The occurrence of rheumatoid factor and other abnormalities in families of patients with agammaglobulinemia.код для вставкиСкачать
The Occurrence of Rheumatoid Factor and Other Abnormalities in Families of Patients with Agammaglobulinemia By HUGHFUDENBERG, JAMESL. GERMAN 111 AND HENRYG. KUNKEL The incidence of rheumatoid agglutinating activity (RAA) in parents and siblings of patients with acquired agammsglobulinemia was found to be considerably higher than the incidence of RAA in the normal population. Elevated y-globulin levels were also occasionally present, as well as alterations in the &, and &N proteins in such relatives. Laboratory findings suggestive of lupus erythematosus were observed in two instances. The high incidence of serum y-globulin abnormalities in relatives of patients with idiopathic “acquired agammaglobulinemia suggests that this is a genetically determined disorder. Le incidentia de rheumatoide activitate agglutinante in parentes e fraternos de patientes con acquirite agammaglobulinemia se revelava como considerabilemente plus aIte que in le population normal. Elevate nivellos de globulina gamma esseva etiam presente in certe casos, si ben como alterationes in le proteinas PaA e BaM. Constatationes laboratorial que poteva suggerer lupus erythematose esseva facite in duo casos. Le alte incidentia de anormalitates de globulina gamma in consanguineos de patientes con idiopathic “acquirite” agammaglobulinemia permitte le supposition que iste disordine es genticamente determinate. R ECENT STUDIES in several laboratories have demonstrated positive scrological reactions for rheumatoid factor in asymptomatic relatives of patients with rheumatoid arthritis.’r2 During the course of similar family studies in patients with various abnormalities in y-globulin, a family was encountered in which several members had positive serological tests although the proband had “acquired agammaglobulinemia”. This finding stimulated an investigation of relatives of patients with both acquired and congenital agammaglobulinemia with tests for rheumatoid factors, certain of the serum factors associated with lupus erythematosus, and other abnormalities of the immunoglobulins. The reports of hypergammaglobulinemia in parents of patients with acquired agammaglobulinemia3 and the findings of hypergammaglobulinemia in some parents of children with Still’s d i ~ e a s eas , ~ well as the observations of an increased incidence of arthritis and other forms of connective tissue disease in agammag l ~ b u l i n e r n i aadded ~ ~ ~ further impetus to this study. It has been clearly shown that patients with agammaglobulinemia and manifestations of arthritis lack rheumatoid factors in their This represents one of a number of arguments against a primary role of these factors in producing the joint lesions. However, with the evidence that rheumatoid factors represent useful markers in family studies in rheumatoid arthritis, it seemed particularly important to test the families of patients showing the combination of agammaglobulinemia and arthritis or other connective tissue disorder. The From The Rockefeller Institute, New York, N . Y. 565 ARTHRITIS AND RHEUMATISM,VOL. 5, No. 6 (DECEMBER), 1962 566 FUDENBERG, GERMAN AND KUNKEL results obtained suggest the possibility that these disorders may be intimately related and perhaps can be associated on a genetic basis. A preliminary report of these findings has been published.x MATERIALS AND METHODS Clinical Material Twenty-two families were studied. In Group I (table I ) probands 1-7 had idiopathic’ “acquired agammaglobulineinia”; in Group I1 (table 2 ) probands 8-22 had typical “congenital agammaglobulinemia” with onset of symptoms prior to age three. Exact classification of individual cases within one or another of the two groups was occasionally very difficult. For example, proband 5 had the onset of clinical symptoiiis of recurrent infections at age one and one-half, but is classed as acquired because y-globulin levels at that time were greater than 200 mg. per cent. Levels, subsequently, slowly diminished, reaching values below 100 mg. per cent for the first time a t age eight. Proband 13, a girl, was included in the “congenital” category, although this may not be justified, both because the occurrence of unequivocal congenital agammaglobulinemia in females has not been fully documented, and because she suffered no verified bacterial infections prior to age five. Patients and family members of families 1-5, 17 and 20 were examined or treated a t the Hospital of the Rockefeller Institute. Families 7-11 were seen at the National Institutes of Health by one of the authors.! All family members personally examined were questioned as to the presence of symptoms of joint pain or morning stiffness, of repeated infection, of chronic respiratory distress, and of alcoholism or liver disease.9 Relatives with positive serologic tests for “rheumatoid factor” had X-rays of the hands and chest. The sera of probands and relatives of the other families included in this study were forwarded by cooperative physicians in New York City and elsewhere in the United States and Canada.’ Clinical and laboratory features of these patients were typical of aganimaglobulinemia. Several of the probands thus included have been the subject of previous case reports,3*43-48and the others were studied at various university hospitals throughout the country. With one exception (proband 5, vide supra), the patients with acquired agammaglobulinemia ranged in age from 36 to 65 at the time of study; their siblings were from 26 to 49 years old with the exception of family 5 (sibling age 8 ) , and family 7 (sibs 49 and 57 years), The patients with congenital agammaglobulineinia were 2 to 18 years old at the time of study, and their parents age 2445. In control families, parents were 24 to 45 years in age, and children 2 to 16 years. Gamma Globulin Measurements Gamma globulin levels were measured by a variety of methods. Electrophoretic assays were carried out on the sera of all the agammaglobulineniic patients and certain of the relatives. These were done primarily by a modification of the zone electrophoretic method in starch-supporting medium.10 The lack of absorption of 7-globulin in the starch-supporting medium allowed more accurate results than were obtainable by paper electrophoresis, particularly in the low range of y-globulin levels. Certain of the determinations were carried out by paper electrophoresis. “No symptoms of leukeumia, lymphoma, or other underlying disease for at least 5 years after the onset of symptoms of agammaglobulinemia. +We are indebted to Dr. John Utz of the NLH for making m m y of these families available for study. ‘We are indebted to the following physicians for making the sera of these patients and their relatives available to us: Drs. I. Zelman (family 6 ) , J. Gould (family 9 ) , J. Nasou (family lo), D. Palmer and W. Hammack (family 12), J. Hardy (family 13), M. Ziff (family 14),T. Tomasi (family 15), W. Henle (family 16), H. Porter (family 17), M. Erlandson (families 19 and 20),J. Bruce and G. Carter (family 21), and C. Brubaker (family 2 2 ) . FAMILIES OF PATIENTS WITH AGAMMAGLOBULINEMIA 567 For purposes of screening, a modified zinc turbidity method of 7-globulin measurement was also employed. This proved uniquely useful for the large-scale testing carried out in this study and therefore is described in some detail. The present modification, herein designated zinc turbidity No. 2 test, was designed to increase the turbidity range encompassing normal and hypogammaglobulinemia levels. Quantitative differences within this range were difficult to evaluate by the older method11 because of the low turbidity obtained, although gross abnormalities were detectable. Various other disadvantagey of the earlier method were also eliminated. A stronger buffer was employed which gave a more stable reagent; higher zinc concentiations were used which were not as sensitive to zinc-binding agents that might be present; also, the distinctly greater precipitation of 7-globulin from normal serum obtained with the newer method reduced the yensitivity to variation in technique. Serum, 0.1 ml., was diluted with 3 ml. of R reagent containing 3.1 Gm. barbital, 2.1 Gm. Na barbital, and 0.8 Gm. (Zn SO,, . 7 H 2 0 ) per one liter distilled water. The standardization and technique of measurement is the same as in the earlier procedure.11 The mean value in 36 normal white adults was 42.2 t 4.2 turbidity units.* All the probands of the present study showed values markedly below the normal range. In Group I (table 1 ) the range was 5 to 20 units,t in Group I1 (table 2) 2 to 18.5 units.+ Values above 55 turbidity units were considered as indicative of a significant elevation of serum 7-globulin. Table 3 illustrates the results of simultaneous turbidity and electrophoretic measurements on the same sera from certain of the individuals of the present study. A plot of these and of additional results indicated a relatively straight line relationship over the range from 1040 units. Below I 0 units, the values appeared less quantitative and simply indicated very low 7-globulin levels. Above 40 units, the correlation became less linear and larger increase in 7-globulin produced relatively smaller increments in turbidity. Occasional discrepancies were encountered in the modcrate hypogammaglobuhernia range such as serum No. 7 in table 3, Also, the sera of some patients with myeloma and macroglobulinemia gave low turbidity values despite high electrophoretic 7-globulin components. Despite these limitations, the procedure proved of considerable value for screening purposes, particularly because of good reproducibility on several specimens obtained at varying time intervals from the same patient. Quantitative determinations of the & and p,, proteins3611 were performed by a modification of the agar diffusion method of Oudin,Zlb using adsorbed antisera specific for these proteins. Gross abnormalities were confirmed by immunoelectrophoresiq. Serologic! Methods All sera of probands and relatives were tested for “rheumatoid agglutinating activity” (RAA) by sensitized human Rho red cell and latex fixation12 methods. Positive sera were further tested by F I1 tanned cell13 and sensitized sheep cell14 methods. Complement fixation reactions with DNA, nucleoprotein, and other antigens were kindly performed by Dr. H. R. Holman. The Rho test was used as the primary screening method in this study since intensive investigations of its mechanism were being carried out during the same period and was performed by the same investigator for both the experimental and control studies. A detailed description follows: A two-tube modification of the test system described independently by Grubblj and by Waller and Vaughanle was employed. This system utilizes human Rho positive red cells as inert carriers and an anti-Rho “incomplete” antibody as the 7-globulin “coating” the indicator cells. The modification employed resulted in a sensitive test which permitted screening of two dozen sera within a two-hour period. *Measured by per cent transmission on a Coleman Jr., spectrophotometer-1 unit = 1% change in transmission. +Inonly two instances ( proband 12, zinc turbidity 10 units and proband 20, zinc turbidity 2 units) was serum obtainable prior to institution of parenteral therapy with y-globulin. Therapeutic doses in the treated patients was widely variable, total monthly dosages ranging from 20 mg. per Kg. to 1400 mg. per Kg. S68 FUDENBERG, GERMAN AND KUNKEL Whole blood was obtained from the same 0 Rh, donor a t periodic intervals, mixed with an equal volume of Alsever’s solution, and stored at 4 C. for not longer than two weeks. For use, 0.5 ml. aliquots were taken, the red cells washed three times with a 50-fold volume of normal saline, and reconstituted to a 2 per cent suspension with saline. An equal volume of an optimal dilution of the standard anti-Rh, “coating” reagent* was then added, the cells mixed by inversion and shaking, and the mixture incubated at 37 C. for one hour. Following incubation, the coated red cells were washed three times with 50-fold volumes of normal saline, and again reconstituted as a 2 per cent suspension in saline. The reactions of the coated cells and of non-coated control cells with 1:5 and 1:40 dilutions of test serum, after incubation for one hour a t 37 C., were read with the aid of a hand lens after spinning for 30 seconds in a serological centrifuge.1 In general, titers obtained with the Rh, test were one fourth as great as those obtained in the latex fixation tests. Identical serologic conclusions (“RAA positive” or “RAA negative”) were obtained with these two methods in 96 per cent of 150 sera in which double blind clinical and serological testings were employed. These sera, supplied by Dr. John Ward, Massachusetts General Hospital, represented 100 patients with rheumatoid arthritis or variants and 50 patients with gout, degenerative arthritis and miscellaneous disorders. RESULTS Serological Parameters Group I : acquired aga~nmaglohri1iiacmia.-Table 1 indicates the high incidence of positive reactions for RAA in relatives of patients with idiopathic acquired agammaglobulinemia. Positive reactions of moderate to high titer were present in two or more test systems in at least one close relative (parent, or sibling) of all of the 7 probands in this group. Although the sensitized human cell (Rhorheumatoid reaction) was the primary test method, positive titers of comparable magnitude were obtained with the latex fixation reaction in all but two instances, and in these the F I1 tanned cell reactions were positive in moderate titer. Positive reactions in the sensitized sheep cell (SSC) tests were obtained in six of the eight sera positive in the other test systems. Of all relatives tested three of eight parents and five of eleven siblings were “seropositive” for RAA; clinical evidence of rheumatoid arthritis was present in one or more parents or siblings in three of the seven families. The relative incidence of positive families in the acquired agammaglobulinemia group was seven out of seven (table 6 ) , despite the fact that relatively few relatives were available per family. Family 1 (fig. 1) is of particular interest and was studied in considerable detail. The presence of RAA and hypergammaglobulinemia in both the mother ( 115) and the sister (111,) of the proband ( 1118) is evident from figure 1. M. Mey, mother of the proband had a zinc turbidity value of 58 units and a y-globd i n level of 1500mg. per cent by starch block electrophoresis quantitation. Her “The same standard diIution of the same anti-Rho serum (MTJ) was used throughout. This reagent (and its optimal dilution) was selected from 36 high-titer anti-Rho sera (titer > 1600) on the basis of results obtained in “checkerboard titrations” with strong and with weak RAA sera. Cells coated with the 1:8 dilution of the MU serum reacted with 1:5 dilutions of weakly positive M A sera (titer 1:20 or 1:40 by latex fixation) and did not give negative (prozone) reactions at dilutions greater than 1:40 with strongly positive rheumatoid sera (titer 1:640,0W by latex fixation). t“Serofuge,” Clay-Adams, New York. 569 FAMILIES OF PATIENTS WITH AGAMMAGLOBULINEMIA Table I.-G?oup I . Acquired Agammaglobulinemia. Some Qualitative and Quantitatiue Parameters of y-Globulin in Acquired Agammaglobulinemia _Families _ _ _ ~ Family members ____ 1. Sca. GSc ( f., age 37) MMey AMey HMey 3. Far. AFar (m., age 56) PFar MSch 3. Ka. LKa (ni., age 41) FKa SKa RKa LKa 4. Al. NA1 (m., age 36) SCa JCa DCa 5. Di. CDi (m., age 10) FDi MDi SDi F I1 CF (nuclei) titer7 Relation to proband Zn No. 2 test* proband mother &ter** brother 13 (156) 58 (1690) 56(1500) 50 0 320 20 0 0 160 640 0 0 0 996 448 0 proband father sister"" 11 (190) 41 45 0 0 320 0 0 2560 0 1024 0 16000 proband father brother brother sister 5 (30) 0 0 0 0 0 0 0 0 448 proband mother half-brother hnlf-brother 45 43 43 41 Itho titer? Latex titerf 0 0 0 0 0 0 6 (70) 43 53 36 0 0 640 0 0 0 0 0 proband 12 (184) 0 0 father mother sister 42.5 39.5 41 40 0 0 16 0 0 proband father mother 17 (345) 0 60 (1950) 40 50 (1450) 0 0 proband ,ister* " 20 (283) 50 brother 52 80 SCC titers 0 32 It; tanned cell test11 0 0 0 0 0 0 0 0 0 0 0 0 0 0 996 16 0 0 0 anticomplementary 448 0 0 0 6. So. JSo (m., age 52) ESo MSo 7 . Hu. RHu (f., age 65) HHu VHu 0 80 128 0 0 0 320 0 0 2560 32 0 0 448 0 0 0 996 0 0 (pos. LE Test) 0 0 "Figure in parenthesis indicates y-globulin, nig. %, determined by starch block electrophoresis. to = neg. at 1:5. 40 = neg. at 1:20. 50 = neg. at 1:8. 110 = neg. at 1:28. 70 = neg. at 1:4. *"Clinical rheumatoid arthritis. titers in the Rho, latex fixation, F I1 tanned cell, and sensitized sheep cell tests were 320,160,996, and 32 respectively. It is evident from figure 1 and from table 4 that RAA and hypergammaglobulinemia were widely distributed throughout 570 FUDENBERG, GERMAN AND KUNKEL Table B.-Group I I . Congenital Agammaglobulinemia. S0m.e Qualitative and Quantitative Parameters of y-Globulin in Congenital Agammaglobulinemia Families Family members 8. Bo. PBo (m., age 17) PBo, sr. FBo JBo PaBo CBo 9. He. DHe (m., age 18) WHe EHe LHe VHe 10. St. RiSt (m., age 11) RaSt Rust RoSt HKo 11. Re. RRe (m., age 18) JRe HRe EHe CRe 12. We. DWe (m., age 4 ) JWe CWe GWe RWe 13. Ha. DHs (f., age 12) FHa MHa RHa RHa PHa 14. Co. ACo (m., age 8) ECo Relation to proband Zn No. 2 teat* proband"" father mother sister sister r;isler 12 (170) 0 59.5 (1980) 0 58 (1670) 80 0 58 0 58 58.5 5 proband father mother sister sister 4.5 58 66 56 52 0 0 0 0 0 proband father mother brother sister 7 43 56 46 47.5 0 0 0 0 0 proband father mother sister hrother 5 45 59 42 66 0 0 0 0 80 0 0 0 0 320 proband f:ither mother brother frat. twin bro 10 34 49 24 23 0 0 0 0 0 0 0 0 0 0 proband father mother sister sister yister 7 (40) 38.5 0 31 37 160 20 0 0 0 0 1280 40 0 0 0 proband"* father 6 (50) 59.5 0 0 49 49 Rho titerf Latex titerj 0 0 80 0 0 20 SCC titer! 0 0 0 8 F I1 CF tanned (nurlei) cell test]] titer11 8 16 0 0 112 0 0 112 64 224 0 0 996 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 16 8 0 0 0 0 0 0 0 0 0 "Figure in parenthesis indicates 7-globulin, nig. 70,determined by starch block elect rophoresis . 10 = neg. at 1:5. 40 = neg. at 1:20. 50 = neg. at 1:8. 110 = neg. at 1:28. 70 = neg. at 1:4. "9Clinical rheumatoid arthritis. 571 FAMILIES OF PATIENTS WITH AGAMMAGLOBULINEMIA Table 2.--Continued Zn Family members ACo PCo (age 21 mos) 15. Da. KDa (in., age 12) KDa VDa EDa GDa 16. An. RbAn (m., age 3 ) RlAn (m., age 3) FAn MAn 17. Ma. RMa (in.,age 8 ) RMa OMa 18. Fo. RFo (m., age 9) FFo JFo 19. Me. LMe (m., age 2) FMe MMe EMe 20. Po. PPo (m., age 4 ) JPO JaPo MPO 21. Vi. CVi (m., age 14) W i MVi 22. Mo. VMo. (m., age I ) JMo MMo Relation to proband No. 2 testC mother sister titert Lattx titer$ 36 21 0 0 0 0 proband father mother sister sister 1 4 (289) 38.5 42 (677) 32.5 49 0 0 0 0 0 0 0 0 0 proband proband (twin) father mother 8 8 (140) 36 43 0 0 0 0 0 0 0 0 proband father mother 11 (230) 55 42 0 0 0 0 0 proband father mother 18.5 41 42 0 0 0 0 0 0 proband father mother brother 6.5 49 50 30 0 proband'* father mother sister 2 45 43 40 0 proband father mother 11 36 0 0 37 0 proband father mother 5.5 42 45 0 0 Rho 0 80 0 0 0 0 0 0 0 0 0 F I1 tanned cell test11 CF (nuclei) titer7 0 0 0 0 0 0 80 0 0 0 SCC titer5 0 112 0 0 0 0 0 0 this kindred. Although the patient's father is not alive, the finding of RAA in one and hypergammaglobulinemiain two of his three living sibs" suggests that similar abnormalities were perhaps present in his serum as well. RAA was present in both paternal and maternal relatives of JFai ( V1 in fig. l), another 'The manner of sudden death in two siblings possibly suggests agammaglobulinemia, though death occurred prior to the introduction of antibiotics. 1. Girl, born deaf, died at age 19, after 3 days of questionable meningitis. 2. Boy, died at age 20, after one day of pulmonary infection. 572 FUDENBERG, GERMAN AND KUNKEL Table 3.-Comparison of Serum y-Globulin Levels of Various Patients of this Series b y Starch Block Electrophoresis and by Zinc No. 2 Turbidity Test Serum 1. L. Ka. 2. A. Co. 3. N. Al. 4. G. Sc. 5. C. Di. 6. A. Far. 7. R. Ma. 8. K. Da. 9. J. Fai. 10. N. Co. 11. G. Ge. 12. Fr. (norm.) 13. A. Mey. 14. F. Bo. ___-_______ ~ ~ ~ Gammaalobulin (electrophoresis) mgm%* Zn SO, Turbidity units 30 50 70 156 184 190 230 289 469 580 617 880 1500 1670 5 6 6 13 12 11 11 _ _ _ 1t _ _ ~ - ~ - 21 27 33 42 36 58 - _ _ -. - “As measured by Fohn reaction in comparison with standard normal 3erum. 0. D. of standard normal serum was converted into mgm of 7-globulin, using a standard curve obtained with micro-Kjeldahl analyses of a preparation of normal fraction I1 human 7-globulin (Lederle No. C-439). Table 4.-Family 1. Aberrations in y-Globulin in Family 1 (Proband G1. Sc., Acquired Agammaglobulinemin) Name Relation G. Sc. Rl. Mey. Ar. Mey. Her. Mey. R. sc. Paternal L. Ku. proband mother sister brother daughter Symbol Fig. 1 Zinc2 Rh* Latext SSCS 0 32 16 F I1 tanned eel& 0 996 448 0 111, 11, 111, 111, I”, 13 58 56 50 36 0 320 20 0 0 0 160 80 0 0 0 0 117 69 20 20 0 0 III,, 1111 142 0 0 10 0 160 40 0 0 0 0 160 160 0 0 0 0 0 0 0 11119 1112, 56 59 56 56 52 49 32 56 64 v* 20 0 0 0 0 0 111, IV2 56 62 0 40 0 40 0 0 256 224 0 0 0 0 TdQi%VeS N. Ku. J. Mey. Ed. Mey. 1). Fr. S. Fr. E. Fai. J. Fai. aunt cousin (son of L. Ku.) uncle cousin cousin cousin cousin grands. of E. Fai. 1% 0 Maternal relatives Ch. Te. Fr. Hen. “0 Comp. fix./( cousin cousin’s son = neg. a t 1:5. to = neg. at 1:20. 40 = neg. at 1:8. $0 = neg. at 1:28. 110 = neg. at 1:4. 573 FAMILIES OF PATIENTS WITH AGAMMAGLOBULINEMIA Table 5.--Incidence of Rheumatoid Agglutinating Activity in Parents Sibs of Patients with Agammaglobulinemia Sibs Parents and Total positive No. tested No. positive No. tested No. positive NO. NO. tested Controls Malc Female 23 23 2 1. 21 25 0 1 92 4 Congenitals Male Female 13 13 2 2 4 14 1 1 44 6* Acquired Male Female 4 4 1 2 'p tp ~ > 0.05. < 0.001. hypogammaglobulinemic subject in this kindred. t EFai ( IIIzO),grandmother of JFai and a cousin of the proband ( IILS) displayed chronic discoid lupus, false positive serology,+ positive complement fixation reactions with nucleoprotein, positive serologic tests for RAA, and joint symptoms suggestive of rheumatoid arthritis. Patient AFar, proband 2, also had clinical features suggestive of disseminated lupus, e.g., acquired hemolytic anemia of moderate severity, severe thrombocytopenia, and splenomegaly ( this combination, often termed Evans' syndrome,l7 is known often to eventuate into clinically recognizable disseminated lupus erythematosus ) . Repeated LE preparations were negative, however, as were complement fixation tests with his serum. Despite the extremely low 7-globulin level, a positive direct antiglobulin ( Coombs' ) reaction was demonstrable. This autoantibody could be recovered by elution and displayed no specificity when tested at 37 c. against a panel of red cells of varying antigenic composition; the eluate failed to react at 20 C. or at 4 C.I* The sister (H.Hu) of one other proband ( 7 , R.Hu) had frank arthritis with Table 6.--Incidence of Rheumatoid Agglutinating Activity in Families of Patients with Aeammuelobulinemia Group Control Congenital agammaglobulinemia Acquired agammaglobulinemia Seropositive families' Familiea with one or both parents seropos. No positive sibs. Families with one or more seropos. sibs. No positive yarents 4/23 3/23 1/23 0/23 4/15" 2115 2/10 1/10 2/6 4/7 7/7t Familiw with sempositive parents and seropositive sibs. _ _ 1At least one seropositive parent or sibling. *p > 0.05 relative to control group. t p < 0.01 relative to control group. t JFai developed recurrent cutaneous, gastro-intestinal and upper respiratory infections without pneumonia at age 10 months; these persisted until age 16 months a t which time a 7-globulin level of 410 mg. per cent was found. Gammaglobulin therapy was then begun and he has subsequently been free of symptoms. +Positive Mazzini and Wasserman. Negative Treponema Pallidum immobilization test. ~ 574 FUDENBERG, GERMAN AND KUNKEL FAMILIES O F PATIENTS WITH AGAMMAGLOBULINEMIA 575 0 Negative serological tests Rheumatoid factor Positive L.E.preparation Increased v-glob. Decreased )-’-glob. t o r N.T Serum not tested R A-Clinical rheumatoid arthritis If Proband Fig. 2.-Aberrations in y-globulin in family 7 (proband RHu acquired agammaglobulinemia) . rheumatoid factor and a positive LE cell test but negative complement fixation reactions against nuclei (fig. 2). The proband ( K.Hu) manifested thrombocytopenic purpura as well as repeated bacterial infection^.^^ Proband 5. ( C.Di ) eventually developed clinical symptoms somewhat suggestive of periarteritis nodosa; tissue biopsy material obtained at exploratory laparotomy was compatible with this diagnosis. The presence of anticomplementary reactions in his serum, despite the marked diminution in y-globillin, may represent an additional qualitative abnormality in y-globulin. Group ZZ: congenital agal7imnglobulin~?~ia.-l’he incidence of RAA in parents and siblings of patients with “congenital” agammaglobulinemia was considerably lower than that observed in families with the “acquired form of the disorder. As is evident from table 2, positive Rho and latex fixation rheumatoid factor reactions were demonstrable in six subjects in 4 of the 15families studied, SSC reactions were weakly positive, and tanned cell reactions moderately positive in almost all of the positive reactions. This 26.7 per cent incidence of families with at least one “seropositive” member is not significantly different from the incidence in the control family groups. The total number of seropositive parents and siblings ( 6 of 44) is also not significantly dihrent. These data are summarized in tables 5 and 6. Three probands in this group had classi- 576 FUDENBERG, GERMAN AND KUNKEL cal rheumatoid arthritis with joint effusions and typical radiologic changes; positive serologic reactions were observed in family members of one of these three probands (proband S ) , and in three additional families as well. Control Studies The incidence of positive Rh,, tests in 28 pairs of “normal control” parents, selected to correspond closely in age and ethnic origin to the parents of the congenital agammaglobulinemic probands and the siblings of the acquired agammaglobulinemic probands, was 6 per cent. Weakly positive reactions were obtained in the Rho test with 3 of these 56 control sera; in two of the three, latex fixation reactions were also positive in low titer. These sera were also used as the control in population for y-globulin measurement. Eleven additional sera were provided by the marital partners of the family members listed in kindred one; none of these sera, tested simultaneously with the sera of family members, gave positive reactions. The value of such marital partner controls in the evaluation of family studies has been emphasized by Leonhart.19 An additional 23 non-selected, presumably normal families in each instance comprising father, mother, and one to four children, were also tested. Only four of these 92 sera gave positive results, an incidence of 4.4 per cent. No control family had more than one seropositive member, so that the incidence of positive families (4/23) was 17.4 per cent. Gamma Globulin Levels Group Z: acquired agammagZobzilinemi~.-The serum 7-globulin concentration, as measured by the zinc turbidity No. 2 test and confirmed by starch block electrophoresis, was elevated in 2 of 7 parents, and 1of 11 adult sibs of patients with acquired agammaglobulinemia, as compared with the concentrations in the 56 normal control sera. Zinc turbidity levels of 56 or more units, corresponding to electrophoretic values of approximately 1500 mg. per cent, were accepted as abnormal. However, there is evidence that 7-globulin concentrations increase in the older-age groups,4Qso that levels even as high as 1500 mg. per cent in the parents of the patients with acquired agammaglobulinemia may perhaps be within normal limits in these elderly individuals. Nonetheless, the values obtained in the siblings of the probands of the acquired group are strictly comparable to those of the control group, as none of these siblings were more than 45 years old.” Tables 7a and 7b demonstrate that hypergammaglobulinemia was present in 2 of the 7 families in this group. In view of the reservations above, statistical analysis of the significance of this finding in this limited number of families was not carried out. Group ZZ: congenital agammaglobu1inemia.-Table 7b also demonstrates that 7-globulin levels were significantly elevated (zinc turbidity of 56 or more units) in 5 of the 15 families of the congenital group. Especially noteworthy *Indeed, the control group was selected to correspond, insofar as possible, to the age range and ethnic distribution of the siblings of the acquired agammaglobulinemia patients and the parents of the congenital agammaglobulinemics. It was not possible to obtain blood samples from the children of the normal donors in the control group. 577 FAMILIES OF PATIENTS WITH AGAMMAGLOBULINEMIA Control population Agammaglobuline m i c patients Ac- Congenuired ital Zinc No.: turbidity units Relatives of congenitals R e l a t i v e s of Sib5 Pa~ents Sibs Parents I I I I * I I I I aJ I I ; 50 I ON; C 6 . . 1 I I I +%+,:a I I I I c--- I I 30 I I I . . , .*.. I I 20 * ; I 2 . 0 : * I I ** 8 I I I I I I . I I I I ' . I I . . I f. I . I . 0. w.: I..*. I I . I 10 I I. I. I I I . I . . I... * I 6 bj I I I I I I 4 I I I I. I I... ..I I r--: I I I 5 I I I 60 acquireds 0. . I. I.. I I I I I I I I I I I I I I I I I I I I I I I I I Fig. 3.-Gamma globulin levels by zinc turbidity test in agammaglobulinemic probands and their relatives. T Negative serological tests Rheumatoid factor 5 Positive L E preparation Increased Y -glob ~ R A Decreased f -glob t o r NT-Serum not tested R A-Clinical rheumatoid arthritis f Proband Fig. 4.-Aberrations in y-globulin in family 8 (proband PBo congenital agammaglobulinemia) . 578 FUDENBERG, GERMAN AND KUNKEL Rheumatoid factor @ Malar butterfly rash, arthralgia negative serologic tests for D.L.E. R.A= Clinical rheumatoid arthritis Fig. 5.-Aberrations maglobulinemia) . in y-globulin in family 13 (proband DHa congenital agam- were the elevated levels in the paternal (3/15) and maternal (4/15) sera. (Elevations occurred in both parents in two families.) The levels in these parents, as a group, were significantly different from that of the control population ( p < 0.01). Values of 56 or more zinc turbidity units were also encountered in 5 of the 19 siblings of the patients with congenital agammaglobulinemia, but such elevations were confined to those families in which elevations also occurred in paternal or maternal sera (table 7 ) . No matched control sera were available comparable to the age range encompassed by the sibs of the congenital agammaglobulinemias. It is noteworthy that positive reactions for RAA were obtained with 4 of the 12 hypergammaglobulinemic sera in the 5 “congenital” families and also in sera with normal 7-globulin levels. Of special interest are the uniformly elevated r-globulin levels in the parents and the three sibs of proband 8, P.Bo. Table 7a.-Zncidence of Elevated y-Globulin Levels” tn Parents and Siblings of Patients with Agammaglobulinemia _ _ Sibs Parents Grouo No. pw./ No. tested Controls Congenital Acquired 0/56 7/30 2/7 D. value: No. pos./ No. tested - - <0.01 6/17 1/11 t D. valuei t >0.05 _ ~ 579 FAMILIES OF PATIENTS WITH AGAMMAGLOBULINEMIA Table 7b.-Families with Zinc Turbidity Values 1 56.0 Group Parents or sibs No. PW. P. value - Control Congenital Acquired 5/15 2/7 Parents No. pos. p. value Sibs No. pos. - 0/28 + <0.01 5.'15 2/6 p. value <0.01 3! 10 115 t >0.05 "Mean for control group 42.2 & 4.2 zinc turbidity #2 units. Elevated y-globulin defined 56.0 units. t t - test comparison with control group. +No adequate control group of comparable age (see text). Hence statistical evaluation not valid. as Table 8.-Group Family members 1. sc. GSc. MMey AMey HMey 2. Far. AFar PFar MSch 3. Ka. LKa FKa SKa RKa LKa RiKa JKa PKa 4. Ale. NA1 SCa JCa 5. Di. CDi FDi MDi SDi 8. So. JSo ESo MSo 7. Hu. RHu HHu VHu I . Acquired Agammaglobulinemia. PBAand Acquired Agammaglobulinemia Families PBarLevels Relation to proband P2.4 P2M proband inother sister brother 0 1.0 1.0 1.0 2.0 1.0 1.0 1.0 proband father sister 0.2 2.0 1.0 proband father brother brother sister son daughter daughter 0 1.0 0.87 1.0 0.67 1.0 0 1.0 1.0 1.0 1.0 3.0 proband mother half-brother 0.125 1.0 1.0 proband father mother sister 0 1.0 1.0 1.0 3.0 1.0 1.0 1.0 proband father mother 0 1.5 1.0 0.06 1.0 1.0 proband sister brother 0 1.0 1.0 in Table 9.-Group ZZ. Congenital Agammaglobulinemia. ,Baal and /IsJu Levels in Congenital Agammaglobulinemia Families ~ Family members 8. Bo. PBo PBo FBo JBo PaBo CBo 9. He. DHe WHe EHe LHe VHe 10. St. RiSt RaSt Rust RoSt HKo 11. Re. RRe JRe HRe EHe CRe 12. We. DWe JWe CWe GWe RWe 13. Ha. 14. Co. 15. Da. KDa KDa, Sr. VDa EDa GDa 16. An. RbAn RlAn FAn MAn 17. Ma. RMa RMa OMa 18. Fo. 19. Me. LMe FMe MMe EMe 20. Po. PPo JPo JaPo 21. vi. 22. Mo. Relation to proband P2* rB2M proband father mother sister sister sister 0 0 1.o 1.0 1.0 1.0 1.0 proband father mother sister sister 0 1.o 0 1.o 1.0 1.0 1.0 proband father mother brother sister 1.0 1.0 1.0 1.0 - proband father mother sister brotlier 0 - proband father mother brother frat. twin brother Not done Not done 0 1.o proband father mother sister sister 0 1.0 1.0 - - 0 1.0 1.o 0.67 1.0 proband proband (Twin) father mother 0 0 1.0 1.0 0 0 1.o 1.0 proband father mother Not done 0 proband father mother brother 0.25 1.0 3.0 - proband father mother Not done Not done 0 1.0 1.o 0 1.o 1.0 1.o 1.0 0.67 1.o 1.5 0.5 0.67 0.5 0.5 580 0 1.0 0.67 - 1.o - - 3.0 1.o 1.0 1.0 1.0 0 1.0 0.33 0.5 1.5 1.o 2.0 FAMILIES OF PATIENTS WITH AGAMMAGLOBULINEMIA 581 It is apparent from figure 4 that quantitative and qualitative abnormalities occurred on both the paternal and maternal side of the family. Qualitative abnormalities without elevation in y-globulin concentration, involving both maternal and paternal relatives was evident in the family of D.Ha, the one proband with questionable congenital agammaglobulinemia included in this series (fig. 5j. The concentrations of the PZAand PZM“immunoglobulins”36*were measured immunochemically in 50 normal sera and the mean value for each of these proteins arbitrarily designated as 1.0 unit.’ Results were then expressed relative to these means. In the normal sera, values ranged from 0.75 to 1.3 units.’ and/or PaM immunoglobulins were ocTables 8 and 9 demonstrate that casionally normal (proband 1) or even elevated ( prohand 5 ) in the acquired agammaglobulinemic patients, but were uniformly absent in the patients with congenital agammaglobulinemia. Similar observations have been made by Bridges and Good.6h Of additional interest are the abnormally high and abnormally low values often encountered in the family members of both groups of probands, with values ranging from three times the normal (e.g., J.Ka, family 3) to complete absence ( CWe mother, family 12). DISCUSSION The serologic evidence obtained in these studies suggests that the incidence of rheumatoid agglutinating activity (RAA) in parents and sibs of patients with idiopathic acquired agammaglobulinemia is significantly higher ( p < 0.001) than that found in normal families. The slight increase in incidence of RAA in parents and sibs of the congenital group is statistically insignificant and in addition may be due to the possible inclusion of acquired cases. The dividing line between the congenital and the acquired groups was frequently difficult to determine and probably a number of subgroups are present. One of the positive families in !he congenital group (Ha, 13) has a female propositus and it is not certain that this family should be classified in this way. Perhaps family investigations will aid in providing better means of classification in the future.. Increased serum 7-globulin levels also were occasionally found in family members of both groups. Such findings, previously cited in acquired agammag l o b ~ l i n e m i awere ~ ~ ~ ~present with even greater frequency in parents of children with congenital agammaglobulinemia. The sibs of children with the congenital form showed an interesting distribution of 7-globulin levels, wme strikingly low, some normal, and some definitely elevated. There was no direct correlation between the presence of RAA and of hypergammaglobulinemia in a given serum; many of the sera with grossly elevated r-globulin values showed no serologic abnormalities. It should be emphasized that the presence of positive serologic tests for RAA in these relatives certainly is not indicative of rheumatoid arthritis. Positive *The normal &A globulin level, as measured by the Oudin tube single diffusion method, appears to be of the order of 1.12 mg/ml. for white adults. A mean value of 0.75 mg/ml. globulin. was found for the 582 FUDENBERG, GERMAN AND KUNKEL results in the latex fixation, tanned cell, and Rho tests have been obtained with considerable frequency in sarcoidosis,21 liver disease,g syphilis,22 and other disease states not associated with joint abnormalities.” Positive serologic tests in these conditions occur primarily with those systems utilizing human 7-globulin as the “reactant”;21the sensitized sheep cell reaction (SSC), employing rabbit y-globulin as the reactant, has been more uniformly negative with such pathologic Positive reactors in the family studies cited herein were positive in the SSC in almost all incidences in which moderate to strong titers were obtained with the other serologic test system. Positive SSC reactions have also been observed in clinically normal subjects, most notably in 5.7 per cent of the 1165 individuals over the age of 25 tested included in the survey by Lawrence and Ball2in a small village in Lancashire, England. This incidence is similar to that observed in normals with the Rh,) method employed in this study. The incidence in “normals” varies greatly in different racial groups inhabiting the same geographic area23but whether or not “normals” with positive results represent familial aberrations cannot be answered from the data presented in this or previous investigators. Ziff’ using the sheep cell inhibition reaction has noted an increased incidence of “rheumatoid factor” in parents and siblings of patients with rheumatoid arthritis. A high familial incidence (20 per cent) of positive reactions was also noted in the Lancashire study in parents and siblings of patients with “seropositive” rheumatoid arthritis”; in “seronegative rheumatoid arthritis” the incidence of positive reactors in relatives did not exceed that of the control population.2 Experience in this laboratory with the syndrome of rheumatoid arthritis accompanied by hepatosplenomegaly, lymphadenopathy, and delayed maturation (Still’s disease) has also indicated a high familial incidence of positive reactor^.^ Also of interest, but of questionable significance, is the appearance of rheumatoid factor and other abnormalities in the 7-globulins in the descendants of the two consanguineous marriages in kindred 1 (see Fig. 1, IIlo and IIIS1). It is noteworthy that y-globulin subgroups under genetic control have been established in both and man,27,28,29 the latter demonstrable by serologic techniques comparable to those employed in the Rho test for RAA. The variations in racial and familial incidence suggest that the protein( s ) responsible for RAA are particularly amenable to genetical control. Assumption of this hypothesis, however, raises fundamental questions as to the interrelationship between “rheumatoid factor” and the disease responsible for its nosologic importance. A possible analogy may be invoked between this laboratory parameter and its corresponding “disease”, and the relationship between erythrocyte-reduced glutathione instability and drug-induced hemolytic anemia. As demonstrated by Beutler30 and others, the glutathione stability test provides a useful laboratory parameter for mensuration of a genetically-determined de“Although some association with the degree of hypergamm;iglobulinemia is present in these latter conditions, the relationship is not a direct one. *“Seropositive” defined as rheumatoid arthritis with positive SSC and “seronegative” defined as rheumatoid arthritis with negative SSC. FAMILIES OF PATIENTS WITH AGAMMAGLOBULINEMIA 583 ficiency in erythrocyte glucose-6-phosphate dehydrogenase. Subjects homozygous for the abnormal gene responsible for the enzymatic defect suffer no clinical symptoms unless exposed to primaquine or other chemical agents,31 but upon such exposure usually develop severe hemolytic anemia. Similwly, the presence of rheumatoid factor perhaps delineates that segment of the population genetically susceptible to development of rheumatoid arthritis following exposure to bacterial, viral, chemical, or other environmental agents. A high prevalence of clinical and/or radiologic signs of mild rheumatoid arthritis was indeed noted among the seropositive relatives of the propositi both in the Lawrence and Ball study,2 and in our own (eg., M.Mey and A.Mey family 1 ) . Thus the rheumatoid factor, like the positive-reduced glutathione stability test, may represent a laboratory index of an inherited biochemical characteristic which predisposes the individual concerned to a certain type of disease. The frequent occurrence of rheumatoid arthritis in multiple family members and especially a frequent simultaneous occurrence in identical twins3z*33lends further credence to this concept. The relationship between the positive serologic reactions and the possible later development of clinical rheumatoid arthritis in these family members is as yet unexplored. However, a number of investigators5*6have noted that rheumatoid arthritis-like syndromes occur not infrequently in patients with agammaglobulinemia of the congenital variety. Good has described the development of manifestations of rheumatoid arthritis or other collagen disease in patients with acquired agammaglobulinemia as Three patients with congenital agammaglobulinemia in our series had typical rheumatoid arthritis. Positive serologic results were obtained in the families of two of these patients. Proband 2, A.Fa. in the acquired group, had clinical findings suggestive of disseminated lupus, and proband 5, C.Di., eventually developed clinical symptoms and tissue biopsy evidence of polyarteritis. Further, the presence of laboratory findings suggestive of lupus erythematosus in relatives of two probands with adult acquired agammaglobulinemia raises the possibility that the diseases rheumatoid arthritis, lupus erythematosus, and agammaglobulinemia are directly interreIated in a manner similar to the familial interrelationships between hypergammaglobulinemia and LE delineated by Leonhardtl9 and by Holman et ales4 Too little data is available to draw any conclusions regarding the mode of inheritance of idiopathic acquired agammaglobulinemia. However, it is apparent that quantitative and qualitative abnormalities in 7-globulin occur at an unusually high rate in family members and probably are present in both the maternal and the paternal relatives of each propositus. The observations reported by Zelman and Lewin3 of elevated 7-globulin levels in parents have been confirmed in this series. In addition, the further abnormality of positive lests for RAA has been found as have occasionally positive tests for lupus. These findings suggest that certain basic immunologic alterations occur in families of acquired agammaglobulinemia patients, resulting in elevation or depression of the 7-globulin levels, as well as in qualitative changes in 7-globulin detectable by various serologic parameters. Although environmental factors may play a role, the great variability of the 7-globulin alterations encountered suggests 584 1 FUDENBERG, GERMAN AND KUNKEL “T” h-0 I1 Rheumatoid factor Increased Y-glob Decraeed J- glob @ mot tested Fig. 6.-Aberrations in y-globulin in the family of a proband with infantile hypogammaglobulinemia (see text). Identical titers for rheumatoid factor in three serologic test systems were present in the sera of the two identical twins, one the mother of the proband. that these are differingphenotypic expressions controlled by more than a single genetic determinant. The multiple abnormalities present in the relatives in family 1 provide further evidence for this concept. The presence of identical titers for RAA in three separate test systems in a set of identical twins, vide infra, is illustrated in figure 6 and is also compatible with an important role of genetic mechanisms in elaboration of the 7-globulin abnormalities cited. These various observations suggest that acquired agammaglobulinemia too is genetically determined, a possibility previously raised by Gitlin et al.35In addition, an absence of the &A and “immunoglobulins”despite normal 7s 7-globulin levels in several relatives of patients with acquired agammaglobulinemia has been observed. These observations lend further support for postulating a hereditary basis for these disorders. The delay in onset of clinical symptoms may reflect a genetic basis for an abiotrophic decline in 7-globulin with aging, differing from the presumed sex-linked recessive mechanism involved in congenital agammaglobulinemia. Further, some subjects with persistent absence of 7-globulin on electrophoresis have shown no increased liability to infecti0n,”7,3~so that the adult form may represent delayed clinical manifestations of a congenital disorder. Observations to be reported elsewhere3gon a group of infants” with moder*These infants did not develop pneumonia, but had unusually frequent cutaneous, gastrointestinal and upper respiratory infections, which no longer recurred following institution of 7-globulin therapy. Hypergammaglobulinemia and RAA have been occasionally noted in their parents and sibling.; and in one instance identical titers of RAA were present in Rho, latex fixation and F I1 tanned cell tests in both the mother of the proband and in her sister, an identical twin (see Fig. 6 ) . FAMILIES O F PATIENTS WITH AGAMMAGLOBULINEMIA 585 ately decreased y-globulin levels (300-450 mg. per cent) at 2-3 years of age, raise the possibility that such infants represent a population reservoir from which clinical cases of adult idiopathic agammaglobulinemia are derived. The presence in kindred 1, figure 1, both of a patient with adult acquired agammaglobulinemia and of an infant with moderate hypogammaglobulinemia and clinical symptoms typical of this group of infants, lend further support to this hypothesis. It is of interest that other known genetically determined disorders, usually congenital, may not become apparent until adulthood. For example, Dent and Harris40 have studied three families originating from a single proband with the De Toni-Fanconi syndrome ( aminoaciduria and renal glycosuria ) ; in one family the disease developed in childhood, but in the other two families it became apparent only in adult life. In particular in one family member aminoaciduria and glycosuria were found at age 42; no glycosuria was noted on three separate urinalyses at age 24 during hospitalization for other reasons. Other familial disorders, presumably genetic, do not express themselves until adult life, e.g., Huntington’s chorea4l and pseudoxanthoma e l a s t i c ~ m . ~ ~ The possible genetic interrelationship between “acquired agammaglobulinemia” and the rheumatoid and lupus factors perhaps raises important therapeutic implications because of the clinical effectiveness of the hydroxy-corticoids in the diseases responsible for the nosologic significance of the above factors. Small doses of prednisone (10 mg. daily) have consequently been administered to the first five of the probands in the “acquired agammaglcbulinemia” group; moderate clinical improvement has subsequently been noted in several respects. Although these patients no longer suffered bacterial pneumonias after introduction of y-globulin therapy, upper respiratory infections characterized by purulent sputum and fever without demonstrable pulmonary changes by physical examination or by X-ray still recurred at regular intervals, ranging from every 23 weeks to every 2-3 months in the individual subjects. A subjective feeling of chronic fatigue was also uniformly present. The addition of hydroxy-corticoids to the therapeutic regimen was followed by the disappearance of both the recurrent upper respiratory infections and the fatigue. The longest period of such therapy is three years (two patients); the other three individuals have been on such medication for one to two years. No adverse effects, particularly as concerns infection, have been noted. Although the mechanism responsible for the actions of the hydroxy-corticoids in such individuals remains uncertain, 7-globulin levels have risen in two of the subjects so treated. A striking increase in 7-globulin levels following institution of corticoid therapy also occurred in JFai family 1 ( Vz in fig. 1 ), the hypogammaglobulinemic infant previously cited (vide supra). Clinical improvement also accompanied his rise in y-globulin. In addition, proband 13 had been treated with steroids for several years by Dr. Janet Hardy at Johns Hopkins Hospital and in her opinion improved on such therapy. These clinical observations, subjective to be sure, suggest that cautious use of hydroxy-corticoids may be of value in selected patients with agammaglobulinemia, when administered in conjunction with parenteral 7-globulin and antibiotics when indicated. 586 FUDENBERG, GERMAN AND KUNKEL CONCLUSIONS The incidence of rheumatoid agglutinating activity (RAA) in parents and siblings of patients with acquired agamrnaglobulinemia has been found to be considerably higher than the incidence of RAA in the normal population. In similar relatives of probands with congenital agamrnaglobulinemia only a few positive families were encountered and these may reflect difficulties in classification. Elevated 7-globulin levels were occasionally found in relatives of both the acquired and congenital patients along with alterations in the and &M proteins. Laboratory findings suggestive of lupus erythematosus were aIso observed in relatives of two of the acquired agammaglobulinemia cases. The high incidence of qualitative and quantitative abnormalities in the serum 7-globulin in relatives of patients with idiopathic “acquired adult agammaglobulinemia suggests that this may also be a genetically determined disorder. The possible relationship of acquired agamrnaglobulinemia to the connective tissue disorders is discussed. ACKNOWLEDGMENTS We are very grateful to Drs. I. Zelman, Janet Hardy, and Morris Ziff for their efforts in collecting sera from the widely scattered family members of families 6, 13, and 14. We are indebted to Dr. A. G. Bearn for helpful discussion and aid with the manuscript, and to Miss Dolores Czerwinski for skilled technical assistance. ADDENDUM Another family has been studied recently with Dr. C . K. Osterland where the mother of a patient with acquired agammaglobulinemia has clinical evidence of systemic lupus erythematosus with positive L.E. cells and nuclear complement fixation reactions. REFERENCES 1. Ziff, M., Schmid, F. R., Lewis, A. J., and Tanner, h.1.: Familial occurrence of the rheumatiod factor. Arth. & Rheumat. 1:392, 1958. 2. Lawrence, J . S., and BalI, J.: Genetic studies on rheumatoid arthritis. Ann. Rheumat. Dis. 17:160, 1958. 3. Zelman, S., and Lewin, H.: Adult agammaglobulinemia associated with multiple congenital anomalies. Am. J. Med. 25:150, 1958. 4. Fudenberg, H. H., and Kunkel, H. G.: To be published. 5. Janeway, C. A,, Gitlin, D., Craig, J. M., and Grice, D. S.: “Collagen disease” in patients with congenital agammaglobulinemia. Tr. A. Am. Physicians 69:93, 1956. 621. Good, R. A., and Rotstein, J.: Rheumatoid arthritis and agammaglobulinemia. Bull. Rheumat. Dis. 10:203, 1960. 6b. Bridges, R., and Good, R. A.: Connec- tive tissue diseases and certain serum protein components in patients with agammaglobulinemia. Ann. New York Acad. Sc. 86:1089, 1960. 7. Vaughan, J. H., and Good, R. A.: Relation of agaminaglobulinemia sera to rheumatoid agglutination reactions. Arth. & Rheumat. 1:99, 1959. 8. Fudenberg, H. H., German, I11 and Kunkel, H. G.: Occurrence of rheumatoid factor and other gamma globulin abnormalities in the families of patients with agammaglobulinemia, J. Clin. Invest. 39:987, 1960. 9. Ticktin, H. T., Fudenberg, H. H., and Kunkel, H. G.: Double-blind serologic studies on 100 consecutive cases of hepatic disease. In preparation. LO. Wallenius, G., Trautman, R., Kunkel, H. G., and Franklin, E. C.: Ultracentrifugal studies of major non-lipide electrophoretic components of normal human serum. J. Biol. Chem. FAMILIES O F PATIENTS WITH AGAMMAGLOBULINEMIA 225:253, 1957. 11. Kunkel, H. G., Ahrens, E. H., Jr., and Eisenmenger, W. J.: Application of turhidimetric methods for estimation of 7-globulin and total lipid to the study of patients with liver disease. Gastroenterology 11:499, 1948. 12. Singer, J. M., and Plotz, C. M.: The latex fixation test. I. Application to the serologic diagnosis of rheumatoid arthritis. Am. J. Med. 21:888, 1956. 13. Heller, G., Jacobson, A. S., Kolodny, M. H., and Kammerer, W. H.: Hemagglutination test for rheumatoid arthritis; influence of human plasma fraction I1 (gammaglohulin) on reaction. J. Immunol. 72:66, 1954. 14. Heller, G., Kolodny, M. H., Lepow, I. H., Jacobson, A. S., Rivera, M. E., and Marks, G. H.: The hemagglutination test for rheumatoid arthritis with characterization of the rheumatoid agglutinating factors by analysis of serum fractions prepared by ethanol fractionation. J. Immunol. 74:340, 1955. 15. Gruhb, R.: Agglutination of erythrocytes coated with “incomplete” anti-Rh by certain rheumatoid arthritic sera and some other sera. The existence of human serum groups. Acta. path. et. microbiol. scandinav. 39: 195, 1956. 16. Waller, hl. V., and Vaughan, J. H.: Use of anti-Rh sera for demonstrating agglutination activating factor in rheumatoid arthritis. Proc. SOC. Exper. Biol. & Med. 92:198, 1956. 17. Dameshek, W.: Hemolytic anemia: Direct and indirect indications, pathogenetic mechanisms and classifications. Am. J. Med. 18:325, 1955. 18. Fudenberg, H. H., and Soloman, A.: Acquired agammaglobulinemia with autoimmune hemolytic disease: graftversus-host reaction? Vox Sang. 6:68, 1961. 19. Leonhardt, T.: Familial hypergammaglobulinemia and systemic lupus erythematosus. Lancet 2: 1200, 1957. 20. Young, I. I., Wolfson, W. G., and Cohn, C.: Studies in serum proteins: agammaglobulinemia in the adult. Am, J. Med. 19:222, 1955. 21. Kunkel, H. G., Simon, H. J., and Fudenberg, H.: Observations concerning 587 positive serologic reactions for rheumatoid factor in certain patients with sarcoidosis and other hypergammaglobulinemic states. Arthr. and Rheumat. 1:289, 1958. 22. Peltier, A., and Christian, C. L.: The presence of “rheumatoid factor” in sera from patients with syphilis. Arth. & Rheumat. 2:1, 1959. 23. Fndenherg, H. H., and Butler, V.: Unpublished observations. 24a. Oudin, J.: L’ “Allotypic” de certains antighnes prothidiques du s6rum. Compt. rend. 242:2606, 2489, 1956. 24h. -: Specific precipitation in gels and its application to iinmunochemical analysis. Methods in M. Res. 5:335378, 1952. 25a. Dray, S., ancl Young, G. 0.: Two antigenically different 7-globulins in domestic rabbits revealed by isoprecipitins. Science 129:1023, 1959. 25h. -, and -: Genetic control of two y-globulin isoantigenic sites in domestic rabbits. Science 131:738, 1960. 26. Duhiski, S., and Skdha, D., Duhiski, A., and Kclus, A,. Antigens of rabbit 7-globulin. Natnre, London 184: 1811, 1959. 27. Gruhb, R., and Laurell, A. B.: Hereditary serological human serum groups. Acta path. et microbiol. Scandinav. 39:390, 1956. 28. Harboe, M.: A new hemagglutinating substance in the gm-system, anti-gm. Nature, London 183:1468, 1959. 29. Harboe, M., and Lundevall, J.: A new type in the gm-system. Acta path. et microbiol. Scandinav. 45:357, 1958. 30. Beutler, E.: The hemolytic effect of priinaquine and related compounds: a review. Blood 14: 103, 1959. 31. Childs, B., Zinkam, W., Browne, E. A., Kimhrall, E. L., and Torbert, J. V.: Genetic btudy of a defect in glutathione metabolism in the erythrocyte. Bull. Johns Hopkins Hosp. 102:21, 1958. 32. Thymann, G.: Polyarthritis in twins. Acta Genet. 7:148, 1957. 33. Berglund, S.: Enaggiga twillingar med kronisk polyarthrit och cancer mammae. Nord. Med. 8:2272, 1940. 34. Holman, H. R., and Deicher, H. R.: The appearance of hypergammaglohu- SS8 FUDENBERG, GERMAN AND KUNKEL linemia, positive serologic reactions for rheumatoid arthritis and complement fixation reactions with tissue constituents in the sera of relatives of patients with systemic lupus erythematoms. To be published. 35. Gitlin, D., Janeway, C. A., Apt. L., and Craig, J. M.: Agammaglobulinemia. In Cellular and Humoral Aspects of Hypersensitive States. H. S. Lawrence, ed., Hoeber-Harper, New York, 1959. 3Ga. Heremans, J. F.: Immunochemical studies on protein pathology. The immunoglobulin concept. Clin. Chim. Acta 4:639, 1959. 36b. Heremans, J. F., Fudenberg, H., German, J. L., and Knnkel, H. G.: In preparation. 37. Brem, T. H., and Morton, M. E.: Defective serum y-globulin formation. Ann. Int. Med. 43:465, 1955. 38. Chaplin, H.: Personal communication. 39. Fudenberg, H. H., and Goldfarb, A.: Hypogammaglobulinemia and its relationship to the allergic diathesis. New York J . Med. 61:2721, 1961. 40. Dent, C. E., and Harris, H.: The genetics of cystinuria. Ann. Eugenics, 16: 60, 1951. 41. 1)enny-Brown, D.: Diseases of the basal ganglia and subthalamic nuclei. In Oxford System of Medicine edited by H. Chrutian, revised ed. New York, Oxford University Press, 1946, Vol. VI, Chapter 11. 42. McKusick, V.: Heritable Disorders of Connective Tissue. St. Louis, C. V. Mosby Co., 1956. 43. Oberman, J. J.: Agammaglobulinemia, Clin. Proc. Child. Hosp. Washington. 12:3, 1956. 44. Henley, L.: Hypogammaglobulinemia and hypergammaglobulinemia, J. Mt. Sinai Hosp. 2G:138, 1959. 45. Carson, M. J., and Donnell, G. N.: Vaccinia gangrenosa. California Med. 85: 335, 1956. 46. Sorenson, C., and Partington, C. N.: Primary agaminaglobulinemia as a cause of repeated infection in childhood. Canad. M. A. J. 75:579, 1956. 47a. Martin, C. M., Gordon, R. S., and McCullough, N. B.: Acquired hypogammaglobulinemia in an adult. New England J. Med. 254:449, 1956. 47b. German, J. L., and Utz, J.: To be published. 38. Porter, H. M.: Immunologic studies in congenital agammaglobulinemia with emphasis on delayed hypersensitivity. J. Pediat. 20:958, 1957. 49. Bearn, A. G.: Personal communication. H u g h Fudenberg, hl.D., Associate Professor, University of Californh Medical Center, Sun Francisco, Calif. J U ~ J W S L. Gennccn 111, M.D., Assistant Professor, The Rockefeller Institute, New York, N . Y. Henry G. Kunkel, M.D., Professor, The Rockefeller Institute, New York, N . Y.