Interferon production in primary immunodeficiencies

Journal of Clinical Immunology, Vol. 4, No. 5, 1984

Interferon Production in Primary Immunodeficiencies P. M. MATRICARDI, 1 M. R. CAPOBIANCHI, 2 R. PAGANELLI, 1 J. FACCHINI, 2 M. C. SIRIANNI, 1 R. SEMINARA, ~ F. DIANZANI, 2 and F. AIUTI 1

Accepted: April 26, 1984

INTRODUCTION

Alpha- and gamma-interferon (IFN) production by peripheral blood mononuclear cells (PBMC) from 18 patients affected by primary immunodeficiency syndromes was examined and compared with that of 20 normal donors. Patients included 8 with common variable immunodeficiency (CVI), 2 with congenital agammaglobulinemia, 4 with ataxia-telangiectasia, 2 with hyper-IgE syndrome, 1 with chronic EBV infection, 1 with combined immunodeficiency, and 1 with immunodeficiency with hyper-IgM. No spontaneous IFN production was observed in either patients and controls. Newcastle disease virus-induced alpha-IFN production was found to be normal in all patients. Gamma-IFN was induced by both galactose oxidase and staphylococcal enterotoxin (B). Gamma-interferon production was low or undetectable in patients with ataxia-telangiectasia, in immunodeficiency with hyper-IgM, and in hyper-IgE syndrome. No major defect of gamma-IFN was found in other types of immunodeficiency, despite the presence of occasional low producers (1 of 8 CVI patients and 1 case of congenital agammaglobulinemia).No correlation was found between IFN production and natural killer activity in individual patients. The analysis of lymphocyte subsets by monoclonal antibodies revealed gross imbalances of helper/inducer and suppressor/cytotoxic subpopulations, but no overall correlation could be established with gamma-IFN production. The observation of major defects in gammaIFN yield only in diseases with depression of T cellmediated immunity might contribute to a better understanding of the pathogenetical mechanisms in these diseases. Moreover, future studies should monitor these in vitro functions and their modifications by in vitro or in vivo manipulations. KEY WORDS: Immunodeficiency;gamma-interferon;alpha-interferon; monoclonalantibodies.

The interferon (IFN) system consists of a series of proteins newly produced and secreted by body cells stimulated by foreign substances. These IFN proteins react with other cells to induce new effector molecules, which in turn lead to specific biochemical effects resulting in antiviral action, immunoregulation, and antiproliferative activity (1). Among these proteins, alpha-IFN is produced by leukocytes and induced by viruses, xenogenic or tumor cells, bacteria, and B-cell mitogens. Immune or gamma-IFN is produced by T lymphocytes stimulated by specific antigens or T-cell mitogens (2) and macrophages are required for induction. Lymphoid cells also carry a preformed mRNA for a gammalike IFN whose translation is normally prevented by a rapid turnover mechanism which requires ongoing RNA synthesis (3). Pathogenetic mechanisms of primary immunodeficiency syndromes are still largely unknown, although there is evidence that functional defects of T lymphocytes may be related to defective production of soluble mediators, as we have previously reported for interleukin-2 (IL-2) (4). Since gamma-IFN has been reported to exert several immunoregulatory effects on T, B, and natural killer (NK) cells (5-7) and occasional defects of IFN production (either alpha, gamma, or not characterized) have been found in infectious, autoimmune, and miscellaneous immunodeficiency disorders (8-17), we analyzed spontaneous as well as stimulated alpha- and gamma-IFN release in vitro by mononuclear cells from patients with primary immunodeficiencies. These patients were classified according to WHO criteria (18) and compared to matched healthy controls. An additional goal of our study was the possibility to correlate abnormal secretion of IFN with functional defects and imbalances of lymphocyte subpopulations de-

1Department of Allergology and Clinical Immunology, University of Rome "La Sapienza," Rome, Italy. Zlnstitute of Virology.

388 0271-9142/84/0900-0388503.50/0 © 1984 Plenum Publishing Corporation

389

INTERFERON IN IMMUNODEFICIENCIES

fined by monoclonal antibodies to distinctive surface antigens.

MATERIALS AND METHODS Patients and Controls. Nineteen patients (9 males and 10 females aged between 3 and 52 years) were studied: diagnosis was established according to WHO criteria (18). The cases included common variable immunodeficiency (CVI) (8 cases), congenital agammaglobulinemia (2 cases), ataxia-telangiectasia (4 cases), hyper-IgE syndrome (2 cases), chronic EBV infection (1 case), combined immunodeficiency (CID) (1 case), and immunodeficiency with hyper-IgM (1 case). Blood samples from agammaglobulinemic patients were taken on the 21st day after iv infusion of -/-globulin. The CID case was studied while under thymic hormone therapy with the synthetic pentapeptide TP-5. Diagnosis was previously established as described (18). Ten of ten hypogammaglobulinemia patients had upper respiratory tract infections, four had bronchiectasias, and one had intestinal giardiasis. Viral infections were observed in four other cases. During the follow-up one case of ataxia-telangiectasia died of severe viral hepatitis, and one patient with hyper-IgE syndrome died a few months later of epithelioma of the skin. The other patient with hyper-IgE syndrome had severe recurrent herpetic keratitis and suffered from interstitial pneumonia and chronic cutaneous candidiasis. The experiments reported here were performed when the patients were free from infections. Twenty age-matched donors were examined as healthy controls. Paired specimens from patients and controls were processed and tested under identical conditions on the same day. Isolation of Peripheral Blood Mononuclear Cells (PBMC). Heparinized venous blood was collected from patients and controls; PBMC were isolated by Ficoll-Hypaque gradient sedimentation, washed three times, and resuspended at 10 6 cells/ml in RPMI 1640 (GIBCO) containing 10% heat-inactivated fetal calf serum (Flow). Monoclonal Antibodies. Total T lymphocytes were identified by reactivity of PBMC with OKT3 monoclonal antibody, by indirect immunofluorescence as previously described (19). The same method was applied to quantitation of T-lymphocyte subsets by monoclonal antibodies to the helper/inducer (OKT4) or cytotoxic/suppressor (OKT8) phe-

Journal of Clinical Immunology, Vol. 4, No. 5, 1984

notypes. NK cells were quantitated by monoclonal antibody HNK-1 (20). NK Activity and PHA Response. NK activity was tested as previously described (20) on the K 562 cell line, by 51Cr release in a standard cytotoxic assay, and results are expressed as the percentage of lysis at an effector/target ratio of 100:1. The in vitro lymphoproliferative response to phytohemagglutinin (PHA) was measured in 3-day microcultures by tritiated thymidine uptake, as previously described (4), and results are expressed as counts per minute (cpm). IFN Induction. Alpha-IFN induction was performed with Newcastle disease virus (NDV) 10 HA/106 cells: after 1 hr cells were washed, resuspended at 106/ml, and kept at 37°C. Supernates were collected after 20 hr and kept at pH 3 for 48 hr before titrations. Gamma-IFN was induced either with staphylococcal enterotoxin B (SEB) (Sigma) (0.2 ixg/ml/106 cells) or with galactose oxidase (GO; Worthington Biochemicals) (10 U/107) for 30 min at 24°C as previously described (21, 22). After GO treatment, cells were washed, resuspended at 106/ml, and kept at 37°C. Supernates were collected after 20 hr and titrated immediately. Interferon Titration. Supernates were titrated for IFN activity on human WISH cells by Sindbis virus hemagglutination yield reduction after a single growth cycle (22). A laboratory standard of alphaIFN was included in the titrations (1 U of alpha-IFN equals 1 IU of alpha-IFN standard; gamma-IFN values are expressed as laboratory units). The antiviral activity found in the samples was identified as alpha- or gamma-IFN according to the current criteria including neutralization with specific antisera (23). The antisera used were sheep anti-IFN alpha (Shering Corp., Bloomfield, N J, lot No. 08093) and P2 monospecific anti-human IFN gamma (Immunomodulators Laboratories, Inc., TX). Lymphocytes of each patient were tested with control lymphocytes on at least two separate occasions. Neuraminidase Treatment. Neuraminidase treatment of PBMC was performed as previously described (22). Viability of the cells after treatment was tested by the trypan blue dye uptake method and was found to exceed 85%. RESULTS Alpha-IFN production in all patients and controls was equal to or above 3000 IU/ml, without signifi-

390

MATRICARDI, CAPOBIANCHI, PAGANELLI, FACCHINI~ SIRIANNI, SEMINARA, DIANZANI, AND AIUTI

cant differences between the two groups. No relevant production of spontaneous alpha- and gammaIFN was observed in either controls or patients• Gamma-IFN yields obtained by PBMC from controls, donors, and immunodeficiency patients are reported in Fig. 1. The gamma-IFN production after GO stimulation in 16 healthy donors ranged from 100 to 3162 U/ml, with an arithmetic mean of 741. This broad range of production is the result of individual, not technical variability, as repeated experiments on the same donors gave almost identical results (data not shown)• On the contrary, gamma-IFN yields of congenital agammaglobulinemia, combined immunodeficiency, and chronically EBV infected patients fell at the lower levels of the control range• Gamma-IFN production by CVI patients covered a broad spectrum, varying from 30 to 3000 U/ml. The four cases of ataxia-telangiectasia, the two cases of hyper-IgE syndrome, and the patient with immunodeficiency with hyper-IgM gave very low or undetectable gamma-IFN yields (Fig. 1). In order to determine whether the low gammaIFN yields stimulated by GO treatment were due to the masking of D-galactose residues by N-acetyl neuraminic acid, we pretreated the lymphocytes with neuraminidase in some cases before GO induction. Among low producers, normal gamma-IFN production was restored only in one patient with CVI (data not shown).

Average IFN production after the two induction procedures described was calculated to compare IFN yields from patients and normal donors (Fig. 2). When expressed as the percentage of gammaIFN produced by simultaneously tested matched controls, our data showed that immunodeficiency with hyper-IgM and ataxia-telangiectasia patients were defective gamma-IFN producers (P < 0.001 by Student's t test); combined immunodeficiency, hyper-IgE, and congenttal agammaglobulinemic patients and the case of chronic EBV infection were intermediate gamma-IFN producers, while CVI patients were mostly normal P = NS). All patients had almost-normal levels of NK activity, as well as NK cells, detected by the monoclonal antibody HNK-1 (Leu-7). No statistically significant correlation was found between gamma-IFN production and NK activity. Ataxiatelangiectasia, hyper-IgE syndrome, and immunodeficiency with hyper-IgM patients had a low lymphoproliferative response to PHA, with a range of 420-11,620 cpm, compared to control values of above 18,000 cpm. Analysis of T-lymphocyte subsets by monoclonal antibodies OKT3, OKT4, and OKT8 showed no major abnormalities in hypogammaglobulinemic patients, while ataxia-telangiectasia and hyper-IgE patients had low OKT3 and OKT4 levels, with the OKT4/OKT8 ratio grossly below normal values (1.7 -+ 0.3) (Table I).

SEB[a] [~]

G O[b]

100cO

E E O 0

50,

z

Hyper IgM[1]

A.T.[4]

CID[1]

Hyper Ig E [21

Conqenital agar~nlaglob.

CVl[8]-

[ a ] = Staphylococcal Enterotoxin B [ b] = Galactose oxidase Fig. l. Gamma-IFN production after GO treatment by PBMC from patients and controls.

Journal o f Clinical Immunology, Vol. 4, No. 5, 1984

391

INTERFERON IN iMMUNODEFICIENCIES

300

........

CONTROLS •

20009

II

•

000

1000

Z

PATIENTS~

i

500

u-

Ill

300 2(X

10(

Ai

I•

i 0

A

i

A

20

0

10

oi__

oo¢~

Gatactose- oxidase induction

O A T. z~ Hyper t 9 E I

CVt

tD with hyper IgM

O Chronic E.B.V ,i, Congenital agammaglob. i, CID • Controls

Fig. 2. Comparison of patients' gamma-IFN production with that of healthy donors. Arithmetic means of gamma-IFN production for each group of patients are referred to as the percentage of the normal mean (made equal to 100%). Normal arithmetic mean gamma-IFN production after GO (16 healthy donors) and SEB (20 healthy donors) stimulation was, respectively, 741 (range, 100-3162) and 2127 (range, 316-10000) IU/mt.

DISCUSSION In the present study We investigated both spontaneous and induced alpha- and gamma-IFN production by comparing PBMC from 19 patients with primary immunodeficiencies to those of normal donors. However, the possibility of discriminating between Npha- and gamma-IFN at the induction level and the characterization of the end product

Journal of Clinical Immunology, Vol. 4, No. 5, 1984

were not always shown. We aimed at the identification of the type of IFNs produced after in vitro selective induction of PBMC from any given patient. No spontaneous IFN production was observed in any of the donors tested. No absolute or relative defects were found in NDV-induced alpha-IFN by lymphocytes from either healthy donors or immunodeficient patients. Gamma-IFN production stimulated by SEB or GO treatment was clearly defective or undetectabte in the patient with hyper-IgM, in four with ataxiatelangiectasia, and in two with hyper4gE syndrome. Pretreatment of PBMC with neuraminidase did not restore gamma-IFN production in low producers, demonstrating that their low response was not due to an excess of sialic acid on the cell membrane, which may mask galactose residues needed for induction (24). A moderate deficiency of gamma-IFN production was observed in congenital agammaglobulinemia, CID, and chronic EBV infection. The case of combined immunodeficiencyhad been on long-term thymic hormone therapy with the synthetic pentapeptide TP-5, which proved to be effective in restoring cell-mediated immunity and NK activity as previously reported (25). CVI patients had normal gamma-IFN production with only two exceptions, and in one of them neuraminidase treatment fully restored gamma-IFN production. Other authors have previously described studies of in vitro IFN production in various diseases, and moderate or marked defects were found in some patients (8-I7). Spontaneous production of gammaIFN and acid-labile alpha-IFN has been described (26, 27). Low production of alpha-IFN has been described in some diseases, including congenital cytomegalovirus infections (8), chronic hepatitis (9, 10), multiple sclerosis (11), and systemic lupus erythematosus (12). In addition, a primary defect of alpha-IFN has been reported in children with recurrent respiratory tract infections (13), and a combined defect of gamma- and alpha-IFN has been reported in a case of moderate SCID and a case of cartilage hair syndrome with immunodeficiency (14). An IFN-like activity induced by Raji cells in culture has been observed to be low or absent in children with chronic EBV infection, severe viral and bacterial infections, or partial albinism, whose only ascertained immune defect was low natural killer (NK) activity (15). A defect of gamma-IFN

392

MATRICARDI, CAPOBIANCHI, PAGANELLI, FACCHINI, SIRIANNI, SEMINARA, DIANZANI, AND AIUTI

Table I. PBMC Surface Markers and NK Activity in Primary Immunodeficiency Patients and Controls Patient

Diagnosis

% T3

% T4

% T8

% HNK-1

NK % lysis -

A.G. P.F. B.C. M.S. A.F. F.A. D.A. O.R. F.F. T.M. M.G. A.O. G.A. M.T. C.M. D.P. B.A. D.A. M.M. Controls

Hyper-IgE Hyper-IgE ID with hyper-lgE Combined ID Ataxia-telang. Ataxia-telang. Ataxia-telang. Ataxia-telang. Chronic EBV Congenital agammaglob. Congenital agammaglob. CVI CVI CVI CVI CVI CVI CVI CVI

42 52 71 43 54 36 40 33 63 72 56 ND 63 52 72 67 80 58 72 63 -+ 6

19 37 10 50 13 18 29 48 36 65 41 38 40 39 33 21 52 40 48 45 -+ 3

14 30 61 19 25 40 21 13 43 15 24 28 37 36 ND 47 36 19 24 27 -+ 4

20 20 24 18 47 8 6 ND 9 4 10 32 6 1 19 13 37 24 10 Ad,