Ultrastructural and immunological demonstration of the nodulation of the European Alnus glutinosa (L.) Gaertn. host plant by the North-American Alnus crispa var. mollis Fern. root nodule endophyte' M. L A L O N DAEN~D A. QUISPEL Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Texas A&M University on 11/13/14 For personal use only.
Cllnrles F . Ketteritlg Resenrctl Lohorntory, Yello\v Springs, O H , U.S.A. 45387 Botntzislz Lnhorntorirrt?~der Rijksrrtliversiteit. Not~tletlsteeg3, Leiden, Nederlntrd Accepted July 20, 1977 L A L O N D EM., , and A. QUISPEL.1977. Ultrastructural and immunological demonstration of the nodulation of the European AInrrs glrrtinosn (L.) Gaertn. host plant by the North-American Altlrrs o i s p n var. ttzollis Fern. root nodule endophyte. Can. J. Microbiol. 23: 1529-1547. The inoculation of the European Alt~rrsglirtitlosn (L.) Gaertn. host plant by a crushed-nodule inoculum, prepared with the North-American Alnrrs o i s p n var. tnollis Fern. root nodule, was successful. Fluorescein- and ferritin-labelled antibodies, specific against the A. crispcr var. mollis root nodule endophyte (Lalonde et 01. 1975), demonstrated the identity of this endophyte in the resulting nodules. The nodulation process of this abnormal host-endophyte system was studied by light and electron microscopy. An excretion of host blebs containing electron-dense polysaccharide material, resulting in the formation of exo-encapsulation threads containing presumptive endophytic bacterial cells, was associated with deformed root hairs. Originating from an exoencapsulation thread, the endophyte penetrates the root hair cell and then migrates as a hypha toward the cortical cells ofthe root. Its migration in the cortical cells of the primary nodule results in the induction of a lateral root which develops a s the true nodule. The ultrastructure of the A. crispn var. tnollis endophyte developing in the primary and true nodule of the abnormal A. gl~rtitlosnhost was similar to the one induced inside its normal A. crisper var. tnollis host. The actinomycetal intruder was a branched and septate hypha able to produce septate vesicles. The endophyte was always encapsulated in an electron-dense polysaccharide material surrounded by a host plasma membrane envelope. However, in this abnormal host-endophyte system, the number of primary nodules formed per root system was drastically reduced, and their appearance was delayed by 1 to 2 weeks. The delayed nodules were effective in fixing nitrogen and able to support satisfactory plant growth in a nitrogen-free medium. L A L O N D EM., , et A. QUISPEL.1977. Ultrastructural and immunological demonstration of the nodulation of the European Altlrrs glrrtinosn (L.) Gaertn. host plant by the North-American Altrrrs oispcr var. tnollis Fern. root nodule endophyte. Can. J. Microbiol. 23: 1529-1547. L'inoculation d e la plante hbte europeenne AItlrrs glrrtinosrr (L.) Gaertn., avec des nodules racinaires de la plante hBte nord-americaine Altzus crispcr var. ttrollis Fern.. fut un succes. L'emploi d'anticorps marques H la fluoresceine et a la ferritine, anticorps specifiques contre I'endophyte du nodule racinaire d ' A . crispcr var. tnollis (Lalonde et 01. 1975), a demontrC I'identite de cet endophyte dans les nodules produits sur le systeme racinaire d e la plante hBte A. glrrtitrosu. Le processus d e nodulation de c e systkme hbte-endophyte a ete CtudiC par microscopie photonique et electronique. L'excretion de globules appartenant a I'hbte, et contenant un materiel polysaccharidien opaque aux electrons participe a la formation de tubulures exoencapsulaires incorporant des cellules bactkriennes supposees endophytiques de nature. Ces tubulures se retrouvent associies aux poils absorbant dCformCs. Originaire d'une tubulure exoencapsulaire, I'endophyte penetre la paroi cellulaire d'un poi1 absorbant et migre comme un hyphe vers les cellules corticales d e la racine. Cette migration d e I'endophyte dans les cellules corticales du nodule primaire provoque I'induction d'une racine laterale qui devient un nodule veritable. L'ultrastructure d e I'endophyte d ' A . crispn var. tnollis croissant dans le nodule primaire et dans le nodule veritable de la plante hbte anormale A. glutinosa est similaire celle produite dans la plante hate normale A. o i s p n var. mollis. L'intrus microbien est un hyphe septC et ramifit capable d e produire des vCsicules sept6es. Cet endophyte est toujours encapsule par un materiel polysaccharidique opaque aux electrons, le tout Ctant entour6 par une enveloppe membraneuse appartenant I'hbte. NCanmoins, dans ce systkme anormal h8te-endophyte, le nombre d e nodules primaires form& sur le systeme racinaire est rCduit considCrablement, et leur apparition est retardee d'une a deux semaines. NCanmoins, ces nodules retard& sont efficaces pour fixer I'azote et sont capable d e soutenir une croissance normale de la plante hbte se developpant dans un milieu exempt d'azote. 'Contribution No. C-590 from Charles F. Kettering Research Laboratory. 'Author t o whom inquiries a n d requests for reprints should be addressed.
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Texas A&M University on 11/13/14 For personal use only.
1530
C A N . J. MlCROBl OL. VOL. 23. 1977
Introduction Nitrogen-fixing root nodules occur on nonleguminous angiosperms (13, 14). Over the years, there is more and more evidence of the ecological importance of this nitrogen-fixing symbiosis (37). In microscopic studies, it is indicated that the root nodule endophyte is a pleomorphic prokaryotic microorganism which is assumed to be an actinomycete (20, 8, 9). However, the isolation, the subculture, and the demonstration of the nodulation of the aseptic host plant with an infective pure culture of the isolated prokaryotic endophyte, cultured in vitro, had not been achieved according to Koch's postulates. Nevertheless, the non-leguminous root nodule endophytes were speculatively classified as Frankia species by Becking (6, 7). The proposed taxonomic position of the Frankiaceae family is based on morphological and physiological characteristics of the endophytes as observed in the root nodules borne by the different host plants. The 10 proposed species were based mainly on the results of cross-inoculation studies which suggest that the non-legume root nodule endophyte is specific for a certain host plant genus or a group of related genera. In these crossinoculation assays, because of the lack of virulent pure cultures of the endophytes for inoculation purposes, root nodules are taken from a nodulated host plant, surface-sterilized, crushed to make a water-suspension, and used as a crushednodule inoculum. However, as cautioned by Bond (13), the possibility exists that infective particles of other endophytes may still be present as a contaminant in or on the surface-sterilized nodule. This possibility of endophytic contaminants involved in a cross-inoculation study was sustained in the discussion of cross-inoculation trials between Alnus glutinosa and Myrica gale (35). Until recently, no way was known to bypass this possibility of endophytic contamination during the cross-inoculation assays. However, Lalonde et al. (30) have recently demonstrated that the Alnus crispa var. mollis root nodule endophyte can be isolated in pure culture in a non-infective state and that this endophytic isolate can be used to produce an immunomarker (fluorescein- and ferritin-labelled antibody) which can specifically bind to the normal endophyte present in the A. crispa root nodule. The normal endophyte of a given host plant is the endophyte which occurs in the nodules of a given nodulated host plant growing in its typical
habitat and within its usual geographical distribution. Within the limits of the immunolabelling technique, the A. crispa endophyte can be recognized and differentiated from some of the other non-legume root nodule endophytes (30). We have studied the nodulation of the European A. glutinosa host plant by the North-American A. crispa root nodule endophyte, using the immunomarker to determine the identity of the endophyte in the used A. crispa crushed-nodule inoculum and later to demonstrate the identity of the endophyte in the nodules induced by this inoculum on the root system of the A. glutinosa host plants. As it was known that a certain degree of incompatibility exists between North-American and European Alnus with respect to their endophyte (13), we also studied in detail the nodulation process and the ultrastructure of the root nodules induced on the A. glutinosa host plant by the A. crispa endophyte compared with normal host-endophyte systems (e.g. the A. glutinosa endophyte on the A. glutinosa host plant and the A. crispa endophyte on the A. crispa host plant).
Materials and Methods Culture of Plants The seeds of the Alnusglutir~osa(L.) Gaertn., collected from native trees, were washed with tap water, surfacesterilized with a bromine solution (0.1% v/v) for 5 min, and washed with sterile distilled water. Seeds were germinated on wet gravel for 10 days in the dark at 24°C and transferred into a growth chamber a t 20°C under 10 000 Ix for 16 h each day and in darkness for 8 h each night. With the appearance of the first leaves the seedlings were transferred to a modified Crone's solution (12) containing 1 g/! of K N 0 3 and trace elements according to Allen and Arnon (2). After 10 days of cultivation in this nutrient solution, lots of three seedlings were transferred in beakertype vessels containing 350 ml of the modified Crone's solution without nitrogen. After 2 to 3 weeks, plants with four to five exposed leaves were inoculated with a crushednodule inoculum prepared with root nodules obtained from A. crispa var. mollis Fern. host plants grown in glass tubes (20 x 150 mm) containing 15 cm3 of infective air-dried sand collected from Tadoussac sand dune (Quebec, Canada). These normal A. crispa root nodules were of the spore- type (41). The crushed-nodule inoculum was prepared with a Virtis homogenizer according to Lalonde and Fortin (27). Between 15 and 50mg (fresh weight) of the nodular material was used per 350-millilitre vessel containing three seedlings of the A. glutinosa host plant. Microscopic Obserrmtiorzs For the electron-microscopic observations, the lateral roots were sectioned at 1-3 cm from the root tip. These root pieces, contained in 50 mM N a phosphate buffer, p H 7.4, were then transversely sectioned in 0.5-1.5 c m
153 1
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Texas A&M University on 11/13/14 For personal use only.
LALONDE AND QUISPEL
lengths, depending on their characteristics, e.g., root cap, zone of elongation, zone of growing root hairs, zone of the nodule, etc. This fine sectioning was done with a single-edge razor blade under a dissecting microscope. These root pieces were then immersed in 50 m M N a phosphate-buffered 6% glutaraldehyde, pH 7.4, for 30 min at room temperature. These fixed specimens were then washed 5 times in the Na phosphate buffer and further 0 ~30 min at room temperafixed in buffered 3% 0 . ~ for ture. After five washes in distilled water, these fixed samples were dehydrated in 30% ethanol, followed by a 5-min staining treatment in uranyl acetate (1% in 30% ethanol), followed by a dehydration series of ethanol in water (30, 70,95, and 100%) and propylene oxide (100%) for 10 min each. The root samples were embedded in Epon 812 for 24 h at room temperature and then polymerized for 48 h at 60°C. Ultrathin sections were cut with a diamond knife, stained on Formvar-coated grids with uranyl acetate - lead citrate, and examined in a Ph~lipsE M 300 electron microscope operating at 60 Kv. For the light-microscopic observations, the root samples, embedded in Epon 812, were sectioned on a LKB Pyramitome with a glass knife t o a thickness of 1-10 Hm, fixed to glass slides on a hot plate, and stained with 1% toluidine blue in 1% borax solution. The PAS (periodic acid - Schiff's) reaction (24) was used on Epon sections of root nodule to detect total carbohydrates of insoluble polysaccharides. These pink, PAS-treated sections were then differentially stained with the toluidine blue - borax stain. Immunotechnique The specific gamma globulin against the whole cell antigens of the A . crispa root nodule isolates were those of Lalonde er a / . (30). The stock serum was stored at 4°C after the addition of sodium azide t o give a final concentration of 0.1%. The preparation of fluorescein isothiocyanate (FITC) conjugated gamma globulins and their utilization were described by Lalonde et al. (30). However, for the fluorescent-antibody (FA) staining of the A . glutinosa endophyte we first had to degrade the intensely autofluorescent pectic capsule of the endophyte. For this pectinase degradation of the endophyte capsule, the procedure of Lalonde er al. (30) was used except that their antibiotics were replaced by 100 ppm of thimerosal in phosphate-buffered saline solution (PBS, 50 m M N a phosphate, p H 6.0, 0.6% NaCI). Stock solutions of F A conjugates were diluted in PBS. Adsorption of antisera was performed at 37°C for 2 h using equal volumes of antiserum, conjugated or unconjugated, and packed, Formol-fixed endophyte suspension (30) obtained from the spore- type or spore+ type root nodules (41) of A . glutinosa and from the spore- type A . crispa root nodules. The immunoferritin reactions were done using the described technique of Lalonde er al. (30). The ultrathin sections containing the ferritin-labelled specimens were observed with a Philips E M 300 electron microscope operating at 60 Kv Acetylene Reduction Assay The acetylene reduction assay was carried out on root systems sectioned from the plant shoot. One t o three root systems of inoculated plants were placed in 65-11-11glass tubes and kept under a gas phase of 10% v/v acetylene in air. After a 1- t o 4-h-incubation period at room tempera-
ture, a sample of gas phase was withdrawn and assayed for acetylene reduction by gas chromatography (10).
Results Immunoreactions The immunofluorescence and immunoferritin reactions, summarized in Table 1, indicated that the homology of the A. crispa endophyte (spore- type) with the A. glutinosa endophyte (spore- and spore+ types) was limited to a small number of common specific antigens. The fluorescein- and ferritin-labelled anti- A . crispa isolate conjugate ; reacted only slightly with the A . glutinosa endophyte. This point is confirmed by the observation that anti- A, crispa - isolate conjugates, preabsorbed with the A . glutinosa endophyte (spore - and spore +), still react strongly with the A . crispa endophyte and A . crispa isolate. These findings demonstrate that the endophytes of A. crispa and A. glutinosa belong to different specific serotypes. The limited serological homology between these Alnus endophytes, however, allows the specific recognition of the A . crispa endophyte in nodules induced on the A . glutinosa host plant by an A . crispa crushed-nodule inoculum in the absence or presence of the A. nlutinosa endophyte. It should be noted that the-reverse reaction-, recognition of the A . glutinosa endophyte in the presence of the A . crispa endophyte or isolate, cannot be done with the anti- A . crispa - isolate conjugates. Nodulation Process After exposure to the A . crispa crushed-nodule inoculum, the A . glutinosa root hairs, which were initially straight, become deformed. This deformation of the root hairs, which is visible 1 day after the inoculation, takes place behind the root tip and in the elongation zone of the root where the root hairs are growing (32). These growing root hairs include very young ones just appearing on the root and older ones almost at the end of their growth. This pattern of deformation of the growing root hairs is seen as a slope of deformed root hairs along the elongation zone of the root. In addition, all the root hairs which are produced after the initial inoculation are deformed and stay very short. After inoculation on the A . glutinosa root system, the A. crispa endophyte was seen as an encapsulated vesicle attached to a short parental hypha. Then the endophyte seemed to proliferate in the rhizosphere surrounding the zone of growing root hairs of the
1532
CAN. 1. MICROBIOL. VOL. 23, 1977
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Texas A&M University on 11/13/14 For personal use only.
TABLE 1. Immunological reactions of Alnus root nodule endophytes stained with FITC and ferritin-labelled conjugates of non-infective A . crispa var. tnollis isolates
Antigen pectinase treated"
Conjugatesb against :
A . crispa endophyte from A . crispa nodule
Isolatese Isolates/Ads. A . crispa endophytef Isolate/Ads. A . glutit~osaendophyteg Normal serum
Fluorescein reactionc
Ferritin reactiond
A . glutinosa endophyte from A . glutinosa
nodule
Isolates Isolates/Ads. A. crispa endophyte Isolates/Ads. A . glutinosa endophyte Normal serum
Endophyte from A . glutinosa nodule induced with crushed-nodule inoculum of A . crispa
Isolates Isolates/Ads. A. crispa endophyte Isolates/Ads. A . gl~itinosaendophyte Normal serum
'All endophyte suspensions were pretreated with pectinase to remove the pectic capsule. as described in Methods and Materials. fluorescein isothiocyanate (FITC) conjugates were used at a I : 6 dilution. For the ferritin reaction, all immune sera, diluted to 1 :6, were followed with ferritin-labelled goat antibody against crude rabbit gamma globulin.
Cllnrles F . Ketteritlg Resenrctl Lohorntory, Yello\v Springs, O H , U.S.A. 45387 Botntzislz Lnhorntorirrt?~der Rijksrrtliversiteit. Not~tletlsteeg3, Leiden, Nederlntrd Accepted July 20, 1977 L A L O N D EM., , and A. QUISPEL.1977. Ultrastructural and immunological demonstration of the nodulation of the European AInrrs glrrtinosn (L.) Gaertn. host plant by the North-American Altlrrs o i s p n var. ttzollis Fern. root nodule endophyte. Can. J. Microbiol. 23: 1529-1547. The inoculation of the European Alt~rrsglirtitlosn (L.) Gaertn. host plant by a crushed-nodule inoculum, prepared with the North-American Alnrrs o i s p n var. tnollis Fern. root nodule, was successful. Fluorescein- and ferritin-labelled antibodies, specific against the A. crispcr var. mollis root nodule endophyte (Lalonde et 01. 1975), demonstrated the identity of this endophyte in the resulting nodules. The nodulation process of this abnormal host-endophyte system was studied by light and electron microscopy. An excretion of host blebs containing electron-dense polysaccharide material, resulting in the formation of exo-encapsulation threads containing presumptive endophytic bacterial cells, was associated with deformed root hairs. Originating from an exoencapsulation thread, the endophyte penetrates the root hair cell and then migrates as a hypha toward the cortical cells ofthe root. Its migration in the cortical cells of the primary nodule results in the induction of a lateral root which develops a s the true nodule. The ultrastructure of the A. crispn var. tnollis endophyte developing in the primary and true nodule of the abnormal A. gl~rtitlosnhost was similar to the one induced inside its normal A. crisper var. tnollis host. The actinomycetal intruder was a branched and septate hypha able to produce septate vesicles. The endophyte was always encapsulated in an electron-dense polysaccharide material surrounded by a host plasma membrane envelope. However, in this abnormal host-endophyte system, the number of primary nodules formed per root system was drastically reduced, and their appearance was delayed by 1 to 2 weeks. The delayed nodules were effective in fixing nitrogen and able to support satisfactory plant growth in a nitrogen-free medium. L A L O N D EM., , et A. QUISPEL.1977. Ultrastructural and immunological demonstration of the nodulation of the European Altlrrs glrrtinosn (L.) Gaertn. host plant by the North-American Altrrrs oispcr var. tnollis Fern. root nodule endophyte. Can. J. Microbiol. 23: 1529-1547. L'inoculation d e la plante hbte europeenne AItlrrs glrrtinosrr (L.) Gaertn., avec des nodules racinaires de la plante hBte nord-americaine Altzus crispcr var. ttrollis Fern.. fut un succes. L'emploi d'anticorps marques H la fluoresceine et a la ferritine, anticorps specifiques contre I'endophyte du nodule racinaire d ' A . crispcr var. tnollis (Lalonde et 01. 1975), a demontrC I'identite de cet endophyte dans les nodules produits sur le systeme racinaire d e la plante hBte A. glrrtitrosu. Le processus d e nodulation de c e systkme hbte-endophyte a ete CtudiC par microscopie photonique et electronique. L'excretion de globules appartenant a I'hbte, et contenant un materiel polysaccharidien opaque aux electrons participe a la formation de tubulures exoencapsulaires incorporant des cellules bactkriennes supposees endophytiques de nature. Ces tubulures se retrouvent associies aux poils absorbant dCformCs. Originaire d'une tubulure exoencapsulaire, I'endophyte penetre la paroi cellulaire d'un poi1 absorbant et migre comme un hyphe vers les cellules corticales d e la racine. Cette migration d e I'endophyte dans les cellules corticales du nodule primaire provoque I'induction d'une racine laterale qui devient un nodule veritable. L'ultrastructure d e I'endophyte d ' A . crispn var. tnollis croissant dans le nodule primaire et dans le nodule veritable de la plante hbte anormale A. glutinosa est similaire celle produite dans la plante hate normale A. o i s p n var. mollis. L'intrus microbien est un hyphe septC et ramifit capable d e produire des vCsicules sept6es. Cet endophyte est toujours encapsule par un materiel polysaccharidique opaque aux electrons, le tout Ctant entour6 par une enveloppe membraneuse appartenant I'hbte. NCanmoins, dans ce systkme anormal h8te-endophyte, le nombre d e nodules primaires form& sur le systeme racinaire est rCduit considCrablement, et leur apparition est retardee d'une a deux semaines. NCanmoins, ces nodules retard& sont efficaces pour fixer I'azote et sont capable d e soutenir une croissance normale de la plante hbte se developpant dans un milieu exempt d'azote. 'Contribution No. C-590 from Charles F. Kettering Research Laboratory. 'Author t o whom inquiries a n d requests for reprints should be addressed.
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Texas A&M University on 11/13/14 For personal use only.
1530
C A N . J. MlCROBl OL. VOL. 23. 1977
Introduction Nitrogen-fixing root nodules occur on nonleguminous angiosperms (13, 14). Over the years, there is more and more evidence of the ecological importance of this nitrogen-fixing symbiosis (37). In microscopic studies, it is indicated that the root nodule endophyte is a pleomorphic prokaryotic microorganism which is assumed to be an actinomycete (20, 8, 9). However, the isolation, the subculture, and the demonstration of the nodulation of the aseptic host plant with an infective pure culture of the isolated prokaryotic endophyte, cultured in vitro, had not been achieved according to Koch's postulates. Nevertheless, the non-leguminous root nodule endophytes were speculatively classified as Frankia species by Becking (6, 7). The proposed taxonomic position of the Frankiaceae family is based on morphological and physiological characteristics of the endophytes as observed in the root nodules borne by the different host plants. The 10 proposed species were based mainly on the results of cross-inoculation studies which suggest that the non-legume root nodule endophyte is specific for a certain host plant genus or a group of related genera. In these crossinoculation assays, because of the lack of virulent pure cultures of the endophytes for inoculation purposes, root nodules are taken from a nodulated host plant, surface-sterilized, crushed to make a water-suspension, and used as a crushednodule inoculum. However, as cautioned by Bond (13), the possibility exists that infective particles of other endophytes may still be present as a contaminant in or on the surface-sterilized nodule. This possibility of endophytic contaminants involved in a cross-inoculation study was sustained in the discussion of cross-inoculation trials between Alnus glutinosa and Myrica gale (35). Until recently, no way was known to bypass this possibility of endophytic contamination during the cross-inoculation assays. However, Lalonde et al. (30) have recently demonstrated that the Alnus crispa var. mollis root nodule endophyte can be isolated in pure culture in a non-infective state and that this endophytic isolate can be used to produce an immunomarker (fluorescein- and ferritin-labelled antibody) which can specifically bind to the normal endophyte present in the A. crispa root nodule. The normal endophyte of a given host plant is the endophyte which occurs in the nodules of a given nodulated host plant growing in its typical
habitat and within its usual geographical distribution. Within the limits of the immunolabelling technique, the A. crispa endophyte can be recognized and differentiated from some of the other non-legume root nodule endophytes (30). We have studied the nodulation of the European A. glutinosa host plant by the North-American A. crispa root nodule endophyte, using the immunomarker to determine the identity of the endophyte in the used A. crispa crushed-nodule inoculum and later to demonstrate the identity of the endophyte in the nodules induced by this inoculum on the root system of the A. glutinosa host plants. As it was known that a certain degree of incompatibility exists between North-American and European Alnus with respect to their endophyte (13), we also studied in detail the nodulation process and the ultrastructure of the root nodules induced on the A. glutinosa host plant by the A. crispa endophyte compared with normal host-endophyte systems (e.g. the A. glutinosa endophyte on the A. glutinosa host plant and the A. crispa endophyte on the A. crispa host plant).
Materials and Methods Culture of Plants The seeds of the Alnusglutir~osa(L.) Gaertn., collected from native trees, were washed with tap water, surfacesterilized with a bromine solution (0.1% v/v) for 5 min, and washed with sterile distilled water. Seeds were germinated on wet gravel for 10 days in the dark at 24°C and transferred into a growth chamber a t 20°C under 10 000 Ix for 16 h each day and in darkness for 8 h each night. With the appearance of the first leaves the seedlings were transferred to a modified Crone's solution (12) containing 1 g/! of K N 0 3 and trace elements according to Allen and Arnon (2). After 10 days of cultivation in this nutrient solution, lots of three seedlings were transferred in beakertype vessels containing 350 ml of the modified Crone's solution without nitrogen. After 2 to 3 weeks, plants with four to five exposed leaves were inoculated with a crushednodule inoculum prepared with root nodules obtained from A. crispa var. mollis Fern. host plants grown in glass tubes (20 x 150 mm) containing 15 cm3 of infective air-dried sand collected from Tadoussac sand dune (Quebec, Canada). These normal A. crispa root nodules were of the spore- type (41). The crushed-nodule inoculum was prepared with a Virtis homogenizer according to Lalonde and Fortin (27). Between 15 and 50mg (fresh weight) of the nodular material was used per 350-millilitre vessel containing three seedlings of the A. glutinosa host plant. Microscopic Obserrmtiorzs For the electron-microscopic observations, the lateral roots were sectioned at 1-3 cm from the root tip. These root pieces, contained in 50 mM N a phosphate buffer, p H 7.4, were then transversely sectioned in 0.5-1.5 c m
153 1
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Texas A&M University on 11/13/14 For personal use only.
LALONDE AND QUISPEL
lengths, depending on their characteristics, e.g., root cap, zone of elongation, zone of growing root hairs, zone of the nodule, etc. This fine sectioning was done with a single-edge razor blade under a dissecting microscope. These root pieces were then immersed in 50 m M N a phosphate-buffered 6% glutaraldehyde, pH 7.4, for 30 min at room temperature. These fixed specimens were then washed 5 times in the Na phosphate buffer and further 0 ~30 min at room temperafixed in buffered 3% 0 . ~ for ture. After five washes in distilled water, these fixed samples were dehydrated in 30% ethanol, followed by a 5-min staining treatment in uranyl acetate (1% in 30% ethanol), followed by a dehydration series of ethanol in water (30, 70,95, and 100%) and propylene oxide (100%) for 10 min each. The root samples were embedded in Epon 812 for 24 h at room temperature and then polymerized for 48 h at 60°C. Ultrathin sections were cut with a diamond knife, stained on Formvar-coated grids with uranyl acetate - lead citrate, and examined in a Ph~lipsE M 300 electron microscope operating at 60 Kv. For the light-microscopic observations, the root samples, embedded in Epon 812, were sectioned on a LKB Pyramitome with a glass knife t o a thickness of 1-10 Hm, fixed to glass slides on a hot plate, and stained with 1% toluidine blue in 1% borax solution. The PAS (periodic acid - Schiff's) reaction (24) was used on Epon sections of root nodule to detect total carbohydrates of insoluble polysaccharides. These pink, PAS-treated sections were then differentially stained with the toluidine blue - borax stain. Immunotechnique The specific gamma globulin against the whole cell antigens of the A . crispa root nodule isolates were those of Lalonde er a / . (30). The stock serum was stored at 4°C after the addition of sodium azide t o give a final concentration of 0.1%. The preparation of fluorescein isothiocyanate (FITC) conjugated gamma globulins and their utilization were described by Lalonde et al. (30). However, for the fluorescent-antibody (FA) staining of the A . glutinosa endophyte we first had to degrade the intensely autofluorescent pectic capsule of the endophyte. For this pectinase degradation of the endophyte capsule, the procedure of Lalonde er al. (30) was used except that their antibiotics were replaced by 100 ppm of thimerosal in phosphate-buffered saline solution (PBS, 50 m M N a phosphate, p H 6.0, 0.6% NaCI). Stock solutions of F A conjugates were diluted in PBS. Adsorption of antisera was performed at 37°C for 2 h using equal volumes of antiserum, conjugated or unconjugated, and packed, Formol-fixed endophyte suspension (30) obtained from the spore- type or spore+ type root nodules (41) of A . glutinosa and from the spore- type A . crispa root nodules. The immunoferritin reactions were done using the described technique of Lalonde er al. (30). The ultrathin sections containing the ferritin-labelled specimens were observed with a Philips E M 300 electron microscope operating at 60 Kv Acetylene Reduction Assay The acetylene reduction assay was carried out on root systems sectioned from the plant shoot. One t o three root systems of inoculated plants were placed in 65-11-11glass tubes and kept under a gas phase of 10% v/v acetylene in air. After a 1- t o 4-h-incubation period at room tempera-
ture, a sample of gas phase was withdrawn and assayed for acetylene reduction by gas chromatography (10).
Results Immunoreactions The immunofluorescence and immunoferritin reactions, summarized in Table 1, indicated that the homology of the A. crispa endophyte (spore- type) with the A. glutinosa endophyte (spore- and spore+ types) was limited to a small number of common specific antigens. The fluorescein- and ferritin-labelled anti- A . crispa isolate conjugate ; reacted only slightly with the A . glutinosa endophyte. This point is confirmed by the observation that anti- A, crispa - isolate conjugates, preabsorbed with the A . glutinosa endophyte (spore - and spore +), still react strongly with the A . crispa endophyte and A . crispa isolate. These findings demonstrate that the endophytes of A. crispa and A. glutinosa belong to different specific serotypes. The limited serological homology between these Alnus endophytes, however, allows the specific recognition of the A . crispa endophyte in nodules induced on the A . glutinosa host plant by an A . crispa crushed-nodule inoculum in the absence or presence of the A. nlutinosa endophyte. It should be noted that the-reverse reaction-, recognition of the A . glutinosa endophyte in the presence of the A . crispa endophyte or isolate, cannot be done with the anti- A . crispa - isolate conjugates. Nodulation Process After exposure to the A . crispa crushed-nodule inoculum, the A . glutinosa root hairs, which were initially straight, become deformed. This deformation of the root hairs, which is visible 1 day after the inoculation, takes place behind the root tip and in the elongation zone of the root where the root hairs are growing (32). These growing root hairs include very young ones just appearing on the root and older ones almost at the end of their growth. This pattern of deformation of the growing root hairs is seen as a slope of deformed root hairs along the elongation zone of the root. In addition, all the root hairs which are produced after the initial inoculation are deformed and stay very short. After inoculation on the A . glutinosa root system, the A. crispa endophyte was seen as an encapsulated vesicle attached to a short parental hypha. Then the endophyte seemed to proliferate in the rhizosphere surrounding the zone of growing root hairs of the
1532
CAN. 1. MICROBIOL. VOL. 23, 1977
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Texas A&M University on 11/13/14 For personal use only.
TABLE 1. Immunological reactions of Alnus root nodule endophytes stained with FITC and ferritin-labelled conjugates of non-infective A . crispa var. tnollis isolates
Antigen pectinase treated"
Conjugatesb against :
A . crispa endophyte from A . crispa nodule
Isolatese Isolates/Ads. A . crispa endophytef Isolate/Ads. A . glutit~osaendophyteg Normal serum
Fluorescein reactionc
Ferritin reactiond
A . glutinosa endophyte from A . glutinosa
nodule
Isolates Isolates/Ads. A. crispa endophyte Isolates/Ads. A . glutinosa endophyte Normal serum
Endophyte from A . glutinosa nodule induced with crushed-nodule inoculum of A . crispa
Isolates Isolates/Ads. A. crispa endophyte Isolates/Ads. A . gl~itinosaendophyte Normal serum
'All endophyte suspensions were pretreated with pectinase to remove the pectic capsule. as described in Methods and Materials. fluorescein isothiocyanate (FITC) conjugates were used at a I : 6 dilution. For the ferritin reaction, all immune sera, diluted to 1 :6, were followed with ferritin-labelled goat antibody against crude rabbit gamma globulin.
