a | I M H II ‘tl ‘ i 1, I t It 1 V m EDENTEFICATION 0F SALMONELLA ENTERITIDES IN SECTEONS BY FLUORESCENT ANTIBODY Thesis {or the Degree of M. S. MICHIGAN STATE UNIVERSITY Horst-Hermann Schimmelpfennig 1965 TH ESlS LIBRARY Michigan State * University __,_ 4 4 _— m ABSTRACT mENTIFICATIOfi'OF SALMONELLA mmnmxs IN SECTIONS BY FLUORESCENT ANTIBODY by Horst-Hermann Schimmelpfennig Experiments were performed in order to determine whether the fluorescent antibody technique is of value in the histologic diag- nosis of Salmonella infections. Salmonella enteritidis antiserum was prepared in rabbits and labeled with fluorescein isothiocyanate. The conjugate was purified by dialysis and Sephadex gel filtration. In addition it was absorbed with tissue powder. In order to provide a good color contrast between organisms and tissue elements, rhodamin counterstain was employed. By means of 2 reactions with unabsorbed and specifically absorbed conjugate, §, enteritidis could be identified in alcohol-fixed Paraplast- embedded tissues. Twenty mice‘were inoculated intravenously'with‘§, enteritidis. Tissues of these animals plus 5 controls were examined grossly, histo- pathologically, and by means of the fluorescent antibody technique. Salmonella enteritidis was detected in the tissues of 12 mice which died within 12 days after inoculation. With one exception §, enteritidis could not be found in 7 other inoculated mice which survived the in- fection for 14 days. From 6 animals of this group, §, enteritidis were reisolated by conventional diagnostic methods. Microscopic lesions in animals which died from the infection included necrotizing Horst-Hermann Schimmelpfennig hepatitis, splenitis, and lymphadenitis, embolism of the renal glomeruli and generalized thrombosis. There was good correlation between histo- logic lesions and demonstrable antigen and between negative results of the microsc0pic examination and those of the fluorescent antibody technique. Based on these experiments and earlier work done by the author, it is suggested that the method described should be useful in the histopathologic diagnosis of septicemic cases of salmonellosis. IDENTIFICATION OF SALMONELLA EN‘I‘ERITIDIS IN SECTIONS BY FLUORESCENT ANTIBODY By Horst-Hermann Schimmelpfennig A THESIS Submitted to Mishigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Pathology 1965 ACKNOWLEDGEMENTS The writer is highly indebted to Dr. P. R. Edwards, Miss L. M. Neu, Dr. J. P. Newman, Dr. V. H. Hallmann, Dr. J. A. Ray and, last but not least, Dr. M. A. Richardson for supplying himwwith the equip- ment and the material to do this work. It is a pleasure for him to thank Dr. C. C. Merrill, Dr. W. L. ‘Mallmann, Dr. R. F. Langham and Dr. S. D. Sleight for their guidance in writing this thesis and their constructive criticism of the‘lanu° script. The writer wishes to express his special gratitude to Dr. C. C. Morrill for offering him the opportunity to do this work. The technical assistance of Dr. T. E. Staley and Mr. Koslowski will be always appreciated. ii TABLE OF CONTENTS Page INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 REVIEW OF LITERATURE. . . . . . . . . . . . . . . . . . . . . . . . . 2 MATERIALS AND METHODS . . . . . . . . . . . . . . . . . . . . . . . . 6 cultmea O O O O O O O O O O O O O O O O O O O O O O O O O O O O 6 Antigen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Antisera . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Conjugates . . . . . .‘. . . . . . . . . . . . . . . . . . . . . 7 Mouse inoculations . . . . . . . . . . . . . . . . . . . . . . . 8 Histologic technique . . . . . . . . . . . . . . . . . . . . . . 8 Bacteriologic technique. . . . . . . . . . . . . . . . . . . . . 8 FA technique, optical equipment, photography . . . . . . . . . . 9 RESULTS . . . . e . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Preliminary experiment . . . . . . . . . . . . . . . . . . . . . 10 Main experiment. . . . . . . . . . . .i. . . . . . . . . . . . . 10 .Group I . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Group II. . . . . . . . . . . . . . . . . . . . . . . . . . 13 Group III . . . . . . . . . . . . . . . . . . . . . . . . . 16 Group IV. . . . . . . . . . . . . . . . . . . . . . . . . . 22 Group V . . . . . . . . . . . . . . . . . . . . . . . . . . 22 DISCUSSION. . . . . . . . . . . . . . . . . . s . . . . . . . . . . . 24 SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . 27 REFERENCES. . . . . . . . . . . . . . . . . . . . . e . e . . . . . . 28 VITA iii LIST OF TABLES Table Page 1 Times of death of mice inoculated with various dilutions ofa_S_. enteritidis suspension. . . . . . . . . . . . . . . . 11 2 Results of the gross, histological, FA, and bacteriological examination of 20 mice inoculated with §_. enteritidis . . . . 12 iv Figure 10 ll 12 13 LIST or FIGURES Spleen from Mouse 5, 2 days after inoculation. Necrosis of the white pulp. Hematoxylin and eosin. x 250 . . . . . Liver from Mouse 3, 1 day after inoculation. Necrosis of hepatic and Kupffer cells, early stage. Hematoxylin and .oainO x 400 O O O O O O O O O O O O O O O O O O O O O O O Kidney from Mouse 5, 2 days after inoculation. Embolism . of glomerulus. Hematoxylin and eosin. x 250 . . . . . . . Heart from Mouse 5, 2 days after inoculation. §_. enteriti- dis in capillaries of the myocardium. FA. 3 320 . . . . . Kidney from Mouse 5, 2 days after inoculation. g. enteri- tidi. in ‘ glmrUIM. FAQ I 320. O O O O O O O O O O O 0 Liver from Mouse 12, 10 days\after inoculation. Thru- bosis of a vein in the liver. Hematoxylin and eosin. x 100. O O O O O O O O O O O O O O O O O O O O O O O O O 0 Liver. from Mouse 8, 3 days after inoculation. Focal necrosis of the hepatic parenchyma. Hematoxylin and eoain. x 250 a a a a a s a a e a e e s e e a a a e s.e a 0 Liver from Mouse 9, 3 days after inoculation. Area of. destruction of the hepatic parenchyma. Hematoxylin and eosin. x 250 O 0 O O O O O O O D O O O O O O O O O O O O 0 Lung from Mouse 8, 3 days after inoculation. Thrombosis of a vein in the lung. Hematoxylin and eosin.,x 100 . . . Lung from Mouse 8, 3 days after inoculation. _S_. enteriti- dis within a thrombus of a vein in the lung. FA. 1: 320. . Liver from Mouse 8, 3 days after inoculation. .S_. enteriti- dis within the sinusoids. FA. 1: 320 . . . . . . . . . . . Spleen from Mouse 10, 4 days after inoculation. Phagocy- tosedantigen.. FA.x320................ Liver from Mouse 9, 3 days after inoculation. S. enteriti- dis Within 81DU‘01d3a. FA. x 320 a a a a a e o e s e e a e Page 14 14 15 17 17 l9 19 20 20 21 21 23 23 INTRODUCT I ON The diagnosis of infections caused by SaLmonella is based upon the isolation of the organisms in appropriate culture media followed by their biochemical and serological identification. Gross and micro- scOpic lesions in dead animals do not allow a definite diagnosis. In this paper it will be demonstrated that, by means of immunofluores- cence, §, enteritidis can be detected in tissues of experimentally infected mice. The technique described will enable the pathologist tO‘llkO an etiologically confirmed diagnosis in septicemic cases of salmonellosis. REVIEW (E LITERATURE Thomason, Cherry and Moody (1957) stained selectively O, H, and Vi antigens of Salmonella typhosa and other Salmonella species by means of fluorescent antibody (FA). The specificity of staining established by inhibition and absorption tests with FA agreed closely with that of agglutination tests for all investigated serotypes. In some strains they found differences between the results of the agglu- tination reaction and those obtained with the FA technique: 5 inagé glutinable §, typhosg strains with Vi antigen stained readily with labeled 0 antibody. Three S, bredeney strains agglutinating with IS. typhosa antiserum could not be stained with labeled 0 antiserum. With mixtures of labeled and unlabeled sera, only those antigens could be stained which did not react with unlabeled antibody. Absorption of the conjugates with heterologous antigen had no influence on the staining reactions. Thomason, Cherry and Edwards (1959) applied the FA technique as a diagnostic method for the identification of g, typhosa. Three con- jugates were used: labeled polyvalent serum containing antibodies againm:O 1 through 0 45 (Polyvalent 1), labeled polyvalent serum con- taining antibodies against the first 8 somatic 0 groups (Polyvalent 2), and labeled g, typhosg O antiserum. A large number of cross reactions with many Arizona, E, gg}i,l§. freundii and Paracolon strains was observed. These cross reactions could not be explained in all cases by the antigenic structure of the investigated strains. The number 2 3 of cross reactions was larger with sera containing a variety of anti- bodies. Fecal smears from healthy and infected persons examined with the FA technique yielded inconsistent results. In 9 out of 20 smears from persons suffering from acute gastroenteritis, fluorescent organisms were found. Only 2 of these cases could be confirmed by conventional diagnostic methods. Seventeen out of 25 stool samples from healthy persons were positive by immunofluorescence. From none of them could §,'typhosa be isolated by culture. Previous personal investigations (Schimmelpfennig, 1964) carried out with 9 0H sera and l omnivalent serum from the Behring Co. and 18 Salmonella strains of different antigenic structure confirmed the results of Thomason 55,3}, (1957, 1959). With a few exceptions the FA reactions followed the antigenic structure of the investigated strains. Omnivalent Salmonella antiserum stained all strains examined. However, there were strong cross reactions with other Gram-negative organisms, such as Arizona, Enterobacter, Shigella, Citrobacter, and some §,Iggli strains. Because of these cross reactions, it appeared to be impossible to identify Salmonella in samples of feces, meat, or sewage. I Recently Thomason, MeWhorter and Sanders (1964) described a Specific method for the identification of S, typhosa by applying labeled Vi antibody. One hundred thirty samples from carriers were examined. Using the FA technique, 68.51 of the samples were positive, whereas 69.2% positive results were obtained when conventional culo tural and serological methods were employed. Combining both methods detected more carriers than either of them individually. 4 Tanaka 25.33, (1959) investigated the localization of heat-killed S, egteritidis cells and of a chrome vaccine in mice after intravenous and intraperitoneal injections. The authors employed the indirect staining technique. In the liver the antigen of S, enteritidis was observed in the cytoplasm of the Kupffer cells and that of the leuko- cytes infiltrating around the sinusoids and other vessels. When granu- lomas were formed Salmonella antigen was demonstrated in their cells in various amounts. In the spleen, antigen was localized in the cyto- plasm of large round mononuclear cells of the red pulp. Using heat- killed cells the antigen accumulated in a few layers of cells lining the lymphatic follicles. After administering chrome vaccine, antigen was found mainly in the cells of the red pulp. Salmonella typhimurium and S, enteritidis are the 2 species most frequently found in Salmonella infections in mice. Both species cause identical clinical and pathological changes. Other species of Salmonella are less important in this regard (Dingle, 1941; Parish, 1950; Tuffery, 1956). The pathogenesis of S, typhimurium infection of white mice serves as a model for typhoid fever in man and has, therefore, been thoroughly studied (Oerskov 35.51,, 1928; Oerskov and Moltke, 1928; Seiffert $5.31., 1928; Bakken and Vogelsang, 1950). Following oral infection, the organisms first settle in the intestinal tract, from which they spread into the mesenteric lymph nodes. Lymphohematogenous spread occurs between the 2nd (Oerskov and Mbltke, 1928) and 4th days after inocu- lation (Bakken and Vogelsang, 1950). Mueller (1912) claimed that Salmonella is able to penetrate from the intestinal lumen into the capillaries of the intestinal wall, an Opinion which is not shared by L 5 other authors. Tissue damage in all infected organs is caused by the Salmonella endotoxin (Topley and Wilson, 1965). Infection of the bile ducts and gall bladder occurs after focal necrosis in the liver causes comunication between the sinusoids and the bile ducts (Oerskov St, 91..., 1928). A few days after inoculation the organisms may disappear from the intestine. A 2nd infection of the intestinal tract usually starts from the infected liver-gall bladder system. In animals which die during the acute stage of the disease the organisms can be found in all tissues. In chronic cases they often localize in a few organs, such as the intestine, gall bladder, and kidney (Topley and Wilson, 1965). In the acute infection, gross lesions include enteritis, enlarge- ment and congestion of liver and spleen, serosanguineous fluid in the peritoneal cavity, and enlarged congested lymph nodes. Subacute and chronic infections may show extreme hyperplastic enlargement of the liver, spleen and lymph nodes, findings which resemble the changes in leukemia (Loeffler, 1892; Worden, 1941, Parish, 1950). MicroscOpic manifestations of salmonellosis in mice are usually observed in the liver, Spleen, lymph nodes and intestine. The earliest lesions consist of miliary necrosis. The host reacts with a histio- cytic proliferation, the so-called Salmonella granuloma. Thrombo- phlebitis in the lungs, liver and spleen may occur. The liver and kidneys may show parenchymatous degeneration (Pallaske, 1933; Bakken t 1., 1950). MATERIAL AND METHODS Cultures Ten S, enteritidis strains were received from Dr. P. Edwards, Communicable Disease Center, Atlanta, Georgia. One S, £2l;_strain of unknown antigenic structure was obtained from Miss L. Neu, Department of Microbiology and Public Health, Michigan State University. Cultures were maintained on nutrient agar at 4 C. Antigen One S, enteritidis strain (2585 - 64) was prOpagated in nutrient broth for 4 days. Cultures were killed by heating in the autoclave for 1 hour at 104 C. After centrifugation the sediment was washed 3 trees with saline (0.85%) solution. It was brought to an optical density (OD) of 0.5, measured by a Bausch and Lamb photometer at a wave length of 600‘millnmicrons, and stored at -16 C. The S,‘gg;; strain was propagated in nutrient broth supplemented with 0.5% peptone and 1% dextrose. After good growth, cultures were autoclaved for 20 minutes at 121 C., centrifuged, washed 3 times with saline (0.852) solution, and stored at -16 C. Antisera Salmonella enteritidis antigen was given intravenously to 3 rabbits twice weekly. The dose was gradually increased from 0.2 to 1.0 ml. Animals were bled twice from the ear vein, 4 and 6 weeks reapectively after injections were started. Sera were evaluated by 6 7 the tube agglutination test using a 0.3 OD dilution of the antigen de- scribed above. Four weeks after the beginning of the immunization pro- cedure, all 3 animals had titers of 1:160‘which did not increase after further injections of antigen. After 7 weeks of immunization, the animals were exsanguinated and their sera were stored at -16 C. Con ates Antisera were pooled and labeled with fluorescein isothiocyanate (FITC).* The precipitation of the globulins and their labeling with . FITC, followed by dialysis in phosphate-buffered saline solution (0.85%, pH 7.4), were carried out according to a previously used procedure (Schimmelpfennig and Mutscherlich, 1964). In addition to dialysis, the conjugate was purified by gel filtration with Sephadex * m G 25 coarse* and absorbed twice with mouse powder. A control con- jugate was prepared by absorbing the labeled Salmonella antiglobulin twice with the homologous strain. For all absorptions the technique described by Schimmelpfennig and Mitscherlich (1964) was employed, with the exception that a highspeed centrifuge (27,000 x g) was used in place of an ultracentrifuge (272,500 x g). In order to pro? vide good suspensions 1 part of S, gglipantigen'was added to 2 parts of mouse powder. For counterstaning, Bacto FA Rhodamin counterstain*** was used. * Manufactured by Sylvana Chemical Co., Orange, N.J. ** Manufactured by Pharmacia, Uppsala, Sweden. *** Manufactured by Difco Laboratories, Detroit, Michigan. Mouse inoculations Forty-five mice were received from.a colony without a history of Salmonella infection. In a prelimingry egperiment the infectious dose that would kill 50% of a number of mice within 3 days was determined. Salmonella enteritidis was rinsed-off blood-agar plates and washed 3 times with saline (0.85%) solution. The suspension‘was diluted with saline (0.85%) solution to an optical density of 0.3. One part of the suspension was reserved for the main experiment and stored at -16 C. Tenfold dilutions were made from the other part. Twenty mice were divided into groups of 4 each. Each group was infected with l dilu° tion. Each mouse received 0.2 ml. of the diluted suspension intra- venously. A In the main experiment 25 mice were employed. To each of 20 mice the LDSO dose was given intravenously. The remaining 5 animals were used as controls. Animals that died were examined grossly, histo- pathologically and by the FA technique. Controls and mace that survived the 14th day after inoculation were killed on the 15th day and examined by the same methods as well as bacteriologically. Histolggical techniques Sections from the lungs, hearts, livers, spleens, kidneys and superficia1_inguinal lymph nodes were fixed in 952 ethyl alcohol and embedded in Paraplast.* Sections were cut at 4 to 6 microns thickness and stained with hematoxylin and eosin (H & E). Bacteriological technigues Attempts were made to reisolate S, enteritidis from all animals that survived the 14th day of the experiment. Pieces of lung, liver, w— *Manufactured by Aloe Scientific, St. Louis, Mo. 9 and kidney were inoculated into Selenite broth and incubated for 24 hours at 37 C.; 0.1 ml. of each Selenite broth tube was plated on SS agar. Lactose-negative colonies were subcultured and tested in Kligler iron agar. Strains able to ferment dextrose but not lactose, that did produce H28, were investigated biochemically. Strains that were lactose-negative, dextrose-positive, maltose-positive, mannitol- positive, sucrose-negative, methyl red-positive, indol-negative, citrate-positive, and motile were tested with a 1:10 dilution of S, enteritidis serum prepared by the author. Strains which fulfilled all biochemical requirements and did agglutinate were regarded as S, enteritidis. FA techniqueI Optical eguipggnt, photoggaphz Sections were stained in a moist chamber for 30 minutes at room temperature. The anti-Salmonella conjugate was diluted 1:2 or 1:4 with buffered saline solution (0.85%, pH 7.4), whereas the control conju- gate was used undiluted. Sections were counterstained with 1:5 saline-diluted Bacto FA.Rhodamin counterstain. After staining the sections were rinsed in phosphate-buffered saline solution (0.85%, pH 7.4) and coverslips mounted with buffered glycerol. A Zeiss Standard microscope equipped with dark-field condenser, HBO mercury arc lamp and 2 BG 12 exciter filters (3 and 4 mm.) was used for fluorescence microscopy. Barrier filters were employed in different combinations. Photographic work was done with HighSpeed Ektachrome film. Photomicrographs of H & E-stained sections were taken on ADOX KB 14 fihn. RESUDTS Prelhminagy egperiment The data obtained in the preliminary experiment to determine the LD50 of a §, enteritidis suspension are given in TABLE 1. All mice of f9 Groups I and II died within 3 days. Two animals in Group III and 2 in Group IV outlived the period of observation. In Group V, 3 mice sur- EA. vived. For the main experiment the antigen dilution of 10'4 was chosen. Main egpgriment Death rates, gross and histopathologic findings as well as the results of the bacteriologic and FA investigations are recorded in TABLE 2. Within 14 days after inoculation, a total of 13 mice died. One mouse died on the day of inoculation, 2 on the let, 3 on the 2nd, 3 on the 3rd, and 1 mouse each on the 4th, 5th, 10th, and 12th days after inoculation. The controls and 7 of the inoculated animals survived the 14th day of the experiment. Based on macrosc0pic, histologic, and FA findings, the experimental animals can be divided into 5 groups. Group I contains only Animal 1; Group II contains Animals 2 through 7; Group III contains Animals 8 through 13; Group IV contains Animals 14 through 20; and Group V con- tains all control animals (21 through 25). 10 11 TABLE 1. Times of death of mice inoculated with various dilutions of a §, enteritidis suspension Dead Animals Total of Antigen 1st 2nd 3rd Total of Surviving Group Dilution day day day Dead Animals Animals I 10"1 3 1 o 4 o 11 10"2 1 3 o 4 0 III 10'"3 o 1 1 2 2 IV 10"4 o 2 o 2 2 v 10'5 o o 1 1 3 Results of the gross, histological, FA, and bacteriological examination of 20 mice inoculated with S. enteritidis TABLE 2. Gross and Microscopic Lesions lymph nodes' Kidney Spleen Liver Heart '3sanu1 '3osg VJ srsoaosu ,‘penssflan pus peBasIug VJ ‘WSIIOQUE norasasuoo VJ SIBOQWOJQL , statuetds Surzruoaosnw pafiastug VJ sIsoqmoaql sIsoaoan Itostonu paSasIug Iatonu go msrqdaomostg u012333u03 VJ sIsoqmoaqL saBsqaaomsH uornssfiuoa u‘ 8121p1330&u fWIa } wuwmw 12 ooooooooooooo+++ I +-i:i:1:CD+-+-I ¢:+-C’l I+1Ii0 I I++¢O '++°¢¢ I I I CD1:+ I I +-CI+-+ '+iiiiiiii'i+* IiIiIOI I +'I +-I CII IiiilIl I + I I I I I l I++I+Il 'iii'i' 'i+iiiiiii'i' '++¢+++i+i+i+ 'iiiiiiiii+ii +++++ I I +-+ I % 'i+I‘ . . ' . -l'r r . I u I 'iiii' . I i‘+ I i:I +I+iii eiiiii +IIIII OHHNNNMMNQ'mS HNM¢MONw0\S 'iiiiii 'iii' I I I