‘5‘" s. g). x o t in. I 2‘ \— Ga .- . . .2 .3. , i a. 62% 9 , a L . Ia... .. 9 . . {3, . $3509.“ ,_.. s \ Suing! ‘ . a «2“. it. 4. ‘ 5.?“5’34. 5)) ‘5.“(3 hum}: 3%3’4 5333\KFI ’: 3.5)- ‘ a v . l 58555351255 i v 1 , ,. . , \ ,akiizfikgfiaast v , v I. o .. t 4v? , > 1 0.. . . Vk {.V. zbnvhrufln . at 1 4 , ~ . .Wwdgfli .. . ‘ , \ . x: 7!; r q ‘ 0917.???“ g x s 79!. o h... 4..- , g": 4&fibp‘fl’aflefiuuth . . It». . r e ‘ g . . . r.- F I o a 4... ,. , . i .. . 4.. fin. : . $3, .11.; it .v . a u. a. , ligag‘i’v 2: 3‘ I .7 i q a . .A . all“? ‘ 512‘) >315!!! 253 : 1 17:21.1)...“ 55.2.3141}! 1.53.336)» §57§5§3t§21iv 3%}; H5 1 55‘ igfliggin IKgJ§nuuuWfl 52 #415. “4013.... \cygrflzggivgg 5; . sill: )\. . ‘1. 3 is); 2.: .5! 1.53.. . 1521;. 15.}... yzfiflfifsa fln‘s‘vflyiiigvzugfiamwflzaaxs. .2. ‘ I19. 1 if”; .{f 1;)?35 25“. A A ‘1 gi‘)... 1 1.131%? . 3 , 51")». ‘1') 3 )3 Cl. .1 ‘5 .3. 3? gluing“. «I If .. «1s . ‘52:. v. . . . ‘ , ~54...- ) 1 I V 35?3?u}'.\w 21") 3 , . 1.11.9:1!!!)32' v A. (It! s Q v ‘ , V 79.0.... . Q "awe: ggéfx it..." If . .r a n u s , #13:: we; , , 1 . ; . ~< a. 3572?; Ht ‘ x , n 4. I p v ‘ 1 , t 3“ t l L. .. . 5 I ‘ . ‘ to! r .4! {Ar-Aggy ‘ . a . _ . . 3. . n , , - . §I.m.f:u‘fdwofizmm.hftfirdflltt 1mm. . n x . ..u .: 3 . . . . .35: t , £355 V I 1 , WangWflVfilhfigh! . ‘ .wgflfn.; g5. ( V .7 . 5 .. ,. , . . . .. , V . 5345;533:123» a 1 I I... .52“?! 1 ‘ ... A. .n l I 7 fr . . . .55... fsggktg; u... 9... vhlgulflllfmfg ‘ a £1 : ., 1.41.: I ,LvL! A‘ . {fiqug 9 . .9 , 31¢. Avg; 1532.32.35.04}!!! iigiv’ \‘ ‘15; l FAIII§ . g§§f u 3 . .. I]!!! 1321).; . git!!! , ggfif.’xium 34 1 It'll»: 1. . .. . 1.555.? 5:31.? .13: sixilxa.§lx! .35”: .. . n... .. llgnnsg :1... 6-3;? ¢ . .1 r\ V I . n . . , . . £3 V iii-$1) g5¥4§élfllfl§§2§ . 5:35): . 4 . I 1 0 . , , .) . ‘ hall. ital... , IL+1A . , . 'IM ‘1 ' i; ‘ 1. 5- :.r. 4 , .' ‘ ‘ s. {my * z ' I ' ‘ . . ‘Llnlol..fl|llltlr l . ‘ ‘.(ll..ra.r\.J.f . {114:}. NH. - .inf« H. I I4... {fut/1 lot! 1.". I A 1 . It‘ll .1y('r.>s4\|...\llr¢| l ,tllinl 1.5 4 t |fi.y Lttt a . . :.I w . . i .1-.. 91.1! . n 1.! LILAC! ‘ ‘I 14.11 I .C . . .. .TZIt’TIIIL :3 . IAL ; ‘ . ‘ 3ft}... . .. . I J . g‘._kv.l¢11rf\:v ((6.? Inl.fl..\lfrl...l. Ill; ‘ 1.79:; nirlraulrpts’svlbu (C(i nlt’.‘.(;II)Ls|rL.l.lll.l 511:1.f1LtlLtlllflf. ACIII?‘1 21.1)? Ll\({\l. .IAI)»’I£.(¢I¢Itlv)I.t&n-txl ‘ | In. (I, It: ‘11:. k . {5’ LL: 1!! 5)\‘¢IAEI| » u 11.? 3|, I...) ’loi, ‘ . 1 «(I .L 1“:\ . .51... 1 . ‘ Iltk. )1.\Ilv{vf:|. ... . v u]: g‘ A"... :L ,[i( . Tl:fL\.. » Ixtitfitly, , {If f. . vt... t. .1 WA \ v . llIthll.,.l. . \Q not}. . . i. (.35: {1&‘clc. .1 : 4ylt. fittf‘ \(CAIA . £11.12 uféA.y.||t- .01! (V.I1I\|..yxnluf alt ,‘I/y I . A t i. lfrrn IL AAAAA . Er , f V‘lflllfAL’l££-\y..f\§\.3 ltwlll..a. 1.... . . . 1. a v, u ‘ y ‘ ‘ IIP‘.~..HI4K ‘ g, LIBRARY ‘” Michigan State ' University ill}IllllllllljlflllllLllljllflllllllllflllal|l ’. This is to certify that the thesis entitled PRODUCTION, ISOLATION AND IDENTIFICATION OF UNIQUE LIPIDS BY AZOTOBACTER_V‘INELANDII presented by Chung-Jey Su has been accepted towards fulfillment of the requirements for Ph.D. degree in Microbiology Major professor Date November 81 1979 0-7639 l OVERDUE FINES: 25¢ per day per item RETURNING LIBRARY‘MATERIALS: P1ace in book return to remove charge from circulation records PRODUCTION, ISOLATION AND IDENTIFICATION OF UNIQUE LIPIDS BY AZOTOBACTER VINELANDII BY Chung-Jey Su A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Microbioloqy and Public Health 1979 ABSTRACT PRODUCTION, ISOLATION AND IDENTIFICATION OF UNIQUE LIPIDS BY AZOTOBACTER VINELANDII BY Chung-Jey Su Encystment in Azotobacter vinelandii can be induced by transfering exponentially growing cells from glucose to B-hydroxybutyrate (BHB) as the carbon and energy source in the medium. Because of its specificity as an inducer, the metabolism of BHB during the encystment was studied. Using l4C-labeled BHB, two—thirds of the radioactivity incorporated into five—day old cysts was recovered in the lipid fraction. Thin layer chromatographic (TLC) analysis of cyst lipids revealed at least nine diazoreagent-reacting compounds. These were designated as AR3, ARS’ ARl, AR4, APl’ AP2, AP3, AP and AR according to their motility on the chromatogram. 4 2 Beside ARl and AR2, which have already been identi— fied as mixture of 5-n-heneicosylresorcinol, 5—n—tricosyl- resorcinol and their galactose derivatives, AR was identi— 3 fied as mixture of 6—n—heneicosylresorcylic acid methyl ester and 6-n—tricosylresorcylic acid methyl ester. AR5 was identified as mixture of 5—n—heneicosyl-4-acetyl—resorcinol Chung-Jey Su and S-n-tricosyl-4-acetyl-resorcinol. AR4 was identified as mixture of 5—n-(2-hydroxy)-heneicosylresorcinol and 5-n- (2—hydroxy)—tricosylresorcinol, APl was identified as mixture of 6-n—heneicosyl—4-hydroxy-pyran-2-one and 6—n—tricosyl-4- hydroxy-pyran-Z-one and AP2 identified as mixture of 6- (2-oxotricosyl)-4—hydroxy—pyran-2—one and 6-(2-ox0pentacosyl) -4—hydroxy-pyran-2-one. AP3 and AP4 were not identified, but their reactions to various Spray reagents were similar and AP to AP These nine components account for 81% of l 2‘ the total radioactivity in the lipid fraction of mature cysts. The carbon atoms in these compounds were established to be mostly derived from BHB used to induce encystment. The biosynthesis of these compounds during encystment were studied by inducing with 14 C-BHB, extracting the lipids, separating them by TLC and then determining the radioactivity in each compound. The synthesis of these unique lipids started at 8-12 hours after the induction of encystment and reached a plateau 2 days after the induction of encystment. After their formation, these compounds remained relatively stable for the rest of encystment period. Most of the phenolic compounds (ARl, AR2, AR3 and AR4) remained relatively stable during the germination but the pyronic compounds (APl and APZ) were degraded during the germination. About 70% of these unique lipids were located in the cyst central body, 23% in the exine and 7% in the intine. The phenolic compounds were evenly distributed in exine, Chung-Jey Su intine and central body according to the proportion of lipid distribution in these three fractions but the pyronic compounds were concentrated in central body fraction. The physiological significance of these compounds for the cysts is unknown but they may function as structural components. Their biosynthesis pathways are presumed to be via polyketides like similar natural products. Dedicated to my parents, my wife and my daughter. ii ACKNOWLEDGMENTS I wish to express my sincere appreciation to my academic advisor, Dr. Harold L. Sadoff, for the guidance, encouragement and patience throughout this study. I also wish to express my thanks to Dr. Rosetta N. Reusch. Without her enthusiasm and expertise, the identi- fication of these compounds would not be possible. The excellent service provided by the M.S.U. mass Spectrometry facility and the NMR spectra and chemical ionization mass spectra taken in the Department of Chemistry, M.S.U. are gratefully acknowledged. Sincere thanks are due to Dr. William Reusch for his valuable input into the interpretation of certain NMR spectra. The advice and encouragement of my guidance committee, Drs. L. L. Bieber, T. R. Corner and R. N. Costilow is also gratefully acknowledged. Finally, I wish to express my appreciation for the financial support from the Department of Microbiology and Public Health, Michigan State University. iii TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . LIST OF FIGURES. . . . . . . . . . . . INTRODUCTION. . . . . . . . . . . . . LITERATURE REVIEW . . . . . . . . . . . A. Physiology of Azotobacter vinelandii. . . B. Cysts of Azotobacter vinelandii . . Induction of encystment and germination Morphological changes during encystment and germination. . . Metabolic changes during encystment and germination . . . . . . . . Compositional differences between vegetative cells and cysts of Azotobacter vinelandii . . . . . C. Poly-B-hydroxybutyrate (PHB) Metabolism in Azotobacter . . . . . . . . . MATERIALS AND METHODS. . . . . . . . . . Growth of Bacteria . . . . . . . . . Induction of Encystment . . . . . . . . Germination of Cysts . . . . . . . . . Lipid Extraction. . . . . . . . . . . Removal of PHB . . . . . . . . . . Methylation of Samples. . . . . . . . . Chroamtographic Methods . . . . . . . . A. Column chromatography . . . . . B. Thin layer chromatography . . . . . C. Gas chromatography . . . . D. High pressure liquid chromatography. . Spectrometric Methods . . . . . . . . . iv Page vi vii 10 12 15 17 23 23 23 23 24 24 25 25 25 27 28 28 29 DOW?” Melting Point Determination Radioactive Isotope Labeling Study Fractionation of Cysts . Ultraviolet Infrared . Mass spectra. Nuclear magnetic resonance RESULTS AND DISCUSSION. Page . . . . . 29 . . . . . . 29 . . . . . . 30 spectra (NMR) . 3O . . . . 31 . . . . . 31 . . . . . 33 . . . . . . 34 Comparison of Lipid From Vegetative Cells and Cysts . . . . . . . . . . 34 Purification of AR3, AR5, AR4, APl and APZ . . . 37 Identification of APl . . . . . . . . . . 40 Identification of AP2 . . . . . . . . . 54 Identification of AR3 . . . . . . . . . 58 Identification of AR5 . . . . . . . . 63 Identification of AR4 . . . . . . . . . . 65 The Origin of Carbon Atoms in Unique Cyst Lipids . 70 Relative Abundance of These Compounds in Cyst Lipids. . . . . . . . . . . . . 75 Biosynthesis of Unique Lipids During Encystment. . 77 Distribution of Unique Lipids in the Cyst. . . . 8O Fates of Unique Lipids During Germination. . 81 Possible Biosynthetic Pathway of the Unique Lipids. 86 Possible Physiological Roles of These Compounds. . 90 Occurrence of Similar Compounds in Nature. . . . 91 REFERENCES. . . . . . . . . . . . . . 94 APPENDIX . . . . . . . . 103 Identification of C-AMP in Azotobacter Vinelandii . 103 Introduction Materials and Methods Strain and Culture. C-AMP Assay Preparation of Samples for Identification of C- AMP 3',5'-Cyclic Nucleotide Treatment Protein Assay Results . . Discussion . References . . . . . . . 103 . . . . . . 104 . . . . . . 104 . . . 105 C-AMP Assay . . . . 105 . . . . . 106 Phosphodiesterase . . . . . . 106 . . . . . . 107 . . . . . . 107 . . . . . . 110 . . . . . . 115 LIST OF TABLES Table Page 1. Structures and PrOperties of Unique Lipids From Encysting A. vinelandii . . . . . . 71 2. Composition of Lipids From 5 Day Old Cysts . . 76 3. The Relative Composition of Unique Lipids in Each Fraction of Cysts . . . . . . . . 82 4. The Distribution of Unique Lipids in Cysts . . 83 A-l. Cyclic-AMP Levels in A. vinelandii Cells and Culture Fluids. . . . . . . . . . . 109 A-2. Phosphodiesterase-sensitive C-AMP Levels in Cells and Media . . . . . . . . . . lll vi LIST OF FIGURES Figure 1. 10. 11. 12. 13. LITERATURE REVIEW Cyclic scheme for the metabolism of poly-B- hydroxybutyrate and its control in A. beijerinckii . . . . . . . . . . . RESULTS AND DISCUSSION Chromatogram of lipid extracted from A. vinelandii. . . . . . . . . . . . Separation of A. vinelandii cyst lipid on a florisil column . . . . . . . . . . Mass spectrum of APl (mixture of two homologues) . . . . . . . . . . . High pressure liquid chromatography separation profile of two APl homologues . . . . . UV absorption spectrum of APl . . . . . . Infrared absorption spectrum of APl . . . . NMR spectrum of AP 1 . . . . . . . . . UV absorption spectrum of AR3 . . . . . UV absorption spectrum of AR4 . . . . . . l4 Incorporation of C-BHB into the lipid fraction during encystment of A. vinelandii. The biosynthesis of unique lipids during the encystement of A. vinelandii. . . . . . Fates of unique lipids during the germination of A. vinelandii cysts. . . . . . . . Possible biosynthesis pathway of unique lipids in Azotobacter vinelandii. . . . . . . vii Page 21 36 39 42 45 48 51 53 61 69 73 79 85 89 INTRODUCTION Azotobacter vinelandii are gram negative, free- living, nitogen-fixing bacteria which inhabit the soil (53). EXponentially growing cells appear as large ovoid rods under the microscope. At the end of exponential growth, a small percentage of cells undergo differentiation which results in the formation of cysts (lOl). Cysts are metabolically dormant and are more resistant to deleterious physical con— ditions than vegetative cells (83). Under suitable environ- mental conditions, they can germinate readily and convert back into vegetative cells (54) and thus complete the life cycle of A. vinelandii. Encystment can be induced by transfering exponenti- ally growing vegetative cells from glucose to 0.2% B-hydro- xybutyrate (BHB) (51). The entire encystment process is completed in 4-6 days and results in the conversion of more than 90% of the vegetative cells into cysts (51). BHB serves both as carbon source and energy source during the encyst- ment (35). Like other kinds of cell differentiation processes, encystment involves a programmed activation of some parti- cular genes and turn-off of some other genes during its course. BHB is a specific inducer of encystment whereas other compounds such as amino acids, carbohydrates and short chain fatty acids such as acetate or prOpionate are without effect. n-butanol and crotonic acid do induce encystment but probably exert their effect by being first converted into BHB or its metabolic product by bacteria (51). An understanding of how BHB functions to trigger the whole encystment process is critical to our understanding of cellular differentiation in this organism. BHB is a normal metabolite in Azotobacter and these cells accumulate poly-B-hydroxybutyrate (PHB) as an energy reserve during vegetative growth (87). PHB is synthesized and degraded via acetyl CoA (21), a key intermediate in l4c-BHB labeled at lipid metabolism and TCA cycle. Using the C-3 position to study the BHB metabolism during encyst- ment, two-thirds of the total radioactivity incorporated into the cysts can be recovered in lipid fraction after 5 days (75). Thus, the importance of lipid metabolism during encystment cannot be overemphasized. Lin and Sadoff (49) found cysts contain twice as much lipid as vegetative cells on the dry weight basis. A wide variety of fatty acids were found in the cysts whereas only four major fatty acids are present in vegetative cells (39). Cis-9,lO-methy1ene hexadecanoic and lactobacillic acids are two unique fatty acids synthesized during encystment (77). These compounds can be detected as early as 6 hours after the induction of encystment and are synthesized at the expense of palmitoleic and cis-vaccenic acids by the addition of a methylene group across a carbon—carbon double bond (104). They eventually account for 25% of total fatty acids in mature cysts. In a recent investigation of cyst lipids, Reusch and Sadoff (75) fractionated the cyst lipids into neutral, "g1yco-" and phospholipids on a silicic acid column (45). They found that 82% of lipids in mature cysts were in the acetone or "glycolipid" fraction and most of lipids in this fraction were derived from BHB used to induce encystment. Further analysis of this fraction by thin layer chromato- graphy revealed two major molecular species where each species consisted of two homologues. The four compounds were identified as 5—n-heneicosy1resorcinol, 5-n—tricosylresor- cinol and their galactoside derivatives. In my research, I have isolated several new compounds in the acetone fraction (glycolipid) and established their structure by ultraviolet (UV), infrared (IR), nuclear magnetic resonance (NMR) and mass spectrosc0py. The time course for the appearance of these lipids was also established in encysting cells. The origins and possible roles of these compounds is the subject of the Discussion section of this disserta- tion. The dissertation is organized into the following sections: an Abstract of the dissertation; Introduction; a Literature Review which contains information about the physiological prOperties of the organism, studies of encystment and metabolic patterns in A. vinelandii with special reference to B-hydroxybutyrate; Materials and Methods utilized in the research; and Results and Discussion of the research. An Appendix is included which contains a description of unpublished research results. LITERATURE REVIEW A. Physiology of Azotobacter vinelandii A. vinelandii are gram negative, free-living, nitrOgen—fixing bacteria originally isolated by Lipman in 1903 from a soil sample from Vineland, New Jersey (53). Exponentially growing cells of A. vinelandii appear as large ovoid rods, about 2 x 4 um in size, frequently in pairs under microsc0pe. The organism produces large amounts of extracellular polysaccharide to give its colony a mucoid appearance on agar surfaces. Fresh cultures of A. vinelandii are motile due to the presence of flagella (37), but the motility is lost gradually as the culture approaches the end of exponential growth. Cells accumulate poly-B-hydroxy- butyric acid apparently as an energy reserve. The morphology of A. vinelandii varies with growth conditions and as many as five morphological forms have been described (38). Azotobacter is widely distributed in the soil with pH between 6-10 (38). In the laboratory, it can be easily grown in synthetic medium in which potassium, magnesium, calcium, iron, sulfate and phosphate are essential mineral nutrients (96). When fixed nitrOgen is not available in the medium and cells have to fix atmosphere nitrogen, then molybdenum is also essential for its growth because this element is one of the components of nitrogenase (14). Alcohols such as ethanol, propanol, butanol and glycerol, organic acids such as acetate, propionate, butyrate and succinate and carbohydrate such as glucose, sucrose, fructose, galactose and maltose all can be used as both carbon and energy source by this organism (38). Azotobacter is a strict aerobe with a high respiration rate (52). Carbohydrates such as glucose are metabolized through the Entner-Doudoroff pathway into 2 pyruvate with the gen- eration of 1 adenosine triphosphate (ATP) and 2 NADH (62, 85). Pyruvate is further oxidized into CO2 and H20 in the tricarboxylic acid (TCA) cycle (85) with the generation of metabolic intermediates and ATP. However, an excess of 02 inhibits the growth of Azotobacter by inhibiting nitrogen fixation which requires a reducing condition (103). Although Azotobacter grows well in simple synthetic medium, it does not grow well in nutrient broth (38) because its utilization of organic nitrogen is very poor (30). Some amino acids (e.g., glycine and leucine) inhibit the growth of Azotobacter at a concentration of 0.1% (24). On the other hand, ammonium ion, nitrate and urea can all serve as nitrogen sources for Azotobacter. When fixed nitrogen is not available, Azotobacter can fix atmospheric nitrogen under aerobic growth conditions. Nitrate and urea are reduced or degraded into ammonium ion and, when present in high concentration, are assimilated into the cell 411! ! 1|! - ! w -J components via the glutamate dehydrogenase pathway (39, 96). When the ammonium ion concentration is low and cells have to fix nitrogen from the air, ammonia is assimilated into the cell via the enzymes glutamine synthetase and glutamate synthetase (25, 40). During nitrogen fixation, N2 is reduced to 2 NH3 by nitrogenase using reduced nicotinamide adenine dinucleotide phosphate (NADPH) as the reductant and ATP as the energy source (88). Nitrogenase is an enzyme complex consisting of 2 kinds of metal-containing proteins. In A. vinelandii, complex I (Mo-Fe protein) is an enzyme containing molybdenum and iron with a molecular weight of 245,000 daltson (89). Component II (Fe-protein) is a component I reductase which contains iron and has a molecular weight of 60,000 daltons (90). Both proteins are needed for the enzymatic activity and a ratio of 1:1 is optimal for the highest activity. A third protein, iron-sulphur protein II, with a molecular weight of 24,000 dalton is also associated with nitrogenase in Azotobacter and may have a regulatory role (79) or con- tribute to the stability of the Azotobacter nitrogenase in the presence of oxygen (32). A. vinelandii DNA has a G + C content of 63 to 66% (56). Each cell in early exponential growth phase contains about 15x10.l4 g of DNA which is equivalent to 10 times that of E. coli (78). Possible DNA redundancy in the Azotobacter gene might explain the difficulty encountered in isolating mutants from this organism. Recently, Page and Sadoff (67, 68) have developed a method for the transformation of Azotobacter. A. vinelandii becomes competent for a period of time approaching the end of exponential growth. Transformation competence decreases rapidly as cells enter stationary phase. Optimal transfor- mation occurs at 30°C at a pH of 7.0-7.1. Magnesium and phosphate are essential for the transformation. High con- centrations of calcium in the medium enhance the production of capsule and hence are inhibitory to the transformation (70). Ammonium ion in the iron-free medium increases the transformation frequency. Only NH+, N03 or urea can serve as nitrogen sources for growth and, at the same time, allow the induction of competence. Other nitrogen sources which do not inhibit glutamine synthetase inhibit the induction of competence (70). Glucose, sucrose, manitol and glycerol all serve as good carbon sources in the induction of com- petence but not acetate or B—hydroxybutyrate. Competence also can be increased lOOO-fold in early exponentially growing cells by the addition of 1mM c-AMP (68). A. vinelandii can also be transformed with Rhizobium spp. DNA (10, 66). B. Cysts of Azotobacter vinelandii 1. Induction of encystment and germination Among the four genera of Azotobacteriaceae, only Azotobacter sp. are able to form cysts (3). Cysts are metabolically dormant cells that are more resistant than vegetative cells to radiation, sonic treatment, heat and dessication (83). Thus, cysts can survive under unfavorable conditions where the survival of vegetative cells is impossible. Encystment occurs at the end of exponential growth when one or more of the growth factors become limiting. At this time, vegetative cells start to accumulate poly-B-hydroxybutyrate (PHB) and some cells eventually differentiate into cysts (98, 101). The extent of encyst— ment in a culture is related to the intracellular level of PHB and normally less than 0.1% of the total population become cysts (86) in cultures grown on a carbohydrate such as glucose. Encystement also can be induced by growing cells in n—butanol (100) or by shifting the exponentially growing vegetative cells from glucose to 0.2% BHB as carbon source (51). This latter two-step encystment procedure developed by Lin and Sadoff (51) results in a more than 90% yield of cysts after 5 days of incubation. If the vegetative cells are not washed free of glucose before the addition of BHB, abortive encystment occurs resulting in a low yield of cysts and release of a viscous plymer into the medium (51). The BHB used to induce encystment is absorbed and metabolized by the cells (35). Tracer studies with l4 C-3- BHB showed that about 10% of the BHB in the medium was released as CO2 and about 25% was incorporated into cyst materials. Radioactive BHB was incorporated into the RNA, protein and lipid fractions (35). The radioactivity in the RNA fraction increased steadily, reached a peak at 9 hours 10 after the induction and then decreased slowly. The radio- activity in the protein fraction increased steadily up to 9 hours and continued thereafter at a lower rate. The radio- activity of the lipid fraction increased slowly up to 9 hours, then sharply up to 24 hours and then leveled off. No radioactive BHB was incorporated into the DNA fraction. When metabolizable carbon sources become available in the cyst environment and given suitable pH, temperature, moisture, minerals and carbon source, cysts germinate and grow into vegetative cells. Germination can be induced by suspending cysts in Burk's N-free medium plus glucose, sucrose or acetate as the carbon source (48). The whole process takes about 12 hours and can be roughly divided into 3 stages of 4 hours each: Germination, outgrowth and transi- tion (48). 2. Morphological changes during encystment and germination The morphological changes during the encystment and germination of Azotobacter cysts have been studied micro- sc0pically (78, 98) and electron microscopically (36, 48, 84, 100). When viewed by phase contrast microscopy, vegeta- tive cells are large rods, actively motile, and in many stages of division. The cytoplasm is quite homogeneous and not refractile, although old cells do contain refractile granules. When examined under the electron microscope, the cells ribosomes are seen as dark granules dispersed through- out the cytoplasm. DNA can be seen in the electron 11 transparent areas of cytoplasm and membranous vesicles can be seen within the cytoplasm and along the periphery of cell. Surrounding the cells is a double layered cytoplasmic membrane and a triple layered cell wall. Extrusions pro- truding from the cell surface also can be observed (36). Microscopically, mature cysts appear as round, refractile and non-motile cells. Thin sections, viewed in the electron microscope, show cysts to be made up of three major components; (a) an electron dense central body con- taining some PHB grandules surrounded by (b) a less dense intine and (c) a multilayered exine (36, 50, 73). Upon the induction of encystment, cells undergo a last round of cell division (78) which takes about 4 hours. During this period, cells start to "round up" and accumulate PHB. As encystment progresses, each cell is gradually surrounded by a layer of capsular material and a number of protrusions or "blebs" are excreted from the cell surface. These aggregate and flatten into sheets which ultimately become the exine. After 36 hours, the exine, intine and central body can be easily seen but approximately 80 more hours are required for the completion of intine and exine structure. As the cysts mature, the central body becomes rather compact (36). Upon the induction of germination, the cysts lose their refractility gradually over a 4 hour period (54). In the first 2 hours after induction, intines become more electron transparent, a fibrillar structure around central 12 body can be observed, but little change occurs in exine and central body. As the germination progresses, the intine becomes less electron dense and the vesicular and fibrillar structure of intine can be easily observed. Four hours after germination commences, the PHB in the central body starts to be metabolized. The central body starts to elongate at 6 hours signalling the beginning of outgrowth. The central body continues to grow and starts to divide, and at 8 hours it emerges from the disintegrating exine and becomes a young vegetative cell. According to Lin et a1. (48) the young vegetative cells which emerge are still more resistant to sonication than exponentially growing vegetative cells and it takes another 4 hours of transitional period for the outgrown cells to become normal vegetative cells (48). 3. Metabolic changes during encystment and germination When glucose-grown vegetative cells of A. vinelandii are induced to encyst with BHB, they must shift from car- bohydrate to lipid metabolism. A readjustment of metabolic rate and induction of new metabolic pathways is necessary to c0pe with this kind of change. Some enzymes of the Entner-Doudoroff pathway are not required for the metabolism of BHB, but at the same time, glyoxylate and gluconeogenesis pathway enzymes must be induced for the continuous operation of TCA cycle and to provide carbohydrate for the synthesis of certain cyst components. 13 A number of the metabolic events which occur during encystment of A. vinelandii have been studied (35, 78). The initiation of DNA synthesis stops immediately and no syn— thesis can be detected at 4 hours after induction of encysts (35). The DNA content per cell drops during the course of 14 to 3.4 x 10'14 g (78). The encystment from 15 x 10— nitrogen fixing ability of cells also drops sharply and losses of 30% of the original activity occur within 15 minutes after the induction of encystment with complete loss after 1 hour (35). RNA synthesis continues at a reduced rate until 12 hours into encystment. Protein syn— thesis during encystment occurs at about one—third the rate at which it occurs in vegetative cells (35) and continues throughout the entire encystment period. Because of the early cessation of nitrogen fixation during the encystment of A. yigglandii, macromolecular synthesis must depend on the turnover of existing protein and RNA. The Specific activity of glucose-6—phosphate dehydrogenase, which is a key enzyme in glucose degradation, decreases steadily after the induction of encystment and is almost completely gone after 9 hours. At the same time. BHB dehydrogenase, the first enzyme in BHB metabolism, is being synthesized. BHB dehydrogenase reaches a peak activity at 6 hours after induction of encystment, decreases slightly, then increases to another peak at 21 hours. Isocitrate dehydrogenase activity follow the same pattern as BHB dehydrogenase. The two glyoxylate shunt enzymes, isocitrate lyase and malate l4 synthetase, are derepressed immediately after induction of encystment, reach a peak at 3 hours, decrease to a low value by 9 hours, then increase again at 15 hours to reach a high level during the late encystment. Aldolase and fructose diphosphatase (FDPase), two key enzymes in the gluconeogenesis pathway, also have a temporal relationship during encystment. Aldolase reaches a maximum at 6 hours, decreases to the original value at 18 hours and then in- creases to another peak at 24 hours. FDPase activity increases and reaches a maximum at 9 hours. After that, it decreases to a low point at 18 hours and then increases to another peak of activity at 27 hours. Cysts in Burk's buffer at 30°C start to germinate upon the addition of glucose. The germination and outgrowth phases each have their characteristic rates of 02 consumption, CO evolution and RNA and protein synthesis (54). The 0 2 uptake and CO 2 2 evolution begin very soon after the addition of glucose and constitute the first signs of germination. Both the O uptake and CO2 evolution start at low rates 2 initially and increase to constant values at about 4 hours (54). RNA synthesis occurs within 20 minutes after the initiation of germination and proceeds at a low rate up to 3.5 hours. It then increases sevenfold by 5 hours post- induction and remains at this rate throughout outgrowth.. Protein synthesis occurs very soon after the initiation of germination and proceeds at a constant rate up to 5 hours. It then increases 3 fold and remains at that rate until the 15 completion of outgrowth (28). Both DNA synthesis and N2 fixation start at about 5 hours after induction signalling the beginning of the outgrowth (54). 4. Compositional differences between vegetative cells and cysts of Azotobacter vinelandii Vegetative cells of A. vinelandii contain 28% car- bohydrate, 52% protein, 9.2% lipid and 7.1% ash by dry weight. By comparison, cysts contain 45% carbohydrate, 26% protein, 16% lipid and 8.8% ash (49). Cysts have been further fractionated into exine, intine and central body (50). Exines were prepared by rupturing the cysts in Tris (hydroxymethyl)aminomethane (Tris) buffer, pH 7.8 plus 3 mM of ethylenediaminetetraacetic acid and isolated by differ- ential and is0pycnic centrifugation. Intines were soluble in the buffer and were prepared by rupture of cysts in the Tris buffer plus EDTA, collecting the supernatant fluid after centrifugation and then dialyzing it against distilled water to remove small molecules. An analysis of the purified exines and intines indicates that exine contains 44% of carbohydrate, 9.1% protein, 36% lipid and 4.1% ash whereas intine contains 32% carbohydrate, 28% protein, 28% lipid and 3.2% ash (49). The slime of Azotobacter vegetative cells contains mostly partially acetylated polyuronic acids (D-mannuronic and L-guluronic acids) plus small amounts of glucose and rhamnose (18, 46). The two uronic acids are distributed along the polymer chain in the typical block wise fashion l6 characteristic of alginates from brown algae (46). The ratio of the uronic acids in the slime depends on the calcium concentration in the medium. High calcium con— centrations result in the production of slime rich in L- guluronic acid. An enzyme "5-epimerase" capable of con— verting D-mannuronic into L-guluronic has been demonstrated in medium in which A. vinelandii has grown (33). The activity of the epimerase depends on the calcium concentra- tion of the medium. Uronic acid accounts for 40% and 72% of the car- bohydrate in the exine and intine, respectively, of Azotobacter cysts (69). D-Mannuronic and L-guluronic acids are the only two uronic acids present. The exine is rich in polyguluronic acid and the intine is rich in polyman- nuronic acid (69). This may be an effect of the high con- centration of calcium at the beginning of encystment when the exine is formed. Calcium enhances the 5-epimerase activity and converts the D-mannuronic into L-guluronic acid. The resulting polymers bind calcium ions selectively (82) and help form the sheet like structures of exine. The completed exine contains a relatively large amount of calcium and forms a physical barrier which prevents the influx of calcium ions. Exponentially growing vegetative cells of A. yggg- landii contain about 10% free lipids and an additional 2% bound lipids (39, 49). About 80% of the free lipids are phospholipids of which 64% is phosphatidylethanolamine, 27% 17 is phosphatidylglycerol, plus minor amounts of phosphatidyl- N-methyl-ethanolamine, phosphatidylcholine and cardiolipin (74). In stationary phase cells, cardiolipin increases to 23% of the total phospholipid while phosphatidylglycerol content decreases to 13% of the total phospholipid (74). Major fatty acids in the vegetative cells are myristric acid (7%), palmitic acid (35%), palmitoleic acid (41%) and octadecenoic acid (17%) (39). Cysts contain twice as much lipid as vegetative cells. A wide variety of fatty acids have been identified in the cysts (49, 77). Among these, 2 cyclOprOpane fatty acids, cis-9,lO-methylenehexadecanoic and lactobacillic acid, are unique in Azotobacter cysts (77). They are synthesized at the expense of palmitoleic acid and octadecenoic acid after induction of encystment (77) and appear as early as 6 hours after induction of encystment. Eventually each of the cyclopropane fatty acids reaches 10% of the total fatty acids in the mature cysts. C. Poly—B-hydroxybutyrate (PHB) Metabolism in Azotobacter BHB is a normal metabolite in cells which can be synthesized by the condensation of 2 acetyl CoA into ace- toacetate followed by reduction to BHB. BHB is also a specific inducer of encystment of A. vinelandii. The pro- cedure used to induce encystment probably results in a sudden rise of the intracellular BHB concentration. How such a concentration increase of a normal metabolite can trigger the sequence of events resulting in the formation 18 of dormant cysts is an important question. A complete understanding of BHB metabolism in the cell help us to answer this question. A wide variety of bacteria accumulate BHB as a polymer (PHB) intracellularly as an energy reserve when their carbon supply is plentiful (20). PHB is a straight chain homopolymer of D(-)-3-hydroxybutyrate. Natural PHB has a molecular weight between 60,000-250,000 daltons (55) and can be observed by light microscopy as refractile granules in bacteria or as dark bodies when stained by Sudan black (60). Under the electron microscope, a mem- branous structure can be observed surrounding the PHB granules (56). This coat can be removed by treating the native PHB with aqueous acetone (26) and has been found to consist of lipids and protein, the protein often possessing PHB polymerization activity (31). PHB is an excellent storage form of carbon and energy sources because it is in a highly reduced state, virtually insoluble in water, and thus does not exert any osmotic effect. Macrae and Wilkinson (57) observed in Baccilus megaterium that, as the ratio of carbon and energy source to the nitroqen source of the medium was increased, the amount of PHB accumulated per cell also increased. Pyruvate, 3-hydroxybutyrate and glucose increase the yield of PHB in the same organism. When acetate was added together with any of these substrates the extent of PHB synthesis was greatly increased (58). In continuous culture experiments 19 with B. megaterium (94), it was observed that PHB accumulated when any of the nutrients in the culture become a limiting factor. Senior et a1. (80) studied the accumulation of PHB in batch cultures of A. beijerinckii grown in nitrogen-free medium. They found that cells started to accumulate PHB at the end of their exponential growth and continued to do so during the stationary phase. The amount of PHB accumulated in the cell was proportional to the amount of glucose in the medium. Continuous culture studies revealed that neither nitrogen nor carbon source limitation in the culture led to the accumulation of large amounts of PHB but that oxygen limitation was the major factor in its accumulation in Azotobacter. PHB accumulation ceased when the 02 limita- tion in the culture was relaxed. The possibility of using PHB as reducing power sink to regulate the redox potential in Azotobacter had been suggested (21). The biosynthesis of PHB in A. beijerinckii is dependent on three enzymes (21). B-Ketothiolase catalyzes the condensation of 2 acetyl CoA into acetoacetyl CoA with the release of CoASH. Acetoacetyl CoA is then reduced into D(-)—3-hydroxybutyryl CoA by acetoacetyl CoA reductase. NADPH is the preferred coenzyme in this reaction. D(—)-3- hydroxybutyryl CoA is then incorporated into PHB by PHB synthetase which is usually associated with the PHB granules. The degradation of PHB in this organism involves another pathway. PHB is degraded into BHB by a PHB depoly- merase after which BHB is then oxidized into acetoacetate .. “1.. 1.1 _.—_4. --..__ . 20 by BHB dehydrogenase using NAD as the coenzyme. The enzyme "thiOphorase" converts acetoacetate into acetoacetyl CoA using succinyl CoA as CoA donnor. Acetoacetyl CoA is further cleaved into 2 acetyl CoA by the B-ketothiolase which catalyzes a reversible reaction. The regulation of PHB metabolism in A. beijerinckii is shown in the following scheme (Fig. 1). When 02 becomes limiting in an Azotobacter culture, NADPH and NADH increase and inhibit citrate synthetase and isocitrate dehydrogenase activity. The inhibition of these two enzymes results in the accumulation of acetyl CoA and reduces the CoASH concentration in cells. This condition favors the formation of acetoacetyl CoA in the reaction catalyzed by B-ketothiolase and results in the synthesis of PHB. High concentrations of CoA inhibit the formation of acetoacetyl CoA. Acetoacetyl CoA inhibits the acetoacetyl CoA reductase (76). Only when the NADPH concentration is high and CoASH is low, (e.g.: during 02 limitation) would acetoacetyl CoA reduction occur and PHB be synthesized. Simultaneously, high NADH inhibits the depolymerization of PHB by inhibiting the BHB dehydrogenase and thus preventing unrestricted cycling of PHB. When 02 limitation is relaxed, the decreased concentration of NADPH and NADH will restrict the biosynthesis of PHB but not until the carbon source is also limited e.g., decreased pyruvate concentration. This will result in the decrease of acetyl CoA and increase of CoASH. B-Ketothiolase will then catalyze the degradation 21 Pyruvate Acetoacetyl CoA é CoASH ‘%—’———_‘T 1 I 1 ~1 )1 : Acetyl CoA _ g 1 1 L ---------- > T <- ----------------- ' Acetoacetyl CoA Succinate Acetoacetyl CoA l 1 Acetanetate NADP NADPH , é____ n <—-- I Succ1nyl CoA Acetoacetate D(-)-3-Hydroxylbutyryl CoA NAD NADH V// Poly-B—Hydroxybutyrate _J NADH Pyruvate > 2-Oxoglutarate D(-)—3-Hydroxybutyrate Fig. l.--Cyclic scheme for the metabolism of poly—B-hydroxybutyrate and its control in A. beijerinckii. (After Dawes E.A. and P.J. Seniopr) (21) ------ indicates inhibitor 22 of acetoacetyl CoA into acetyl CoA with subsequent PHB degradation. Slepecky and Law (81) concluded that while PHB accumulation is not a prerequisite for spore formation in A. megaterium, the polymer serves as a source of carbon and and energy during sporulation. Kominek and Halvorson (42) came to the similar conclusion in studies of sporulation in A. cereus. Stevenson and Socolofsky (86) studied cyst formation and PHB accumulation in Azotobacter and found that cells accumulated large amounts of PHB before the on-set of encystment. The accumulated PHB was subsequently degraded to provide carbon and energy for encystment. The percentage of encystment in a culture was directly proportional to the amount of PHB accumulated. Good carbon sources which stimulate PHB accumulation, e.g., butanol or glucose, also result in a higher percentage of encystment. MATERIALS AND METHODS Growth of Bacteria A. vinelandii ATCC 12837 was used throughout this study. Cells were cultivated in Burk's nitrOgen-free medium (97) with either 1% glucose or 1% sodium acetate as the carbon source. The culture was incubated at 30°C with shaking. Bacterial growth was monitored by measuring the absorbance at 620 nm (OD ) with a spectrophotometer. 620 Induction of Encystment Encystment was induced according to the method of Lin and Sadoff (51). Exponentially growing cells (OD = 0.7) were harvested by centrifugation, washed once with Burk's buffer (medium without carbon source) and suspended in the same volume of Burk's buffer. B-Hydroxybutyrate (BHB) was added to a final concentration of 0.2% and the culture was then incubated at 30°C with shaking. Mature cysts were formed by the fifth day after the shift to BHB. Cysts were harvested by centrifugation, washed once with Burk's buffer and stored at -20°C. Germination of Cysts The fate of unique cyst lipids was studied during germination. Sufficient cysts were suspended in 100 ml 23 24 sterile Burk's buffer to produce a final OD = 0.7 and incubated at 30°C. Germination was started by adding 2 ml of a sterile 50% solution of glucose to the cyst suspension and incubating at 30°C with shaking. Complete outgrowth of vegetative cells occurred by 8 hours after germination. Lipid Extraction Total lipids in the vegetative cells or cysts were extracted according to the method of Bligh and Dyer (11). Cells were harvested by centrifugation, washed once and then suspended in a small volume of water. To each 0.8 ml of cell suspension, 1 ml of chloroform and 2 ml of methanol were added. The mixture was shaken vigorously (vortexed) for 1 minute to form one phase and kept at 4°C overnight. The cell residue was filtered out and the solution was separated into 2 phases by the addition of 1 ml of chloroform and 1 ml of H20 to every 3.8 ml of solution. The biphasic system was shaken in a vortex mixer and then centrifuged briefly to again separate it into two layers. The chloroform layer was washed once with 0.7% NaCl solution, rinsed twice with distilled water and taken to dryness under a hood. Removal of PHB Occasionally, PHB is co-extracted with other lipids in the chloroform layer and interferes with the chromato- graphic separation of lipids. The removal of PHB was euflxieved by drying the chloroform layer slowly in a hOOd- 25 PHB forms a membrane-like sheet under these conditions (95) which does not go into solution when lipids other than PHB are dissolved in a chloroform:methanol (2:1) mixture. Methylation of Samples Lipids usually become more volatile and easier to separate in gas chromatograph columns after methylation. Methylation also can help to determine the number of "active exchangable protons" in a molecule. Methylation was carried out with diazomethane. Diazomethane was prepared by distillation from an ethereal solution resulting when its precusor N—N-dinitroso-N,N—dimethylterephthalamide was mixed with 10% KOH. Enough diazomethane was added to the compounds undergoing methylation so that a yellow color persisted for at least 20 hours. Completeness of methyla- tion was checked by thin layer chromatography (TLC). Chromatographic Methods A. Column chromatography Primary separation of crude lipids extracted from cysts was carried out on florisil (Matheson Coleman and Bell, 60-100 mesh) in a 2.5 x 50 cm column. About 80 g of florisil was used for 2 g of crude lipid extract. The absorbant was suSpended in chloroform, degassed under vacuum, and then packed into a column. Samples were dis- solved in a small amount of hot chloroform and methanol mixture (2/l,v/v) and loaded onto the column. The column 26 was eluted in sequence with chloroform (200 ml), 20% acetone in chloroform (100 ml), 60% acetone in chloroform (50 ml), 100 ml acetone (50 ml), 25% methanol in chloroform (100 ml), 50% methanol in chloroform (150 ml) and methanol (100 ml). The flow rate was 0.5 ml/minute and 10 ml fractions of eluate were collected in test tubes. The solvent was evaporated under a stream of N2 gas and the lipid in each fraction was dissolved in chloroformzmethanol (2:1, v/v) and then analyzed by thin layer chromatography (TLC). Fractions with similar composition were pooled and further fractionated on a silicic acid column (silic AR cc—7 Mallinckrodt Analytic Reagent). Silicic acid was activated at 120°C for 16 hours, cooled in a dessicator anui then made into a slurry with chloroform. This was degassed under vacuum and then packed into a 1.3 x 25 cm column (lg/10 mg of sample). Samples were loaded onto the column in the same way as in the florisil columns. Elution of fractions was stepwise beginning with chloroform, then increasing concentrations of acetone in chloroform followed by increasing concentrations of methanol in chloroform. The volume of each solvent employed was approximately the volume of packed column bed at a flow rate of 0.3 ml/minute. Approximately 10 ml volumes of eluate were collected in test tubes and the lipids in each tube were analyzed by TLC. 27 B. Thin layer chromatography Precoated 2000 u silica gel G glass plates (Analtech Labs) were used for preparative separation of lipids. From 2 to 5 mg of sample was applied as a strip on each plate. Precoated silica gel 60 aluminum sheets (E.M. Labs.) were used for analytical separation. Development was in either solvent system - 1) Carbon tetrachloride : Acetone (90:10, v/v) or 2) Chloroform : acetone : Methanol (12:4:1, v/v/v/) Total lipids were detected by exposure of plates to iodine vapor. Diazotized sulphanilamide reagent (43) was used to locate compounds that were of interest in tfliis study. The vanillin test (43) was used to identify phenolic compounds. Preparative plates were dried under nitrogen. after being developed. At this point, the plate was partially covered with a glass sheet leaving about a 2 cm strip at the edge to be Sprayed with the appropriate reagent. After spraying, the remaining areas corresponding to the Rf of compounds of interest were scraped into indi— vidual flasks and extracted with chloroform : methanol (2:1, v/v). The solvent was evaporated under nitrogen and the residue analyzed by analytical TLC. The procedure was repeated until there was only one compound present as detected by analytical TLC. Spray reagent : Diazotized sulfanilamide reagent : 3 g of sulfanilamide were dissolved in a solution of 200 ml H20, 6 ml of 36% HCl and 14 ml of n-butanol. 0.3 g Of 28 sodium nitrite was added to 20 ml of this cold solution immediately before spraying chromatograms. Phenolic com- pounds appeared as red—orange areas immediately after spraying. Pyronic compounds reacted Slower and turned yellow to orange color. The background became yellow several hours after spraying the chromatograms. Spraying with 10% of sodium carbonate prevented this but may have decreased the color intensity of some spots (pyrones). Vanillin test : Samples on plates which had been sprayed with 1% vanillin in glacial acetic acid were dried. Plates were then sprayed with H2804 in ethanol (1 part of concentrated H2804 mixed with 4 parts of absolute alcohol) and heated at 100°C for 5 minutes. Phenolic compounds turned pink. C. Gas chromatography Gas chromatography (GC) was conducted in a Varian Aerograph 1440 gas chromatograph equipped with a flame ionization detecter. Helium was used as the carrier gas at a flow rate of 30 cc/min. A 183 cm x 0.32 cm or 61 cm x 0.32 cm stainless steel column was used. Packing was 1.5% SE—30 on chrom G A/W, 100/120 mesh. Methylated derivatives of unknown lipids were chromatographed either isothermally or with temperature programming. D. High pressure liquid chromatography (HPLC) Because samples purified from preparative TLC were mixtures of 2 homologues of different chain length, some 29 samples were further separated by HPLC. These separations were achieved on a preparative reverse phase column (LiChrosorb RP-8, 5 u particle size, packed in a 10 (ID) x 250 mm stainless steel column (Altex Associates Inc.). The sample injector used was a Rheodyne model 7120 fitted with a 200 pl 100p. Pressure was applied with a Milton Roy model 396 minipump. Samples were detected with an Altex analytical model 153 UV-Vis detector at 280 nm connected to a Linear recorder model 255. The mobile phase solvent was either 10% H20 in methanol or 20% H20 in methanol depending on the compounds being separated. About 500 ug of sample in chloroform: methanol (2:1) were applied to the column. The mobile phase was pumped at 2 ml/min, eluates were detected and each peak concentration of compound was collected in a flask. Samples were extracted from the flasks with chloro- form. Spectrometric Methods A. Ultraviolet (UV) UV absorption spectra were taken with a Beckman DU spectrOphotometer model 2400 employing 95% ethanol as the solvent. B. Infrared (IR) IR spectra were taken as Nujol mulls or dilute solutions in chloroform or carbontetrachloride depending 30 on the solubility of the sample. A Perkin-Elmer 700 infra— red spectrophotometer was used. C. Mass spectra A Hewlett-Packard 5985 GC-MS system was used to take low resolution mass spectra (MS). A direct probe was used for non-volatile samples. Methylated samples were run in GC-MS combination. Electron impact was the mode of ionization with the ionization voltage at 70 eV. Electron multiplier voltage was at 2000 V. Methane was used in chemical ionization to obtain the molecular ions of unstable samples. Peak matching for precise ion masses were run in a Varian MAT CH-SDF mass spectrometer with a direct probe. The respective operating conditions were: accelerating voltage at 3 KV, ionization voltage at 70 eV and electron multiplier voltage at 2000 V. Perfluorokerosine was used as the reference. D. Nuclear magnetic resonance spectra (NMR) A WH-180 Bruker Fourier transform spectrometer was used to take NMR spectra; CDCl3 with small amount of CHCl3 was the Solvent. The CHCl3 peak at 7.27 ppm was used as a reference point. After the NMR spectrum was taken, a few drops of D20 were added to the solution and another spectrum was taken to determine the extent of DZO-exchangable protons present. 31 Melting Point Determination The melting points of some compounds were taken with a Thomas-Hoover capillary melting apparatus (A.H. Thomas Co., Philadelphia, PA). Radioactive IsotOpe Labeling Study l4C-glucose and 3-13C-BHB were Radioactive 2- purchased from New England Nuclear Company. In order to determine what percentage of carbon in the cyst lipids was derived from glucose, an analysis was made of cysts derived from cells grown in that labeled sugar. A. vinelandii cells were grown in 50 ml of Burk's buffer plus 1% glucose plus 3.8 uCi of 2—14 C-glucose to an OD = 0.63. Cells were harvested and washed twice and then suspended in same amount of Burk's buffer. Total radioactivity per ml of cells was determined by collecting cells on a filter and counting. Cells were then induced to encyst with unlabeled BHB, and after 5 days they were harvested and washed. The total radioactivity remaining and the radioactivity in each lipid fraction was analyzed by extracting the lipids, separating them on analytical TLC plates and counting the radioactivity in each lipid fraction. 10 ml of counting scintillant (ACS, Amersham/Searle) was shaken with lipid fraction in a vial which then was incubated in the cold overnight prior to counting in a Packard Tricarb 3320 scintillation Spectrometer. 32 The time course of synthesis of each compound in the cyst lipids was studied by growing cells in glucose, l4C-BHB to induce encystment, sampling at shifting to 3- various intervals up to 5 days during the encystment, and extracting and analyzing lipids as above. The fate of unique cyst lipids during germination 14C of cysts was also studied by inducing encystment in 3- BHB. The cysts were then germinated in unlabeled glucose, sampled for 12 hours after germination and extracted and analyzed as above. Because the diazotized sulfanilamide spray reagent turns compounds of interest into red-orange or yellow dyes, various degrees of quenching occur when radioactivity is monitored by scintillation. Quenching was corrected by an internal standard method. After determining the radio- activity of sprayed samples, 0.1 m1 of ACS solution con— taining about 1000 cpm of 14C-glucose was added to each vial. After mixing and allowing the Silica gel to settle, the vial was counted again and the counting efficiency of each vial was calculated as follows: CPM after addition of standard - CPM before addition of standard CPM of standard Counting Efficiency = Thus, the radioactivity in each vial could be obtained by: CPM before addition of standard / counting efficiency = Corrected CPM 33 Fractionation of Cysts In order to study the distribution of unique cyst l4C-BHB) were lipids, l4C-labeled cyst (induced with fractionated into exine, intine and central bodies (50) and the lipids in each fraction analyzed. RESULTS AND DISCUSSION Comparison of Lipids From Vegetative Cells and Cysts The Chromatogram of lipids extracted from young vegetative cells growing exponentially in glucose or acetate, and from 5 day old cysts is Shown in Fig. 1. Under the chromatographic conditions employed, all phOSpholipids stayed at the origin. After the TLC had developed, it was sprayed with diazotized sulphanilamide. Lipids from young vegetative cells grown in glucose or acetate do not react with the spray reagent but lipid extracts from mature cysts contain at least 9 components which react with the spray reagent. Two major components, ARl and ARZ' have already been identified as mixtures of 5-n-heneicosylresorcinol and 5—n-tricosylresorcinol (ARl) and their galactoside derivatives (ARZ) respectively (75). The other components reacting with diazosulphanilamide were designated AR , AR 3 AR4, APl' APZ’ AP3 and AP4 respectively in order of in— 5] creasing mobilities. The biosynthesis of these unique lipids is correlated with the appearance of cysts or with B-hydroxybutyrate metabolism in cultures of this organism. In the solvent system I, Rf values were AR = 0.2, AR = l 5 0.48 and AR3 = 0.75. Other components remained very close 34 'Fig. l. 35 Chromatogram of lipid extracted from A. vinelandii. Chromatogram of lipid extracted from (1) young vegetative cells of A. vinelandii growing in 1% glucose, (2) young vegetative cells growing in 1% acetate, and (3) 5 day old cysts. Lipid extracted from 20 mg of wet cells of each type was spotted on the Silica gel plate. The Chromatogram was then developed in solvent system 2 (chloroform: acetonezmethanol 12:4:1 v/v/v) for about 100 minutes at room temperature. After drying the plate was sprayed with dizotized sulphanilamide reagent. The 2 young vegetative cells do not have diazoreagent— reacting compounds in their lipid. But in cysts, there are at least 9 compounds in the lipid fraction reacting with diazoreagent. These were designated as ARl, AR2, AR3 AR4, AR5, APl, AP2, AP3 and AP4. This solvent system does not resolve AR3 and AR5 well. On this particular plate, AP3 and AP4 were not well separated. 36 Al“3+5 An] AR4 AP] AP3+4 An2 _ 37 to the origin. In solvent system II, the Rf values were ARl APl 0.66, AR2 = 0.07, AR3 = 0.78, AR4 = 0.56, AR5 = 0.74, 0.31 and AP2 = 0.165. Purification of ARq, AR5, AR4, APl and AP2 In order to determine their chemical structures, AR3, AR5, AR4, AP and AP were purified from the mixture 1 2 of lipids in the extract. AP3 and AP4, which have the same color reaction as AP2 with different Spray reagents on TLC plates, were not purified because they are minor components and probably have structures similar to APZ' Primary separation was carried out on florisil columns. The results are shown in Fig. 2 which is a TLC analysis of the unique lipid content of the various eluates. ARl' AR3, AR4 and AR5 were eluted by chloroform and a chloroform acetone mixture. One hundred percent acetone starts to elute ARZ’ APl’ AP4 but elution was completed with a 50% mixture of methanol in chloroform. After this treatment, most of the lipids in column had been eluted. Fractions containing ARl' AR3, AR4 and AR5 were pooled and rechromatographed on a silicic acid column (cc—7). AR3 and AR5 were eluted tOgether by 5% acetone in chloroform, ARl followed by elution with 5%-10% acetone in chloroform and AR4 was usually eluted by 20% acetone in chloroform. Tubes from florisil columns containing APl, APZI AP3, AP4 plus AR2 were pooled and rechromatographed on silic acid column (cc-7). A 5-10% acetone in chloroform .pso Nm¢ tam vm< .mm< .mm< .Hmd madam QHSDxHE Hocmnemfi EHOMOSOHQO wHH£3 vm< .Hm¢ .mm< .mm< mpsao musuxfla EMOMOHOHQU mcoumom paw EH0moHoH£U .AN Empmmm ucw>aomv Ode QDHB pmuwawcm cam AHHNV musuxfle HosmnuoEuEHOMOMOHno ca ©o>H0mmewu was goapomum zoom ca mDSHOm mnu .pwpmsomm>m qu>HOm may Houmfl .ms0fluowum HE OH usonm CH topomaaoo was ohmsam .msousm gees topmoapnfl mm wuHMmHom msflmwwnocfl mo mucw>aom gees pmpus swap pow sEsHoo on» so topmoa was pflmfla one .EEDHOO HHmHHon m so Uflmfla #mwo HflpsmHmcfl> .m mo coflumummmm11.m .on Showouofiso Showomflno Hocmfi we Hocmmwwe occumom occumom EpomouoHso NOOH New wa NOOH Noonm Nooa « e e . _ e ? m< q+mm< we... 5... 39 a? my? 1mm... 40 mixture was used to elute APl but this solvent also tended to remove APZ’ Thus the separation of these 2 compounds was not very sharp on this kind of column. AP3 and AP4 were eluted after AP2. AR2 was eluted by 50% acetone in chloroform and other more polar solvents. Tubes containing lipids rich in AR3, AR5, AR4, APl and AP2 were selected and the final purification of each component was carried out on preparative TLC. Solvent system I was used to purify AR3 and AR5 by TLC. Without pre-equilibration, AR3, AR5 and ARl have Rf values of 0.75, 0.48 and 0.2 respectively in this Solvent which permitted a clear separation of these 3 compounds. ARl usually was the major contaminant in the sample of AR3 plus AR5. TLC with solvent system II and preequilibration was used to separate APl’ AP and AR . 2 4 These compounds have Rf values of 0.31, 0.17 and 0.56 respectively under these conditions. Identification of APl APl is a white to slightly yellow crystal which is relatively stable in air. It iS soluble in hot chloroform or in a chloroform-methanol mixture but its solubility is low in cold solvent. A direct probe mass spectrum of APl is shown in Fig. 3. Upon ionization, the compound yielded a base peak at m/e 69. Other major peaks are m/e 139, 126, 111 and 98. Also of importance is the series of peaks l4 molecular weight units apart indicating a long chain hydrocarbon 41 Fig. 3.——Mass spectrum of APl (mixture of 2 homologues). This Spectrum was taken at M.S.U. NIH mass Spectrum facility. Direct probe was the mode of sample insertion. Ionization voltage was 70 eV, and the electron multiplier was 2000 V. M/e 69 was the base peak. Other major peaks are at m/e 84, 98, 111, 126 and 139. Two molecular ions are at m/e 406 and m/e 434 respectively. A series of well spaced peaks with 14 molecular weight units apart can be observed between mass unit 180 to 300 indicating a long chain alkyl structure in the molecule. 42 38515 ZRDOFV 1 29/79 IRS! 4800*10**2 SCRLE 1.00 RT 1.81 IK 0.00 FE) 27,1979 5:05P HT 100 IRK 1000 "UL 5.00 SCI. 33 $000 0 2T1 0.00 l 1 1 50 70 90 110 130 150 170 190 210 230 _ 90515 SRDOFF/1/29/79 303E 4800*10**2 SCRLE 1.00 RT 1.81 IK 0.00 FE) 27,1979 5:05P HT 100 )RK 1000 HUL 5.00 SCH! 33 SCH. 0 2T1 0.00 390 410 430 450 470 490 510 530 550 570 00515 SRDOFF/1/29/79 008E 4800*10*+2 SCRLE 1.00 RT 1.81 0K 0.00 FE) 27,1979 5:05P HT 100 IRK 1000 "UL 5.00 SCI. 33 SCH! 0 2T1 0.00 560 580 600 620 640 660 2;: 70: 720 740 43 structure. Two molecular ions were observed, m/e 406 and 434, with an intensity ratio of about 1:3. Because I suSpected that sample APl was a mixture of 2 homologues, APl which had been purified from preparative TLC was put through a reverse phase HPLC column (c-8). A separation of this fraction into two major components was achieved, the results of which are shown in Fig. 4. Two major peaks which had a ratio roughly 1:2.5. These were each collected and reextracted and when dried, they appeared as silver-white flakes. Mass spectral data from samples of peak I and II are Similar to APl with molecular ions at m/e 406 and m/e 434, respectively. The melting points of samples from peaks I and II were 107-108°C and lll-112°C respectively. Peak matching on the m/e 126, 139 and 434 peaks yielded empirical formulas for each respective peak of C6H6O3, C7H7O3 and C28H5003' Based on this information, the structure of APl was proposed as a mixture of 6-n-heneicosyl- 4-hydroxy-pyran-2-one and 6-n-tricosyl-4-hydroxy-pyran-2- one (I). OH OH OCH3 \. \\ 1 (1 Li 0 0 C) o orc211143 CH3 CH3 C23H47 (I) (II) (III) Peak In/e 126 arises from the loss of either alkene (C20H40 Or 022H44) and results in the formation of 6-methyl-4- hydrtaxybpyran—Z-one (II). The mass Spectrum of this 44 .qu o>flnonomonm no ponmnnsm onEom and an po>nomno oHoB mnoom manonHEon Inoo HHoEm N hano cno uson H ndono conflswou mmooonm oHonB one .m.Nua noono mo oHnon o o>on mxoom N omone .>Ho>flnoommon ozonmlnonemlexonc>n1w1Hmmooflnn1o pno on01m1non>m1mxoncwn1vlameOHononlm mo ponmnnnocfl oHoB .HH pno H .mnoom HofloE ose .onEom onn o>H0mmnc on cows nno>H0m onn on one wanononm mo: noon muonm nmuflm one .Bonno Uncoom onn no vo.o on comoonooc Uno non onn no onensflmon onn no onoom Hasw mos mnnmnoc Hoonnmo mH.o no nom mos mnfl>flnnmnom nonoonop one .En 0mm no now mos nnmnono>o3 nonoonop onn pno nson Mom mononfl m no moB coomm nnono one .nonH onosqm uom canoe oomlooma mo oHSmmon o no onanE\HE N no pomsdm mos nno>HOm one .HononnoE on nonoB won mos omonm oaflnoa one .cows mos nwlmm nHOmonnonAV nfidaoo omonm omno>on o>nnonomonm < .mosmoHoEon Hmn OBn no oHHwonm noflnonomom mnmoumonofionno UHSwHH ousmmoum nmnm .mnm 45 l a- who concoE T 8 3 8 8 46 compound has 5 major peaks at m/e 69, 43, 98, 126 and 85 (91). The compounds I have isolated also produce these peaks and therefore match very well with this structure. The methylation of APl results in the formation of a yellowish crystal. Gas chromatography separates this mixture into 4 peaks. GC-MS analysis of each peak yielded molecular ions of m/e 420, 420, 448 and 448 indicating that the addition of 1 methyl group on each of homologues had occurred. This confirmed my proposed structure of only one -OH group. Peaks of m/e 43, 41, 125, 153, 69 and 140 were obtained which also corresponded to the mass spectrum of 6-methyl-4—methoxy—pyran-2-one (III) which has major peaks at m/e 112, 69, 125, 43 and 140 (91). The 4 derivatives of methylation can be explained on the basis that when diazomethane was used to methylate pyrone, it yielded a mixture of d and y methylated pyrones (34) because pyrone exists as a mixture of 2 tautomers under acidic conditions. OH ._____1 a—pyrone v-—-- y-pyrone R \O R OH Thus with 2 homologues, methylation produced 4 derivatives which was confirmed by GC-MS analysis. The UV absorption spectrum of API in 95% ethanol is shown in Fig. 5. It had an absorption maximum at 283 nm and a minimum at 241 nm. Using a molecular weight of 434, 5A 47 .2 wloa x mom.H moB COHnounnoonoo oamfiom .A 11111 v Hononno an mom weo.o pno A v Hononno wmm onos mnco>H0m one .Hmd no Esnnoomm noHanomno >3 .m .mnm 48 HES Ihmvzmn m><>> owm com com "000000 l o.— 0N BONVEHOSQV 49 the calculated extinction coefficient was 6 = log 4.09 at 283 nm. It also had an abnormal hypsochromic Shift in alkaline solution AEtOH’KOH max 278 nm e = log 4.32. This phenomenon has been observed in the 4-hydroxy-2-pyrones and their enol tautomers (7, 8). The vanillin test on APl was negative. Both observations exclude the possibility that API is a phenolic compound (75). The high extinction coeffi- cient indicated that APl was a ring structure, but the value was too high for an aromatic compound. The absorption wave- length indicated that APl was a conjugated chromophore. The UV spectrum is similar to the compounds triacetic lactone (6-methylpyranone) (AHZO 288 nm e = 6750) (61, 99) and 3,6- max EtOH dimethyl-4-hydroxy-2-pyrone (Amax 288 nm e = 8300) (1) which are Similar to our prOposed structure. Because y- pyrones have absorption peaks around 240 nm, my spectral results suggests that APl exists principally as the a—pyrone. The infra red (IR) absorption spectrum is shown in Fig. 6. APl has absorption peaks at 2940 and 2880 cm-1 1 1 (0:0), 1645 and 1565 cm’ (C=C), l (aliphatic C-H), 1680 cm- 1300 cm"1 (c-o stretching), 1250 cm‘ (Ar-O vibration) and 1145 cm.1 (C-O stretching). After methylation, IR absorption was observed at 2950 and 2880 cm-1 (aliphatic C—H). 1725 cm--'1'(C=O), 1570 cm'1 1 (C=C) and 1140 cm_ (C-O stretching). The nuclear magnetic resonance (NMR) spectrum of AEEL in CDCl3 is Shown in Fig. 7. Using the CHCl3 peak as a reference at 7.27 ppm, 5 absorption peaks were observed at 5.5) (singlet), 5.5 (Singlet), 2.48 (triplet), 1.29 (multiple). 50 .HHSE Hoflsz mo nonon mos Enhnoomm one .Hmn no Esnnoomm GOanHOmno cononnnH .o .mnm 51 IWRABKNMFTTABK3E(%J nooo wmoo .__n!_ » mnoo - ammocmzo3 .m .mnm 61 AECVIFOZm; mZSS 06 OM EONVBUOSGV 62 (64) reported that the UV absorption Spectrum of orsellinic ac1d had a Amax 260 nm, e = 9750 and Amax 296 nm in 0.1 N HCl. These results are similar to values for AR3. The IR Spectrum of AR in CCl4 has absorption peaks 3 at 3590 and 3270 cm’1 (-OH), 2950 and 2875 cm“1 (aliphatic C-H), 1440 cm‘1 (CH-C), 1720, 1650, 1320, and 1110 cm‘1 (C=O and ester), 1620, 1585, 950, and 850 cm.1 (benzene substitution) and 1260 cm”1 (aromatic -OH). These results compare well to the IR absorption of B-resorcylic acid methyl ester. The NMR Spectrum of AR had 6 peaks at 11.53 (sing- 3 let), 6.27 and 6.24 (multiplet), 3.93 (singlet), 2.85 (triplet), 1.28 (multiplet) and 0.9 (triplet) ppm respec— tively. The intensity ratio of the peaks was approximately 1:2:3:2:40:3. The assignment of these protons was as follows: (f) a = 0.90 ppm (g) OH OH (e) b = 1.28 ppm c = 2.85 ppm - H (“H \S°C3 d=3.93ppm O CH3-(CH2%;CH2 (d) e = 11.53 ppm kn (b) (C) f = 6.27 and 6.24 ppm The 2 aromatic ring protons absorb at 2 different ppm indicating an asymmetric molecule. Thus, the position of methylcarbonyl group had to be at 4 position on the ring. Proton e shifts to the high field because of hydrogen bonding between the —OH and the C=O group. The intensity 63 of the peak diminished when D20 was added to the solution. The other —OH was nothetected in the NMR. Thus all data obtained confirm the prOposed structure of AR3. Identification of AR5 AR5 purified from TLC was a white to pale yellow crystal which turned yellow on exposure to air. It had a melting point of 94-95°C and was stable in a mixture of chloroform and methanol. On silica gel plates, it turned red to purple upon exposure to air or iodine vapor. The vanillin test for phenol was positive and its reaction when sprayed with the diazoreagent yielded a red dye. Direct probe mass Spectral analysis of AR5 yielded a base peak at m/e 124. Other major peaks were m/e 123, 137, 166 and 404. A series of peaks, 14 molecular weight units apart, also was present between 180-300 indicating the presence of a long chain hydrocarbon on the molecule. Two small peaks at m/e 446 and 474 were suspected to be the molecular ions. The mass spectrum of chemically ionized AR5 confirmed that the molecular ions are m/e 446 and 474 in a ratio of 6:4 respectively. Methylation of AR5 results in the formation of a yellowish crystal. Mass Spectral analysis of this material yielded a base peak at m/e 138 (1 methylation of 124 peaks). Other important peaks were m/e 152 (2 methylations), 124 and 43. Other peaks present were m/e 180 (1 methylation of 166 peak); 418 and 460 (l methylation of 404 and 446 64 respectively) and 432 (2 methylations of 404) indicating there were 2 exchangable protons in the molecule. This compound was thought to be very similar in structure to AR3 and a derivative of ARl with an extra group of 42 molecular weight. The peak at m/e 166 was thought to be the ion resulting from the B-cleavage and loss of side chain as alkenes (C20H40 or C22H44). The data suggest that the functional group CHz-C was the substituent of mass 42 and the prOposed structurg of AR5 homologues are 5-n-heneicosyl- 4-acetyl-resorcinol and 5-n—tricosyl-4-acety1-resorcinol (VI) OH R = C21H43 or C23H47 (VI) 3 The UV absorption spectrum of AR5 has 2 absorption EtOH peaks at Amax 271 nm, log e = 3.935 and Agng 305 nm, 109 E = 3.57. Two absorption minima are at AEEEH and AEFOH m1n 242.5 nm 295 nm. A bathochromic shift and increase of extinction coefficient was observed upon the addition of KOH to the solution of AR5 indicating the presence of a phenolic group. The UV absorption Spectrum and extinction coeffi- cient of AR5 was very Similar to AR3 indicating a Similar chromophore. 65 The IR spectrum of AR in CCl 1 has absorption peaks 1 5 (-OH), 2960 and 2870 cm“ 1 4 l at 3620 cm‘ (C-H), 1720 cm' (c=0), 1605 and 1585 cm" 1 (aromatic ring substitution), 1 1455 cm” (C-H) and 1215 cm’ (phenolic group). The NMR Spectrum of AR5 has six peaks at 11.38 (Singlet), 6.5 (singlet), 6.31 (singlet), 3.0 (triplet), 2.60 (triplet), 1.28 (multiplet) and 0.9 (triplet) ppm respectively. The intensity ratio of these peaks is about l:l:l:3:2:40:3. The assignment of these peaks are as follows: = 0.90 ppm = 1.28 ppm = 2.60 ppm 3.0 ppm = 11.38 ppm = 6.31 and 6.50 ppm 00 Q. momoo‘m II (a) (b) (c) Two different absorption peaks of protons on the benzene ring indicate an asymmetric molecule and thus the location of the acetyl group has to be at carbon 4 of the ring. The proton absorption at high field indicated hydro- gen bonding between an -OH and a group (C=O) ortho to it. Identification of AR4 AR4 which had been purified by preparative TLC was a white crystalline compound of melting point 95—101°C. It was unstable and turned brown upon exposure to air. It was insoluble in a chloroform but soluble in a 66 chloroform—methanol mixture. On silica gel plates, it turned red or purple upon exposure to air or iodine vapor. Its vanillin test for phenol was positive and it reacted with diazosulfanilamide to produce a red dye. The direct probe mass spectrum of AR4 had a base peak at m/e 124. Other important peaks were m/e 123, and 137, results which were very similar to the ARl Spectrum (75). Two molecular ions were noted at m/e 420 and m/e 448 respectively. Peak matching on peak 448 yielded an empirical formula of C29H5203. AR4 consumed large amounts of diazomethane upon methylation. After methylation, it became a white to yellowish crystal. A short GC column (61 cm, SE-30) was used to separate the methylated AR4 into 2 broad components (A and B) with a peak ratio of 3:4 for the 2 homologues. .A longer column (183 cm SE-30) resolved each individual component (A or B) into 3 fractions for a total of 6 components in methylated AR4. GC-MS analysis of each of the three compounds derived from fraction A revealed two or three methyl groups on each homologue and a small amount of a tetramethyl derivative. The mass data for these methylated derivates were m/e 152 (base peak) and 488 (mole- cular ion), m/e 166 (base peak) and 462 (molecular ion) and m/e 180 (base peak) and 476 (molecular ion). These data indicate that the compounds separated by the GC peaks are 2, 3 and 4 methyl group-containing derivatives of the fraction A. The other 3 components (methylated derivates of the AR4 fraction B) were 28 molecular weight units 67 heavier than those derived from fraction A. The fact that derivatives containing 3 methyl groups had base peaks of 166 suggested that only 2 hydroxy groups could exist on the ring (the molecular weight of tri-methoxy toluene is 182) and that the other oxygen was probably on the Side chain as an alcohol. The following structure of AR4 was proposed (VII) H R‘C-CH = . 2 R C19H39 or C21H43 OH OH (VII) 5-n-(2—hydroxy)—heneicosy1resorcinol and 5-n-(2-hydroxy)— tricosylresorcinol. The UV absorption spectrum of AR4 was shown in Fig. 9, it was very similar to the spectrum of ARl (75) having 2 absorption peaks at Agng and Agzgfl 275 nm, 10g 8 = 3.19. Upon the addition of 281 nm, log 6 = 3.199 alkali (1N KOH) there occurred a shift in the absorption peak to 297.5 and 292.5 nm with an increase of extinction coefficient. These data indicate the presence of phenolic groups. The IR spectrum of AR4 had a very broad and strong 1 (-OH group). 2890 and 2820 cm—1 (aliphatic l 1 peak at 3300 cm— (benzene substitution), 1160 cm- 1 C-H), 1600 and 830 cm" (aromatic alcohol), and 1470 cm- (CH-C). 68 .2 enon x mm.m mos noHnonnnoonoo oamsom .A 11111 v Hononno wmm on mom 2 H.o poo Alllllv Hononno wmm onoz mnno>H0m one .vmn no Esnnoomm n0nnmn0mno >3 .m .mnm 69 HES IPGZN... m><>> 060 OCN 132 0 Q 1- EONVBHOSEV 70 The NMR Spectrum of AR4 had 5 major peak at 6.24 (multiplet), 3.75 (Singlet), 2.50 (multiplet), 1.27 (multi- plet) and 0.89 (triplet) ppm each. The ratio of these peaks was approximately 3:1:2:40:3. Assignment of these protons is as follows: OH a = 0.89 ppm H (e) H b = 1.27 ppm CH3-(CH2)n-C-CH2-OH (e) c = 3.75 ppm OH d = 2.50 ppm (a) (b) (C) (d)(:) OH e = 6.24 ppm The structures and properties of unique lipids from encysting A. vinelandii are presented in Table 1. The Origin of Carbon Atoms in Unique Cyst Lipids The synthesis of unique lipids occurred after the induction of encystment at which time BHB was the only carbon source available. Therefore, it is very likely these compounds were derived from BHB. This possibility was investigated by using 3-14 C-BHB to induce encystment and determining the distribution of radioactive carbon in cyst components. The total uptake of BHB, the radioactivity incorporated in the lipid fraction, as well as the change in culture turbidity were followed for 5 days and the results are shown in Fig. 10. The culture turbidity increased from OD = 0.6 to 1.0 within 24 hours after induction of encystment and reflects an increase of cell number. After reaching the pOpulation peak, the culture Table l.--Structures and Properties of Unique Lipids from Encysting A. vinelandii. 71 Designated Chemical Molecular Melting UV Absorption Mass Spectrum Vanillin Name Structure Weight Point Peak Base Peak Test ’01-! ARl R / Q) 404 and 432 90-92°C 27S log €=3.25 124 (+) ___. 87:13 281 log €=3.24 OH OH / 584 and 612 loo-102°C 279—281 log 124 (+) AR2 R / \ 86:14 8:3.20 O-galactose OH AR R / \ 462 and 490 95—97°c 265 log e=3.99 182 (+) 3 1:1 300 109 €=3.59 C=O OH OCH3 OH OH // ‘\ 420 and 488 95-101°C 281 log €=3.199 124 (+) D . = AR R-C-CH 3.4 275 log 6 3.19 4 . 2 ___ H OH OH 446 and 474 94-95°C 271 log €=3.935 124 (+) AR5 R 6:4 305 log €=3.57 C=O on I CH R 3 “" 406 and 434 107-108 283 log €=4.09 126 (-) AP 0 H 2:5 (406) 1 / 111-112 4 o (43 ) R-E=O H2 AP2 O // OH \ 448 and 476 97-101 286 log 93.72 153 H / 1:1 72 .onsnaso no HE\on:nHE nom nnsoo mo pommonmxo mos >na>nnoooflpon one .mommsn m.nn5m on mHHoo mo HE com on poppo mo: mmmIOvH no no; m .annononn> .m no nnoEnmmono onn mnnnnp .1111. mflaoo no noHnoonn UHQHH onn once pononomnoonn mmm o>flnoooflpon onn pno OIIIO mmmnuvH mo ononmd onn rolllo mnflmnop HoOflnmo mo omnono one .Hflpnononn> .m mo nnoEnwmono mnHHSU coanoonn UHQHH onn onnn mmmlvva no QOHnonomnoonH .OH .mnm A 73 IKEIEBCIFIEIIKIUISIE / C’ V! —= 1o l I {.l [I I I T I I I .fir“\\. .\\\\\‘\~ ‘ ° “\\\\\\ q I . Q\fi\ 41 l 1 1 1 3‘5: 1 . . . i j :2 e '0 w 'd '0 ,0: 96 120 18 24 36 48 60 72 FQCIIJIR El 12 6 74 turbidity remained constant for 36 hours and then started to decrease, probably due to the mobilization of PHB with a resultant decrease of refractility of cells. BHB uptake started at the induction of encystment and increased linearly for 48 hours. At this time, the total radioactivity in the cells amounted to 30% of total radioactivity added to the medium. A gradual decrease of radioactivity then occurred which coincided with the decrease of turbidity. A significant increase of the radioactivity in the total lipid began at 8 hours after the induction of encystment. It then increased sharply after 12 hours and continued to increase up to 48 hours after induction, after which time it remained relatively constant during the maturation period of cysts. Five days after the induction of encystment, the total lipid fraction contained 68% of the total radioactivity in the cysts. The possibility that some of the carbon Skeleton in these lipids was derived from sources other than BHB was also studied. Vegetative cells were labeled with 2—14C- glucose and then induced to encyst with non-radioactive BHB. The total radioactivity of cells before and after encystment was determined. These cells had incorporated 10% of the l4C-glucose added to the medium. Five days after the induction of encystment, 75% of the radioactivity from vegetative cells remained in the cysts and, of this, 14% 75 was in the lipid fraction. The lipid fraction was separated by TLC and the radioactivity of each compound was counted. The phospholipid fraction contained 45% of the total radio- activity in the lipid or 4.7% of the radioactivity of vegetative cells (RVC). In the lipid compounds synthesized during the encystment, only ARl had a Significant amount of radioactivity (2.9% RVC). Other compounds only had counts slightly above background, AR3 + AR5 (0.8% RVC), APl (0.6% RVC) and AP2 (0.2% RVC). The radioactivity in these com— pounds could have been derived from radioactive PHB accumulated during vegetative growth (87). I concluded that these unique lipids were synthesized from BHB used to induce the encystment. Relative Abundance of These Compounds in Cyst Lipid The relative amounts of each component in the lipids of 5 day old cysts were estimated by labeling these com- pounds with 14 C—BHB during encystment, separating them on TLC, and then measuring the radioactivity in each compound.. The results are shown in Table 2 (arranged according to the Rf value in solvent II). Table 2 compositions of lipids Total counts in these 9 compounds account for 81% of total counts in the extractable lipid fraction. The rest of the radioactivity carbon counts was recovered mostly in the phospholipd fraction. 76 pagan O o o o p o W o o o o o Hmpoa aim” me amm pH n o m H om m omH mH mom m we mm wvw H we 0 nnsoo mo omonnounom nnsoo oEoz no a mom was moo was Hes ems Hmo moo moo ossOdEOU Honoe .nnmso one moo m sonn mpaoaa no scanamoosoouu.m mnnme 77 Biosynthesis of Unique Lipids During Encystment The time course for the biosynthesis of these com- pounds during the encystment was followed by inducing l4C—BHB, sampling at various times and encystment with 3- extracting the lipids. The lipids were fractionated on TLC and the radioactivity in each component counted. The results are shown in Fig. 11. The biosynthesis of ARl started at about 8 hours after induction of encystment and increased sharply until it reached a peak at 36 hours. A small decrease in the level of ARl occurred in the period 36 to 48 hours after which time it remained relatively constant up to 120 hours. Significant synthesis of ARZ' AR3, AR4, AR AP and AP2 was observed 12 hours after encystment. 1 reached a plateau of concentration at 24 hours after 5. AR3 encystment and stayed at a constant level for the rest of encystment process. This compound, although low in con— centration in the developing cyst, was the first of the unique lipids to reach a plateau or steady state level. APl and AP2 reached their concentration peaks 48 hours after encystment and thereafter decreased somewhat. The amount of AR2 reached a peak at 60 hours, remained at its high level for about 2 days and then decreased slightly. The biosynthesis of AR4 was difficult to estimate and subject to large fluctuations because it migrated close to ARl during chromatography. A slight contamination with AR 1 could result in a large error in estimating the relative Fig. 11. 78 The biosynthesis of unique lipids during the encystment of A. vinelandii. l4C—BHB labeled at C-3 position was used to monitor the biosynthesis of these unique lipids. Enc stment was induced with 0.2% BHB plus 9 uCi of 4C labeled BHB in 300 ml of cell suspension. Samples were taken at various times during the encystment process and the lipid extracted and separated on TLC. The radioactivity incorporated into each component was determined and expressed as counts per minute/ml of culture. Bottom H AR4 ,k—A AR5 , q—fi AR3 100 GR M. 300 200 50' 30 20 79 120 80 amount of AR4. Nevertheless, the time course of the bio- synthesis of this compound was estimated. AR4 reached its peak concentration about 48 hours after initiation of encystment and decreased slightly as the cyst matured. A Significant increase of AR5 synthesis occurred 12 hours after encystment and the compound continued to increase in concentration until 72 hours, after which time its rate of synthesis decreased. AR5 was the only lipid that was still being synthesized after the uptake of BHB had stopped. Except for AR and AR5, the synthesis of these unique lipids 3 was parallel to the BHB uptake and only a slight decrease occurred in concentration of any of them after they reached their plateau concentration. It is unknown if these com- pounds were being degraded or simply released into the medium. Distribution of Unique Lipids in the Cysts The distribution of unique lipids in the cysts was studied by inducing encystment with l4C-BHB, fractionating cysts into exine, intine and central bodies, and extracting and analyzing the lipid in each fraction. Central bodies contained 70% of the extractable radioactive lipids, 23.2% were found in the exine and 6.8% in intine (using the total extractable lipid count from all 3 fractions as 100%). The amount in central bodies could have been slightly over- estimated because some unbroken cysts were co-precipitated with the central body fraction during its collection by 81 centrifugation. A few exine fragments were also observed microsc0pically in this fraction. On the other hand, while a few central bodies were also observed microsc0pically in the exine fraction, the overall cross-contamination was not severe. The lipid composition in each fraction is shown in Table 3 and the distribution of each component in the cyst is Shown in Table 4. ARl and AR2 were the most abundant components accounting for 35% and 20% respectively of the total lipid in all three fractions. Central bodies seemed to contain a Slightly higher percentage of ARl whereas AR2 was uniformly distributed. Of the pyrone derivatives, APl and AP2 were located principally in the central body fraction which contains 87.47% of total APl and 82.97% of total APZ' AR5 occurred primarily in the exine and intine. Fates of Unique Lipids During Germination The fates of the unique lipids were investigated by labeling them with 14 C-BHB during encystment, germinating the cysts with glucose, and analyzing the unique lipid distribution at various times during the germination. The results of this experiment are shown in Fig. 12. None of the phenolic compounds (ARI, AR2, AR3, AR4 or AR5) underwent appreciable changes in concentration during germination of cysts. However, there was a decrease in pyronic compounds (APl, APZ’ AP3 and AP4) during the outgrowth period. AP 1 decreased by 70% and AP2 by 60% during this period. Since 82 Table 3.-—The Relative Composition of Unique Lipids in Each Fraction of Cysts.albrcrd Fraction Exine Intine Central Bodies Compound Percentage AR3 and above 13.1 15.7 11.1 AR5 9.7 11.8 4.1 ARl 33.7 36.1 36.5 AR4 3.8 5.3 2.7 APl 7.5 2.2 15.6 AP2 5.3 2.6 7.9 AP3 and AP4 4.0 5.6 2.9 AR2 22.9 20.7 19.2 aThe percentages were calculated using the total radioactive counts from 8 compounds in each structure as 100%. bSome neutral lipid co-chromatographed with AR3 and was not further separated. CCompounds are listed according to the sequence of migration distance on TLC. dThe distribution of total lipid count in 3 cyst structures: exine 23.2%, intine 6.8%, and central bodies 70.0%. 83 Table 4.--The Distribution of Unique Lipids in Cysts.a'b'c'd Fraction Exine Intine Central Bodies 1 Compound Percentage ! AR3 and above 23.2 6.4 70.4 AR5 35.7 10.1 54.2 ARl 19.5 4.9 75.6 AR4 25.6 8.3 66.1 APl 11.7 0.8 87.5 AP2 15.3 1.7 83.0 AP3 and AP4 25.8 8.4 65.8 AR2 23.7 5.0 71.3 aThe percentages were calculated using the total radioactive counts of each compound in these three structures as 100%. bSome neutral lipids co-chromatographed with AR3 and was not further separated. cCompounds are listed according to the sequence of migration distance on TLC. dThe distribution of total lipid count in 3 cyst structures: exine 23.2%, intine 6.8%, and central bodies 70.0%. 83 Table 4.--The Distribution of Unique Lipids in Cysts.a'brcrd Fraction Exine Intine Central Bodies Compound Percentage AR3 and above 23.2 6.4 70.4 AR5 35.7 10.1 54.2 ARl 19.5 4.9 75.6 AR4 25.6 8.3 66.1 APl 11.7 0.8 87.5 AP2 15.3 1.7 83.0 AP3 and AP4 25.8 8.4 65.8 AR2 23.7 5.0 71.3 aThe percentages were calculated using the total radioactive counts of each compound in these three structures as 100%. bSome neutral lipids co-chromatographed with AR3 and was not further separated. CCompounds are listed according to the sequence of migration distance on TLC. dThe distribution of total lipid count in 3 cyst structures: exine 23.2%, intine 6.8%, and central bodies 70.0%. Fig. 12. 84 Fates of unique lipids during the germination of A. vinelandii cysts. The unique lipids in the cyst were labeled by the addition of 200 mg BHB plus 10 uCi of C14—3-BHB into into 100 ml of vegetative cell suspension of A. vinelandii. After 5 days of incubation, cystS were harvested, washed twice and suspended in the same volume of Burk's buffer. The cyst germina- tion was started by the addition of 2 ml 50% glucose into the suSpension and samples were taken at 2 hour interval. Cells were harvested by the centrifugation and lipid extracted and separated on TLC. The radioactivity in each component of unique lipid was determined and expressed as counts per minute/ml. O—OARl V—V ARZ H AR3 C1—C1 AR4 O—O AR5 _ * APl ‘—AAP2 A‘AAP3+AP4 * C.P. M./ml 1500 500 85 nl wommm on“ Eonm pmnomuumndm mmz msHm> mflna .uomupxm Hmuaflm CH OHOEQ mm.o Ucm mumuuaflm CH maoam m.o mo moam> m on: Mamas Hmmmsn m.xH5m maanmpm N.NOH MH.N m.ma OH m.HH mHHOU m.mm m.om mma m.mv OH Acflmuoum mE\mHOEmv m2¢lo m.mm m.o m.m m.m m.m maamo ucmumcHOQSm m.e m.mm m.mm m.m m.m Aaa\maosmc mzmuo on omm on cam on mam ms omm m: omm Hs\cflmuoum mm.o mn.o ms.o mm.o He.o cmumm>umn no Houmowaw vmz+mpmumo¢ mumumo¢ vmz+wmoosaw omoous Esflpmz nusouo .mpflsam OHSDHDU pom mHHmU Hflpcmamcfl> .m CH mHm>mH mS¢I0HHo>UII.HI¢ OHQMB 110 I then decided to assay for the substance in the partially purified samples that was competing with c-AMP in the binding assay and which was also sensitive to 3',5' cyclic nucleotide phosphodiesterase. The results are shown in Table A-2. From the table, it can be seen that cells contain 0.2 pmole c-AMP/ml of culture and the medium contained 3.2 pmole c—AMP/ml. Setlow has obtained 50% recovery of c-AMP (23) using this purification procedure. Taking this factor into account, A. vinelandii culture contained 0.4 pmole c—AMP/ml of cells and 6.4 pmole c-AMP/ml in the supernatant medium. The intracellular c-AMP level was therefore 0.9 pmole/mg of protein. Discussion The values of the intracellular and extracellular c-AMP level in A. vinelandii, Table A-1, are in the same range as those reported for E. coli (1, 16). After assaying only the phosphodiesterase-sensitive component, Table A-2, I estimate there are about 0.9 pmole c-AMP/mg protein intra- cellularly and 6.4 pmole c-AMP/ml in the A. Vinelandii culture medium. This value is considerably lower than that obtained by the binding assay alone because of the very high background level in the cellular extract of compounds capable of non-specific binding. The discrepancy could be partially due to different growth conditions, but methods used to extract and assay for c—AMP also affect the values 111 h.H wm.o mm.o m.o HE Hoo.o mm.m m.m H m.s as Ho.o ucmumcummsm v.m em 0 ow HE mo.o m.o wo.o m.o mm.o HE moo.o o o N.H N.H HE Ho.o uomuuxm Hamo mm.o m.m v m.m HE mo.o ma.o m.m m.m m.m HE H.o HE\mHoEm mocmHOMMHQ Umummua DcmEummHB wmmumpmmflp asocuflz Aomwmupcmoaou nonsmoam mafia ooa. OHQEMm Udamausm mHoz ovam CH mzmlo >HHMHDHmm mo pcDOE¢ .mapmz paw mHHmU CH me>®A mz¢lo O>Huflmammlmmmumgmmflponmmonmll.mud OHQMB 112 obtained. There are 20-fold differences in c-AMP concentra- tion in the same organism reported by different researchers (15). I conclude that the protein binding assay for c-AMP is not very specific and requires that controls be run in order to substract for the background. This is particularly true for estimating intracellular c-AMP levels. It would appear that A. vinelandii has about l/10 of c-AMP content of E. coli K-12 (1) both intracellularly and extracellularly. The presence of c-AMP in A. vinelandii has never been shown directly, although there is evidence suggesting the existence of c-AMP in this organism. Reuser and Postma (19) found c-AMP could overcome glucose repression in the induction of C4-dicarboxylic acid anion translocator and C6-tricarboxylic acid translocator for the Kreb's-cycle intermediates. Lepo and Wyss (9) found the derepression of nitrogenase was accelerated by c-AMP. More recently, Page and Sadoff (14) found c-AMP induced competence of genetic transformation. Goldenbaum (7) found c-AMP inhibited the encystment process in this organism. Using a protein binding assay, he also obtained some c-AMP values in the culture but he was not able to detect 3',5' cyclic nucleo- tide phosphodiesterase activity in this organism. From my experiments, I conclude that c-AMP is syn- thesized by this organism. Although c-AMP identification with TLC was not very conclusive, the binding assay using 3',5' cyclic monOphosphate phosphodiesterase treated sample as control was good evidence for the presence of cyclic 113 nucleotides. In View of the fact that c-GMP is usually present in quantities far less than c-AMP, most of the phosphodiesterase-sensitive substance could have been c-AMP. A search for adenyl cyclase activity in this organism is essential and, if it can be detected, will be conclusive evidence for the presence of c-AMP in this organism. C—AMP has been suggested as a possible mediator during the cell differentiation process in Caulobacter crescentus (24), A. vinelandii (7) and Dictyostelium dicoideum (3). Thus it will be of interest to study the changes in intracellular c-AMP levels during encystment and germination in this organism and correlate them with the induction of new enzymes during these processes. There is also a good possibility that repression and derepression of nitrogenase is mediated by cyclic nucleotides. Prusiner et a1. (17) reported that several enzymes involved in ammonia assimilation in E. 39;; are regulated by c-AMP. Lepo and Wyss (9) found that c-AMP accelerated the derepression of nitrogenase when it was inhibited by NH+. More recently, Botsford and Villa (2) studied the metabolism of c-AMP in Klebsiella pneumonia and found that the accumulation of c-AMP was not influenced by the cells nitrogen source. They suggested c-AMP was not involved in the regulation of nitrogen fixation in this organism. Lim et a1. (10) found the expression of nitro- genase in Rhizobium japonicum was completely inhibited by c-GMP. Hydrogenase and nitrate reductase also were markedly 114 inhibited by this compound. The intracellular level of c-GMP decreased when nitrogenase was induced but c-AMP had no effect on nitrogenase. It may be that in the regulation of nitrogen fixation, c-GMP plays a more important role than c-AMP. 10. 11. 115 References Botsford, J. L. 1975. Metabolism of cyclic adenosine 3',5'-mon0phosphate and induction of tryptOphanase in Escherichia coli. J. Bacteriol. 124:380-390. Botsford, J. L., and G. G. Villa. 1979. Metabolism of cyclic-3',5'-adenosine monOphosphate in Klebsiella pneumonia. Abstract of the Annual Meeting of the American Society for Microbiology. K.12. pp. 165. Cooper, S., D. A. Chambers, S. Scanlon, and J. B. Respers. 1979. Detection of developmentally regulated c-AMP binding protein in Dictyostelium discoideum by photoaffinity labeling. XI th international congress of biochemistry abstract. p. 509. Dobrogosz, W. J., and P. B. Hamilton. 1971. The role of cyclic AMP in chemotaxis in Escherichia coli. Biochem. BiOphys. Res. Comm. 42:202-207. Epstein, W., L. B. Rothman-Denes, and J. Hesse. 1975. Adenosine 3',5'-cyclic monOphosphate as mediator of catabolite repression in Escherichia coli. Proc. Natl. Acad. Sci. U.S. 72:2300-2304. Gilman, A. G. 1970. A protein binding assay for adenosine 3',5'-cyclic monophosphate. Proc. Natl. Acad. Sci. U.S. 67:305-312. Goldenbaum, P. E. 1979. Inhibition of Azotobacter vinelandii morphogenesis by exogenous cyclic AMP. XI th international congress of biochemistry abstract. p. 509. Hylemon, P. B., and P. V. Phibbs, Jr. 1974. Evidence against the presence of cyclic AMP and related enzymes in selected strains of Bacterioids fragilis. Biochem. BiOphys. Res. Comm. 60:76-80. Lepo, J. E., and O. Wyss. 1974. Depression of nitro- genase in Azotobacter. Biochem. Biophys. Res. Comm. 60:76-80. Lim, S. T., H. Hennecke, and D. B. Scott. 1979. Effect of cyclic guanosine 3',5'-mon0phosphate on nitrogenase fixation in Rhizobium japonicum. J. Bacteriol. 139:256-263. Lowry, O. W., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 116 Makman, R. S., and E. W. Sutherland. 1965. Adenosine 3',5'-phosphate in Escherichia coli. J. Biol. Chem. 240:1309—1314. Okabayashi, T., M. Ide, and A. Yashimoto. 1963. Extraction of Adenosine-3',5'-phosphate in the culture broth of Brevibacterium liquefaciens. Arch. Biochem. BiOphys. 100:158-159. Page, W. J., and H. L. Sadoff. 1976. Control of transformation competence in Azotobacter vine- landii by nitrogen catabolite derepression. J. Bacteriol. 125:1088-1095. Pastan, I., and S. Adhya. 1976. Cyclic adenosine 3', 5'-mon0phosphate in Escherichia coli. Bacterial. Rev. 40:527-551. Polglass, W. J., D. Iwacha, and M. Thomson. 1978. Elevated cyclic AMP concentration in streptomycin dependent Escherichia coli. J. Bacteriol. 133: 422-423. Pursiner, S., R. E. Miller, and R. C. Valentine. 1972. Adenosine 3',5'-cyclic monOphosphate control of the enzymes of glutamine metabolism in Escherichia coli. Proc. Natl. Acad. Sci. U.S. 69:2922-2926. Rall, T. W., E. W. Sutherland, and J. Berthet. 1957. The relationship of epinephrine and glucagon to liver phosphorylase IV Effect of epinephrine and glucagon on the reactivation of phosphorylase in liver homogenate. J. Biol. Chem. 224:463-475. Reuser, A. J., and P. W. Postma. 1973. The induction of translocator for di and tricarboxylic acid anions in Azotobacter vinelandii. Eur. J. Biochem. 33:584-592. Rickenberg, H. W. 1974. Cyclic AMP in prokaryotes. Ann. Rev. Microbiol. 28:353-369. Robison, G. A., R. W. Butcher, and E. W. Sutherland. 1968. Cyclic AMP. Ann. Rev. Biochem. 37:149-174. Sayhoun, N., and I. F. Durr. 1972. Evidence against the presence of 3',5'-cyclic adenosine monOphos- phate and relevant enzymes in Lactobacillus plantarum. J. Bacteriol. 112:421-426. 117 23. Setlow, P. 1973. Inability to detect cyclic AMP in vegetative or sporulating cells or dormant spores of Bacillus megaterium. Biochem. BiOphys. Res. Comm. 52:365-372. 24. Shapiro, L., N. Azabian-Keshishian, A. Hirsch, and O. M. Rosen. 1972. Effect of dibutyryladenosine 3',5'-cyclic monophosphate on growth and differ- entiation in Caulobacter crescentus. Proc. Natl. Acad. Sci. U.S. 69:1225-1229. 25. Wise, E. M., S. P. Alexander, and M. Powers. 1973. Adenosine 3',5'-cyclic monophosphate as a regula- tor of bacterial transformation. Proc. Natl. Acad. Sci. U.S. 70:471-474. "1111111117111111113