Content uploaded by Olivier Leroux
Author content
All content in this area was uploaded by Olivier Leroux
Content may be subject to copyright.
INVITED REVIEW
Collenchyma: a versatile mechanical tissue with dynamic cell walls
Olivier Leroux1,2,*
1
Botany and Plant Science and Ryan Institute, School of Natural Sciences, National University of Ireland Galway,
University Road, Galway, Ireland and
2
Pteridology, Department of Biology, Ghent University, KL Ledeganckstraat 35,
B-9000 Ghent, Belgium
* E-mail: Olivier.Leroux@UGent.be
Received: 29 April 2012 Returned for revision: 11 June 2012 Accepted: 6 July 2012 Published electronically: 29 August 2012
†Background Collenchyma has remained in the shadow of commercially exploited mechanical tissues such as
wood and fibres, and therefore has received little attention since it was first described. However, collenchyma
is highly dynamic, especially compared with sclerenchyma. It is the main supporting tissue of growing organs
with walls thickening during and after elongation. In older organs, collenchyma may become more rigid due
to changes in cell wall composition or may undergo sclerification through lignification of newly deposited
cell wall material. While much is known about the systematic and organographic distribution of collenchyma,
there is rather less information regarding the molecular architecture and properties of its cell walls.
†Scope and conclusions This review summarizes several aspects that have not previously been extensively dis-
cussed including the origin of the term ‘collenchyma’ and the history of its typology. As the cell walls of col-
lenchyma largely determine the dynamic characteristics of this tissue, I summarize the current state of
knowledge regarding their structure and molecular composition. Unfortunately, to date, detailed studies specif-
ically focusing on collenchyma cell walls have not been undertaken. However, generating a more detailed under-
standing of the structural and compositional modifications associated with the transition from plastic to elastic
collenchyma cell wall properties is likely to provide significant insights into how specific configurations of
cell wall polymers result in specific functional properties. This approach, focusing on architecture and functional
properties, is likely to provide improved clarity on the controversial definition of collenchyma.
Key words: Collenchyma, histology, plant anatomy, mechanical tissue, plant cell wall, primary and secondary
cell walls, plant biomechanics.
INTRODUCTION
The emergence of mechanical tissues was a key innovation in
the evolution of land plants and a prerequisite for the appear-
ance of large terrestrial species. By the middle Devonian,
many plant species developed a hypodermal sterome consist-
ing of heavily thickened sclerenchyma cells (Rowe and
Speck, 2004). Biomechanical investigations indicated that
the sterome significantly contributed to the stiffness of
stems and allowed them to reach great heights and evolve
diverse branched architectures compared with plants with
turgor-based support systems (Rowe and Speck, 2004).
While sclerenchyma tissues confer rigidity and tensile and
shear strength to many plant organs (Niklas, 1992;Jarvis,
2007), their properties are incapable of supporting growing
plant organs which undergo extensive turgor-driven elong-
ation. Indeed, sclerified tissues generally consist of dead
cells with non-extensible rigid cell walls which are unable
to undergo mitotic divisions. In small slowly growing plant
organs, turgor pressure generated in parenchyma cells may
provide sufficient support, but many plant stems grow fast
and are fragile, and therefore they cannot fully rely on
turgor pressure for support. Partly because they are non-
sclerified and only minimally lignified, young plant tissues
are preferentially selected by grazing animals and plant
bugs. For this reason, supporting tissues in these regions
should not be terminally differentiated but capable of under-
going wound healing or tissue regeneration. Moreover, as sec-
ondary growth increases the diameter of stems, the ability to
transdifferentiate and initiate periderm formation is an add-
itional advantage. Finally, young above-ground organs are
photosynthetic and the reinforcing tissues should ideally be
as translucent as possible to enable light to reach the chloro-
plasts in tissues deeper in the plant. To meet most of the
above-mentioned requirements, and to provide support
without preventing cell elongation, many plants – eudicotyle-
dons in particular – develop collenchyma: a mechanical
tissue composed of elongated cells with thick flexible and
translucent cell walls and with protoplasts capable of resum-
ing meristematic activity.
In this review, I describe the remarkable origin of the term
‘collenchyma’ and discuss some of the controversies asso-
ciated with the description of this tissue. In contrast to scleri-
fied mechanical tissues such as wood and fibres, which are
economically important raw materials, collenchyma tissues
have received little attention. Consequently, a clear definition
of collenchyma has never been given. It is not surprising
that this has resulted in some confusion with respect to its des-
ignation in certain cases. However, it needs to be highlighted
that one of the factors that makes collenchyma so unique is
also the reason why it is difficult to define: the dynamic
nature of its cell walls.
#The Author 2012. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved.
For Permissions, please email: journals.permissions@oup.com
Annals of Botany 110: 1083–1098, 2012
doi:10.1093/aob/mcs186, available online at www.aob.oxfordjournals.org
by guest on January 4, 2013http://aob.oxfordjournals.org/Downloaded from
HISTORY
Several textbooks (e.g. Esau, 1965;Fahn, 1990) report that
‘collenchyma’ is derived from the Greek word ‘ko
´lla’,
meaning glue and referring to the thick, glistening appearance
of unstained collenchyma cell walls. Although this explanation
seems perfectly acceptable, confusion exists because the first
use of ‘collenchyma’ was by Link (1837) who used it to de-
scribe the sticky substance on Bletia (Orchidaceae, monocots)
pollen. Two years later, in an anatomical survey of Cactaceae
(eudicots), Schleiden (1839) criticized Link’s (1837) excessive
nomenclature and stated mockingly that the term ‘collen-
chyma’ could have more easily been used to describe elon-
gated sub-epidermal cells with unevenly thickened cells.
Although Schleiden (1839) himself used ‘a
¨ussere Rinden-
lage’ or ‘Zellen der a
¨ussere Rindenschicht’ rather than ‘collen-
chyma’, the term seems to have stuck as a way to describe
elongated and thickened sub-epidermal cells similarly to cur-
rently accepted usage. Others such as Meyen (1830) used
‘prosenchyma’ to describe elongated cells with tapering
ends, without distinguishing between vascular/ground tissue
and even between sclerenchyma-like and collenchyma-like
tissues. Common usage of ‘collenchyma’ can perhaps be
attributed to Harting (1844) as he repetitively used ‘collen-
chyma’ sensu Schleiden in his anatomical survey of annual
dicotyledonous angiosperms. French and English translations
of his work soon followed (Giltay, 1882), spreading the new
definition or appropriation of ‘collenchyma’. That collen-
chyma was not in common use in the mid-19th century is
perhaps suggested by von Mohl (1844) who described collen-
chyma tissues as ‘jelly-like subepidermal cells’ adding paren-
thetically ‘the so-called collenchyma cells’. By the end of the
19th century, the term ‘collenchyma’ was incorporated in
some prominent and influential plant anatomy text books and
publications (e.g. Sachs, 1868;de Bary, 1877;Ambronn,
1881;Giltay 1882;van Tieghem, 1886–1888) and became
more widely accepted.
GENERAL MORPHOLOGY AND ONTOGENY
The three most characteristic morphological features of collen-
chyma are (i) their axially elongated cells; (2) their cell wall
thickenings; and (3) their living protoplasts (Fig. 1A–D).
During elongation, collenchyma cells do not divide as much
as the surrounding parenchyma cells, which explains their pro-
senchymatic nature. However, cell size and shape still can vary
from short isodiametric and prismatic cells to long, fibre-like
cells with tapering ends. The latter may even reach lengths
of up to 2.5mminHeracleum sphondylium (Apiaceae, eudi-
cots) (Majumdar and Preston, 1941). In some cases, transverse
divisions take place after or during elongation, and the result-
ing daughter cells often remain together enclosed by a shared
cell wall derived from the mother cell, giving it the appearance
of a septate fibre with non-thickened cross walls (Fig. 1D).
Nonetheless, collenchyma shares more morphological and
physical characteristics with parenchyma tissues, and therefore
intermediate types are not uncommon. The similarities
between both tissues even led several researchers to categorize
collenchyma as thick-walled parenchyma (e.g. de Bary, 1877).
Collenchyma and parenchyma cell walls both have the ability
to stretch and/or grow during differentiation, but in the case of
collenchyma the walls thicken throughout elongation and often
post-elongation (Jarvis, 2007). Cell wall material is generally
not distributed equally so that most collenchyma cells have ir-
regular thickenings (see Histological typology). Similarly to
parenchyma, collenchyma cells have living protoplasts, essen-
tial for controlling the hydration state of the cell wall, but also
to enable transdifferentiation and cell wall thickening and
modification. Many textbooks (e.g. Esau, 1965;Fahn, 1990)
mention that chloroplasts are present in collenchyma, but in
typical collenchyma tissue with a clear mechanical function,
chloroplasts are rarely found (Evert, 2006). However, to
allow photosynthesis, collenchyma cell walls are generally
translucent, enabling light to be transmitted to the chloroplasts
in tissues below.
Controversy remains regarding the ontogeny of collenchyma
as it has been the focus of very few studies (Ambronn, 1881;
van Wisselingh, 1882;Esau, 1936;Majumdar, 1941). Acc-
ording to Esau (1936), collenchyma of celery (Apium graveo-
lens, Apiaceae, eudicots) originates in the ground meristem
close to or against the protoderm. Periclinal divisions initially
predominate but are soon followed by anticlinal longitudinal
sections. As divisions rapidly follow each other, cells
enlarge only moderately, appearing smaller than the surround-
ing ground meristem cells. The rapid succession of divisions
generally prevents the formation of intercellular spaces,
which are numerous in the ground tissue at that stage.
Ambronn (1881), on the other hand, observed that in the
Apiaceae collenchyma and vascular tissue arise from the
same procambium strand, while in most other families both
tissues arise independently from each other. van Wisselingh
(1882), who did not study Apiaceae, never found a common
origin for collenchyma and vascular tissue in any of the
species he studied (including Aucuba,Euonymus and
Lamium). A later investigation of Heracleum (Majumdar,
1941) failed to provide further clarity as it was reported that
the inner parts of each collenchyma strand are derived in the
earliest stages from the same meristem as the vascular
bundles, whereas the outer parts are derived from the ground
meristem. It needs to be noted that Esau (1936),Majumdar
(1941) and Ambronn (1881) did not study the same species,
and variation at species level may occur. Esau (1936) also
studied the ontogeny and origin of collenchyma tissue asso-
ciated with the vascular bundles in celery and showed that
they are composed of phloem parenchyma cells that have
enlarged and thickened their cell walls subsequent to the oblit-
eration of sieve tubes and companion cells (Esau, 1936).
SYSTEMATIC AND ORGANOGRAPHIC
DISTRIBUTION IN THE PLANT
Position in the plant
Collenchyma is a supporting tissue characteristic of the
growing organs of many herbaceous and woody plants, and
it is also found in stems and leaves of mature herbaceous
plants, including those that are only slightly modified by sec-
ondary growth. Although the localization of collenchyma has
been described by many authors, only Duchaigne (1955)
proposed a typology which has been adopted here (Table 1).
Leroux —Collenchyma: a review1084
by guest on January 4, 2013http://aob.oxfordjournals.org/Downloaded from
In stems and petioles, collenchyma typically occurs in a per-
ipheral position and can be found immediately beneath the epi-
dermis or separated from it by one to several layers of
parenchyma. If collenchyma is located adjacent to the epider-
mis, its inner tangential walls may be collenchymatously
thickened, or in some cases all epidermal walls may develop
thickenings. It is not uncommon to observe one cell layer of
collenchyma under the epidermis where only the cell wall
facing the epidermis is thickened. Collenchyma can occur as
a continuous peripheral layer (Fig. 2A), but may occasionally
be interspersed by intercellular space-rich and/or parenchy-
matous tissues opposite stomata. In other cases, collenchyma
is organized in discrete axial strands, and in stems or petioles
with protruding ribs it is usually well developed in the ribs
(Fig. 2B). Apart from its peripheral distribution, collenchyma
is sometimes associated with vascular bundles (fascicular col-
lenchyma), occurring at the phloem side (supracribral)
(Fig. 2C), xylem side (infraxylary) or surrounding the vascular
bundle completely (circumfascicular). Whereas most research-
ers recognize both peripheral and fascicular collenchyma,
Esau (1965) suggested that only cells in a peripheral location
in the plant should be called collenchyma. As mentioned
earlier, celery collenchyma bundle caps differ in their on-
togeny (Esau, 1936). Other observations such as differences
in their biomechanical properties (Esau, 1936) and response
to boron deficiency (Spurr, 1957) led Esau (1936) to conclude
that this tissue should be referred to as ‘collenchymatous’, a
term she also suggested to apply to any parenchyma in a
BA
D
C
FIG. 1. General morphology of celery collenchyma (Apium graveolens, eudicot, Apiaceae). (A) Transverse vibratome-cut section of a fresh petiole triple-stained
with acridine red, chrysoidine and astra blue showing collenchyma strands in prominent abaxial ribs. Vascular bundles are positioned opposite the collenchyma
strands. (B) Detail of a collenchyma strand indicated in A. (C) Transverse section of a resin-embedded petiole stained with toluidine blue showing that collen-
chyma accompanies the vascular bundles at the phloem side. Note that dehydration, which is required for resin embedding, resulted in a decreased thickness of
the collenchyma cell walls. (D) Longitudinal resin section stained with toluidine blue showing elongated collenchyma cells and isodiametric ground parenchyma
and epidermis cells. Note that most collenchyma cells are septate with thin cross-walls (inset, arrowhead). Abbreviations: c, collenchyma; p, parenchyma; e,
epidermis; ph, phloem; x, xylem. Scale bars: (A) ¼500 mm; B –D, D inset) ¼100 mm.
Leroux —Collenchyma: a review 1085
by guest on January 4, 2013http://aob.oxfordjournals.org/Downloaded from
non-peripheral position resembling collenchyma (Esau, 1965).
Although this logic might be acceptable, few have adopted it.
Therefore, I refer to the collenchymatous bundle caps as
collenchyma as they are composed of elongated cells with
collenchymatous thickenings.
In the lamina, collenchyma occurs in the ribs associated
with the major veins where it can be found under the epidermis
or as a cap at the phloem side of the vascular bundle, and/or
along the leaf margins. Some leaves, such as these of
Robinia pseudoacacia (Fabaceae, eudicots), have the ability
to move due to the presence of joint-like thickenings at the
base of the petiole. These structures, called pulvini, can
contain a central perivascular collenchyma ring surrounded
by cortical motor cells that swell asymmetrically to bend the
petiole (Moysset and Simon, 1991). Sclerenchyma is also
often replaced by collenchyma at the transition from blade to
sheath in grass leaves (Dayanandan et al., 1976,1977;Paiva
and Machado, 2003;Evert, 2006).
Collenchyma has been reported in roots (Kroemer, 1903;
Ba
¨secke, 1908;von Alten, 1909;Turner, 1934;van Fleet,
1946) and, although this appears anomalous as roots are un-
likely to require the type of support that collenchyma offers,
von Guttenberg (1940) and van Fleet (1950) highlighted that
collenchyma was especially apparent in aerial roots.
Systematic distribution
Collenchyma is most commonly observed in eudicots (for
an exhaustive overview, I refer to Metcalfe and Chalk, 1950,
1979). Interestingly, collenchyma is absent in stems and/or
leaves of many of the ferns and monocots (grasses, including
cereals) that develop sclerenchyma early (Falkenberg, 1876;
Giltay, 1882;Metcalfe and Chalk, 1979). Tissues with
similar properties, either in appearance or in function, have
been reported to occur in representatives of other plant
groups. However, some of these tissues have not been
studied in detail and, while they may share some characteris-
tics, it is unclear if they are homologous to collenchyma
described from angiosperms. In bryophytes, collenchyma-like
cells have been reported in Dendroligotrichum (Polytricha-
ceae, mosses) (Scheirer, 1977), and Physcomitrium collenchy-
matum (Funariaceae, mosses) was categorized on the basis of
the collenchymatous nature of the exothecial cells of its cap-
sules (Gier, 1955). However, Crum and Anderson (1964) did
not observe these thickenings in a more mature sample and
concluded that they are an unimportant expression of develop-
ment. Rolleri and Prada (2007) observed collenchyma tissue in
the lycophyte Isoe
¨tes (Isoetaceae, lycophytes) but only pro-
vided drawings and no photographic evidence. In ferns, collen-
chyma has only been infrequently reported. Russow (1872)
reported collenchyma in the petiole of the eusporangiate fern
Marattia (Marattiaceae), but this observation was not comple-
mented with drawings. In Equisetum (Equisetaceae), strength-
ening tissue under the ridges has been described as (annular)
collenchyma by some authors (e.g. Hauke, 1963;Brown,
1976), while others referred to it as sclerenchyma (e.g.
de Bary, 1877;Ogura, 1972;Johnson, 1933;Sørensen et al.,
2008). To avoid confusion some preferred more neutral
terms such as hypodermis (Brown, 1976), hypodermal
sterome (Gierlinger et al., 2008) or strengthening tissue
(Spatz et al., 1998;Speck et al., 1998). In the more advanced
leptosporangiate ferns, collenchymatous tissues have been
observed in Asplenium rutifolium (Aspleniaceae) (O. Leroux
et al., unpubl. res.). This fern contains annular collenchyma-
tous tissues, which, at maturity, sclerify and become impreg-
nated with brown phenolic compounds. Chaerle and Viane
(2004) described false veins in Asplenium (Aspleniaceae),
composed of moderately thickened annular collenchyma
cells. Nayar and Bajpai (1970) and Nayar (1965) reported
collenchyma-like thickenings in the corners of wing cells in
the prothalli of Hypodematium crenatum (Hypodematiaceae)
and drynarioid (Polypodiaceae) ferns, respectively. Unfortu-
nately, neither study provided photographic evidence so their
observations are questionable. Moreover, some reports,
including Alston (1956) who stated that collenchyma is well
developed in the black stipes of Adiantaceae, incorrectly
TABLE 1. Distribution and histological types of collenchyma
Position of collenchyma in plant stems and petioles
1. Peripheral collenchyma (‘collenchyme pe
´riphe
´rique ou cortical’,
Duchaigne, 1955): immediately beneath the epidermis or separated from
it by one or more layers of parenchyma.
(a) Continuous collenchyma (‘cylindre continu’, Duchaigne, 1955):
occurring as a continuous layer (although parenchymatous interruptions
can occur below the stomata).
(b) Strand collenchyma (‘cordon distinct’, Duchaigne, 1955):
occurring as axial strands separated from one another by parenchyma,
often in externally visible stem ridges.
2. Fascicular collenchyma* (‘collenchyme profound ou fasciculaire’,
Duchaigne, 1955; ‘perivascular collenchyma’, Metcalfe, 1979).
(a) Supracribral (‘supralibe
´riens’, Duchaigne, 1955): bordering the
vascular bundle at the phloem pole.
(b) Infraxylary (‘infralignieux’, Duchaigne, 1955): bordering the
vascular bundle at the xylem pole.
(c) Circumfascicular (‘circumfasciculaire’, Duchaigne, 1955):
completely surrounding the vascular bundle.
Histological types of collenchyma
1. Angular collenchyma (Esau, 1965;Metcalfe, 1979;Mauseth, 1988;
Fahn, 1990) (syn. ‘Eckencollenchym’, Mu
¨ller 1890; ‘collenchyme
angulaire’, Duchaigne, 1955): extra wall material is deposited on the
vertical walls where cells meet.
2. Tangential collenchyma (Metcalfe, 1979) (syn. ‘Plattencollenchym’,
Mu
¨ller 1890; ‘collenchyme tangentiel’, Duchaigne, 1955; ‘lamellar
collenchyma’, Esau, 1965;Mauseth, 1988;Fahn, 1990): thickenings
mainly located on the inner and outer tangential cell walls.
3. Annular collenchymas
†
(Metcalfe, 1979;Mauseth, 1988) (syn.
‘collenchyma annulaire’, Duchaigne, 1955): uniformly thickened cell
walls.
4. Lacunar collenchyma (Esau, 1965;Mauseth, 1988;Fahn, 1990)
(‘Lu
¨ckencollenchym’, Mu
¨ller, 1890; ‘Lacunate collenchyma’, Metcalfe,
1979): walls facing the intercellular spaces are thickened.
5. Collenchymatous thickenings (Esau, 1936,1965): collenchyma-like
cell wall thickenings which cannot be categorized in the four types
mentioned above [e.g. thickened radial cell walls of sub-epidermal cells
in Mamillaria magnimamma (Mauseth, 1988) or epidermal cell walls
with thickened inner tangential walls]. By using this term it is implied
neither that the cells are prosenchymatous, nor that they contribute to the
mechanical support of the organs in which they occur.
* Some authors do not recognize perivascular collenchyma and refer to
this tissue as ‘collenchymatous tissue’.
†
The distinction between angular and annular collenchyma is often
difficult, especially when massive thickening occurs, causing the lumen to
lose its angular appearance.
Leroux —Collenchyma: a review1086
by guest on January 4, 2013http://aob.oxfordjournals.org/Downloaded from
attributed collenchyma to ferns. In Gymnosperms, collen-
chyma cells have been reported in the leaves of Chigua restre-
poi (Zamiaceae) (Stevenson et al., 1996) and Abies grandis
(Pinaceae) (Larsen, 1927). Although some reports were sup-
ported with microphotographs, most are unclear. It is yet to
be determined if these collenchyma-like tissues in non-
flowering plants are homologous or analogous to the collen-
chyma tissues commonly found in flowering plants.
HISTOLOGICAL TYPOLOGY
Since the late 19th century collenchyma tissues have received
more attention and several typologies have been presented. As
different names were often applied to the same tissue type, I
have adopted the typology of Metcalfe (1979) and mention
some alternative names reported in other publications
(Mu
¨ller, 1890;Duchaigne, 1955) or in commonly used plant
anatomy text books in which collenchyma is extensively dis-
cussed (Esau, 1965;Mauseth, 1988;Fahn, 1990; Table 1).
In the section on the history of the term collenchyma I men-
tioned that by the end of the 19th century, ‘collenchyma’ was
included in most plant anatomy text books. However, in Sachs’
popular ‘Lehrbuch der Botanik’(Sachs, 1870), ‘collenchyma’
was mentioned solely in the figure legend for a drawing of a
transverse section through a Begonia (Begoniaceae, eudicots)
petiole, reporting that ‘collenchyma, adjoining the epidermis,
consists of cell thickenings where three cells adjoin each
other’. In later editions Sachs promoted this description to
the main text, further explaining that collenchyma cells are
prosenchymatous, but different types were not distinguished.
Vesque (1876) defined ‘typical’ collenchyma as a prosenchy-
matous tissue devoid of intercellular spaces with thickenings
most apparent in the cell corners. He described this type as
‘convex’ collenchyma and distinguished it from ‘concave’ col-
lenchyma, with the latter type giving the lumen a rather
rounded appearance. de Bary (1877) described collenchyma
in more detail, as a specialized type of thick-walled paren-
chyma, reproducing Sachs’ image of Begonia collenchyma.
Although he reported similar patterns to Vesque (1876),he
did not distinguish different types. Haberlandt (1879) pro-
posed a different typology by discriminating ‘provisorisches
collenchymgewebe’ from ‘dauercollenchym’, with the latter
consisting of cells with thickenings mainly located in the
cell corners, and the first with all walls moderately thickened.
In a comprehensive study, Giltay (1882) reported many differ-
ent patterns of thickening in collenchyma tissues, but did not
DA
E
B
FC
POSITION IN THE STEM HISTOLOGICAL TYPES
FIG. 2 . Collenchyma diversity: position in the stem (A–C) and histological types (D – F). Vibratome sections triple-stained with acridine red, chrysoidine and
astra blue. (A) Coprosma repens (Rubiaceae, eudicots) with a continuous peripheral layer of collenchyma. (B, C) Levisticum officinale (Apiaceae, eudicots) with
collenchyma in the ribs (B) and at the phloem side of the vascular bundles (C). (D) Angular collenchyma in Plectranthus fruticosus (Lamiaceae, eudicots). Note
the sub-epidermal periderm tissue. (E) Intermediate type between tangential and lacunar collenchyma in Geranium sobolifolium (Geraniaceae, eudicots). Note the
many intercellular spaces (arrows). (F) Peperomia sp. (Piperaceae, basal angiosperms) with annular collenchyma. Abbreviations: c, collenchyma; p, parenchyma;
pe, periderm; ph, phloem. Scale bars ¼50 mm.
Leroux —Collenchyma: a review 1087
by guest on January 4, 2013http://aob.oxfordjournals.org/Downloaded from
propose a typology. Instead, he highlighted that collenchyma
displays a natural gradient of shape and form towards both par-
enchyma and sclerenchyma.
The first exhaustive overview of the different types of col-
lenchyma was published by Mu
¨ller (1890). He distinguished
several types based on the pattern of cellular thickening:
‘Eckencollenchym’ (angular collenchyma) with more pro-
nounced wall thickening in the cell corners; ‘Lu
¨ckencollen-
chym’ (lacunar or lacunate collenchyma) with only that
portion of the wall thickened which borders intercellular
spaces; ‘Bastcollenchym’ (‘bast collenchyma’), characterized
by cells grouped in sub-epidermal strands with no intercellular
spaces and cells thickened all around; ‘Knorpelcollenchym’
(‘cartilage collenchyma’), with walls thickened strongly all
around and with a sharply distinguishable inner lamella
giving the tissue the appearance of a transverse section of car-
tilage, with separate tubes imbedded in a homogeneous matrix;
‘Plattencollenchym’ (tangential, plate or lamellar collen-
chyma), with the thickenings on the tangential walls;
‘Metacollenchym’, formed very late in the differentiation of
organs, by the obliteration of primary phloem and xylem
cells (called ‘keratenchym’ by Wigand, 1863) and; finally, a
type which resembles sclerenchyma in shape, ‘Protoscleren-
chym’, a transitionary collenchymatous phase prior to the de-
velopment of sclerenchyma (cf. Haberlandt’s ‘Provisorisches
Collenchym’). Later, Duchaigne (1955) simplified earlier clas-
sifications by recognizing only the three types of collenchyma
which are still distinguished in most contemporary plant
anatomy text books (e.g. Esau, 1965;Fahn, 1990;Mauseth,
1988;Dickison, 2000;Beck, 2005;Evert, 2006). The first
type, ‘angular collenchyma’ (‘collenchyme angulaire’)
(Figs 2D and 3A), is the common, classical type of collen-
chyma where the cell corners appear more heavily thickened.
This type is seen most as sub-epidermal tissue in many
stems and petioles of herbaceous dicots. The second type, ‘tan-
gential collenchyma’ (‘collenchyme tangential’) (Figs 2E and
3B), also known as lamellar or plate collenchyma, is character-
ized by thickening of the inner and outer tangential cell walls.
I favour the term ‘tangential collenchyma’ as it best suits the
actual distribution of the thickenings and avoids confusion
with the lamellar structure of collenchyma cell walls (see
Cell wall structure). Finally, the third type, ‘annular collen-
chyma’ (‘collenchyme annulaire’) (Figs 2F and 3C), is distin-
guished by having uniformly thickened walls. Although these
types show clear-cut differences, in reality there appears to be
a continuum, and separation of these types is not always clear.
For example, the distinction between angular and annular col-
lenchyma is often difficult, especially when massive thicken-
ing occurs causing the lumen to lose its angular appearance
(Fig. 2F). Therefore, some authors (Esau, 1965;Fahn, 1990;
Beck, 2005) do not recognize this type. Several textbooks
also distinguish lacunar (or lacunate) collenchyma (Mu
¨ller’s
‘Lu
¨ckencollenchym’) (Fig. 3D) when thickened cell walls
occur adjacent to intercellular spaces (Esau, 1965;Mauseth,
1988;Dickison, 2000;Beck, 2005;Evert, 2006). Duchaigne
(1955) and Fahn (1990) did not distinguish this type, as they
state that intercellular spaces often occur in other collenchyma
types. Therefore, intermediate forms occur where, for
example, tangential collenchyma can be lacunate (Fig. 2E),
and these are often found at the interface with parenchyma
tissues. As mentioned in the previous section, the term ‘col-
lenchymatous tissue’ was introduced by Esau (1936) to de-
scribe the bundle caps composed of collenchyma-like cells.
Although this terminology is not adopted by many authors,
it is still applicable to parenchyma resembling collenchyma
in any location in the plant, e.g. collenchymatous thickenings
occurring in epidermal and secretory cells (Evert, 2006). Often
parenchymatous cell types can have thickened walls which can
be referred to as ‘collenchymatous thickenings’. By using this
term, it is implied neither that the cells are prosenchymatous
nor that they contribute to the mechanical support of the
organs in which they occur.
In tissues lacking intercellular spaces, cell wall material is
often accumulated in the cell junctions and, therefore, they
have frequently been mistakenly referred to as collenchyma.
During the formation of intercellular spaces these accumula-
tions are generally reorganized as filamentous or wart-like
protrusions, and have been named intercellular pectic pro-
tuberances (Carlquist, 1956;Potgieter and Van Wyck, 1992;
Leroux et al., 2007). Carlquist (1956) observed such structures
in the intercellular spaces of the peripheral collenchyma of
Fitchia speciosa (Asteraceae, eudicots), explaining that some
intercellular spaces were occluded by ‘centrifugal extrusion
of pectic materials’. These tissues with large, occluded inter-
cellular spaces resemble, but may not be referred to as,
collenchyma.
COLLENCHYMA: A PRIMARY OR SECONDARY
CELL WALL?
Collenchyma cell walls are generally described as being
primary walls which they resemble in properties and compos-
ition. However, the terms ‘primary wall’ and ‘secondary wall’
have been employed in several fundamentally different senses,
often designating different structures or cell wall layers. Jarvis
(2007) pointed out that collenchyma does not fit comfortably
in most definitions as it is unclear how much of the thickening
is deposited after cells have ceased elongation. Kerr and Bailey
(1934) described a terminology based on morphology, reserv-
ing the term ‘primary cell wall’ for the original wall of the cell
which is formed in the meristematic region after cytokinesis,
and the ‘secondary cell wall’ for all subsequent layers depos-
ited during differentiation. According to these definitions, col-
lenchyma cell walls are secondary. This terminology was
based on investigations of tracheids and fibres, and did not
consider cell walls such as those of collenchyma which in-
crease simultaneously in surface area and in thickness during
the growth of young tissues. Wardrop et al. (1979) proposed
an alternative concept and recognized a primary cell wall, in
the sense of a meristematic one, a growing cell wall, present
during elongation, and a secondary cell wall, representing
the wall material deposited after surface expansion has
ceased. In this case, a part of the collenchyma cell wall is sec-
ondary. A more widely adopted concept (Fry, 2008) defines a
primary cell wall as ‘a wall in which microfibrils were laid
down while it was still capable of growing in area’, and sec-
ondary cell walls as being composed of ‘any additional micro-
fibrils deposited after the cell has stopped growing’. This
means that all growing cell walls, including collenchyma cell
walls, are primary. However, it is not clear if collenchyma
Leroux —Collenchyma: a review1088
by guest on January 4, 2013http://aob.oxfordjournals.org/Downloaded from
cell walls lose their capability to grow in area after cell elong-
ation has ceased and, how much, if any, cell wall material is
deposited after termination of cell wall expansion. Therefore,
a part of the collenchyma cell wall might be referred to as sec-
ondary. Moreover, this concept is problematic when describing
thickened xylem cell walls. Lignified cell walls of protoxylem
elements, organized in ring or helix patterns, are deposited
during elongation and should therefore be called primary,
whereas the lignified metaxylem thickenings are deposited
when elongation has ceased and hence should be referred to
as secondary. The latter problem can be solved by defining
cell walls in terms of their extensibility (Lee et al., 2011)–
with primary cell walls being extensible and secondary cell
walls being non-extensible – and allowing application of
this concept either locally or over the entire cell surface. For
example, the cell walls between the rings or helix structure
of protoxylem elements are extensible and should be referred
to as primary, whereas the locally lignified cell wall layers
are non-extensible and therefore should be called secondary.
Nonetheless, the confusion with regards to the nature (i.e.
primary and/or secondary) of collenchyma cell walls remains
as there is no clear view on the architecture and properties
of the collenchyma cell wall layers that have been deposited
after elongation has ceased.
Regardless of which concept is preferred, one needs to bear
in mind that cell walls are complex biomaterials. While many
concepts attempt to define boundaries, in reality there is more
of a gradient of architectures and properties between primary
and secondary cell walls (Lee et al., 2011).
CELL WALL STRUCTURE
To the best of my knowledge, Giltay (1882) was the first to
report the lamellation of collenchymatous cell walls, and
Anderson (1927) the first to have documented it. The latter
researcher also suggested that pectin-rich and cellulose-poor
lamellae alternated with pectin-poor and cellulose-rich lamel-
lae. While some researchers (Majumdar, 1941;Majumdar and
Preston, 1941) presented similar results for the collenchyma
cell walls in Heracleum sphondyllum (Apiaceae, eudicots),
others, including Preston and Duckworth (1946), reported a
uniform distribution of cellulose in the collenchyma cell
walls of Petasites vulgaris (Asteraceae, eudicots). It was not
until the development of transmission electron microscopy
and its subsequent use in biological disciplines in the mid
1950s that detailed studies were possible, and, after investigat-
ing the ultrastructure of collenchyma cell walls, Beer and
Setterfield (1958) and Roland (1964,1965,1966) reported
that the cellulose microfibril orientation in collenchyma cell
walls was predominantly longitudinal. However, both Beer
and Setterfield (1958) and Roelofsen (1951,1958,1965)
found that, in elongating cell walls such as these of collen-
chyma tissue the microfibrils adjacent to the cell membrane
had a more or less transverse orientation. These observations
led to the ‘multi-net growth’ hypothesis (Roelofsen and
Houwing, 1953;Roelofsen, 1959), which states that microfi-
brils, deposited transversely (or as transverse helices with a
flat shallow pitch; Lloyd, 2011) adjacent to the cell membrane,
change in orientation to a degree depending on the extent and
polarity of wall growth. As well as disagreement on the distri-
bution of cellulose and pectins in each of the lamellae, some
reported that the lamellae were continuous (Majumdar and
Preston, 1941) while others believed they were discontinuous
(Beer and Setterfield, 1958;Roland, 1965) as extra layers
appeared to arise at the corners inside and outside continuous
lamellae. These controversies led Chafe (1970) to undertake a
comparative study of different types of collenchyma, some of
which had previously been studied, such as celery and
Petasites. Instead of providing improved clarity, he observed
that the transverse orientation of fibrils that appeared to be
AB
CD
FIG. 3 . Schematic drawings of the most common types of collenchyma. (A) Angular collenchyma. (B) Tangential collenchyma. (C) Annular collenchyma. (D)
Lacunar collenchyma. This type often occurs as an intermediate type with angular and lamellar collenchyma, in which the size of the intercellular spaces can vary
from minute spaces (1) to large cavities surrounded by collenchymatous walls (2).
Leroux —Collenchyma: a review 1089
by guest on January 4, 2013http://aob.oxfordjournals.org/Downloaded from
restricted to the innermost layer in earlier works occurred
throughout the cell wall, alternating with lamellae in which
the orientation was longitudinal, the so-called crossed polyla-
mellated wall. In addition, he showed that the lamellae were
continuous. Roland et al. (1975,1977) later mentioned that
the observations that served as a basis for the multi-net
growth hypothesis were made on macerated material (partial
removal of matrix components). The cell walls in which the
crossed polylamellation was shown, on the other hand, had
either been stained with histochemical dyes or were investi-
gated by applying shadow-casting techniques. The removal
of matrix material probably disrupted the orientation of the
microfibrils and may have caused the differences in observa-
tions. The crossed-polylamellated cell walls in collenchyma
(Wardrop, 1969;Chafe, 1970;Wardrop et al., 1979) chal-
lenged the multi-net growth hypothesis. Moreover, Roland
et al. (1975) showed that orientation of microfibrils near the
cell membrane could be either transverse or parallel to the
long axis of the cell. These observations led to the ‘ordered
fibril hypothesis’ (Roland et al., 1975) in which microfibrils
are considered to be deposited in alternating transverse and
longitudinal orientations (Lloyd, 2011). During elongation,
previously formed layers decrease in thickness due to stretch-
ing (Chafe and Wardrop, 1972). In fact, the crossed-
polylamellated structure represents helices of shallow and
steep pitch, respectively, as shown by the results obtained by
Vian et al. (1993) (Fig. 4A, B). The pattern of deposition
might thus not alternate discontinuously between left-hand
oblique and right-hand oblique, but change more progressive-
ly, with the angle of each layer being regularly offset from its
predecessor. During cell expansion or elongation this helicoid-
al organization of the expanding wall may become partly dis-
persed. When extension of collenchyma cells has ceased, the
helicoidal pattern of deposition continues to thicken the wall
(Vian et al., 1993) and the transitional strata may become
clearly observable, resulting in a cell wall structure consisting
of consecutive bow-shaped arcs (Fig. 4B). In some cases these
transitional layers are very thin and only a criss-crossed poly-
lamellated pattern can be observed.
Some authors reported that the outermost layer of collen-
chyma cell walls shows a more distorted and random orienta-
tion of microfibrils (Wardrop, 1956a,b,1969;Roland, 1965;
Deshpande, 1976a,b,c;Matar and Catesson, 1988) and sug-
gested that this layer represents the cell wall formed during
cytokinesis. Furthermore, Beer and Setterfield (1958) observed
a distinct inner layer in which the cellulose microfibrils have a
transverse orientation. They suggested that this layer is depos-
ited after cell elongation has ceased and therefore might repre-
sent a thin secondary cell wall.
CELL WALL COMPOSITION
Early studies focused on the heterogeneity in collenchyma cell
wall composition
Apart from observations that collenchyma cell walls swell in
water, most early researchers including de Bary (1877) and
Giltay (1882) did not mention the chemical nature of collen-
chyma cell walls. After performing only a few histochemical
tests, Ambronn (1881) concluded that they were composed
of cellulose and lacked lignins. As late as the early 1900s,
plant anatomy books (e.g. Haberland, 1914) only specified cel-
lulose as the main constituent of collenchyma cell walls.
However, Giltay (1882) mentioned that some researchers, in-
cluding Harting, Mulder and Schacht, reported that collenchyma
cells were not entirely composed of cellulose. Vesque, on the
other hand, suggested that cellulose in older collenchyma
walls is modified into a gum-like substance (Giltay, 1882).
To the best of my knowledge, Anderson (1927) was the first
to report that collenchyma cell walls contain pectins in add-
ition to cellulose. He suggested that the wall consists of alter-
nating and closely packed cellulose and pectin lamellae. After
retting (a process employing the action of moist and decay-
producing bacteria to dissolve cellular components) and treat-
ment with chromic acid to remove pectic substances,
Majumdar and Preston (1941) concluded that pectin-rich la-
mellae alternated with cellulose-rich lamellae. This compos-
itional heterogeneity was later affirmed (Beer and Setterfield,
1958;Roland, 1964,1965,1966). Deshpande (1976b)found
that pectinase treatment of Curcurbita petiole collenchyma
caused cells to separate at the middle lamellae and layers of
A
B
FIG. 4 . Structure of celery collenchyma cell walls revealed by transmission
electron microscopy. (A) Transverse section of collenchyma tissue treated
with dimethylsulfoxide (DMSO) to extract matrix polysaccharides showing
unevenly thickened cell walls. (B) Detail of a thickened cell wall showing
its lamellated structure. While the helicoidal pattern is obvious near the
plasma membrane, it is dispersed in the outward direction (arrow). Note the
lateral thinning of the lamellae (squared area). Abbreviations: ct, collenchyma
thickenings; n, nucleus; pm: plasma membrane; pp, periplasmatic space.
Reproduced with minor modifications from Vian et al. (1993) with permission
from The University of Chicago Press.
Leroux —Collenchyma: a review1090
by guest on January 4, 2013http://aob.oxfordjournals.org/Downloaded from
microfibrils to separate within the cell walls, which led him to
suggest that pectins serve as ‘glue’ between adjacent cells as
well as lamellae. However, in the collenchyma walls of
Petasites vulgaris, cellulose was found to be distributed
evenly within the cell wall, while pectin was found in alternat-
ing lamellae (Preston and Duckworth, 1946). To obtain more
insight into the distribution of pectins in collenchyma cell
walls, Chafe (1970) stained collenchyma of different species
with pectin-specific dyes and showed that the distribution of
pectic substances within the collenchymatous cell wall was
relatively continuous in some species, but associated with spe-
cific lamellae in other species. In the latter case, he found the
pectins in the lamellae with microfibrils in longitudinal orien-
tation. The lack of a pronounced heterogeneity in pectin distri-
bution in some species led Chafe and Wardrop (1972) to
suggest that the lamellate appearance of the walls could be
an optical effect caused by differences in microfibrillar
orientation.
Majumdar and Preston (1941) reported that the inner layer
of collenchyma cell walls in Heracleum is chemically distinct,
being composed of cellulose only. Later, Preston (1952) men-
tioned that histochemical tests failed to provide evidence for
the presence of lignins, pectins, cellulose and lipids in this
layer. To date, no detailed study has been undertaken to
confirm or complete these observations.
Recent studies provided more insight into the molecular
composition of collenchyma cell walls
Since the 1990s more detailed compositional analyses of the
cell walls of specific tissues have been undertaken and also
provided greater insight into the molecular composition of col-
lenchyma. This was largely made possible by the increased use
and number of available cell wall-directed monoclonal anti-
bodies and carbohydrate-binding modules (CBMs; Biosupp-
lies Australia, CarboSource Services, Complex Carbohydrate
Research Center, University of Georgia, USA and
PlantProbes, University of Leeds, UK) which have facilitated
improved knowledge of in situ cell wall composition (Herve
´
et al., 2011;Lee et al., 2011). Unfortunately, to date, no
detailed immunocytochemical study specifically focused on
collenchyma cell walls has been undertaken. However, as col-
lenchyma occurs in the stems of many dicots such as tobacco,
which has been included in detailed analyses of cell wall com-
position, some data are available and are summarized below
and shown in Fig. 5. Details of the molecular composition of
collenchyma walls obtained by methods other than immuno-
cytochemistry are also discussed.
Collenchyma walls have a similar composition to primary
cell walls (Jarvis, 1992). These cell walls surround growing
cells and are made up of cellulose microfibrils embedded in
a hydrated matrix of complex polysaccharides classified as
hemicelluloses and pectins (Cosgrove, 2005). Hemicelluloses
are cellulose-binding polysaccharides, tethering cellulose
microfibrils together in order to form a strong network.
Pectins, on the other hand, are complex polysaccharides
forming hydrated gels that could affect the physical properties
of the cell wall. They are also important factors for controlling
wall porosity and wall thickness and they are the main compo-
nent of the middle lamella (Albersheim et al., 2010).
In addition to these polysaccharides, cell walls also contain
small amounts of structural proteins. Secondary cell walls,
which are thick and rigid, contain larger proportions of hemi-
celluloses, lower amounts of pectins, and are generally ligni-
fied (Albersheim et al., 2010).
Cellulose. Cellulose was one of the first components reported
to be present in collenchyma cell walls (Giltay, 1882). It is
found in the form of linear insoluble microfibrils occurring
in highly ordered crystalline, semi-ordered para-crystalline
and disordered amorphous states (O’Sullivan, 1997). Using
13
C nuclear magnetic resonance (NMR) spectroscopy on
living tissues, Jarvis and Apperley (1990) found high amo-
unts (in comparison with cotton, wood and other secondary
cell walls) of amorphous (possible crystallite-surface) cellu-
lose relative to cellulose I in celery collenchyma. Recently,
the repertoire of cell wall-directed probes has been extended
by the development of CBMs obtained from microbial plant
cell wall hydrolases (Boraston et al., 2004;Shoseyov et al.,
2006) (Fig. 5A). The variation in binding patterns of some
cellulose-directed CBMs (Blake et al., 2006) showed that cel-
lulose chains in collenchymatous cell walls in celery petioles
do not form highly ordered crystalline structures, as CBM17,
binding to internal regions of amorphous cellulose, labelled
the collenchymatous thickenings, especially after enzymatic
removal of pectic homogalacturonan.
Pectins. Pectic polysaccharides are abundant in primary cell
walls and include homogalacturonans, rhamnogalacturonans,
xylogalacturonans, galactans, arabinans and arabinogalactans
(Harholt et al., 2010). Pectic polymers display variation in
terms of both glycosyl structure and polysaccharide modifica-
tions, such as methyl-esterification and acetylation, and these
may vary within tissues and even single cell walls and are
often developmentally regulated (Albersheim et al., 2010).
Jones et al. (1997) showed that JIM5 and JIM7, binding to
pectic homogalacturonan with low and high degrees of
methyl-esterification, respectively, bound to collenchymatous
thickenings in tomato petioles, with JIM5 displaying a stronger
binding. JIM5 also labelled collenchyma in elderberry
(Sambucus, Adoxaceae, eudicots) (Fig. 5C) and tobacco
(Fig. 5G). Pectins with low degrees of methyl-esterification
have the ability to form gels (Willats et al., 2001). For in-
stance, in parenchyma, high-esterified pectins are generally
present throughout the cell wall, while low-esterified homoga-
lacturonans are generally found in the middle lamellae where
they can participate in calcium cross-linking and gel formation
(Albersheim et al., 2010). In a study estimating the polymer
rigidity of growing and non-growing celery collenchyma
walls through in vivo solid-state NMR, Fenwick et al. (1997)
showed a decrease in the proportion of methyl-esterified
pectin in the collenchyma cell walls when growth ceased. If
the pectin matrix, in which the layers of cellulose microfibrils
are embedded, are rich in methyl-esterified pectins, the latter
may facilitate a degree of shear between the lamellae and
enable elongation. Jarvis (1992) reported that pectic polysac-
charides might control the thickness of collenchyma cell
walls by tethering the lamellae in its walls. He also suggested
that the cell wall pectins in the collenchymatous thickenings
form a gel continuous with that of the middle lamellae, pre-
venting the layers from delaminating.
Leroux —Collenchyma: a review 1091
by guest on January 4, 2013http://aob.oxfordjournals.org/Downloaded from
In contrast to the abundance of pectic homogalacturonan
throughout collenchyma cell walls of tomato petioles, LM5,
an antibody recognizing (1 4)-b-D-galactan which occurs
as side chains on pectic rhamnogalacturonan-I (RG-I), was
specifically detected in the inner cell wall layer of collenchyma
(Jones et al., 1997), as shown for elderberry in Fig. 5E
(Sambucus, Adoxaceae, eudicots). It is possible that this
layer corresponds to the distinct inner layer observed in collen-
chyma walls by Majumdar and Preston (1941),Preston (1952)
and Beer and Setterfield (1958). The side chains of RG-I also
include arabinan polymers. Weak labelling of collenchyma
cell walls by the LM6 antibody, which recognizes pectic
(1 5)-b-D-arabinan, was shown in tobacco by Marcus
et al. (2008) (Fig. 5J). Willats et al. (2001) suggested that
A
Elderberry
Tobacco
B
CD
E
G
LM
HIJK
F
FIG. 5 . Indirect immunolabelling of cell wall polysaccharides in collenchyma of elderberry (Sambucus nigra, Adoxaceae, eudicots) and tobacco (Nicotiana
tabacum, Solanaceae, eudicots) with monoclonal antibodies and carbohydrate-binding modules (CBMs). (A) CBM3a, targeting crystalline cellulose, binds
strongly to the collenchyma cell walls. (B) Equivalent section to (A) stained with Calcofluor White to show the full extent of cell walls. (C) Binding of the
pectic homogalacturonan antibody JIM5 to all cell walls. Note stronger labelling of the inner cell wall layers of the collenchyma tissue (arrow). (D)
Equivalent section to (C) stained with Calcofluor White to show the full extent of cell walls. (E) LM5, directed against pectic galactan, labels the inner layer
of the collenchyma cell walls (arrow). (F) Equivalent section to (E) stained with Calcofluor White to show the full extent of cell walls. (G) Section immuno-
labelled with the pectic homogalacturonan-directed probe JIM5. (H, I) Weak recognition of the xyloglucan LM15 epitope (H) and its increased detection
after pectate lyase pre-treatment (I). (J) Section immunolabelled with the pectic arabinan probe LM6 after pectate lyase pre-treatment. (K) Calcofluor White
staining of equivalent section to (G– J) showing the full extent of cell walls. (L, M) The anti-xylan monoclonal antibodies LM10 and LM11 bind to the
inner regions (arrow) of cell wall thickenings and to the outer regions (arrow) of the cell wall near the cell junctions, respectively (combined with Calcofluor
White fluorescence in the insets). Abbreviations: e, epidermis; c, collenchyma; p, parenchyma. Scale bars: (A–K) ¼100 mm; (L, M) ¼10 mm. (A–F) are repro-
duced from unpublished results with kind permission of P. Knox (University of Leeds, UK); (G–K) are reproduced from Marcus et al. (2008), and (L, M) are
reproduced from Herve
´et al. (2009) with permission from John Wiley and Sons.
Leroux —Collenchyma: a review1092
by guest on January 4, 2013http://aob.oxfordjournals.org/Downloaded from
the occurrence of RG-I and its structural variants may be
related to mechanical properties. For instance, the appearance
of galactan is correlated with an increase in firmness of pea
cotyledons (McCartney et al., 2000). McCartney et al.
(2003) found that the incorporation of the pectic (1
4)-b-D-galactan epitope in cell walls preceded the main
phase of cell elongation in arabidopsis roots. Moreover, galac-
tans are enriched in primary cell walls of elongating cells of
potato stolons (Bush et al. 2001) as well as in elongating
carrot suspension cells (Willats et al., 1999). Arend (2008)
showed that in tension wood fibres the distribution of the
LM5 epitope is restricted to a narrow cell wall area between
the gelatinous G-layer and the secondary cell wall, further
strengthening the view that galactan-rich pectins may play an
important role in mechanically stressed tissues. Pectic galac-
tans are also known to be some of the most flexible cell wall
polymers (Ha et al., 1996), and the presence of these polymers
decreases the ability of pectin molecules to cross-link and form
a coherent gel network (Hwang and Kokini, 1991). Jones et al.
(2003) showed that enzymic treatments with arabinase pre-
vented either stomatal opening or closing, suggesting that
removal of arabinans, also occurring as side chains on RG-I,
might induce stiffening of the cell wall by preventing the for-
mation of calcium-mediated interactions between homogalac-
turonan domains (Harholt et al., 2010). However, at present,
there is no clear view on the function of these arabinan side
chains. Other studies showed that they might have important
functions in relation to cell wall hydration (Moore et al., 2008).
Spurr (1957) showed that the amount of boron in celery has
a pronounced effect on the cell wall thickness. In boron-
deficient plants the cell walls of different tissues showed con-
trasting responses. While cortical and phloem parenchyma cell
walls were thicker, collenchyma cell walls became markedly
thinner. This effect was not (or not entirely) caused by swel-
ling as Spurr (1957) showed that boron-deficient celery collen-
chyma cell walls contained fewer lamellae. An early symptom
of boron deficiency in flowering plants is the formation of
primary walls with abnormal morphology and mechanical
properties causing stems to become more rigid, sometimes re-
ferred to as the ‘cracked-stem’ symptom (Purvis and Ruprecht,
1937). Kaneko et al. (1997) showed that borate diesters cova-
lently cross-link rhamnogalacturonan-II (RG-II) dimers. RG-II
is a structurally complex pectic polysaccharide present in the
primary walls of lycophytes, ferns, gymnosperms and angios-
perms, and its cross-linking is required for the formation of a
three-dimensional pectic network in muro (O’Neill et al.,
2004). This network contributes to the mechanical properties
of the primary wall and is required for normal plant growth
and development (O’Neill et al., 2004). The changes in wall
properties that result from decreased borate cross-linking of
pectin may have led to some of the symptoms reported by
Spurr (1957).
Since pectins are hydrophilic, collenchyma cell walls
are rich in water. The amount of water may reach 60 %,
based on fresh weight (Cohn, 1892). Dehydration of col-
lenchyma generally results in shrinkage, especially in the
radial direction. By comparing vibratome sections, which
are made without performing chemical fixation or dehy-
dration, and sections of resin-embedded samples, which
are chemically fixed and dehydrated, a notable shrinking
of collenchyma cell walls can be observed in the latter
(Fig. 1B, C).
Hemicelluloses. Hemicelluloses include xyloglucans, xylans,
mannans and glucomannans, and their most important bio-
logical role is their contribution to strengthening the cell
wall by interaction with cellulose (Scheller and Ulvskov,
2010). Although cellulose and pectins have long been seen
as the sole constituents of collenchymatous cell walls,
Preston (1952) reported that collenchyma cell walls also
contain hemicelluloses. In the collenchyma of Petasites,
Roelofsen (1959) detected that the cell walls were composed
of 45 % pectin and 35 % hemicelluloses, meaning that cellu-
lose only accounts for a maximum of 20 % of the wall polysac-
charides. Xyloglucan is one of the most abundant
hemicelluloses found in primary cell walls of non-
graminaceous flowering plants and is proposed to have a func-
tional role in tethering cellulose microfibrils. Moreover, there
is evidence suggesting that hemicelluloses can be linked to
pectins (Popper and Fry, 2008) and therefore might take part
in a complex mechanism for the tethering and spacing of cel-
lulose microfibrils. The LM15 monoclonal antibody, binding
to the XXXG-motif of xyloglucan, labelled tobacco collen-
chyma cell walls, especially the inner cell wall layers adjacent
to the middle lamella. However, this distribution pattern was
only observed after pectate lyase pre-treatment (Marcus
et al., 2008) (Fig. 5H– I). In young celery petioles, xyloglucan
endotransglycosylase action was found to be particularly high
in the thick-walled collenchyma cells (Vissenberg et al.,
2000). As collenchyma tissues support the stem while its
cells are still elongating, the increased XET action suggests
that endotransglycosylase/hydrolase-mediated wall modification
may contribute to the structural integrity of the plant body and/
or that it plays an important role during cell expansion.
Although xylans are one of the major hemicelluloses found
in most secondary cell walls (Carpita and Gibeaut, 1993;
Harris, 2005), they have been detected in collenchyma cell
walls, especially after enzymatic removal of pectic homogalac-
turonan (Herve
´et al., 2009). Herve
´et al. (2009) showed that
the xylan probes CBM15 and LM11 bound to the collenchy-
matous cell thickenings in tobacco stems, while LM10 labelled
the complementary inner cell wall regions (Fig. 5L– M). The
occurrence of xylans in thickened primary cell walls of collen-
chyma and the epidermis, and not in all primary cell walls,
may indicate some similarities in the thickening processes of
primary and secondary cell walls.
Mannans are another group of hemicellulosic polysacchar-
ides associated with both storage and structural properties of
cell walls (Scheller and Ulvskov, 2010). They are proposed
to cross-link cellulose by means of hydrogen bonds, acting
in similar ways to other hemicelluloses. Recently, mannan epi-
topes have been unmasked in collenchyma cell walls of
tobacco after removal of pectic homogalacturonan (Marcus
et al., 2010). Although a link between pectins and mannans
has not been proven, it is possible that this association may
be implicated in cell wall structure and/or modification.
Even though the knowledge of the in situ distribution of cell
wall polymers in collenchyma is fragmentary, the above exam-
ples serve to illustrate that much is to be discovered in future
studies focused on collenchyma.
Leroux —Collenchyma: a review 1093
by guest on January 4, 2013http://aob.oxfordjournals.org/Downloaded from
SCLERIFICATION AND PERIDERM FORMATION
Although collenchyma tissues are especially suited to provide
support to young plant organs, they may also serve as the main
supporting tissue in stems that are only moderately strength-
ened by secondary growth or in herbaceous stems that lack
such growth altogether. In some (flowering) plant species
belonging to the Acanthaceae, Apiaceae, Bignoniaceae,
Fabaceae, Lamiaceae, Piperaceae and Polemoniaceae, periph-
eral collenchyma tissues are known to undergo sclerification
by deposition of lignified secondary cell walls after elongation
has ceased (e.g. Born, 1886;Went, 1924;Lemesle, 1929;
Nobe
´court and Papier, 1951;Duchaigne, 1953,1954a,b,
1955;Roland, 1965;Wardrop, 1969;Calvin and Null, 1977)
(Fig. 6). While most of the researchers listed above agree
that sclerification implies the addition of extra lignified cell
wall layers, much controversy remains when it comes to
what extent, if at all, the original collenchyma wall is lignified.
Some report that the original collenchyma wall remains non-
lignified (Nobe
´court and Papier, 1951), while others believe
that it becomes lignified or that the wall is reabsorbed
(Went, 1924). Duchaigne (1953,1955) reports a double trans-
formation process in which secondary cell wall material is
deposited in a centripetal direction while the collenchyma
wall is modified through deposition of new material and/or
substitution in a transfugal direction. The addition of second-
ary cell wall material reduces the size of the cell lumen
giving the cell the appearance of a thick-walled sclerenchyma
cell. Roland (1965) and Wardrop (1969) both confirmed the
double transformation system of Duchaigne (1953), which
makes lignified collenchyma indistinguishable from scleren-
chyma. A reduction in thickness of the original collenchyma
cell walls does not necessarily imply that parts of it become
lignified, as sclerification could cause the original collenchyma
cell walls to shrink as a result of dehydration. Moreover, the
controversy regarding whether or not the original collenchyma
cell wall undergoes lignification should at least take into
account that variation may exist between species. For compari-
son, when considering sclerenchyma tissues, in some xylem
elements lignin is first deposited in the middle lamellae, some-
times even before elongation has ceased, while in other scler-
ified cells, lignin is not present in the primary cell walls (Evert,
2006). Both Wardrop (1969) and Calvin and Null (1977)
reported that collenchyma tissues at the petiole base of
carrots did not sclerify while higher up in the petiole sclerifi-
cation was observed. Wardrop (1969) concluded that growth
mainly took place in this region, but, as pointed out by
Calvin and Null (1977), carrot leaves can change their angle
with respect to the horizontal, and these adjustments might
be facilitated by collenchyma tissues at the petiole base.
The cases mentioned above are examples of mature collen-
chyma undergoing sclerification. However, in some cases, col-
lenchyma could represent sclerenchyma in which collenchyma
is just a temporal developmental phase. Slowness of matur-
ation relative to initiation and early differentiation is not un-
common in biological systems. This could explain why such
tissues are present in some ferns species such as Asplenium
theciferum (O. Leroux et al. unpubl. res.). In the latter case,
cell walls may thicken as long as elongation takes place. The
collenchymatous nature of these walls enables more light to
reach the chloroplasts of the underlaying cortical tissue, and
differences in growth rate might also enable unfurling of
fern croziers. Once elongation is completed, the lamina takes
over photosynthesis and the petiole can increase its stiffness
through sclerification of some of its tissues, a process often ac-
companied by the impregnation of cell walls with dark phen-
olic compounds. The collenchymatous tissue shown to
replace sclerenchyma at the transition from blade to sheath
in grass leaves (Dayanandan et al., 1976,1977;Paiva and
Machado, 2003;Evert, 2006) is another example of non-
lignified sclerenchyma mimicking the properties of collen-
chyma in an early developmental phase.
Although collenchyma is the main supporting tissue in
growing stems, it is also often implicated in the secondary
growth process. At first, collenchyma cells enlarge in order
to keep up with the increasing diameter of stems during exten-
sive secondary growth. During this process, the cell walls
become thinner. It is not clear if this thinning results from
removal of wall components or from wall stretching and dehy-
dration (Esau, 1965;Evert, 2006). When stretching has
reached its limit, the phellogen initiates in a single cell layer
of the collenchyma through periclinal divisions. While most
divisions are periclinal, periodic anticlinal longitudinal
ABCD
FIG. 6 . Sclerified collenchyma tissue in the petiole of Eryngium campestre (Apiaceae, eudicots). (A) Vibratome section triple-stained with acridine red, chry-
soidine and astra blue showing gross anatomy. (B) Detail of (A) showing sub-epidermal sclerified collenchyma (sc) and sclerenchyma (s). (C) Unstained vibra-
tome section. (D) Vibratome section stained with phloroglucinol/HCl to indicate lignins (red). Only the inner layer of the collenchyma cell walls is lignified. Note
the glistening nature of the non-lignified collenchyma cell walls under the epidermis (arrowheads). Abbreviations: s, sclerenchyma; sc, sclerified collenchyma.
Scale bars ¼50 mm.
Leroux —Collenchyma: a review1094
by guest on January 4, 2013http://aob.oxfordjournals.org/Downloaded from
divisions are necessary to keep up with the increase in circum-
ference of the stem. It needs to be noted that the phellogen can
be initiated in any living, and thus potentially meristematic,
tissue (Evert, 2006). Similar structures are initiated when
collenchyma responds to injuries through formation of a
wound-periderm.
FUNCTIONAL ROLE
Although collenchyma was one of the first tissues that intri-
gued botanists because of its mechanical role in growth and
development (Ambronn, 1881), few studies have focused on
its biomechanical properties (Ambronn, 1881;Esau, 1936).
By the end of the 19th century it became clear that plants
react to gradually increasing strains by developing more mech-
anical tissue ( for a review, see Hibbard, 1907). Many experi-
ments were carried out to investigate the impact of
mechanical stress on plant growth and development, and the
term ‘thigmomorphogenesis’ was later coined by Jaffe
(1973) to describe the response of plants to mechanical stimu-
lation. Venning (1949) and Walker (1957) studied the effect of
wind on celery collenchyma and found that while the number
of collenchyma bundles remained unchanged, larger areas of
collenchyma with heavier cell wall thickenings developed at
an early stage. Walker (1960) measured the length of collen-
chyma cells in Datura stramonium (Solanaceae, eudicots)
and found that mechanical stimulation decreased their size
and thus inhibited elongation. He also examined if etiolation
resulted in reutilization of collenchyma cell wall material
and showed a reduction of wall thickening, which suggested
that collenchyma wall material might be reabsorbed as a re-
spiratory substrate during etiolation. A number of detailed
studies have been performed on isolated collenchyma
strands, and their tensile strength has been estimated by meas-
uring the weight required to break them. Ambronn (1881) mea-
sured the breaking stresses of isolated mechanical tissues of
different species and found that collenchyma and scleren-
chyma could support 10–12 and 10 – 15 kg mm
22
, respective-
ly. Collenchyma was shown to undergo plastic deformations at
relatively low stresses compared with fibres, which regained
their original length after tensile stresses that were up to 18
times higher (Ambronn, 1881). Esau (1936) reported that the
average tensile breaking stress for collenchyma of celery was
roughly 5.7 times higher than that of the primary xylem ele-
ments in the same region, but also showed that variation in
breaking stress was correlated with the age of the tissue. The
latter aspect was also shown by Jaccard and Pilet (1975)
who reported that collenchyma exhibits less strain as it ages,
with a decrease of the plastic to elastic strain ratios. These
experiments indicate that young collenchyma tissues are char-
acterized by their great tensile strength and plasticity, whereas
sclerenchyma combines great tensile strengths with elasticity.
The plastic deformation of young collenchyma walls at rela-
tively low stress levels is important as most of the elongation
occurs after collenchyma cells have started to thicken their
walls. In some plant organs that do not develop sclerenchyma
or secondary growth, collenchyma often persists as the main
mechanical tissue, and, as mentioned above, their walls
become more elastic. Some collenchyma tissues sclerify and
modify their walls in such a way that they are able to withstand
tensile and compressive components of bending stresses
(Jarvis, 2007). In organs that undergo secondary growth, the
xylem becomes the main supporting tissue and therefore the
mechanical properties of collenchyma become less important.
Flexibility of collenchymatous tissues is an advantage for
any plant part that is subjected to mechanical stress that
would result in damaging its tissues. For example, the col-
lenchymatous tissue found at the sheath – leaf blade transition
area of the grass Panicum maximum (Paiva and Machado,
2003) enables greater flexibility, reducing the risk of ruptures.
Likewise, the central position of the collenchyma in some
pulvini (Moysset and Simon, 1991) facilitates bending and
prevents damage to the vascular tissue.
FUTURE PERSPECTIVES
Although there is a general consensus that collenchyma is a
mechanical tissue supporting growing organs, there remains
some controversy about its precise definition. Collenchyma
cell walls thicken during elongation, but also after cell expan-
sion has ceased, and therefore they do not fit comfortably in
the definitions of either primary or secondary cell walls.
Moreover, as their shape varies from short prismatic-like to
elongated cells with tapering ends and some tissues undergo
sclerification, efforts for sharp distinction of collenchyma
may lead to partial overlap with other tissues. Regardless of
its definition, many questions pertaining to collenchyma cell
walls remain and these should be addressed by employing
techniques that have become possible or more easily accessible
in recent years. The distribution patterns of some cell wall
glycan epitopes discussed in this review indicate that immuno-
cytochemistry has significant potential to contribute to knowl-
edge of collenchyma cell wall architecture. Is the distribution
of glycan polymers uniform or restricted to specific lamellae?
To what extent are polymers and their structural variants asso-
ciated with specific developmental stages of collenchyma cell
walls? How are the cell walls remodelled during and after cell
elongation? Detailed analysis of the chemical composition of
collenchyma cell walls will also provide increased insight
into the structural– functional relationships of cell wall archi-
tectures, especially when its biomechanical properties are the
subject of concurrent investigations. Which architectural mod-
ifications are associated with the transition from plastic to
elastic properties? As mechanical properties are determined
not only by collenchyma cell wall architecture, but also by
the shape and arrangement of cells in the tissue, a combinator-
ial approach is preferred, correlating structural and compos-
itional information with biomechanical properties, at both
the tissue and cell level.
ACKNOWLEDGEMENTS
This work was supported by an IRCSET-EMPOWER post-
doctoral fellowship grant at the National University of
Ireland, Galway (NUI Galway) as well as by the Systematics
Research Fund (The Linnean Society of London) awarded to
the author. The NUI Galway James Hardiman Library pro-
vided support through its interlibrary loan service, handling
many requests for literature quickly and efficiently. I thank
Z. A. Popper for critically reading the manuscript.
Leroux —Collenchyma: a review 1095
by guest on January 4, 2013http://aob.oxfordjournals.org/Downloaded from
LITERATURE CITED
Albersheim P, Darvill A, Roberts K, Sederoff R, Staehelin A. 2010. Plant
cell walls: from chemistry to biology. New York: Garland Science.
Alston HG. 1956. The subdivision of the Polypodiaceae. Taxon 5: 23–25.
von Alten H. 1909. Wurzelstudien. Botanische Zeitung 67: 175– 198.
Ambronn H. 1881. U
¨ber die Entwickelungsgeschichte und die mechanishen
Eigenschaften des Collenchyms. Jahrbuch fu
¨r Wissenschaftliche
Botanik 12: 473–541.
Anderson D. 1927. U
¨ber die Struktur der Kollenchymzellwand auf Grund
mikrochemischer Untersuchungen. Sitzungsberichte der Akademie der
Wissenschaften in Wien. Mathematisch-Naturwissenschaftliche Klasse
136: 429– 440.
Arend M. 2008. Immunolocalization of (14)-b-galactan in tension wood
fibers of poplar. Tree Physiology 28: 1263– 1267.
de Bary A. 1877. Vergleichende Anatomie der Vegetationsorgane der
Phanerogamen und Farne. Leipzig: Engelmann.
Ba
¨secke P. 1908. Beitra
¨ge zur lenntnis der physiologischen Scheiden der
Filicinen-Achsen und -Wedel sowie uber den Ersatz des Korkes Bei den
Filicinen. Dissertation, Marburg.
Beck CB. 2005. An introduction to plant structure and development.
Cambridge: Cambridge University Press.
Beer M, Setterfield G. 1958. Fine structure in thickened primary walls of col-
lenchyma cells of celery petioles. American Journal of Botany 45:
571–580.
Blake AW, McCartney L, Flint JE, et al. 2006. Understanding the biological
rationale for the diversity of cellulose-directed carbohydrate-binding
modules in prokaryotic enzymes. Journal of Biological Chemistry 281:
29321– 29329.
Boraston AB, Bolam DN, Gilbert HJ, Davies GJ. 2004. Carbohydrate
binding modules: fine-tuning polysaccharide recognition. Biochemical
Journal 382:769– 781.
Born A. 1886. Vergleichende systematische Anatomie des Stengels der
Labiaten und Scrophulariaceen. Dissertation, Berlin.
Brown JT. 1976. Observations on the hypodermis of Equisetum.South
African Journal of Science 72: 303–305.
Bush MS, Marry M, Huxham IM, Jarvis MJ, McCann MC. 2001.
Developmental regulation of pectic epitopes during potato tuberisation.
Planta 213: 869– 880.
Calvin CL, Null RL. 1977. On the development of collenchyma in carrot.
Phytomorphology 27:323– 331.
Carlquist S. 1956. On the occurrence of intercellular pectic warts in
Compositae. American Journal of Botany 43: 425– 429.
Carpita NC, Gibeaut DM. 1993. Structural models of primary-cell walls in
flowering land plants – consistency of molecular structure with the phys-
ical properties of the walls during growth. The Plant Journal 3: 1 –30.
Chaerle P, Viane RLL. 2004. Leaf anatomy and the occurrence of false veins
in Asplenium (Aspleniaceae, Pteridophyta). Botanical Journal of the
Linnean Society 145: 187–194.
Chafe SC, Wardrop B. 1972. Fine structural observations on the
epidermis. 1. The epidermal cell wall. Planta 92, 13– 24.
Chafe SC. 1970. The fine structure of the collenchyma cell wall. Planta 90:
12– 21.
Cohn J. 1892. Beitrage zur Physiologie des Collenchyms. Jahrbu
¨cher fu
¨r wis-
senschaftliche Botanik 24: 144– 172.
Cosgrove DJ. 2005. Growth of the plant cell wall. Nature Reviews Molecular
Cell Biology 6: 850– 861.
Crum HA, Anderson LE. 1964. Notes on Physcomitrium collenchymatum.
Bryologist 67: 350– 355.
Dayanandan P, Herbard FV, Kaufman PB. 1976. Cell elongation in the
grass pulvinus in response to geotropic stimulation and auxin application.
Planta 132: 245– 252.
Dayanandan P, Hebard FV, Baldwin Van D, Kaufman PB. 1977. Structure
of gravity-sensitive sheath and internodal pulvini in grass shoots.
American Journal of Botany 64: 1189– 1199.
Deshpande BP. 1976a. Observations on the fine structure of plant cell
walls. I. Use of permanganate staining. Annals of Botany 40: 433– 437.
Deshpande BP. 1976b. Observations on the fine structure of plant cell walls.
II. The microfibrillar framework of the parenchymatous cell wall in
Cucurbita.Annals of Botany 40: 439– 442.
Deshpande BP. 1976c. Observations on the fine structure of plant cell walls.
III. The sieve tube wall in Cucurbita.Annals of Botany 40: 443 – 446.
Dickison WC. 2000. Integrative plant anatomy. New York: Harcourt
Academic Press.
Duchaigne A. 1953. Sur la transformation du collenchyme en scle
´renchyme
chez certaines Ombellife
`res. Comptes Rendus Hebdomadaires des
Se
´ances de l’Acade
´mie des Sciences, Paris 236: 839– 841.
Duchaigne A. 1954a. Nouvelles observations sur la scle
´rification du collench-
ymes chez les Ombellife
`res. Comptes Rendus Hebdomadaires des
Se
´ances de l’Acade
´mie des Sciences, Paris 238: 375– 377.
Duchaigne A. 1954b. La scle
´rification du collenchymes chez les labie
´es.
Bulletin de la Socie
´te
´Botanique de France 101: 235–237.
Duchaigne A. 1955. Les divers types de collenchymes chez les dicotyle
´dones:
leur ontoge
´nie et leur lignification. Annales des Sciences
Naturelles-Botanique Biologie Ve
´ge
´tales 16: 455– 479.
Esau K. 1936. Ontogeny and structure of collenchyma and of vascular tissues
in celery petioles. Hilgardia 10: 431–476.
Esau K. 1965. Plant anatomy. New York: John Wiley.
Evert RF. 2006. Esau’s plant anatomy. Meristems, cells, and tissues of the
plant body: their structure, function, and development, 3rd edn. New
Jersey: John Wiley and Sons, Inc.
Fahn A. 1990. Plant anatomy, 4th edn. New York: Pergamon Press.
Falkenberg P. 1876. Vergleichende Untersuchungen u
¨ber den Bau der
Vegetationsorgane der Monocotyledonen. Stuttgart: Ferdinand Enke.
Fenwick KM, Jarvis MC, Apperley DC. 1997. Estimation of polymer rigid-
ity in cell walls of growing and nongrowing celery collenchyma by solid-
state nuclear magnetic resonance in vivo.Plant Physiology 115: 587–592.
van Fleet DS. 1946. An oxidation and absorption reaction for differentiating
the endodermis and the collenchyma. Stain Technology 21: 95– 98.
van Fleet DS. 1950. A comparison of histochemical and anatomical character-
istics of the hypodermis with the endodermis in vascular plants. American
Journal of Botany 37:721–725.
Fry SC. 2008. Plant cell walls. In: Roberts K. ed. Handbook of plant science.
New York: John Wiley, 266–276.
Gier LJ. 1955. Physcomitrium collenchymatum. Transactions of the Kansas
Academy of Science 58: 330 –333.
Gierlinger N, Sapei L, Paris O. 2008. Insights into the chemical composition
of Equisetum hyemale by high resolution Raman imaging. Planta 227:
969–980.
Giltay E. 1882. Het collenchym. PhD dissertation, University of Leiden.
von Guttenberg H. 1940. Der prima
¨re Bau der Angiospermenwurzel. In: Von
Lindbauer K. ed. Handbuch der Pflanzenanatomie. Berlin: Gebru
¨der
Borntraeger.
Ha MA, Evans BW, Jarvis MC, Apperley DC, Kenwright AM. 1996.
CP-MAS NMR of highly mobile hydrated biopolymers: polysaccharides
of