Circulating Immune Complexes in Patients with Cystic Fibrosis

Circulating Immune Complexes in Patients with Cystic Fibrosis* Joseph A. Church, M.D.,t Stanley C. Jordan, M.D.;:j: T'lwmo.s G. Keens, M.D.;§ and Chun-1. Wang, M.D.II

Circulating immune complexes (CICs) were measured iD

the sera of clinicaUy stable and acutely infected patients with cystic fibrosis (CF). Twenty CF patients were stndied wben dinicaUy stable; an additional 18 patients were stndied during an acute exacerbation of pnlmonary infection as evidenced by fever, tachypnea, increased white blood ceO count, increased sputnm production, and acnte chest x-ray film changes. Three methods for determining CICs were employed: polyethylene glycol precipitation, a C1q solid phase assay, and the Raji ceO radioimmnne assay. Ten of 20 clini-

A !though the primary pathogenetic event in cystic

fibrosis is unknown, the pulmonary pathology appears to be associated with chronic infection. 1•2 Previous investigations suggest that systemic specific For editorial comment see page 389

and nonspecific immune responses are normal in these patients. a.s On the contrary, cystic fibrosis patients often have elevated serum immunoglobulin concentrations and high levels of circulating antibodies to their infecting organisms.8•7 Circulating immune complexes ( CICs) are associated with a variety of chronic diseases characterized by immunologic hyperreactivity and are known to mediate tissue damage. 8•9 The purposes of this study were to determine the presence of CICs in clinicaily stable and acutely infected patients with cystic fibrosis and to evaluate and compare these assays for CICs in these patients. SUBJECI'S AND METHODS

Serum for immune complex and complement determinations •From the Department of Pediatrics, University of Southern California School of Medicine, and the Children's Haspital of Los Angeles, Los Angeles. tAssistant Professor of Pediatrics, Division of Allergy-Clinical Immunology. tAssistant Clinical Professor of Pediatrics, Dialysis and Transplantation. §Assistant Professor of Pediatrics, Neonatology-Respiratory Diseases.

!!Associate Professor of Pediatrics, Cystic Fibrosis Program. This study was funded in part by the Allergy Research Fund and the Research Program, Cbildrens Haspital of Los Angeles. Reprint reqvests: Dr. Church, Children's Hospital of Los Angeles, PO Bo:t 54700, Los Angeles 90054

CHEST, 80: 4, OCTOBER, 1981

caDy stable CF patients bad one or two positive assays for CICs; two of 20 bad two positive ~RBys. In contrast, 16 of 18 acutely infected CF patients bad a positive CIC test, and 12 of these were positive with two or tbree of the assays employed. Serum C3 and C4 concentrations and total hemolytic complement activity did not correlate with the presence of CICs in either patient group. These findings suggest that immune complex formation may mediate some of the tissne damage characteristic of CF, althongb this usuaDy does not involve intravascular complement activation.

was obtained from 20 clinically stable cystic fibrosis patients. All patients were older than eight years, and the mean age ( ±SD) was 17 ± six years. All patients were afebrile, with no clinical or laboratory indications of acute infection. An additional 18 patients, aged 16 ± six years, were similarly studied during an acute exacerbation of pulmonary infection as evidenced by fever, tachypnea, increased white blood cell count, increased sputum production, and acute chest x-ray film changes. Eleven of these required hOSPitalization. Control sera were obtained from healthy young adult laboratory and hoSPital personnel. Serum was separated from clotted blood within one hour of venipuncture, aliquoted, and stored at -70"C until used for individual laboratory determinations. Serum availability limited the performance of assays in some patients.

Detection of CICs by Polyethylene Glycol (PEG) Precipitation

This method of CIC determination depends on the differential solubility of CICs and other serum proteins.1o,n To 0.1 ml of each test serum was added 0.1 ml of a cold 8 percent PEG solution in phosphate buffered saline (PBS) (final concentration of PEG, 4 percent). This mixture was incubated for one hour at 4"C. The precipitate was washed once with cold 4 percent PEG in PBS, and the washed precipitate was solubilized in 0.1 ml of PBS. The concentrations of IgC in an aliquot of the original test serum and in the solubilized PEG precipitate were then measured with kinetic nephelometry (Beckman Immunochemistry System, Beckman Instruments, Fullerton, Calif). The percentage of IgG precipitated in 4 percent PEG was calculated with the formula: S IgG precipitated

=

[IgG] in solubilized PEG precipitate X 100 [IgG] in test serum

CIRCULATING IMMUNE COMPLEXES IN CF 405

Detection of CIC1 (&;f-.IUA)

br/ the '&Jfl Cell Htulloimmune Allarf

Raji cells were cultured in Eagle's minimal essential media ( MEM), with cell viability determined by trypan blue exclusion. Human sera were assayed for CICs as follows: 2 X IO' Raji cells in 200 "g MEM were reacted with 2.5 "g of a I:4 dilution of test serum. After an incubation period of 45 minutes at 37"C with periodic shaking, the cells were washed three times with MEM and then reacted for 30 minutes at 4"C with an optimum amount of I:2 dilution of 125J goat antihuman lgG in MEM containing I percent bovine serum aJbmnin (BSA) (I percent MEM-BSA). The cells were then washed three times with I percent MEMBSA, and the radioactivity of the cell pellets was determined by counting in a gamma counter. The amount of uptake was then referred to a standard curve of radioactive antibody uptake by cells incubated with 2.5 "g of I :4 dilution of normal human sera {NHS), or a pool of NHS as a source of complement. to which various amounts of aggregated human IgG ( AHG) were added. The mixture of NHS to which increasing amounts of AHG were added was preincubated for 30 minutes at 37"C before being added to the Raji cells. The amount of CICs in the sera tested was expressed as "g of AHG-equivalent per ml of serum.1• Detection of CICa

br/

Clq Solid Phase AIICifl (Clq-SPA)

The method of Ciq-SPA was a modification of the method described by Hay et al.1a C1q was isolated from NHS by the method of Yonemasu and Stroud,U with the substitution of relative salt concentrations of 0.4 M and 0.78 M for the 6rst and second precipitation steps, respectively. The purity of the isolated C1q was confirmed by immunoelectrophoresis using rabbit antisera to whole human sera and to human C1q. The assay was pedormed by incubating 50 1£) of test serum with 100 1£) 0.2 m ethylenediamine tetracetic acid (EDTA), pH 7.5, for 30 minutes at 37"C. Sixty microliters of the EDTA-treated serum was added to C1q-coated tubes in 1.0 ml PBS. Tests were pedormed in triplicate. The tubes were incubated for 60 minutes at 37"C and then for 20 hours at 4"C. At the end of the incubation period, the tubes were washed three times with PBS. Approximately 2.5 "g of 12 51 goat antihuman IgG anbbody was added to each tube in 1 ml of 1 percent BSA-PBS, and incubated for 60 minutes at 37"C and 30 minutes at 4"C. Tubes were washed three times in PBS, and radioactivity bound to the tubes was counted in a gamma counter. The results were expressed as percentage of counts bound to the tubes.

fu:suLTS

Results of the PEG precipitation assays for CICs in 17 normal control subjects, 17 clinically stable cystic fibrosis patients, and 16 cystic fibrosis patients during acute exacerbations are illustrated in Figure 1. The mean percentage of IgG precipitated from the sera of cystic fibrosis patients, clinically stable and acutely infected, is significantly greater than that from control sera. Positivity for CICs using the PEG precipitation assay was arbitrarily defined as a test result greater than 2 SD above the mean value for concurrently performed controls. Tables 1 and 2 detail the results of the assays for CICs in individual cystic fibrosis patients. Seven of 17 clinically stable cystic fibrosis patients had CICs by the PEG precipitation method; 12 of 16 acutely infected cystic fibrosis patients had positive tests. Mean total serum IgG concentrations did not differ between clinically stable and acutely infected patients, and percent IgG precipitated did not correlate with serum IgG concentrations in either group. Raji cell-RIA for CICs were performed on the sera from 16 clinically stable cystic fibrosis patients and concurrently on the sera of ten healthy adult subjects. Similarly, 18 acutely infected cystic fibrosis patients were studied. again concurrently with ten control subjects. Figure 2 illustrates the results of these tests. The mean "g AHG-equivalent/ml found in the sera of clinically stable cystic fibrosis patients did not differ significantly from that noted in control sera. However, cystic fibrosis patients 20

~

Serum complement components C3 and C4 were measured with standard radial in1munodiffusion plates (Meloy Laboratories, Springfield, Va). Total hemolytic complement activity ( CH 50 ) was determined with a standard assay employing hemolysin-sensitized sheep red blood ceiis.u

Statiatical Analy.ns Comparisons were made between normal and CF groups and between the two CF groups using analysis of variance and independent t tests. Correlations of positive results with the three CIC assays were statistically tested using x 2 analysis.

•

CHURCH ET AL

16 12

~

10

g

at

•

14

~

(!)

Determination of C3 and C4 Concentrations and Total Hemolytic Complement CH10 Actit>ity

•• •

18

• •

0

II.I....!..L-

• •

• • •

:

.I.

•

STABLE CF PATIENTS

CF PATIENTS

6.4--

6 2

• •

•

8 4

•• ••

• 3.1~

I

•

•

NORMAL

CONTROLS t •

•

2.316

INFECTED

3.276

p. (.025 (.005 FicvBE 1. Results of polyethylene glycol precipitation assays for circulating in1mune complexes in normal adult controls, clinically stable CF patients, and CF patients during acute infectious exacerbations.

CHEST, 80: 4, OCTOBER, 1981

Table 1-Auay•/or Cireuladqlmmune Complesa -.tJ Serum C3, C4, -.tJ Hernol7de Complemem (CHrso) Leeeh in Clinieall,- S1t16le Patie"" ath C,-llic l'i6rwil• C4,

CH.., units

91

36

111

103

42

135

86

21

152

112

34

132

100

66

152

88

36

172

105

38

102

59

22

102

110

34

128

10

106

36

161

11

122

nd

263

+

128

48

192

nd

93

48

102

14

112

48

167

15

86

42

132

89

46

192

106

48

200

100

52

172

132

68

102

99

54

208

Patient No.

PEG

Raji

1

nd

+

2

+

C3, mg/dc mg/dc

Clq

3

+

4 5

8

+ + +

9

nd

6

7

12

nd

+

nd

+

13

+

16 17

+

18

nd

19

+

+

nd

20 Number of assays positive Number of assays performed

7

4

1

17

20

16 105-220

55-120

Normal values and ranges

•Assays for circulating immune complexes: PEG-polyethylene glycol precipitation assay; Raji-Raji cell radioimmune assay; Clq-Clq solid phase assay; and nd=not done. 70

during acute infectious exacerbation have significantly greater pg AHG-equivalent/ml in their sera than do control subjects. A test was considered positive for CICs by the Raji-RIA when the result was greater than 2 SD above the mean value for concurrent control sera. Of the 18 clinically stable cystic fibrosis patient tests, four were positive (Table 1). In contrast, during acute infection, 12 of 18 cystic fibrosis patients had CICs by this technique (Table 2). Cfcs as determined with the C1q-SPA were measured in 16 stable cystic fibrosis patients and concurrently in ten control subjects. Similarly, 18 acutely infected cystic fibrosis patients and ten conFIGURE 2. Results of Raji cell radioimmune assays for circulating immune complexes in clinically stable CF patients compared with concurrent normal adult control subjects, and in CF patients during acute infectious exacerbations compared with concurrent normal adult control subjects.

CHEST, 80: 4, OCTOBER, 1981

65

60 ~

55

• .I.

If 50 ~

0

45

§

40

~

30

~

25 20

e

~

a ~

0

.••

35

...

al-+-

2~.8--

.

42.0-!--

1

•

•:• ·•·

.

15 10

5

0~----~------~------~----~--NORMAL CONTROLS

I P

• •

STAEl.E

a: ~TIENTS

0.222 0.80 (N.S.)

NORMAL CXlNTROLS

INFECTED 0: PATIENTS

5.636 <.001

CIRCULATING IMMUNE COMPLEXES IN CF 407

Table 2--.4...,.. /or Cireulad,.. lrnrnu- Cornplesa _ , Serurn CS, C4, _ , Hernolyrie Complement (CHrso) 'LewJ& in .4eulelyln/eeled Pllliem. willa Cyarie Fibroaia • Patient No.

PEG

Raji

Clq

21 22

++ +

23

28

+++ ++ + +++ ++

29

nd

24 25 26 27

30 31 32 33 34 35

+ + ++ + + +

+

++ +

+ +

38

nd

+ ++ +

Number of assays positive

12

12

7

Number of assays performed

16

18

18

36 37

Normal values and ranges

C3, mg/dc

C4, mg/dc

CH10, units

nd

nd

nd

166

43

227

112

40

172

94

30

156

103

68

208

108

23

nd

169

102

nd

94

38

185

84

58

222

61

17

102

87

19

93

130

39

222

101

52

135

200

98

278

93

32

185

110

61

172

90

26

50

105

30

98

55-120

20-50

105-220

*Assays for circulating immune complexes: PEG=-polyethylene glycol precipitation assay; Raji=Raji cell radioimmune assay;

Clq = Clq solid phase assay; and nd =not done.

current control subjects were tested. As shown in Figure 3, there is no difference in the mean percentage of radioactivity bound to the solid phase between either group of cystic fibrosis patients and the respective control subjects. A test was considered positive for CICs by the C1q-SPA when the result was greater than 2 SD above the mean value for concurrent control sera. Only one of 16 clinically stable cystic fibrosis patients had CICs by this technique (Table 1); seven of 18 acutely infected cystic fibrosis patients had CICs with this assay. Ten of 20 clinically stable cystic fibrosis patients had one or more positive assays for CICs, and two of 20 had two assays positive. In contrast, 16 of 18 acutely infected cystic fibrosis patients had a positive CIC test, and 12 of these were positive with two or three of the assays employed. The x" analyses of the results of the PEG precipitation, Raji cell radioimmune, and C1q-SPA assays failed to reveal positive correlations when -

CHURCH ET AL

any two of these assays for CIC detection were compared. Clinically stable and acutely infected cystic fibrosis patients did not differ in their serum C3 and C4 concentrations and total hemolytic complement activities (Fig 4). These levels generally were within the range of normal, although one acutely infected cystic fibrosis patient had very low total hemolytic complement activity. Further, there was no association between the presence of CICs and low serum complement levels, and elevated complement levels were noted in sera with CICs by two or three assays. DISCUSSION

Systemic immune responses in patients with cystic fibrosis are normal or hyperreactive, and immunologic hyperreactivity is commonly noted in disease states with CICs. That CICs may mediate the tissue damage in cystic fibrosis was suggested in 1975 when McFarlane et al18 described deposits of

CHEST, 80: 4, OCTOBER, 1981

100

~ 0

z

~~

90 70

~ ~

•

60

>O 50 tid 40

gcn

•

80

I

•••

47.8-J.-

44.9--$-••

•

I

I 45.4T • •

39.3-..J..-

•

30

••

20 10 0

t p

NORMAL

STABLE

NORMAL

INFECTED

CONTROLS

CF PATIENTS

CONTROLS

CF PATIENTS

. .

1.084

1.608

0.30(N.S.)

0.12 (N.S.)

immunoglobulin and complement in a variety of tissues from cystic fibrosis patients, including lung, liver, pancreas, and thymus, but not kidney. Strunk et aP 7 reported serum complement component C3 and C4 depression during viral lower respiratory tract infection in cystic fibrosis. Those authors postulated that antigen-antibody complex activation of complement occurred, but they did not test for CICs. Using a complement consumption assay, Schif6tz et al18 found CICs in cystic fibrosis patients, and those infected with Pseudomonas had a much high-

FtcVBE 3. Results of Clq solid phase assays for circulating immune complexes in clinically stable CF patients compared with concurrent normal adult control subjects, and in CF patients during acute infectious exacerbations compared with concurrent normal adult control subjects.

er incidence of CICs than those not so infected. Also in this study, immunofluorescent staining of skin biopsy specimens revealed granular deposits of lgM at the dermal-epidermal junction in 17 of 21 cystic fibrosis patients, although this finding appeared unrelated to Pseudomonas infection. In an extension of this work, immune complexes were detected with complement consumption and solid phase rheumatoid factor binding assays in sputum from Pseudomonas-infected cystic fibrosis patients. te Moss and Lewiston20 found CICs in 51 percent of cystic fibrosis patients tested during acute exacerba-

.-300

98;_102

~280

!'260

•

0

~240

• 0

•

6e

180

160 15 140

=

CHEST, 80: 4, OCTOBER, 1981

160.3

1120 @ 100 ~

80

~

60 40

0

!200

~

0

16 lb

---:.:e.

.

153.9

112.2

~P--·

t•

S

t:6

I I I I

I

:~ 61 •A:

i

*f t: •

;oeqo

• 1 - 43.1 10A

•

:~

6

45.6 - 1I 6

••

0

I I

• I

•

6

:66

6e:

0

F'IcVBE 4. Serum CH50 activities and C3 and C4 concentrations in acutely infected CF patients (I) compared with those of clinically stable CF patients ( S). Vertical rectangles indicate normal ranges for the respective test results. Horizontal bars and numbers indicate mean group values. Open circles = no circulating immune complexes ( CICs) detected with any method. Triangles = CICs detected with CICs deone assay. Closed circles tected with two or three of the assays employed.

:

•

s

L......J

1014

• • •

C3

soi 0 ~

40

I I I

:

s

L......J

0.372 1.154 0~ > ______________________ __o_.7_o_.
CH~

~

30

•

I

70 60

0

~:

§

6 0

20

~

I ~

.& ~

10 0

C4

CIRCULATING IMMUNE COMPLEXES IN CF -

ff

tions or routine clinic visits. Those investigators used a Clq binding assay and discerned no correlation between a positive test for CICs and Pseudomonas colonization or antibody response, and no evidence of in vivo complement activation. Recently, Berdischewski et al21 demonstrated CICs in 18 of 20 hospitalized CF patients with a PEG precipitation assay. 21 Analysis of the precipitated complexes revealed enrichment of antibodies against Pseudomonas lipopolysaccharide. In the present investigation, three techniques were employed to detect CICs. These studies confirm a high incidence of CICs in clinically stable cystic fibrosis patients; and patients during acute infectious exacerbation have an even greater incidence of CICs. That these assays correlate poorly with one another has been noted previously,8 and has been attributed to the different immunochemical or physical properties of CICs in various disorders. The PEG precipitation method as used in these studies depends on the differential solubility in 4 percent polyethylene glycol solution between aggregated and native IgG. That positive PEG assays for CICs did not reflect nonspecific aggregation is suggested by the lack of correlation between individual PEG assay results and serum IgG concentrations. The Raji cell-RIA detects CICs by quantitating immune complexes which have already activated complement and which bind to Raji cell C3b and C3a receptors. This assay is very sensitive and appears to be selective for immune complexes in the liS to 19S range. The Clq-SPA is based on binding of Clq to antigen-complexed IgG1, lg~. IgGs and lgM. It is more sensitive to immune complexes 9S to liS in size and it does not detect complexes composed of noncomplement binding immunoglobulin, such as lgG4 and IgA. These three assays detect different families of CICs that might not be detected if only one assay were used, and the ClqSPA appears less sensitive than the PEG precipitation assay or the Raji cell-RIA in detecting CICs in stable and acutely infected patients with cystic fibrosis. Depression of C3, C4 or CH50 levels did not correlate with the presence of CICs. In addition, serum from clinically stable and acutely infected cystic fibrosis patients generate normal chemotactic responses, a complement-dependent function, when activated with zymosan.1·a Normal levels of specific complement proteins and total hemolytic complement activities may falsely suggest lack of complement activation if synthetic rates of these components are increased. Further, specific assays for detection of activation products or turnover studies would be required to rule out complement involve410 CHURCH ET AL

ment in cystic fibrosis. However, the present studies do not support intravascular complement activation in the pathogenesis of this disorder. Although there may be a subset of patients with immune complexes who get sick more frequently, the present studies suggest that clinically stable and acutely infected CF patients are discrete populations. A prospective study of the development of CICs in CF patients will be necessary to clarify the role of CICs in this disorder. The detection of CICs in a high proportion of acutely infected cystic fibrosis patients and the previously noted finding of immunoglobulin and complement deposits in the lungs of these patients suggest that immune complex formation mediates at least some of the tissue damage characteristic of cystic fibrosis. ACKNOWLEDGMENT: We gratefully acknowledge the expert technical assistance of Rebecca Sakai, Carol Nye, the Childrens Hospital of Los Angeles Blood Bank under the supervision of Jacque Dellinger, and the excellent manuscript preparation of Geri Coldewey.

I Lamb D, Reid L. The tracheobronchial submucosa glands in cystic fibrosis, a qualitative and quantitative histochemical study. Br J Dis Chest 1972; 66:239-47 2 Oppenheimer EH. Tracheal and bronchial mucus glands and epithelium in infants with cystic fibrosis and controls. CF Club 19th Meeting, Atlanta, CF Foundation, 1978 3 Talamo RC, Stiehm ER, Schwartz RH. Immunologic aspects of cystic fibrosis. In: Mangos JA, Talamo RC, eds. Cystic fibrosis: projections into the future. New York: Stratton Intercontinental, 1976 4 Conover JH, Conod EJ, Hirschhorn K. Complement components in cystic fibrosis. Lancet 1973; 2:1501 5 Churoo JA, Keens TG, Wang C, et al. Normal neutrophil and monocyte chemotaxis in patients with cystic fibrosis. J Pediatr 1979; 95:272-74 6 Bums MW, May JR. Haemophilus influenzae precipitins in the serum of patients with chronic bronchial disorders. Lancet 1967; 1:354-58 7 Hoiby N, Axelsen NH. Identification and quantitation of precipitins against Pseudomonas aeruginosa in patients with cystic &brosis by means of crossed immunoelectrophoresis with intermediate gel. Acta Pathol Microbiol Scand 1973; 81:298-308 8 Barnett EV, moderator. Circulating immune complexes: their immunochemistry, detection and importance. Ann Intem Med 1979; 91:430-40 9 Theofilopoulos AN. Evaluation and clinical significance of circulating immune complexes. Prog Clio Immunol 1980; 4:63-106 10 Barnett EV, Chia D. Quantification of immunoglobulin G in serum and immune complexes isolated in polyethylene glycol. Ann Rheum Dis 1973; 36( suppl 1) :26 11 Digeon M, Laver M, Riza J, Bach JF. Detection of circulating immune complexes in human sera by simplified assays with polyethylene glycol. J Immunol Methods 1977; 16:165-83 12 Theofilopoulos AN, Wilson CB, Dixon FJ. The Raji cell radioimmune assay for detecting immune complexes in

CHEST, 80: 4, OCTOBER, 1981

human sera. J Clin Invest 1977; 57: 169-82 13 Hay FC, Ninehma L, Roitt IM. Routine assay for the detection of immune complexes of known immunoglobulin class using solid phase C1q. Clin Exp Immunol 1976; 24:396-400

14 Yonemasu K, Stroud RM. C1q: rapid puriflcation method for preparation of monospecific antisera and for biochemical studies. J Immunol 1971; 106:304-13 15 Ruddy S, Austen KF. Complement and its components. In: Cohen AS, ed. Laboratory diagnostic procedures in the rheumatic diseases. Boston: Little, Brown and Co, 1975 16 McFarlane H, Holzel A, Brenchley P, et al. Immune complexes in cystic fibrosis. Br Med J 1975; 1:423-28 17 Strunk RC, Sieber OF, Taussig LM, Gall EP. Serum complement depression during viral lower respiratory tract illness in cystic fibrosis. Arch Dis Child 1977; 52:

CHEST, 80: 4, OCTOBER, 1981

687-90 18 Schillttz PO, Hoiby N, Juhl F, et al. Immune complexes in cystic fibrosis. Acta Pathol Microbiol Scand 1977; 85:57-64 19 Sohil\tz PO, Nielsen H, Hoiby N, et al. Immune complexes in the sputum of patients with cystic fibrosis suffering from chronic Pseudomonas aeruginosa lung infection. Acta Pathol Microbiol Scand 1978; 86:37-40 20 Moss RB, Lewiston NJ. Immune complexes and humoral responses to Pseudomonas aeruginosa in cystic fibrosis. Am Rev Respir Dis 1980; 121 :23-29 21 Berdischewsky M, Pollack M, Young LS, et al. Circulating immune complexes in cystic fibrosis. Pediatr Res 1980; 12:830-33 22 Church JA, Keens TG, Wang C. Neutrophil and monocyte chemotaxis in acutely infected patients with cystic fibrosis. Ann Allergy 1980; 45:217-19

CIRCULAnNG IMMUNE COMPLEXES IN CF 411