (MPLC) system (Radke et al., 1980). Extracts were dried
227 under nitrogen gas, thereafter dissolved in 500 µL n-hexane and injected into the MPLC
228 system. Aliphatic and aromatic compounds were separated chromatographically using n-
229 hexane as solvent (Radke et al. 1980). More polar compounds with functional groups
230 containing nitrogen, sulphur or oxygen (NSO-compounds) remained on the pre-columns
231 of the MPLC system and were rinsed from the columns later using DCM/MeOH (95:5) as
232 solvent. Only the aliphatic fraction, containing n-alkanes, was further investigated.
233 Constituents of the aliphatic fraction were identified and quantified using a GC-FID
234 (Agilent GC6890N, Agilent, Santa Clara, CA, USA) equipped with a DB5ms column (30
235 m, ID:0.32 mm, film thickness: 0.5 µm, Agilent, Palo Alto, U.S.A.). 5α-androstane was
236 used as an internal standard for lipid quantification.
237 For the δD analyses of individual n-alkanes 1 µl of the n-hexane dissolved aliphatic
238 hydrocarbon fraction was injected into a HP6890N GC (Agilent Technologies, Palo Alto,
239 U.S.A.), equipped with a HP Ultra 1 column (50m, ID:0.2mm, film thickness: 0.33µm,
240 Agilent). During injection the PTV injector was heated with 700°C/min to 300°C and
241 held at this temperature for the remaining run. The injector was operated in splitless
242 mode. The oven was maintained for 1 min at 80°C then heated at 10°C/min to 150°C,
243 then at 3°C/min to 300°C and held for 25 min at the final temperature. The column flow
244 was held constant at 1.0 ml/min throughout the run. The eluting compounds were
245 transferred via a GC-C/TC III combustion interface (ThermoFisher Scientific, Bremen,
246 Germany) to a high-temperature conversion furnace operated at 1440°C (Hilkert et al.,
11 247 1999b) and quantitatively converted to H2, which was introduced into an isotope ratio
248 mass spectrometer (IRMS) (Delta V plus, ThermoFisher Scientific, Bremen, Germany)
249 for compound-specific analysis of δD values. Three replicate measurements were
250 performed on each sample. After the measurement of three samples (6 GC runs), a
251 mixture of n-alkanes (n-C16 to n-C30) with known δD values (‘Mix A’), supplied by A.
+ 252 Schimmelmann (University of Indiana), was injected 3 times. The H3 factor was
253 determined once a day and stayed constant within the analytical error of the instrument
254 during the measurement period, indicating stable ion source conditions.
255 70% of the δD analyses of individual n-alkanes were performed at the CSIB at UC
256 Berkeley, the remaining 30% of the samples were analyzed at the Geoforschungszentrum
257 in Potsdam (GFZ). The methods that we employed for the analyses of n-alkane δD values
258 were comparable in both laboratories. Further, we detected no systematic differences
259 between n-alkane δD values analyzed at the CSIB in Berkeley or the GFZ in Potsdam.
260 The results from both labs were normalized to the VSMOW scale using the same
261 methodology, i.e. using the Mix A standard supplied by A. Schimmelmann (University of
262 Indiana). The relationship between measured δD values (δD vs. lab gas) and known δD
263 values of Mix A standard was typically linear and has an explanatory power between r2=
264 0.90 and 0.99. The slope of the relationship was constant during the 24h measurement
265 sequences. The overall precision of the measurements, evaluated by the average standard
266 deviation of repeated measurements of samples and standards was 2.8‰.
267
268 Statistics
12 269 Statistical differences between leaf wax n-alkane δD values of treatment and control
270 plants were tested using a one-way ANOVA with a LSD post-hoc test. We used the
271 software Aabel (Gigawiz, Ltd. Co.) for these calculations.
272
273 Results
274 Leaf water δD values
275 Midday leaf water δD values of the control plants did not differ among the three different
276 leaf types and no temporal trend was detected for midday leaf water δD values in any of
277 the three leaf types of the control plants during the experiment (Fig. 2). Only on the first
278 sampling date the leaf water δD values of the leaves developed during the tracer addition
279 (DDT-leaves) were approximately 10‰ less enriched in deuterium than the leaf water δD
280 values of the leaves developed before the tracer addition (DBT-leaves) or the youngest
281 mature leaves (YM-leaves). This is because DDT-leaves had just started to expand at the
282 first sampling date and had probably not yet reached full metabolic activity leading to
283 less evaporative enrichment of the leaf water as compared to older and metabolically
284 fully active leaves. Without this first value of the DDT- leaves, temporally averaged δD
285 values in the control plants were -7.7 ±3.9‰ for the DDT-leaves, -8.0 ±6.0‰ for the
286 DBT-leaves and -6.1 ±4.8‰ for the YM-leaves (Fig. 2).
287 Treatment with deuterium-enriched water resulted in a substantial enrichment of leaf
288 water δD values at midday in all leaf types (Fig. 2). The tracer induced leaf water
289 enrichment in deuterium at midday reached its highest level at the end of the tracer
290 application period 6 days after the beginning of the experiment with δD values of
291 +52.7‰ above control values for the DDT-leaves, +48.6‰ above control values for the
13 292 DBT-leaves and +51.1‰ above control values for the YM-leaves. After the tracer
293 application was stopped the deuterium enrichment in the leaf water stayed high for at
294 least another two days in all three leaf types but declined steadily thereafter and returned
295 to control levels 21 days after the tracer application had first started (Fig. 2).
296
297 Concentration of leaf wax n-alkanes
298 Leaf wax n-alkane concentrations were identical for treatment and control leaves in any
299 of the three leaf types (Fig. 3). Treatment and control leaves combined showed mean
300 seasonal values of total n-alkane concentrations of 2.7 ±0.6 mg g-1 dry leaf material for
301 the DDT-leaves, 2.8 ±0.2 mg g-1 dry leaf material for the DBT-leaves and 2.9 ±0.5 mg g-1
302 dry leaf material for the YM-leaves (Fig. 3). The n-alkane concentrations were low for
303 DDT-leaves at the beginning of the experiment when DDT-leaves had just started to
304 expand. Leaf wax n-alkane concentrations of the DDT-leaves increased to levels that
305 were comparable to DBT-leaves or YM-leaves 6 days after the experiment had started
306 (Fig. 3). Except for these DDT-leaves at the beginning of the experiment, total n-alkane
307 concentrations were remarkably constant throughout the experiment in DDT- and DBT-
308 leaves. Only the YM-leaves showed a slight seasonal trend with increasing n-alkane
309 concentrations towards the end of the experiment.
310 The n-alkanes in the leaf extracts were mainly composed of nC25, nC27, nC29 and
311 nC31, with nC29 as the dominant compound, constituting more than 50% of all n-alkanes
312 (Fig. 4). Overall, the composition of n-alkanes was consistent across all three leaf types
313 and we did not observe any differences in n-alkane composition between treatment and
314 control plants. Also, no temporal changes in the composition of n-alkanes were observed
14 315 throughout the experiment in any of the three leaf types. The only exception was for the
316 n-alkane composition of the DDT-leaves, which deviated on the first sampling date
317 slightly from the overall pattern. Here, nC29 contributed less than average to the total n-
318 alkane mix while nC31 contributed more than average to the total n-alkane mix (Fig. 4).
319
320 Hydrogen isotope composition of n-alkanes
321 The hydrogen isotope composition was analyzed for nC25, nC27 and nC29 n-alkanes. The
322 concentration of the nC31 n-alkane in our extracts was too low to yield reproducible δD
323 values from the IRMS analyses. Also, due to their small size, the amount of all n-alkanes
324 that was extracted from the DDT-leaves on the first sampling date was too low to allow
325 reproducible δD analyses on the IRMS.
326 δD values for nC25, nC27 and nC29 n-alkanes from the control plants showed no
327 temporal trends during the experiment in either of the three leaf types (Fig. 5). The mean
328 δD values of the individual compounds over the course of the experiment did not differ
329 from each other, nor did the mean seasonal δD values of the individual compounds differ
330 among the three leaf types (Table 1). We calculated εbio, the biosynthetic fractionation
331 between midday leaf water δD values and the δD values of the individual n-alkane
332 compounds, for the three leaf types of the control plants (Table 1). We obtained mean
333 seasonal values for εbio for nC25, nC27 and nC29 that were comparable for the different leaf
334 types. When averaged across leaf types mean εbio was -160.1‰ for nC25, -167.3‰ for
335 nC27 and -165.1‰ for nC29 (Table 1), which is in the range of previously estimated
336 values for εbio for aceitogenic lipids from photosynthetic organisms (Sessions et al., 1999,
337 Sachse et al., 2004, Zhang & Sachs, 2007).
15 338 In the treatment plants n-alkane δD values of the three leaf types responded
339 differently to the deuterium tracer addition (Fig. 5). In DDT-leaves δD values of nC25,
340 nC27 and nC29 n-alkanes of the treatment plants were enriched in deuterium compared to
341 the control plants throughout the entire duration of the experiment (Fig. 5). The
342 deuterium enrichment of leaf wax n-alkanes in DDT-leaves was highest 6 days after the
343 first tracer addition for nC25, nC27 and nC29 (Fig. 6). Tracer induced deuterium
344 enrichment in leaf wax n-alkanes of the DDT-leaves declined, however, from day 6 to
345 day 13 in nC25 and nC29 and from day 6 to day 21 in nC27 (Fig. 6). Following day 13 for
346 nC25 and nC29 and following day 21 for nC27 no significant temporal trend could be
347 observed for tracer induced deuterium enrichment in leaf wax n-alkanes of the DDT-
348 leaves as tested with linear regression analyses (Fig. 6). In contrast to the DDT-leaves,
349 δD values of nC25, nC27 and nC29 n-alkanes from the DBT-leaves were not significantly
350 different between treatment and control plants at any point in time of the experiment (Fig.
351 5). Finally, the enrichment of nC25, nC27 and nC29 n-alkane δD values in the YM-leaves
352 showed a strong temporal trend with no significant enrichment of leaves from the
353 treatment plants at the beginning of the experiment. The three leaf samples collected on
354 days 6, 13 and 21 after the beginning of the tracer application showed a significant
355 enrichment in deuterium as compared to control samples, while no significant enrichment
356 was observed 21 days after the beginning of the experiment (Fig. 5). The deuterium
357 enrichment in the YM-leaves on days 6, 13 and 21 was similar to the values determined
358 for the DDT-leaves collected at the same time.
359
360 Discussion
16 361 Concentration of leaf wax n-alkanes
362 The concentration of total leaf wax n-alkanes was constant over the entire duration of the
363 experiment in the DBT-leaves and also in the DDT-leaves after these leaves were 6 days
364 old and had fully expanded (Fig. 3). Also, the contribution of the different compounds
365 nC25, nC27, nC29 and nC31 to the total n-alkane mix was remarkably consistent in the
366 different leaf types and throughout the entire experiment (Fig. 4). Stable concentrations
367 of leaf wax n-alkanes and consistent contributions of the individual compounds to the
368 total n-alkane mix for the entire duration of our study suggests that the cuticle of P.
369 trichocarpa leaves and the n-alkanes embedded therein are synthesized early in the
370 ontogeny of these leaves and that no additional leaf wax n-alkanes are synthesized
371 thereafter. This said and shown, we urge caution in concluding that this pattern will
372 always be true because abrasion of n-alkanes from the leaves could mask the de-novo
373 synthesis of leaf wax n-alkanes throughout our experiment (Baker & Hunt, 1986,
374 Cameron et al., 2006, Shepherd & Griffiths, 2006). Such unaccounted losses could result
375 in a zero net accumulation and consistent overall concentration of leaf wax n-alkanes
376 over time despite continuous de-novo synthesis. Only the isotope data that we discuss
377 below can distinguish if the consistent concentrations in leaf wax n-alkanes that we
378 observed throughout the experiment are in fact the result of no further leaf wax n-alkane
379 synthesis after leaf maturation or if leaf wax n-alkanes are continuously synthesized, but
380 that this synthesis is offset by continuous losses of leaf wax n-alkanes.
381
382 Tracer induced deuterium enrichment of leaf water
17 383 The application of a deuterium enriched water tracer to the treatment plants resulted in
384 midday leaf water δD values in all leaf types that were up to ~50‰ enriched in deuterium
385 compared to the δD values of midday leaf water of the control plants (Fig. 2). A
386 maximum deuterium enrichment of midday leaf water of 50‰ seems small considering
387 the ∼172‰ difference between the δD values of Berkeley tap water (-90‰) that was
388 received by control plants and the δD values of deuterium enriched water (+99‰) that we
389 applied as tracer to the treatment plants (Fig. 2). Two possible mechanisms can explain
390 the relatively low levels of tracer-induced leaf water deuterium enrichment: i) Leaf water
391 δD values are influenced by the atmospheric vapor that surrounds the leaf. The influence
392 of vapor on leaf water δD values becomes particularly strong at a relative humidity above
393 50% which the P. trichocarpa plants had experienced throughout this experiment
394 (Farquhar et al., 2007, Sachse et al., 2009). Full or partial isotopic equilibration between
395 leaf water and the deuterium depleted atmospheric vapor is therefore a likely reason for
396 the reduction of the original tracer signal in the treatment plants to a ~50‰ deuterium
397 enrichment in the midday leaf water. ii) There is a chance that the deuterium enriched
398 irrigation water that we used to water the treatment plants did not replace all of the non-
399 enriched Berkeley tap water in the soil that these plant grew in (Brooks et al., 2010).
400 Residual non-enriched Berkeley tap water could have therefore also dampened the
401 anticipated tracer-induced leaf water deuterium enrichment. Despite the dampening effect
402 that depleted atmospheric vapor and/or residual soil water had on the tracer signal in the
403 leaves of the treatment plants, the tracer application yet generated a distinct deuterium
404 pulse in the leaf water of the treatment plants that we could follow into the n-alkanes of
405 leaves using compound-specific stable hydrogen isotope analyses (Fig. 2).
18 406
407 Tracer induced deuterium enrichment of leaf wax n-alkanes
408 We found that leaf wax n-alkanes in the DDT-leaves of treatment plants were enriched in
409 deuterium compared to control plants over the entire duration of the experiment but that
410 DBT-leaves were not affected by the tracer applications (Fig. 5). This suggests that the
411 deuterium tracer that we applied to the plants was effectively incorporated into the leaf
412 wax n-alkanes of leaves that emerged and developed during the time of tracer
413 applications (DDT-leaves) but that the tracer was not incorporated into the leaf wax n-
414 alkanes of leaves that had developed and matured before the tracer applications had
415 started (DBT-leaves).
416 The maximum deuterium enrichment of leaf wax n-alkane δD values of the DDT-
417 leaves was 40.5‰, 43.6‰, and 47.0‰ for nC25, nC27, and nC29, respectively. The
418 enrichment of leaf wax n-alkanes was therefore lower than the maximum tracer induced
419 leaf water enrichment, which was 55.0‰ (Fig. 2). The lower deuterium enrichment in the
420 n-alkanes may have been the result of an isotopic memory effect in the pool of
421 compounds that are utilized in the biosynthesis of leaf wax n-alkanes. Using compounds
422 that had been assimilated prior to the tracer addition in the biosynthesis of leaf wax n-
423 alkanes, could lead to δD values of leaf wax n-alkanes that are less enriched in deuterium
424 than expected from the tracer-induced deuterium enrichment of leaf water in the DDT-
425 leaves of the treatment plants. Such an isotopic memory effect in the biosynthetic pool
426 should, however, also be visible in the leaf wax n-alkanes that were synthesized after the
427 tracer had disappeared from the leaf water. We tested therefore, if leaves that developed
428 immediately after the tracer-induced deuterium enrichment of leaf water had returned to
19 429 control values showed such a memory effect in their leaf wax n-alkane δD values. We
430 tested this by sampling the youngest fully matured leaf (YM-leaves) at each sampling
431 time (Fig. 1). We found that YM-leaves that had developed immediately after the tracer-
432 induced deuterium enrichment of leaf water had already returned to control values and
433 showed no signs of deuterium enrichment in their leaf wax n-alkane δD values (Fig. 6).
434 We conclude from this, that the pool of compounds that is utilized in the biosynthesis of
435 leaf wax n-alkanes turns over quickly and that an isotopic memory effect from the
436 biosynthetic pool should therefore not affect the δD values of leaf wax n-alkanes.
437 An alternative explanation why the tracer-induced deuterium enrichment of leaf wax
438 n-alkanes was less than the observed tracer-induced deuterium enrichment of leaf water
439 at midday is the diurnal variability in the tracer-induced leaf water deuterium enrichment.
440 The effects of the tracer application on the deuterium enrichment of leaf water are most
441 likely highest during midday when we measured leaf water (Fig. 2). Since the
442 biosynthesis of leaf wax n-alkanes is likely to occur throughout the entire day, δD values
443 of leaf wax n-alkanes in the treatment plants should integrate the entire diurnal variability
444 of the tracer-induced deuterium enrichment of leaf water. This could explain why leaf
445 wax n-alkane δD values of the treatment plants are less enriched in deuterium than the
446 midday leaf water δD values that we measured in the course of this experiment.
447
448 Duration of de-novo leaf wax n-alkane synthesis
449 The deuterium enrichment of all three n-alkanes in the DDT-leaves was highest 6 days
450 after the first tracer application and declined from day 6 to day 13 for nC25 and nC29 and
451 from day 6 to day 21 for nC27 (Fig. 6). The decline in deuterium enrichment in the n-
20 452 alkanes of DDT-leaves suggests that the de-novo synthesis of leaf wax n-alkanes
453 continued in the DDT-leaves after the tracer application had stopped and that n-alkanes
454 that were synthesized after the tracer application had stopped "diluted" the deuterium
455 enriched leaf wax n-alkanes that had been synthesized at the time of maximum tracer
456 induced leaf water enrichment in deuterium (Figs. 5 and 6). Importantly, we found no
457 decline in deuterium enrichment of leaf wax n-alkane δD values in the DDT-leaves after
458 day 13 (nC25 and nC29) or day 21 (nC27) of the experiment (Fig. 6), when DDT-leaf n-
459 alkane δD values were still enriched compared to the control. Our data suggest therefore
460 that the de-novo synthesis of leaf wax n-alkanes continued until day 13 or 21 of the
461 experiment (when leaves were between 13 - 21 days old) but stopped thereafter. This
462 conclusion is supported by the fact that we could not find any significant effects of the
463 tracer addition on δD values of leaf wax n-alkanes of the DBT-leaves, which were 21
464 days old at the time of the initial tracer application (Fig. 5).
465
466 Integration time of leaf wax n-alkane δD values
467 The consistent concentration and composition of leaf wax n-alkanes throughout the entire
468 duration of our experiment in combination with the results that we obtained from the
469 isotope pulse-chase experiment provide evidence that leaf wax n-alkanes are synthesized
470 only early in the life of a leaf and that the de-novo synthesis of leaf wax n-alkanes is
471 terminated once a leaf has fully matured. Previous work has shown that the cuticle is
472 established early in the life of a leaf (Kolattukudy, 1970, Hauke & Schreiber, 1998, Jetter
473 et al., 2006). Our work corroborates these findings of previous studies and shows in
474 addition that under controlled environmental conditions leaf wax n-alkanes are not de-
21 475 novo synthesized once a leaf has fully matured. As such, our study suggests that the δD
476 values of leaf waxes n-alkanes are established early in the life of a leaf and should show
477 little environmentally induced variation once a leaf has fully developed. This finding
478 supports the recent work by Sachse et al. (2010) on grass leaves, who have shown that
479 the δD values of leaf wax n-alkanes in barley leaves differ among different leaf
480 generations but that once a δD value had been established and "locked in" within a leaf
481 generation, it remains consistent for the rest of the season.
482 The plants that we used for the experiment presented here were purposely grown
483 under controlled and stable environmental conditions. This allowed testing if de-novo
484 synthesis of leaf wax n-alkanes occurs in the natural ontogeny of leaves in the absence of
485 environmental stressors. Environmental stress or physical damage of the cuticle can,
486 however, stimulate the synthesis of leaf waxes even in mature leaves (Baker, 1974, Baker
487 & Hunt, 1986, Maffei et al., 1993, Jenks et al., 2001, Cameron et al., 2006, Shepherd &
488 Griffiths, 2006). Environmental stress or the physical abrasion of the cuticle could
489 therefore explain, why leaf wax n-alkanes δD values from plants that are exposed to
490 natural environmental conditions have shown substantial variation in the hydrogen
491 isotope composition of their leaf wax n-alkanes (Sachse et al., 2006, Pedentchouk et al.,
492 2008, Sachse et al., 2010). While our study clearly shows that leaf wax n-alkanes are not
493 de-novo synthesized in the natural course of leaf ontogeny when P. tricoracpa leaves
494 were more than 21 days old, future studies should test if and how environmental stress or
495 physical damage affect the δD values of leaf wax n-alkanes throughout the life of fully
496 matured leaves under natural environmental conditions.
22 497 Our finding that δD values of leaf wax n-alkanes are established early in the life of a
498 leaf has important implications for the interpretation of δD values in leaf wax n-alkanes
499 found in a range of sample types. The results we present here can, for example, explain
500 some of the variation that has been observed in leaf wax n-alkanes δD values across
501 different plant species even between plants grown at the same site (Liu et al., 2006,
502 Sachse et al., 2006, Smith & Freeman, 2006, Hou et al., 2007, Liu & Yang, 2008,
503 Pedentchouk et al., 2008, Sachse et al., 2009, Feakins & Sessions, 2010). Species differ
504 in their phenologies and establish their leaves at different times of the season. The
505 observed differences in leaf wax n-alkane δD values across species could at least partly
506 be a result of contrasting environmental conditions that the different plant species
507 experienced during their leaf development. Additionally, our results have implications for
508 the interpretation of paleohydrological changes from leaf wax n-alkane δD records from
509 sediments (Schefuss et al., 2005, Tierney et al., 2008). Such records may therefore not
510 document the entire seasonal variability of their environmental or physiological drivers
511 but mostly the conditions during a brief period early in the life of leaves. This needs to be
512 considered when leaf wax n-alkane δD values are used as proxies of modern or past
513 climate regimes, particularly in regions characterized by strong hydrological seasonality.
514
515 Acknowledgments
516 We would like to thank Kevin Simonin for his help with sample collections in the
517 greenhouse and Paul Brooks for assistance with the compound specific isotope analyses
518 in the Center for Stable Isotope Analyses at UC Berkeley. We thank Robert Glasmacher
519 for help with the lipid extraction and purification at UP. Alex Sessions, Graham Farquhar
23 520 and one anonymous referee helped to improve an earlier version of this manuscript.
521 Ansgar Kahmen was supported by a Marie Curie Outgoing International Fellowship of
522 the European Union (BiWaClim) and Dirk Sachse was supported by an Emmy-Noether
523 Research grant by the German Science Foundation (DFG) SA-1889/1-1.
524
24 524 References
525 Baker E.A. (1974) Influence of environment on leaf wax development in Brassica 526 oleracea var. Gemmifera. New Phytologist, 73, 955-&. 527 Baker E.A. & Hunt G.M. (1986) Erosion of waxes from leaf surfaces by simulated rain. 528 New Phytologist, 102, 161-173. 529 Bengtson C., Larsson S. & Liljenberg C. (1978) Effects of water stress on cuticular 530 transpiration rate and amount and composition of epicuticular wax in seedlings of 531 six oat varieties. Physiologia Plantarum, 44, 319-324. 532 Brooks J.R., Barnard H.R., Coulombe R. & McDonnell J.J. (2010) Ecohydrologic 533 separation of water between trees and streams in a Mediterranean climate. Nature 534 Geoscience, 3, 100-104. 535 Brooks J.R. & Coulombe R. (2009) Physiological responses to fertilization recorded in 536 tree rings: isotopic lessons from a long-term fertilization trial. Ecological 537 Applications, 19, 1044-1060. 538 Burgoyne T.W. & Hayes J.M. (1998) Quantitative production of H2 by pyrolysis of gas 539 chromatographic effluents. Analytical Chemistry, 70, 5136-5141. 540 Cameron K.D., Teece M.A. & Smart L.B. (2006) Increased accumulation of cuticular 541 wax and expression of lipid transfer protein in response to periodic drying events 542 in leaves of tree tobacco. Plant Physiology, 140, 176-183. 543 Craig H. & Gordon L.I. (1965) Deuterium and oxygen 18 variations in the ocean and 544 marine atmosphere. Consiglio Nationale Delle Ricerche Laboratorio Di Geologia 545 Nucleare, Pisa, Italy. 546 Dawson T.E., Mambelli S., Plamboeck A.H., Templer P.H. & Tu K.P. (2002) Stable 547 isotopes in plant ecology. Annual Review of Ecology and Systematics, 33, 507- 548 559. 549 Dawson T.E. & Siegwolf R. (2007) Stable isotopes as indicators of ecological change. 550 Elsevier. 551 Farquhar G.D., Cernusak L.A. & Barnes B. (2007) Heavy water fractionation during 552 transpiration. Plant Physiology, 143, 11-18. 553 Feakins S.J. & Sessions A.L. (2010) Controls on the D/H ratios of plant leaf waxes in an 554 arid ecosystem. Geochimica et Cosmochimica Acta, 74, 2128-2141. 555 Gat J.R. (1996) Oxygen and hydrogen isotopes in the hydrologic cycle. Annual Review of 556 Earth and Planetary Sciences, 24, 225-262. 557 Greule M., Mosandl A., Hamilton J.T.G. & Keppler F. (2008) A rapid and precise 558 method for determination of D/H ratios of plant methoxyl groups. Rapid 559 Communications in Mass Spectrometry, 22, 3983-3988. 560 Hauke V. & Schreiber L. (1998) Ontogenetic and seasonal development of wax 561 composition and cuticular transpiration of ivy (Hedera helix L.) sun and shade 562 leaves. Planta, 207, 67-75. 563 Hilkert A.W., Douthitt C.B., Schluter H.J. & Brand W.A. (1999a) Isotope ratio 564 monitoring gas chromatography mass spectrometry of D H by high temperature 565 conversion isotope ratio mass spectrometry. Rapid Communications in Mass 566 Spectrometry, 13, 1226-1230. 567 Hilkert A.W., Douthitt C.B., Schlüter H.J. & Brand W.A. (1999b) Isotope Ratio 568 Monitoring Gas Chromatography/Mass Spectrometry of D/H by High
25 569 Temperature Conversion Isotope Ratio Mass Spectrometry. Rapid 570 Communications in Mass Spectrometry, 13, 1226-1230. 571 Hou J.Z., D'Andrea W.J. & Huang Y.S. (2008) Can sedimentary leaf waxes record D/H 572 ratios of continental precipitation? Field, model, and experimental assessments. 573 Geochimica et Cosmochimica Acta, 72, 3503-3517. 574 Hou J.Z., D'Andrea W.J., MacDonald D. & Huang Y.S. (2007) Hydrogen isotopic 575 variability in leaf waxes among terrestrial and aquatic plants around Blood Pond, 576 Massachusetts (USA). Organic Geochemistry, 38, 977-984. 577 Huang Y.S., Shuman B., Wang Y. & Webb T. (2004) Hydrogen isotope ratios of 578 individual lipids in lake sediments as novel tracers of climatic and environmental 579 change: a surface sediment test. Journal of Paleolimnology, 31, 363-375. 580 Jenks M.A., Andersen L., Teusink R.S. & Williams M.H. (2001) Leaf cuticular waxes of 581 potted rose cultivars as affected by plant development, drought and paclobutrazol 582 treatments. Physiologia Plantarum, 112, 62-70. 583 Jenks M.A., Tuttle H.A. & Feldmann K.A. (1996) Changes in epicuticular waxes on 584 wildtype and Eceriferum mutants in Arabidopsis during development. 585 Phytochemistry, 42, 29-34. 586 Jetter R., Kunst L. & Samuels A. (2006) Composition of plant cuticular waxes. In: 587 Biology of the plant cuticle (eds M. Riederer & C. Mueller), pp. 145-181. 588 Blackwell Publishing. 589 Jetter R. & Schaffer S. (2001) Chemical composition of the Prunus laurocerasus leaf 590 surface. Dynamic changes of the epicuticular wax film during leaf development. 591 Plant Physiology, 126, 1725-1737. 592 Kahmen A., Sachse D., Arndt S.K., Farrington H., Vitousek P.M. & Dawson T.E. (2011) 593 Cellulose δ18O is an index of leaf to air vapor pressure difference (VPD) in 594 tropical plants. Proceedings of the National Academy of Sciences of the United 595 States of America (PNAS), 108, 1981-1986. 596 Kahmen A., Simonin K., Tu K., Goldsmith G.R. & Dawson T.E. (2009) The influence of 597 species and growing conditions on the 18-O enrichment of leaf water and its 598 impact on 'effective path length'. New Phytologist, 184, 619-630. 599 Keppler F., Harper D.B., Kalin R.M., Meier-Augenstein W., Farmer N., Davis S., 600 Schmidt H.L., Brown D.M. & Hamilton J.T.G. (2007) Stable hydrogen isotope 601 ratios of lignin methoxyl groups as a paleoclimate proxy and constraint of the 602 geographical origin of wood. New Phytologist, 176, 600-609. 603 Kolattukudy P. (1970) Cutin biosynthesis in Vicia faba leaves - Effect of age. Plant 604 Physiology, 46, 759-&. 605 Liu W.G. & Yang H. (2008) Multiple controls for the variability of hydrogen isotopic 606 compositions in higher plant n-alkanes from modern ecosystems. Global Change 607 Biology, 14, 2166-2177. 608 Liu W.G., Yang H. & Li L.W. (2006) Hydrogen isotopic compositions of n-alkanes from 609 terrestrial plants correlate with their ecological life forms. Oecologia, 150, 330- 610 338. 611 Lockheart M.J., VanBergen P.F. & Evershed R.P. (1997) Variations in the stable carbon 612 isotope compositions of individual lipids from the leaves of modern angiosperms: 613 Implications for the study of higher land plant-derived sedimentary organic 614 matter. Organic Geochemistry, 26, 137-153.
26 615 Maffei M., Mucciarelli M. & Scannerini S. (1993) Environmental factors affecting the 616 lipid metabolism in Rosmarinus officinalis (L). Biochemical Systematics and 617 Ecology, 21, 765-784. 618 McInerney F.A., Helliker B.R. & Freeman K.H. (2011) Hydrogen isotope ratios of leaf 619 wax n-alkanes in grasses are insensitive to transpiration. Geochimca et 620 Cosmochimica Acta, 75, 541-554. 621 Pedentchouk N., Sumner W., Tipple B. & Pagani M. (2008) delta C-13 and delta D 622 compositions of n-alkanes from modern angiosperms and conifers: An 623 experimental set up in central Washington State, USA. Organic Geochemistry, 624 39, 1066-1071. 625 Radke J., Bechtel A., Gaupp R., Puttmann W., Schwark L., Sachse D. & Gleixner G. 626 (2005) Correlation between hydrogen isotope ratios of lipid biomarkers and 627 sediment maturity. Geochimica et Cosmochimica Acta, 69, 5517-5530. 628 Radke M., Willsch H. & Welte D.H. (1980) Preparative hydrocarbon group type 629 determination by automated medium pressure liquid chromatography. Analytical 630 Chemistry, 52, 406-411. 631 Riederer M. & Markstaedter C. (1996) Cuticular Waxes: a critical assessment of current 632 knowledge. In: Plant cuticles - an integrated functional approach (ed K. G), pp. 633 189 - 200. BIOS Scientific Publishers, Oxford. 634 Roden J.S., Lin G.G. & Ehleringer J.R. (2000) A mechanistic model for interpretation of 635 hydrogen and oxygen isotope ratios in tree-ring cellulose. Geochimica et 636 Cosmochimica Acta, 64, 21-35. 637 Sachse D., Gleixner G., Wilkes H. & Kahmen A. (2010) Leaf wax n-alkane δD values of 638 field-grown barley reflect leaf water δD values at the time of leaf formation. 639 Geochimca et Cosmochimica Acta, 74, 6741-6750. 640 Sachse D., Kahmen A. & Gleixner G. (2009) Significant seasonal variation in the 641 hydrogen isotopic composition of leaf-wax lipids for two deciduous tree 642 ecosystems (Fagus sylvativa and Acer pseudoplatanus). Organic Geochemistry, 643 40, 732-742. 644 Sachse D., Radke J. & Gleixner G. (2004) Hydrogen isotope ratios of recent lacustrine 645 sedimentary n-alkanes record modern climate variability. Geochimica et 646 Cosmochimica Acta, 68, 4877-4889. 647 Sachse D., Radke J. & Gleixner G. (2006) delta D values of individual n-alkanes from 648 terrestrial plants along a climatic gradient - Implications for the sedimentary 649 biomarker record. Organic Geochemistry, 37, 469-483. 650 Sauer P.E., Eglinton T.I., Hayes J.M., Schimmelmann A. & Sessions A.L. (2001) 651 Compound-specific D/H ratios of lipid biomarkers from sediments as a proxy for 652 environmental and climatic conditions. Geochimica et Cosmochimica Acta, 65, 653 213-222. 654 Schefuss E., Schouten S. & Schneider R.R. (2005) Climatic controls on central African 655 hydrology during the past 20,000 years. Nature, 437, 1003-1006. 656 Schimmelmann A., Sessions A.L. & Mastalerz M. (2006) Hydrogen isotopic (D/H) 657 composition of organic matter during diagenesis and thermal maturation. Annual 658 Review of Earth and Planetary Sciences, 34, 501-533. 659 Sessions A.L., Burgoyne T.W., Schimmelmann A. & Hayes J.M. (1999) Fractionation of 660 hydrogen isotopes in lipid biosynthesis. Organic Geochemistry, 30, 1193-1200.
27 661 Shepherd T. & Griffiths D.W. (2006) The effects of stress on plant cuticular waxes. New 662 Phytologist, 171, 469-499. 663 Smith F.A. & Freeman K.H. (2006) Influence of physiology and climate on delta D of 664 leaf wax n-alkanes from C-3 and C-4 grasses. Geochimica et Cosmochimica Acta, 665 70, 1172-1187. 666 Sternberg L. (2009) Oxygen stable isotope ratios of tree-ring cellulose: the next phase of 667 understanding. New Phytologist, 181, 553-562. 668 Sternberg L.D.L. (1988) D/H ratios of environmental water recorded by D/H ratios of 669 plant lipids. Nature, 333, 59-61. 670 Tierney J.E., Russell J.M., Huang Y.S., Damste J.S.S., Hopmans E.C. & Cohen A.S. 671 (2008) Northern hemisphere controls on tropical southeast African climate during 672 the past 60,000 years. Science, 322, 252-255. 673 van Maarseveen C., Han H. & Jetter R. (2009) Development of the cuticular wax during 674 growth of Kalanchoe daigremontiana (Hamet et Perr. de la Bathie) leaves. Plant 675 Cell and Environment, 32, 73-81. 676 West A.G., Patrickson S.J. & Ehleringer J.R. (2006) Water extraction times for plant and 677 soil materials used in stable isotope analysis. Rapid Communications in Mass 678 Spectrometry, 20, 1317-1321. 679 Xie S., Nott C.J., Avsejs L.A., Volders F., Maddy D., Chambers F.M., Gledhill A., Carter 680 J.F. & Evershed R.P. (2000) Palaeoclimate records in compound-specific delta D 681 values of a lipid biomarker in ombrotrophic peat. Organic Geochemistry, 31, 682 1053-1057. 683 Yakir D. (1992) Variations in the natural abundance of O-18 and deuterium in plant 684 carbohydrates. Plant Cell and Environment, 15, 1005-1020. 685 Zhang Z.H. & Sachs J.P. (2007) Hydrogen isotope fractionation in freshwater algae: I. 686 Variations among lipids and species. Organic Geochemistry, 38, 582-608. 687 Zhou Y., Grice K., Chikaraishi Y., Stuart-Williams H., Farquhar G.D. & Ohkouchi N. 688 (2011) Temperature effect on leaf water deuterium enrichment and isotopic 689 fractionation during leaf lipid biosynthesis: Results from controlled growth of C3 690 and C4 land plants. Phytochemistry, 72, 207-213. 691 692
693
28 693 Tables
694 Table 1: Average leaf wax n-alkane δD values and εbio for nC25, nC27 and nC29 of the
695 three different leaf types of the control plants. εbio is the biosynthetic fractionation
696 between midday leaf water δD values and the δD values of the individual n-alkane
697 compounds and was calculated as the difference between δD values nC25, nC27 and
698 nC29 and the corresponding midday leaf water δD values.
699
Leaves developed Leaves developed Youngest during tracer before tracer mature addition addition Leaves (DDT-Leaves) (DBT-Leaves) (YM-Leaves) δD nC25 -147.5 ±4.4 -149.1 ±3.8 -144.5 ±3.4 δD nC27 -142.7 ±3.8 -149.0 ±3.9 -148.8 ±6.8 δD nC29 -144.2 ±5.9 -150.8 ±3.8 -152.3 ±7.7
εbio nC25 -164.1 ±5.9 -165.8 ±4.7 -161.7 ±7.4
εbio nC27 -157.5 ±5.5 -165.8 ±10.3 -167.6 ±5.1
εbio nC29 -159.5 ±6.9 -172.5 ±8.8 -168.2 ±9.4 700
701
702
29 702 Figures
703 Fig. 1: Sampling design illustrating the the three different leaf types that we sampled
704 throughout the experiment. For space reasons we show only days 1, 21 and 51 after the
705 experiment had started. Lower case letters indicate the different leaf generations. Leaves
706 that emerged and developed during the tracer addition (DDT-leaves) were always
707 sampled from leaves of the generation “g”, leaves that had developed before the tracer
708 addition (DBT-leaves) were always sampled from leaf generation “c”. In addition, we
709 always sampeled the youngest fully matured leaf at the top of each plant (YM-leaf).
710
711 Fig. 2: Midday leaf water δD values of the control and the treatment plants in the
712 different leaf types. Shaded area indicated the time when the deuterium-enriched tracer
713 was applied to the treatment plants. DDT-leaves are leaves that have developed during
714 the tracer addition; DBT-leaves are leaves that had developed before the tracer addition
715 and were at least 21 days old at the beginning of the experiment; YM leaves are the
716 youngest fully matured leaf at the top of the plant. Error bars are one standard deviation,
717 n=4.
718
719 Fig. 3: Total n-alkane concentration in leaf waxes in the tree leaf types of the treatment
720 and control plants over the time of the experiment. Concentrations are shown per gram
721 leaf dry weight. DDT-leaves are leaves that have developed during the tracer addition;
722 DBT-leaves are leaves that had developed before the tracer addition and were at least 21
723 days old at the beginning of the experiment; YM leaves are the youngest fully matured
724 leaf at the top of the plant. Error bars are one standard deviation, n=3.
30 725
726 Fig. 4: Relative contribution of nC25, nC27, nC31 and nC29 n-alkanes to the total n-
727 alkane mix in the tree leaf types of the treatment and control plants shown in %. DDT-
728 leaves are leaves that have developed during the tracer addition; DBT-leaves are leaves
729 that had developed before the tracer addition and were at least 21 days old at the
730 beginning of the experiment; YM leaves are the youngest fully matured leaf at the top of
731 the plant. Standard deviation was < 5% and is therefore not shown, n=3.
732
733 Fig. 5: Hydrogen isotope rations (δD values) of nC25, nC27 and nC29 leaf wax n-
734 alkanes in the tree different leaf types of the treatment and control plants. Dark shaded
735 area indicated the time of tracer applications, lightly shaded area indicates the time when
736 tracer applications had stopped but when the tracer was still detectable in the leaf water
737 (cf. Fig. 2). DDT-leaves are leaves that have developed during the tracer addition; DBT-
738 leaves are leaves that had developed before the tracer addition and were at least 21 days
739 old at the beginning of the experiment; YM leaves are the youngest fully matured leaf at
740 the top of the plant. Asterisks indicate significant difference (p<0.05) between treatment
741 and control plants.
742
743 Fig. 6: Tracer-induced deuterium enrichment in nC25, nC27 and nC29 leaf wax n-
744 alkanes in DDT-leaves of treatment plants as compared to control plants. The deuterium
745 enrichment was highest 6 days after the first tracer application and declined from day 6 to
746 day 13 for nC25 and nC29 and from day 6 to day 21 for nC27. If these first sampling dates
747 are excluded from the analysis (symbols in parenthesis), linear regression analyses
31 748 indicate no further significant decline in deuterium enrichment in any of the three n-
749 alkanes in the DDT-leaves of treatment plants.
750
32 750 Fig. 1 (Kahmen et al.):
751
33 751 Fig. 2 (Kahmen et al.):
752
34 752 Fig. 3 (Kahmen et al.):
753
35 753 Fig. 4 (Kahmen et al.):
754
36 754 Fig. 5 (Kahmen et al.):
755
37 755 Fig. 6 (Kahmen et al.):
38