._:},_E_E_v_:z,,E::355.25;: mmm . RN Q? "m: 4 J- T I 33.13 'ON ‘5‘ L M 4‘ -l S 5- "Ci ‘.I I i: .h .1‘ 1.11 1h '2 g, 1.1! Q I n 4", 3':- 9 L’; Thisistocertlfgthntthe thesis entitled Dissociation Pattern of Salmonella Pnllonm presented by Bonnie Wolfran has been accepted towards fulfillment of the requirements for 14.3. d in Bacteriology Wfl/w [Major professor 0-169 _ I —L-.-___, . --—.- -b-———__' . ..— —- :51 .. . .0135 I a fit.h-fi, a. 3.3.5.719 .1 2 ...,r. .... .z. ‘ 3.. .. £2.71. ¢ Lairfv.» .... . 11.36“. aosff THE DISSOCIATION PATTERN OF SALMONELLA PULLORUM By Bonnie Redmond Wolfram A THESIS submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in.partial fulfillment of the requirements for the degree of NMSTER OF SCIENCE Department of Bacteriology and.Public Health 1951 THESIS I ' I '-U ‘TABLE or CONTENTS Acknowledgment8..............o.....s............o..................1 I II III IntrOdHCtionooooooeooooocoooooooooooooooo0.0000000000000002 HiStorical BaCkgroundooooooooooooooooooooooocoo000000000004 Experimental Procedureaoooooooooooooooooooooooe000000000013 A. Cultures B. Media 0. Techniques R38u1t3 and Discu381onoooooooooooooooooooooooo.00.000.00.20 SWOOOOOOOOOOOO...OOOOOOOOOOOOOOOOO00......00.0.0.0..43 BibliographyOOOOOOOOOOOOOOOOOOOOOOOOCOOOOOOOOOO0.0.0.000000000000044 ‘J ACKNOWLEDGMENTS The author wishes to express her sincere thanks to Dr. H. J. Stefseth, Head and Professor of Bacteriology and Public Health, under whose guidance and supervision this investigation was undertaken. Grateful acknowledgment is due Dr. I. F. Huddleson, Professor of Bacteriology and Public Health, for his kind guidance and valuable help in checking culture plates and for taking the microphotographs. The author is greatly indebted to her family for their encouragement and assistance in making possible this work. -1.’ INTRODUCTION Since the turn of the century, study of bacterial dissociation has gained great prominence in the field of research. Enlighement concerning phases achievable by microorganisms has been sought for a multitude of reasons, some of which are: the relating of the phenomenon of dissociation to the phases of the life cycle of bac- teria, and an explanation of variant behavior of certain bacteria in biochemical activity, antigenicity and virulence. Numerous labora- tory methods have been divised that have enhanced the knowledge of microbic dissociation. Colonial appearance on solid media has for a long time been given precedence over other criteria in measuring the dissociation status of a culture. For many years, gross changes in bacterial behavior and appearance have been recognised, but few techniques have been proposed for differentiating variations that occur in minute colonial morphology within a species. in entirely new approach to the identification of bacterial species and variation is indicated by the use of the dye 2,3,5— triphenyltetrasolium chloride. Colonies grown on clear agar in which this compound has been incorporated develop color in varying shades of red, or one shade of red with different tints of pastel colors in the borders of the colony (Huddleson 1950). Because of the difficulties that have been encountered in agglutination tests with Salmonella pgllorum, this organism was -2- selected for dissociation studies in the hope that 2,3,5-triphenyl- tetrazolium chloride might assist in the recognition of slight variant changes not detectable by ordinary cultural methods. For the sake of brevity this compound will be referred to hereafter in this paper as tetrazolium. -3- HISTORICAL BACKGROUND The doctrine of variability of bacterial species had its inception before the monomorphic theory introduced by Cohn in 1872 and Koch in 1876. The principle of pleomorphism.was accepted by bacteriologists in the early days, until Cohn and Koch made known their claims that bacterial species are characterised by'a marked fixity in regard to their morphology and physiology. Because results of’much of the earlier work were established as being due to faulty techniques, the maxim of Cohn and Koch was unequivocally accepted for several generations to the extent that any bacteria which deviated from the "normal" type were considered to be contaminants. Dubos (1945) pointed out that such acceptance discouraged the study of variability of bacteria, and that it was not until the beginning of the twentieth century that any serious attempt was made to study bacterial variation. Much of the terminology used to express variation of bacteria has been borrowed from studies of higher plants and animals. The variety of names used to describe bacterial modifications include mutation, variation, saltation, dauermodifikation, anpassung, adaptation, dissociation, etc. (Dubos 1945). . Dobell (1913) stated that mutation is "any permanent change which is transmitted to subsequent generations of bacteria without any impli- cation in regard to the suddenness or gradualness of the change or the manner of its acquisition". -4— The term, "bacterial dissociation", frequently denotes a particular type of variation. The appearance in a bacterial culture of a form distinctly different from the "normal" form.and which is sufficiently stable to maintain its characters over several generations, is commonly referred to as a bacterial dissociate. Such digression from the "normal" may be in colony form, in antigenic structure, in biochemical properties, or in some other way (Topley and Wilson 1946). Under the influence of the doctrine of monomorphism, one so-called "normal" aspect of a given culture has been se- lected for description. There are unfortunately no criteria to define the "normal" aspect of a culture, and in practice bacteriologists have selected as normal the forms which most readily become acclimatised to the standard laboratory con- ditions. Mereover, for a number of irrelevant reasons, chiefly on historical grounds, the forms which are considered "typical" or "normal" vary from one species to another (Dubos 1945). Colonial morphology is most frequently described as smooth (3), rough (R), mucoid (M), diphtheroid (D) and gonidial (G). The mmooth colonies are round, convex and have an even edged margin and a smooth to glistening appearance (Dubos 1945). "Organisms belonging to the ccli-typhoid-paratyphoid-dysentery group usually are in the smooth (8) phase when isolated from the patient.“ The cells are short plump rods (Zinnser 1948) . The rough (R) colonies are larger and flatter and have varying degrees of roughness on the surface. Cells are long, filamentous and often twisted in mycelial fashion (Dubos 1945). The mucoid (M) colonies are characterized by'a viscous glistening appearance on the surface (Dubos 1945 and Zinsser 1948). Virulence is commonly associated with the smooth type colony. There are some classical exception to this, such as the anthrax bacilli and the human and bovine types of tubercule bacilli in which the rough type colony is the more virulent. Then too, one must recognize that there are degrees of virulence which may, but not neces- sarily, correspond to the degree of smoothness or roughness of a colony (Zinsser 1948). The usual succession of change in morphological characteristics is from the mucoid (M) or smooth (S) to the rough (R). This change may occur suddenly in one step or by a series of steps, which result in the appearance of intermediate colonies. Terminology describing inter- mediate colonies has not been agreed upon (Topley and Wilson 1946). Arkwright (1921) first noted that certain bacterial cultures, particularly members of the enteric group, undergo changes in colonial morphology. Such variants were characterised by the appearance of rough (R) colonies on solid media. When such colonies were grown in broth, the medium took on a granular appearance with frequent settling of sediment to the bottom of the tube. Rough colonies were also characterized by a stability in saline different from that of the smooth colonies. Autoagglutination in physiological saline has been accepted for a long time as a means of confirming the roughness of a colony. White (1927) reported that the reason for rough.bacilli having a tendency to agglutinate spontaneously in physiological saline is the presence of some lipoidal substance on the surface of the bacteria. Topley and Wilson (1946) pointed out that the extraction of the lipoidal substance with alcohol at a temperature of 50—6000. removes the salt sensitiveness of rough strains. —6-, From studies by White (1929) it appears that there is good reason for believing that the surface of the normal smooth form of the bac- terial cell contains a polysaccharide component. The change from smoothness to roughness is characterized by'a loss of this normal smooth polysaccharide antigen and other antigenic components are ex- posed. These include a polysaccharide which differs from.the one that characterizes the smooth antigen. Topley and Wilson (1946) in their discussion of variation stated that the lipoid fraction present at the surface of the cell is increased in the rough variant as compared with the polysaccharide components, thereby explaining the consequent change in colloidan behavior from hydrophile to hydrophobe. Pampana (1933) found that trypaflavine in a 1-500 dilution in physiological saline could be used to detect roughness. Edwards and Bruner (1942) stated that in their experience, “the trypaflavine test more closely follows serological behavior of S and R antigens than does any other indicator of roughness". Topley and Wilson (1946) stated that the hydrophcbe qualities of rough variants are expressed by agglutinability with salt and other non—specific agents like the dye trypaflavine. They also stated that there is not any dependable connection between this degree of agglutinability and roughness. According to Zamenkof (1946), the frequency of occurrence of a given bacteriological mutant is at the rate of one per million, and that mutant makes its appearance after a delay of a least three days following inoculation. The reasOn he gave for this delay is that time must elapse before the medium will be changed so as to become "prefer— ential' for the eventual mutant. If this change did not occur in the -7- medium to favor the divisions of the mutant,:more than the million surrounding mother type cells, a mutant would never appear. Mallmann (1932) published results of several years of work on dissociation of‘§ pullorum.and related species. Fiftyhsix cultures were studied with regard to the effect of varying conditions on the dissociation of the organisms. Colonial appearance was the original criterion for the selection of the cultures. Of the 56 cultures, 27 were characterized by their smooth colony formation, 13 by their marked rough colonial appearance, and 16 by their lack of stability in colonial appearance. Various plating media were used for determining the influence of the type of medium on the stable R and S type cultures. Smooth and intermediate strains were inoculated into a peptone meat extract medium to which a one per cent aqueous solution of brilliant green was added, and incubated for a varying number of days in order to ascertain the effect of a chemical incitant on dissociation. Rapid transferring of cultures in plain broth, i.e., twice daily for 102 days was tested as a check on Arkwright's claim (1921) that rapid transferring in a desirable medium.favored the S property of an organism at the expense of the R properties. An eight month experiment was conducted to determine growth in- fluence of rough and smooth anti-serums on broth cultures. The effect of animal passage and storage was also studied. From the results of this complete study of dissociation, Mallmann (1932) found that dissociation changes occurred only in the intermediate type cultures. Pseudo rough and smooth forms were obtained temporarily, resulting from the environmental conditions, but reversion to the original type soon took place when the particular incitant was eliminated. -8- hallmann concluded that the colony types are of a hereditary nature and not merely environmental. In most cases a change from smooth to rough form is associated with a loss of virulence as well as a change in antigenic structure. This variation consists in the "loss of the heat stable somatic antigen that characterizes the surface of the normal virulent bacterial cell". Some other somatic antigen may be exposed and a new antigenic behavior dis- closed (Topley and Wilson 1946). Bruner and Edwards (1942) in their work on serological identification of Salmonella cultures, found that colonial morphology is not always an accurate indicator of antigenic structure. Yet, prior to the time Card (1937) perfected the technique of adding agglutinating anti—serum to semi-solid agar to immObilize a dissociated phase, laborious colony selection was used for isolating a phase for serum.production. Because of the obvious hmportance of identification of the virulent phase of an organism, which is usually the smooth form, Crossley, Ferguson, and Brydson (1946) experimented with the addition of starch to culture media to see if it had any effect on reversing the smooth to rough variation of Salnonella antigens and also if it aided in preserving cultures in the smooth state. Variation was effected temporarily, but reversion took place even when cultures were stored in a cold room. These results correspond to the findings of Mallmann. A tube technique has been devised by Braun and Howell (1950) that permits automatic separation of antigenically smooth as well as various non-smooth types. The apparatus they used is a U tube that contains a fritted glass disk at the base. Broth or saline solution is pipetted -9-_ into the side arms of the tube; the unit is then sterilized by moist heat. They found that by adding as little as 0.04 ml. of a cell suspension containing both smooth and nonpsmooth types to the saline side of the tube, pure smooth cultures could be isolated from the uninoculated side 16 to 90 hours later. Only smooth types could be isolated from the uninoculated side. The reason for this separation is that the non-smooth types clump in the saline solution or the broth and clumps are unable to pass through the fritted glass disk. Smooth types do not clump, and are therefore, able to migrate through the disk. Their studies showed efficient separation with populations containing as high as ninetyanine percent nonpsmooth types. Huddleson (1950) described a new method of identifying bacterial species and variation that occur within a species using tetrazolium. This salt in its oxidised form.is colorless and becomes a red compound upon reduction. Colonies grown on clear agar to which a solution to tetrazolium chloride has been added develop a red coloring in varying degrees of intensity. Bacterial colonies exposed to this compound seemingly adhere to the pattern of a deeply dyed center varying within the limits of the red spectrum but the borders may be seen in various tints of pastel shades. Huddleson stated that the property of tetrazolium.that makes it valuable in the study of variation is its ability to color colonies either in the center, the border, or both sufficiently distinctively to permit identification of bacterial species and variations that occur within a species. Although this compound had been previously described, Huddleson was the first to make the application of its usage in bacterial differentiation. —10- Mattson, Jensen, and Dutcher (1947) reported that tetrazolium was first prepared in Germany‘by Peckman and Runge in 1894. Dutcher noted the useage of this salt as a test reagent for seed germinability when on a tour of Germany in 1945, and foresaw that the compound might have wider application as a test reagent for vitality of tissues other than seeds. This new application suggested itself to him.be- cause only those parts of seed embryos which were capable of growth were stained when exposed to the tetrazolium. ‘ Triphenyltetrazolium salts were not available in this country at the time Dutcher returned from Germany in 1945. The potentialities of this compound as a test reagent for living tissues in general, motivated Dutcher and his colleagues to synthesize the triphenyl comp pound and the furfural derivative. Mattson, Jensen, and Dutcher also reported that the colorless solution of tetrazolium chloride forms the insoluble red triphenyl formazan by the following reaction. 0 QT—Nmr 2.2.2.3., rib-N c -N= N" H’Cl’ N\\’N 2,3, 5.1-" P“"‘.‘l' tafrazolcum chloride TH plush)! forwazan From their work to determine the underlying reason for this reduction, ' they credited the enzyme systems present in viable tissue for having the power of reduction. Huddleson (1950) also indicated that color differentiation in colonies of various species of bacteria and in their -11- variants may be due to differences in the ability of bacterial enzyme systems to reduce the compound to insoluble red triphenyl formazan; this is taken up by the bacterial cells as they multiply. Huddleson (1950) reported his studies on fifteen different species or varieties of bacteria of seven different genera. Brucella cultures were purposely dissociated. Colonies of the smooth phase and related phases of the three Brucella species were found to be readily dis- tinguishable on agar medium containing tetrazolium chloride. ~12- EXPERIHEHTAL PROCEDURES Cultures S pullorum stock cultures used for study were obtained from the Poultry Diagnostic Laboratory, Department of Bacteriology and.Public Health, Michigan State College. .A 11 except one (17) of these cultures were isolated from chickens or turkeys sent in for diagnosis. Culture 17, a turkey strain, was obtained from the Bureau of Animal Industry, the U. S. Department of Agriculture and recommended for use in pullorum antigen. Four Salmonella cultures were first selected forstudyt Routine identification tests performed as part of the diagnostic procedure of the Poultry Diagnostic Laboratory were repeated before starting dis— sociation studies. These tests consisted of Gran stains, sugar fer- mentations and motility tests. Cultures 17, 963, 957 and 898 were first studied. The fermenta- tion reactions of these cultures are recorded in Table I. Two of these cultures were hydrogen sulfide producers and two were not. This varia— tion and the fermentation deviations provided an interesting aspect for study. Several weeks were spent on emperimentation with these cultures. Cultures 116, 134 and 32% were studies in detail by various techp niques in regard to dissociation. Initial steps of study were also started with cultures 331 and 3M8, but were discontinued because their variation patterns were the same as that of culture 32h. Cultures 116, . 3:4 and 3&8 were standard variants and cultures 135 and 331 were Canadian -13- variants. This was determined by the Department of Vet rinary Science, University of Massachusetts, 1951. The age of the stock cultures, except that of culture 17, was approximately three days following isolation from the infected birds. iedia The original stock cultures obtained from the Poultry Diagnostic Laboratory were lactose motility stab cultures. Difco dehydrated culture media were used throughout the experimental work, with.a few alterations and additions. Bacto Nutrient Agar was originally used for Petri plates and slants on which colonies in various nhases were seeded for making suspensions and dilutions. Use of plain nutrient agar was continued for slants on which stock reference cultures of all strains and nhase variants were kep. Slants were made with a deep butt and short slant. Bacto Nutrient Agar was rehydrated as directed by the manufacturer. 4 A substitution was made for Petri plate medium used for culture 32% from plain nutrient agar to Bacto Tryptose gar. The latter medium afforded more luxuriant growth. Tryptose Agar plates inoculated by the streak method were more successful than plain nutrient agar. Three grams per liter of Bacto Agar was added to the regular dehydrated tryptose agar to increase the agar content to 1.8 per cent, thereby providing a firmer medium for streaking. .A one per cent solution of tetrazolium was prepared.in distilled water and sterilized at 115°C. for 15 minutes. Ten m1. of this solution was added to one liter of melted tryptose agar (initially nutrient agar was used) and mixed.well just before pouring into Petri plates. The depth of the agar in the plates approximated 0.5 cm. and the plates were incup ~14- bated at 37°C for twenty-four hours to evaporate excess moisture from the surface of the agar and to assure sterility before inoculation. Bacto Tryptose Broth was used for inciting dissociation in a liquid medium. Twentybfive grams of dehydrated Bacto Tryptose Broth.wes dis- solved in 1,000 ml. of distilled water and dispensed into test tubes in approximately 15 ml. quantities. Filled, plugged tubes were sterilized .in the autoclave for 15 minutes at 15 pounds pressure (121°C). Broth tubes were incubated for he hours to ascertain sterility and refrigerated, then again incubated for 2% hours before inoculation. The fermentation base used was made up as follows: 1,000 ml. of distilled.water, 10 grams of Bacto Tryptose, five grams of sodium chloride and 10 m1. of Andrade's indicator. This base was sterilized at 15 pounds pressure (121°C) for 15 minutes. The following carbohydrates, dextrose, lactose, maltose, marrite, sucrose and dulcitol were added to the base in sufficient quantity to make a one percent solution. With the exception of maltose, sterilization was at 12 pounds for twenty minutes. Maltose was sterilized for ten minutes at 10 pounds pressure. The reagent used for checking stability of colony variants consisted of a 1:500 solution of acriflavine in physiological saline (Pampana 1933). Physiological saline was also used for determining stability of bacterial suspensions. Techniques The pattern used in the experimental work was followed with the origi- nal stock cultures and all phases isolated from those cultures. Immedi- ately upon receipt of a new culture, a transfer was made from the lactose motility stab to three plain nutrient agar slants for stock reference -15.. cultures and a streak plate made to determine the variant status of the culture to be studied. Simultaneously, a Gram stain was made and the following media inoculated: lactose motility agar, dextrose, lactose, maltose, mannite, sucrose, and dulcitol broths.r A.strip of hydrogen sulfide paper was inserted in the dextrose fermentation tube for ascer— taining HES production. Tryptose broth was inoculated from the original stock lactose motility culture. Plates were made from broth transfers at varying intervals and colony selection and isolation studies begun. New cultures and isolated colonial variants were checked for the follow- ing: stability in acriflavine and saline, cell morphology by a Gram stain and fermentation reactions. In the initial phase of the experimental work, plates were inoculated by spreading 1 m1. of bacterial suspension of varying dilutions over the surface of the Petri plate. Colony'phases were transferred to nutrient agar slants, incubated at 37°C. for forty-eight hours andéilutions made as follows: 1 ml. of suSpension, approximating in turbidity that of a tube one McFarland's nephelometer, was added to a dilution blank of sterile physiological saline making a 1-100 suspensions and serial transfers made through 1-100,000 dilutions. One ml. of each of these bacterial suSpen- sion was seeded on duplicate agar plates. The dilution technique soon became too cumbersome because of the great number of plates involved. Streak plates were substituted for phase isolation, and dilution plates made only for final colony descrip- tions. Plates used were as free as possible from irregularities and the surface of the medium was perfectly dry. A dissecting microscOpe (low power, 12X), was used when selecting colonies for transferring, and a -16- microns wire with flattened spatula end used for sicking up the material from the center of a colony. Inoculation of streak plates was performed with a bent wire 100p with care tah n not to scratch the surface of the agar. Multiple colonies of the phase to be transferred were marked on the Petri plates. One colony was used for transferring to a new plate, some for inoculating nutrient agar slants, and at least one was left on the plate for comparison with the new growth. Reference plates were refrig» erated until all comparisons were completed. Color differentiation in colonies is so important in work with tetrazolium chloride, that a stan- dard is ess ntial for classifying and describing colony phases. Several nutrient agar slants with fortyaeight hour growth of the inoculated phases were also refrigerated. These served as stock reference cultures and.were used for mixing with the bacterial growth of other phases isolated. Because of the high degree of differentiation of colony phases afforded by using tetrazolium chloride, many phases could not be distinguished by comparing colonies of one plate with those of another, or even those on the same plate, if the two phases in question were not side by side. The nutrient agar slants were used for mixing phases of close resemblance to ascertain whether the phases were identical or different. Growth on the slants was also used for inoculating carbohydrate broths, making Gram FA. stains and for checking stains and stabilit es in saline and acriflavine. Isolation of variants in a "pure" form was accomplished by continued plating and selection of colonies. Some variants were found to be a mix- ture and could not be obtained in separate phases even with continous colony selection. Work on cultures 135 and 116 was started simultaneously. -17- ;Culture 135 diSplayed the same variants as 116 and several cthers, and the number of plates needed for proner transferring of the dissociates occurring in this culture made it necessary to discontinue work with culture 116. The dilution method was used with culture 135, therefore, each plate illustrated in Tables II, III, IV and V, represent one of eight plates made with dilutions ranging from l-lOO through l-lO0,000. (Each dilution was plated on two Petri plates). Culture 32M was selected by plating lactose motility medium cultures 32%, 331 and 348. The colony phases of these three cultures appeared to be the same and this was verified by streak plates of mixtures of all cultures. The effect of bacterial growth in a liquid mediumxas determined by inoculating tryptose broth tubes with stock culture the first day study of a new strain commenced, and plating a lOOpful of broth suSpension on Petri plates at varying intervals. Colony variants, obtained on these plates, were selected and transferred in an effort to isolate them in pure colony phases. Selection and transfer of phase variants was directed toward obtaining new, not yet seen dissociates. All Petri plates were incubated at 37°C. for fortyeeight hours before growth was examined and described. The growth on the Petri plates was examined by low power (12X) of a dissecting microscOpe and a new Spencer adjustable microscopic lamp, item #66178, catalOg'# J-lSO, manufactured by the American Optical Company, was used for illumination. The light was focused at an angle of hooon the concave side of a microscOpe mirror which was placed on a wooden block -18- -— one and one-half inches from the base of th lamp. This light arrange- ment permits the rays to be rdlected at an oblique angle from the mirror and to strike the undersurface of the Petri plate. A clear glass stage made of a cleaned photographic plate was used for plate examination. The technique described by Pampana (1933) was used for determining colonial stability in acriflavine of original stock cultures and each colonial phase that was obtained in.pure form. A drOp of acriflavine sol- ution was placed on a.plain microsc0pe slide and a minute fraction of the bacterial growth to be examined placed adjacent to the drOp but not into it. The material was broken up by using a.wire 100p and gradually worked into the drOplet of acriflavine solution. The slide was examined by using the dissecting microsc0pe and the oblique reflected light set-up, used in reading Petri plates. Agglutination.within a few seconds denoted the pre- sence of a rough.variant. Stability in salt solution of original stock cultures and colonial variants was determined by washing a forty-eight hour growth of the phase being examined from a nutrient agar slant with 5 ml. of physiolOgica sa- line. This suspension was diluted to the density of Tube one McFarland's Nephelometer (300,000,000 bacterial suspension per cc.) in a clear scratch— less test tube and left undisturbed for seventy—two hours. Carbohydrate fermentation tubes were inoculated in the usual manner; however, more broth was used than ordinarily and the tubes were incubated at 37°C. for thirty days. Rubber stoppers were placed in each tube above the cotton.plugs to reduce evaporation. Fermentation reactions were checked on original stock cultures and all colonial variants obtained to see if dissociation had any effect on the biochemical prOperties of the organisms. -19- RESULTS AND DISCUSSION Tetrasolium chloride incorporated in clear agar provides a refined method for distinguishing variation within a bacterial species. Differ- entiation is made possible by color taken up by colonies. The color gbsorbed‘byl§‘pgllgggg.colonies varies fron.a rose red spot in the center of a colony with a gray opaque margin to a wine to brown.red center and irridescent margin of varying pastel tints. The difference of some colony phases was so slight that unless the two phases were side by side in the miscroscopic field, an assumption might have been.made that only one phase was present. The reverse of this was also true. Sole colonies displayed sufficient irregularities in color, consistency and shape, so that one might suppose that more than one phase was present. However, by mixing the questiOnable phases and streaking a new plate, sub— sequent growth demonstrated that the two original phases were Just one. Alterations in colony appearance, unless gross, were difficult or impossible to detect or plain agar. This trouble, familiar to all bacteriologists, enhances the value of tetrazolium as an aid to phase differentiation. Six distinct colony types were isolated through colony selection and transfer. Three of these types are pictured on Plate I. These types were designated by number and a phase description given to each. However, this was an arbitrary assignment of phases based on colony description and stability in saline and acriflavine. This suggests the need for serolo- gical studies to be correlated with tetrazolium chloride findings. Variants differed in size of colony, texture, color and contour. Description of -20- EXPLANATION OF PLATE 1 SALMONELLA PULLORUM COLONY PHASES Figure l - Smooth Phase (S) Figure 2 - Rough Phase (R2) Figure 3 - Rough Phase (R3) PLATE 1 “I: 3 Figure 2 r, «’1. Q. 2‘s r . ' " these findings are recorded in Table I. Culture 135 on original transfer to a nutrient agar tetrazolium chloride plate displayed two colony variants. Both these phases were ‘of a smooth nature but a plated mixture of the two demonstrated that they were distinct. Type 1 designated S, on subsequent transfer de- monstrated that this variant was capable of producing other than its own likeness. The plates streaked from he type 1 (S) chase resulted in colony types 1 (S), 2 (St and 3 (SR). Type 2 and type 3 colonies were respectively designated as S or SR (slightly rough or smooth rough). A new mutant, tyne M (Bi) appeared on the Petri plate streaked with the type 3 (SR) phase. Colony selection and transferring of each of types 1, 2 and h resulted in pure plates of these phases. Tgpe 3 (SR) on continued transfer, deveIOped both type 3 (SR) and type h (R1) colonies. The assumption here must be that the SR phase was at the borderline of dissociation into an.R phase, and, at that stage of develOpment, the SR variant had as part of it the R mutant. Table II. Initial transfer of three-day tryptose broth culture demonstrated the type 3 (SR) colony not seen on the first plating from lactose motility medium, in addition to type 1 (S) and type 2 (81) colonies. Plates strea — ed with type 2 (51) and type 3 (SR) had a new mutant, type 1!» (R1). SR and R phases were not obtained in pure form by subsequent transfers of selected colonies. Table III. Dissociation incited by the liquid medium had prOgressed in the five- day broth culture. The four phases (type 1, 2, 3 and M) now appeared on first transfer from the broth to a Petri plate. An additional rough phase, type 5 (R2) was demonstrated on plates streaked.with a selected type H (R1) -239 COLOE‘ TYPE I'D SALMCEELLA ULLCRUM COL HY PHASE DESCRIPTIOE CHART COLOHY PHASE SR R1 11 T3“ - I SIZE OF COLONY medium medium medium large medium medium to large -24- SIZE or MARGIN 1/3 of colony 1/2 of colony outer border 1/3 of colony inner ring 1/3 of colony 1/3 of colony 1/2 of colony 1/3 of colony TEXTURE OF MARGIN smooth, glistenp ing gelatinous smooth fine granular crusty lacy crystalline fluffy TEXTURE OF DYE CENTER smooth, gelatinous some granular, others smooth fine granular smooth.with fine dyes precipitates crystalline jelly like center to pillar growth (raspberry appearance) TABLE I (continued) pastel shades yellow to blue gray Opaque outer border pastel blue, inner border pastel yellow opalescent (milky appear- ance) irredescent irredescent COLOR or CENTER wine red rose red to deep red pinkish rose rose red deep red brown red 925- CONTOUR and RLEVATI CII circular, low convex, even edged mushroom round to slightly irre- gular, flat disc shaped circular, even edged, mushroom shape low convex, irregular shapes border low convex, elevated pillar growth center SKETCH .c: .43 ’62. .43 gal @& TABLE II COLONY PATTERN CULTURE NUMBER 135 ' 02.44200 zzzzzz . ma «$22.2 $3.50 zmmtzn. >zonoo a mnmfi. colony'but did not become stable until the inoculated broth had been incubated eleven days. Tables IV and V. Culture 116 on original transfer form lactose motility medium demon- strated two phases: type 1 (S) and type 2 (81). Further study of this strain was discontinued. The pattern of the dissociation of this culture is illustrated in Table VI. Culture 324, on first transfer from lactose motility medium, con- tained one type of colony, type 1 (S). This smooth phase was transferred and remained smooth. Table VII. Three days incubation in tryptose broth developed a second phase. It was evident that the broth was inciting variation in the type 1 (S) colony and environment favored the appearance of the type 2 (31) dissociate. Table VIII. Continued incubations in tryptose broth favored a new mutant, type 5 (R2). This was isolated in the pure phase and subsequent transfers did not indicate the appearance of additional variants. Table IX. Ten days in tryptose broth disclosed an additional phase not yet encountered. The smooth and slightly dis- sociated variants (type 1 and 2) made an appearance on the first transfer from the tenpday‘broth culture. However, there had been in the type 4 (R1) and type 5 (R2) colonies on previous plates, a suggestion that a pillar colony similar to type 6 (R3) might develop. These questionable phases remained as transferred when mixed with one another, and it was not until the streaking of the ten—day tryptose broth cultures that phase 6 (R3) appeared as a separate distinct phase, Table X. Twelve days in- cubation in tryptose was required before the rough phase only'was demon- strated, Table XI. Inasmuch as the obtaining of new variants in pure form was an objective, only type 5 (R2) and type! -ZQ. TABLE III COLONY PATTERN CULTURE NUMBER 135 “ ooooooooo «U 1 '''''''' a G) 6) Q) o 0 O G G 6) ® (3) C9 C53 0 0 O 0 Q 0 ~29- TABLE I COLONY PATTERN CULTURE NUMBER 135 U ll DAY TRYPTOSE ' BROTH INOCULATED FROM ORIGINAL CULTURE. -30- TABLE III COLONY PATTERN CULTURE NUMBER 116 .' LACTOSE MOTILITY STOCK MEDIUM (D (2) 0 0 G G) 0 0 C9 0 @ ? O , ® 0 09.9. TABLE III COLONY PATTERN CULTURE NUMBER 324 .L. L2 1? LACTOSE MOTILITY ' STOCK MEDIUM x Q, -32- TABLE 12m COLONY PATTERN CULTURE NUMBER 324 '4” 1 v 0 3 DAY TRYPTOSE BROTH. ® ® 09090 069.699 TABLE 1X COLONY PATTERN CULTURE NUMBER 324 +++++ 6 DAY TRYPTOSE IBROTQ , -34— 6 (Rs) were transferred to additional plates. The stabilities of the colony phases in saline and acriflavine correlated with the findings on Petri plates. This means of ascertain- ing the presence of an 3 mutant is not quantitative, however, it does serve to substantiate other dissociation criteria, Table III. The fermentation reactions of colony phases did not differ from those of the stock cultures. This is not to be construed as meaning that mutants of‘§ pgllorun.do not alter in their biochemical proper- ties. A complete study of all types of incited dissociates should be checked with a multitude of carbohydrates before such a statement could be made with any degree of certainty, Tables XIII and XIV. Gram stains demonstrated a change in cell morphology consistent with the alterated appearance of the colony on the Petri plate. Stains of a loopful of the cultures when first obtained showed a morphology agreeing with the text book description of the salmonellae. As dissociation occured, cell filaments appeared and colonies took on a variant appearance, Table IV. -317- TABLE XII COMPARISON OF STABILITY OF SAHIOIELIA PULLORUH COLOIY PEASES IN ACRIFLAVIITE AID SALUTE CULTURE COLOIY COLONY STABILITY IN STABILITY IN NUMBER TIPE PHASE PHYSIOLOGICAL SALIEE ACRIFLAVIHE l S Stable at 72 hours Stable Partially settled at 120 hours 2 S1 Stable at 72 hours Stable Partially settled at 120 hours 135 3 SR Stable at 72 hours Slight Settled at 120 hours agglutination u Bl Settled at 72 hours Slight agglutination 5 R2 Settled at #8 hours Instant agglutination l S Same as 135-1 Same as 135-1 32k 2 31 Same as 135—2 Same as 135—2 5 R2 Same as 135-5 Same as 135-5 6 33 Settled at an hours Instant agglutination NOTE: Saline stability determined by test tube technique. Number 1 McFarland's Nephelometer suspension of phase being examined was left undisturbed for 72 hours. Settling refers to normal sedimentation. Acriflavine stability determined.by slide technique described'by E. J. Pampana (1933) -38- .mse .E; zofifizmzmun mm £25535“. .+ “zofifizuzcuu oz .I ”5.. I I I IT I I + I o e n IT I I ® I I Q I . mm + I I ® I I ® I e ~ n I I I . ® I I ® I n n . .T I I ® I I I ® I o _ _ IT I I + I I + I o no IT I I ® I I ® I a. o e I I I I ® I I ® I n o a I I I I I I ® I a _ 29538,... m «1 8:38 3088 utzzsa 33.2: $085 mmomexma >58: $085 Sosa: #553 Am>._.m_I._. mo“. 0 on E4. 20.53302: mzzeem 228:8 sjmzozém so mzofismm 20:52”..sz EN 39: ~39- + I I ® I I we I .. . + I I ® I I ® I s . + I I ® I I ® I _. . + I I am I I ® I . _ I I I ® I I ® I .. . I I I me I I me I _. .. I I I ® I I me I ._ . . I I I ® I I me I _. . I I I ® I I me I . . + I I. me I I mm I _. . + I I ® I I ® I . . 20.5.52... m a: .5533 39.03.. 3.224: 30.54: 30.54.. umomhxuo $3.5: umoeoj umSE >233 we: >233 59.5: 95.58 320 >521... mo“. 0 {In .2 zo_.rzonoo 2:10.33 <._._mzoz._121 STAIN CULTURE COLONY COLOIY Lactose Motility SUI-BER TYPE PHASE too}: Culture 116 took Rods a Culture ' 1 (S) 2 S ( l) 135 Stock Culture 2 ($1) 3 (SR) h (R1) 5 (Re ) 32 Stock minus mms! 1 (S) 2 (51) 5 (32) 6 (33 * Rods were long and slender with slightly rounded ends. HORMAL MORPHOLOGY -41— TABLE XV (continued) APPEARALTCE or ORGAITISI-I (Gram Stain) 3 days rods"I rods with few filap ments rods‘ rods* rods"I filaments with few rods filaments filaments rods* rods* rods with few fila- ments filaments 6 days rods with few filap ments filaments rods with few filap ments filaments filaments with few rods * Rods were long and slender Tryptose Broth 10 days rods with few filaments filaments filaments rods with few filap ments rods with few filaments filaments 15 days filaments with few rods Filaments filaments filaments with few rods filaments with slightly rounded ends. IIOPJLAL ICOBPESOLOGY 42.- SUMMARY Tetrasolium.chloride salt incorporated in clear agar medium provides a new method of identifying variation within a bacterial species. This dye is colorless in its oxidised form and becomes a red compound upon reduction. Bacterial colonies exposed to this compound take on differ— ent tints of color in the colony center and margin, thereby permitting minute differentiation of colonial variants. Each phase had a distinct appearance. A line of demarcation can be seen between two colonies lo- Cated side by side that are not of the same phase. The experimental studies on variation of § pullorum.resulted in the establishment of a dissociation pattern of six distinct colony phases that ranged from smooth to rough. Fermentation studies of original cul- tures and isolated colony phases did not disclose any variation. As far as was observed, there was a noticeable correlation between changes in cellular and colonial morphology. Stability of the original cultures and isolated phases was checked using saline and acriflavine and the results corresponded to colony morphology. The Obvious importance of identification of the virulent form of an organism enhances the value of tetrazolium Chloride as an indicator of dissociation. The use of this compound, as far as was Observed, indicated that it is beneficial in determining the various phases of'§ pullorum and it is reasonable to assume that it would be valuable in selecting smooth strains for pullorum.antigen. -43.. . 5. 6. 10. ll. 13. BIBLIOGRAPHY Arkwright, J. A. 1921 as cited by D. II. Bruner and P. R. Edwards. Braun, III. and Howell, E. V. 1950 A simple method for the auto—' matic separation of smooth bacterial types from mixed populations. J. Bact. _6_Q 366-367. Crossley, V., Ferguson, M., Brydson, L. 1946 The use of soluble starch medium in the preparation of smooth "0" Salmonella Antigens. J. Bact. §_2_ 367-371. Dobell, C. .1913 as cited by Topley and G. S. Wilson. Dubos, R. J. 1945 The Bacterial Cell, ed. 1, Harvard University Press, Cambridge, mesachusetts. 48-112, 135-187. Edwards, P. R., Bruner, D. II. 1942 Serological Identification of Salmonella cultures. Kentucky Agricultural Experimental Station, College Station, Circular 54, 4-6. Card, 8. As cited by P. R. Edwards and E. W. Bruner. Huddleson, Forest I. 1950 Differentiation of bacterial species and variation within species by means of 2, 3, 5 - triphenyl— tetrazolium chloride in culture medium. Report of the School of Veterinary Medicine, Michigan State College. 50-51. Mallmann, W. L. 1932 The dissociation of Salmonella pullorum and related species. Michigan State College Agricultm'al Ex- perimental Station, Section of Bacteriology. Technical Bull. No. 122, 1—38. mttson, A. 14., Jensen, C. 0., Dutcher, R. A. 1947 Triphenyl— tetrazolium chloride as a dye for vital tissues. Sci. _]_._Q_§ 294. Pampanon, E. J. 1933 Microbic dissociation: Detection of the "R" variant by means of a specific drop agglutionation. J. Hyg. 33, 404-403. Topley, Wilson, G. S. Principles of Bacteriology and Immunity. ed. 3. The Williams and Uilkins Company, Baltimore, 288-309. White, P. B. 1929 Notes on intestional bacilli with special re- ference to smooth and rough races. J. Path. Bact. ,2, 85-94. I'll I'll I I'll I l I 3178 4444 31293 O IIIIIIIIIIIIIIIIIIII I l