Proc. Nati. Acad. Sci. USA Vol. 84, pp. 3826-3830, June 1987 Immunology Prematurity, hypogammaglobulinemia, and neuropathology with human immunodeficiency virus (HIV) infection (acquired immunodeficiency syndrome/human immunodeficiency virus antibody/intracranial calcification) RAJENDRA PAHWA*t, ROBERT A. GOOD*, AND SAVITA PAHWAt *Immunobiology Department, All Children's Hospital, University of South Florida, St. Petersburg, FL 33701; and tPediatric Immunology Division, North Shore University Hospital, Cornell University Medical Center, Manhasset, NY 11030 Contributed by Robert A. Good, February 4, 1987

ABSTRACT Infection with the human immunodeficiency that develops in infants with HIV infection may be a bad virus (HIV) is characteristically associated with hypergam- prognostic feature. Furthermore, neurological complications maglobulinemia in both adult and pediatric cases. We report occur and appear to be severe in such hypogammaglobulin- herein four infants who had an HIV infection in association emic infants. with severe hypogammaglobulinemia and did not exhibit antibodies against HIV. HIV was isolated antemortem or AND METHODS postmortem in all four infants from either peripheral blood, MATERIALS cerebrospinal fluid, or body tissues. HIV infection could be Study Subjects, The four patients described herein were presumed to be acquired transplacentally in two infants and by each born prematurely between 26 and 30 weeks of gestation way of infected blood transfusions during the neonatal period (Table 1). Two infants (patients 1 and 2) were born to HIV in the other two. Each infant became symptomatic within the antibody-positive, asymptomatic mothers; the first was an first year of life and developed rapidly progressive manifesta- intravenous drug user, and the second probably had sexually tions of the disease. Features that were common to all four acquired the infection. The other two infants (patients 3 and infants include premature birth, failure to thrive, hepatomeg- 4) were born to healthy mothers. The HIV serology was aly, and progressive neurological abnormalities that were negative on both parents in each ofthe latter infants who had, associated with intracranial calcifications. We conclude that, however, been given repeated transfusions in the neonatal when infection occurs early in development either by trans- period prior to availability of HIV antibody screening. These placental exposure to the virus or from blood transfusion in infants were presumed to be infected by HIV-infected blood small premature infants, hypogammaglobulinemia and defi- transfusions; the blood donor of patient 3 was subsequently ciency of antibody production leading to the absence of proven to have an HIV infection. antibody responses on which diagnosis is usually based can The clinical manifestations in the four patients are recorded occur. Furthermore, progressive central nervous system dis- in Table 2. Patient 1 became symptomatic between 7 and 8 ease may be a frequent finding in such infants, and this may months ofage. Patient 1 was followed from birth, and patients lead to cerebral calcifications that must be attributed to the 2, 3, and 4 were followed from the ages of5, 8, and 10 months, HIV infection itself and not to complicating infections-e.g., respectively. Patient 3, who had radiological evidence of toxoplasmosis. It is suggested that patients with hypogamma- interstitial pneumonitis, was found by biopsy to have globulinemia, antibody deficiency syndrome, and central ner- lymphoid interstitial pneumonitis negative for Pneumocystis vous system disease have an extremely bad prognosis. carinii pneumonia, bacteria, fungi, and all other identifiable viruses. All infants had progressive retardation of their Infection with the retrovirus, human T-lymphotropic virus growth and development, and each exhibited clear evidence type III/lymphadenopathy-associated virus (HTLV-III/ of central nervous system (CNS) involvement. Evidence of LAV), now known as human immunodeficiency virus (HIV), brain atrophy and CNS calcification was present in the results in a wide spectrum of clinical manifestations, ranging computerized tomographic scans of each ofthe four patients; from asymptomatic states to the severest disease forms examples of the CNS calcifications are illustrated in Fig. 1. associated with the acquired immunodeficiency syndrome Cytomegalovirus was repeatedly isolated from the urine and (AIDS) (1-12). This virus infects preferentially T lympho- saliva of only one infant, patient 2. Severe and recurrent cytes bearing the T4 antigen and also seems to exhibit bacterial infections resulting in pneumonia, meningitis, or preference for cells, especially macrophages, of the central bacteremia ultimately occurred in each of the patients. All nervous system (13, 14). Among the immunological distur- patients were treated with intravenous gamma globulin be- bances resulting from infection with HIV, hypergammaglob- cause they exhibited hypogammaglobulinemia. The dose of ulinemia appears to be a consistent finding in adults as well gamma globulin was gradually increased over a 4-week as in most children (5, 15-17). The presence of antibodies to period from 100 to 300 mg/kg and was given at this dose every HIV in serum is considered to be a hallmark ofthis infection. 2-4 weeks, depending upon compliance. Patients 1 and 2 In this report four infants are described in whom HIV were started on intravenous injections of gamma globulin at infection was documented by positive virological cultures; 6 months of age, whereas patients 3 and 4 first received they also exhibited profound hypogammaglobulinemia and injections at 13 months of age. Only patient 2 had adverse did not produce HIV antibodies. These infants were born reactions to the gamma globulin, which often precluded prematurely and had illnesses characterized by severe and administration of the desired dose. All patients died of sepsis progressive neurological disease. Although the protective at 7, 22, 21, and 17 months of age, respectively. Autopsy was role of HIV antibodies has not been established, these performed on patient 1. This analysis revealed generalized observations suggest that severe hypogammaglobulinemia Abbreviations: HIV, human immunodeficiency virus; AIDS, ac- The publication costs of this article were defrayed in part by page charge quired immunodeficiency syndrome; CNS, central nervous system. payment. This article must therefore be hereby marked "advertisement" 1To whom reprint requests should be addressed at: All Children's in accordance with 18 U.S.C. §1734 solely to indicate this fact. Hospital, 801 6th Street South, St. Petersburg, FL 33701.

Downloaded by guest on September 29, 2021 3826 Immunology: Pahwa et al. Proc. Natl. Acad. Sci. USA 84 (1987) 3827

Table 1. Patient profiles Gestational Birth weight, Neonatal Neonatal Onset of other Patient age, wk g complications transfusion symptoms, mo Clinical outcome Route of infection 1 26 1090 IRDS, sepsis + 3 Died, 7 mo.; sepsis Intrauterine/prenatal; HIV Ab+ mother; IVDA 2 30 1040 Sepsis - 7 Died, 22 mo.; sepsis Intrauterine/prenatal; HIV Ab+ mother; sexually acquired infection 3 28 1080 Sepsis + 8 Died, 21 mo.; sepsis; Infected blood transfusion respiratory failure 4 26 850 IRDS + 7 Died, 17 mo.; sepsis; Infected blood transfusion respiratory failure IRDS, idiopathic respiratory distress syndrome; IVDA, intravenous drug abuser; Ab+, antibody-positive. lymphocyte depletion and the presence ofmicroglial nodules, Quantitative serum immunoglobulin levels were determined calcification, and atrophy of cerebral cortex of the brain. by radial immunodiffusion. Parents refused to permit autopsies on the other three patients. RESULTS Laboratory Methods. Serologic testing for HIV antibodies was performed by immunological blot analysis (18). Specif- Serum immunoglobulin levels and results of HIV antibody ically, the presence of the anti-p41 antibody to HIV was testing are shown in Table 3 for all four infants. In patients 1 considered as evidence of HIV infection. HIV cultures were and 2, serum IgG levels were within normal limits for age at performed on peripheral blood lymphocytes of all four 1 and 5 months, respectively, but they were decreased at 3 months of age in patient 1 and at 7 months of age in patient patients and on cerebrospinal fluid of patient 4. For these 2. Patients 3 and 4 had reduced IgG levels at the time of studies, lymphocytes from peripheral blood or cerebrospinal presentation (i.e., at 8 and 10 months of age, respectively). fluid were cultured with phytohemagglutinin, grown in the Patients 1 and 2, who were initially positive for HIV anti- presence of interleukin 2-stimulated lymphocytes, and sub- body, became negative after 6 months of age. It is presumed sequently cocultivated with an uninfected clone of the H-9 that the antibodies in the infants' serum were passively cell line, as previously described (19). Virus production in the transferred to the infants from their HIV-positive mothers. infected cell clones was monitored by assaying for reverse Patients 3 and 4 were negative for HIV antibodies when they transcriptase activity in culture supernatants and by electron were first tested at 8 and 15 months of age, respectively. microscopy. In patient 1, tissues obtained at autopsy from Serum IgA levels were ultimately low in all four patients. brain and lung were cultured for HIV; the virus was isolated Serum IgM levels were elevated in patients 2 and 3 at the time from primary cultures of lung tissue and was propagated in of initial diagnosis and became low after 14 months of age. peripheral blood-derived macrophages ofnormal donors (20). Sometime during the course of the disease, three of the Virus expression was tested by an immunofluorescence patients had low levels of IgM. The fourth patient, patient 2, assay using a monoclonal antibody to p17 and by measure- had elevated or normal IgM levels early in life; these levels ment of reverse transcriptase activity (20). dropped in subsequent months, but they were still in the T and B lymphocytes, and cells of the major T-cell subsets normal range prior to death. that bear T4 or T8 antigens on their surfaces, were quantified All of the children were treated with intravenous gamma in whole blood using monoclonal antibodies (Ortho-immune globulin (IgG) for their hypogammaglobulinemia as described and Coulter clone reagents) and flow cytometry (21). The above. functional capacity of peripheral blood lymphocytes was HIV was isolated from peripheral blood lymphocytes of assessed investigating the proliferative responses of the patients 2, 3, and 4 and from cerebrospinal fluid of patient 4. by In patient 1, virus was also isolated from autopsy material in lymphocytes to phytomitogens, namely phytohemagglutinin, lung macrophages as described above (20). concanavalin A, and pokeweed mitogen, as determined by Table 4 summarizes the results of analyses of lymphocyte incorporation of [14C]thymidine (New England Nuclear) (22). surface markers. Absolute lymphocyte counts were de- creased in two patients, and a depression in the number ofT4 Table 2. Clinical characteristics of patients under study cells, with absolute counts less than 400/mm3, was noted in Patient three ofthe four patients. The T4/T8 ratios were less than 1.0 Manifestation 1 2 3 4 in all four infants. B-cell numbers, which are usually in- creased in HIV infection, were decreased in each of the three Severe infections infants tested. Results of immune testing are summarized in Opportunistic* - - - + Table 5. Functional lymphoproliferative responses to the Bacterial + + + + phytomitogens phytohemagglutinin, concanavalin A, and Lymphocytic interstitial pneumonitist - - + pokeweed mitogen were severely depressed in two infants Neurological abnormalities (brain (patients 1 and 2), and response to concanavalin A was atrophy, calcifications) + + + + depressed in patient 3. Normal proliferative responses were Hepatomegaly + + + + observed in patient 4 at the time of testing. Splenomegaly + - - + Failure to thrive + + + + Growth retardation + + + + DISCUSSION Lymnphadenopathy + + + + Documentation of HIV antibodies in serum has proved to be Thrush + - + + a reliable and practical means of establishing the presence of Clubbing of nails - + + HIV infection (18). On rare occasions, a transient or persis- *Pneumocystis carini pneumonia. tent seronegative virus-infected state occurs and poses a tBiopsy-proven lymphocytic interstitial pneumonitis. potential problem in serodiagnosis of HIV infection (23). In Downloaded by guest on September 29, 2021 3828 Immunology: Pahwa et al. Proc. Natl. Acad. Sci. USA 84 (1987)

FIG. 1. Computerized tomographs of the brain from patient 3. Cuts at various levels show cortical atrophy, ventricular dilatation, and intracranial calcifications.

this report, we describe four infants with HIV infection, who infected with HIV. Presumably this immunity status is a developed severe hypogammaglobulinemia and were HIV consequence ofthe hypogammaglobulinemia. The features of antibody-negative to this virus, although they were clearly all four of these infants included premature birth and pro- gressive development of severe neurological abnormalities. The latter consisted of intracranial calcifications, cerebral Table 3. Immunoglobulin levels and HIV infection atrophy, and clinical evidence of cerebral palsy. Test results B-lymphocyte dysfunction as an early manifestation of Patient HIV infection is usually characterized by profound polyclo- and age IgG, mg/dI IgA, mg/dl IgM, mg/dl HIV-Ab* nal hypergammaglobulinemia (5, 15-17, 24). In a previous Patient 1 study (5), hypergammaglobulinemia was consistently noted 1 mo. 382 9 33 ND in HIV-infected infants, ranging in age from 3 months to 1 3 mo. 133 <7 20 + year, who had typical clinical features of AIDS or AIDS- 6 mo. 25 8 14 related complex and were seropositive for HIV antibodies. 7 mo.t 1220 <7 35 Hypogammaglobulinemia occurring in association with HIV Patient 2 infection represents a new phenomenon, and its relationship 5 mo. 333 26 158 + to those situations in which HIV-infected patients are sero- 7 mo.t 181 34 127 negative despite normal or increased levels ofimmunoglobins 11 mo.t 149 25 97 ND needs to be investigated. Already, however, it has been 14 mo.t 169 14 37 shown that the levels of IgG2 and IgG4 subclasses may be 18 mo.t 374 <7 55 low in patients with HIV infection, while hypergammaglob- Patient 3 ulinemia attributable to increased levels of IgGi and IgG3 8 mo. 147 16 52 occurs in the same patient (25). Furthermore, hypogam- 9 mo. 72 ND ND ND maglobulinemia has been encountered in association with 10 mo. 42 ND ND selective T-helper cell immunodeficiency in patients with 14 mo.t 704 <7 18 severe combined immunodeficiencies or common variable Patient 4 immunodeficiencies (26, 27). Thus, it seems quite possible 10 mo. 100 11 23 ND that the hypogammaglobulinemia seen in these infants may 13 mo.t 364 <4 26 ND) be attributable to profoundly defective functions of T-helper 15 mo.t 250 <2 10 cells, defective T cell-B cell interactions, or even deficiencies Values less than 2 SD for age are in boldface. ND, not done. in the function of B cells themselves (24). *Antibodies to HIV in serum, determined by immunological blot. In regard to the B-cell pathology in HIV infection, it has tPatient on intravenous gamma globulin infusions. been shown already that proliferating B cells infected with Downloaded by guest on September 29, 2021 Immunology: Pahwa et al. Proc. Natl. Acad. Sci. USA 84 (1987) 3829 Table 4. Lymphocyte markers, percentages, and absolute numbers Percentage and absolute counts of markers T3 T4 T8 T11 B1 Patient and age % No. % No. % No. % No. % No. T4/T8 ratio Patient 1 3 mo. 35 441 24 307 54 684 75 942 2.1 27 0.4 Patient 2 7 mo. 68 3002 34 1501 43 1898 76 3356 3.6 159 0.8 Patient 3 15 mo. 34 275 9 73 52 421 48 389 ND ND 0.2 Patient 4 15 mo. 38 775 10 204 25 511 36 736 0.9 18 0.4 Values falling outside of the normal range for age are in boldface. For T8, boldface numbers indicate increased values; for all others, boldface numbers indicate decreased values. ND, not done. another virus can grow the HIV (28). It is possible that series (34). As yet, an association between hypogammaglob- different subpopulations of B lymphocytes react differently ulinemia and tropism of the HIV for macrophages has not to the influence ofthe virus; some are stimulated while others been described. It is possible that monocyte dysfunction are inhibited. Even in the absence ofdirect infection ofB cells might further contribute to the failure of the development of by HIV, different components of the HIV might indirectly a normal capacity to produce antibodies in such patients. exert either stimulatory and/or inhibitory influences (29). In An additional concurrent viral infection might be respon- fact, two patients reported herein, in spite of having low sible for the CNS disease observed in these children. This is, levels of serum IgG, had high levels IgM at some time during however, an unlikely possibility, but it is one that cannot be the first year of life. This recalls the dysgammaglobulinemia conclusively ruled out in each ofthese patients because ofthe first described with virus infections in patients with the in hypogammaglobulinemia and consequent absence of sero- utero infections caused by the rubella virus (30, 31). The logical responses that otherwise might be expected-for infants described herein also had a paucity of B cells, which example, toxoplasma, rubella, cytomegalovirus, or Epstein- is unusual in patients with HIV infection. This abnormality in Barr virus. However, cerebrospinal fluid cultures in one B-cell numbers was unrelated to the mode of acquisition of patient were negative for cytomegalovirus and also were the HIV infection-i.e., whether it occurred as a conse- negative for other herpes or enteric viruses and, at the same quence of in utero infection or after birth. It is possible that time, were positive for HIV. The presence of HIV in the in the premature infant infected with HIV the B lymphocytes, cerebrospinal fluid of patient 4, the autopsy findings of the which are characteristically immature in the newborn period brain in patient 1, and the radiological and clinical abnormal- (32, 33), are more readily compromised in their development ities of the brain in all these patients raises the likelihood that than at a later time in life. Only further studies will ascertain a primary HIV infection involving the brain has occurred in just how and in what cells the primary influence of the virus all four of these children. The mechanism whereby such an to inhibit the development of a normal capacity for immu- infection would result in intracranial calcifications is as yet noglobulin production is exercised. unclear, but this finding should be included in the pathologic Accumulating clinical, pathological, and virological evi- features of HIV infection. dence has firmly established the neurotropic potential ofHIV Thus, these patients who were infected with HIV virus (3, 13, 14). In brain tissues of AIDS patients, HIV sequences very early in life present evidence both of a threatening have been demonstrated by Southern blot analysis of host neurotropism of the HIV virus, especially for very young cell DNA. Further, HIV messenger RNA has been detected children, and challenging new evidence of the complexity of by in situ hybridization (13). In a recent study, Popovic and the lymphoid cellular involvement in this disease, which now coworkers (20) demonstrated selective tropism by HIV viral must be recognized as potentially leading to hypogam- isolates for macrophages obtained either from brain tissue of maglobulinemia or agammaglobulinemia. Although a protec- an AIDS patient or from lung macrophages ofan infant. It has tive role for HIV antibodies has not been conclusively been suggested that in certain patients, such as patient 1, the established, the absence of such antibodies in the cases macrophages represent a primary target for transmission of described herein was associated in each instance with rapidly infection to the CNS. In agreement with this concept is progressive clinical deterioration. A heightened awareness of another report that shows that virus-positive cells in the brain HIV seronegativity occurring as a consequence of hypogam- tissue ofAIDS patients belong to the mononuclear phagocyte maglobulinemia has important consequences for the serodi- agnosis of HIV infection in children. Table 5. Summary of immunologic abnormalities Patient We are indebted to C. Leombrumo and M. Marecki for technical Abnormality 1 2 3 4 assistance and to A. Johnston for secretarial assistance. We thank Drs. L. Resnick and M. Popovic for HIV viral isolation experiments Hypergammaglobulinemia (>2 SD for age) - - - and Dr. M. Sarngadharan for HIV serology. This work was support- Hypogammaglobulinemia (<2 SD for age) + + + + ed by a grant from the New York State AIDS Society (C001545) and Absolute lymphopenia (<1500/mm3) + - + a grant from the National Institute of Aging (AG05625). T4 lymphopenia (<400/mm3) + - + + Reversed T4/T8 ratio (<1.0) + + + + 1. Gallo, R. C., Salahuddin, S. Z., Popovic, M., Kaplan, G. M., Depressed lymphoproliferative responses Haynes, M., Palker, B. F., Redfield, T. J., Oleske, R., Safai, (<2 SD)* + + + - J., White, G., Foster, R. & Markham, P. D. (1984) Science *Decrease of more than 2 SD from the normal mean lymphoprolif- 224, 500-502. erative responses to phytomitogens, phytohemagglutinin, con- 2. Broder, S. & Gallo, R. C. (1984) N. Engl. J. Med. 511, canavalin A, or pokeweed mitogen. 1292-1297. Downloaded by guest on September 29, 2021 3830 Immunology: Pahwa et al. Proc. Natl. Acad. Sci. USA 84 (1987)

3. Fauci, A. S., Macher, A. M., Longo, D. L., Lane, H. C., 19. Popovic, M., Sarngadharan, M. G., Read, E. & Gallo, R. C. Rook, A. H., Masur, H. & Gelmann, E. P. (1984) Ann. Intern. (1984) Science 224, 497-500. Med. 100, 92-106. 20. Gartner, S., Marcovits, P., Markovitz, D. M., Kaplan, M. H., 4. Levy, J. A. & Shimabukuro, J. (1985) J. Infect. Dis. 152, Gallo, R. C. & Popovic, M. (1986) Science 233, 215-219. 734-738. 21. Pahwa, S., Sia, C., Te, C. & Pahwa, R. (1985) Pediatrics 76, 5. Pahwa, S., Kaplan, M., Fikrig, S., Pahwa, R., Sarngadharan, 914-917. M. G., Popovic, M. & Gallo, R. C. (1986) J. Am. Med. Assoc. 22. Cunningham-Rundles, S., Hanson, J. A. & Dupont, B. (1976) 255, 2299-3305. in Clinical Immunology, eds. Bach, F. H. & Good, R. A. 6. Thomas, P. A., Jaffe, H. W., Spira, T. J., Reiss, R., Guerrero, (Academic, New York), Vol. 3, pp. 646-672. I. C. & Auerbach, D. (1984) J. Am. Med. Assoc. 252, 639-644. 23. Salahuddin, S. Z., Groopman, J. E., Markham, P. D., Sarn- 7. Levy, J. A., Kaminsky, L. S., Morrow, W. J., Steimer, K., gadharan, M. G., Redfield, R. R., McLane, M. F., Essex, M., Luciw, P., Dina, D., Hoxie, J. & Oshiro, L. (1985) Ann. Sliski, A. & Gallo, R. C. (1984) Lancet fi, 1418-1420. Intern. Med. 103, 694-696. 24. Pahwa, S. G., Fikrig, S., Menez, R. & Pahwa, R. (1986) 8. Oleske, J., Minnefor, A., Cooper, R., Thomas, K., Dela Cruz, Diagn. Immunol. 4, 20-24. A., Ahdieh, H., Guerrero, I., Joshi, V. V. & Desposito, F. 25. Aucouturier, A., Bremard-Oury, C., Clauvel, J. P., Debre, (1983) J. Am. Med. Assoc. 249, 2345-2349. M., Griscelli, C., Seligmann, M., Preud'homme, J. L. (1986) in 9. Scott, G. B., Buck, B. E., Leterman, J. G., Bloom, F. L. & Monographs in Allergy, eds. Dukor, P., Kallos, P., Trnka, Z. Parks, W. P. (1984) N. Engl. J. Med. 310, 76-81. & Klaksman, B. H. (Krager, Basel), Vol. 20, pp. 203-225. 10. Rubinstein, A. (1983) Am. J. Dis. Child. 137, 825-827. 26. Pahwa, S. G., Pahwa, R. N. & Good, R. A. (1980) J. Clin. 11. Rogers, M. F. (1985) Pediatr. Infect. Dis. 4, 230-236. Invest. 66, 543-550. 12. Ammann, A. J., Cowan, M. J., Wara, D. W., Weintrub, P., 27. Siegal, F. P., Siegal, M. & Good, R. A. (1978) N. Engl. J. Dritz, S., Goldman, H. & Perkins, H. (1983) Lancet i, 956-958. Med. 299, 172-177. 13. Shaw, G. M., Harper, M. E., Hahn, B. M., Epstein, L. G., 28. Montagnier, L., Gruest, J., Chamaret, S., Dauguet, C., Axler, Gajdusek, D. C., Price, R. W., Navia, B. A., Petito, C. K., C., Guetard, D., Nugeyre, M. T., Barre-Sinoussi, F., Cher- O'Hara, C. J., Cho, E., Oleske, J. M., Wong-Staal, F. & mann, J. C., Brunet, J. B., Klatzmann, D. & Gluckman, J. C. Gallo, R. C. (1985) Science 227, 177-181. (1984) Science 225, 63-66. 14. Epstein, L. G., Sharer, L. R., Joshi, V. V., Fojas, M. M., 29. Pahwa, S., Pahwa, R., Saxinger, C., Gallo, R. C. & Good, Koenigsberger, M. R. & Oleske, J. M. (1985) Ann. Neurol. 17, R. A. (1985) Proc. Natl. Acad. Sci. USA 82, 8198-8202. 488-4%. 30. Dent, P. B., Olson, G. B., Good, R. A., Rawls, W. E., South, 15. Seligmann, M., Chess, J. L., Fahey, A. S., Fauci, A. S., M. A. & Melnick, J. L. (1968) Lancet i, 291-293. Lachmann, P. J., L'Age-Stehr, J., Ngu, J., Pinching, A. J., 31. Soothill, J. E., Haynes, K. & Dudgeon, J. A. (1966) Lancet i, Rosen, F. S., Spira, T. J. & Wybran, J. (1984) N. Engl. J. 1385-1387. Med. 311, 1286-1292. 32. Wu, L. Y., Bianco, A., Cooper, M. D. & Lawton, A. R. 16. Lane, C. H., Masur, H., Edgar, L. C., Whalen, G., Rook, (1976) Clin. Immunol. Immunopathol. 5, 208-217. A. H. & Fauci, A. S. (1983) N. Engl. J. Med. 309, 453-458. 33. Tosato, G., Magrath, I. T., Koski, I. R., Dooley, N. J. & 17. Pahwa, S. G., Quilop, M. T. J., Lange, M., Pahwa, R. & Blaese, M. R. (1980) J. Clin. Invest. 66, 383-388. Grieco, M. (1984) Ann. Intern. Med. 101, 757-763. 34. Koenig, S., Gendelman, H. E., Orenstein, J. M., Dal Canto, 18. Sarngadharan, M. G., Popovic, M., Schupback, J. & Gallo, M. C., Pezeshkpour, Yungbluth, M., Janotta, F., Aksamit, A., R. C. (1984) Science 224, 506-508. Martin, M. & Fauci, A. S. (1986) Science 233, 1089-1093. Downloaded by guest on September 29, 2021