U ‘-~ ~- .. .. .. - - ~:&J(~I~USE ”A" i: 8 3?.L-'£:tffi£‘§: 3:“)? 9‘ kCEi E-Et’s CERF; J3 . A. .. il Hi "-3 Hunt's fa: 15:: Bar‘s-“ca: 0f 1"}! .‘7. MP 5“}!th 32353:. ... v'Ji'JiVERiiT'Y' ‘ . Haw-Am Lcuise En e! 1966: if.) 7 5086 I WWW, L MI W!!! H: 1293 0105 Will/1 WI] WWW ' .x r '1 i {but ‘5 55,“. ”a $169: {3J- ‘i My? 5‘9 i m: ABSTRACT INTER-CTION OF HEflflUEY LABELED ANTIBODY WITH THE GLUCOSE DEHYDROGENASE OF BATILLUS CEEEUS By Helen L. Engelbrecht Stains for the detection of antigens by electron micro- scopic techniques can be produced by coupling heavy metals to specific gamma globulin. Mercury was coupled to anti- glucose dehydrogenase and the immunological properties of the modified globulin were investigated. The enzyme specific globulin was produced in rabbits, carboxy methylated and coupled at pH 8.6 to the diazonium derivative of para—aminophenylmercuric acetate. The extent of labeling achieved ranged from seven to twenty-one atoms of mercury per molecule of antibody. The mercurated globulin was separated from the reaction products by gel filtration with Sephadex G-25. The antienzyme activity of untreated, carboxy methylated, and mercurated antibody (14 atoms mercury per molecule) and mercurated normal gamma globulin were compared. The enzyme specific globulin lost 80% of its original activity on acetylation but 50% of its activity was recovered on mer- curation. Hercurated normal gamma globulin did not precipitate glucose dehydrogenaseo Spores of Bacillus cereus, which had been fixed in formalin and partially disrupted, were reacted with mercurated specific globulin , enzyme specific globulin, and normal gamma globulin. Only mercurated specific globulin Helen L. Engelbrecht produced discrete staining of the spore cytoplasm on electron microscopic examination. INTERACTION OF MERCURY LABE ED ANTIBODY WITH L THE GLUCOSE DLHYDBOGENASE OF BACILLUS CEBEUS By Helen Louise Engelbrecht A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Microbiology and Public Health 1966 ACKN WLEDGMENTS I wish to acknowledge with sincere appreciation the inspiring guidance of Dr. Harold L. Sadoff. His original thinking and patience, with my impatience, have been very generous. I also express my gratitude to Dr. Frank J. Bauscher of the National Cancer Institute and to Dr. John J. Ray, Jr. of Fort Detrick, Maryland, without whose wise counsel this thesis would never have been produced. 11 O TABLE OF CONTENTS IIJTRODUCTION o o o o o o o o o o o o o o o o o o o o LITEBALT BE IZEVIEL‘J. o o o o o o o o o o o o o o o o o Fractionation of Serum Proteins . . . . . . . . Preparation of Protein Derivatives. . . . . . . Hercurated Antibody as an Electron Dense Stain. IIATEBIALS AND IIETEIODSO o o o o o o o o o o o o o o 0 Preparation and Assay of Glucose Dehydrogenase. Preparation of Medium. . . . . . . . . . Preparation of Inoculum and Spore Crops. Assay of Glucose Dehydrogenase . . . . . Purification of Glucose Dehydrogenase. . Estimation of Protein . . . . . . . . . . . Preparation of Specific Gamma Globulin. . . Assay for I‘Iercury o o o o o o o o o o o o 0 Synthesis of P-aminophenylmercuric Acetate. Diazotization and Coupling of Mercurated Ani to Gamfla. GlObUlin o o o o o o o o o o o o Immunochemical Methods. . . . . . . . . . . o o In ooHoooo Modification of the Procedure of Cohn and TOI‘I‘ianl o o o o o o o o o o o o o o o o mar‘gel DifoSion Studies 0 o o o o o o o Electrophoresis of Gamma Globulin Fractions . . EXPEB IEEITAIJ RE SUI-IT S O O O O O O O O O O O O O O O 0 Production and Assay of Antibody . . . . . . . Hercuration of Gamma Globulin . . . . . . . . . Determination of Changes in Properties of Gamma Globulin by Mercuration . . . . . . . . . . . Spectrophotometric Studies . . . . . . . . ElectrOphOI'eth I'IO-bility o o o o o o o o o 3112 mile Inhibition o o o o o o o o o o o o 0 Electron Density . . . . . . . . . . . . . DISCUSSIOIJ o o o o o o o o o o o o o o o o o o o o o SUI'H'IAB o o o o o o o o o o o o o o o o o o o o o o REFEEIJCE S O O O O O O O O O O O O O O O O O O O O 0 17 18 19 20 20 20 26 26 26 29 31 39 45 47 Table 1. LIST OF TABLES Titration of Glucose Dehydrogenase by Rabbit Anti-Glucose Dehydrogenase . Titration of Antibody and Antigen in dimensional Diffusion (72 hours) Effect of PAPMA Concentration on the of Gamma Globulin. . . . Electrophoretic Mobility Box Titration of Untreated and Mercurated Anti- enzyme vs. Glucose Dehydrogenase in Ouchterlony Double Diffusion Plates. iv Two- Mercuration Page Figure l. 2. Effect of Various Concentrations of PAPMA on Antienzyme Activity . . . . . . . . . . . . . . Effect of the Concentration of PAPMA on the Mercury Content and Immunochemical Properties of Gamma GlObUlin. O O O O O O O O O O O O O O O 0 Spectral Analysis of Gamma Globulin . . . . . . Effect of Treated and Untreated Normal Gamma Globulin and Antienzyme on Glucose Dehydrogenase ACtiVity. O O O O O O O O O 0 O O O O O O O O O Titration of Antienzyme Gamma Globulin by 2-Dimensional Immunodiffusion . . . . . . . . . Electron Micrographs of Aggregates of Enzyme- Antienzyme Globulin which had been Beacted One Hinute and Dried on Formvar-Coated, Electron Microscope Grids. o o o o o o o o o o o o o o 0 Electron Micrographs of Spores which had been Disrupted in a Nossal Shaker, Formaldehyde-Fixed and Treated with Antienzyme Gamma Globulins . . Page 24 25 28 32 33 36 38 INTRODUCTION Microbiology like many other sciences has been dependent for its development on advances in instrumentation. Leewoenhoek's superior ability in lens grinding permitted him to develop a microscOpe which allowed him to visualize the "animacules", some of which are now known as bacteria. The development of stain technology was a consequence of the need to distinguish microorganisms from their environment. The Gram stain, used to differentiate bacterial cells from tissue was a major contribution in this field. The application of modern cytological techniques to the study of microorganisms with the light microscope was exploited for example, by Bisset (1950), Knaysi (1951), and Robinow (1960, 1960a). Their contributions consisted of the standardization of fixa- tion techniques in order to minimize morphological distortion and the development of staining techniques for the maximal visualization of the fine structure. With the widespread use and improvement of electron microscopic techniques, stress has been placed on the deveIOp— ment of specific stains for use with this instrument. These stains are utilized for the identification of organelles within the cells. Electron micrographic investigations have supplemented information already acquired by metabolic studies from which one can obtain an understanding of the integrated processes of the cell. Singer, (1959, 1960) borrowing from the fluorescent anti- body technique of Coons (l9h2), coupled an electron dense l 2 moiety to specific antibody and established a means for localizing specific proteins. He coupled ferritin to specific gamma globulins producing electron dense antibody and reacted the conjugate with cells. Antigen-antibody complexes were formed in which the specific sites of the antigen could be seen. It is likely that the conjugation of any heavy metal to antibody could be used as an electron dense stain if (a) the atomic number of the metal is sufficiently high and (b) the antigen-antibody aggregate is of sufficient size. The molecular weight of the ferritin molecule is approximately four times that of gamma globulin. The high molecular weight complex might limit the resolution and could also affect the specificity of the antigen-antibody reaction. The objectives of current research in staining technique for electron microsc0py are to achieve lower resolu- tion and higher specificity of the stain, yet not alter the reactivity or specificity of the conjugated antibody. Kulberg and Azadova (196A) stained viral particles with virus-specific gamma globulin which had been coupled to the diazonium salt of p-aminophenylmercuric acetate (PAPMA). This compound is of considerable interest for it is possible to couple several of these molecules to the protein and still maintain its immunochemical properties. Immunochemical techniques have been widely applied to the study of cell differentiation. The success of the pro- cedure lies in the fact that in the course of differentiation, uniquely different proteins are synthesized within the cells 3 and these can be readily detected. Sporulation is a differen- tiation process which occurs in the Bacillaceae and the physiol— ogy of the process should be ammenable to study by immuno- chemical procedures. The following criteria should be met in order to apply immunochemistry to the Sporulation process. The proteins investigated must be either spore-specific or vegetative cell specific. They should be antigenic in suit- able laboratory animals, be readily assayed and be capable of purification so that highly specific antibody against them can be prepared. Glucose dehydrogenase from spores of Bacillus cereus was selected as the antigen in this study because it is a Spore specific protein. The enzyme is stable on storage and a sensitive, yet simple assay is available. Furthermore, highly purified preparations of glucose dehydro- genase can be prepared with relative ease. The object of the work discussed in this thesis is two- fold: (a) to establish the conditions for the preparation of highly mercurated, specific antibody to glucose dehydrogenase; (b) to investigate the feasibility of its use as an electron dense stain. LITERATURE REVIEW One of the objects of the present study was to modify the structure of specific gamma globulin by coupling to it mercurated diazonium salts. In order to determine the pro- perties of the resulting conjugate, it was essential that a separation of the modified protein from the native protein be achieved. Fractionation of Serum Proteins Various systems of fractionation of serum proteins have been used. Landsteiner (1914) and Nichols et. a1. (l9#8) described methods utilizing alcohol precipitation. These procedures were conducted at 0 to 5 C. Alcohol precipitation above these temperatures frequently results in protein denatur- ation. Marenzi and Barrow (1960) investigated the fractiona- tion of serum proteins using 2.13 to 2.5 M ammonium sulfate solution as precipitating agents. They found that there was no influence on the total globulin precipitation by either time or temperature and that the globulins recovered were native when examined electrophoretically. A chromatographic method of separation of proteins was developed by Peterson and Sober (1956) using substituted cel- 1uloses. The substituent groups were either anionic or cationic substances. The proteins were eluted by change of pH or ionic strength of the eluant. Humphrey and Porter (1957) separated proteins from human serum on diethyl amino ethyl cellulose columns using a sodium phOSphate concentration gradient for 4 5 elution. They obtained one principal fraction, gamma globulin, and three smaller fractions. Human serums have also been fractionated by gel filtra- tion. Fireman et. a1. (196M) used Sephadex G-200 to study the elution characteristics of nineteen known serum proteins. They obtained three major peaks which eluted in order depend- ent on their molecular size with the larger proteins eluting first. Smaller molecules appeared in later effluent fractions depending on their penetration of the gel and their reaction with the gel. Zone electrophoresis has been used for separation of blood proteins. Cellulose acetate films were used by Kohn (1958) who separated the serum into twelve distinct protein bands. Similar experiments on paper yielded a maximum of five proteins. Only very small quantities of protein can be separated by zone electrOphoresis. That is, it would be difficult to utilize this procedure as a preparative method, but it is useful for the identification of proteins on a semi-micro scale. Preparation of Protein Derivatives various modifications of purified proteins have been attempted. Changes in protein by addition of various chemical compounds have been investigated by Landsteiner and Gablons (1912, Landsteiner (1914), Iandsteiner and Lampl (19lh, 1915, 1917, and 1918), and by Iandsteiner and Prasek (1916). The forementioned investigators found that the treatment of protein and with acetic anhydride changed their antigenicity 6 and reduced their species specificity. When serum proteins were treated with alcohol and H280“, alcohol and HCl and diazomethane, the derivatives obtained differed antigenically from the original proteins. They also found that the extent of acetylation varied with the proteins employed. They con- cluded that the acetate groups had attached to the nitrogen atoms. Landsteiner and Lampel (1918) coupled horse serum to diazo derivatives of aniline, sulfanilic acid,1xma-amino- benzoic acid and atoxyl. The conjugates were injected into rabbits and antibodies were produced whose specificities were directed at the aromatic ring structure. The various azo- proteins were found to be distinguishable serologically. Landsteiner's studies have shown that injection of a modified antigen into suitable animals produces modified antibodies. Marrack and Orlans (1954) showed that the precipitating character of various rabbit antibodies was destroyed by acetylation. However, the acetylated antibody still combined with the antigen since antigen-antibody aggregates were formed by COprecipitation. Karush (1957) showed that the disulfide bonds are essential in the maintainance of the specific con- figuration of the combining region of the antibody molecule. He also showed that antibody globulins which had been treated with an excess of monoiodoacetate did not change their serologic properties. Sulfhydral groups on antibody globulin do not appear to be involved in the primary antigen-antibody reaction. 7 Mercurated Antibody as an Electron Dense Stain The use of mercury-antibody conjugates as immunochemical stains for electron microscopy was a modification of the work of Singer (1959, 1960). He utilized ferritin as an electron dense label for anti—bovine serum albumin, anti-bovine RNase, and anti-tobacco mosiac virus. Its size and iron content are the basis for its success as an electron dense stain. Many attempts have been made to introduce organic mercu- rials into proteins and to examine the resulting changes in protein properties. Cook et. a1. (1946) studied the effects of phenylmercuric nitrate on enzyme systems. They found that the activity of cytochrome oxidase, succinic oxidase, succinic, lactic, and glucose dehydrogenase as well as catalase were depressed due to the involvement of sulfhydral groups. Pepe (1961) used the diazonium salt of tetraacetoxymercuriarsanilic acid for conjugation to rabbit gamma-globulin. The resultant conjugate was electron dense, but less than six sites of conjugation were obtained with antibody gamma globulin. Kochler and Dowben (1961) used p-aminophenylmercury acetate (PAPMA) to examine the interactions of a nonionic detergent with proteins using a bovine plasma albumin- azomercurial complex. The PAPMA was prepared according to the method of Dimroth (1902) and diazotized with sodium nitrite. The coupling was primarily to the tyrosine and tryptophan residues of the protein. This same azomercurial was used by Kulberg and Azadova (1964) for specific contrast in electron microscOpy. They prepared an azo-mercurial 8 complex from virus-specific serum. The complex was more stable and greater yields of antibody were obtained than with the method of Pepe. Bizok and Kallos (1965) introduced mercury into the active site of chymotrypsin by using p-chloromercuribenzene- sulfonyl fluoride. The reagent combined irreversibly with a serine residue and the enzyme was irreversibly inactivated. Kendall (1965) labeled the amino groups of rabbit anti- serum (anti-PH 8 influenza virus) with N-acetyl-homocysteine thiolactone. He reacted the conjugate with 0.1% aqueous methyl mercuric hydroxide and produced an electron dense stain. Koike et. a1 (1964) coupled diazotized anthranilic acid to tyrosine residues of anti-PR-B influenza virus. The tyrosine-anthranilate complex was an effective chelating agent and treatment with heavy metals was used to produce electron dense stains. MATERIALS AND METHODS This thesis describes the mercuration of antibody to glucose dehydrogenase and the effects of antibody on the enzyme. The following is a description of those specialized techniques used in the preparation of antigen and antibody and in the measurement of the effects of mercuration. Preparation and Assay of Glucose Dehydrogenase Bacillus cereus is a member of the Bacillaceae and forms endospores upon completion of its exponential growth. The organism employed in this research was originally isolated at the University of Illinois and has been variously known as Hacillus terminalis (Stewart and Halvorson, 1953) and Bacing§ cereus T (Church and Halvorson, 1955). This strain produces an enzyme which lyses the sporangia so that free spores are easily obtained. Preparation of Medium Sporulation levels of 95% or greater are achieved in semi-synthetic growth medium, G medium, described by Stewart and Halvorson, 1953 and Hashimoto et al., 1960. The modified 'G' medium contained the following materials per liter: KZHPou, lg; (NH4)ZSOQ, 4g; MnSOu'HgO, 0.1g; MgSOu, 0.8g; yeast extrast,2g; glucose, 4g; ZnSOu, 0.01g; CuSOg-SHZO, 0.01g; CaClZ, 0.1g; and FeSOu-7H20, 0.001g. Dow Corning antifoam AF was also added at the rate of 1 ml/liter. The organism grew and sporulated well in either large or small 9 10 quantities at 30 C. For convenience in extraction and preparation of enzyme, one pound batches of spores were produced. Preparation of inoculum and Spore CrOps To obtain the levels of Sporulation described, it was necessary to attain some level of synchrony in the culture. This was achieved by developing a schedule of transfers in order to prepare an active inoculum for one hundred liters of medium. The stock culture was maintained in the sporulated state at 4 C on nutrient agar. The inoculation and transfer schedule was as follows: A fresh nutrient agar slant was inoculated from stock and incubated for eight hours. 'G' medium, 50 ml in a 500 m1 Erlenmeyer flask was inoculated from the slant and shaken for two hours. Two similar flasks were inoculated with 10 ml each of the culture and shaken for two hours. These cultures were then used to inoculate six flasks. After two hours, the 300 milliliters of culture were introduced into 2700 milli- liters of 'G' medium in a New Brunswick fermenter. (New Brunswick Scientific Co., New Brunswick, N. J.) An active inoculum for the 100 liter culture was attained after two hours of growth. The culture prepared in the 100 liter stainless steel fermenter (Stainless Steel Products Co.) sporulated, and lysed to produce free spores in approximately nineteen hours. The medium was cooled, the spores harvested in a Sharples 11 centrifuge, Model A5-l2, and stored as a frozen paste at ~15 C. Assay of Glucose Dehydrogenase Church and Halvorson (1955) first reported the nucleotide- linked glucose dehydrogenase in extracts of Bacillus cereus spores. The enzyme was partially purified and characterized by Doi, Church, and Halvorson (1959). These formentioned authors were unaware of the manganese requirement for the enzyme because they had not achieved sufficiently high levels of purification. A spectrOphotometric assay was used to measure the activity of glucose dehydrogenase. One molecule of nicotinamide adenine dinucleotide (NAD) is reduced per molecule of glucose oxidized, therefore, the rate or progress of the reaction may be followed by observing the light absorption by reduced NAD at 340 millimicrons. These assays were conducted in a Beckman DU spectrOphotometer equipped with a Ledland log converter and a SH Sargent recorder. The sample compartment of the spectrophotometer was maintained at 25 C. The assay cuvettes contained 0.5 milliliters 1 M tris buffer, pH 8, 0.2 milliliter 0.001 M Mn SOQ'HZO, 0.1 milli- liter 1 M glucose, 0.1 milliliter 0.02 uM/ml NAD, and 0.1 milliliter of enzyme preparation. One unit of enzyme catalyzed the reaction producing an Optical absorbance change of 0.001 per minute. Specific activity was defined as the activity per milligram of protein. Pure enzyme has a Specific activity of 500,000 (Bach, 1963). 12 Purification of Glucose Dehydrogenase Glucose dehydrogenase was extracted from disrupted spores of Bacillus cereus. It was purified by successive ammonium sulfate fractionations and sequential gel filtrations. Five hundred grams of frozen spores were thawed and suspended in 1200 milliliters of 0.05 M acetate buffer, pH 5.0. The suspension was placed in an Eppenbach colloid mill (model MV-6-3 Gifford-Wood Co., Hudson, N. Y.) which was cooled by circulating alcohol at -20 C. The initial rotor-stator distance was 0.06 inches (wide separation). Four milliliters of antifoam were added to the suspension and the mill was brought to about 25% of the full speed. Seven hundred and forty grams of glass beads (No. 110 pavement marking beads, Minnesota Mining and Manufacturing Co.) were added slowly. The speed of the unit was increased to approximately 85 per cent of the full Speed and the rotor-stator gap reduced to 0.03 inches. The spores were subjected to this treatment for 10 to 30 minutes and drained from the mill. Glucose dehydro- genase was present in the supernatant, obtained on centrifuga- tion of the extract. The enZyme precipitated in solutions of ammonium sulfate between 0.5 and 0.8 saturation. The precipitates were resuspended in 150 milliliters of cold 0.05 M acetate buffer, pH 5.0. On refractionation, the enzyme precipitated in the 0.35-O.60 saturation range. The precipitate was resuspended in 15 milliliters of cold buffer and then treated with one milliliter of 2% solution of protamine sulfate per 100 13 milligrams of protein to remove the nucleic acids. Further purification of the enzyme was accomplished by gel filtration using Sephadex G-100 (Pharmacia). The columns were 4 x 45 centimeters. Fractions of 5 milliliter volume were collected at ten minute intervals. The tubes containing the highest specific activity were pooled and dialyzed against 0.05 M acetate buffer, pH 5.0. To obtain more highly purified enzyme, the dialysate was concentrated by 1yophilization and re-chromatographed on a Sephadex G-200 column. By this procedure, enzyme of 60% purity in 50% yield was obtained (Sadoff, unpublished). Estimation of Protein The protein content of glucose dehydrogenase was estimated spectrophotometrically by the 280/260 ratio according to the method of Warburg and Christian (1942). The estimation of the protein content of rabbit gamma- globulin preparations was performed by use of the biuret test according to Kabat and Meyer (1961). One half milliliter of the protein sample was added to .75 m1 of the biuret reagent. The mixture was incubated at 37 C for one half hour and the optical absorbance read at 555g in Beckman DU Spectrophotometer using one milliliter matched cuvettes. A standard curve for protein content was prepared by using normal, electrOphoreti- cally pure rabbit gamma globulin. The protein content was determined in terms of milligrams per milliliter and converted to micromoles per milliliter using 167,000 as the molecular weight of gamma globulin. 14 Preparation of Specific Gamma-Globulin Ten Dutch belt rabbits were inoculated for the pro- duction of antibody to glucose dehydrogenase. Equal quanti- ties of glucose dehydrogenase (specific activity 100,000) and complete Freund's adjuvant (Difco Laboratories) were emulsified in a 1 cubic centimeter syringe with a 23 gauge needle. One tenth milliliter of the enzyme-adjuvant was injected subcutaneously into each of ten sites in the regions of the inguinal and brachial nodes of each rabbit. This amounted to the injection of 0.02 - 0.025 milligrams of pro- tein per rabbit. The pattern of inoculation was repeated after two weeks. After the second series of injections, twenty milliliters of blood were drawn by cardiac puncture. The third and subsequent injections were given in four or six sites. The enzyme, however, was emulsified in incomplete Freund adjuvant (Difco Laboratories). This was done to prevent severe reactions to the Mycobacterium lipOproteins which are in the complete adjuvant. After four bleedings, the animals could no longer sustain cardiac puncture and all subsequent bleedings were done from the marginal ear veins. The blood samples were collected in centrifuge tubes, placed in a slanted position, and allowed to clot. The clots were freed from the sides of the tubes and stored for one hour at 37 C prior to 12-24 hours refrigeration at 4 C. All serums were decanted and both serums and clots were subjected to centrifugation at 2000xg in a Survall-SS-l centrifuge. Additional serum was recovered from the clots by this pro- cedure. The serum was stored at -15 C. 15 To obtain gamma globulin from the serum, .65 milliliters of saturated ammonium sulfate solution per milliliter of whole serum was added dropwise. The final concentration of ammonium sulfate was .39 saturation. The precipitate was allowed to stand in the cold for one half hour, then removed by centri- fugation at 10,000 x g for twenty minutes. The supernant was decanted and the precipitate brought to the original serum volume with phosphate buffered saline, pH 7.2 (Merchant, 1964). The preparation was stable at either 4 C or at —15 C. Assay for Mercury The mercury assay was performed according to a modifica- tion of the method of Abadi and Wilcox (1960). The solvent for the mercurials including mercurated protein was a solution of five parts dimethyl formamide to one part glacial acetic acid. Diphenylthiocarbazone (Dithizone, Fisher Scientific, 06H5NH NH-CS°N:NC6H5) was used for the colorimetric titration. (l milligram dithizone per 200 milliliters chloroform.) The mercury standard contained one macromole HgC12 per milliliter. Mercuric compounds appear intense orange with the titration procedure while mercuri compounds are lighter. At their respective end points, each mercuric ion is bound to two dithizone molecules as keto complexes and each mercuri com- pound to one dithizone as an enol complex. The endpoint of the mercury titration is difficult to discern for an untrained technician. Cross-lighting with a micros00pe illuminator and a white background facilitated the titration. 16 Synthesis of P-aminophenylmercuric Acetate Para-aminOphenylmercuric acetate was prepared according to the method of Dimerot (1902). Mercuric acetate, 15.4 grams, was dissolved in 80 milliliters distilled water and stirred into 25 milliliters freshly distilled aniline (b. pt. 180°C) in 35 milliliters 90% ethyl alcohol. After forty-five minutes, crystals of product were formed. Five hours later, the crys- tals were filtered in a Buchner funnel with Whatman #1 filter paper, washed with distilled water, and air-dried overnight. The product was assayed for its mercury content and found to contain 57%. Assuming one atom of mercury per molecule, the product had a molecular weight of 352 which compares to 352 theoretical. Diazotization and Coupling_of Mercurated Aniline to Gamma Globulin Fifty milligrams of para-aminOphenylmercuric acetate were dissolved in 0.5 milliliters of 50% glacial acetic acid and placed in an ice bath. An excess of sodium nitrite was added slowly (approximately 10 milligrams) with constant stirring. Potassium iodide-starch test paper was used to ascertain that an excess of nitrite existed. The reaction proceded for one half hour. The sulfhydral groups of rabbit gamma globulin were carboxy methylated in the following manner. Two milliliters Of gamma globulin, 15-25 milligrams per milliliter, were incu- bated at room temperature for two hours with one milliliter of 0.1 M iodoacetic acid. The modified protein was dialyzed 1? overnight against 0.1 M Veronal buffer, pH 8.6. Three milliliters of protein solution were coupled to a 1:50 dilution of the diazotized mercurial at pH 8.6 in a Sargent pH stat. Forty moles of diazonium were used per mole of protein. The temperature was maintained in the range of 12 - 15 C. The reaction mixture was placed in an ice bath for thirty minutes and then applied to a Sephadex G-25 (coarse) column (1 x 20 centimeters). Upon elution with saline, the mercurated protein moved with the solvent front. Those fractions contain- ing the maximal concentration of protein (biuret determination) were saved. Immunochemical Methods Modification of the Procedure of Cohn and Torriani A modification of the method of Cohn and Torriani (1952) was used to test for antibody specific for glucose dehydrogenase. Two-fold serial dilutions of the enzyme, 15000 units per milli- liter were prepared in 13 x 100 millimeter tubes. The final activity was 235 units per milliliter which was 1/64 of the initial enzyme activity. One tenth milliliter of undiluted anti-enzyme globulin was added to each tube. Normal gamma globulin was used as a control. Phosphate buffered saline, 005 milliliter was added to another set of tubes containing enzyme as an antigen control. The volume of all tubes was then adjusted to one milliliter with saline. The reaction mixtures were incubated 1-2 hours at room temperature, over- n1Sht at 4 C, and then clarified for twenty minutes (Servall 18 33-1) at 5000 x g. The supernatant solutions were tested for the presence of glucose dehydrogenase. Agar-Gel Diffusion Studies A gel diffusion technique was deve10ped by Ouchterlony (1948) based upon the single diffusion method of Oudin (1946). The Ouchterlony procedure involves double diffusion in two dimensions. A layer of agar was poured into petri dishes and holes punched in the agar. The reagents were added to the holes and diffused from the holes, toward each other. Pre-cut Ouchterlony plates containing six center wells of five millimeters diameter and two outer rows of five wells of seven millimeters diameter were obtained from Dr. John G. Hay Jr. of Fort Detrick, Md. The gel, 12 milliliters per plate, contained 10 grams monobasic potassium phosphate, 9 grams sodium chloride, 10 grams Noble Agar (Difco), and 0.1% thimersal (E. Lilly) per liter of distilled water. Dilutions of 1:1, 1:2, 1:4, 1:8, and 1:16 of enzyme, 69,000 units per milliliter and Specific activity 150,000 were placed in the center wells of the five plates. Serial dilutions of untreated antibody gamma globulin was added to the t0p row of large wells. Serial dilutions of mercurated antibody gamma globulin were added to the lower rows. Similar control plates were prepared using normal serum gamma globulin. The plates were incubated at room temperature in a plastic bag, in a water-saturated atmosphere. The plates were observed at 24, 48, and 72 hours and the precipitation lines were photographed. l9 Electrophoresis of Gamma Globulin Fractions Electrophoretic studies were carried out on cellulose acetate strips to detect changes in the mobility of the pro- tein due to acetylation and mercuration. (Sepraphore II and Sepraphore III, Gelman Instrument Co.) The proteins were stained on the strips with nigrosin and Ponceau S as described by Consden and Kohn (1959). Ten microliters of the protein solution were applied to the strips. Owen buffer (1958) of the following composition was used: 5.0 grams Sodium diethyl barbitone 3.25 grams Sodium acetate (hydrated) 34.2 milliliters 0.1 N HCl Distilled H20 to one liter The pH of this buffer was 8.6. The potential gradient of 27 volts per centimeter was maintained. EXPERIMENTAL RESULTS Prpduction and Assay of Antibody Antibodies specific for glucose dehydrogenase were elicited in rabbits as described in materials and methods. In the initial attempts where enzymes of 2% purity was used, no detectable antibody response was elicited. The rabbits which had been injected with this protein were removed from the test group. All subsequent antigens used were at least 20% pure. These rabbits produced high titer enzyme specific globulin according to the Cohn and Torriani assay (1952) (Refer Table l) and the Ouchterlony plate method of Ray and Kadull (1964) (Refer Table 2). The titration data Show that one milligram of globulin protein, when suspended in phOSphate buffered saline, will complex with 10,600 units of glucose dehydrogenase. Mercuration of Gamma Globulin The mercuration procedures of Kulberg and Azadova (1964) were unsatisfactory for preparation of labeled gamma globulin. These authors employed approximately two thousand fold excess 0f diazotized PAPMA with the result that excessive precipita- tion of the gamma globulin occurred. A study was instituted to examine the effects of the con- centration of PAPMA on the extent of mercuration of gamma globulin. The mercuration of the gamma globulin was performed with diazotized PAPMA in molar excesses of 200, 100, 67, 40, and 29 fold. The yellow azoproteins were separated from the 20 03p mo .H8\mamawaaada o>a . do me poaoom N :oHpmapsmosoo cameosm m Spas mpappma mach Soak Saasnoam mssmm mean an finance to soapaeea .H l .oahmnm 0:9 90 hpa>apom one momdmhofla naasnoaw msamm 21 o o 50H o o 0 mo H so a :m.o o o so.a ow.o mmm ma.0 0 o mo.H Wm.w msm.a m.o o o.H . aw.o mam.o o ao.a oa.o ooonm mm»ommm No.0 maa.o o wo.H om.o 000 m mac.o o mo.a am: a:m.o o oo.a m m. mum.o o H.H oom.H s msw.o 350.0 H.H oos.m mw\w\ mm.o mmo.o oo.a com a ma cam bom.H ooH»H omuHI rbHuH N.Hj canal adamm oz saasnoaw Ngaasnoaw «88mm meandoaps4 amasmw Hoapsoo HmpHHfiHHHS Hmshoz semapdg Hem caamso namepmoap amped m>apom msHsHSSma manage no soapomam Hence mm¢zmuommwmmm mmODDQmIHBZ4 BHmmfim Mm mmdzmoommwmmm mmovaqw mo ZOHemmBHE «H mqmfiB 22 TABLE 2: TITRATION OF ANTIBODY AND ANTIGEN IN TWO-DIMENSIONAL DIFFUSION (72 HOURS) Enzyme (Specific activity 100,000) Antienzyme gamma-fraction 1:1 1:2 1:4 1:8 1:16 1:1 + + + + + 1:2 + + + + + 1:4 + + + + + 1:8 + + - - - 1:16 - — - - - Normal gamma - - - - - globulin 1 MED = 1:8 11 S 2 MED = 1 16 1 A : l. MHDS is defined as the minimal reacting dilution of serum. 2. MEDA is defined as the minimal reacting dilution of antigen. 23 reaction mixtures on G-25 Sephadex columns. The mercury content of the protein was assayed with dithizone reagent. Immunochemical prOperties of the gamma globulin were determined by the method of Cohn and Torriani (1952). Typical results are shown in figure 1. The controls were normal gamma globulin. The enzyme precipitating titer of the indivi- dual globulins can be obtained by extrapolating these curves to zero soluble enzyme. The titer is the point of inter- section of the extrapolated line and the abcissa. The effect of the concentration of PAPMA on the mercury content and the immunochemical properties of gamma globulin are shown in figure 2. It should be noted at forty-fold molar excess of PAPMA that an average of thirteen atoms of mercury are introduced into the gamma globulin molecule with concomitant reduction of precipitating activity of 60%. Increases of PAPMA to 200 molar excess increased the mercura- tion to twenty—four mercury atoms per gamma globulin with little change in precipitating activity. The yield of mercurated protein was also dependent on the concentration of PAPMA. The Optimal concentration of PAPMA for coupling is in the range of 40-67 molar excess. Table 3 presents data showing the yield of mercurated gamma globulin decreases as the molar excess of PAPMA.increases. Furthermore, the highly mercurated gamma globulins are unstable and tend to precipitate on storage. All studies requiring mercurated gamma globulin were therefore carried out with gamma globulin which had been 24 Molar i3 PAPMA 40001. 200 ‘5’ L - 28 2 23000 . 4o 2 - 67 m a: (J >2000b 0 t: 2. [- 2 1000 *- l J J 0 IOOO 2000 '3 000 4000 E NZYME ADDED -UN!TS Figure 1: The effect of concentration of PAPMA on antienzyme activity. 25 100 01 0 ‘70 Pp? Act 5~ —o— Hq/GG ~--o—- °/. Ppt Ac: L J 1 0 O ICC 200 Molar SEE PAPMA Figure 2: The effect of the concentration of PAPMA on the mercury content and immunochemical properties of gamma globulin. 26 treated with a forty-molar excess of diazotized PAPMA. The resultant protein remained in solution at least twenty-four hours at 4 C and for prolonged periods at -20 C. Determination of Changes in PrOperties of Gamma Globulin by Mercuration Spectrophotometric Studies The absorption spectra of antienzyme globulin and its carboxy methylated and mercurated derivatives were examined in an effort to determine the effects of the experimental procedures on the protein. One milliliter aliquots contain- ing 14.4 milligrams of gamma globulin were used. Absorbance values were read for each sample vs. a saline blank in a Beckman DU spectrOphotometer at increments of 5 mu from 200- 240 mu, of 2 mu from 240-310 mu, of 10 mu from 310-350 mu and of 25 mu from 350-600 mu. It can be observed in figure 3 that iodoacetate treat- ment and mercuration produce specific spectral changes. A peak occurs at 235 mu on the carboxy methylated and mercurated materials. This represents a change from -SH to S—ether linkage by carboxy methylation or modification of histadyl side chains. An absorbance peak at 340 mu in the mercurated protein is due to the presence of the azomercurial. Electrophoretic Mobility Modifications in a protein cause changes in its net charge which can be seen as changes in its electrOphoretic mobility. Mobility studies were carried out on cellulose acetate strips, Sepraphore III (Gelman Instrument Co.). At 27 TABLE 3: EFFECT OF PAPMA CONCENTRATION ON THE MERCURATION OF GAMMA GLOBULIN Molar Mercury per mole % Protein excess of gamma globulin Recovery 29 6.5 95 40 13 95 66.7 17 90 100 19 50 200 23.8 65 28 L4 1 ' ‘ T |.2 |.0 in ABSORBANCE \/\ \1 .2 b -——— GLOBULIN 7* ---- MODIFIED GLOBULIN 0 220 260 300 340 380 x - mp Figure 3: Spectral analysis of gamma globulin. Key: AcGG-carb. meth.gamma globulin, Hg-Gdeercurated gamma globulin. 29 pH 8.6, 27 volts per centimeter, it was noted that carboxy methylated gamma globulin moved at a greater rate than normal gamma globulin (Refer Table 4). The higher mobility is due to the ionization of the carboxyl groups. The intro- duction of mercury into the gamma globulin then reduced the electrophoretic mobility to a point approximately equivalent to normal gamma globulin. This was attributed to the posi- tive charge of the mercury which balanced the negative charge from the carboxyl groups. EnZyme Inhibition Modified Cohn and Torriani (1952) tests were performed on normal and enzyme Specific gamma globulins and on carboxy methylated and mercurated derivatives. The data presented in Figure 4 show that the enzyme is not precipitated by normal gamma globulin. Neither carboxy methylated nor mercurated gamma globulin precipitate the enzyme. Enzyme specific globulin precipitated glucose dehydrogenase but the precipitate does not readily dissociate in an excess of antigen. The antienzyme in this experiment removed 8,600 units of enzyme per milligram of gamma globulin. Treatment with iodoacetate produced about a 75% reduction in the precipitating titer and the antigen-antibody complex solubilized in the presence Of excess antigen. Mercuration appeared to compensate for the effect of acetylation by a two-fold increase in the precipitating ability of the anti— enzyme. The mercurated antibody-antigen complex did not become completely soluble in the presence of excess of antigen. TABLE 4: Sample Normal gamma globulin Normal, carboxy methylated gamma globulin Normal, PAPMA-gamma globulin Antienzyme gamma globulin Carboxy methylated antienzyme gamma globulin PAPMA-antienzyme gamma globulin 30 ELECTROPHORETIC MOBILITY* Average mobility 210 V/Z hrs. 2 cm volt-sec 270 v/l hr. 1.13 x 10‘5 9.3 x 10'5 19.1 .82 x 10‘5 9.8 .82 x 10‘5 10.3 10.8 967 X 10-5 9'3 X X X a Not corrected for electro-endoosmotic flow. 10"5 10"5 10'5 11.9 x 10‘5 10‘5 13.7 x 110’5 10‘5 11.3 x 10‘5 31 The antienzyme in this experiment removed 8,600 units of enzyme per milligram of gamma globulin. Treatment with iodoacetate produced about a 75% reduction in the precipitating titer and the antigen-antibody complex solubilized in the presence of excess antigen. Mercuration appeared to compensate for the effect of acetylation by a two-fold increase in the precipitating ability of the antienzyme. The mercurated antibody-antigen complex did not become completely soluble in the presence of excess Of antigen. A second examination of the effects of mercuration on the enzyme specific rabbit gamma globulin was made using the Ouchterlony immunodiffusion technique. Figure 5 and Table 5 indicate the results of these studies. NO apparent antigen- antibody reaction occurred in plates containing normal serum and enzyme. An examination of figure 5 showed both anti- enzyme and mercurated antienzyme did produce precipitation lines in the agar. It should be noted that the serum fraction contains at least two precipitating antibodies of different concentrations. The minor component disappears on dilution of the serum. The data in Table 5 show that the precipitating ability of mercurated and non-mercurated antienzyme are approximately the same. Electron Density In order for the mercurated enzyme specific globulin to be effective as a stain for electron microscopy, the product 0f its antigen-antibody reaction must be electron dense. This property was tested in the following manner. ACTIVITY - UNITS x103 nor 32 Hg-AB A859 1 4 6 8 l0 l2 ENZYME ADDED - UNITS x Io‘ Figure 4: Effects of treated and untreated normal gamma globulin and antienzyme on glucose dehydrogenase activity. Key: NG-normal gamma globulin, Ac-NG- carb.meth. normal gamma globulin, HG-NG- mercurated normal gamma globulin, AB- normal antienzyme Ac-AB-carb.meth. anti- enzyme, and Hg-Ab—mercurated antienzyme. Plate 1. Glucose dehydrogenase Plate 2. Glucose dehydrogenase in center wells, 69,000 units/ml. in center wells, 34,500 units/m1. Plate 3. Glucose dehydrogenase in center wells, 17,250 units/m1. Plate 4. Glucose dehydrogenase Plate 5. Glucose dehydrogenaSe in center wells, 8,625 units/m1. in center wells, 4,312 units/m1. Figure 5. Titration of antienzyme gamma—globulin by 2- dimensional immunodiffusion. Top rows contain (1 to r) rabbit antienzyme gamma- globulin diluted 1:1, 1:2, 1:4, 1:8, and 1:16. Bottom rows contain (1 to r) mercurated rabbit antienzyme gamma-globulin diluted 1:1, 1:2, 1:4, 1:8, and 1:16. 34 TABLE 5: BOX TITRATION OF UNTREATED AND MERCURATED ANTIENZYME VS GLUCOSE DEHYDROGENASE IN OUCHTERLONY DOUBLE DIFFUSION PLATES Enzyme Untreated 1:1 1:2 1:4 1:8 1:16 antienzyme 1:1 + + + + + 1:2 + + + + + 1:4 + + + + + 1:8 + + + - - 1:16 - - - - - Mercurated antienzyme 1:1 + + + + + 1:2 + + + + + 1:4 + + + + + 1:8 + + + + - 1:16 - - ~ - - precipitation band Key: + no precipitation :5 One tenth milliliter of PAPMA-labeled antienZyme gamma globulin was mixed in a test tube with one half milliliter of glucose dehydrogenase (5,000 units per milliliter) and droplets were immediately placed on formvar-coated electron microscope grids (Ernest F. Fullam, Inc.). The films were dried about thirty minutes, and inverted on distilled water for ten minutes. Following washing, the grids were dried for two to five hours. A similar treatment was made using non-mercurated antibody gamma globulin. Figure 6, Plate A shows an electron micrograph of enzyme-antienzyme complex using unlabeled antienzyme. Figure 6, Plate B shows an electron micrograph of antigen-antibody complex using mercury- labeled antienzyme. There are considerable electron dense areas, especially noticeable toward the borders of the complex which are not seen in Plate A. Bacillus cereus spores were fixed in 5% formaldehyde in 0.1 M phosphate buffer, pH 7.4 for one hour. The spores were. washed in buffer, shaken with Super-Brits beads in a Nossal shaker, and treated with mercurated and non-mercurated anti- body. The partially disrupted spore-antibody complexes were washed in saline and then in distilled water. They were placed on grids and treated in the same manner as previously described. Figure 7, Plate A, is an electron micrograph of a Spore Of Bacillus cereus which had been fixed, ruptured, and treated with unlabeled antienzyme. Extruded spore "cytOplasm" appears amorphous and ill-defined. Figure 7, Plate B is an electron Plate A. Enzyme-antienzyme complex. (32,000 x)— ”Plate 3. Enzyme—mercurated antienzyme complex. (38.750 X) Figure 6. Electron micrographs of aggregates of enzyme-antienzyme globulin which had been reacted one minute and dried on formvar— coated, electron microscope grids. micrc spor boun (16115 the bod 001’} E7 micrograph Of the extruded "cytoplasm" from a Bacillus cereus spore that was treated with mercurated antienzyme. The boundary of the spore material was well-defined (electron dense). Electron dense areas of 2002 diameter occurred in the "cytoplasm". If these represent the interaction of anti- body and intraSpore glucose dehydrogenase, each aggregate contains approximately 25 molecules of antibody or 425 atoms of mercury. Plate A. Disrupted Bacillus cereus spore, treated with antienzyme gamma-globulin. (63,500 x) *— ———~—~_— - ~—r_—_—~_‘._. _.__ ._._.__~_ M W -W_._' .h W Plate 3. Disrupted Bacillus cereus spore, treated with mercurated antienzyme gamma-globulin. (63,500 x) Figure 7. Electron micrographs of spores which had been disrupted in a Nossal shaker, formaldehyde-fixed, and treated with antienzyme gamma-globulins. rf‘ DISCUSSION The purpose Of the research described in this thesis was to develop an electron dense, highly specific stain which could be used to identify intracellular antigens; specifically glucose dehydrogenase of Bacillus cereus. An attempt was made to couple para-aminOphenyhmercurnz acetate r to glucose dehydrogenase-specific globulin in order to achieve these objectives. The serologic prOperties of the normal and ( mercurated antibodies were investigated to determine if mercuration changed the properties of the antienzyme. The L ability to combine with specific antigens, nonspecific bind- ing caused by mercuration, and changes in physical-chemical prOperties Of the mercurated antienzyme were investigated. The method of Cohn and Torriani (1952) was used as an assay because the Ouchterlony plates showed that more than one precipitating antibody existed in the serums. The advan- tage of the Cohn and Torriani assay was that it selected for only one antigen-antibody system. Furthermore, it was quan- titative and the results were not affected by other precipitat- ing systems. Reduced nicotinamide adenine dinucleotide (NADH) oxidase was present in rabbit serum. This reacted with the NADH generated in the course of the enzyme assay and prevented accurate antibody assays. The antibody globulin and NADH oxidase were separated by precipitation of the antibody gamma globulin with 39% (NH4)2304. This is not a specific procedure. 39 0th in 1.x" 40 Other proteins were precipitated by the salt and could account in part for the low specific activity of the gamma globulin. The Cohn and Torriani method was applied to determine the specificity of mercurated antibody. A comparison was made of the effects on the enzyme of untreated and mercurated normal gamma globulin and antienzyme and mercurated antienzyme. Neither mercurated nor untreated normal gamma globulin reacted with glucose dehydrogenase. The enzyme has been shown to be insensitive to -SH reagents and the mercury coupled to the protein therefore, did not produce a non—specific effect. Mercuration reduced the antibody titer and neither mercurated nor untreated antienzyme globulin reacted like normal rabbit antibody. That is, the insoluble rabbit antibody-antigen complex solubilizes in the presence of excess antigen. In the experiments performed, this did not appear to be the case because the antigen-antibody complex did not solubilize in the presence of excess enzyme. Either the enZyme was denatured during the solubilization or the soluble antigen-antibody complex was inactive. One reason for the inactivity of the antigen-antibody complex could be that the antibody was coupled near the active center of the enzyme. The changes caused by mercuration of the antibody could be caused by steric effects due to the addition of ten mercurated diazonium molecules to the gamma globulin. Accord- ing to the lattice theory most precipitating antibodies must have at least two sites for combination with their respective antigens. The mercury atoms or aromatic rings could prevent combination by shielding one or both of these sites. 41 Alternatively, this decrease on mercuration could be the result of a conformational change. There are three orders of structural integrity in protein molecules. “The tertiary structure probably is the level of involvment of this reaction. This is maintained co-Operatively by covalent disulfide linkages, salt bonds, hydrOgen bonding and hydro- phobic bonds. The carboxy methylation covers free sulfhydral groups which could cause a change in the configuration. Attachment of aromatic mercurials to the tyrosine and trypto- phan residues could change the tertiary structure by modifica- tion of hydrOphobic and ionic bonding at the active sites of the antibody. Spectral analyses of the untreated, carboxy methylated, and mercurated antienzyme protein was performed to verify the mercuration of the protein by changes in light absorbency. Three modifications in the Spectrum of the carboxy methylated and mercurated protein were Observed. Increased absorbency at 235 mu and 260 mu were observed in both the carboxy methy- lated and mercurated proteins and at 340 mu in the mercurated protein. The 340 mu peak was attributable to the presence of the azomercurial. The 235 mu peak was the result of iodo- acetate treatment Of the protein and the 260 mu peak was enhanced by carboxy methylation. Though iodoacetate treatment was performed under alkaline conditions and the reaction with the sulfhydral groups was rapid, it is possible that other groups such as tyrosine were carboxy methylated. These Spectral shifts prevented the use of a simple photometric assay for estimation of protein. 42 Attempts were made to fractionate mercurated antibody by column chromatography using DEAE cellulose and carboxymethyl cellulose. The proteins could not be removed from the DEAE column by change of ionic strength or pH of the eluant. This could have been due to the negative charges of the additional carboxyl groups. The cationic exchanger, carboxymethyl cellu- lose was also ineffective. This could probably by attributed to the mercury binding to the column. Therefore, another method for separation of mercurated and non-mercurated gamma globulin was sought. The electrophoretic mobility of gamma globulin and its derivatives were examined on cellulose acetate strips. Mercurated and non—mercurated gamma globulins have approxi- mately identical mobility. However, the negatively charged carboxy methylated protein had 1.3 - 2 times the mobility of the untreated gamma globulin. On the basis of this data, it was concluded that electrOphoresis could not be readily used to separate normal and mercurated gamma globulin or globulins mercurated at various levels. The average number of mercury atoms per molecule of gamma globulin was used as the criterion of degree of mercuration for all gamma globulin preparations. The primary question to be resolved for electron micro- scopy is the number of heavy atoms necessary to be visible under the electron microscope. Beer and Moudrianakes (1962) labeled the guanosine moiety of deoxyribonucleic acid (DNA) with uranyl ions. They were able to resolve individual uranyl ions in electron microscopic studies. They found that one 43 uranyl per 100 32 could be detected with ultra-high resolution electron microscOpy. These results then formed the basis for the following considerations: Bethe's Law states that the rate of energy loss in the scattering of beta-particles is proportional to the atomic number of the atom on which the electron impinges. Therefore, had Beer and Moudrianakis used mercury rather than uranium, they would have required 92 _ 0 32 (1) _80_ x 1 — 1.15 Hg atoms per 100 In Figure 7, Plate B, small uniform electron dense areas 0 of 200 A diameter were Observed. They possess a projected 6 O “(A2 and a volume of 4.2 x 10 A3. In this area of 3.14 x 10 volume, with 66; packing efficiency, 25 mercurated-antibody molecules could be fitted. This calculation is based on the assumption that gamma globulin molecules are 60 %.diameter. These aggregates therefore contain an average of 425 mercury atoms because the gamma globulin contained an average of seventeen mercury atoms per molecule. The average number of On mercury atoms per 100 A‘ would therefore be: 425 Q 02 (2) “3.14 x 102 = 1.35 Hg atoms/100 A This is within the range of concentration necessary for visualization in the electron microscOpe. The very sharp definition of the edge of the cytoplasm of those Spores treated with mercurated antibody appears to support this calculation. 4U Beer and Houdrianakis employed photographic amplifica- tions in printing technique to resolve individual atoms. The long-range objectives of this research could be fulfilled by the resolution of aggregates of mercurated antibody- antigen complex. SUNNAEY The serologic properties of mercurated glucose dehydro- genase were investigated. Enzyme Specific globulins were produced in rabbits. One milligram of globulin protein, resuspended in a serum volume of phosphate buffered saline complexes with 10,600 units of glucose dehydrOgenase. These were carboxy methylated and coupled to the diazonium derivative of para-aminophenyl-mercuric acetate. It was observed that the Optimal concentration of diazonium was forty to sixty-seven molar excess. Increasing levels of diazonium caused increas- ing levels of mercuration. However high levels of mercuration decreased the specificity and yield of antienzyme protein. A range of seven to twenty-one atoms of mercury could be attached to each molecule. The antienzyme activities of untreated, carboxy methy- lated, and mercurated antibody were compared. Normal gamma globulin treated in the same manner served as controls. The antienzyme globulin lost 80% of its original activity on iodoacetate treatment, but 50% of the activity was recovered on mercuration. Neither the normal gamma globulin nor its derivatives precipitated glucose dehydrogenase. Chromatographic and electrophoretic methods were attempted to separate antienzyme on the basis of degree of mercuration. These methods were unsuccessful. An average number of mercury atoms per molecule of gamma globulin was, therefore, used as the basis for estimations of degree of mercuration. 45 a .D val- ..n u as Antigen-antibody aggregates produced from mercurated antibody and glucose dehydrogenase were shown to be electron dense when compared to non-mercurated controls. Spores of Bacillus cereus which had been formalin-fixed and ruptured were reacted with mercurated specific globulin, untreated specific globulin, and mercurated normal gamma glob- ulin. Only mercurated enzyme specific globulin produced discrete staining of the spore cytOplasm on electron micro- scopic examination. REFERENCES Abadi, D. M., and P. E. Wilcox. 1960. Chemical derivatives of alpha-chymotrypsinogen. III. Reaction with n-acetyl- DL-homocysteine thiolactone. J. Biol. Chem. 235:396-404. Bach, J. A. 1960. The glucose dehydrogenase of Bacillus cereus: A model for the study of spore heat resistance. M. S. Thesis, Michigan State University, East Lansing. Bach, J. A. and H. L. Sadoff. 1962. Aerobic sporulating bacteria. I. Glucose dehydrogenase of Bacillus cereus. J. Bacteriol. Q3:699-707. Bach, J. A. 1963. Some properties of the glucose dehydro- genase from Spores of Bacillus cereus. P‘. D. Thesis, Michigan State University, East Lansing. Beer, Michael and Evangelos N. Moudrianakis. 1962. Deter- mination of base sequence in nucleic acids with the electron microscope: Visibility of a marker. Proc. Natl. Acad. Sci. U. s. g§:409-416o Bisset, K. A. 1950. The cytology and life-history of bacteria. E. and S. Livingstone, Ltd., Edinburgh. Church, B. D. and H. Halvorson. 1955. Glucose metabolism by resting spores of aerobic bacilli. Bacteriol. Proc. l9§§z4l. Cohn, M. and A. H. Torriani. 1952. Immunological studies with beta-galactosidase and structurally related pro- teins of Escherichia coli. J. Immunol. 62:471-491. Consden, R. and J. Kohn. 1959. Cellulose acetate as a medium for immunodiffusion. Nature 133:1512-1513. Cook, E. 8., C. W. Kreke, S. M. L. McDevitt, and S. M. . Bartlett. 1946. The action of phenylmercuric nitrate. I. Effects on enzyme systems. J. Biol. Chem. 162: 43-49 0 Coons, A. H., H. G. Creech, B. N. Jones, and E. Berliner. 1942. The demonstration of pneumococcal antigens in tissues by the use of fluorescent antibody. J. Immunol. 32:159-170. Dimroth, O. l902. Ueber die Hercurirung Aromatischer Verbindung. Ber. 32:2032-2045. Dixon, M. and E. C. Webb. 1961. Enzyme fractionation by salting-out: a theoretical note. Advan. Protein. Chem. lié:l97-219. 4? 48 D01, 3., H. Halvorson, and B. Church. 1959. Intermediate metabolism of aerobic spores. III. The mechanisms of glucose and hexose phosphate oxidation in extracts of Bacillus spores. J. Bacteriol. 11:43-54. Fireman, P., w. E. Vannier, and H. C. Goodman. 1964. Immunochemical studies of human serum fractionated by gel filtration with Sephadex G-200. Proc. Soc. Exptl. Biol. Red. 115:845-849. Flodin, P. 1962. Dextran gels and their applications in gel filtration. Dissertation, Uppsala, AB Pharmacia, Uppsala, Sweden. Freund, J. and M. V. Bonanto. 1944. The effect of paraffin oil, lanolin-like substances and killed tubercle bacilli on immunization with diphtheric toxoid and Bacterium tvnhosum. J. Immunol. 4§:325-334. Green, J. i. 1963. Comparison of soluble reduced nicotinamide adenine dinucleotide oxidases from vegetative cells and spores of Clostridium botulinum 62-A. Ph. D. Thesis. Michigan State University, East Lansing. Hanson, H. 8., V. H. Srinivasan, and H. O. Halvorson. 1963. Biochemistry of Sporulation. II. Enzymic changes dur- ing Sporulation of Bacillus cereus. J. Bacteriol. gg: 5-50. Kabat, E. A. and M. M. Mayer. 1961. Experimental Immuno- chemistry. C. C. Thomas, Springfield, Illinois. Karush, F. 1957. The role of disulfide bonds in antibody specificity. J. Am. Chem. Soc. 1255323. Kendall, P. A. 1965. Labeling of thiolated antibody with mercury for electron microscopy. Biochem. Biophys. Acta. 213174-176. Knaysi, G. 1951. Elements of Bacterial Cytology. Comstock Publ. Co., Inc. Ithaca, New York. Koehler, w. R. and B. M. Dowben. 1961. Interaction of a nonionic detergent with protein. 11. Effects on a bovine plasma albumin-azomercurial complex. Arch. Biochem. Biophys. 23:501-507. Kohn, J. 1957. An immunoelectrophoretic technique. Nature Kohn, J. 1958. A microelectrophoretic method. Nature 181: 839—840 0 Kohn, J. 1964. Technique of cellulose acetate electrophoresis. Aerztl. Lab. 1Q3232-248. / “9 / Koike, N., A. Nagayama, H. Nakayama, K. Amako, T. Natsuo, and K. Takeya. 1964. A new immuno-electron micrOSCOpe technique with use of metal chelate antibody. J. Electron Microscopy (Tokyo) 13:133-138. Kulberg: A. Y. and N. B. Azadova. 1964. Use of mercury labeled antibodies for Specific contrast in electron microscopy. Fed. Proc. Transl. Supp. 23:69-70. Landsteiner, K. and B. Gablons. 1912. Antigenic prOperties of acetylized protein. Z. Immunitat. 21:193-201. Landsteiner, K. 1914. Some protein derivatives. Biochem. [$71.0 2:2:302-3621". Landsteiner, K. and E. Prasek. 1916. Acetylated proteins. Biochem. Z. 24:388-393. Landsteiner, K. and H. Lampl. 1915. Investigations of the Specificity of serum reactions after introduction of different groups into proteins. Zentr. Physiol. 39: 329-330. Landsteiner, K. and H. Lampl. 1917. Antigenic prOperties of az0proteins. Z. Immunitat. 26:293-304. Landsteiner, K. and H. Lampl. 1918. Antigens. XII. The relationship between serological specificity and chemical structure. (Preparation of antigens with specific groups of known chemical structure.). Biochem. Z; 86:343-394. Marenzi, A. D. and L. Barrera. 1960. Serum protein fraction- ation by ammonium sulfate solution. Hev. Assoc. Bioqu1mo 44.1.30 $31.63*].680 Marrack, J. H. and E. S. Orlans. 1954. The effects of acetylation of amino groups on the reactions of antigens and antibodies. Brit. J. EXptl. Path. 23:389-401. I‘fferChant, Do Jo, Ho Ho Kahn, and. W. Ho Murphy Jr. 1964. Handbook of Cell and Organ Culture. Burgess Publ. Co., Minneapolis. foudrianakis, Evangelos N. and Michael Beer. 1965. Base sequence determination in nucleic acids with the electron microsc0pe, III. Chemistry and microscopy of guanine-labeled DNA. Proc. Natl. Acad. Sci. U. S. 22:564-571. Nichols, I. C. and H. F. Deutsch. 1948. Biophysical studies of blood plasma proteins. VII. Separation of gamma- globulin from the sera of various animals. J. Am. Chem. SOC. 19:80’83 o 5o O'Connor, 3. J., H. T. D01, and H. Halvorson. 1960. Preparation of bacterial spore extracts by use of the Eppenbach colloid mill. Can. J. Microbiol. 6:233-235. Ouchterlony, O. 1953. Antigen-antibody reactions in gels. IV. Types of reactions in co-ordinated systems of diffusion. Acta. Pathol. Microbiol. Scand. 32:231-240. Owen, J. A. 1958. Paper electrophoresis of proteins. Adv. C1111. Chemo 21:273‘300 o Pepe, F. A. 1961. Use of specific antibody in electron microscopy. I. Preparation of mercury-labeled antibody. J; Biophys. Biochem. Cytol. 11:515—520. Peterson, E. A. and H. A. Sober. 1962. Column chromatography of_proteins: substituted celluloses. p. 3-27. In S. P. Colowick and N. 0. Kaplan (ed.), Kethods in enzymalogy, vol. 5. Academic Press, Inc., New York. Baffel, S. 1963. Immunity. Appleton Century Crofts, Inc., New York. Ray, J. G. and P. J. Kadull. 1964. Agar-gel Precipitin Technique in Anthrax-antibody determinations. Appl. Microbiol. 12:349-354. Bizok, D. and J. Kallos. 1965. Introduction of mercury into active site of chymotrypsin, p-chloromercuribenzene- sulfonyl-chymotrypsin. Biochem. Biophys. Res. Commun. Bobinow, C. F. 1960. Morphology of bacterial spores, their development and germination, p. 207-248. In I. C. Gunsalus and B. Y. Stanier (ed.), The bacteria, vol. 1. Academic Press, Inc., New York. Bobinow, C. F. 1960. Outline of the visible organization of bacteria. p. 46:108. 13 Brachet and Niersky (ed.), The cell, vol. 4. Academic Press, Inc., New York. Sadoff, H. L., J. W. Kools, and H. S. Bagheb. 1959. Heat resistant catalase of Bacillus cereus. Bacteriol. Proc., 1959. 117- Setlow, Richard B. and Ernest C. Pollard. 1962. Molecular Biophysics. Addison-Wesley Publ. Co. Singer, S. J. 1959. Preparation of an electron dense antibody conjugate. Nature 18331523. Singer, 8. J. and A. F. Schick. 1960. The properties of specific stains for electron microscopy prepared by the conjugation of antibody molecules with ferritin. J. Biophys. Biochem. Cytol. 23519-537. 51 Stewart, B. T. and H. O. Halvorson. 1953. Studies on the spores of aerobic bacteria. I. The occurrence of alanine racemase. J. Bacteriol. fizzloo-léé. Stewart, B. T. and H. O. Halvorson. 1954. Studies on the spores of aerobic bacteria. II. The prOperties of an extracted heat stable enzyme. Arch. Biochem. Biophys. Harburg, O. and w. Christian. 1942. Isolierung und Kristallisation des Garungsferments Enolase. Biochem. g. 310:384-421. Wilson, G. B. and J. H. Morrison. 1961. Cytology. Reinhold Publ. Corp., New York. ICHIGRN STQTE UNIV. LIBRRRIES t | I ll llllllll H 31293010575086