Content uploaded by Rowan F Sage
Author content
All content in this area was uploaded by Rowan F Sage on May 27, 2016
Content may be subject to copyright.
Content uploaded by Tammy L Sage
Author content
All content in this area was uploaded by Tammy L Sage on Feb 09, 2016
Content may be subject to copyright.
Journal of Experimental Botany, Page 1 of 13
doi:10.1093/jxb/err059
RESEARCH PAPER
The occurrence of C
2
photosynthesis in Euphorbia subgenus
Chamaesyce (Euphorbiaceae)
Tammy L. Sage, Rowan F. Sage*, Patrick J. Vogan, Beshar Rahman, Daniel C. Johnson, Jason C. Oakley and
Marta A. Heckel
Department of Ecology and Evolutionary Biology, The University of Toronto, 25 Willcocks Street, Toronto, ON M5S3B2, Canada
* To whom correspondence should be addressed: E-mail: r.sage@utoronto.ca
Received 5 January 2011; Revised 11 February 2011; Accepted 11 February 2011
Abstract
This study investigated whether Euphorbia subgenus Chamaesyce subsection Acutae contains C
3
–C
4
intermediate
species utilizing C
2
photosynthesis, the process where photorespired CO
2
is concentrated into bundle sheath cells.
Euphorbia species in subgenus Chamaesyce are generally C
4
, but three species in subsection Acutae (E. acuta,
E. angusta, and E. johnstonii) have C
3
isotopic ratios. Phylogenetically, subsection Acutae branches between basal
C
3
clades within Euphorbia and the C
4
clade in subgenus Chamaesyce.Euphorbia angusta is C
3
, as indicated by
a photosynthetic CO
2
compensation point (G)of69mmol mol
21
at 30 C, a lack of Kranz anatomy, and the
occurrence of glycine decarboxylase in mesophyll tissues. Euphorbia acuta utilizes C
2
photosynthesis, as indicated
byaGof 33 mmol mol
21
at 30 C, Kranz-like anatomy with mitochondria restricted to the centripetal (inner) wall of
the bundle sheath cells, and localization of glycine decarboxlyase to bundle sheath mitochondria. Low activities of
PEP carboxylase, NADP malic enzyme, and NAD malic enzyme demonstrated no C
4
cycle activity occurs in E. acuta
thereby classifying it as a Type I C
3
–C
4
intermediate. Kranz-like anatomy in E. johnstonii indicates it also utilizes C
2
photosynthesis. Given the phylogenetically intermediate position of E. acuta and E. johnstonii, these results support
the hypothesis that C
2
photosynthesis is an evolutionary intermediate condition between C
3
and C
4
photosynthesis.
Key words: C
3
–C
4
intermediate, C
4
photosynthesis, gas exchange, Kranz anatomy, photorespiration, photosynthetic evolution.
Introduction
C
4
photosynthesis has independently evolved over 60 times
in divergent families of the angiosperms, making it one of
the most convergent of evolutionary phenomena in the
biosphere (Conway Morris, 2003;Sage et al., 2011a). The
process of C
4
evolution has been widely studied, largely in
genera with species that exhibit traits intermediate between
C
3
and C
4
plants. The best-studied genera with C
3
–C
4
intermediate species are Flaveria (Asteraceae) and Stein-
chisma (¼Panicum sensu lato, Poaceae) (Monson and
Rawsthorne, 2000). C
3
–C
4
intermediates have also been
identified in Heliotropium (Boraginaceae), Neurachne (Poa-
ceae), Alternathera (Amaranthaceae), Cleome (Cleomea-
ceae), and Mollugo (Molluginaceae) (reviewed in Bauwe,
2011, and Sage et al., 2011a). C
3
–C
4
intermediate plants
with no close relationship to C
4
lineages are also recognized
in the Brassicaceae (Moricandia, Diplodoxis) and Asteraceae
(Parthenium)(Sage et al., 1999). Based on extensive research
in Flaveria,Moricandia,andPanicum sensu lato, elaborate
models of C
4
evolution have been proposed (Monson and
Moore, 1989; Monson, 1999;Monson and Rawsthorne,
2000;Sage, 2004). A central feature of these models is the
appearance of photorespiratory CO
2
concentration, a pro-
cess where photorespired CO
2
accumulates in the bundle
sheath (BS) compartment, leading to enhanced efficiency
of BS Rubisco (von Caemmerer, 1989;Monson and
Rawsthorne, 2000). Photorespiratory CO
2
concentration
occurs when glycine decarboxylase (GDC) is localized to
the BS tissue, such that all of the glycine produced in
mesophyll (M) cells by Rubisco oxygenation must move to
the BS tissue for conversion to serine and CO
2
by GDC
(Monson and Rawsthorne, 2000;Bauwe, 2011). The CO
2
released by GDC is trapped in the BS cell, where it can
ªThe Author [2011]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved.
For Permissions, please e-mail: journals.permissions@oup.com
Journal of Experimental Botany Advance Access published March 31, 2011
at University of Toronto Library on April 1, 2011jxb.oxfordjournals.orgDownloaded from
accumulate and be refixed by BS Rubisco at much higher
efficiency than is possible in M cells (von Caemmerer,
1989). GDC localization typically follows a mutation that
prevents GDC expression in the M tissue (Hylton et al.,
1988;Voznesenskaya et al., 2007;Bauwe, 2011). Associated
with the GDC localization are a series of traits which are
thought to enhance the capture and fixation of photo-
respired CO
2
by BS Rubisco. These include an enlargement
of BS cells to form a Kranz-like anatomy, and the
positioning of mitochondria and chloroplasts along the cen-
tripetal (inner) wall of the BS cell adjacent to the vascular
tissue. These traits indicate that photorespiratory CO
2
concentration is a distinct carbon concentrating mechanism
that specifically evolved to compensate for high rates of
photorespiration. In recognition of this, Vogan et al. (2007)
have proposed the term ‘C
2
photosynthesis’ to refer to
photorespiratory CO
2
concentration. The use of ‘C
2
photo-
synthesis’ is logically and historically consistent with the
prior use of the terms C
2
metabolism and C
4
photosynthe-
sis. Tolbert (1997) used ‘C
2
metabolism’ to refer to the
photorespiratory metabolic cycle, and hence it makes sense
to label a carbon concentration mechanism based on
photorespiratory metabolism as C
2
photosynthesis. Logical
consistency is apparent in that ‘C
2
’ refers to the number of
carbons in the Rubisco product that shuttles CO
2
into the
BS, just as ‘C
4
’ refers to the number of carbons in the
carboxylation product that shuttles CO
2
into the BS cells of
C
4
plants.
While C
2
photosynthesis is widely thought to be the key
intermediate stage in the evolution of C
4
photosynthesis
(Monson and Rawsthorne, 2000;Sage, 2004), there is limited
phylogenetic support for this hypothesis. Only in the genera
Flaveria,Cleome,andMollugo have detailed, molecular
phylogenies demonstrated that C
2
photosynthesis occurs in
species that branch at phylogenetic nodes between C
3
and C
4
clades (McKown et al.,2005;Feodorova et al.,2010;Christin
et al.,2011). In the other clades where C
3
–C
4
intermediate
taxa occur, phylogenies with enough detail to resolve the
relationships between C
3
,C
4
,andtheC
3
–C
4
species have yet
to be published. Hence, to facilitate the study of C
4
evolution,
it is useful to identify additional clades where C
2
occurs in
species branching between C
3
and C
4
nodes within a phylog-
eny. In such studies, C
2
photosynthesis is indicated by reduced
CO
2
compensation points of photosynthesis, non-linearity in
the response of the CO
2
compensation point to O
2
, localiza-
tion of GDC to the BS tissue, high refixation of photorespired
CO
2
, and a line-up of mitochondria along the centripetal end
of the BS cells adjacent to the vascular tissue (Bauwe et al.,
1987;von Caemmerer, 1989;Monson, 1999;Monson and
Rawsthorne, 2000).
One promising group to look for evolutionary intermedi-
ates is the subsection Acutae in the genus Euphorbia
subgenus Chamaesyce (Euphorbiaceae). Based on morpho-
logical characters, Webster et al. (1975) hypothesized that
the species in subsection Acutae make up the ancestral clade
of subgenus Chamaesyce, which is a large group of C
4
species that derive from C
3
species in Euphorbia (Steinmann
and Porter, 2002; Tropicos, 2010). Chamaesyce has been
recognized as an independent genus (Webster, 1994), but
this makes Euphorbia paraphyletic and thus its status as
a distinct genus is not currently favoured (Tropicos, 2010).
Webster et al. (1975) showed that two species in subsection
Acutae—E. angusta and E. acuta—have C
3
isotopic values,
while a third, E. lata,hasaC
4
isotopic value. All other
species in Euphorbia subgenus Chamaesyce are believed to
be C
4
plants based on isotopic values, Kranz anatomy, gas
exchange, or phylogenetic affinity to known C
4
species
(Webster et al., 1975;Koutnik, 1987). Subsequently,
Euphorbia johnstonii has been identified as a fourth species
in subsection Acutae, and is thought, on morphological
grounds, to be very close to E. acuta (Mayfield, 1991).
Recent molecular phylogenies support the placement of
subsection Acutae between C
3
species of Euphorbia and C
4
species of Euphorbia subgenus Chamaesyce (Steinmann and
Porter, 2002;Yang and Berry, 2007), raising the possibility
that the species in subsection Acutae may express evolution-
arily intermediate traits between the C
3
and C
4
conditions.
All four species of Euphorbia subgenus Chamaesyce
subsection Acutae grow in south-western Texas, USA and
adjacent Mexico, and are abundant enough for easy
collection and study in their field habit (Correll and
Johnston, 1979). The purpose of this study was to collect
live specimens of these species, describe their field ecology,
and analyse their physiology and cell biology to determine if
any of the species are C
3
–C
4
intermediates conducting C
2
photosynthesis. The Caribbean species Euphorbia mesem-
bryanthemifolia was included in the study as a representative
of subgenus Chamaesyce (Herndon, 1996; Tropicos, 2010),
which we assumed would have a well-developed C
4
pathway. The growth form of E. mesembryanthemifolia
resembles that of E. acuta.
Materials and methods
Field work
Seeds of E. acuta Engelm. (synonymous with the Mexican species
E. georgei Oudejans; Tropicos, 2010) and E. lata Engelm. were
collected on 17 August 2007 from plants growing on a limestone
outcrop along Highway 18, 8 km north of the Interstate 10
interchange at Fort Stockton, Texas. Seeds of E. angusta Engelm.
were gathered on 7 April 2007 from plants in a limestone road cut
along Highway 12, about 24 km north of Uvalde in Uvalde
County, Texas. Euphorbia mesembryanthemifolia Jacq. seeds were
collected near Progreso, Yucatan, Mexico on 26 June 2008. Sarah
Taylor (University of Texas) collected E. johnstonii Mayfield plants
on 11 October 2007 near Nuevo Leon, Mexico. These plants were
couriered to Toronto. They lacked seeds and did not survive
transplanting. It was possible to sample viable leaves of
E. johnstonii for qualitative observations using light and trans-
mission electron microscopy (TEM) as well as immunolocalization
but living material could not be preserved for physiological/
biochemical analyses. Leaves of E. johnstonii were not used for
quantitative analyses of leaf anatomy because they were not grown
in the same conditions as the other four species.
To evaluate diurnal leaf temperatures of E. acuta in its natural
habitat,Veriteq Spectrum data loggers (SP1700, Veriteq Corpora-
tion, Vancouver, British Columbia, Canada) were set up to log fine
wire thermocouples in July 2010. Thirty-six gauge thermocouple
junctions were placed against the bottom of attached leaves from
2of13 |Sage et al.
at University of Toronto Library on April 1, 2011jxb.oxfordjournals.orgDownloaded from
plants growing at the collection site 8 km north of Fort Stockton,
Texas.
Carbon isotope ratios were assayed on dried leaves from field-
collected plants by the University of California, Davis stable
isotope facility (http://stableisotopefacility.ucdavis.edu).
Plant growth, whole-leaf gas exchange, and enzyme assays
Plants were grown from seed in a rooftop greenhouse located at
the Earth Sciences Centre of the University of Toronto. Plants
were grown in 4.0 l or 8.0 l pots in equal parts sand, Pro-Mix
potting soil (Premier Horticulture, Inc., Quakertown, PA, USA),
and sterilized topsoil. Plants were watered as necessary to avoid
drought and fertilized weekly with equal parts of two commercial
fertilizers (1.4 g l
1
each of 24-8-16 Miracle-Grow All-Purpose
plant food and 28-10-10 Miracle Grow Evergreen Tree and Shrub
Food; Scotts-Canada, Mississauga, Ontario), with bimonthly
supplements of approximately 300 ml of a 1 mM magnesium
sulphate and 5 mM calcium nitrate solution. The approximate day/
night temperature of the greenhouse was 30/25 C and peak
midday irradiance exceeded 1600 lmol photons m
2
s
1
on sunny
days. Plants were periodically trimmed to a canopy area approx-
imately equal to the pot width.
All measurements were conducted on recently expanded healthy,
leaves. Photosynthetic gas exchange was measured using a LI-6400
portable photosynthesis system (Li-Cor, Inc., Lincoln, NE, USA)
during August and September 2008. Measurements were con-
ducted at a leaf temperature of 30 C, a photon flux density of
1300 lmol m
2
s
1
, and leaf-to-air vapour pressure deficit (VPD)
of 2.0 kPa. Because leaves were small (about 0.5 cm
2
), multiple
leaves on a branch were placed in the chamber. CO
2
compensation
points were calculated as the x-intercept of a linear regression
through the five lowest intercellular CO
2
(C
i
) values on a graph of
net CO
2
assimilation rate (A) versus C
i
. The initial point in each A/
C
i
curve was determined at 370 lmol CO
2
mol
1
air after which
ambient CO
2
was lowered in four to five measurement steps to 30
lmol mol
1
, then raised to ambient again and allowed to stabilize
at the original values of Aand stomatal conductance. The CO
2
level was then raised to 600 and 800 lmol mol
1
to assess the
maximum rate of photosynthesis.
The activities of PEP carboxylase (PEPC), NADP-malic en-
zyme, NAD-malic enzyme, and PEP carboxykinase (PEPCK) were
assayed using a coupled enzyme assay on freshly extracted leaf
material. PEP carboxylase was assayed spectrophotometrically at
340 nm by coupling the production of OAA to NADH oxidation
via malate dehydrogenase (modified from Ashton et al., 1990). The
reaction mixture contained 50 mm Bicine (pH 8.0), 5 mM MgCl
2
,
2 mM DTT, 2 mM NaHCO
3
, 1 mm glucose 6-phosphate, 5 mM
PEP, 0.25 mM NADH, 2.5 units ml
1
malate dehydrogenase, and
enzyme extract. The reaction was initiated by the addition of PEP.
NAD-ME and NADP-ME were assayed by measuring the
formation of NADH and NADPH, respectively. The reaction
mixture for NADP-ME contained 50 mM TRIS-HCl (pH 8.2), 1
mM EDTA, 20 mM MgCl
2
, 2 mM DTT, 0.5 mM NADP
+
,5mM
Na-malate, and enzyme extract (modified from Kanai and
Edwards, 1973). The reaction was initiated by the addition of
malate. The reaction mixture for the assay of NAD-ME contained
25 mM HEPES (pH 7.2), 5 mM DTT, 0.2 mM EDTA, 2.5 mM
NAD
+
, 5 mM Na-malate, 8 mM (NH
4
)
2
SO
4
,75lM coenzyme A,
2 mM MnCl
2
,25lM NADH, and enzyme extract (modified from
Hatch and Kagawa, 1974;Hatch et al., 1982). The reaction was
initiated by the addition of MnCl
2
. Activity of PEPCK was
assayed in the carboxylating direction by measuring the nucleotide-
dependent production of OAA, coupled to NADH oxidation via
malate dehydrogenase. The reaction mixture contained 80 mM
MES (pH 6.7), 0.25 mM NADH, 5 mM DTT, 5 mM MnCl
2
,2mM
PEP, 2 mM ADP, 10 mM KHCO
3
, and 5 units ml
1
malate
dehydrogenase (modified from Walker et al., 1995). Interfering
PEPC activity was measured with PEP only, and the PEPCK
reaction was initiated with ADP. Activity of PEPCK was taken as
the change in absorbance after the addition of ADP (Edwards
et al., 1971).
All chemicals for enzyme assays were obtained from Sigma-
Aldrich, St Louis, Missouri USA.
Leaf anatomy and ultrastructure
Leaf tissue was sampled for anatomical observations from the
most recent, fully expanded, mature leaves. The internal anatomy
of leaves was assessed on leaf sections harvested from the middle
of the leaf (1 leaf plant
1
; three plants) and prepared for light and
TEM as described by Sage and Williams (1995). Images of leaf
cross-sections taken with a light microscope were used to measure
leaf thickness, % M, % BS, and the ratio of M-to-BS tissue. Leaf
thickness, % M, and % BS were determined from four separate
sections per leaf. Leaf thickness was measured using Image J
software (National Institutes of Health, Bethesda, MD, USA). The
percentage of the leaf cross-section covered by M or BS tissue was
determined by laying a grid of 100 random points over cross-
sections of images and calculating the proportion of points falling
on M or BS cells (Parkhurst, 1982;McKown and Dengler, 2007).
TEM images were prepared from three BS and three M cells from
each of three leaf sections from three separate plants. TEM images
were used to quantify the area and number of chloroplasts and
mitochondria per M and BS cell area using Image J software.
Interveinal distances were determined from five images of leaves
taken from three plants and cleared as described by McKown and
Dengler (2007). Interveinal distance was determined by laying
a grid of randomized points on an image and measuring the length
of the shortest line that connected two adjacent veins through one
of the randomized points. This was repeated ten times for each
image. The 10 values determined for an image were averaged to
produce an image value that was used in the statistical analysis of
the data.
In situ immunolocalization
The preparation of tissue samples for immunolocalization of
Rubisco, PEPC, and the P-subunit of the GDC complex follows
Marshall et al. (2007). Polyclonal antisera used were anti-Nicotiana
tabacum Rubisco (provided by NG Dengler, University of
Toronto, Canada), anti-maize PEPC (obtained from J Berry, State
Univeristy of New York, USA), and anti-pea GDC (P-subunit;
received from S Rawsthorne, John Innes Centre, Norwich, UK).
The cross-reactivity of each antibody was verified by running
control labelling experiments on cross-sections of paraffin-embed-
ded leaf tissues of Flaveria pringlei (C
3
), F. ramossisima (C
3
–C
4
),
and F. trinervia (C
4
, NADP-ME) in which the enzymes were
known to be expressed and accumulated in a tissue-specific manner
(Edwards et al., 2001).
Data analysis
Results were analysed with Sigmaplot version 11.0 using a one-
way analysis of variance and Tukey’s pairwise multiple compari-
son. Sample sizes varied between two and five for physiological
and enzyme assays, and between 9 and 15 for anatomical and
ultrastructural analysis. In the anatomical and ultrastructural
study, an individual microscope section was considered as the
sampling unit.
Results
Ecological habit
The four species listed for Euphorbia subgenus Chamaesyce
subsection Acutae by Webster et al. (1975) and Mayfield
C
2
photosynthesis in Euphorbia |3of13
at University of Toronto Library on April 1, 2011jxb.oxfordjournals.orgDownloaded from
(1991), are E. acuta,E. angusta,E. johnstonii, and E. lata.
Euphorbia acuta occurs on dry limestone uplands of the
Edwards plateau and semi-arid scrublands of Brewster,
Crane and Pecos Counties in Western Texas, USA (Fig. 1;
Correll and Johnston, 1979). Euphorbia angusta grows on
dry limestone outcrops in the Edwards plateau of west-
central Texas, while E. lata occurs on dry calcareous soils
and sandy plains of the western end of the Edwards plateau
and adjacent trans-Pecos region of Texas (Fig. 1;Correll
and Johnston, 1979). Euphorbia johnstonii is restricted to
calcareous soils and caliche outcrops in northern regions of
the Mexican states of Nuevo Leon and Tamulaipas (Mayfield,
1991). Euphorbia mesembryanthemifolia is widespread in
warm, seasonally dry, coastal regions of the Caribbean
basin and Gulf of Mexico, where it is found on sandy or
rocky areas affected by salt spray (Herndon, 1996). All
species are presumed to be photosynthetically active in the
summer, based by the presence of a robust, fleshy leaf
canopy observed for each species during the summer
collection periods. Flowering for each species is greatest in
mid-to-late summer, when monsoon rains provide periodic
precipitation.
Midday leaf temperatures of E. acuta peaked near 39 C
between 12–14 July 2010 (Fig. 2). Leaf and air temperatures
were usually similar, although on 13 July the mid-morning
leaf temperatures were 3–4 C above air temperature.
Maximum summer temperatures from 1971–2010 in Fort
Stockton, Texas, averaged 31.5 C for June, 33.5 C for
July, and 32.2 C for August (PRISM, 2010). Peak temper-
atures for July exceeded 36 C in 15 of the 40 years in the
1971–2010 record (PRISM, 2010). Comparison of the
results in Fig. 2 with the long-term climate records indicate
that the leaf and air temperatures observed for E. acuta
represent typical thermal profiles on average (13 July) to
warmer than average (14 July) summer days. From this, it is
concluded that leaf temperatures for E. acuta are above
30 C for much of the day during summer, and will
frequently peak above 35 C. Similar temperature profiles
probably exist for the other members of Euphorbia sub-
section Acutae in western Texas, given similarity in leaf size
and shapes (Fig. 1).
Gas exchange, enzyme activities, and carbon isotopes
Gas exchange characteristics were determined for
E. angusta,E. acuta, and E. lata. All three species have
similar maximum rates of net CO
2
assimilation in the low-
to-mid 30 lmol m
2
s
1
range (Table 1). Euphorbia lata had
a typical C
4
type response of net photosynthesis rate to
intercellular CO
2
concentration (the A/C
i
response), with
a steep initial slope, a sharp transition to CO
2
saturation,
and a CO
2
compensation point of photosynthesis (G)of
6lmol mol
1
(Fig. 3;Table 1). By comparison, E. angusta
and E. acuta had similarly shaped A/C
i
responses, similar
initial slopes, and a gradual transition to CO
2
saturation;
however, while E. angusta exhibited a C
3
-like Gof 69 lmol
mol
1
at 30 C, the Gof E. acuta was half this value, being
33 lmol mol
1
(Fig. 3;Table 1). Gvalues of 20–35 lmol
mol
1
at 30 C are characteristic of C
2
species (Monson and
Rawsthorne, 2000;Vogan et al., 2007). Water use efficiency
Fig. 1. Photographs of Texas Euphorbia species classified into subgenus Chamaesyce subsection Acutae by Webster et al. (1975). (A)
E. angusta; (B) E. acuta; (C) E. lata. (D) A habitat photograph showing limestone-derived soils and the dominant vegetation in the region
where E. acuta and E. lata were collected. Habitat photograph and photographs of E. acuta and E. lata taken in Crockett County, Texas,
along Live Oak Road, 1 km north of Interstate 10 (3043’ N, 10140’’ W). The E. angusta photo was taken along Highway 83 in Uvalde
County Texas, 5 km south of the junction with road 1051 (2928’ N, 9947’ W).
4of13 |Sage et al.
at University of Toronto Library on April 1, 2011jxb.oxfordjournals.orgDownloaded from
(WUE) was higher in E. lata than in E. angusta and E. acuta,
which had statistically similar WUE values (Table 1).
The activities of the major enzymes of the C
4
metabolic
cycle in E. angusta and E. acuta were similar and well below
those typically seen in C
4
plants (Table 1;Edwards and
Walker, 1983;Muhaidat et al., 2007). PEP carboxylase and
NADP-malic enzyme activity in E. angusta and E. acuta
were less than 8% of the values for these enzymes in E. lata
and E. mesembryanthemifolia. Activities of NAD-malic
enzyme and PEPCK were similar and very low in all four
species.
Carbon isotope ratios were C
3
-like in E. angusta and
E. acuta averaging between –26.0&and –28.5&(Table 1).
In E. johnstonii, the d
13
C value was –27.960.5 (mean 6SE,
n¼4). Euphorbia lata had a typical C
4
d
13
C value of 15.0&
(Table 1).
Leaf anatomy and ultrastructure
Leaves of E. angusta,E.acuta, and E.johnstonii are
unifacial with single layers of adaxial and abaxial palisade
parenchyma (Figs 4A, B, 5A). Leaves of E. lata and
E. mesembryanthemifolia are bifacial with a single layer of
adaxial palisade parenchyma and either two layers of
abaxial spongy mesophyll (E.lata;Fig. 4C) or non-
chlorenchymatous water-storage tissue and internal chlor-
enchymatous palisade parenchyma (E.mesembryanthemfo-
lia;Fig. 4D). The distance between veins is lowest in
E. angusta, intermediate in E. acuta and greatest in E. lata
and E. mesembryanthemifolia, being about 50% greater in
the latter two species than in E. angusta (Table 1).
Euphorbia angusta has about 40% greater relative area of
M tissue in cross-section than E. acuta and E. lata, while
relative BS area of E. angusta is 40–50% lower than the BS
area of E. acuta and E. lata (Table 1). The % BS area is
similar between E. angusta and E. mesembryanthemifolia,
Table 1. Summary of gas exchange, activities of C
4
enzymes, and leaf anatomical properties in Euphorbia species of the subgenus
Chamaesyce
Mean 6SE for physiological data. Mean 6SD for anatomical data. Letters indicate statistical differences between species at P<0.05. ND, not
determined. The maximum rate of net CO
2
assimilation (A
max
) was determined at elevated CO
2
, while water use efficiency was determined at
an ambient CO
2
of 350 lbar. Abbreviations: G,CO
2
compensation point of photosynthesis; A/C
i
, response of net CO
2
assimilation rate to
intercellular partial pressure of CO
2
; PEPC, PEP carboxylase; ME, malic enzyme; PEPCK, PEP carboxykinase.
Parameter Units nSpecies
E. angusta E. acuta E. lata E. mesembryanthemifolia
Physiological data
A
max
lmol m
2
s
1
2–3 37.661.8 a 32.962.5 a 35.461.6 a ND
Glmol mol
1
2–3 69.461.8 c 33.262.5 b 6.160.4 a ND
A/C
i
initial slope mol m
2
s
1
2–3 0.1760.02 a 0.1260.02 a 1.3360.4 b ND
Water use efficiency mmol mol
1
2–3 2.660.3 a 3.360.2 a 7.660.1 b ND
d
13
C&3–5 –28.560.4 a –26.060.2 b –15.060.5 c ND
PEPC activity lmol m
2
s
1
51361.1 a 16.260.9 a 190.261.7 b 310616 c
NADP-ME activity lmol m
2
s
1
5 3.063a 2.260.3 a 68.664.2 b 119.966.9 b
NAD-ME activity lmol m
2
s
1
51865a 1.560.5 a 1.760.2 a 3.860.7 a
PEPCK activity lmol m
2
s
1
526a2.461.6 a 4.862.9 a 1.760.9 a
Chlorophyll content lmol m
2
5630639 a 645661 ab 512615 a 806638 b
Anatomical data
Interveinal distance lm 15 80.0612.1 a 101.8617.7 b 124.4613.7 c 131.2620.4 c
Leaf thickness lm12196614 c 17668 b 145611 a 26569d
% Mesophyll – 12 41.164.4 c 29.764.9 b 28.465.3 b 18.863.0 a
% Bundle sheath – 12 14.962.7 a 24.061.7 b 29.861.2 c 14.360.3 a
Mesophyll/bundle sheath Tissue ratio 12 2.860.5 c 1.360.3 b 1.060.02 b 0.560.2 b
Mesophyll cell size lm
2
93876145 b 188655 a 1996118 a 287695 ab
Bundle sheath cell size lm
2
92476100 a 397680 b 4546134 b 4216115 b
Mesophyll:bundle sheath Cell area ratio 9 1.860.8 b 0.560.2 a 0.460.2 a 0.760.2 a
re, °CemperatuTe
Leaf 1
Leaf 2
M
M
M
7/12/10 7/13/10 7/14/10
35
40
30
35
25
15
20
10
A
ir
12 A
M
12 A
M
12 P
M
Date
Fig. 2. Leaf and air temperatures of two E. acuta leaves measured
on 12–14 July 2010. Leaves were from two separate plants
growing along Highway 18, 8 km north of the Interstate 10
overpass at Fort Stockton, Texas.
C
2
photosynthesis in Euphorbia |5of13
at University of Toronto Library on April 1, 2011jxb.oxfordjournals.orgDownloaded from
due to the presence of water storage tissue in E. mesem-
bryanthemifolia which reduced its relative BS area. Individual
McellsizeofE. angusta is about twice that of E. acuta and
E. lata, while E. angusta has the smallest BS size of all four
species (Table 1). These differences lead to a greater M-to-BS
tissue ratio in E. angusta. Notably, E. acuta has a similar
M-to-BS tissue ratio as E. lata and E. mesembryanthemifolia.
The BS cells of E.angusta contain numerous peripherally
situated chloroplasts, mitochondria, and peroxisomes, while
all BS organelles of E.lata and E.mesembryanthemifolia are
positioned along the centripetal cell wall adjacent to the
vascular tissue (Figs 4, 6, 7). Euphorbia acuta also has
a peripheral distribution of chloroplasts in both M and BS
cells (Figs 4B,6B), but the BS mitochondria and perox-
isomes are centripetally situated along the inner BS cell wall
(Figs 6B,7B). Chloroplasts and mitochondria in BS cells of
E.acuta are more numerous than in the BS cells of the
other three species (Table 2). Mesophyll chloroplasts are
fewer in E. lata and E. mesembryanthemifolia than the other
two species and E. lata has the fewest BS mitochondria of
the four species (Table 2). The organelle distribution within
M and BS cells in E.johnstonii mirrors that of E.acuta at
both the light level (Fig. 5) and the TEM level (data not
shown).
The size of the M chloroplasts does not differ greatly
between the four species, while the BS chloroplasts of
E. lata and E. mesembryanthemifolia are two to three times
larger than those of E. angusta and E. acuta (Table 2).
E. angusta and E. acuta have larger chloroplasts in the M
than the BS tissue, while the reverse was true for E. lata.
When the differences in chloroplast number are combined
with the size of individual chloroplasts, the % of the cell
area covered by chloroplasts is greater in the M than BS
cells of E. angusta and E. acuta (Table 2). The percentage
of cell area covered by chloroplasts tended to be greater
in M cells of E. angusta and E. acuta than E. lata and
E. mesembryanthemifolia, while in the BS cells, it was
greater in E. lata and E mesembryanthemifolia (Table 2).
Bundle sheath chloroplasts in E. lata and E. mesembryan-
themifolia had low granal stacking while granal stacking in
the chloroplasts of E.angusta and E.acuta BS and M cells
was common (Fig. 7).
Individual mitochondria in the BS cells of E. acuta are
two to three times larger than BS mitochondria of
E. angusta, E. lata, and E. mesembryanthemifolia (Figs 6,7;
Table 2). Mesophyll mitochondria are smallest in E. lata
and E. mesembryanthemifolia (Table 2). Mitochondria of all
species occupy a similar percentage of the M cell area, but
the combination of more and larger mitochondria in the BS
Fig. 4. Light micrographs of cross-sections through leaves of (A) E. angusta (Euan); (B) E.acuta (Euac); (C) E.lata (Eula); (D) E.
mesembryanthemifolia (Eume). Arrowheads denote centripetal localization of chloroplasts in Kranz bundle sheath. BS, Bundle sheath; M,
palisade mesophyll; S, spongy mesophyll; W, water storage cell. Bars¼20 lm.
40
30
20
10
0
C. angusta
C. acuta
C. lata
Intercellular CO2concentration, mol mol-1
0100200300400500
lm
-2 s-1
rate, momiilationCO2assNet
Fig. 3. The response of net CO
2
assimilation rate to intercellular
CO
2
concentration in E. angusta,E. acuta, and E. lata at 30 C.
Curves shown are representative of two (E. lata) or three individual
curves (E. acuta and E. angusta).
6of13 |Sage et al.
at University of Toronto Library on April 1, 2011jxb.oxfordjournals.orgDownloaded from
cells of E. acuta result in a percentage of the BS cell area
occupied by mitochondria that is 4–15 times greater than
that of E. angusta,E. lata, and E. mesembryanthemifolia
(Table 2).
The presence of brown stain demonstrated the occurrence
of Rubisco, PEPC, and GDC in the immunolocalizations of
the species studied here. Immunolocalization images
showed Rubisco to be scattered around the periphery of
the M tissue in E. angusta and E. acuta, and absent from
the M tissue of E. lata and E. mesembryanthemifolia (Fig.
8). The stain for Rubsico is peripherally located in the BS of
E. angusta, while in E. acuta it forms a pronounced layer
along the inner walls of the BS cells. In E. lata and
E.mesembryanthemifolia, Rubisco stain is present in the
inner half of the BS cells. PEPC stain is very faint in all leaf
cells of E. angusta (not shown) and E. acuta (Fig. 9A), and
is pronounced in the M cells of E. lata (Fig. 9B) and
E. mesembryanthemifolia (Fig. 9C).
The GDC stain is pronounced in M cells of E. angusta,
where it forms dark spots that correspond to individual
mitochondria (see arrows in Fig. 10A). By contrast, the
GDC stain is restricted to the inner region of the BS cells in
Table 2. Organelle size and number for four Euphorbia subgenus Chamaesyce species
Means 6SD. Letters indicate the statistical differences between species at P<0.05, n¼9.
Units Species
E. angusta E. acuta E. lata E. mesembryanthemifolia
Chloroplast number
Per mesophyll cell – 1664.7 b 12.863.6 b 5.962.3 a 6.862.4 a
Per bundle sheath cell – 7.662.1 a 20.762.9 c 9.763.2 a 13.962.8 b
Per mesophyll cell area lm
2
310
3
44.2614.6 b 67.0612.7 c 36.3618.0 a 25.362.6 a
Per bundle sheath cell area lm
2
310
3
32.568.3 a 54.5615.5 b 23.1610.1 a 35.1611.2 a
Mesophyll to bundle sheath ratio – 1.560.7 ab 1.460.5 ab 1.660.6 b 0.860.4 a
Chloroplast size
Mesophyll lm
2
7.361.7 ab 5.560.6 a 6.261.6 a 8.762.6 b
Bundle sheath lm
2
5.361.8 a 3.560.4 a 14.565.6 c 10.061.7 b
mesophyll:bundle sheath size ratio – 1.560.5 b 1.660.2 b 0.560.2 a 0.960.4 a
Chloroplast area per mesophyll area % 31.268.4 b 37.465.8 b 21.6610.1 ab 21.062.0 a
Chloroplast area per bundle sheath area % 17.166.8 a 18.764.9 a 29.366.5 b 34.7610.9 b
Total chloroplast area ratio, M to BS – 2.160.9 b 2.160.6 b 0.860.4 a 0.660.2 b
Mitochondria number
Per mesophyll cell – 10.168.5 b 4.663.6 ab 3.462.4 a 8.363.1 ab
Per bundle sheath cell – 7.063.2 ab 22.467.5 c 3.862.2 a 13.365.3 b
Per mesophyll cell area lm
2
310
3
24.6614.4 a 24.3618.9 a 21.8617.4 a 31.2615.0 a
Per bundle sheath cell area mm–2310–3 31.0614.6 b 60.3626.8 c 9.066.2 a 32.0611.2 b
Mesophyll to bundle sheath ratio – 1.261.4 a 0.460.1 a 1.961.3 a 1.261.1 a
Mitochondria size
Mesophyll lm
2
0.560.1 b 0.460.3 ab 0.260.2 a 0.260.1a
Bundle sheath lm
2
0.460.1 a 0.760.1 b 0.360.2 a 0.260.1 a
Mesophyll:bundle sheath size ratio – 1.560.7 b 0.660.3 a 0.960.6 ab 1.0360.5 ab
Mitochondria area per mesophyll area % 1.260.7 a 1.061.1 a 0.760.6 a 0.660.3 a
Mitochondria area per bundle sheath cell % 1.060.5 a 4.462.0 b 0.360.2 a 0.760.3 a
Total mitochondria area ratio, M to BS – 1.561.6 ab 0.260.2 a 2.461.9 b 1.060.5 ab
Fig. 5. Light micrographs of cross-sections through leaves of E. johnstonii (Eujo). Arrowheads mark centripetal location of chloroplasts in
bundle sheath cells. B, Bundle sheath; M, mesophyll. Bars¼20 lm.
C
2
photosynthesis in Euphorbia |7of13
at University of Toronto Library on April 1, 2011jxb.oxfordjournals.orgDownloaded from
E. lata and E. mesembryanthemifolia (Fig. 10C, D). Diffuse
staining was apparent in the mesophyll of E. lata, but it is
non-specific background staining and does not correspond
to individual mitochondria. In E. acuta, the GDC stain is
evident as a dark band along the inner BS wall. In this
region, individual mitochondria stand out as dark spots of
stained GDC (Fig. 10B). Immunolocalization of GDC for
E. johnstonii was also attempted and a dark-staining area
was identified only in the inner portion of the BS cells (data
not shown).
Discussion
In this study, evidence is presented that Euphorbia subgenus
Chamaesyce contains species that concentrate photorespired
CO
2
into the BS cells using the C
2
metabolic cycle.
Euphorbia acuta is a C
2
species based on its Gvalue of 33.2
lmol mol
1
, Kranz-like arrangement of enlarged isodiamet-
ric BS cells, higher density of chloroplasts along the
centripetal BS wall, and localization of GDC to BS cells.
Bundle sheath cells in E. acuta also have increased numbers
of mitochondria and larger mitochondria than the C
3
species E. angusta, and mitochondria are located centripe-
tally within BS cells. Increased size and number of BS
mitochondria are common in C
2
species, presumably
reflecting a need for greater GDC capacity to process all of
the photorespiratory metabolites produced by the leaf
(Monson and Rawsthorne, 2000). Euphorbia johnstonii also
exhibits a Kranz-like anatomy with GDC-positive staining
and enhanced aggregation of chloroplasts and other organ-
elles along the inner wall of the enlarged BS cells. Therefore,
E.johnstonii is probably a C
2
species as well, although
definitive confirmation of this will require measurements of
photosynthetic characteristics. In contrast to E.acuta,
E. angusta is clearly a C
3
species, based on its Gvalue of 69
lmol mol
1
,aC
3
carbon isotope ratio, lack of well-defined
and enlarged BS cells, low activity of C
4
-cycle enzymes, and
no evidence of chloroplast, mitochondria, and GDC
localization to the inner wall of the BS cells. Euphorbia lata
is a NADP-ME type of C
4
species based on a low Gvalue,
a carbon isotope ratio of –15.0&, high activities of PEPC
and NADP-malic enzyme, and the obvious presence of
Kranz anatomy. Euphorbia mesembryanthemifolia is also
a NADP-ME C
4
species, as shown by the high PEPC and
NADP-ME activities and Kranz anatomy. These results for
E. lata and E. mesembryanthemifolia, in combination with
prior work by Gutierrez et al. (1974), indicate all C
4
species
Fig. 6. Transmission electron micrographs illustrating bundle sheath cell structure in leaves of Euphorbia species. Arrows mark
mitochondria. (A) E.angusta; (B) E.acuta; (C) E.lata; (D) E.mesembryanthemifolia. B, bundle sheath; C, chloroplast; M, mesophyll; VT,
vascular tissue. Bars¼2lm.
8of13 |Sage et al.
at University of Toronto Library on April 1, 2011jxb.oxfordjournals.orgDownloaded from
in Euphorbia subgenus Chamaesyce are NADP-ME species.
Both E. lata and E. mesembryanthemifolia have low granal
stacking in the BS chloroplasts relative to M chloroplasts.
Chloroplast dimorphism with respect to thylakoid develop-
ment has been reported for other C
4
Euphorbia species (Kim
et al., 2000) and is typical of NADP-ME subtypes (Edwards
and Walker, 1983;Dengler and Nelson, 1999). Mitochon-
drial number and size is reduced in the two C
4
species
relative to E. acuta, presumably a result of the decreased
metabolic role of the mitochondria in leaves of NADP-ME
type C
4
species. Photorespiration is low in NADP-ME
subtypes due to high BS CO
2
and relatively low BS O
2
levels (Kanai and Edwards, 1999;Sage et al., 2011b).
Many C
2
species express a C
4
metabolic cycle of varying
strength. Edwards and Ku (1987) have called C
2
species
with a modest C
4
cycle Type II C
3
–C
4
intermediates, while
C
2
species lacking a C
4
cycle have been termed Type I C
3
–
C
4
intermediates. There is no evidence that E. acuta
operates a C
4
metabolic cycle thereby classifying it as a Type
IC
3
–C
4
intermediate. Euphorbia acuta has a C
3
-like carbon
isotope ratio and C
3
-level activities of PEPC, NADP-ME,
and NAD-ME. C
4
cycle activity is required to raise d
13
C
values in C
3
species above –23&to –21&; otherwise,
carbon isotope ratios reflect the discrimination of Rubisco
and are C
3
-like (von Caemmerer, 1992). It is also suspected
that E. johnstonii is a Type I C
3
–C
4
species given its C
3
isotopic ratio.
Phylogenetic data confirm that subsection Acutae is basal
to all the C
4
species in subgenus Chamaesyce and thus
support the hypothesis that photorespiratory CO
2
concen-
tration (C
2
photosynthesis) is a key intermediate stage for
the evolution of C
4
photosynthesis. Euphorbia angusta and
E. acuta branch at a node between the more basal C
3
species in the genus Euphorbia, and more distal C
4
species in
subgenus Chamaesyce (Steinmann and Porter, 2002;Yang
and Berry, 2007). With the placement of E. acuta between
C
3
and C
4
nodes in the Euphorbia phylogeny, the Euphorbia
subsection Acutae represents a fourth confirmed case where
C
2
photosynthesis appears in an evolutionary intermediate
position between C
3
and C
4
clades. Other confirmed cases
include Flaveria,Cleome,andMollugo (McKown et al.,
2005;Feodorova et al., 2010;Christin et al., 2011). These
multiple observations provide strong evidence that C
2
photosynthesis is an obligatory step for C
4
evolution.
However, the existence of numerous C
2
species that are
unrelated to C
4
clades demonstrates that C
2
photosynthesis
Figure 7. Transmission electron micrographs of chloroplasts and mitochondria in the centripetal regions of bundle sheath cells of
Euphorbia species. (A) E.angusta; (B) E.acuta; (C) E.lata; (D) E.mesembryanthemifolia. B, bundle sheath; C, chloroplast; M,
mitochondrion; P, peroxisome; VT, vascular tissue. Bars¼0.5 lm.
C
2
photosynthesis in Euphorbia |9of13
at University of Toronto Library on April 1, 2011jxb.oxfordjournals.orgDownloaded from
does not automatically lead to the C
4
condition (Hylton
et al., 1988, for Moricandia;Christin et al., 2011, for
Mollugo;Sage et al., 2011a). This, along with the ecological
success of C
2
species such as E. acuta, Mollugo verticillata,
and M. nudicaulis (Christin et al., 2011),demonstrates the
value of recognizing C
2
photosynthesis as a distinct photo-
synthetic adaptation in its own right. In light of this, it is
proposed that the term ‘C
3
–C
4
intermediate’ commonly
used for C
2
species be restricted in use to species with
intermediate traits that can be phylogeneticaly placed
between C
3
and C
4
clades. Such a convention could include
all species that are evolutionary intermediates, regardless of
whether they express C
2
photosynthesis.
The evolutionary transition from C
3
to C
4
photosynthesis
is marked by modifications in the characteristics of the leaf
tissues and cell structure (Dengler and Taylor, 2000).
Relative to C
3
species, the majority of species exhibiting C
4
photosynthesis have closer vein spacing, reduced M-to-BS
tissue ratio, and an asymmetric positioning of organelles in
the BS cells (Dengler and Taylor, 2000;Muhaidat et al.,
2007;Voznesenskaya et al., 2007). Using a phylogenetic
analysis, McKown and Dengler (2007) proposed a stepwise
acquisition of anatomical traits during the evolution of C
4
photosynthesis from C
3
ancestors in Flaveria. Initial de-
velopmental innovations leading to C
3
–C
4
intermediacy
include increases in BS chloroplast numbers, decreases in
the M-to-BS tissue ratio and increases in BS cell size. These
features possibly enabled subsequent C
4
evolution by
facilitating the acquisition of the C
4
metabolic steps
(Monson, 1999;Sage, 2004). Results from this study
demonstrate that a number of the critical traits associated
with the C
4
pathway in E. lata and E.mesembryantemifolia
(for example, enlarged BS size, reduced M cell size, and
asymmetric positioning of organelles) are also present in
E.acuta and E.johnstonii. Chloroplast and mitochondria
number as well as mitochondrial size are also enhanced in the
BS of E. acuta. As noted for Flaveria (McKown and Dengler,
2007), the C
4
-like traits in E.acuta occur in tandem with
changes in GDC localization but prior to expression of
PEPC in M tissue. The presence of these traits in E.acuta are
consistent with those reported in C
2
species in Cleome,
Fig. 8. Light micrographs illustrating in situ immunolocalization of Rubisco in leaves of Euphorbia species. Brown precipitate indicates
positive immunolabelling. Arrowheads denote chloroplasts in bundle sheath cells. (A) E.angusta; (B) E.acuta; (C) E.lata; (D)
E.mesembryanthemifolia. B, Bundle sheath; M, mesophyll. Bars¼20 lm.
Fig. 9. Light micrographs illustrating in situ immunolocalization of
PEPC in leaves of Euphorbia species. Brown precipitate indicates
positive immunolabelling. Arrowheads denote chloroplasts in
bundle sheath cells. (A) E.acuta; (B) E.lata; (C) E.mesembryan-
themifolia. B, Bundle sheath; M, mesophyll. Bars¼20 lm.
10 of 13 |Sage et al.
at University of Toronto Library on April 1, 2011jxb.oxfordjournals.orgDownloaded from
Flaveria,andMollugo (Marshall et al.,2007;McKown and
Dengler, 2007;Voznesenskaya et al.,2007;Christin et al.,
2011). Together, these examples provide strong support that
reduced M-to-BS tissue ratio, increased organelle number
and size in enlarged BS cells, and the polar distribution of BS
organelles are common intermediate steps in the evolution of
C
4
photosynthesis in the eudicots.
Modifications in vein patterning that led to increases in
vein density in C
4
species precede changes in BS organelle
content and cell size in Flaveria; it has thus been hypothe-
sized that changes in vein patterning may be a precondition
for the evolution of C
3
-C
4
intermediacy (McKown and
Dengler, 2007). Euphorbia angusta has a number of traits
which may facilitate the shift from the C
3
to C
2
and then
the C
4
pathway. Interveinal distance in E. angusta is less
than in E. acuta and either of the C
4
species, and the BS
tissue is enlarged relative to what is considered typical in C
3
plants. In a survey of 21 C
3
to C
4
lineages, Muhaidat et al.
(2007) observed that the C
3
species had a mean % area of
BS-like tissue of 8.1%, which is less than the 14.9% observed
here in E. angusta. Notably, chloroplast numbers on a BS
area basis in E. angusta are statistically similar to the values
in E. acuta and both C
4
species, indicating significant
involvement of the E. angusta BS cells in leaf photosynthe-
sis. The combination of close vein spacing, enlarged BS size,
and enough chloroplasts for significant photosynthetic
activity may provide the BS tissue of E. angusta with the
potential to carry the photorespiratory load of the leaf
following a mutation that knocks out GDC expression in
the M tissue. The knockout of M GDC expression is
proposed to be the key mutation that establishes photo-
respiratory CO
2
concentration and hence C
2
photosynthesis
in the leaf (Monson et al., 1984;Bauwe, 2011). To evaluate
whether E. angusta has increased potential to facilitate C
2
evolution, it would be useful to examine its relatives in the
more basal C
3
clades of Euphorbia.
In addition to supporting physiological, molecular, and
cell biology studies of C
4
origins, Euphorbia subgenus
Chamaesyce is well positioned to support studies address-
ing the ecological factors facilitating C
4
evolution. The
leading explanations for the origin of C
4
photosynthesis
postulate that the C
4
pathway arose from increasingly
sophisticated modifications that served to compensate for
high levels of photorespiration (Monson et al., 1984;
Ehleringer et al., 1991;Bauwe, 2011). The initial innova-
tions facilitated the refixation of photorespired CO
2
in the
BS, leading to an optimized C
2
pathway following loss of
mesophyll GDC expression (Monson and Rawsthorne,
2000). As such, it is hypothesized that the C
4
pathway
evolvedinhabitatswherephotorespiration represents
a large drag on C
3
photosynthesis. The ecological habitat
of the species studied here supports this hypothesis. The
semi-arid landscapes of western Texas and north-eastern
Mexico are most likely to be the location where C
4
photosynthesis evolved in Euphorbia, given the restriction
of species from subsection Acutae to this region. As shown
in Fig. 2, daily summer temperatures can climb well above
30 C, soil water can be episodically scarce, and the vapour
pressure difference between leaf and air is high. High VPD
would restrict stomatal conductance, thus aggravating
photorespiratory inhibitions. Elevated temperatures would
therefore favour high rates of photorespiration, particu-
larly at the lower CO
2
levels of recent geological time
(Sage, 2004).
Fig. 10. Light micrographs illustrating in situ immunolocalization of glycine decarboxylase in leaves of Euphorbia species. Arrows mark
mitochondria with positive dark-brown immunolabelling. (A) E.angusta; (B) E.acuta; (C) E.lata; (D) E.mesembryanthemifolia. B, Bundle
sheath; M, mesophyll. Bars¼20 lm.
C
2
photosynthesis in Euphorbia |11 of 13
at University of Toronto Library on April 1, 2011jxb.oxfordjournals.orgDownloaded from
Photorespiration was unlikely to be the only factor
promoting C
4
evolution, as indicated by the presence of C
4
-
like traits in the C
3
plant E. angusta, notably close vein
spacing. Close vein spacing may have arisen in C
3
plants in
response to very high evaporative demands, which occur in
hot, semi-arid to arid climates (Sage, 2004). High VPD may
favour close vein spacing in order to sustain water flux to
the M tissue under high evaporative demand. The presence
of a summer monsoon in the habitat of E. angusta also
appears to be critical, as it would provide enough soil
moisture to allow for summertime photosynthesis, even in
hot conditions promoting high photorespiration and tran-
spiration. Consistently, other lineages where C
2
and C
4
photosynthesis evolved also occur in hot, summer monsoon
climates with some level of water or salinity stress (Sage
et al., 2011a). C
2
and C
3
species of Flaveria and Helio-
tropium are active in summer monsoon habitats of west
Texas and subtropical Mexico (Frohlich, 1978;Powell,
1978;Vogan et al., 2007); Mollugo is generally a tropical
genus from monsoon-affected regions with hot summers
(Christin et al., 2011), and the close relatives of the C
4
clades in Cleome occur on hot, rocky or sandy soils of the
subtropics and tropics (Feodorova et al., 2010). Together,
these patterns present in Euphorbia subgenus Chamaesyce
and the other C
2
groups indicate that C
4
evolution is
favoured by environments where photorespiration in C
3
plants is high; however, selection pressures that compensate
for high evaporative demand establish the preliminary
conditions such as close vein spacing that subsequently
facilitate the origin of the C
2
pathway.
Acknowledgements
This research was supported by grants from the Natural
Science and Engineering Research Council of Canada
(NSERC) and the International Rice Research Institute C
4
Rice programme to RFS and TLS. This paper is dedicated
to the memory of Professor Grady Webster (1927–2005).
Grady Webster was a renowned expert on the Euphorbia-
ceae and an inspiration to TLS and RFS while they were
graduate students at UC Davis. Grady inspired this project
when he pointed out the evolutionary significance of
Euphorbia subsection Acutae to RFS in 2002. We are also
grateful to contributions made by Kathy Sault (sectioning),
Riyadh Muhaidat (immunolocalizations), Debbie Tam
(greenhouse care), and Greta Chu (tissue preparation). We
are especially grateful for the assistance of Debra Hansen,
who arranged for the collection of Euphorbia johnstonii.
References
Ashton AR, Burnell JN, Furbank RT, Jenkins CLD, Hatch MD.
1990. Enzymes of C4 photosynthesis. Methods in Plant Biochemistry
3, 39–70.
Bauwe H. 2011. Photorespiration: the bridge to C
4
photosynthesis.
In: Raghavendra AS, Sage RF, eds. C
4
photosynthesis and related
CO
2
concentrating mechanisms. Dordrecht, The Netherlands:
Springer, 81–108.
Bauwe H, Keerbert O, Bass}
uner R, Parnik T, Bass}
uner B. 1987.
Reassimilation of carbon dioxide by Flaveria (Asteraceae) species
representing different types of photosynthesis. Planta 172, 214–218.
Christin PA, Sage TL, Edwards EJ, Ogburn RM, Khoshravesh R,
Sage RF. 2011. Complex evolutionary transitions and the significance
of C
3
–C
4
intermediate forms of photosynthesis in Molluginaceae.
Evolution 10.1111/j.1558–5646.2010.01168.x.
Conway Morris S. 2003. Life’s solutions: inevitable humans in
a lonely universe. Cambridge: Cambridge University Press.
Correll DS, Johnston MC. 1979. Manual of the vascular plants of
Texas. Richardson, Texas, USA: University of Texas Press.
Dengler NG, Nelson T. 1999. Leaf structure and development in C
4
plants. In: Sage RF, Monson RK, eds. C
4
plant biology. San Diego:
Academic Press, 133–172.
Dengler NG, Taylor WC. 2000. Developmental aspects of C
4
photosynthesis. In: Leegood RC, Sharkey TD, von Caemmerer S, eds.
Photosynthesis: physiology and metabolism. Dordrecht, The
Netherlands: Kluwer, 471–495.
Edwards G, Franceschi VR, Ku MSB, Voznesenskaya EV,
Pyankov VI, Andreo CS. 2001. Compartmentation of photosynthesis
in cells and tissues of C
4
plants. Journal of Experimental Botany 52,
577–590.
Edwards GE, Kanai R, Black CC. 1971. Phosphoenolpyruvate
carboxykinase in leaves of certain plants which fix CO
2
by the C
4
-
dicarboxylic acid cycle of photosynthesis. Biochemical and Biophysical
Research Communications 45, 278–285.
Edwards GE, Ku MSB. 1987. Biochemistry of C
3
–C
4
intermediates.
In: Hatch MD, Boardman NK, eds. Biochemistry of plants:
photosynthesis, Vol. 14. San Diego: Academic Press, 275–325.
Edwards GE, Walker DA. 1983. C
3
,C
4
: mechanisms, and cellular
regulation, of photosynthesis. Oxford: Blackwell Scientific.
Ehleringer JR, Sage RF, Flanagan LB, Pearcy RW. 1991. Climate
change and the evolution of C
4
photosynthesis. Trends in Ecology and
Evolution 6, 95–99.
Feodorova TA, Voznesenskaya EV, Edwards GE, Roalson EH.
2010. Biogeographic patterns of diversification and the origins of C
4
in
Cleome (Cleomaceae). Systematic Botany 35, 1–16.
Frohlich MW. 1978. Systematics of Heliotropium section
Orthostachys in Mexico. PhD thesis. Cambridge, MA: Harvard
University.
Gutierrez M, Gracen VE, Edwards GE. 1974. Biochemical and
cytological relationships in C
4
plants. Planta 119, 279–300.
Hatch MD, Kagawa T. 1974. Activity, location, and role of NAD malic
enzyme in leaves with C
4
-pathway photosynthesis. Australian Journal
of Plant Physiology 1, 357–369.
Hatch MD, Tsuzuki M, Edwards GE. 1982. Determination of NAD
malic enzyme in leaves of C
4
plants. Effects of malate dehydrogenase
and other factors. Plant Physiology 69, 483–491.
Herndon A, Acevedo-Rodriguez P. 1996. Chamaesyce. Flora of St
John, US Virgin Islands. Memoirs of the New York Botanical Garden,
78, 205–208.
12 of 13 |Sage et al.
at University of Toronto Library on April 1, 2011jxb.oxfordjournals.orgDownloaded from
Hylton CM, Rawsthorne S, Smith AM, Jones DA, Woolhouse HW.
1988. Glycine decarboxylase is confined to the bundle sheath cells of
leaves of C
3
–C
4
intermediate species. Planta 175, 452–459.
Kanai R, Edwards G. 1999. The biochemistry of C
4
photosynthesis.
In: Sage RF, Monson RK, eds. C
4
plant biology. San Diego, USA:
Academic Press, 49–87.
Kanai R, Edwards GE. 1973. Separation of mesophyll protoplasts
and bundle sheath cells from maize laves for photosynthetic studies.
Plant Physiology 51, 1133–1137.
Kim I, Pak J, Seo B, Song S. 2000. Development of the Kranz
structure during leaf growth in C
4
Euphorbia maculata.Journal of Plant
Biology 43, 238–246.
Koutnik DL. 1987. A taxonomic revision of the Hawaiian species of
the genus. Chamaesyce (Euphorbiaceae). Allertonia 4, 331–388.
Mayfield M. 1991. Euphorbia johnstonii (Euphorbiaceae) a new
species from Tamaulipas. Mexico with notes on Euphorbia subsection
Acutae. Sida 14, 573–579.
Marshall DM, Muhaidat R, Brown NJ, Liu Z, Stanley S, Griffiths H,
Sage RF, Hibberd JM. 2007. Cleome, a genus closely related to
Arabidopsis, contains species spanning a developmental progression
from C
3
to C
4
photosynthesis. The Plant Journal 51, 886–896.
McKown AD, Dengler NG. 2007. Key innovations in the evolution of
Kranz anatomy and C
4
vein pattern in Flaveria (Asteraceae). American
Journal of Botany 94, 382–399.
McKown AD, Moncalvo JM, Dengler NG. 2005. Phylogeny of
Flaveria (Asteraceae) and inference of C
4
photosynthesis evolution.
American Journal of Botany 92, 1911–1928.
Monson RK. 1999. The origins of C
4
genes and evolutionary pattern
in the C
4
metabolic phenotype. In: Sage RF, Monson RK, eds. C
4
plant biology. San Diego CA: Academic Press, 377–410.
Monson RK, Edwards GE, Ku MSB. 1984. C
3
–C
4
intermediate
photosynthesis in plants. Bioscience 34, 563–574.
Monson RK, Moore Bd. 1989. On the significance of C
3
–C
4
intermediate photosynthesis to the evolution of C
4
photosynthesis.
Plant, Cell and Environment 12, 689–699.
Monson RK, Rawsthorne S. 2000. Carbon dioxide assimilation in
C
3
–C
4
intermediate plants. In: Leegood RC, Sharkey TD, von
Caemmerer S, eds. Photosynthesis: physiology and metabolism.
Dordrecht, The Netherlands: Kluwer Academic, 533–550.
Muhaidat R, Sage RF, Dengler NG. 2007. Diversity of Kranz
anatomy and biochemistry in C
4
eudicots. American Journal of Botany
94, 362–381.
Parkhurst DF. 1982. Stereological methods for measuring internal
leaf structure variables. American Journal of Botany 69, 31–39.
Powell AM. 1978. Sytematics of Flaveria (Flaveriinae-Asteraceae).
Annals of the Missouri Botanical Garden 65, 590–636.
PRISM. 2010. PRISM climate group. Oregon State University, http://
www.prismclimate.org, created 4 February 2004. Accessed 30
December 2010.
Sage RF. 2004. The evolution of C
4
photosynthesis. New Phytologist
161, 341–370.
Sage RF, Christin PA, Edwards EA. 2011 a. The lineages of C
4
photosynthesis on planet Earth. Journal of Experimental Botany 62,
(in press).
Sage RF, Kocacinar F, Kubien DK. 2011 b.C
4
photosynthesis and
temperature. In: Raghavendra AS, Sage RF, eds. C
4
photosynthesis
and related CO
2
concentrating mechanisms. Dordrecht, The
Netherlands: Springer, 161–195.
Sage RF, Li M, Monson RK. 1999. The taxonomic distribution of C
4
photosynthesis. In: Sage RF, Monson RK, eds. C
4
plant biology. San
Diego, CA: Academic Press, 551–584.
Sage TL, Williams EG. 1995. Structure, ultrastructure, and
histochemistry of the pollen-tube pathway in the milkweed Asclepias
exaltata L. Sexual Plant Reproduction 8, 257–265.
Steinmann VW, Porter JM. 2002. Phylogenetic relationships in
Euphorbieae (Euphorbiaceae) based on ITS and ndhF sequence data.
Annals of the Missouri Botanical Garden 89, 453–490.
Tolbert NE. 1997. The C
2
oxidative photosynthetic carbon cycle.
Annual Review of Plant Physiology and Plant Molecular Biology 48,
1–23.
Tropicos. 2010. Tropicos botanical information system at the
Missouri Botanical Garden: www.tropicos.org. Accessed 31
December 2010.
Vogan PJ, Frohlich MW, Sage RF. 2007. The functional significance
of C
3
–C
4
intermediate traits in Heliotropium L. (Boraginaceae): gas
exchange perspective. Plant, Cell and Environment 30, 1337–1345.
von Caemmerer S. 1989. A model of photosynthetic CO
2
assimilation and carbon-isotope discrimination in leaves of certain C
3
–
C
4
intermediates. Planta 178, 463–474.
von Caemmerer S. 1992. Carbon isotope discrimination in C
3
–C
4
intermediates. Plant, Cell and Environment 15, 1063–1072.
Walker RP, Trevanion SJ, Leegood RC. 1995.
Phosphoenolpyruvate carboxykinase from higher plants: purification
from cucumber and evidence of rapid proteolytic cleavage in extracts
from a range of plant tissues. Planta 196, 58–63.
Webster GL. 1994. Synopsis of the genera and suprageneric taxa of
Euphorbiaceae. Annals of the Missouri Botanical Garden 81, 33–144.
Webster GL, Brown WV, Smith BN. 1975. Systematics of
photosynthetic carbon fixation pathways in. Euphorbia. Taxon 24,
27–33.
Voznesenskaya EV, Koteyeva NK, Chuong SDX, Ivanova AN,
Barroca J, Craven LA, Edwards GE. 2007. Physiological,
anatomical and biochemical characterisation of photosynthetic types
in genus Cleome (Cleomaceae). Functional Plant Biology 34,
247–267.
Yang Y, Berry PE. 2007. Phylogenetics and evolution of Euphorbia
subgenus Chamaesyce (Euphorbiaceae). Botany and plant biology
2007 joint congress proceedings. Abstract 1963.
C
2
photosynthesis in Euphorbia |13 of 13
at University of Toronto Library on April 1, 2011jxb.oxfordjournals.orgDownloaded from
