Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

Localization of the guinea pig eosinophil major basic protein to the core of the granule

Localization of the guinea pig eosinophil major basic protein to the core of the granule LOCALIZATION OF THE GUINEA PIG EOSINOPHIL MAJOR BASIC PROTEIN TO THE CORE OF THE GRANULE DANIEL M. LEWIS, JON C. LEWIS, DAVID A. LOEGERING, and GERALD J. GLEICH From the Departments of Immunology, Internal Medicine, Pathology and Anatomy, Mayo Clinic and Mayo Foundation, and Mayo Medical School, Rochester, Minnesota 55901. Dr. D. M. Lewis' present address is the Department of Otolaryngology, Ohio State University Hospitals, Columbus, Ohio 43210, and Dr. J. C. Lewis' present address is the Department of Pathology, Bowman-Gray School of Medicine, Winston-Salem, North Carolina 27103. ABSTRACT The localization of the guinea pig eosinophil major basic protein (MBP) within the cell was investigated by the use of immunoelectron microscopy and by isolation of the granule crystalloids. First, by immunoperoxidase electron micros- copy, we found that the MBP of eosinophil granules is contained within the crystalloid core of the granule. Specific staining of cores was present when rabbit antiserum to MBP was used as the first stage antibody in a double antibody staining procedure, whereas staining was not seen when normal rabbit serum was used as the first stage antibody. Second, crystalloids were isolated from eosinophil granules by disruption in 0.1% Triton X-100 and centrifugation through a cushion of 50% sucrose. Highly purified core preparations yielded essentially a single band when analyzed by electrophoresis on polyacrylamide gels containing 1% sodium dodecyl sulfate (SDS). The of the core protein was 26.8 _+ 1.0 (X _+ SEM); the for the MBP was 26.3. The core protein could not be distinguished from the MBP by radioimmunoassay (RIA) and essentially all of the protein in the core preparations could be accounted for as MBP. The results indicate that the MBP is contained in the core of the guinea pig eosinophil granule and that it is probably the only protein present in the core. KEY WORDS eosinophilia basic protein these have been localized by histochemical tech- eosinophil granule crystalloid 9 niques. For example, eosinophil peroxidase and guinea pig localization acid phosphatase are located in the matrix of the primary granule (5, 7, 29), and in human eosino- phils arylsulfatase and acid phosphatase are found When examined under the electron microscope, chiefly in small granules (4, 26). The nature of the material contained within the core of the primary the granules of mature eosinophils possess a char- granule has not been identified although several acteristic morphology consisting of an electron- authors have postulated that the Charcot-Leyden dense crystalloid core surrounded by a less dense crystal could originate from the core of human matrix (21). Numerous proteins have been iso- eosinophils (6, 30). lated from eosinophil granules (1,9), and some of 701 J. CELL BIOLOGY 9 The Rockefeller University Press 9 0021-9525/78/0601-070251.00 ~lcm~l~Tc "-'Icm~l'~c by mixing equal volumes of NRS (Pelfreeze Inc., We have isolated a major basis protein (MBP) 1 Rogers, Ariz.) with 28% Na2SO4. The resulting precipi- from eosinophil granules of guinea pigs (12, 13), tate was washed twice with 14% Na2SO4, dissolved in humans (14), and rats (20), and in the guinea pig 0.01 M K2HPOcKH2PO4, pH 7.4, dialyzed, and chro- this material comprises over 50% of the protein matographed on a DEAE-Sephadex column equili- content of the granule. The MBP of the guinea brated with the phosphate buffer. The protein fraction pig has a mol wt of 11,000, a high arginine which did not absorb to the column was concentrated by content (13%), and readily aggregates through ultrafiltration to 12 mg/ml. Immunoelectrophoretic oxidation of its two free sulfhydryl groups. We analysis using a potent goat anti-rabbit whole serum and have produced an antiserum in rabbits to guinea goat anti-rabbit IgG revealed only a single band in the pig MBP and have characterized the reactivity of slow gamma region. The rabbit IgG was stored at this antiserum by quantitative microcomplement -20~ until used for either immunization or immuno- absorption procedures. fixation (19). In this report, we present results Antiserum to rabbit IgG was produced in a burro by obtained by immunoelectron microscopy and by subcutaneous injection of 1.0 mg of rabbit lgG emulsi- analyses of isolated cores which indicate that the fied in Freund's complete adjuvant. A second injection MBP is contained in the core of the eosinophil of 2.0 mg of rabbit lgG in complete adjuvant was given granule and that it is probably the only protein subcutaneously 2 wk after the first injection. 3 wk later, present in the core. the burro was serially bled and the serum was collected, pooled, and stored at -20~ MATERIALS AND METHODS For preparation of Fab-peroxidase conjugates, the procedure suggested by Kraehenbuhl and Jamieson (18) Materials was modified as follows: Burro anti-rabbit lgG was Horseradish peroxidase (HRP) (type II), o-dianisi- fractionated by mixing serum with an equal volume of dine, diaminobenzidine, crystalline human serum albu- 28% Na2SO4. The resulting precipitate was washed min, protamine sulfate, and bovine serum albumin frac- twice with 14% NazSO4, dissolved in 0.01 M K~HPO4- tion V were obtained from Sigma Chemical Co. (St. KHzPO4, 0.13 M NaCI, pH 7.5 (phosphate-buffered Louis, Mo.). Hydrogen peroxide, acetic acid, and su- saline [PBS]), and dialyzed against PBS. Burro gamma crose were obtained from Fisher Scientific Co. (Pitts- globulin was digested with papain according to the burgh, Pa.). Triton X-100 and sodium dodecyl sulfate method of Porter (28). Briefly, 1 mg of papain was (SDS) were purchased from Schwarz-Mann (Orange- added to 100 mg of burro gamma globulin (assumed burg, N. J.). Sepharose 2B, DEAE-Sephadex, and 1,~ = 14.5) in PBS containing 0.01 M cysteine and Elcm Sephadex G-50 were obtained from Pharmacia Fine 0.002 M EDTA and incubated overnight at 37~ Fab Chemicals Inc. (Piscataway, N. J.). fragments possessing antibody activity to rabbit lgG were isolated by passing the digest over an immuno- Antiserum to MBP absorbent column of purified rabbit IgG coupled to Sepharose 2B by a modification of the method of Axen The production and characterization of the antiserum et al. (3, 32). The column was eluted with 0.05 M to guinea pig MBP have been described elsewhere (19). glycine-HCl, pH 2.8, and the purified Fab fraction was Normal rabbit sera (NRS) used in the immunoelectron pooled. The purified Fab fragments were conjugated microscopy experiments were preimmunization bleed- with HRP by glutaraldehyde (2) at a ratio of 2 mg of ings from the same rabbits injected with MBP. These HRP/mg Fab protein. The HRP was dissolved in the Fab NRS lacked antibody to MBP as judged by complement pool, and 25% glutaraldehyde was added slowly with fixation assay (19). constant stirring to a final concentration of 0.45%. The mixture was allowed to react for 2 h at 23~ After Preparation of the HRP-Anti-Rabbit dialysis to remove glutaraldehyde, the conjugate was IgG Conjugate passed over the rabbit IgG immunoabsorption column a second time to remove free HRP and Fab fragments Purified rabbit immunoglobulin G (IgG) used for which had lost antibody activity during the conjugation immunization and as an immunoabsorbent was prepared procedure. The immune Fab-HRP conjugate was eluted with 0.05 M glycine HCI, pH 2.8; the pH was adjusted ~ Abbreviations used in this paper: HRP, horseradish with twice-concentrated PBS; and the conjugate was peroxidase; MBP, major basic protein; NRS, normal stored at -70~ To demonstrate that the Fab-HRP rabbit serum; PAGE, polyacrylamide gel electrophore- conjugate possessed antibody activity, dilutions of the sis; PBS, phosphate-buffered saline; Phos-BSA-P, phos- conjugate were mixed with an equal volume of a 1/20 phate buffer containing bovine serum albumin and prot- dilution of normal rabbit serum and incubated for 1 h at amine sulfate; RIA, radioimmunoassay; SDS, sodium room temperature. The mixture, presumably containing dodecyl sulfate. complexes of rabbit IgG and the univalent Fab-HRP LI~wls ~T AL. Localization of Major Basic Protein 703 conjugate, was tested by Ouchterlony analysis using goat ml of 0.25 M sucrose and washed twice. The cells were anti-rabbit IgG to precipitate the rabbit IgG-Fab HRP resuspended in 0.25 M sucrose and disrupted in a complexes. The resulting precipitin lines were stained for Tenbroeck tissue grinder (Kontes Glass Co., Evanston, peroxidase with a saturated solution of diaminobenzidine Ill.) by 10 passes of the pestle. The disrupted cells were in a pH 7.5 Tris-HCI buffer containing 0.05% H~O2. centrifuged at 400 g for 5 min and the supernate was The Fab-HRP conjugate had an antibody titer of 1/16 placed in a Tenbroeck tissue grinder, and 10% Triton by this method. Peroxidase activity associated with the X-100 was added to achieve a final concentration of Fab-HRP was measured by the rate of decomposition of 0.1%. The granules were disrupted by 10 passes of the hydrogen peroxide with o-dianisidine as the hydrogen pestle and the suspension was centrifuged at 400 g for donor, and the conjugate possessed 2.57 U/mg protein 20 min. The supernate was layered over a cushion of of enzyme activity (31). By comparison of the absorb- 50% sucrose and centrifuged at 20,000 g for 30 min in ance at 277 and 400 nm of fresh HRP to the HRP Fab a Beckman Instruments model L ultracentrifuge (Beck- conjugate, we estimated the molar ratio of HRP to Fab man Instruments, Inc., Spinco Div., Palo Alto, Calif.) protein to be 1:7 (assuming E]~ = 15 for the Fab equipped with an SW 50.1 rotor. The supernatant protein). The protein concentration of the conjugate was solution above the sucrose layer, the material at the 1.6 mg/ml. interface, and the sucrose layer were aspirated. The tube was gently rinsed with 0.25 M sucrose, and the pellet Localization of the MBP by was suspended in 0.4 ml of 0.25 M sucrose. Portions of the pellet suspension as well as the layers above were lmmunoelectron Microscopy fixed in 3% glutaraldehyde in 0.1 M KH~PO~-Na2HPO4, Eosinophils were obtained from guinea pigs by peri- pH 7.4, postfixed in 1% osmium tetroxide, dehydrated toneal lavage with saline, and granules were prepared as in graded alcohols, and embedded in Epon 812. Thin previously described (11, 13). The lavage fluids contain- sections were stained with uranyl acetate and lead citrate ing -45% eosinophils or the purified eosinophil granules before examination with the electron microscope. were fixed in 4% formalin in 0.1 M KH2PO4-Na2HPO4, pH 7.4, for 2 or 4 h at 23~ After fixation, the Physicochemical Analysis of Solubilized suspensions were centrifuged and the pellets were sus- Core Proteins pended in 0.15 M NaCI containing 10% acetic acid and Eosinophil cores were dissolved in 0.01 M HCI, 0.15 incubated overnight at 23~ The samples were washed M NaCI, and centrifuged at 20,000 g for 10 min. The thrice with PBS, divided into two aliquots, and exposed resulting solutions were analyzed for their content of to either NRS or rabbit anti-MBP, each diluted 1/10, for peroxidase by the rate of decomposition of hydrogen 1 h at 23~ After three additional washes with PBS, the peroxide with o-dianisidine as hydrogen donor (13, 31), samples were incubated with undiluted Fab HRP conju- for the molecular weights and heterogeneity of proteins gate for 1 h at room temperature. The samples were by electrophoresis in polyacrylamide gels containing 1% washed twice with PBS, twice with 0.1 M Tris-HCl, pH SDS (SDS-PAGE) (8), for the dye staining associated 7.4, and stained for peroxidase using saturated diami- with the MBP band by scanning of SDS-PAGE gels nobenzidine in Tris-HCI containing 0.5% H202 for 4 stained with Coomassie brilliant blue in a Beckman min (15). The samples were then washed twice with model 25 spectrophotometer (Beckman Instruments, Tris-HCl, and postfixed with 2% osmium tetroxide. Inc., Palo Alto, Calif.), for their protein content by a After dehydration though graded alcohols, they were scaled down biuret procedure (16), for their absorbance embedded in Epon 812. Thin sections were examined in at 277 nm using a Gilford model 252 spectrophotometer a Hitachi H U- 12 electron microscope (Hitachi American (Gilford Instrument Laboratories Inc., Oberlin, Ohio), Ltd., Indianapolis, Ind.) without further staining. This and for their content of MBP by radioimmunoassay study was conducted as a single blind study in that the (RIA) as described below. In the RIA experiments, electron microscope examination was performed on either freshly prepared core proteins were compared to coded samples and the code was not broken until all freshly prepared MBP or alkylated core proteins were specimens of an experiment were examined and the compared to alkylated MPB. Although no difference results recorded. was found between the reactivity of freshly isolated MBP and that of alkylated MBP in the RIA, nonalkylated Isolation of Cores from MBP readily polymerizes (12), and we found that stored Eosinophil Granules nonalkylated MBP which had polymerized was a less Eosinophil granules were prepared from guinea pig potent inhibitor in the RIA than alkylated MBP. There- eosinophils essentially as described previously (12, 13). fore, comparisons of solubilized core proteins to MBP Briefly, peritoneal cells were subjected to hypotonic lysis were always performed either using nonalkylated prepa- by suspension in 0.046 M sodium chloride for 1 min, rations freshly prepared on the day of the RIA experi- ment or using alkylated preparations. MBP or core after which the osmolality was reconstituted by addition protein was alkylated at pH 8 by treatment with iodo- of 10-fold concentrated PBS. After centrifugation at 400 g for 5 min, the sedimented cells were suspended in -5 acetamide, 1.7 /.~M, in the presence of 0.002 M ethyl- 704 THE JOURNAL OF CELL BIOLOGY' VOLUME 77, 1978 FIGURE 1 Purified eosinophil granules stained for peroxidase activity with diaminobenzidine. (a) (left) Granules fixed in formalin show peroxidase activity in the matrix portion of the granule, (b) (right) while granules that had been fixed in formalin and exposed to 10% acetic acid show a marked reduction in peroxidase activity. � 46,500. enediamine tetraacetate in the dark for 20 min at 20~ 10 � 75 glass tube. After incubation for 30 min at (17). 37~ and for 15 min at 4~ -2 ng of ~311-MBP was added to each tube and the solution was incubated at Radioimmunoassay for M B P 4~ overnight. The resulting immune complexes were precipitated by addition of 0.1 ml of a 1:20 dilution of This procedure will be described in detail in a normal rabbit serum in 0.1 M K2HPO4-KH2PO~ con- subsequent separate report. ~ Briefly, alkylated MBP taining 1% bovine serum albumin and 0.1 ml of burro was radioiodinated with 13~I by a modification of the anti-rabbit IgG. All solutions except the burro anti-IgG procedure described by McConahey and Dixon (23). contained 0.1% sodium azide. The tubes were mixed, After completion of the iodination reaction, a solution incubated at room temperature for 2 h, and centrifuged of 0.1 M K..,HPO4-KH~,PO~, pH 7.4, containing 1% for 20 rain at 4~ and 2,500 g. The supernates were bovine serum albumin and 0.1% protamine sulfate decanted and the sediment was suspended in 0.8 ml of (Phos-BSA-P), was added to the vial, and the contents Phos-BSA-P buffer and transferred to a fresh tube. were transferred to a 3,500 dalton cut-off dialysis After centrifugation and decantation of the supernate, casing (Spectrum Medical Industries, Inc., Los Ange- the precipitates were counted in Nuclear-Chicago les, Calif.) and dialyzed overnight. The specific activity gamma scintillation counter (G. D. Searle & Co., Des of the ~3q-MBP averaged 33 #Ci//zg, and -95% of Plaines, I11). counts was precipitated by 10% tungstic acid. The RIA was performed by additions of 0.I ml of a 1:6,000 dilution of rabbit anti-MBP in Phos-BSA-P buffer, 0.3 Statistical Analyses ml of Phos-BSA-P buffer, and 0.1 ml of inhibitor to a The results of analysis of MBP and core protein by Wassom, D. L., D. A. Loegering, and G. J. Gleich. R1A were tested by logit-log transformation of the Measurement of the guinea pig eosinophil major basic inhibition curves and by comparison of the slopes of protein by radioimmunoassay. Manuscript in prepara- the resulting regression lines using analysis of covari- tion. ance with the aid of a programmable Hewlett-Packard LEWIS ET AL. Localization of Major Basic Protein 705 F~GURE 2a Purified eosinophil granules after irnmunoperoxidase staining. On the left are granules that had been exposed to NRS as the first-stage antibody, � 28,000. 9810 A calculator (Hewlett-Packard Co., Palo Alto, tions of acetic acid and monitored their peroxi- Calif.). dase activity by staining with diaminobenzidine and by examination under the light microscope. RESULTS The peroxidase activity of the cells was dimin- ished by 3 h of incubation in 5% acetic acid in Localization of MBP by 0.15 M NaC1, while overnight incubation in 10% Immunoelectron Microscopy acetic acid almost totally abolished the peroxi- In initial experiments, we attempted to localize dase activity. In control experiments, we found the MBP within the eosinophil by immunoelec- that MBP was not extracted from formalin-fixed tron microscopy. However, before immuno- granules by acetic acid as judged by analyses peroxidase staining could be attempted, it was using SDS-PAGE and that its antigenicity was necessary to devise a fixation procedure which not reduced by exposure to 10% acetic acid as would reduce the native peroxidase activity while measured by RIA. Fig. 1 shows the comparison maintaining cellular architecture and protein an- of purified granules stained for peroxidase with tigenicity. Earlier studies had suggested that, at and without the acetic acid treatment. In the low pH, eosinophil peroxidase activity was labile untreated granules, peroxidase activity is appar- while MBP was stable (13). Therefore, we ex- ent in the matrix portion of the granule, but after posed formalin-fixed cells to various concentra- acetic acid treatment this peroxidase activity is 706 THE JOURNAL OF CELL BIOLOGY' VOLUME 77, 1978 FIGUaE 2b On the right, anti-MBP was used as the first-stage antibody. Note the deposition of peroxidase reaction product in the core of the granules on the right. � 28,000. virtually abolished. a slight deposition of peroxidase reaction product To localize the MBP, isolated granules and within the matrix of the granule. Controls in- intact eosinophils were fixed in formalin, treated cluded granules treated with NRS as the first- with 10% acetic acid, and examined by the stage antibody, granules not exposed to either immunoperoxidase staining procedure. In initial antibody, and granules that were exposed to Fab- experiments with whole cells, reaction product HRP and not washed before reaction with dia- was present over the granule core and also at the minobenzidine and H202. In the latter control, interface between the core and the matrix; no the peroxidase reaction product did not nonspe- cifically adhere to the cores, strengthening our deposition of peroxidase reaction product was seen in other areas of the eosinophil cytoplasm conclusion that the staining of the core protein or in other cell types that were present. As with anti-MBP was specific. shown in Fig. 2, electron microscope examina- To obtain a less subjective interpretation of tion of isolated granules revealed that the elec- the results, the negatives of the electron micro- tron-dense peroxidase reaction product was de- graphs were examined in a spot densitometer posited throughout the crystalline core of sam- (Macbeth TDS04 Densitometer, Macbeth Co., ples which had been exposed to anti-MBP as the Newburgh, N. Y.). The optical density of the matrix portion and the core portion of a number first-stage antibody. In contrast, granules ex- of granules was determined and the relative posed to NRS as the first-stage antibody showed LEWIS ET AL. Localization of Major Basic Protein 707 TABLE 1 Comparison of the Optical Densities of the Matrix and Core of Granules after lmmunoperoxidase Staining Optical Density Core Matrix Differencew NRS (12)* 0.926 - 0.047~: 0.853 - 0.036 -+0.073 -+ 0.033 Anti-MBP (10) 0.831 -+ 0.017 0.886 -+ 0.02 -0.055 -+ 0.015 The optical densities were performed on photographic negatives so that areas of high electron density show lower values than do areas of low electron density. * No. of granules examined. ~: Mean -- SD. w P < 0.001 byt test. difference in density between the two portions of sedimentation through a cushion of 50% sucrose. the granule was calculated. These results are After centrifugation, the pellet and supernate summarized in Table I and indicate that the core layers were analyzed by electron microscopy, by of granules stained with anti-MBP is more dense SDS-PAGE, and for peroxidase activity. The relative to the matrix, while in the case of results indicated that the supernatant material granules stained with the NRS control the core is was free of cores and contained 98% of the less dense than the matrix (P < 0.001). peroxidase activity present in the starting prepa- In another series of experiments, we measured ration, and little MBP was found by SDS-PAGE. the peroxidase activity in isolated granules after In contrast, the pellets contained many cores, exposure to antibody and Fab HRP. For these were devoid of peroxidase activity, and were experiments, the granules were carried through enriched in MBP as judged by SDS-PAGE. the staining protocol up to the reaction step with However, the pellets always contained organelles diaminobenzidine, at which point o-dianisidine other than crystalloids, and the electrophoretic was added as a hydrogen donor instead of dia- analyses showed numerous bands in addition to minobenzidine. Because of the low level of en- the MBP. Therefore, we attempted to purify zyme activity, the reaction was allowed to pro- cores using isolated eosinophil granules as the ceed for 1 h at room temperature. The granule starting material. Fig. 3 a shows a preparation of preparations were solubilized by the addition of cores obtained by centrifugation of granules after treatment with 0.1% Triton X-100 in 0.25 M SDS, final concentration 0.1%, and after centrif- ugation to remove particulate matter, the ab- sucrose. Numerous blunt rod-shaped bodies were sorbance of the supernates at 460 nm was deter- recovered which resembled granule crystalloids; mined. In two experiments, granules which had in many areas, these bodies are surrounded by been exposed to anti-MBP as the first-stage granular debris. In Fig. 3b, the core preparation antibody showed approximately three times as is shown at higher magnification, and it appears much enzyme activity as granules exposed to that much of the granular debris might be derived from the cores. In particular, one core has NRS. rounded edges and is surrounded by clumps of Isolation of Granule Cores irregular material. These experiments were re- peated on several occasions, with essentially the The experiments employing immunoelectron same results. microscropy were consistent with the view that the MBP is localized in the core of the granules. Physicochemical and Immunochemical To obtain further support for this conclusion and Analyses of Core Proteins to determine whether the MBP was the only protein present in the core, we attempted to Cores were dissolved in 0.1 M HC1, 0.15 M purify cores from eosinophil granules. In initial NaCI and analyzed for their content of MBP. In experiments, we used the approach taken by the first experiment, samples from the isolation Gessner et al. (10) in which purified eosinophils of cores shown in Fig. 3 were analyzed by SDS- are disrupted in 0.25 M sucrose containing 0.1% PAGE. As shown in Fig. 4, a band in the expected position for the MBP is present in all Triton X-100 and the cores are obtained by 708 THE JOURNAL OF CELL BIOLOGY" VOLUME 77, 1978 FIGURE 3 Electron micrographs of eosinophil granule cores. After centrifugation of disrupted granules through a cushion of 50% sucrose, the pellets were fixed with glutaraldehyde and osmium tetroxide and sections were stained with lead citrate and uranyl acetate. (a) Portion of a field showing numerous roughly rectangular bodies resembling eosinophil cores, x 10,120. (b) Higher magnification of eosinophil cores showing some with irregular edges associated with irregular material around these sites, x 51,415. samples, and it is essentially the only band in the double antibody RIA was established to identify sample of the pellet. This experiment was re- and quantitate the MBP. Utilizing this proce- peated several times, and in every case the MBP dure, we compared the inhibition produced by was the principal band. Quantitative scanning of solubilized core protein to that produced by the SDS-PAGE gels loaded with 5-20 /zg of core MBP. Alkylated MBP was used as the standard, protein revealed that between 79 and 91% (83.7 and its inhibitory ability was compared to that of -+ 5.1; 5~ ___ SD) of the dye binding was associ- alkylated core protein. As shown in Fig. 5, the ated with the band in the position of the MBP. inhibiton curves produced by these materials To determine whether the protein derived were superimposable, and statistical analysis of from the core preparations was the MBP, a the logit-log regressions did not reveal a differ- LEwls I~T AL. Localization of Major Basic Protein 709 the extinction coefficient at 277 nm of core protein should be similar to that of the MBP. The extinction coefficient, Elc~l~c, of the MBP is 26.3 by biuret analysis using human serum albu- min as a standard (12). Using the same analytical conditions in eight experiments, the extinction coefficient of the core protein was 26.8 --- 1.0 ()( - SEM). DISCUSSION Miller and his associates observed that the core of the eosinophil granule is a crystal which has a cubic lattice with a repeat of -30 /k in rodents and -40 /~ in man (24). On the basis of their observations as well as the observation by Cotran and Litt that the granule core does not possess peroxidase activity (5), we hypothesized that the MBP was derived from the granule core (13). The results presented here support this hypothe- sis. In experiments employing rabbit antibody to MBP, we found that more antibody activity was associated with granules exposed to antibody than granules exposed to NRS. This result is expected, in that our prior work had shown that the MBP was derived from the eosinophil gran- ule (12, 13). Evidence for the localization of the FIGURE 4 Analysis of eosinophil cores by SDS-PAGE. MBP within the granule was provided by immu- Eosinophil granules were disrupted by homogenization noelectron microscopy. No deposition of peroxi- in 0.25 M sucrose containing 0.1% Triton X-100, and dase reaction product was seen in other organ- cores were purified by centrifugation through a layer of elles in the eosinophil cytoplasm or in other cells, 50% sucrose. The pellet (left), interface (middle), and supernate (right) were analyzed by SDS-PAGE. These and as shown in Fig. 2, antibody to MBP was preparations were derived from the experiment shown localized to the core of the granule. The conclu- in Fig. 3. sion that the staining of the core protein was specific is further supported by the absence of ence. Thus, the core protein could not be distin- staining with the NRS control and the failure of guished immunologically from the MBP, and the peroxidase reaction product to nonspecifi- essentially all of the protein in the alkylated core cally localize in the core when excess Fab-HRP preparation could be accounted for as MBP. was not washed away before the diaminobenzi- Furthermore, alkylated core protein radiolabeled dine reaction step. with 1811 was reactive with antibody to the MBP, In another approach to the localization of the and the binding of the alkylated core protein was MBP, we isolated cores from granules and deter- inhibited equally well by the MBP and the core mined the characteristics of the proteins derived protein (results not shown). In another series of from the core. Early experiments with whole experiments, MBP and core protein were freshly eosinophils indicated that eosinophil cores were prepared and immediately analyzed by RIA. The concentrated in the pellet after centrifugation results of a typical experiment are shown in Fig. through a cushion of 50% sucrose and that the 6. The inhibition curves produced by MBP and interface layer did not contain crystalloids. How- ever, in these early experiments, the cores were core protein were virtually identical, and statisti- cal analysis of the logit-log regressions did not contaminated by considerable debris, and analy- sis of the proteins by SDS-PAGE revealed the reveal a difference. Again, essentially all of the presence of numerous proteins in addition to the protein in the solubilized core preparations could be accounted for as MBP. MBP. Therefore, we used granules as the starting Finally, if the core protein is the MBP, then material for the preparation of cores and were 710 THE JOURNAL OF CELL BIOLOGY 9 VOLUME 77, 1978 \ (o 25 I- \\. -a -,., \\ \\ i i J I l I I I I ~ o 0.5 1 2 3 4 5 6 7 8 -0.3 0 0.2 0.4 0.6 0.8 1.0 Protein, n9 Leo protein, ng FIGURE 5 Comparison of alkylated MBP with alkylated core protein by RIA. Rabbit anti-MBP was reacted with ~aq-MBP and the inhibitory potencies of alkylated MBP and alkylated core protein were tested. In the absence of inhibitor, 36% of the counts bound to antibody, whereas in the absence of antibody 4.8% of counts were precipitated. The concentration of MBP was determined by absorbance at 277 nm, and the concentration of core protein was determined by biuret analysis using human serum albumin as a standard. The results with the MBP are shown by 9 and with the core protein by x. On the left the binding curves are shown, and on the right the logit-log regressions are shown. The correlation coefficients for the MBP inhibition regression line was -0.98 and for the core protein -0.96. Comparison of the logit-log regression lines revealed that the null hypothesis of a common line could not be rejected (F2. 6 = 1.46; NS). 3O "6 -1.8 i~ 20 r~ -2.0 X .~ I I I t i I t I I I I I I I I I 0 4 8 12 16 20 0.3 0.5 0.7 0.9 1.1 1.3 Protein, ng Log protein, ng FIGURE 6 Comparison of freshly prepared MBP with freshly prepared core protein by RIA. The ability of MBP and core protein, both freshly prepared, to inhibit the binding of ~aq-MBP to antibody was tested. In the absence of inhibitor, 33% of counts bound to antibody, whereas in the absence of antibody, 7.4% of counts were precipitated. The concentrations of MBP and core protein were determined by absorbance at 277 nm and by biuret analysis, respectively. On the left the binding curves are shown, and on the right, the logit-log regressions. The results with the MBP are shown by O and with the core protein by x. The correlation coefficients for the logit-log regressions of MBP and core protein were -0.97 and -0.99, respectively. Analysis of these regression lines revealed that the null hypothesis of a common line could not be rejected (F2,4 = 0.61; NS). LEWIS ET AL. Localization of Major Basic Protein 711 We thank Dr. Harold L. Moses and Dr. Robert E. able to obtain highly purified preparations as Scott for helpful discussions of these results. Dr. Jorge shown in Fig. 3. Even the apparent debris among E. Maldonado participated in some of the early exper- the crystalloids may be core protein, as inspec- iments to purify cores from disrupted eosinophils. tion of Fig. 3 b indicates that this irregular mate- This work was supported by grants from the National rial is found around the edges of cores which are Institute of Allergy and Infectious Diseases, A1 9728 losing their characteristic shape. Numerous anal- and A1 11483, and from the Mayo Foundation. yses of the core protein by SDS-PAGE showed a Received for publication 20 September 1976, and in band in the expected position for the MBP, and revised form 24 January 1978. this accounted for up to 90% of the dye binding in the gel. More conclusive evidence that the REFERENCES crystalloid protein is the MBP was provided by 1. ARCHER, G. T., and J. G. Hmscn. 1963. Isolation the RIA experiments in which we could not show of granules from eosinophil leucocytes and study a difference between the inhibitory properties of of their enzyme content. J. Exp. Med. 118:277- the core protein and the MBP. Moreover, we could account for virtually all of the protein in 2. AVRAMEAS, S. 1969. Coupling of enzymes to the solubilized core as MBP. The results of the proteins with glutaraldehyde. Use of the conjugate two sets of experiments employing immunoelec- for the detection of antigens and antibodies. Im- tron microscopy and analysis of isolated crystal- munochemistry. 6:43-52. loids indicate that the MBP is contained within 3. AXEN, R., J. PORATH, and S. ERNBACK. 1967. the core of the eosinophil granule, and further- Chemical coupling of peptides and proteins to polysaccharides by means of cyanogen halides. more, that it appears to be the only protein in Nature (Lond.). 214:1302-1304. the crystalloid. 4. BAINTON, D. F., and M. G. FARQUHAR. 1970. Our results do not exclude the possibility sug- Segregation and packaging of granule enzymes in gested by certain authors (6, 30) that the Char- eosinophilic leukocytes. J. Cell. Biol. 45:54-73. cot-Leyden crystal protein is contained in the 5. COTRAN, R. S., and M. LITT. 1969. The entry of human eosinophil granule core. Although human granule-associated peroxidase into phagocytic vac- MBP and Charcot-Leyden crystal protein differ uoles of eosinophils. J. Exp. Med. 129:1291- in their molecular weights and amino acid com- positions and are distinct proteins (14), it is 6. EL-HASHIMI, W. 1971. Charcot-Leyden crystals. possible that the human and guinea pig eosino- Formation from primate and lack of formation phil crystalloids differ. Therefore, if the Charcot- from non-primate eosinophils. Am. J. Path. 65:311. Leyden crystal protein is derived from the core 7. ENOMOTO, T., and T. KITANI. 1966. Electron of the human eosinophil granule, then it presum- microscopic studies on peroxidase and acid phos- ably coexists there with the MBP. phatase reaction in human leukocytes. Acta Hae- Finally, Okun et al. (25) reported that the matol. Japn. 29:554-570. crystalloid core of human eosinophil granules 8. FAIRBANKS, G., T. L. STECK, and D. F. H. contained melanin. Their conclusion was based WALLACH. 1971. Electrophoretic analysis of the on results obtained with histochemical staining major polypeptides of the human erythrocyte techniques, and in particular with the Fontana- membrane. Biochemistry 10:2606-2617. Masson argentaffin stain. Because we have not 9. GESSNER, T. P., S. R. HIMMELHOCH, and E. seen pigment in any protein preparation made SHELTON. 1973. Partial characterization of the protein components of eosinophil granules isolated from purified granules, we do not believe that from guinea pig exudates. Arch. Biochem. Bio- melanin is present in guinea pig eosinophil gran- phys. 156:383-389. ules. Moreover, MacRae and Spitznagel (22) 10. GESSNER, T. P., E. SHELTON, and S. R. HIMMEL- have presented evidence that the ammoniacal HOCH. 1972. Liberation of crystalloid from eo- silver reaction, which is the basis of the Fontana- sinophil granules. Fed. Proc. 31:253 (Abstr.). Mason argentaffin stain (27), can be used as a 11. GLEICH, G. J., and D. A. LOEGERING. 1973. cytochemical marker for the detection of cationic Selective stimulation and purification of eosino- proteins by electron microscopy. Thus, the ob- phils and neutrophils from guinea pig peritoneal servations of Okun and his associates might be fluids. J. Lab. Clin. Med. 82:522-528. viewed by supportive evidence for the presence 12. GLEICrt, G. J., D. A. LOEGERING, F. KUEVPERS, S. P. BAJAJ, and K. G. MANN. 1974. Physi- of the cationic MBP in the granule core. 712 ThE JOURNAL OF CELL B1OLOCV" VOLUME 77,1978 ochemical and biological properties of the major Blood. McGraw-Hill Book Co., New York. 22. MACRAE, E. K., and J. K. SPITZNAGEL. 1975. basic protein from guinea pig eosinophil granules. UItrastructural localization of cationic proteins in J. Exp. Med. 140:313-332. cytoplasmic granules of chicken and rabbit poly- 13. GLEICH, G. J., D. A. LOEGERING, and J. E. morphonuclear leukocytes. J. Cell Sci. 17"79-94. MALDONADO. 1973. Identification of a major basic 23. McCONAHEY, P. J., and F. J. DIXON. 1966. A protein in guinea pig eosinophil granules. J. Exp. method of trace iodination of proteins for immu- Med. 137:1459-1471. nologic studies. Int. Arch. Allergy' Appl. Immunol. 14. GLEICH, G. J., D. A. LOEGERING, K. G. MANN, 29:185-189. and J. E. MALDONADO. 1976. Comparative prop- 24. MILLER, F., E, DEHARVEN, and G. E. PALADE. erties of the Charcot-Leyden crystal protein and 1966. The structure of the eosinophil leukocyte the major basic protein from human eosinophils. granules in rodents and in man. J. Cell Biol. J. Clin. Invest. 57:633-640. 31:349-362. 15. GRAHAM, R. C., JR., and M. J. KARNOVSKY. 25. OKUN, M. R., B. DONNELLAN, S. H. PERASON, 1966. The early stages of absorption of injected and L. M. EDELSTEIN. 1974. Melanin: A normal horseradish peroxidase in the proximal tubules of component of human eosinophils. Lab. Invest. mouse kidney: ultrastructural cytochemistry by a 30:681-685. new technique. J. Histochem. Cytochem. 14:291- 26. PARMLEY, R. T., and S. S. SPICER. 1974. Cyto- chemical and ultrastructural identification of a 16. KABAT, E. A., and M. N. MAYER. 1961. Estima- small type granule in human late eosinophils. Lab. tion of protein with the biuret and ninhydrin Invest. 30:557-567. reactions. In Experimental Immunochemistry. Charles C Thomas, Publisher, Springfield, Ill. 2nd 27. PEARSE, A. G. E. 1972. Histochemistry, Theoret- ical and Applied. Vol 2. Williams & Wilkins Co., edition. 559. 17. KONIGSBERG, W. 1972. Reduction of disulfide Baltimore. 3rd edition. 1485. bonds in proteins with dithiothreitol. Methods 28. PORTER, R. R. 1959. The hydrolysis of rabbit gamma globulin and antibodies with crystalline Enzymol. 25:185-188. 18. KRAEHENBUHL, J. P., and J. D. JAMIESON. 1974. papain. Biochem. J. 73:119-126. Localization of intracellular antigens by immuno- 29. SEEMAN, P. M., and G. E. PALADE. 1967. Acid electron microscopy. Int. Rev. Exp. Path. 13:1-53. phosphatase localization in rabbit eosinophils. J. 19. LEWIS, D. M., D. A. LOEGERIN6, and G. J. Cell Biol. 34:745-756. GLEICH. 1976. Antiserum to the major basic 30. WELSH, R. A. 1959. The genesis of the Charcot- Leyden crystal in the eosinophilic leukocyte of protein of eosinophil granules, lmrnunochemistry. 13:743-746. man. Am. J. Pathol. 35:1091-1104. 20. LEWIS, D. M., D. A. LOEGERING, and G. J. 31. WORTHINGTON, Enzyme manual. 1972. Worth- ington Biochemical Corporation, Freehold, N. J. GLEICH. 1976. Isolation and partial characteriza- tion of a major basic protein from rat eosinophil 32. YUNGINGER, J. W., and G. J. GLEICH. 1972. granules. Proc. Soc. Exp. Biol. Med. 152:512- Comparison of the protein binding capacities of cyanogen bromide-activated polysaccharides. J. 21. Low, F. N., and J. A. FREEMAN. 1958. Electron Microscopic Atlas of Normal and Leukemic Human Allergy Clin. Immunol. 50:109-116. LEWIS ET AL. Localization of Major Basic Protein 713 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journal of Cell Biology Pubmed Central

Localization of the guinea pig eosinophil major basic protein to the core of the granule

The Journal of Cell Biology , Volume 77 (3) – Jun 1, 1978

Loading next page...
 
/lp/pubmed-central/localization-of-the-guinea-pig-eosinophil-major-basic-protein-to-the-MnBBFoZU0A

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Pubmed Central
ISSN
0021-9525
eISSN
1540-8140
Publisher site
See Article on Publisher Site

Abstract

LOCALIZATION OF THE GUINEA PIG EOSINOPHIL MAJOR BASIC PROTEIN TO THE CORE OF THE GRANULE DANIEL M. LEWIS, JON C. LEWIS, DAVID A. LOEGERING, and GERALD J. GLEICH From the Departments of Immunology, Internal Medicine, Pathology and Anatomy, Mayo Clinic and Mayo Foundation, and Mayo Medical School, Rochester, Minnesota 55901. Dr. D. M. Lewis' present address is the Department of Otolaryngology, Ohio State University Hospitals, Columbus, Ohio 43210, and Dr. J. C. Lewis' present address is the Department of Pathology, Bowman-Gray School of Medicine, Winston-Salem, North Carolina 27103. ABSTRACT The localization of the guinea pig eosinophil major basic protein (MBP) within the cell was investigated by the use of immunoelectron microscopy and by isolation of the granule crystalloids. First, by immunoperoxidase electron micros- copy, we found that the MBP of eosinophil granules is contained within the crystalloid core of the granule. Specific staining of cores was present when rabbit antiserum to MBP was used as the first stage antibody in a double antibody staining procedure, whereas staining was not seen when normal rabbit serum was used as the first stage antibody. Second, crystalloids were isolated from eosinophil granules by disruption in 0.1% Triton X-100 and centrifugation through a cushion of 50% sucrose. Highly purified core preparations yielded essentially a single band when analyzed by electrophoresis on polyacrylamide gels containing 1% sodium dodecyl sulfate (SDS). The of the core protein was 26.8 _+ 1.0 (X _+ SEM); the for the MBP was 26.3. The core protein could not be distinguished from the MBP by radioimmunoassay (RIA) and essentially all of the protein in the core preparations could be accounted for as MBP. The results indicate that the MBP is contained in the core of the guinea pig eosinophil granule and that it is probably the only protein present in the core. KEY WORDS eosinophilia basic protein these have been localized by histochemical tech- eosinophil granule crystalloid 9 niques. For example, eosinophil peroxidase and guinea pig localization acid phosphatase are located in the matrix of the primary granule (5, 7, 29), and in human eosino- phils arylsulfatase and acid phosphatase are found When examined under the electron microscope, chiefly in small granules (4, 26). The nature of the material contained within the core of the primary the granules of mature eosinophils possess a char- granule has not been identified although several acteristic morphology consisting of an electron- authors have postulated that the Charcot-Leyden dense crystalloid core surrounded by a less dense crystal could originate from the core of human matrix (21). Numerous proteins have been iso- eosinophils (6, 30). lated from eosinophil granules (1,9), and some of 701 J. CELL BIOLOGY 9 The Rockefeller University Press 9 0021-9525/78/0601-070251.00 ~lcm~l~Tc "-'Icm~l'~c by mixing equal volumes of NRS (Pelfreeze Inc., We have isolated a major basis protein (MBP) 1 Rogers, Ariz.) with 28% Na2SO4. The resulting precipi- from eosinophil granules of guinea pigs (12, 13), tate was washed twice with 14% Na2SO4, dissolved in humans (14), and rats (20), and in the guinea pig 0.01 M K2HPOcKH2PO4, pH 7.4, dialyzed, and chro- this material comprises over 50% of the protein matographed on a DEAE-Sephadex column equili- content of the granule. The MBP of the guinea brated with the phosphate buffer. The protein fraction pig has a mol wt of 11,000, a high arginine which did not absorb to the column was concentrated by content (13%), and readily aggregates through ultrafiltration to 12 mg/ml. Immunoelectrophoretic oxidation of its two free sulfhydryl groups. We analysis using a potent goat anti-rabbit whole serum and have produced an antiserum in rabbits to guinea goat anti-rabbit IgG revealed only a single band in the pig MBP and have characterized the reactivity of slow gamma region. The rabbit IgG was stored at this antiserum by quantitative microcomplement -20~ until used for either immunization or immuno- absorption procedures. fixation (19). In this report, we present results Antiserum to rabbit IgG was produced in a burro by obtained by immunoelectron microscopy and by subcutaneous injection of 1.0 mg of rabbit lgG emulsi- analyses of isolated cores which indicate that the fied in Freund's complete adjuvant. A second injection MBP is contained in the core of the eosinophil of 2.0 mg of rabbit lgG in complete adjuvant was given granule and that it is probably the only protein subcutaneously 2 wk after the first injection. 3 wk later, present in the core. the burro was serially bled and the serum was collected, pooled, and stored at -20~ MATERIALS AND METHODS For preparation of Fab-peroxidase conjugates, the procedure suggested by Kraehenbuhl and Jamieson (18) Materials was modified as follows: Burro anti-rabbit lgG was Horseradish peroxidase (HRP) (type II), o-dianisi- fractionated by mixing serum with an equal volume of dine, diaminobenzidine, crystalline human serum albu- 28% Na2SO4. The resulting precipitate was washed min, protamine sulfate, and bovine serum albumin frac- twice with 14% NazSO4, dissolved in 0.01 M K~HPO4- tion V were obtained from Sigma Chemical Co. (St. KHzPO4, 0.13 M NaCI, pH 7.5 (phosphate-buffered Louis, Mo.). Hydrogen peroxide, acetic acid, and su- saline [PBS]), and dialyzed against PBS. Burro gamma crose were obtained from Fisher Scientific Co. (Pitts- globulin was digested with papain according to the burgh, Pa.). Triton X-100 and sodium dodecyl sulfate method of Porter (28). Briefly, 1 mg of papain was (SDS) were purchased from Schwarz-Mann (Orange- added to 100 mg of burro gamma globulin (assumed burg, N. J.). Sepharose 2B, DEAE-Sephadex, and 1,~ = 14.5) in PBS containing 0.01 M cysteine and Elcm Sephadex G-50 were obtained from Pharmacia Fine 0.002 M EDTA and incubated overnight at 37~ Fab Chemicals Inc. (Piscataway, N. J.). fragments possessing antibody activity to rabbit lgG were isolated by passing the digest over an immuno- Antiserum to MBP absorbent column of purified rabbit IgG coupled to Sepharose 2B by a modification of the method of Axen The production and characterization of the antiserum et al. (3, 32). The column was eluted with 0.05 M to guinea pig MBP have been described elsewhere (19). glycine-HCl, pH 2.8, and the purified Fab fraction was Normal rabbit sera (NRS) used in the immunoelectron pooled. The purified Fab fragments were conjugated microscopy experiments were preimmunization bleed- with HRP by glutaraldehyde (2) at a ratio of 2 mg of ings from the same rabbits injected with MBP. These HRP/mg Fab protein. The HRP was dissolved in the Fab NRS lacked antibody to MBP as judged by complement pool, and 25% glutaraldehyde was added slowly with fixation assay (19). constant stirring to a final concentration of 0.45%. The mixture was allowed to react for 2 h at 23~ After Preparation of the HRP-Anti-Rabbit dialysis to remove glutaraldehyde, the conjugate was IgG Conjugate passed over the rabbit IgG immunoabsorption column a second time to remove free HRP and Fab fragments Purified rabbit immunoglobulin G (IgG) used for which had lost antibody activity during the conjugation immunization and as an immunoabsorbent was prepared procedure. The immune Fab-HRP conjugate was eluted with 0.05 M glycine HCI, pH 2.8; the pH was adjusted ~ Abbreviations used in this paper: HRP, horseradish with twice-concentrated PBS; and the conjugate was peroxidase; MBP, major basic protein; NRS, normal stored at -70~ To demonstrate that the Fab-HRP rabbit serum; PAGE, polyacrylamide gel electrophore- conjugate possessed antibody activity, dilutions of the sis; PBS, phosphate-buffered saline; Phos-BSA-P, phos- conjugate were mixed with an equal volume of a 1/20 phate buffer containing bovine serum albumin and prot- dilution of normal rabbit serum and incubated for 1 h at amine sulfate; RIA, radioimmunoassay; SDS, sodium room temperature. The mixture, presumably containing dodecyl sulfate. complexes of rabbit IgG and the univalent Fab-HRP LI~wls ~T AL. Localization of Major Basic Protein 703 conjugate, was tested by Ouchterlony analysis using goat ml of 0.25 M sucrose and washed twice. The cells were anti-rabbit IgG to precipitate the rabbit IgG-Fab HRP resuspended in 0.25 M sucrose and disrupted in a complexes. The resulting precipitin lines were stained for Tenbroeck tissue grinder (Kontes Glass Co., Evanston, peroxidase with a saturated solution of diaminobenzidine Ill.) by 10 passes of the pestle. The disrupted cells were in a pH 7.5 Tris-HCI buffer containing 0.05% H~O2. centrifuged at 400 g for 5 min and the supernate was The Fab-HRP conjugate had an antibody titer of 1/16 placed in a Tenbroeck tissue grinder, and 10% Triton by this method. Peroxidase activity associated with the X-100 was added to achieve a final concentration of Fab-HRP was measured by the rate of decomposition of 0.1%. The granules were disrupted by 10 passes of the hydrogen peroxide with o-dianisidine as the hydrogen pestle and the suspension was centrifuged at 400 g for donor, and the conjugate possessed 2.57 U/mg protein 20 min. The supernate was layered over a cushion of of enzyme activity (31). By comparison of the absorb- 50% sucrose and centrifuged at 20,000 g for 30 min in ance at 277 and 400 nm of fresh HRP to the HRP Fab a Beckman Instruments model L ultracentrifuge (Beck- conjugate, we estimated the molar ratio of HRP to Fab man Instruments, Inc., Spinco Div., Palo Alto, Calif.) protein to be 1:7 (assuming E]~ = 15 for the Fab equipped with an SW 50.1 rotor. The supernatant protein). The protein concentration of the conjugate was solution above the sucrose layer, the material at the 1.6 mg/ml. interface, and the sucrose layer were aspirated. The tube was gently rinsed with 0.25 M sucrose, and the pellet Localization of the MBP by was suspended in 0.4 ml of 0.25 M sucrose. Portions of the pellet suspension as well as the layers above were lmmunoelectron Microscopy fixed in 3% glutaraldehyde in 0.1 M KH~PO~-Na2HPO4, Eosinophils were obtained from guinea pigs by peri- pH 7.4, postfixed in 1% osmium tetroxide, dehydrated toneal lavage with saline, and granules were prepared as in graded alcohols, and embedded in Epon 812. Thin previously described (11, 13). The lavage fluids contain- sections were stained with uranyl acetate and lead citrate ing -45% eosinophils or the purified eosinophil granules before examination with the electron microscope. were fixed in 4% formalin in 0.1 M KH2PO4-Na2HPO4, pH 7.4, for 2 or 4 h at 23~ After fixation, the Physicochemical Analysis of Solubilized suspensions were centrifuged and the pellets were sus- Core Proteins pended in 0.15 M NaCI containing 10% acetic acid and Eosinophil cores were dissolved in 0.01 M HCI, 0.15 incubated overnight at 23~ The samples were washed M NaCI, and centrifuged at 20,000 g for 10 min. The thrice with PBS, divided into two aliquots, and exposed resulting solutions were analyzed for their content of to either NRS or rabbit anti-MBP, each diluted 1/10, for peroxidase by the rate of decomposition of hydrogen 1 h at 23~ After three additional washes with PBS, the peroxide with o-dianisidine as hydrogen donor (13, 31), samples were incubated with undiluted Fab HRP conju- for the molecular weights and heterogeneity of proteins gate for 1 h at room temperature. The samples were by electrophoresis in polyacrylamide gels containing 1% washed twice with PBS, twice with 0.1 M Tris-HCl, pH SDS (SDS-PAGE) (8), for the dye staining associated 7.4, and stained for peroxidase using saturated diami- with the MBP band by scanning of SDS-PAGE gels nobenzidine in Tris-HCI containing 0.5% H202 for 4 stained with Coomassie brilliant blue in a Beckman min (15). The samples were then washed twice with model 25 spectrophotometer (Beckman Instruments, Tris-HCl, and postfixed with 2% osmium tetroxide. Inc., Palo Alto, Calif.), for their protein content by a After dehydration though graded alcohols, they were scaled down biuret procedure (16), for their absorbance embedded in Epon 812. Thin sections were examined in at 277 nm using a Gilford model 252 spectrophotometer a Hitachi H U- 12 electron microscope (Hitachi American (Gilford Instrument Laboratories Inc., Oberlin, Ohio), Ltd., Indianapolis, Ind.) without further staining. This and for their content of MBP by radioimmunoassay study was conducted as a single blind study in that the (RIA) as described below. In the RIA experiments, electron microscope examination was performed on either freshly prepared core proteins were compared to coded samples and the code was not broken until all freshly prepared MBP or alkylated core proteins were specimens of an experiment were examined and the compared to alkylated MPB. Although no difference results recorded. was found between the reactivity of freshly isolated MBP and that of alkylated MBP in the RIA, nonalkylated Isolation of Cores from MBP readily polymerizes (12), and we found that stored Eosinophil Granules nonalkylated MBP which had polymerized was a less Eosinophil granules were prepared from guinea pig potent inhibitor in the RIA than alkylated MBP. There- eosinophils essentially as described previously (12, 13). fore, comparisons of solubilized core proteins to MBP Briefly, peritoneal cells were subjected to hypotonic lysis were always performed either using nonalkylated prepa- by suspension in 0.046 M sodium chloride for 1 min, rations freshly prepared on the day of the RIA experi- ment or using alkylated preparations. MBP or core after which the osmolality was reconstituted by addition protein was alkylated at pH 8 by treatment with iodo- of 10-fold concentrated PBS. After centrifugation at 400 g for 5 min, the sedimented cells were suspended in -5 acetamide, 1.7 /.~M, in the presence of 0.002 M ethyl- 704 THE JOURNAL OF CELL BIOLOGY' VOLUME 77, 1978 FIGURE 1 Purified eosinophil granules stained for peroxidase activity with diaminobenzidine. (a) (left) Granules fixed in formalin show peroxidase activity in the matrix portion of the granule, (b) (right) while granules that had been fixed in formalin and exposed to 10% acetic acid show a marked reduction in peroxidase activity. � 46,500. enediamine tetraacetate in the dark for 20 min at 20~ 10 � 75 glass tube. After incubation for 30 min at (17). 37~ and for 15 min at 4~ -2 ng of ~311-MBP was added to each tube and the solution was incubated at Radioimmunoassay for M B P 4~ overnight. The resulting immune complexes were precipitated by addition of 0.1 ml of a 1:20 dilution of This procedure will be described in detail in a normal rabbit serum in 0.1 M K2HPO4-KH2PO~ con- subsequent separate report. ~ Briefly, alkylated MBP taining 1% bovine serum albumin and 0.1 ml of burro was radioiodinated with 13~I by a modification of the anti-rabbit IgG. All solutions except the burro anti-IgG procedure described by McConahey and Dixon (23). contained 0.1% sodium azide. The tubes were mixed, After completion of the iodination reaction, a solution incubated at room temperature for 2 h, and centrifuged of 0.1 M K..,HPO4-KH~,PO~, pH 7.4, containing 1% for 20 rain at 4~ and 2,500 g. The supernates were bovine serum albumin and 0.1% protamine sulfate decanted and the sediment was suspended in 0.8 ml of (Phos-BSA-P), was added to the vial, and the contents Phos-BSA-P buffer and transferred to a fresh tube. were transferred to a 3,500 dalton cut-off dialysis After centrifugation and decantation of the supernate, casing (Spectrum Medical Industries, Inc., Los Ange- the precipitates were counted in Nuclear-Chicago les, Calif.) and dialyzed overnight. The specific activity gamma scintillation counter (G. D. Searle & Co., Des of the ~3q-MBP averaged 33 #Ci//zg, and -95% of Plaines, I11). counts was precipitated by 10% tungstic acid. The RIA was performed by additions of 0.I ml of a 1:6,000 dilution of rabbit anti-MBP in Phos-BSA-P buffer, 0.3 Statistical Analyses ml of Phos-BSA-P buffer, and 0.1 ml of inhibitor to a The results of analysis of MBP and core protein by Wassom, D. L., D. A. Loegering, and G. J. Gleich. R1A were tested by logit-log transformation of the Measurement of the guinea pig eosinophil major basic inhibition curves and by comparison of the slopes of protein by radioimmunoassay. Manuscript in prepara- the resulting regression lines using analysis of covari- tion. ance with the aid of a programmable Hewlett-Packard LEWIS ET AL. Localization of Major Basic Protein 705 F~GURE 2a Purified eosinophil granules after irnmunoperoxidase staining. On the left are granules that had been exposed to NRS as the first-stage antibody, � 28,000. 9810 A calculator (Hewlett-Packard Co., Palo Alto, tions of acetic acid and monitored their peroxi- Calif.). dase activity by staining with diaminobenzidine and by examination under the light microscope. RESULTS The peroxidase activity of the cells was dimin- ished by 3 h of incubation in 5% acetic acid in Localization of MBP by 0.15 M NaC1, while overnight incubation in 10% Immunoelectron Microscopy acetic acid almost totally abolished the peroxi- In initial experiments, we attempted to localize dase activity. In control experiments, we found the MBP within the eosinophil by immunoelec- that MBP was not extracted from formalin-fixed tron microscopy. However, before immuno- granules by acetic acid as judged by analyses peroxidase staining could be attempted, it was using SDS-PAGE and that its antigenicity was necessary to devise a fixation procedure which not reduced by exposure to 10% acetic acid as would reduce the native peroxidase activity while measured by RIA. Fig. 1 shows the comparison maintaining cellular architecture and protein an- of purified granules stained for peroxidase with tigenicity. Earlier studies had suggested that, at and without the acetic acid treatment. In the low pH, eosinophil peroxidase activity was labile untreated granules, peroxidase activity is appar- while MBP was stable (13). Therefore, we ex- ent in the matrix portion of the granule, but after posed formalin-fixed cells to various concentra- acetic acid treatment this peroxidase activity is 706 THE JOURNAL OF CELL BIOLOGY' VOLUME 77, 1978 FIGUaE 2b On the right, anti-MBP was used as the first-stage antibody. Note the deposition of peroxidase reaction product in the core of the granules on the right. � 28,000. virtually abolished. a slight deposition of peroxidase reaction product To localize the MBP, isolated granules and within the matrix of the granule. Controls in- intact eosinophils were fixed in formalin, treated cluded granules treated with NRS as the first- with 10% acetic acid, and examined by the stage antibody, granules not exposed to either immunoperoxidase staining procedure. In initial antibody, and granules that were exposed to Fab- experiments with whole cells, reaction product HRP and not washed before reaction with dia- was present over the granule core and also at the minobenzidine and H202. In the latter control, interface between the core and the matrix; no the peroxidase reaction product did not nonspe- cifically adhere to the cores, strengthening our deposition of peroxidase reaction product was seen in other areas of the eosinophil cytoplasm conclusion that the staining of the core protein or in other cell types that were present. As with anti-MBP was specific. shown in Fig. 2, electron microscope examina- To obtain a less subjective interpretation of tion of isolated granules revealed that the elec- the results, the negatives of the electron micro- tron-dense peroxidase reaction product was de- graphs were examined in a spot densitometer posited throughout the crystalline core of sam- (Macbeth TDS04 Densitometer, Macbeth Co., ples which had been exposed to anti-MBP as the Newburgh, N. Y.). The optical density of the matrix portion and the core portion of a number first-stage antibody. In contrast, granules ex- of granules was determined and the relative posed to NRS as the first-stage antibody showed LEWIS ET AL. Localization of Major Basic Protein 707 TABLE 1 Comparison of the Optical Densities of the Matrix and Core of Granules after lmmunoperoxidase Staining Optical Density Core Matrix Differencew NRS (12)* 0.926 - 0.047~: 0.853 - 0.036 -+0.073 -+ 0.033 Anti-MBP (10) 0.831 -+ 0.017 0.886 -+ 0.02 -0.055 -+ 0.015 The optical densities were performed on photographic negatives so that areas of high electron density show lower values than do areas of low electron density. * No. of granules examined. ~: Mean -- SD. w P < 0.001 byt test. difference in density between the two portions of sedimentation through a cushion of 50% sucrose. the granule was calculated. These results are After centrifugation, the pellet and supernate summarized in Table I and indicate that the core layers were analyzed by electron microscopy, by of granules stained with anti-MBP is more dense SDS-PAGE, and for peroxidase activity. The relative to the matrix, while in the case of results indicated that the supernatant material granules stained with the NRS control the core is was free of cores and contained 98% of the less dense than the matrix (P < 0.001). peroxidase activity present in the starting prepa- In another series of experiments, we measured ration, and little MBP was found by SDS-PAGE. the peroxidase activity in isolated granules after In contrast, the pellets contained many cores, exposure to antibody and Fab HRP. For these were devoid of peroxidase activity, and were experiments, the granules were carried through enriched in MBP as judged by SDS-PAGE. the staining protocol up to the reaction step with However, the pellets always contained organelles diaminobenzidine, at which point o-dianisidine other than crystalloids, and the electrophoretic was added as a hydrogen donor instead of dia- analyses showed numerous bands in addition to minobenzidine. Because of the low level of en- the MBP. Therefore, we attempted to purify zyme activity, the reaction was allowed to pro- cores using isolated eosinophil granules as the ceed for 1 h at room temperature. The granule starting material. Fig. 3 a shows a preparation of preparations were solubilized by the addition of cores obtained by centrifugation of granules after treatment with 0.1% Triton X-100 in 0.25 M SDS, final concentration 0.1%, and after centrif- ugation to remove particulate matter, the ab- sucrose. Numerous blunt rod-shaped bodies were sorbance of the supernates at 460 nm was deter- recovered which resembled granule crystalloids; mined. In two experiments, granules which had in many areas, these bodies are surrounded by been exposed to anti-MBP as the first-stage granular debris. In Fig. 3b, the core preparation antibody showed approximately three times as is shown at higher magnification, and it appears much enzyme activity as granules exposed to that much of the granular debris might be derived from the cores. In particular, one core has NRS. rounded edges and is surrounded by clumps of Isolation of Granule Cores irregular material. These experiments were re- peated on several occasions, with essentially the The experiments employing immunoelectron same results. microscropy were consistent with the view that the MBP is localized in the core of the granules. Physicochemical and Immunochemical To obtain further support for this conclusion and Analyses of Core Proteins to determine whether the MBP was the only protein present in the core, we attempted to Cores were dissolved in 0.1 M HC1, 0.15 M purify cores from eosinophil granules. In initial NaCI and analyzed for their content of MBP. In experiments, we used the approach taken by the first experiment, samples from the isolation Gessner et al. (10) in which purified eosinophils of cores shown in Fig. 3 were analyzed by SDS- are disrupted in 0.25 M sucrose containing 0.1% PAGE. As shown in Fig. 4, a band in the expected position for the MBP is present in all Triton X-100 and the cores are obtained by 708 THE JOURNAL OF CELL BIOLOGY" VOLUME 77, 1978 FIGURE 3 Electron micrographs of eosinophil granule cores. After centrifugation of disrupted granules through a cushion of 50% sucrose, the pellets were fixed with glutaraldehyde and osmium tetroxide and sections were stained with lead citrate and uranyl acetate. (a) Portion of a field showing numerous roughly rectangular bodies resembling eosinophil cores, x 10,120. (b) Higher magnification of eosinophil cores showing some with irregular edges associated with irregular material around these sites, x 51,415. samples, and it is essentially the only band in the double antibody RIA was established to identify sample of the pellet. This experiment was re- and quantitate the MBP. Utilizing this proce- peated several times, and in every case the MBP dure, we compared the inhibition produced by was the principal band. Quantitative scanning of solubilized core protein to that produced by the SDS-PAGE gels loaded with 5-20 /zg of core MBP. Alkylated MBP was used as the standard, protein revealed that between 79 and 91% (83.7 and its inhibitory ability was compared to that of -+ 5.1; 5~ ___ SD) of the dye binding was associ- alkylated core protein. As shown in Fig. 5, the ated with the band in the position of the MBP. inhibiton curves produced by these materials To determine whether the protein derived were superimposable, and statistical analysis of from the core preparations was the MBP, a the logit-log regressions did not reveal a differ- LEwls I~T AL. Localization of Major Basic Protein 709 the extinction coefficient at 277 nm of core protein should be similar to that of the MBP. The extinction coefficient, Elc~l~c, of the MBP is 26.3 by biuret analysis using human serum albu- min as a standard (12). Using the same analytical conditions in eight experiments, the extinction coefficient of the core protein was 26.8 --- 1.0 ()( - SEM). DISCUSSION Miller and his associates observed that the core of the eosinophil granule is a crystal which has a cubic lattice with a repeat of -30 /k in rodents and -40 /~ in man (24). On the basis of their observations as well as the observation by Cotran and Litt that the granule core does not possess peroxidase activity (5), we hypothesized that the MBP was derived from the granule core (13). The results presented here support this hypothe- sis. In experiments employing rabbit antibody to MBP, we found that more antibody activity was associated with granules exposed to antibody than granules exposed to NRS. This result is expected, in that our prior work had shown that the MBP was derived from the eosinophil gran- ule (12, 13). Evidence for the localization of the FIGURE 4 Analysis of eosinophil cores by SDS-PAGE. MBP within the granule was provided by immu- Eosinophil granules were disrupted by homogenization noelectron microscopy. No deposition of peroxi- in 0.25 M sucrose containing 0.1% Triton X-100, and dase reaction product was seen in other organ- cores were purified by centrifugation through a layer of elles in the eosinophil cytoplasm or in other cells, 50% sucrose. The pellet (left), interface (middle), and supernate (right) were analyzed by SDS-PAGE. These and as shown in Fig. 2, antibody to MBP was preparations were derived from the experiment shown localized to the core of the granule. The conclu- in Fig. 3. sion that the staining of the core protein was specific is further supported by the absence of ence. Thus, the core protein could not be distin- staining with the NRS control and the failure of guished immunologically from the MBP, and the peroxidase reaction product to nonspecifi- essentially all of the protein in the alkylated core cally localize in the core when excess Fab-HRP preparation could be accounted for as MBP. was not washed away before the diaminobenzi- Furthermore, alkylated core protein radiolabeled dine reaction step. with 1811 was reactive with antibody to the MBP, In another approach to the localization of the and the binding of the alkylated core protein was MBP, we isolated cores from granules and deter- inhibited equally well by the MBP and the core mined the characteristics of the proteins derived protein (results not shown). In another series of from the core. Early experiments with whole experiments, MBP and core protein were freshly eosinophils indicated that eosinophil cores were prepared and immediately analyzed by RIA. The concentrated in the pellet after centrifugation results of a typical experiment are shown in Fig. through a cushion of 50% sucrose and that the 6. The inhibition curves produced by MBP and interface layer did not contain crystalloids. How- ever, in these early experiments, the cores were core protein were virtually identical, and statisti- cal analysis of the logit-log regressions did not contaminated by considerable debris, and analy- sis of the proteins by SDS-PAGE revealed the reveal a difference. Again, essentially all of the presence of numerous proteins in addition to the protein in the solubilized core preparations could be accounted for as MBP. MBP. Therefore, we used granules as the starting Finally, if the core protein is the MBP, then material for the preparation of cores and were 710 THE JOURNAL OF CELL BIOLOGY 9 VOLUME 77, 1978 \ (o 25 I- \\. -a -,., \\ \\ i i J I l I I I I ~ o 0.5 1 2 3 4 5 6 7 8 -0.3 0 0.2 0.4 0.6 0.8 1.0 Protein, n9 Leo protein, ng FIGURE 5 Comparison of alkylated MBP with alkylated core protein by RIA. Rabbit anti-MBP was reacted with ~aq-MBP and the inhibitory potencies of alkylated MBP and alkylated core protein were tested. In the absence of inhibitor, 36% of the counts bound to antibody, whereas in the absence of antibody 4.8% of counts were precipitated. The concentration of MBP was determined by absorbance at 277 nm, and the concentration of core protein was determined by biuret analysis using human serum albumin as a standard. The results with the MBP are shown by 9 and with the core protein by x. On the left the binding curves are shown, and on the right the logit-log regressions are shown. The correlation coefficients for the MBP inhibition regression line was -0.98 and for the core protein -0.96. Comparison of the logit-log regression lines revealed that the null hypothesis of a common line could not be rejected (F2. 6 = 1.46; NS). 3O "6 -1.8 i~ 20 r~ -2.0 X .~ I I I t i I t I I I I I I I I I 0 4 8 12 16 20 0.3 0.5 0.7 0.9 1.1 1.3 Protein, ng Log protein, ng FIGURE 6 Comparison of freshly prepared MBP with freshly prepared core protein by RIA. The ability of MBP and core protein, both freshly prepared, to inhibit the binding of ~aq-MBP to antibody was tested. In the absence of inhibitor, 33% of counts bound to antibody, whereas in the absence of antibody, 7.4% of counts were precipitated. The concentrations of MBP and core protein were determined by absorbance at 277 nm and by biuret analysis, respectively. On the left the binding curves are shown, and on the right, the logit-log regressions. The results with the MBP are shown by O and with the core protein by x. The correlation coefficients for the logit-log regressions of MBP and core protein were -0.97 and -0.99, respectively. Analysis of these regression lines revealed that the null hypothesis of a common line could not be rejected (F2,4 = 0.61; NS). LEWIS ET AL. Localization of Major Basic Protein 711 We thank Dr. Harold L. Moses and Dr. Robert E. able to obtain highly purified preparations as Scott for helpful discussions of these results. Dr. Jorge shown in Fig. 3. Even the apparent debris among E. Maldonado participated in some of the early exper- the crystalloids may be core protein, as inspec- iments to purify cores from disrupted eosinophils. tion of Fig. 3 b indicates that this irregular mate- This work was supported by grants from the National rial is found around the edges of cores which are Institute of Allergy and Infectious Diseases, A1 9728 losing their characteristic shape. Numerous anal- and A1 11483, and from the Mayo Foundation. yses of the core protein by SDS-PAGE showed a Received for publication 20 September 1976, and in band in the expected position for the MBP, and revised form 24 January 1978. this accounted for up to 90% of the dye binding in the gel. More conclusive evidence that the REFERENCES crystalloid protein is the MBP was provided by 1. ARCHER, G. T., and J. G. Hmscn. 1963. Isolation the RIA experiments in which we could not show of granules from eosinophil leucocytes and study a difference between the inhibitory properties of of their enzyme content. J. Exp. Med. 118:277- the core protein and the MBP. Moreover, we could account for virtually all of the protein in 2. AVRAMEAS, S. 1969. Coupling of enzymes to the solubilized core as MBP. The results of the proteins with glutaraldehyde. Use of the conjugate two sets of experiments employing immunoelec- for the detection of antigens and antibodies. Im- tron microscopy and analysis of isolated crystal- munochemistry. 6:43-52. loids indicate that the MBP is contained within 3. AXEN, R., J. PORATH, and S. ERNBACK. 1967. the core of the eosinophil granule, and further- Chemical coupling of peptides and proteins to polysaccharides by means of cyanogen halides. more, that it appears to be the only protein in Nature (Lond.). 214:1302-1304. the crystalloid. 4. BAINTON, D. F., and M. G. FARQUHAR. 1970. Our results do not exclude the possibility sug- Segregation and packaging of granule enzymes in gested by certain authors (6, 30) that the Char- eosinophilic leukocytes. J. Cell. Biol. 45:54-73. cot-Leyden crystal protein is contained in the 5. COTRAN, R. S., and M. LITT. 1969. The entry of human eosinophil granule core. Although human granule-associated peroxidase into phagocytic vac- MBP and Charcot-Leyden crystal protein differ uoles of eosinophils. J. Exp. Med. 129:1291- in their molecular weights and amino acid com- positions and are distinct proteins (14), it is 6. EL-HASHIMI, W. 1971. Charcot-Leyden crystals. possible that the human and guinea pig eosino- Formation from primate and lack of formation phil crystalloids differ. Therefore, if the Charcot- from non-primate eosinophils. Am. J. Path. 65:311. Leyden crystal protein is derived from the core 7. ENOMOTO, T., and T. KITANI. 1966. Electron of the human eosinophil granule, then it presum- microscopic studies on peroxidase and acid phos- ably coexists there with the MBP. phatase reaction in human leukocytes. Acta Hae- Finally, Okun et al. (25) reported that the matol. Japn. 29:554-570. crystalloid core of human eosinophil granules 8. FAIRBANKS, G., T. L. STECK, and D. F. H. contained melanin. Their conclusion was based WALLACH. 1971. Electrophoretic analysis of the on results obtained with histochemical staining major polypeptides of the human erythrocyte techniques, and in particular with the Fontana- membrane. Biochemistry 10:2606-2617. Masson argentaffin stain. Because we have not 9. GESSNER, T. P., S. R. HIMMELHOCH, and E. seen pigment in any protein preparation made SHELTON. 1973. Partial characterization of the protein components of eosinophil granules isolated from purified granules, we do not believe that from guinea pig exudates. Arch. Biochem. Bio- melanin is present in guinea pig eosinophil gran- phys. 156:383-389. ules. Moreover, MacRae and Spitznagel (22) 10. GESSNER, T. P., E. SHELTON, and S. R. HIMMEL- have presented evidence that the ammoniacal HOCH. 1972. Liberation of crystalloid from eo- silver reaction, which is the basis of the Fontana- sinophil granules. Fed. Proc. 31:253 (Abstr.). Mason argentaffin stain (27), can be used as a 11. GLEICH, G. J., and D. A. LOEGERING. 1973. cytochemical marker for the detection of cationic Selective stimulation and purification of eosino- proteins by electron microscopy. Thus, the ob- phils and neutrophils from guinea pig peritoneal servations of Okun and his associates might be fluids. J. Lab. Clin. Med. 82:522-528. viewed by supportive evidence for the presence 12. GLEICrt, G. J., D. A. LOEGERING, F. KUEVPERS, S. P. BAJAJ, and K. G. MANN. 1974. Physi- of the cationic MBP in the granule core. 712 ThE JOURNAL OF CELL B1OLOCV" VOLUME 77,1978 ochemical and biological properties of the major Blood. McGraw-Hill Book Co., New York. 22. MACRAE, E. K., and J. K. SPITZNAGEL. 1975. basic protein from guinea pig eosinophil granules. UItrastructural localization of cationic proteins in J. Exp. Med. 140:313-332. cytoplasmic granules of chicken and rabbit poly- 13. GLEICH, G. J., D. A. LOEGERING, and J. E. morphonuclear leukocytes. J. Cell Sci. 17"79-94. MALDONADO. 1973. Identification of a major basic 23. McCONAHEY, P. J., and F. J. DIXON. 1966. A protein in guinea pig eosinophil granules. J. Exp. method of trace iodination of proteins for immu- Med. 137:1459-1471. nologic studies. Int. Arch. Allergy' Appl. Immunol. 14. GLEICH, G. J., D. A. LOEGERING, K. G. MANN, 29:185-189. and J. E. MALDONADO. 1976. Comparative prop- 24. MILLER, F., E, DEHARVEN, and G. E. PALADE. erties of the Charcot-Leyden crystal protein and 1966. The structure of the eosinophil leukocyte the major basic protein from human eosinophils. granules in rodents and in man. J. Cell Biol. J. Clin. Invest. 57:633-640. 31:349-362. 15. GRAHAM, R. C., JR., and M. J. KARNOVSKY. 25. OKUN, M. R., B. DONNELLAN, S. H. PERASON, 1966. The early stages of absorption of injected and L. M. EDELSTEIN. 1974. Melanin: A normal horseradish peroxidase in the proximal tubules of component of human eosinophils. Lab. Invest. mouse kidney: ultrastructural cytochemistry by a 30:681-685. new technique. J. Histochem. Cytochem. 14:291- 26. PARMLEY, R. T., and S. S. SPICER. 1974. Cyto- chemical and ultrastructural identification of a 16. KABAT, E. A., and M. N. MAYER. 1961. Estima- small type granule in human late eosinophils. Lab. tion of protein with the biuret and ninhydrin Invest. 30:557-567. reactions. In Experimental Immunochemistry. Charles C Thomas, Publisher, Springfield, Ill. 2nd 27. PEARSE, A. G. E. 1972. Histochemistry, Theoret- ical and Applied. Vol 2. Williams & Wilkins Co., edition. 559. 17. KONIGSBERG, W. 1972. Reduction of disulfide Baltimore. 3rd edition. 1485. bonds in proteins with dithiothreitol. Methods 28. PORTER, R. R. 1959. The hydrolysis of rabbit gamma globulin and antibodies with crystalline Enzymol. 25:185-188. 18. KRAEHENBUHL, J. P., and J. D. JAMIESON. 1974. papain. Biochem. J. 73:119-126. Localization of intracellular antigens by immuno- 29. SEEMAN, P. M., and G. E. PALADE. 1967. Acid electron microscopy. Int. Rev. Exp. Path. 13:1-53. phosphatase localization in rabbit eosinophils. J. 19. LEWIS, D. M., D. A. LOEGERIN6, and G. J. Cell Biol. 34:745-756. GLEICH. 1976. Antiserum to the major basic 30. WELSH, R. A. 1959. The genesis of the Charcot- Leyden crystal in the eosinophilic leukocyte of protein of eosinophil granules, lmrnunochemistry. 13:743-746. man. Am. J. Pathol. 35:1091-1104. 20. LEWIS, D. M., D. A. LOEGERING, and G. J. 31. WORTHINGTON, Enzyme manual. 1972. Worth- ington Biochemical Corporation, Freehold, N. J. GLEICH. 1976. Isolation and partial characteriza- tion of a major basic protein from rat eosinophil 32. YUNGINGER, J. W., and G. J. GLEICH. 1972. granules. Proc. Soc. Exp. Biol. Med. 152:512- Comparison of the protein binding capacities of cyanogen bromide-activated polysaccharides. J. 21. Low, F. N., and J. A. FREEMAN. 1958. Electron Microscopic Atlas of Normal and Leukemic Human Allergy Clin. Immunol. 50:109-116. LEWIS ET AL. Localization of Major Basic Protein 713

Journal

The Journal of Cell BiologyPubmed Central

Published: Jun 1, 1978

There are no references for this article.