Cartilage hair hypoplasia: Immunological aspects and their clinical implications

CLINICAL

IMMUNOLOGY

Cartilage

AND

IMMUNOPATHOLOGY

40, 87-93 (1986)

Hair Hypoplasia: Immunological Clinical Implications’

Aspects and Their

STEPHEN H. POLMAR*AND GLENN Department.\ of Pediatrics. Microbiology and Imnztrnolog~. Medicine. Departments of Puthology und Internal Medicine. Medicine, St. Louis, Missorrri

F. PIERCE

trnd The Di~~ision of Lahorutory Wa.shington Uniwrsity School 63178

of

Cartilage hair hypoplasia (CHH) is an autosomal recessive form of short-limbed dwarfism prevalent among the Old Order Amish. Mild to moderately severe cellular immunodeficiency is associated with this disorder. Antibody synthesis is, however. normal in CHH. Individuals affected with CHH were found to have marked impairment of T-lymphocyte function due to an intrinsic defect in cell proliferation. Defective proliferation was also found in B cells and fibroblasts from CHH individuals suggesting that impaired T-cell function reflects a generalized defect in cell proliferation in this syndrome. Studies of cytotoxic mechanisms in CHH patients revealed that proliferation-dependent mechanisms (e.g., cell-mediated cytotoxicity and natural killer [NKj-like activity) were markedly impaired while proliferation-independent NK activity was normal. In spite of impairment of T-cell function, an increased incidence of malignancy was not observed in CHH patients. These observations suggest that NK activity is vital in host defense against malignancies and that marked impairment of T-cell-mediated immunity need not be associated with an increased susceptibility to malignancy if NK function is preserved. v lY8h Academtc Prss. Im

INTRODUCTION

The immunodeficiency diseases of man have provided unique insights into the functioning of the normal immune system. These “experiments of nature” (1). in which one or more of the essential components of bodily defense mechanisms are defective, have permitted correlation of immunologic defects with their clinical consequences (e.g., infection, malignancy). In many instances the immune defect in a primary immunodeficiency disease is so severe that if the disorder is not corrected, death will result in a relatively short period of time. Cartilage hair hypoplasia (CHHJ3 is a disorder in which a mild to moderately severe cellular ’ Supported in part by National Institute of Health Grants AII8527, AI30082. and RR00036 (General Clinical Research Center). Presented as part of a symposium entitled “Childhood lmmunodeficiency Disorders: Diagnosis, Prevention, and Management,” February 25-26. 1985, National Academy of Sciences, Washington, D.C. z To whom reprint requests should be addressed: Division of Immunology. Department of Pediatrics. Washington University School of Medicine. 400 S. Kingshighway Blvd., P.O. Box 14871. St, Louis, MO. 63178. 3 Abbreviations used: CHH. cartilage hair hypoplasia: PHA. phytohemagglutinin: ConA. concanavalin A: MLR. mixed leukocyte response; PBMC, peripheral blood mononuclear cells: IL-2. interleukin-2: CTCL. continuous T cell line: SCID, severe combined immune deficiency disease: ADA. adenosine deaminase; PNP. purine nucleoside phosphorylase: NK. natural killer: CML, cell-mediated lympholysir: EB, Epstein-Barr virus.

87

xx

POLMAR

AND

PIERCE

immunodeficiency exists and yet most affected individuals Iivc well into adult-~ hood. For this reason. this disorder is of particular interest since it permits us to examine the consequences of long-term partial cellular immunodeficiency. BACKGROUND Cartilage Hair Hypoplasia is an autosomal recessive form of short-limbed dwarfism. This disorder was first described by McKusick or trl. who reported 77 cases in the Old Order Amish (3). Affected individuals have short limbs due to metaphyseal chondrodysplasia. Histologically. there is a paucity of chondrocytex as well as a failure of chondrocytes to form orderly columns at the growth plates. The disease is phenotypically heterogeneous and displays incomplete penetrance (7Oc/c). In the most severely affected individuals. hair is fine, sparse, unpigmented, and lacks a central pigment core. Unlike other forms of dwarfism, there are no dysmorphic facial features and intelligence is usually normal. CHH is most prevalent among the Amish communities of Ohio and Indiana where the gent frequency for CHH has been estimated to be Tcf (I). CHH also occurs with high frequency in the Finnish population (3. 4). In the Amish. 80% of the affected individuals can trace their ancestry to one of two individuals. Catherine or Jacob Hochstetler. thought to be siblings. who emigrated to the United States in the mid-1700s. Thus. CHH in the Amish is most likely due to genes identical by descent. Other cases of CHH have been reported sporadically in individuals of nonAmish descent: however. these patients often have a disease quite different from that found in the Amish and Finnish populations. The association of an immune defect with CHH was first suggested by McKusick, who noted an increased morbidity and mortality from varicella in Amish children with CHH compared to their normal siblings. Mild to moderate lymphopenia, decreased delayed hypersensitivity, and impaired leukocyte responses in the presence of normal immunoglobulin levels and antibody synthesis was observed in two patients by Lux and co-workers (5). Two studies of immunological function of Finnish subjects with CHH showed impairment of cell-mediated immunity. but normal antibody-mediated immunity (3.3). Absent delayed hypersensitivity reactions and decreased irl IB~~IYIlymphocyte proliferative responses to mitogens and antigens were observed in the majority of patients studied. The absolute lymphocyte count of CHH subjects was approximately 50% that of the normal controls. In contrast. serum immunoglobulin levels were found to be normal for age and antibody synthesis was not defective. increased susceptibility to infection was not prominent among individuals affected with CHH in these Finnish studies. Although the severity of impairment of cellular immunity was variable, it was present in virtually all Finnish individuals affected with CHH. On the basis of these findings, Virolainen and co-workers concluded that impairment of cellular immunity was an integral part of the CHH syndrome (4). STUDIES OF IMMUNOLOGICAL FUNCTION We undertook studies of the immunological aspects of CHH approximately 6 years ago in an attempt to identify the cellular basis of the immunodeficiency seen in CHH and to correlate those findings with the clinical features of this

CARTILAGE

HAIR

89

HYPOPLASIA

disorder. Thirty-two individuals affected with CHH in 16 sibships (6-8) from the Ohio Amish community were studied. The mean age was 20.6 -+ 14.0 years and subjects ranged from age 1.5 to 52 years. A similar number of age- and sexmatched sibling controls were also studied. Lymphocytes from all CHH subjects studied showed a marked impairment in their ability to proliferate in response to phytohemagglutinin (PHA), concanavalin A (ConA), allogeneic cells in the mixed leukocyte response (MLR) and the mitogenic monoclonal antibody (OKT3) (Fig. 1) (6). The proliferative responses of CHH lymphocytes were only 20-30s of those seen in normal sibling controls. This pronounced decrease in lymphocyte proliferation did not appear to be due to a significant imbalance in T-lymphocyte subpopulations in the peripheral blood of CHH subjects as determined by immunotluorescent analysis using subpopulation-specific monoclonal antibodies. The circulating lymphocyte counts were less than 50% of normal and the percentages of total OKT3+ cells were reduced compared to normal; however. the proportions of the major T-lymphocyte subpopulations were normal in CHH individuals. The mononuclear cell populations obtained from CHH patients used in these proliferation studies had an increased proportion of OKMI positive cells and nonspecific esterase-staining cells (6). Thus, to determine if monocyte contamination could account for decreased proliferative responses. or if CHH monocytes were less capable of providing accessory cell function than normal monocytes. CHH and normal monocytes were studied with respect to their ability to enhance OKT3-induced proliferation. Proliferation of T cells initiated by stimulation of the ‘r3-Ti antigen receptor complex with this monoclonal antibody is absolutely dependent upon macrophages (monocytes). As shown in Fig. 2. CHH macrophages facilitated the proliferation of normal T cells as well as autologous

MLR

FIG. I. Proliferation monoclonal antibody striped bars. Source:

CON A

PHA

of CHH and normal cells to allogeneic (OKT3). CHH cell proliferation is shown Ref. (6).

OKT3

cells (MLR). mitogens and a mitogenic in the solid bars and normal cells in the

POLMAK

M@ (~10~)

ADDED

ANDPIEKCk

T O T CELL

CULTURES

2. Macrophage (M+) helper function assessed in CHH and normal T cells using mitogenic monoclonal antibody OKT3. Source: Ref. (6). FIG.

normal macrophages. Conversely, CHH T cells proliferated poorly in the presence of both normal and CHH macrophages. These data suggest that defective proliferation was due to an abnormality in the CHH T cell and not in CHH macrophage function. Furthermore, CHH cells did not suppress proliferation in coculture experiments in which CHH and normal sibling peripheral blood mononuclear cells (PBMC) were mixed (6). Thus it appears that the defect in T-lymphocyte proliferation resides in the T cell and is not due to excess suppressor-ceil activity or impaired accessory-cell function. The defect in cell proliferation described above is not strictly limited to T lymphocytes. B lymphocytes isolated from CHH subjects were found to proliferate very poorly when stimulated with StLlphqlo~oc.c.rrsMIIWIIS Cowan I, a B-cell mitogen. Furthermore, the growth rate of fibroblasts from CHH subjects was found to be less than 50% that of normal controls (6). McKusick had observed that the histological appearance of the cartilage of a child with CHH resembled that of cartilage in which growth had ceased (2). Thus the defect in lymphocyte proliferation which underlies the impairment in cellular immunity seen in CHH appears to be part of a more generalized defect in cellular proliferation in the CHH ryndrome. The biochemical basis of defective cell proliferation in CHH remains unknown. Attempts to induce normal lymphocyte proliferation using membrane-perturbing and tumor-promoting agents such as phorbol myristate acetate have been unsuccessful (6). Induction of lymphocyte proliferation with the calcium ionophore A23187 was also defective in CHH lymphocytes. Thus. the defect in cell proliferation in CHH appears to exist later in the sequence of lymphocyte activation than the early membrane activation and calcium influx stages (6). We observed that interleukin-2 (IL-2) production by CHH lymphocytes was reduced, but proliferation of CHH continuous T cell lines (CTCL) could not be normalized by providing excess exogenous IL-2 (7). However. the ability of CHH CTCL to “up regulate” IL-2 receptors as measured by binding of radiolabeled IL-2 was found to be comparable to that of normal CTCL. Thus. while defects in the G, stage of

CARTILAGE

HAIR

HYPOPLASIA

91

the cell cycle have been observed in CHH (e.g., decreased IL-2 production), the fact that growth of CHH T cells remains abnormal even in the presence of exogenous IL-2 and normal IL-2 receptor expression suggests that the basic lesion in CHH may express itself at multiple stages of the cell cycle. Radiologically, the bone abnormalities in CHH resemble those seen in patients with severe combined immune deficiency disease (SCID) and adenosine deaminase (ADA) deficiency. For this reason we examined ADA activity in the erythrocytes of 15 CHH subjects and II sibling controls. Erythrocyte ADA activity was 108 + 12 nM/mg Hgb/Hr for CHH subjects versus 36.3 2 3.0 for controls; P < 0.0005. In contrast, purine nucleoside phosphorylase (PNP) levels in CHH versus normal erythrocytes were not significantly different (2203 ? 154 nM/mg HgbiHr in CHH versus 2332 ? 228 nM/mg HgbiHr in controls). In spite of these interesting differences in the enzyme activity, no differences in erythrocyte adenosine nucleotide pools in CHH erythrocytes or lymphocytes were found, nor were any alterations of adenosine metabolism in CHH continuous T cell lines observed in comparison to normal T cell lines. It is interesting to note that similarly elevated levels of ADA in erythrocytes have been reported in the red blood cells of patients with congenital hypoplastic anemia (9). Several cases of congenital hypoplastic anemia have also been observed among Amish CHH patients. The significance of these associations are unknown at present. A defect in cell proliferation, such as that which exists in CHH, has important consequences for lymphocyte function since so many immune responses require cell division. Particularly interesting are the implications for cytotoxic mechanisms thought to be important in immunity against virally infected ceils and tumors. Natural killer (NK) cell activity is a spontaneous nonspecific cytotoxicity which does not require previous exposure to antigen or cell proliferation and is mediated by OKMl-positive large granular lymphocytes t IO). Specific cell-mediated cytotoxicity or lympholysis (CML) requires prior antigen exposure and cell proliferation and is mediated by T cells. NK-like activity is a nonspecific cytotoxicity also requiring proliferation but not prior antigen exposure and also appears to be mediated by T cells. We carried out studies of these three in I)itro cytotoxicity mechanisms using PBMC from CHH subjects. NK activity and NKlike activity assays utilized the K562 cell line as target cells, while the CML assays used the sensitizing allogeneic PBMC as the targets. CHH individuals were found to have marked impairment of proliferation-dependent cytotoxic mechanisms (CML and NK-like activity) while proliferation-independent NK activity was normal or even above normal (Fig. 3) (8). These data, correlated with results of ceil surface phenotype studies. demonstrate a significant deficiency of T cells and retention of OKMI-positive lymphocytes, i.e.. NK cells (6. 8). CLINICAL

STUDIES

These observations have interesting implications for the role of these cytotoxic mechanisms in immunity to viral infections and malignancy. The clinical characteristics of the CHH subjects in our study are summarized in Table 1. As in previous reports (3, 5), an increased frequency of severe and prolonged varicella was observed in our patients. While morbidity and mortality to varicella is clearly

POLMAR

AND

PIEKC’t:

NKLIKE

% CYTOTOXICITY

CPM 3WTdR

FIG. 3. Cytotoxic individuals. Source:

mechanisms Ref. (8).

present

in eight

(E/T

3O:ll

B

lx iO-3l

CHH

(solid

bars1 and eight

normal

(striped

bar-\1

increased in CHH. there is no apparent increase in susceptibility to severe or fatal infections with other viruses. Unlike other patients with T-cell-mediated immune deficiency, we have not found any cases of malignancy in our CHH subjects (Table 2) and our Finnish colleagues (I I) are not aware of any malignancies in their approximately 30 CHH patients. CHH is thus precisely the opposite of the Chediak-Higashi syndrome in which patients lack NK activity but have normal T-cell-mediated cytotoxicity and have a high incidence of lymphoproliferative disease (12). Epstein-Barr (EB) virus has been associated with the development of the accelerated phase in the Chediak-Higashi syndrome: no unusual susceptibility or consequences of EB virus infection have been observed in CHH subjects. Thus our observations in CHH are consistent with the concept that NK activity is of vital importance in host defense against malignancies (13) and that quite marked impairment of T-cell-mediated immunity need not be associated with an TABLE CLINICALFEATURESOF

1 CHH

I, = 30

Feature(s) Severe varicella Recurrent otitis media Increased frequency of UKla Allergies and/or sinusitis Childhood anemiah Hirschsprung’s disease Malignancy Li Upper respiratory tract infection. b Includes two cases of congenital

INDIVIDUALS

I2 4 Ii 6 Y I 0

hypoplastic

anemia

Percentage 40 ifI 43 20 30 3 0

CARTILAGE

93

HAIR HYPOPLASIA

TABLE 2 ANALYSIS OF PERIPHERALBLOOD LYMPHOCYTESIN CHH AND NORMAL INDIVIDUALS Cell Population

CHH

Normal

P

value”

Lymphocytes

0.001

OKT3 (pan-T) OKTWOKTS ratio OKMI tNKlh

0.02 NS 0.005

ClTwo-tailed Student’s t test; NS. not significant h Only OKMI + lymphocytes were counted.

increased susceptibility to malignancy if the function of NK cells is preserved. Other explanations are possible, for example, the proliferative defect in CHH may impair the growth of neoplastic cells as well. While we think this is unlikely. we are currently examining this possibility. In conclusion, the study of rare disorders and even those with subtle immunologic defects remains a worthwhile undertaking. The findings from entities such as CHH have important implications for much more common diseases of man. REFERENCES I. Good, R. A.. In “immunodeficiency in Man and Animals” (D. Bergsma. Ed.). Vol. I I. No. I. pp. 13-20. Birth Defects Original Article Series, March of Dimes. White Plains. N.Y.. 1975. 2. McKusick. V. A., Eldridge, R.. Hostetler. J. A.. Ruangwit, U., and Egelund. J.. &r/l. J&n.\ HopXirr.s Hosp. 116, 285, 1965. 3. Ranki. A.. Perheentupa, J.. Anderson. L. C.. and Hayry. P.. Clirt. E.xp. I~nrmrnol. 32, 352. 1978. 4. Virolainen. M.. Savilahti, E.. Kaitila. 1.. and Perheentupa. J.. Pedirrtr. Re.s. 12, 961, 1978. 5. Lux. S. E.. Johnston, R. B.. August. C. S., Say. B.. Penchazadeh. V. B.. Rosen. E S.. and McKusick. V. A.. N. f%gI. J. Med. 282, 331. 1970. 6. Pierce. G. F.. and Polmar, S. H.. J. Irmmtr~ol. 129. 570. 1982. 7. Pierce. G. F.. and Polmar. S. H.. C/k. Exp. bnn7w7ol. 50, 621. 1982. 8. Pierce. G. F.. Brovall. C.. Schacter. B. Z.. and Polmar, S. H.. J. Clitt. It7wsr. 71, 1737. 1983. 9. Glader. B. E., Backer, K.. and Diamond. L. K.. N. En,q/. J. Med. 309. 1486, 1983. IO. Ferrarini, M.. and Grossi. C. E.. Swnir7. Hrmrrtol. 21, 270. 1984. I I. Kaitila. I.. Personal communication. I?. Roder. J. C., Haliotis. T.. Klein. M.. Korec. S.. Jett. J. R.. Ortaldo. J.. Herberman. R. B., Katz, P.. and Fauci. A. S.. N~?r(re (Londor71 284, S53, 1980. 13. Herherman. R. B., and Ortaldo. J. R.. Science 214. 14. 1981.