THE ROLE OF SPECIFIC GENES IN ~ PRIMARY INFECTION 0F WHEAT AND. .BARLEY BY ERYSIPHE ‘GRAMINIS Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY SHFAU-LOH-YANG 1971 .__.-- ~o--,-A- 3‘ L .IL' ._. I L I B R A R Y ‘ Michigan State University “this This is to certify that the thesis entitled THE ROLE OF SPECIFIC GENES IN PRIMARY INFECTION 0F WHEAT AND BARLEY BY ERYSIPHE GRAMINIS presented by S‘I‘fé'fla ~LOH V420 6' has been accepted towards fulfillment of the requirements for Mdegree inim— Pm‘&% , ‘ 4 "7” ,7 I.) FLE’Z’flz/f/{A/Z/yaT/w Major profess Dam [LT/5’, W}, A; 7/ 0-169 M'.‘i““‘h-.T——_ g amomo av, Va?“ NONE 8: SONS: ,IIQ‘: mm m I. ' BIND: ‘ ' ABSTRACT THE ROLE OF SPECIFIC GENES IN PRIMARY INFECTION OF WHEAT AND BARLEY BY ERYSIPHE GRAMINIS BY Sheau-loh Yang The process of primary infection of wheat and barley by grysiphe graminis consists of a number of morphologically identifiable stages of development: spore germination, for- mation of appressoria, penetration into host cells, formation of haustoria in host cells, and formation of elongating sec- condary hyphae (ESH) which are capable of initiating secon- dary and tertiary infections. Conidia will germinate on a number of different natural and artificial surfaces. A high percentage of normal appearing, mature appressoria was observed on host leaves, the upper surface of epidermal strips, or on the upper surface of enzymatically isolated cuticles. No or few normal appearing, mature appressoria were formed on the lower surface of epidermal strips, the lower surface of isolated cuticles, on reconstructed wax layers, or on a number of different artificial surfaces. The presence or absnece of specific M_£ or :3 genes in the plants from which the epidermal strip or cuticles were isolated did not affect the formation of mature appressoria. Malformed appressoria Sheau-loh Yang were observed on plants that possessed eceriferum (cer) mutations that affect the chemistry and physical structure of wax layers. The g&_and Pm genes in barley and wheat, respec- tively, did not appear to interact with the corresponding 2_ genes in the parasites to affect the morphological develOp- ment of the parasites prior to penetration. The two effects of the incompatible parasite/host genotypes were: (1) the reduction in the percentage of parasite units that formed elongating secondary hyphae, and (2) the reduction of the infection type six days after inoculation. In all four near-isogenic barley lines tested, the range of the percentages of ESH on the homozygous dominant (M M and the homozygous recessive (M _m_@ plants derived from selfing of heterozygous parents (M_£ M was consider- ably larger than on homozygous dominant (M_£ M or recessive (M M plants derived from homozygous parents. It appears that there is a carry-over effect.from the genotype of the parent to the progenies. No differences in the antigenicity of the proto- plasts with different gm genes were detected by homologous and heterologous agar gel double diffusion tests. The differences between rabbits of sensitivity to produce anti— bodies against isolated protOplasts were greater than the differences between tested protoplasts with different gm genes. THE ROLE OF SPECIFIC GENES IN PRIMARY INFECTION OF WHEAT AND BARLEY BY ERYSIPHE GRAMINIS BY Sheau-loh Yang A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Botany and Plant Pathology 1971 2‘" ‘\ ACKNOWLEDGMENTS Sincere appreciation is due to the numerous faculty members and graduate students that I have known at Michigan .State University that have made my graduate education a precious and lasting experience. A special thanks to Dr. A. H. Ellingboe for his guidance, encouragement and understanding throughout the author's graduate program. My thanks are due to my graduate committee, Dr. R. L. Anderson, Dr. E. C. Cantino, Dr..J. A. Boezi, Dr. R. P. Scheffer and Dr. W. G. Fields, for serving on my committee and their critical reading of the thesis. The author is indebted to Mr. Joseph L. Clayton for his technical assistance throughout the author's research program and Dr. S. C. Hsu for his valuable help to prepare the thesis. Financial assistance for this investigation was obtained from the National Institute of Health for which the author is indebted. The continuous moral support of my father and mother from thousands of miles away are above all beyond thanks. ii TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . . . . . . . LIST OF FIGURES . . . . . . . . . . . . . . . . . INTRODUCTION . . . . . . . . . . . . . . . . . . LITERATURE REVIEW . . . . . . . . . . . . . . . . MATERIALS AND METHODS . . . . . . . . . . . . . . Culturing of powdery mildew . . . . . . . Lines of wheat and barley used . . . . . . Designation of genotypes . . . . . . . . . Method of controlled inoculation . . . . Environmental conditions for experiments . Examination of fungal deveIOpment . . . . Test of the segregation of M genes in barley . . . . . . . . . . . . . . . . . AEnzyme solutions for the preparation of cuticle layers and protOplasts . . . . . Preparation of natural intact cuticles . . Reconstruction of wax layer . . . . . . . Preparation of protOplasts from coleOptiles . . . . . . . . . . . . . . Serological techniques . . . . . . . . . Sample size and replication of experiments Statistical analyses . . . . . . . . . . . RESULTS O O O O O O O O O C O O O O O O O O O O 0 CONFIRMATION OF PREVIOUS WORK . . . . . . . . EFFECT OF THE CUTICLE ON THE FORMATION OF MATURE APPRESSORIA . . . . . . . . . . . . . Formation of mature appressoria on isolated natural cuticles from near-isogenic barley and wheat lines . . . . . . . . Formation of appressoria of Erysiphe graminis f. sp. tritici on various surfaces of isolated cuticles and epiderman strips . . . . . . . . . . . . iii Page vi 17 17 ,18 19 20 21 22 23 23 24 24 25 26 ‘27 27 29 29 34 34 36 Formation of mature appressoria of E, graminis f. sp. tritici on various artificial and reconstructed surfaces . . Formation of mature appressoria and elon- gating secondary hyphae by E, graminis f. sp. hordei on mutants affecting the wax layer of barley . . . . . . . . . . . 'Formation of mature appressorium of E, graminis f. sp. tritici on isolated natural cuticles after washing with organic solvents . . . . . . . . EFFECT OF J GENES IN BARLEY AND Pm GENES IN WHEAT ON PRIMARY INFECTION . . . . . . . . . Effect ofM_1 genes on the formation of elongating secondary hyphae of E. graminis f. sp. hordei . . . . . . . . . Effect of Pm genes in wheat on the forma- tion of ESH of E. graminis f. sp. tritici . . . . . . . . . . . . . . .Effect of Pm andM_& genes on the infection type produced by the powdery mildew fungi on wheat and barley . . . . . . . . .Effect of specific genotypes for incompat- ibility on the growth rate of elongating secondary hyphae produced by barley mildew . . . . . . . . . . . . . . Formation of elongating secondary hyphae as a criterion for the identification of segregatingM_£ genes . . . . . . . . ANTIGENIC PROPERTIES ON THE SURFACE OF CELL PROTOPLASTS ISOLATED FROM.WHEAT COLEOPTILES WITH DIFFERENT gngENES . . . . . . . . . . . DISCUSSION . . . . . . . . . . . . . . . . . . . . . S MMY O O O O O O 0 O O O O O O O O O O O O O O 0 LITERATURE CITED . . . . . . . . . . . . . . . . . . iv Page 37 38 43 43 43 47 52 52 54 67 73 84 87 Table 1. LIST OF TABLES Formation of mature appressoria of Erysiphe graminis on isolated natural cuticles from near-isogenic barley and wheat lines . . . Germination and formation of appressoria of E, graminis f. sp. tritici on various sur- faces of isolated cuticles and epidermal strips . . . . .y. . . . . . . . . . . . . Germination and formation of appressoria of E, graminis f. sp. tritici on various arti— ficial and reconstructed surfaces . . . . . Germination, formation of appressoria and elongating secondary hyphae of Erysiphe graminis f. sp. hordei on barley seedlings with wax mutations . . . . . . . . . . . . Spore germination and formation of mature appressoria on isolated natural cuticles after leaching with chloroform or ether . . .Action of different genes in the hosts on the sequential development of barley and wheat powdery mildew, cultures CR—3 and MS—l on barley and wheat, respectively . . Page 35 37 39 42 44 53 Figure 1. LIST OF FIGURES _Diagram of Erysiphe graminis development during the stages of primary infection . Diagram of the four possible parasite/host genotypes involving single loci governing compatibility in the plant and the pathogen DeveIOpment of Erysiphe graminis f. sp. tritici during primary infection . . . . . Formation of mature and malformed appres- soria on different natural and synthetic surfaces . . . . . . . . . . . . . . . . . .Formation of elongating secondary hyphae (ESH) by Erysiphe graminis f. sp. hordei (CR-3) on near-isogenic barley lines that possessed the differentiMj genes . . . . . Formation of elongating secondary hyphae (ESH) by Erysiphe graminis f. sp. hordei (CR-3) on five nearéisogenic barley lines all with recessive m1 genes . . . . . . . . .Formation of elongating secondary hyphae (ESH) by Erysiphe graminis f. sp. tritici culture MS-l on five“near-isogenic wheat lines with different gm genes . . . . . . . The length of elongating secondary hyphae (ESH) produced by Ervgiphe graminig f. sp. hordei on near-isogenic paired barley lines which possessed different Ml genes . . . . The number of plants homozygous ML; Mia, heterozygous M M, and homozygous EUE.EJE.0D which a particular range of percentage ESH was produced. .All plants were derived from the selfing of hetero- zygous Mza mza plants . . . . . . . . . . . vi Page 14 31 33 46 49 51 56 59 Figure Page 10. The number of plants homozygous Mjg_MJgJ heterozygous Mjg gig, and homozygous mgg_ggg_on which a particular range of percentage ESH was produced. All plants were derived from the selfing of hetero- zygousMngfgplants............. 61 11. The number of plants homozygous MA MA, heterozygous Mg}; mg, and homozygous M M on which a particular range of percentage ESH was produced. All plants were derived from the selfing of hetero- zygousMgmflkplants............. 63 12. The number of plants homozygous Mjprip, heterozygous Mia M, and homozygous M M on which a particular range of percentage ESH was produced. All plants were derived from the selfing of hetero— zygous Mgp mgp plants . . . . . . . . . . . . . 65 13. Agar gel double diffusion tests of differ- ent wheat protOplasts with different gm_ genes against their antiserum produced . . . . 70 vii I NTRODUCTI O N The powdery mildew diseases of barley (Hordeum vulgare L.) and wheat (Triticum aestivum L.) are caused by ‘Erysiphe graminis D.C. f. sp. hordei Em. Marchal and Erysiphe graminis D.C. f. sp. tritici Em..Marcha1, respec- tively. These fungi are obligate ectoparasites. They are prevalent in most of the barley and wheat growing areas of the world. The diseases cause a very significant reduction in yield of grain in some areas of the world (46, 47, 83, 86). The principal means of control of the mildew diseases have been through the selection and breeding of resistant cultivars. As a result, there is available a large array of host genotypes affecting mildew development, which give dif— ferent degrees of resistance. The genes in the host inter- act with the genes in the parasite, a prediction of the gene-for-gene hypothesis, to affect parasite development and host response (29, 30, 31, 32, 52, 61, 79). There are numerous reports on the comparative physiology and biochemistry of healthy and diseased tissues. -Studies have also been made on the comparative physiology and biochemistry of "resistant" and "susceptible" infection types. The relationships of the differences observed to the genes determining the various types of interactions are not clear. The confusion that surrounds the different inter- pretations of data by the various researchers is probably affected by several factors. Use of different cultures, environmental regimes, inocula, "resistant" and "susceptible" plants of different species, times of observations, etc., have probably contributed to the confusion in attempts by different workers to determine the critical processes in interactions between host and parasite, both for parasitism and pathogenesis. A reasonably defined System for examining the early infection process of E, graminis on wheat and barley has been reported (52, 62, 64). The develOpment of the fungus was shown to consist of a number of develOpmental stages which can be distinguished by changes in morphology and dif- ferential sensitivity to various environmental conditions. With the apprOpriate environmental conditions, the synchrony of morphogenesis can be increased and the percentage of successful infections can be maximized. Conidia germinate under favorable environmental conditions and produce appressoria. Penetration pegs pro- duced on appressoria provide the means for the fungi to get into the epidermal cells and produce haustoria. .Further develOpment of the fungus is by growth of the fungus struc- ture on the surface of the host leaves and the production of additional penetration pegs and haustoria in the epidermal cell layer (Figure l). A.moma .wuamuo>HcD mumum Cmmanoaz .mflmmnu m.flxmcflmoam .m EOHmV .EdAHOmmoummm mumccoomm Ahv .Esfluoumcmc ousume 0cm menu»: mumccoumm mcflummcon AHV .mmmmccmmmm Hmwnoumsmn mcflmon>m© can omen»: mnmccoomm maeummcoam Amy .Hmwuflcfl Hmnmmn mumccooom cam anon HMHHODmSmn comumacm AOV .mvon Hmauoumsmn Amv .mmm newumuumcwm Amv .Eswuommmummm musume.AQv .amfluwca Hmwuommmummm ADV .mnomm cmumcflanom Amv .ouomm woumcwEHomcs Adv .coHuommcw ammEHHm mo mommum mnu mcauso ucoEmoHo>oc mwcfiemum ocmamaum mo Emummwn .H musmam a musmam ZO_._.<.SOOZ_ mmhud. E). E. £8 .8 I 28-2 1.. I o :1 «3 II :36 I. The objectives of this study were the following: (1) to determine the role of the host cuticle layer during the primary infection process, particularly on the formation of mature appressoria by the parasite p0pu1ation before penetration; (2) to examine the action of genes for incom- patibility in the parasite and host on the sequential devel- opment of the parasite: (3) to determine the relationship between the percentage of elongating secondary hyphae to the segregation of genes for incompatibility in the host: and (4) to determine if antigenic differences on the cell proto- plasts could be detected among near-isogenic host lines with and without genes for incompatibility. LI TERATURE REVIEW Numerous articles have been published on the physio- logical, genetic and biochemical aspects of powdery mildew diseases. Many of the articles have contradictory conclu- sions. This review is to summarize the developments con- sidered to be pertinent to the understanding of the diseases and to be related to the results reported herein. Environmental conditions, i.e., light intensity, light period, relative humidity, temperature, etc., greatly affect powdery mildew development (12, 14, 33, 76, 94). Several different optimal conditions for mildew development have been reported (12, 36, 64, 76). Germination and sub- sequent development of conidia occur over a wide range of temperatures and relative humidities. The development of E, graminis spores after depo- sition on the plant surface can be divided into several morphologically distinguishable stages: (1) germination, (2) production of appressorial initials, (3) formation of mature appressoria, (4) penetration of the cuticle and epidermal cells, (5) formation of haustoria, (6) development of elongating secondary hyphae (ESH), (7) initiation of secondary, tertiary, etc., infections, and (8) sporulation (12, 36, 51, 94). It has been shown that particular temperature, relative humidity, and light conditions are required for synchronous development of each of the differ- ent stages of primary infection (51, 64). With Optimum conditions for each stage, over 75% of the spores applied onto the host leaves proceed through the successive stages with a high degree of synchrony (51, 52, 53, 55). The pro- duction of elongating secondary hyphae (ESH) by the parasite indicates that the fungus is obtaining nutrients from the host which permit the continued growth of the parasite (52, 55). .Furthermore, the number of ESH formed on the host surface corresponded to the number of haustoria in the host epidermal cells (52). The percentage of the applied conidia which develop ESH and, therefore, form a functional relation- ship with the host is defined as the infection efficiency (25). Under natural conditions, spores of the fungus are conveyed to leaves by wind. There they germinate and form appressoria as the initial steps toward invasion of epider- mal cells. The stimulus for forming appressoria was thought to be provided simply by a physical contact with the host surface (6, l6). Gold-leaf and collodion membranes were demonstrated to excite appressorium formation of Botrytis by physical contact (6). Physical contact in exciting rust spores to form appressoria was also emphasized (16). Chem— ical stimuli, too, may be important in the formation of appressoria (40, 41, 77). Spores of Puccinia coronata formed appressoria on gelatin if zinc ions were present (77). A chemical stimulus is also apparent in the production of infection cushions by the soilborne pathogen Rhizgctonia solani (40, 41). The hyphae of the strain of R, solani specifically pathogenic to each host produced an appres- sorium-like structure on the surface of the cellophane placed over roots and indicated that the fungus responded to substances diffusing from the roots through the cello- phane (40, 41, 27). Strains not pathogenic to a particular host showed no such response. The wax on living leaves of onion seemed to have no effect on the germination of spores of Alternaria pgggi, but the formation of appressoria was markedly influenced by the presence of wax on the leaves (1). The host leaf surface penetrated is complicated both structurally and chemically. The pegs produced by the appressoria of a pathogen have to penetrate through several distinct layers, with indistinct interfaces, constructed of wax, cutin, pectin, cellulose, and protein (15, 54) to reach the phospholipid plasma membrane and to where the haustoria are formed.‘ .The mechanism of wall penetration by obligately parasitic fungi has long been a subject of speculation. Many plant pathologists have believed that the mode of entry was mechanical (3, 7, 88). Some work on Puccinia graminis on barberries also supported the mechanical theory (58). However, indirect evidence for degradation of cuticle by powdery mildew comes from analyses of the surface layers of healthy and mildewed leaves of apple and turnip (93). The cutin content of mildewed leaves is considerably lower than that of symptomless leaves. Rose leaves infected with §phaerotheca pannosa have only one-fourth of the cutin content of healthy leaves (93). Penetration of the cell wall of the host plant by a pathogen may require the partic- ipation of cutinase, pectinases, pectin-methylesterase, cellulases, peptidases and proteases besides mechanical force (35). Polygalacturonase, cellulase and a hemicellu- lase had been reported in uredospores of Puccinia graminig var. tritici (81). Electron micrographs of some workers showed a crack across the center of the halo where penetra- tion occurred (1) by Erysiphe qraminis, while others did not find evidence for cracks in the cuticular layer (4). Cyto- chemical studies have also suggested that the cellulose wall in the zone around the infection peg is degraded by enzymes produced by the powdery mildew fungi (56). Only recently, observations of the penetration process of powdery mildew into the epidermal cells of barley by electron microsc0py indicate the enzymatic digestion of the cuticle and cellu- lose portion of the epidermal wall by enzymes secreted from the develOping mildew infection peg, and a mechanical push- ‘ ing of the infection peg through a layer of material which has been deposited on the underside of the epidermal walls (22) . The electron microsc0pe has been used for studies on the ultrastructure of the haustorial apparatus and the boundary region of the haustoria and host cytoplasm, the 10 so-called host-parasite interface (4, 23, 24, 57, 68, 78). Little attention has been given to the age of the haustoria examined and to the ultrastructural features of the develop- ing haustoria. Haustoria are thought to be specialized fungal structures for the uptake and transport of nutrients from the plant to the mycelium on the surface of the host (9, 12, 94). ESH never form if haustoria do not form (52). .Fungus develOpment on the leaf surface ceases after the destruction of the haustoria (9). .Electron microsc0py has shown that the haustorial wall of various pathogens is separated from the host cytOplasm by a sac-like membrane (4, 23, 24). The. cytoplasm of the host and the cytoplasm of the parasite appear never to come into direct contact. Biochemical studies on the nature of compatible or incompatible relationships between host and parasite (21, 37, 42, 92) are difficult, since two inseparable metabolic machineries are involved and both are changing continuously. Even if changes in an infected plant could be observed by various chemical or biochemical assays, the cause and effect relationships of these alterations as determinants of dis- ease develOpment is difficult to assess. The most convinc- ing data of a known primary disease determinant comes from a nonobligate fungal parasite Helminthosporium victoriae. A host-specific toxin necessary for disease production was liberated by the fungus prior to the penetration (70, 75). The toxin affects only varieties of oats possessing the Eb 11 gene, which is thought to code for a receptor site for the toxin molecule (75, 87). No such determinants have been found in the physiological and biochemical studies of the powdery mildew diseases. Extracts of E, graminis conidia have been demonstrated to produce disease-like symptoms (8). No specificity of these extracts was found, however. Other substances, such as yeast extract, were shown to produce similar results (82). Many toxic compounds have been iso- lated from 'resistant' plants (82), but again they lack sufficient specificity and have not been shown to correlate with the specific genes conditioning incompatibility. It is clear from many investigations (32, 65) that the interaction of host andparasite are controlled by complementary genetic information possessed by both host and parasite. The ability of Melampggra lini to grow and produce disease symptoms on flax lines containing genes for resistance was determined by specific corresponding genes in the pathogen (29). The existence of one gene in the patho- gen for each gene in the host led to the develOpment of the gene-for-gene hypothesis (31, 66). The gene-for-gene con- cept simply states that each 3 gene in the plant interacts with a specific corresponding §_gene in the pathogen to determine mildew develOpment or parasite/host incompatibil- ity (26, 29, 66). With most diseases studied to date, resistance (R) and avirulence (P) are dominant and virulence (p) and susceptibility (r) are-recessive (26, 48). This 12 concept has been found to apply to many other host-parasite systems (60, 67, 71) (Figure 2). A biological test termed the 'quadratic check' has been proposed to study the physiological and biochemical effects of disease (72) (Figure 2). Incompatibility, or a low infection type, is specified only when the corresponding parasite/host genotype contains at least one §_gene in the fungus and the corresponding 5 gene in the plant (P/R). Compatibility, or high infection type, is specified with the remaining P/r, p/R, and p/r, parasite/host genotypes (48, 72). No difference in the development of Puccinia sorghi on near-isogenic corn lines was observed with compatible parasite/host genotypes. With incompatible genotypes, however, reduced numbers of haustoria, encapsulation of haustoria in the cells, and fungus lysis were observed (38). The rapidity of cell collapse has also been suggested as a basis for resistance to disease development (70). .Measure- ments of the rate ofBSSO transferred from a wheat leaf to 4 the develOping fungus during primary infection suggested that the rate of transfer was lowest with P/R, intermediate with p/R, and greatest with P/r and p/r (79, 80). The concept that an immune response of a host may exert selective pressure on fitness for survival of many parasitic species has recently been supported by several lines of evidence in both animal and plant disease (17, 18, 19. 20, 39). The parasitic worm, Haemonchus contortus was found to display a greater antigenic disparity with rabbit 13 .muflawnaummeoo mmeoomm momxuocom AMKMm.tcm .AflKflm xflM\flM .muwaenaummeooce amaommm momwuocom_flm\flm .ucmam ecu cw ocom.flm onu ou mcatcommouuoo comonumm on» GM moamHam mumcnouam mnu mnm.flm t:m.flM tam woman can cw moaoaam oumcuouam onu mum Am tcm_flfi .comonumm ecu tam Damam ocu Ga xuwaanwummaoo mcacno>om “00H mamcam mcw>ao>ca mommuocom umoc\muwmmnmm manwmmom Hsom may no Emu0mafl. .N wusmflm 14 .54 mm. H.5— aha—ZOO Qua—H4m200 HF.» man—e4.— 2602. HQ 8.. 5: ES. 28 mmhszmnv EGO: HJLLONHO 14390111“! 15 than with sheep, its natural host (18). The failure of sheep to produce antibodies to antigens common to both larval and adult worms was interpreted to indicate that these antigens were common to the host (sheep) and that such antigens constitute 'fitness characters.‘ At present there is no conclusive evidence that similar immune re- sponses are produced in plants directed against pathogenic agents. .Several authors, however, have alluded to the pos- sible Operation of an immune response in plants (10, 20, 73). Others (84, 85, 89, 90) have presented evidence to support the notion that new 'proteins' are formed by plant tissue following infection by pathogenic fungi and bacteria. Evidence was presented to suggest that a specific antigen in each of four races of the obligately parasitic rust fungus Melampsora lini was commonly shared by only those lines of flax that were susceptible to those races. A race of rust tested was avirulent to flax lines lacking the spe- cific antigen (19, 20). The idea is plausible that common antigens between a host and a parasite might provide a less hostile environment for the parasite during the infection process and improve the chance for a successful parasitic relationship. However, more conclusive evidence is needed that the common antigens between host and parasite have some role in determining compatibility between host and parasite.‘ The research approaches to the problem of obligate pparasitism have been varied. They range from studies of gross chemical differences in host varieties resistant and '4 .. I:- I5 16 susceptible to a given strain of pathogen to studies of the host-parasite interface by electron microscopy. Each approach has specific advantages, but none seem to be free from either conceptual or technical problems. It has become increasingly obvious that there will be no royal and direct road to an understanding of infections caused by obligate parasites. Much of the earlier work must be reevaluated as new techniques are available and more information is obtained. It is encouraging, however, that we are becoming aware of the limitations that many of our analytical proce- dures and experimental designs have imposed. MATERIALS AND METHODS Culturing of powdery mildew.--The strain CR-3 of Erysiphe graminis D.C. f. sp. hordei Em. Marchal was main- tained on susceptible barley (Hordeum vulgare L. 'Manchuria‘). ,The strain MS-l of Egygiphe qraminis D.C. f. Sp. tritici Em. .Marchal was maintained on Susceptible wheat (Triticum aestivum L. 'Little Club'). Wheat and barley plants were grown in 4-inch pots and were inoculated when they were 6-7 days old. Sets of wheat and barley plant were inoculated daily by dusting conidia produced 7 days after inoculation onto the leaves of the susceptible wheat and barley. Inoc- ulated plants were maintained in a controlled environment chamber provided with adequate air circulation under the following conditions: Light intensity: 700 to 800 ft-c (650-705 ft-c from.white VHO-fluorescent tubes and 50 ft-c from 25 watt incandescent bulbs). Light period: 15 hr/day. Temperature: 18:10C during the light period and l6il°C dur- ing darkness. The relative humidity (RH) was 65:§%.during the light period and 95:5%.during darkness. Mycelial growth of the culture was first macrosc0pically evident 3-4 days after inoculation. Conidia for experimental uses were abundant by 6 days after inoculation. 17 . v- 18 Line§;of wheat and barley used.-€Five lines of wheat were used. Chancellor, which contains no known major genes affecting mildew development, was used as the standard to which mildew develOpment on other lines was compared. The four backcross-derived lines of wheat which contained gm genes affecting mildew development were obtained from Dr. L. W. Briggle. They were designated as follows: gml 8 8 (Axminster x Cc), Pm2 (Ulka x Cc), Pm3 (Asosan x8 Cc), Pm4 (Khapli x8 Cc). The symbol, x8 Cc, refers to the original cross and seven backcrosses to the cultivar Chancellor (2, 5). All barley lines were obtained from Dr. J. G. Moseman. Manchuria barley was used as the standard to which other lines were compared. Four other lines, each possessing a different gene affecting mildew develOpment, were back- crossed to the variety Manchuria. The four near-isogenic lines and their derivations were as follows: Mfla (Algerian 2C.I. 1179 X: Manchuria C.I. 2330), M29 (Goldfoil C.I. 928 x: Manchuria C.I. 2330), Mflp (Psaknon C.I. 6305 x4 Manchuria 7 C.I. 2330), Mfik (Kwan C.I. 1016 x4 Manchuria C.I. 2330). 8 The symbol, x4 refers to the original cross with Manchuria, 7. three generations of backcrossing to Manchuria, and then the selfing of the heterozygous progeny in each of seven genera- tions. After 7 to 9 generations of selfing, homozygous dominant and homozygous recessive lines were selected. These lines were highly isogenic to each other and are referred to as paired lines. Homozygous dominant, homo- zygous recessive and heterozygous plants were selected for .; .uu .«v I . u o 19 use in this study. Five lines of eceriferum mutants (eceriferum loci control the synthesis and/or excretion of the organic-specific wax components), cer-J59, cer-J7l, cer-zd67, oer-2e81, and cer-zj78, which were induced by either ionizing radiations or chemical mutagens, as well as wild type cultivar Bonus, were kindly supplied by Dr. P. von Wettstein-Knowles (49). Designation of qenotypes.--The barley variety Manchuria contains no known major genes affecting develop- ment of Erygiphe qraminis. By definition, therefore, it contains the recessive alleles at the four loci that are known to determine mildew development. Rather than write out the complete genotype for each host line the following system of abbreviation is used: Designation Actual Genotype g§§g_ Barley: mm mm mm MM Manchuria Marie. mm mm 31.25% 1.4.2: my; mm mm MM 21.29 mm mm 11221122 MM 114.22 new mm mm MM Ms 20 ' , Designation Actual Genotype Used Wheat: pm}; p_m_l_ ng pin; m m3 m4; Lm4_ Chancellor 2:212:11. mam; 3:13.22; meme. Pml mm mm. mine. Btu—49124 .1192 21:12:11 mam; may». stimuli. 1:13. .Eml.Eml EE2.EEZ .Eml.2mi EB£.£E£ BEE The genotype of the strain (MS-1) of E, graminis f. sp. tritici used in this study that is pertinent to this work is g; 23 g; 33, The genotype of the strain (CR-3) of E, graminis f. sp. hordei used in this study is ga_gg,§p_§k, Method of controlled inoculation.--Single 5 to 6 day-old plants grown in 2-inch pots were inoculated by the rolling method (63) in all experiments where the development of the powdery mildew fungus during primary infection was studied. Conidia were dusted onto a clean glass slide and transferred to a single wheat or barley plant with a cotton swab. Only the lower (abaxial) leaf surface of each plant was inoculated. The progress of the infection process is similar on either side of the leaf, but microscopic observa- 'tions and the removal of the host epidermis are much easier um mo ucoEmon>ma .m mucmwm 31 m madmam ZO—hddDOOzn flflhb< GI anonvflfiflcuo— O— Q—N— O— Q 0 Q N BOVINE 3215111 Figure 4. 32 Formation of mature and malformed appressoria on different natural and synthetic surfaces. E, graminis f. sp. tritici on wheat epidermis (A), on isolated wheat cuticle (C), on the reverse surface of isolated wheat cuticle (D), on 2% water agar (E), on reversed surface of wheat epidermal Strip (F), and on reconstructed wax layer (G)- E, graminis f. sp. hordei on wheat epidermis (B) and on wax mutant cer-Zj78 (H). Figure 4 , -s. u.E.‘\ . 7." ‘ £~ - }f$;::~ . , Mo ' 3:... i“ .» s: vgw.f‘..‘.__ ’- 34 EEG: OF THE CUflCLE ON THE FORMATIgm OF MATURE APPREEEQR;§,--One of the long range objectives of this research is to develop procedures whereby all host cells can be infected with the fungus. The advantages of obtain— ing all cells infected, rather than less than 1%, is very obvious for studies on molecular changes associated with or prerequisite to the establishment of compatible, functional relationships between host and parasite. .A series of exper- iments described below were designed to determine if it would be possible to get infection of different types of cell surfaces, to determine if the cuticle is necessary to initiate infection, and whether the chemical composition or physical structure is the determining factor. .gggggtionggmature appreggor;§ on isolated natural Iggggcle§7from near-igogenic barley7and wheat ling§,--Intact natural cuticle layers were enzymatically isolated from five .near-isogenic wheat as well as five near-isogenic barley lines. Each of the five wheat lines possessed different g3 genes affecting mildew develOpment. .Each of the five barley 'lines possessed different E1 genes. The cuticles isolated from.Manchuria barley and Chancellor wheat were used as controls. The formation of mature appressoria of either E, graminis f. sp. hordei or E, graminis f. sp. tritici on different isolated cuticles is shown in Table 1. No sig- nificant differences were found in the formation of mature appressoria among cuticles isolated from plants with 35 Table 1. .Formation of mature appressoria of Erysiphe graminis on isolated natural cuticles from near- isogenic barley and wheat lines Appressoria formed by Erygiphe graminig Near isogenic Barley mildew Wheat mildew barley lines (f. sp. hordei) (f. sp. tritici) %. ‘% Manchuria (31.x) 88 85 Mza 92 -- big; 89 -- iiflg 86 -- M £2 90 -- Near-isogenic wheat lines: Chancellor (pg_x) 83 86 Pmla -- I 94 PmZa -- 90 Pm3a -- 91 Pm4a -- 86 aEight hours after inoculation under standard condition. 36 different genes for reaction. This suggests the genes tested do not act to affect develOpment of the parasite prior to penetration. These genes do not appear to impart a biological specificity to the cuticle layers. Moreover, the data also indicate that the formation of mature appres- soria is not species specific, since E, graminis f. sp. hordei produced nearly the same per cent of mature appres- soria on isolated wheat cuticle as on its natural host barley. Similar results were also observed with E, graminis f. sp. tritici inoculated onto barley. The results obtained with the isolated cuticles were consistent with results obtained with intact plants. Formation of appressoria of Erysiphe graminis f. sp. tritici on vgrious surfaceg of isolated cuticles and epgger- mal strips.-4The development of mature appressoria following inoculations of both sides 0f epidermal strips and both sides of isolated cuticle layers is presented in Table 2. About 85%.mature appressoria were formed at the surfaces with wax layers on both isolated cuticles and epidermal strips 8 hrs after inoculation. Less than 5% mature appres- soria were formed on the surface without the wax layer, and over 40% of the appressoria formed were malformed (Figure 4). The per cent mature appressoria on isolated cuticles grafted onto the lower surface of epidermal strips was essentially the same as on the upper surface of intact epidermal strips. 37 Table 2. Germination and formation of appressoria of E, graminis f. sp. tritici on various surfaces of isolated cuticles and epidermal strips . a Appressoria Surface Germinationa Malformed Mature % %» ‘% Isolated cuticle: upper face (with wax) 93 3 84 lOwer face (without wax) 91 42 l Epidermal strip: upper face (with wax) 93 2 89 lower face (without wax) 90 45 2 Isolated cuticle grafted onto lower surface of the epidermal strip 89 5 82 aEight hours after inoculation under standard conditions. This result is also consistent with the interpretation that the wax layer is a major determining factor for the forma- tion of mature appressoria. Formation of mature appressoria of E.¥graminis f. sp. tritici on variogg artifggial and reconstructed surface§.-- Urediospores of many species of rust fungi germinated on isolated host or nonhost cuticle and sequentially developed appressoria, infection pegs, vesicles, and infection hyphae, similar to those produced in nature during host infection. The effects of cuticle were duplicated by nitrocellulose membranes containing hydrocarbons isolated from the surface 38 wax of snapdragon leaves or containing mineral oil. The surface wax of the leaf, in addition, promoted the formation of haustorial mother cells and the branching of infection hyphae in bean and snapdragon rust fungi (50). In the experiments described here seven different artificial surfaces were inoculated with conidia. Isolated cuticles were also inoculated and used as controls. The percentages of mature appressoria formed on these surfaces are shown in Table 3. The percentages of mature appressoria formed ranged from.0% to 17% on the artificial surfaces tested. .Eleven per cent of the conidia produced mature appressoria on paraffin coated cellulose paper. .Ninety per cent mature appressoria were formed on the isolated cuticles. Only 9 to 17% of the applied conidia formed mature appres- soria on the reconstructed wax layers. Similar results were obtained whether the wax used for the reconstructed layers was extracted from fresh leaves or isolated cuticles. .Since few parasite units produced mature appressoria on the recon- structed wax layers (without fractionation of the cuticular components), it appeared that the specific physical confor- mation of the wax layer may be one of the major factors affecting the formation of mature appressoria. .Formation of magure appresgoria and elongating ggcondgry hyphae by E. graminis f. sp. horde;_on mutants ggfecting the ng layer of barley.--If the chemical composi- tion and the physical structure of the wax layer is important 39 Table 3. Germination and formation of appressoria of graminis f. sp. tritici on various artificia and reconstructed surfaces 1;. 1 Appressoriaa Surface Germinationa Malformed Mature % % % Isolated cuticle 98 2 90 Water agar (2%) 96 32 0 Glass slide: plain 30 O 0 paraffin coated 83 10 O Cellulose paper plain 93 15 0 paraffin coated 96 42 ll Reconstructed wax layer: wax from fresh leaf blade 62 43 9 wax from iso- lated cuticle 90 54 17 aEight hours after inoculation under standard conditions. 40 for the normal differentiation of appressoria of the para- site, mutations which affect the chemical composition and physical structure of the wax layer may differ in their effects upon the parasite. Several hundred independent mutations which give a glossy appearance to the leaves have been induced in barley by a variety of different mutagens (49). These mutations, called eceriferum mutants because they affect the wax layer on the leaves, map at several different loci. The loci are designated by the letter and 59 the allele number by a superscript, e.g., cer-J is muta- tion number 59 at the g locus. .Five ecerferum.(wax) mutations, cer-J59, cer-J7l, 67 81 .78 . cer-Zd , cer-Ze , cer-Zj , were ava1lable for use. These mutations have been partially characterized both by chemical analysis and with a scanning electron microsc0pe. Primary leaves of barley with the mutations gg£3g§9 and g§£3E71 pro- duce per unit area 30% and 56%.less wax, respectively, than the wild type Bonus barley. The wax mutations at the Eggzg, locus give a much smaller percentage of primary alcohols and a somewhat larger percentage of esters than that of wild type Bonus (49, 91). Approximately a 20% reduction in the amount of wax was found on the seedling leaves of barley with mutations cer-Zd67 and cer-Ze81 (91). .Morphologically quite different wax coats were found on plants with either cer-Zd67 or car-2e81, in addition to the wax bodies similar to those found on wild type Bonus. No obvious difference in the pr0portion of the lipid classes composing the wax could 41 be recognized with the latter two mutations, as compared to those of Bonus, the wild type barley (49, 91). Approx- imately 40% less wax was found on the primary leaves of 78 than on wild type Bonus. plants with mutation cer-Zj The wax bodies were smaller and irregularly distributed over the surface of the leaf (91). A higher percentage of aldehydes was found in plants with cer-Zj78, but the struc- tural arrangements of the fibrils in the wax coating were less obvious. .Five to six-day-old seedlings with the different mutations were inoculated. Inoculations of Manchuria as well as wild type Bonus were also made as controls. The results are shown in Table 4. .Eight hours after inoculation the percentage of malformed appressoria produced on barley plants with the various mutations ranged from 14% to 27%. The percentage of malformed appressoria depended on which mutation was involved. Less than F% of malformed appres- soria were produced on the controls. Reduction in the percentage of mature appressoria caused by the wax mutations ranged from approximately 15% to 35%. Reduction of the per- centage of ESH 28 hrs after inoculation ranged from basically no differences to 40%. This result suggested the chemical components and the physical conformation, such as distribu- tion of the wax bodies, are all important factors for the formation of mature appressoria. 42 Table 4. Germination, formation of appressoria and elongat- ing secondary hyphae of Erysiphe graminis f. sp. hordei on barley seedlings with wax mutations ! _ Appressoriaa Host Germinationa Malformed Mature ESHb % % 96 % Manchuria 99 < l 97 77 Bonus 97 < 1 96 71 Ecerferum mutations: cer-J59 96 15 80 68 oer—J71 92 19 72 62 cer—Zd67 97 14 82 59 oer-2e81 97 14 83 55 cer-Zj78 93 27 60 30 aEight hours after inoculation. bTwenty-eight hours after inoculation. 43 Eggmation of mature appressorium of E. graminis f. sp. tritici on isoEQted natural cuticleg after wagging with organic solvents.--The hypothesis that the wax layer is a determininngactor for the formation of mature appressoria was also tested by intensive washing of isolated cuticle with organic solvents capable of desolving chloroform and ether which are waxy substances. Approximately 2 liters of an organic solvent were washed over a single 2.5 cm long isolated cuticle over a period of 6 hrs. Cuticles washed by the same method but with distilled water were used as control. The percentage of mature appressoria 8 hrs after inoculation was approximately 15% less on the cuticles washed with either chloroform or ether (Table 5). The mate- rial on or in the cuticle that stimulates the parasite to develOp mature appressoria must not be entirely removed by the organic solvents used. This result also is in agreement with the hypothesis that some of the wax bodies are firmly impregnated into the cuticle structure (54). EFFECT o_§:_M£ GENMN BARLELAND Pm GENES IN WHEAT ON PREMARY'INFECTION.--EffeEt of Mg genes on the formation of elongating secondary hyphae of E. gramgnis f. gp. hordei.-- Earlier work (52) has shown that ESH are formed.only by para- site units that have formed haustoria. The formation of ESH has been used, therefore, as evidence that the parasite and host have established a compatible relationship. The effect of different E£_genes carried by four near-isogenic barley 44 Table 5. .Spore germination and formation of mature appres- soria on isolated natural cuticles after leaching with chloroform or ether Organic solvent Germinationa Mature appressoriaa % % Chloroform 93 77 Ether 95 72 Control 98 ' 89 aEight hours after inoculation. lines were tested by inoculation with CR-3 of E, graminis f. sp. hordei, a strain that possesses the complimentary genes ,Eg,,Eg, EE, and £2 for incompatibility with the four genes Mia, Mfg, M, and flip, respectively. The formation of elongating secondary hyphae was used as a criterion of the establishment of compatible relations between host and para- site. The results of each gene pair, i.e., Eg/EJE_(parasite/ host genotype) on the formation of elongating secondary hyphae of the barley mildew fungus were recorded from 20 to 26 hours after inoculation (Figure 5). Approximately 8, 16, 32, and 37% ESH were observed 26 hrs after inoculation with the genotypes (parasite/host) Eel/111a, fl/Efig, gig/M, and EE/EJE: respectively, specifying incompatibility. The pos- sibility that genes other than the ones specified contributed to the results obtained was considered unlikely because essentially the same results were obtained on paired lines 45 Figure 5. Formation of elongating secondary hyphae (ESH) by Erysiphe graminis f. sp. hordei (CR-3) on near- isogenic barley lines that possessed the different I‘d—figeneS: M (O-O). M2 (*->I<). 111% (0-0). Mfg (*-*), and Mfia (fi-fi). ESH PERCENT 46 2O 21 22 23 24 25. 26 HR AFTER INOCULA‘I'ION Figure 5 47 with no known dominant El genes, which were known to be very highly isogenic to the lines with E1 genes, as with Manchuria. The kinetics of formation of ESH on each near— isogenic homozygous recessive line was shown to closely parallel the results on Manchuria (Figure 6). Effect of nggenes in wheat on the formation of E§H of E. graminis f. sp. tritici.-—The formation of ESH by E, graminis f. sp. tritici culture MS-l on four near- isogenic wheat lines, each of which possessed a different 2E gene, was also tested. The culture MS-l possesses the four complimentary genes 2;, E2, 2;, and Eg_which specify incompatibility in the presence of Efll, Egg, E31, and Egg, respectively. The results of each gene affecting the formation of ESH were recorded from 20 to 26 hours after inoculation (Figure 7). Approximately 28, 84, 18, and 5% ESH were produced 26 hrs after inoculation in the presence of the parasite/host genotypes EE/EQE, gg/ggg, EE/EQE, and gg/ggg, respectively. The kinetics of ESH formation with EE/Egg genotype closely followed What was observed when the vari- ety Chancellor was inoculated with strain MS-l. This indicated that the Eg/EQE genotype does not alter the primary infection process, at least up to 26 hours after inoculation. .Figure 6. 48 Formation of elongating secondary hyphae (ESH) by Eryglphe graminis f. sp. hordei (CR-3) on five near-isogenic barley lines all with reces- sive M genes. The line designated m, for example, was the homozygous recessive derived from the cross Algerian x: Manchuria. PERCENT E SH 90 80 7O 60 50 4O 3O 20 IO 49 l HR AFTER INOCULATION Figure 6 .Infl *Lnfla * nflg O nflk * mlp 50 Figure 7. Formation of elongating secondary hyphae (ESH) by Erysiphe graminis f. sp. tritici culture MS-l on five near-isogenic wheat lines with different fl). genes. Chancellor (CC) (C-O), m (O-O), P_m2_ (*-*). £111; (*-*), and Egg]; (fi-fi). ESH PERCENT 51 HR AFTER INOCULATION Figure 7 52 .ngect of Pm and.M£ genes on the infection type pro- duced by the powdery mildew fungi on wheat and barley.--Near- isogenic lines of barley and wheat with and without 33 or E1 genes, respectively, were inoculated by dusting the conidia of the apprOpriate fungus onto 5 to 6-day-old seedlings. The inoculated plants were incubated under environmental conditions similar to that used for the maintenance of stock culture. The infection types were recorded 6 days after inoculation. Table 6 summarizes the effect of all the dif- ferent genes on final infection type as well as the primary infection process. Incompatibility is expressed as the reduction of ESH formed during primary infection and the reduction of infection type observed 6 days after inocu- lation. Eggect of specific genotypes for incompatibility on the growth rate of elongating gecondary hyphae produced by barley mildew.--One effect of 7 of the 8 different genotypes specifying incompatibility is to reduce the percentage of the conidia applied to the host surface that produce ESH by 26-28 hrs after inoculation. These results are consistent with published data (55). The fate of the approximately 17% of the parasite units that do form ESH with the EgAEjg|geno— type, for example, is important in the determination of when the genes act to affect the interactions between host and parasite. iClearly the genes do more than reduce primary infection efficiency because the final infection type is reduced. 53 .Aoanaummeoov ucoemoHo>mo zmcaae ucmocsnmlle .o no oceumu ecu Bonn ucoEmoHo>mt Booaaa cw coHuosoou ucmoamacmamllm scofluomou oeuonoocllm .mcaxomam oeuouoanolla .ucoemon>mo Booawe manm>uomno ocllo woman cofluoomcHn .coflum>uomno mo meflu may no ucmfimon>mo mo ommum umnu unmoa um um oo>uomno Godumasmom ouammnmm amuou mo ommucoouomm H.o 0H cm A mm A mean m mmlma om A mm A mmEm N mu om A mm A mmEm H.o omlom om A mm A mHEm v mu - om A mm A HoaaoocmsoH "menu: pawns» vasomOmHIHmoz NJ mvlmm om A mm A mus N oelom om A mm A $2 .10 owned om A mm A $2 0 0.... cm A mm A was v m5 om A mm A mwuonocmz: "mocfla moaumn DecomOmHIHmoz wow we.» new mom» codumasuomm ooze»: memo“ coin canon coflumc muco>o c0auoomcH codumuwcoaoo mumocooom Imsmm -Imuu Imoummm IHEHOO . Hmaucoovom Q mcwummcoam locum macaw: oma omalmm mNION nu ma w v Godumasoocw Houmm muoom hao>auoommou .umonz 0cm moanmn co Hum: com mlmo mwusuado .3ooaaa humozom umon3 tam moaumn mo ucoEmoHo>oo Hmflucosoom onu co mumon can CH mocom ucouommwo mo coauom. .o manna 54 The effect of different genotypes on the length of ESH 32 to 36 hrs after inoculation is presented in Figure 8. The data on paired, highly isogenic host lines which differ in the presence or absence of dominant M genes are also presented to aid in determining whether genes other than the M genes affect the growth of secondary hyphae. The lepe of the lines suggested no significant differences between growth rates of ESH formed on plants of different genotypes (Figure 8), at least up to 36 hrs after inoculation. The inhibition of mildew development, as indicated by the reduction of infection type, by the different genes for incompatibility must be at a stage of development of the parasite other than the elongation of secondary‘hyphae. The genes for incompatibility must, therefore, func- tion by two means. .The formation of haustoria is an all or none develOpment. Haustoria either form or they do not. No deficient or malformed haustoria were found during pri- mary infection. The subsequent develOpment of the parasite units that do form haustoria is more subtly affected by the different genotypes, at least as presently understood. Formation of elongating secondary hyphae as a crite- gion for thegidentification of segregating,M11genes.--Since the percentage of the total applied parasite units which eventually form ESH is dependent upon different genes, it should be possible to identify the segregation of genes in- Figure 8. 55 The length of elongating secondary hyphae (ESH) produced by Egysiphe ggaminis f. sp. hordei on near-isogenic paired barley lines which possessed different El genes. .The line designated Egg, for example, was the homozygous recessive derived from the cross Algerian X3 Manchuria. LENGTH OF ESH (MICRON) 36 32 28 24 20 l6 I2 56 32 33 34 35 36 HR AFTER Figure 8 I NOCULATION 57 the host by the formation of ESH. The purpose of this experiment was to see whether, by use of the percentage ESH formed as a criterion, one can identify the three types of progenies expected, the homozygous dominant.fl1, the homozy- gous recessive ml, and the heterozygote, from the selfing of heterozygous plants. .Seeds derived from the selfing of four near-isogenic heterozygous barley parents were planted individually in 2 inch pots. Approximately 50 seedlings of each line, 5 to 6-days-old, were inoculated with isolate CR-3. A one cm section of each seedling was removed, and the percentage of applied parasite units which formed ESH on that section 28 hr after inoculation was recorded. Infection types on the remaining portion of the inoculated leaves were observed 6 days after inoculation. The plants were grown to maturity and the genotype of each seedling was determined by the segregation of infection types among its progenies. The ranges of the percentages of ESH formed on seedlings were plotted against the number of seedlings which produced the same range of percentage of ESH. The percent ESH on each segregant derived by selfing of plants of the genotypes gig r_n_£_a_, .1113 m, M M, and M M are shown in .Figures 9, 10, 11, and 12, respectively. The percentages of ESH produced by control homozygous recessiVe plants derived from homozygous recessive parents of all four near-isogenic lines were always in the range of 75%.to 90%. Per cent of ESH 58 .mucmam awe sz mdom>N0umum£ m0 mchHmm mcu Eoum ©m>mump mnt mucmam Ham .cmosboum mmB mmm mmmucmonmm mo mmCmH Hmaooflunmm m £3.23 :04 D v qu qu msommuoeos 98 .A a v mesa was. 2 msommmouwumc .A —-— v mm: mm: msomhmoEon mucmam m0 umflEDC one .0 musmfim 59 m whomwm mag—.ZfiOu—Nm 7: IO”.— 00— Do cm ON 60 On Ow an ON O— O IIIIIIIIIV _ WWIIBDBEIII II BEBE EB BBB BBB W B B U 3E3: x 0.50.2 NOflOQN l-l- SlNV'Id :10 “CHIN!!!“ 6O .mucmam meme mm: msomhmouwumn wo mammamm map Eouw Um>flumc mumB mucmam Hafl .Umosboum mmB mmm mmmucmuumm mo mommy nmasofluumm m £033 so A D v me MT: msommmoeos cam .A a V my“. E 9...: msomxuonmumn .A I v was: mewz moomxuoeoz mucmam wo gonads one .oa muomflm 61 OH onsmum u94hzm—Oflmh Zn ION cap no a» 9n 00 On ow On ON O— I VIV BDW WUEB DDIIDB B EB REGIME U WEBB BBB 929.2 x 2.: 32 o SINV’IJ :10 “HENGN 62 .mucmam was: $2 msomxuoumuwc mo mammamm on» Eoum Um>flum© mumB mucmam HH< .Umosooum mmB mmm mmmucmoumm mo mmcmu Hmasofluumm m 3333 so A D V as: wme msomxuoeoz cam .A a V $3: $me msomwmonmumn .A . v xwz up“: woommmoeoc mummam wo Hones: och. . HH wHDmH .m 63 HA mnsmam H°flum© mumB magmam HH4 .omosooum mmB mmm mmmucmoumm Mo mmcmn umasoflunmm m Suits co A D v muse mm): msomxuoaon cam .A a v mmE mm 2 msommuoumumn .A v mm: mm: msommmoaon mquHm wo Honcho: one .NH musmflh 65 NH magmas QO