Re view

Cell walls are at the basis of a structural, four-dimensional framework of plant form and growth time. Recent rapid progress of cell wall research has led to the situation where the old, long-lasting juxtaposition: "living" protoplast--"dead" cell wall, had to be dropped. Various attempts of re-interpretation cast, however, some doubts over the very nature of plant cell and the status of the walls within such a cell. Following a comparison of exocellular matrices of plants and animals, their position in relation to cells and organisms is analysed. A multitude of perspectives of the biological organisation of living beings is presented with particular attention paid to the cellular and organismal theories. Basic tenets and resulting corollaries of both theories are compared, and evolutionary and developmental implications are considered. Based on these data, "The Plant Body"--an organismal concept of plants and plant cells is described.

When Rob ert Hooke first ob served the cork un der mi cro scope and de scribed its struc ture as com posed of small units -"the cells" as he called them [1], he did not real ise that in fact he was ob serv ing a net work of cell walls in a dead tis sue.Nev er the less, the idea and the term "cell" per sisted and later found its place within the cel lu lar the ory of bi o log i cal or ganisa tion.Cell walls (CW) are con sid ered one of the ma jor struc tural el e ments that dis tin guish plant cells from other eukaryotic cells.Taking ad van tage of the mod els elab o rated for an imal sys tems, and on the ba sis of re cent rapid prog ress of CW re search, it is of ten sug gested that there are func tional sim i lar i ties of plant walls to the an i mal exocellular ma trix (ECM; e.g.[2,3]).This, how ever, casts some doubts over the very def i ni tion of "plant cell" and the po si tion of cell walls within such a cell.
Cell walls are in dis pens able el e ments of plant cells de ter min ing their shape and af fecting their func tion.Within a plant they form a struc tural and func tional con tin uum -the apoplast.On the other hand, CW could be consid ered as a cel lu lar "organelle" un der go ing dy namic changes in re sponse to a pleth ora of stim uli [4].In that re spect, they could be defined as part of yet an other struc tural and func tional con tin uum, com posed of cell walls, plasma mem brane, and the cytoskeleton [5][6][7][8].Taken to gether, CW are at the ba sis of a struc tural, three-dimensional or in deed four-dimensional frame work of plant form and growth time [9].

EXOCELLULAR MA TRI CES OF PLANTS AND AN I MALS -A COM PAR I SON
Plants and an i mals adopted two dif fer ent strat e gies of life and this found a re flec tion in the prop er ties and be hav iour of their cells, and also in the struc ture and func tions of their exocellular ma tri ces.From the chem i cal point of view, an i mal ECM is com posed mainly of glycoproteins and proteoglycans, while poly sac cha rides are the ma jor build ing el e ments of plant CW with (glyco)pro teins and phe no lic com pounds be ing mi nor com ponents.In both cases, the cur rently used models de scribe exocellular ma tri ces as com plex net works of macromolecules.For ex am ple, in plant CW at least three in ter twin ing networks: cel lu lose/hemicellulose, pec tin, protein, and lignin (in some types of walls), could be dis tin guished [10].Al though chem i cally dif fer ent, CW and some types of an i mal ECM (es pe cially in con nec tive tis sue) are rel a tively sim i lar when their me chan i cal prop er ties are con sid ered.Both are com pos ite ma te ri als with rigid, stretch-resistant rods (cel lu lose microfibrils or var i ous types of col la gen) embed ded in an amor phous, com pres sion-re sistant, ge lat i nous ma trix [11].This amor phous gel in plants is formed mainly by pectins and sta bi lised by (glyco)pro teins and phenolics [12].It should be noted, how ever, that the occur rence and dis tri bu tion of exocellular ma trices dif fer in or gan isms from both King doms, and this re flects their dif fer en ti ated func tional ity with re spect to the main te nance of in ternal chem i cal com po si tion of the cells and the re sult ing os motic gra di ent be tween cells and their en vi ron ment.In an i mal cells, op er at ing sys tems of ac tive ion trans port pro vide the required os motic bal ance [13].In ef fect, ECM oc cur rence is dif fer en ti ated, de pend ing on the tis sue, and within a tis sue ECM might be shared by many cells and cell types.In contrast, plant cells are im mo bi lised within the bound aries of their walls, which have to be strong enough to con strain the hy dro static pres sure evoked by os motic gra di ents between cells and their mi lieu.Thus CW forms a struc tural el e ment which is both an in te gral part of each cell and a con tin uum span ning the en tire body of the plant [13,14].
The for ma tion of exocellular ma tri ces is roughly a two-step pro cess, com pris ing 1) biosynthesis of the build ing blocks, and 2) assem bly of these el e ments into a func tional matrix, with these steps sep a rated spa tially and tem po rally.Due to the dif fer ences in chem i cal com po si tion be tween CW and ECM, these pro cesses are un der dif fer ent lev els of ge netic con trols.ECM glycoproteins could be regarded as pri mary prod ucts of gene ac tiv ity and thus transcriptional con trols are of primary im por tance, with some post-translational con trol.On the other hand, poly sac charides (and phenolics) of plant CW are products of en zyme ac tion.This shifts the weight of biosynthetic con trols to the post-translational level.The ma jor dif fer ence, how ever, re gards the se cre tion and as sem bly of functional ma tri ces.In an i mal cells, ECM is formed as a re sult of a co-operative ef fort of groups of cells.In plants, CW of in di vid ual cells are formed only by their re spec tive protoplasts.The pro cess of CW for ma tion is un der very pre cise struc tural and phys i o log i cal as well as organismal con trols.CW of var i ous tissues, and of dif fer ent cells within a tis sue, and even do mains of the walls around in di vid ual cells may be formed in dif fer ent ways, at differ ent speeds and with the use of dif fer ent com po nents (for re view see [9]).

BI O LOG I CAL OR GANI SA TION -A MUL TI TUDE OF PER SPEC TIVES
The dis cus sion on "What is a plant cell?" carried out through out 1991 in The Plant Cell revealed the ex is tence of two op po site ways of think ing about a cell in gen eral and a plant cell in par tic u lar.The first one at tempts to de fine the cell as "a ba sic unit of life" com mon to all liv ing or gan isms.This is ex em pli fied by the def i ni tion of Alberts et al. [3] that cells are "small mem brane-bounded com part ments filled with a con cen trated aque ous so lu tion of chem i cals".Hence, ev ery thing which is outside the plasma mem brane should be con sidered as a prod uct of the cell, but not a part of the cell.This would mean that CW of plants, fungi or bac te ria as well as an i mal ECM should not be in cluded into the def i ni tion of the cell.The op po site ap proach con sid ers the dif fer en ti ated or gani sa tion of cells as a de termi nant of di ver sity ob served in or gan isms belong ing to var i ous King doms.In this re spect, plant or fun gal cell walls con sti tute a char acter is tic fea ture which, when com bined with other fea tures, en able us to dis tin guish a partic u lar type of cell or or gan ism built from such cells as a plant or a fun gus.This ap proach is also more rooted in the tra di tional un derstand ing of cells, par tic u larly in plant research where the en tity en closed within the plasma mem brane is called "a protoplast".As the abil ity to di vide is con sid ered an in dispens able fea ture of a liv ing cell, it is worth to in di cate that plant proto plasts are un able to di vide be fore the for ma tion of a func tional sur round ing wall [15].Con se quently, within this ap proach cell walls are usu ally con sid ered to be a part of plant cell.Al though in such a def i ni tion the ex act re la tion ship be tween the in ner protoplast and the outer CW is not precisely iden ti fied, his tor i cal per spec tive in dicates that CW are the prod uct, but not the part ner of the protoplast.
The dis cus sion on the def i ni tion of the plant cell is em bed ded in a much broader con troversy over the way of in ter pret ing bi o log i cal or gani sa tion of liv ing or gan isms.There is a gen eral agree ment that the phe nom e non of Life orig i nated with the for ma tion of the first cell and that it still is in ti mately re lated to the cell (Virchow: "Omnis cellula e cellula").The dif fi cul ties ap pear when at tempts to ex plain the or gani sa tion of multicellular or gan isms are un der taken.Two the o ries have been proposed, cel lu lar and organismal.The cell theory was orig i nally in tended by Matthias Schleiden for de scrib ing struc tural dif ferences be tween the in ter nal or gani sa tion of plants and an i mals [13].In 1839 this the ory was trans formed by Theodor Schwann into a uni fy ing prin ci ple that all liv ing things are made up of cells -el e men tary units of structure, phys i ol ogy, and or gani sa tion [16].This idea be came one of the foun da tions of mod ern bi ol ogy un der ly ing many other con cepts aimed at the elu ci da tion of bi o log i cal phe nomena.Ba sic ten ets of this the ory state also that each cell is ini tially an in di vid ual of equal morpho log i cal rank and that each multicellular organ ism is an ag gre gate (a "re pub lic") of cells [14,17].At about the same time the organismal the ory has been for mu lated, based on the ob ser va tions that the cell the ory is not well suited to de scribe plant de vel op ment (de Bary: "Die Pflanze bildet Zellen, nicht die Zelle bildet Pflanzen"; cited af ter [14]).Ac cord ing to this the ory, a liv ing thing is a con tin u ous proto plas mic en tity of com plex or gani sa tion which may or may not be par ti tioned into smaller units re cog nised as cells.In any case, such par ti tion ing is a sec ond ary event and if it takes place, the re sult ing units ("cells") are sub or di nate parts of the whole [14,16,17].
It was rightly pointed out by Korn [16] that both the o ries orig i nated as, and still are, struc tural con cepts.The de vel op ment of bi ology and other con tem po rary fields of nat u ral sci ences al lowed other points of view on Life it self or on bi o log i cal or gani sa tion to de velop.Well known ex am ples are: the ther mo dy namic con cept de fin ing cell as an open sys tem or the cy ber netic one de scrib ing cell as a unit of self-control and self-reproduction.It is worth not ing, how ever, that all the o ries are based to a con sid er able de gree on a qual i ta tive core, while the mul ti tude of life forms and functions is a man i fes ta tion of sub tle quan ti ta tive changes and in ter ac tions be tween var i ous mol e cules build ing cells and or gan isms [18].Ac cord ingly, none of those per spec tives of fers a full ex pla na tion of such in ter est ing phe nomena as the tran si tion of one-dimen sional in forma tion, car ried by genes, into three-dimensional ar chi tec ture of a liv ing be ing [9].Although there have been for mu lated first modern con cepts of the cell and bi o log i cal or ganisa tion (e.g.[16,19]), for the sake of clar ity these two struc tural the o ries will be used here to de scribe the pos si ble ex tremes of in ter preta tion.

CELLULARITY -EVO LU TION ARY AND DE VEL OP MEN TAL IM PLI CA TIONS
Both the cel lu lar and organismal the o ries attempt to ex plain the re la tion be tween the organ ism and the cell, be tween the whole and the part.Their ba sic ten ets as well as cor ol laries de rived from them are for mu lated in an exclu sive man ner.This will be il lus trated by an anal y sis of evo lu tion ary and de vel op men tal im pli ca tions.Ac cord ing to the cel lu lar the ory, multicellularity arises as a re sult of the ag grega tion of in di vid ual or gan isms, the cells.This places the em pha sis on the qual ity of build ing units, and the prop er ties of the or gan ism would then be viewed as the sum of the proper ties of many cells [20].Con se quently, de velop men tal pro cesses (on tog eny) would be consid ered as an ef fect of the co-operative ef fort of many cells.On the other hand, the organismal the ory views each or gan ism as a pro toplas mic unit which might be cham bered second arily into in di vid ual parts (sub units) recog nised as cells.Ac cord ingly, the the ory places the ba sic de vel op men tal con trols at the level of the or gan ism and con sid ers ontogenesis as the res o lu tion of the whole into parts.More over, as the em pha sis is put on the or gan ism as a whole, po si tional cri te ria become a pri mary de ter mi nant when ana lys ing re la tion ships be tween the build ing sub units [20].Com par a tive stud ies of plants and an imals re veal that the cell the ory pro vides the best de scrip tion of an i mal de vel op ment.Here, di vi sion of cells in volves com plete sep a ra tion of daugh ter cells en abling their dis tinct mo bility and in de pend ence of be hav iour.The gen era tion of the fi nal three-dimensional shape of, e.g., a mam ma lian em bryo is a re sult of cell mi gra tion.In con trast, cell di vi sion in higher plants in volves the in ser tion of the newly formed cell plate be tween daugh ter protoplasts.How ever, this cell di vi sion is in complete and the cell plate does not fully sep a rate the daugh ter cells, giv ing rise to cy to plas mic and endoplasmic sys tem con nec tions through plasmodesmata.And such dy namic struc tural ar chi tec ture is best de scribed by the organismal the ory (see [9,13,14,17,21]).
The ac cep tance of ei ther the ory in flu ences also our views on the evo lu tion of multicellularity.As the cel lu lar the ory is deeply impli cated in the con cepts of mod ern bi ol ogy, it of fers a fa mil iar per spec tive known from many text books.This the ory views uni cel lu lar or gan isms as prim i tive ("el e men tary") forms of life and multicellular or gan isms as advanced ones.In this way, phy log eny ap pears as a rel a tively straight for ward pro cess leading from uni cel lu lar or gan isms, through a colo nial stage, to multicellular or gan isms charac ter ised by spe ciali sa tion and in de pend ence of their build ing cells.The organismal the ory sug gests an other ex pla na tion.As each liv ing thing is a pro to plas mic whole, uni cel lu lar and multicellular or gan isms are of the same rank, and they shall be con sid ered as nonseptate and septate in di vid u als, re spec tively.In the course of evo lu tion, even tual cham ber ing (cellu larization) of the pro to plas mic unit en abled the di vi sion of la bour among var i ous subunits.In par tic u lar, the uti li sa tion of me chan ical ben e fits re sult ing from the in clu sion of CW into their bod ies per mit ted plants to success fully colo nise land (or more prop erly, air) [14,22,23].
The pres ent state of knowl edge in di cates that multicellularity evolved in de pend ently in a few evo lu tion ary lin eages.Both plants and an i mals started with the same cel lu lar toolkit as their last com mon an ces tor, most prob a bly a uni cel lu lar eukaryote con tain ing a mi tochon drial endosymbiont and func tional el ements of the cytoskeleton [24].They di verged how ever, about one bil lion years be fore they be came multicellular or gan isms [25].As multicel lu larity pres ents a clear evo lu tion ary ad van tage [22], it emerged in both lin eages but the mech a nisms by which this was achieved ap pear to be lin eage-specific.The differ ent life styles of plants and an i mals seem to be the con se quence of the so lu tions uti lised in both King doms to solve the prob lem: how to main tain the chem i cal com po si tion of cells and en able cell-to-cell com mu ni ca tion and exchange of ma te ri als at the same time?In an imals, ac tive ion trans port sys tem, con trol ling chem i cal gra di ents across plasma membranes, has been em ployed [13].Di rect phys ical con tact be tween neigh bour ing cells is achieved in one of three non ex clu sive ways: tight junc tions, desmosomes, and gap junctions [22].Only the lat ter two en able the exchange of sol utes con tain ing small me tab olites and ions.It should be noted, how ever, that due to the com plete sep a ra tion of daughter cells, all forms of intercellular con tacts between an i mal cells are formed de novo, of ten fol low ing mi gra tion of cells.Thus, an i mals are truly multicellular in the sense given by the cel lu lar the ory.In plants, os mot i cally driven wa ter in flux cre ates intracellular hy dro static pres sure (turgor) which is coun ter acted by me chan i cally and struc tur ally sta ble cell walls.The ac qui si tion of such a reg u la tory mech a nism has im por tant evo lu tion ary con sequences re flected in plants' life styles.As summa rised by Pe ters et al. [13], these could be listed as fol lows: 1) the hy dro static pres sure across the plasma mem brane ex ceed ing 2 MPa could be used for me chan i cal sta bili sation of plant bod ies, 2) func tional cell walls become in dis pens able el e ments of plant cells, 3) proto plasts sur rounded by the walls are unable to move, and thus plants have to be sessile, 4) the pres ence of a her metic ma trix around proto plasts lim its the pos si bil i ties of en ergy and nu tri ent ac qui si tion, and thus fungi are saprophytic and plants are autotrophic, and 5) the ex is tence of the walls fixes the po si tion of each cell in re la tion to its neigh bours [13].These con straints have been to some ex tent over come, how ever, by a new mech a nism of cell di vi sion lead ing to the forma tion of the cell plate and the de vel op ment of a new type of intercellular com mu ni ca tion through plasmodesmata [26].The ef fect was a true symplasmic con ti nu ity be tween plant cells, en abling them to ex change not only low mo lec u lar mass sol utes, but also pro teins and in for ma tional macro molecules.As the number and lo ca tion of plasmo desmatal con nections could be ac tively reg u lated, this al lowed for the ex er tion of a new, supracellular control of gene ex pres sion in symplastic do mains dur ing plant de vel op ment [27].

"THE PLANT BODY" -AN ORGANISMAL CON CEPT OF PLANTS AND PLANT CELLS
The fun da men tally dif fer ent ten ets and result ing cor ol lar ies of both the o ries seem to pre clude the ap pli ca tion of mod els elab o rated in an i mal sys tems for the ex pla na tion of, e.g., plant de vel op ment.The same is true for the re spec tive exocellular ma tri ces.Al though some of the me chan i cal and func tional char acter is tics of ECM and CW seem to be sim i lar (see above), in gen eral they ap pear to be noncomparable, and the mech a nisms by which plants and an i mals in ter act with the out side world seem also to be quite dif fer ent [28].The re la tions be tween plant cells and their walls are much eas ier to com pre hend in terms of the organismal the ory which views CW as a means for cham ber ing plants into smaller sub units re cog nised as cells [23].Thus, CW could be de fined both in re la tion to the proto plasts that pro duce them, and in re lation to the whole plant.Con se quently, CW are in dis pens able fea tures of plant cells, not only prod ucts, but also part ners of plant protoplasts.As they are both cham ber ing de vices and de ter mi nants of the bound ary of the organ ism, their biosynthesis and func tion are also un der or gan is mic con trol [9].
Cell walls seem to be one of the crit i cal factors af fect ing cel lu lar fate and de vel op ment.Plant proto plasts re quire a sur round ing, functional CW (or other ma trix) in or der to di vide at all [15], and the com po si tion of such ma trix de ter mines the fate of the em bed ded protoplasts (e.g.[29]).The same is true for, e.g., a newly formed na ked Fu cus zy gote, where the first cel lu lar event is the for ma tion of the walls.These CW are then re quired for the fixa tion of the polar ised cell axis and next function as a res er voir of morphogenic in for mation which can be ac tively de liv ered back to the protoplast to di rect its fur ther de vel opment [30].In plant cells, through the con tinuum of cell walls -plasma mem branecytoskeleton, CW pro vide an an chor sta bi lising the "tensegrally" or gan ised cytoskeletal net work [31][32][33] thus pro vid ing the scaf folding onto which the cel lu lar struc tures could be at tached [8,34,35].In this way, the con tinuum con sti tutes a ma jor source of ar chi tectural bi o log i cal in for ma tion and an en vi ronment for a part of the bio chem i cal ma chin ery [32].
At the higher level of or gani sa tion, the structural and func tional CW con tin uum forms an ar chi tec tural frame work en sur ing the de velop men tal in teg rity of plants.It pro vides the means and the route for inter-and intracellular sig nal ling events.Changes in wall mechan ics and/or changes in turgor pres sure will phys i cally af fect the ge om e try of cells [36] lead ing to changes in the fate of in di vid ual cells.Ex per i ments with cal luses and cells cultured in vi tro in di cate that the phys i cal en viron ment found in planta is cru cial for nor mal plant de vel op ment [37,38].More over, they dem on strate also that such me chan i cal stimuli are de ci sive for the proper or gani sa tion of cel lu lar met a bolic net works [39], re flected in CW com po si tion as well [40].On the other hand, changes in biomechanical prop er ties of the walls are im por tant for cell growth and for or gan gen er a tion [41].Local ised expansin-depend ent weak en ing of cell walls in duces the for ma tion of root hairs [42], while local ised ap pli ca tion of expansins onto the stem api cal meristem in duces gen er a tion of leaf-like outgrowths and re ver sal of the di rec tion of phyllotaxis [43].When com bined with the devel op men tal con trol of symplasmic con ti nu ity through the reg u la tion of the ex tent of plasmodesmatal com mu ni ca tion [26], these mech a nisms pro vide the ba sis of plant morpho genesis.
This par tic u lar du al ity of cell wall func tioning in re la tion to the whole or gan ism and its parts -the cells, seems to be a unique fea ture of plants.It also pro vides the stron gest ar gument for the organismal na ture of plants [17].Using this point of view, cell walls, al though ex ter nal to proto plasts, be come in te gral parts of plant cells.These cells, how ever, lose their in di vid u al ity and are treated as sub or di nate parts of the or gan ism, the plant.Nev er theless, the walls func tion as the de ter mi nants of not only cel lu lar, but also or gan is mic boundaries.The par tic u lar mech a nism of cell di vision, used for in ter nal cham ber ing, does not dis rupt the pro to plas mic con ti nu ity of a plant giv ing rise to two con tinua, re cog nised early by plant bi ol o gists: the symplast and the apoplast."The Plant Body" con cept [9] treats thus a plant as a unit filled with sev eral in tertwin ing "net works" with cell walls pro vid ing a liv ing ar chi tec tural scaf fold ing.These networks in clude: (1) the symplast un der stood as a cy to plas mic con ti nu ity of plant body lined with a con tin u ous plasma mem brane, (2) the endomembrane sys tem, (3) the cytoskeletal con tin uum, (4) com mu ni ca tion con tact points pro vided by the cell wall -plasma mem brane -cytoskeleton con tin uum [9].Sev eral mod els of the or gani sa tion of plant cells, such as the plasmalemmal con trol cen tre [44], the endomembrane sheath [45], and the cell body and the cell pe riph ery com plex [8,46], have been pro posed re cently.They were ac com pa nied by two other mod els in di cat ing the pos si ble supracellular na ture of plants due to the intercellular com mu ni ca tion through plasmodesmata [27] or the ex is tence of a mo bile endoplasmic net work span ning the whole plant [47].All these con cepts are rooted in the cel lu lar the ory of bi o log i cal or gani sa tion.The lat ter two mod els could, how ever, be treated as a means for over com ing se ri ous dif fi cul ties in in ter pret ing some ex per i men tal data concern ing plants when us ing this the o ret i cal basis.More over, in all those con cepts cell walls are not pres ent.In that re spect, the pro posed idea of "The Plant Body" pro vides a broader per spec tive of bi o log i cal or gani sa tion and in -cludes the unique du al ity of cell wall func tioning in plants.
As it was men tioned be fore, both the cel lu lar and the organismal the ory orig i nated as structural ideas, and now they pro vide prob a bly the most ex treme ways of in ter pret ing data concern ing bi o log i cal or gani sa tion of liv ing beings.At the pres ent state of knowl edge, each of the the o ries ap pears to be most suit able to de scribe or gan isms be long ing to dif fer ent King doms: the cel lu lar the ory for an i mals, and the organismal the ory for plants and most prob a bly also fungi.Both the o ries, however, have their own lim i ta tions and Life it self seems to be more com pli cated.For ex am ple, plants are able to de ter mine the size and shape of their or gans, and they reg u late accord ingly the num ber and po si tions of cell divi sions.This phe nom e non could be eas ily explained by the organismal the ory.How ever, such pro cesses are also ob served dur ing an imal de vel op ment which in di cates that at least some level of organismal con trol ex ists also in these multicellular or gan isms [48].Thus, there is still a need for much more gen eral unify ing con cept(s) which would be able to embrace the ten ets and the re sult ing cor ol lar ies of both the o ries.Based on the rapid prog ress of mod ern bi ol ogy, these con cept(s) would most prob a bly orig i nate as cel lu lar [19] or infor ma tional [16] ideas ex plain ing the won derful mul ti tude of Life forms.