Flow cytometric determination of circulating immune complexes with ...

0 1985 Alan R. Liss, Inc.

Cytometry 6316-320 (1985)

Flow Cytometric Determination of Circulating Immune Complexes With the Indirect Granulocyte Phagocytosis Test L.W.M.M. Terstappen, B.G. de Grooth, G.M.J. Nolten, C.H.H. ten Napel, W. van Berkel, and J. Greve Department of Applied Physics, Twente University of Technology (L.W.M.M.T., B.G.d.G., G.M.J.N., J.G.) and Department of Internal Medicine, Ziekenzorg Hospital (C.H.H.t.N., W.V.B.) Received for publication March 7, 1984; accepted February 3,1985

A method for the determination of circulating immune complexes (CIC) was adapted for flow cytometric analysis. Human granulocytes were used to phagocytose I&-bearing CIC of serum from systemic lupus erythematosus (SLE) patients. A method for labeling the phagocytosed CIC with FITC-conjugated anti-human IgG was developed where the granulocytes remain in suspension during fixation and labeling. The fluorescence per cell, Quantitative determination of the amount of circulating immune complexes (CIC) in human blood can be of great diagnostic and prognostic value for patients with auto-immunological disorders, infectious diseases (1,3,4,9,15), and some malignant diseases (5,6,12,18). It appears that a reproducible and rapid method for the determination of CIC is useful. One of the methods currently in use is the indirect granulocyte phagocytosis test (11,17,19). In this method the patient’s serum is added to a suspension of granulocytes isolated from a healthy donor. Under carefully controlled conditions, the CIC of the patient’s serum are phagocytosed by the granulocytes. After drying on a microscope slide, the granulocytes are fixed. IgG-bearing CIC can then be labeled with FITC-conjugated anti-human IgG antibodies, whereupon the phagocytosed CIC are visible as small fuorescent inclusions in the cells. The number of inclusions per cell is taken as a measure for the relative amount of CIC in the patient’s serum. This microscopic analysis is time-consuming, which greatly limits a more frequent use of this technique. Therefore we have adapted the granulocyte phagocytosis test so that the fluorescent inclusions of the granulocytes can be measured by flow cytometry. This greatly reduces the time required for analysis. MATERIALS AND METHODS Preparation of Granulocytes donors. A ‘loo‘ (6.25 m/ml heparin) Of Of 12*5m1 Of the huffy ‘Oat was to 50 m1 with buffer (0.11 M NaCl, 1.0 mM NaH2P04.H20, 7.3 mM

measured with a flow cytometer, is a measure of the total amount of the phagocytosed IgG. The results indicate that a rapid and quantitative method for the detection and measurement of phagocytosed CIC is possible using the flow cytometer. Key terms: Flow cytometry, phagocytosis, circulating immune complexes, granulocytes Na2HP04, 0.5% [w/v] bovine serum albumin (BSA) and 0.38% [w/v] sodium citrate, pH 7.4). The sample was then divided between two tubes. For venous blood 12.5 ml of blood was diluted to 25 ml with the same buffer. In each tube we carefully layered 15ml Percoll (Pharmacia Fine Chemicals, Uppsala, Sweden) solution (density 1.077 g/cm3, suspended in 0.2 M NaC1, 10 mM NaH2P04*H20,4.4 mM Na2HP04, 27.5 mM sodium citrate and 1%[w/v] BSA) with a syringe under the cell suspensions. After centrifugation at 20°C for 20 min at 1,000 x g, granulocytes and erythrocytes were pelleted and separated from platelets and mononuclear leucocytes by the Percoll layer. The pellet was then resuspended in 35 ml of 0.155 M NH4C1 and 9.9 mM KHCO3 to lyse the erythrocytes. After 20 min gentle shaking on ice the solution was centrifuged at 4°C for 5 min at 500 x g. The pellets were diluted with 10 ml of a solution of 0.9%(w/v) NaCl and 2.5 W/ml heparin in order to avoid clotting. The cells were centrifuged at 20°C for 5 min at 500 g, and the pellet was resuspended in Hanks’ balanced salt solution (HBSS) containing 5% (w/v) BSA. The concentration was adjusted to lo7 celldml. The overall recovery of the granulocytes was about 70%. The viability of the granulocytes was more than 99% as measured by flow cytometry using the FDA-ethidium bromide method. Phagocytosis Assay

Address reprint requests to L.W.M.M.Terstappen, Twente University of Technology, Department of Applied Physics, P.O. Box 217,7500 AE Enschede, The Netherlands.

FLOW CYTOMETRIC DETECTION OF IMMUNE COMPLEXES

donor with 0.5 ml serum of a systemic lupus erythematosus (SLE)patient. To obtain the complement necessary for phagocytosis, we added 0.125 ml fresh serum of the granulocyte donor. This mixture was incubated for 60 min at 37°C while being shaken to promote contact between granulocytes and CIC. Then the cells were pelleted and washed twice with 2 ml phosphate-buffered saline (PBS), centrifuged for 0.5 min at 700 x g, and resuspended in 0.1 ml PBS. While being shaken (vortexed), the cells were fixed in formaldehyde by adding 0.15 ml 10% formaldehyde buffered in PBS (PH 7.2) at 1-min intervals during 10 min at room temperature. Thereafter the cells were shaken during 15 min in the fixative and washed twice with PBS. Phagocytosed IgGbearing CIC were demonstrated as inclusions within the granulocytes using a direct immunofluorescence method. The pellet was suspended in 1 ml of a solution of FITC labeled goat anti-human IgG, F(ab’)2 fragment (Kallestad, Austin, TX) (1:lOO diluted with HBSS containing 1%[w/v] BSA). The cells were then incubated for 60 rnin a t room temperature while being shaken, then they were washed with PBS and resuspended in 1 ml HBSS containing 5% (wh) BSA. The cell suspension was analysed within 2 h by fluorescence microscope and flow cytometry. Fluorescence Microscopy For fluorescence microscopy we used a Nikon microscope (type Optiphot) equipped with a n Osram HBO 50W Mercury lamp, a 480-500-nm excitation filter, and 515-nm long-pass eyepiece side-absorption filter. Flow Cytometry The flow cytometer was built in our laboratory. It contained a flow cell with a 250 x 250-pm square flow channel (Precision Cells, Hicksville, NY). For the phagocytosis assay, a 3-watt argon-ion laser (model CR3, Coherent Radiation, Palo Alto, CA) tuned at 488nm was used as a light source. The small-angle forward light scatter was detected with a photodiode (model Pin 10-D, United Detector Technology). Fluorescence emission at 90“ was focused by means of a Leitz microscope object-lens (H32, NA 0.6) through a KV 520 and OG 530 filter combination (Scott) onto a Hamamatsu R 928 photomultiplier. The signals were analysed with a homemade multichannel analyser connected to a LSI 11/23 minicomputer (Digital Equipment Corporation). For each experiment at least 20,000 cells were measured.

RESULTS In the microscopic determination of phagocytosed CIC currently in use, the granulocytes are fixed and stained while they are on a microscope slide (3,18,19).Fixation of the granulocytes is necessary in order to allow the fluorescent stain to pass through the cell membrane. The fixation procedure used for microscopic analysis cannot be used for flow cytometry since cells must remain in suspension. In order to fix cells in suspension,

317

background fluorescence and clotting of cells must be avoided, whereas cell morphology must be preserved. Thus, we have tried several fixatives at different concentrations: water-free acetone, ethanol, formaldehyde, and glacial acetic acid. To minimize cell clotting the fixatives were added slowly to the cell suspension under continuous shaking. We have obtained the best results with a 10%solution of formaldehyde. The flow-cytometric indirect granulocyte phagocytosis assay was tested with sera from five SLE patients as positive samples and sera from five healthy donors as negative controls. The sera of the SLE patients were scored in the Clq-binding assay (10) by the Blood Transfusion Service of the Netherlands (Amsterdam). In this test, soluble high molecular immune complexes are measured. Microscopic examination revealed that all granulocytes showed faint green cytoplasmic fluorescence both for positive samples and controls. Stained IgG-bearing immune complexes (IC) were visible as fluorescent inclusions (diameter 1-2 pm). The results obtained from flow cytometric analysis of serum from a n SLE patient with a high score in the Clq-binding assay and those obtained with serum of a healthy donor are shown in Figures 1 and 2. In Figure 1, the small-angle light scatter and fluorescence histograms of granulocytes incubated with serum of the control and serum of the SLE patient are plotted. The figure clearly shows that the cells incubated with SLE serum have a much higher fluorescence intensity than those incubated with the control serum. In Figure 2 the same results are presented as twodimensional dot plots. Both figures reveal the presence of a minor subpopulation of cells with a very faint fluorescence. In Table 1the results of the flow cytometric determination are given for five SLE patients and five healthy donors. Granulocytes with a fluorescence intensity above channel 35 (see Fig. 2) are designated as positive cells. When fixation was omitted, less than 1%of the cells were positive using patient serum and less than 0.1% using control serum. To test the day-to-day and sample-to-sample variation, the assay was carried out with one positive serum and granulocytes of a single donor on different days. The variation in the percentage of positive cells was less than 10%. The results of the granulocyte phagocytosis test may be dependent on the donor granulocytes. Therefore we have applied the test to serum of a single SLE patient and a single healthy donor, using granulocytes of four different healthy donors. The results are given in Table 2. To determine the origin of the fluorescence of the granulocytes incubated with control serum C‘background” fluorescence), we have applied the following variations to our standard procedure: 1)When the complete procedure was followed except for the final immunofluorescence staining, the fluorescence of the granulocytes was negligible. Thus the background fluorescence was not due to autofluorescence of granulocytes fixed by formal-

3 18

TEFGTAPPEN ET AL.

Serum of a healthy donor Number of cells (a.u.1

: : # : l' Number of cells 1a.u.l

0

0

LO 80 Small angle light scotter

0

0

LO

80 fluorescence

0

LO

a0 fluoresceoce

Serum of an SLE patient

; :#: Number of cells lo u.1

Number of cells (a.u 1

0l

.

O

L

z

C

I

l

0

LO 80 Small angle light scatter

0

Fig. 1. The upper curves show the results aRer incubating with serum from a healthy donor, the lower with serum from a n SLE patient. The fluorescence is proportional to the total amount of IgG-

bearing IC. The t y e s clearly show that the SLE serum contained more CIC than the serum from the healthy donor.

Serum of a healthy donor 80 7

Serum of an SLE patient.

1 8o

E W

u.

0

+ LO

r--.

'

0

80 Fluorescence

0

f

LO

80 FIuorescence

Fig. 2. Two-dimensional dot plots of the same results as in Figure 1. Granulocytes with a fluorescence intensity above channel 35 (f) are designated as positive cells. The figurea clearly show the difference

between the fluorescence intensity of granulocytes incubated with healthy donor serum (4% positive cells) and SLE serum (38% positive cells).

dehyde. 2) When the procedure was followed without furation, the median fluorescence dropped to less than 10%. Therefore the background fluorescence is mainly due to the combination of fixation and staining. 3) Omission of the incubation with the control serum did not significantly change the background fluorescence (see Table 2). This suggests that the background fluorescence is caused by IgG already present in the donor granulocytes before incubation with serum. Interestingly, we have noticed a significant difference between patient and control when the granulocyte phag-

ocytosis assay was carried out without fixation. Under these conditions serum of a patient with SLE showed a median fluorescence intensity, which was about 20% of that obtained with the standard procedure. With control sera this value was less than 10%.as stated above.

DISCUSSION We have demonstrated that flow cytometric determination of CIC with the indirect granulocyte phagocytosis test is possible. Sera of SLE patients and controls differ significantly with respect to the percentage of granulo-

3 19

FLOW CYTOMETRIC DETECTION OF IMMUNE COMPLEXES

Table 2 Flow Cytometric Determination of CIC for One SLE Patient

Table 1 Flow Cvtometric Determination o f CIC for SLE Patients and Healthy Ddnors ' Healthy donor

1 2 3 4 5

Control sera Positive cells (%) Clq(%) 4 3 4

5 4

aNormalvalues of C l q

a a

a a a

SLE sera Positive Patients cells (%) 1 2 3 4 5

14 38 10 14 8

and One Healthy Donor Using Four Different Healthy Granulocyte Donors

Granulocytes

not incubated

Clq(%)

90 87 53 25

16

< 7%.

positive cells

Control sera positive cells

SLE serum positive cells

(%I

(%I

(9%)

3 2 3 2

54 34 53

with sera

Granulocyte donor

1 2 3 4

3 2 2 N.D."

46

"N.D., not done.

cytes with a high fluorescence intensity (Figs. 1, 2, and Table 1).The high fluorescence of positive cells is mainly due to intracellular IgG. This can be concluded from the observation that the percentage of positive cells is negligible when we omitted the fixation procedure that is necessary to introduce FI'I'C-labeled anti-IgG into the cell. Whether or not the phagocytose determination of CIC can be used as a rapid clinical method depends on the sensitivity and reproducibility of the test. The sensitivity of the test is limited by the background fluorescence of cells incubated with control sera. From our results we may conclude that the background fluorescence is due to intracellular FITC-labeled goat anti-human IgG F(ab') 2 fragments. These fragments may be bound nonspecifically to intracellular components or specifically to intracellular IgG. In the latter case this IgG was already present in the donor granulocytes before phagocytosis, since incubation with control serum did not increase the background fluorescence (see Table 2). The sensitivity of the assay will then be determined by the amount of IgG present in the donor granulocytes. If, on the other hand, the background is caused by nonspecifically bound antihuman IgG, a n increase of the sensitivity can be expected if a monoclonal antibody is used instead of the polyclonal antibody used in the present study. The reproducibility of the test is not limited by the sample-to-sample or day-to-day variation, but by the observed variation when granulocytes of different donors are used (Table 2). This problem can be met through the use of standard sera for calibration. The results in Table 2 indicate that sera with a positive score in the Clq-binding assay were also positive with the indirect granulocyte phagocytosis assay, although significant differences in the level of the scores are observed. In general, there is a poor correlation between the several detection methods for CIC (7,8,13,14,16,20,21). It is interesting to note that the difference obtained for control and patient sera with unfxed granulocytes suggests that a procedure without fixation may also be used for detection of CIC. In this case membrane-bound immune complexes are detected. The results shown in Figures 1 and 2 indicate the existence of two populations of cells. A minor fraction of

about 15% shows very little fluorescence, whereas the remaining cells have more fluorescence. In view of the isolation procedure used, and, since both fractions revealed the same small-angle light scatter intensity (Fig.21, both cell populations must be granulocytes. This is in accordance with our microscopic observations that 10-20% of the granulocytes showed no inclusions, and with the recent observations of Bassoe et al. (2), who also observed a peak with low fluorescence due to nonphagocytes. It should be noted, however, that the relative number of nonphagocytes depends on the donor granulocytes used. Compared with the microscopic analysis, the flow cytometric method is less time-consuming and can be more quantitative. However, one should be aware of the differences between the two methods. With the microscopic method, the total number of inclusions are counted, whereas the flow cytometric method determines the intensity of the fluorescence per cell, which is a measure of the total amount of IC per cell. In order to find objective criteria from which the score of the test can be determined, a more elaborate study is needed. In such a study, standard positive and negataive sera will be used in order to eliminate the influence of the use of granulocytes of different donors, and reliable quantitative data would then be obtained. Also, a comparison of the flow cytometric method and other methods for detection of CIC of patients with different diseases is required.

ACKNOWLEDGMENTS The authors wish to thank Dr. H.R. de Vries for supplying the b&y coats, E. Masseus for skillful laboratory work, and Dr. M. van der Giessen for helpful criticism and advice.

LITERATURE CITED 1. Barnett EV, Knutson DW, Abrass CK, Chia DS, Young LS: Circu-

lating immune complexes: Their immunochemistry,detection, and importance. Ann Intern Med 91:430-440, 1979. 2. Bassoe CF,Laerum OD, Glette J, Hopen G, Haneberg B, Solberg CO: Simultaneous measurement of phagocytosis and phagosomal pH by flow cytometry: Role of polymorphonuclear neutrophilic leukocyte granules in phagosome acidification.Cytometry 4:254262, 1983. 3. Cats A, Lafeber GJM, Klein F: Immunoglobulin phagocytosis by granulocytes from sera and synovial fluids in various rhematoid and nonrhematoid diseases. Ann Rheum Dis 34:146-155,1975. 4. Dowd PM, Kirby JD,Holborow EJ, Cooke ED, Bowcock SA: Detec-

320

TERSTAPPEN ET AL.

tion of immune complexes in systemic sclerosis and Raynaud’s phenomenon. Br J Dermatol 105:179-188,1981. 5. Friedrich A, Schraning B, Sengebusch B, Glass W In-vitro-Untersuchungen der Granulozytenphagozytose bei Patienten mit malignen Tumoren. Folia Haematol 110:402412,1983. 6. Friedrich A, Schaning B, Glass W Zum Wert des Granulozytenphagozytosetests in der onkologischen Nachsorge Geschwulstforsch 53:139-152,1983. 7. Halla JT,Volanakis JE,Schrohenloher RE: Immunecomplexes in rheumatoid-arthritis sera and synovial fluids. Arthritis Rheum 22:440448,1979. 8. Halla JT,Volanakis JE,Schrohenloher RE: Circulating immune complexes in mixed connective-tissue disease. Arthritis Rheum 22:484-489, 1979. 9. Hurd ER, Andreis M, Ziff M Phagocytosis of immune complexes by polymorphonuclear leucocytes in patients with Felty’s syndrome. Clin Exp Immunol28413425,1977. 10. Hurd ER, Jasin HE, Gilliam JN: Correlation of disease activity and Clq-binding immune complexes with the neutrophil inclusions which form in the presence of SLE sera. Clin Exp Immunol 40283-291,1980, 11. Hurd E R Use of neutrophil phagocytosis for detection of immune complexes. Contr Nephrol 35:128-133, 1983. 12. Jansen HM, The TH, de Gast GC,Huiges HA, Esselink MT, van der Wal AM, Orie NGM Immunoglobulin and complement inclusions with peripheral blood polymorphonuclear leuocytes of patients with bronchial carcinoma. Thorax 32306-710, 1977. 13. Lambert, PH, Dixon FJ, Zubler RH, Agnello V, Cambiaso C, Casali P, Clarke J, Cowdery JS, McDuffie FC, Hay FC: WHOcollaborative study for evaluation of 18 methods for detecting immune-

complexes in serum. J Clin Lab Immunol 1:l-15,1978. 14. Macuel SE, Tappeiner G, Brumfield, H: Circulating immune complexes in cutaneous vasculitus-detection with C l Q and monoclonal rheumatoid-factor. J Clin Invest 641652-1660, 1979. 15. Meulen van der J. Wouda AA, Mandema E, The TH: Immune complexes in peripheral blood polymorphonuclear leucocytes of patients with Raynaud’s phenomenon. Clin Exp Immunol 355266, 1979. 16. Nydegger UE, Davis JS, Rossen RD: Soluble immunecomplexes in human disease (critical review). Clin Lab Sci 12 (2),123-170, 1980. 17. Steffelaar JW, ten Kate JW, Nap M, Swaak JG, de Graaf-Reitsema CB, van Elven EH, Feltkamp-Vroom TM:Immune complex detection by immunofluorescence on polymorphonuclear leucocytes. Clin Exp Immunol27,391-396,1977. 18. The TH, van der Giessen M, Huiges HA, Schrdordt Koops H, van Wingerden I Immune complexes in peripheral blood polymorphonuclear leucocytes of malignant melanoma patients. Clin Exp Immunol32387-391,1978. 19. Van Wingerden I, The TH, van der Giessen M, Rumke PH: Demonstration of circulating immune complexes in melanoma patients by means of a granulocyte phagocytosis test. In: Protides of the Biological Fluids, Proceedings of the Colloquium, Peeters H (ed), Vol. 26 Pergamon, Oxford, 1979, pp 345-348. 20. Whitsed H, McCarthy WH, Hersey P Nephelometric detection of circulating immunecomplexes using monoclonal rheumatoid factor. J Immunol Meth 29:311-321,1979. 21. Yoshinoya S, Pope RM: Detection of immunecomplexes in acute rheumatic-fever and their relationship to HLA-B5. J Clin Invest 65:136-145, 1980.