Acquired hyper-IgM syndrome with necrotizing granuloma

Acquired hyper-IgM syndrome with necrotizing granuloma

Acquired hyper-lgM syndrome with necrotizing granuloma M. F. Goldstein, M.D., M. J. Kornstein, M.D., S. Talbot, M.D., and A. I. L e v i n s o n , M.D...

732KB Sizes 0 Downloads 40 Views

Acquired hyper-lgM syndrome with necrotizing granuloma M. F. Goldstein, M.D., M. J. Kornstein, M.D., S. Talbot, M.D., and A. I. L e v i n s o n , M.D. Philadelphia, Pa.

This article is highlighted by the finding of striking cervical lymphadenopathy in a patient with acquired hyper-lgM syndrome and the pathologic description of the involved nodes. Routine hematoxylin-eosin stains demonstrated the presence of idiopathic necrotizing granulomas in the nodal tissue, a finding not previously reported in this syndrome. Immunoperoxidase techniques were used to further characterize these granulomas and delineate the cellular composition of the nodal architecture. We found that the necrotizing granulomas consisted of a peripheral rim of la positive palisaded, epithelioid histiocytes and central areas of debris and scattered inflammatory cells that were T l l positive. In the uninvolved areas of the node, we observed a lack of lgG-bearing lymphocytes in germinal centers as well as an absence of lgG-containing and decreased lgA-containing plasma cells in interfollicular areas. In conjunction with these in situ observations, there was a lack of lgA and IgG immunoglobulin-secreting cell responses in pokeweed mitogen-stimulated cultures of the patient's peripheral blood mononuclear cells. Unique features of this article include: (1) the association of necrotizing granulomas with the hyper-lgM syndrome and (2) the use of monoclonal antibodies to characterize the distributions of nodal lymphocytes in a patient with this disorder. ( J ALLERGY CLIN IMMUNOL 75.'472-8, 1985.)

Immunodeficiency with hyper-IgM is a dysgammaglobulinemia that was first reported in 19611 and is recognized by the World Health Organization as one of the primary immunodeficiency syndromes. 2 This syndrome is characterized by a very low or absent serum concentration of IgG and IgA and either normal or more frequently a markedly elevated polyclonal serum IgM concentration. It has been reported in children in both X-linked and acquired forms 1' 3, 4 as well as in adult subjects) Clinically, patients experience mild to severe recurrent pyogenic infections and may develop a lymphoproliferative disorder 6 and autoimmune cytopenias. 7 The immunopathogenic mechanisms for this syndrome appear to be varied 3' 6. s-ll with evidence for intrinsic B abnormalities as well as regulatory T cell dysfunction. In this article we describe a man with the acquired form of the hyper-IgM syndrome who presented with

From the Departments of Medicine, Allergy and ImmunologySection, and Pathology and Laboratory Medicine, University of Pennsylvania, School of Medicine, Philadelphia, Pa. Received for publication May 4, 1984. Accepted for publication Aug. 11, 1984. Reprint requests: A. I. Levinson, M.D., Universityof Pennsylvania School of Medicine, 515 Johnson Pavilion, 36th & Hamilton Walk, Philadelphia, PA 19104.

472

Abbreviations used PBM: Peripheral blood mononuclear cells PWM: Pokeweed mitogen IgSC: Immunoglobulin secreting cells PAP: Peroxidase anti-peroxidase IgGSC: Immunoglobulin G secreting cells IgMSC: Immunoglobulin M secreting cells IgASC: Immunoglobulin A secreting cells

striking cervical lymphadenopathy. A lymph node biopsy specimen revealed necrotizing granulomas, a feature not previously reported in this syndrome. Moreover, the need for a lymph node biopsy enabled us to carry out a detailed immunohistologic analysis of the nodal tissue that also has not previously been described in this syndrome. This unusual clinical presentation along with the immunohistologic studies and immunologic studies of our patient's PBM cells are the subjects of this article.

Clinical case The patient is a 19-yr-old black man who was admitted to the hospital of the University of Pennsylvania with a complaint of fever and a neck mass on the left side. He had been healthy until the age of 14 yr after which he had yearly

VOLUME75

Acquired hyper-lgM syndrome 473

NUMBER 4

episodes of radiographically documented pneumonia, each of which cleared with outpatient antibiotic therapy. At the age of 17 yr, a laparotomy performed at another hospital for presumed appendicitis revealed mesenteric adenitis. Histopathology revealed many centrally necrotic granulomas, whereas microbiologic evaluation was unrevealing. The resuits of quantitative immunoglobulins (IgM, 677 mg/dl, normal 70 to 210 mg/dl; undetectable lgG and lgA) returned after his discharge. The patient was lost to follow-up, and no specific therapy was instituted. One month before his current admission the patient noticed a cervical mass on the left side that became progressively larger and more indurated but remained nontender. There was no past history of sinusitis, otitis media, sepsis, cytopenias, or autoimmune phenomena. Family history was unremarkable for immunodeficiency, infant deaths, infectious mononucleosis, lymphoproliferative disease, or autoimmune disease. He was a heterosexual, smoked marijuana on a daily basis, but did not smoke cigarettes or use illicit drugs. He was currently employed in loading vegetable trucks and had no unusual environmental exposures. On his admission rectal temperature was 105.3 ~ F, pulse was regular at 92, unlabored respiratory rate was 18, and blood pressure was 100/64. There was a 4 by 6 cm firm, nontender neck mass on the left side, a 2 by 3 cm freely mobile submandibular node on the right side of the neck, and shotty axillary, inguinal, and cervical adenopathy on the right side. Tonsils, lingual lymphoid tissue, and epiglottis were moderately enlarged. Examination of the heart, lungs, abdomen, extremities, and skin was normal. His hemoglobin was 13.9 gm/dl, platelet count 335,000/mm3, and white blood cell count 17,800/ mm~with 44% polymorphonuclear leukocytes, 36% bands, 9% lymphocytes, and 10% monocytes. Serum electrolytes, liver function tests, creatine phosphokinase, albumin, calcium, urine analysis, electrocardiogram, and chest x-ray film were normal. Despite negative blood, urine, and throat cultures, his fever slowly defervesced while he was receiving intravenous ampicillin therapy, and his white blood cell count and differential returned to normal. Serum protein electrophoresis performed on cellulose acetate demonstrated hypogammaglobulinemia. Quantitative immunoglobulins by radial immunodiffusion demonstrated an IgM of 465 mg/dl (nl 50 to 280 mg/dl), IgA <52 mg/dl (nl 60 to 340 mg/dl) and IgG <256 mg/dl (nl 620 to 1420 mg/ dl). His ABO blood type was B with anti-A titers of 1 : 16 at room temperature. Epstein-Barr serology determination demonstrated no evidence of acute, chronic, or persistent Epstein-Barr virus infection (kindly performed by Dr. W. Henle, Children's Hospital of Philadelphia). Hepatitis B virus serologies revealed no hepatitis B surface antigen and no antibodies to hepatitis B surface antigen or core antigen. C3 was 210 mg/dl (nl 100 to 233 mg/dl), C4 was 24 mg/ dl (nl 15 to 48 mg/dl), and total hemolytic complement was 265 txm/ml (nl 110 to 190 ixm/ml). He had positive cutaneous delayed hypersensitivity reactions to 0.1 ml injections of Candida (500 PNU/ml), tetanus toxoid (1.5 Loeffler units per milliliter), and mumps (20 complement fixing units per milliliter) but was nonreactive to 0.1 ml of 5 TU of purified protein derivative. After 175 mg of intravenous gamma

globulin on the tenth day of hospitalization, he had a direct laryngoscopy and biopsy of the neck mass of the left side. Grossly, the neck mass was made up of a large firm group of cervical nodes. A superficial node 2.5 cm in diameter was excised for histologic and microbiologic examination. Cultures were negative for routine bacteria, mycobacteria, and fungi. On microscopic examination there were numerous well circumscribed, centrally necrotic granulomas (Fig. 1). The granulomas were irregularly shaped and were composed of palisaded epithelioid histiocytes at the periphery with neutrophils, rare lymphocytes, histiocytes, and inflammatory debris in the center. Scattered secondary follicles were also noted with normal architecture. Within the germinal centers were lymphocytes and macrophages surrounded by mantle zones of small lymphocytes. Prominent fibrosis and numerous plasma cells were evident in the subcapsular and interfolticular areas. A review of the lymph node biopsy specimen obtained from the exploratory laparotomy 21/2 yr earlier revealed similar pathology but with less fibrosis. During this admission leukocyte function studies including random migration and chemotaxis, bactericidal killing, phagocytosis, and a nitroblue tetrazolium reduction test were all normal (kindly performed by Dr. S. Douglas, Children's Hospital of Philadelphia). The patient was then lost to follow-up and untreated for the next 7 mo during which time he noted a decrease in the size of the cervical mass on the left side. He reappeared with a productive cough, and an examination was remarkable only for a nontender, firm 4 by 4 cm cervical mass on the left side and 2 by 2 cm cervical mass on the right side. Laboratory evaluation revealed IgM 105 mg/dl (nl 56 to 352 mg/dl); undetectable IgG and IgA; white cell count of 12,300 cc/mm with 77% neutrophils, 16% lymphocytes, 7% monocytes, and 1% eosinophils; and a chest x-ray film that illustrated chronic inflammatory changes suggestive of bronchiectasis. Febrile agglutinins were undetectable. An angiotensin-converting enzyme was normal. Intravenous gamma globulin therapy was reinstituted without complication.

MATERIAL AND METHODS In vitro studies of PBM Before initial institution of gamma globulin therapy, analysis of the phenotype and function of the patient's PBM was undertaken to provide insight into the cellular basis of the patient's immunodeficiency. PBM were prepared on Ficoll-Hypaque density gradients. In addition, a single cell suspension was prepared from the excised lymph node. A tissue fragment was minced and then passed through a no. 50 stainless wire mesh. Cells were then separated on a Ficoll-Hypaque gradient. Lymphocyte subsets were identified in the PBM and lymph node cell suspension after incubation with murine monoclonal antibodies and staining with a fluorescein conjugated goat anti-mouse IgG antibody (Tago, Burlingame, Calif.) as previously described.'2 Lymphocytes were isolated for analysis by optimal gain settings on a forward and right angle light scatter cytogram by use of an Ortho Spectrum III fluorescent cell analyzer (Ortho

474 Goldstein et al.

Diagnostic Instruments, Westwood, Mass.), and a fluorescence histogram was then obtained. Positive and negative fluorescence thresholds were adjusted so that less than 1% of the cells were considered positive with a negative control monoclonal antibody. The murine monoclonal antibodies used included T3, a pan-T-lymphocyte marker; T4, a helper/ inducer T-lymphocyte marker; T8, a suppressor/cytotoxic T-lymphocyte marker; and B 1, a peripheral B cell marker (all obtained from Coulter Corp., Hialeah, Fla.). For assessment of mitogen-induced lymphocyte proliferation, PBM were cultured under our standard conditions,t3 and 3H-thymidine incorporation was measured in cells stimulated with PWM, concanavalin A, or phytohemagglutinin. Incorporation of radioactivity into newly synthesized DNA was determined by liquid scintillation spectrometry at day 4. For determination of polyclonal B cell differentiation responses, aliquots of the same PBM specimen were cultured as previously described'3 in round bottom microtiter wells in the presence and absence of optimal (1:50 dilution) and suboptimal (1:1000 dilution) concentrations of PWM. The cultures were harvested on day 6, at which time the number of IgSC, i.e., cells secreting IgG, IgM, or IgA, were determined by a protein A reverse hemolytic plaque assay as previously described. 13.~4Responses of IgSC were expressed as IgSC/106 cultured cells. PBM, obtained from a healthy volunteer and run in parallel, served as a control.

Immunohistologic studies In vitro characterization of intracytoplasmic immunoglobulin in lymph node cells was carried out on paraffinembedded Bouin's fixed tissue by the standard PAP technique, k~ This technique was performed by use of heteroantisera as primary antibodies (Dako Corp., Santa Barbara, Calif.) to lgG, IgM, IgA, kappa chain, or lambda chain.' In addition, the avidin-biotin immunoperoxidase tech-

j. ALLERGYCLIN,IMMUNOL. APRIL 1985

nique was performed on acetone-fixed frozen sections to demonstrate lymphocyte surface markers.'6 The following monoclonal reagents were used as primary antibodies: T11, a marker for E rosette + cells (Coulter Corp.), T4, T8, OK Ia, a marker for a gene product encoded in the major histocompatibility complex (Ortho Pharmaceuticals, Raritan, N. J.), anti-IgD (Cappel Laboratories, Cochranville, Pa.), and anti-IgM (Courtesy Dr. M. Conley, Children's Hospital of Philadelphia). After a 15-minute incubation with the primary antibody at room temperature, sections were washed in phosphate-buffered saline and then were incubated for 15 min with biotinylated horse anti-mouse antibody (Vector Laboratories, Burlingame, Calif.). After another phosphate-buffered saline wash, slides were incubated with avidin-biotinylated peroxidase complex (Vector Laboratories). Slides were then washed in PBS and incubated for 10 min with 0.05% diaminobenzidine tetrahydrochloride in 0.05 mol/L Tris buffer, pH 7.2, and 0.03% hydrogen peroxide. Sections were counterstained in hematoxylin for 30 sec and mounted in Elvanol (Dupont Co., Wilmington, Del.).

RESULTS In vitro studies of PBM Characterization of PBM revealed normal percentages of total B cells and T cells and normal percentages of T cells expressing the T4 or T8 phenotype (Table I). The proliferative responses of the patient's PBM to concanavalin A, phytohemagglutinin, and P W M were comparable to those of the normal control subject (data not shown). By contrast, P W M - i n d u c e d generation of IgSC in cultures of the patient's PBM was depressed (Table II). The use of isotype-specific developing antisera revealed that most IgSC generated were IgMSC in contrast to culture of normal PBM in

FIG. 1. Low power photomicrograph of surgically excised cervical lymph node demonstrating irregularly shaped necrotizing granulomas (open arrow) and secondary follicles (closed arrows). (Hematoxylin-eosin stain. Original magnification x 40.) FIG. 2. Section of lymph node interfollicular zone demonstrates numerous IgM-containing plasma cells with brownish-red cytoplasmic staining. Closed arrow is a representative cell. (PAP technique with rabbit anti-human IgM antibody. Hematoxylin counter stain. Original magnification • 400.) FIG. 3. Section of lymph node interfollicular zone demonstrates rare IgA-containing plasma cells (closed arrows). (PAP with rabbit anti-human IgA antibody. Hematoxylin counterstain. Original magnification x 400.) FIG. 4. Section of lymph node interfollicular zone demonstrates no IgG-containing plasma cells. (PAP with rabbit anti-human IgG antibody. Hematoxylin counterstain. Original magnification x 400.) FIG. 5, Frozen section of lymph node was examined for T l l surface antigen. Numerous T l l positive cells with brownish-red peripheral ring staining are present predominantly in the interfollicular areas. Closed arrows are secondary follicles. (Avidin-biotin immunoperoxidase with anti-T11 monoclonal antibody. Hematoxylin counterstain. Original magnification x 200.) FIG. 6. Frozen section of lymph node demonstrates localization of la antigen. A necrotizing granuloma surrounded by la positive histiocytes is illustrated (closed arrow). Surrounding the granuloma are numerous la positive lymphocytes. (Avidin-biotin immunoperoxidase with antila monoclonal antibody. Hematoxylin counterstain. Original magnification • 200.)

VOLUME75

Acquired hyper-lgM syndrome 475

NUMBER 4

FIGS. 1-6. For legend see opposite page.

which the usual mixed pattern of IgGSC, IgMSC, and IgASC was observed. ~7 I m m u n o h i s t o l o g i c studies

Microscopic examination of the cervical node biopsy specimen from the left side revealed necrotizing granulomas throughout the specimen as previously described (Fig. 1). Routine PAP studies on the present material revealed nearly all the interfollicular plasma cells to contain cytoplasmic IgM (Fig. 2). Rare IgA positive plasma cells and no IgG positive cells were present (Figs. 3 and 4). Kappa and lambda light chain-contaming plasma cells were present in approximately equal numbers. Of note, germinal centers (areas of B

cell predominance) identified on hematoxylin-eosin stains revealed cells containing cytoplasmic IgM but not IgG or IgA on PAP studies. Surface marker studies on frozen sections revealed a normal distribution of T11 positive cells predominantly in the interfollicular areas with occasional positive cells in the germinal center (Fig. 5). T8 positive cells had a similar distribution to that of the TI 1 population and comprised about 30% of the T11 population (consistent with what has been described in hyperplastic nodal tissue). ~8The remainder of the T 11 population in the interfollicular zones were T4 positive. Anti-Ia and B 1 antibody stained the lymphocytes of the secondary follicles. Surface lgD positive cells were located predominantly in the mantle zones and

476

G o l d s t e i n et al.

J. ALLERGYCLIN, IMMUNOL. APRIL 1985

TABLE I. Phenotypic analysis of blood lymphocytes Patient (%)

Phenotype

Cell type

B1 T3 T4 T8 T4/T8 ratio

Bcells Total peripheral T cells Helper/inducer cells Suppressor/cytotoxic cells

7 72 37 24 1.5

Controls (%)

7to 61 to 38 to 18 to 1.1 to

11 75 53 30 2.9

TABLE II. In vitro polyclonal B cell differentiation responses* IgSC/10s Cultured cells Subject

Control

Patient

Isotype of IgSCl"

IgG IgM IgA lgG IgM IgA

RPMI-1640

1638 2400 5562 0 312 50

PWM (1 : 1000)

PWM (1:50)

10,163 7,900 2,913 50 525 0

12,200 10,238 8,475 0 3,463 300

*Cells cultured in presence of medium (RPMI-1640) or PWM and mean numbers of IgSC/10~cultured cells determined after 6 days of culture. tlsotype of IgSC determined by use of heavy chain-specific anti-human immunoglobulinantisera as developingreagents in the reverse hemolytic plaque assay. primary follicles with scattered positive cells in the interfollicular zone. Mantle lymphocytes and germinal center lymphocytes also stained positively for surface IgM. Antibody to IgM appeared to stain extracellular material in the germinal centers as well. In regard to the necrotizing granuloma, anti-Ia antibody stained epithelioid histiocytes in the peripheral rim of the granulomas (Fig. 6), whereas rare scattered T11 positive cells were present in the central aspect of these granulomas. B 1 positive cells were not present in these lesions. Analysis of cell suspension of the lymph node revealed the following percentages of T cell subsets: T l l , 42%; T8, 7.6%; and T4, 31%. The T4/T8 ratio was 4: 1. DISCUSSION

The clinical diagnosis of the hyper-IgM syndrome in our patient is based on the pattern of his humoral immunodeficiency (low IgG, low IgA, and elevated IgM), the presence of isohemagglutinins, and the history of recurrent pulmonary bacterial infections. In vitro functional studies were consistent with previous observations made in this syndrome, 6' 8.9. ~9.zo i.e., the patient's PBM demonstrated normal proliferative responses to T and B cell mitogens. However, they manifested an impaired IgSC response to PWM with IgMSC comprising the bulk of responding cells. This in vitro defect has been previously reported and is

thought to reflect impaired B-lymphocyte isotype switching.6, J9 Further evidence of this defect has been suggested by the reduction or lack of IgGand IgA-bearing peripheral blood B cells in these patients? 6, ~9 This case is highlighted by an unusual clinical presentation and the histopathologic features of the biopsy specimen of the lymphoid tissue. This immunodeficient patient presented because of a large neck mass rather than the usual chief complaint of recurrent bacterial infections. Although lymphoid hyperplasia is observed in the hyper-IgM syndrome, the asymmetry and magnitude of the nodal enlargement observed in our patient is unusual in the absence of frank lymphoma. On routine microscopic analysis of this mass, there was no evidence of lymphoma. The mass represented a cluster of lymph nodes that contained necrotizing granulomas. Granulomas have been reported in the lymph nodes and bone marrow in only two patients with the hyperIgM syndrome 2~ 21 as well as in occasional patients with other hypogammaglobulinemias, including common variable hypogammaglobulinemia22 and rare cases of hypogammaglobulinemia with coexistent sarcoidosis. 2325 The granulomas in these cases have all been of a noncaseating, noninfectious variety. It was therefore surprising to find necrotizing granulomas on the biopsy specimens from our patient, particularly without evidence for an underlying infectious etiology.

VOLUME75

Acquired hyper-lgM syndrome

477

NUMBER 4

TABLE III. Summary of PAP stains Stain

Hyper-lgM lymph node

Reactive lymph node

PAP Cytoplasmic IgM Cytoplasmic IgA Cytoplasmic IgG

Interfollicular plasma cells; germinal center Interfollicular plasma cells None

Interfollicular plasma27cells; germinal center Interfollicular plasma27cells; germinal center Interfollicular plasma:7 cells; germinal center

Disorders associated with necrotizing granulomas include: lymphogranuloma venereum, cat scratch disease, listeriosis, glanders, melioidosis, tularemia, tuberculosis, and syphilis. The absence of any positive microbiologic stains or cultures made all these diagnoses unlikely except for cat scratch disease; however, the chronicity of the lesions evidenced by the fibrotic reaction and the presence of necrotizing granulomas 21/2 yr earlier in the appendix were against this possibility. The availability of a lymph node specimen from our patient provided us with the opportunity to comprehensively study the nodal architecture with a new, sensitive immunoperoxidase technique linked with monoclonal antibodies that enables in situ characterization of cellular constituents. These techniques have not been reported in the analysis of lymphoid tissue of patients with the hyper-lgM syndrome. Previous immunofluorescent studies of lymphoid tissue in patients with the hyper-IgM syndrome have used specific heteroantisera, and the results have been varied. Lymphoid tissue has generally demonstrated poor organization with few or absent germinal centers and decreased numbers of plasma cells containing IgM and no IgG- or IgA-containing plasma cells, similar to the observations in our case. 22 With regard to the distribution of cells belonging to the B cell lineage, we observed a normal pattern of IgD- and IgM-bearing lymphocytes in the primary follicles and mantle zones of secondary follicle cells. Such cells are considered to be unstimulated virgin B cells.26.27 Features atypical for reactive nodes but consistent with what one might expect in the hyper-IgM immunodeficiency syndrome were: (1) lack of IgGcontaining B-lymphocytes in the germinal centers, and (2) an absence of IgG-containing and decreased IgA-containing plasma cells in the interfollicular areas where normally a mixture of IGA-, IgG-, and IgMcontaining plasma cells are usually observed 26-28 (Table III). These findings are consistent with the in vitro and in vivo observations of our patient's failure to produce anything but IgMSC on polyclonal B cell stimulation and the presence of only IgM in the circulation, respectively. In regard to the distribution of T cells, the anatomic localization of T cell subsets in areas uninvolved with

necrotizing granuloma was similar to that previously described in reactive nodes, ~8i.e., T cells were present predominantly in interfollicular areas or cortical regions of the lymph node. The ratio of helper T/suppressor T cells in these areas was also consistent with a ratio of approximately 3 to 4:1 previously reported in reactive nodesfl 9 The cell suspension data, albeit from nodal tissue cor,taining necrotizing granulomas, also revealed a similar percentage distribution of T cell subsets (i.e., T4/T8 ratio of 4: 1). The latter resuits are similar to those reported recently for suspensions of lymphoid cells from reactive nodes. 3~ In that article the magnitude of the T4/T8 ratio was correlated with the morphologic pattern of reactive lymphoid hyperplasia. The localization of T- and B-lymphocytes and macrophages in and around granulomas secondary to leprosy,29. 31 sarcoid,:9, 32 and rhinoscleroma29 has been described. No consistent findings have been reported among these disease states. In our case Ia positive, B1 negative, T3 negative, T8 negative, and T l l negative epithelioid histiocytes rimmed the necrotizing granuloma with a few T1 1 positive cells in the center of these lesions. This phenotypic analysis of mononuclear cell infiltrates is similar to that described in other mature epithelioid cell granulomatous diseases, i.e., tuberculoid leprosy and sarcoidosis? 9 The heterogeneity of the patterns of T cell and macrophage proportions in these various granulomatous diseases may reflect differences in pathogenesis and/or stages of immune responses. In summary we report an unusual case of acquired hyper-IgM syndrome associated with necrotizing granulomas. This histopathologic picture may represent either a finding unrelated to the patient's dysgammaglobulinemia, an unusual inflammatory response in a patient with impaired humoral immunity, or it may be secondary to some unknown agent capable of impairing the patient's humoral immune system and affecting the reticuloendothelial system in such a way as to stimulate production of nonspecific necrotizing granulomas. Most of the evidence conceming immunologic perturbation in the hyper-IgM syndrome has been on the basis of in vitro studies on PBM. The availability of monoclonal antibodies coupled with immunoperoxidase techniques allowed for

478

G o l d s t e i n et al.

in situ characterization of immunocompetent cells in the lymphoid tissue of our patient. This technique may help define various pathologic lymph node microenvironments in other immunodeficiency states as well and may represent a useful link between the histopathology and the in vitro immunologic analysis of lymphocyte phenotype and function in these patients. REFERENCES I. Rosen FJ, Kevy SV, Merler E, Janeway CA, Gitlin D: Recurrent bacterial infections and dysgammaglobulinemia: deficiency of 7S gammaglobulins in the presence of elevated 19S gammaglobulins. Pediatrics 28:182, 1961 2. Immunodeficiency report. WHO scientific group. Clin Immunol Immunopathol 13:297, 1979 3. Stiehm ER, Fundenberg HH: Clinical and immunologic features of dysgammaglobulinemia type I. Am J Ped 40:805, 1966 4. Kyong CV, Virella D, Fundenberg HH, Darby PC: X-linked immunodeficiency with increased IgM: clinical, ethic, and immunologic heterogeneity. Pediatr Res 12:1024, 1978 5. Rosen FS, Bougas JA: Acquired dysgammaglobulinemia: elevation of 19S gammaglobulin and deficiency of the 7S gammaglobulin in a woman with chronic progressive bronchiectosis. N Engl J Med 269:1336, 1963 6. Geha RS, Hyslop N, Alami S, Farah F, Scheeberger EE, Rosen T: Hyperimmunoglobulin M immunodeficiency (dysgammaglobulinemia). J Clin Invest 64:385, 1979 7. Hinz CF, Boyer JT: Dysgammaglobulinemia in an adult manifested as autoimmune hemolytic anemia: serologic and immunochemical characterization of antibody of unusual specificity. N Engl J Med 269:1329, 1963 8. Krantman HJ, Stiehm ER, Stevens RH, Saxon A, Seeger RC: Abnormal B cell differentiation and variable increased T cell suppression in immunodeficiency with hyper-IgM. Clin Exp Immunol 40:147, 1980 9. Pascual-Salcedo D, DelaCoucher EG, Gacia-Rodriquez MC, Zabay JM, Saint T, Fontan G: Cellular basis of hyper-lgM immunodeficiency with hyper-lgM. J Clin Lab Immunol 2:337, 1979 10. Mitsuya H, Tomino S, Hisanitsu S, Kishimoto S: Evidence for the failure of IgA specific T helper activity in a patient with immunodeficiency with hyper-lgM. J Clin Lab Immunol 2:337, 1979 11. Herrod HG, Perlman DB, Buckley RH: Evidence of excessive T suppressor activity in x-linked immunodeficiency with hyperIgM. Pediatr Res 12:480, 1978 12. Reinherz EL, Kung PC, Goldstein G, Schlossman SF: Separation of functional subsets of human T cells by a monoclonal antibody. Proc Natl Acad Sci USA 76:4061, 1979 13. Levinson AI, Dziarski A, Pincus T, DeHoratius RJ, Zweiman 13: Heterogeneity of polyclonal B cell activity in systemic lupus erythematosus. J Clin Lab Immunol 5:17, 1981 14. Granowicz E, Couthino A, Melchers F: A plaque assay for all

J. ALLERGY CLIN. IMMUNOL. APRIL 1985

15.

16.

17. 18.

19.

20.

21. 22. 23.

24.

25. 26. 27.

28.

29.

30.

31.

32.

cells secreting lg of a given type or class. Eur J Immunol 6:588, 1976 DeLellis RA, Sternberger LA, Mann RB, Banks PM, Nakane PK: Immunoperoxidase techniques in diagnostic pathology. Workshop sponsored by the National Cancer Institute. Am J Clin Pathol 71:483, 1979 Kornstein MJ, Brooks JSJ, Elder DE: Immunoperoxidase localization of lymphocyte subsets in the host response to melanoma and nevi. Cancer Res 43:2749, 1983 Levitt D, Dagg MK: Human B-lymphocyte subpopulations. Clin Immunol Immunopathol 21:50, 1981 Poppema S, Bhan AK, Reinherz EL, McCloskey RT, Schlossman SF: Distribution of T cell subsets in human lymph nodes. J Exp Med 153:30, 1981 Levitt D, Haber P, Rich K, Cooper MD: Hyper-IgM immunodeficiency a primary dysfunction of B lymphocyte isotype switching. J Clin Invest 72:1650, 1983 13eall GN, Ashman RF, Miller ME, Easwaran C, Raghunatha R, Louie J, Yoshikawa T: Hypogammaglobulinemia in mother and son. J ALLERGYCLIN IMMUNOL65:471, 1980 Hobbs JR, Russel A, Worlledge SM: Dysgammaglobulinemia type IV. Clin Exp Immunol 2:581, 1967 Buckley RH: Immunodeficiency. J ALLERGYCLIN IMMUNOL 72:627, 1983 Friedman R, Ackerman M, Mallory G, et al: Hypogammaglobulinemia with sarcoid-like granulomas. Am J Dis Child 137:774, 1983 Sharma OP, James DG: Hypogammaglobulinemia, depression of delayed-type hypersensitivity, and granulomata formation. Am Rev Respir Dis 104:228, 1971 Bronsky D, Dunn Y: Sarcoidosis with hypogammaglobulinemia. Am J Med Sci 11:18, 1965 Bhan AK, Nadler LM, Stashenko P: Stages of B cell difterentiation in human lymphoid tissue. J Exp Med 154:737, 1981 Hoffman FM, Yanagihara E, Byrne B, et ah Analysis of B cell antigens in normal reactive lymphoid tissue using four B cell monoclonal antibodies. Blood 62:775, 1983 Morris HB, Mason DY, Stein H, Lennert K: An immunohistological study of reactive lymphoid tissue. Histopathology 7:809, 1983 Modlin RL, Hofman FM, Meyer PR, et al: In situ demonstration of T lymphocyte subsets in granulomatous inflammation: leprosy, rhinoscleroma, and sarcoidosis. Clin Exp Immunol 51:430, 1983 Ruco LP, Stoppacciaro A, Valtieri M, et ah T-lymphocyte subsets in human lymph nodes: relative increase of OKT8 + cells in neoplastic and reactive B cell proliferation. Clin lmmunol lmmunopathol 30:337, 1984 Narayanan RB, Bhutani LK, Sharma AK, et ah T cell subsets in leprosy lesions: in situ characterization using monoclonal antibodies. Clin Exp Immunol 51:421, 1983 Koattinen YT, Tolvanen E, Visa-Tolvanen K, et al: Inflammatory cells in sarcoid granulomas detected by monoclonal antibodies and an esterase technique. Clin Immunol Immunopathol 26:380, 1983