Overexpression of Poplar Cellulase Accelerates Growth and

Overexpression of Poplar Cellulase Accelerates Growthand Disturbs the Closing Movements of Leavesin Sengon1[OA]

Sri Hartati, Enny Sudarmonowati, Yong Woo Park, Tomomi Kaku, Rumi Kaida,Kei’ichi Baba, and Takahisa Hayashi*

Research Centre for Biotechnology, LIPI, Cibinong 16911, Indonesia (S.H., E.S.); and Kyoto University,RISH, Uji 611–0011, Japan (Y.W.P., T.K., R.K., K.B., T.H.)

In this study, poplar (Populus alba) cellulase (PaPopCel1) was overexpressed in a tropical Leguminosae tree, sengon (Paraserianthesfalcataria), by the Agrobacterium tumefaciens method. PaPopCel1 overexpression increased the length and width of stems with largerleaves, which showed a moderately higher density of green color than leaves of the wild type. The pairs of leaves on thetransgenic plants closed more slowly during sunset than those on the wild-type plants. When main veins from each genotypewere excised and placed on a paper towel, however, the leaves of the transgenic plants closed more rapidly than those of the wild-type plant. Based on carbohydrate analyses of cell walls, the leaves of the transgenic plants contained less wall-bound xyloglucanthan those of the wild-type plants. In situ xyloglucan endotransglucosylase activity showed that the incorporation of wholexyloglucan, potentially for wall tightening, occurred in the parenchyma cells (motor cells) of the petiolule pulvinus attached to themain vein, although the transgenic plant incorporated less whole xyloglucan than the wild-type plant. These observationssupport the hypothesis that the paracrystalline sites of cellulose microfibrils are attacked by poplar cellulase, which loosensxyloglucan intercalation, resulting in an irreversible wall modification. This process could be the reason why the overexpressionof poplar cellulase both promotes plant growth and disturbs the biological clock of the plant by altering the closing movements ofthe leaves of the plant.

Overexpression of plant cellulase in plants does notlead to a lack of cellulose; rather, it modifies the cellwalls by trimming off disordered Glc chains from themicrofibrils. This action has been demonstrated inArabidopsis (Arabidopsis thaliana) overexpressing pop-lar (Populus alba) cellulase (Park et al., 2003). Trans-genic poplar overexpressing Arabidopsis cellulase(cel1) had longer internodes and longer fiber cells(Shani et al., 2004). Overexpression does not increasexyloglucan depolymerization (Harpster et al., 2002),because the reaction efficiency of plant cellulase forxyloglucan is very low compared with that for amor-phous 1,4-b-glucans (the amorphous regions of cellu-lose microfibrils). This is confirmed by the fact thatthe enzyme has high reactive efficiency for artificialsubstrates such as carboxymethylcellulose, phospho-

swollen cellulose, and (1/3),(1/4)-b-glucan (Nakamuraand Hayashi, 1993). Nevertheless, the trimming ofmicrofibrils by cellulase might solubilize some xylo-glucan that would otherwise have been intercalatedwithin the disordered paracrystalline domains of themicrofibrils (Hayashi, 1989). This kind of cell wallmodification in Arabidopsis causes an increase in thesize of cells in the petioles and blades of rosette leavesand stems (Park et al., 2003). Based on relative loadextension curves, a cross-linking component appearedto be reduced in the walls of the transgenic plantscompared with those of the wild-type plants. Thedecreased cross-linking component is attributable to adecreased amount of tethering xyloglucan and couldin turn accelerate growth by increasing plastic ex-tensibility under turgor pressure. Thus, the overex-pression of cellulase could affect wall dynamics,particularly the turgor-related movements of plantorgans such as the opening and closing movements ofleaves and stomata, et cetera.

Poplar cellulase cDNA with 35S promoter has beenused in sengon (Paraserianthes falcataria) because theoverexpression of poplar cellulase in Arabidopsis andpoplar produced more visible effects in their leaves(Ohmiya et al., 2003; Park et al., 2003). In these species,leaf length and width were increased to the same extentas the length of the blade and petiole. The question,therefore, is whether Leguminosae plants that overex-press cellulase (in this case, sengon) show any phenotypefor leaf movements related to the walls of motor cells.

1 This work was supported by the Program for the Promotion ofBasic Research Activities for Innovative Biosciences and by JSPSKAKENHI (grants nos. 19208016 and 19405030). This work is alsopart of the outcome of the JSPS Global COE Program (E–04): InSearch of Sustainable Humanosphere in Asia and Africa.

* Corresponding author; e-mail [email protected] author responsible for distribution of materials integral to the

findings presented in this article in accordance with the policydescribed in the Instructions for Authors (www.plantphysiol.org) is:Takahisa Hayashi ([email protected]).

[OA] Open Access articles can be viewed online without a sub-scription.

www.plantphysiol.org/cgi/doi/10.1104/pp.108.116970

552 Plant Physiology, June 2008, Vol. 147, pp. 552–561, www.plantphysiol.org � 2008 American Society of Plant Biologists www.plantphysiol.orgon March 25, 2018 - Published by Downloaded from

Copyright © 2008 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

Sengon belongs to the subfamily Mimosoideae ofLeguminosae and is native to Haiti, Indonesia, andPapua New Guinea. The sengon variety used forreforestation is the fastest growing tree in industrialforests. It even thrives in marginal land, where itgrows symbiotically with nitrogen-fixing Rhizobiumand phosphorus-promoting mycorrhizal fungi. There-fore, it is a suitable species for industrial timber estatesin southeast Asian countries (Binkley et al., 2003;Shively et al., 2004; Kurinobu et al., 2007; Siregaret al., 2007). The sengon tree typically gains 7 m inheight per year and reaches a mean height of 25.5 mand a bole diameter of 17 cm after 6 years. After 15years, it reaches 39 m in height and 63.5 cm indiameter. The tree is useful not only for timber mate-rial but also for pulp and paper. Due to its soft timberand leaves that can be used as animal feed (Merkelet al., 2000), the tree has a wider range of end uses thanAcacia species (Otsamo, 1998). Therefore, it is expectedto be one of the most useful tree species for industrialforests. However, despite attempts with tree cuttings,tissue culture techniques with multiple propagations,and stable gene transfer using Agrobacterium tumefa-ciens, attempts to propagate the tree clonally havefailed. In this article, we demonstrate, to our knowl-edge for the first time, the production of transgenicsengon.

Cellulose is an important component of plants andserves as the most abundant biopolymer on earth, withabout 100 billion tons produced annually. In addition,it is a significant biological sink for CO2. It has beensuggested that cellulases may have originally beeninvolved in either the repair or arrangement of cellu-lose microfibrils during their biosynthesis, rather thanin cellulose degradation (Hayashi et al., 2005). It hasalso been reported that membrane-bound cellulase(Korrigan) is required for cellulose biosynthesis (Nicolet al., 1998), but nothing is known about its role incellulose biosynthesis.

In this study, we used the overexpression of poplarcellulase in the sengon tree in order to increase itsgrowth rate. We hope, thereby, to increase the plant’sproduction of raw material, not only for timber, pulp,and paper, but also for use as a biofuel (Ragaukas et al.,2006). Ultimately, experiments like this that result in anincreased deposition of cellulose in the stem couldproduce the fastest growing tree in the world.

RESULTS

Transformation

Two-week-old hypocotyls germinated from seedswere cut into 2- to 4-mm-long stems, the explants ofwhich were used for transformation in the samemanner as leaf discs for the Agrobacterium-mediatedtransformation. The explants from the cut hypocotylsformed a callus-like tissue, which was followed bygreen color and shoot formation on Murashige and

Skoog (MS) medium containing 4 mM benzylamino-purine. The transformants were selected in MS me-dium containing kanamycin.

During several transfers of explants to fresh me-dium each month, shoots were induced by the addi-tion of benzylaminopurine (Fig. 1A) under light (4,000lux). Benzylaminopurine alone was used as a planthormone to induce shoots because auxin did not affectthe formation of callus tissue or shoots in the presenceof benzyladenine (Bon et al., 1998).

During direct adventitious shoot formation, partsof the explants turned green and produced greennodular structures to form adventitious buds at theapical ends. The adventitious shoots (5 mm long) weretransplanted into the medium in the absence ofbenzylaminopurine. After the shoots elongated to 3 cm(Fig. 1B), they were again transplanted into fresh MSmedium in the absence of any plant hormone for theinduction of roots (Fig. 1C). No plant hormone wasused for rooting because auxin and cytokinin preventedthe induction of roots (Bon et al., 1998).

Pinnate leaflets were formed during shoot elonga-tion and root formation, although young trees andshoot apical meristems in adult trees form pinnateleaves. About 30 shoots were regenerated from 400cocultivated explants; 10 of these shoots producedroots. Ultimately, seven independent seedlings wereobtained. Shoots and roots were also induced from theyoung shoots of wild-type plants in the presence andabsence of benzylaminopurine, respectively (Fig. 1D).

It should be noted that it took about 5 to 6 months toproduce transgenic seedlings and about 2 to 3 monthsto produce wild-type seedlings.

Cellulase Expression

To study the effects of cellulase on cell wall structureand growth, we generated transgenic sengon plantsthat expressed poplar cellulase (PaPopCel1) underthe control of a constitutive promoter. To assay theexpression of the transgene, we performed reversetranscription-PCR Southern-blot analysis of mRNAsderived from small sections excised from the petio-lule pulvinus (Fig. 2A). PopCel1 mRNA was accumu-lated in transgenic lines 1 to 7 (trg1–trg7), and weaksignals were detected in trg4 to trg7. Also, we used anantibody against a 15-amino acid sequence (163-CWERPEDMDTPRNVY-167) for the PaPopCel1 geneproduct (Ohmiya et al., 2003).

In each transgenic line (trg1–trg7), the antibodyrecognized a single, 50-kD band on a western blot,present in the petiolule pulvinus, running at a positioncorresponding to the expected and actual size of themature cellulase (Fig. 2B). No signal was detected inthe wild-type plants.

In the pulvinus attached to the veins of the trans-genic plants, cell wall fractions showed cellulase ac-tivity that was approximately 1.05- to 5.25-fold higherthan that of the wild type (Fig. 2C). The activity ofcellulase was also assessed by measuring soluble cello-

Overexpression of Poplar Cellulose

Plant Physiol. Vol. 147, 2008 553 www.plantphysiol.orgon March 25, 2018 - Published by Downloaded from



oligosaccharides, which are presumably released bythe enzyme (Fig. 2D). These oligosaccharides accumu-lated in the transgenic plants, as all of the transgenicplants were found to contain far more oligosaccha-rides than the wild-type plants contained. The amountof oligosaccharides was closely related to cellulaseactivity levels in each of the seven nic lines. Thus, thelevels of expression and activity varied among thetransgenic plants: they were relatively high in trg1 totrg3 and relatively low in trg7, although a trace ofcello-oligosaccharides were detected even in trg7. Thiswas probably because the poplar cellulase expressedwas discretely localized in the cellulose microfibrils ofthe apoplastic spaces.

Based on the carbohydrate analyses of cell walls, itappears that the petiolule pulvinus and the main veinin the transgenic plants contained less wall-boundxyloglucan than those in the wild-type plants (Table I).Increased cellulase activity in the wall did not decreasethe levels of cellulose; rather, cellulose content perplant increased with plant growth (i.e. cellulose permilligram dry weight was not changed). The methyl-ated sugars due to the minor components consistedof 4-linked Xyl, 4-linked Gal, and 4-linked Man at aconstant proportion in both the transgenic and thewild-type plants (data not shown). These methylatedsugars are probably derived from xylan, galactan, andmannan at a ratio of 4.7:3.2:1. Therefore, the transgenicplants differ from the wild-type plants only in theamount of xyloglucan present in the cell walls.

Admittedly, there is no correlation between cellulaseactivity (Fig. 2B) and the extent of xyloglucan solubi-lization (Table I). Nevertheless, the levels of solublecello-oligosaccharides, which could be related to invivo cellulase activity, were closely related to cellulaseactivity levels across the seven transgenic lines andwere consistently higher in the transgenic plants thanin the wild-type plants, which again corresponds tocellulase activity.

Growth Response of Transgenic Plants

We generated seven independent transgenic sengonlines, four of which (trg1–trg4) grew significantlybetter than the wild type, although the overall mor-phology of these transgenic plants was similar to thatof the wild type (Fig. 3). Two other transgenic lines(trg5 and trg6) grew slightly better than the wild type,and one (trg7) grew about as well as the wild type.Based on the expression of the transgene, the four linesthat showed a high growth rate (approximately 20-cmstem length) were selected for further analysis.

The transgenic sengon plants (trg1–trg4) grew fasterthan the wild-type plants, although the young seed-lings (less than 30 cm in height) of both types some-times grew at the same rate. The stems of the transgenicplants elongated faster than those of the wild-typeplants and had larger diameters (Fig. 3). The overallmorphology of the transgenic plants was similar tothat of the wild type (Fig. 4A). As with stem growth,

Figure 1. Regenerating shoots and roots in transgenic and wild-typeplants. A, Regenerating shoots of the transgenic plants. Arrows indicateadventitious buds. Bar 5 3 mm. B, Growth of regenerated transgenicshoots. The shoots had pinnate leaflets during elongation. Bar 5 2.0cm. C, Regenerating transgenic roots. The plantlets producing roots hadpinnate leaves. Bar 5 1 cm. D, Regenerated plantlets of wild-typeplants. Bar 5 1 cm.

Hartati et al.

554 Plant Physiol. Vol. 147, 2008 www.plantphysiol.orgon March 25, 2018 - Published by Downloaded from



the leaves of the transgenic plants were greener andlarger than those of the wild type (Fig. 4B). In bothtypes, the length and width of the leaves increased tothe same extent as the length of the main and minorveins, and this increase in size was even distributedamong all leaves of the plant. Parenchyma cells inleaves of both types were identified in the central partof the petioles. Finally, both palisade and epidermalcells were a little larger in the leaves of the transgenicplants than in the leaves of the wild-type plants (datanot shown).

Figure 5 shows the times at which the transgenicand wild-type plants started leaf opening before sun-rise and completed leaf closure during sunset. Therewas no difference between transgenic and wild-typeplants in the starting time of leaf opening (Fig. 5A);the leaves in both types of plants started openingaround midnight and completed opening by 5:25 AM.In contrast, the leaf pairs closed more slowly in thetransgenic plants than in the wild-type plants. Thisdifference was visible in the upper, middle, and lowerparts of the petioles (Fig. 5B). In the transgenic plants,older leaves located at the middle and bottom part ofthe stems started closing 30 min later than the corre-sponding leaves in the wild-type plants. Likewise, inthe transgenic plants, leaves at the bottom part of the

stem completed closing more than 1 h later than thecorresponding wild-type leaves. However, both the trans-genic and wild-type plants closed their leaves within afew minutes when they were placed in darkness atnoon (data not shown). In spite of this change to theirnormal conditions, they also started opening theirleaves at almost the same time (midnight) and finishedopening completely by 5:25 AM (Fig. 5A).

Interestingly, the transgene had the opposite effecton excised main veins. When the main vein withleaves was excised and placed on a paper towel, thepairs of leaves from the transgenic plants closed fasterthan those of the wild-type plants (Fig. 6). The youngerleaves in the upper part of each plant started closingimmediately; leaves in the middle part completedclosing more than 1 h later; and leaves in the lowerpart completed closing 2 h later. The older leaves in themiddle and lower parts of the transgenic plants com-pleted closing more than 30 min earlier than those ofthe wild-type plants. Thus, when the main vein wasexcised, closing was faster in the leaves of the trans-genic plants than in those of the wild-type plants,whereas in vivo, closing was slower in the transgenicleaves.

Xyloglucan endotransglucosylase activity was de-tected in situ on the transverse sections of the petiolulepulvinus attached to the main vein using either fluo-rescent whole xyloglucan (50 kD) or fluorescent xylo-glucan heptasaccharide (XXXG; Takeda et al., 2002).Whole xyloglucan was incorporated into the paren-chyma cells (motor cells) of the pulvinus in bothgenotypes, although the transgenic plants incorpo-rated less xyloglucan than the wild-type plants (Fig. 7).The incorporation of whole xyloglucan into the vas-cular bundle of the main vein was also greater in thewild type than in the transgenic plants, although it wasincorporated into the connection between the petiolulepulvinus and the main vein at the same rate in bothgenotypes.

In contrast, the incorporation of XXXG, either intothe parenchyma cells of the pulvinus or into thevascular bundle of the main vein, was not observed

Figure 2. Analysis of PaPopCel1 expression in the main vein withpetiolule pulvinus. A, Reverse transcription-PCR Southern-blot anal-ysis of PaPopCel1 mRNA. The relative amounts of mRNAs by reversetranscription-PCR analysis at 15 cycles are shown. B, Western-blotanalysis of cell wall proteins. Five micrograms of protein was used foreach. C, Level of cellulase activity. D, Level of cello-oligosaccharides.Three separate main veins for each plant were used for the determi-nation.

Table I. Xyloglucan and cellulose content in cell walls

Xyloglucan was determined from 24% KOH-soluble fractions. Threeseparate main veins for each plant were used for the determination. SE

values were calculated from three samples per line.

PlantPetiolule Pulvinus Main Vein

Xyloglucan Cellulose Xyloglucan Cellulose

mg mg21 dry weight

Wild type 41.1 6 2.2 370 6 32 16.0 6 2.2 454 6 23trg1 9.1 6 1.4 371 6 18 8.1 6 0.9 460 6 55trg2 8.3 6 1.3 372 6 20 9.1 6 1.2 455 6 32trg3 8.8 6 1.8 378 6 21 8.2 6 0.8 461 6 31trg4 10.4 6 1.5 377 6 28 8.7 6 1.6 451 6 43trg5 11.6 6 1.8 374 6 32 8.9 6 0.9 452 6 53trg6 10.1 6 2.1 385 6 38 7.9 6 1.3 460 6 47trg7 12.0 6 2.8 366 6 24 8.2 6 1.4 458 6 44





in either genotype. These results show that the walls ofthe parenchyma cells (motor cells) can incorporatewhole xyloglucan but not XXXG. The level of incor-poration was higher in the wild-type plants than in thetransgenic plants, probably because the walls of thetransgenic plants contain less endogenous xyloglucanmolecules to act as donors. Another possible explana-tion is that the increased cellulase in the transgenicplants cleaves the glucan chains to which xyloglucanbinds, so that the glucan chains are washed out,making it more difficult for xyloglucan molecules toattach firmly to the wall.

DISCUSSION

We have succeeded in producing transgenic sengonplants for the first time. Seven independent shootsregenerated from 400 cocultivated explants were dem-onstrated to be transgenic; this represents a trans-formation frequency of 1.75%. So far, we have notsucceeded in either multiple propagation of the trans-genic shoots or clonal propagation by cutting theirstems. Nevertheless, shoots and roots were induced in

MS medium in both the presence and absence of 4 mM

benzylaminopurine. The shoots always formed fromthe callus-like tissues of explants from the cut hypo-cotyls, although the shoots and roots formed directlyfrom the explants of hypocotyls in the case of Acaciasinuate (Vengadesan et al., 2006). Now sengon can be

Figure 3. Effects of PaPopCel1 transgenes on stem growth. A, Increasein stem length. B, Increase in stem diameter. Black circles, trg1; whitecircles, trg2; black squares, trg3; white squares, trg4; white triangles,wild type.

Figure 4. Wild-type and transgenic (trg1) plants. The wild-type andtransgenic plants are shown on the left and right, respectively, at 390 dafter adventitious shoot formation. A, Whole plants. Bar 5 10 cm. B,Leaves. Bar 5 1.5 cm.

Hartati et al.




genetically modified to exhibit desirable traits, such aseasy degradation of cellulose microfibrils as well aspathogen and insect resistance, qualities that are dif-ficult to achieve by traditional breeding. It would alsobe expected to increase the transformation frequency,particularly the induction of roots from shoots. Theincreased frequency of root induction should acceler-ate the biotechnological application of sengon as abiomass resource. It should be noted that wild-typesengon is already one of the fastest growing treespecies in the world; the transgenic sengon developedin this study grows even faster and may ultimately bethe fastest growing tree in the world.

Transgenic sengon overexpressing poplar cellulase(PaPopCel1) increased the size of leaves by increasingcell volume, as other authors have demonstrated inArabidopsis leaves (Park et al., 2003). This phenome-non has been attributed to an increase in the specificactivity of cellulase, similar to the increase observed intransgenic Arabidopsis and poplar (Park et al., 2003;Shani et al., 2004). Nevertheless, no bulk degradationof cellulose was confirmed, because the amount ofcellulose per dry weight was nearly the same in thetransgenic and the wild-type plants. Therefore, we

believe that cellulase promotes increased growth intransgenic sengon by trimming off disordered Glcchains from cellulose microfibrils, where some xylo-glucan would otherwise be solubilized. Residual xy-loglucan can adhere tightly to cellulose microfibrils,perhaps as a monolayer coating the surface, but thetransgenic plants contained less xyloglucan bound tocellulose microfibrils. This agrees with a previousfinding that a decrease in xyloglucan tethers acceler-ates cell elongation (Takeda et al., 2002).

We have found that leaf movements are somehowdisturbed by the transgene expression. The transgenicplants opened their leaf pairs at the same time as thewild-type plants (midnight) but started closing theirleaves 30 min later and completed closing them morethan 1 h later (Fig. 5). Nevertheless, transgenic sengonstill retains a type of circadian rhythm in the openingand closing movements of leaves, although sterilizedsengon (in vitro culture) did not show closing move-ments at night, even if the plant was placed in dark-ness. The opening and closing movements of leavesoccur in the leaf bases (petiolule pulvinus) and arecaused by the expansion and contraction of the motorcells. Motor cells occupy most of the space in the

Figure 5. Leaf movements of upper, middle, andlower parts of petioles in the wild-type and transgenicplants. A, Opening. All of the leaves are open (left).Bar 5 4 cm. B, Closing. Closing leaves are distin-guishable as yellow and white leaves (left). Bar 5 4cm. The lower part of the petiole was defined as thesecond petiole from the bottom, the middle part as thefifth or sixth petiole from the bottom, and the upperpart as the ninth or tenth (or newest) petiole. All of theleaves in the petiole were observed to determine theopening and closing times of leaf pairs from start tofinish. SE values were calculated from four lines oftrg1, trg2, trg3, and trg4.





pulvinus, surrounding the central ring of the vascularbundle. The expansion and contraction of these cells isbelieved to result from changes in their turgor, whichis regulated in turn by the flow of K1 ions across thecells’ thin walls.

In the case of transgenic sengon that overexpressespoplar cellulase, an increase in wall plasticity maycause changes in normal leaf movements, because themovements correspond to the balance between turgorpressure and wall pressure. Therefore, the turgorpressure in transgenic sengon motor cells might in-crease during the day and decrease at night, while thewall pressure remains constant. In the case of pea(Pisum sativum) hypocotyls, the wall pressure in grow-ing cells is decreased during elongation, while theturgor pressure remains constant. Since the walls ofthe motor cells in the pulvinus incorporated wholexyloglucan but not XXXG, wall tightening rather thanloosening could be required to prevent the expansionof the cells at a cut surface (Takeda et al., 2002).

In this case, the xyloglucan endotransglucosylaseactivity could occur between high-molecular-size xy-loglucans in the cell walls, where the enzyme has bothenzyme-donor and enzyme-acceptor complexes. Uedaand Nakamura (2007) suggested that leaf movementsare controlled by the balance of the concentrations ofchemical substances involving an aglycon or a gluco-

side, which induce opening and closing, respectively.Nakanishi et al. (2005) showed that in excised leaves ofOxalis corymbosa, the opening movement was sensitiveto blue light but not to red light. It should be noted thatcellulase tends to affect the down-regulation of leafmovements, whereas chemical substances and lighttend to affect the up-regulation of leaf movements.

Darwin (1880) postulated that the nyctinastic move-ments of plants occurred in order to hide the uppersurface of the leaves at night, because the upper leafepidermis was believed to be a sense organ for adap-tive responses. Bunning and Moser (1969), on thecontrary, suggested that adaptive leaf movementsoccurred to protect the plant’s photoperiodic rhythmagainst radiation from moonlight rather than againstradiation from the leaf surfaces into the sky. Thiscannot be considered a satisfactory explanation in thecase of sengon, however, because pairs of sengonleaves start to open at night. It should be noted thatlight does not induce the opening movements duringnormal growth.

It is possible that the transgenic sengon plants haveslightly higher photosynthetic capability than thewild-type plants, since the leaf pairs of the transgenicplants close more slowly than those of the wild-typeplants. The sun always sets in Indonesia by 6:40 PM,while the transgenic plants keep their leaf pairs open

Figure 6. Closing movements of leaves whose mainvein was excised. Leaves in the vein at the upper,middle, and lower parts of petioles are shown from leftto right. SE values were calculated from four lines oftrg1, trg2, trg3, and trg4. Bar 5 2 cm.

Hartati et al.




for about 1 h after the normal sunset time. We areperforming detailed analyses of the plant’s leaf move-ments to determine their photosynthetic efficiency.

MATERIALS AND METHODS

Transgenic Constructs

The PaPopCel1 cDNA fragment was excised from pBluescript SK1 by

digestion with BamHI and KpnI (Nakamura et al., 1995). The GUS-coding

sequence of pBE2113 was removed from the fragment by digestion with

BamHI and SacI, and the cDNA fragment was inserted into the pBE vector

between the cauliflower mosaic virus 35S promoter and the Agrobacterium

tumefaciens NOS transcription terminator. The chimeric construct was intro-

duced into the disarmed Agrobacterium strain LBA4404.

Plant Transformation

Agrobacterium carrying plasmid-harboring poplar (Populus alba) cellulase

cDNA (PaPopCel1) and selectable marker NPTII genes were cultured in YES

medium (0.1% yeast extract, 0.5% polypeptone, 0.5% Suc, and 0.0246%

MgSO4) containing 50 mg mL21 kanamycin. The bacterial suspension was

pelleted and resuspended with sterilized water. Seeds were germinated for

2 weeks to produce hypocotyls elongating 1 to 2 cm in length.

Pieces of sengon (Paraserianthes falcataria) stems (2–4 mm in length) excised

from their hypocotyls were dipped in diluted Agrobacterium solution (optical

density at 600 nm 5 0.1) for 5 to 10 min and put on sterile filter paper, then

cocultivated for 1 d on half-strength hormone-free MS medium. Next, the

pieces of stems were placed on MS agar medium containing 600 mg mL21

kanamycin for 2 weeks, after which they were washed with a water solution

containing 400 mg mL21 Claforan. The stems were cultured several times by

transplantation on MS agar medium containing 600 mg mL21 kanamycin and

4 mM benzylaminopurine for 2 to 4 months, under 14-h-day (4,000 lux)/10-h-

night cycles, after which they were placed on a medium containing 300 mg

mL21 kanamycin for 2 weeks. Shoots 5 mm in length were excised from the

medium and cultured again on a medium containing 300 mg mL21 kana-

mycin in the absence of plant hormone. Roots were then formed in the

medium for 2 to 4 weeks, and the plantlets were further cultured for growth

in the medium for 2 months. Plantlets approximately 10 to 15 cm long were

planted in soil.

RNA Isolation and Reverse Transcription-PCR Analysis

Total RNA was isolated from the main vein with petiolule pulvinus

(Ohmiya et al., 2000). The first-strand cDNA was synthesized using 5 mg of

total RNA at 42�C for 1 h using oligo(dT) (n 5 20) and SuperScript II reverse

transcriptase (Gibco BRL). PCR was performed with final volumes of 20 mL

containing 0.5 unit of cDNA polymerase mix (Clontech), 0.2 mM dNTPs, 3.5

mM Mg(OAc)2, and 0.4 mM gene-specific primers. The forward primer was

5#-CACCACGCAATGTGTACAAAGTAACCATC-3# (nucleotide positions 512–

540), and the reverse primer was 5#-GGGTGTATATTGGGCCTGGAAACTA-

GAAGTT-3# (nucleotide positions 993–1,023) with the first-strand cDNA. The

PCR was initially denatured at 94�C for 5 min and in the subsequent cycles at

94�C for 30 s. Annealing and elongation cycles were both performed for 3 min

at 68�C. PCR products were size separated by electrophoresis on a 0.9%

agarose gel and blotted onto nylon membranes (Hybond-N1; Amersham-

Pharmacia Biotech).

Membranes were hybridized in 53 SSC, with 1.0% blocking reagent, 0.1%

lauroylsarcosine, and 0.02% SDS at 42�C, to digoxigenin-dUTP-labeled

probes. Probes were labeled using the DIG-DNA labeling kit (Roche Diag-

nostics) and were synthesized from PopCel1 cDNA by gene-specific primers.

Following hybridization, the membranes were washed in 23 SSC for 5 min

at room temperature and then two times in 0.13 SSC with 0.1% SDS at 68�C

for 15 min each time. The washed membranes were developed using the DIG-

DNA detection kit (Roche Diagnostics) for chemiluminescent detection.

Western-Blot Analysis

After the leaves were removed, the main vein with petiolule pulvinus in

the middle part of the petiole was homogenized in 20 mM sodium phosphate

buffer (pH 6.2) in a mortar, and the wall residue was washed three times. The

wall-bound proteins were extracted from the wall residue with a buffer

containing 1 M NaCl. The proteins were then subjected to electrophoresis with

10% SDS-PAGE, electrotransferred to Hybond-C Extra (Amersham), and

probed with an antibody against the PopCel sequence, followed by a second

antibody using the Toyobo ABC High-HRP immunostaining kit. Seven lines of

transgenic plants were assayed.

Cellulase Activity Assay

Each enzyme preparation was obtained from the wall residue of the main

vein with petiolule pulvinus in the middle part of the petiole with a buffer

containing 1 M NaCl, and its activity was assayed viscometrically at 35�C for

2 h, using 0.1 mL of the enzyme preparation plus 0.9 mL of 10 mM sodium

phosphate buffer (pH 6.2) containing 0.65% (w/v) carboxymethylcellulose in

semimicroviscometers from Cannon Instruments. One unit of activity is

defined as the amount of enzyme required to cause 0.1% loss in viscosity in

2 h under such conditions (Ohmiya et al., 1995). Protein was determined using

the Coomassie Plus protein assay reagent (Pierce), according to the method

described by Bradford (1976).

Determination of Cello-Oligosaccharides

After the leaves were removed, the main vein with petiolule pulvinus in

the middle part of the petiole was homogenized in 20 mM sodium phosphate

buffer (pH 6.2) in a mortar. The soluble extract was boiled for 5 min and left at

room temperature for 24 h to equilibrate the anomer configuration between

a- and b-types. After centrifugation, the amount of cello-oligosaccharides was

determined by cellobiose dehydrogenase purified from conidia spores of

Phanerochaete chrysosporium (Tominaga et al., 1999). The reaction mixture

contained 90 mU (10 mL) of cellobiose dehydrogenase, 50 mM cytochrome c (10

mL), and sample solution (70 mL) in 100 mM sodium acetate buffer, with a pH

of 4.2. After incubation for 5 min at room temperature, the A550 was deter-

mined. A linear standard curve was obtained with a standard cellobiose

solution, and an absorbance of 0.5% corresponded to approximately 270 ng

per 100 mL of reaction mixture for cellobiose.

Fractionation and Measurement of Wall Components

The main vein and petiolule pulvinus in the middle part of the petiole were

separated after the leaves had been removed. Each sample was ground in

liquid nitrogen and freeze-dried before its dry weight was determined. The

Figure 7. In situ xyloglucan endotransglucosylase activity incorporat-ing green fluorescent whole xyloglucan for 15 min on the transversesection (200 mm) of the petiolule pulvinus attached to the main veinof trg1. The tissues of the pulvinus and vein are shown in the upperand lower areas, respectively, in the images of the wild-type (A) andtransgenic (B) plants. The arrows indicate the incorporated wholexyloglucan in the parenchyma motor cells. The red color is due to theautofluorescence of chloroplasts. Bars 5 0.5 mm.





sample was successively extracted six times with 10 mM sodium phosphate

buffer (pH 7.0) and three times with 24% KOH containing 0.1% NaBH4 at less

than 45�C for 3 h in an ultrasonic bath. The insoluble wall residue (the

cellulose fraction) was washed with water and solubilized with ice-cold 72%

sulfuric acid. Total sugar in each fraction was determined by the phenol/

sulfuric acid method (Dubois et al., 1956). The xyloglucan level was deter-

mined by the iodine/sodium sulfate method (Kooiman, 1961).

Growth Measurements

The growth response of the transgenic plants was monitored after they

were transplanted in soil and habituated for 3 weeks under nonsterile

conditions. Each stem (around 15 cm) was marked at a height of 5 cm, which

was used as a reference point for measuring the height, diameter, and number

of internodes every third day. The length of the stem was determined from the

top to the reference point. Dry weight was determined after freeze drying the

samples.

The timing of the leaf movement was determined by observing the move-

ments every day for 20 d during both the rainy (January) and dry (May)

seasons. The closing movements of the leaves with the base of their main vein

excised occurred between 10 and 11 AM during both the rainy and dry seasons.

The veins with leaves attached were placed on paper towels immediately after

excision.

Four transgenic lines, trg1, trg2, trg3, and trg4, were used to observe the

opening and closing movements of leaves. These plants had nine or 10 petioles

each, all of which were more than 10 cm long. The lower part of the petiole

was defined as the second petiole from the bottom, the middle part as the fifth

or sixth petiole from the bottom, and the upper part as the ninth or tenth (or

newest) petiole. All of the pairs of leaves attached to the main vein in the

petiole were observed to determine the opening and closing movements.

Fluorescent Xyloglucans

Fluorescent xyloglucan (Takeda et al., 2002) was prepared by dissolving 10

mg of cyanogen bromide and 20 mg of pea (Pisum sativum) xyloglucan (50 kD)

in 1 mL of water and adjusting the pH to 11.0 by adding NaOH. The activated

polysaccharide was incubated with 4 mg of fluoresceinamine overnight at

room temperature. The fluorescein-labeled xyloglucan was purified by gel

filtration on a Sephadex G-50 column (Amersham-Pharmacia Biotech). Cal-

culations showed that 1 mmol of fluorescein incorporated into 110 mmol of

sugar residues, corresponding to a substitution rate of 3.7 mol of fluorescein

per mole of xyloglucan. To prepare fluorescent XXXG (Fry, 1997), 40 mg of

XXXG was aminated with 2 mL of 0.4 M sodium cyanotrihydroborate in 2.0 M

ammonium chloride at 100�C for 120 min. The aminated oligosaccharide was

purified by gel filtration on Bio-Gel P-2 and incubated with 50 mg of

fluorescein isothiocyanate in sodium carbonate bicarbonate buffer (pH 9.0)

for 2 h at room temperature. The oligosaccharide-fluorescein isothiocyanate

conjugate was purified by gel filtration on Bio-Gel P-2.

In Situ Xyloglucan Endotransglucosylase Activity

Transverse sections (200 mm) of the main vein including petiolule pulvinus

with fluorescent derivatives were incubated for 15 min in 300 mL of 2 mM

MES/KOH buffer (pH 6.2) containing 0.2 mM fluorescent whole xyloglucan or

9 mM fluorescent XXXG while being shaken in darkness at 23�C. The sections

that were incubated with whole xyloglucan were washed three times in 0.01 M

NaOH for 30 min. Those incubated with XXXG were washed with 5% formic

acid in 90% ethanol for 5 min followed by 5% formic acid for 5 min (Takeda

et al., 2002). The sections were washed twice with water and examined using a

Zeiss Axioscope microscope equipped with epifluorescence illumination.

Sequence data from this article can be found in the GenBank/EMBL data

libraries under accession number D32166 (PopCel1 cDNA).

ACKNOWLEDGMENTS

We thank Takahide Tsuchiya and Nobuyuki Kanzawa (Department of

Chemistry, Sophia University) for valuable discussions during the final

preparation of this article.

Received February 4, 2008; accepted March 11, 2008; published April 16, 2008.

LITERATURE CITED

Binkley D, Senock R, Bird S, Cole TG (2003) Twenty years of stand

development in pure and mixed stands of Eucalyptus saligna and

nitrogen-fixing Facaltaria moluccana. For Ecol Manage 182: 93–102

Bon MC, Bonal D, Goh DK, Monteuuis O (1998) Influence of different

macronutrient solutions and growth regulators on micropropagation of

juvenile Acacia mangium and Paraserianthes falcataria explants. Plant Cell

Tissue Organ Cult 53: 171–177

Bradford MN (1976) A rapid and sensitive method for the quantitation of

microgram quantities of protein utilizing the principle of protein-dye

binding. Anal Biochem 72: 248–254

Bunning E, Moser I (1969) Interference of moonlight with the photoperi-

odic measurement of time by plants, and their adaptive reaction. Proc

Natl Acad Sci USA 62: 1018–1022

Darwin C (1880) The Power of Movement in Plants. John Murray, London

Dubois M, Gilles KA, Hamilton JK, Rebers PA, Smith F (1956) Colori-

metric method for determination of sugars and related substances. Anal

Chem 28: 350–356

Fry SC (1997) Novel ‘dot-blot’ assays for glycosyltransferases and glyco-

sylhydrolases: optimization for xyloglucan endotransglycosylase (XET)

activity. Plant J 11: 1141–1150

Harpster MH, Dawson DM, Nevins DJ, Dunsmuir P, Brummell DA (2002)

Constitutive overexpression of ripening-related pepper endo-1,4-beta

glucanase in transgenic tomato fruit does not increase xyloglucan

depolymerization of fruit softening. Plant Mol Biol 50: 357–369

Hayashi T (1989) Xyloglucans in the primary cell wall. Annu Rev Plant

Physiol Plant Mol Biol 40: 139–168

Hayashi T, Yoshida K, Park YW, Konishi T, Baba K (2005) Cellulose

metabolism in plants. Int Rev Cytol 247: 1–34

Kooiman P (1961) The constitution of Tamarindus-amyloid in plant seeds.

Recl Trav Chim Pays Bas 80: 849–865

Kurinobu S, Prehatin D, Mohanmad N, Matsune K, Chigira O (2007) A

provisional growth model with a size-density relationship for a plan-

tation of Paraserianthes falcataria derived from measurements taken over

2 years in Pare, Indonesia. J For Res 12: 230–236

Merkel RC, Pond KR, Burns JC, Fisher DS (2000) Rate and extent of dry

matter digestibility in sacco of both oven- and freeze-dried Paraser-

ianthes falcataria, Calliandra calothyrsus, and Gliricidia sepium. Trop Agric

77: 1–5

Nakamura S, Hayashi T (1993) Purification and properties of extracellular

endo-1,4-b-glucanase from suspension-cultured poplar cells. Plant Cell

Physiol 34: 1009–1013

Nakamura S, Mori H, Sakai F, Hayashi T (1995) Cloning and sequencing of

cDNA for poplar endo-1,4-b-glucanase. Plant Cell Physiol 36: 1229–1235

Nakanishi F, Nakazawa M, Katayama N (2005) Opening and closing of

Oxalis leaves in response to light stimuli. J Biol Educ 39: 87–91

Nicol F, His I, Jauneau A, Vernhettes S, Canut H, Hofte H (1998) A plasma

membrane-bound putative endo-1,4-b-glucanase is required for normal

wall assembly and cell elongation in Arabidopsis. EMBO J 17: 5563–5576

Ohmiya Y, Nakai T, Park YW, Aoyama T, Oka A, Sakai F, Hayashi T (2003)

The role of PopCel1 and PopCel2 in poplar leaf growth and cellulose

biosynthesis. Plant J 33: 1087–1097

Ohmiya Y, Nakamura S, Sakai F, Hayashi T (1995) Purification and prop-

erties of wall-bound endo-1,4-b-glucanase from suspension-cultured

poplar cells. Plant Cell Physiol 36: 607–614

Ohmiya Y, Samejima M, Amano Y, Kanda T, Sakai F, Hayashi T (2000)

Evidence that endo-1,4-b-glucanase act on cellulose in suspension-

cultured poplar cells. Plant J 24: 147–158

Otsamo R (1998) Effect of nurse tree species on early growth of Anisoptera

marginata Korth. (Dipterocarpaceae) on an Imperata cylindrica (L.) Beauv.

grassland site in south Kalimantan, Indonesia. For Ecol Manage 105:

303–311

Park YW, Tominaga R, Sugiyama J, Furuta Y, Tanimoto E, Samejima M,

Sakai F, Hayashi T (2003) Enhancement of growth by expression of

poplar cellulase in Arabidopsis thaliana. Plant J 33: 1099–1106

Ragaukas AJ, Williams CK, Davison BH, Britovsek G, Cairney J, Eckert

CA, Frederick WJ Jr, Hallett JP, Leak DJ, Liotta CL, et al (2006) The path

forward for biofuels and biomaterials. Science 311: 484–489

Shani Z, Dekel M, Tsabary G, Goren R, Shoseyov O (2004) Growth

enhancement of transgenic poplar plants by overexpression of Arabi-

dopsis thaliana endo-1,4-b-glucanase (cel1). Mol Breed 14: 321–330

Shively GE, Zelek CA, Midmore DJ, Nissen TM (2004) Carbon seques-

Hartati et al.




tration in a tropical landscape: an economic model to measure its incre-

mental cost. Agrofor Syst 60: 189–197

Siregar UJ, Rachmi A, Massijaya MY, Ishibashi N, Ando K (2007) Economic

analysis of sengon (Paraserianthes falcataria) community forest plantation, a

fast growing species in East Java, Indonesia. For Policy Econ 9: 822–829

Takeda T, Furuta Y, Awano T, Mizuno K, Mitsuishi Y, Hayashi T (2002)

Suppression and acceleration of cell elongation by integration of xylo-

glucans in pea stem segments. Proc Natl Acad Sci USA 99: 9055–9060

Tominaga R, Samejima M, Sakai F, Hayashi T (1999) Occurrence of cello-

oligosaccharides in the apoplast of auxin-treated pea stems. Plant

Physiol 199: 249–254

Ueda M, Nakamura Y (2007) Chemical basis of plant leaf movement. Plant

Cell Physiol 48: 900–907

Vengadesan G, Amutha S, Muruganantham M, Anand RP, Ganapathi A

(2006) Transgenic Acacia sinuata from Agrobacterium tumefaciens-mediated

transformation of hypocotyls. Plant Cell Rep 25: 1174–1180





Documents

Overexpression of Poplar Cellulase Accelerates Growth and