Plant Soil (2015) 386:331340

DOI 10.1007/s11104-014-2182-x

REGULAR ARTICLE

Modelling the relative contribution of seed nitrogen reserves

and external nitrogen uptake during heterotrophic growth
in Medicago truncatula
Sophie Brunel-Muguet & Bruno Mary & Carolyne Drr

Received: 19 March 2014 / Accepted: 23 June 2014 / Published online: 19 September 2014
# Springer International Publishing Switzerland 2014

Abstract
Background and aims Heterotrophic growth relies on
remobilisation of seed reserves and mineral absorption.
We used a compartmental model to investigate the
fluxes of N absorption and remobilisation of N reserves
in a legume seed with high protein content.
Methods Seedling growth was studied during the heterotrophic stage in two genotypes of Medicago
truncatula as a function of N supply. N absorption and
seed remobilisation fluxes were distinguished in a 15 N
labelling experiment.
Results Remobilisation of seed N reserves was high
during germination, but N uptake started as soon as
the radicle protruded. Both sources contributed to high
elongation rates of the radicle and hypocotyl. When
organ lengths stabilised, there was an efflux of N from
the cotyledons and roots indicating that seedling growth
Responsible Editor: Ad C. Borstlap.
S. Brunel-Muguet
INRA UMR 950 Ecophysiologie Vgtale, Agronomie et
nutritions N, C, S,
Caen F-14032, France
e-mail: sbrmuguet@rennes.inra.fr
B. Mary
INRA UR Agro-Impact, Ple du Griffon,
Barenton-Bugny F-02000, France
e-mail: bruno.mary@laon.inra.fr
C. Drr (*)
INRA UMR 1345 Institute of Research on Horticulture and
seeds,
Beaucouz F-49045, France
e-mail: durr@angers.inra.fr

was limited by carbohydrate production. No significant

differences between genotypes were observed except
for early N uptake, which was lower in the genotype
with the highest initial seed N content.
Conclusions N fluxes were similar to those of other
non-legume dicotyledonous species but differed from
monocotyledonous species. These results improve our
understanding of the effects of mineral fertilisation on
crop establishment. The compartmental model is a useful tool to analyse N fluxes patterns within and between
diverse species, in relation to seed characteristics and
soil N availability.
Keywords Nitrogen . Seedling . Absorption .
Remobilization . Compartment model . Medicago
truncatula

Introduction
During the heterotrophic stage, seedling growth is enabled by depletion of seed reserves until the seedling
emerges and photosynthesis starts. The supply of carbon
(C) depends exclusively on seed reserves during heterotrophic growth, but the contribution of other major
minerals such as nitrogen (N) from seed reserves and
from external sources merits further investigation, especially to assess the importance of early mineral
fertilisation for crop stand establishment. Early localised
mineral fertilisation has been shown to improve stand
establishment and favour early growth (Rosati and
Magnifico 2001; Andrews et al. 1991). But few studies

332

have focused on the timing of early N uptake with

respect to the period of seed N remobilisation, or considered possible differences among species. In sugar
beet (Beta vulgaris L.), a starchy dicotyledonous seed,
absorption begins as soon as the radicle protrudes from
the teguments (Drr and Mary 1998), which stimulates
hypocotyl elongation (Durrant and Mash 1989). By
contrast, in wheat (Triticum aestivum L.), which is also
a starchy but monocotyledonous seed, N absorption
begins later, when the leaves break through the coleoptile (Drr and Mary 1998). The different categories of
species and seed morphology may thus influence early
patterns of N fluxes i.e. absorption and remobilisation.
As legume seeds are rich in reserve proteins (possibly
more than 35 % of seed dry weight, Monti and Grillo
1983), they are often expected to remain self-sufficient
longer to fulfil early seedling N requirements during the
heterotrophic stage and even after the beginning of
photosynthesis until efficient symbiotic N2 fixation with
Rhizobium is established. Previous studies have shown
that nodule activity is efficient after about 200-days
from emergence under low soil nitrate availability
(Pisum sativum L., Voisin et al. 2002; Medicago
truncatula G., Moreau et al. 2008). Studies on mineral
absorption by several forage legume species (Cooper
1977) suggested that the absorption of N and other
minerals begins early in seedling life, but the exact
timing of uptake has never been established. To this
end, we studied N absorption and remobilisation fluxes
in Medicago truncatula, one of the model legumes
which provides molecular tools to elucidate the genetic
determinism of legume characteristics (Cook 1999;
Thoquet et al. 2002; Tivoli et al. 2006; Young et al.
2011). In this study, we characterised two genotypes,
Jemalong A17, the reference for genomic studies, and
Paraggio, a cultivar mainly grown in Australia, which
have different seed N contents (Brunel et al. 2009). A
compartmental model was built to estimate the respective contribution of N sources, i.e. absorption and seed
reserves, to the different parts of the seedling. The
objectives of the present study were to (i) determine the timing of N absorption by seedlings in
two genotypes of Medicago truncatula under varying N supplies, (ii) estimate the respective contribution of external N and N seed reserves to the
different seedling parts during heterotrophic
growth, and (iii) analyse legume N uptake and
allocation patterns with regard to seed characteristics by comparison with non-legumes species.

Plant Soil (2015) 386:331340

Material and methods

Seedling growth experiments
Two genotypes were studied: Jemalong A17, the reference line used for the genome sequencing of Medicago
truncatula (Young et al. 2011) and Paraggio, a cultivar
mainly grown in southern Australia. Mother plants of
A17 seeds were grown in a greenhouse in INRA
Montpellier (43.61N, 3.87E, South of France) in
2005 and Paraggio seeds were produced under field
conditions in Australia in 2004 (Seedco Australia CoOperative Ltd). Pots were filled with 500 g of sand
(SIFRACO quality NE34: SiO2>99.7 %, solid density
2.65, mean particle size 200 m). Before sowing, the
sand water gravimetric content was adjusted to
0.200 kg kg1 with either deionised water or nutrient
solution (Saglio and Pradet 1980) and was maintained
constant throughout the experiment by watering regularly with deionised water. Before dilution, the initial
nutrient solution contained 700 mmol L -1 KNO3,
750 mmol L -1 NH4NO3, 650 mmol L -1 Ca (NO3) 2,
275 mmol L -1 KH 2 PO 4 , 50 mmol L -1 NaCl,
157 mmol L -1MgSO4, plus micronutrient supplements.
We used a 15 N labelled solution to differentiate N
absorption from seed N remobilisation. Differences in
natural 15 N abundance are not usually sufficient to
ensure 15 N tracking which requires the external source
to be enriched with 15 N (Duc et al. 1988). This is the
case in legumes, whose natural abundance in the seed is
close to that of atmospheric N2 because of the ability of
mother plants to fix atmospheric N2. For that reason, we
used a 15 N labelled solution (2 atom % excess, APE) to
detect the exact time absorption began and to distinguish
the allocation of N originating from absorption to the
different seedling parts: radicle, hypocotyl and cotyledons. The solution was diluted with deionised water at a
rate of 2, 3, 4 mL L -1 to obtain N concentrations of 7
(S1), 10.5 (S2) and 14 (S3) mmol L -1, the last concentration being usually used to grow plants in pots under
non-limiting mineral nutrition conditions. The experiment with only deionised water and no added nutrient
solution is hereafter referred to as S0. Five seeds were
sown per pot (6.5 cm in diameter, 10 cm high) at a depth
of 1.5 cm in a pre-defined position in the pot so we were
able to refer the seedling weight to its initial dry seed
weight. Seeds were first rubbed to overcome seed physical dormancy (Uzun and Aydin 2004; Zeng et al. 2005)
and then individually weighed. Seed water content was

Plant Soil (2015) 386:331340

measured in a sub-sample of 100 seeds taken from the

same seed lot to estimate seed dry weight from the initial
seed fresh weight. Pots were incubated in a growth
chamber at 20 C in the dark to mimic heterotrophic
growth in the soil.
Phenological measurements and 15 N analyses
For each concentration of the nutrient solution and each
genotype, three pots were sampled 24, 48, 96, 144, 192
and 264 h (h) after sowing until maximum elongation was
reached. On each sampling occasion, the length of the
hypocotyl and radicle was measured. The seedling parts
were dried at 80 C for 48 h and then weighed. Changes in
the dry mass (DM) of the different parts of the seedling
over time are expressed as the percentage of the initial
seed dry mass (SDMi) so as to account for the effects of
the initial seed mass on variations in the mass of the
different parts of the seedling. Total N content and 15 N
isotopic composition were measured using a mass spectrometer (VG SIRA 9) connected to an automatic combustion analyser (Carlo Erba NA 1500). On each sampling occasion, three replicates were obtained by pooling
the five plants grown in each sample pot. The seedling
parts (radicle, hypocotyl, cotyledons) were analysed separately after rinsing with deionised water to ensure there
was no 15 N left on the organs surface. Results are
expressed as 15 N content which is the product of N
content and 15 N atom excess, i.e. 15 N abundance of the
sample minus 15 N abundance in the air (0.3663 % 15 N).
Data fitting and statistical analyses
Two-way ANOVA was performed to test the effects of
genotype (G), N supply (S) and G x S interactions of the
measured variables (STATGRAPHICS Plus 3.1. software). The N supply in each genotype was compared
using Bonferronis multiple comparison procedure. For
a given level of N supply, the genotype effect was tested
using a one-way ANOVA. A model was built to calculate N fluxes to estimate the relative proportions of
nitrogen originating from mobilisation of seed reserves
and N absorption using 15 N labelling (Fig. 1). To estimate N fluxes, the non-parametric Kruskal-Wallis procedure was performed to test the genotype and the N
supply effects because sample sizes were reduced (two
replicates per modality i.e. n=8 and n=4 for testing the
genotype and the N supply effects respectively) and the
variance equality was verified (Levenes test).

333

Results
Seedling growth and the length of the radicle
and hypocotyl depend on nutrient supply
The hypocotyl and the radicle reached plateau values
150 h after germination independent of the amount of
nutrients in the solution. The final length of the hypocotyl ranged from 52 to 76 mm (Fig. 2). No genotype
effect was observed, but in both genotypes, final lengths
were shorter in seeds supplied with the S0 solution
compared to those supplied with the S1, S2 and S3
solution. By contrast, the final length of the radicle
ranged from 45 to 57 mm and significant effects of the
nutrient supply and genotype were observed. Paraggio
radicles were longer than A17 radicles independent of
the nutrient solution (Fig. 2). The effect of the nutrient
supply was visible in the longer radicle in Paraggio
seeds grown in the S0 solution, leading to a significant
G x S interaction effect, since there was no significant S
effect on the final length of the radicle in A17
(Bonferronis comparison, data not shown). Regarding
changes in dry mass (DM) with time (Fig. 3), a slow
decrease in seedling DM was observed independent of
the genotype and solution. The decrease is expressed as
a percentage of the initial seed dry mass (SDMi) and
ranged between 10 % and 15 % except in seeds grown in
the S0 solution, in which case it reached about 20 %
(Fig. 3). Irrespective of the genotype and nutrient supply, there was a sharp decrease in cotyledon DM of from
25 % (after 24 h) to 75 % (after 264 h) of the initial
SDMi (Fig. 3). During the same period, the DM of the
hypocotyl and radicle increased (Fig. 3). Hypocotyl DM
reached between 30 % and 45 % of the SDMi 144 h
after sowing. Radicle DM reached its peak value, 10 %
of the SDMi, sooner i.e. 96 h after sowing. When
estimating the effect of different nutrient solutions and
genotypes, we found a significant effect of the nutrient
solution on seedling DM from 96 h after sowing
(P<0.001; Fig. 3). This effect was due to a decrease in
cotyledon DM (at 96 and 264 h after sowing) concomitant with an increase in hypocotyl DM (from 144 h after
sowing) and an increase in root DM (at 144 and 264 h
after sowing) when solution was added, independent of
the concentration (S1, S2 and S3). The decrease in
cotyledon DM at 96 h after sowing was not linked with
an increase in the DM of the hypocotyl or radicle,
meaning C losses from the cotyledons were only used
for respiration at this time. Genotype effects were

334

Plant Soil (2015) 386:331340

Seedling S, Es
a

Nutrient Soluon
N, E

Radicle
R, Er

E: isotopic excess in the nutrient soluon

Ec: isotopic excess in the cotyledons
Er: isotopic excess in the radicle
Eh: isotopic excess in the hypocotyl

b
g

Cotyledons
C, Ec

Calculated variables
N* = N.E 15N amount (g 15N plant-1) in the nutrient soluon
C* = C.Ec 15N amount (g 15N plant-1) in the cotyledons
R* = R.Er 15N amount (g 15N plant-1) in the radicle
H* = H.Eh 15N amount (g 15N plant-1) in the hypocotyl

d
c
Hypocotyl
H, Eh

Equaons
g = b+c
R/t = a+b-d
H/t = c+d
S/t = a+g

Measured variables
N : N amount (g 15N plant -1) in the nutrient soluon
C : N amount (g 15N plant -1) in the cotyledons
R : N amount (g 15N plant -1) in the radicle
H : N amount (g 15N plant -1) in the hypocotyl

Esmated uxes
a : N uptake ux (g day -1 plant -1)
g : N remobilizaon ux from the cotyledons (g day -1 plant -1)
b : N remobilizaon ux from cotyledons to radicle (g day -1 plant -1)
c : N remobilizaon ux from cotyledons to hypocotyl (g day -1 plant -1)
d : N transfer ux from radicle to hypocotyl (g day -1 plant -1)

R*/t = a.E+b.Ec-d.Er
H*/t = c.Ec
S*/t = a.E+g.Ec

a = (S*/t - S/t.Ec)/ (E-Ec)

g = S/t - a

Fig. 1 Compartmental model for estimating the fluxes and relative proportions of N originating from seed reserves and absorption

had germinated about 24 h after sowing, 15 N content

increased in all the seedling parts (radicle, hypocotyl and
cotyledons, Figs. 4a, b, c and d) in treatments with S1, S2
and S3 solutions, whereas it remained close to zero in S0.
The increase in 15 N content was earlier and higher in the
radicle than in the hypocotyl and was slowest and very
small in the cotyledons until 192 h after sowing. The effect
of the N supply was significant from 24 h after sowing in
the radicle and the cotyledons, and from 48 h after sowing
in the hypocotyl, with lower 15 N contents in S1 and S2
than in S3 (Figs. 4b, c and d). On each sampling occasion,
the differences in 15 N contents between treatments (S1, S2
and S3) were larger in A17 than in Paraggio. For a given

mainly observed in root and cotyledon DM from 144 h

after sowing, when higher values were measured in
Paraggio irrespective of the solution.
Patterns of seedling 15 N enrichment of seedling
highlights early absorption just after radicle protrusion
The initial N content in seeds was significantly higher in
Paraggio (7.60.1 % w/w) than in A17 (7.00.1 % w/w).
Initial %15 N excess in seeds was also significantly different but very low in the two genotypes (0.00012
0.00003 % in Paraggio and 0.000250.00002 % in
A17 respectively). In both genotypes, as soon as the seeds

100

100

Length (mm)

Hypocotyl

75

75

A17 S0

50

50

25

Radicle

Paraggio S0

A17 S1

Paraggio S1

A17 S2

Paraggio S2

A17 S3

Paraggio S3

25

0
00

48
48

9696

144 144 192 192240

240
288

00

48
48

96 96

144 144 192 192

240

240
288

Time aer germinaon (hours)

Fig. 2 Time course of elongation of the radicle and the hypocotyl
after sowing in A17 (black symbols, solid line) and Parragio
(empty symbols, dashed line) genotypes according to the concentration of the nutrient solution (S0, S1, S2, S3). Lines are fitted to a

Weibull function. The effects of N supply and of genotype are

denoted * and respectively when significant (P<0.05). Vertical
bars denote s.e. (n=15)

Plant Soil (2015) 386:331340

335

100%

100%

A17 S0

90%

Dry mass (in % of inial seed mass)

75%
80%

*
70%

Paraggio S1

A17 S2

Paraggio S2

A17 S3

Paraggio S3

50%

* *
*

60%

Paraggio S0

A17 S1

25%

Cotyledons

Whole seedling

50%

0%
0

48

96

144

192

240

288

75%

48

96

144

192

240

288

20%

Hypocotyl

Radicle

* *

50%

15%

10%
25%

5%

0%

0%
0

48

96

144

192

240

288

48

96

144

192

240

288

Time aer sowing (hours)

Fig. 3 Changes in dry mass with respect to the initial seed dry
mass with time after sowing in A17 (black symbols, solid line) and
Paraggio (empty symbols, dashed line) genotypes according to the
concentration of the nutrient solution (S0, S1, S2, S3) in the whole

seedling, cotyledons, hypocotyl and radicle. The solution and

genotype effects are denoted * and respectively when significant
(P<0.05). Vertical bars denote s.e. (n=15)

genotype, the effect of N supply was only significant in

Paraggio radicle, hypocotyl, and cotyledons 24 h after
sowing, whereas in A17 it was significant at 24 h in all
three organs, at 144 h in the radicle and the hypocotyl, and
at 192 h in the hypocotyl and the cotyledons, which were
the last organs supplied with 15 N (Bonferronis comparison, data not shown). 15 N content stopped increasing
concomitantly with the end of elongation of the seedling
parts (Figs. 2 and 4b and c).

mean germination time), a large daily flux of N came

from the seed reserve compartment (g), reaching about
30 g day1 plant 1 24 h after sowing, and remaining at
this level until 48 h after sowing in both genotypes
(Fig. 5). Absorption started about 24 h after sowing
and daily N amounts from absorption (a) increased until
96 h after sowing whereas the flux (g) from the seed
reserves started decreasing. From 24 to 96 h after sowing, daily N absorption and N remobilisation allowed
the flow of N towards the growing organs i.e. radicle
and hypocotyl, to be maintained at about 30 g day1
plant 1. After 96 h, both N remobilisation and absorption rates decreased, which was concomitant with the
end of hypocotyl elongation. Finally, the N
remobilisation flux (g) reached negative values in
Paraggio in the S3 treatment 264 h after sowing, and
in A17 in the S1, S2 and S3 treatments at 144, 192 and
264 h after sowing, meaning the plants released N when
the hypocotyl and root were no longer growing (Fig. 2).
No significant differences in the fluxes from seed

Seed reserves are the main source of N for seedling

growth and are supported by exogenous N uptake
after germination
Figure 5 shows daily fluxes of N from external uptake
and from remobilisation of seed reserves based on the
calculations performed with the compartmental model
in both genotypes (a and g fluxes respectively, Fig. 1).
The two genotypes followed the same general pattern
(Fig. 5). During the first 24 h after sowing (before the

Plant Soil (2015) 386:331340

1 000

600

400
300
200
100

48

96

144

192

240

Cotyledons15N content

(g15N gDW-1)

800

*
*

600

400

200

0
0

48

96

144

192

240

*
b
0

288

48

96

144

192

240

A17 S0
250
200

Paraggio S0

A17 S1

Paraggio S1

A17 S2

Paraggio S2

A17 S3

Paraggio S3

150

*
*

100
50

* *

*
d

0
288

288

300

(g15N gDW-1)

1 000

Hypocotyl 15N content

400
200

600

800

(g15N gDW-1)

Radicle15N content

500

(g15N gDW-1)

Whole seedling 15N content

336

48

96

144

192

240

288

Time aer sowing (hours)

15

15

Fig. 4 Changes in N content (g N g DW ) in the whole

seedling (a) in the radicle (b), hypocotyl (c) and the cotyledons (d)
in A17 (black symbols, solid line) and Paraggio (empty symbols,
dashed line) genotypes according to the concentration of the

nutrient solution (S0, S1, S2, S3). The effects of N supply and of
genotype are denoted * and respectively when significant
(P<0.05). Vertical bars denote s.e. (n=15)

reserves and N absorption were observed between the

two genotypes over the sampling period (data not
shown).

24 h after sowing: mainly a higher value in A17 under no

N supply). N allocation to the hypocotyl was lower (c,
ca.10 g day1 plant 1) and almost zero under no N
supply (Fig. 6). The radicle also absorbed N directly from
the solution as shown by the 15 N content (Fig. 3). Fig. 6
shows a huge flux from the radicle to the hypocotyl (d)
from germination (about 24 h after sowing) to 96 h after
sowing, when the radicle reached its final length (Fig. 2).
The flux from the radicle was supported by the N flux
from the cotyledons (c), both fluxes helped maintain the
total flux to the hypocotyl at about 30 g day1 plant 1.
From 96 h after sowing onwards, N fluxes from the
cotyledons to the radicle (b) and to the hypocotyl (c)
decreased significantly. At 192 h after sowing, fluxes
from the cotyledons to the hypocotyl (c) and from the
radicle to the hypocotyl (d) were negative in A17 with the
S3 and S2 treatments, meaning that the radicle and the
cotyledons released N under high N supply conditions,
which was concomitant with the end of hypocotyl elongation. Considering the sampling period as a whole, a

The hypocotyl is the main sink for N from seed reserves

and from external absorption
The compartmental model enabled us to calculate net
fluxes between the different seedling parts (b, c, d in
Fig. 1). These calculations enabled finer estimation of
the two components of the daily N remobilisation flux
(g), i.e. b (from cotyledons to the radicle) and c (from the
cotyledons to the hypocotyl) and of the daily N transfer
between the radicle and the hypocotyl (d). Seed N reserves from the cotyledons were allocated to the growing
parts of the embryo i.e. radicle and hypocotyl (Fig. 6).
Total daily N fluxes from the cotyledons were mainly
allocated to the radicle during the first 48 h after sowing
(b, 2040 g day1 plant 1 depending on N supplies and
genotypes, but only one genotype effect was observed at

Daily N uxes g N day-1 plant -1

Plant Soil (2015) 386:331340

337

Paraggio

50

A17

50

S0

S0

40

40

S1
S2

30

30

S3

20

20

10

10

0
0

-10

48

96

144

192

S1
S2
S3

0
288 0
-10

240

-20

48

96

144

192

240

288

-20

Time aer sowing (hours)

1

treatments. No significant effect of N supply on uptake and

remobilisation were observed (Kruskal-Wallis procedure, n=8).
Vertical bars denote s.e

significant genotype effect was only observed 24 h after

sowing in the daily N fluxes b, c and d with lower values
in Paraggio (Fig. 6). A significant effect of N supply was

observed in daily N fluxes from the cotyledons to the

radicle 96 h after sowing, with lower values with solution
S0. Indeed, 96 h after sowing, the N from the solution

Amount of N in % of inial seed N

140

*
*

120

100

80

60
0

48

96

144

192

240

288

Daily N uxes g N plant-1 day-1

Fig. 5 Daily N uptake (a, g N day plant , dotted line) and

remobilisation (g, g N day1 plant1, solid line) fluxes in
Paraggio (A) and A17 (B) genotypes under S0, S1, S2 and S3

40
30
20

A17 S0

Paraggio S0

A17 S1

Paraggio S1

A17 S2

Paraggio S2

A17 S3

Paraggio S3

10
0
0

48

96

144

192

240

288

-10
-20

Daily N uxes g N plant-1 day-1

Daily N uxes g N plant-1 day-1

Cotyledons towards hypocotyl (c)

50

Cotyledons towards radicle (b)

50

40

30
20
10

0
0

48

96

144

192

240

288

-10
-20

Radicle towards hypocotyl (d)

50
40
30

20
10
0
0

48

96

144

192

240

288

-10
-20

Time aer sowing (hours)

Fig. 6 Amount of N in the whole seedling as a proportion of
initial seed N and daily N fluxes from the cotyledons to the radicle
(b flux, g N day1 plant1), from the cotyledons to the hypocotyl
(c flux, g N day1 plant1), and from the radicle to the hypocotyl
(d flux, g N day1 plant1) in Paraggio and A17 genotypes under

S0, S1, S2 and S3 treatments. The N supply and genotypes effects

are denoted with * and respectively when significant (KruskalWallis procedure, n=8 and n=4 for testing genotype and solution
effects, respectively). Vertical bars denote s.e

338

(quantified by 15 N content) in the cotyledons started

being a source of N for this plant part with the S1, S2
and S3 treatments (Fig. 4d). Therefore, with the S0 solution, the amount of N in the cotyledons only came from
initial seed reserves with no additional N from the solution, leading to a net depletion of N from the cotyledons.
Figure 6 shows that the global net N balance (ratio of
N seedling content to initial seed N content) decreased
progressively from 192 h after sowing with the S0
solution in both genotypes, indicating that, from then
on, there was a N net efflux from the seedling (i.e.
radicle and hypocotyl, Fig. 1), which reached about
20 % of the seed initial content. This efflux occurred
in both the cotyledons and the radicle as shown in Fig. 6.
By contrast, where N absorption occurred in treatments
with the other solutions (S1, S2 and S3), the net balance
remained close to 100 %, meaning N uptake from
external solution compensated for N losses.

Discussion
The present study enabled us to calculate the relative
contribution of N from seed reserves and from absorption during the first stages of crop establishment in
Medicago truncatula, a legume species with high seed
N content. Previous studies reported early absorption in
legume species but without mentioning the exact time N
absorption began (Cooper 1977). In our experiment,
when the level of nutrient solution supply varied, the
concentrations of all the nutrients in the solution were
changed in the same time, which could have influenced
seedling growth. However, knowing this limit of the
experiment, by 15 N labelling the nutrient solution, we
were able to check the origin of the different sources of
N in the different seedling parts more precisely. Our
results demonstrate that with an external supply of N,
N absorption began as soon as the radicle protruded
from the teguments (i.e. at germination) and that
chronogically, N was first allocated to the radicle and
then to the hypocotyl for elongation. The cotyledons
were also supplied with external N but to a lesser extent
during the early days of elongation. When the radicle
and hypocotyl reached their final lengths, the amount of
external N absorbed decreased as a consequence of a
decrease in demand for N by both the radicle and the
hypocotyl. The release of N from the cotyledons and the
radicle observed when elongation stopped highlights the
fate of absorbed N, which was neither stored nor

Plant Soil (2015) 386:331340

assimilated for tissue growth. This could be linked to

the allocation of carbohydrate resources which might be
oriented towards more C demanding metabolic functions such as respiration and/or cellulose accumulation
in the cell wall, and to the lack of newly produced
carbohydrate resources under heterotrophic conditions.
These observations in the legume species Medicago
truncatula are in agreement with results observed in
sugar beet (Beta vulgaris L.), another dicotyledonous
but non-legume species: N absorption in sugar beet
started just after germination with absorbed N cumulating first in the root and then in the hypocotyl (Drr and
Mary 1998). These authors showed that absorption began later in wheat (Triticum aestivum L), a monocotyledonous species, after the leaves broke out of the coleoptile. Other studies on the allocation of remobilised
seed phosphorus (P) reserves and external P uptake
during early growth in maize, another monocotyledonous species, showed that the timing of P uptake was
similar to that in wheat, since P uptake began as soon as
the seedling leaves developed (Nadeem et al. 2012,
Nadeem et al. 2013). These observed differences in
early external mineral uptake among groups of species
are also consistent with earlier studies in dicotyledonous
pasture species and grasses (MacWilliam et al. 1970),
showing that dicotyledonous species are more responsive to an early external N supply (when available), than
monocotyledonous species, irrespective of the nature of
the seed reserves. Interestingly, Fayaud et al. (2014)
reported similar effects on early N supply to those
observed in monocotyledonous species, in pea (Pisum
sativum L.), which is a dicotyledonous legume species
but whose seedling growth and emergence resemble
those of monocotyledonous species (pea cotyledons
remain in the soil and the seedling has an epicotyl like
many monocotyledonous species rather than a hypocotyl like most dicotyledons). Finally, the timing of the
beginning of mineral absorption and its effects on seedling growth could be linked to seed morphology and
embryo growth, and seed classification (as proposed for
instance in Finch-Savage and Leubner-Metzger 2006),
could help predict these effects.
Our study revealed no differences between the two
genotypes except for very early N uptake which was
lower in Paraggio, the genotype with the highest N
initial seed content. This genotype had a lower %15 N
excess in the root just after germination, meaning the
uptake flux to the root was lower than in A17, perhaps
due to the larger amount of readily available N from the

Plant Soil (2015) 386:331340

seed reserves. The larger amount of N in the seeds could

be due to the seed production conditions, which were
not the same for the two genotypes. However, this
difference in N content in the two genotypes was already
observed in other seed lots produced under the same
conditions and could also be a genotypic difference as
reported by Brunel et al. (2009). In agreement with these
observations, a study on maize seedlings also showed
that initial P seed contents affected the intensity of P
uptake and the allocation to roots just after the radicle
protruded (Nadeem et al. 2012).

Conclusion
This study describes patterns of early N uptake and
remobilisation during heterotrophic growth in the legume species Medicago truncatula. Because of its high
seed N content, contrasting patterns with other nonlegume seeds might be expected, but the observed differences were more pronounced between monocotyledonous and dicotyledonous species than between legumes and dicotyledonous non-legumes. The modelling
approach used in this study allowed N uptake and N
remobilisation from seed reserve to be analysed separately, and the seedling N budget to be interpreted using
both sources during early growth. The flux compartment
model we developed is thus a powerful tool to analyse
(i) differences among species in the allocation of internal
and external sources of N and of other minerals before
autotrophic growth and (ii) the putative effects of genetic diversity in early mineral absorption related to initial
seed characteristics or to differences in seedling growth.
Acknowledgements This research was funded by the Region
Pays de Loire and INRA (French National Institute for Agricultural Research). The authors thank MH. Wagner and L. Ledroit for
technical assistance with measurements on the plants, and O.
Delfosse for mass spectrometry analysis. The authors are also
grateful to E. Personeni for her helpful reading of the manuscript
and to Jean-Paul Maalouf for his help with statistical analyses.

References
Andrews M, Scott WR, McKenzie BA (1991) Nitrate effects on
pre-emergence growth and emergence percentage of wheat
(Triticum aestivum L.) from different sowing depths. J Exp
Bot 42:14491454

339
Brunel S, Teulat-Merah B, Wagner MH, Huguet T, Prosperi JM,
Durr C (2009) Using a model-based framework for analysing
genetic diversity during germination and heterotrophic
growth of Medicago truncatula. Ann Bot 103:11031117
Cook DR (1999) Medicago truncatula - a model in the making!
Curr Opin Plant Biol 2:301304
Cooper CS (1977) Growth of the legume seedling. Adv Agron 29:
119139
Duc G, Mariotti A, Amarger N (1988) Measurements of genetic
variability for symbiotic dinitrogen fixation in field-grown
faba bean (Vicia faba L.) using a low level 15 N-tracer
technique. Plant Soil 106:269276
Drr C, Mary B (1998) Effects of nutrient supply on preemergence growth and nutrient absorption of wheat
(Triticum aestivum L.) and sugarbeet (Beta vulgaris L.).
Ann Bot 81: 665672
Durrant MJ, Mash SJ (1989) Stimulation of sugarbeet hypocotyl
extension with potassium nitrate. Ann Appl Biol 115:367
374
Fayaud B, Coste F, Corre-Hellou G, Gardarin A, Drr C (2014)
Modelling early growth under different sowing conditions: A
tool to predict variations in intercrop early stages. Eur J
Agron 52:180190
Finch-Savage WE, Leubner-Metzger G (2006) Seed dormancy
and the control of germination. New Phytol 171:501523
MacWilliam JR, Clements RJ, Dowling PM (1970) Some factors
influencing the germination and early seedling development
of pasture plants. Aust J Agric Res 21:1932
Monti LM, Grillo S (1983) Legume seed improvement for protein
content and quality. Qual Plant Foods Hum Nutr 32:253266
Moreau D, Voisin AS, Salon C, Munier-Jolain N (2008) The
model symbiotic association between Medicago truncatula
cv. Jemalong and Rhizobium meliloti strain 2011 leads to Nstressed plants when symbiotic N2 fixation is the main N
source for plant growth. J Exp Bot 59:35093522
Nadeem M, Mollier A, Morel C, Shahid M, Aslam M, Zia-urRehman M, Ashfaq Wahid M, Pellerin S (2013) Maize
seedling phosphorus nutrition: Allocation of remobilized
seed phosphorus reserves and external phosphorus uptake
to seedling roots and shoots during early growth stages.
Plant Soil 371:327338
Nadeem M, Mollier A, Morel C, Vives A, Prudhomme L, Pellerin
S (2012) Seed phosphorus remobilization is not a major
limiting step for phosphorus nutrition during early growth
of maize. J Plant Nutr Soil Sci 175:805809
Rosati A, Magnifico V (2001) Effect of early and localised nitrogen fertilisation on growth, yield, earliness and fertiliser use
efficiency of field eggplant. Acta Horticult 563:187194
Saglio P, Pradet M (1980) Soluble sugars, respiration and energy
charge during aging of excised maize root tips. Plant Physiol
66:516519
Thoquet P, Ghrardi M, Journet ET, Kereszt A, An JM, Prospri
JM, Huguet T (2002) The molecular genetic linkage map of
the model legume Medicago truncatula: an essential tool for
comparative legume genomics and the isolation of agronomically important genes. BMC Plant Biol 2:113
Tivoli B, Barange A, Sivasithamparam K, Barbetti J (2006)
Annual Medicago: From a Model Crop Challenged by a
Spectrum of Necrotrophic Pathogens to a Model Plant to
Explore the Nature of Disease Resistance. Ann Bot 98:
11171128

340
Uzun F, Aydin I (2004) Improving germination rate of
Medicago and Trifolium species. Asian J Plant Sci 3:
714717
Voisin AS, Salon C, Munier-Jolain N, Ney B (2002) Quantitative
effects of soil nitrate, growth potential and phenology on
symbiotic nitrogen fixation of pea (Pisum sativum L.). Plant
Soil 243:3142

Plant Soil (2015) 386:331340

Young ND, Debell F, Oldroyd GE, Geurts R, Cannon SB,
Udvardi MK, Benedito VA, Mayer KFX, Gouzy J, Schoof
H et al (2011) The Medicago genome provides insight into
the evolution of rhizobial symbioses. Nature 480:520524
Zeng LW, Cocks PS, Kailis SG, Kuo J (2005) Structure of the seed
coat and its relationship to seed softening in Mediterranean
annual legumes. Seed Sci Technol 33:351362