Root Hair Infection in Actinomycete-Induced Root Nodule Initiation in Casuarina, Myrica, and Comptonia Author(s): Dale Callaham, William Newcomb, John G. Torrey, R. L. Peterson Source: Botanical Gazette, Vol. 140, Supplement: Symbiotic Nitrogen Fixation in Actinomycete-Nodulated Plants (Mar., 1979), pp. S1-S9 Published by: The University of Chicago Press Stable URL: http://www.jstor.org/stable/2474196 . Accessed: 30/08/2011 15:59
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to Botanical Gazette.
http://www.jstor.org BOT.GAZ. 14O(SUPP1.): S1-S9. 1979. (¢ 1979by The Universityof Chicago.0006-8071/79/40OS-0001$00.88
ROOT HAIR INFECTION IN ACTINOMYCETE-INDUCED ROOT NODULE INITIATION IN CASUARINA, MYRICA, AND COMPTONIA
DALE CALLAHAM,1* WILLIAM NEWCOMB,2t JOHN G. TORREY,* AND R. L. PETERSONt *CabotFoundation, Harvard University, Petersham, Massachusetts 01366, and tDepartmentof Botany and Genetics Universityof Guelph,Guelph, Ontario, Canada N1G 2W1
The infection process leading to the developmentof root nodules of Comptonia peregrina, Casuarina cunninghamiana, Myrica gale, and M. cerifera was studied by light and electronmicroscopy. Deformed growth of root hairs was observedas early as 24 h after seedlingsgrown aeroponically or hydroponically wereinoculated with suspensionsof crushednodules or culturesof the actinomycetousendophyte of Comp- tonia. The extent of root hair deforIrJationshowed a positive correlationwith the numberof noduleswhich subsequentlydeveloped. The essentialfeatures of infectionin each of these specieswere very similar.The actinomyceteentered a deformedroot hair of the host in a regionof foldingof the cell wall. A convoluted elaborationof the root hair wall which occurredat this presumptivepenetration site was continuouswith the moreevenly depositedcapsule of the endophyticactinomycete. An associatedfeature of this wall deposi- tion was thickeningof the cell wall of the infectedroot hair and the adjacentprenodular cells. The actino- mycete encapsulationwas thickest at the presumedsite of penetrationand thinnerin later stages of endo- phytic growthaway from this site. These observationssuggest a periodof initial disequilibriumcaused by the infection,followed by more harmonioussymbiotic growth. The observationof a morphologicallyand cytologicallysimilar root hair infection process in these three genera indicates that root hair infection involves a specificand orderlyinteraction which representsthe commonmode of invasionin the initiation of actinomycete-inducedroot nodules.
Introduction The ultrastructuralstudies of LALONDE(1977) The developmentof actinomycete-inducednod- showed the endophyte within a root hair to be ules on the roots of certainwoody dicots represents encapsulatedby host-derivedwall materialas the a complexseries of events. Three major stages in actinomycetegrows tonard the root hair base. The developmentcan be distinguished:the infectionof infectionof C. peregrina roots observedin the light the root hair, the induction and invasion of a microscope(CALLAHAM and TORREY1977) occurred prenodularproliferation of the cortex, and the by penetrationof a roothair at a site of invagination inductionand invasionof primaryand higher-order or foldingof the cell wall. nodule lobe primordia.The anatomicaldetails of Despite these observationsof the actinomycete root nodule morphogenesishave become better enteringthe host throughroot hairs,there has been understoodwith studies of this process in A Inus little understandingof how this processoccurs. The glutinosa (POMMER1956; TAUBERT1956; ANGULO direct involvementof root hair deformationin the CARMONA1974), Casuarina cunninghamiana (TOR- infectionprocess was questionedby QUISPEL(1955, REY1976), Comptoniaperegrina (BOWES, CALLAHAM, 1974), w ho claimed his experimentsshowed an and TORREY1977; CALLAHAMand TORREY1977; inabilits7of the actinomaceteto proliferateoutside NEWCOMBet al. 1978), and Myrica gale (FLETCHER of host cells;thus, he eliminatedthe possibilitythat 1955; TORREYand CALLAHAM1978, 1979). the actinomyceteeas responsiblefor root hair de- The processof root infectionby the actinom)cete formation. Conclusionsdrawn from these results which initiates this developmentalsequence is not shouldbe reconsideredin light of new evidencethat clearlyunderstood. Compelling evidence by TAUBERT at least one of these nodule endophytescan grow (1956) and ANGULOCARMONTA (1974) established outside of host tissues (CALLAHAM,DEL TREDICI, that the actinomyceteinitially enters the rootsof A. and TORREY1978). glutinosa by penetrationof a deformedroot hair. The observationsof LALONDE(1977) of a specific LALONDE(1977) observed root hair deformation "exoencapsulation"process preliminary to the root within 24 h after inoculationof the plants. A11 hair infection in A. glutinosa were based on the elongatingroot hairs appearedto be affected,and assumptionthat the noduleendophyte exhibits rod- filamentpleomorphism (LALONDE, KNOWLES, and eachexhibited deformed growth with furtherelonga- FORTIN1975; LALONDE 1977) when growing outside tion occurringin branchesof the originalroot hair as opposedto eithin the host cells. However,the axis. The result of this responsewas a "slope"of rhizospherebacterium discussed by LALONDE(1977) deformedroot hairswith newly initiatedroot hairs was not identifiedby availableimmunological meth- distal to this regionremaining short and branched. ods (LALONDEet al. 1975)or by directobservations l Present address: Department of Botany, University of of root hair penetration.In his exoencapsulation Massachusetts,Amherst, Massachusetts 01003. theory, LALONDEdid not interpret the root hair 2 Present address: Department of Biology, Queen's Uni- deformationas directlyinvolved in the mechanism versity, Kingston, Ontario,Canada K7L 3N6. of penetrationof the root hairwall, althoughhe did S1 :_^^ \'X.IS.,,1 ts / /F > f f t.Xit _ v>:- ::vx;fJk -- . - _Z 1u - _
S2 BOTANICALGAZETTE view the deformationas a plant responseto the Materialand methods actinomycete. PLANTCULTURE. Locally collected fruit of Comp- There are consistent observationsof root hair tonia peregrina were scarified,soaked for 24 h in deformationassociated with nodulationin Alnus, 500 ppm gibberellicacid (GA3),and germinatedin Comptonia,Cas?sarina, and M. gale. Convincing flats of washed sand in the greenhouseat the evidence for root hair infection associated with HarvardForest (DEL TREDICIand TORREY1976). deformationwas providedfor A. glutinosa(POM1HER Locallycollected fruit of MyricagaZe and M. cerifers 1956; TAUBERT1956; ANGULOCARMONA 1974), A. were treated similarlybut without scarificationof crispa (LALONDE1977), and Comptonia(CALLAHAM the fruit of M. gale. Seeds of Casuarinacunning- and TORREY1977). The researchreported here is an hamianawere germinated in flats of washedsand in investigationof the infectionprocess in the initiation the greenhousewithout scarification.All seedlings of the root nodulesof M. gale, M. cerifera, and C. weretransferred to aeroponicculture tanks (ZOBEL, cunninghamtanafor whichobservations of the initial DEL TREDICI,and TORREY1976) when the shoots infectionhave not been recorded.Further evidence wereabout 3 cm high and weregrown under condi- of root hair infection of C. peregrina, extending tions specifiedby CALLAHAMand TORREY(1977). earlierobservations (CALLAHAM and TORREY1977), Two weeks after germination,seedlings of M. gale is also reported. weretransferred to smalltest tubewater cultures for
1 w} ,= t . |
v.e t>.>*t.6 K # #e g ^_ .<. 8 l#^.Z*. F.F_ t t fi{ x
# r : 1 r
-*> r j . | # drP | _ -* i ; tl " ; _ " n 'i ,,' *j- lf,ljifs X f -
FIGS.1-4. Root hairs of Myricagale from seedlingsgrown in water cultures.Fig. 1, Uninoculatedroot hairs which developed long and straight;scale = 200 ,um.Fig. 2, FU11Yelongated root hair 24 h after applicationof CI inoculum;fragments of the inocu- lum are entwinedabout the root hair which still exhibitedactive streamingbut failed to branch;phase contrast; scale = 50 ,am. Fig. 3, Elongatingroot hair 24 h after applicationof CI inoculum,showing branching and continuedgrowth from several points; phase contrast;scale = 50 ,um.Fig. 4, Deformedroot hair 24 h after inoculationwith the CI; branchingcan be quite ex- tensive; CI filamentsare not associatedwith each branchpoint; anoptralphase contrast;scale = 20 ,um.
FIGS. 5-10. Fig. 5, Sectionof infectedroot hair (IRH) of Casuarinacunninghamiana cut longitudinally;the endophytewithin the root hair (arrows)was traceableto the infectednodule cortex in other sections;scale = 10 ,am. Fig. 6, Section throughthe highlybranched distal part of the infectedroot hair of Casuarinain fig. S; the filamentousendophyte (arrows) is presentthroughout this lobed root hair tip; scale = 10 ,um.Fig. 7, Transmissionelectron micrograph (TEM) of a section of the Casxarinaroot hair cut adjacent to the section in fig. 6; the endophyte (E) within the root hair is encapsulatedby a layer (c) which joins several filamentsat the lowerright into a largestrand; scale = 1 ,am.Fig. 8, TEM of the endophyte(EN) within an infected root hair of CompGoniaperegrsna. The hyphaeof the endophyteare encapsulatedwithin a fibrillarencapsulation material (c) which is continu- ous with the root hair wall as in later stages; the host plasmalemma(PM) separatesthe encapsulatedhyphae from the host cyto- plasm;scale = 1 ,um.Fig. 9, Crosssection of the root of CompGoniaperegrina at the level of the prenodulewhich shows two root hair infections (IRH-1 and IRH-2) which have occurredafter inoculationwith culturesof the highly invasive Comptonxaendo- phyte. The infectinghyphae (en) fromIRH-2 penetratedeeply into the prenodule(infected cortex) while the infectinghyphae of IRH-1 (arrows)are localizedto a few cells just below the root hair; scale = 25 ,um.Fig. 10, Light micrographof an infectedroot hairof M. gale,cut longitudinally;the endophyte(EN) has enteredthe root hairwall and appearsclosely associated with a densely stainingbody tDB) of segmentedor particulatesubstructure just adjacentto the invasionsite; the root hair nucleus(N) is at the root hair base and is not closely associatedwith hyphal growth. Note the thicknessof the root hair wall relative to the adjacent uninfectedepidermal cells; scale = 10 ,um. _{_sFsLsx"niy;f'--- I N *#wt i ::f /...... KllX4"*"
t-f:fu
_s-*
t0f C. X X tt > a I ^ O C = y'"ff"''.ERt'''.l',,;;'.'-ni'M ' i Si:':.'S, i '0 f r _J ' * * / O
t Er is W-uSy / f lt 0 S= t ;f;-_ I tF 5 1
w - _ U lsl S iFMF>' H- ;-- a; g tU E st ts 10 _|_S,2 fuS -{!A R s_
._ Z . lf t A;_.1 ! * S C! f-giti-0F -#; !!s S " ox i FIG.11. Fig. lla-e, A seriesof sectionscut throughan infectedroot hairof M. gale demonstratingthe differentimages obtained in variousplanes of sections; the distal portioncontaining convoluted wall materialis continuouswith the proximalportion of root hair; endophyte(en) hyphae are presentin the proximaland distal portionsand pass into the adjacentcortical cell- bar = 20 ,um.Fig. 1lf, TEM montageof the distal and proximalportions of the same infectedroot hair and adjacentinfected and un- infectedcortical cells in fig. lla-e; the distal portionof the root hair containsconvoluted elaborations of wall materialand en- capsulatedendophyte (en); in the proximalportion the encapsulatedhyphae are continuouswith the host cell wall; comparethe thicknessof the encapsulatingmaterial in the root hair and in infectedcortical cells; the cell wall betweenthe infectedand unin- fected cells contiguouswith the root hair showssculptured thickenings; bar = S ,um.Fig. 1lg, High magnificationof a portionof the distalpart of the roothair in fig. 1lf; the outerlayers (arrows) of the cell wall areelectron dense; the convolutedwall materialis continuouswith the host cell wall and containsmany randomlyarranged electron-dense fibrils; the folded portion (FW) of the host wall and a transectionof an encapsulatedhyphae (en) are also shown;bar = 2 ,um.Fig. llh, High magnificationof an en- capsulatedhypha in fig. 1lf showingthat the host wall materialis comprisedof manyelectron-dense fibrils embedded in a less dense matrix;bar = 1 ,um.Fig. lli, High magnificationof the commoncell wall betweenthe infectedand uninfectedcells continuous with the infectedroot hair in fig. 1lf; many polyribosomesand profilesof roughendoplasmic reticulum are presentnear the sculp- tured cell walls of the infectedcell; bar = 1 ,um. CALLAHAMET AL. ROOT HAIR INFECTION S5 studiesof root hairdeformation and the time course tions were of plants of Comptoniaand M. gale of infectionby the "Comptonia Isolate." inoculatedwith a crushed,washed suspension of the INOCULATION.At about 1 wk after transferto culturedactinomycete isolated from Comptonia nod- aeroponicculture, the plants were inoculatedwith ules, that is, the "ComptoniaIsolate" or "CI" crushednodule suspension(BOND, FLETCHER, and (CALLAHAMet al. 1978).This microorganismforms FERGUSON1954) from soil-grownor aeroponically noduleson Comptoniawith much higherfrequency grownnodules of the host plant, eitherby brushing than crushed nodule suspensions and produces the inoculumonto the root system in M. gale or by effective nitrogen-fisingnodules on M. gale, M. pouringit into the growthmedium. Some observa- cerifera,and Comptonia.
FIG. 1 l (Continud ) so BOTANICALGAZETTE
MICROSCOPY.Tissues were preparedfor light trated), and Comptonia peregrina (figs. 8, 9). In microscopyand transmissionelectron microscopy as every plant infectedroot hairs were deformed,and describedby CALLAHAMand TORREY(1977) and the initial entry of the actinomycetewas always NEWCOMBet al. ( 1978). traceableto a crookor sharplyfolded region of the root hair (figs. llf, 12a). At such a site, the lobes of Observations the deformedroot hair come together at a point ROOTHAIR DEFORMATION. All plants in either filled with dense polysaccharidedeposits in which aeroponicor hydroponicculture developed long, filamentsmay be embedded(figs. 11a-g, 12a). The straightroot hairs if the roots were not exposedto presumedsite of penetrationis withinsuch a folded an active inoculum(fig. 1). Within 24 h after the root hair. In M. gale and Comptoniaa pronounced applicationof the CI inoculum,root hair deforma- depositionof wallmaterial was frequentlyassociated tion was observedin both Myricagale (figs. 3, 4) and withthe siteof penetration.This wall material was ar- Comptonia.This rapidresponse was observedalso in rangedin elaborateconvolutions (figs. llf, g, 12a, c), M. gale seedlingsinoculated with crushedM. gale andsometimes endophyte hyphae were encapsulated noduleinoculum, and such seedlings developed many with it (fig. 12c). Randomlyarranged electron-dense root noduleswithin 8 days afterinoculation. Comp- fibrillarmaterial was present in the convoluted tonia, however, formed few deformedroot hairs structures(fig. 12c) and in the outer layers of the when inoculated with Comptoniacrushed nodule root hair cell wall (figs. 11g, 12a, c). With toluidine suspensions,and very few nodules subsequently blue O staining,both the convolutedwall configura- developed (CALLAHAMand TORREY1977). After tions and the outer layersof the root hair cell wall inoculationof seedling roots of M. gale with CI appeareddark blue, differingfrom the light reddish- suspensions,the filamentsof the actinomyceteinter- blue of the typical host cell walls of the root hair twinedabout many root hairs (figs. 2-4). Elongating and the encapsulatingmaterial surroundingthe hairswere deformed (figW 3); apparentlythe stimulus actinomycete.The encapsulatingmaterial in Alnus causing branchingcould act over short distances is believed to consist of pectins (LALONDEand since deformedroot hairs not associatedwith fila- KNOWLES1975) . mentswere observed. Thus our observations indicate Within the root hair and away from the site of a positivecorrelation between the extentof roothair penetration,the deposition of wall material was deformationand the numbers of nodules which restrictedto the capsulesurrounding the endophyte develop. hyphae(figs. llf, g, 12a-c); in somesites (figs.8, llJ) The deformedgrowth of root hairs appearedto the encapsulatingmaterial was continuouswith the result from branchingand wall foldingwhich was host cell wall, as was observedin root hairinfections limitedto youngand still elongatingroot hairs(figs. in Alnus (LALONDE1977). The wall materialencap- 3, 4). Fully elongatedroot hairswhich still showed sulating the hyphae consists of fibrillarmaterial cytoplasmicstreaming failed to branchin response arrangedless randomlythan in the convolutedwall to the inoculum(fig. 2). Root hair deformationwas material(cf. figs. 11g, h, 12b, c) and appearsto be a welldescribed by LALONDE(1977) for Alnus seedlings moreorderly continuation of the initialwall deposi- and was not significantlydifferent here; just as in tion. This stage may representa periodof equilibra- Alnus,a "slope"of deformedroot hairs was observed tion in the establishmentof an associationbetween to be related to root hair length at the time of the host and endophyte;during this periodthe host inoculation. and endophytemay not be in completeharmony. Ourobservations of M. ceriferawere not as exten- Further evidence favoring this interpretationis- sive as for Comptoniaand M. gale. Myricacerifera providedby the even thinnerencapsulation of the inoculatedwith CI suspensionsshowed root hair endophyte in subsequentlyinfected cortical cells deformationand effective nodules developed in (fig. llf) and the pronouncedthickening of the cell abundance.When roots of M. ceriferaseedlings wall in the infectedcell contiguouswith the infected grownin aeroponicswere inoculated with a crushed root hair cell but not in the adjacentuninfected nodulesuspension from M. gale,root hair deforma- cortical cell (fig. 1lf, i). Because most of the tion occurred,but many small ineffectivenodules infectedroot hairswere senescent, few observations wereproduced, as reportedby GARDNERand BOND were made on the organellesof these cells. One (1966). interestingfinding, however, was the localizationof INFECTION.Using later stagesof the same-plant the nucleusat the base of the infectedroot hair (fig. material,very young prenodularstages of nodule 10), apparentlynot associatedwith the growthof developmentwere examinedto determinethe site the penetratingendophyte. This is in contrastto the of infectionleading to noduleinitiation. Actinomy- situationin infectedroot hairs of leguminousroot cete invasion of the root occurredexclusively by nodules(DART 1977) and in Alnus (LALONDE1977). root hair infection in Casuarinacunninghamiana It is not clearhow many root hairinfections may (figs.5-7), M. gale (figs. 1F12), M. cerifera(inocu- be associatedwith a single root nodule. In earlier lated with nodulesuspensions of M. gale,not illus- studieswith Comptonia, which were conducted under . - - ...... *i
12b
. * re s e
.- : ^
* . :
i
# w - -
F1P 3i \| ?; SF
K
.. 8R f . te OS 5;r*;
b * Al/cb . */ + . t: - . iet *''' t E .v N ..,] t t ^ s sE. ... .v2v iZ' .Z .
F :^t.s:: : :S- s : si < :s sC .ffiv1
:EN
FIG.12. Fig. 12a, TEM of a deformed,infected root hair and contiguouscortical cells of capsulatedendophyte hyphae (en), the M. gale, showingtransection of en- fold (arrows)which is the presumptivesite of endophyte ofconvoluted wall material;bar = 10 ,um. penetration,and the nearbymass Fig. 12b, High magnificationof encapsulatedhypha fig.12a; the wall materialis fibrillarand is bounded indicatedby double arrowsin (large by the host plasma membrane(small arrows)and the arrows);bar = 0.5 ,um.Fig. 12c,High magnificationof the endophytecell wall infig. 12a; many convolutedmaterial from an adjacentsection to that illustrated electron-densefibrils are presentin the convolutedmaterial tionsof encapsulated and in the outer layers of the cell walP severalsec- endophytehyphae (EN) are also shown;bar = 1 jum. S8 BOTANICAL GAZETTE [MARCH (SUPPL.) inoculum-limitingconditions and thereforeonly led deformedroot hair growth as well as nodulation to the establishmentof few root nodules,only one underthese conditions. root hair infectionwas associatedwith each nodule Root hair infections by actinomycetesappear (CALLAHAMand TORREY1977). In other plants, analogousin certainways to the infectionprocess in such as M. galeinoculated with a nodulesuspension legume-Rhizobium association,since they involve (TORREYand CALLAHAM1979) or Comptoniaseed- both a deformationof young growingroot hairs lings inoculatedwith culturesof Comptoniaendo- attributableto the presenceof activitiesof the micro- phyte, numerousroot noduleswere formed and root symbiont in the rhizosphere(QUISPEL 1974) and hair infectionswere sometimesobserved close to- alsothe penetrationof the roothair in a foldedregion getherbut radiallyseparated even within the same of the wall (NAPOLIand HUBBELL 1975; DART 1977). 1 ,umthick section (fig. 9). One of these discrete In root noduleinitiation in legumesa role has been infectionscould be tracedto the infectedcells of the suggestedfor root hair-Rhizobium interactionsin prenoduleand nodulelobes, while the secondinfec- determinationof the host-microsymbiontspecificity tion was limited to the root hair or outer cortical through specific binding (BOHLOOL and SCHMIDT parenchyma.Since these observationsare of fixed 1974;BHUVANESWARI, PUEPPKE, and BAUER1977) materials,it is not knownif one infectionoccurred or infection thread formation (LI and HUBBELL later than the other or was halted in development 1969).A commonfeature of both actinomyceteand by an influencefrom the adjacentinfection (LALONDE Rhizobiumroot hair infectionsis the penetrationat 1977) . It seems clear that single infections are a foldedsite of the root hair ce]l wall whichin both sufficientfor noduleinitiation and this is most likely casesmay be involvedin a localizationof biochemi- the situation occurringin the field. The situation, cal interactionsso as to intensifytheir effect and aid however, may be more complex and endophytes wallpenetration (HUNTER and ELKAN1975; NAPOLI fromtwo sources might be presentwithin one nodule. and HUBBELL1975). The modelproposed by LALONDE(1977) for root Discussion hairinteractions and penetration in A lnus nodulation Includingthe presentreport, a total of five species suggeststhat an "exoencapsulationthread" is formed in fourgenera from three plant orders are now known by a specificinteraction of the Alnus root hairwith to be infected by this method, which appearsto a bacterialform of the actinomyceteand that this involve an orderlyinteraction of the actinomycete structuremay function in penetrationof the root with the root hair cell. The repeatedobservations hair wall by concentratinghydrolytic enzymes pro- of root hair infectionin the initiationof actinomy- ducedby the invadingbacteria. Our studies reported cete-inducedroot nodules providestrong support here have not producedevidence of such a mecha- for the view that this is the common mode of nism in these plants. Preliminarystudies now in associationestablishment in this type of symbiosis. progressindicate that the "ComptoniaIsolate" exists The sequenceof events in the infectionof each of in the hyphalform in the rhizosphereand that such theseplants seems to be the same:a deformationof hyphal forms invade deformed root hairs from the hair elicited by substancesproduced by the within the fold, as suggestedby the evidencepre- microorganism,followed by penetrationof the root sented here. Furtherwork is neededto understand hair cell wall accompaniedby extensiveformation the processin detail. of wall-like material by the host cell, including Acknowledgments encapsulationof the endophyte filaments within the invadedcells. We gratefully acknowledgethe technical assis- tance of P. DEL TREDICI,S. LAPOINTE,and D. The actual processesby which the root hairs GRENVILLE.This study was madepossible through becomedeformed and the actinomycetepenetrates the financialsupport of the Maria Moors Cabot the cell wall arenot known.Evidence here indicates Foundation for Botanical Research at Harvard that in Comptoniaand Myricagale the deformation University,National Science Foundation grant DEB is relatedto inoculationwith active inoculawhich 77-02249, an operatinggrant from the National can includepure washedcultures of the Comptonia ResearchCouncil of Canadato R. L. PETERSON, nodule actinomycete,in which case no extraneous and materialsand laboratoryspace made available materialsare present. Such evidence argues in favor by P. K. HEPLER,Department of Botany,University of a direct role of the actinomycetein causingthe of Massachusetts,Amherst, Massachusetts.
LITERATURE CITED
ANGULO CARMONA, A. F. 1974.La formation des nodules tions. I. Binding of soybean lectin to Rhizobia. Plant fixateurs dXazote chez Alnusglutinosa (L.) Vill. Acta Bot. Physiol. 60:486-492. Neerlandica 23:257-303. BOHLOOL,B. B., and E. L. SCHMIDT.1974. Lectins, a possible BHUVANESWARI, T. V., S. G. PUEPPKE, and W. D. BAUER. basis of specificity in the Rhizobium-legumeroot nodule 1977.The role of lectins in plant-microorganism interac- symbiosis.Science 185:269-271. 1979] CALLAHAMET AL. ROOT HAIR INFECTION
BOND,G., W. W. FLETCHER,and T. P. FERGUSON.1954. The var. molZisFern. nodule endophyte by morphological, im- developmentand function of the root nodules of Alnus, munolabelling and whole cell composition studies. Can. J. s9 Myricaand Hippophae.Plant Soil 5:309-323. Microbiol. 21:1901-1920. BOWES,B., D. CALLAHAM,and J. G. TORREY.1977. Time- LI, D., and D. H. HUBBELL.1969. Infection thread formation lapse photographicobservations of morphogenesisin root as a basis of nodulation specificity of Rhizobium-strawberry nodulesof Comptoniaperegrina (Myricaceae). Amer. J. Bot. clover associations. Can. J. Microbiol. 15:1133-1136. 64:516-525. NAPOLI,C. A., and D. H. HUBBELL.1975. Ultrastructure of CALLAHAM,D., P. DEL TREDICI,and J. G. TORREY.1978. Rhizobium-induced infection threads in clover root hairs. Isolation and cultivation in vitro of the actinomycete Appl. Microbiol. 30:1003-1009. causingroot nodulationin Comptonia.Science 199:899-902. NEWCOMB,W., R. L. PETERSON,D. CALLAHAM,and J. G. CALLAHAM,D., and J. G. TORREY.1977. Prenodule formation TORREY.1978. Structure and host-actinomycete interac- and primary nodule developmentin roots of Comptonia tions in developing root nodules of Comptonia peregrina. (Myricaceae).Can. J. Bot. 55:230S2318. Can. J. Bot. 56:502-531. DART,P. J. 1977. Infection and developmentof leguminous POMMER,E. H. 1956. Beitrage zur Anatomie und Biologie der nodules. Pages 367-472 in R. W. F. HARDYand W. S. Wurzelknollchen von Alnus glutinosa Gaertn. Flora 143: SILVER,eds. A treatise on dinitrogenfiwation. Section III. 603-634. Biology. Wiley, New York. QUISPEL,A. 1955. Symbiotic nitrogen fixation in non-legumi- DELTREDICI, P., and J. G. TORREY.1976. On the germination nous plants. III. Experiments on the growth in vitro of the of seeds of Comptoniaperegrina, the sweet fern. BOT.GAZ. endophyte of Alnus glutinosa. Acta Bot. Neerlandica 4: 137:262-268. 671-689. FLETCHER,W. W. 1955. The developmentand structureof . 1974. The endophytes of the root nodules in the non- root-nodulesof Myricagale L. Ann. Bot., N.S., 19:501-513. leguminous plants. Pages 499-520 in A. QUISPEL,ed. The GARDNER,I. C., and G. BOND.1966. Host plant-endophyte biology of nitrogen fixation. North-Holland, Amsterdam. adaptationin Myrica. Naturwissenschaften53:161. TAUBERT,H. 1956. Uber den Infektionsvorgang und die HUNTER,W. J., and G. H. ELKAN.1975. Rokeof pectic and Entwicklung der Knollchen bei A Inus glutinosa Gaertn . cellulolytic enzymes in the invasion of the soybean by RAlizobiumjaponicum. Can. J. Microbiol.21: 1254-1258. Planta 48:135-156. LALONDE,M. 1977. The infectionprocess of the Alnus root TORREY,J. G. 1976. Initiation and development of root nodulesymbiosis. Pages 569-589 in W. NEWTON,J. R. POST- nodules of Casuarina (Casuarinaceae). Amer. J. Bot. 63: GATE,and C. RoDRIGuEz-BARRueco,eds. Recent develop- 335-344. ments in nitrogenfixation. Academic Press, London. TORREY,J. G., and D. CALLAHAM.1978. Determinate develop- LALONDE,M., and R. KNOWLES.1975. Ultrastructure, ment of nodule roots in actinomycete-induced root nodules composition,and biogenesisof the encapsulationmaterial of Myrica gale Can. J. Bot. 56: 1357-1364. surroundingthe endophyteof Alnus crispavar. mollisroot . 1979. Early nodule development in Myrica gale. BOT. nodules.Can. J. Bot. 53: 1951-1971. GAZ. 140(Suppl.):S1.0-S14. LALONDE,M., R. KNOWLES,and J. ANDREFORTIN. 1975. ZOBEL,R., P. DEL TREDICI,and J. G. TORREY.1976. Method Demonstrationof the isolationof non-infectiveAlnus crispa for growing plants aeroponically. Plant Physiol. 57: 344-346.