Untitled Impaired Culture Generated Cytotoxicity with Preservation of Spontaneous Natural Killer-Cell Activity in Cartilage-Hair Hypoplasia GLENN F. PIERCE, CHARLOTTE BROVALL, BERNICE Z. SCHACTER, and STEPHEN H. POLMAR, Departments of Pathology and Pediatrics, Case Western Reserve University School of Medicine, Cleveland, Ohio 44106 A B S T R A C T Recent studies of cartilage-hair hypo- plasia (CHH), a form of short-limbed dwarfism, have shown that all affected individuals have a cellular pro- liferation defect that results in a cellular immunode- ficiency. However, only a minority of CHH individuals suffer from severe, life-threatening infections. For this reason, relevant immune defense mechanisms that may be responsible for maintaining intact host defen- ses in the majority of CHH individuals were studied. Spontaneous and allogeneic culture-induced (mixed lymphocyte response-MLR) specific and nonspecific (NK-like) cytotoxic mechanisms were analyzed and correlated with lymphocyte subpopulations present in CHH and normal individuals. Spontaneous natural- killer (NK) activity was present at or above normal levels, but culture-induced specific cytotoxicity and NK-like cytotoxicity as well as NK-like activity by T cell lines were significantly reduced in CHH individ- uals. The generation of radiation-resistant cytotoxicity, which normally occurs during allogeneic MLR, was markedly diminished in CHH, and was correlated with the decreased proliferation observed in CHH cultures. Preservation of spontaneous NK activity and loss of all forms of culture-induced cytotoxicity was associated with an increase in the proportion of lymphocytes bearing a thymic independent NK phenotype (OKM1+ OKT3- Fcy+ low-affinity E+), and a significant de- crease in thymic derived OKT3+ cytolytic T cell sub- populations in CHH individuals. Therefore, an intact cellular cytotoxic effector mechanism has been iden- tified in CHH (i.e., NK activity). Natural cytotoxicity may be of importance in maintaining host resistance to viral infections despite diminished thymic-derived effector mechanisms in cartilage-hair hypoplasia. Address reprint requests to Dr. S. H. Polmar, Department of Pediatrics, Washington University School of Medicine, P.O. Box 14871, St. Louis, MO 63178. Received for publication 28 July 1982 and in revised form 28 December 1982. INTRODUCTION Cartilage-hair hypoplasia (CHH)' is an autosomal re- cessive form of short-limbed dwarfism, which is found in increased frequency in the Old Order Amish of the United States, as well as in the Finnish population (1, 2). It is characterized by short stature, sparse unpig- mented hair, and cellular immunodeficiency (1-4). CHH individuals have decreased numbers of circu- lating B and T lymphocytes, negative delayed-type hypersensitivity reactions, and severely depressed pro- liferative responses to B and T cell specific antigens and mitogens. Impaired interleukin 2 (IL2) production and utilization have been found in studies on contin- uous T cell lines (CTCL), suggesting that the presence of a GC phase defect in the activation of CHH T cells (5). The Amish cartilage-hair hypoplasia subjects inves- tigated in our previous studies were healthy and in- fection free, in spite of their significant in vivo and in vitro immunologic abnormalities. Only a minority of CHH individuals suffer severe, sometimes fatal, viral infections (1-3). Thus, we sought to identify im- munological mechanisms present in CHH that might account for this apparent paradox. Cytotoxicity against virally infected altered-self has been demonstrated by OKT3+ OKT8+ HLA-restricted T cytotoxic cells as well as by Fc'yy natural killer (NK) cells (6-9). Our previous studies noted decreased numbers of OKT3+ (pan-T), OKT4+ (helper/inducer), and OKT8+ (sup- pressor/cytotoxic) T lymphocytes, and a large increase in the proportion of OKM1+ lymphocytes in CHH in- ' Abbreviations used in this paper: CHH, cartilage-hair hypoplasia; CML, cell-mediated lympholysis; CTCL, con- tinuous T cell lines; EAE1, sheep erythrocytes pretreated with 2-amino-ethylisothiouronium; Fcy', cells bearing receptors for the Fc portion of IgG; IL2, interleukin 2; MLR, allo- geneic mixed lymphocyte reaction; NK, natural killer; PBMC, peripheral blood mononuclear cells. J. Clin. Invest. © The American Society for Clinical Investigation, Inc. * 0021-9738/83/06/1737/07 $1.00 1737 Volume 71 June 1983 1737-1743 dividuals (4). Since many OKM1+ lymphocytes me- diate spontaneous NK activity (10-12), and OKT3+ T cells are responsible for proliferation and specific al- logeneic cell-mediated lympholysis (CML) in the MLR (12-14), we have evaluated and compared spontaneous NK activity with allogeneic culture-induced specific CML and nonspecific (NK-like) cytotoxicity in CHH and normal individuals. We found that spontaneous NK activity was preserved in CHH. However, allo- geneic culture induced specific and NK-like cytotox- icity were markedly diminished in CHH. In addition, the induction of radiation-resistant cytotoxicity in the MLR did not occur in lymphocyte cultures from CHH patients. Since most CHH individuals are healthy and survive well into adulthood, these findings suggest a clinically significant role for spontaneous NK cytotox- icity in host defense. METHODS Patients. The nine Amish CHH individuals (mean age 17.4 yr, range 13-26) who participated in these studies were originally diagnosed by field teams from the Johns Hopkins Hospital, under the direction of Dr. V. A. McKusick, using physical examinations, family histories, pedigree analyses, and, in some cases, radiographic and hair sample analyses (1, 3). All individuals were clinically healthy at the time of study. Six affected subjects had a history of severe, prolonged varicella infection, and four of these individuals also had histories of severe repeated chronic respiratory tract infec- tions. None of the patients studied received vaccinia or polio immunizations in childhood. Eight patients had blond body hair, and in three it was exceptionally sparse (lanugo). Age- and sex-matched unrelated normal individuals, as well as four unaffected siblings, served as controls. Blood samples were drawn in accordance with the guidelines established by the Institutional Review Board of University Hospitals of Cleveland. Cell separation. Peripheral blood mononuclear cells (PBMC) were obtained from heparinized whole blood by Ficoll-Hypaque density gradient centrifugation. PBMC were washed three times, counted, and used for the NK assay and the mixed lymphocyte culture induced cytotoxicity studies. For surface marker analysis, macrophages were removed from PBMC by adherence to plastic in 10% fetal calf serum (FCS) in RPMI 1640 and 5% CO2 for 30 min. Nonadherent cells, designated as peripheral blood lymphocytes (PBL), were decanted and used immediately for cell surface antigen studies. For surface marker analysis of purified T cells, PBMC were rosetted with 2-amino-ethylisothiouronium- treated sheep erythrocytes (EAET), and EAET+ cells were iso- lated on Ficoll-Hypaque gradients, as previously described (4). Cell recovery after separation procedures was similar for both CHH and normal individuals. Fluorescence staining. Subpopulation analysis was per- formed using monoclonal antibodies against cell surface de- terminants, as previously described (4). Briefly, OKT3 (pan- T), OKT4 (helper/inducer), and OKT8 (suppressor/cyto- toxic) were purchased from Ortho Pharmaceutical, Raritan, NJ. OKT5 (suppressor/cytotoxic), OKIal (anti-Ia), OKM1, and directly fluorescein-labeled OKT3, OKT4, OKT8, OKM1 were generously provided by Dr. Patrick Kung (Ortho Phar- maceutical). Incubation of PBL and EAET+ cells in 10% FCS with latex particles (0.81 gm, Difco Laboratories, Detroit, MI) for 30 min at 37°C was performed before staining to identify re- sidual macrophages. For staining, 3-5 X 105 PBL or EAET+ cells were incubated in duplicate with unlabeled monoclonal antibody for 30 min at 4°C, washed three times in 5% bovine serum albumin, then incubated with rhodamine-conjugated goat anti-mouse IgG (Cappel Laboratories, Inc., Cochran- ville, PA) for 30 min at 4°C. Nonspecific fluorescence was assessed by use of a mouse monoclonal IgG of irrelevant specificity, and was usually <3%. Sheep erythrocyte (E), EAET, and Fcy rosettes were de- termined on duplicate samples. For E and EAET rosettes, 3 ul of 5% (vol/vol) untreated or AET-treated sheep eryth- rocytes, respectively, were added to 5 X 104 lymphocytes in 25% FCS. Cells were centrifuged at 100 g, incubated at 4°C for 30 min, and viable cells (>97%) were read under epi- fluorescence after the addition of acridine orange-ethidium bromide (4). Fc-y rosettes were determined after overnight incubation of 5 X 104 lymphocytes in serum-free media with 5 gl of 5% IgG-coated ox erythrocytes, as previously de- scribed (5). Culture of CML effectors. One-way MLR were initiated by culturing PBMC in round-bottomed tubes in 3-ml vol at a cell concentration of 3 X 106/ml with either an equal number of irradiated (3,000 rad) autologous or allogeneic PBMC in RPMI 1640 containing 20% pooled AB serum. After 5 d, proliferation was assessed in 96 well microcultures by tritiated thymidine incorporation ([3HJTdR; 0.5 MCi/well, 25 Ci/mmol, New England Nuclear, Boston, MA), as described previously (15). After 6 d, effector cells were washed, counted, and diluted for the cytotoxic assays. Cytotoxicity targets. Autologous and allogeneic PBMC targets were maintained in culture for a total of 6 d in 20% AB serum in the absence of mitogens until they were re- quired for the cytotoxic assays. The erythroleukemia cell line K562 was maintained in exponential growth culture. Cytotoxic assays. Fresh or cultured effector PBMC were placed in round-bottomed wells of a microtiter tray with 5 X 103 51chromium-labeled (>450 mCi/mg, 1 mCi/ml, New England Nuclear) target cells at effector to target cell (E/T) ratios of 30:1, 15:1, and 7.5:1 in 150 Ml RPMI 1640 and 20% FCS. The targets used were autologous or allogeneic PBMC, or K562. The tray was spun at 50 g for 5 min, incubated for 4 h at 37°C in 5% CO2 in air, and then centrifuged at 150 g for 10 min. Aliquots (50 ul) of the supernatant were then removed and placed in glass tubes for counting in a gamma counter. Cytotoxic activity was quantitated as percentage of lysis: % lysis cpm (experimental release) -cpm (spontaneous release) x 100. cpm (total release) - cpm (spontaneous release) Spontaneous release of 5"chromium from target cells was determined in the presence of medium alone, whereas total release was determined by lysis in 1% sodium dodecyl sul- fate. Spontaneous release from targets was always <10%. Activity was linear over the range of effector-to-target ratios used. Cytotoxicity for autologous or allogeneic targets in autologous MLR-CML cultures was generally less than 5% of the cytotoxicity found in primed allogeneic MLR cultures directed against allogeneic targets. Autologous cytotoxicity in primed allogeneic cultures was less than 10% of the al- logeneic cytotoxicity. Radiation-resistant killing. Aliquots of fresh or cultured PBMC were irradiated (3,000 rad with a cobalt source) to 1738 G. F. Pierce, C. Brovall, B. Z. Schacter, and S. H. Polmar assess radiation-resistant killing, as previously described (15). Irradiated cells were treated identically to nonirradiated cells, except that radiation resistance experiments were per- formed using E/T ratios of 30:1 only. Intraassay coefficients of variation among triplicate samples were <3% for unir- radiated and irradiated samples. CTCL. Cell lines were initiated as previously described (5). Briefly, T cell blasts derived from phytohemaglutinin stimulated PBMC were maintained in continuous prolifer- ation by feeding at 5-d intervals with 40% (vol/vol) inter- leukin-2-containing medium. IL2 was prepared from an ir- radiated, pooled MLR from 10 allogeneic individuals cul- tured in the presence of 3.75 ug/ml PHA-M (Difco Labo- ratories). After 48 h, the supernatant was harvested, filtered through a 0.45-Am filter, and stored at -20°C until used. Supernatants were tested for IL2 potency in a CTCL mi- croassay, as previously described (5). After 20 d of culture, CTCL from two CHH and two normal individuals were tested for cytotoxicity against K562 targets as described above. Statistical analysis. A two-tailed unpaired Student's t test was used to evaluate significance levels. Critical values of the sample correlation coefficient, r, were determined in a two-tailed test. RESULTS Cytotoxic mechanisms in CHH and normal indi- viduals. Spontaneous NK activity was at or above normal levels in all CHH individuals studied (Fig. 1A). Fresh PBMC from eight CHH individuals lysed 44.3±5.0% (mean±SE) of K562 target cells in the 4-h NK assay at a 30:1 E/T ratio, compared with 36.4±3.9% lysis by fresh PBMC from eight normal individuals (Fig. IA). The killing was linear for all E/T ratios tested, and no significant differences were found be- tween the two groups. In contrast, minimal MLR-induced specific cell me- diated lympholysis could be generated in CHH indi- viduals (4.6±2.1% at E/T 30:1), whereas PBMC from normal subjects sensitized in an MLR lysed 26.3±2.4% of the targets at the same E/T ratio (P < 0.001, Fig. lA). Three CHH individuals had no detectable CML. Decreased MLR-induced proliferation was found against each of the allogeneic individuals used as spe- cific targets in the CML assay (CHH = 1,548±335 cpm, normal controls = 13,007±4,629 cpm; P < 0.005; Fig. 1B). This was not due to a lack of HLA-D locus differences, since in this series, as in previous studies (3, 4), the proliferative response to a pool of 10 allo- geneic individuals was significantly diminished (CHH = 3,531±609 cpm; normal = 24,262±2858 cpm; P <0.001). NK-like cytotoxicity against K562 was also assayed in the 6-d allogeneic MLR cultures. NK-like activity against K562 was significantly decreased in CHH in- dividuals, compared with normals (P < 0.001, Fig. 1A). Thus, CHH individuals showed a markedly de- creased specific CML, as well as NK-like cytotoxicity L PNMS\Wllll' _llm~__S 4 ANK NKC- I-..5 lll OMEEEMMM IKE MLR -CML 10 20 30 40 50 % CYTOTOXICITY (E/T 30:1)) B MLR 5 10 15 20 CPM "H-TdR (x 10-) FIGURE 1 Cytotoxic mechanisms present in eight CHH (solid bars) and eight normal (striped bars) individuals. (A) Spontaneous NK cytotoxic activity was measured on day 0 against K562. Culture-induced nonspecific (NK-like, P < 0.025) and specific (MLR-CML, P < 0.001) cytotoxicity was measured after 6 d using K562 or MLR stimulator cells, respectively. (B) Allogeneic MLR-induced proliferation was assessed in microcultures after 5 d (P < 0.005). after 6 d of allogeneic culture, but showed normal spontaneous activity in the NK assay. Radiation sensitivity and the generation of radia- tion-resistant killing. Previous studies revealed that radiation sensitivity of spontaneous NK activity from healthy control subjects was under the control of an X-linked gene, with radiation resistance found in 10% of Caucasian subjects (15). NK-like and specific killing by MLR-generated effector cells prepared from ra- diation-sensitive subjects was found to be radiation resistant. Therefore, it was of interest to determine the relative ability of CHH effectors to develop radiore- sistance in allogeneic MLR cultures. Spontaneous NK activity was sensitive to irradiation in all CHH indi- viduals tested, as well as in radiosensitive control sub- jects (data not shown). Culture-induced CML effectors from CHH individuals retained minimal cytotoxic ac- tivity after irradiation, compared with those from nor- mal individuals, P < 0.001 (Table I). In contrast, the amounts of radiosensitive cytotoxicity of CHH and normal CML effectors were not significantly different. Similarly decreased radiation resistance was also found for NK-like activity against K562 induced in allogeneic cultures in CHH cells, compared with normal cells P < 0.02, (Table II). Radiation-sensitive NK-like activity was comparable in CHH and normal cells. The gen- eration of radioresistant NK-like and specific cytotox- icity was correlated with allogeneic MLR-induced pro- liferation in seven CHH and normal individuals (pro- liferation vs. specific radioresistant cytotoxicity, CHH: r = 0.94, P < 0.01; normal: r = 0.99; P < 0.01; pro- Cytotoxic Activity in Cartilage-Hair Hypoplasia 1739 TABLE I Radiation Resistance of CML Activity in CHH and Nornml Individuals Cytotoxic activity' Cytotoxic activity' Radiation Radiation Normal Radiation Radiation CHIt subject Total resistantI sensitive§ subject Total resistant sensitive 1 14.5 1.1 13.4 1 21.0 14.8 6.2 2 3.3 0 3.3 2 26.6 22.1 4.5 3 5.4 2.5 2.9 3 37.9 36.7 1.2 4 8.8 5.5 3.3 4 33.7 28.5 5.2 5 0 0 0 5 20.1 15.6 4.5 6 10.1 0 10.1 6 18.8 17.2 1.6 7 0 0 0 7 28.2 22.1 6.1 Mean±SEM 6.0±2.2 1.3±0.9 4.7±2.0 26.6±3.0 22.4±3.2 4.2±0.8 Pl" <0.001 <0.001 NS - % 5'Cr release. Radiation-resistant cytotoxic activity is the cytotoxic activity remaining after effector cells have been irradiated with 3,000 rad. § Radiation-sensitive cytotoxic activity is the difference between the total cytotoxic activity minus the radiation-resistant activity. Comparison of CHH vs. normal by two-tailed unpaired Student's t tests. liferation vs. radioresistant NK-like activity; CHH: r = 0.88, P < 0.01; normal: r = 0.87, P < 0.05). Lymphocyte subpopulation analysis. It was of in- terest to relate the cytotoxic activities present in CHH PBMC to the lymphocyte subpopulations known to mediate cytotoxicity that are present in these individ- uals. In earlier work (4), we found a decrease in OKT3+ (pan-T), OKT4+ (helper/inducer), and OKT8+ (sup- pressor/cytotoxic) T lymphocytes, normal proportions of EAET+ cells, and a marked increase in percentages of OKM1+ lymphocytes in CHH. The monoclonal an- tibody analyses of PBL of the CHH and normal in- dividuals in the present study is shown in Table III and are similar to our previous findings (4). OKT5+ cells, reactive with a monoclonal antibody that iden- tifies a subset of suppressor/cytotoxic T cells similar but not identical to the OKT8+ subpopulation, were also decreased in CHH individuals (P < 0.05). E+ cells were significantly lower, but EAET+ cells, which con- tain more low-affinity rosette-forming cells, were in normal range, in spite of a marked decrease in OKT3+ cells in CHH patients, compared with controls. In purified EAET+ cell fractions made from CHH PBMC, OKT3+ cells were decreased (P < 0.001), and TABLE II Radiation Resistance of NK-like Activity in CHH and Normal Individuals Cytotoxic activity' Cytotoxic activity' Radiation Radiation Normal Radiation Radiation CHH subject Total resistantI sensitive§ subject Total resistant sensitive 1 16.2 2.3 13.9 1 13.6 3.2 10.4 2 5.0 1.2 3.8 2 36.5 30.6 5.9 3 13.4 11.2 2.1 3 49.5 45.0 4.5 4 9.3 7.1 2.2 4 22.2 5.9 16.3 5 4.5 0 4.5 5 23.8 15.6 8.2 6 26.7 2.7 24.0 6 16.2 3.7 12.5 7 4.9 0 4.9 7 17.5 14.7 2.8 Mean±SEM 11.4±3.3 3.5±1.7 7.9±3.3 25.6±5.3 17.0±6.4 8.7±1.9 pl" <0.01 <0.02 NS % 5'Cr release. Radiation-resistant cytotoxic activity is the cytotoxic activity remaining after effector cells have been irradiated with 3,000 rad. § Radiation-sensitive cytotoxic activity is the difference between the total cytotoxic activity minus the radiation resistant activity. Comparison of CHH vs. normal by two-tailed unpaired Student's t tests. 1740 G. F. Pierce, C. Brovall, B. Z. Schacter, and S. H. Polmar TABLE III Surface Marker Analysis of Subpopulations in CHH and Normal Individuals % positive peripheral blood lymphocytes' Subjects OKT3 OKT4 OKT5 OKT8 OKM1 ElEAEtt CHH (n = 9) 52±7 36±5 11±4 18±3 44±7 69±2 87±2 Normal (n = 9) 77±3 55±2 30±3 32±3 18±2 79±1 88±1 p§ <0.005 <0.005 <0.05 <0.005 <0.001 <0.01 NS ° The absolute lymphocyte count, calculated from the WBC and differential, for CHH individuals was 1023+95/mm3; normals 1952±183; P < 0.001. t E rosettes were determined using untreated sheep erythrocytes; EAET rosettes were determined using sheep erythrocytes pretreated with AET. Both were read following a 30-min incubation at 4°C. § Significance levels were determined by two-tailed unpaired Student's t tests. both Fc-y' and OKM1+ lymphocytes were significantly increased (P < 0.05, Table IV). In CHH individuals, there was a strong positive correlation between the increased percentages of Fc"y cells and OKM1+ cells (r = 0.83, P < 0.01). In normal subjects, Fcy'y and OKM1+ cells represented only a small proportion of the total lymphocytes. In CHH, NK activity was sig- nificantly correlated with the number of OKM1+ cells (r = 0.80, P < 0.05), as well as with the percentage of Fc-y' cells (r = 0.74, P < 0.05). In normal subjects, all NK activity does not appear to be mediated by OKM1+ cells (11, 16). Thus, no significant correlations would be expected and none were detected. Therefore, it appears that the subpopulation bearing an NK cell phenotype (Fcy' OKM1+ OKT3- OKT8- low affinity E+) is proportionally increased in CHH individuals, whereas thymus derived OKM1- OKT3+ OKT8+ EAET+ cells, which mediate cytotoxic T-cell activity, are selectively depleted. These findings are consistent with the cytotoxicity studies in CHH in which NK activity is preserved while CML is markedly reduced. TABLE IV Subpopulation Analysis of EAET + Cells in CHH and Normal Individuals % positive EAET + cells' Subjects Fc7t OKMl OKT3 CHH (n = 9) 30±5§ 38±1 70±6 Normal (n = 6) 14±2 12±1 92±1 P§ <0.05 <0.005 <0.001 * EAFT cells were obtained after rosetting PBMC with AET-treated sheep erythrocytes. t Fcy rosettes were determined after overnight incubation with IgG-coated ox erythrocytes in serum-free media. § Significance levels were determined by unpaired two-tailed t tests. Anti-K562 activity generated in CTCL. Analysis of CTCL confirmed the absence of culture-generated NK-like activity in CHH individuals. CTCL depen- dent on IL2 were established simultaneously from two normal and two CHH subjects. Both CHH CTCL showed diminished cytotoxicity against K562 targets when tested after 20 d of culture (Table V). No OKM1+ or Fc"y cells were detected in either CHH or normal CTCL cultures. Cells from both groups were greater than 90% EAET+ and OKT3+. Both CHH and normal CTCL were -50% OKT4+ and 50% OKT8+ and mor- phologically resembled 3-d phytohemaglutinin blasts. OKIal detected -70% of the cells from both groups of individuals. As in allogeneic MLR-CML cultures, radioresistant cytotoxicity could not be generated in CHH CTCL (Table V). DISCUSSION Marked impairment of lymphocyte proliferation has been described in cartilage-hair hypoplasia (2-5). In TABLE V Radiation Resistance of NK-like Cytotoxic Activity of CTCL from CHH and Normal Individuals Cytotoxic activity' Radiation Radiation CTCL Total resistant I sensitivel CHH 1 4.2±0.4 1.5±0.1 2.7±0.4 CHH 2 4.7±0.2 0.7+0.1 4.0±0.2 Normal 1 21.0±2.1 17.2±0.3 3.8±2.1 Normal 2 39.9±1.1 27.4±1.6 12.5±2.6 % 5"Cr release. Radiation-resistant cytotoxic activity is the cytotoxic activity re- maining after effector cells have been irradiated with 3,000 rad. § Radiation-sensitive cytotoxic activity is the difference between the total cytotoxic activity minus the radiation-resistant activity. Cytotoxic Activity in Cartilage-Hair Hypoplasia 1741 this paper, we report that cellular cytotoxicity effector mechanisms are impaired but not constitutively ab- sent. Culture-induced specific and NK-like cytotox- icity mediated by OKT3+ T-cells is proliferation dependent, and was found to be absent or markedly diminished in CHH cultures. However, proliferation- independent spontaneous natural cytotoxicity is intact in CHH. Individuals with cartilage-hair hypoplasia appear to have a selective loss of high-affinity E-ro- setting cells which are OKT3+, whereas low-affinity E rosette-positive cells bearing the OKM1 and Fc-y receptor markers are preserved. Spontaneous NK activity against K562 has been shown to be mediated by Fcy' OKM1+ OKT3- E+ and E- (null) large granular lymphocytes (LGL) (10-12, 17-19). In contrast, allogeneic MLR-induced specific cytotoxicity is mediated by the OKT3+ T5/T8+ T lym- phocyte subpopulation (12-14), whereas the pheno- type of the cells mediating NK-like culture-induced cytotoxicity has not been established. However, recent studies by Lopez-Botet et al. (20) suggest that NK-like activity is mediated by an IL2-responsive OKT3+ T- cell blast. Seeley et al. (21) found that MLR-CML and NK-like against K562 were mediated by different cell types. They, and others, also reported that the anti- K562 culture-induced killing was not mediated by Fc-y' cells, in contrast to spontaneous NK activity (21- 23). Our results support the concept that the cells mediating spontaneous NK activity are distinct from NK-like and MLR-CML, since (a) NK-like and specific cytolysis could not be generated in significant amounts in CHH individuals with impaired T cell function after culture, but NK activity was found in freshly isolated PBMC, and (b) Fcy' and OKM1+ cells decrease rap- idly in MLR cultures, as culture-induced cytotoxicity increases, and are not present on CTCL, which bear strictly T-cell phenotypes (OKT3+T4+ or OKT3+T8+). Using clonally selected specific cytotoxic CTCL, others have also demonstrated the T cell nature of nonspecific anti-K562 cytotoxicity (24, 25). Moretta et al. (24) re- ported three out of four anti-K562 clones were OKT8+ and all were E+ and Fcy-. The inability of CHH individuals to recruit OKT5+/ T8+ cells to function in both HLA-restricted (MLR- CML) and nonrestricted killing (NK-like) is due, at least in part, to decreased production and utilization of interleukin 2 by CHH T cells (5). The clonal ex- pansion of MLR-generated specific CML is dependent upon helper factors (IL2) synthesized by some OKT4+ cells during the MLR sensitization (6, 13, 26). Radio- resistant nonspecific cytotoxicity can be recruited by IL2-containing media (27). However, decreased IL2 production by CHH OKT4+ cells cannot entirely ex- plain the persistent radiosensitivity and severely de- pressed levels of anti-K562 activity by CHH CTCL, since excess exogenously supplied IL2 permits the re- cruitment of radioresistant effectors in normal lines, which themselves do not produce IL2. The failure of CHH PBMC and CTCL to develop culture-induced radioresistance supports the concept of defective re- cruitment due to an intrinsic defect in CHH T cells, which is also associated with decreased responsiveness to IL2 (5). In contrast, thymic-independent sponta- neous NK activity does not require cell proliferation or soluble recruitment factors. Interestingly, the athymic nude mouse also has normal spontaneous NK activity and diminished culture generated cytolysis. However, nu/nu mice appear to have an isolated IL2 production deficit, and can generate cytolytic T cell effectors when exogenous IL2 is provided (28). The in vivo role of spontaneous NK activity, as well as antibody-dependent cellular cytotoxicity, which is mediated by the same cell population, remains to be elucidated. Deficient spontaneous NK activity has been found in X-linked lymphoproliferative syndrome (29) and Chediak-Higashi syndrome (C-HS) (30). In both diseases, as well as in the beige mouse model of C-HS, the selective loss of NK effectors is associated with an increased susceptibility to infections and lym- phoproliferative disorders (31). Only a minority of in- dividuals with cartilage-hair hypoplasia have a history of severe viral infections (usually varicella, vaccinia, or polio) as well as recurrent, persistent respiratory infections, and in only a minority of these individuals is the outcome lethal (unpublished observations). Most CHH individuals are clinically healthy and survive well into adulthood despite profound in vitro cellular immune dysfunction (3, 4). Moreover, rates of lym- phoproliferative and autoimmune diseases do not ap- pear to be increased in CHH individuals (unpublished observations). Spontaneous natural cytotoxicity has been proposed to mediate host-defense against certain viral infections (i.e., vaccinia, cytomegalovirus, mea- sles) (7-9), as well as immune surveillance against malignancies (31), and our studies of CHH support this theory. Our observations in CHH lend support to the importance of NK activity in human host defenses and raise questions regarding the relative importance of proliferation-dependent T cell mediated cytotoxic mechanisms. Thus, cartilage-hair hypoplasia provides a unique human model of lifelong partial cellular im- munodeficiency for the analysis of the in vivo require- ments for immunocompetence. ACKNOWLEDGMENTS We thank Dr. Patrick Kung and Dr. Gideon Goldstein for their advice relating to some of the immunofluorescence studies. This study was supported in part by National Institutes of Health grants Al 18527, AI 20082, and RR 00080. Glenn F. Pierce was supported by Medical Scientist Training Pro- gram grant GM07250. 1742 G. F. Pierce, C. Brovall, B. Z. Schacter, and S. H. Polmar REFERENCES 1. McKusick, V. A., R. Eldridge, J. A. Hostetler, U. Ruang- wit, and J. A. Egeland. 1965. Dwarfism in the Amish. II. Cartilage-hair hypoplasia. Bull. Johns Hopkins Hosp. 116: 231-272. 2. Virolainen, M., E. Savilahti, I. Kaitila, and J. Perheen- tupa. 1978. Cellular and humoral immunity in cartilage- hair hypoplasia. Pediatr. Res. 12: 961-966. 3. Trojak, J. E., S. H. Polmar, J. A. Winkelstein, S. Hsu, C. Francomano, G. F. Pierce, J. J. Scillian, A. N. Gale, and V. A. McKusick. 1981. Immunologic studies of car- tilage-hair hypoplasia in the Amish. Johns Hopkins Med. J. 148: 157-164. 4. Pierce, G. F., and S. H. Polmar. 1982. Lymphocyte dys- function in cartilage-hair hypoplasia: I. Evidence for an intrinsic defect in cellular proliferation. J. Immunol. 129: 570-575. 5. Pierce, G. F., and S. H. Polmar. 1982. Lymphocyte dys- function in cartilage-hair hypoplasia: II. Evidence for a cell cycle specific defect in T-cell growth. Clin. Exp. Immunol. 50: 621-628. 6. Biddison, W. E., S. 0. Sharrow, and G. M. Shearer. 1981. T-cell subpopulations required for the human cytotoxic T-lymphocyte response to influenza virus: evidence for T-cell help. J. Immunol. 127: 487-491. 7. Perrin, L. H., R. M. Zinkernagel, and M. B. A. Oldstone. 1977. Immune response in humans after vaccination with vaccinia virus: Generation of a virus-specific cy- totoxic activity by human peripheral lymphocytes. J. Exp. Med. 146: 949-969. 8. Quinnan Jr., G. V., N. Kirmani, A. H., Rook, J. F. Man- ischewitz, L. Jackson, G. Moreschi, G. W. Santos, R. Saral, and W. H. Burns. 1982. Cytotoxic T-cells in cy- tomegalovirus infection: HLA restricted T-lymphocyte and non-T-lymphocyte cytotoxic responses correlate with recovery from cytomegalovirus infection with bone-marrow-transplant recipients. N. Engl. J. Med. 307: 7-13. 9. Perrin, L. H., A. Tishon, and M. B. A. Oldstone. 1977. Immunologic injury in measles virus infection: III. Pres- ence and characterization of human cytotoxic lympho- cytes. J. Immunol. 118: 282-289. 10. Kay, H. D., and D. A. Horwitz. 1980. Evidence by reac- tivity with hybridoma antibodies for a probable myeloid origin of peripheral blood cells active in natural cyto- toxicity and antibody-dependent cell-mediated cytotox- icity. J. Clin. Invest. 66: 847-851. 11. Ortaldo, J. R., S. 0. Sharrow, T. Timonen, and R. B. Herberman. 1981. Determination of surface antigens on highly purified human NK cells by flow cytometry with monoclonal antibodies. J. Immunol. 127: 2401-2409. 12. Zarling, J. M., and P. C. Kung. 1980. Monoclonal anti- bodies which distinguish between human NK cells and cytotoxic T-lymphocytes. Nature (Lond.). 288: 394-396. 13. Reinherz, E. L., and S. F. Schlossman. 1980. The dif- ferentiation and function of human T-lymphocytes. Cell. 19: 821-827. 14. Fast, L. D., J. A. Hansen, and W. Newman. 1981. Ev- idence for T-cell nature and heterogeneity within nat- ural killer (NK) and antibody-dependent cellular cyto- toxicity (ADCC) effectors: A comparison with cytolytic T-lymphocytes (CTL). J. Immunol. 127: 448-452. 15. Brovall, C., and B. Schacter. 1981. Radiation sensitivity of human natural killer cell activity: Control by X-linked genes. J. Immunol. 126: 2236-2239. 16. Abo, T., and C. M. Balch. 1981. A differentiation antigen of human NK and K cells identified by a monoclonal antibody (HNK-1). J. Immunol. 127: 1024-1029. 17. Timonen, T., J. R. Ortaldo, and R. B. Herberman. 1981. Characteristics of human large granular lymphocytes and relationship to natural killer and K cells. J. Exp. Med. 153: 569-582. 18. Ferrarini, M., A. Cadoni, A. T. Franzi, C. Ghigliotti, A. Leprini, A. Zicca, and C. E. Grossi. 1980. Ultrastructure and cytochemistry of human peripheral blood lympho- cytes. Similarities between the cells of the third popu- lation and TG lymphocytes. Eur. J. Immunol. 10: 562- 570. 19. Van de Griend, R. J., I. T. Berge, H. J. Tanke, D. Ross, P. Th. A. Schellekens, C. J. M. Melief, W. P. Zeijlemaker, and A. Astaldi. 1982. Characterization of two subsets of human TG cells. J. Immunol. 128: 1979-1985. 20. Lopez-Botet, M., A. Silva, J. Rodriguez, and M. 0. de Landazuri. 1982. Generation of T-cell blasts with NK- like activity in human MLC: Cellular precursors, IL2 responsiveness and phenotype expression. J. Immunol. 129: 1109-1115. 21. Seeley, J. K., G. Masucci, A. Poros, E. Klein, and S. H. Golub. 1979. Studies on cytotoxicity generated in human mixed lymphocyte cultures. II. Anti-K562 effectors are distinct from allospecific CTL and can be generated from NK-depleted T cells. J. Immunol. 123: 1303-1310. 22. Ortaldo, J. R., G. D. Bonnard, P. D. Kind, and R. B. Herberman. 1979. Cytotoxicity by cultured human lym- phocytes: Characteristics of effector cells and specificity of cytotoxicity. J. Immunol. 122: 1489-1494. 23. Bolhuis, R. L. H., and H. Schellekens. 1981. Induction of natural killer activity and allocytotoxicity in human peripheral blood lymphocytes after mixed lymphocyte culture. Scand. J. Immunol. 13: 401-412. 24. Moretta, L., M. C. Mingari, P. R. Sekaly, A. Moretta, B. Chapuis, and J. C. Cerottini. 1981. Surface markers of cloned human T cells with various cytolytic activities. J. Exp. Med. 154: 569-574. 25. Pawelec, G. P., M. R. Hadam, A. Ziegler, J. Lohmeyer, A. Rehbein, I. Kumbier, and P. Wernet. 1982. Long term culture, cloning, and surface markers of mixed leukocyte culture-derived human T lymphocytes with natural killer-like cytotoxicity. J. Immunol. 128: 1892-1896. 26. Kung, P. C., and G. Goldstein. 1980. Functional and developmental compartments of human T lymphocytes. Vox Sang. 39: 121-127. 27. Schacter, B., C. Brovall, G. Pierce, and J. Ellner. 1982. Non-specific cellular cytotoxicity differs from natural killer activity in its susceptibility to y radiation. Im- munobiology. 163: 410. 28. Wagner, H., C. Hardt, H. Stockinger, K. Pfizenmaier, R. Bartlett, and M. Rollinghoff. 1981. Impact of thymus on the generation of immunocompetence and diversity of antigen specific MHC-restricted cytotoxic T-lympho- cyte precursors. Immunol. Rev. 58: 95-129. 29. Sullivan, J. L., K. S. Byron, F. E. Brewster, and D. T. Purtillo. 1980. Deficient natural killer cell activity in X- linked lymphoproliferative syndrome. Science (Wash. DC). 210: 543-545. 30. Roder, J. C., T. Haliotis, M. Klein, S. Korec, J. R. Jett, J. Ortaldo, R. B. Herberman, P. Katz, and A. S. Fauci. 1980. A new immunodeficiency disorder in humans in- volving NK cells. Nature (Lond.). 284: 553-555. 31. Herberman, R. B., and J. R. Ortaldo. 1981. Natural killer cells: Their role in defenses against disease. Science (Wash. DC). 214: 24-30. Cytotoxic Activity in Cartilage-Hair Hypoplasia 1743